diff options
297 files changed, 21375 insertions, 5723 deletions
diff --git a/.gitignore b/.gitignore index de6344e15706..efab0ebec859 100644 --- a/.gitignore +++ b/.gitignore @@ -36,6 +36,7 @@ modules.builtin # tags TAGS +linux vmlinux vmlinuz System.map diff --git a/Documentation/ABI/stable/sysfs-devices-node b/Documentation/ABI/stable/sysfs-devices-node new file mode 100644 index 000000000000..49b82cad7003 --- /dev/null +++ b/Documentation/ABI/stable/sysfs-devices-node @@ -0,0 +1,7 @@ +What: /sys/devices/system/node/nodeX +Date: October 2002 +Contact: Linux Memory Management list <linux-mm@kvack.org> +Description: + When CONFIG_NUMA is enabled, this is a directory containing + information on node X such as what CPUs are local to the + node. diff --git a/Documentation/cpu-freq/pcc-cpufreq.txt b/Documentation/cpu-freq/pcc-cpufreq.txt new file mode 100644 index 000000000000..9e3c3b33514c --- /dev/null +++ b/Documentation/cpu-freq/pcc-cpufreq.txt @@ -0,0 +1,207 @@ +/* + * pcc-cpufreq.txt - PCC interface documentation + * + * Copyright (C) 2009 Red Hat, Matthew Garrett <mjg@redhat.com> + * Copyright (C) 2009 Hewlett-Packard Development Company, L.P. + * Nagananda Chumbalkar <nagananda.chumbalkar@hp.com> + * + * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or NON + * INFRINGEMENT. See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License along + * with this program; if not, write to the Free Software Foundation, Inc., + * 675 Mass Ave, Cambridge, MA 02139, USA. + * + * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + */ + + + Processor Clocking Control Driver + --------------------------------- + +Contents: +--------- +1. Introduction +1.1 PCC interface +1.1.1 Get Average Frequency +1.1.2 Set Desired Frequency +1.2 Platforms affected +2. Driver and /sys details +2.1 scaling_available_frequencies +2.2 cpuinfo_transition_latency +2.3 cpuinfo_cur_freq +2.4 related_cpus +3. Caveats + +1. Introduction: +---------------- +Processor Clocking Control (PCC) is an interface between the platform +firmware and OSPM. It is a mechanism for coordinating processor +performance (ie: frequency) between the platform firmware and the OS. + +The PCC driver (pcc-cpufreq) allows OSPM to take advantage of the PCC +interface. + +OS utilizes the PCC interface to inform platform firmware what frequency the +OS wants for a logical processor. The platform firmware attempts to achieve +the requested frequency. If the request for the target frequency could not be +satisfied by platform firmware, then it usually means that power budget +conditions are in place, and "power capping" is taking place. + +1.1 PCC interface: +------------------ +The complete PCC specification is available here: +http://www.acpica.org/download/Processor-Clocking-Control-v1p0.pdf + +PCC relies on a shared memory region that provides a channel for communication +between the OS and platform firmware. PCC also implements a "doorbell" that +is used by the OS to inform the platform firmware that a command has been +sent. + +The ACPI PCCH() method is used to discover the location of the PCC shared +memory region. The shared memory region header contains the "command" and +"status" interface. PCCH() also contains details on how to access the platform +doorbell. + +The following commands are supported by the PCC interface: +* Get Average Frequency +* Set Desired Frequency + +The ACPI PCCP() method is implemented for each logical processor and is +used to discover the offsets for the input and output buffers in the shared +memory region. + +When PCC mode is enabled, the platform will not expose processor performance +or throttle states (_PSS, _TSS and related ACPI objects) to OSPM. Therefore, +the native P-state driver (such as acpi-cpufreq for Intel, powernow-k8 for +AMD) will not load. + +However, OSPM remains in control of policy. The governor (eg: "ondemand") +computes the required performance for each processor based on server workload. +The PCC driver fills in the command interface, and the input buffer and +communicates the request to the platform firmware. The platform firmware is +responsible for delivering the requested performance. + +Each PCC command is "global" in scope and can affect all the logical CPUs in +the system. Therefore, PCC is capable of performing "group" updates. With PCC +the OS is capable of getting/setting the frequency of all the logical CPUs in +the system with a single call to the BIOS. + +1.1.1 Get Average Frequency: +---------------------------- +This command is used by the OSPM to query the running frequency of the +processor since the last time this command was completed. The output buffer +indicates the average unhalted frequency of the logical processor expressed as +a percentage of the nominal (ie: maximum) CPU frequency. The output buffer +also signifies if the CPU frequency is limited by a power budget condition. + +1.1.2 Set Desired Frequency: +---------------------------- +This command is used by the OSPM to communicate to the platform firmware the +desired frequency for a logical processor. The output buffer is currently +ignored by OSPM. The next invocation of "Get Average Frequency" will inform +OSPM if the desired frequency was achieved or not. + +1.2 Platforms affected: +----------------------- +The PCC driver will load on any system where the platform firmware: +* supports the PCC interface, and the associated PCCH() and PCCP() methods +* assumes responsibility for managing the hardware clocking controls in order +to deliver the requested processor performance + +Currently, certain HP ProLiant platforms implement the PCC interface. On those +platforms PCC is the "default" choice. + +However, it is possible to disable this interface via a BIOS setting. In +such an instance, as is also the case on platforms where the PCC interface +is not implemented, the PCC driver will fail to load silently. + +2. Driver and /sys details: +--------------------------- +When the driver loads, it merely prints the lowest and the highest CPU +frequencies supported by the platform firmware. + +The PCC driver loads with a message such as: +pcc-cpufreq: (v1.00.00) driver loaded with frequency limits: 1600 MHz, 2933 +MHz + +This means that the OPSM can request the CPU to run at any frequency in +between the limits (1600 MHz, and 2933 MHz) specified in the message. + +Internally, there is no need for the driver to convert the "target" frequency +to a corresponding P-state. + +The VERSION number for the driver will be of the format v.xy.ab. +eg: 1.00.02 + ----- -- + | | + | -- this will increase with bug fixes/enhancements to the driver + |-- this is the version of the PCC specification the driver adheres to + + +The following is a brief discussion on some of the fields exported via the +/sys filesystem and how their values are affected by the PCC driver: + +2.1 scaling_available_frequencies: +---------------------------------- +scaling_available_frequencies is not created in /sys. No intermediate +frequencies need to be listed because the BIOS will try to achieve any +frequency, within limits, requested by the governor. A frequency does not have +to be strictly associated with a P-state. + +2.2 cpuinfo_transition_latency: +------------------------------- +The cpuinfo_transition_latency field is 0. The PCC specification does +not include a field to expose this value currently. + +2.3 cpuinfo_cur_freq: +--------------------- +A) Often cpuinfo_cur_freq will show a value different than what is declared +in the scaling_available_frequencies or scaling_cur_freq, or scaling_max_freq. +This is due to "turbo boost" available on recent Intel processors. If certain +conditions are met the BIOS can achieve a slightly higher speed than requested +by OSPM. An example: + +scaling_cur_freq : 2933000 +cpuinfo_cur_freq : 3196000 + +B) There is a round-off error associated with the cpuinfo_cur_freq value. +Since the driver obtains the current frequency as a "percentage" (%) of the +nominal frequency from the BIOS, sometimes, the values displayed by +scaling_cur_freq and cpuinfo_cur_freq may not match. An example: + +scaling_cur_freq : 1600000 +cpuinfo_cur_freq : 1583000 + +In this example, the nominal frequency is 2933 MHz. The driver obtains the +current frequency, cpuinfo_cur_freq, as 54% of the nominal frequency: + + 54% of 2933 MHz = 1583 MHz + +Nominal frequency is the maximum frequency of the processor, and it usually +corresponds to the frequency of the P0 P-state. + +2.4 related_cpus: +----------------- +The related_cpus field is identical to affected_cpus. + +affected_cpus : 4 +related_cpus : 4 + +Currently, the PCC driver does not evaluate _PSD. The platforms that support +PCC do not implement SW_ALL. So OSPM doesn't need to perform any coordination +to ensure that the same frequency is requested of all dependent CPUs. + +3. Caveats: +----------- +The "cpufreq_stats" module in its present form cannot be loaded and +expected to work with the PCC driver. Since the "cpufreq_stats" module +provides information wrt each P-state, it is not applicable to the PCC driver. diff --git a/Documentation/device-mapper/snapshot.txt b/Documentation/device-mapper/snapshot.txt index e3a77b215135..0d5bc46dc167 100644 --- a/Documentation/device-mapper/snapshot.txt +++ b/Documentation/device-mapper/snapshot.txt @@ -122,3 +122,47 @@ volumeGroup-base: 0 2097152 snapshot-merge 254:11 254:12 P 16 brw------- 1 root root 254, 11 29 ago 18:15 /dev/mapper/volumeGroup-base-real brw------- 1 root root 254, 12 29 ago 18:16 /dev/mapper/volumeGroup-base-cow brw------- 1 root root 254, 10 29 ago 18:16 /dev/mapper/volumeGroup-base + + +How to determine when a merging is complete +=========================================== +The snapshot-merge and snapshot status lines end with: + <sectors_allocated>/<total_sectors> <metadata_sectors> + +Both <sectors_allocated> and <total_sectors> include both data and metadata. +During merging, the number of sectors allocated gets smaller and +smaller. Merging has finished when the number of sectors holding data +is zero, in other words <sectors_allocated> == <metadata_sectors>. + +Here is a practical example (using a hybrid of lvm and dmsetup commands): + +# lvs + LV VG Attr LSize Origin Snap% Move Log Copy% Convert + base volumeGroup owi-a- 4.00g + snap volumeGroup swi-a- 1.00g base 18.97 + +# dmsetup status volumeGroup-snap +0 8388608 snapshot 397896/2097152 1560 + ^^^^ metadata sectors + +# lvconvert --merge -b volumeGroup/snap + Merging of volume snap started. + +# lvs volumeGroup/snap + LV VG Attr LSize Origin Snap% Move Log Copy% Convert + base volumeGroup Owi-a- 4.00g 17.23 + +# dmsetup status volumeGroup-base +0 8388608 snapshot-merge 281688/2097152 1104 + +# dmsetup status volumeGroup-base +0 8388608 snapshot-merge 180480/2097152 712 + +# dmsetup status volumeGroup-base +0 8388608 snapshot-merge 16/2097152 16 + +Merging has finished. + +# lvs + LV VG Attr LSize Origin Snap% Move Log Copy% Convert + base volumeGroup owi-a- 4.00g diff --git a/Documentation/fault-injection/provoke-crashes.txt b/Documentation/fault-injection/provoke-crashes.txt new file mode 100644 index 000000000000..7a9d3d81525b --- /dev/null +++ b/Documentation/fault-injection/provoke-crashes.txt @@ -0,0 +1,38 @@ +The lkdtm module provides an interface to crash or injure the kernel at +predefined crashpoints to evaluate the reliability of crash dumps obtained +using different dumping solutions. The module uses KPROBEs to instrument +crashing points, but can also crash the kernel directly without KRPOBE +support. + + +You can provide the way either through module arguments when inserting +the module, or through a debugfs interface. + +Usage: insmod lkdtm.ko [recur_count={>0}] cpoint_name=<> cpoint_type=<> + [cpoint_count={>0}] + + recur_count : Recursion level for the stack overflow test. Default is 10. + + cpoint_name : Crash point where the kernel is to be crashed. It can be + one of INT_HARDWARE_ENTRY, INT_HW_IRQ_EN, INT_TASKLET_ENTRY, + FS_DEVRW, MEM_SWAPOUT, TIMERADD, SCSI_DISPATCH_CMD, + IDE_CORE_CP, DIRECT + + cpoint_type : Indicates the action to be taken on hitting the crash point. + It can be one of PANIC, BUG, EXCEPTION, LOOP, OVERFLOW, + CORRUPT_STACK, UNALIGNED_LOAD_STORE_WRITE, OVERWRITE_ALLOCATION, + WRITE_AFTER_FREE, + + cpoint_count : Indicates the number of times the crash point is to be hit + to trigger an action. The default is 10. + +You can also induce failures by mounting debugfs and writing the type to +<mountpoint>/provoke-crash/<crashpoint>. E.g., + + mount -t debugfs debugfs /mnt + echo EXCEPTION > /mnt/provoke-crash/INT_HARDWARE_ENTRY + + +A special file is `DIRECT' which will induce the crash directly without +KPROBE instrumentation. This mode is the only one available when the module +is built on a kernel without KPROBEs support. diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 31575e220f3b..a5cc0db63d7a 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -449,12 +449,6 @@ Who: Alok N Kataria <akataria@vmware.com> ---------------------------- -What: adt7473 hardware monitoring driver -When: February 2010 -Why: Obsoleted by the adt7475 driver. -Who: Jean Delvare <khali@linux-fr.org> - ---------------------------- What: Support for lcd_switch and display_get in asus-laptop driver When: March 2010 Why: These two features use non-standard interfaces. There are the diff --git a/Documentation/filesystems/00-INDEX b/Documentation/filesystems/00-INDEX index 875d49696b6e..5139b8c9d5af 100644 --- a/Documentation/filesystems/00-INDEX +++ b/Documentation/filesystems/00-INDEX @@ -62,6 +62,8 @@ jfs.txt - info and mount options for the JFS filesystem. locks.txt - info on file locking implementations, flock() vs. fcntl(), etc. +logfs.txt + - info on the LogFS flash filesystem. mandatory-locking.txt - info on the Linux implementation of Sys V mandatory file locking. ncpfs.txt diff --git a/Documentation/filesystems/logfs.txt b/Documentation/filesystems/logfs.txt new file mode 100644 index 000000000000..e64c94ba401a --- /dev/null +++ b/Documentation/filesystems/logfs.txt @@ -0,0 +1,241 @@ + +The LogFS Flash Filesystem +========================== + +Specification +============= + +Superblocks +----------- + +Two superblocks exist at the beginning and end of the filesystem. +Each superblock is 256 Bytes large, with another 3840 Bytes reserved +for future purposes, making a total of 4096 Bytes. + +Superblock locations may differ for MTD and block devices. On MTD the +first non-bad block contains a superblock in the first 4096 Bytes and +the last non-bad block contains a superblock in the last 4096 Bytes. +On block devices, the first 4096 Bytes of the device contain the first +superblock and the last aligned 4096 Byte-block contains the second +superblock. + +For the most part, the superblocks can be considered read-only. They +are written only to correct errors detected within the superblocks, +move the journal and change the filesystem parameters through tunefs. +As a result, the superblock does not contain any fields that require +constant updates, like the amount of free space, etc. + +Segments +-------- + +The space in the device is split up into equal-sized segments. +Segments are the primary write unit of LogFS. Within each segments, +writes happen from front (low addresses) to back (high addresses. If +only a partial segment has been written, the segment number, the +current position within and optionally a write buffer are stored in +the journal. + +Segments are erased as a whole. Therefore Garbage Collection may be +required to completely free a segment before doing so. + +Journal +-------- + +The journal contains all global information about the filesystem that +is subject to frequent change. At mount time, it has to be scanned +for the most recent commit entry, which contains a list of pointers to +all currently valid entries. + +Object Store +------------ + +All space except for the superblocks and journal is part of the object +store. Each segment contains a segment header and a number of +objects, each consisting of the object header and the payload. +Objects are either inodes, directory entries (dentries), file data +blocks or indirect blocks. + +Levels +------ + +Garbage collection (GC) may fail if all data is written +indiscriminately. One requirement of GC is that data is seperated +roughly according to the distance between the tree root and the data. +Effectively that means all file data is on level 0, indirect blocks +are on levels 1, 2, 3 4 or 5 for 1x, 2x, 3x, 4x or 5x indirect blocks, +respectively. Inode file data is on level 6 for the inodes and 7-11 +for indirect blocks. + +Each segment contains objects of a single level only. As a result, +each level requires its own seperate segment to be open for writing. + +Inode File +---------- + +All inodes are stored in a special file, the inode file. Single +exception is the inode file's inode (master inode) which for obvious +reasons is stored in the journal instead. Instead of data blocks, the +leaf nodes of the inode files are inodes. + +Aliases +------- + +Writes in LogFS are done by means of a wandering tree. A naïve +implementation would require that for each write or a block, all +parent blocks are written as well, since the block pointers have +changed. Such an implementation would not be very efficient. + +In LogFS, the block pointer changes are cached in the journal by means +of alias entries. Each alias consists of its logical address - inode +number, block index, level and child number (index into block) - and +the changed data. Any 8-byte word can be changes in this manner. + +Currently aliases are used for block pointers, file size, file used +bytes and the height of an inodes indirect tree. + +Segment Aliases +--------------- + +Related to regular aliases, these are used to handle bad blocks. +Initially, bad blocks are handled by moving the affected segment +content to a spare segment and noting this move in the journal with a +segment alias, a simple (to, from) tupel. GC will later empty this +segment and the alias can be removed again. This is used on MTD only. + +Vim +--- + +By cleverly predicting the life time of data, it is possible to +seperate long-living data from short-living data and thereby reduce +the GC overhead later. Each type of distinc life expectency (vim) can +have a seperate segment open for writing. Each (level, vim) tupel can +be open just once. If an open segment with unknown vim is encountered +at mount time, it is closed and ignored henceforth. + +Indirect Tree +------------- + +Inodes in LogFS are similar to FFS-style filesystems with direct and +indirect block pointers. One difference is that LogFS uses a single +indirect pointer that can be either a 1x, 2x, etc. indirect pointer. +A height field in the inode defines the height of the indirect tree +and thereby the indirection of the pointer. + +Another difference is the addressing of indirect blocks. In LogFS, +the first 16 pointers in the first indirect block are left empty, +corresponding to the 16 direct pointers in the inode. In ext2 (maybe +others as well) the first pointer in the first indirect block +corresponds to logical block 12, skipping the 12 direct pointers. +So where ext2 is using arithmetic to better utilize space, LogFS keeps +arithmetic simple and uses compression to save space. + +Compression +----------- + +Both file data and metadata can be compressed. Compression for file +data can be enabled with chattr +c and disabled with chattr -c. Doing +so has no effect on existing data, but new data will be stored +accordingly. New inodes will inherit the compression flag of the +parent directory. + +Metadata is always compressed. However, the space accounting ignores +this and charges for the uncompressed size. Failing to do so could +result in GC failures when, after moving some data, indirect blocks +compress worse than previously. Even on a 100% full medium, GC may +not consume any extra space, so the compression gains are lost space +to the user. + +However, they are not lost space to the filesystem internals. By +cheating the user for those bytes, the filesystem gained some slack +space and GC will run less often and faster. + +Garbage Collection and Wear Leveling +------------------------------------ + +Garbage collection is invoked whenever the number of free segments +falls below a threshold. The best (known) candidate is picked based +on the least amount of valid data contained in the segment. All +remaining valid data is copied elsewhere, thereby invalidating it. + +The GC code also checks for aliases and writes then back if their +number gets too large. + +Wear leveling is done by occasionally picking a suboptimal segment for +garbage collection. If a stale segments erase count is significantly +lower than the active segments' erase counts, it will be picked. Wear +leveling is rate limited, so it will never monopolize the device for +more than one segment worth at a time. + +Values for "occasionally", "significantly lower" are compile time +constants. + +Hashed directories +------------------ + +To satisfy efficient lookup(), directory entries are hashed and +located based on the hash. In order to both support large directories +and not be overly inefficient for small directories, several hash +tables of increasing size are used. For each table, the hash value +modulo the table size gives the table index. + +Tables sizes are chosen to limit the number of indirect blocks with a +fully populated table to 0, 1, 2 or 3 respectively. So the first +table contains 16 entries, the second 512-16, etc. + +The last table is special in several ways. First its size depends on +the effective 32bit limit on telldir/seekdir cookies. Since logfs +uses the upper half of the address space for indirect blocks, the size +is limited to 2^31. Secondly the table contains hash buckets with 16 +entries each. + +Using single-entry buckets would result in birthday "attacks". At +just 2^16 used entries, hash collisions would be likely (P >= 0.5). +My math skills are insufficient to do the combinatorics for the 17x +collisions necessary to overflow a bucket, but testing showed that in +10,000 runs the lowest directory fill before a bucket overflow was +188,057,130 entries with an average of 315,149,915 entries. So for +directory sizes of up to a million, bucket overflows should be +virtually impossible under normal circumstances. + +With carefully chosen filenames, it is obviously possible to cause an +overflow with just 21 entries (4 higher tables + 16 entries + 1). So +there may be a security concern if a malicious user has write access +to a directory. + +Open For Discussion +=================== + +Device Address Space +-------------------- + +A device address space is used for caching. Both block devices and +MTD provide functions to either read a single page or write a segment. +Partial segments may be written for data integrity, but where possible +complete segments are written for performance on simple block device +flash media. + +Meta Inodes +----------- + +Inodes are stored in the inode file, which is just a regular file for +most purposes. At umount time, however, the inode file needs to +remain open until all dirty inodes are written. So +generic_shutdown_super() may not close this inode, but shouldn't +complain about remaining inodes due to the inode file either. Same +goes for mapping inode of the device address space. + +Currently logfs uses a hack that essentially copies part of fs/inode.c +code over. A general solution would be preferred. + +Indirect block mapping +---------------------- + +With compression, the block device (or mapping inode) cannot be used +to cache indirect blocks. Some other place is required. Currently +logfs uses the top half of each inode's address space. The low 8TB +(on 32bit) are filled with file data, the high 8TB are used for +indirect blocks. + +One problem is that 16TB files created on 64bit systems actually have +data in the top 8TB. But files >16TB would cause problems anyway, so +only the limit has changed. diff --git a/Documentation/filesystems/nfs/nfs41-server.txt b/Documentation/filesystems/nfs/nfs41-server.txt index 1bd0d0c05171..6a53a84afc72 100644 --- a/Documentation/filesystems/nfs/nfs41-server.txt +++ b/Documentation/filesystems/nfs/nfs41-server.txt @@ -17,8 +17,7 @@ kernels must turn 4.1 on or off *before* turning support for version 4 on or off; rpc.nfsd does this correctly.) The NFSv4 minorversion 1 (NFSv4.1) implementation in nfsd is based -on the latest NFSv4.1 Internet Draft: -http://tools.ietf.org/html/draft-ietf-nfsv4-minorversion1-29 +on RFC 5661. From the many new features in NFSv4.1 the current implementation focuses on the mandatory-to-implement NFSv4.1 Sessions, providing @@ -44,7 +43,7 @@ interoperability problems with future clients. Known issues: trunking, but this is a mandatory feature, and its use is recommended to clients in a number of places. (E.g. to ensure timely renewal in case an existing connection's retry timeouts - have gotten too long; see section 8.3 of the draft.) + have gotten too long; see section 8.3 of the RFC.) Therefore, lack of this feature may cause future clients to fail. - Incomplete backchannel support: incomplete backchannel gss diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt index 0d07513a67a6..96a44dd95e03 100644 --- a/Documentation/filesystems/proc.txt +++ b/Documentation/filesystems/proc.txt @@ -164,6 +164,7 @@ read the file /proc/PID/status: VmExe: 68 kB VmLib: 1412 kB VmPTE: 20 kb + VmSwap: 0 kB Threads: 1 SigQ: 0/28578 SigPnd: 0000000000000000 @@ -188,6 +189,12 @@ memory usage. Its seven fields are explained in Table 1-3. The stat file contains details information about the process itself. Its fields are explained in Table 1-4. +(for SMP CONFIG users) +For making accounting scalable, RSS related information are handled in +asynchronous manner and the vaule may not be very precise. To see a precise +snapshot of a moment, you can see /proc/<pid>/smaps file and scan page table. +It's slow but very precise. + Table 1-2: Contents of the statm files (as of 2.6.30-rc7) .............................................................................. Field Content @@ -213,6 +220,7 @@ Table 1-2: Contents of the statm files (as of 2.6.30-rc7) VmExe size of text segment VmLib size of shared library code VmPTE size of page table entries + VmSwap size of swap usage (the number of referred swapents) Threads number of threads SigQ number of signals queued/max. number for queue SigPnd bitmap of pending signals for the thread @@ -430,6 +438,7 @@ Table 1-5: Kernel info in /proc modules List of loaded modules mounts Mounted filesystems net Networking info (see text) + pagetypeinfo Additional page allocator information (see text) (2.5) partitions Table of partitions known to the system pci Deprecated info of PCI bus (new way -> /proc/bus/pci/, decoupled by lspci (2.4) @@ -584,7 +593,7 @@ Node 0, zone DMA 0 4 5 4 4 3 ... Node 0, zone Normal 1 0 0 1 101 8 ... Node 0, zone HighMem 2 0 0 1 1 0 ... -Memory fragmentation is a problem under some workloads, and buddyinfo is a +External fragmentation is a problem under some workloads, and buddyinfo is a useful tool for helping diagnose these problems. Buddyinfo will give you a clue as to how big an area you can safely allocate, or why a previous allocation failed. @@ -594,6 +603,48 @@ available. In this case, there are 0 chunks of 2^0*PAGE_SIZE available in ZONE_DMA, 4 chunks of 2^1*PAGE_SIZE in ZONE_DMA, 101 chunks of 2^4*PAGE_SIZE available in ZONE_NORMAL, etc... +More information relevant to external fragmentation can be found in +pagetypeinfo. + +> cat /proc/pagetypeinfo +Page block order: 9 +Pages per block: 512 + +Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10 +Node 0, zone DMA, type Unmovable 0 0 0 1 1 1 1 1 1 1 0 +Node 0, zone DMA, type Reclaimable 0 0 0 0 0 0 0 0 0 0 0 +Node 0, zone DMA, type Movable 1 1 2 1 2 1 1 0 1 0 2 +Node 0, zone DMA, type Reserve 0 0 0 0 0 0 0 0 0 1 0 +Node 0, zone DMA, type Isolate 0 0 0 0 0 0 0 0 0 0 0 +Node 0, zone DMA32, type Unmovable 103 54 77 1 1 1 11 8 7 1 9 +Node 0, zone DMA32, type Reclaimable 0 0 2 1 0 0 0 0 1 0 0 +Node 0, zone DMA32, type Movable 169 152 113 91 77 54 39 13 6 1 452 +Node 0, zone DMA32, type Reserve 1 2 2 2 2 0 1 1 1 1 0 +Node 0, zone DMA32, type Isolate 0 0 0 0 0 0 0 0 0 0 0 + +Number of blocks type Unmovable Reclaimable Movable Reserve Isolate +Node 0, zone DMA 2 0 5 1 0 +Node 0, zone DMA32 41 6 967 2 0 + +Fragmentation avoidance in the kernel works by grouping pages of different +migrate types into the same contiguous regions of memory called page blocks. +A page block is typically the size of the default hugepage size e.g. 2MB on +X86-64. By keeping pages grouped based on their ability to move, the kernel +can reclaim pages within a page block to satisfy a high-order allocation. + +The pagetypinfo begins with information on the size of a page block. It +then gives the same type of information as buddyinfo except broken down +by migrate-type and finishes with details on how many page blocks of each +type exist. + +If min_free_kbytes has been tuned correctly (recommendations made by hugeadm +from libhugetlbfs http://sourceforge.net/projects/libhugetlbfs/), one can +make an estimate of the likely number of huge pages that can be allocated +at a given point in time. All the "Movable" blocks should be allocatable +unless memory has been mlock()'d. Some of the Reclaimable blocks should +also be allocatable although a lot of filesystem metadata may have to be +reclaimed to achieve this. + .............................................................................. meminfo: diff --git a/Documentation/gpio.txt b/Documentation/gpio.txt index 1866c27eec69..c2c6e9b39bbe 100644 --- a/Documentation/gpio.txt +++ b/Documentation/gpio.txt @@ -253,6 +253,70 @@ pin setup (e.g. controlling which pin the GPIO uses, pullup/pulldown). Also note that it's your responsibility to have stopped using a GPIO before you free it. +Considering in most cases GPIOs are actually configured right after they +are claimed, three additional calls are defined: + + /* request a single GPIO, with initial configuration specified by + * 'flags', identical to gpio_request() wrt other arguments and + * return value + */ + int gpio_request_one(unsigned gpio, unsigned long flags, const char *label); + + /* request multiple GPIOs in a single call + */ + int gpio_request_array(struct gpio *array, size_t num); + + /* release multiple GPIOs in a single call + */ + void gpio_free_array(struct gpio *array, size_t num); + +where 'flags' is currently defined to specify the following properties: + + * GPIOF_DIR_IN - to configure direction as input + * GPIOF_DIR_OUT - to configure direction as output + + * GPIOF_INIT_LOW - as output, set initial level to LOW + * GPIOF_INIT_HIGH - as output, set initial level to HIGH + +since GPIOF_INIT_* are only valid when configured as output, so group valid +combinations as: + + * GPIOF_IN - configure as input + * GPIOF_OUT_INIT_LOW - configured as output, initial level LOW + * GPIOF_OUT_INIT_HIGH - configured as output, initial level HIGH + +In the future, these flags can be extended to support more properties such +as open-drain status. + +Further more, to ease the claim/release of multiple GPIOs, 'struct gpio' is +introduced to encapsulate all three fields as: + + struct gpio { + unsigned gpio; + unsigned long flags; + const char *label; + }; + +A typical example of usage: + + static struct gpio leds_gpios[] = { + { 32, GPIOF_OUT_INIT_HIGH, "Power LED" }, /* default to ON */ + { 33, GPIOF_OUT_INIT_LOW, "Green LED" }, /* default to OFF */ + { 34, GPIOF_OUT_INIT_LOW, "Red LED" }, /* default to OFF */ + { 35, GPIOF_OUT_INIT_LOW, "Blue LED" }, /* default to OFF */ + { ... }, + }; + + err = gpio_request_one(31, GPIOF_IN, "Reset Button"); + if (err) + ... + + err = gpio_request_array(leds_gpios, ARRAY_SIZE(leds_gpios)); + if (err) + ... + + gpio_free_array(leds_gpios, ARRAY_SIZE(leds_gpios)); + GPIOs mapped to IRQs -------------------- diff --git a/Documentation/hwmon/adt7411 b/Documentation/hwmon/adt7411 new file mode 100644 index 000000000000..1632960f9745 --- /dev/null +++ b/Documentation/hwmon/adt7411 @@ -0,0 +1,42 @@ +Kernel driver adt7411 +===================== + +Supported chips: + * Analog Devices ADT7411 + Prefix: 'adt7411' + Addresses scanned: 0x48, 0x4a, 0x4b + Datasheet: Publicly available at the Analog Devices website + +Author: Wolfram Sang (based on adt7470 by Darrick J. Wong) + +Description +----------- + +This driver implements support for the Analog Devices ADT7411 chip. There may +be other chips that implement this interface. + +The ADT7411 can use an I2C/SMBus compatible 2-wire interface or an +SPI-compatible 4-wire interface. It provides a 10-bit analog to digital +converter which measures 1 temperature, vdd and 8 input voltages. It has an +internal temperature sensor, but an external one can also be connected (one +loses 2 inputs then). There are high- and low-limit registers for all inputs. + +Check the datasheet for details. + +sysfs-Interface +--------------- + +in0_input - vdd voltage input +in[1-8]_input - analog 1-8 input +temp1_input - temperature input + +Besides standard interfaces, this driver adds (0 = off, 1 = on): + + adc_ref_vdd - Use vdd as reference instead of 2.25 V + fast_sampling - Sample at 22.5 kHz instead of 1.4 kHz, but drop filters + no_average - Turn off averaging over 16 samples + +Notes +----- + +SPI, external temperature sensor and limit registers are not supported yet. diff --git a/Documentation/hwmon/adt7473 b/Documentation/hwmon/adt7473 deleted file mode 100644 index 446612bd1fb9..000000000000 --- a/Documentation/hwmon/adt7473 +++ /dev/null @@ -1,74 +0,0 @@ -Kernel driver adt7473 -====================== - -Supported chips: - * Analog Devices ADT7473 - Prefix: 'adt7473' - Addresses scanned: I2C 0x2C, 0x2D, 0x2E - Datasheet: Publicly available at the Analog Devices website - -Author: Darrick J. Wong - -This driver is depreacted, please use the adt7475 driver instead. - -Description ------------ - -This driver implements support for the Analog Devices ADT7473 chip family. - -The ADT7473 uses the 2-wire interface compatible with the SMBUS 2.0 -specification. Using an analog to digital converter it measures three (3) -temperatures and two (2) voltages. It has four (4) 16-bit counters for -measuring fan speed. There are three (3) PWM outputs that can be used -to control fan speed. - -A sophisticated control system for the PWM outputs is designed into the -ADT7473 that allows fan speed to be adjusted automatically based on any of the -three temperature sensors. Each PWM output is individually adjustable and -programmable. Once configured, the ADT7473 will adjust the PWM outputs in -response to the measured temperatures without further host intervention. -This feature can also be disabled for manual control of the PWM's. - -Each of the measured inputs (voltage, temperature, fan speed) has -corresponding high/low limit values. The ADT7473 will signal an ALARM if -any measured value exceeds either limit. - -The ADT7473 samples all inputs continuously. The driver will not read -the registers more often than once every other second. Further, -configuration data is only read once per minute. - -Special Features ----------------- - -The ADT7473 have a 10-bit ADC and can therefore measure temperatures -with 0.25 degC resolution. Temperature readings can be configured either -for twos complement format or "Offset 64" format, wherein 63 is subtracted -from the raw value to get the temperature value. - -The Analog Devices datasheet is very detailed and describes a procedure for -determining an optimal configuration for the automatic PWM control. - -Configuration Notes -------------------- - -Besides standard interfaces driver adds the following: - -* PWM Control - -* pwm#_auto_point1_pwm and temp#_auto_point1_temp and -* pwm#_auto_point2_pwm and temp#_auto_point2_temp - - -point1: Set the pwm speed at a lower temperature bound. -point2: Set the pwm speed at a higher temperature bound. - -The ADT7473 will scale the pwm between the lower and higher pwm speed when -the temperature is between the two temperature boundaries. PWM values range -from 0 (off) to 255 (full speed). Fan speed will be set to maximum when the -temperature sensor associated with the PWM control exceeds temp#_max. - -Notes ------ - -The NVIDIA binary driver presents an ADT7473 chip via an on-card i2c bus. -Unfortunately, they fail to set the i2c adapter class, so this driver may -fail to find the chip until the nvidia driver is patched. diff --git a/Documentation/hwmon/asc7621 b/Documentation/hwmon/asc7621 new file mode 100644 index 000000000000..7287be7e1f21 --- /dev/null +++ b/Documentation/hwmon/asc7621 @@ -0,0 +1,296 @@ +Kernel driver asc7621 +================== + +Supported chips: + Andigilog aSC7621 and aSC7621a + Prefix: 'asc7621' + Addresses scanned: I2C 0x2c, 0x2d, 0x2e + Datasheet: http://www.fairview5.com/linux/asc7621/asc7621.pdf + +Author: + George Joseph + +Description provided by Dave Pivin @ Andigilog: + +Andigilog has both the PECI and pre-PECI versions of the Heceta-6, as +Intel calls them. Heceta-6e has high frequency PWM and Heceta-6p has +added PECI and a 4th thermal zone. The Andigilog aSC7611 is the +Heceta-6e part and aSC7621 is the Heceta-6p part. They are both in +volume production, shipping to Intel and their subs. + +We have enhanced both parts relative to the governing Intel +specification. First enhancement is temperature reading resolution. We +have used registers below 20h for vendor-specific functions in addition +to those in the Intel-specified vendor range. + +Our conversion process produces a result that is reported as two bytes. +The fan speed control uses this finer value to produce a "step-less" fan +PWM output. These two bytes are "read-locked" to guarantee that once a +high or low byte is read, the other byte is locked-in until after the +next read of any register. So to get an atomic reading, read high or low +byte, then the very next read should be the opposite byte. Our data +sheet says 10-bits of resolution, although you may find the lower bits +are active, they are not necessarily reliable or useful externally. We +chose not to mask them. + +We employ significant filtering that is user tunable as described in the +data sheet. Our temperature reports and fan PWM outputs are very smooth +when compared to the competition, in addition to the higher resolution +temperature reports. The smoother PWM output does not require user +intervention. + +We offer GPIO features on the former VID pins. These are open-drain +outputs or inputs and may be used as general purpose I/O or as alarm +outputs that are based on temperature limits. These are in 19h and 1Ah. + +We offer flexible mapping of temperature readings to thermal zones. Any +temperature may be mapped to any zone, which has a default assignment +that follows Intel's specs. + +Since there is a fan to zone assignment that allows for the "hotter" of +a set of zones to control the PWM of an individual fan, but there is no +indication to the user, we have added an indicator that shows which zone +is currently controlling the PWM for a given fan. This is in register +00h. + +Both remote diode temperature readings may be given an offset value such +that the reported reading as well as the temperature used to determine +PWM may be offset for system calibration purposes. + +PECI Extended configuration allows for having more than two domains per +PECI address and also provides an enabling function for each PECI +address. One could use our flexible zone assignment to have a zone +assigned to up to 4 PECI addresses. This is not possible in the default +Intel configuration. This would be useful in multi-CPU systems with +individual fans on each that would benefit from individual fan control. +This is in register 0Eh. + +The tachometer measurement system is flexible and able to adapt to many +fan types. We can also support pulse-stretched PWM so that 3-wire fans +may be used. These characteristics are in registers 04h to 07h. + +Finally, we have added a tach disable function that turns off the tach +measurement system for individual tachs in order to save power. That is +in register 75h. + +-- +aSC7621 Product Description + +The aSC7621 has a two wire digital interface compatible with SMBus 2.0. +Using a 10-bit ADC, the aSC7621 measures the temperature of two remote diode +connected transistors as well as its own die. Support for Platform +Environmental Control Interface (PECI) is included. + +Using temperature information from these four zones, an automatic fan speed +control algorithm is employed to minimize acoustic impact while achieving +recommended CPU temperature under varying operational loads. + +To set fan speed, the aSC7621 has three independent pulse width modulation +(PWM) outputs that are controlled by one, or a combination of three, +temperature zones. Both high- and low-frequency PWM ranges are supported. + +The aSC7621 also includes a digital filter that can be invoked to smooth +temperature readings for better control of fan speed and minimum acoustic +impact. + +The aSC7621 has tachometer inputs to measure fan speed on up to four fans. +Limit and status registers for all measured values are included to alert +the system host that any measurements are outside of programmed limits +via status registers. + +System voltages of VCCP, 2.5V, 3.3V, 5.0V, and 12V motherboard power are +monitored efficiently with internal scaling resistors. + +Features +- Supports PECI interface and monitors internal and remote thermal diodes +- 2-wire, SMBus 2.0 compliant, serial interface +- 10-bit ADC +- Monitors VCCP, 2.5V, 3.3V, 5.0V, and 12V motherboard/processor supplies +- Programmable autonomous fan control based on temperature readings +- Noise filtering of temperature reading for fan speed control +- 0.25C digital temperature sensor resolution +- 3 PWM fan speed control outputs for 2-, 3- or 4-wire fans and up to 4 fan + tachometer inputs +- Enhanced measured temperature to Temperature Zone assignment. +- Provides high and low PWM frequency ranges +- 3 GPIO pins for custom use +- 24-Lead QSOP package + +Configuration Notes +=================== + +Except where noted below, the sysfs entries created by this driver follow +the standards defined in "sysfs-interface". + +temp1_source + 0 (default) peci_legacy = 0, Remote 1 Temperature + peci_legacy = 1, PECI Processor Temperature 0 + 1 Remote 1 Temperature + 2 Remote 2 Temperature + 3 Internal Temperature + 4 PECI Processor Temperature 0 + 5 PECI Processor Temperature 1 + 6 PECI Processor Temperature 2 + 7 PECI Processor Temperature 3 + +temp2_source + 0 (default) Internal Temperature + 1 Remote 1 Temperature + 2 Remote 2 Temperature + 3 Internal Temperature + 4 PECI Processor Temperature 0 + 5 PECI Processor Temperature 1 + 6 PECI Processor Temperature 2 + 7 PECI Processor Temperature 3 + +temp3_source + 0 (default) Remote 2 Temperature + 1 Remote 1 Temperature + 2 Remote 2 Temperature + 3 Internal Temperature + 4 PECI Processor Temperature 0 + 5 PECI Processor Temperature 1 + 6 PECI Processor Temperature 2 + 7 PECI Processor Temperature 3 + +temp4_source + 0 (default) peci_legacy = 0, PECI Processor Temperature 0 + peci_legacy = 1, Remote 1 Temperature + 1 Remote 1 Temperature + 2 Remote 2 Temperature + 3 Internal Temperature + 4 PECI Processor Temperature 0 + 5 PECI Processor Temperature 1 + 6 PECI Processor Temperature 2 + 7 PECI Processor Temperature 3 + +temp[1-4]_smoothing_enable +temp[1-4]_smoothing_time + Smooths spikes in temp readings caused by noise. + Valid values in milliseconds are: + 35000 + 17600 + 11800 + 7000 + 4400 + 3000 + 1600 + 800 + +temp[1-4]_crit + When the corresponding zone temperature reaches this value, + ALL pwm outputs will got to 100%. + +temp[5-8]_input +temp[5-8]_enable + The aSC7621 can also read temperatures provided by the processor + via the PECI bus. Usually these are "core" temps and are relative + to the point where the automatic thermal control circuit starts + throttling. This means that these are usually negative numbers. + +pwm[1-3]_enable + 0 Fan off. + 1 Fan on manual control. + 2 Fan on automatic control and will run at the minimum pwm + if the temperature for the zone is below the minimum. + 3 Fan on automatic control but will be off if the temperature + for the zone is below the minimum. + 4-254 Ignored. + 255 Fan on full. + +pwm[1-3]_auto_channels + Bitmap as described in sysctl-interface with the following + exceptions... + Only the following combination of zones (and their corresponding masks) + are valid: + 1 + 2 + 3 + 2,3 + 1,2,3 + 4 + 1,2,3,4 + + Special values: + 0 Disabled. + 16 Fan on manual control. + 31 Fan on full. + + +pwm[1-3]_invert + When set, inverts the meaning of pwm[1-3]. + i.e. when pwm = 0, the fan will be on full and + when pwm = 255 the fan will be off. + +pwm[1-3]_freq + PWM frequency in Hz + Valid values in Hz are: + + 10 + 15 + 23 + 30 (default) + 38 + 47 + 62 + 94 + 23000 + 24000 + 25000 + 26000 + 27000 + 28000 + 29000 + 30000 + + Setting any other value will be ignored. + +peci_enable + Enables or disables PECI + +peci_avg + Input filter average time. + + 0 0 Sec. (no Smoothing) (default) + 1 0.25 Sec. + 2 0.5 Sec. + 3 1.0 Sec. + 4 2.0 Sec. + 5 4.0 Sec. + 6 8.0 Sec. + 7 0.0 Sec. + +peci_legacy + + 0 Standard Mode (default) + Remote Diode 1 reading is associated with + Temperature Zone 1, PECI is associated with + Zone 4 + + 1 Legacy Mode + PECI is associated with Temperature Zone 1, + Remote Diode 1 is associated with Zone 4 + +peci_diode + Diode filter + + 0 0.25 Sec. + 1 1.1 Sec. + 2 2.4 Sec. (default) + 3 3.4 Sec. + 4 5.0 Sec. + 5 6.8 Sec. + 6 10.2 Sec. + 7 16.4 Sec. + +peci_4domain + Four domain enable + + 0 1 or 2 Domains for enabled processors (default) + 1 3 or 4 Domains for enabled processors + +peci_domain + Domain + + 0 Processor contains a single domain (0) (default) + 1 Processor contains two domains (0,1) diff --git a/Documentation/hwmon/it87 b/Documentation/hwmon/it87 index f9ba96c0ac4a..8d08bf0d38ed 100644 --- a/Documentation/hwmon/it87 +++ b/Documentation/hwmon/it87 @@ -5,31 +5,23 @@ Supported chips: * IT8705F Prefix: 'it87' Addresses scanned: from Super I/O config space (8 I/O ports) - Datasheet: Publicly available at the ITE website - http://www.ite.com.tw/product_info/file/pc/IT8705F_V.0.4.1.pdf + Datasheet: Once publicly available at the ITE website, but no longer * IT8712F Prefix: 'it8712' Addresses scanned: from Super I/O config space (8 I/O ports) - Datasheet: Publicly available at the ITE website - http://www.ite.com.tw/product_info/file/pc/IT8712F_V0.9.1.pdf - http://www.ite.com.tw/product_info/file/pc/Errata%20V0.1%20for%20IT8712F%20V0.9.1.pdf - http://www.ite.com.tw/product_info/file/pc/IT8712F_V0.9.3.pdf + Datasheet: Once publicly available at the ITE website, but no longer * IT8716F/IT8726F Prefix: 'it8716' Addresses scanned: from Super I/O config space (8 I/O ports) - Datasheet: Publicly available at the ITE website - http://www.ite.com.tw/product_info/file/pc/IT8716F_V0.3.ZIP - http://www.ite.com.tw/product_info/file/pc/IT8726F_V0.3.pdf + Datasheet: Once publicly available at the ITE website, but no longer * IT8718F Prefix: 'it8718' Addresses scanned: from Super I/O config space (8 I/O ports) - Datasheet: Publicly available at the ITE website - http://www.ite.com.tw/product_info/file/pc/IT8718F_V0.2.zip - http://www.ite.com.tw/product_info/file/pc/IT8718F_V0%203_(for%20C%20version).zip + Datasheet: Once publicly available at the ITE website, but no longer * IT8720F Prefix: 'it8720' Addresses scanned: from Super I/O config space (8 I/O ports) - Datasheet: Not yet publicly available. + Datasheet: Not publicly available * SiS950 [clone of IT8705F] Prefix: 'it87' Addresses scanned: from Super I/O config space (8 I/O ports) @@ -136,6 +128,10 @@ registers are read whenever any data is read (unless it is less than 1.5 seconds since the last update). This means that you can easily miss once-only alarms. +Out-of-limit readings can also result in beeping, if the chip is properly +wired and configured. Beeping can be enabled or disabled per sensor type +(temperatures, voltages and fans.) + The IT87xx only updates its values each 1.5 seconds; reading it more often will do no harm, but will return 'old' values. @@ -150,11 +146,38 @@ Fan speed control ----------------- The fan speed control features are limited to manual PWM mode. Automatic -"Smart Guardian" mode control handling is not implemented. However -if you want to go for "manual mode" just write 1 to pwmN_enable. +"Smart Guardian" mode control handling is only implemented for older chips +(see below.) However if you want to go for "manual mode" just write 1 to +pwmN_enable. If you are only able to control the fan speed with very small PWM values, try lowering the PWM base frequency (pwm1_freq). Depending on the fan, it may give you a somewhat greater control range. The same frequency is used to drive all fan outputs, which is why pwm2_freq and pwm3_freq are read-only. + + +Automatic fan speed control (old interface) +------------------------------------------- + +The driver supports the old interface to automatic fan speed control +which is implemented by IT8705F chips up to revision F and IT8712F +chips up to revision G. + +This interface implements 4 temperature vs. PWM output trip points. +The PWM output of trip point 4 is always the maximum value (fan running +at full speed) while the PWM output of the other 3 trip points can be +freely chosen. The temperature of all 4 trip points can be freely chosen. +Additionally, trip point 1 has an hysteresis temperature attached, to +prevent fast switching between fan on and off. + +The chip automatically computes the PWM output value based on the input +temperature, based on this simple rule: if the temperature value is +between trip point N and trip point N+1 then the PWM output value is +the one of trip point N. The automatic control mode is less flexible +than the manual control mode, but it reacts faster, is more robust and +doesn't use CPU cycles. + +Trip points must be set properly before switching to automatic fan speed +control mode. The driver will perform basic integrity checks before +actually switching to automatic control mode. diff --git a/Documentation/hwmon/lm90 b/Documentation/hwmon/lm90 index 93d8e3d55150..6a03dd4bcc94 100644 --- a/Documentation/hwmon/lm90 +++ b/Documentation/hwmon/lm90 @@ -84,6 +84,10 @@ Supported chips: Addresses scanned: I2C 0x4c Datasheet: Publicly available at the Maxim website http://www.maxim-ic.com/quick_view2.cfm/qv_pk/3500 + * Winbond/Nuvoton W83L771AWG/ASG + Prefix: 'w83l771' + Addresses scanned: I2C 0x4c + Datasheet: Not publicly available, can be requested from Nuvoton Author: Jean Delvare <khali@linux-fr.org> @@ -147,6 +151,12 @@ MAX6680 and MAX6681: * Selectable address * Remote sensor type selection +W83L771AWG/ASG + * The AWG and ASG variants only differ in package format. + * Filter and alert configuration register at 0xBF + * Diode ideality factor configuration (remote sensor) at 0xE3 + * Moving average (depending on conversion rate) + All temperature values are given in degrees Celsius. Resolution is 1.0 degree for the local temperature, 0.125 degree for the remote temperature, except for the MAX6657, MAX6658 and MAX6659 which have a @@ -163,6 +173,18 @@ The lm90 driver will not update its values more frequently than every other second; reading them more often will do no harm, but will return 'old' values. +SMBus Alert Support +------------------- + +This driver has basic support for SMBus alert. When an alert is received, +the status register is read and the faulty temperature channel is logged. + +The Analog Devices chips (ADM1032 and ADT7461) do not implement the SMBus +alert protocol properly so additional care is needed: the ALERT output is +disabled when an alert is received, and is re-enabled only when the alarm +is gone. Otherwise the chip would block alerts from other chips in the bus +as long as the alarm is active. + PEC Support ----------- diff --git a/Documentation/init.txt b/Documentation/init.txt new file mode 100644 index 000000000000..535ad5e82b98 --- /dev/null +++ b/Documentation/init.txt @@ -0,0 +1,49 @@ +Explaining the dreaded "No init found." boot hang message +========================================================= + +OK, so you've got this pretty unintuitive message (currently located +in init/main.c) and are wondering what the H*** went wrong. +Some high-level reasons for failure (listed roughly in order of execution) +to load the init binary are: +A) Unable to mount root FS +B) init binary doesn't exist on rootfs +C) broken console device +D) binary exists but dependencies not available +E) binary cannot be loaded + +Detailed explanations: +0) Set "debug" kernel parameter (in bootloader config file or CONFIG_CMDLINE) + to get more detailed kernel messages. +A) make sure you have the correct root FS type + (and root= kernel parameter points to the correct partition), + required drivers such as storage hardware (such as SCSI or USB!) + and filesystem (ext3, jffs2 etc.) are builtin (alternatively as modules, + to be pre-loaded by an initrd) +C) Possibly a conflict in console= setup --> initial console unavailable. + E.g. some serial consoles are unreliable due to serial IRQ issues (e.g. + missing interrupt-based configuration). + Try using a different console= device or e.g. netconsole= . +D) e.g. required library dependencies of the init binary such as + /lib/ld-linux.so.2 missing or broken. Use readelf -d <INIT>|grep NEEDED + to find out which libraries are required. +E) make sure the binary's architecture matches your hardware. + E.g. i386 vs. x86_64 mismatch, or trying to load x86 on ARM hardware. + In case you tried loading a non-binary file here (shell script?), + you should make sure that the script specifies an interpreter in its shebang + header line (#!/...) that is fully working (including its library + dependencies). And before tackling scripts, better first test a simple + non-script binary such as /bin/sh and confirm its successful execution. + To find out more, add code to init/main.c to display kernel_execve()s + return values. + +Please extend this explanation whenever you find new failure causes +(after all loading the init binary is a CRITICAL and hard transition step +which needs to be made as painless as possible), then submit patch to LKML. +Further TODOs: +- Implement the various run_init_process() invocations via a struct array + which can then store the kernel_execve() result value and on failure + log it all by iterating over _all_ results (very important usability fix). +- try to make the implementation itself more helpful in general, + e.g. by providing additional error messages at affected places. + +Andreas Mohr <andi at lisas period de> diff --git a/MAINTAINERS b/MAINTAINERS index 51d8b5221dd8..c8a8b1fd58b3 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -71,6 +71,7 @@ Descriptions of section entries: M: Mail patches to: FullName <address@domain> L: Mailing list that is relevant to this area W: Web-page with status/info + Q: Patchwork web based patch tracking system site T: SCM tree type and location. Type is one of: git, hg, quilt, stgit. S: Status, one of the following: Supported: Someone is actually paid to look after this. @@ -182,6 +183,7 @@ M: Ron Minnich <rminnich@sandia.gov> M: Latchesar Ionkov <lucho@ionkov.net> L: v9fs-developer@lists.sourceforge.net W: http://swik.net/v9fs +Q: http://patchwork.kernel.org/project/v9fs-devel/list/ T: git git://git.kernel.org/pub/scm/linux/kernel/git/ericvh/v9fs.git S: Maintained F: Documentation/filesystems/9p.txt @@ -238,6 +240,7 @@ ACPI M: Len Brown <lenb@kernel.org> L: linux-acpi@vger.kernel.org W: http://www.lesswatts.org/projects/acpi/ +Q: http://patchwork.kernel.org/project/linux-acpi/list/ T: git git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux-acpi-2.6.git S: Supported F: drivers/acpi/ @@ -428,7 +431,6 @@ P: Jordan Crouse L: linux-geode@lists.infradead.org (moderated for non-subscribers) W: http://www.amd.com/us-en/ConnectivitySolutions/TechnicalResources/0,,50_2334_2452_11363,00.html S: Supported -F: arch/x86/kernel/geode_32.c F: drivers/char/hw_random/geode-rng.c F: drivers/crypto/geode* F: drivers/video/geode/ @@ -966,6 +968,13 @@ W: http://www.arm.linux.org.uk/ S: Maintained F: arch/arm/vfp/ +ASC7621 HARDWARE MONITOR DRIVER +M: George Joseph <george.joseph@fairview5.com> +L: lm-sensors@lm-sensors.org +S: Maintained +F: Documentation/hwmon/asc7621 +F: drivers/hwmon/asc7621.c + ASUS ACPI EXTRAS DRIVER M: Corentin Chary <corentincj@iksaif.net> M: Karol Kozimor <sziwan@users.sourceforge.net> @@ -1332,6 +1341,7 @@ BTRFS FILE SYSTEM M: Chris Mason <chris.mason@oracle.com> L: linux-btrfs@vger.kernel.org W: http://btrfs.wiki.kernel.org/ +Q: http://patchwork.kernel.org/project/linux-btrfs/list/ T: git git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable.git S: Maintained F: Documentation/filesystems/btrfs.txt @@ -1496,6 +1506,7 @@ M: Steve French <sfrench@samba.org> L: linux-cifs-client@lists.samba.org (moderated for non-subscribers) L: samba-technical@lists.samba.org (moderated for non-subscribers) W: http://linux-cifs.samba.org/ +Q: http://patchwork.ozlabs.org/project/linux-cifs-client/list/ T: git git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6.git S: Supported F: Documentation/filesystems/cifs.txt @@ -1782,6 +1793,7 @@ DEVICE-MAPPER (LVM) P: Alasdair Kergon L: dm-devel@redhat.com W: http://sources.redhat.com/dm +Q: http://patchwork.kernel.org/project/dm-devel/list/ S: Maintained F: Documentation/device-mapper/ F: drivers/md/dm* @@ -2126,6 +2138,7 @@ M: "Theodore Ts'o" <tytso@mit.edu> M: Andreas Dilger <adilger@sun.com> L: linux-ext4@vger.kernel.org W: http://ext4.wiki.kernel.org +Q: http://patchwork.ozlabs.org/project/linux-ext4/list/ S: Maintained F: Documentation/filesystems/ext4.txt F: fs/ext4/ @@ -2502,13 +2515,6 @@ L: linux-parisc@vger.kernel.org S: Maintained F: sound/parisc/harmony.* -HAYES ESP SERIAL DRIVER -M: "Andrew J. Robinson" <arobinso@nyx.net> -W: http://www.nyx.net/~arobinso -S: Maintained -F: Documentation/serial/hayes-esp.txt -F: drivers/char/esp.c - HEWLETT-PACKARD SMART2 RAID DRIVER M: Chirag Kantharia <chirag.kantharia@hp.com> L: iss_storagedev@hp.com @@ -2717,6 +2723,7 @@ F: drivers/scsi/ips.* IDE SUBSYSTEM M: "David S. Miller" <davem@davemloft.net> L: linux-ide@vger.kernel.org +Q: http://patchwork.ozlabs.org/project/linux-ide/list/ T: git git://git.kernel.org/pub/scm/linux/kernel/git/davem/ide-2.6.git S: Maintained F: Documentation/ide/ @@ -2771,6 +2778,7 @@ M: Sean Hefty <sean.hefty@intel.com> M: Hal Rosenstock <hal.rosenstock@gmail.com> L: linux-rdma@vger.kernel.org W: http://www.openib.org/ +Q: http://patchwork.kernel.org/project/linux-rdma/list/ T: git git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband.git S: Supported F: Documentation/infiniband/ @@ -2790,6 +2798,7 @@ INPUT (KEYBOARD, MOUSE, JOYSTICK, TOUCHSCREEN) DRIVERS M: Dmitry Torokhov <dmitry.torokhov@gmail.com> M: Dmitry Torokhov <dtor@mail.ru> L: linux-input@vger.kernel.org +Q: http://patchwork.kernel.org/project/linux-input/list/ T: git git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input.git S: Maintained F: drivers/input/ @@ -3046,6 +3055,13 @@ W: http://www.melware.de S: Maintained F: drivers/isdn/hardware/eicon/ +IT87 HARDWARE MONITORING DRIVER +M: Jean Delvare <khali@linux-fr.org> +L: lm-sensors@lm-sensors.org +S: Maintained +F: Documentation/hwmon/it87 +F: drivers/hwmon/it87.c + IVTV VIDEO4LINUX DRIVER M: Andy Walls <awalls@radix.net> L: ivtv-devel@ivtvdriver.org (moderated for non-subscribers) @@ -3099,6 +3115,7 @@ F: drivers/hwmon/k8temp.c KCONFIG M: Roman Zippel <zippel@linux-m68k.org> L: linux-kbuild@vger.kernel.org +Q: http://patchwork.kernel.org/project/linux-kbuild/list/ S: Maintained F: Documentation/kbuild/kconfig-language.txt F: scripts/kconfig/ @@ -3312,6 +3329,7 @@ M: Benjamin Herrenschmidt <benh@kernel.crashing.org> M: Paul Mackerras <paulus@samba.org> W: http://www.penguinppc.org/ L: linuxppc-dev@ozlabs.org +Q: http://patchwork.ozlabs.org/project/linuxppc-dev/list/ T: git git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc.git S: Supported F: Documentation/powerpc/ @@ -3432,6 +3450,13 @@ S: Maintained F: Documentation/ldm.txt F: fs/partitions/ldm.* +LogFS +M: Joern Engel <joern@logfs.org> +L: logfs@logfs.org +W: logfs.org +S: Maintained +F: fs/logfs/ + LSILOGIC MPT FUSION DRIVERS (FC/SAS/SPI) M: Eric Moore <Eric.Moore@lsi.com> M: support@lsi.com @@ -3568,6 +3593,7 @@ M: Mauro Carvalho Chehab <mchehab@infradead.org> P: LinuxTV.org Project L: linux-media@vger.kernel.org W: http://linuxtv.org +Q: http://patchwork.kernel.org/project/linux-media/list/ T: git git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6.git S: Maintained F: Documentation/dvb/ @@ -3603,8 +3629,9 @@ F: mm/memcontrol.c MEMORY TECHNOLOGY DEVICES (MTD) M: David Woodhouse <dwmw2@infradead.org> -W: http://www.linux-mtd.infradead.org/ L: linux-mtd@lists.infradead.org +W: http://www.linux-mtd.infradead.org/ +Q: http://patchwork.ozlabs.org/project/linux-mtd/list/ T: git git://git.infradead.org/mtd-2.6.git S: Maintained F: drivers/mtd/ @@ -3864,6 +3891,7 @@ S: Maintained NETWORKING [WIRELESS] M: "John W. Linville" <linville@tuxdriver.com> L: linux-wireless@vger.kernel.org +Q: http://patchwork.kernel.org/project/linux-wireless/list/ T: git git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-2.6.git S: Maintained F: net/mac80211/ @@ -3956,6 +3984,7 @@ M: Tony Lindgren <tony@atomide.com> L: linux-omap@vger.kernel.org W: http://www.muru.com/linux/omap/ W: http://linux.omap.com/ +Q: http://patchwork.kernel.org/project/linux-omap/list/ T: git git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap-2.6.git S: Maintained F: arch/arm/*omap*/ @@ -4182,6 +4211,7 @@ M: Helge Deller <deller@gmx.de> M: "James E.J. Bottomley" <jejb@parisc-linux.org> L: linux-parisc@vger.kernel.org W: http://www.parisc-linux.org/ +Q: http://patchwork.kernel.org/project/linux-parisc/list/ T: git git://git.kernel.org/pub/scm/linux/kernel/git/kyle/parisc-2.6.git S: Maintained F: arch/parisc/ @@ -4224,6 +4254,7 @@ F: Documentation/powerpc/eeh-pci-error-recovery.txt PCI SUBSYSTEM M: Jesse Barnes <jbarnes@virtuousgeek.org> L: linux-pci@vger.kernel.org +Q: http://patchwork.kernel.org/project/linux-pci/list/ T: git git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci-2.6.git S: Supported F: Documentation/PCI/ @@ -4265,7 +4296,9 @@ M: Ingo Molnar <mingo@elte.hu> S: Supported F: kernel/perf_event.c F: include/linux/perf_event.h -F: arch/*/*/kernel/perf_event.c +F: arch/*/kernel/perf_event.c +F: arch/*/kernel/*/perf_event.c +F: arch/*/kernel/*/*/perf_event.c F: arch/*/include/asm/perf_event.h F: arch/*/lib/perf_event.c F: arch/*/kernel/perf_callchain.c @@ -4599,6 +4632,7 @@ F: include/linux/rtc.h REAL TIME CLOCK (RTC) SUBSYSTEM M: Alessandro Zummo <a.zummo@towertech.it> L: rtc-linux@googlegroups.com +Q: http://patchwork.ozlabs.org/project/rtc-linux/list/ S: Maintained F: Documentation/rtc.txt F: drivers/rtc/ @@ -4966,6 +5000,7 @@ F: drivers/*/*/*s3c2410* TI DAVINCI MACHINE SUPPORT P: Kevin Hilman M: davinci-linux-open-source@linux.davincidsp.com +Q: http://patchwork.kernel.org/project/linux-davinci/list/ S: Supported F: arch/arm/mach-davinci @@ -5131,6 +5166,7 @@ F: include/sound/soc* SPARC + UltraSPARC (sparc/sparc64) M: "David S. Miller" <davem@davemloft.net> L: sparclinux@vger.kernel.org +Q: http://patchwork.ozlabs.org/project/sparclinux/list/ T: git git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6.git T: git git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-next-2.6.git S: Maintained @@ -5146,6 +5182,7 @@ SPI SUBSYSTEM M: David Brownell <dbrownell@users.sourceforge.net> M: Grant Likely <grant.likely@secretlab.ca> L: spi-devel-general@lists.sourceforge.net +Q: http://patchwork.kernel.org/project/spi-devel-general/list/ S: Maintained F: Documentation/spi/ F: drivers/spi/ @@ -5201,7 +5238,7 @@ F: drivers/net/starfire* STARMODE RADIO IP (STRIP) PROTOCOL DRIVER S: Orphan -F: drivers/net/wireless/strip.c +F: drivers/staging/strip/strip.c F: include/linux/if_strip.h STRADIS MPEG-2 DECODER DRIVER @@ -5222,6 +5259,7 @@ SUPERH M: Paul Mundt <lethal@linux-sh.org> L: linux-sh@vger.kernel.org W: http://www.linux-sh.org +Q: http://patchwork.kernel.org/project/linux-sh/list/ T: git git://git.kernel.org/pub/scm/linux/kernel/git/lethal/sh-2.6.git S: Supported F: Documentation/sh/ @@ -5989,7 +6027,7 @@ L: linux-wireless@vger.kernel.org W: http://www.hpl.hp.com/personal/Jean_Tourrilhes/Linux/ S: Maintained F: Documentation/networking/wavelan.txt -F: drivers/net/wireless/wavelan* +F: drivers/staging/wavelan/ WD7000 SCSI DRIVER M: Miroslav Zagorac <zaga@fly.cc.fer.hr> @@ -6185,6 +6223,7 @@ F: drivers/serial/zs.* THE REST M: Linus Torvalds <torvalds@linux-foundation.org> L: linux-kernel@vger.kernel.org +Q: http://patchwork.kernel.org/project/LKML/list/ T: git git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git S: Buried alive in reporters F: * diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c index 62619f25132f..53c213f70fcb 100644 --- a/arch/alpha/kernel/osf_sys.c +++ b/arch/alpha/kernel/osf_sys.c @@ -361,7 +361,7 @@ osf_procfs_mount(char *dirname, struct procfs_args __user *args, int flags) SYSCALL_DEFINE4(osf_mount, unsigned long, typenr, char __user *, path, int, flag, void __user *, data) { - int retval = -EINVAL; + int retval; char *name; name = getname(path); @@ -379,6 +379,7 @@ SYSCALL_DEFINE4(osf_mount, unsigned long, typenr, char __user *, path, retval = osf_procfs_mount(name, data, flag); break; default: + retval = -EINVAL; printk("osf_mount(%ld, %x)\n", typenr, flag); } putname(name); diff --git a/arch/cris/Kconfig b/arch/cris/Kconfig index 7adac388a771..059eac6abda1 100644 --- a/arch/cris/Kconfig +++ b/arch/cris/Kconfig @@ -20,6 +20,12 @@ config RWSEM_GENERIC_SPINLOCK config RWSEM_XCHGADD_ALGORITHM bool +config GENERIC_TIME + def_bool y + +config ARCH_USES_GETTIMEOFFSET + def_bool y + config GENERIC_IOMAP bool default y diff --git a/arch/cris/arch-v32/drivers/cryptocop.c b/arch/cris/arch-v32/drivers/cryptocop.c index fd529a0ec758..b70fb34939d9 100644 --- a/arch/cris/arch-v32/drivers/cryptocop.c +++ b/arch/cris/arch-v32/drivers/cryptocop.c @@ -628,9 +628,9 @@ static int create_output_descriptors(struct cryptocop_operation *operation, int cdesc->dma_descr->buf = (char*)virt_to_phys(operation->tfrm_op.indata[*iniov_ix].iov_base + *iniov_offset); cdesc->dma_descr->after = cdesc->dma_descr->buf + dlength; + assert(desc_len >= dlength); desc_len -= dlength; *iniov_offset += dlength; - assert(desc_len >= 0); if (*iniov_offset >= operation->tfrm_op.indata[*iniov_ix].iov_len) { *iniov_offset = 0; ++(*iniov_ix); diff --git a/arch/cris/arch-v32/mach-fs/arbiter.c b/arch/cris/arch-v32/mach-fs/arbiter.c index 84d31bd7b692..82ef293c4c81 100644 --- a/arch/cris/arch-v32/mach-fs/arbiter.c +++ b/arch/cris/arch-v32/mach-fs/arbiter.c @@ -332,7 +332,7 @@ int crisv32_arbiter_unwatch(int id) if (id == 0) intr_mask.bp0 = regk_marb_no; else if (id == 1) - intr_mask.bp2 = regk_marb_no; + intr_mask.bp1 = regk_marb_no; else if (id == 2) intr_mask.bp2 = regk_marb_no; else if (id == 3) diff --git a/arch/cris/kernel/time.c b/arch/cris/kernel/time.c index 074fe7dea96b..a05dd31f3efb 100644 --- a/arch/cris/kernel/time.c +++ b/arch/cris/kernel/time.c @@ -42,75 +42,11 @@ unsigned long loops_per_usec; extern unsigned long do_slow_gettimeoffset(void); static unsigned long (*do_gettimeoffset)(void) = do_slow_gettimeoffset; -/* - * This version of gettimeofday has near microsecond resolution. - * - * Note: Division is quite slow on CRIS and do_gettimeofday is called - * rather often. Maybe we should do some kind of approximation here - * (a naive approximation would be to divide by 1024). - */ -void do_gettimeofday(struct timeval *tv) -{ - unsigned long flags; - signed long usec, sec; - local_irq_save(flags); - usec = do_gettimeoffset(); - - /* - * If time_adjust is negative then NTP is slowing the clock - * so make sure not to go into next possible interval. - * Better to lose some accuracy than have time go backwards.. - */ - if (unlikely(time_adjust < 0) && usec > tickadj) - usec = tickadj; - - sec = xtime.tv_sec; - usec += xtime.tv_nsec / 1000; - local_irq_restore(flags); - - while (usec >= 1000000) { - usec -= 1000000; - sec++; - } - - tv->tv_sec = sec; - tv->tv_usec = usec; -} - -EXPORT_SYMBOL(do_gettimeofday); - -int do_settimeofday(struct timespec *tv) +u32 arch_gettimeoffset(void) { - time_t wtm_sec, sec = tv->tv_sec; - long wtm_nsec, nsec = tv->tv_nsec; - - if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC) - return -EINVAL; - - write_seqlock_irq(&xtime_lock); - /* - * This is revolting. We need to set "xtime" correctly. However, the - * value in this location is the value at the most recent update of - * wall time. Discover what correction gettimeofday() would have - * made, and then undo it! - */ - nsec -= do_gettimeoffset() * NSEC_PER_USEC; - - wtm_sec = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec); - wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec); - - set_normalized_timespec(&xtime, sec, nsec); - set_normalized_timespec(&wall_to_monotonic, wtm_sec, wtm_nsec); - - ntp_clear(); - write_sequnlock_irq(&xtime_lock); - clock_was_set(); - return 0; + return do_gettimeoffset() * 1000; } -EXPORT_SYMBOL(do_settimeofday); - - /* * BUG: This routine does not handle hour overflow properly; it just * sets the minutes. Usually you'll only notice that after reboot! diff --git a/arch/frv/include/asm/pci.h b/arch/frv/include/asm/pci.h index 492b5c4dfed6..8c7260a3cd41 100644 --- a/arch/frv/include/asm/pci.h +++ b/arch/frv/include/asm/pci.h @@ -68,41 +68,4 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev, #define PCIBIOS_MIN_IO 0x100 #define PCIBIOS_MIN_MEM 0x00010000 -/* Make physical memory consistent for a single - * streaming mode DMA translation after a transfer. - * - * If you perform a pci_map_single() but wish to interrogate the - * buffer using the cpu, yet do not wish to teardown the PCI dma - * mapping, you must call this function before doing so. At the - * next point you give the PCI dma address back to the card, the - * device again owns the buffer. - */ -static inline void pci_dma_sync_single(struct pci_dev *hwdev, - dma_addr_t dma_handle, - size_t size, int direction) -{ - BUG_ON(direction == PCI_DMA_NONE); - - frv_cache_wback_inv((unsigned long)bus_to_virt(dma_handle), - (unsigned long)bus_to_virt(dma_handle) + size); -} - -/* Make physical memory consistent for a set of streaming - * mode DMA translations after a transfer. - * - * The same as pci_dma_sync_single but for a scatter-gather list, - * same rules and usage. - */ -static inline void pci_dma_sync_sg(struct pci_dev *hwdev, - struct scatterlist *sg, - int nelems, int direction) -{ - int i; - BUG_ON(direction == PCI_DMA_NONE); - - for (i = 0; i < nelems; i++) - frv_cache_wback_inv(sg_dma_address(&sg[i]), - sg_dma_address(&sg[i])+sg_dma_len(&sg[i])); -} - #endif /* _ASM_FRV_PCI_H */ diff --git a/arch/ia64/include/asm/elf.h b/arch/ia64/include/asm/elf.h index 4c41656ede87..b5298eb09adb 100644 --- a/arch/ia64/include/asm/elf.h +++ b/arch/ia64/include/asm/elf.h @@ -219,54 +219,6 @@ do { \ NEW_AUX_ENT(AT_SYSINFO_EHDR, (unsigned long) GATE_EHDR); \ } while (0) - -/* - * These macros parameterize elf_core_dump in fs/binfmt_elf.c to write out - * extra segments containing the gate DSO contents. Dumping its - * contents makes post-mortem fully interpretable later without matching up - * the same kernel and hardware config to see what PC values meant. - * Dumping its extra ELF program headers includes all the other information - * a debugger needs to easily find how the gate DSO was being used. - */ -#define ELF_CORE_EXTRA_PHDRS (GATE_EHDR->e_phnum) -#define ELF_CORE_WRITE_EXTRA_PHDRS \ -do { \ - const struct elf_phdr *const gate_phdrs = \ - (const struct elf_phdr *) (GATE_ADDR + GATE_EHDR->e_phoff); \ - int i; \ - Elf64_Off ofs = 0; \ - for (i = 0; i < GATE_EHDR->e_phnum; ++i) { \ - struct elf_phdr phdr = gate_phdrs[i]; \ - if (phdr.p_type == PT_LOAD) { \ - phdr.p_memsz = PAGE_ALIGN(phdr.p_memsz); \ - phdr.p_filesz = phdr.p_memsz; \ - if (ofs == 0) { \ - ofs = phdr.p_offset = offset; \ - offset += phdr.p_filesz; \ - } \ - else \ - phdr.p_offset = ofs; \ - } \ - else \ - phdr.p_offset += ofs; \ - phdr.p_paddr = 0; /* match other core phdrs */ \ - DUMP_WRITE(&phdr, sizeof(phdr)); \ - } \ -} while (0) -#define ELF_CORE_WRITE_EXTRA_DATA \ -do { \ - const struct elf_phdr *const gate_phdrs = \ - (const struct elf_phdr *) (GATE_ADDR + GATE_EHDR->e_phoff); \ - int i; \ - for (i = 0; i < GATE_EHDR->e_phnum; ++i) { \ - if (gate_phdrs[i].p_type == PT_LOAD) { \ - DUMP_WRITE((void *) gate_phdrs[i].p_vaddr, \ - PAGE_ALIGN(gate_phdrs[i].p_memsz)); \ - break; \ - } \ - } \ -} while (0) - /* * format for entries in the Global Offset Table */ diff --git a/arch/ia64/kernel/Makefile b/arch/ia64/kernel/Makefile index 4138282aefa8..db10b1e378b0 100644 --- a/arch/ia64/kernel/Makefile +++ b/arch/ia64/kernel/Makefile @@ -45,6 +45,8 @@ endif obj-$(CONFIG_DMAR) += pci-dma.o obj-$(CONFIG_SWIOTLB) += pci-swiotlb.o +obj-$(CONFIG_BINFMT_ELF) += elfcore.o + # fp_emulate() expects f2-f5,f16-f31 to contain the user-level state. CFLAGS_traps.o += -mfixed-range=f2-f5,f16-f31 diff --git a/arch/ia64/kernel/elfcore.c b/arch/ia64/kernel/elfcore.c new file mode 100644 index 000000000000..bac1639bc320 --- /dev/null +++ b/arch/ia64/kernel/elfcore.c @@ -0,0 +1,80 @@ +#include <linux/elf.h> +#include <linux/coredump.h> +#include <linux/fs.h> +#include <linux/mm.h> + +#include <asm/elf.h> + + +Elf64_Half elf_core_extra_phdrs(void) +{ + return GATE_EHDR->e_phnum; +} + +int elf_core_write_extra_phdrs(struct file *file, loff_t offset, size_t *size, + unsigned long limit) +{ + const struct elf_phdr *const gate_phdrs = + (const struct elf_phdr *) (GATE_ADDR + GATE_EHDR->e_phoff); + int i; + Elf64_Off ofs = 0; + + for (i = 0; i < GATE_EHDR->e_phnum; ++i) { + struct elf_phdr phdr = gate_phdrs[i]; + + if (phdr.p_type == PT_LOAD) { + phdr.p_memsz = PAGE_ALIGN(phdr.p_memsz); + phdr.p_filesz = phdr.p_memsz; + if (ofs == 0) { + ofs = phdr.p_offset = offset; + offset += phdr.p_filesz; + } else { + phdr.p_offset = ofs; + } + } else { + phdr.p_offset += ofs; + } + phdr.p_paddr = 0; /* match other core phdrs */ + *size += sizeof(phdr); + if (*size > limit || !dump_write(file, &phdr, sizeof(phdr))) + return 0; + } + return 1; +} + +int elf_core_write_extra_data(struct file *file, size_t *size, + unsigned long limit) +{ + const struct elf_phdr *const gate_phdrs = + (const struct elf_phdr *) (GATE_ADDR + GATE_EHDR->e_phoff); + int i; + + for (i = 0; i < GATE_EHDR->e_phnum; ++i) { + if (gate_phdrs[i].p_type == PT_LOAD) { + void *addr = (void *)gate_phdrs[i].p_vaddr; + size_t memsz = PAGE_ALIGN(gate_phdrs[i].p_memsz); + + *size += memsz; + if (*size > limit || !dump_write(file, addr, memsz)) + return 0; + break; + } + } + return 1; +} + +size_t elf_core_extra_data_size(void) +{ + const struct elf_phdr *const gate_phdrs = + (const struct elf_phdr *) (GATE_ADDR + GATE_EHDR->e_phoff); + int i; + size_t size = 0; + + for (i = 0; i < GATE_EHDR->e_phnum; ++i) { + if (gate_phdrs[i].p_type == PT_LOAD) { + size += PAGE_ALIGN(gate_phdrs[i].p_memsz); + break; + } + } + return size; +} diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c index b81e46b1629b..703062c44fb9 100644 --- a/arch/ia64/kernel/perfmon.c +++ b/arch/ia64/kernel/perfmon.c @@ -2315,6 +2315,7 @@ pfm_smpl_buffer_alloc(struct task_struct *task, struct file *filp, pfm_context_t DPRINT(("Cannot allocate vma\n")); goto error_kmem; } + INIT_LIST_HEAD(&vma->anon_vma_chain); /* * partially initialize the vma for the sampling buffer diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c index ca3335ea56cc..ed41759efcac 100644 --- a/arch/ia64/mm/init.c +++ b/arch/ia64/mm/init.c @@ -117,6 +117,7 @@ ia64_init_addr_space (void) */ vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); if (vma) { + INIT_LIST_HEAD(&vma->anon_vma_chain); vma->vm_mm = current->mm; vma->vm_start = current->thread.rbs_bot & PAGE_MASK; vma->vm_end = vma->vm_start + PAGE_SIZE; @@ -135,6 +136,7 @@ ia64_init_addr_space (void) if (!(current->personality & MMAP_PAGE_ZERO)) { vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); if (vma) { + INIT_LIST_HEAD(&vma->anon_vma_chain); vma->vm_mm = current->mm; vma->vm_end = PAGE_SIZE; vma->vm_page_prot = __pgprot(pgprot_val(PAGE_READONLY) | _PAGE_MA_NAT); diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c index b037d95eeadc..64c00227b997 100644 --- a/arch/powerpc/mm/numa.c +++ b/arch/powerpc/mm/numa.c @@ -451,7 +451,7 @@ static int __cpuinit numa_setup_cpu(unsigned long lcpu) nid = of_node_to_nid_single(cpu); if (nid < 0 || !node_online(nid)) - nid = any_online_node(NODE_MASK_ALL); + nid = first_online_node; out: map_cpu_to_node(lcpu, nid); @@ -1114,7 +1114,7 @@ int hot_add_scn_to_nid(unsigned long scn_addr) int nid, found = 0; if (!numa_enabled || (min_common_depth < 0)) - return any_online_node(NODE_MASK_ALL); + return first_online_node; memory = of_find_node_by_path("/ibm,dynamic-reconfiguration-memory"); if (memory) { @@ -1125,7 +1125,7 @@ int hot_add_scn_to_nid(unsigned long scn_addr) } if (nid < 0 || !node_online(nid)) - nid = any_online_node(NODE_MASK_ALL); + nid = first_online_node; if (NODE_DATA(nid)->node_spanned_pages) return nid; diff --git a/arch/um/.gitignore b/arch/um/.gitignore new file mode 100644 index 000000000000..a73d3a1cc746 --- /dev/null +++ b/arch/um/.gitignore @@ -0,0 +1,3 @@ +kernel/config.c +kernel/config.tmp +kernel/vmlinux.lds diff --git a/arch/um/drivers/line.c b/arch/um/drivers/line.c index cf8a97f34518..64cda95f59ca 100644 --- a/arch/um/drivers/line.c +++ b/arch/um/drivers/line.c @@ -18,10 +18,10 @@ static irqreturn_t line_interrupt(int irq, void *data) { struct chan *chan = data; struct line *line = chan->line; - struct tty_struct *tty = line->tty; + struct tty_struct *tty; if (line) - chan_interrupt(&line->chan_list, &line->task, tty, irq); + chan_interrupt(&line->chan_list, &line->task, line->tty, irq); return IRQ_HANDLED; } diff --git a/arch/um/sys-i386/Makefile b/arch/um/sys-i386/Makefile index 1b549bca4645..804b28dd0328 100644 --- a/arch/um/sys-i386/Makefile +++ b/arch/um/sys-i386/Makefile @@ -6,6 +6,8 @@ obj-y = bug.o bugs.o checksum.o delay.o fault.o ksyms.o ldt.o ptrace.o \ ptrace_user.o setjmp.o signal.o stub.o stub_segv.o syscalls.o sysrq.o \ sys_call_table.o tls.o +obj-$(CONFIG_BINFMT_ELF) += elfcore.o + subarch-obj-y = lib/semaphore_32.o lib/string_32.o subarch-obj-$(CONFIG_HIGHMEM) += mm/highmem_32.o subarch-obj-$(CONFIG_MODULES) += kernel/module.o diff --git a/arch/um/sys-i386/asm/elf.h b/arch/um/sys-i386/asm/elf.h index 770885472ed4..e64cd41d7bab 100644 --- a/arch/um/sys-i386/asm/elf.h +++ b/arch/um/sys-i386/asm/elf.h @@ -116,47 +116,4 @@ do { \ } \ } while (0) -/* - * These macros parameterize elf_core_dump in fs/binfmt_elf.c to write out - * extra segments containing the vsyscall DSO contents. Dumping its - * contents makes post-mortem fully interpretable later without matching up - * the same kernel and hardware config to see what PC values meant. - * Dumping its extra ELF program headers includes all the other information - * a debugger needs to easily find how the vsyscall DSO was being used. - */ -#define ELF_CORE_EXTRA_PHDRS \ - (vsyscall_ehdr ? (((struct elfhdr *)vsyscall_ehdr)->e_phnum) : 0 ) - -#define ELF_CORE_WRITE_EXTRA_PHDRS \ -if ( vsyscall_ehdr ) { \ - const struct elfhdr *const ehdrp = (struct elfhdr *)vsyscall_ehdr; \ - const struct elf_phdr *const phdrp = \ - (const struct elf_phdr *) (vsyscall_ehdr + ehdrp->e_phoff); \ - int i; \ - Elf32_Off ofs = 0; \ - for (i = 0; i < ehdrp->e_phnum; ++i) { \ - struct elf_phdr phdr = phdrp[i]; \ - if (phdr.p_type == PT_LOAD) { \ - ofs = phdr.p_offset = offset; \ - offset += phdr.p_filesz; \ - } \ - else \ - phdr.p_offset += ofs; \ - phdr.p_paddr = 0; /* match other core phdrs */ \ - DUMP_WRITE(&phdr, sizeof(phdr)); \ - } \ -} -#define ELF_CORE_WRITE_EXTRA_DATA \ -if ( vsyscall_ehdr ) { \ - const struct elfhdr *const ehdrp = (struct elfhdr *)vsyscall_ehdr; \ - const struct elf_phdr *const phdrp = \ - (const struct elf_phdr *) (vsyscall_ehdr + ehdrp->e_phoff); \ - int i; \ - for (i = 0; i < ehdrp->e_phnum; ++i) { \ - if (phdrp[i].p_type == PT_LOAD) \ - DUMP_WRITE((void *) phdrp[i].p_vaddr, \ - phdrp[i].p_filesz); \ - } \ -} - #endif diff --git a/arch/um/sys-i386/elfcore.c b/arch/um/sys-i386/elfcore.c new file mode 100644 index 000000000000..6bb49b687c97 --- /dev/null +++ b/arch/um/sys-i386/elfcore.c @@ -0,0 +1,83 @@ +#include <linux/elf.h> +#include <linux/coredump.h> +#include <linux/fs.h> +#include <linux/mm.h> + +#include <asm/elf.h> + + +Elf32_Half elf_core_extra_phdrs(void) +{ + return vsyscall_ehdr ? (((struct elfhdr *)vsyscall_ehdr)->e_phnum) : 0; +} + +int elf_core_write_extra_phdrs(struct file *file, loff_t offset, size_t *size, + unsigned long limit) +{ + if ( vsyscall_ehdr ) { + const struct elfhdr *const ehdrp = + (struct elfhdr *) vsyscall_ehdr; + const struct elf_phdr *const phdrp = + (const struct elf_phdr *) (vsyscall_ehdr + ehdrp->e_phoff); + int i; + Elf32_Off ofs = 0; + + for (i = 0; i < ehdrp->e_phnum; ++i) { + struct elf_phdr phdr = phdrp[i]; + + if (phdr.p_type == PT_LOAD) { + ofs = phdr.p_offset = offset; + offset += phdr.p_filesz; + } else { + phdr.p_offset += ofs; + } + phdr.p_paddr = 0; /* match other core phdrs */ + *size += sizeof(phdr); + if (*size > limit + || !dump_write(file, &phdr, sizeof(phdr))) + return 0; + } + } + return 1; +} + +int elf_core_write_extra_data(struct file *file, size_t *size, + unsigned long limit) +{ + if ( vsyscall_ehdr ) { + const struct elfhdr *const ehdrp = + (struct elfhdr *) vsyscall_ehdr; + const struct elf_phdr *const phdrp = + (const struct elf_phdr *) (vsyscall_ehdr + ehdrp->e_phoff); + int i; + + for (i = 0; i < ehdrp->e_phnum; ++i) { + if (phdrp[i].p_type == PT_LOAD) { + void *addr = (void *) phdrp[i].p_vaddr; + size_t filesz = phdrp[i].p_filesz; + + *size += filesz; + if (*size > limit + || !dump_write(file, addr, filesz)) + return 0; + } + } + } + return 1; +} + +size_t elf_core_extra_data_size(void) +{ + if ( vsyscall_ehdr ) { + const struct elfhdr *const ehdrp = + (struct elfhdr *)vsyscall_ehdr; + const struct elf_phdr *const phdrp = + (const struct elf_phdr *) (vsyscall_ehdr + ehdrp->e_phoff); + int i; + + for (i = 0; i < ehdrp->e_phnum; ++i) + if (phdrp[i].p_type == PT_LOAD) + return (size_t) phdrp[i].p_filesz; + } + return 0; +} diff --git a/arch/x86/kernel/cpu/cpufreq/Kconfig b/arch/x86/kernel/cpu/cpufreq/Kconfig index f138c6c389b9..870e6cc6ad28 100644 --- a/arch/x86/kernel/cpu/cpufreq/Kconfig +++ b/arch/x86/kernel/cpu/cpufreq/Kconfig @@ -10,6 +10,20 @@ if CPU_FREQ comment "CPUFreq processor drivers" +config X86_PCC_CPUFREQ + tristate "Processor Clocking Control interface driver" + depends on ACPI && ACPI_PROCESSOR + help + This driver adds support for the PCC interface. + + For details, take a look at: + <file:Documentation/cpu-freq/pcc-cpufreq.txt>. + + To compile this driver as a module, choose M here: the + module will be called pcc-cpufreq. + + If in doubt, say N. + config X86_ACPI_CPUFREQ tristate "ACPI Processor P-States driver" select CPU_FREQ_TABLE diff --git a/arch/x86/kernel/cpu/cpufreq/Makefile b/arch/x86/kernel/cpu/cpufreq/Makefile index 509296df294d..1840c0a5170b 100644 --- a/arch/x86/kernel/cpu/cpufreq/Makefile +++ b/arch/x86/kernel/cpu/cpufreq/Makefile @@ -4,6 +4,7 @@ obj-$(CONFIG_X86_POWERNOW_K8) += powernow-k8.o obj-$(CONFIG_X86_ACPI_CPUFREQ) += acpi-cpufreq.o +obj-$(CONFIG_X86_PCC_CPUFREQ) += pcc-cpufreq.o obj-$(CONFIG_X86_POWERNOW_K6) += powernow-k6.o obj-$(CONFIG_X86_POWERNOW_K7) += powernow-k7.o obj-$(CONFIG_X86_LONGHAUL) += longhaul.o diff --git a/arch/x86/kernel/cpu/cpufreq/pcc-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/pcc-cpufreq.c new file mode 100644 index 000000000000..ff36d2979a90 --- /dev/null +++ b/arch/x86/kernel/cpu/cpufreq/pcc-cpufreq.c @@ -0,0 +1,620 @@ +/* + * pcc-cpufreq.c - Processor Clocking Control firmware cpufreq interface + * + * Copyright (C) 2009 Red Hat, Matthew Garrett <mjg@redhat.com> + * Copyright (C) 2009 Hewlett-Packard Development Company, L.P. + * Nagananda Chumbalkar <nagananda.chumbalkar@hp.com> + * + * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or NON + * INFRINGEMENT. See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License along + * with this program; if not, write to the Free Software Foundation, Inc., + * 675 Mass Ave, Cambridge, MA 02139, USA. + * + * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + */ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/smp.h> +#include <linux/sched.h> +#include <linux/cpufreq.h> +#include <linux/compiler.h> + +#include <linux/acpi.h> +#include <linux/io.h> +#include <linux/spinlock.h> +#include <linux/uaccess.h> + +#include <acpi/processor.h> + +#define PCC_VERSION "1.00.00" +#define POLL_LOOPS 300 + +#define CMD_COMPLETE 0x1 +#define CMD_GET_FREQ 0x0 +#define CMD_SET_FREQ 0x1 + +#define BUF_SZ 4 + +#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ + "pcc-cpufreq", msg) + +struct pcc_register_resource { + u8 descriptor; + u16 length; + u8 space_id; + u8 bit_width; + u8 bit_offset; + u8 access_size; + u64 address; +} __attribute__ ((packed)); + +struct pcc_memory_resource { + u8 descriptor; + u16 length; + u8 space_id; + u8 resource_usage; + u8 type_specific; + u64 granularity; + u64 minimum; + u64 maximum; + u64 translation_offset; + u64 address_length; +} __attribute__ ((packed)); + +static struct cpufreq_driver pcc_cpufreq_driver; + +struct pcc_header { + u32 signature; + u16 length; + u8 major; + u8 minor; + u32 features; + u16 command; + u16 status; + u32 latency; + u32 minimum_time; + u32 maximum_time; + u32 nominal; + u32 throttled_frequency; + u32 minimum_frequency; +}; + +static void __iomem *pcch_virt_addr; +static struct pcc_header __iomem *pcch_hdr; + +static DEFINE_SPINLOCK(pcc_lock); + +static struct acpi_generic_address doorbell; + +static u64 doorbell_preserve; +static u64 doorbell_write; + +static u8 OSC_UUID[16] = {0x63, 0x9B, 0x2C, 0x9F, 0x70, 0x91, 0x49, 0x1f, + 0xBB, 0x4F, 0xA5, 0x98, 0x2F, 0xA1, 0xB5, 0x46}; + +struct pcc_cpu { + u32 input_offset; + u32 output_offset; +}; + +static struct pcc_cpu *pcc_cpu_info; + +static int pcc_cpufreq_verify(struct cpufreq_policy *policy) +{ + cpufreq_verify_within_limits(policy, policy->cpuinfo.min_freq, + policy->cpuinfo.max_freq); + return 0; +} + +static inline void pcc_cmd(void) +{ + u64 doorbell_value; + int i; + + acpi_read(&doorbell_value, &doorbell); + acpi_write((doorbell_value & doorbell_preserve) | doorbell_write, + &doorbell); + + for (i = 0; i < POLL_LOOPS; i++) { + if (ioread16(&pcch_hdr->status) & CMD_COMPLETE) + break; + } +} + +static inline void pcc_clear_mapping(void) +{ + if (pcch_virt_addr) + iounmap(pcch_virt_addr); + pcch_virt_addr = NULL; +} + +static unsigned int pcc_get_freq(unsigned int cpu) +{ + struct pcc_cpu *pcc_cpu_data; + unsigned int curr_freq; + unsigned int freq_limit; + u16 status; + u32 input_buffer; + u32 output_buffer; + + spin_lock(&pcc_lock); + + dprintk("get: get_freq for CPU %d\n", cpu); + pcc_cpu_data = per_cpu_ptr(pcc_cpu_info, cpu); + + input_buffer = 0x1; + iowrite32(input_buffer, + (pcch_virt_addr + pcc_cpu_data->input_offset)); + iowrite16(CMD_GET_FREQ, &pcch_hdr->command); + + pcc_cmd(); + + output_buffer = + ioread32(pcch_virt_addr + pcc_cpu_data->output_offset); + + /* Clear the input buffer - we are done with the current command */ + memset_io((pcch_virt_addr + pcc_cpu_data->input_offset), 0, BUF_SZ); + + status = ioread16(&pcch_hdr->status); + if (status != CMD_COMPLETE) { + dprintk("get: FAILED: for CPU %d, status is %d\n", + cpu, status); + goto cmd_incomplete; + } + iowrite16(0, &pcch_hdr->status); + curr_freq = (((ioread32(&pcch_hdr->nominal) * (output_buffer & 0xff)) + / 100) * 1000); + + dprintk("get: SUCCESS: (virtual) output_offset for cpu %d is " + "0x%x, contains a value of: 0x%x. Speed is: %d MHz\n", + cpu, (pcch_virt_addr + pcc_cpu_data->output_offset), + output_buffer, curr_freq); + + freq_limit = (output_buffer >> 8) & 0xff; + if (freq_limit != 0xff) { + dprintk("get: frequency for cpu %d is being temporarily" + " capped at %d\n", cpu, curr_freq); + } + + spin_unlock(&pcc_lock); + return curr_freq; + +cmd_incomplete: + iowrite16(0, &pcch_hdr->status); + spin_unlock(&pcc_lock); + return -EINVAL; +} + +static int pcc_cpufreq_target(struct cpufreq_policy *policy, + unsigned int target_freq, + unsigned int relation) +{ + struct pcc_cpu *pcc_cpu_data; + struct cpufreq_freqs freqs; + u16 status; + u32 input_buffer; + int cpu; + + spin_lock(&pcc_lock); + cpu = policy->cpu; + pcc_cpu_data = per_cpu_ptr(pcc_cpu_info, cpu); + + dprintk("target: CPU %d should go to target freq: %d " + "(virtual) input_offset is 0x%x\n", + cpu, target_freq, + (pcch_virt_addr + pcc_cpu_data->input_offset)); + + freqs.new = target_freq; + freqs.cpu = cpu; + cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); + + input_buffer = 0x1 | (((target_freq * 100) + / (ioread32(&pcch_hdr->nominal) * 1000)) << 8); + iowrite32(input_buffer, + (pcch_virt_addr + pcc_cpu_data->input_offset)); + iowrite16(CMD_SET_FREQ, &pcch_hdr->command); + + pcc_cmd(); + + /* Clear the input buffer - we are done with the current command */ + memset_io((pcch_virt_addr + pcc_cpu_data->input_offset), 0, BUF_SZ); + + status = ioread16(&pcch_hdr->status); + if (status != CMD_COMPLETE) { + dprintk("target: FAILED for cpu %d, with status: 0x%x\n", + cpu, status); + goto cmd_incomplete; + } + iowrite16(0, &pcch_hdr->status); + + cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); + dprintk("target: was SUCCESSFUL for cpu %d\n", cpu); + spin_unlock(&pcc_lock); + + return 0; + +cmd_incomplete: + iowrite16(0, &pcch_hdr->status); + spin_unlock(&pcc_lock); + return -EINVAL; +} + +static int pcc_get_offset(int cpu) +{ + acpi_status status; + struct acpi_buffer buffer = {ACPI_ALLOCATE_BUFFER, NULL}; + union acpi_object *pccp, *offset; + struct pcc_cpu *pcc_cpu_data; + struct acpi_processor *pr; + int ret = 0; + + pr = per_cpu(processors, cpu); + pcc_cpu_data = per_cpu_ptr(pcc_cpu_info, cpu); + + status = acpi_evaluate_object(pr->handle, "PCCP", NULL, &buffer); + if (ACPI_FAILURE(status)) + return -ENODEV; + + pccp = buffer.pointer; + if (!pccp || pccp->type != ACPI_TYPE_PACKAGE) { + ret = -ENODEV; + goto out_free; + }; + + offset = &(pccp->package.elements[0]); + if (!offset || offset->type != ACPI_TYPE_INTEGER) { + ret = -ENODEV; + goto out_free; + } + + pcc_cpu_data->input_offset = offset->integer.value; + + offset = &(pccp->package.elements[1]); + if (!offset || offset->type != ACPI_TYPE_INTEGER) { + ret = -ENODEV; + goto out_free; + } + + pcc_cpu_data->output_offset = offset->integer.value; + + memset_io((pcch_virt_addr + pcc_cpu_data->input_offset), 0, BUF_SZ); + memset_io((pcch_virt_addr + pcc_cpu_data->output_offset), 0, BUF_SZ); + + dprintk("pcc_get_offset: for CPU %d: pcc_cpu_data " + "input_offset: 0x%x, pcc_cpu_data output_offset: 0x%x\n", + cpu, pcc_cpu_data->input_offset, pcc_cpu_data->output_offset); +out_free: + kfree(buffer.pointer); + return ret; +} + +static int __init pcc_cpufreq_do_osc(acpi_handle *handle) +{ + acpi_status status; + struct acpi_object_list input; + struct acpi_buffer output = {ACPI_ALLOCATE_BUFFER, NULL}; + union acpi_object in_params[4]; + union acpi_object *out_obj; + u32 capabilities[2]; + u32 errors; + u32 supported; + int ret = 0; + + input.count = 4; + input.pointer = in_params; + input.count = 4; + input.pointer = in_params; + in_params[0].type = ACPI_TYPE_BUFFER; + in_params[0].buffer.length = 16; + in_params[0].buffer.pointer = OSC_UUID; + in_params[1].type = ACPI_TYPE_INTEGER; + in_params[1].integer.value = 1; + in_params[2].type = ACPI_TYPE_INTEGER; + in_params[2].integer.value = 2; + in_params[3].type = ACPI_TYPE_BUFFER; + in_params[3].buffer.length = 8; + in_params[3].buffer.pointer = (u8 *)&capabilities; + + capabilities[0] = OSC_QUERY_ENABLE; + capabilities[1] = 0x1; + + status = acpi_evaluate_object(*handle, "_OSC", &input, &output); + if (ACPI_FAILURE(status)) + return -ENODEV; + + if (!output.length) + return -ENODEV; + + out_obj = output.pointer; + if (out_obj->type != ACPI_TYPE_BUFFER) { + ret = -ENODEV; + goto out_free; + } + + errors = *((u32 *)out_obj->buffer.pointer) & ~(1 << 0); + if (errors) { + ret = -ENODEV; + goto out_free; + } + + supported = *((u32 *)(out_obj->buffer.pointer + 4)); + if (!(supported & 0x1)) { + ret = -ENODEV; + goto out_free; + } + + kfree(output.pointer); + capabilities[0] = 0x0; + capabilities[1] = 0x1; + + status = acpi_evaluate_object(*handle, "_OSC", &input, &output); + if (ACPI_FAILURE(status)) + return -ENODEV; + + if (!output.length) + return -ENODEV; + + out_obj = output.pointer; + if (out_obj->type != ACPI_TYPE_BUFFER) { + ret = -ENODEV; + goto out_free; + } + + errors = *((u32 *)out_obj->buffer.pointer) & ~(1 << 0); + if (errors) { + ret = -ENODEV; + goto out_free; + } + + supported = *((u32 *)(out_obj->buffer.pointer + 4)); + if (!(supported & 0x1)) { + ret = -ENODEV; + goto out_free; + } + +out_free: + kfree(output.pointer); + return ret; +} + +static int __init pcc_cpufreq_probe(void) +{ + acpi_status status; + struct acpi_buffer output = {ACPI_ALLOCATE_BUFFER, NULL}; + struct pcc_memory_resource *mem_resource; + struct pcc_register_resource *reg_resource; + union acpi_object *out_obj, *member; + acpi_handle handle, osc_handle; + int ret = 0; + + status = acpi_get_handle(NULL, "\\_SB", &handle); + if (ACPI_FAILURE(status)) + return -ENODEV; + + status = acpi_get_handle(handle, "_OSC", &osc_handle); + if (ACPI_SUCCESS(status)) { + ret = pcc_cpufreq_do_osc(&osc_handle); + if (ret) + dprintk("probe: _OSC evaluation did not succeed\n"); + /* Firmware's use of _OSC is optional */ + ret = 0; + } + + status = acpi_evaluate_object(handle, "PCCH", NULL, &output); + if (ACPI_FAILURE(status)) + return -ENODEV; + + out_obj = output.pointer; + if (out_obj->type != ACPI_TYPE_PACKAGE) { + ret = -ENODEV; + goto out_free; + } + + member = &out_obj->package.elements[0]; + if (member->type != ACPI_TYPE_BUFFER) { + ret = -ENODEV; + goto out_free; + } + + mem_resource = (struct pcc_memory_resource *)member->buffer.pointer; + + dprintk("probe: mem_resource descriptor: 0x%x," + " length: %d, space_id: %d, resource_usage: %d," + " type_specific: %d, granularity: 0x%llx," + " minimum: 0x%llx, maximum: 0x%llx," + " translation_offset: 0x%llx, address_length: 0x%llx\n", + mem_resource->descriptor, mem_resource->length, + mem_resource->space_id, mem_resource->resource_usage, + mem_resource->type_specific, mem_resource->granularity, + mem_resource->minimum, mem_resource->maximum, + mem_resource->translation_offset, + mem_resource->address_length); + + if (mem_resource->space_id != ACPI_ADR_SPACE_SYSTEM_MEMORY) { + ret = -ENODEV; + goto out_free; + } + + pcch_virt_addr = ioremap_nocache(mem_resource->minimum, + mem_resource->address_length); + if (pcch_virt_addr == NULL) { + dprintk("probe: could not map shared mem region\n"); + goto out_free; + } + pcch_hdr = pcch_virt_addr; + + dprintk("probe: PCCH header (virtual) addr: 0x%p\n", pcch_hdr); + dprintk("probe: PCCH header is at physical address: 0x%llx," + " signature: 0x%x, length: %d bytes, major: %d, minor: %d," + " supported features: 0x%x, command field: 0x%x," + " status field: 0x%x, nominal latency: %d us\n", + mem_resource->minimum, ioread32(&pcch_hdr->signature), + ioread16(&pcch_hdr->length), ioread8(&pcch_hdr->major), + ioread8(&pcch_hdr->minor), ioread32(&pcch_hdr->features), + ioread16(&pcch_hdr->command), ioread16(&pcch_hdr->status), + ioread32(&pcch_hdr->latency)); + + dprintk("probe: min time between commands: %d us," + " max time between commands: %d us," + " nominal CPU frequency: %d MHz," + " minimum CPU frequency: %d MHz," + " minimum CPU frequency without throttling: %d MHz\n", + ioread32(&pcch_hdr->minimum_time), + ioread32(&pcch_hdr->maximum_time), + ioread32(&pcch_hdr->nominal), + ioread32(&pcch_hdr->throttled_frequency), + ioread32(&pcch_hdr->minimum_frequency)); + + member = &out_obj->package.elements[1]; + if (member->type != ACPI_TYPE_BUFFER) { + ret = -ENODEV; + goto pcch_free; + } + + reg_resource = (struct pcc_register_resource *)member->buffer.pointer; + + doorbell.space_id = reg_resource->space_id; + doorbell.bit_width = reg_resource->bit_width; + doorbell.bit_offset = reg_resource->bit_offset; + doorbell.access_width = 64; + doorbell.address = reg_resource->address; + + dprintk("probe: doorbell: space_id is %d, bit_width is %d, " + "bit_offset is %d, access_width is %d, address is 0x%llx\n", + doorbell.space_id, doorbell.bit_width, doorbell.bit_offset, + doorbell.access_width, reg_resource->address); + + member = &out_obj->package.elements[2]; + if (member->type != ACPI_TYPE_INTEGER) { + ret = -ENODEV; + goto pcch_free; + } + + doorbell_preserve = member->integer.value; + + member = &out_obj->package.elements[3]; + if (member->type != ACPI_TYPE_INTEGER) { + ret = -ENODEV; + goto pcch_free; + } + + doorbell_write = member->integer.value; + + dprintk("probe: doorbell_preserve: 0x%llx," + " doorbell_write: 0x%llx\n", + doorbell_preserve, doorbell_write); + + pcc_cpu_info = alloc_percpu(struct pcc_cpu); + if (!pcc_cpu_info) { + ret = -ENOMEM; + goto pcch_free; + } + + printk(KERN_DEBUG "pcc-cpufreq: (v%s) driver loaded with frequency" + " limits: %d MHz, %d MHz\n", PCC_VERSION, + ioread32(&pcch_hdr->minimum_frequency), + ioread32(&pcch_hdr->nominal)); + kfree(output.pointer); + return ret; +pcch_free: + pcc_clear_mapping(); +out_free: + kfree(output.pointer); + return ret; +} + +static int pcc_cpufreq_cpu_init(struct cpufreq_policy *policy) +{ + unsigned int cpu = policy->cpu; + unsigned int result = 0; + + if (!pcch_virt_addr) { + result = -1; + goto pcch_null; + } + + result = pcc_get_offset(cpu); + if (result) { + dprintk("init: PCCP evaluation failed\n"); + goto free; + } + + policy->max = policy->cpuinfo.max_freq = + ioread32(&pcch_hdr->nominal) * 1000; + policy->min = policy->cpuinfo.min_freq = + ioread32(&pcch_hdr->minimum_frequency) * 1000; + policy->cur = pcc_get_freq(cpu); + + dprintk("init: policy->max is %d, policy->min is %d\n", + policy->max, policy->min); + + return 0; +free: + pcc_clear_mapping(); + free_percpu(pcc_cpu_info); +pcch_null: + return result; +} + +static int pcc_cpufreq_cpu_exit(struct cpufreq_policy *policy) +{ + return 0; +} + +static struct cpufreq_driver pcc_cpufreq_driver = { + .flags = CPUFREQ_CONST_LOOPS, + .get = pcc_get_freq, + .verify = pcc_cpufreq_verify, + .target = pcc_cpufreq_target, + .init = pcc_cpufreq_cpu_init, + .exit = pcc_cpufreq_cpu_exit, + .name = "pcc-cpufreq", + .owner = THIS_MODULE, +}; + +static int __init pcc_cpufreq_init(void) +{ + int ret; + + if (acpi_disabled) + return 0; + + ret = pcc_cpufreq_probe(); + if (ret) { + dprintk("pcc_cpufreq_init: PCCH evaluation failed\n"); + return ret; + } + + ret = cpufreq_register_driver(&pcc_cpufreq_driver); + + return ret; +} + +static void __exit pcc_cpufreq_exit(void) +{ + cpufreq_unregister_driver(&pcc_cpufreq_driver); + + pcc_clear_mapping(); + + free_percpu(pcc_cpu_info); +} + +MODULE_AUTHOR("Matthew Garrett, Naga Chumbalkar"); +MODULE_VERSION(PCC_VERSION); +MODULE_DESCRIPTION("Processor Clocking Control interface driver"); +MODULE_LICENSE("GPL"); + +late_initcall(pcc_cpufreq_init); +module_exit(pcc_cpufreq_exit); diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 641ccb9dddbc..b1fbdeecf6c9 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -676,7 +676,7 @@ static int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign) if (c->weight != w) continue; - for_each_bit(j, c->idxmsk, X86_PMC_IDX_MAX) { + for_each_set_bit(j, c->idxmsk, X86_PMC_IDX_MAX) { if (!test_bit(j, used_mask)) break; } diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index cf6590cf4a5f..977e7544738c 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -757,7 +757,7 @@ again: inc_irq_stat(apic_perf_irqs); ack = status; - for_each_bit(bit, (unsigned long *)&status, X86_PMC_IDX_MAX) { + for_each_set_bit(bit, (unsigned long *)&status, X86_PMC_IDX_MAX) { struct perf_event *event = cpuc->events[bit]; clear_bit(bit, (unsigned long *) &status); diff --git a/drivers/Kconfig b/drivers/Kconfig index 368ae6d3a096..a2b902f4d437 100644 --- a/drivers/Kconfig +++ b/drivers/Kconfig @@ -96,8 +96,6 @@ source "drivers/edac/Kconfig" source "drivers/rtc/Kconfig" -source "drivers/clocksource/Kconfig" - source "drivers/dma/Kconfig" source "drivers/dca/Kconfig" diff --git a/drivers/acpi/processor_core.c b/drivers/acpi/processor_core.c index 9863c98c81ba..e9b7b402dbfb 100644 --- a/drivers/acpi/processor_core.c +++ b/drivers/acpi/processor_core.c @@ -123,6 +123,8 @@ static const struct file_operations acpi_processor_info_fops = { #endif DEFINE_PER_CPU(struct acpi_processor *, processors); +EXPORT_PER_CPU_SYMBOL(processors); + struct acpi_processor_errata errata __read_mostly; /* -------------------------------------------------------------------------- diff --git a/drivers/clocksource/Kconfig b/drivers/clocksource/Kconfig deleted file mode 100644 index 08f726c5fee5..000000000000 --- a/drivers/clocksource/Kconfig +++ /dev/null @@ -1,9 +0,0 @@ -config CS5535_CLOCK_EVENT_SRC - tristate "CS5535/CS5536 high-res timer (MFGPT) events" - depends on GENERIC_TIME && GENERIC_CLOCKEVENTS && CS5535_MFGPT - help - This driver provides a clock event source based on the MFGPT - timer(s) in the CS5535 and CS5536 companion chips. - MFGPTs have a better resolution and max interval than the - generic PIT, and are suitable for use as high-res timers. - diff --git a/drivers/cpuidle/governors/menu.c b/drivers/cpuidle/governors/menu.c index 73655aeb3a60..1aea7157d8ff 100644 --- a/drivers/cpuidle/governors/menu.c +++ b/drivers/cpuidle/governors/menu.c @@ -100,8 +100,8 @@ struct menu_device { int needs_update; unsigned int expected_us; - u64 predicted_us; unsigned int measured_us; + u64 predicted_us; unsigned int exit_us; unsigned int bucket; u64 correction_factor[BUCKETS]; diff --git a/drivers/dma/ioat/dma.c b/drivers/dma/ioat/dma.c index 5d0e42b263df..af14c9a5b8d4 100644 --- a/drivers/dma/ioat/dma.c +++ b/drivers/dma/ioat/dma.c @@ -71,7 +71,7 @@ static irqreturn_t ioat_dma_do_interrupt(int irq, void *data) } attnstatus = readl(instance->reg_base + IOAT_ATTNSTATUS_OFFSET); - for_each_bit(bit, &attnstatus, BITS_PER_LONG) { + for_each_set_bit(bit, &attnstatus, BITS_PER_LONG) { chan = ioat_chan_by_index(instance, bit); tasklet_schedule(&chan->cleanup_task); } diff --git a/drivers/eisa/eisa-bus.c b/drivers/eisa/eisa-bus.c index 66958b3f10b4..806c77bfd434 100644 --- a/drivers/eisa/eisa-bus.c +++ b/drivers/eisa/eisa-bus.c @@ -39,10 +39,10 @@ static unsigned int enable_dev_count; static int disable_dev[EISA_MAX_FORCED_DEV]; static unsigned int disable_dev_count; -static int is_forced_dev (int *forced_tab, - int forced_count, - struct eisa_root_device *root, - struct eisa_device *edev) +static int is_forced_dev(int *forced_tab, + int forced_count, + struct eisa_root_device *root, + struct eisa_device *edev) { int i, x; @@ -55,21 +55,21 @@ static int is_forced_dev (int *forced_tab, return 0; } -static void __init eisa_name_device (struct eisa_device *edev) +static void __init eisa_name_device(struct eisa_device *edev) { #ifdef CONFIG_EISA_NAMES int i; for (i = 0; i < EISA_INFOS; i++) { - if (!strcmp (edev->id.sig, eisa_table[i].id.sig)) { - strlcpy (edev->pretty_name, - eisa_table[i].name, - sizeof(edev->pretty_name)); + if (!strcmp(edev->id.sig, eisa_table[i].id.sig)) { + strlcpy(edev->pretty_name, + eisa_table[i].name, + sizeof(edev->pretty_name)); return; } } /* No name was found */ - sprintf (edev->pretty_name, "EISA device %.7s", edev->id.sig); + sprintf(edev->pretty_name, "EISA device %.7s", edev->id.sig); #endif } @@ -91,7 +91,7 @@ static char __init *decode_eisa_sig(unsigned long addr) */ outb(0x80 + i, addr); #endif - sig[i] = inb (addr + i); + sig[i] = inb(addr + i); if (!i && (sig[0] & 0x80)) return NULL; @@ -106,17 +106,17 @@ static char __init *decode_eisa_sig(unsigned long addr) return sig_str; } -static int eisa_bus_match (struct device *dev, struct device_driver *drv) +static int eisa_bus_match(struct device *dev, struct device_driver *drv) { - struct eisa_device *edev = to_eisa_device (dev); - struct eisa_driver *edrv = to_eisa_driver (drv); + struct eisa_device *edev = to_eisa_device(dev); + struct eisa_driver *edrv = to_eisa_driver(drv); const struct eisa_device_id *eids = edrv->id_table; if (!eids) return 0; - while (strlen (eids->sig)) { - if (!strcmp (eids->sig, edev->id.sig) && + while (strlen(eids->sig)) { + if (!strcmp(eids->sig, edev->id.sig) && edev->state & EISA_CONFIG_ENABLED) { edev->id.driver_data = eids->driver_data; return 1; @@ -141,61 +141,71 @@ struct bus_type eisa_bus_type = { .match = eisa_bus_match, .uevent = eisa_bus_uevent, }; +EXPORT_SYMBOL(eisa_bus_type); -int eisa_driver_register (struct eisa_driver *edrv) +int eisa_driver_register(struct eisa_driver *edrv) { edrv->driver.bus = &eisa_bus_type; - return driver_register (&edrv->driver); + return driver_register(&edrv->driver); } +EXPORT_SYMBOL(eisa_driver_register); -void eisa_driver_unregister (struct eisa_driver *edrv) +void eisa_driver_unregister(struct eisa_driver *edrv) { - driver_unregister (&edrv->driver); + driver_unregister(&edrv->driver); } +EXPORT_SYMBOL(eisa_driver_unregister); -static ssize_t eisa_show_sig (struct device *dev, struct device_attribute *attr, char *buf) +static ssize_t eisa_show_sig(struct device *dev, struct device_attribute *attr, + char *buf) { - struct eisa_device *edev = to_eisa_device (dev); - return sprintf (buf,"%s\n", edev->id.sig); + struct eisa_device *edev = to_eisa_device(dev); + return sprintf(buf, "%s\n", edev->id.sig); } static DEVICE_ATTR(signature, S_IRUGO, eisa_show_sig, NULL); -static ssize_t eisa_show_state (struct device *dev, struct device_attribute *attr, char *buf) +static ssize_t eisa_show_state(struct device *dev, + struct device_attribute *attr, + char *buf) { - struct eisa_device *edev = to_eisa_device (dev); - return sprintf (buf,"%d\n", edev->state & EISA_CONFIG_ENABLED); + struct eisa_device *edev = to_eisa_device(dev); + return sprintf(buf, "%d\n", edev->state & EISA_CONFIG_ENABLED); } static DEVICE_ATTR(enabled, S_IRUGO, eisa_show_state, NULL); -static ssize_t eisa_show_modalias (struct device *dev, struct device_attribute *attr, char *buf) +static ssize_t eisa_show_modalias(struct device *dev, + struct device_attribute *attr, + char *buf) { - struct eisa_device *edev = to_eisa_device (dev); - return sprintf (buf, EISA_DEVICE_MODALIAS_FMT "\n", edev->id.sig); + struct eisa_device *edev = to_eisa_device(dev); + return sprintf(buf, EISA_DEVICE_MODALIAS_FMT "\n", edev->id.sig); } static DEVICE_ATTR(modalias, S_IRUGO, eisa_show_modalias, NULL); -static int __init eisa_init_device (struct eisa_root_device *root, - struct eisa_device *edev, - int slot) +static int __init eisa_init_device(struct eisa_root_device *root, + struct eisa_device *edev, + int slot) { char *sig; - unsigned long sig_addr; + unsigned long sig_addr; int i; - sig_addr = SLOT_ADDRESS (root, slot) + EISA_VENDOR_ID_OFFSET; + sig_addr = SLOT_ADDRESS(root, slot) + EISA_VENDOR_ID_OFFSET; - if (!(sig = decode_eisa_sig (sig_addr))) + sig = decode_eisa_sig(sig_addr); + if (!sig) return -1; /* No EISA device here */ - memcpy (edev->id.sig, sig, EISA_SIG_LEN); + memcpy(edev->id.sig, sig, EISA_SIG_LEN); edev->slot = slot; - edev->state = inb (SLOT_ADDRESS (root, slot) + EISA_CONFIG_OFFSET) & EISA_CONFIG_ENABLED; - edev->base_addr = SLOT_ADDRESS (root, slot); + edev->state = inb(SLOT_ADDRESS(root, slot) + EISA_CONFIG_OFFSET) + & EISA_CONFIG_ENABLED; + edev->base_addr = SLOT_ADDRESS(root, slot); edev->dma_mask = root->dma_mask; /* Default DMA mask */ - eisa_name_device (edev); + eisa_name_device(edev); edev->dev.parent = root->dev; edev->dev.bus = &eisa_bus_type; edev->dev.dma_mask = &edev->dma_mask; @@ -210,42 +220,45 @@ static int __init eisa_init_device (struct eisa_root_device *root, #endif } - if (is_forced_dev (enable_dev, enable_dev_count, root, edev)) + if (is_forced_dev(enable_dev, enable_dev_count, root, edev)) edev->state = EISA_CONFIG_ENABLED | EISA_CONFIG_FORCED; - if (is_forced_dev (disable_dev, disable_dev_count, root, edev)) + if (is_forced_dev(disable_dev, disable_dev_count, root, edev)) edev->state = EISA_CONFIG_FORCED; return 0; } -static int __init eisa_register_device (struct eisa_device *edev) +static int __init eisa_register_device(struct eisa_device *edev) { - int rc = device_register (&edev->dev); + int rc = device_register(&edev->dev); if (rc) return rc; - rc = device_create_file (&edev->dev, &dev_attr_signature); - if (rc) goto err_devreg; - rc = device_create_file (&edev->dev, &dev_attr_enabled); - if (rc) goto err_sig; - rc = device_create_file (&edev->dev, &dev_attr_modalias); - if (rc) goto err_enab; + rc = device_create_file(&edev->dev, &dev_attr_signature); + if (rc) + goto err_devreg; + rc = device_create_file(&edev->dev, &dev_attr_enabled); + if (rc) + goto err_sig; + rc = device_create_file(&edev->dev, &dev_attr_modalias); + if (rc) + goto err_enab; return 0; err_enab: - device_remove_file (&edev->dev, &dev_attr_enabled); + device_remove_file(&edev->dev, &dev_attr_enabled); err_sig: - device_remove_file (&edev->dev, &dev_attr_signature); + device_remove_file(&edev->dev, &dev_attr_signature); err_devreg: device_unregister(&edev->dev); return rc; } -static int __init eisa_request_resources (struct eisa_root_device *root, - struct eisa_device *edev, - int slot) +static int __init eisa_request_resources(struct eisa_root_device *root, + struct eisa_device *edev, + int slot) { int i; @@ -263,17 +276,19 @@ static int __init eisa_request_resources (struct eisa_root_device *root, if (slot) { edev->res[i].name = NULL; - edev->res[i].start = SLOT_ADDRESS (root, slot) + (i * 0x400); + edev->res[i].start = SLOT_ADDRESS(root, slot) + + (i * 0x400); edev->res[i].end = edev->res[i].start + 0xff; edev->res[i].flags = IORESOURCE_IO; } else { edev->res[i].name = NULL; - edev->res[i].start = SLOT_ADDRESS (root, slot) + EISA_VENDOR_ID_OFFSET; + edev->res[i].start = SLOT_ADDRESS(root, slot) + + EISA_VENDOR_ID_OFFSET; edev->res[i].end = edev->res[i].start + 3; edev->res[i].flags = IORESOURCE_BUSY; } - if (request_resource (root->res, &edev->res[i])) + if (request_resource(root->res, &edev->res[i])) goto failed; } @@ -281,99 +296,100 @@ static int __init eisa_request_resources (struct eisa_root_device *root, failed: while (--i >= 0) - release_resource (&edev->res[i]); + release_resource(&edev->res[i]); return -1; } -static void __init eisa_release_resources (struct eisa_device *edev) +static void __init eisa_release_resources(struct eisa_device *edev) { int i; for (i = 0; i < EISA_MAX_RESOURCES; i++) if (edev->res[i].start || edev->res[i].end) - release_resource (&edev->res[i]); + release_resource(&edev->res[i]); } -static int __init eisa_probe (struct eisa_root_device *root) +static int __init eisa_probe(struct eisa_root_device *root) { int i, c; struct eisa_device *edev; - printk (KERN_INFO "EISA: Probing bus %d at %s\n", - root->bus_nr, dev_name(root->dev)); + printk(KERN_INFO "EISA: Probing bus %d at %s\n", + root->bus_nr, dev_name(root->dev)); /* First try to get hold of slot 0. If there is no device * here, simply fail, unless root->force_probe is set. */ - if (!(edev = kzalloc (sizeof (*edev), GFP_KERNEL))) { - printk (KERN_ERR "EISA: Couldn't allocate mainboard slot\n"); + edev = kzalloc(sizeof(*edev), GFP_KERNEL); + if (!edev) { + printk(KERN_ERR "EISA: Couldn't allocate mainboard slot\n"); return -ENOMEM; } - if (eisa_request_resources (root, edev, 0)) { - printk (KERN_WARNING \ - "EISA: Cannot allocate resource for mainboard\n"); - kfree (edev); + if (eisa_request_resources(root, edev, 0)) { + printk(KERN_WARNING \ + "EISA: Cannot allocate resource for mainboard\n"); + kfree(edev); if (!root->force_probe) return -EBUSY; goto force_probe; } - if (eisa_init_device (root, edev, 0)) { - eisa_release_resources (edev); - kfree (edev); + if (eisa_init_device(root, edev, 0)) { + eisa_release_resources(edev); + kfree(edev); if (!root->force_probe) return -ENODEV; goto force_probe; } - printk (KERN_INFO "EISA: Mainboard %s detected.\n", edev->id.sig); + printk(KERN_INFO "EISA: Mainboard %s detected.\n", edev->id.sig); - if (eisa_register_device (edev)) { - printk (KERN_ERR "EISA: Failed to register %s\n", - edev->id.sig); - eisa_release_resources (edev); - kfree (edev); + if (eisa_register_device(edev)) { + printk(KERN_ERR "EISA: Failed to register %s\n", + edev->id.sig); + eisa_release_resources(edev); + kfree(edev); } force_probe: for (c = 0, i = 1; i <= root->slots; i++) { - if (!(edev = kzalloc (sizeof (*edev), GFP_KERNEL))) { - printk (KERN_ERR "EISA: Out of memory for slot %d\n", - i); + edev = kzalloc(sizeof(*edev), GFP_KERNEL); + if (!edev) { + printk(KERN_ERR "EISA: Out of memory for slot %d\n", i); continue; } - if (eisa_request_resources (root, edev, i)) { - printk (KERN_WARNING \ - "Cannot allocate resource for EISA slot %d\n", - i); - kfree (edev); + if (eisa_request_resources(root, edev, i)) { + printk(KERN_WARNING \ + "Cannot allocate resource for EISA slot %d\n", + i); + kfree(edev); continue; } - if (eisa_init_device (root, edev, i)) { - eisa_release_resources (edev); - kfree (edev); + if (eisa_init_device(root, edev, i)) { + eisa_release_resources(edev); + kfree(edev); continue; } - printk (KERN_INFO "EISA: slot %d : %s detected", - i, edev->id.sig); + printk(KERN_INFO "EISA: slot %d : %s detected", + i, edev->id.sig); switch (edev->state) { case EISA_CONFIG_ENABLED | EISA_CONFIG_FORCED: - printk (" (forced enabled)"); + printk(" (forced enabled)"); break; case EISA_CONFIG_FORCED: - printk (" (forced disabled)"); + printk(" (forced disabled)"); break; case 0: - printk (" (disabled)"); + printk(" (disabled)"); break; } @@ -381,15 +397,15 @@ static int __init eisa_probe (struct eisa_root_device *root) c++; - if (eisa_register_device (edev)) { - printk (KERN_ERR "EISA: Failed to register %s\n", - edev->id.sig); - eisa_release_resources (edev); - kfree (edev); + if (eisa_register_device(edev)) { + printk(KERN_ERR "EISA: Failed to register %s\n", + edev->id.sig); + eisa_release_resources(edev); + kfree(edev); } } - printk (KERN_INFO "EISA: Detected %d card%s.\n", c, c == 1 ? "" : "s"); + printk(KERN_INFO "EISA: Detected %d card%s.\n", c, c == 1 ? "" : "s"); return 0; } @@ -403,7 +419,7 @@ static struct resource eisa_root_res = { static int eisa_bus_count; -int __init eisa_root_register (struct eisa_root_device *root) +int __init eisa_root_register(struct eisa_root_device *root) { int err; @@ -417,35 +433,35 @@ int __init eisa_root_register (struct eisa_root_device *root) root->eisa_root_res.end = root->res->end; root->eisa_root_res.flags = IORESOURCE_BUSY; - if ((err = request_resource (&eisa_root_res, &root->eisa_root_res))) + err = request_resource(&eisa_root_res, &root->eisa_root_res); + if (err) return err; root->bus_nr = eisa_bus_count++; - if ((err = eisa_probe (root))) - release_resource (&root->eisa_root_res); + err = eisa_probe(root); + if (err) + release_resource(&root->eisa_root_res); return err; } -static int __init eisa_init (void) +static int __init eisa_init(void) { int r; - if ((r = bus_register (&eisa_bus_type))) + r = bus_register(&eisa_bus_type); + if (r) return r; - printk (KERN_INFO "EISA bus registered\n"); + printk(KERN_INFO "EISA bus registered\n"); return 0; } module_param_array(enable_dev, int, &enable_dev_count, 0444); module_param_array(disable_dev, int, &disable_dev_count, 0444); -postcore_initcall (eisa_init); +postcore_initcall(eisa_init); int EISA_bus; /* for legacy drivers */ -EXPORT_SYMBOL (EISA_bus); -EXPORT_SYMBOL (eisa_bus_type); -EXPORT_SYMBOL (eisa_driver_register); -EXPORT_SYMBOL (eisa_driver_unregister); +EXPORT_SYMBOL(EISA_bus); diff --git a/drivers/firmware/memmap.c b/drivers/firmware/memmap.c index 56f9234781fa..20f645743ead 100644 --- a/drivers/firmware/memmap.c +++ b/drivers/firmware/memmap.c @@ -122,29 +122,53 @@ static int firmware_map_add_entry(u64 start, u64 end, return 0; } +/* + * Add memmap entry on sysfs + */ +static int add_sysfs_fw_map_entry(struct firmware_map_entry *entry) +{ + static int map_entries_nr; + static struct kset *mmap_kset; + + if (!mmap_kset) { + mmap_kset = kset_create_and_add("memmap", NULL, firmware_kobj); + if (!mmap_kset) + return -ENOMEM; + } + + entry->kobj.kset = mmap_kset; + if (kobject_add(&entry->kobj, NULL, "%d", map_entries_nr++)) + kobject_put(&entry->kobj); + + return 0; +} + /** - * firmware_map_add() - Adds a firmware mapping entry. + * firmware_map_add_hotplug() - Adds a firmware mapping entry when we do + * memory hotplug. * @start: Start of the memory range. * @end: End of the memory range (inclusive). * @type: Type of the memory range. * - * This function uses kmalloc() for memory - * allocation. Use firmware_map_add_early() if you want to use the bootmem - * allocator. - * - * That function must be called before late_initcall. + * Adds a firmware mapping entry. This function is for memory hotplug, it is + * similar to function firmware_map_add_early(). The only difference is that + * it will create the syfs entry dynamically. * * Returns 0 on success, or -ENOMEM if no memory could be allocated. **/ -int firmware_map_add(u64 start, u64 end, const char *type) +int __meminit firmware_map_add_hotplug(u64 start, u64 end, const char *type) { struct firmware_map_entry *entry; - entry = kmalloc(sizeof(struct firmware_map_entry), GFP_ATOMIC); + entry = kzalloc(sizeof(struct firmware_map_entry), GFP_ATOMIC); if (!entry) return -ENOMEM; - return firmware_map_add_entry(start, end, type, entry); + firmware_map_add_entry(start, end, type, entry); + /* create the memmap entry */ + add_sysfs_fw_map_entry(entry); + + return 0; } /** @@ -154,7 +178,7 @@ int firmware_map_add(u64 start, u64 end, const char *type) * @type: Type of the memory range. * * Adds a firmware mapping entry. This function uses the bootmem allocator - * for memory allocation. Use firmware_map_add() if you want to use kmalloc(). + * for memory allocation. * * That function must be called before late_initcall. * @@ -214,19 +238,10 @@ static ssize_t memmap_attr_show(struct kobject *kobj, */ static int __init memmap_init(void) { - int i = 0; struct firmware_map_entry *entry; - struct kset *memmap_kset; - - memmap_kset = kset_create_and_add("memmap", NULL, firmware_kobj); - if (WARN_ON(!memmap_kset)) - return -ENOMEM; - list_for_each_entry(entry, &map_entries, list) { - entry->kobj.kset = memmap_kset; - if (kobject_add(&entry->kobj, NULL, "%d", i++)) - kobject_put(&entry->kobj); - } + list_for_each_entry(entry, &map_entries, list) + add_sysfs_fw_map_entry(entry); return 0; } diff --git a/drivers/gpio/Kconfig b/drivers/gpio/Kconfig index 1f1d88ae68d6..a4cdbd51b1c6 100644 --- a/drivers/gpio/Kconfig +++ b/drivers/gpio/Kconfig @@ -65,8 +65,17 @@ config GPIO_SYSFS # put expanders in the right section, in alphabetical order +config GPIO_MAX730X + tristate + comment "Memory mapped GPIO expanders:" +config GPIO_IT8761E + tristate "IT8761E GPIO support" + depends on GPIOLIB + help + Say yes here to support GPIO functionality of IT8761E super I/O chip. + config GPIO_PL061 bool "PrimeCell PL061 GPIO support" depends on ARM_AMBA @@ -87,6 +96,13 @@ config GPIO_VR41XX comment "I2C GPIO expanders:" +config GPIO_MAX7300 + tristate "Maxim MAX7300 GPIO expander" + depends on I2C + select GPIO_MAX730X + help + GPIO driver for Maxim MAX7301 I2C-based GPIO expander. + config GPIO_MAX732X tristate "MAX7319, MAX7320-7327 I2C Port Expanders" depends on I2C @@ -124,6 +140,13 @@ config GPIO_PCA953X This driver can also be built as a module. If so, the module will be called pca953x. +config GPIO_PCA953X_IRQ + bool "Interrupt controller support for PCA953x" + depends on GPIO_PCA953X=y + help + Say yes here to enable the pca953x to be used as an interrupt + controller. It requires the driver to be built in the kernel. + config GPIO_PCF857X tristate "PCF857x, PCA{85,96}7x, and MAX732[89] I2C GPIO expanders" depends on I2C @@ -226,8 +249,9 @@ comment "SPI GPIO expanders:" config GPIO_MAX7301 tristate "Maxim MAX7301 GPIO expander" depends on SPI_MASTER + select GPIO_MAX730X help - gpio driver for Maxim MAX7301 SPI GPIO expander. + GPIO driver for Maxim MAX7301 SPI-based GPIO expander. config GPIO_MCP23S08 tristate "Microchip MCP23S08 I/O expander" diff --git a/drivers/gpio/Makefile b/drivers/gpio/Makefile index 48687238edb1..128abf8a98da 100644 --- a/drivers/gpio/Makefile +++ b/drivers/gpio/Makefile @@ -7,6 +7,8 @@ obj-$(CONFIG_GPIOLIB) += gpiolib.o obj-$(CONFIG_GPIO_ADP5520) += adp5520-gpio.o obj-$(CONFIG_GPIO_ADP5588) += adp5588-gpio.o obj-$(CONFIG_GPIO_LANGWELL) += langwell_gpio.o +obj-$(CONFIG_GPIO_MAX730X) += max730x.o +obj-$(CONFIG_GPIO_MAX7300) += max7300.o obj-$(CONFIG_GPIO_MAX7301) += max7301.o obj-$(CONFIG_GPIO_MAX732X) += max732x.o obj-$(CONFIG_GPIO_MC33880) += mc33880.o @@ -20,5 +22,6 @@ obj-$(CONFIG_GPIO_UCB1400) += ucb1400_gpio.o obj-$(CONFIG_GPIO_XILINX) += xilinx_gpio.o obj-$(CONFIG_GPIO_CS5535) += cs5535-gpio.o obj-$(CONFIG_GPIO_BT8XX) += bt8xxgpio.o +obj-$(CONFIG_GPIO_IT8761E) += it8761e_gpio.o obj-$(CONFIG_GPIO_VR41XX) += vr41xx_giu.o obj-$(CONFIG_GPIO_WM831X) += wm831x-gpio.o diff --git a/drivers/gpio/cs5535-gpio.c b/drivers/gpio/cs5535-gpio.c index 0fdbe94f24a3..0c3c498f2260 100644 --- a/drivers/gpio/cs5535-gpio.c +++ b/drivers/gpio/cs5535-gpio.c @@ -154,7 +154,7 @@ static int chip_gpio_request(struct gpio_chip *c, unsigned offset) static int chip_gpio_get(struct gpio_chip *chip, unsigned offset) { - return cs5535_gpio_isset(offset, GPIO_OUTPUT_VAL); + return cs5535_gpio_isset(offset, GPIO_READ_BACK); } static void chip_gpio_set(struct gpio_chip *chip, unsigned offset, int val) @@ -172,6 +172,7 @@ static int chip_direction_input(struct gpio_chip *c, unsigned offset) spin_lock_irqsave(&chip->lock, flags); __cs5535_gpio_set(chip, offset, GPIO_INPUT_ENABLE); + __cs5535_gpio_clear(chip, offset, GPIO_OUTPUT_ENABLE); spin_unlock_irqrestore(&chip->lock, flags); return 0; @@ -184,6 +185,7 @@ static int chip_direction_output(struct gpio_chip *c, unsigned offset, int val) spin_lock_irqsave(&chip->lock, flags); + __cs5535_gpio_set(chip, offset, GPIO_INPUT_ENABLE); __cs5535_gpio_set(chip, offset, GPIO_OUTPUT_ENABLE); if (val) __cs5535_gpio_set(chip, offset, GPIO_OUTPUT_VAL); diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c index 350842ad3632..9006fdb26fea 100644 --- a/drivers/gpio/gpiolib.c +++ b/drivers/gpio/gpiolib.c @@ -1237,6 +1237,64 @@ void gpio_free(unsigned gpio) } EXPORT_SYMBOL_GPL(gpio_free); +/** + * gpio_request_one - request a single GPIO with initial configuration + * @gpio: the GPIO number + * @flags: GPIO configuration as specified by GPIOF_* + * @label: a literal description string of this GPIO + */ +int gpio_request_one(unsigned gpio, unsigned long flags, const char *label) +{ + int err; + + err = gpio_request(gpio, label); + if (err) + return err; + + if (flags & GPIOF_DIR_IN) + err = gpio_direction_input(gpio); + else + err = gpio_direction_output(gpio, + (flags & GPIOF_INIT_HIGH) ? 1 : 0); + + return err; +} +EXPORT_SYMBOL_GPL(gpio_request_one); + +/** + * gpio_request_array - request multiple GPIOs in a single call + * @array: array of the 'struct gpio' + * @num: how many GPIOs in the array + */ +int gpio_request_array(struct gpio *array, size_t num) +{ + int i, err; + + for (i = 0; i < num; i++, array++) { + err = gpio_request_one(array->gpio, array->flags, array->label); + if (err) + goto err_free; + } + return 0; + +err_free: + while (i--) + gpio_free((--array)->gpio); + return err; +} +EXPORT_SYMBOL_GPL(gpio_request_array); + +/** + * gpio_free_array - release multiple GPIOs in a single call + * @array: array of the 'struct gpio' + * @num: how many GPIOs in the array + */ +void gpio_free_array(struct gpio *array, size_t num) +{ + while (num--) + gpio_free((array++)->gpio); +} +EXPORT_SYMBOL_GPL(gpio_free_array); /** * gpiochip_is_requested - return string iff signal was requested diff --git a/drivers/gpio/it8761e_gpio.c b/drivers/gpio/it8761e_gpio.c new file mode 100644 index 000000000000..753219cf993a --- /dev/null +++ b/drivers/gpio/it8761e_gpio.c @@ -0,0 +1,231 @@ +/* + * it8761_gpio.c - GPIO interface for IT8761E Super I/O chip + * + * Author: Denis Turischev <denis@compulab.co.il> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License 2 as published + * by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; see the file COPYING. If not, write to + * the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. + */ + +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/io.h> +#include <linux/errno.h> +#include <linux/ioport.h> + +#include <linux/gpio.h> + +#define SIO_CHIP_ID 0x8761 +#define CHIP_ID_HIGH_BYTE 0x20 +#define CHIP_ID_LOW_BYTE 0x21 + +static u8 ports[2] = { 0x2e, 0x4e }; +static u8 port; + +static DEFINE_SPINLOCK(sio_lock); + +#define GPIO_NAME "it8761-gpio" +#define GPIO_BA_HIGH_BYTE 0x60 +#define GPIO_BA_LOW_BYTE 0x61 +#define GPIO_IOSIZE 4 +#define GPIO1X_IO 0xf0 +#define GPIO2X_IO 0xf1 + +static u16 gpio_ba; + +static u8 read_reg(u8 addr, u8 port) +{ + outb(addr, port); + return inb(port + 1); +} + +static void write_reg(u8 data, u8 addr, u8 port) +{ + outb(addr, port); + outb(data, port + 1); +} + +static void enter_conf_mode(u8 port) +{ + outb(0x87, port); + outb(0x61, port); + outb(0x55, port); + outb((port == 0x2e) ? 0x55 : 0xaa, port); +} + +static void exit_conf_mode(u8 port) +{ + outb(0x2, port); + outb(0x2, port + 1); +} + +static void enter_gpio_mode(u8 port) +{ + write_reg(0x2, 0x7, port); +} + +static int it8761e_gpio_get(struct gpio_chip *gc, unsigned gpio_num) +{ + u16 reg; + u8 bit; + + bit = gpio_num % 7; + reg = (gpio_num >= 7) ? gpio_ba + 1 : gpio_ba; + + return !!(inb(reg) & (1 << bit)); +} + +static int it8761e_gpio_direction_in(struct gpio_chip *gc, unsigned gpio_num) +{ + u8 curr_dirs; + u8 io_reg, bit; + + bit = gpio_num % 7; + io_reg = (gpio_num >= 7) ? GPIO2X_IO : GPIO1X_IO; + + spin_lock(&sio_lock); + + enter_conf_mode(port); + enter_gpio_mode(port); + + curr_dirs = read_reg(io_reg, port); + + if (curr_dirs & (1 << bit)) + write_reg(curr_dirs & ~(1 << bit), io_reg, port); + + exit_conf_mode(port); + + spin_unlock(&sio_lock); + return 0; +} + +static void it8761e_gpio_set(struct gpio_chip *gc, + unsigned gpio_num, int val) +{ + u8 curr_vals, bit; + u16 reg; + + bit = gpio_num % 7; + reg = (gpio_num >= 7) ? gpio_ba + 1 : gpio_ba; + + spin_lock(&sio_lock); + + curr_vals = inb(reg); + if (val) + outb(curr_vals | (1 << bit) , reg); + else + outb(curr_vals & ~(1 << bit), reg); + + spin_unlock(&sio_lock); +} + +static int it8761e_gpio_direction_out(struct gpio_chip *gc, + unsigned gpio_num, int val) +{ + u8 curr_dirs, io_reg, bit; + + bit = gpio_num % 7; + io_reg = (gpio_num >= 7) ? GPIO2X_IO : GPIO1X_IO; + + it8761e_gpio_set(gc, gpio_num, val); + + spin_lock(&sio_lock); + + enter_conf_mode(port); + enter_gpio_mode(port); + + curr_dirs = read_reg(io_reg, port); + + if (!(curr_dirs & (1 << bit))) + write_reg(curr_dirs | (1 << bit), io_reg, port); + + exit_conf_mode(port); + + spin_unlock(&sio_lock); + return 0; +} + +static struct gpio_chip it8761e_gpio_chip = { + .label = GPIO_NAME, + .owner = THIS_MODULE, + .get = it8761e_gpio_get, + .direction_input = it8761e_gpio_direction_in, + .set = it8761e_gpio_set, + .direction_output = it8761e_gpio_direction_out, +}; + +static int __init it8761e_gpio_init(void) +{ + int i, id, err; + + /* chip and port detection */ + for (i = 0; i < ARRAY_SIZE(ports); i++) { + spin_lock(&sio_lock); + enter_conf_mode(ports[i]); + + id = (read_reg(CHIP_ID_HIGH_BYTE, ports[i]) << 8) + + read_reg(CHIP_ID_LOW_BYTE, ports[i]); + + exit_conf_mode(ports[i]); + spin_unlock(&sio_lock); + + if (id == SIO_CHIP_ID) { + port = ports[i]; + break; + } + } + + if (!port) + return -ENODEV; + + /* fetch GPIO base address */ + enter_conf_mode(port); + enter_gpio_mode(port); + gpio_ba = (read_reg(GPIO_BA_HIGH_BYTE, port) << 8) + + read_reg(GPIO_BA_LOW_BYTE, port); + exit_conf_mode(port); + + if (!request_region(gpio_ba, GPIO_IOSIZE, GPIO_NAME)) + return -EBUSY; + + it8761e_gpio_chip.base = -1; + it8761e_gpio_chip.ngpio = 14; + + err = gpiochip_add(&it8761e_gpio_chip); + if (err < 0) + goto gpiochip_add_err; + + return 0; + +gpiochip_add_err: + release_region(gpio_ba, GPIO_IOSIZE); + gpio_ba = 0; + return err; +} + +static void __exit it8761e_gpio_exit(void) +{ + if (gpio_ba) { + gpiochip_remove(&it8761e_gpio_chip); + + release_region(gpio_ba, GPIO_IOSIZE); + gpio_ba = 0; + } +} +module_init(it8761e_gpio_init); +module_exit(it8761e_gpio_exit); + +MODULE_AUTHOR("Denis Turischev <denis@compulab.co.il>"); +MODULE_DESCRIPTION("GPIO interface for IT8761E Super I/O chip"); +MODULE_LICENSE("GPL"); diff --git a/drivers/gpio/max7300.c b/drivers/gpio/max7300.c new file mode 100644 index 000000000000..9d74eef1157a --- /dev/null +++ b/drivers/gpio/max7300.c @@ -0,0 +1,94 @@ +/* + * drivers/gpio/max7300.c + * + * Copyright (C) 2009 Wolfram Sang, Pengutronix + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * Check max730x.c for further details. + */ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/platform_device.h> +#include <linux/mutex.h> +#include <linux/i2c.h> +#include <linux/spi/max7301.h> + +static int max7300_i2c_write(struct device *dev, unsigned int reg, + unsigned int val) +{ + struct i2c_client *client = to_i2c_client(dev); + + return i2c_smbus_write_byte_data(client, reg, val); +} + +static int max7300_i2c_read(struct device *dev, unsigned int reg) +{ + struct i2c_client *client = to_i2c_client(dev); + + return i2c_smbus_read_byte_data(client, reg); +} + +static int __devinit max7300_probe(struct i2c_client *client, + const struct i2c_device_id *id) +{ + struct max7301 *ts; + int ret; + + if (!i2c_check_functionality(client->adapter, + I2C_FUNC_SMBUS_BYTE_DATA)) + return -EIO; + + ts = kzalloc(sizeof(struct max7301), GFP_KERNEL); + if (!ts) + return -ENOMEM; + + ts->read = max7300_i2c_read; + ts->write = max7300_i2c_write; + ts->dev = &client->dev; + + ret = __max730x_probe(ts); + if (ret) + kfree(ts); + return ret; +} + +static int __devexit max7300_remove(struct i2c_client *client) +{ + return __max730x_remove(&client->dev); +} + +static const struct i2c_device_id max7300_id[] = { + { "max7300", 0 }, + { } +}; +MODULE_DEVICE_TABLE(i2c, max7300_id); + +static struct i2c_driver max7300_driver = { + .driver = { + .name = "max7300", + .owner = THIS_MODULE, + }, + .probe = max7300_probe, + .remove = __devexit_p(max7300_remove), + .id_table = max7300_id, +}; + +static int __init max7300_init(void) +{ + return i2c_add_driver(&max7300_driver); +} +subsys_initcall(max7300_init); + +static void __exit max7300_exit(void) +{ + i2c_del_driver(&max7300_driver); +} +module_exit(max7300_exit); + +MODULE_AUTHOR("Wolfram Sang"); +MODULE_LICENSE("GPL v2"); +MODULE_DESCRIPTION("MAX7300 GPIO-Expander"); diff --git a/drivers/gpio/max7301.c b/drivers/gpio/max7301.c index 480956f1ca50..965d9b1ea13e 100644 --- a/drivers/gpio/max7301.c +++ b/drivers/gpio/max7301.c @@ -1,98 +1,41 @@ -/** +/* * drivers/gpio/max7301.c * * Copyright (C) 2006 Juergen Beisert, Pengutronix * Copyright (C) 2008 Guennadi Liakhovetski, Pengutronix + * Copyright (C) 2009 Wolfram Sang, Pengutronix * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as * published by the Free Software Foundation. * - * The Maxim's MAX7301 device is an SPI driven GPIO expander. There are - * 28 GPIOs. 8 of them can trigger an interrupt. See datasheet for more - * details - * Note: - * - DIN must be stable at the rising edge of clock. - * - when writing: - * - always clock in 16 clocks at once - * - at DIN: D15 first, D0 last - * - D0..D7 = databyte, D8..D14 = commandbyte - * - D15 = low -> write command - * - when reading - * - always clock in 16 clocks at once - * - at DIN: D15 first, D0 last - * - D0..D7 = dummy, D8..D14 = register address - * - D15 = high -> read command - * - raise CS and assert it again - * - always clock in 16 clocks at once - * - at DOUT: D15 first, D0 last - * - D0..D7 contains the data from the first cycle - * - * The driver exports a standard gpiochip interface + * Check max730x.c for further details. */ +#include <linux/module.h> #include <linux/init.h> #include <linux/platform_device.h> #include <linux/mutex.h> #include <linux/spi/spi.h> #include <linux/spi/max7301.h> -#include <linux/gpio.h> - -#define DRIVER_NAME "max7301" - -/* - * Pin configurations, see MAX7301 datasheet page 6 - */ -#define PIN_CONFIG_MASK 0x03 -#define PIN_CONFIG_IN_PULLUP 0x03 -#define PIN_CONFIG_IN_WO_PULLUP 0x02 -#define PIN_CONFIG_OUT 0x01 - -#define PIN_NUMBER 28 - - -/* - * Some registers must be read back to modify. - * To save time we cache them here in memory - */ -struct max7301 { - struct mutex lock; - u8 port_config[8]; /* field 0 is unused */ - u32 out_level; /* cached output levels */ - struct gpio_chip chip; - struct spi_device *spi; -}; -/** - * max7301_write - Write a new register content - * @spi: The SPI device - * @reg: Register offset - * @val: Value to write - * - * A write to the MAX7301 means one message with one transfer - * - * Returns 0 if successful or a negative value on error - */ -static int max7301_write(struct spi_device *spi, unsigned int reg, unsigned int val) +/* A write to the MAX7301 means one message with one transfer */ +static int max7301_spi_write(struct device *dev, unsigned int reg, + unsigned int val) { + struct spi_device *spi = to_spi_device(dev); u16 word = ((reg & 0x7F) << 8) | (val & 0xFF); + return spi_write(spi, (const u8 *)&word, sizeof(word)); } -/** - * max7301_read - Read back register content - * @spi: The SPI device - * @reg: Register offset - * - * A read from the MAX7301 means two transfers; here, one message each - * - * Returns positive 8 bit value from device if successful or a - * negative value on error - */ -static int max7301_read(struct spi_device *spi, unsigned int reg) +/* A read from the MAX7301 means two transfers; here, one message each */ + +static int max7301_spi_read(struct device *dev, unsigned int reg) { int ret; u16 word; + struct spi_device *spi = to_spi_device(dev); word = 0x8000 | (reg << 8); ret = spi_write(spi, (const u8 *)&word, sizeof(word)); @@ -108,125 +51,13 @@ static int max7301_read(struct spi_device *spi, unsigned int reg) return word & 0xff; } -static int max7301_direction_input(struct gpio_chip *chip, unsigned offset) -{ - struct max7301 *ts = container_of(chip, struct max7301, chip); - u8 *config; - int ret; - - /* First 4 pins are unused in the controller */ - offset += 4; - - config = &ts->port_config[offset >> 2]; - - mutex_lock(&ts->lock); - - /* Standard GPIO API doesn't support pull-ups, has to be extended. - * Hard-coding no pollup for now. */ - *config = (*config & ~(3 << (offset & 3))) | (1 << (offset & 3)); - - ret = max7301_write(ts->spi, 0x08 + (offset >> 2), *config); - - mutex_unlock(&ts->lock); - - return ret; -} - -static int __max7301_set(struct max7301 *ts, unsigned offset, int value) -{ - if (value) { - ts->out_level |= 1 << offset; - return max7301_write(ts->spi, 0x20 + offset, 0x01); - } else { - ts->out_level &= ~(1 << offset); - return max7301_write(ts->spi, 0x20 + offset, 0x00); - } -} - -static int max7301_direction_output(struct gpio_chip *chip, unsigned offset, - int value) -{ - struct max7301 *ts = container_of(chip, struct max7301, chip); - u8 *config; - int ret; - - /* First 4 pins are unused in the controller */ - offset += 4; - - config = &ts->port_config[offset >> 2]; - - mutex_lock(&ts->lock); - - *config = (*config & ~(3 << (offset & 3))) | (1 << (offset & 3)); - - ret = __max7301_set(ts, offset, value); - - if (!ret) - ret = max7301_write(ts->spi, 0x08 + (offset >> 2), *config); - - mutex_unlock(&ts->lock); - - return ret; -} - -static int max7301_get(struct gpio_chip *chip, unsigned offset) -{ - struct max7301 *ts = container_of(chip, struct max7301, chip); - int config, level = -EINVAL; - - /* First 4 pins are unused in the controller */ - offset += 4; - - mutex_lock(&ts->lock); - - config = (ts->port_config[offset >> 2] >> ((offset & 3) * 2)) & 3; - - switch (config) { - case 1: - /* Output: return cached level */ - level = !!(ts->out_level & (1 << offset)); - break; - case 2: - case 3: - /* Input: read out */ - level = max7301_read(ts->spi, 0x20 + offset) & 0x01; - } - mutex_unlock(&ts->lock); - - return level; -} - -static void max7301_set(struct gpio_chip *chip, unsigned offset, int value) -{ - struct max7301 *ts = container_of(chip, struct max7301, chip); - - /* First 4 pins are unused in the controller */ - offset += 4; - - mutex_lock(&ts->lock); - - __max7301_set(ts, offset, value); - - mutex_unlock(&ts->lock); -} - static int __devinit max7301_probe(struct spi_device *spi) { struct max7301 *ts; - struct max7301_platform_data *pdata; - int i, ret; - - pdata = spi->dev.platform_data; - if (!pdata || !pdata->base) { - dev_dbg(&spi->dev, "incorrect or missing platform data\n"); - return -EINVAL; - } + int ret; - /* - * bits_per_word cannot be configured in platform data - */ + /* bits_per_word cannot be configured in platform data */ spi->bits_per_word = 16; - ret = spi_setup(spi); if (ret < 0) return ret; @@ -235,90 +66,35 @@ static int __devinit max7301_probe(struct spi_device *spi) if (!ts) return -ENOMEM; - mutex_init(&ts->lock); - - dev_set_drvdata(&spi->dev, ts); + ts->read = max7301_spi_read; + ts->write = max7301_spi_write; + ts->dev = &spi->dev; - /* Power up the chip and disable IRQ output */ - max7301_write(spi, 0x04, 0x01); - - ts->spi = spi; - - ts->chip.label = DRIVER_NAME, - - ts->chip.direction_input = max7301_direction_input; - ts->chip.get = max7301_get; - ts->chip.direction_output = max7301_direction_output; - ts->chip.set = max7301_set; - - ts->chip.base = pdata->base; - ts->chip.ngpio = PIN_NUMBER; - ts->chip.can_sleep = 1; - ts->chip.dev = &spi->dev; - ts->chip.owner = THIS_MODULE; - - /* - * tristate all pins in hardware and cache the - * register values for later use. - */ - for (i = 1; i < 8; i++) { - int j; - /* 0xAA means input with internal pullup disabled */ - max7301_write(spi, 0x08 + i, 0xAA); - ts->port_config[i] = 0xAA; - for (j = 0; j < 4; j++) { - int offset = (i - 1) * 4 + j; - ret = max7301_direction_input(&ts->chip, offset); - if (ret) - goto exit_destroy; - } - } - - ret = gpiochip_add(&ts->chip); + ret = __max730x_probe(ts); if (ret) - goto exit_destroy; - - return ret; - -exit_destroy: - dev_set_drvdata(&spi->dev, NULL); - mutex_destroy(&ts->lock); - kfree(ts); + kfree(ts); return ret; } static int __devexit max7301_remove(struct spi_device *spi) { - struct max7301 *ts; - int ret; - - ts = dev_get_drvdata(&spi->dev); - if (ts == NULL) - return -ENODEV; - - dev_set_drvdata(&spi->dev, NULL); - - /* Power down the chip and disable IRQ output */ - max7301_write(spi, 0x04, 0x00); - - ret = gpiochip_remove(&ts->chip); - if (!ret) { - mutex_destroy(&ts->lock); - kfree(ts); - } else - dev_err(&spi->dev, "Failed to remove the GPIO controller: %d\n", - ret); - - return ret; + return __max730x_remove(&spi->dev); } +static const struct spi_device_id max7301_id[] = { + { "max7301", 0 }, + { } +}; +MODULE_DEVICE_TABLE(spi, max7301_id); + static struct spi_driver max7301_driver = { .driver = { - .name = DRIVER_NAME, - .owner = THIS_MODULE, + .name = "max7301", + .owner = THIS_MODULE, }, - .probe = max7301_probe, - .remove = __devexit_p(max7301_remove), + .probe = max7301_probe, + .remove = __devexit_p(max7301_remove), + .id_table = max7301_id, }; static int __init max7301_init(void) @@ -336,7 +112,6 @@ static void __exit max7301_exit(void) } module_exit(max7301_exit); -MODULE_AUTHOR("Juergen Beisert"); +MODULE_AUTHOR("Juergen Beisert, Wolfram Sang"); MODULE_LICENSE("GPL v2"); -MODULE_DESCRIPTION("MAX7301 SPI based GPIO-Expander"); -MODULE_ALIAS("spi:" DRIVER_NAME); +MODULE_DESCRIPTION("MAX7301 GPIO-Expander"); diff --git a/drivers/gpio/max730x.c b/drivers/gpio/max730x.c new file mode 100644 index 000000000000..c9bced55f82b --- /dev/null +++ b/drivers/gpio/max730x.c @@ -0,0 +1,244 @@ +/** + * drivers/gpio/max7301.c + * + * Copyright (C) 2006 Juergen Beisert, Pengutronix + * Copyright (C) 2008 Guennadi Liakhovetski, Pengutronix + * Copyright (C) 2009 Wolfram Sang, Pengutronix + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * The Maxim MAX7300/1 device is an I2C/SPI driven GPIO expander. There are + * 28 GPIOs. 8 of them can trigger an interrupt. See datasheet for more + * details + * Note: + * - DIN must be stable at the rising edge of clock. + * - when writing: + * - always clock in 16 clocks at once + * - at DIN: D15 first, D0 last + * - D0..D7 = databyte, D8..D14 = commandbyte + * - D15 = low -> write command + * - when reading + * - always clock in 16 clocks at once + * - at DIN: D15 first, D0 last + * - D0..D7 = dummy, D8..D14 = register address + * - D15 = high -> read command + * - raise CS and assert it again + * - always clock in 16 clocks at once + * - at DOUT: D15 first, D0 last + * - D0..D7 contains the data from the first cycle + * + * The driver exports a standard gpiochip interface + */ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/platform_device.h> +#include <linux/mutex.h> +#include <linux/spi/max7301.h> +#include <linux/gpio.h> + +/* + * Pin configurations, see MAX7301 datasheet page 6 + */ +#define PIN_CONFIG_MASK 0x03 +#define PIN_CONFIG_IN_PULLUP 0x03 +#define PIN_CONFIG_IN_WO_PULLUP 0x02 +#define PIN_CONFIG_OUT 0x01 + +#define PIN_NUMBER 28 + +static int max7301_direction_input(struct gpio_chip *chip, unsigned offset) +{ + struct max7301 *ts = container_of(chip, struct max7301, chip); + u8 *config; + u8 offset_bits; + int ret; + + /* First 4 pins are unused in the controller */ + offset += 4; + offset_bits = (offset & 3) << 1; + + config = &ts->port_config[offset >> 2]; + + mutex_lock(&ts->lock); + + /* Standard GPIO API doesn't support pull-ups, has to be extended. + * Hard-coding no pollup for now. */ + *config = (*config & ~(PIN_CONFIG_MASK << offset_bits)) + | (PIN_CONFIG_IN_WO_PULLUP << offset_bits); + + ret = ts->write(ts->dev, 0x08 + (offset >> 2), *config); + + mutex_unlock(&ts->lock); + + return ret; +} + +static int __max7301_set(struct max7301 *ts, unsigned offset, int value) +{ + if (value) { + ts->out_level |= 1 << offset; + return ts->write(ts->dev, 0x20 + offset, 0x01); + } else { + ts->out_level &= ~(1 << offset); + return ts->write(ts->dev, 0x20 + offset, 0x00); + } +} + +static int max7301_direction_output(struct gpio_chip *chip, unsigned offset, + int value) +{ + struct max7301 *ts = container_of(chip, struct max7301, chip); + u8 *config; + u8 offset_bits; + int ret; + + /* First 4 pins are unused in the controller */ + offset += 4; + offset_bits = (offset & 3) << 1; + + config = &ts->port_config[offset >> 2]; + + mutex_lock(&ts->lock); + + *config = (*config & ~(PIN_CONFIG_MASK << offset_bits)) + | (PIN_CONFIG_OUT << offset_bits); + + ret = __max7301_set(ts, offset, value); + + if (!ret) + ret = ts->write(ts->dev, 0x08 + (offset >> 2), *config); + + mutex_unlock(&ts->lock); + + return ret; +} + +static int max7301_get(struct gpio_chip *chip, unsigned offset) +{ + struct max7301 *ts = container_of(chip, struct max7301, chip); + int config, level = -EINVAL; + + /* First 4 pins are unused in the controller */ + offset += 4; + + mutex_lock(&ts->lock); + + config = (ts->port_config[offset >> 2] >> ((offset & 3) << 1)) + & PIN_CONFIG_MASK; + + switch (config) { + case PIN_CONFIG_OUT: + /* Output: return cached level */ + level = !!(ts->out_level & (1 << offset)); + break; + case PIN_CONFIG_IN_WO_PULLUP: + case PIN_CONFIG_IN_PULLUP: + /* Input: read out */ + level = ts->read(ts->dev, 0x20 + offset) & 0x01; + } + mutex_unlock(&ts->lock); + + return level; +} + +static void max7301_set(struct gpio_chip *chip, unsigned offset, int value) +{ + struct max7301 *ts = container_of(chip, struct max7301, chip); + + /* First 4 pins are unused in the controller */ + offset += 4; + + mutex_lock(&ts->lock); + + __max7301_set(ts, offset, value); + + mutex_unlock(&ts->lock); +} + +int __devinit __max730x_probe(struct max7301 *ts) +{ + struct device *dev = ts->dev; + struct max7301_platform_data *pdata; + int i, ret; + + pdata = dev->platform_data; + if (!pdata || !pdata->base) { + dev_err(dev, "incorrect or missing platform data\n"); + return -EINVAL; + } + + mutex_init(&ts->lock); + dev_set_drvdata(dev, ts); + + /* Power up the chip and disable IRQ output */ + ts->write(dev, 0x04, 0x01); + + ts->chip.label = dev->driver->name; + + ts->chip.direction_input = max7301_direction_input; + ts->chip.get = max7301_get; + ts->chip.direction_output = max7301_direction_output; + ts->chip.set = max7301_set; + + ts->chip.base = pdata->base; + ts->chip.ngpio = PIN_NUMBER; + ts->chip.can_sleep = 1; + ts->chip.dev = dev; + ts->chip.owner = THIS_MODULE; + + /* + * tristate all pins in hardware and cache the + * register values for later use. + */ + for (i = 1; i < 8; i++) { + int j; + /* 0xAA means input with internal pullup disabled */ + ts->write(dev, 0x08 + i, 0xAA); + ts->port_config[i] = 0xAA; + for (j = 0; j < 4; j++) { + int offset = (i - 1) * 4 + j; + ret = max7301_direction_input(&ts->chip, offset); + if (ret) + goto exit_destroy; + } + } + + ret = gpiochip_add(&ts->chip); + if (ret) + goto exit_destroy; + + return ret; + +exit_destroy: + dev_set_drvdata(dev, NULL); + mutex_destroy(&ts->lock); + return ret; +} +EXPORT_SYMBOL_GPL(__max730x_probe); + +int __devexit __max730x_remove(struct device *dev) +{ + struct max7301 *ts = dev_get_drvdata(dev); + int ret; + + if (ts == NULL) + return -ENODEV; + + dev_set_drvdata(dev, NULL); + + /* Power down the chip and disable IRQ output */ + ts->write(dev, 0x04, 0x00); + + ret = gpiochip_remove(&ts->chip); + if (!ret) { + mutex_destroy(&ts->lock); + kfree(ts); + } else + dev_err(dev, "Failed to remove GPIO controller: %d\n", ret); + + return ret; +} +EXPORT_SYMBOL_GPL(__max730x_remove); diff --git a/drivers/gpio/pca953x.c b/drivers/gpio/pca953x.c index 6a2fb3fbb3d9..ab5daab14bc2 100644 --- a/drivers/gpio/pca953x.c +++ b/drivers/gpio/pca953x.c @@ -14,6 +14,8 @@ #include <linux/module.h> #include <linux/init.h> #include <linux/gpio.h> +#include <linux/interrupt.h> +#include <linux/irq.h> #include <linux/i2c.h> #include <linux/i2c/pca953x.h> #ifdef CONFIG_OF_GPIO @@ -26,23 +28,28 @@ #define PCA953X_INVERT 2 #define PCA953X_DIRECTION 3 +#define PCA953X_GPIOS 0x00FF +#define PCA953X_INT 0x0100 + static const struct i2c_device_id pca953x_id[] = { - { "pca9534", 8, }, - { "pca9535", 16, }, + { "pca9534", 8 | PCA953X_INT, }, + { "pca9535", 16 | PCA953X_INT, }, { "pca9536", 4, }, - { "pca9537", 4, }, - { "pca9538", 8, }, - { "pca9539", 16, }, - { "pca9554", 8, }, - { "pca9555", 16, }, + { "pca9537", 4 | PCA953X_INT, }, + { "pca9538", 8 | PCA953X_INT, }, + { "pca9539", 16 | PCA953X_INT, }, + { "pca9554", 8 | PCA953X_INT, }, + { "pca9555", 16 | PCA953X_INT, }, { "pca9556", 8, }, { "pca9557", 8, }, { "max7310", 8, }, - { "max7315", 8, }, - { "pca6107", 8, }, - { "tca6408", 8, }, - { "tca6416", 16, }, + { "max7312", 16 | PCA953X_INT, }, + { "max7313", 16 | PCA953X_INT, }, + { "max7315", 8 | PCA953X_INT, }, + { "pca6107", 8 | PCA953X_INT, }, + { "tca6408", 8 | PCA953X_INT, }, + { "tca6416", 16 | PCA953X_INT, }, /* NYET: { "tca6424", 24, }, */ { } }; @@ -53,6 +60,15 @@ struct pca953x_chip { uint16_t reg_output; uint16_t reg_direction; +#ifdef CONFIG_GPIO_PCA953X_IRQ + struct mutex irq_lock; + uint16_t irq_mask; + uint16_t irq_stat; + uint16_t irq_trig_raise; + uint16_t irq_trig_fall; + int irq_base; +#endif + struct i2c_client *client; struct pca953x_platform_data *dyn_pdata; struct gpio_chip gpio_chip; @@ -202,6 +218,210 @@ static void pca953x_setup_gpio(struct pca953x_chip *chip, int gpios) gc->names = chip->names; } +#ifdef CONFIG_GPIO_PCA953X_IRQ +static int pca953x_gpio_to_irq(struct gpio_chip *gc, unsigned off) +{ + struct pca953x_chip *chip; + + chip = container_of(gc, struct pca953x_chip, gpio_chip); + return chip->irq_base + off; +} + +static void pca953x_irq_mask(unsigned int irq) +{ + struct pca953x_chip *chip = get_irq_chip_data(irq); + + chip->irq_mask &= ~(1 << (irq - chip->irq_base)); +} + +static void pca953x_irq_unmask(unsigned int irq) +{ + struct pca953x_chip *chip = get_irq_chip_data(irq); + + chip->irq_mask |= 1 << (irq - chip->irq_base); +} + +static void pca953x_irq_bus_lock(unsigned int irq) +{ + struct pca953x_chip *chip = get_irq_chip_data(irq); + + mutex_lock(&chip->irq_lock); +} + +static void pca953x_irq_bus_sync_unlock(unsigned int irq) +{ + struct pca953x_chip *chip = get_irq_chip_data(irq); + + mutex_unlock(&chip->irq_lock); +} + +static int pca953x_irq_set_type(unsigned int irq, unsigned int type) +{ + struct pca953x_chip *chip = get_irq_chip_data(irq); + uint16_t level = irq - chip->irq_base; + uint16_t mask = 1 << level; + + if (!(type & IRQ_TYPE_EDGE_BOTH)) { + dev_err(&chip->client->dev, "irq %d: unsupported type %d\n", + irq, type); + return -EINVAL; + } + + if (type & IRQ_TYPE_EDGE_FALLING) + chip->irq_trig_fall |= mask; + else + chip->irq_trig_fall &= ~mask; + + if (type & IRQ_TYPE_EDGE_RISING) + chip->irq_trig_raise |= mask; + else + chip->irq_trig_raise &= ~mask; + + return pca953x_gpio_direction_input(&chip->gpio_chip, level); +} + +static struct irq_chip pca953x_irq_chip = { + .name = "pca953x", + .mask = pca953x_irq_mask, + .unmask = pca953x_irq_unmask, + .bus_lock = pca953x_irq_bus_lock, + .bus_sync_unlock = pca953x_irq_bus_sync_unlock, + .set_type = pca953x_irq_set_type, +}; + +static uint16_t pca953x_irq_pending(struct pca953x_chip *chip) +{ + uint16_t cur_stat; + uint16_t old_stat; + uint16_t pending; + uint16_t trigger; + int ret; + + ret = pca953x_read_reg(chip, PCA953X_INPUT, &cur_stat); + if (ret) + return 0; + + /* Remove output pins from the equation */ + cur_stat &= chip->reg_direction; + + old_stat = chip->irq_stat; + trigger = (cur_stat ^ old_stat) & chip->irq_mask; + + if (!trigger) + return 0; + + chip->irq_stat = cur_stat; + + pending = (old_stat & chip->irq_trig_fall) | + (cur_stat & chip->irq_trig_raise); + pending &= trigger; + + return pending; +} + +static irqreturn_t pca953x_irq_handler(int irq, void *devid) +{ + struct pca953x_chip *chip = devid; + uint16_t pending; + uint16_t level; + + pending = pca953x_irq_pending(chip); + + if (!pending) + return IRQ_HANDLED; + + do { + level = __ffs(pending); + handle_nested_irq(level + chip->irq_base); + + pending &= ~(1 << level); + } while (pending); + + return IRQ_HANDLED; +} + +static int pca953x_irq_setup(struct pca953x_chip *chip, + const struct i2c_device_id *id) +{ + struct i2c_client *client = chip->client; + struct pca953x_platform_data *pdata = client->dev.platform_data; + int ret; + + if (pdata->irq_base && (id->driver_data & PCA953X_INT)) { + int lvl; + + ret = pca953x_read_reg(chip, PCA953X_INPUT, + &chip->irq_stat); + if (ret) + goto out_failed; + + /* + * There is no way to know which GPIO line generated the + * interrupt. We have to rely on the previous read for + * this purpose. + */ + chip->irq_stat &= chip->reg_direction; + chip->irq_base = pdata->irq_base; + mutex_init(&chip->irq_lock); + + for (lvl = 0; lvl < chip->gpio_chip.ngpio; lvl++) { + int irq = lvl + chip->irq_base; + + set_irq_chip_data(irq, chip); + set_irq_chip_and_handler(irq, &pca953x_irq_chip, + handle_edge_irq); + set_irq_nested_thread(irq, 1); +#ifdef CONFIG_ARM + set_irq_flags(irq, IRQF_VALID); +#else + set_irq_noprobe(irq); +#endif + } + + ret = request_threaded_irq(client->irq, + NULL, + pca953x_irq_handler, + IRQF_TRIGGER_FALLING | IRQF_ONESHOT, + dev_name(&client->dev), chip); + if (ret) { + dev_err(&client->dev, "failed to request irq %d\n", + client->irq); + goto out_failed; + } + + chip->gpio_chip.to_irq = pca953x_gpio_to_irq; + } + + return 0; + +out_failed: + chip->irq_base = 0; + return ret; +} + +static void pca953x_irq_teardown(struct pca953x_chip *chip) +{ + if (chip->irq_base) + free_irq(chip->client->irq, chip); +} +#else /* CONFIG_GPIO_PCA953X_IRQ */ +static int pca953x_irq_setup(struct pca953x_chip *chip, + const struct i2c_device_id *id) +{ + struct i2c_client *client = chip->client; + struct pca953x_platform_data *pdata = client->dev.platform_data; + + if (pdata->irq_base && (id->driver_data & PCA953X_INT)) + dev_warn(&client->dev, "interrupt support not compiled in\n"); + + return 0; +} + +static void pca953x_irq_teardown(struct pca953x_chip *chip) +{ +} +#endif + /* * Handlers for alternative sources of platform_data */ @@ -286,7 +506,7 @@ static int __devinit pca953x_probe(struct i2c_client *client, /* initialize cached registers from their original values. * we can't share this chip with another i2c master. */ - pca953x_setup_gpio(chip, id->driver_data); + pca953x_setup_gpio(chip, id->driver_data & PCA953X_GPIOS); ret = pca953x_read_reg(chip, PCA953X_OUTPUT, &chip->reg_output); if (ret) @@ -301,6 +521,9 @@ static int __devinit pca953x_probe(struct i2c_client *client, if (ret) goto out_failed; + ret = pca953x_irq_setup(chip, id); + if (ret) + goto out_failed; ret = gpiochip_add(&chip->gpio_chip); if (ret) @@ -317,6 +540,7 @@ static int __devinit pca953x_probe(struct i2c_client *client, return 0; out_failed: + pca953x_irq_teardown(chip); kfree(chip->dyn_pdata); kfree(chip); return ret; @@ -345,6 +569,7 @@ static int pca953x_remove(struct i2c_client *client) return ret; } + pca953x_irq_teardown(chip); kfree(chip->dyn_pdata); kfree(chip); return 0; diff --git a/drivers/gpio/pl061.c b/drivers/gpio/pl061.c index 4ee4c8367a3f..3ad1eeb49609 100644 --- a/drivers/gpio/pl061.c +++ b/drivers/gpio/pl061.c @@ -219,7 +219,7 @@ static void pl061_irq_handler(unsigned irq, struct irq_desc *desc) if (pending == 0) continue; - for_each_bit(offset, &pending, PL061_GPIO_NR) + for_each_set_bit(offset, &pending, PL061_GPIO_NR) generic_handle_irq(pl061_to_irq(&chip->gc, offset)); } desc->chip->unmask(irq); diff --git a/drivers/gpio/timbgpio.c b/drivers/gpio/timbgpio.c index a4d344ba8e5c..d4295fa5369e 100644 --- a/drivers/gpio/timbgpio.c +++ b/drivers/gpio/timbgpio.c @@ -23,6 +23,7 @@ #include <linux/module.h> #include <linux/gpio.h> #include <linux/platform_device.h> +#include <linux/irq.h> #include <linux/io.h> #include <linux/timb_gpio.h> #include <linux/interrupt.h> @@ -37,6 +38,8 @@ #define TGPIO_ICR 0x14 #define TGPIO_FLR 0x18 #define TGPIO_LVR 0x1c +#define TGPIO_VER 0x20 +#define TGPIO_BFLR 0x24 struct timbgpio { void __iomem *membase; @@ -125,17 +128,23 @@ static int timbgpio_irq_type(unsigned irq, unsigned trigger) struct timbgpio *tgpio = get_irq_chip_data(irq); int offset = irq - tgpio->irq_base; unsigned long flags; - u32 lvr, flr; + u32 lvr, flr, bflr = 0; + u32 ver; if (offset < 0 || offset > tgpio->gpio.ngpio) return -EINVAL; + ver = ioread32(tgpio->membase + TGPIO_VER); + spin_lock_irqsave(&tgpio->lock, flags); lvr = ioread32(tgpio->membase + TGPIO_LVR); flr = ioread32(tgpio->membase + TGPIO_FLR); + if (ver > 2) + bflr = ioread32(tgpio->membase + TGPIO_BFLR); if (trigger & (IRQ_TYPE_LEVEL_HIGH | IRQ_TYPE_LEVEL_LOW)) { + bflr &= ~(1 << offset); flr &= ~(1 << offset); if (trigger & IRQ_TYPE_LEVEL_HIGH) lvr |= 1 << offset; @@ -143,21 +152,27 @@ static int timbgpio_irq_type(unsigned irq, unsigned trigger) lvr &= ~(1 << offset); } - if ((trigger & IRQ_TYPE_EDGE_BOTH) == IRQ_TYPE_EDGE_BOTH) - return -EINVAL; - else { + if ((trigger & IRQ_TYPE_EDGE_BOTH) == IRQ_TYPE_EDGE_BOTH) { + if (ver < 3) + return -EINVAL; + else { + flr |= 1 << offset; + bflr |= 1 << offset; + } + } else { + bflr &= ~(1 << offset); flr |= 1 << offset; - /* opposite compared to the datasheet, but it mirrors the - * reality - */ if (trigger & IRQ_TYPE_EDGE_FALLING) - lvr |= 1 << offset; - else lvr &= ~(1 << offset); + else + lvr |= 1 << offset; } iowrite32(lvr, tgpio->membase + TGPIO_LVR); iowrite32(flr, tgpio->membase + TGPIO_FLR); + if (ver > 2) + iowrite32(bflr, tgpio->membase + TGPIO_BFLR); + iowrite32(1 << offset, tgpio->membase + TGPIO_ICR); spin_unlock_irqrestore(&tgpio->lock, flags); @@ -174,7 +189,7 @@ static void timbgpio_irq(unsigned int irq, struct irq_desc *desc) ipr = ioread32(tgpio->membase + TGPIO_IPR); iowrite32(ipr, tgpio->membase + TGPIO_ICR); - for_each_bit(offset, &ipr, tgpio->gpio.ngpio) + for_each_set_bit(offset, &ipr, tgpio->gpio.ngpio) generic_handle_irq(timbgpio_to_irq(&tgpio->gpio, offset)); } diff --git a/drivers/hwmon/Kconfig b/drivers/hwmon/Kconfig index 68cf87749a42..e4595e6147b4 100644 --- a/drivers/hwmon/Kconfig +++ b/drivers/hwmon/Kconfig @@ -170,6 +170,16 @@ config SENSORS_ADM9240 This driver can also be built as a module. If so, the module will be called adm9240. +config SENSORS_ADT7411 + tristate "Analog Devices ADT7411" + depends on I2C && EXPERIMENTAL + help + If you say yes here you get support for the Analog Devices + ADT7411 voltage and temperature monitoring chip. + + This driver can also be built as a module. If so, the module + will be called adt7411. + config SENSORS_ADT7462 tristate "Analog Devices ADT7462" depends on I2C && EXPERIMENTAL @@ -190,20 +200,6 @@ config SENSORS_ADT7470 This driver can also be built as a module. If so, the module will be called adt7470. -config SENSORS_ADT7473 - tristate "Analog Devices ADT7473 (DEPRECATED)" - depends on I2C && EXPERIMENTAL - select SENSORS_ADT7475 - help - If you say yes here you get support for the Analog Devices - ADT7473 temperature monitoring chips. - - This driver is deprecated, you should use the adt7475 driver - instead. - - This driver can also be built as a module. If so, the module - will be called adt7473. - config SENSORS_ADT7475 tristate "Analog Devices ADT7473, ADT7475, ADT7476 and ADT7490" depends on I2C && EXPERIMENTAL @@ -216,6 +212,19 @@ config SENSORS_ADT7475 This driver can also be build as a module. If so, the module will be called adt7475. +config SENSORS_ASC7621 + tristate "Andigilog aSC7621" + depends on HWMON && I2C + help + If you say yes here you get support for the aSC7621 + family of SMBus sensors chip found on most Intel X48, X38, 975, + 965 and 945 desktop boards. Currently supported chips: + aSC7621 + aSC7621a + + This driver can also be built as a module. If so, the module + will be called asc7621. + config SENSORS_K8TEMP tristate "AMD Athlon64/FX or Opteron temperature sensor" depends on X86 && PCI && EXPERIMENTAL @@ -563,9 +572,10 @@ config SENSORS_LM90 depends on I2C help If you say yes here you get support for National Semiconductor LM90, - LM86, LM89 and LM99, Analog Devices ADM1032 and ADT7461, and Maxim + LM86, LM89 and LM99, Analog Devices ADM1032 and ADT7461, Maxim MAX6646, MAX6647, MAX6648, MAX6649, MAX6657, MAX6658, MAX6659, - MAX6680, MAX6681 and MAX6692 sensor chips. + MAX6680, MAX6681 and MAX6692, and Winbond/Nuvoton W83L771AWG/ASG + sensor chips. This driver can also be built as a module. If so, the module will be called lm90. @@ -909,7 +919,8 @@ config SENSORS_W83793 select HWMON_VID help If you say yes here you get support for the Winbond W83793 - hardware monitoring chip. + hardware monitoring chip, including support for the integrated + watchdog. This driver can also be built as a module. If so, the module will be called w83793. diff --git a/drivers/hwmon/Makefile b/drivers/hwmon/Makefile index 4bc215c0953f..4aa1a3d112ad 100644 --- a/drivers/hwmon/Makefile +++ b/drivers/hwmon/Makefile @@ -29,12 +29,13 @@ obj-$(CONFIG_SENSORS_ADM1029) += adm1029.o obj-$(CONFIG_SENSORS_ADM1031) += adm1031.o obj-$(CONFIG_SENSORS_ADM9240) += adm9240.o obj-$(CONFIG_SENSORS_ADS7828) += ads7828.o +obj-$(CONFIG_SENSORS_ADT7411) += adt7411.o obj-$(CONFIG_SENSORS_ADT7462) += adt7462.o obj-$(CONFIG_SENSORS_ADT7470) += adt7470.o -obj-$(CONFIG_SENSORS_ADT7473) += adt7473.o obj-$(CONFIG_SENSORS_ADT7475) += adt7475.o obj-$(CONFIG_SENSORS_APPLESMC) += applesmc.o obj-$(CONFIG_SENSORS_AMS) += ams/ +obj-$(CONFIG_SENSORS_ASC7621) += asc7621.o obj-$(CONFIG_SENSORS_ATXP1) += atxp1.o obj-$(CONFIG_SENSORS_CORETEMP) += coretemp.o obj-$(CONFIG_SENSORS_DME1737) += dme1737.o diff --git a/drivers/hwmon/adcxx.c b/drivers/hwmon/adcxx.c index 5e9e095f1136..74d9c5195e44 100644 --- a/drivers/hwmon/adcxx.c +++ b/drivers/hwmon/adcxx.c @@ -62,18 +62,23 @@ static ssize_t adcxx_read(struct device *dev, struct spi_device *spi = to_spi_device(dev); struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); struct adcxx *adc = dev_get_drvdata(&spi->dev); - u8 tx_buf[2] = { attr->index << 3 }; /* other bits are don't care */ + u8 tx_buf[2]; u8 rx_buf[2]; int status; - int value; + u32 value; if (mutex_lock_interruptible(&adc->lock)) return -ERESTARTSYS; - status = spi_write_then_read(spi, tx_buf, sizeof(tx_buf), - rx_buf, sizeof(rx_buf)); + if (adc->channels == 1) { + status = spi_read(spi, rx_buf, sizeof(rx_buf)); + } else { + tx_buf[0] = attr->index << 3; /* other bits are don't care */ + status = spi_write_then_read(spi, tx_buf, sizeof(tx_buf), + rx_buf, sizeof(rx_buf)); + } if (status < 0) { - dev_warn(dev, "spi_write_then_read failed with status %d\n", + dev_warn(dev, "SPI synch. transfer failed with status %d\n", status); goto out; } diff --git a/drivers/hwmon/adt7411.c b/drivers/hwmon/adt7411.c new file mode 100644 index 000000000000..3471884e42d2 --- /dev/null +++ b/drivers/hwmon/adt7411.c @@ -0,0 +1,366 @@ +/* + * Driver for the ADT7411 (I2C/SPI 8 channel 10 bit ADC & temperature-sensor) + * + * Copyright (C) 2008, 2010 Pengutronix + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * TODO: SPI, support for external temperature sensor + * use power-down mode for suspend?, interrupt handling? + */ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/err.h> +#include <linux/delay.h> +#include <linux/mutex.h> +#include <linux/jiffies.h> +#include <linux/i2c.h> +#include <linux/hwmon.h> +#include <linux/hwmon-sysfs.h> + +#define ADT7411_REG_INT_TEMP_VDD_LSB 0x03 +#define ADT7411_REG_EXT_TEMP_AIN14_LSB 0x04 +#define ADT7411_REG_VDD_MSB 0x06 +#define ADT7411_REG_INT_TEMP_MSB 0x07 +#define ADT7411_REG_EXT_TEMP_AIN1_MSB 0x08 + +#define ADT7411_REG_CFG1 0x18 +#define ADT7411_CFG1_START_MONITOR (1 << 0) + +#define ADT7411_REG_CFG2 0x19 +#define ADT7411_CFG2_DISABLE_AVG (1 << 5) + +#define ADT7411_REG_CFG3 0x1a +#define ADT7411_CFG3_ADC_CLK_225 (1 << 0) +#define ADT7411_CFG3_REF_VDD (1 << 4) + +#define ADT7411_REG_DEVICE_ID 0x4d +#define ADT7411_REG_MANUFACTURER_ID 0x4e + +#define ADT7411_DEVICE_ID 0x2 +#define ADT7411_MANUFACTURER_ID 0x41 + +static const unsigned short normal_i2c[] = { 0x48, 0x4a, 0x4b, I2C_CLIENT_END }; + +struct adt7411_data { + struct mutex device_lock; /* for "atomic" device accesses */ + struct mutex update_lock; + unsigned long next_update; + int vref_cached; + struct device *hwmon_dev; +}; + +/* + * When reading a register containing (up to 4) lsb, all associated + * msb-registers get locked by the hardware. After _one_ of those msb is read, + * _all_ are unlocked. In order to use this locking correctly, reading lsb/msb + * is protected here with a mutex, too. + */ +static int adt7411_read_10_bit(struct i2c_client *client, u8 lsb_reg, + u8 msb_reg, u8 lsb_shift) +{ + struct adt7411_data *data = i2c_get_clientdata(client); + int val, tmp; + + mutex_lock(&data->device_lock); + + val = i2c_smbus_read_byte_data(client, lsb_reg); + if (val < 0) + goto exit_unlock; + + tmp = (val >> lsb_shift) & 3; + val = i2c_smbus_read_byte_data(client, msb_reg); + + if (val >= 0) + val = (val << 2) | tmp; + + exit_unlock: + mutex_unlock(&data->device_lock); + + return val; +} + +static int adt7411_modify_bit(struct i2c_client *client, u8 reg, u8 bit, + bool flag) +{ + struct adt7411_data *data = i2c_get_clientdata(client); + int ret, val; + + mutex_lock(&data->device_lock); + + ret = i2c_smbus_read_byte_data(client, reg); + if (ret < 0) + goto exit_unlock; + + if (flag) + val = ret | bit; + else + val = ret & ~bit; + + ret = i2c_smbus_write_byte_data(client, reg, val); + + exit_unlock: + mutex_unlock(&data->device_lock); + return ret; +} + +static ssize_t adt7411_show_vdd(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct i2c_client *client = to_i2c_client(dev); + int ret = adt7411_read_10_bit(client, ADT7411_REG_INT_TEMP_VDD_LSB, + ADT7411_REG_VDD_MSB, 2); + + return ret < 0 ? ret : sprintf(buf, "%u\n", ret * 7000 / 1024); +} + +static ssize_t adt7411_show_temp(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct i2c_client *client = to_i2c_client(dev); + int val = adt7411_read_10_bit(client, ADT7411_REG_INT_TEMP_VDD_LSB, + ADT7411_REG_INT_TEMP_MSB, 0); + + if (val < 0) + return val; + + val = val & 0x200 ? val - 0x400 : val; /* 10 bit signed */ + + return sprintf(buf, "%d\n", val * 250); +} + +static ssize_t adt7411_show_input(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int nr = to_sensor_dev_attr(attr)->index; + struct i2c_client *client = to_i2c_client(dev); + struct adt7411_data *data = i2c_get_clientdata(client); + int val; + u8 lsb_reg, lsb_shift; + + mutex_lock(&data->update_lock); + if (time_after_eq(jiffies, data->next_update)) { + val = i2c_smbus_read_byte_data(client, ADT7411_REG_CFG3); + if (val < 0) + goto exit_unlock; + + if (val & ADT7411_CFG3_REF_VDD) { + val = adt7411_read_10_bit(client, + ADT7411_REG_INT_TEMP_VDD_LSB, + ADT7411_REG_VDD_MSB, 2); + if (val < 0) + goto exit_unlock; + + data->vref_cached = val * 7000 / 1024; + } else { + data->vref_cached = 2250; + } + + data->next_update = jiffies + HZ; + } + + lsb_reg = ADT7411_REG_EXT_TEMP_AIN14_LSB + (nr >> 2); + lsb_shift = 2 * (nr & 0x03); + val = adt7411_read_10_bit(client, lsb_reg, + ADT7411_REG_EXT_TEMP_AIN1_MSB + nr, lsb_shift); + if (val < 0) + goto exit_unlock; + + val = sprintf(buf, "%u\n", val * data->vref_cached / 1024); + exit_unlock: + mutex_unlock(&data->update_lock); + return val; +} + +static ssize_t adt7411_show_bit(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct sensor_device_attribute_2 *attr2 = to_sensor_dev_attr_2(attr); + struct i2c_client *client = to_i2c_client(dev); + int ret = i2c_smbus_read_byte_data(client, attr2->index); + + return ret < 0 ? ret : sprintf(buf, "%u\n", !!(ret & attr2->nr)); +} + +static ssize_t adt7411_set_bit(struct device *dev, + struct device_attribute *attr, const char *buf, + size_t count) +{ + struct sensor_device_attribute_2 *s_attr2 = to_sensor_dev_attr_2(attr); + struct i2c_client *client = to_i2c_client(dev); + struct adt7411_data *data = i2c_get_clientdata(client); + int ret; + unsigned long flag; + + ret = strict_strtoul(buf, 0, &flag); + if (ret || flag > 1) + return -EINVAL; + + ret = adt7411_modify_bit(client, s_attr2->index, s_attr2->nr, flag); + + /* force update */ + mutex_lock(&data->update_lock); + data->next_update = jiffies; + mutex_unlock(&data->update_lock); + + return ret < 0 ? ret : count; +} + +#define ADT7411_BIT_ATTR(__name, __reg, __bit) \ + SENSOR_DEVICE_ATTR_2(__name, S_IRUGO | S_IWUSR, adt7411_show_bit, \ + adt7411_set_bit, __bit, __reg) + +static DEVICE_ATTR(temp1_input, S_IRUGO, adt7411_show_temp, NULL); +static DEVICE_ATTR(in0_input, S_IRUGO, adt7411_show_vdd, NULL); +static SENSOR_DEVICE_ATTR(in1_input, S_IRUGO, adt7411_show_input, NULL, 0); +static SENSOR_DEVICE_ATTR(in2_input, S_IRUGO, adt7411_show_input, NULL, 1); +static SENSOR_DEVICE_ATTR(in3_input, S_IRUGO, adt7411_show_input, NULL, 2); +static SENSOR_DEVICE_ATTR(in4_input, S_IRUGO, adt7411_show_input, NULL, 3); +static SENSOR_DEVICE_ATTR(in5_input, S_IRUGO, adt7411_show_input, NULL, 4); +static SENSOR_DEVICE_ATTR(in6_input, S_IRUGO, adt7411_show_input, NULL, 5); +static SENSOR_DEVICE_ATTR(in7_input, S_IRUGO, adt7411_show_input, NULL, 6); +static SENSOR_DEVICE_ATTR(in8_input, S_IRUGO, adt7411_show_input, NULL, 7); +static ADT7411_BIT_ATTR(no_average, ADT7411_REG_CFG2, ADT7411_CFG2_DISABLE_AVG); +static ADT7411_BIT_ATTR(fast_sampling, ADT7411_REG_CFG3, ADT7411_CFG3_ADC_CLK_225); +static ADT7411_BIT_ATTR(adc_ref_vdd, ADT7411_REG_CFG3, ADT7411_CFG3_REF_VDD); + +static struct attribute *adt7411_attrs[] = { + &dev_attr_temp1_input.attr, + &dev_attr_in0_input.attr, + &sensor_dev_attr_in1_input.dev_attr.attr, + &sensor_dev_attr_in2_input.dev_attr.attr, + &sensor_dev_attr_in3_input.dev_attr.attr, + &sensor_dev_attr_in4_input.dev_attr.attr, + &sensor_dev_attr_in5_input.dev_attr.attr, + &sensor_dev_attr_in6_input.dev_attr.attr, + &sensor_dev_attr_in7_input.dev_attr.attr, + &sensor_dev_attr_in8_input.dev_attr.attr, + &sensor_dev_attr_no_average.dev_attr.attr, + &sensor_dev_attr_fast_sampling.dev_attr.attr, + &sensor_dev_attr_adc_ref_vdd.dev_attr.attr, + NULL +}; + +static const struct attribute_group adt7411_attr_grp = { + .attrs = adt7411_attrs, +}; + +static int adt7411_detect(struct i2c_client *client, + struct i2c_board_info *info) +{ + int val; + + if (!i2c_check_functionality(client->adapter, I2C_FUNC_SMBUS_BYTE_DATA)) + return -ENODEV; + + val = i2c_smbus_read_byte_data(client, ADT7411_REG_MANUFACTURER_ID); + if (val < 0 || val != ADT7411_MANUFACTURER_ID) { + dev_dbg(&client->dev, "Wrong manufacturer ID. Got %d, " + "expected %d\n", val, ADT7411_MANUFACTURER_ID); + return -ENODEV; + } + + val = i2c_smbus_read_byte_data(client, ADT7411_REG_DEVICE_ID); + if (val < 0 || val != ADT7411_DEVICE_ID) { + dev_dbg(&client->dev, "Wrong device ID. Got %d, " + "expected %d\n", val, ADT7411_DEVICE_ID); + return -ENODEV; + } + + strlcpy(info->type, "adt7411", I2C_NAME_SIZE); + + return 0; +} + +static int __devinit adt7411_probe(struct i2c_client *client, + const struct i2c_device_id *id) +{ + struct adt7411_data *data; + int ret; + + data = kzalloc(sizeof(*data), GFP_KERNEL); + if (!data) + return -ENOMEM; + + i2c_set_clientdata(client, data); + mutex_init(&data->device_lock); + mutex_init(&data->update_lock); + + ret = adt7411_modify_bit(client, ADT7411_REG_CFG1, + ADT7411_CFG1_START_MONITOR, 1); + if (ret < 0) + goto exit_free; + + /* force update on first occasion */ + data->next_update = jiffies; + + ret = sysfs_create_group(&client->dev.kobj, &adt7411_attr_grp); + if (ret) + goto exit_free; + + data->hwmon_dev = hwmon_device_register(&client->dev); + if (IS_ERR(data->hwmon_dev)) { + ret = PTR_ERR(data->hwmon_dev); + goto exit_remove; + } + + dev_info(&client->dev, "successfully registered\n"); + + return 0; + + exit_remove: + sysfs_remove_group(&client->dev.kobj, &adt7411_attr_grp); + exit_free: + i2c_set_clientdata(client, NULL); + kfree(data); + return ret; +} + +static int __devexit adt7411_remove(struct i2c_client *client) +{ + struct adt7411_data *data = i2c_get_clientdata(client); + + hwmon_device_unregister(data->hwmon_dev); + sysfs_remove_group(&client->dev.kobj, &adt7411_attr_grp); + i2c_set_clientdata(client, NULL); + kfree(data); + return 0; +} + +static const struct i2c_device_id adt7411_id[] = { + { "adt7411", 0 }, + { } +}; + +static struct i2c_driver adt7411_driver = { + .driver = { + .name = "adt7411", + }, + .probe = adt7411_probe, + .remove = __devexit_p(adt7411_remove), + .id_table = adt7411_id, + .detect = adt7411_detect, + .address_list = normal_i2c, + .class = I2C_CLASS_HWMON, +}; + +static int __init sensors_adt7411_init(void) +{ + return i2c_add_driver(&adt7411_driver); +} +module_init(sensors_adt7411_init) + +static void __exit sensors_adt7411_exit(void) +{ + i2c_del_driver(&adt7411_driver); +} +module_exit(sensors_adt7411_exit) + +MODULE_AUTHOR("Sascha Hauer <s.hauer@pengutronix.de> and " + "Wolfram Sang <w.sang@pengutronix.de>"); +MODULE_DESCRIPTION("ADT7411 driver"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/hwmon/adt7473.c b/drivers/hwmon/adt7473.c deleted file mode 100644 index 434576f61c84..000000000000 --- a/drivers/hwmon/adt7473.c +++ /dev/null @@ -1,1180 +0,0 @@ -/* - * A hwmon driver for the Analog Devices ADT7473 - * Copyright (C) 2007 IBM - * - * Author: Darrick J. Wong <djwong@us.ibm.com> - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - */ - -#include <linux/module.h> -#include <linux/jiffies.h> -#include <linux/i2c.h> -#include <linux/hwmon.h> -#include <linux/hwmon-sysfs.h> -#include <linux/err.h> -#include <linux/mutex.h> -#include <linux/delay.h> -#include <linux/log2.h> - -/* Addresses to scan */ -static const unsigned short normal_i2c[] = { 0x2C, 0x2D, 0x2E, I2C_CLIENT_END }; - -/* ADT7473 registers */ -#define ADT7473_REG_BASE_ADDR 0x20 - -#define ADT7473_REG_VOLT_BASE_ADDR 0x21 -#define ADT7473_REG_VOLT_MIN_BASE_ADDR 0x46 - -#define ADT7473_REG_TEMP_BASE_ADDR 0x25 -#define ADT7473_REG_TEMP_LIMITS_BASE_ADDR 0x4E -#define ADT7473_REG_TEMP_TMIN_BASE_ADDR 0x67 -#define ADT7473_REG_TEMP_TMAX_BASE_ADDR 0x6A - -#define ADT7473_REG_FAN_BASE_ADDR 0x28 -#define ADT7473_REG_FAN_MIN_BASE_ADDR 0x54 - -#define ADT7473_REG_PWM_BASE_ADDR 0x30 -#define ADT7473_REG_PWM_MIN_BASE_ADDR 0x64 -#define ADT7473_REG_PWM_MAX_BASE_ADDR 0x38 -#define ADT7473_REG_PWM_BHVR_BASE_ADDR 0x5C -#define ADT7473_PWM_BHVR_MASK 0xE0 -#define ADT7473_PWM_BHVR_SHIFT 5 - -#define ADT7473_REG_CFG1 0x40 -#define ADT7473_CFG1_START 0x01 -#define ADT7473_CFG1_READY 0x04 -#define ADT7473_REG_CFG2 0x73 -#define ADT7473_REG_CFG3 0x78 -#define ADT7473_REG_CFG4 0x7D -#define ADT7473_CFG4_MAX_DUTY_AT_OVT 0x08 -#define ADT7473_REG_CFG5 0x7C -#define ADT7473_CFG5_TEMP_TWOS 0x01 -#define ADT7473_CFG5_TEMP_OFFSET 0x02 - -#define ADT7473_REG_DEVICE 0x3D -#define ADT7473_VENDOR 0x41 -#define ADT7473_REG_VENDOR 0x3E -#define ADT7473_DEVICE 0x73 -#define ADT7473_REG_REVISION 0x3F -#define ADT7473_REV_68 0x68 -#define ADT7473_REV_69 0x69 - -#define ADT7473_REG_ALARM1 0x41 -#define ADT7473_VCCP_ALARM 0x02 -#define ADT7473_VCC_ALARM 0x04 -#define ADT7473_R1T_ALARM 0x10 -#define ADT7473_LT_ALARM 0x20 -#define ADT7473_R2T_ALARM 0x40 -#define ADT7473_OOL 0x80 -#define ADT7473_REG_ALARM2 0x42 -#define ADT7473_OVT_ALARM 0x02 -#define ADT7473_FAN1_ALARM 0x04 -#define ADT7473_FAN2_ALARM 0x08 -#define ADT7473_FAN3_ALARM 0x10 -#define ADT7473_FAN4_ALARM 0x20 -#define ADT7473_R1T_SHORT 0x40 -#define ADT7473_R2T_SHORT 0x80 - -#define ALARM2(x) ((x) << 8) - -#define ADT7473_VOLT_COUNT 2 -#define ADT7473_REG_VOLT(x) (ADT7473_REG_VOLT_BASE_ADDR + (x)) -#define ADT7473_REG_VOLT_MIN(x) (ADT7473_REG_VOLT_MIN_BASE_ADDR + ((x) * 2)) -#define ADT7473_REG_VOLT_MAX(x) (ADT7473_REG_VOLT_MIN_BASE_ADDR + \ - ((x) * 2) + 1) - -#define ADT7473_TEMP_COUNT 3 -#define ADT7473_REG_TEMP(x) (ADT7473_REG_TEMP_BASE_ADDR + (x)) -#define ADT7473_REG_TEMP_MIN(x) (ADT7473_REG_TEMP_LIMITS_BASE_ADDR + ((x) * 2)) -#define ADT7473_REG_TEMP_MAX(x) (ADT7473_REG_TEMP_LIMITS_BASE_ADDR + \ - ((x) * 2) + 1) -#define ADT7473_REG_TEMP_TMIN(x) (ADT7473_REG_TEMP_TMIN_BASE_ADDR + (x)) -#define ADT7473_REG_TEMP_TMAX(x) (ADT7473_REG_TEMP_TMAX_BASE_ADDR + (x)) - -#define ADT7473_FAN_COUNT 4 -#define ADT7473_REG_FAN(x) (ADT7473_REG_FAN_BASE_ADDR + ((x) * 2)) -#define ADT7473_REG_FAN_MIN(x) (ADT7473_REG_FAN_MIN_BASE_ADDR + ((x) * 2)) - -#define ADT7473_PWM_COUNT 3 -#define ADT7473_REG_PWM(x) (ADT7473_REG_PWM_BASE_ADDR + (x)) -#define ADT7473_REG_PWM_MAX(x) (ADT7473_REG_PWM_MAX_BASE_ADDR + (x)) -#define ADT7473_REG_PWM_MIN(x) (ADT7473_REG_PWM_MIN_BASE_ADDR + (x)) -#define ADT7473_REG_PWM_BHVR(x) (ADT7473_REG_PWM_BHVR_BASE_ADDR + (x)) - -/* How often do we reread sensors values? (In jiffies) */ -#define SENSOR_REFRESH_INTERVAL (2 * HZ) - -/* How often do we reread sensor limit values? (In jiffies) */ -#define LIMIT_REFRESH_INTERVAL (60 * HZ) - -/* datasheet says to divide this number by the fan reading to get fan rpm */ -#define FAN_PERIOD_TO_RPM(x) ((90000 * 60) / (x)) -#define FAN_RPM_TO_PERIOD FAN_PERIOD_TO_RPM -#define FAN_PERIOD_INVALID 65535 -#define FAN_DATA_VALID(x) ((x) && (x) != FAN_PERIOD_INVALID) - -struct adt7473_data { - struct device *hwmon_dev; - struct attribute_group attrs; - struct mutex lock; - char sensors_valid; - char limits_valid; - unsigned long sensors_last_updated; /* In jiffies */ - unsigned long limits_last_updated; /* In jiffies */ - - u8 volt[ADT7473_VOLT_COUNT]; - s8 volt_min[ADT7473_VOLT_COUNT]; - s8 volt_max[ADT7473_VOLT_COUNT]; - - s8 temp[ADT7473_TEMP_COUNT]; - s8 temp_min[ADT7473_TEMP_COUNT]; - s8 temp_max[ADT7473_TEMP_COUNT]; - s8 temp_tmin[ADT7473_TEMP_COUNT]; - /* This is called the !THERM limit in the datasheet */ - s8 temp_tmax[ADT7473_TEMP_COUNT]; - - u16 fan[ADT7473_FAN_COUNT]; - u16 fan_min[ADT7473_FAN_COUNT]; - - u8 pwm[ADT7473_PWM_COUNT]; - u8 pwm_max[ADT7473_PWM_COUNT]; - u8 pwm_min[ADT7473_PWM_COUNT]; - u8 pwm_behavior[ADT7473_PWM_COUNT]; - - u8 temp_twos_complement; - u8 temp_offset; - - u16 alarm; - u8 max_duty_at_overheat; -}; - -static int adt7473_probe(struct i2c_client *client, - const struct i2c_device_id *id); -static int adt7473_detect(struct i2c_client *client, - struct i2c_board_info *info); -static int adt7473_remove(struct i2c_client *client); - -static const struct i2c_device_id adt7473_id[] = { - { "adt7473", 0 }, - { } -}; - -static struct i2c_driver adt7473_driver = { - .class = I2C_CLASS_HWMON, - .driver = { - .name = "adt7473", - }, - .probe = adt7473_probe, - .remove = adt7473_remove, - .id_table = adt7473_id, - .detect = adt7473_detect, - .address_list = normal_i2c, -}; - -/* - * 16-bit registers on the ADT7473 are low-byte first. The data sheet says - * that the low byte must be read before the high byte. - */ -static inline int adt7473_read_word_data(struct i2c_client *client, u8 reg) -{ - u16 foo; - foo = i2c_smbus_read_byte_data(client, reg); - foo |= ((u16)i2c_smbus_read_byte_data(client, reg + 1) << 8); - return foo; -} - -static inline int adt7473_write_word_data(struct i2c_client *client, u8 reg, - u16 value) -{ - return i2c_smbus_write_byte_data(client, reg, value & 0xFF) - && i2c_smbus_write_byte_data(client, reg + 1, value >> 8); -} - -static void adt7473_init_client(struct i2c_client *client) -{ - int reg = i2c_smbus_read_byte_data(client, ADT7473_REG_CFG1); - - if (!(reg & ADT7473_CFG1_READY)) { - dev_err(&client->dev, "Chip not ready.\n"); - } else { - /* start monitoring */ - i2c_smbus_write_byte_data(client, ADT7473_REG_CFG1, - reg | ADT7473_CFG1_START); - } -} - -static struct adt7473_data *adt7473_update_device(struct device *dev) -{ - struct i2c_client *client = to_i2c_client(dev); - struct adt7473_data *data = i2c_get_clientdata(client); - unsigned long local_jiffies = jiffies; - u8 cfg; - int i; - - mutex_lock(&data->lock); - if (time_before(local_jiffies, data->sensors_last_updated + - SENSOR_REFRESH_INTERVAL) - && data->sensors_valid) - goto no_sensor_update; - - for (i = 0; i < ADT7473_VOLT_COUNT; i++) - data->volt[i] = i2c_smbus_read_byte_data(client, - ADT7473_REG_VOLT(i)); - - /* Determine temperature encoding */ - cfg = i2c_smbus_read_byte_data(client, ADT7473_REG_CFG5); - data->temp_twos_complement = (cfg & ADT7473_CFG5_TEMP_TWOS); - - /* - * What does this do? it implies a variable temperature sensor - * offset, but the datasheet doesn't say anything about this bit - * and other parts of the datasheet imply that "offset64" mode - * means that you shift temp values by -64 if the above bit was set. - */ - data->temp_offset = (cfg & ADT7473_CFG5_TEMP_OFFSET); - - for (i = 0; i < ADT7473_TEMP_COUNT; i++) - data->temp[i] = i2c_smbus_read_byte_data(client, - ADT7473_REG_TEMP(i)); - - for (i = 0; i < ADT7473_FAN_COUNT; i++) - data->fan[i] = adt7473_read_word_data(client, - ADT7473_REG_FAN(i)); - - for (i = 0; i < ADT7473_PWM_COUNT; i++) - data->pwm[i] = i2c_smbus_read_byte_data(client, - ADT7473_REG_PWM(i)); - - data->alarm = i2c_smbus_read_byte_data(client, ADT7473_REG_ALARM1); - if (data->alarm & ADT7473_OOL) - data->alarm |= ALARM2(i2c_smbus_read_byte_data(client, - ADT7473_REG_ALARM2)); - - data->sensors_last_updated = local_jiffies; - data->sensors_valid = 1; - -no_sensor_update: - if (time_before(local_jiffies, data->limits_last_updated + - LIMIT_REFRESH_INTERVAL) - && data->limits_valid) - goto out; - - for (i = 0; i < ADT7473_VOLT_COUNT; i++) { - data->volt_min[i] = i2c_smbus_read_byte_data(client, - ADT7473_REG_VOLT_MIN(i)); - data->volt_max[i] = i2c_smbus_read_byte_data(client, - ADT7473_REG_VOLT_MAX(i)); - } - - for (i = 0; i < ADT7473_TEMP_COUNT; i++) { - data->temp_min[i] = i2c_smbus_read_byte_data(client, - ADT7473_REG_TEMP_MIN(i)); - data->temp_max[i] = i2c_smbus_read_byte_data(client, - ADT7473_REG_TEMP_MAX(i)); - data->temp_tmin[i] = i2c_smbus_read_byte_data(client, - ADT7473_REG_TEMP_TMIN(i)); - data->temp_tmax[i] = i2c_smbus_read_byte_data(client, - ADT7473_REG_TEMP_TMAX(i)); - } - - for (i = 0; i < ADT7473_FAN_COUNT; i++) - data->fan_min[i] = adt7473_read_word_data(client, - ADT7473_REG_FAN_MIN(i)); - - for (i = 0; i < ADT7473_PWM_COUNT; i++) { - data->pwm_max[i] = i2c_smbus_read_byte_data(client, - ADT7473_REG_PWM_MAX(i)); - data->pwm_min[i] = i2c_smbus_read_byte_data(client, - ADT7473_REG_PWM_MIN(i)); - data->pwm_behavior[i] = i2c_smbus_read_byte_data(client, - ADT7473_REG_PWM_BHVR(i)); - } - - i = i2c_smbus_read_byte_data(client, ADT7473_REG_CFG4); - data->max_duty_at_overheat = !!(i & ADT7473_CFG4_MAX_DUTY_AT_OVT); - - data->limits_last_updated = local_jiffies; - data->limits_valid = 1; - -out: - mutex_unlock(&data->lock); - return data; -} - -/* - * Conversions - */ - -/* IN are scaled acording to built-in resistors */ -static const int adt7473_scaling[] = { /* .001 Volts */ - 2250, 3300 -}; -#define SCALE(val, from, to) (((val) * (to) + ((from) / 2)) / (from)) - -static int decode_volt(int volt_index, u8 raw) -{ - return SCALE(raw, 192, adt7473_scaling[volt_index]); -} - -static u8 encode_volt(int volt_index, int cooked) -{ - int raw = SCALE(cooked, adt7473_scaling[volt_index], 192); - return SENSORS_LIMIT(raw, 0, 255); -} - -static ssize_t show_volt_min(struct device *dev, - struct device_attribute *devattr, - char *buf) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct adt7473_data *data = adt7473_update_device(dev); - return sprintf(buf, "%d\n", - decode_volt(attr->index, data->volt_min[attr->index])); -} - -static ssize_t set_volt_min(struct device *dev, - struct device_attribute *devattr, - const char *buf, - size_t count) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct adt7473_data *data = i2c_get_clientdata(client); - long volt; - - if (strict_strtol(buf, 10, &volt)) - return -EINVAL; - - volt = encode_volt(attr->index, volt); - - mutex_lock(&data->lock); - data->volt_min[attr->index] = volt; - i2c_smbus_write_byte_data(client, ADT7473_REG_VOLT_MIN(attr->index), - volt); - mutex_unlock(&data->lock); - - return count; -} - -static ssize_t show_volt_max(struct device *dev, - struct device_attribute *devattr, - char *buf) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct adt7473_data *data = adt7473_update_device(dev); - return sprintf(buf, "%d\n", - decode_volt(attr->index, data->volt_max[attr->index])); -} - -static ssize_t set_volt_max(struct device *dev, - struct device_attribute *devattr, - const char *buf, - size_t count) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct adt7473_data *data = i2c_get_clientdata(client); - long volt; - - if (strict_strtol(buf, 10, &volt)) - return -EINVAL; - - volt = encode_volt(attr->index, volt); - - mutex_lock(&data->lock); - data->volt_max[attr->index] = volt; - i2c_smbus_write_byte_data(client, ADT7473_REG_VOLT_MAX(attr->index), - volt); - mutex_unlock(&data->lock); - - return count; -} - -static ssize_t show_volt(struct device *dev, struct device_attribute *devattr, - char *buf) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct adt7473_data *data = adt7473_update_device(dev); - - return sprintf(buf, "%d\n", - decode_volt(attr->index, data->volt[attr->index])); -} - -/* - * This chip can report temperature data either as a two's complement - * number in the range -128 to 127, or as an unsigned number that must - * be offset by 64. - */ -static int decode_temp(u8 twos_complement, u8 raw) -{ - return twos_complement ? (s8)raw : raw - 64; -} - -static u8 encode_temp(u8 twos_complement, int cooked) -{ - u8 ret = twos_complement ? cooked & 0xFF : cooked + 64; - return SENSORS_LIMIT(ret, 0, 255); -} - -static ssize_t show_temp_min(struct device *dev, - struct device_attribute *devattr, - char *buf) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct adt7473_data *data = adt7473_update_device(dev); - return sprintf(buf, "%d\n", 1000 * decode_temp( - data->temp_twos_complement, - data->temp_min[attr->index])); -} - -static ssize_t set_temp_min(struct device *dev, - struct device_attribute *devattr, - const char *buf, - size_t count) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct adt7473_data *data = i2c_get_clientdata(client); - long temp; - - if (strict_strtol(buf, 10, &temp)) - return -EINVAL; - - temp = DIV_ROUND_CLOSEST(temp, 1000); - temp = encode_temp(data->temp_twos_complement, temp); - - mutex_lock(&data->lock); - data->temp_min[attr->index] = temp; - i2c_smbus_write_byte_data(client, ADT7473_REG_TEMP_MIN(attr->index), - temp); - mutex_unlock(&data->lock); - - return count; -} - -static ssize_t show_temp_max(struct device *dev, - struct device_attribute *devattr, - char *buf) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct adt7473_data *data = adt7473_update_device(dev); - return sprintf(buf, "%d\n", 1000 * decode_temp( - data->temp_twos_complement, - data->temp_max[attr->index])); -} - -static ssize_t set_temp_max(struct device *dev, - struct device_attribute *devattr, - const char *buf, - size_t count) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct adt7473_data *data = i2c_get_clientdata(client); - long temp; - - if (strict_strtol(buf, 10, &temp)) - return -EINVAL; - - temp = DIV_ROUND_CLOSEST(temp, 1000); - temp = encode_temp(data->temp_twos_complement, temp); - - mutex_lock(&data->lock); - data->temp_max[attr->index] = temp; - i2c_smbus_write_byte_data(client, ADT7473_REG_TEMP_MAX(attr->index), - temp); - mutex_unlock(&data->lock); - - return count; -} - -static ssize_t show_temp(struct device *dev, struct device_attribute *devattr, - char *buf) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct adt7473_data *data = adt7473_update_device(dev); - return sprintf(buf, "%d\n", 1000 * decode_temp( - data->temp_twos_complement, - data->temp[attr->index])); -} - -static ssize_t show_fan_min(struct device *dev, - struct device_attribute *devattr, - char *buf) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct adt7473_data *data = adt7473_update_device(dev); - - if (FAN_DATA_VALID(data->fan_min[attr->index])) - return sprintf(buf, "%d\n", - FAN_PERIOD_TO_RPM(data->fan_min[attr->index])); - else - return sprintf(buf, "0\n"); -} - -static ssize_t set_fan_min(struct device *dev, - struct device_attribute *devattr, - const char *buf, size_t count) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct adt7473_data *data = i2c_get_clientdata(client); - long temp; - - if (strict_strtol(buf, 10, &temp) || !temp) - return -EINVAL; - - temp = FAN_RPM_TO_PERIOD(temp); - temp = SENSORS_LIMIT(temp, 1, 65534); - - mutex_lock(&data->lock); - data->fan_min[attr->index] = temp; - adt7473_write_word_data(client, ADT7473_REG_FAN_MIN(attr->index), temp); - mutex_unlock(&data->lock); - - return count; -} - -static ssize_t show_fan(struct device *dev, struct device_attribute *devattr, - char *buf) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct adt7473_data *data = adt7473_update_device(dev); - - if (FAN_DATA_VALID(data->fan[attr->index])) - return sprintf(buf, "%d\n", - FAN_PERIOD_TO_RPM(data->fan[attr->index])); - else - return sprintf(buf, "0\n"); -} - -static ssize_t show_max_duty_at_crit(struct device *dev, - struct device_attribute *devattr, - char *buf) -{ - struct adt7473_data *data = adt7473_update_device(dev); - return sprintf(buf, "%d\n", data->max_duty_at_overheat); -} - -static ssize_t set_max_duty_at_crit(struct device *dev, - struct device_attribute *devattr, - const char *buf, - size_t count) -{ - u8 reg; - struct i2c_client *client = to_i2c_client(dev); - struct adt7473_data *data = i2c_get_clientdata(client); - long temp; - - if (strict_strtol(buf, 10, &temp)) - return -EINVAL; - - mutex_lock(&data->lock); - data->max_duty_at_overheat = !!temp; - reg = i2c_smbus_read_byte_data(client, ADT7473_REG_CFG4); - if (temp) - reg |= ADT7473_CFG4_MAX_DUTY_AT_OVT; - else - reg &= ~ADT7473_CFG4_MAX_DUTY_AT_OVT; - i2c_smbus_write_byte_data(client, ADT7473_REG_CFG4, reg); - mutex_unlock(&data->lock); - - return count; -} - -static ssize_t show_pwm(struct device *dev, struct device_attribute *devattr, - char *buf) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct adt7473_data *data = adt7473_update_device(dev); - return sprintf(buf, "%d\n", data->pwm[attr->index]); -} - -static ssize_t set_pwm(struct device *dev, struct device_attribute *devattr, - const char *buf, size_t count) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct adt7473_data *data = i2c_get_clientdata(client); - long temp; - - if (strict_strtol(buf, 10, &temp)) - return -EINVAL; - - temp = SENSORS_LIMIT(temp, 0, 255); - - mutex_lock(&data->lock); - data->pwm[attr->index] = temp; - i2c_smbus_write_byte_data(client, ADT7473_REG_PWM(attr->index), temp); - mutex_unlock(&data->lock); - - return count; -} - -static ssize_t show_pwm_max(struct device *dev, - struct device_attribute *devattr, - char *buf) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct adt7473_data *data = adt7473_update_device(dev); - return sprintf(buf, "%d\n", data->pwm_max[attr->index]); -} - -static ssize_t set_pwm_max(struct device *dev, - struct device_attribute *devattr, - const char *buf, - size_t count) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct adt7473_data *data = i2c_get_clientdata(client); - long temp; - - if (strict_strtol(buf, 10, &temp)) - return -EINVAL; - - temp = SENSORS_LIMIT(temp, 0, 255); - - mutex_lock(&data->lock); - data->pwm_max[attr->index] = temp; - i2c_smbus_write_byte_data(client, ADT7473_REG_PWM_MAX(attr->index), - temp); - mutex_unlock(&data->lock); - - return count; -} - -static ssize_t show_pwm_min(struct device *dev, - struct device_attribute *devattr, - char *buf) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct adt7473_data *data = adt7473_update_device(dev); - return sprintf(buf, "%d\n", data->pwm_min[attr->index]); -} - -static ssize_t set_pwm_min(struct device *dev, - struct device_attribute *devattr, - const char *buf, - size_t count) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct adt7473_data *data = i2c_get_clientdata(client); - long temp; - - if (strict_strtol(buf, 10, &temp)) - return -EINVAL; - - temp = SENSORS_LIMIT(temp, 0, 255); - - mutex_lock(&data->lock); - data->pwm_min[attr->index] = temp; - i2c_smbus_write_byte_data(client, ADT7473_REG_PWM_MIN(attr->index), - temp); - mutex_unlock(&data->lock); - - return count; -} - -static ssize_t show_temp_tmax(struct device *dev, - struct device_attribute *devattr, - char *buf) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct adt7473_data *data = adt7473_update_device(dev); - return sprintf(buf, "%d\n", 1000 * decode_temp( - data->temp_twos_complement, - data->temp_tmax[attr->index])); -} - -static ssize_t set_temp_tmax(struct device *dev, - struct device_attribute *devattr, - const char *buf, - size_t count) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct adt7473_data *data = i2c_get_clientdata(client); - long temp; - - if (strict_strtol(buf, 10, &temp)) - return -EINVAL; - - temp = DIV_ROUND_CLOSEST(temp, 1000); - temp = encode_temp(data->temp_twos_complement, temp); - - mutex_lock(&data->lock); - data->temp_tmax[attr->index] = temp; - i2c_smbus_write_byte_data(client, ADT7473_REG_TEMP_TMAX(attr->index), - temp); - mutex_unlock(&data->lock); - - return count; -} - -static ssize_t show_temp_tmin(struct device *dev, - struct device_attribute *devattr, - char *buf) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct adt7473_data *data = adt7473_update_device(dev); - return sprintf(buf, "%d\n", 1000 * decode_temp( - data->temp_twos_complement, - data->temp_tmin[attr->index])); -} - -static ssize_t set_temp_tmin(struct device *dev, - struct device_attribute *devattr, - const char *buf, - size_t count) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct adt7473_data *data = i2c_get_clientdata(client); - long temp; - - if (strict_strtol(buf, 10, &temp)) - return -EINVAL; - - temp = DIV_ROUND_CLOSEST(temp, 1000); - temp = encode_temp(data->temp_twos_complement, temp); - - mutex_lock(&data->lock); - data->temp_tmin[attr->index] = temp; - i2c_smbus_write_byte_data(client, ADT7473_REG_TEMP_TMIN(attr->index), - temp); - mutex_unlock(&data->lock); - - return count; -} - -static ssize_t show_pwm_enable(struct device *dev, - struct device_attribute *devattr, - char *buf) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct adt7473_data *data = adt7473_update_device(dev); - - switch (data->pwm_behavior[attr->index] >> ADT7473_PWM_BHVR_SHIFT) { - case 3: - return sprintf(buf, "0\n"); - case 7: - return sprintf(buf, "1\n"); - default: - return sprintf(buf, "2\n"); - } -} - -static ssize_t set_pwm_enable(struct device *dev, - struct device_attribute *devattr, - const char *buf, - size_t count) -{ - u8 reg; - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct adt7473_data *data = i2c_get_clientdata(client); - long temp; - - if (strict_strtol(buf, 10, &temp)) - return -EINVAL; - - switch (temp) { - case 0: - temp = 3; - break; - case 1: - temp = 7; - break; - case 2: - /* Enter automatic mode with fans off */ - temp = 4; - break; - default: - return -EINVAL; - } - - mutex_lock(&data->lock); - reg = i2c_smbus_read_byte_data(client, - ADT7473_REG_PWM_BHVR(attr->index)); - reg = (temp << ADT7473_PWM_BHVR_SHIFT) | - (reg & ~ADT7473_PWM_BHVR_MASK); - i2c_smbus_write_byte_data(client, ADT7473_REG_PWM_BHVR(attr->index), - reg); - data->pwm_behavior[attr->index] = reg; - mutex_unlock(&data->lock); - - return count; -} - -static ssize_t show_pwm_auto_temp(struct device *dev, - struct device_attribute *devattr, - char *buf) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct adt7473_data *data = adt7473_update_device(dev); - int bhvr = data->pwm_behavior[attr->index] >> ADT7473_PWM_BHVR_SHIFT; - - switch (bhvr) { - case 3: - case 4: - case 7: - return sprintf(buf, "0\n"); - case 0: - case 1: - case 5: - case 6: - return sprintf(buf, "%d\n", bhvr + 1); - case 2: - return sprintf(buf, "4\n"); - } - /* shouldn't ever get here */ - BUG(); -} - -static ssize_t set_pwm_auto_temp(struct device *dev, - struct device_attribute *devattr, - const char *buf, - size_t count) -{ - u8 reg; - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct i2c_client *client = to_i2c_client(dev); - struct adt7473_data *data = i2c_get_clientdata(client); - long temp; - - if (strict_strtol(buf, 10, &temp)) - return -EINVAL; - - switch (temp) { - case 1: - case 2: - case 6: - case 7: - temp--; - break; - case 0: - temp = 4; - break; - default: - return -EINVAL; - } - - mutex_lock(&data->lock); - reg = i2c_smbus_read_byte_data(client, - ADT7473_REG_PWM_BHVR(attr->index)); - reg = (temp << ADT7473_PWM_BHVR_SHIFT) | - (reg & ~ADT7473_PWM_BHVR_MASK); - i2c_smbus_write_byte_data(client, ADT7473_REG_PWM_BHVR(attr->index), - reg); - data->pwm_behavior[attr->index] = reg; - mutex_unlock(&data->lock); - - return count; -} - -static ssize_t show_alarm(struct device *dev, - struct device_attribute *devattr, - char *buf) -{ - struct sensor_device_attribute *attr = to_sensor_dev_attr(devattr); - struct adt7473_data *data = adt7473_update_device(dev); - - if (data->alarm & attr->index) - return sprintf(buf, "1\n"); - else - return sprintf(buf, "0\n"); -} - - -static SENSOR_DEVICE_ATTR(in1_max, S_IWUSR | S_IRUGO, show_volt_max, - set_volt_max, 0); -static SENSOR_DEVICE_ATTR(in2_max, S_IWUSR | S_IRUGO, show_volt_max, - set_volt_max, 1); - -static SENSOR_DEVICE_ATTR(in1_min, S_IWUSR | S_IRUGO, show_volt_min, - set_volt_min, 0); -static SENSOR_DEVICE_ATTR(in2_min, S_IWUSR | S_IRUGO, show_volt_min, - set_volt_min, 1); - -static SENSOR_DEVICE_ATTR(in1_input, S_IRUGO, show_volt, NULL, 0); -static SENSOR_DEVICE_ATTR(in2_input, S_IRUGO, show_volt, NULL, 1); - -static SENSOR_DEVICE_ATTR(in1_alarm, S_IRUGO, show_alarm, NULL, - ADT7473_VCCP_ALARM); -static SENSOR_DEVICE_ATTR(in2_alarm, S_IRUGO, show_alarm, NULL, - ADT7473_VCC_ALARM); - -static SENSOR_DEVICE_ATTR(temp1_max, S_IWUSR | S_IRUGO, show_temp_max, - set_temp_max, 0); -static SENSOR_DEVICE_ATTR(temp2_max, S_IWUSR | S_IRUGO, show_temp_max, - set_temp_max, 1); -static SENSOR_DEVICE_ATTR(temp3_max, S_IWUSR | S_IRUGO, show_temp_max, - set_temp_max, 2); - -static SENSOR_DEVICE_ATTR(temp1_min, S_IWUSR | S_IRUGO, show_temp_min, - set_temp_min, 0); -static SENSOR_DEVICE_ATTR(temp2_min, S_IWUSR | S_IRUGO, show_temp_min, - set_temp_min, 1); -static SENSOR_DEVICE_ATTR(temp3_min, S_IWUSR | S_IRUGO, show_temp_min, - set_temp_min, 2); - -static SENSOR_DEVICE_ATTR(temp1_input, S_IRUGO, show_temp, NULL, 0); -static SENSOR_DEVICE_ATTR(temp2_input, S_IRUGO, show_temp, NULL, 1); -static SENSOR_DEVICE_ATTR(temp3_input, S_IRUGO, show_temp, NULL, 2); - -static SENSOR_DEVICE_ATTR(temp1_alarm, S_IRUGO, show_alarm, NULL, - ADT7473_R1T_ALARM | ALARM2(ADT7473_R1T_SHORT)); -static SENSOR_DEVICE_ATTR(temp2_alarm, S_IRUGO, show_alarm, NULL, - ADT7473_LT_ALARM); -static SENSOR_DEVICE_ATTR(temp3_alarm, S_IRUGO, show_alarm, NULL, - ADT7473_R2T_ALARM | ALARM2(ADT7473_R2T_SHORT)); - -static SENSOR_DEVICE_ATTR(fan1_min, S_IWUSR | S_IRUGO, show_fan_min, - set_fan_min, 0); -static SENSOR_DEVICE_ATTR(fan2_min, S_IWUSR | S_IRUGO, show_fan_min, - set_fan_min, 1); -static SENSOR_DEVICE_ATTR(fan3_min, S_IWUSR | S_IRUGO, show_fan_min, - set_fan_min, 2); -static SENSOR_DEVICE_ATTR(fan4_min, S_IWUSR | S_IRUGO, show_fan_min, - set_fan_min, 3); - -static SENSOR_DEVICE_ATTR(fan1_input, S_IRUGO, show_fan, NULL, 0); -static SENSOR_DEVICE_ATTR(fan2_input, S_IRUGO, show_fan, NULL, 1); -static SENSOR_DEVICE_ATTR(fan3_input, S_IRUGO, show_fan, NULL, 2); -static SENSOR_DEVICE_ATTR(fan4_input, S_IRUGO, show_fan, NULL, 3); - -static SENSOR_DEVICE_ATTR(fan1_alarm, S_IRUGO, show_alarm, NULL, - ALARM2(ADT7473_FAN1_ALARM)); -static SENSOR_DEVICE_ATTR(fan2_alarm, S_IRUGO, show_alarm, NULL, - ALARM2(ADT7473_FAN2_ALARM)); -static SENSOR_DEVICE_ATTR(fan3_alarm, S_IRUGO, show_alarm, NULL, - ALARM2(ADT7473_FAN3_ALARM)); -static SENSOR_DEVICE_ATTR(fan4_alarm, S_IRUGO, show_alarm, NULL, - ALARM2(ADT7473_FAN4_ALARM)); - -static SENSOR_DEVICE_ATTR(pwm_use_point2_pwm_at_crit, S_IWUSR | S_IRUGO, - show_max_duty_at_crit, set_max_duty_at_crit, 0); - -static SENSOR_DEVICE_ATTR(pwm1, S_IWUSR | S_IRUGO, show_pwm, set_pwm, 0); -static SENSOR_DEVICE_ATTR(pwm2, S_IWUSR | S_IRUGO, show_pwm, set_pwm, 1); -static SENSOR_DEVICE_ATTR(pwm3, S_IWUSR | S_IRUGO, show_pwm, set_pwm, 2); - -static SENSOR_DEVICE_ATTR(pwm1_auto_point1_pwm, S_IWUSR | S_IRUGO, - show_pwm_min, set_pwm_min, 0); -static SENSOR_DEVICE_ATTR(pwm2_auto_point1_pwm, S_IWUSR | S_IRUGO, - show_pwm_min, set_pwm_min, 1); -static SENSOR_DEVICE_ATTR(pwm3_auto_point1_pwm, S_IWUSR | S_IRUGO, - show_pwm_min, set_pwm_min, 2); - -static SENSOR_DEVICE_ATTR(pwm1_auto_point2_pwm, S_IWUSR | S_IRUGO, - show_pwm_max, set_pwm_max, 0); -static SENSOR_DEVICE_ATTR(pwm2_auto_point2_pwm, S_IWUSR | S_IRUGO, - show_pwm_max, set_pwm_max, 1); -static SENSOR_DEVICE_ATTR(pwm3_auto_point2_pwm, S_IWUSR | S_IRUGO, - show_pwm_max, set_pwm_max, 2); - -static SENSOR_DEVICE_ATTR(temp1_auto_point1_temp, S_IWUSR | S_IRUGO, - show_temp_tmin, set_temp_tmin, 0); -static SENSOR_DEVICE_ATTR(temp2_auto_point1_temp, S_IWUSR | S_IRUGO, - show_temp_tmin, set_temp_tmin, 1); -static SENSOR_DEVICE_ATTR(temp3_auto_point1_temp, S_IWUSR | S_IRUGO, - show_temp_tmin, set_temp_tmin, 2); - -static SENSOR_DEVICE_ATTR(temp1_auto_point2_temp, S_IWUSR | S_IRUGO, - show_temp_tmax, set_temp_tmax, 0); -static SENSOR_DEVICE_ATTR(temp2_auto_point2_temp, S_IWUSR | S_IRUGO, - show_temp_tmax, set_temp_tmax, 1); -static SENSOR_DEVICE_ATTR(temp3_auto_point2_temp, S_IWUSR | S_IRUGO, - show_temp_tmax, set_temp_tmax, 2); - -static SENSOR_DEVICE_ATTR(pwm1_enable, S_IWUSR | S_IRUGO, show_pwm_enable, - set_pwm_enable, 0); -static SENSOR_DEVICE_ATTR(pwm2_enable, S_IWUSR | S_IRUGO, show_pwm_enable, - set_pwm_enable, 1); -static SENSOR_DEVICE_ATTR(pwm3_enable, S_IWUSR | S_IRUGO, show_pwm_enable, - set_pwm_enable, 2); - -static SENSOR_DEVICE_ATTR(pwm1_auto_channels_temp, S_IWUSR | S_IRUGO, - show_pwm_auto_temp, set_pwm_auto_temp, 0); -static SENSOR_DEVICE_ATTR(pwm2_auto_channels_temp, S_IWUSR | S_IRUGO, - show_pwm_auto_temp, set_pwm_auto_temp, 1); -static SENSOR_DEVICE_ATTR(pwm3_auto_channels_temp, S_IWUSR | S_IRUGO, - show_pwm_auto_temp, set_pwm_auto_temp, 2); - -static struct attribute *adt7473_attr[] = -{ - &sensor_dev_attr_in1_max.dev_attr.attr, - &sensor_dev_attr_in2_max.dev_attr.attr, - &sensor_dev_attr_in1_min.dev_attr.attr, - &sensor_dev_attr_in2_min.dev_attr.attr, - &sensor_dev_attr_in1_input.dev_attr.attr, - &sensor_dev_attr_in2_input.dev_attr.attr, - &sensor_dev_attr_in1_alarm.dev_attr.attr, - &sensor_dev_attr_in2_alarm.dev_attr.attr, - - &sensor_dev_attr_temp1_max.dev_attr.attr, - &sensor_dev_attr_temp2_max.dev_attr.attr, - &sensor_dev_attr_temp3_max.dev_attr.attr, - &sensor_dev_attr_temp1_min.dev_attr.attr, - &sensor_dev_attr_temp2_min.dev_attr.attr, - &sensor_dev_attr_temp3_min.dev_attr.attr, - &sensor_dev_attr_temp1_input.dev_attr.attr, - &sensor_dev_attr_temp2_input.dev_attr.attr, - &sensor_dev_attr_temp3_input.dev_attr.attr, - &sensor_dev_attr_temp1_alarm.dev_attr.attr, - &sensor_dev_attr_temp2_alarm.dev_attr.attr, - &sensor_dev_attr_temp3_alarm.dev_attr.attr, - &sensor_dev_attr_temp1_auto_point1_temp.dev_attr.attr, - &sensor_dev_attr_temp2_auto_point1_temp.dev_attr.attr, - &sensor_dev_attr_temp3_auto_point1_temp.dev_attr.attr, - &sensor_dev_attr_temp1_auto_point2_temp.dev_attr.attr, - &sensor_dev_attr_temp2_auto_point2_temp.dev_attr.attr, - &sensor_dev_attr_temp3_auto_point2_temp.dev_attr.attr, - - &sensor_dev_attr_fan1_min.dev_attr.attr, - &sensor_dev_attr_fan2_min.dev_attr.attr, - &sensor_dev_attr_fan3_min.dev_attr.attr, - &sensor_dev_attr_fan4_min.dev_attr.attr, - &sensor_dev_attr_fan1_input.dev_attr.attr, - &sensor_dev_attr_fan2_input.dev_attr.attr, - &sensor_dev_attr_fan3_input.dev_attr.attr, - &sensor_dev_attr_fan4_input.dev_attr.attr, - &sensor_dev_attr_fan1_alarm.dev_attr.attr, - &sensor_dev_attr_fan2_alarm.dev_attr.attr, - &sensor_dev_attr_fan3_alarm.dev_attr.attr, - &sensor_dev_attr_fan4_alarm.dev_attr.attr, - - &sensor_dev_attr_pwm_use_point2_pwm_at_crit.dev_attr.attr, - - &sensor_dev_attr_pwm1.dev_attr.attr, - &sensor_dev_attr_pwm2.dev_attr.attr, - &sensor_dev_attr_pwm3.dev_attr.attr, - &sensor_dev_attr_pwm1_auto_point1_pwm.dev_attr.attr, - &sensor_dev_attr_pwm2_auto_point1_pwm.dev_attr.attr, - &sensor_dev_attr_pwm3_auto_point1_pwm.dev_attr.attr, - &sensor_dev_attr_pwm1_auto_point2_pwm.dev_attr.attr, - &sensor_dev_attr_pwm2_auto_point2_pwm.dev_attr.attr, - &sensor_dev_attr_pwm3_auto_point2_pwm.dev_attr.attr, - - &sensor_dev_attr_pwm1_enable.dev_attr.attr, - &sensor_dev_attr_pwm2_enable.dev_attr.attr, - &sensor_dev_attr_pwm3_enable.dev_attr.attr, - &sensor_dev_attr_pwm1_auto_channels_temp.dev_attr.attr, - &sensor_dev_attr_pwm2_auto_channels_temp.dev_attr.attr, - &sensor_dev_attr_pwm3_auto_channels_temp.dev_attr.attr, - - NULL -}; - -/* Return 0 if detection is successful, -ENODEV otherwise */ -static int adt7473_detect(struct i2c_client *client, - struct i2c_board_info *info) -{ - struct i2c_adapter *adapter = client->adapter; - int vendor, device, revision; - - if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_BYTE_DATA)) - return -ENODEV; - - vendor = i2c_smbus_read_byte_data(client, ADT7473_REG_VENDOR); - if (vendor != ADT7473_VENDOR) - return -ENODEV; - - device = i2c_smbus_read_byte_data(client, ADT7473_REG_DEVICE); - if (device != ADT7473_DEVICE) - return -ENODEV; - - revision = i2c_smbus_read_byte_data(client, ADT7473_REG_REVISION); - if (revision != ADT7473_REV_68 && revision != ADT7473_REV_69) - return -ENODEV; - - strlcpy(info->type, "adt7473", I2C_NAME_SIZE); - - return 0; -} - -static int adt7473_probe(struct i2c_client *client, - const struct i2c_device_id *id) -{ - struct adt7473_data *data; - int err; - - data = kzalloc(sizeof(struct adt7473_data), GFP_KERNEL); - if (!data) { - err = -ENOMEM; - goto exit; - } - - i2c_set_clientdata(client, data); - mutex_init(&data->lock); - - dev_info(&client->dev, "%s chip found\n", client->name); - - /* Initialize the ADT7473 chip */ - adt7473_init_client(client); - - /* Register sysfs hooks */ - data->attrs.attrs = adt7473_attr; - err = sysfs_create_group(&client->dev.kobj, &data->attrs); - if (err) - goto exit_free; - - data->hwmon_dev = hwmon_device_register(&client->dev); - if (IS_ERR(data->hwmon_dev)) { - err = PTR_ERR(data->hwmon_dev); - goto exit_remove; - } - - return 0; - -exit_remove: - sysfs_remove_group(&client->dev.kobj, &data->attrs); -exit_free: - kfree(data); -exit: - return err; -} - -static int adt7473_remove(struct i2c_client *client) -{ - struct adt7473_data *data = i2c_get_clientdata(client); - - hwmon_device_unregister(data->hwmon_dev); - sysfs_remove_group(&client->dev.kobj, &data->attrs); - kfree(data); - return 0; -} - -static int __init adt7473_init(void) -{ - pr_notice("The adt7473 driver is deprecated, please use the adt7475 " - "driver instead\n"); - return i2c_add_driver(&adt7473_driver); -} - -static void __exit adt7473_exit(void) -{ - i2c_del_driver(&adt7473_driver); -} - -MODULE_AUTHOR("Darrick J. Wong <djwong@us.ibm.com>"); -MODULE_DESCRIPTION("ADT7473 driver"); -MODULE_LICENSE("GPL"); - -module_init(adt7473_init); -module_exit(adt7473_exit); diff --git a/drivers/hwmon/asc7621.c b/drivers/hwmon/asc7621.c new file mode 100644 index 000000000000..7f948105d8ad --- /dev/null +++ b/drivers/hwmon/asc7621.c @@ -0,0 +1,1255 @@ +/* + * asc7621.c - Part of lm_sensors, Linux kernel modules for hardware monitoring + * Copyright (c) 2007, 2010 George Joseph <george.joseph@fairview5.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + */ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/slab.h> +#include <linux/jiffies.h> +#include <linux/i2c.h> +#include <linux/hwmon.h> +#include <linux/hwmon-sysfs.h> +#include <linux/err.h> +#include <linux/mutex.h> + +/* Addresses to scan */ +static unsigned short normal_i2c[] = { + 0x2c, 0x2d, 0x2e, I2C_CLIENT_END +}; + +enum asc7621_type { + asc7621, + asc7621a +}; + +#define INTERVAL_HIGH (HZ + HZ / 2) +#define INTERVAL_LOW (1 * 60 * HZ) +#define PRI_NONE 0 +#define PRI_LOW 1 +#define PRI_HIGH 2 +#define FIRST_CHIP asc7621 +#define LAST_CHIP asc7621a + +struct asc7621_chip { + char *name; + enum asc7621_type chip_type; + u8 company_reg; + u8 company_id; + u8 verstep_reg; + u8 verstep_id; + unsigned short *addresses; +}; + +static struct asc7621_chip asc7621_chips[] = { + { + .name = "asc7621", + .chip_type = asc7621, + .company_reg = 0x3e, + .company_id = 0x61, + .verstep_reg = 0x3f, + .verstep_id = 0x6c, + .addresses = normal_i2c, + }, + { + .name = "asc7621a", + .chip_type = asc7621a, + .company_reg = 0x3e, + .company_id = 0x61, + .verstep_reg = 0x3f, + .verstep_id = 0x6d, + .addresses = normal_i2c, + }, +}; + +/* + * Defines the highest register to be used, not the count. + * The actual count will probably be smaller because of gaps + * in the implementation (unused register locations). + * This define will safely set the array size of both the parameter + * and data arrays. + * This comes from the data sheet register description table. + */ +#define LAST_REGISTER 0xff + +struct asc7621_data { + struct i2c_client client; + struct device *class_dev; + struct mutex update_lock; + int valid; /* !=0 if following fields are valid */ + unsigned long last_high_reading; /* In jiffies */ + unsigned long last_low_reading; /* In jiffies */ + /* + * Registers we care about occupy the corresponding index + * in the array. Registers we don't care about are left + * at 0. + */ + u8 reg[LAST_REGISTER + 1]; +}; + +/* + * Macro to get the parent asc7621_param structure + * from a sensor_device_attribute passed into the + * show/store functions. + */ +#define to_asc7621_param(_sda) \ + container_of(_sda, struct asc7621_param, sda) + +/* + * Each parameter to be retrieved needs an asc7621_param structure + * allocated. It contains the sensor_device_attribute structure + * and the control info needed to retrieve the value from the register map. + */ +struct asc7621_param { + struct sensor_device_attribute sda; + u8 priority; + u8 msb[3]; + u8 lsb[3]; + u8 mask[3]; + u8 shift[3]; +}; + +/* + * This is the map that ultimately indicates whether we'll be + * retrieving a register value or not, and at what frequency. + */ +static u8 asc7621_register_priorities[255]; + +static struct asc7621_data *asc7621_update_device(struct device *dev); + +static inline u8 read_byte(struct i2c_client *client, u8 reg) +{ + int res = i2c_smbus_read_byte_data(client, reg); + if (res < 0) { + dev_err(&client->dev, + "Unable to read from register 0x%02x.\n", reg); + return 0; + }; + return res & 0xff; +} + +static inline int write_byte(struct i2c_client *client, u8 reg, u8 data) +{ + int res = i2c_smbus_write_byte_data(client, reg, data); + if (res < 0) { + dev_err(&client->dev, + "Unable to write value 0x%02x to register 0x%02x.\n", + data, reg); + }; + return res; +} + +/* + * Data Handlers + * Each function handles the formatting, storage + * and retrieval of like parameters. + */ + +#define SETUP_SHOW_data_param(d, a) \ + struct sensor_device_attribute *sda = to_sensor_dev_attr(a); \ + struct asc7621_data *data = asc7621_update_device(d); \ + struct asc7621_param *param = to_asc7621_param(sda) + +#define SETUP_STORE_data_param(d, a) \ + struct sensor_device_attribute *sda = to_sensor_dev_attr(a); \ + struct i2c_client *client = to_i2c_client(d); \ + struct asc7621_data *data = i2c_get_clientdata(client); \ + struct asc7621_param *param = to_asc7621_param(sda) + +/* + * u8 is just what it sounds like...an unsigned byte with no + * special formatting. + */ +static ssize_t show_u8(struct device *dev, struct device_attribute *attr, + char *buf) +{ + SETUP_SHOW_data_param(dev, attr); + + return sprintf(buf, "%u\n", data->reg[param->msb[0]]); +} + +static ssize_t store_u8(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + SETUP_STORE_data_param(dev, attr); + long reqval; + + if (strict_strtol(buf, 10, &reqval)) + return -EINVAL; + + reqval = SENSORS_LIMIT(reqval, 0, 255); + + mutex_lock(&data->update_lock); + data->reg[param->msb[0]] = reqval; + write_byte(client, param->msb[0], reqval); + mutex_unlock(&data->update_lock); + return count; +} + +/* + * Many of the config values occupy only a few bits of a register. + */ +static ssize_t show_bitmask(struct device *dev, + struct device_attribute *attr, char *buf) +{ + SETUP_SHOW_data_param(dev, attr); + + return sprintf(buf, "%u\n", + (data->reg[param->msb[0]] >> param-> + shift[0]) & param->mask[0]); +} + +static ssize_t store_bitmask(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + SETUP_STORE_data_param(dev, attr); + long reqval; + u8 currval; + + if (strict_strtol(buf, 10, &reqval)) + return -EINVAL; + + reqval = SENSORS_LIMIT(reqval, 0, param->mask[0]); + + reqval = (reqval & param->mask[0]) << param->shift[0]; + + mutex_lock(&data->update_lock); + currval = read_byte(client, param->msb[0]); + reqval |= (currval & ~(param->mask[0] << param->shift[0])); + data->reg[param->msb[0]] = reqval; + write_byte(client, param->msb[0], reqval); + mutex_unlock(&data->update_lock); + return count; +} + +/* + * 16 bit fan rpm values + * reported by the device as the number of 11.111us periods (90khz) + * between full fan rotations. Therefore... + * RPM = (90000 * 60) / register value + */ +static ssize_t show_fan16(struct device *dev, + struct device_attribute *attr, char *buf) +{ + SETUP_SHOW_data_param(dev, attr); + u16 regval; + + mutex_lock(&data->update_lock); + regval = (data->reg[param->msb[0]] << 8) | data->reg[param->lsb[0]]; + mutex_unlock(&data->update_lock); + + return sprintf(buf, "%u\n", + (regval == 0 ? -1 : (regval) == + 0xffff ? 0 : 5400000 / regval)); +} + +static ssize_t store_fan16(struct device *dev, + struct device_attribute *attr, const char *buf, + size_t count) +{ + SETUP_STORE_data_param(dev, attr); + long reqval; + + if (strict_strtol(buf, 10, &reqval)) + return -EINVAL; + + reqval = + (SENSORS_LIMIT((reqval) <= 0 ? 0 : 5400000 / (reqval), 0, 65534)); + + mutex_lock(&data->update_lock); + data->reg[param->msb[0]] = (reqval >> 8) & 0xff; + data->reg[param->lsb[0]] = reqval & 0xff; + write_byte(client, param->msb[0], data->reg[param->msb[0]]); + write_byte(client, param->lsb[0], data->reg[param->lsb[0]]); + mutex_unlock(&data->update_lock); + + return count; +} + +/* + * Voltages are scaled in the device so that the nominal voltage + * is 3/4ths of the 0-255 range (i.e. 192). + * If all voltages are 'normal' then all voltage registers will + * read 0xC0. This doesn't help us if we don't have a point of refernce. + * The data sheet however provides us with the full scale value for each + * which is stored in in_scaling. The sda->index parameter value provides + * the index into in_scaling. + * + * NOTE: The chip expects the first 2 inputs be 2.5 and 2.25 volts + * respectively. That doesn't mean that's what the motherboard provides. :) + */ + +static int asc7621_in_scaling[] = { + 3320, 3000, 4380, 6640, 16000 +}; + +static ssize_t show_in10(struct device *dev, struct device_attribute *attr, + char *buf) +{ + SETUP_SHOW_data_param(dev, attr); + u16 regval; + u8 nr = sda->index; + + mutex_lock(&data->update_lock); + regval = (data->reg[param->msb[0]] * asc7621_in_scaling[nr]) / 256; + + /* The LSB value is a 2-bit scaling of the MSB's LSbit value. + * I.E. If the maximim voltage for this input is 6640 millivolts then + * a MSB register value of 0 = 0mv and 255 = 6640mv. + * A 1 step change therefore represents 25.9mv (6640 / 256). + * The extra 2-bits therefore represent increments of 6.48mv. + */ + regval += ((asc7621_in_scaling[nr] / 256) / 4) * + (data->reg[param->lsb[0]] >> 6); + + mutex_unlock(&data->update_lock); + + return sprintf(buf, "%u\n", regval); +} + +/* 8 bit voltage values (the mins and maxs) */ +static ssize_t show_in8(struct device *dev, struct device_attribute *attr, + char *buf) +{ + SETUP_SHOW_data_param(dev, attr); + u8 nr = sda->index; + + return sprintf(buf, "%u\n", + ((data->reg[param->msb[0]] * + asc7621_in_scaling[nr]) / 256)); +} + +static ssize_t store_in8(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + SETUP_STORE_data_param(dev, attr); + long reqval; + u8 nr = sda->index; + + if (strict_strtol(buf, 10, &reqval)) + return -EINVAL; + + reqval = SENSORS_LIMIT(reqval, 0, asc7621_in_scaling[nr]); + + reqval = (reqval * 255 + 128) / asc7621_in_scaling[nr]; + + mutex_lock(&data->update_lock); + data->reg[param->msb[0]] = reqval; + write_byte(client, param->msb[0], reqval); + mutex_unlock(&data->update_lock); + + return count; +} + +static ssize_t show_temp8(struct device *dev, + struct device_attribute *attr, char *buf) +{ + SETUP_SHOW_data_param(dev, attr); + + return sprintf(buf, "%d\n", ((s8) data->reg[param->msb[0]]) * 1000); +} + +static ssize_t store_temp8(struct device *dev, + struct device_attribute *attr, const char *buf, + size_t count) +{ + SETUP_STORE_data_param(dev, attr); + long reqval; + s8 temp; + + if (strict_strtol(buf, 10, &reqval)) + return -EINVAL; + + reqval = SENSORS_LIMIT(reqval, -127000, 127000); + + temp = reqval / 1000; + + mutex_lock(&data->update_lock); + data->reg[param->msb[0]] = temp; + write_byte(client, param->msb[0], temp); + mutex_unlock(&data->update_lock); + return count; +} + +/* + * Temperatures that occupy 2 bytes always have the whole + * number of degrees in the MSB with some part of the LSB + * indicating fractional degrees. + */ + +/* mmmmmmmm.llxxxxxx */ +static ssize_t show_temp10(struct device *dev, + struct device_attribute *attr, char *buf) +{ + SETUP_SHOW_data_param(dev, attr); + u8 msb, lsb; + int temp; + + mutex_lock(&data->update_lock); + msb = data->reg[param->msb[0]]; + lsb = (data->reg[param->lsb[0]] >> 6) & 0x03; + temp = (((s8) msb) * 1000) + (lsb * 250); + mutex_unlock(&data->update_lock); + + return sprintf(buf, "%d\n", temp); +} + +/* mmmmmm.ll */ +static ssize_t show_temp62(struct device *dev, + struct device_attribute *attr, char *buf) +{ + SETUP_SHOW_data_param(dev, attr); + u8 regval = data->reg[param->msb[0]]; + int temp = ((s8) (regval & 0xfc) * 1000) + ((regval & 0x03) * 250); + + return sprintf(buf, "%d\n", temp); +} + +static ssize_t store_temp62(struct device *dev, + struct device_attribute *attr, const char *buf, + size_t count) +{ + SETUP_STORE_data_param(dev, attr); + long reqval, i, f; + s8 temp; + + if (strict_strtol(buf, 10, &reqval)) + return -EINVAL; + + reqval = SENSORS_LIMIT(reqval, -32000, 31750); + i = reqval / 1000; + f = reqval - (i * 1000); + temp = i << 2; + temp |= f / 250; + + mutex_lock(&data->update_lock); + data->reg[param->msb[0]] = temp; + write_byte(client, param->msb[0], temp); + mutex_unlock(&data->update_lock); + return count; +} + +/* + * The aSC7621 doesn't provide an "auto_point2". Instead, you + * specify the auto_point1 and a range. To keep with the sysfs + * hwmon specs, we synthesize the auto_point_2 from them. + */ + +static u32 asc7621_range_map[] = { + 2000, 2500, 3330, 4000, 5000, 6670, 8000, 10000, + 13330, 16000, 20000, 26670, 32000, 40000, 53330, 80000, +}; + +static ssize_t show_ap2_temp(struct device *dev, + struct device_attribute *attr, char *buf) +{ + SETUP_SHOW_data_param(dev, attr); + long auto_point1; + u8 regval; + int temp; + + mutex_lock(&data->update_lock); + auto_point1 = ((s8) data->reg[param->msb[1]]) * 1000; + regval = + ((data->reg[param->msb[0]] >> param->shift[0]) & param->mask[0]); + temp = auto_point1 + asc7621_range_map[SENSORS_LIMIT(regval, 0, 15)]; + mutex_unlock(&data->update_lock); + + return sprintf(buf, "%d\n", temp); + +} + +static ssize_t store_ap2_temp(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + SETUP_STORE_data_param(dev, attr); + long reqval, auto_point1; + int i; + u8 currval, newval = 0; + + if (strict_strtol(buf, 10, &reqval)) + return -EINVAL; + + mutex_lock(&data->update_lock); + auto_point1 = data->reg[param->msb[1]] * 1000; + reqval = SENSORS_LIMIT(reqval, auto_point1 + 2000, auto_point1 + 80000); + + for (i = ARRAY_SIZE(asc7621_range_map) - 1; i >= 0; i--) { + if (reqval >= auto_point1 + asc7621_range_map[i]) { + newval = i; + break; + } + } + + newval = (newval & param->mask[0]) << param->shift[0]; + currval = read_byte(client, param->msb[0]); + newval |= (currval & ~(param->mask[0] << param->shift[0])); + data->reg[param->msb[0]] = newval; + write_byte(client, param->msb[0], newval); + mutex_unlock(&data->update_lock); + return count; +} + +static ssize_t show_pwm_ac(struct device *dev, + struct device_attribute *attr, char *buf) +{ + SETUP_SHOW_data_param(dev, attr); + u8 config, altbit, regval; + u8 map[] = { + 0x01, 0x02, 0x04, 0x1f, 0x00, 0x06, 0x07, 0x10, + 0x08, 0x0f, 0x1f, 0x1f, 0x1f, 0x1f, 0x1f, 0x1f + }; + + mutex_lock(&data->update_lock); + config = (data->reg[param->msb[0]] >> param->shift[0]) & param->mask[0]; + altbit = (data->reg[param->msb[1]] >> param->shift[1]) & param->mask[1]; + regval = config | (altbit << 3); + mutex_unlock(&data->update_lock); + + return sprintf(buf, "%u\n", map[SENSORS_LIMIT(regval, 0, 15)]); +} + +static ssize_t store_pwm_ac(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + SETUP_STORE_data_param(dev, attr); + unsigned long reqval; + u8 currval, config, altbit, newval; + u16 map[] = { + 0x04, 0x00, 0x01, 0xff, 0x02, 0xff, 0x05, 0x06, + 0x08, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x0f, + 0x07, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, + 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x03, + }; + + if (strict_strtoul(buf, 10, &reqval)) + return -EINVAL; + + if (reqval > 31) + return -EINVAL; + + reqval = map[reqval]; + if (reqval == 0xff) + return -EINVAL; + + config = reqval & 0x07; + altbit = (reqval >> 3) & 0x01; + + config = (config & param->mask[0]) << param->shift[0]; + altbit = (altbit & param->mask[1]) << param->shift[1]; + + mutex_lock(&data->update_lock); + currval = read_byte(client, param->msb[0]); + newval = config | (currval & ~(param->mask[0] << param->shift[0])); + newval = altbit | (newval & ~(param->mask[1] << param->shift[1])); + data->reg[param->msb[0]] = newval; + write_byte(client, param->msb[0], newval); + mutex_unlock(&data->update_lock); + return count; +} + +static ssize_t show_pwm_enable(struct device *dev, + struct device_attribute *attr, char *buf) +{ + SETUP_SHOW_data_param(dev, attr); + u8 config, altbit, minoff, val, newval; + + mutex_lock(&data->update_lock); + config = (data->reg[param->msb[0]] >> param->shift[0]) & param->mask[0]; + altbit = (data->reg[param->msb[1]] >> param->shift[1]) & param->mask[1]; + minoff = (data->reg[param->msb[2]] >> param->shift[2]) & param->mask[2]; + mutex_unlock(&data->update_lock); + + val = config | (altbit << 3); + newval = 0; + + if (val == 3 || val >= 10) + newval = 255; + else if (val == 4) + newval = 0; + else if (val == 7) + newval = 1; + else if (minoff == 1) + newval = 2; + else + newval = 3; + + return sprintf(buf, "%u\n", newval); +} + +static ssize_t store_pwm_enable(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + SETUP_STORE_data_param(dev, attr); + long reqval; + u8 currval, config, altbit, newval, minoff = 255; + + if (strict_strtol(buf, 10, &reqval)) + return -EINVAL; + + switch (reqval) { + case 0: + newval = 0x04; + break; + case 1: + newval = 0x07; + break; + case 2: + newval = 0x00; + minoff = 1; + break; + case 3: + newval = 0x00; + minoff = 0; + break; + case 255: + newval = 0x03; + break; + default: + return -EINVAL; + } + + config = newval & 0x07; + altbit = (newval >> 3) & 0x01; + + mutex_lock(&data->update_lock); + config = (config & param->mask[0]) << param->shift[0]; + altbit = (altbit & param->mask[1]) << param->shift[1]; + currval = read_byte(client, param->msb[0]); + newval = config | (currval & ~(param->mask[0] << param->shift[0])); + newval = altbit | (newval & ~(param->mask[1] << param->shift[1])); + data->reg[param->msb[0]] = newval; + write_byte(client, param->msb[0], newval); + if (minoff < 255) { + minoff = (minoff & param->mask[2]) << param->shift[2]; + currval = read_byte(client, param->msb[2]); + newval = + minoff | (currval & ~(param->mask[2] << param->shift[2])); + data->reg[param->msb[2]] = newval; + write_byte(client, param->msb[2], newval); + } + mutex_unlock(&data->update_lock); + return count; +} + +static u32 asc7621_pwm_freq_map[] = { + 10, 15, 23, 30, 38, 47, 62, 94, + 23000, 24000, 25000, 26000, 27000, 28000, 29000, 30000 +}; + +static ssize_t show_pwm_freq(struct device *dev, + struct device_attribute *attr, char *buf) +{ + SETUP_SHOW_data_param(dev, attr); + u8 regval = + (data->reg[param->msb[0]] >> param->shift[0]) & param->mask[0]; + + regval = SENSORS_LIMIT(regval, 0, 15); + + return sprintf(buf, "%u\n", asc7621_pwm_freq_map[regval]); +} + +static ssize_t store_pwm_freq(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + SETUP_STORE_data_param(dev, attr); + unsigned long reqval; + u8 currval, newval = 255; + int i; + + if (strict_strtoul(buf, 10, &reqval)) + return -EINVAL; + + for (i = 0; i < ARRAY_SIZE(asc7621_pwm_freq_map); i++) { + if (reqval == asc7621_pwm_freq_map[i]) { + newval = i; + break; + } + } + if (newval == 255) + return -EINVAL; + + newval = (newval & param->mask[0]) << param->shift[0]; + + mutex_lock(&data->update_lock); + currval = read_byte(client, param->msb[0]); + newval |= (currval & ~(param->mask[0] << param->shift[0])); + data->reg[param->msb[0]] = newval; + write_byte(client, param->msb[0], newval); + mutex_unlock(&data->update_lock); + return count; +} + +static u32 asc7621_pwm_auto_spinup_map[] = { + 0, 100, 250, 400, 700, 1000, 2000, 4000 +}; + +static ssize_t show_pwm_ast(struct device *dev, + struct device_attribute *attr, char *buf) +{ + SETUP_SHOW_data_param(dev, attr); + u8 regval = + (data->reg[param->msb[0]] >> param->shift[0]) & param->mask[0]; + + regval = SENSORS_LIMIT(regval, 0, 7); + + return sprintf(buf, "%u\n", asc7621_pwm_auto_spinup_map[regval]); + +} + +static ssize_t store_pwm_ast(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + SETUP_STORE_data_param(dev, attr); + long reqval; + u8 currval, newval = 255; + u32 i; + + if (strict_strtol(buf, 10, &reqval)) + return -EINVAL; + + for (i = 0; i < ARRAY_SIZE(asc7621_pwm_auto_spinup_map); i++) { + if (reqval == asc7621_pwm_auto_spinup_map[i]) { + newval = i; + break; + } + } + if (newval == 255) + return -EINVAL; + + newval = (newval & param->mask[0]) << param->shift[0]; + + mutex_lock(&data->update_lock); + currval = read_byte(client, param->msb[0]); + newval |= (currval & ~(param->mask[0] << param->shift[0])); + data->reg[param->msb[0]] = newval; + write_byte(client, param->msb[0], newval); + mutex_unlock(&data->update_lock); + return count; +} + +static u32 asc7621_temp_smoothing_time_map[] = { + 35000, 17600, 11800, 7000, 4400, 3000, 1600, 800 +}; + +static ssize_t show_temp_st(struct device *dev, + struct device_attribute *attr, char *buf) +{ + SETUP_SHOW_data_param(dev, attr); + u8 regval = + (data->reg[param->msb[0]] >> param->shift[0]) & param->mask[0]; + regval = SENSORS_LIMIT(regval, 0, 7); + + return sprintf(buf, "%u\n", asc7621_temp_smoothing_time_map[regval]); +} + +static ssize_t store_temp_st(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + SETUP_STORE_data_param(dev, attr); + long reqval; + u8 currval, newval = 255; + u32 i; + + if (strict_strtol(buf, 10, &reqval)) + return -EINVAL; + + for (i = 0; i < ARRAY_SIZE(asc7621_temp_smoothing_time_map); i++) { + if (reqval == asc7621_temp_smoothing_time_map[i]) { + newval = i; + break; + } + } + + if (newval == 255) + return -EINVAL; + + newval = (newval & param->mask[0]) << param->shift[0]; + + mutex_lock(&data->update_lock); + currval = read_byte(client, param->msb[0]); + newval |= (currval & ~(param->mask[0] << param->shift[0])); + data->reg[param->msb[0]] = newval; + write_byte(client, param->msb[0], newval); + mutex_unlock(&data->update_lock); + return count; +} + +/* + * End of data handlers + * + * These defines do nothing more than make the table easier + * to read when wrapped at column 80. + */ + +/* + * Creates a variable length array inititalizer. + * VAA(1,3,5,7) would produce {1,3,5,7} + */ +#define VAA(args...) {args} + +#define PREAD(name, n, pri, rm, rl, m, s, r) \ + {.sda = SENSOR_ATTR(name, S_IRUGO, show_##r, NULL, n), \ + .priority = pri, .msb[0] = rm, .lsb[0] = rl, .mask[0] = m, \ + .shift[0] = s,} + +#define PWRITE(name, n, pri, rm, rl, m, s, r) \ + {.sda = SENSOR_ATTR(name, S_IRUGO | S_IWUSR, show_##r, store_##r, n), \ + .priority = pri, .msb[0] = rm, .lsb[0] = rl, .mask[0] = m, \ + .shift[0] = s,} + +/* + * PWRITEM assumes that the initializers for the .msb, .lsb, .mask and .shift + * were created using the VAA macro. + */ +#define PWRITEM(name, n, pri, rm, rl, m, s, r) \ + {.sda = SENSOR_ATTR(name, S_IRUGO | S_IWUSR, show_##r, store_##r, n), \ + .priority = pri, .msb = rm, .lsb = rl, .mask = m, .shift = s,} + +static struct asc7621_param asc7621_params[] = { + PREAD(in0_input, 0, PRI_HIGH, 0x20, 0x13, 0, 0, in10), + PREAD(in1_input, 1, PRI_HIGH, 0x21, 0x18, 0, 0, in10), + PREAD(in2_input, 2, PRI_HIGH, 0x22, 0x11, 0, 0, in10), + PREAD(in3_input, 3, PRI_HIGH, 0x23, 0x12, 0, 0, in10), + PREAD(in4_input, 4, PRI_HIGH, 0x24, 0x14, 0, 0, in10), + + PWRITE(in0_min, 0, PRI_LOW, 0x44, 0, 0, 0, in8), + PWRITE(in1_min, 1, PRI_LOW, 0x46, 0, 0, 0, in8), + PWRITE(in2_min, 2, PRI_LOW, 0x48, 0, 0, 0, in8), + PWRITE(in3_min, 3, PRI_LOW, 0x4a, 0, 0, 0, in8), + PWRITE(in4_min, 4, PRI_LOW, 0x4c, 0, 0, 0, in8), + + PWRITE(in0_max, 0, PRI_LOW, 0x45, 0, 0, 0, in8), + PWRITE(in1_max, 1, PRI_LOW, 0x47, 0, 0, 0, in8), + PWRITE(in2_max, 2, PRI_LOW, 0x49, 0, 0, 0, in8), + PWRITE(in3_max, 3, PRI_LOW, 0x4b, 0, 0, 0, in8), + PWRITE(in4_max, 4, PRI_LOW, 0x4d, 0, 0, 0, in8), + + PREAD(in0_alarm, 0, PRI_LOW, 0x41, 0, 0x01, 0, bitmask), + PREAD(in1_alarm, 1, PRI_LOW, 0x41, 0, 0x01, 1, bitmask), + PREAD(in2_alarm, 2, PRI_LOW, 0x41, 0, 0x01, 2, bitmask), + PREAD(in3_alarm, 3, PRI_LOW, 0x41, 0, 0x01, 3, bitmask), + PREAD(in4_alarm, 4, PRI_LOW, 0x42, 0, 0x01, 0, bitmask), + + PREAD(fan1_input, 0, PRI_HIGH, 0x29, 0x28, 0, 0, fan16), + PREAD(fan2_input, 1, PRI_HIGH, 0x2b, 0x2a, 0, 0, fan16), + PREAD(fan3_input, 2, PRI_HIGH, 0x2d, 0x2c, 0, 0, fan16), + PREAD(fan4_input, 3, PRI_HIGH, 0x2f, 0x2e, 0, 0, fan16), + + PWRITE(fan1_min, 0, PRI_LOW, 0x55, 0x54, 0, 0, fan16), + PWRITE(fan2_min, 1, PRI_LOW, 0x57, 0x56, 0, 0, fan16), + PWRITE(fan3_min, 2, PRI_LOW, 0x59, 0x58, 0, 0, fan16), + PWRITE(fan4_min, 3, PRI_LOW, 0x5b, 0x5a, 0, 0, fan16), + + PREAD(fan1_alarm, 0, PRI_LOW, 0x42, 0, 0x01, 0, bitmask), + PREAD(fan2_alarm, 1, PRI_LOW, 0x42, 0, 0x01, 1, bitmask), + PREAD(fan3_alarm, 2, PRI_LOW, 0x42, 0, 0x01, 2, bitmask), + PREAD(fan4_alarm, 3, PRI_LOW, 0x42, 0, 0x01, 3, bitmask), + + PREAD(temp1_input, 0, PRI_HIGH, 0x25, 0x10, 0, 0, temp10), + PREAD(temp2_input, 1, PRI_HIGH, 0x26, 0x15, 0, 0, temp10), + PREAD(temp3_input, 2, PRI_HIGH, 0x27, 0x16, 0, 0, temp10), + PREAD(temp4_input, 3, PRI_HIGH, 0x33, 0x17, 0, 0, temp10), + PREAD(temp5_input, 4, PRI_HIGH, 0xf7, 0xf6, 0, 0, temp10), + PREAD(temp6_input, 5, PRI_HIGH, 0xf9, 0xf8, 0, 0, temp10), + PREAD(temp7_input, 6, PRI_HIGH, 0xfb, 0xfa, 0, 0, temp10), + PREAD(temp8_input, 7, PRI_HIGH, 0xfd, 0xfc, 0, 0, temp10), + + PWRITE(temp1_min, 0, PRI_LOW, 0x4e, 0, 0, 0, temp8), + PWRITE(temp2_min, 1, PRI_LOW, 0x50, 0, 0, 0, temp8), + PWRITE(temp3_min, 2, PRI_LOW, 0x52, 0, 0, 0, temp8), + PWRITE(temp4_min, 3, PRI_LOW, 0x34, 0, 0, 0, temp8), + + PWRITE(temp1_max, 0, PRI_LOW, 0x4f, 0, 0, 0, temp8), + PWRITE(temp2_max, 1, PRI_LOW, 0x51, 0, 0, 0, temp8), + PWRITE(temp3_max, 2, PRI_LOW, 0x53, 0, 0, 0, temp8), + PWRITE(temp4_max, 3, PRI_LOW, 0x35, 0, 0, 0, temp8), + + PREAD(temp1_alarm, 0, PRI_LOW, 0x41, 0, 0x01, 4, bitmask), + PREAD(temp2_alarm, 1, PRI_LOW, 0x41, 0, 0x01, 5, bitmask), + PREAD(temp3_alarm, 2, PRI_LOW, 0x41, 0, 0x01, 6, bitmask), + PREAD(temp4_alarm, 3, PRI_LOW, 0x43, 0, 0x01, 0, bitmask), + + PWRITE(temp1_source, 0, PRI_LOW, 0x02, 0, 0x07, 4, bitmask), + PWRITE(temp2_source, 1, PRI_LOW, 0x02, 0, 0x07, 0, bitmask), + PWRITE(temp3_source, 2, PRI_LOW, 0x03, 0, 0x07, 4, bitmask), + PWRITE(temp4_source, 3, PRI_LOW, 0x03, 0, 0x07, 0, bitmask), + + PWRITE(temp1_smoothing_enable, 0, PRI_LOW, 0x62, 0, 0x01, 3, bitmask), + PWRITE(temp2_smoothing_enable, 1, PRI_LOW, 0x63, 0, 0x01, 7, bitmask), + PWRITE(temp3_smoothing_enable, 2, PRI_LOW, 0x64, 0, 0x01, 3, bitmask), + PWRITE(temp4_smoothing_enable, 3, PRI_LOW, 0x3c, 0, 0x01, 3, bitmask), + + PWRITE(temp1_smoothing_time, 0, PRI_LOW, 0x62, 0, 0x07, 0, temp_st), + PWRITE(temp2_smoothing_time, 1, PRI_LOW, 0x63, 0, 0x07, 4, temp_st), + PWRITE(temp3_smoothing_time, 2, PRI_LOW, 0x63, 0, 0x07, 0, temp_st), + PWRITE(temp4_smoothing_time, 3, PRI_LOW, 0x3c, 0, 0x07, 0, temp_st), + + PWRITE(temp1_auto_point1_temp_hyst, 0, PRI_LOW, 0x6d, 0, 0x0f, 4, + bitmask), + PWRITE(temp2_auto_point1_temp_hyst, 1, PRI_LOW, 0x6d, 0, 0x0f, 0, + bitmask), + PWRITE(temp3_auto_point1_temp_hyst, 2, PRI_LOW, 0x6e, 0, 0x0f, 4, + bitmask), + PWRITE(temp4_auto_point1_temp_hyst, 3, PRI_LOW, 0x6e, 0, 0x0f, 0, + bitmask), + + PREAD(temp1_auto_point2_temp_hyst, 0, PRI_LOW, 0x6d, 0, 0x0f, 4, + bitmask), + PREAD(temp2_auto_point2_temp_hyst, 1, PRI_LOW, 0x6d, 0, 0x0f, 0, + bitmask), + PREAD(temp3_auto_point2_temp_hyst, 2, PRI_LOW, 0x6e, 0, 0x0f, 4, + bitmask), + PREAD(temp4_auto_point2_temp_hyst, 3, PRI_LOW, 0x6e, 0, 0x0f, 0, + bitmask), + + PWRITE(temp1_auto_point1_temp, 0, PRI_LOW, 0x67, 0, 0, 0, temp8), + PWRITE(temp2_auto_point1_temp, 1, PRI_LOW, 0x68, 0, 0, 0, temp8), + PWRITE(temp3_auto_point1_temp, 2, PRI_LOW, 0x69, 0, 0, 0, temp8), + PWRITE(temp4_auto_point1_temp, 3, PRI_LOW, 0x3b, 0, 0, 0, temp8), + + PWRITEM(temp1_auto_point2_temp, 0, PRI_LOW, VAA(0x5f, 0x67), VAA(0), + VAA(0x0f), VAA(4), ap2_temp), + PWRITEM(temp2_auto_point2_temp, 1, PRI_LOW, VAA(0x60, 0x68), VAA(0), + VAA(0x0f), VAA(4), ap2_temp), + PWRITEM(temp3_auto_point2_temp, 2, PRI_LOW, VAA(0x61, 0x69), VAA(0), + VAA(0x0f), VAA(4), ap2_temp), + PWRITEM(temp4_auto_point2_temp, 3, PRI_LOW, VAA(0x3c, 0x3b), VAA(0), + VAA(0x0f), VAA(4), ap2_temp), + + PWRITE(temp1_crit, 0, PRI_LOW, 0x6a, 0, 0, 0, temp8), + PWRITE(temp2_crit, 1, PRI_LOW, 0x6b, 0, 0, 0, temp8), + PWRITE(temp3_crit, 2, PRI_LOW, 0x6c, 0, 0, 0, temp8), + PWRITE(temp4_crit, 3, PRI_LOW, 0x3d, 0, 0, 0, temp8), + + PWRITE(temp5_enable, 4, PRI_LOW, 0x0e, 0, 0x01, 0, bitmask), + PWRITE(temp6_enable, 5, PRI_LOW, 0x0e, 0, 0x01, 1, bitmask), + PWRITE(temp7_enable, 6, PRI_LOW, 0x0e, 0, 0x01, 2, bitmask), + PWRITE(temp8_enable, 7, PRI_LOW, 0x0e, 0, 0x01, 3, bitmask), + + PWRITE(remote1_offset, 0, PRI_LOW, 0x1c, 0, 0, 0, temp62), + PWRITE(remote2_offset, 1, PRI_LOW, 0x1d, 0, 0, 0, temp62), + + PWRITE(pwm1, 0, PRI_HIGH, 0x30, 0, 0, 0, u8), + PWRITE(pwm2, 1, PRI_HIGH, 0x31, 0, 0, 0, u8), + PWRITE(pwm3, 2, PRI_HIGH, 0x32, 0, 0, 0, u8), + + PWRITE(pwm1_invert, 0, PRI_LOW, 0x5c, 0, 0x01, 4, bitmask), + PWRITE(pwm2_invert, 1, PRI_LOW, 0x5d, 0, 0x01, 4, bitmask), + PWRITE(pwm3_invert, 2, PRI_LOW, 0x5e, 0, 0x01, 4, bitmask), + + PWRITEM(pwm1_enable, 0, PRI_LOW, VAA(0x5c, 0x5c, 0x62), VAA(0, 0, 0), + VAA(0x07, 0x01, 0x01), VAA(5, 3, 5), pwm_enable), + PWRITEM(pwm2_enable, 1, PRI_LOW, VAA(0x5d, 0x5d, 0x62), VAA(0, 0, 0), + VAA(0x07, 0x01, 0x01), VAA(5, 3, 6), pwm_enable), + PWRITEM(pwm3_enable, 2, PRI_LOW, VAA(0x5e, 0x5e, 0x62), VAA(0, 0, 0), + VAA(0x07, 0x01, 0x01), VAA(5, 3, 7), pwm_enable), + + PWRITEM(pwm1_auto_channels, 0, PRI_LOW, VAA(0x5c, 0x5c), VAA(0, 0), + VAA(0x07, 0x01), VAA(5, 3), pwm_ac), + PWRITEM(pwm2_auto_channels, 1, PRI_LOW, VAA(0x5d, 0x5d), VAA(0, 0), + VAA(0x07, 0x01), VAA(5, 3), pwm_ac), + PWRITEM(pwm3_auto_channels, 2, PRI_LOW, VAA(0x5e, 0x5e), VAA(0, 0), + VAA(0x07, 0x01), VAA(5, 3), pwm_ac), + + PWRITE(pwm1_auto_point1_pwm, 0, PRI_LOW, 0x64, 0, 0, 0, u8), + PWRITE(pwm2_auto_point1_pwm, 1, PRI_LOW, 0x65, 0, 0, 0, u8), + PWRITE(pwm3_auto_point1_pwm, 2, PRI_LOW, 0x66, 0, 0, 0, u8), + + PWRITE(pwm1_auto_point2_pwm, 0, PRI_LOW, 0x38, 0, 0, 0, u8), + PWRITE(pwm2_auto_point2_pwm, 1, PRI_LOW, 0x39, 0, 0, 0, u8), + PWRITE(pwm3_auto_point2_pwm, 2, PRI_LOW, 0x3a, 0, 0, 0, u8), + + PWRITE(pwm1_freq, 0, PRI_LOW, 0x5f, 0, 0x0f, 0, pwm_freq), + PWRITE(pwm2_freq, 1, PRI_LOW, 0x60, 0, 0x0f, 0, pwm_freq), + PWRITE(pwm3_freq, 2, PRI_LOW, 0x61, 0, 0x0f, 0, pwm_freq), + + PREAD(pwm1_auto_zone_assigned, 0, PRI_LOW, 0, 0, 0x03, 2, bitmask), + PREAD(pwm2_auto_zone_assigned, 1, PRI_LOW, 0, 0, 0x03, 4, bitmask), + PREAD(pwm3_auto_zone_assigned, 2, PRI_LOW, 0, 0, 0x03, 6, bitmask), + + PWRITE(pwm1_auto_spinup_time, 0, PRI_LOW, 0x5c, 0, 0x07, 0, pwm_ast), + PWRITE(pwm2_auto_spinup_time, 1, PRI_LOW, 0x5d, 0, 0x07, 0, pwm_ast), + PWRITE(pwm3_auto_spinup_time, 2, PRI_LOW, 0x5e, 0, 0x07, 0, pwm_ast), + + PWRITE(peci_enable, 0, PRI_LOW, 0x40, 0, 0x01, 4, bitmask), + PWRITE(peci_avg, 0, PRI_LOW, 0x36, 0, 0x07, 0, bitmask), + PWRITE(peci_domain, 0, PRI_LOW, 0x36, 0, 0x01, 3, bitmask), + PWRITE(peci_legacy, 0, PRI_LOW, 0x36, 0, 0x01, 4, bitmask), + PWRITE(peci_diode, 0, PRI_LOW, 0x0e, 0, 0x07, 4, bitmask), + PWRITE(peci_4domain, 0, PRI_LOW, 0x0e, 0, 0x01, 4, bitmask), + +}; + +static struct asc7621_data *asc7621_update_device(struct device *dev) +{ + struct i2c_client *client = to_i2c_client(dev); + struct asc7621_data *data = i2c_get_clientdata(client); + int i; + +/* + * The asc7621 chips guarantee consistent reads of multi-byte values + * regardless of the order of the reads. No special logic is needed + * so we can just read the registers in whatever order they appear + * in the asc7621_params array. + */ + + mutex_lock(&data->update_lock); + + /* Read all the high priority registers */ + + if (!data->valid || + time_after(jiffies, data->last_high_reading + INTERVAL_HIGH)) { + + for (i = 0; i < ARRAY_SIZE(asc7621_register_priorities); i++) { + if (asc7621_register_priorities[i] == PRI_HIGH) { + data->reg[i] = + i2c_smbus_read_byte_data(client, i) & 0xff; + } + } + data->last_high_reading = jiffies; + }; /* last_reading */ + + /* Read all the low priority registers. */ + + if (!data->valid || + time_after(jiffies, data->last_low_reading + INTERVAL_LOW)) { + + for (i = 0; i < ARRAY_SIZE(asc7621_params); i++) { + if (asc7621_register_priorities[i] == PRI_LOW) { + data->reg[i] = + i2c_smbus_read_byte_data(client, i) & 0xff; + } + } + data->last_low_reading = jiffies; + }; /* last_reading */ + + data->valid = 1; + + mutex_unlock(&data->update_lock); + + return data; +} + +/* + * Standard detection and initialization below + * + * Helper function that checks if an address is valid + * for a particular chip. + */ + +static inline int valid_address_for_chip(int chip_type, int address) +{ + int i; + + for (i = 0; asc7621_chips[chip_type].addresses[i] != I2C_CLIENT_END; + i++) { + if (asc7621_chips[chip_type].addresses[i] == address) + return 1; + } + return 0; +} + +static void asc7621_init_client(struct i2c_client *client) +{ + int value; + + /* Warn if part was not "READY" */ + + value = read_byte(client, 0x40); + + if (value & 0x02) { + dev_err(&client->dev, + "Client (%d,0x%02x) config is locked.\n", + i2c_adapter_id(client->adapter), client->addr); + }; + if (!(value & 0x04)) { + dev_err(&client->dev, "Client (%d,0x%02x) is not ready.\n", + i2c_adapter_id(client->adapter), client->addr); + }; + +/* + * Start monitoring + * + * Try to clear LOCK, Set START, save everything else + */ + value = (value & ~0x02) | 0x01; + write_byte(client, 0x40, value & 0xff); + +} + +static int +asc7621_probe(struct i2c_client *client, const struct i2c_device_id *id) +{ + struct asc7621_data *data; + int i, err; + + if (!i2c_check_functionality(client->adapter, I2C_FUNC_SMBUS_BYTE_DATA)) + return -EIO; + + data = kzalloc(sizeof(struct asc7621_data), GFP_KERNEL); + if (data == NULL) + return -ENOMEM; + + i2c_set_clientdata(client, data); + data->valid = 0; + mutex_init(&data->update_lock); + + /* Initialize the asc7621 chip */ + asc7621_init_client(client); + + /* Create the sysfs entries */ + for (i = 0; i < ARRAY_SIZE(asc7621_params); i++) { + err = + device_create_file(&client->dev, + &(asc7621_params[i].sda.dev_attr)); + if (err) + goto exit_remove; + } + + data->class_dev = hwmon_device_register(&client->dev); + if (IS_ERR(data->class_dev)) { + err = PTR_ERR(data->class_dev); + goto exit_remove; + } + + return 0; + +exit_remove: + for (i = 0; i < ARRAY_SIZE(asc7621_params); i++) { + device_remove_file(&client->dev, + &(asc7621_params[i].sda.dev_attr)); + } + + i2c_set_clientdata(client, NULL); + kfree(data); + return err; +} + +static int asc7621_detect(struct i2c_client *client, + struct i2c_board_info *info) +{ + struct i2c_adapter *adapter = client->adapter; + int company, verstep, chip_index; + struct device *dev; + + dev = &client->dev; + + if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_BYTE_DATA)) + return -ENODEV; + + for (chip_index = FIRST_CHIP; chip_index <= LAST_CHIP; chip_index++) { + + if (!valid_address_for_chip(chip_index, client->addr)) + continue; + + company = read_byte(client, + asc7621_chips[chip_index].company_reg); + verstep = read_byte(client, + asc7621_chips[chip_index].verstep_reg); + + if (company == asc7621_chips[chip_index].company_id && + verstep == asc7621_chips[chip_index].verstep_id) { + strlcpy(client->name, asc7621_chips[chip_index].name, + I2C_NAME_SIZE); + strlcpy(info->type, asc7621_chips[chip_index].name, + I2C_NAME_SIZE); + + dev_info(&adapter->dev, "Matched %s\n", + asc7621_chips[chip_index].name); + return 0; + } + } + + return -ENODEV; +} + +static int asc7621_remove(struct i2c_client *client) +{ + struct asc7621_data *data = i2c_get_clientdata(client); + int i; + + hwmon_device_unregister(data->class_dev); + + for (i = 0; i < ARRAY_SIZE(asc7621_params); i++) { + device_remove_file(&client->dev, + &(asc7621_params[i].sda.dev_attr)); + } + + i2c_set_clientdata(client, NULL); + kfree(data); + return 0; +} + +static const struct i2c_device_id asc7621_id[] = { + {"asc7621", asc7621}, + {"asc7621a", asc7621a}, + {}, +}; + +MODULE_DEVICE_TABLE(i2c, asc7621_id); + +static struct i2c_driver asc7621_driver = { + .class = I2C_CLASS_HWMON, + .driver = { + .name = "asc7621", + }, + .probe = asc7621_probe, + .remove = asc7621_remove, + .id_table = asc7621_id, + .detect = asc7621_detect, + .address_list = normal_i2c, +}; + +static int __init sm_asc7621_init(void) +{ + int i, j; +/* + * Collect all the registers needed into a single array. + * This way, if a register isn't actually used for anything, + * we don't retrieve it. + */ + + for (i = 0; i < ARRAY_SIZE(asc7621_params); i++) { + for (j = 0; j < ARRAY_SIZE(asc7621_params[i].msb); j++) + asc7621_register_priorities[asc7621_params[i].msb[j]] = + asc7621_params[i].priority; + for (j = 0; j < ARRAY_SIZE(asc7621_params[i].lsb); j++) + asc7621_register_priorities[asc7621_params[i].lsb[j]] = + asc7621_params[i].priority; + } + return i2c_add_driver(&asc7621_driver); +} + +static void __exit sm_asc7621_exit(void) +{ + i2c_del_driver(&asc7621_driver); +} + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("George Joseph"); +MODULE_DESCRIPTION("Andigilog aSC7621 and aSC7621a driver"); + +module_init(sm_asc7621_init); +module_exit(sm_asc7621_exit); diff --git a/drivers/hwmon/fschmd.c b/drivers/hwmon/fschmd.c index fa0728232e71..0627f7a5b9b8 100644 --- a/drivers/hwmon/fschmd.c +++ b/drivers/hwmon/fschmd.c @@ -267,7 +267,7 @@ struct fschmd_data { struct list_head list; /* member of the watchdog_data_list */ struct kref kref; struct miscdevice watchdog_miscdev; - int kind; + enum chips kind; unsigned long watchdog_is_open; char watchdog_expect_close; char watchdog_name[10]; /* must be unique to avoid sysfs conflict */ @@ -325,8 +325,7 @@ static ssize_t show_in_value(struct device *dev, int index = to_sensor_dev_attr(devattr)->index; struct fschmd_data *data = fschmd_update_device(dev); - /* fscher / fschrc - 1 as data->kind is an array index, not a chips */ - if (data->kind == (fscher - 1) || data->kind >= (fschrc - 1)) + if (data->kind == fscher || data->kind >= fschrc) return sprintf(buf, "%d\n", (data->volt[index] * dmi_vref * dmi_mult[index]) / 255 + dmi_offset[index]); else @@ -492,7 +491,7 @@ static ssize_t show_pwm_auto_point1_pwm(struct device *dev, int val = data->fan_min[index]; /* 0 = allow turning off (except on the syl), 1-255 = 50-100% */ - if (val || data->kind == fscsyl - 1) + if (val || data->kind == fscsyl) val = val / 2 + 128; return sprintf(buf, "%d\n", val); @@ -506,7 +505,7 @@ static ssize_t store_pwm_auto_point1_pwm(struct device *dev, unsigned long v = simple_strtoul(buf, NULL, 10); /* reg: 0 = allow turning off (except on the syl), 1-255 = 50-100% */ - if (v || data->kind == fscsyl - 1) { + if (v || data->kind == fscsyl) { v = SENSORS_LIMIT(v, 128, 255); v = (v - 128) * 2 + 1; } @@ -1037,7 +1036,7 @@ static int fschmd_detect(struct i2c_client *client, else return -ENODEV; - strlcpy(info->type, fschmd_id[kind - 1].name, I2C_NAME_SIZE); + strlcpy(info->type, fschmd_id[kind].name, I2C_NAME_SIZE); return 0; } @@ -1065,6 +1064,7 @@ static int fschmd_probe(struct i2c_client *client, (where the client is found through a data ptr instead of the otherway around) */ data->client = client; + data->kind = kind; if (kind == fscpos) { /* The Poseidon has hardwired temp limits, fill these @@ -1085,9 +1085,6 @@ static int fschmd_probe(struct i2c_client *client, } } - /* i2c kind goes from 1-6, we want from 0-5 to address arrays */ - data->kind = kind - 1; - /* Read in some never changing registers */ data->revision = i2c_smbus_read_byte_data(client, FSCHMD_REG_REVISION); data->global_control = i2c_smbus_read_byte_data(client, diff --git a/drivers/hwmon/g760a.c b/drivers/hwmon/g760a.c index 19c01a49f6be..09ea12e0a551 100644 --- a/drivers/hwmon/g760a.c +++ b/drivers/hwmon/g760a.c @@ -68,7 +68,7 @@ struct g760a_data { #define PWM_FROM_CNT(cnt) (0xff-(cnt)) #define PWM_TO_CNT(pwm) (0xff-(pwm)) -unsigned int rpm_from_cnt(u8 val, u32 clk, u16 div) +static inline unsigned int rpm_from_cnt(u8 val, u32 clk, u16 div) { return ((val == 0x00) ? 0 : ((clk*30)/(val*div))); } diff --git a/drivers/hwmon/it87.c b/drivers/hwmon/it87.c index 0ffe84d190bb..1002befd87d5 100644 --- a/drivers/hwmon/it87.c +++ b/drivers/hwmon/it87.c @@ -1,40 +1,40 @@ /* - it87.c - Part of lm_sensors, Linux kernel modules for hardware - monitoring. - - The IT8705F is an LPC-based Super I/O part that contains UARTs, a - parallel port, an IR port, a MIDI port, a floppy controller, etc., in - addition to an Environment Controller (Enhanced Hardware Monitor and - Fan Controller) - - This driver supports only the Environment Controller in the IT8705F and - similar parts. The other devices are supported by different drivers. - - Supports: IT8705F Super I/O chip w/LPC interface - IT8712F Super I/O chip w/LPC interface - IT8716F Super I/O chip w/LPC interface - IT8718F Super I/O chip w/LPC interface - IT8720F Super I/O chip w/LPC interface - IT8726F Super I/O chip w/LPC interface - Sis950 A clone of the IT8705F - - Copyright (C) 2001 Chris Gauthron - Copyright (C) 2005-2007 Jean Delvare <khali@linux-fr.org> - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2 of the License, or - (at your option) any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; if not, write to the Free Software - Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. -*/ + * it87.c - Part of lm_sensors, Linux kernel modules for hardware + * monitoring. + * + * The IT8705F is an LPC-based Super I/O part that contains UARTs, a + * parallel port, an IR port, a MIDI port, a floppy controller, etc., in + * addition to an Environment Controller (Enhanced Hardware Monitor and + * Fan Controller) + * + * This driver supports only the Environment Controller in the IT8705F and + * similar parts. The other devices are supported by different drivers. + * + * Supports: IT8705F Super I/O chip w/LPC interface + * IT8712F Super I/O chip w/LPC interface + * IT8716F Super I/O chip w/LPC interface + * IT8718F Super I/O chip w/LPC interface + * IT8720F Super I/O chip w/LPC interface + * IT8726F Super I/O chip w/LPC interface + * Sis950 A clone of the IT8705F + * + * Copyright (C) 2001 Chris Gauthron + * Copyright (C) 2005-2010 Jean Delvare <khali@linux-fr.org> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + */ #include <linux/module.h> #include <linux/init.h> @@ -128,6 +128,7 @@ superio_exit(void) #define IT87_SIO_GPIO5_REG 0x29 #define IT87_SIO_PINX2_REG 0x2c /* Pin selection */ #define IT87_SIO_VID_REG 0xfc /* VID value */ +#define IT87_SIO_BEEP_PIN_REG 0xf6 /* Beep pin mapping */ /* Update battery voltage after every reading if true */ static int update_vbat; @@ -187,9 +188,13 @@ static const u8 IT87_REG_FANX_MIN[] = { 0x1b, 0x1c, 0x1d, 0x85, 0x87 }; #define IT87_REG_VIN_ENABLE 0x50 #define IT87_REG_TEMP_ENABLE 0x51 +#define IT87_REG_BEEP_ENABLE 0x5c #define IT87_REG_CHIPID 0x58 +#define IT87_REG_AUTO_TEMP(nr, i) (0x60 + (nr) * 8 + (i)) +#define IT87_REG_AUTO_PWM(nr, i) (0x65 + (nr) * 8 + (i)) + #define IN_TO_REG(val) (SENSORS_LIMIT((((val) + 8)/16),0,255)) #define IN_FROM_REG(val) ((val) * 16) @@ -246,6 +251,7 @@ struct it87_sio_data { /* Values read from Super-I/O config space */ u8 revision; u8 vid_value; + u8 beep_pin; /* Features skipped based on config or DMI */ u8 skip_vid; u8 skip_fan; @@ -279,9 +285,21 @@ struct it87_data { u8 vid; /* Register encoding, combined */ u8 vrm; u32 alarms; /* Register encoding, combined */ + u8 beeps; /* Register encoding */ u8 fan_main_ctrl; /* Register value */ u8 fan_ctl; /* Register value */ - u8 manual_pwm_ctl[3]; /* manual PWM value set by user */ + + /* The following 3 arrays correspond to the same registers. The + * meaning of bits 6-0 depends on the value of bit 7, and we want + * to preserve settings on mode changes, so we have to track all + * values separately. */ + u8 pwm_ctrl[3]; /* Register value */ + u8 pwm_duty[3]; /* Manual PWM value set by user (bit 6-0) */ + u8 pwm_temp_map[3]; /* PWM to temp. chan. mapping (bits 1-0) */ + + /* Automatic fan speed control registers */ + u8 auto_pwm[3][4]; /* [nr][3] is hard-coded */ + s8 auto_temp[3][5]; /* [nr][0] is point1_temp_hyst */ }; static inline int has_16bit_fans(const struct it87_data *data) @@ -296,6 +314,15 @@ static inline int has_16bit_fans(const struct it87_data *data) || data->type == it8720; } +static inline int has_old_autopwm(const struct it87_data *data) +{ + /* The old automatic fan speed control interface is implemented + by IT8705F chips up to revision F and IT8712F chips up to + revision G. */ + return (data->type == it87 && data->revision < 0x03) + || (data->type == it8712 && data->revision < 0x08); +} + static int it87_probe(struct platform_device *pdev); static int __devexit it87_remove(struct platform_device *pdev); @@ -352,7 +379,10 @@ static ssize_t set_in_min(struct device *dev, struct device_attribute *attr, int nr = sensor_attr->index; struct it87_data *data = dev_get_drvdata(dev); - unsigned long val = simple_strtoul(buf, NULL, 10); + unsigned long val; + + if (strict_strtoul(buf, 10, &val) < 0) + return -EINVAL; mutex_lock(&data->update_lock); data->in_min[nr] = IN_TO_REG(val); @@ -368,7 +398,10 @@ static ssize_t set_in_max(struct device *dev, struct device_attribute *attr, int nr = sensor_attr->index; struct it87_data *data = dev_get_drvdata(dev); - unsigned long val = simple_strtoul(buf, NULL, 10); + unsigned long val; + + if (strict_strtoul(buf, 10, &val) < 0) + return -EINVAL; mutex_lock(&data->update_lock); data->in_max[nr] = IN_TO_REG(val); @@ -441,7 +474,10 @@ static ssize_t set_temp_max(struct device *dev, struct device_attribute *attr, int nr = sensor_attr->index; struct it87_data *data = dev_get_drvdata(dev); - int val = simple_strtol(buf, NULL, 10); + long val; + + if (strict_strtol(buf, 10, &val) < 0) + return -EINVAL; mutex_lock(&data->update_lock); data->temp_high[nr] = TEMP_TO_REG(val); @@ -456,7 +492,10 @@ static ssize_t set_temp_min(struct device *dev, struct device_attribute *attr, int nr = sensor_attr->index; struct it87_data *data = dev_get_drvdata(dev); - int val = simple_strtol(buf, NULL, 10); + long val; + + if (strict_strtol(buf, 10, &val) < 0) + return -EINVAL; mutex_lock(&data->update_lock); data->temp_low[nr] = TEMP_TO_REG(val); @@ -483,8 +522,9 @@ static ssize_t show_sensor(struct device *dev, struct device_attribute *attr, int nr = sensor_attr->index; struct it87_data *data = it87_update_device(dev); - u8 reg = data->sensor; /* In case the value is updated while we use it */ - + u8 reg = data->sensor; /* In case the value is updated while + we use it */ + if (reg & (1 << nr)) return sprintf(buf, "3\n"); /* thermal diode */ if (reg & (8 << nr)) @@ -498,7 +538,10 @@ static ssize_t set_sensor(struct device *dev, struct device_attribute *attr, int nr = sensor_attr->index; struct it87_data *data = dev_get_drvdata(dev); - int val = simple_strtol(buf, NULL, 10); + long val; + + if (strict_strtol(buf, 10, &val) < 0) + return -EINVAL; mutex_lock(&data->update_lock); @@ -511,9 +554,9 @@ static ssize_t set_sensor(struct device *dev, struct device_attribute *attr, } /* 3 = thermal diode; 4 = thermistor; 0 = disabled */ if (val == 3) - data->sensor |= 1 << nr; + data->sensor |= 1 << nr; else if (val == 4) - data->sensor |= 8 << nr; + data->sensor |= 8 << nr; else if (val != 0) { mutex_unlock(&data->update_lock); return -EINVAL; @@ -531,6 +574,19 @@ show_sensor_offset(2); show_sensor_offset(3); /* 3 Fans */ + +static int pwm_mode(const struct it87_data *data, int nr) +{ + int ctrl = data->fan_main_ctrl & (1 << nr); + + if (ctrl == 0) /* Full speed */ + return 0; + if (data->pwm_ctrl[nr] & 0x80) /* Automatic mode */ + return 2; + else /* Manual mode */ + return 1; +} + static ssize_t show_fan(struct device *dev, struct device_attribute *attr, char *buf) { @@ -538,7 +594,7 @@ static ssize_t show_fan(struct device *dev, struct device_attribute *attr, int nr = sensor_attr->index; struct it87_data *data = it87_update_device(dev); - return sprintf(buf,"%d\n", FAN_FROM_REG(data->fan[nr], + return sprintf(buf, "%d\n", FAN_FROM_REG(data->fan[nr], DIV_FROM_REG(data->fan_div[nr]))); } static ssize_t show_fan_min(struct device *dev, struct device_attribute *attr, @@ -548,8 +604,8 @@ static ssize_t show_fan_min(struct device *dev, struct device_attribute *attr, int nr = sensor_attr->index; struct it87_data *data = it87_update_device(dev); - return sprintf(buf,"%d\n", - FAN_FROM_REG(data->fan_min[nr], DIV_FROM_REG(data->fan_div[nr]))); + return sprintf(buf, "%d\n", FAN_FROM_REG(data->fan_min[nr], + DIV_FROM_REG(data->fan_div[nr]))); } static ssize_t show_fan_div(struct device *dev, struct device_attribute *attr, char *buf) @@ -560,14 +616,14 @@ static ssize_t show_fan_div(struct device *dev, struct device_attribute *attr, struct it87_data *data = it87_update_device(dev); return sprintf(buf, "%d\n", DIV_FROM_REG(data->fan_div[nr])); } -static ssize_t show_pwm_enable(struct device *dev, struct device_attribute *attr, - char *buf) +static ssize_t show_pwm_enable(struct device *dev, + struct device_attribute *attr, char *buf) { struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; struct it87_data *data = it87_update_device(dev); - return sprintf(buf,"%d\n", (data->fan_main_ctrl & (1 << nr)) ? 1 : 0); + return sprintf(buf, "%d\n", pwm_mode(data, nr)); } static ssize_t show_pwm(struct device *dev, struct device_attribute *attr, char *buf) @@ -576,7 +632,7 @@ static ssize_t show_pwm(struct device *dev, struct device_attribute *attr, int nr = sensor_attr->index; struct it87_data *data = it87_update_device(dev); - return sprintf(buf,"%d\n", data->manual_pwm_ctl[nr]); + return sprintf(buf, "%d\n", PWM_FROM_REG(data->pwm_duty[nr])); } static ssize_t show_pwm_freq(struct device *dev, struct device_attribute *attr, char *buf) @@ -593,15 +649,24 @@ static ssize_t set_fan_min(struct device *dev, struct device_attribute *attr, int nr = sensor_attr->index; struct it87_data *data = dev_get_drvdata(dev); - int val = simple_strtol(buf, NULL, 10); + long val; u8 reg; + if (strict_strtol(buf, 10, &val) < 0) + return -EINVAL; + mutex_lock(&data->update_lock); reg = it87_read_value(data, IT87_REG_FAN_DIV); switch (nr) { - case 0: data->fan_div[nr] = reg & 0x07; break; - case 1: data->fan_div[nr] = (reg >> 3) & 0x07; break; - case 2: data->fan_div[nr] = (reg & 0x40) ? 3 : 1; break; + case 0: + data->fan_div[nr] = reg & 0x07; + break; + case 1: + data->fan_div[nr] = (reg >> 3) & 0x07; + break; + case 2: + data->fan_div[nr] = (reg & 0x40) ? 3 : 1; + break; } data->fan_min[nr] = FAN_TO_REG(val, DIV_FROM_REG(data->fan_div[nr])); @@ -616,10 +681,13 @@ static ssize_t set_fan_div(struct device *dev, struct device_attribute *attr, int nr = sensor_attr->index; struct it87_data *data = dev_get_drvdata(dev); - unsigned long val = simple_strtoul(buf, NULL, 10); + unsigned long val; int min; u8 old; + if (strict_strtoul(buf, 10, &val) < 0) + return -EINVAL; + mutex_lock(&data->update_lock); old = it87_read_value(data, IT87_REG_FAN_DIV); @@ -651,6 +719,32 @@ static ssize_t set_fan_div(struct device *dev, struct device_attribute *attr, mutex_unlock(&data->update_lock); return count; } + +/* Returns 0 if OK, -EINVAL otherwise */ +static int check_trip_points(struct device *dev, int nr) +{ + const struct it87_data *data = dev_get_drvdata(dev); + int i, err = 0; + + if (has_old_autopwm(data)) { + for (i = 0; i < 3; i++) { + if (data->auto_temp[nr][i] > data->auto_temp[nr][i + 1]) + err = -EINVAL; + } + for (i = 0; i < 2; i++) { + if (data->auto_pwm[nr][i] > data->auto_pwm[nr][i + 1]) + err = -EINVAL; + } + } + + if (err) { + dev_err(dev, "Inconsistent trip points, not switching to " + "automatic mode\n"); + dev_err(dev, "Adjust the trip points and try again\n"); + } + return err; +} + static ssize_t set_pwm_enable(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { @@ -658,7 +752,16 @@ static ssize_t set_pwm_enable(struct device *dev, int nr = sensor_attr->index; struct it87_data *data = dev_get_drvdata(dev); - int val = simple_strtol(buf, NULL, 10); + long val; + + if (strict_strtol(buf, 10, &val) < 0 || val < 0 || val > 2) + return -EINVAL; + + /* Check trip points before switching to automatic mode */ + if (val == 2) { + if (check_trip_points(dev, nr) < 0) + return -EINVAL; + } mutex_lock(&data->update_lock); @@ -669,16 +772,18 @@ static ssize_t set_pwm_enable(struct device *dev, it87_write_value(data, IT87_REG_FAN_CTL, tmp | (1 << nr)); /* set on/off mode */ data->fan_main_ctrl &= ~(1 << nr); - it87_write_value(data, IT87_REG_FAN_MAIN_CTRL, data->fan_main_ctrl); - } else if (val == 1) { + it87_write_value(data, IT87_REG_FAN_MAIN_CTRL, + data->fan_main_ctrl); + } else { + if (val == 1) /* Manual mode */ + data->pwm_ctrl[nr] = data->pwm_duty[nr]; + else /* Automatic mode */ + data->pwm_ctrl[nr] = 0x80 | data->pwm_temp_map[nr]; + it87_write_value(data, IT87_REG_PWM(nr), data->pwm_ctrl[nr]); /* set SmartGuardian mode */ data->fan_main_ctrl |= (1 << nr); - it87_write_value(data, IT87_REG_FAN_MAIN_CTRL, data->fan_main_ctrl); - /* set saved pwm value, clear FAN_CTLX PWM mode bit */ - it87_write_value(data, IT87_REG_PWM(nr), PWM_TO_REG(data->manual_pwm_ctl[nr])); - } else { - mutex_unlock(&data->update_lock); - return -EINVAL; + it87_write_value(data, IT87_REG_FAN_MAIN_CTRL, + data->fan_main_ctrl); } mutex_unlock(&data->update_lock); @@ -691,15 +796,19 @@ static ssize_t set_pwm(struct device *dev, struct device_attribute *attr, int nr = sensor_attr->index; struct it87_data *data = dev_get_drvdata(dev); - int val = simple_strtol(buf, NULL, 10); + long val; - if (val < 0 || val > 255) + if (strict_strtol(buf, 10, &val) < 0 || val < 0 || val > 255) return -EINVAL; mutex_lock(&data->update_lock); - data->manual_pwm_ctl[nr] = val; - if (data->fan_main_ctrl & (1 << nr)) - it87_write_value(data, IT87_REG_PWM(nr), PWM_TO_REG(data->manual_pwm_ctl[nr])); + data->pwm_duty[nr] = PWM_TO_REG(val); + /* If we are in manual mode, write the duty cycle immediately; + * otherwise, just store it for later use. */ + if (!(data->pwm_ctrl[nr] & 0x80)) { + data->pwm_ctrl[nr] = data->pwm_duty[nr]; + it87_write_value(data, IT87_REG_PWM(nr), data->pwm_ctrl[nr]); + } mutex_unlock(&data->update_lock); return count; } @@ -707,9 +816,12 @@ static ssize_t set_pwm_freq(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { struct it87_data *data = dev_get_drvdata(dev); - unsigned long val = simple_strtoul(buf, NULL, 10); + unsigned long val; int i; + if (strict_strtoul(buf, 10, &val) < 0) + return -EINVAL; + /* Search for the nearest available frequency */ for (i = 0; i < 7; i++) { if (val > (pwm_freq[i] + pwm_freq[i+1]) / 2) @@ -724,6 +836,132 @@ static ssize_t set_pwm_freq(struct device *dev, return count; } +static ssize_t show_pwm_temp_map(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); + int nr = sensor_attr->index; + + struct it87_data *data = it87_update_device(dev); + int map; + + if (data->pwm_temp_map[nr] < 3) + map = 1 << data->pwm_temp_map[nr]; + else + map = 0; /* Should never happen */ + return sprintf(buf, "%d\n", map); +} +static ssize_t set_pwm_temp_map(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); + int nr = sensor_attr->index; + + struct it87_data *data = dev_get_drvdata(dev); + long val; + u8 reg; + + /* This check can go away if we ever support automatic fan speed + control on newer chips. */ + if (!has_old_autopwm(data)) { + dev_notice(dev, "Mapping change disabled for safety reasons\n"); + return -EINVAL; + } + + if (strict_strtol(buf, 10, &val) < 0) + return -EINVAL; + + switch (val) { + case (1 << 0): + reg = 0x00; + break; + case (1 << 1): + reg = 0x01; + break; + case (1 << 2): + reg = 0x02; + break; + default: + return -EINVAL; + } + + mutex_lock(&data->update_lock); + data->pwm_temp_map[nr] = reg; + /* If we are in automatic mode, write the temp mapping immediately; + * otherwise, just store it for later use. */ + if (data->pwm_ctrl[nr] & 0x80) { + data->pwm_ctrl[nr] = 0x80 | data->pwm_temp_map[nr]; + it87_write_value(data, IT87_REG_PWM(nr), data->pwm_ctrl[nr]); + } + mutex_unlock(&data->update_lock); + return count; +} + +static ssize_t show_auto_pwm(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct it87_data *data = it87_update_device(dev); + struct sensor_device_attribute_2 *sensor_attr = + to_sensor_dev_attr_2(attr); + int nr = sensor_attr->nr; + int point = sensor_attr->index; + + return sprintf(buf, "%d\n", PWM_FROM_REG(data->auto_pwm[nr][point])); +} + +static ssize_t set_auto_pwm(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + struct it87_data *data = dev_get_drvdata(dev); + struct sensor_device_attribute_2 *sensor_attr = + to_sensor_dev_attr_2(attr); + int nr = sensor_attr->nr; + int point = sensor_attr->index; + long val; + + if (strict_strtol(buf, 10, &val) < 0 || val < 0 || val > 255) + return -EINVAL; + + mutex_lock(&data->update_lock); + data->auto_pwm[nr][point] = PWM_TO_REG(val); + it87_write_value(data, IT87_REG_AUTO_PWM(nr, point), + data->auto_pwm[nr][point]); + mutex_unlock(&data->update_lock); + return count; +} + +static ssize_t show_auto_temp(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct it87_data *data = it87_update_device(dev); + struct sensor_device_attribute_2 *sensor_attr = + to_sensor_dev_attr_2(attr); + int nr = sensor_attr->nr; + int point = sensor_attr->index; + + return sprintf(buf, "%d\n", TEMP_FROM_REG(data->auto_temp[nr][point])); +} + +static ssize_t set_auto_temp(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + struct it87_data *data = dev_get_drvdata(dev); + struct sensor_device_attribute_2 *sensor_attr = + to_sensor_dev_attr_2(attr); + int nr = sensor_attr->nr; + int point = sensor_attr->index; + long val; + + if (strict_strtol(buf, 10, &val) < 0 || val < -128000 || val > 127000) + return -EINVAL; + + mutex_lock(&data->update_lock); + data->auto_temp[nr][point] = TEMP_TO_REG(val); + it87_write_value(data, IT87_REG_AUTO_TEMP(nr, point), + data->auto_temp[nr][point]); + mutex_unlock(&data->update_lock); + return count; +} #define show_fan_offset(offset) \ static SENSOR_DEVICE_ATTR(fan##offset##_input, S_IRUGO, \ @@ -744,7 +982,36 @@ static SENSOR_DEVICE_ATTR(pwm##offset, S_IRUGO | S_IWUSR, \ show_pwm, set_pwm, offset - 1); \ static DEVICE_ATTR(pwm##offset##_freq, \ (offset == 1 ? S_IRUGO | S_IWUSR : S_IRUGO), \ - show_pwm_freq, (offset == 1 ? set_pwm_freq : NULL)); + show_pwm_freq, (offset == 1 ? set_pwm_freq : NULL)); \ +static SENSOR_DEVICE_ATTR(pwm##offset##_auto_channels_temp, \ + S_IRUGO | S_IWUSR, show_pwm_temp_map, set_pwm_temp_map, \ + offset - 1); \ +static SENSOR_DEVICE_ATTR_2(pwm##offset##_auto_point1_pwm, \ + S_IRUGO | S_IWUSR, show_auto_pwm, set_auto_pwm, \ + offset - 1, 0); \ +static SENSOR_DEVICE_ATTR_2(pwm##offset##_auto_point2_pwm, \ + S_IRUGO | S_IWUSR, show_auto_pwm, set_auto_pwm, \ + offset - 1, 1); \ +static SENSOR_DEVICE_ATTR_2(pwm##offset##_auto_point3_pwm, \ + S_IRUGO | S_IWUSR, show_auto_pwm, set_auto_pwm, \ + offset - 1, 2); \ +static SENSOR_DEVICE_ATTR_2(pwm##offset##_auto_point4_pwm, \ + S_IRUGO, show_auto_pwm, NULL, offset - 1, 3); \ +static SENSOR_DEVICE_ATTR_2(pwm##offset##_auto_point1_temp, \ + S_IRUGO | S_IWUSR, show_auto_temp, set_auto_temp, \ + offset - 1, 1); \ +static SENSOR_DEVICE_ATTR_2(pwm##offset##_auto_point1_temp_hyst, \ + S_IRUGO | S_IWUSR, show_auto_temp, set_auto_temp, \ + offset - 1, 0); \ +static SENSOR_DEVICE_ATTR_2(pwm##offset##_auto_point2_temp, \ + S_IRUGO | S_IWUSR, show_auto_temp, set_auto_temp, \ + offset - 1, 2); \ +static SENSOR_DEVICE_ATTR_2(pwm##offset##_auto_point3_temp, \ + S_IRUGO | S_IWUSR, show_auto_temp, set_auto_temp, \ + offset - 1, 3); \ +static SENSOR_DEVICE_ATTR_2(pwm##offset##_auto_point4_temp, \ + S_IRUGO | S_IWUSR, show_auto_temp, set_auto_temp, \ + offset - 1, 4); show_pwm_offset(1); show_pwm_offset(2); @@ -775,7 +1042,10 @@ static ssize_t set_fan16_min(struct device *dev, struct device_attribute *attr, struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr); int nr = sensor_attr->index; struct it87_data *data = dev_get_drvdata(dev); - int val = simple_strtol(buf, NULL, 10); + long val; + + if (strict_strtol(buf, 10, &val) < 0) + return -EINVAL; mutex_lock(&data->update_lock); data->fan_min[nr] = FAN16_TO_REG(val); @@ -805,7 +1075,8 @@ show_fan16_offset(4); show_fan16_offset(5); /* Alarms */ -static ssize_t show_alarms(struct device *dev, struct device_attribute *attr, char *buf) +static ssize_t show_alarms(struct device *dev, struct device_attribute *attr, + char *buf) { struct it87_data *data = it87_update_device(dev); return sprintf(buf, "%u\n", data->alarms); @@ -836,27 +1107,78 @@ static SENSOR_DEVICE_ATTR(temp1_alarm, S_IRUGO, show_alarm, NULL, 16); static SENSOR_DEVICE_ATTR(temp2_alarm, S_IRUGO, show_alarm, NULL, 17); static SENSOR_DEVICE_ATTR(temp3_alarm, S_IRUGO, show_alarm, NULL, 18); -static ssize_t -show_vrm_reg(struct device *dev, struct device_attribute *attr, char *buf) +static ssize_t show_beep(struct device *dev, struct device_attribute *attr, + char *buf) +{ + int bitnr = to_sensor_dev_attr(attr)->index; + struct it87_data *data = it87_update_device(dev); + return sprintf(buf, "%u\n", (data->beeps >> bitnr) & 1); +} +static ssize_t set_beep(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + int bitnr = to_sensor_dev_attr(attr)->index; + struct it87_data *data = dev_get_drvdata(dev); + long val; + + if (strict_strtol(buf, 10, &val) < 0 + || (val != 0 && val != 1)) + return -EINVAL; + + mutex_lock(&data->update_lock); + data->beeps = it87_read_value(data, IT87_REG_BEEP_ENABLE); + if (val) + data->beeps |= (1 << bitnr); + else + data->beeps &= ~(1 << bitnr); + it87_write_value(data, IT87_REG_BEEP_ENABLE, data->beeps); + mutex_unlock(&data->update_lock); + return count; +} + +static SENSOR_DEVICE_ATTR(in0_beep, S_IRUGO | S_IWUSR, + show_beep, set_beep, 1); +static SENSOR_DEVICE_ATTR(in1_beep, S_IRUGO, show_beep, NULL, 1); +static SENSOR_DEVICE_ATTR(in2_beep, S_IRUGO, show_beep, NULL, 1); +static SENSOR_DEVICE_ATTR(in3_beep, S_IRUGO, show_beep, NULL, 1); +static SENSOR_DEVICE_ATTR(in4_beep, S_IRUGO, show_beep, NULL, 1); +static SENSOR_DEVICE_ATTR(in5_beep, S_IRUGO, show_beep, NULL, 1); +static SENSOR_DEVICE_ATTR(in6_beep, S_IRUGO, show_beep, NULL, 1); +static SENSOR_DEVICE_ATTR(in7_beep, S_IRUGO, show_beep, NULL, 1); +/* fanX_beep writability is set later */ +static SENSOR_DEVICE_ATTR(fan1_beep, S_IRUGO, show_beep, set_beep, 0); +static SENSOR_DEVICE_ATTR(fan2_beep, S_IRUGO, show_beep, set_beep, 0); +static SENSOR_DEVICE_ATTR(fan3_beep, S_IRUGO, show_beep, set_beep, 0); +static SENSOR_DEVICE_ATTR(fan4_beep, S_IRUGO, show_beep, set_beep, 0); +static SENSOR_DEVICE_ATTR(fan5_beep, S_IRUGO, show_beep, set_beep, 0); +static SENSOR_DEVICE_ATTR(temp1_beep, S_IRUGO | S_IWUSR, + show_beep, set_beep, 2); +static SENSOR_DEVICE_ATTR(temp2_beep, S_IRUGO, show_beep, NULL, 2); +static SENSOR_DEVICE_ATTR(temp3_beep, S_IRUGO, show_beep, NULL, 2); + +static ssize_t show_vrm_reg(struct device *dev, struct device_attribute *attr, + char *buf) { struct it87_data *data = dev_get_drvdata(dev); return sprintf(buf, "%u\n", data->vrm); } -static ssize_t -store_vrm_reg(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) +static ssize_t store_vrm_reg(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) { struct it87_data *data = dev_get_drvdata(dev); - u32 val; + unsigned long val; + + if (strict_strtoul(buf, 10, &val) < 0) + return -EINVAL; - val = simple_strtoul(buf, NULL, 10); data->vrm = val; return count; } static DEVICE_ATTR(vrm, S_IRUGO | S_IWUSR, show_vrm_reg, store_vrm_reg); -static ssize_t -show_vid_reg(struct device *dev, struct device_attribute *attr, char *buf) +static ssize_t show_vid_reg(struct device *dev, struct device_attribute *attr, + char *buf) { struct it87_data *data = it87_update_device(dev); return sprintf(buf, "%ld\n", (long) vid_from_reg(data->vid, data->vrm)); @@ -931,51 +1253,176 @@ static const struct attribute_group it87_group = { .attrs = it87_attributes, }; -static struct attribute *it87_attributes_opt[] = { +static struct attribute *it87_attributes_beep[] = { + &sensor_dev_attr_in0_beep.dev_attr.attr, + &sensor_dev_attr_in1_beep.dev_attr.attr, + &sensor_dev_attr_in2_beep.dev_attr.attr, + &sensor_dev_attr_in3_beep.dev_attr.attr, + &sensor_dev_attr_in4_beep.dev_attr.attr, + &sensor_dev_attr_in5_beep.dev_attr.attr, + &sensor_dev_attr_in6_beep.dev_attr.attr, + &sensor_dev_attr_in7_beep.dev_attr.attr, + + &sensor_dev_attr_temp1_beep.dev_attr.attr, + &sensor_dev_attr_temp2_beep.dev_attr.attr, + &sensor_dev_attr_temp3_beep.dev_attr.attr, + NULL +}; + +static const struct attribute_group it87_group_beep = { + .attrs = it87_attributes_beep, +}; + +static struct attribute *it87_attributes_fan16[5][3+1] = { { &sensor_dev_attr_fan1_input16.dev_attr.attr, &sensor_dev_attr_fan1_min16.dev_attr.attr, + &sensor_dev_attr_fan1_alarm.dev_attr.attr, + NULL +}, { &sensor_dev_attr_fan2_input16.dev_attr.attr, &sensor_dev_attr_fan2_min16.dev_attr.attr, + &sensor_dev_attr_fan2_alarm.dev_attr.attr, + NULL +}, { &sensor_dev_attr_fan3_input16.dev_attr.attr, &sensor_dev_attr_fan3_min16.dev_attr.attr, + &sensor_dev_attr_fan3_alarm.dev_attr.attr, + NULL +}, { &sensor_dev_attr_fan4_input16.dev_attr.attr, &sensor_dev_attr_fan4_min16.dev_attr.attr, + &sensor_dev_attr_fan4_alarm.dev_attr.attr, + NULL +}, { &sensor_dev_attr_fan5_input16.dev_attr.attr, &sensor_dev_attr_fan5_min16.dev_attr.attr, + &sensor_dev_attr_fan5_alarm.dev_attr.attr, + NULL +} }; + +static const struct attribute_group it87_group_fan16[5] = { + { .attrs = it87_attributes_fan16[0] }, + { .attrs = it87_attributes_fan16[1] }, + { .attrs = it87_attributes_fan16[2] }, + { .attrs = it87_attributes_fan16[3] }, + { .attrs = it87_attributes_fan16[4] }, +}; +static struct attribute *it87_attributes_fan[3][4+1] = { { &sensor_dev_attr_fan1_input.dev_attr.attr, &sensor_dev_attr_fan1_min.dev_attr.attr, &sensor_dev_attr_fan1_div.dev_attr.attr, + &sensor_dev_attr_fan1_alarm.dev_attr.attr, + NULL +}, { &sensor_dev_attr_fan2_input.dev_attr.attr, &sensor_dev_attr_fan2_min.dev_attr.attr, &sensor_dev_attr_fan2_div.dev_attr.attr, + &sensor_dev_attr_fan2_alarm.dev_attr.attr, + NULL +}, { &sensor_dev_attr_fan3_input.dev_attr.attr, &sensor_dev_attr_fan3_min.dev_attr.attr, &sensor_dev_attr_fan3_div.dev_attr.attr, - - &sensor_dev_attr_fan1_alarm.dev_attr.attr, - &sensor_dev_attr_fan2_alarm.dev_attr.attr, &sensor_dev_attr_fan3_alarm.dev_attr.attr, - &sensor_dev_attr_fan4_alarm.dev_attr.attr, - &sensor_dev_attr_fan5_alarm.dev_attr.attr, + NULL +} }; + +static const struct attribute_group it87_group_fan[3] = { + { .attrs = it87_attributes_fan[0] }, + { .attrs = it87_attributes_fan[1] }, + { .attrs = it87_attributes_fan[2] }, +}; + +static const struct attribute_group * +it87_get_fan_group(const struct it87_data *data) +{ + return has_16bit_fans(data) ? it87_group_fan16 : it87_group_fan; +} +static struct attribute *it87_attributes_pwm[3][4+1] = { { &sensor_dev_attr_pwm1_enable.dev_attr.attr, - &sensor_dev_attr_pwm2_enable.dev_attr.attr, - &sensor_dev_attr_pwm3_enable.dev_attr.attr, &sensor_dev_attr_pwm1.dev_attr.attr, - &sensor_dev_attr_pwm2.dev_attr.attr, - &sensor_dev_attr_pwm3.dev_attr.attr, &dev_attr_pwm1_freq.attr, + &sensor_dev_attr_pwm1_auto_channels_temp.dev_attr.attr, + NULL +}, { + &sensor_dev_attr_pwm2_enable.dev_attr.attr, + &sensor_dev_attr_pwm2.dev_attr.attr, &dev_attr_pwm2_freq.attr, + &sensor_dev_attr_pwm2_auto_channels_temp.dev_attr.attr, + NULL +}, { + &sensor_dev_attr_pwm3_enable.dev_attr.attr, + &sensor_dev_attr_pwm3.dev_attr.attr, &dev_attr_pwm3_freq.attr, + &sensor_dev_attr_pwm3_auto_channels_temp.dev_attr.attr, + NULL +} }; + +static const struct attribute_group it87_group_pwm[3] = { + { .attrs = it87_attributes_pwm[0] }, + { .attrs = it87_attributes_pwm[1] }, + { .attrs = it87_attributes_pwm[2] }, +}; +static struct attribute *it87_attributes_autopwm[3][9+1] = { { + &sensor_dev_attr_pwm1_auto_point1_pwm.dev_attr.attr, + &sensor_dev_attr_pwm1_auto_point2_pwm.dev_attr.attr, + &sensor_dev_attr_pwm1_auto_point3_pwm.dev_attr.attr, + &sensor_dev_attr_pwm1_auto_point4_pwm.dev_attr.attr, + &sensor_dev_attr_pwm1_auto_point1_temp.dev_attr.attr, + &sensor_dev_attr_pwm1_auto_point1_temp_hyst.dev_attr.attr, + &sensor_dev_attr_pwm1_auto_point2_temp.dev_attr.attr, + &sensor_dev_attr_pwm1_auto_point3_temp.dev_attr.attr, + &sensor_dev_attr_pwm1_auto_point4_temp.dev_attr.attr, + NULL +}, { + &sensor_dev_attr_pwm2_auto_point1_pwm.dev_attr.attr, + &sensor_dev_attr_pwm2_auto_point2_pwm.dev_attr.attr, + &sensor_dev_attr_pwm2_auto_point3_pwm.dev_attr.attr, + &sensor_dev_attr_pwm2_auto_point4_pwm.dev_attr.attr, + &sensor_dev_attr_pwm2_auto_point1_temp.dev_attr.attr, + &sensor_dev_attr_pwm2_auto_point1_temp_hyst.dev_attr.attr, + &sensor_dev_attr_pwm2_auto_point2_temp.dev_attr.attr, + &sensor_dev_attr_pwm2_auto_point3_temp.dev_attr.attr, + &sensor_dev_attr_pwm2_auto_point4_temp.dev_attr.attr, + NULL +}, { + &sensor_dev_attr_pwm3_auto_point1_pwm.dev_attr.attr, + &sensor_dev_attr_pwm3_auto_point2_pwm.dev_attr.attr, + &sensor_dev_attr_pwm3_auto_point3_pwm.dev_attr.attr, + &sensor_dev_attr_pwm3_auto_point4_pwm.dev_attr.attr, + &sensor_dev_attr_pwm3_auto_point1_temp.dev_attr.attr, + &sensor_dev_attr_pwm3_auto_point1_temp_hyst.dev_attr.attr, + &sensor_dev_attr_pwm3_auto_point2_temp.dev_attr.attr, + &sensor_dev_attr_pwm3_auto_point3_temp.dev_attr.attr, + &sensor_dev_attr_pwm3_auto_point4_temp.dev_attr.attr, + NULL +} }; + +static const struct attribute_group it87_group_autopwm[3] = { + { .attrs = it87_attributes_autopwm[0] }, + { .attrs = it87_attributes_autopwm[1] }, + { .attrs = it87_attributes_autopwm[2] }, +}; + +static struct attribute *it87_attributes_fan_beep[] = { + &sensor_dev_attr_fan1_beep.dev_attr.attr, + &sensor_dev_attr_fan2_beep.dev_attr.attr, + &sensor_dev_attr_fan3_beep.dev_attr.attr, + &sensor_dev_attr_fan4_beep.dev_attr.attr, + &sensor_dev_attr_fan5_beep.dev_attr.attr, +}; + +static struct attribute *it87_attributes_vid[] = { &dev_attr_vrm.attr, &dev_attr_cpu0_vid.attr, NULL }; -static const struct attribute_group it87_group_opt = { - .attrs = it87_attributes_opt, +static const struct attribute_group it87_group_vid = { + .attrs = it87_attributes_vid, }; /* SuperIO detection - will change isa_address if a chip is found */ @@ -1035,6 +1482,10 @@ static int __init it87_find(unsigned short *address, if (sio_data->type == it87) { /* The IT8705F doesn't have VID pins at all */ sio_data->skip_vid = 1; + + /* The IT8705F has a different LD number for GPIO */ + superio_select(5); + sio_data->beep_pin = superio_inb(IT87_SIO_BEEP_PIN_REG) & 0x3f; } else { int reg; @@ -1068,7 +1519,11 @@ static int __init it87_find(unsigned short *address, pr_info("it87: in3 is VCC (+5V)\n"); if (reg & (1 << 1)) pr_info("it87: in7 is VCCH (+5V Stand-By)\n"); + + sio_data->beep_pin = superio_inb(IT87_SIO_BEEP_PIN_REG) & 0x3f; } + if (sio_data->beep_pin) + pr_info("it87: Beeping is supported\n"); /* Disable specific features based on DMI strings */ board_vendor = dmi_get_system_info(DMI_BOARD_VENDOR); @@ -1093,14 +1548,46 @@ exit: return err; } +static void it87_remove_files(struct device *dev) +{ + struct it87_data *data = platform_get_drvdata(pdev); + struct it87_sio_data *sio_data = dev->platform_data; + const struct attribute_group *fan_group = it87_get_fan_group(data); + int i; + + sysfs_remove_group(&dev->kobj, &it87_group); + if (sio_data->beep_pin) + sysfs_remove_group(&dev->kobj, &it87_group_beep); + for (i = 0; i < 5; i++) { + if (!(data->has_fan & (1 << i))) + continue; + sysfs_remove_group(&dev->kobj, &fan_group[i]); + if (sio_data->beep_pin) + sysfs_remove_file(&dev->kobj, + it87_attributes_fan_beep[i]); + } + for (i = 0; i < 3; i++) { + if (sio_data->skip_pwm & (1 << 0)) + continue; + sysfs_remove_group(&dev->kobj, &it87_group_pwm[i]); + if (has_old_autopwm(data)) + sysfs_remove_group(&dev->kobj, + &it87_group_autopwm[i]); + } + if (!sio_data->skip_vid) + sysfs_remove_group(&dev->kobj, &it87_group_vid); +} + static int __devinit it87_probe(struct platform_device *pdev) { struct it87_data *data; struct resource *res; struct device *dev = &pdev->dev; struct it87_sio_data *sio_data = dev->platform_data; - int err = 0; + const struct attribute_group *fan_group; + int err = 0, i; int enable_pwm_interface; + int fan_beep_need_rw; static const char *names[] = { "it87", "it8712", @@ -1118,7 +1605,8 @@ static int __devinit it87_probe(struct platform_device *pdev) goto ERROR0; } - if (!(data = kzalloc(sizeof(struct it87_data), GFP_KERNEL))) { + data = kzalloc(sizeof(struct it87_data), GFP_KERNEL); + if (!data) { err = -ENOMEM; goto ERROR1; } @@ -1146,120 +1634,60 @@ static int __devinit it87_probe(struct platform_device *pdev) it87_init_device(pdev); /* Register sysfs hooks */ - if ((err = sysfs_create_group(&dev->kobj, &it87_group))) + err = sysfs_create_group(&dev->kobj, &it87_group); + if (err) goto ERROR2; + if (sio_data->beep_pin) { + err = sysfs_create_group(&dev->kobj, &it87_group_beep); + if (err) + goto ERROR4; + } + /* Do not create fan files for disabled fans */ - if (has_16bit_fans(data)) { - /* 16-bit tachometers */ - if (data->has_fan & (1 << 0)) { - if ((err = device_create_file(dev, - &sensor_dev_attr_fan1_input16.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_fan1_min16.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_fan1_alarm.dev_attr))) - goto ERROR4; - } - if (data->has_fan & (1 << 1)) { - if ((err = device_create_file(dev, - &sensor_dev_attr_fan2_input16.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_fan2_min16.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_fan2_alarm.dev_attr))) - goto ERROR4; - } - if (data->has_fan & (1 << 2)) { - if ((err = device_create_file(dev, - &sensor_dev_attr_fan3_input16.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_fan3_min16.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_fan3_alarm.dev_attr))) - goto ERROR4; - } - if (data->has_fan & (1 << 3)) { - if ((err = device_create_file(dev, - &sensor_dev_attr_fan4_input16.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_fan4_min16.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_fan4_alarm.dev_attr))) - goto ERROR4; - } - if (data->has_fan & (1 << 4)) { - if ((err = device_create_file(dev, - &sensor_dev_attr_fan5_input16.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_fan5_min16.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_fan5_alarm.dev_attr))) - goto ERROR4; - } - } else { - /* 8-bit tachometers with clock divider */ - if (data->has_fan & (1 << 0)) { - if ((err = device_create_file(dev, - &sensor_dev_attr_fan1_input.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_fan1_min.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_fan1_div.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_fan1_alarm.dev_attr))) - goto ERROR4; - } - if (data->has_fan & (1 << 1)) { - if ((err = device_create_file(dev, - &sensor_dev_attr_fan2_input.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_fan2_min.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_fan2_div.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_fan2_alarm.dev_attr))) - goto ERROR4; - } - if (data->has_fan & (1 << 2)) { - if ((err = device_create_file(dev, - &sensor_dev_attr_fan3_input.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_fan3_min.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_fan3_div.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_fan3_alarm.dev_attr))) + fan_group = it87_get_fan_group(data); + fan_beep_need_rw = 1; + for (i = 0; i < 5; i++) { + if (!(data->has_fan & (1 << i))) + continue; + err = sysfs_create_group(&dev->kobj, &fan_group[i]); + if (err) + goto ERROR4; + + if (sio_data->beep_pin) { + err = sysfs_create_file(&dev->kobj, + it87_attributes_fan_beep[i]); + if (err) goto ERROR4; + if (!fan_beep_need_rw) + continue; + + /* As we have a single beep enable bit for all fans, + * only the first enabled fan has a writable attribute + * for it. */ + if (sysfs_chmod_file(&dev->kobj, + it87_attributes_fan_beep[i], + S_IRUGO | S_IWUSR)) + dev_dbg(dev, "chmod +w fan%d_beep failed\n", + i + 1); + fan_beep_need_rw = 0; } } if (enable_pwm_interface) { - if (!(sio_data->skip_pwm & (1 << 0))) { - if ((err = device_create_file(dev, - &sensor_dev_attr_pwm1_enable.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_pwm1.dev_attr)) - || (err = device_create_file(dev, - &dev_attr_pwm1_freq))) - goto ERROR4; - } - if (!(sio_data->skip_pwm & (1 << 1))) { - if ((err = device_create_file(dev, - &sensor_dev_attr_pwm2_enable.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_pwm2.dev_attr)) - || (err = device_create_file(dev, - &dev_attr_pwm2_freq))) + for (i = 0; i < 3; i++) { + if (sio_data->skip_pwm & (1 << i)) + continue; + err = sysfs_create_group(&dev->kobj, + &it87_group_pwm[i]); + if (err) goto ERROR4; - } - if (!(sio_data->skip_pwm & (1 << 2))) { - if ((err = device_create_file(dev, - &sensor_dev_attr_pwm3_enable.dev_attr)) - || (err = device_create_file(dev, - &sensor_dev_attr_pwm3.dev_attr)) - || (err = device_create_file(dev, - &dev_attr_pwm3_freq))) + + if (!has_old_autopwm(data)) + continue; + err = sysfs_create_group(&dev->kobj, + &it87_group_autopwm[i]); + if (err) goto ERROR4; } } @@ -1268,10 +1696,8 @@ static int __devinit it87_probe(struct platform_device *pdev) data->vrm = vid_which_vrm(); /* VID reading from Super-I/O config space if available */ data->vid = sio_data->vid_value; - if ((err = device_create_file(dev, - &dev_attr_vrm)) - || (err = device_create_file(dev, - &dev_attr_cpu0_vid))) + err = sysfs_create_group(&dev->kobj, &it87_group_vid); + if (err) goto ERROR4; } @@ -1284,8 +1710,7 @@ static int __devinit it87_probe(struct platform_device *pdev) return 0; ERROR4: - sysfs_remove_group(&dev->kobj, &it87_group); - sysfs_remove_group(&dev->kobj, &it87_group_opt); + it87_remove_files(dev); ERROR2: platform_set_drvdata(pdev, NULL); kfree(data); @@ -1300,8 +1725,7 @@ static int __devexit it87_remove(struct platform_device *pdev) struct it87_data *data = platform_get_drvdata(pdev); hwmon_device_unregister(data->hwmon_dev); - sysfs_remove_group(&pdev->dev.kobj, &it87_group); - sysfs_remove_group(&pdev->dev.kobj, &it87_group_opt); + it87_remove_files(&pdev->dev); release_region(data->addr, IT87_EC_EXTENT); platform_set_drvdata(pdev, NULL); @@ -1387,15 +1811,18 @@ static void __devinit it87_init_device(struct platform_device *pdev) int tmp, i; u8 mask; - /* initialize to sane defaults: - * - if the chip is in manual pwm mode, this will be overwritten with - * the actual settings on the chip (so in this case, initialization - * is not needed) - * - if in automatic or on/off mode, we could switch to manual mode, - * read the registers and set manual_pwm_ctl accordingly, but currently - * this is not implemented, so we initialize to something sane */ + /* For each PWM channel: + * - If it is in automatic mode, setting to manual mode should set + * the fan to full speed by default. + * - If it is in manual mode, we need a mapping to temperature + * channels to use when later setting to automatic mode later. + * Use a 1:1 mapping by default (we are clueless.) + * In both cases, the value can (and should) be changed by the user + * prior to switching to a different mode. */ for (i = 0; i < 3; i++) { - data->manual_pwm_ctl[i] = 0xff; + data->pwm_temp_map[i] = i; + data->pwm_duty[i] = 0x7f; /* Full speed */ + data->auto_pwm[i][3] = 0x7f; /* Full speed, hard-coded */ } /* Some chips seem to have default value 0xff for all limit @@ -1436,7 +1863,8 @@ static void __devinit it87_init_device(struct platform_device *pdev) if ((data->fan_main_ctrl & mask) == 0) { /* Enable all fan tachometers */ data->fan_main_ctrl |= mask; - it87_write_value(data, IT87_REG_FAN_MAIN_CTRL, data->fan_main_ctrl); + it87_write_value(data, IT87_REG_FAN_MAIN_CTRL, + data->fan_main_ctrl); } data->has_fan = (data->fan_main_ctrl >> 4) & 0x07; @@ -1461,30 +1889,32 @@ static void __devinit it87_init_device(struct platform_device *pdev) /* Fan input pins may be used for alternative functions */ data->has_fan &= ~sio_data->skip_fan; - /* Set current fan mode registers and the default settings for the - * other mode registers */ - for (i = 0; i < 3; i++) { - if (data->fan_main_ctrl & (1 << i)) { - /* pwm mode */ - tmp = it87_read_value(data, IT87_REG_PWM(i)); - if (tmp & 0x80) { - /* automatic pwm - not yet implemented, but - * leave the settings made by the BIOS alone - * until a change is requested via the sysfs - * interface */ - } else { - /* manual pwm */ - data->manual_pwm_ctl[i] = PWM_FROM_REG(tmp); - } - } - } - /* Start monitoring */ it87_write_value(data, IT87_REG_CONFIG, (it87_read_value(data, IT87_REG_CONFIG) & 0x36) | (update_vbat ? 0x41 : 0x01)); } +static void it87_update_pwm_ctrl(struct it87_data *data, int nr) +{ + data->pwm_ctrl[nr] = it87_read_value(data, IT87_REG_PWM(nr)); + if (data->pwm_ctrl[nr] & 0x80) /* Automatic mode */ + data->pwm_temp_map[nr] = data->pwm_ctrl[nr] & 0x03; + else /* Manual mode */ + data->pwm_duty[nr] = data->pwm_ctrl[nr] & 0x7f; + + if (has_old_autopwm(data)) { + int i; + + for (i = 0; i < 5 ; i++) + data->auto_temp[nr][i] = it87_read_value(data, + IT87_REG_AUTO_TEMP(nr, i)); + for (i = 0; i < 3 ; i++) + data->auto_pwm[nr][i] = it87_read_value(data, + IT87_REG_AUTO_PWM(nr, i)); + } +} + static struct it87_data *it87_update_device(struct device *dev) { struct it87_data *data = dev_get_drvdata(dev); @@ -1494,24 +1924,22 @@ static struct it87_data *it87_update_device(struct device *dev) if (time_after(jiffies, data->last_updated + HZ + HZ / 2) || !data->valid) { - if (update_vbat) { /* Cleared after each update, so reenable. Value - returned by this read will be previous value */ + returned by this read will be previous value */ it87_write_value(data, IT87_REG_CONFIG, - it87_read_value(data, IT87_REG_CONFIG) | 0x40); + it87_read_value(data, IT87_REG_CONFIG) | 0x40); } for (i = 0; i <= 7; i++) { data->in[i] = - it87_read_value(data, IT87_REG_VIN(i)); + it87_read_value(data, IT87_REG_VIN(i)); data->in_min[i] = - it87_read_value(data, IT87_REG_VIN_MIN(i)); + it87_read_value(data, IT87_REG_VIN_MIN(i)); data->in_max[i] = - it87_read_value(data, IT87_REG_VIN_MAX(i)); + it87_read_value(data, IT87_REG_VIN_MAX(i)); } /* in8 (battery) has no limit registers */ - data->in[8] = - it87_read_value(data, IT87_REG_VIN(8)); + data->in[8] = it87_read_value(data, IT87_REG_VIN(8)); for (i = 0; i < 5; i++) { /* Skip disabled fans */ @@ -1519,7 +1947,7 @@ static struct it87_data *it87_update_device(struct device *dev) continue; data->fan_min[i] = - it87_read_value(data, IT87_REG_FAN_MIN[i]); + it87_read_value(data, IT87_REG_FAN_MIN[i]); data->fan[i] = it87_read_value(data, IT87_REG_FAN[i]); /* Add high byte if in 16-bit mode */ @@ -1532,11 +1960,11 @@ static struct it87_data *it87_update_device(struct device *dev) } for (i = 0; i < 3; i++) { data->temp[i] = - it87_read_value(data, IT87_REG_TEMP(i)); + it87_read_value(data, IT87_REG_TEMP(i)); data->temp_high[i] = - it87_read_value(data, IT87_REG_TEMP_HIGH(i)); + it87_read_value(data, IT87_REG_TEMP_HIGH(i)); data->temp_low[i] = - it87_read_value(data, IT87_REG_TEMP_LOW(i)); + it87_read_value(data, IT87_REG_TEMP_LOW(i)); } /* Newer chips don't have clock dividers */ @@ -1551,9 +1979,13 @@ static struct it87_data *it87_update_device(struct device *dev) it87_read_value(data, IT87_REG_ALARM1) | (it87_read_value(data, IT87_REG_ALARM2) << 8) | (it87_read_value(data, IT87_REG_ALARM3) << 16); + data->beeps = it87_read_value(data, IT87_REG_BEEP_ENABLE); + data->fan_main_ctrl = it87_read_value(data, IT87_REG_FAN_MAIN_CTRL); data->fan_ctl = it87_read_value(data, IT87_REG_FAN_CTL); + for (i = 0; i < 3; i++) + it87_update_pwm_ctrl(data, i); data->sensor = it87_read_value(data, IT87_REG_TEMP_ENABLE); /* The 8705 does not have VID capability. @@ -1628,7 +2060,7 @@ exit: static int __init sm_it87_init(void) { int err; - unsigned short isa_address=0; + unsigned short isa_address = 0; struct it87_sio_data sio_data; memset(&sio_data, 0, sizeof(struct it87_sio_data)); @@ -1640,7 +2072,7 @@ static int __init sm_it87_init(void) return err; err = it87_device_add(isa_address, &sio_data); - if (err){ + if (err) { platform_driver_unregister(&it87_driver); return err; } @@ -1661,7 +2093,8 @@ MODULE_DESCRIPTION("IT8705F/8712F/8716F/8718F/8720F/8726F, SiS950 driver"); module_param(update_vbat, bool, 0); MODULE_PARM_DESC(update_vbat, "Update vbat if set else return powerup value"); module_param(fix_pwm_polarity, bool, 0); -MODULE_PARM_DESC(fix_pwm_polarity, "Force PWM polarity to active high (DANGEROUS)"); +MODULE_PARM_DESC(fix_pwm_polarity, + "Force PWM polarity to active high (DANGEROUS)"); MODULE_LICENSE("GPL"); module_init(sm_it87_init); diff --git a/drivers/hwmon/lm90.c b/drivers/hwmon/lm90.c index 7c9bdc167426..7cc2708871ab 100644 --- a/drivers/hwmon/lm90.c +++ b/drivers/hwmon/lm90.c @@ -1,7 +1,7 @@ /* * lm90.c - Part of lm_sensors, Linux kernel modules for hardware * monitoring - * Copyright (C) 2003-2009 Jean Delvare <khali@linux-fr.org> + * Copyright (C) 2003-2010 Jean Delvare <khali@linux-fr.org> * * Based on the lm83 driver. The LM90 is a sensor chip made by National * Semiconductor. It reports up to two temperatures (its own plus up to @@ -93,7 +93,8 @@ static const unsigned short normal_i2c[] = { 0x18, 0x19, 0x1a, 0x29, 0x2a, 0x2b, 0x4c, 0x4d, 0x4e, I2C_CLIENT_END }; -enum chips { lm90, adm1032, lm99, lm86, max6657, adt7461, max6680, max6646 }; +enum chips { lm90, adm1032, lm99, lm86, max6657, adt7461, max6680, max6646, + w83l771 }; /* * The LM90 registers @@ -151,6 +152,7 @@ static int lm90_detect(struct i2c_client *client, struct i2c_board_info *info); static int lm90_probe(struct i2c_client *client, const struct i2c_device_id *id); static void lm90_init_client(struct i2c_client *client); +static void lm90_alert(struct i2c_client *client, unsigned int flag); static int lm90_remove(struct i2c_client *client); static struct lm90_data *lm90_update_device(struct device *dev); @@ -173,6 +175,7 @@ static const struct i2c_device_id lm90_id[] = { { "max6659", max6657 }, { "max6680", max6680 }, { "max6681", max6680 }, + { "w83l771", w83l771 }, { } }; MODULE_DEVICE_TABLE(i2c, lm90_id); @@ -184,6 +187,7 @@ static struct i2c_driver lm90_driver = { }, .probe = lm90_probe, .remove = lm90_remove, + .alert = lm90_alert, .id_table = lm90_id, .detect = lm90_detect, .address_list = normal_i2c, @@ -201,6 +205,9 @@ struct lm90_data { int kind; int flags; + u8 config_orig; /* Original configuration register value */ + u8 alert_alarms; /* Which alarm bits trigger ALERT# */ + /* registers values */ s8 temp8[4]; /* 0: local low limit 1: local high limit @@ -758,6 +765,14 @@ static int lm90_detect(struct i2c_client *new_client, && reg_convrate <= 0x07) { name = "max6646"; } + } else + if (address == 0x4C + && man_id == 0x5C) { /* Winbond/Nuvoton */ + if ((chip_id & 0xFE) == 0x10 /* W83L771AWG/ASG */ + && (reg_config1 & 0x2A) == 0x00 + && reg_convrate <= 0x08) { + name = "w83l771"; + } } if (!name) { /* identification failed */ @@ -794,6 +809,19 @@ static int lm90_probe(struct i2c_client *new_client, new_client->flags &= ~I2C_CLIENT_PEC; } + /* Different devices have different alarm bits triggering the + * ALERT# output */ + switch (data->kind) { + case lm90: + case lm99: + case lm86: + data->alert_alarms = 0x7b; + break; + default: + data->alert_alarms = 0x7c; + break; + } + /* Initialize the LM90 chip */ lm90_init_client(new_client); @@ -830,7 +858,7 @@ exit: static void lm90_init_client(struct i2c_client *client) { - u8 config, config_orig; + u8 config; struct lm90_data *data = i2c_get_clientdata(client); /* @@ -842,7 +870,7 @@ static void lm90_init_client(struct i2c_client *client) dev_warn(&client->dev, "Initialization failed!\n"); return; } - config_orig = config; + data->config_orig = config; /* Check Temperature Range Select */ if (data->kind == adt7461) { @@ -860,7 +888,7 @@ static void lm90_init_client(struct i2c_client *client) } config &= 0xBF; /* run */ - if (config != config_orig) /* Only write if changed */ + if (config != data->config_orig) /* Only write if changed */ i2c_smbus_write_byte_data(client, LM90_REG_W_CONFIG1, config); } @@ -875,10 +903,46 @@ static int lm90_remove(struct i2c_client *client) device_remove_file(&client->dev, &sensor_dev_attr_temp2_offset.dev_attr); + /* Restore initial configuration */ + i2c_smbus_write_byte_data(client, LM90_REG_W_CONFIG1, + data->config_orig); + kfree(data); return 0; } +static void lm90_alert(struct i2c_client *client, unsigned int flag) +{ + struct lm90_data *data = i2c_get_clientdata(client); + u8 config, alarms; + + lm90_read_reg(client, LM90_REG_R_STATUS, &alarms); + if ((alarms & 0x7f) == 0) { + dev_info(&client->dev, "Everything OK\n"); + } else { + if (alarms & 0x61) + dev_warn(&client->dev, + "temp%d out of range, please check!\n", 1); + if (alarms & 0x1a) + dev_warn(&client->dev, + "temp%d out of range, please check!\n", 2); + if (alarms & 0x04) + dev_warn(&client->dev, + "temp%d diode open, please check!\n", 2); + + /* Disable ALERT# output, because these chips don't implement + SMBus alert correctly; they should only hold the alert line + low briefly. */ + if ((data->kind == adm1032 || data->kind == adt7461) + && (alarms & data->alert_alarms)) { + dev_dbg(&client->dev, "Disabling ALERT#\n"); + lm90_read_reg(client, LM90_REG_R_CONFIG1, &config); + i2c_smbus_write_byte_data(client, LM90_REG_W_CONFIG1, + config | 0x80); + } + } +} + static int lm90_read16(struct i2c_client *client, u8 regh, u8 regl, u16 *value) { int err; @@ -966,6 +1030,21 @@ static struct lm90_data *lm90_update_device(struct device *dev) } lm90_read_reg(client, LM90_REG_R_STATUS, &data->alarms); + /* Re-enable ALERT# output if it was originally enabled and + * relevant alarms are all clear */ + if ((data->config_orig & 0x80) == 0 + && (data->alarms & data->alert_alarms) == 0) { + u8 config; + + lm90_read_reg(client, LM90_REG_R_CONFIG1, &config); + if (config & 0x80) { + dev_dbg(&client->dev, "Re-enabling ALERT#\n"); + i2c_smbus_write_byte_data(client, + LM90_REG_W_CONFIG1, + config & ~0x80); + } + } + data->last_updated = jiffies; data->valid = 1; } diff --git a/drivers/hwmon/tmp401.c b/drivers/hwmon/tmp401.c index a13b30e8d8d8..d14a1af9f550 100644 --- a/drivers/hwmon/tmp401.c +++ b/drivers/hwmon/tmp401.c @@ -134,7 +134,7 @@ struct tmp401_data { struct mutex update_lock; char valid; /* zero until following fields are valid */ unsigned long last_updated; /* in jiffies */ - int kind; + enum chips kind; /* register values */ u8 status; @@ -524,7 +524,7 @@ static int tmp401_detect(struct i2c_client *client, if (reg > 15) return -ENODEV; - strlcpy(info->type, tmp401_id[kind - 1].name, I2C_NAME_SIZE); + strlcpy(info->type, tmp401_id[kind].name, I2C_NAME_SIZE); return 0; } @@ -572,8 +572,7 @@ static int tmp401_probe(struct i2c_client *client, goto exit_remove; } - dev_info(&client->dev, "Detected TI %s chip\n", - names[data->kind - 1]); + dev_info(&client->dev, "Detected TI %s chip\n", names[data->kind]); return 0; diff --git a/drivers/hwmon/tmp421.c b/drivers/hwmon/tmp421.c index 4f7c051e2d7b..738c472ece27 100644 --- a/drivers/hwmon/tmp421.c +++ b/drivers/hwmon/tmp421.c @@ -61,9 +61,9 @@ static const u8 TMP421_TEMP_LSB[4] = { 0x10, 0x11, 0x12, 0x13 }; #define TMP423_DEVICE_ID 0x23 static const struct i2c_device_id tmp421_id[] = { - { "tmp421", tmp421 }, - { "tmp422", tmp422 }, - { "tmp423", tmp423 }, + { "tmp421", 2 }, + { "tmp422", 3 }, + { "tmp423", 4 }, { } }; MODULE_DEVICE_TABLE(i2c, tmp421_id); @@ -73,21 +73,23 @@ struct tmp421_data { struct mutex update_lock; char valid; unsigned long last_updated; - int kind; + int channels; u8 config; s16 temp[4]; }; static int temp_from_s16(s16 reg) { - int temp = reg; + /* Mask out status bits */ + int temp = reg & ~0xf; return (temp * 1000 + 128) / 256; } static int temp_from_u16(u16 reg) { - int temp = reg; + /* Mask out status bits */ + int temp = reg & ~0xf; /* Add offset for extended temperature range. */ temp -= 64 * 256; @@ -107,7 +109,7 @@ static struct tmp421_data *tmp421_update_device(struct device *dev) data->config = i2c_smbus_read_byte_data(client, TMP421_CONFIG_REG_1); - for (i = 0; i <= data->kind; i++) { + for (i = 0; i < data->channels; i++) { data->temp[i] = i2c_smbus_read_byte_data(client, TMP421_TEMP_MSB[i]) << 8; data->temp[i] |= i2c_smbus_read_byte_data(client, @@ -166,7 +168,7 @@ static mode_t tmp421_is_visible(struct kobject *kobj, struct attribute *a, devattr = container_of(a, struct device_attribute, attr); index = to_sensor_dev_attr(devattr)->index; - if (data->kind > index) + if (index < data->channels) return a->mode; return 0; @@ -252,9 +254,9 @@ static int tmp421_detect(struct i2c_client *client, return -ENODEV; } - strlcpy(info->type, tmp421_id[kind - 1].name, I2C_NAME_SIZE); + strlcpy(info->type, tmp421_id[kind].name, I2C_NAME_SIZE); dev_info(&adapter->dev, "Detected TI %s chip at 0x%02x\n", - names[kind - 1], client->addr); + names[kind], client->addr); return 0; } @@ -271,7 +273,7 @@ static int tmp421_probe(struct i2c_client *client, i2c_set_clientdata(client, data); mutex_init(&data->update_lock); - data->kind = id->driver_data; + data->channels = id->driver_data; err = tmp421_init_client(client); if (err) diff --git a/drivers/hwmon/vt8231.c b/drivers/hwmon/vt8231.c index d47b4c9949c2..e6078c9f0e27 100644 --- a/drivers/hwmon/vt8231.c +++ b/drivers/hwmon/vt8231.c @@ -948,8 +948,7 @@ static int __devinit vt8231_pci_probe(struct pci_dev *dev, address = val & ~(VT8231_EXTENT - 1); if (address == 0) { - dev_err(&dev->dev, "base address not set -\ - upgrade BIOS or use force_addr=0xaddr\n"); + dev_err(&dev->dev, "base address not set - upgrade BIOS or use force_addr=0xaddr\n"); return -ENODEV; } diff --git a/drivers/hwmon/w83793.c b/drivers/hwmon/w83793.c index 9a2022b67495..9de81a4c15a2 100644 --- a/drivers/hwmon/w83793.c +++ b/drivers/hwmon/w83793.c @@ -3,6 +3,10 @@ Copyright (C) 2006 Winbond Electronics Corp. Yuan Mu Rudolf Marek <r.marek@assembler.cz> + Copyright (C) 2009-2010 Sven Anders <anders@anduras.de>, ANDURAS AG. + Watchdog driver part + (Based partially on fschmd driver, + Copyright 2007-2008 by Hans de Goede) This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by @@ -35,6 +39,16 @@ #include <linux/hwmon-sysfs.h> #include <linux/err.h> #include <linux/mutex.h> +#include <linux/fs.h> +#include <linux/watchdog.h> +#include <linux/miscdevice.h> +#include <linux/uaccess.h> +#include <linux/kref.h> +#include <linux/notifier.h> +#include <linux/reboot.h> + +/* Default values */ +#define WATCHDOG_TIMEOUT 2 /* 2 minute default timeout */ /* Addresses to scan */ static const unsigned short normal_i2c[] = { 0x2c, 0x2d, 0x2e, 0x2f, @@ -51,6 +65,18 @@ static int reset; module_param(reset, bool, 0); MODULE_PARM_DESC(reset, "Set to 1 to reset chip, not recommended"); +static int timeout = WATCHDOG_TIMEOUT; /* default timeout in minutes */ +module_param(timeout, int, 0); +MODULE_PARM_DESC(timeout, + "Watchdog timeout in minutes. 2<= timeout <=255 (default=" + __MODULE_STRING(WATCHDOG_TIMEOUT) ")"); + +static int nowayout = WATCHDOG_NOWAYOUT; +module_param(nowayout, int, 0); +MODULE_PARM_DESC(nowayout, + "Watchdog cannot be stopped once started (default=" + __MODULE_STRING(WATCHDOG_NOWAYOUT) ")"); + /* Address 0x00, 0x0d, 0x0e, 0x0f in all three banks are reserved as ID, Bank Select registers @@ -72,6 +98,11 @@ MODULE_PARM_DESC(reset, "Set to 1 to reset chip, not recommended"); #define W83793_REG_VID_LATCHB 0x08 #define W83793_REG_VID_CTRL 0x59 +#define W83793_REG_WDT_LOCK 0x01 +#define W83793_REG_WDT_ENABLE 0x02 +#define W83793_REG_WDT_STATUS 0x03 +#define W83793_REG_WDT_TIMEOUT 0x04 + static u16 W83793_REG_TEMP_MODE[2] = { 0x5e, 0x5f }; #define TEMP_READ 0 @@ -223,8 +254,37 @@ struct w83793_data { u8 tolerance[3]; /* Temp tolerance(Smart Fan I/II) */ u8 sf2_pwm[6][7]; /* Smart FanII: Fan duty cycle */ u8 sf2_temp[6][7]; /* Smart FanII: Temp level point */ + + /* watchdog */ + struct i2c_client *client; + struct mutex watchdog_lock; + struct list_head list; /* member of the watchdog_data_list */ + struct kref kref; + struct miscdevice watchdog_miscdev; + unsigned long watchdog_is_open; + char watchdog_expect_close; + char watchdog_name[10]; /* must be unique to avoid sysfs conflict */ + unsigned int watchdog_caused_reboot; + int watchdog_timeout; /* watchdog timeout in minutes */ }; +/* Somewhat ugly :( global data pointer list with all devices, so that + we can find our device data as when using misc_register. There is no + other method to get to one's device data from the open file-op and + for usage in the reboot notifier callback. */ +static LIST_HEAD(watchdog_data_list); + +/* Note this lock not only protect list access, but also data.kref access */ +static DEFINE_MUTEX(watchdog_data_mutex); + +/* Release our data struct when we're detached from the i2c client *and* all + references to our watchdog device are released */ +static void w83793_release_resources(struct kref *ref) +{ + struct w83793_data *data = container_of(ref, struct w83793_data, kref); + kfree(data); +} + static u8 w83793_read_value(struct i2c_client *client, u16 reg); static int w83793_write_value(struct i2c_client *client, u16 reg, u8 value); static int w83793_probe(struct i2c_client *client, @@ -1063,14 +1123,349 @@ static void w83793_init_client(struct i2c_client *client) /* Start monitoring */ w83793_write_value(client, W83793_REG_CONFIG, w83793_read_value(client, W83793_REG_CONFIG) | 0x01); +} + +/* + * Watchdog routines + */ + +static int watchdog_set_timeout(struct w83793_data *data, int timeout) +{ + int ret, mtimeout; + + mtimeout = DIV_ROUND_UP(timeout, 60); + + if (mtimeout > 255) + return -EINVAL; + + mutex_lock(&data->watchdog_lock); + if (!data->client) { + ret = -ENODEV; + goto leave; + } + + data->watchdog_timeout = mtimeout; + + /* Set Timeout value (in Minutes) */ + w83793_write_value(data->client, W83793_REG_WDT_TIMEOUT, + data->watchdog_timeout); + + ret = mtimeout * 60; + +leave: + mutex_unlock(&data->watchdog_lock); + return ret; +} + +static int watchdog_get_timeout(struct w83793_data *data) +{ + int timeout; + + mutex_lock(&data->watchdog_lock); + timeout = data->watchdog_timeout * 60; + mutex_unlock(&data->watchdog_lock); + + return timeout; +} + +static int watchdog_trigger(struct w83793_data *data) +{ + int ret = 0; + + mutex_lock(&data->watchdog_lock); + if (!data->client) { + ret = -ENODEV; + goto leave; + } + + /* Set Timeout value (in Minutes) */ + w83793_write_value(data->client, W83793_REG_WDT_TIMEOUT, + data->watchdog_timeout); + +leave: + mutex_unlock(&data->watchdog_lock); + return ret; +} + +static int watchdog_enable(struct w83793_data *data) +{ + int ret = 0; + + mutex_lock(&data->watchdog_lock); + if (!data->client) { + ret = -ENODEV; + goto leave; + } + + /* Set initial timeout */ + w83793_write_value(data->client, W83793_REG_WDT_TIMEOUT, + data->watchdog_timeout); + + /* Enable Soft Watchdog */ + w83793_write_value(data->client, W83793_REG_WDT_LOCK, 0x55); + +leave: + mutex_unlock(&data->watchdog_lock); + return ret; +} + +static int watchdog_disable(struct w83793_data *data) +{ + int ret = 0; + + mutex_lock(&data->watchdog_lock); + if (!data->client) { + ret = -ENODEV; + goto leave; + } + + /* Disable Soft Watchdog */ + w83793_write_value(data->client, W83793_REG_WDT_LOCK, 0xAA); + +leave: + mutex_unlock(&data->watchdog_lock); + return ret; +} + +static int watchdog_open(struct inode *inode, struct file *filp) +{ + struct w83793_data *pos, *data = NULL; + int watchdog_is_open; + + /* We get called from drivers/char/misc.c with misc_mtx hold, and we + call misc_register() from w83793_probe() with watchdog_data_mutex + hold, as misc_register() takes the misc_mtx lock, this is a possible + deadlock, so we use mutex_trylock here. */ + if (!mutex_trylock(&watchdog_data_mutex)) + return -ERESTARTSYS; + list_for_each_entry(pos, &watchdog_data_list, list) { + if (pos->watchdog_miscdev.minor == iminor(inode)) { + data = pos; + break; + } + } + + /* Check, if device is already open */ + watchdog_is_open = test_and_set_bit(0, &data->watchdog_is_open); + + /* Increase data reference counter (if not already done). + Note we can never not have found data, so we don't check for this */ + if (!watchdog_is_open) + kref_get(&data->kref); + + mutex_unlock(&watchdog_data_mutex); + + /* Check, if device is already open and possibly issue error */ + if (watchdog_is_open) + return -EBUSY; + + /* Enable Soft Watchdog */ + watchdog_enable(data); + + /* Store pointer to data into filp's private data */ + filp->private_data = data; + + return nonseekable_open(inode, filp); +} + +static int watchdog_close(struct inode *inode, struct file *filp) +{ + struct w83793_data *data = filp->private_data; + if (data->watchdog_expect_close) { + watchdog_disable(data); + data->watchdog_expect_close = 0; + } else { + watchdog_trigger(data); + dev_crit(&data->client->dev, + "unexpected close, not stopping watchdog!\n"); + } + + clear_bit(0, &data->watchdog_is_open); + + /* Decrease data reference counter */ + mutex_lock(&watchdog_data_mutex); + kref_put(&data->kref, w83793_release_resources); + mutex_unlock(&watchdog_data_mutex); + + return 0; +} + +static ssize_t watchdog_write(struct file *filp, const char __user *buf, + size_t count, loff_t *offset) +{ + size_t ret; + struct w83793_data *data = filp->private_data; + + if (count) { + if (!nowayout) { + size_t i; + + /* Clear it in case it was set with a previous write */ + data->watchdog_expect_close = 0; + + for (i = 0; i != count; i++) { + char c; + if (get_user(c, buf + i)) + return -EFAULT; + if (c == 'V') + data->watchdog_expect_close = 1; + } + } + ret = watchdog_trigger(data); + if (ret < 0) + return ret; + } + return count; +} + +static int watchdog_ioctl(struct inode *inode, struct file *filp, + unsigned int cmd, unsigned long arg) +{ + static struct watchdog_info ident = { + .options = WDIOF_KEEPALIVEPING | + WDIOF_SETTIMEOUT | + WDIOF_CARDRESET, + .identity = "w83793 watchdog" + }; + + int val, ret = 0; + struct w83793_data *data = filp->private_data; + + switch (cmd) { + case WDIOC_GETSUPPORT: + if (!nowayout) + ident.options |= WDIOF_MAGICCLOSE; + if (copy_to_user((void __user *)arg, &ident, sizeof(ident))) + ret = -EFAULT; + break; + + case WDIOC_GETSTATUS: + val = data->watchdog_caused_reboot ? WDIOF_CARDRESET : 0; + ret = put_user(val, (int __user *)arg); + break; + + case WDIOC_GETBOOTSTATUS: + ret = put_user(0, (int __user *)arg); + break; + + case WDIOC_KEEPALIVE: + ret = watchdog_trigger(data); + break; + + case WDIOC_GETTIMEOUT: + val = watchdog_get_timeout(data); + ret = put_user(val, (int __user *)arg); + break; + + case WDIOC_SETTIMEOUT: + if (get_user(val, (int __user *)arg)) { + ret = -EFAULT; + break; + } + ret = watchdog_set_timeout(data, val); + if (ret > 0) + ret = put_user(ret, (int __user *)arg); + break; + + case WDIOC_SETOPTIONS: + if (get_user(val, (int __user *)arg)) { + ret = -EFAULT; + break; + } + + if (val & WDIOS_DISABLECARD) + ret = watchdog_disable(data); + else if (val & WDIOS_ENABLECARD) + ret = watchdog_enable(data); + else + ret = -EINVAL; + + break; + default: + ret = -ENOTTY; + } + + return ret; +} + +static const struct file_operations watchdog_fops = { + .owner = THIS_MODULE, + .llseek = no_llseek, + .open = watchdog_open, + .release = watchdog_close, + .write = watchdog_write, + .ioctl = watchdog_ioctl, +}; + +/* + * Notifier for system down + */ + +static int watchdog_notify_sys(struct notifier_block *this, unsigned long code, + void *unused) +{ + struct w83793_data *data = NULL; + + if (code == SYS_DOWN || code == SYS_HALT) { + + /* Disable each registered watchdog */ + mutex_lock(&watchdog_data_mutex); + list_for_each_entry(data, &watchdog_data_list, list) { + if (data->watchdog_miscdev.minor) + watchdog_disable(data); + } + mutex_unlock(&watchdog_data_mutex); + } + + return NOTIFY_DONE; } +/* + * The WDT needs to learn about soft shutdowns in order to + * turn the timebomb registers off. + */ + +static struct notifier_block watchdog_notifier = { + .notifier_call = watchdog_notify_sys, +}; + +/* + * Init / remove routines + */ + static int w83793_remove(struct i2c_client *client) { struct w83793_data *data = i2c_get_clientdata(client); struct device *dev = &client->dev; - int i; + int i, tmp; + + /* Unregister the watchdog (if registered) */ + if (data->watchdog_miscdev.minor) { + misc_deregister(&data->watchdog_miscdev); + + if (data->watchdog_is_open) { + dev_warn(&client->dev, + "i2c client detached with watchdog open! " + "Stopping watchdog.\n"); + watchdog_disable(data); + } + + mutex_lock(&watchdog_data_mutex); + list_del(&data->list); + mutex_unlock(&watchdog_data_mutex); + + /* Tell the watchdog code the client is gone */ + mutex_lock(&data->watchdog_lock); + data->client = NULL; + mutex_unlock(&data->watchdog_lock); + } + + /* Reset Configuration Register to Disable Watch Dog Registers */ + tmp = w83793_read_value(client, W83793_REG_CONFIG); + w83793_write_value(client, W83793_REG_CONFIG, tmp & ~0x04); + + unregister_reboot_notifier(&watchdog_notifier); hwmon_device_unregister(data->hwmon_dev); @@ -1099,7 +1494,10 @@ static int w83793_remove(struct i2c_client *client) if (data->lm75[1] != NULL) i2c_unregister_device(data->lm75[1]); - kfree(data); + /* Decrease data reference counter */ + mutex_lock(&watchdog_data_mutex); + kref_put(&data->kref, w83793_release_resources); + mutex_unlock(&watchdog_data_mutex); return 0; } @@ -1203,6 +1601,7 @@ static int w83793_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct device *dev = &client->dev; + const int watchdog_minors[] = { WATCHDOG_MINOR, 212, 213, 214, 215 }; struct w83793_data *data; int i, tmp, val, err; int files_fan = ARRAY_SIZE(w83793_left_fan) / 7; @@ -1218,6 +1617,14 @@ static int w83793_probe(struct i2c_client *client, i2c_set_clientdata(client, data); data->bank = i2c_smbus_read_byte_data(client, W83793_REG_BANKSEL); mutex_init(&data->update_lock); + mutex_init(&data->watchdog_lock); + INIT_LIST_HEAD(&data->list); + kref_init(&data->kref); + + /* Store client pointer in our data struct for watchdog usage + (where the client is found through a data ptr instead of the + otherway around) */ + data->client = client; err = w83793_detect_subclients(client); if (err) @@ -1380,8 +1787,77 @@ static int w83793_probe(struct i2c_client *client, goto exit_remove; } + /* Watchdog initialization */ + + /* Register boot notifier */ + err = register_reboot_notifier(&watchdog_notifier); + if (err != 0) { + dev_err(&client->dev, + "cannot register reboot notifier (err=%d)\n", err); + goto exit_devunreg; + } + + /* Enable Watchdog registers. + Set Configuration Register to Enable Watch Dog Registers + (Bit 2) = XXXX, X1XX. */ + tmp = w83793_read_value(client, W83793_REG_CONFIG); + w83793_write_value(client, W83793_REG_CONFIG, tmp | 0x04); + + /* Set the default watchdog timeout */ + data->watchdog_timeout = timeout; + + /* Check, if last reboot was caused by watchdog */ + data->watchdog_caused_reboot = + w83793_read_value(data->client, W83793_REG_WDT_STATUS) & 0x01; + + /* Disable Soft Watchdog during initialiation */ + watchdog_disable(data); + + /* We take the data_mutex lock early so that watchdog_open() cannot + run when misc_register() has completed, but we've not yet added + our data to the watchdog_data_list (and set the default timeout) */ + mutex_lock(&watchdog_data_mutex); + for (i = 0; i < ARRAY_SIZE(watchdog_minors); i++) { + /* Register our watchdog part */ + snprintf(data->watchdog_name, sizeof(data->watchdog_name), + "watchdog%c", (i == 0) ? '\0' : ('0' + i)); + data->watchdog_miscdev.name = data->watchdog_name; + data->watchdog_miscdev.fops = &watchdog_fops; + data->watchdog_miscdev.minor = watchdog_minors[i]; + + err = misc_register(&data->watchdog_miscdev); + if (err == -EBUSY) + continue; + if (err) { + data->watchdog_miscdev.minor = 0; + dev_err(&client->dev, + "Registering watchdog chardev: %d\n", err); + break; + } + + list_add(&data->list, &watchdog_data_list); + + dev_info(&client->dev, + "Registered watchdog chardev major 10, minor: %d\n", + watchdog_minors[i]); + break; + } + if (i == ARRAY_SIZE(watchdog_minors)) { + data->watchdog_miscdev.minor = 0; + dev_warn(&client->dev, "Couldn't register watchdog chardev " + "(due to no free minor)\n"); + } + + mutex_unlock(&watchdog_data_mutex); + return 0; + /* Unregister hwmon device */ + +exit_devunreg: + + hwmon_device_unregister(data->hwmon_dev); + /* Unregister sysfs hooks */ exit_remove: @@ -1628,7 +2104,7 @@ static void __exit sensors_w83793_exit(void) i2c_del_driver(&w83793_driver); } -MODULE_AUTHOR("Yuan Mu"); +MODULE_AUTHOR("Yuan Mu, Sven Anders"); MODULE_DESCRIPTION("w83793 driver"); MODULE_LICENSE("GPL"); diff --git a/drivers/i2c/busses/i2c-designware.c b/drivers/i2c/busses/i2c-designware.c index 9e18ef97f156..3e72b69aa7f8 100644 --- a/drivers/i2c/busses/i2c-designware.c +++ b/drivers/i2c/busses/i2c-designware.c @@ -497,13 +497,13 @@ static int i2c_dw_handle_tx_abort(struct dw_i2c_dev *dev) int i; if (abort_source & DW_IC_TX_ABRT_NOACK) { - for_each_bit(i, &abort_source, ARRAY_SIZE(abort_sources)) + for_each_set_bit(i, &abort_source, ARRAY_SIZE(abort_sources)) dev_dbg(dev->dev, "%s: %s\n", __func__, abort_sources[i]); return -EREMOTEIO; } - for_each_bit(i, &abort_source, ARRAY_SIZE(abort_sources)) + for_each_set_bit(i, &abort_source, ARRAY_SIZE(abort_sources)) dev_err(dev->dev, "%s: %s\n", __func__, abort_sources[i]); if (abort_source & DW_IC_TX_ARB_LOST) diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c index 7522008fda86..58463da814d1 100644 --- a/drivers/infiniband/core/mad.c +++ b/drivers/infiniband/core/mad.c @@ -1193,10 +1193,7 @@ static int method_in_use(struct ib_mad_mgmt_method_table **method, { int i; - for (i = find_first_bit(mad_reg_req->method_mask, IB_MGMT_MAX_METHODS); - i < IB_MGMT_MAX_METHODS; - i = find_next_bit(mad_reg_req->method_mask, IB_MGMT_MAX_METHODS, - 1+i)) { + for_each_set_bit(i, mad_reg_req->method_mask, IB_MGMT_MAX_METHODS) { if ((*method)->agent[i]) { printk(KERN_ERR PFX "Method %d already in use\n", i); return -EINVAL; @@ -1330,13 +1327,9 @@ static int add_nonoui_reg_req(struct ib_mad_reg_req *mad_reg_req, goto error3; /* Finally, add in methods being registered */ - for (i = find_first_bit(mad_reg_req->method_mask, - IB_MGMT_MAX_METHODS); - i < IB_MGMT_MAX_METHODS; - i = find_next_bit(mad_reg_req->method_mask, IB_MGMT_MAX_METHODS, - 1+i)) { + for_each_set_bit(i, mad_reg_req->method_mask, IB_MGMT_MAX_METHODS) (*method)->agent[i] = agent_priv; - } + return 0; error3: @@ -1429,13 +1422,9 @@ check_in_use: goto error4; /* Finally, add in methods being registered */ - for (i = find_first_bit(mad_reg_req->method_mask, - IB_MGMT_MAX_METHODS); - i < IB_MGMT_MAX_METHODS; - i = find_next_bit(mad_reg_req->method_mask, IB_MGMT_MAX_METHODS, - 1+i)) { + for_each_set_bit(i, mad_reg_req->method_mask, IB_MGMT_MAX_METHODS) (*method)->agent[i] = agent_priv; - } + return 0; error4: diff --git a/drivers/input/touchscreen/mc13783_ts.c b/drivers/input/touchscreen/mc13783_ts.c index be115b3b65eb..be54fd639aca 100644 --- a/drivers/input/touchscreen/mc13783_ts.c +++ b/drivers/input/touchscreen/mc13783_ts.c @@ -44,7 +44,7 @@ static irqreturn_t mc13783_ts_handler(int irq, void *data) { struct mc13783_ts_priv *priv = data; - mc13783_ackirq(priv->mc13783, irq); + mc13783_irq_ack(priv->mc13783, irq); /* * Kick off reading coordinates. Note that if work happens already @@ -135,7 +135,7 @@ static int mc13783_ts_open(struct input_dev *dev) mc13783_lock(priv->mc13783); - mc13783_ackirq(priv->mc13783, MC13783_IRQ_TS); + mc13783_irq_ack(priv->mc13783, MC13783_IRQ_TS); ret = mc13783_irq_request(priv->mc13783, MC13783_IRQ_TS, mc13783_ts_handler, MC13783_TS_NAME, priv); diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index a93637223c8d..3bdbb6115702 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -1160,8 +1160,7 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv) } cc->start = tmpll; - if (dm_get_device(ti, argv[3], cc->start, ti->len, - dm_table_get_mode(ti->table), &cc->dev)) { + if (dm_get_device(ti, argv[3], dm_table_get_mode(ti->table), &cc->dev)) { ti->error = "Device lookup failed"; goto bad_device; } diff --git a/drivers/md/dm-delay.c b/drivers/md/dm-delay.c index ebe7381f47c8..852052880d7a 100644 --- a/drivers/md/dm-delay.c +++ b/drivers/md/dm-delay.c @@ -156,8 +156,8 @@ static int delay_ctr(struct dm_target *ti, unsigned int argc, char **argv) goto bad; } - if (dm_get_device(ti, argv[0], dc->start_read, ti->len, - dm_table_get_mode(ti->table), &dc->dev_read)) { + if (dm_get_device(ti, argv[0], dm_table_get_mode(ti->table), + &dc->dev_read)) { ti->error = "Device lookup failed"; goto bad; } @@ -177,8 +177,8 @@ static int delay_ctr(struct dm_target *ti, unsigned int argc, char **argv) goto bad_dev_read; } - if (dm_get_device(ti, argv[3], dc->start_write, ti->len, - dm_table_get_mode(ti->table), &dc->dev_write)) { + if (dm_get_device(ti, argv[3], dm_table_get_mode(ti->table), + &dc->dev_write)) { ti->error = "Write device lookup failed"; goto bad_dev_read; } diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index 1d669322b27c..d7500e1c26f2 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -285,7 +285,8 @@ retry: up_write(&_hash_lock); } -static int dm_hash_rename(uint32_t cookie, const char *old, const char *new) +static int dm_hash_rename(uint32_t cookie, uint32_t *flags, const char *old, + const char *new) { char *new_name, *old_name; struct hash_cell *hc; @@ -344,7 +345,8 @@ static int dm_hash_rename(uint32_t cookie, const char *old, const char *new) dm_table_put(table); } - dm_kobject_uevent(hc->md, KOBJ_CHANGE, cookie); + if (!dm_kobject_uevent(hc->md, KOBJ_CHANGE, cookie)) + *flags |= DM_UEVENT_GENERATED_FLAG; dm_put(hc->md); up_write(&_hash_lock); @@ -736,10 +738,10 @@ static int dev_remove(struct dm_ioctl *param, size_t param_size) __hash_remove(hc); up_write(&_hash_lock); - dm_kobject_uevent(md, KOBJ_REMOVE, param->event_nr); + if (!dm_kobject_uevent(md, KOBJ_REMOVE, param->event_nr)) + param->flags |= DM_UEVENT_GENERATED_FLAG; dm_put(md); - param->data_size = 0; return 0; } @@ -773,7 +775,9 @@ static int dev_rename(struct dm_ioctl *param, size_t param_size) return r; param->data_size = 0; - return dm_hash_rename(param->event_nr, param->name, new_name); + + return dm_hash_rename(param->event_nr, ¶m->flags, param->name, + new_name); } static int dev_set_geometry(struct dm_ioctl *param, size_t param_size) @@ -897,16 +901,17 @@ static int do_resume(struct dm_ioctl *param) set_disk_ro(dm_disk(md), 1); } - if (dm_suspended_md(md)) + if (dm_suspended_md(md)) { r = dm_resume(md); + if (!r && !dm_kobject_uevent(md, KOBJ_CHANGE, param->event_nr)) + param->flags |= DM_UEVENT_GENERATED_FLAG; + } if (old_map) dm_table_destroy(old_map); - if (!r) { - dm_kobject_uevent(md, KOBJ_CHANGE, param->event_nr); + if (!r) r = __dev_status(md, param); - } dm_put(md); return r; @@ -1476,6 +1481,7 @@ static int validate_params(uint cmd, struct dm_ioctl *param) { /* Always clear this flag */ param->flags &= ~DM_BUFFER_FULL_FLAG; + param->flags &= ~DM_UEVENT_GENERATED_FLAG; /* Ignores parameters */ if (cmd == DM_REMOVE_ALL_CMD || diff --git a/drivers/md/dm-linear.c b/drivers/md/dm-linear.c index 82f7d6e6b1ea..9200dbf2391a 100644 --- a/drivers/md/dm-linear.c +++ b/drivers/md/dm-linear.c @@ -47,8 +47,7 @@ static int linear_ctr(struct dm_target *ti, unsigned int argc, char **argv) } lc->start = tmp; - if (dm_get_device(ti, argv[0], lc->start, ti->len, - dm_table_get_mode(ti->table), &lc->dev)) { + if (dm_get_device(ti, argv[0], dm_table_get_mode(ti->table), &lc->dev)) { ti->error = "dm-linear: Device lookup failed"; goto bad; } diff --git a/drivers/md/dm-log.c b/drivers/md/dm-log.c index 7035582786fb..5a08be0222db 100644 --- a/drivers/md/dm-log.c +++ b/drivers/md/dm-log.c @@ -543,8 +543,7 @@ static int disk_ctr(struct dm_dirty_log *log, struct dm_target *ti, return -EINVAL; } - r = dm_get_device(ti, argv[0], 0, 0 /* FIXME */, - FMODE_READ | FMODE_WRITE, &dev); + r = dm_get_device(ti, argv[0], FMODE_READ | FMODE_WRITE, &dev); if (r) return r; diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index e81345a1d08f..826bce7343b3 100644 --- a/drivers/md/dm-mpath.c +++ b/drivers/md/dm-mpath.c @@ -69,6 +69,7 @@ struct multipath { struct list_head priority_groups; unsigned pg_init_required; /* pg_init needs calling? */ unsigned pg_init_in_progress; /* Only one pg_init allowed at once */ + wait_queue_head_t pg_init_wait; /* Wait for pg_init completion */ unsigned nr_valid_paths; /* Total number of usable paths */ struct pgpath *current_pgpath; @@ -95,8 +96,6 @@ struct multipath { mempool_t *mpio_pool; struct mutex work_mutex; - - unsigned suspended; /* Don't create new I/O internally when set. */ }; /* @@ -202,6 +201,7 @@ static struct multipath *alloc_multipath(struct dm_target *ti) m->queue_io = 1; INIT_WORK(&m->process_queued_ios, process_queued_ios); INIT_WORK(&m->trigger_event, trigger_event); + init_waitqueue_head(&m->pg_init_wait); mutex_init(&m->work_mutex); m->mpio_pool = mempool_create_slab_pool(MIN_IOS, _mpio_cache); if (!m->mpio_pool) { @@ -235,6 +235,21 @@ static void free_multipath(struct multipath *m) * Path selection *-----------------------------------------------*/ +static void __pg_init_all_paths(struct multipath *m) +{ + struct pgpath *pgpath; + + m->pg_init_count++; + m->pg_init_required = 0; + list_for_each_entry(pgpath, &m->current_pg->pgpaths, list) { + /* Skip failed paths */ + if (!pgpath->is_active) + continue; + if (queue_work(kmpath_handlerd, &pgpath->activate_path)) + m->pg_init_in_progress++; + } +} + static void __switch_pg(struct multipath *m, struct pgpath *pgpath) { m->current_pg = pgpath->pg; @@ -439,7 +454,7 @@ static void process_queued_ios(struct work_struct *work) { struct multipath *m = container_of(work, struct multipath, process_queued_ios); - struct pgpath *pgpath = NULL, *tmp; + struct pgpath *pgpath = NULL; unsigned must_queue = 1; unsigned long flags; @@ -457,14 +472,9 @@ static void process_queued_ios(struct work_struct *work) (!pgpath && !m->queue_if_no_path)) must_queue = 0; - if (m->pg_init_required && !m->pg_init_in_progress && pgpath) { - m->pg_init_count++; - m->pg_init_required = 0; - list_for_each_entry(tmp, &pgpath->pg->pgpaths, list) { - if (queue_work(kmpath_handlerd, &tmp->activate_path)) - m->pg_init_in_progress++; - } - } + if (m->pg_init_required && !m->pg_init_in_progress && pgpath) + __pg_init_all_paths(m); + out: spin_unlock_irqrestore(&m->lock, flags); if (!must_queue) @@ -597,8 +607,8 @@ static struct pgpath *parse_path(struct arg_set *as, struct path_selector *ps, if (!p) return ERR_PTR(-ENOMEM); - r = dm_get_device(ti, shift(as), ti->begin, ti->len, - dm_table_get_mode(ti->table), &p->path.dev); + r = dm_get_device(ti, shift(as), dm_table_get_mode(ti->table), + &p->path.dev); if (r) { ti->error = "error getting device"; goto bad; @@ -890,9 +900,34 @@ static int multipath_ctr(struct dm_target *ti, unsigned int argc, return r; } -static void flush_multipath_work(void) +static void multipath_wait_for_pg_init_completion(struct multipath *m) +{ + DECLARE_WAITQUEUE(wait, current); + unsigned long flags; + + add_wait_queue(&m->pg_init_wait, &wait); + + while (1) { + set_current_state(TASK_UNINTERRUPTIBLE); + + spin_lock_irqsave(&m->lock, flags); + if (!m->pg_init_in_progress) { + spin_unlock_irqrestore(&m->lock, flags); + break; + } + spin_unlock_irqrestore(&m->lock, flags); + + io_schedule(); + } + set_current_state(TASK_RUNNING); + + remove_wait_queue(&m->pg_init_wait, &wait); +} + +static void flush_multipath_work(struct multipath *m) { flush_workqueue(kmpath_handlerd); + multipath_wait_for_pg_init_completion(m); flush_workqueue(kmultipathd); flush_scheduled_work(); } @@ -901,7 +936,7 @@ static void multipath_dtr(struct dm_target *ti) { struct multipath *m = ti->private; - flush_multipath_work(); + flush_multipath_work(m); free_multipath(m); } @@ -1128,8 +1163,7 @@ static int pg_init_limit_reached(struct multipath *m, struct pgpath *pgpath) static void pg_init_done(void *data, int errors) { - struct dm_path *path = data; - struct pgpath *pgpath = path_to_pgpath(path); + struct pgpath *pgpath = data; struct priority_group *pg = pgpath->pg; struct multipath *m = pg->m; unsigned long flags; @@ -1143,8 +1177,8 @@ static void pg_init_done(void *data, int errors) errors = 0; break; } - DMERR("Cannot failover device because scsi_dh_%s was not " - "loaded.", m->hw_handler_name); + DMERR("Could not failover the device: Handler scsi_dh_%s " + "Error %d.", m->hw_handler_name, errors); /* * Fail path for now, so we do not ping pong */ @@ -1181,14 +1215,24 @@ static void pg_init_done(void *data, int errors) m->current_pgpath = NULL; m->current_pg = NULL; } - } else if (!m->pg_init_required) { - m->queue_io = 0; + } else if (!m->pg_init_required) pg->bypassed = 0; - } - m->pg_init_in_progress--; - if (!m->pg_init_in_progress) - queue_work(kmultipathd, &m->process_queued_ios); + if (--m->pg_init_in_progress) + /* Activations of other paths are still on going */ + goto out; + + if (!m->pg_init_required) + m->queue_io = 0; + + queue_work(kmultipathd, &m->process_queued_ios); + + /* + * Wake up any thread waiting to suspend. + */ + wake_up(&m->pg_init_wait); + +out: spin_unlock_irqrestore(&m->lock, flags); } @@ -1198,7 +1242,7 @@ static void activate_path(struct work_struct *work) container_of(work, struct pgpath, activate_path); scsi_dh_activate(bdev_get_queue(pgpath->path.dev->bdev), - pg_init_done, &pgpath->path); + pg_init_done, pgpath); } /* @@ -1276,8 +1320,7 @@ static void multipath_postsuspend(struct dm_target *ti) struct multipath *m = ti->private; mutex_lock(&m->work_mutex); - m->suspended = 1; - flush_multipath_work(); + flush_multipath_work(m); mutex_unlock(&m->work_mutex); } @@ -1289,10 +1332,6 @@ static void multipath_resume(struct dm_target *ti) struct multipath *m = (struct multipath *) ti->private; unsigned long flags; - mutex_lock(&m->work_mutex); - m->suspended = 0; - mutex_unlock(&m->work_mutex); - spin_lock_irqsave(&m->lock, flags); m->queue_if_no_path = m->saved_queue_if_no_path; spin_unlock_irqrestore(&m->lock, flags); @@ -1428,11 +1467,6 @@ static int multipath_message(struct dm_target *ti, unsigned argc, char **argv) mutex_lock(&m->work_mutex); - if (m->suspended) { - r = -EBUSY; - goto out; - } - if (dm_suspended(ti)) { r = -EBUSY; goto out; @@ -1471,8 +1505,7 @@ static int multipath_message(struct dm_target *ti, unsigned argc, char **argv) goto out; } - r = dm_get_device(ti, argv[1], ti->begin, ti->len, - dm_table_get_mode(ti->table), &dev); + r = dm_get_device(ti, argv[1], dm_table_get_mode(ti->table), &dev); if (r) { DMWARN("message: error getting device %s", argv[1]); diff --git a/drivers/md/dm-raid1.c b/drivers/md/dm-raid1.c index 6c1046df81f6..ddda531723dc 100644 --- a/drivers/md/dm-raid1.c +++ b/drivers/md/dm-raid1.c @@ -465,9 +465,17 @@ static void map_region(struct dm_io_region *io, struct mirror *m, static void hold_bio(struct mirror_set *ms, struct bio *bio) { /* - * If device is suspended, complete the bio. + * Lock is required to avoid race condition during suspend + * process. */ + spin_lock_irq(&ms->lock); + if (atomic_read(&ms->suspend)) { + spin_unlock_irq(&ms->lock); + + /* + * If device is suspended, complete the bio. + */ if (dm_noflush_suspending(ms->ti)) bio_endio(bio, DM_ENDIO_REQUEUE); else @@ -478,7 +486,6 @@ static void hold_bio(struct mirror_set *ms, struct bio *bio) /* * Hold bio until the suspend is complete. */ - spin_lock_irq(&ms->lock); bio_list_add(&ms->holds, bio); spin_unlock_irq(&ms->lock); } @@ -737,9 +744,12 @@ static void do_writes(struct mirror_set *ms, struct bio_list *writes) dm_rh_delay(ms->rh, bio); while ((bio = bio_list_pop(&nosync))) { - if (unlikely(ms->leg_failure) && errors_handled(ms)) - hold_bio(ms, bio); - else { + if (unlikely(ms->leg_failure) && errors_handled(ms)) { + spin_lock_irq(&ms->lock); + bio_list_add(&ms->failures, bio); + spin_unlock_irq(&ms->lock); + wakeup_mirrord(ms); + } else { map_bio(get_default_mirror(ms), bio); generic_make_request(bio); } @@ -917,8 +927,7 @@ static int get_mirror(struct mirror_set *ms, struct dm_target *ti, return -EINVAL; } - if (dm_get_device(ti, argv[0], offset, ti->len, - dm_table_get_mode(ti->table), + if (dm_get_device(ti, argv[0], dm_table_get_mode(ti->table), &ms->mirror[mirror].dev)) { ti->error = "Device lookup failure"; return -ENXIO; @@ -1259,6 +1268,20 @@ static void mirror_presuspend(struct dm_target *ti) atomic_set(&ms->suspend, 1); /* + * Process bios in the hold list to start recovery waiting + * for bios in the hold list. After the process, no bio has + * a chance to be added in the hold list because ms->suspend + * is set. + */ + spin_lock_irq(&ms->lock); + holds = ms->holds; + bio_list_init(&ms->holds); + spin_unlock_irq(&ms->lock); + + while ((bio = bio_list_pop(&holds))) + hold_bio(ms, bio); + + /* * We must finish up all the work that we've * generated (i.e. recovery work). */ @@ -1278,22 +1301,6 @@ static void mirror_presuspend(struct dm_target *ti) * we know that all of our I/O has been pushed. */ flush_workqueue(ms->kmirrord_wq); - - /* - * Now set ms->suspend is set and the workqueue flushed, no more - * entries can be added to ms->hold list, so process it. - * - * Bios can still arrive concurrently with or after this - * presuspend function, but they cannot join the hold list - * because ms->suspend is set. - */ - spin_lock_irq(&ms->lock); - holds = ms->holds; - bio_list_init(&ms->holds); - spin_unlock_irq(&ms->lock); - - while ((bio = bio_list_pop(&holds))) - hold_bio(ms, bio); } static void mirror_postsuspend(struct dm_target *ti) diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c index ee8eb283650d..54853773510c 100644 --- a/drivers/md/dm-snap.c +++ b/drivers/md/dm-snap.c @@ -83,10 +83,10 @@ struct dm_snapshot { /* Whether or not owning mapped_device is suspended */ int suspended; - mempool_t *pending_pool; - atomic_t pending_exceptions_count; + mempool_t *pending_pool; + struct dm_exception_table pending; struct dm_exception_table complete; @@ -96,6 +96,11 @@ struct dm_snapshot { */ spinlock_t pe_lock; + /* Chunks with outstanding reads */ + spinlock_t tracked_chunk_lock; + mempool_t *tracked_chunk_pool; + struct hlist_head tracked_chunk_hash[DM_TRACKED_CHUNK_HASH_SIZE]; + /* The on disk metadata handler */ struct dm_exception_store *store; @@ -105,10 +110,12 @@ struct dm_snapshot { struct bio_list queued_bios; struct work_struct queued_bios_work; - /* Chunks with outstanding reads */ - mempool_t *tracked_chunk_pool; - spinlock_t tracked_chunk_lock; - struct hlist_head tracked_chunk_hash[DM_TRACKED_CHUNK_HASH_SIZE]; + /* Wait for events based on state_bits */ + unsigned long state_bits; + + /* Range of chunks currently being merged. */ + chunk_t first_merging_chunk; + int num_merging_chunks; /* * The merge operation failed if this flag is set. @@ -125,13 +132,6 @@ struct dm_snapshot { */ int merge_failed; - /* Wait for events based on state_bits */ - unsigned long state_bits; - - /* Range of chunks currently being merged. */ - chunk_t first_merging_chunk; - int num_merging_chunks; - /* * Incoming bios that overlap with chunks being merged must wait * for them to be committed. @@ -1081,8 +1081,7 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv) argv++; argc--; - r = dm_get_device(ti, cow_path, 0, 0, - FMODE_READ | FMODE_WRITE, &s->cow); + r = dm_get_device(ti, cow_path, FMODE_READ | FMODE_WRITE, &s->cow); if (r) { ti->error = "Cannot get COW device"; goto bad_cow; @@ -1098,7 +1097,7 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv) argv += args_used; argc -= args_used; - r = dm_get_device(ti, origin_path, 0, ti->len, origin_mode, &s->origin); + r = dm_get_device(ti, origin_path, origin_mode, &s->origin); if (r) { ti->error = "Cannot get origin device"; goto bad_origin; @@ -2100,8 +2099,7 @@ static int origin_ctr(struct dm_target *ti, unsigned int argc, char **argv) return -EINVAL; } - r = dm_get_device(ti, argv[0], 0, ti->len, - dm_table_get_mode(ti->table), &dev); + r = dm_get_device(ti, argv[0], dm_table_get_mode(ti->table), &dev); if (r) { ti->error = "Cannot get target device"; return r; diff --git a/drivers/md/dm-stripe.c b/drivers/md/dm-stripe.c index bd58703ee8f6..e610725db766 100644 --- a/drivers/md/dm-stripe.c +++ b/drivers/md/dm-stripe.c @@ -80,8 +80,7 @@ static int get_stripe(struct dm_target *ti, struct stripe_c *sc, if (sscanf(argv[1], "%llu", &start) != 1) return -EINVAL; - if (dm_get_device(ti, argv[0], start, sc->stripe_width, - dm_table_get_mode(ti->table), + if (dm_get_device(ti, argv[0], dm_table_get_mode(ti->table), &sc->stripe[stripe].dev)) return -ENXIO; diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index 4b22feb01a0c..9924ea23032d 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -429,8 +429,7 @@ static int upgrade_mode(struct dm_dev_internal *dd, fmode_t new_mode, * it's already present. */ static int __table_get_device(struct dm_table *t, struct dm_target *ti, - const char *path, sector_t start, sector_t len, - fmode_t mode, struct dm_dev **result) + const char *path, fmode_t mode, struct dm_dev **result) { int r; dev_t uninitialized_var(dev); @@ -527,11 +526,10 @@ int dm_set_device_limits(struct dm_target *ti, struct dm_dev *dev, } EXPORT_SYMBOL_GPL(dm_set_device_limits); -int dm_get_device(struct dm_target *ti, const char *path, sector_t start, - sector_t len, fmode_t mode, struct dm_dev **result) +int dm_get_device(struct dm_target *ti, const char *path, fmode_t mode, + struct dm_dev **result) { - return __table_get_device(ti->table, ti, path, - start, len, mode, result); + return __table_get_device(ti->table, ti, path, mode, result); } @@ -1231,8 +1229,6 @@ void dm_table_unplug_all(struct dm_table *t) struct mapped_device *dm_table_get_md(struct dm_table *t) { - dm_get(t->md); - return t->md; } diff --git a/drivers/md/dm-uevent.c b/drivers/md/dm-uevent.c index c7c555a8c7b2..6b1e3b61b25e 100644 --- a/drivers/md/dm-uevent.c +++ b/drivers/md/dm-uevent.c @@ -187,7 +187,7 @@ void dm_path_uevent(enum dm_uevent_type event_type, struct dm_target *ti, if (event_type >= ARRAY_SIZE(_dm_uevent_type_names)) { DMERR("%s: Invalid event_type %d", __func__, event_type); - goto out; + return; } event = dm_build_path_uevent(md, ti, @@ -195,12 +195,9 @@ void dm_path_uevent(enum dm_uevent_type event_type, struct dm_target *ti, _dm_uevent_type_names[event_type].name, path, nr_valid_paths); if (IS_ERR(event)) - goto out; + return; dm_uevent_add(md, &event->elist); - -out: - dm_put(md); } EXPORT_SYMBOL_GPL(dm_path_uevent); diff --git a/drivers/md/dm.c b/drivers/md/dm.c index aa4e2aa86d49..d21e1284604f 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -635,8 +635,10 @@ static void dec_pending(struct dm_io *io, int error) if (!md->barrier_error && io_error != -EOPNOTSUPP) md->barrier_error = io_error; end_io_acct(io); + free_io(md, io); } else { end_io_acct(io); + free_io(md, io); if (io_error != DM_ENDIO_REQUEUE) { trace_block_bio_complete(md->queue, bio); @@ -644,8 +646,6 @@ static void dec_pending(struct dm_io *io, int error) bio_endio(bio, io_error); } } - - free_io(md, io); } } @@ -2618,18 +2618,19 @@ out: /*----------------------------------------------------------------- * Event notification. *---------------------------------------------------------------*/ -void dm_kobject_uevent(struct mapped_device *md, enum kobject_action action, +int dm_kobject_uevent(struct mapped_device *md, enum kobject_action action, unsigned cookie) { char udev_cookie[DM_COOKIE_LENGTH]; char *envp[] = { udev_cookie, NULL }; if (!cookie) - kobject_uevent(&disk_to_dev(md->disk)->kobj, action); + return kobject_uevent(&disk_to_dev(md->disk)->kobj, action); else { snprintf(udev_cookie, DM_COOKIE_LENGTH, "%s=%u", DM_COOKIE_ENV_VAR_NAME, cookie); - kobject_uevent_env(&disk_to_dev(md->disk)->kobj, action, envp); + return kobject_uevent_env(&disk_to_dev(md->disk)->kobj, + action, envp); } } @@ -2699,23 +2700,13 @@ int dm_suspended_md(struct mapped_device *md) int dm_suspended(struct dm_target *ti) { - struct mapped_device *md = dm_table_get_md(ti->table); - int r = dm_suspended_md(md); - - dm_put(md); - - return r; + return dm_suspended_md(dm_table_get_md(ti->table)); } EXPORT_SYMBOL_GPL(dm_suspended); int dm_noflush_suspending(struct dm_target *ti) { - struct mapped_device *md = dm_table_get_md(ti->table); - int r = __noflush_suspending(md); - - dm_put(md); - - return r; + return __noflush_suspending(dm_table_get_md(ti->table)); } EXPORT_SYMBOL_GPL(dm_noflush_suspending); diff --git a/drivers/md/dm.h b/drivers/md/dm.h index 8dadaa5bc396..bad1724d4869 100644 --- a/drivers/md/dm.h +++ b/drivers/md/dm.h @@ -125,8 +125,8 @@ void dm_stripe_exit(void); int dm_open_count(struct mapped_device *md); int dm_lock_for_deletion(struct mapped_device *md); -void dm_kobject_uevent(struct mapped_device *md, enum kobject_action action, - unsigned cookie); +int dm_kobject_uevent(struct mapped_device *md, enum kobject_action action, + unsigned cookie); int dm_io_init(void); void dm_io_exit(void); diff --git a/drivers/mfd/htc-egpio.c b/drivers/mfd/htc-egpio.c index aa266e1f69b2..addb846c1e34 100644 --- a/drivers/mfd/htc-egpio.c +++ b/drivers/mfd/htc-egpio.c @@ -108,7 +108,7 @@ static void egpio_handler(unsigned int irq, struct irq_desc *desc) ack_irqs(ei); /* Process all set pins. */ readval &= ei->irqs_enabled; - for_each_bit(irqpin, &readval, ei->nirqs) { + for_each_set_bit(irqpin, &readval, ei->nirqs) { /* Run irq handler */ pr_debug("got IRQ %d\n", irqpin); irq = ei->irq_start + irqpin; diff --git a/drivers/mfd/mc13783-core.c b/drivers/mfd/mc13783-core.c index 735c8a4d164f..62a847e4c2d8 100644 --- a/drivers/mfd/mc13783-core.c +++ b/drivers/mfd/mc13783-core.c @@ -225,7 +225,7 @@ int mc13783_reg_rmw(struct mc13783 *mc13783, unsigned int offset, } EXPORT_SYMBOL(mc13783_reg_rmw); -int mc13783_mask(struct mc13783 *mc13783, int irq) +int mc13783_irq_mask(struct mc13783 *mc13783, int irq) { int ret; unsigned int offmask = irq < 24 ? MC13783_IRQMASK0 : MC13783_IRQMASK1; @@ -245,9 +245,9 @@ int mc13783_mask(struct mc13783 *mc13783, int irq) return mc13783_reg_write(mc13783, offmask, mask | irqbit); } -EXPORT_SYMBOL(mc13783_mask); +EXPORT_SYMBOL(mc13783_irq_mask); -int mc13783_unmask(struct mc13783 *mc13783, int irq) +int mc13783_irq_unmask(struct mc13783 *mc13783, int irq) { int ret; unsigned int offmask = irq < 24 ? MC13783_IRQMASK0 : MC13783_IRQMASK1; @@ -267,7 +267,53 @@ int mc13783_unmask(struct mc13783 *mc13783, int irq) return mc13783_reg_write(mc13783, offmask, mask & ~irqbit); } -EXPORT_SYMBOL(mc13783_unmask); +EXPORT_SYMBOL(mc13783_irq_unmask); + +int mc13783_irq_status(struct mc13783 *mc13783, int irq, + int *enabled, int *pending) +{ + int ret; + unsigned int offmask = irq < 24 ? MC13783_IRQMASK0 : MC13783_IRQMASK1; + unsigned int offstat = irq < 24 ? MC13783_IRQSTAT0 : MC13783_IRQSTAT1; + u32 irqbit = 1 << (irq < 24 ? irq : irq - 24); + + if (irq < 0 || irq >= MC13783_NUM_IRQ) + return -EINVAL; + + if (enabled) { + u32 mask; + + ret = mc13783_reg_read(mc13783, offmask, &mask); + if (ret) + return ret; + + *enabled = mask & irqbit; + } + + if (pending) { + u32 stat; + + ret = mc13783_reg_read(mc13783, offstat, &stat); + if (ret) + return ret; + + *pending = stat & irqbit; + } + + return 0; +} +EXPORT_SYMBOL(mc13783_irq_status); + +int mc13783_irq_ack(struct mc13783 *mc13783, int irq) +{ + unsigned int offstat = irq < 24 ? MC13783_IRQSTAT0 : MC13783_IRQSTAT1; + unsigned int val = 1 << (irq < 24 ? irq : irq - 24); + + BUG_ON(irq < 0 || irq >= MC13783_NUM_IRQ); + + return mc13783_reg_write(mc13783, offstat, val); +} +EXPORT_SYMBOL(mc13783_irq_ack); int mc13783_irq_request_nounmask(struct mc13783 *mc13783, int irq, irq_handler_t handler, const char *name, void *dev) @@ -297,7 +343,7 @@ int mc13783_irq_request(struct mc13783 *mc13783, int irq, if (ret) return ret; - ret = mc13783_unmask(mc13783, irq); + ret = mc13783_irq_unmask(mc13783, irq); if (ret) { mc13783->irqhandler[irq] = NULL; mc13783->irqdata[irq] = NULL; @@ -317,7 +363,7 @@ int mc13783_irq_free(struct mc13783 *mc13783, int irq, void *dev) mc13783->irqdata[irq] != dev) return -EINVAL; - ret = mc13783_mask(mc13783, irq); + ret = mc13783_irq_mask(mc13783, irq); if (ret) return ret; @@ -333,17 +379,6 @@ static inline irqreturn_t mc13783_irqhandler(struct mc13783 *mc13783, int irq) return mc13783->irqhandler[irq](irq, mc13783->irqdata[irq]); } -int mc13783_ackirq(struct mc13783 *mc13783, int irq) -{ - unsigned int offstat = irq < 24 ? MC13783_IRQSTAT0 : MC13783_IRQSTAT1; - unsigned int val = 1 << (irq < 24 ? irq : irq - 24); - - BUG_ON(irq < 0 || irq >= MC13783_NUM_IRQ); - - return mc13783_reg_write(mc13783, offstat, val); -} -EXPORT_SYMBOL(mc13783_ackirq); - /* * returns: number of handled irqs or negative error * locking: holds mc13783->lock @@ -422,7 +457,7 @@ static irqreturn_t mc13783_handler_adcdone(int irq, void *data) { struct mc13783_adcdone_data *adcdone_data = data; - mc13783_ackirq(adcdone_data->mc13783, irq); + mc13783_irq_ack(adcdone_data->mc13783, irq); complete_all(&adcdone_data->done); @@ -486,7 +521,7 @@ int mc13783_adc_do_conversion(struct mc13783 *mc13783, unsigned int mode, dev_dbg(&mc13783->spidev->dev, "%s: request irq\n", __func__); mc13783_irq_request(mc13783, MC13783_IRQ_ADCDONE, mc13783_handler_adcdone, __func__, &adcdone_data); - mc13783_ackirq(mc13783, MC13783_IRQ_ADCDONE); + mc13783_irq_ack(mc13783, MC13783_IRQ_ADCDONE); mc13783_reg_write(mc13783, MC13783_REG_ADC_0, adc0); mc13783_reg_write(mc13783, MC13783_REG_ADC_1, adc1); diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig index e3551d20464f..d16af6a423fb 100644 --- a/drivers/misc/Kconfig +++ b/drivers/misc/Kconfig @@ -212,6 +212,15 @@ config CS5535_MFGPT_DEFAULT_IRQ want to use a different IRQ by default. This is here for architectures to set as necessary. +config CS5535_CLOCK_EVENT_SRC + tristate "CS5535/CS5536 high-res timer (MFGPT) events" + depends on GENERIC_TIME && GENERIC_CLOCKEVENTS && CS5535_MFGPT + help + This driver provides a clock event source based on the MFGPT + timer(s) in the CS5535 and CS5536 companion chips. + MFGPTs have a better resolution and max interval than the + generic PIT, and are suitable for use as high-res timers. + config HP_ILO tristate "Channel interface driver for HP iLO/iLO2 processor" depends on PCI diff --git a/drivers/misc/iwmc3200top/main.c b/drivers/misc/iwmc3200top/main.c index dd0a3913bf6d..3b7292a5cea9 100644 --- a/drivers/misc/iwmc3200top/main.c +++ b/drivers/misc/iwmc3200top/main.c @@ -597,8 +597,6 @@ static void iwmct_remove(struct sdio_func *func) struct iwmct_work_struct *read_req; struct iwmct_priv *priv = sdio_get_drvdata(func); - priv = sdio_get_drvdata(func); - LOG_INFO(priv, INIT, "enter\n"); sdio_claim_host(func); diff --git a/drivers/misc/lkdtm.c b/drivers/misc/lkdtm.c index 3648b23d5c92..4a0648301fdf 100644 --- a/drivers/misc/lkdtm.c +++ b/drivers/misc/lkdtm.c @@ -26,21 +26,9 @@ * It is adapted from the Linux Kernel Dump Test Tool by * Fernando Luis Vazquez Cao <http://lkdtt.sourceforge.net> * - * Usage : insmod lkdtm.ko [recur_count={>0}] cpoint_name=<> cpoint_type=<> - * [cpoint_count={>0}] + * Debugfs support added by Simon Kagstrom <simon.kagstrom@netinsight.net> * - * recur_count : Recursion level for the stack overflow test. Default is 10. - * - * cpoint_name : Crash point where the kernel is to be crashed. It can be - * one of INT_HARDWARE_ENTRY, INT_HW_IRQ_EN, INT_TASKLET_ENTRY, - * FS_DEVRW, MEM_SWAPOUT, TIMERADD, SCSI_DISPATCH_CMD, - * IDE_CORE_CP - * - * cpoint_type : Indicates the action to be taken on hitting the crash point. - * It can be one of PANIC, BUG, EXCEPTION, LOOP, OVERFLOW - * - * cpoint_count : Indicates the number of times the crash point is to be hit - * to trigger an action. The default is 10. + * See Documentation/fault-injection/provoke-crashes.txt for instructions */ #include <linux/kernel.h> @@ -53,13 +41,12 @@ #include <linux/interrupt.h> #include <linux/hrtimer.h> #include <scsi/scsi_cmnd.h> +#include <linux/debugfs.h> #ifdef CONFIG_IDE #include <linux/ide.h> #endif -#define NUM_CPOINTS 8 -#define NUM_CPOINT_TYPES 5 #define DEFAULT_COUNT 10 #define REC_NUM_DEFAULT 10 @@ -72,7 +59,8 @@ enum cname { MEM_SWAPOUT, TIMERADD, SCSI_DISPATCH_CMD, - IDE_CORE_CP + IDE_CORE_CP, + DIRECT, }; enum ctype { @@ -81,7 +69,11 @@ enum ctype { BUG, EXCEPTION, LOOP, - OVERFLOW + OVERFLOW, + CORRUPT_STACK, + UNALIGNED_LOAD_STORE_WRITE, + OVERWRITE_ALLOCATION, + WRITE_AFTER_FREE, }; static char* cp_name[] = { @@ -92,7 +84,8 @@ static char* cp_name[] = { "MEM_SWAPOUT", "TIMERADD", "SCSI_DISPATCH_CMD", - "IDE_CORE_CP" + "IDE_CORE_CP", + "DIRECT", }; static char* cp_type[] = { @@ -100,7 +93,11 @@ static char* cp_type[] = { "BUG", "EXCEPTION", "LOOP", - "OVERFLOW" + "OVERFLOW", + "CORRUPT_STACK", + "UNALIGNED_LOAD_STORE_WRITE", + "OVERWRITE_ALLOCATION", + "WRITE_AFTER_FREE", }; static struct jprobe lkdtm; @@ -193,34 +190,66 @@ int jp_generic_ide_ioctl(ide_drive_t *drive, struct file *file, } #endif +/* Return the crashpoint number or NONE if the name is invalid */ +static enum ctype parse_cp_type(const char *what, size_t count) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(cp_type); i++) { + if (!strcmp(what, cp_type[i])) + return i + 1; + } + + return NONE; +} + +static const char *cp_type_to_str(enum ctype type) +{ + if (type == NONE || type < 0 || type > ARRAY_SIZE(cp_type)) + return "None"; + + return cp_type[type - 1]; +} + +static const char *cp_name_to_str(enum cname name) +{ + if (name == INVALID || name < 0 || name > ARRAY_SIZE(cp_name)) + return "INVALID"; + + return cp_name[name - 1]; +} + + static int lkdtm_parse_commandline(void) { int i; - if (cpoint_name == NULL || cpoint_type == NULL || - cpoint_count < 1 || recur_count < 1) + if (cpoint_count < 1 || recur_count < 1) return -EINVAL; - for (i = 0; i < NUM_CPOINTS; ++i) { + count = cpoint_count; + + /* No special parameters */ + if (!cpoint_type && !cpoint_name) + return 0; + + /* Neither or both of these need to be set */ + if (!cpoint_type || !cpoint_name) + return -EINVAL; + + cptype = parse_cp_type(cpoint_type, strlen(cpoint_type)); + if (cptype == NONE) + return -EINVAL; + + for (i = 0; i < ARRAY_SIZE(cp_name); i++) { if (!strcmp(cpoint_name, cp_name[i])) { cpoint = i + 1; - break; - } - } - - for (i = 0; i < NUM_CPOINT_TYPES; ++i) { - if (!strcmp(cpoint_type, cp_type[i])) { - cptype = i + 1; - break; + return 0; } } - if (cpoint == INVALID || cptype == NONE) - return -EINVAL; - - count = cpoint_count; - - return 0; + /* Could not find a valid crash point */ + return -EINVAL; } static int recursive_loop(int a) @@ -235,53 +264,92 @@ static int recursive_loop(int a) return recursive_loop(a); } -void lkdtm_handler(void) +static void lkdtm_do_action(enum ctype which) { - printk(KERN_INFO "lkdtm : Crash point %s of type %s hit\n", - cpoint_name, cpoint_type); - --count; + switch (which) { + case PANIC: + panic("dumptest"); + break; + case BUG: + BUG(); + break; + case EXCEPTION: + *((int *) 0) = 0; + break; + case LOOP: + for (;;) + ; + break; + case OVERFLOW: + (void) recursive_loop(0); + break; + case CORRUPT_STACK: { + volatile u32 data[8]; + volatile u32 *p = data; + + p[12] = 0x12345678; + break; + } + case UNALIGNED_LOAD_STORE_WRITE: { + static u8 data[5] __attribute__((aligned(4))) = {1, 2, + 3, 4, 5}; + u32 *p; + u32 val = 0x12345678; + + p = (u32 *)(data + 1); + if (*p == 0) + val = 0x87654321; + *p = val; + break; + } + case OVERWRITE_ALLOCATION: { + size_t len = 1020; + u32 *data = kmalloc(len, GFP_KERNEL); + + data[1024 / sizeof(u32)] = 0x12345678; + kfree(data); + break; + } + case WRITE_AFTER_FREE: { + size_t len = 1024; + u32 *data = kmalloc(len, GFP_KERNEL); + + kfree(data); + schedule(); + memset(data, 0x78, len); + break; + } + case NONE: + default: + break; + } + +} + +static void lkdtm_handler(void) +{ + count--; + printk(KERN_INFO "lkdtm: Crash point %s of type %s hit, trigger in %d rounds\n", + cp_name_to_str(cpoint), cp_type_to_str(cptype), count); if (count == 0) { - switch (cptype) { - case NONE: - break; - case PANIC: - printk(KERN_INFO "lkdtm : PANIC\n"); - panic("dumptest"); - break; - case BUG: - printk(KERN_INFO "lkdtm : BUG\n"); - BUG(); - break; - case EXCEPTION: - printk(KERN_INFO "lkdtm : EXCEPTION\n"); - *((int *) 0) = 0; - break; - case LOOP: - printk(KERN_INFO "lkdtm : LOOP\n"); - for (;;); - break; - case OVERFLOW: - printk(KERN_INFO "lkdtm : OVERFLOW\n"); - (void) recursive_loop(0); - break; - default: - break; - } + lkdtm_do_action(cptype); count = cpoint_count; } } -static int __init lkdtm_module_init(void) +static int lkdtm_register_cpoint(enum cname which) { int ret; - if (lkdtm_parse_commandline() == -EINVAL) { - printk(KERN_INFO "lkdtm : Invalid command\n"); - return -EINVAL; - } + cpoint = INVALID; + if (lkdtm.entry != NULL) + unregister_jprobe(&lkdtm); - switch (cpoint) { + switch (which) { + case DIRECT: + lkdtm_do_action(cptype); + return 0; case INT_HARDWARE_ENTRY: lkdtm.kp.symbol_name = "do_IRQ"; lkdtm.entry = (kprobe_opcode_t*) jp_do_irq; @@ -315,28 +383,268 @@ static int __init lkdtm_module_init(void) lkdtm.kp.symbol_name = "generic_ide_ioctl"; lkdtm.entry = (kprobe_opcode_t*) jp_generic_ide_ioctl; #else - printk(KERN_INFO "lkdtm : Crash point not available\n"); + printk(KERN_INFO "lkdtm: Crash point not available\n"); + return -EINVAL; #endif break; default: - printk(KERN_INFO "lkdtm : Invalid Crash Point\n"); - break; + printk(KERN_INFO "lkdtm: Invalid Crash Point\n"); + return -EINVAL; } + cpoint = which; if ((ret = register_jprobe(&lkdtm)) < 0) { - printk(KERN_INFO "lkdtm : Couldn't register jprobe\n"); - return ret; + printk(KERN_INFO "lkdtm: Couldn't register jprobe\n"); + cpoint = INVALID; + } + + return ret; +} + +static ssize_t do_register_entry(enum cname which, struct file *f, + const char __user *user_buf, size_t count, loff_t *off) +{ + char *buf; + int err; + + if (count >= PAGE_SIZE) + return -EINVAL; + + buf = (char *)__get_free_page(GFP_KERNEL); + if (!buf) + return -ENOMEM; + if (copy_from_user(buf, user_buf, count)) { + free_page((unsigned long) buf); + return -EFAULT; + } + /* NULL-terminate and remove enter */ + buf[count] = '\0'; + strim(buf); + + cptype = parse_cp_type(buf, count); + free_page((unsigned long) buf); + + if (cptype == NONE) + return -EINVAL; + + err = lkdtm_register_cpoint(which); + if (err < 0) + return err; + + *off += count; + + return count; +} + +/* Generic read callback that just prints out the available crash types */ +static ssize_t lkdtm_debugfs_read(struct file *f, char __user *user_buf, + size_t count, loff_t *off) +{ + char *buf; + int i, n, out; + + buf = (char *)__get_free_page(GFP_KERNEL); + + n = snprintf(buf, PAGE_SIZE, "Available crash types:\n"); + for (i = 0; i < ARRAY_SIZE(cp_type); i++) + n += snprintf(buf + n, PAGE_SIZE - n, "%s\n", cp_type[i]); + buf[n] = '\0'; + + out = simple_read_from_buffer(user_buf, count, off, + buf, n); + free_page((unsigned long) buf); + + return out; +} + +static int lkdtm_debugfs_open(struct inode *inode, struct file *file) +{ + return 0; +} + + +static ssize_t int_hardware_entry(struct file *f, const char __user *buf, + size_t count, loff_t *off) +{ + return do_register_entry(INT_HARDWARE_ENTRY, f, buf, count, off); +} + +static ssize_t int_hw_irq_en(struct file *f, const char __user *buf, + size_t count, loff_t *off) +{ + return do_register_entry(INT_HW_IRQ_EN, f, buf, count, off); +} + +static ssize_t int_tasklet_entry(struct file *f, const char __user *buf, + size_t count, loff_t *off) +{ + return do_register_entry(INT_TASKLET_ENTRY, f, buf, count, off); +} + +static ssize_t fs_devrw_entry(struct file *f, const char __user *buf, + size_t count, loff_t *off) +{ + return do_register_entry(FS_DEVRW, f, buf, count, off); +} + +static ssize_t mem_swapout_entry(struct file *f, const char __user *buf, + size_t count, loff_t *off) +{ + return do_register_entry(MEM_SWAPOUT, f, buf, count, off); +} + +static ssize_t timeradd_entry(struct file *f, const char __user *buf, + size_t count, loff_t *off) +{ + return do_register_entry(TIMERADD, f, buf, count, off); +} + +static ssize_t scsi_dispatch_cmd_entry(struct file *f, + const char __user *buf, size_t count, loff_t *off) +{ + return do_register_entry(SCSI_DISPATCH_CMD, f, buf, count, off); +} + +static ssize_t ide_core_cp_entry(struct file *f, const char __user *buf, + size_t count, loff_t *off) +{ + return do_register_entry(IDE_CORE_CP, f, buf, count, off); +} + +/* Special entry to just crash directly. Available without KPROBEs */ +static ssize_t direct_entry(struct file *f, const char __user *user_buf, + size_t count, loff_t *off) +{ + enum ctype type; + char *buf; + + if (count >= PAGE_SIZE) + return -EINVAL; + if (count < 1) + return -EINVAL; + + buf = (char *)__get_free_page(GFP_KERNEL); + if (!buf) + return -ENOMEM; + if (copy_from_user(buf, user_buf, count)) { + free_page((unsigned long) buf); + return -EFAULT; + } + /* NULL-terminate and remove enter */ + buf[count] = '\0'; + strim(buf); + + type = parse_cp_type(buf, count); + free_page((unsigned long) buf); + if (type == NONE) + return -EINVAL; + + printk(KERN_INFO "lkdtm: Performing direct entry %s\n", + cp_type_to_str(type)); + lkdtm_do_action(type); + *off += count; + + return count; +} + +struct crash_entry { + const char *name; + const struct file_operations fops; +}; + +static const struct crash_entry crash_entries[] = { + {"DIRECT", {.read = lkdtm_debugfs_read, + .open = lkdtm_debugfs_open, + .write = direct_entry} }, + {"INT_HARDWARE_ENTRY", {.read = lkdtm_debugfs_read, + .open = lkdtm_debugfs_open, + .write = int_hardware_entry} }, + {"INT_HW_IRQ_EN", {.read = lkdtm_debugfs_read, + .open = lkdtm_debugfs_open, + .write = int_hw_irq_en} }, + {"INT_TASKLET_ENTRY", {.read = lkdtm_debugfs_read, + .open = lkdtm_debugfs_open, + .write = int_tasklet_entry} }, + {"FS_DEVRW", {.read = lkdtm_debugfs_read, + .open = lkdtm_debugfs_open, + .write = fs_devrw_entry} }, + {"MEM_SWAPOUT", {.read = lkdtm_debugfs_read, + .open = lkdtm_debugfs_open, + .write = mem_swapout_entry} }, + {"TIMERADD", {.read = lkdtm_debugfs_read, + .open = lkdtm_debugfs_open, + .write = timeradd_entry} }, + {"SCSI_DISPATCH_CMD", {.read = lkdtm_debugfs_read, + .open = lkdtm_debugfs_open, + .write = scsi_dispatch_cmd_entry} }, + {"IDE_CORE_CP", {.read = lkdtm_debugfs_read, + .open = lkdtm_debugfs_open, + .write = ide_core_cp_entry} }, +}; + +static struct dentry *lkdtm_debugfs_root; + +static int __init lkdtm_module_init(void) +{ + int ret = -EINVAL; + int n_debugfs_entries = 1; /* Assume only the direct entry */ + int i; + + /* Register debugfs interface */ + lkdtm_debugfs_root = debugfs_create_dir("provoke-crash", NULL); + if (!lkdtm_debugfs_root) { + printk(KERN_ERR "lkdtm: creating root dir failed\n"); + return -ENODEV; + } + +#ifdef CONFIG_KPROBES + n_debugfs_entries = ARRAY_SIZE(crash_entries); +#endif + + for (i = 0; i < n_debugfs_entries; i++) { + const struct crash_entry *cur = &crash_entries[i]; + struct dentry *de; + + de = debugfs_create_file(cur->name, 0644, lkdtm_debugfs_root, + NULL, &cur->fops); + if (de == NULL) { + printk(KERN_ERR "lkdtm: could not create %s\n", + cur->name); + goto out_err; + } + } + + if (lkdtm_parse_commandline() == -EINVAL) { + printk(KERN_INFO "lkdtm: Invalid command\n"); + goto out_err; + } + + if (cpoint != INVALID && cptype != NONE) { + ret = lkdtm_register_cpoint(cpoint); + if (ret < 0) { + printk(KERN_INFO "lkdtm: Invalid crash point %d\n", + cpoint); + goto out_err; + } + printk(KERN_INFO "lkdtm: Crash point %s of type %s registered\n", + cpoint_name, cpoint_type); + } else { + printk(KERN_INFO "lkdtm: No crash points registered, enable through debugfs\n"); } - printk(KERN_INFO "lkdtm : Crash point %s of type %s registered\n", - cpoint_name, cpoint_type); return 0; + +out_err: + debugfs_remove_recursive(lkdtm_debugfs_root); + return ret; } static void __exit lkdtm_module_exit(void) { - unregister_jprobe(&lkdtm); - printk(KERN_INFO "lkdtm : Crash point unregistered\n"); + debugfs_remove_recursive(lkdtm_debugfs_root); + + unregister_jprobe(&lkdtm); + printk(KERN_INFO "lkdtm: Crash point unregistered\n"); } module_init(lkdtm_module_init); diff --git a/drivers/misc/sgi-xp/xpnet.c b/drivers/misc/sgi-xp/xpnet.c index 16f0abda1423..57b152f8d1b9 100644 --- a/drivers/misc/sgi-xp/xpnet.c +++ b/drivers/misc/sgi-xp/xpnet.c @@ -475,7 +475,7 @@ xpnet_dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) if (skb->data[0] == 0xff) { /* we are being asked to broadcast to all partitions */ - for_each_bit(dest_partid, xpnet_broadcast_partitions, + for_each_set_bit(dest_partid, xpnet_broadcast_partitions, xp_max_npartitions) { xpnet_send(skb, queued_msg, start_addr, end_addr, diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c index 30acd5265821..f4b97d3c3d0f 100644 --- a/drivers/mmc/core/core.c +++ b/drivers/mmc/core/core.c @@ -1151,6 +1151,9 @@ void mmc_stop_host(struct mmc_host *host) cancel_delayed_work(&host->detect); mmc_flush_scheduled_work(); + /* clear pm flags now and let card drivers set them as needed */ + host->pm_flags = 0; + mmc_bus_get(host); if (host->bus_ops && !host->bus_dead) { if (host->bus_ops->remove) @@ -1273,12 +1276,13 @@ int mmc_suspend_host(struct mmc_host *host, pm_message_t state) mmc_claim_host(host); mmc_detach_bus(host); mmc_release_host(host); + host->pm_flags = 0; err = 0; } } mmc_bus_put(host); - if (!err) + if (!err && !(host->pm_flags & MMC_PM_KEEP_POWER)) mmc_power_off(host); return err; @@ -1296,8 +1300,10 @@ int mmc_resume_host(struct mmc_host *host) mmc_bus_get(host); if (host->bus_ops && !host->bus_dead) { - mmc_power_up(host); - mmc_select_voltage(host, host->ocr); + if (!(host->pm_flags & MMC_PM_KEEP_POWER)) { + mmc_power_up(host); + mmc_select_voltage(host, host->ocr); + } BUG_ON(!host->bus_ops->resume); err = host->bus_ops->resume(host); if (err) { diff --git a/drivers/mmc/core/sdio.c b/drivers/mmc/core/sdio.c index 06b64085a355..2dd4cfe7ca17 100644 --- a/drivers/mmc/core/sdio.c +++ b/drivers/mmc/core/sdio.c @@ -188,6 +188,40 @@ static int sdio_disable_cd(struct mmc_card *card) } /* + * Devices that remain active during a system suspend are + * put back into 1-bit mode. + */ +static int sdio_disable_wide(struct mmc_card *card) +{ + int ret; + u8 ctrl; + + if (!(card->host->caps & MMC_CAP_4_BIT_DATA)) + return 0; + + if (card->cccr.low_speed && !card->cccr.wide_bus) + return 0; + + ret = mmc_io_rw_direct(card, 0, 0, SDIO_CCCR_IF, 0, &ctrl); + if (ret) + return ret; + + if (!(ctrl & SDIO_BUS_WIDTH_4BIT)) + return 0; + + ctrl &= ~SDIO_BUS_WIDTH_4BIT; + ctrl |= SDIO_BUS_ASYNC_INT; + + ret = mmc_io_rw_direct(card, 1, 0, SDIO_CCCR_IF, ctrl, NULL); + if (ret) + return ret; + + mmc_set_bus_width(card->host, MMC_BUS_WIDTH_1); + + return 0; +} + +/* * Test if the card supports high-speed mode and, if so, switch to it. */ static int sdio_enable_hs(struct mmc_card *card) @@ -224,7 +258,7 @@ static int sdio_enable_hs(struct mmc_card *card) * we're trying to reinitialise. */ static int mmc_sdio_init_card(struct mmc_host *host, u32 ocr, - struct mmc_card *oldcard) + struct mmc_card *oldcard, int powered_resume) { struct mmc_card *card; int err; @@ -235,9 +269,11 @@ static int mmc_sdio_init_card(struct mmc_host *host, u32 ocr, /* * Inform the card of the voltage */ - err = mmc_send_io_op_cond(host, host->ocr, &ocr); - if (err) - goto err; + if (!powered_resume) { + err = mmc_send_io_op_cond(host, host->ocr, &ocr); + if (err) + goto err; + } /* * For SPI, enable CRC as appropriate. @@ -262,7 +298,7 @@ static int mmc_sdio_init_card(struct mmc_host *host, u32 ocr, /* * For native busses: set card RCA and quit open drain mode. */ - if (!mmc_host_is_spi(host)) { + if (!powered_resume && !mmc_host_is_spi(host)) { err = mmc_send_relative_addr(host, &card->rca); if (err) goto remove; @@ -273,7 +309,7 @@ static int mmc_sdio_init_card(struct mmc_host *host, u32 ocr, /* * Select card, as all following commands rely on that. */ - if (!mmc_host_is_spi(host)) { + if (!powered_resume && !mmc_host_is_spi(host)) { err = mmc_select_card(card); if (err) goto remove; @@ -425,6 +461,12 @@ static int mmc_sdio_suspend(struct mmc_host *host) } } + if (!err && host->pm_flags & MMC_PM_KEEP_POWER) { + mmc_claim_host(host); + sdio_disable_wide(host->card); + mmc_release_host(host); + } + return err; } @@ -437,7 +479,13 @@ static int mmc_sdio_resume(struct mmc_host *host) /* Basic card reinitialization. */ mmc_claim_host(host); - err = mmc_sdio_init_card(host, host->ocr, host->card); + err = mmc_sdio_init_card(host, host->ocr, host->card, + (host->pm_flags & MMC_PM_KEEP_POWER)); + if (!err) + /* We may have switched to 1-bit mode during suspend. */ + err = sdio_enable_wide(host->card); + if (!err && host->sdio_irqs) + mmc_signal_sdio_irq(host); mmc_release_host(host); /* @@ -507,7 +555,7 @@ int mmc_attach_sdio(struct mmc_host *host, u32 ocr) /* * Detect and init the card. */ - err = mmc_sdio_init_card(host, host->ocr, NULL); + err = mmc_sdio_init_card(host, host->ocr, NULL, 0); if (err) goto err; card = host->card; diff --git a/drivers/mmc/core/sdio_io.c b/drivers/mmc/core/sdio_io.c index f9aa8a7deffa..ff27c8c71355 100644 --- a/drivers/mmc/core/sdio_io.c +++ b/drivers/mmc/core/sdio_io.c @@ -189,7 +189,12 @@ static inline unsigned int sdio_max_byte_size(struct sdio_func *func) { unsigned mval = min(func->card->host->max_seg_size, func->card->host->max_blk_size); - mval = min(mval, func->max_blksize); + + if (mmc_blksz_for_byte_mode(func->card)) + mval = min(mval, func->cur_blksize); + else + mval = min(mval, func->max_blksize); + return min(mval, 512u); /* maximum size for byte mode */ } @@ -635,3 +640,52 @@ void sdio_f0_writeb(struct sdio_func *func, unsigned char b, unsigned int addr, *err_ret = ret; } EXPORT_SYMBOL_GPL(sdio_f0_writeb); + +/** + * sdio_get_host_pm_caps - get host power management capabilities + * @func: SDIO function attached to host + * + * Returns a capability bitmask corresponding to power management + * features supported by the host controller that the card function + * might rely upon during a system suspend. The host doesn't need + * to be claimed, nor the function active, for this information to be + * obtained. + */ +mmc_pm_flag_t sdio_get_host_pm_caps(struct sdio_func *func) +{ + BUG_ON(!func); + BUG_ON(!func->card); + + return func->card->host->pm_caps; +} +EXPORT_SYMBOL_GPL(sdio_get_host_pm_caps); + +/** + * sdio_set_host_pm_flags - set wanted host power management capabilities + * @func: SDIO function attached to host + * + * Set a capability bitmask corresponding to wanted host controller + * power management features for the upcoming suspend state. + * This must be called, if needed, each time the suspend method of + * the function driver is called, and must contain only bits that + * were returned by sdio_get_host_pm_caps(). + * The host doesn't need to be claimed, nor the function active, + * for this information to be set. + */ +int sdio_set_host_pm_flags(struct sdio_func *func, mmc_pm_flag_t flags) +{ + struct mmc_host *host; + + BUG_ON(!func); + BUG_ON(!func->card); + + host = func->card->host; + + if (flags & ~host->pm_caps) + return -EINVAL; + + /* function suspend methods are serialized, hence no lock needed */ + host->pm_flags |= flags; + return 0; +} +EXPORT_SYMBOL_GPL(sdio_set_host_pm_flags); diff --git a/drivers/mmc/host/Kconfig b/drivers/mmc/host/Kconfig index ce1d28884e29..7b431bbab7f1 100644 --- a/drivers/mmc/host/Kconfig +++ b/drivers/mmc/host/Kconfig @@ -69,20 +69,16 @@ config MMC_SDHCI_PCI If unsure, say N. config MMC_RICOH_MMC - tristate "Ricoh MMC Controller Disabler (EXPERIMENTAL)" + bool "Ricoh MMC Controller Disabler (EXPERIMENTAL)" depends on MMC_SDHCI_PCI help - This selects the disabler for the Ricoh MMC Controller. This + This adds a pci quirk to disable Ricoh MMC Controller. This proprietary controller is unnecessary because the SDHCI driver supports MMC cards on the SD controller, but if it is not disabled, it will steal the MMC cards away - rendering them - useless. It is safe to select this driver even if you don't + useless. It is safe to select this even if you don't have a Ricoh based card reader. - - To compile this driver as a module, choose M here: - the module will be called ricoh_mmc. - If unsure, say Y. config MMC_SDHCI_OF @@ -193,6 +189,7 @@ config MMC_AU1X choice prompt "Atmel SD/MMC Driver" + depends on AVR32 || ARCH_AT91 default MMC_ATMELMCI if AVR32 help Choose which driver to use for the Atmel MCI Silicon @@ -399,7 +396,7 @@ config MMC_VIA_SDMMC config SDH_BFIN tristate "Blackfin Secure Digital Host support" - depends on MMC && ((BF54x && !BF544) || (BF51x && !BF512)) + depends on (BF54x && !BF544) || (BF51x && !BF512) help If you say yes here you will get support for the Blackfin on-chip Secure Digital Host interface. This includes support for MMC and diff --git a/drivers/mmc/host/Makefile b/drivers/mmc/host/Makefile index 3d253dd4240f..f4803977dfce 100644 --- a/drivers/mmc/host/Makefile +++ b/drivers/mmc/host/Makefile @@ -12,7 +12,6 @@ obj-$(CONFIG_MMC_IMX) += imxmmc.o obj-$(CONFIG_MMC_MXC) += mxcmmc.o obj-$(CONFIG_MMC_SDHCI) += sdhci.o obj-$(CONFIG_MMC_SDHCI_PCI) += sdhci-pci.o -obj-$(CONFIG_MMC_RICOH_MMC) += ricoh_mmc.o obj-$(CONFIG_MMC_SDHCI_PLTFM) += sdhci-pltfm.o obj-$(CONFIG_MMC_SDHCI_S3C) += sdhci-s3c.o obj-$(CONFIG_MMC_WBSD) += wbsd.o diff --git a/drivers/mmc/host/at91_mci.c b/drivers/mmc/host/at91_mci.c index 63924e0c7ea9..91dc60cd032b 100644 --- a/drivers/mmc/host/at91_mci.c +++ b/drivers/mmc/host/at91_mci.c @@ -78,6 +78,17 @@ #define DRIVER_NAME "at91_mci" +static inline int at91mci_is_mci1rev2xx(void) +{ + return ( cpu_is_at91sam9260() + || cpu_is_at91sam9263() + || cpu_is_at91cap9() + || cpu_is_at91sam9rl() + || cpu_is_at91sam9g10() + || cpu_is_at91sam9g20() + ); +} + #define FL_SENT_COMMAND (1 << 0) #define FL_SENT_STOP (1 << 1) @@ -88,6 +99,10 @@ #define at91_mci_read(host, reg) __raw_readl((host)->baseaddr + (reg)) #define at91_mci_write(host, reg, val) __raw_writel((val), (host)->baseaddr + (reg)) +#define MCI_BLKSIZE 512 +#define MCI_MAXBLKSIZE 4095 +#define MCI_BLKATONCE 256 +#define MCI_BUFSIZE (MCI_BLKSIZE * MCI_BLKATONCE) /* * Low level type for this driver @@ -200,8 +215,8 @@ static inline void at91_mci_sg_to_dma(struct at91mci_host *host, struct mmc_data size = data->blksz * data->blocks; len = data->sg_len; - /* AT91SAM926[0/3] Data Write Operation and number of bytes erratum */ - if (cpu_is_at91sam9260() || cpu_is_at91sam9263()) + /* MCI1 rev2xx Data Write Operation and number of bytes erratum */ + if (at91mci_is_mci1rev2xx()) if (host->total_length == 12) memset(dmabuf, 0, 12); @@ -227,8 +242,10 @@ static inline void at91_mci_sg_to_dma(struct at91mci_host *host, struct mmc_data for (index = 0; index < (amount / 4); index++) *dmabuf++ = swab32(sgbuffer[index]); } else { - memcpy(dmabuf, sgbuffer, amount); - dmabuf += amount; + char *tmpv = (char *)dmabuf; + memcpy(tmpv, sgbuffer, amount); + tmpv += amount; + dmabuf = (unsigned *)tmpv; } kunmap_atomic(sgbuffer, KM_BIO_SRC_IRQ); @@ -245,80 +262,14 @@ static inline void at91_mci_sg_to_dma(struct at91mci_host *host, struct mmc_data } /* - * Prepare a dma read - */ -static void at91_mci_pre_dma_read(struct at91mci_host *host) -{ - int i; - struct scatterlist *sg; - struct mmc_command *cmd; - struct mmc_data *data; - - pr_debug("pre dma read\n"); - - cmd = host->cmd; - if (!cmd) { - pr_debug("no command\n"); - return; - } - - data = cmd->data; - if (!data) { - pr_debug("no data\n"); - return; - } - - for (i = 0; i < 2; i++) { - /* nothing left to transfer */ - if (host->transfer_index >= data->sg_len) { - pr_debug("Nothing left to transfer (index = %d)\n", host->transfer_index); - break; - } - - /* Check to see if this needs filling */ - if (i == 0) { - if (at91_mci_read(host, ATMEL_PDC_RCR) != 0) { - pr_debug("Transfer active in current\n"); - continue; - } - } - else { - if (at91_mci_read(host, ATMEL_PDC_RNCR) != 0) { - pr_debug("Transfer active in next\n"); - continue; - } - } - - /* Setup the next transfer */ - pr_debug("Using transfer index %d\n", host->transfer_index); - - sg = &data->sg[host->transfer_index++]; - pr_debug("sg = %p\n", sg); - - sg->dma_address = dma_map_page(NULL, sg_page(sg), sg->offset, sg->length, DMA_FROM_DEVICE); - - pr_debug("dma address = %08X, length = %d\n", sg->dma_address, sg->length); - - if (i == 0) { - at91_mci_write(host, ATMEL_PDC_RPR, sg->dma_address); - at91_mci_write(host, ATMEL_PDC_RCR, (data->blksz & 0x3) ? sg->length : sg->length / 4); - } - else { - at91_mci_write(host, ATMEL_PDC_RNPR, sg->dma_address); - at91_mci_write(host, ATMEL_PDC_RNCR, (data->blksz & 0x3) ? sg->length : sg->length / 4); - } - } - - pr_debug("pre dma read done\n"); -} - -/* * Handle after a dma read */ static void at91_mci_post_dma_read(struct at91mci_host *host) { struct mmc_command *cmd; struct mmc_data *data; + unsigned int len, i, size; + unsigned *dmabuf = host->buffer; pr_debug("post dma read\n"); @@ -334,42 +285,39 @@ static void at91_mci_post_dma_read(struct at91mci_host *host) return; } - while (host->in_use_index < host->transfer_index) { - struct scatterlist *sg; + size = data->blksz * data->blocks; + len = data->sg_len; - pr_debug("finishing index %d\n", host->in_use_index); + at91_mci_write(host, AT91_MCI_IDR, AT91_MCI_ENDRX); + at91_mci_write(host, AT91_MCI_IER, AT91_MCI_RXBUFF); - sg = &data->sg[host->in_use_index++]; + for (i = 0; i < len; i++) { + struct scatterlist *sg; + int amount; + unsigned int *sgbuffer; - pr_debug("Unmapping page %08X\n", sg->dma_address); + sg = &data->sg[i]; - dma_unmap_page(NULL, sg->dma_address, sg->length, DMA_FROM_DEVICE); + sgbuffer = kmap_atomic(sg_page(sg), KM_BIO_SRC_IRQ) + sg->offset; + amount = min(size, sg->length); + size -= amount; if (cpu_is_at91rm9200()) { /* AT91RM9200 errata */ - unsigned int *buffer; int index; - - /* Swap the contents of the buffer */ - buffer = kmap_atomic(sg_page(sg), KM_BIO_SRC_IRQ) + sg->offset; - pr_debug("buffer = %p, length = %d\n", buffer, sg->length); - - for (index = 0; index < (sg->length / 4); index++) - buffer[index] = swab32(buffer[index]); - - kunmap_atomic(buffer, KM_BIO_SRC_IRQ); + for (index = 0; index < (amount / 4); index++) + sgbuffer[index] = swab32(*dmabuf++); + } else { + char *tmpv = (char *)dmabuf; + memcpy(sgbuffer, tmpv, amount); + tmpv += amount; + dmabuf = (unsigned *)tmpv; } - flush_dcache_page(sg_page(sg)); - - data->bytes_xfered += sg->length; - } - - /* Is there another transfer to trigger? */ - if (host->transfer_index < data->sg_len) - at91_mci_pre_dma_read(host); - else { - at91_mci_write(host, AT91_MCI_IDR, AT91_MCI_ENDRX); - at91_mci_write(host, AT91_MCI_IER, AT91_MCI_RXBUFF); + kunmap_atomic(sgbuffer, KM_BIO_SRC_IRQ); + dmac_flush_range((void *)sgbuffer, ((void *)sgbuffer) + amount); + data->bytes_xfered += amount; + if (size == 0) + break; } pr_debug("post dma read done\n"); @@ -461,7 +409,7 @@ static void at91_mci_enable(struct at91mci_host *host) at91_mci_write(host, AT91_MCI_DTOR, AT91_MCI_DTOMUL_1M | AT91_MCI_DTOCYC); mr = AT91_MCI_PDCMODE | 0x34a; - if (cpu_is_at91sam9260() || cpu_is_at91sam9263()) + if (at91mci_is_mci1rev2xx()) mr |= AT91_MCI_RDPROOF | AT91_MCI_WRPROOF; at91_mci_write(host, AT91_MCI_MR, mr); @@ -602,10 +550,14 @@ static void at91_mci_send_command(struct at91mci_host *host, struct mmc_command /* * Handle a read */ - host->buffer = NULL; host->total_length = 0; - at91_mci_pre_dma_read(host); + at91_mci_write(host, ATMEL_PDC_RPR, host->physical_address); + at91_mci_write(host, ATMEL_PDC_RCR, (data->blksz & 0x3) ? + (blocks * block_length) : (blocks * block_length) / 4); + at91_mci_write(host, ATMEL_PDC_RNPR, 0); + at91_mci_write(host, ATMEL_PDC_RNCR, 0); + ier = AT91_MCI_ENDRX /* | AT91_MCI_RXBUFF */; } else { @@ -614,27 +566,15 @@ static void at91_mci_send_command(struct at91mci_host *host, struct mmc_command */ host->total_length = block_length * blocks; /* - * AT91SAM926[0/3] Data Write Operation and + * MCI1 rev2xx Data Write Operation and * number of bytes erratum */ - if (cpu_is_at91sam9260 () || cpu_is_at91sam9263()) + if (at91mci_is_mci1rev2xx()) if (host->total_length < 12) host->total_length = 12; - host->buffer = kmalloc(host->total_length, GFP_KERNEL); - if (!host->buffer) { - pr_debug("Can't alloc tx buffer\n"); - cmd->error = -ENOMEM; - mmc_request_done(host->mmc, host->request); - return; - } - at91_mci_sg_to_dma(host, data); - host->physical_address = dma_map_single(NULL, - host->buffer, host->total_length, - DMA_TO_DEVICE); - pr_debug("Transmitting %d bytes\n", host->total_length); at91_mci_write(host, ATMEL_PDC_TPR, host->physical_address); @@ -701,14 +641,6 @@ static void at91_mci_completed_command(struct at91mci_host *host, unsigned int s cmd->resp[2] = at91_mci_read(host, AT91_MCI_RSPR(2)); cmd->resp[3] = at91_mci_read(host, AT91_MCI_RSPR(3)); - if (host->buffer) { - dma_unmap_single(NULL, - host->physical_address, host->total_length, - DMA_TO_DEVICE); - kfree(host->buffer); - host->buffer = NULL; - } - pr_debug("Status = %08X/%08x [%08X %08X %08X %08X]\n", status, at91_mci_read(host, AT91_MCI_SR), cmd->resp[0], cmd->resp[1], cmd->resp[2], cmd->resp[3]); @@ -754,7 +686,8 @@ static void at91_mci_request(struct mmc_host *mmc, struct mmc_request *mrq) host->request = mrq; host->flags = 0; - mod_timer(&host->timer, jiffies + HZ); + /* more than 1s timeout needed with slow SD cards */ + mod_timer(&host->timer, jiffies + msecs_to_jiffies(2000)); at91_mci_process_next(host); } @@ -942,7 +875,8 @@ static irqreturn_t at91_mmc_det_irq(int irq, void *_host) pr_debug("****** Resetting SD-card bus width ******\n"); at91_mci_write(host, AT91_MCI_SDCR, at91_mci_read(host, AT91_MCI_SDCR) & ~AT91_MCI_SDCBUS); } - mmc_detect_change(host->mmc, msecs_to_jiffies(100)); + /* 0.5s needed because of early card detect switch firing */ + mmc_detect_change(host->mmc, msecs_to_jiffies(500)); } return IRQ_HANDLED; } @@ -1006,24 +940,42 @@ static int __init at91_mci_probe(struct platform_device *pdev) mmc->f_min = 375000; mmc->f_max = 25000000; mmc->ocr_avail = MMC_VDD_32_33 | MMC_VDD_33_34; - mmc->caps = MMC_CAP_SDIO_IRQ; + mmc->caps = 0; - mmc->max_blk_size = 4095; - mmc->max_blk_count = mmc->max_req_size; + mmc->max_blk_size = MCI_MAXBLKSIZE; + mmc->max_blk_count = MCI_BLKATONCE; + mmc->max_req_size = MCI_BUFSIZE; + mmc->max_phys_segs = MCI_BLKATONCE; + mmc->max_hw_segs = MCI_BLKATONCE; + mmc->max_seg_size = MCI_BUFSIZE; host = mmc_priv(mmc); host->mmc = mmc; - host->buffer = NULL; host->bus_mode = 0; host->board = pdev->dev.platform_data; if (host->board->wire4) { - if (cpu_is_at91sam9260() || cpu_is_at91sam9263()) + if (at91mci_is_mci1rev2xx()) mmc->caps |= MMC_CAP_4_BIT_DATA; else dev_warn(&pdev->dev, "4 wire bus mode not supported" " - using 1 wire\n"); } + host->buffer = dma_alloc_coherent(&pdev->dev, MCI_BUFSIZE, + &host->physical_address, GFP_KERNEL); + if (!host->buffer) { + ret = -ENOMEM; + dev_err(&pdev->dev, "Can't allocate transmit buffer\n"); + goto fail5; + } + + /* Add SDIO capability when available */ + if (at91mci_is_mci1rev2xx()) { + /* at91mci MCI1 rev2xx sdio interrupt erratum */ + if (host->board->wire4 || !host->board->slot_b) + mmc->caps |= MMC_CAP_SDIO_IRQ; + } + /* * Reserve GPIOs ... board init code makes sure these pins are set * up as GPIOs with the right direction (input, except for vcc) @@ -1032,7 +984,7 @@ static int __init at91_mci_probe(struct platform_device *pdev) ret = gpio_request(host->board->det_pin, "mmc_detect"); if (ret < 0) { dev_dbg(&pdev->dev, "couldn't claim card detect pin\n"); - goto fail5; + goto fail4b; } } if (host->board->wp_pin) { @@ -1132,6 +1084,10 @@ fail3: fail4: if (host->board->det_pin) gpio_free(host->board->det_pin); +fail4b: + if (host->buffer) + dma_free_coherent(&pdev->dev, MCI_BUFSIZE, + host->buffer, host->physical_address); fail5: mmc_free_host(mmc); fail6: @@ -1154,6 +1110,10 @@ static int __exit at91_mci_remove(struct platform_device *pdev) host = mmc_priv(mmc); + if (host->buffer) + dma_free_coherent(&pdev->dev, MCI_BUFSIZE, + host->buffer, host->physical_address); + if (host->board->det_pin) { if (device_can_wakeup(&pdev->dev)) free_irq(gpio_to_irq(host->board->det_pin), host); diff --git a/drivers/mmc/host/bfin_sdh.c b/drivers/mmc/host/bfin_sdh.c index 3343a57355cc..56f7b448b911 100644 --- a/drivers/mmc/host/bfin_sdh.c +++ b/drivers/mmc/host/bfin_sdh.c @@ -115,7 +115,7 @@ static int sdh_setup_data(struct sdh_host *host, struct mmc_data *data) unsigned int length; unsigned int data_ctl; unsigned int dma_cfg; - struct scatterlist *sg; + unsigned int cycle_ns, timeout; dev_dbg(mmc_dev(host->mmc), "%s enter flags: 0x%x\n", __func__, data->flags); host->data = data; @@ -136,8 +136,11 @@ static int sdh_setup_data(struct sdh_host *host, struct mmc_data *data) data_ctl |= ((ffs(data->blksz) - 1) << 4); bfin_write_SDH_DATA_CTL(data_ctl); - - bfin_write_SDH_DATA_TIMER(0xFFFF); + /* the time of a host clock period in ns */ + cycle_ns = 1000000000 / (get_sclk() / (2 * (host->clk_div + 1))); + timeout = data->timeout_ns / cycle_ns; + timeout += data->timeout_clks; + bfin_write_SDH_DATA_TIMER(timeout); SSYNC(); if (data->flags & MMC_DATA_READ) { @@ -151,6 +154,7 @@ static int sdh_setup_data(struct sdh_host *host, struct mmc_data *data) #if defined(CONFIG_BF54x) dma_cfg |= DMAFLOW_ARRAY | NDSIZE_5 | RESTART | WDSIZE_32 | DMAEN; { + struct scatterlist *sg; int i; for_each_sg(data->sg, sg, host->dma_len, i) { host->sg_cpu[i].start_addr = sg_dma_address(sg); diff --git a/drivers/mmc/host/davinci_mmc.c b/drivers/mmc/host/davinci_mmc.c index dd45e7c3517e..3bd0ba294e9d 100644 --- a/drivers/mmc/host/davinci_mmc.c +++ b/drivers/mmc/host/davinci_mmc.c @@ -73,6 +73,7 @@ /* DAVINCI_MMCCTL definitions */ #define MMCCTL_DATRST (1 << 0) #define MMCCTL_CMDRST (1 << 1) +#define MMCCTL_WIDTH_8_BIT (1 << 8) #define MMCCTL_WIDTH_4_BIT (1 << 2) #define MMCCTL_DATEG_DISABLED (0 << 6) #define MMCCTL_DATEG_RISING (1 << 6) @@ -791,22 +792,42 @@ static void calculate_clk_divider(struct mmc_host *mmc, struct mmc_ios *ios) static void mmc_davinci_set_ios(struct mmc_host *mmc, struct mmc_ios *ios) { - unsigned int mmc_pclk = 0; struct mmc_davinci_host *host = mmc_priv(mmc); - mmc_pclk = host->mmc_input_clk; dev_dbg(mmc_dev(host->mmc), "clock %dHz busmode %d powermode %d Vdd %04x\n", ios->clock, ios->bus_mode, ios->power_mode, ios->vdd); - if (ios->bus_width == MMC_BUS_WIDTH_4) { - dev_dbg(mmc_dev(host->mmc), "Enabling 4 bit mode\n"); - writel(readl(host->base + DAVINCI_MMCCTL) | MMCCTL_WIDTH_4_BIT, - host->base + DAVINCI_MMCCTL); - } else { - dev_dbg(mmc_dev(host->mmc), "Disabling 4 bit mode\n"); - writel(readl(host->base + DAVINCI_MMCCTL) & ~MMCCTL_WIDTH_4_BIT, + + switch (ios->bus_width) { + case MMC_BUS_WIDTH_8: + dev_dbg(mmc_dev(host->mmc), "Enabling 8 bit mode\n"); + writel((readl(host->base + DAVINCI_MMCCTL) & + ~MMCCTL_WIDTH_4_BIT) | MMCCTL_WIDTH_8_BIT, host->base + DAVINCI_MMCCTL); + break; + case MMC_BUS_WIDTH_4: + dev_dbg(mmc_dev(host->mmc), "Enabling 4 bit mode\n"); + if (host->version == MMC_CTLR_VERSION_2) + writel((readl(host->base + DAVINCI_MMCCTL) & + ~MMCCTL_WIDTH_8_BIT) | MMCCTL_WIDTH_4_BIT, + host->base + DAVINCI_MMCCTL); + else + writel(readl(host->base + DAVINCI_MMCCTL) | + MMCCTL_WIDTH_4_BIT, + host->base + DAVINCI_MMCCTL); + break; + case MMC_BUS_WIDTH_1: + dev_dbg(mmc_dev(host->mmc), "Enabling 1 bit mode\n"); + if (host->version == MMC_CTLR_VERSION_2) + writel(readl(host->base + DAVINCI_MMCCTL) & + ~(MMCCTL_WIDTH_8_BIT | MMCCTL_WIDTH_4_BIT), + host->base + DAVINCI_MMCCTL); + else + writel(readl(host->base + DAVINCI_MMCCTL) & + ~MMCCTL_WIDTH_4_BIT, + host->base + DAVINCI_MMCCTL); + break; } calculate_clk_divider(mmc, ios); @@ -1189,10 +1210,14 @@ static int __init davinci_mmcsd_probe(struct platform_device *pdev) /* REVISIT: someday, support IRQ-driven card detection. */ mmc->caps |= MMC_CAP_NEEDS_POLL; + mmc->caps |= MMC_CAP_WAIT_WHILE_BUSY; - if (!pdata || pdata->wires == 4 || pdata->wires == 0) + if (pdata && (pdata->wires == 4 || pdata->wires == 0)) mmc->caps |= MMC_CAP_4_BIT_DATA; + if (pdata && (pdata->wires == 8)) + mmc->caps |= (MMC_CAP_4_BIT_DATA | MMC_CAP_8_BIT_DATA); + host->version = pdata->version; mmc->ops = &mmc_davinci_ops; diff --git a/drivers/mmc/host/ricoh_mmc.c b/drivers/mmc/host/ricoh_mmc.c deleted file mode 100644 index f62790513322..000000000000 --- a/drivers/mmc/host/ricoh_mmc.c +++ /dev/null @@ -1,262 +0,0 @@ -/* - * ricoh_mmc.c - Dummy driver to disable the Rioch MMC controller. - * - * Copyright (C) 2007 Philip Langdale, All Rights Reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or (at - * your option) any later version. - */ - -/* - * This is a conceptually ridiculous driver, but it is required by the way - * the Ricoh multi-function chips (R5CXXX) work. These chips implement - * the four main memory card controllers (SD, MMC, MS, xD) and one or both - * of cardbus or firewire. It happens that they implement SD and MMC - * support as separate controllers (and PCI functions). The linux SDHCI - * driver supports MMC cards but the chip detects MMC cards in hardware - * and directs them to the MMC controller - so the SDHCI driver never sees - * them. To get around this, we must disable the useless MMC controller. - * At that point, the SDHCI controller will start seeing them. As a bonus, - * a detection event occurs immediately, even if the MMC card is already - * in the reader. - * - * It seems to be the case that the relevant PCI registers to deactivate the - * MMC controller live on PCI function 0, which might be the cardbus controller - * or the firewire controller, depending on the particular chip in question. As - * such, it makes what this driver has to do unavoidably ugly. Such is life. - */ - -#include <linux/pci.h> - -#define DRIVER_NAME "ricoh-mmc" - -static const struct pci_device_id pci_ids[] __devinitdata = { - { - .vendor = PCI_VENDOR_ID_RICOH, - .device = PCI_DEVICE_ID_RICOH_R5C843, - .subvendor = PCI_ANY_ID, - .subdevice = PCI_ANY_ID, - }, - { /* end: all zeroes */ }, -}; - -MODULE_DEVICE_TABLE(pci, pci_ids); - -static int ricoh_mmc_disable(struct pci_dev *fw_dev) -{ - u8 write_enable; - u8 write_target; - u8 disable; - - if (fw_dev->device == PCI_DEVICE_ID_RICOH_RL5C476) { - /* via RL5C476 */ - - pci_read_config_byte(fw_dev, 0xB7, &disable); - if (disable & 0x02) { - printk(KERN_INFO DRIVER_NAME - ": Controller already disabled. " \ - "Nothing to do.\n"); - return -ENODEV; - } - - pci_read_config_byte(fw_dev, 0x8E, &write_enable); - pci_write_config_byte(fw_dev, 0x8E, 0xAA); - pci_read_config_byte(fw_dev, 0x8D, &write_target); - pci_write_config_byte(fw_dev, 0x8D, 0xB7); - pci_write_config_byte(fw_dev, 0xB7, disable | 0x02); - pci_write_config_byte(fw_dev, 0x8E, write_enable); - pci_write_config_byte(fw_dev, 0x8D, write_target); - } else { - /* via R5C832 */ - - pci_read_config_byte(fw_dev, 0xCB, &disable); - if (disable & 0x02) { - printk(KERN_INFO DRIVER_NAME - ": Controller already disabled. " \ - "Nothing to do.\n"); - return -ENODEV; - } - - pci_read_config_byte(fw_dev, 0xCA, &write_enable); - pci_write_config_byte(fw_dev, 0xCA, 0x57); - pci_write_config_byte(fw_dev, 0xCB, disable | 0x02); - pci_write_config_byte(fw_dev, 0xCA, write_enable); - } - - printk(KERN_INFO DRIVER_NAME - ": Controller is now disabled.\n"); - - return 0; -} - -static int ricoh_mmc_enable(struct pci_dev *fw_dev) -{ - u8 write_enable; - u8 write_target; - u8 disable; - - if (fw_dev->device == PCI_DEVICE_ID_RICOH_RL5C476) { - /* via RL5C476 */ - - pci_read_config_byte(fw_dev, 0x8E, &write_enable); - pci_write_config_byte(fw_dev, 0x8E, 0xAA); - pci_read_config_byte(fw_dev, 0x8D, &write_target); - pci_write_config_byte(fw_dev, 0x8D, 0xB7); - pci_read_config_byte(fw_dev, 0xB7, &disable); - pci_write_config_byte(fw_dev, 0xB7, disable & ~0x02); - pci_write_config_byte(fw_dev, 0x8E, write_enable); - pci_write_config_byte(fw_dev, 0x8D, write_target); - } else { - /* via R5C832 */ - - pci_read_config_byte(fw_dev, 0xCA, &write_enable); - pci_read_config_byte(fw_dev, 0xCB, &disable); - pci_write_config_byte(fw_dev, 0xCA, 0x57); - pci_write_config_byte(fw_dev, 0xCB, disable & ~0x02); - pci_write_config_byte(fw_dev, 0xCA, write_enable); - } - - printk(KERN_INFO DRIVER_NAME - ": Controller is now re-enabled.\n"); - - return 0; -} - -static int __devinit ricoh_mmc_probe(struct pci_dev *pdev, - const struct pci_device_id *ent) -{ - u8 rev; - u8 ctrlfound = 0; - - struct pci_dev *fw_dev = NULL; - - BUG_ON(pdev == NULL); - BUG_ON(ent == NULL); - - pci_read_config_byte(pdev, PCI_CLASS_REVISION, &rev); - - printk(KERN_INFO DRIVER_NAME - ": Ricoh MMC controller found at %s [%04x:%04x] (rev %x)\n", - pci_name(pdev), (int)pdev->vendor, (int)pdev->device, - (int)rev); - - while ((fw_dev = - pci_get_device(PCI_VENDOR_ID_RICOH, - PCI_DEVICE_ID_RICOH_RL5C476, fw_dev))) { - if (PCI_SLOT(pdev->devfn) == PCI_SLOT(fw_dev->devfn) && - PCI_FUNC(fw_dev->devfn) == 0 && - pdev->bus == fw_dev->bus) { - if (ricoh_mmc_disable(fw_dev) != 0) - return -ENODEV; - - pci_set_drvdata(pdev, fw_dev); - - ++ctrlfound; - break; - } - } - - fw_dev = NULL; - - while (!ctrlfound && - (fw_dev = pci_get_device(PCI_VENDOR_ID_RICOH, - PCI_DEVICE_ID_RICOH_R5C832, fw_dev))) { - if (PCI_SLOT(pdev->devfn) == PCI_SLOT(fw_dev->devfn) && - PCI_FUNC(fw_dev->devfn) == 0 && - pdev->bus == fw_dev->bus) { - if (ricoh_mmc_disable(fw_dev) != 0) - return -ENODEV; - - pci_set_drvdata(pdev, fw_dev); - - ++ctrlfound; - } - } - - if (!ctrlfound) { - printk(KERN_WARNING DRIVER_NAME - ": Main Ricoh function not found. Cannot disable controller.\n"); - return -ENODEV; - } - - return 0; -} - -static void __devexit ricoh_mmc_remove(struct pci_dev *pdev) -{ - struct pci_dev *fw_dev = NULL; - - fw_dev = pci_get_drvdata(pdev); - BUG_ON(fw_dev == NULL); - - ricoh_mmc_enable(fw_dev); - - pci_set_drvdata(pdev, NULL); -} - -static int ricoh_mmc_suspend_late(struct pci_dev *pdev, pm_message_t state) -{ - struct pci_dev *fw_dev = NULL; - - fw_dev = pci_get_drvdata(pdev); - BUG_ON(fw_dev == NULL); - - printk(KERN_INFO DRIVER_NAME ": Suspending.\n"); - - ricoh_mmc_enable(fw_dev); - - return 0; -} - -static int ricoh_mmc_resume_early(struct pci_dev *pdev) -{ - struct pci_dev *fw_dev = NULL; - - fw_dev = pci_get_drvdata(pdev); - BUG_ON(fw_dev == NULL); - - printk(KERN_INFO DRIVER_NAME ": Resuming.\n"); - - ricoh_mmc_disable(fw_dev); - - return 0; -} - -static struct pci_driver ricoh_mmc_driver = { - .name = DRIVER_NAME, - .id_table = pci_ids, - .probe = ricoh_mmc_probe, - .remove = __devexit_p(ricoh_mmc_remove), - .suspend_late = ricoh_mmc_suspend_late, - .resume_early = ricoh_mmc_resume_early, -}; - -/*****************************************************************************\ - * * - * Driver init/exit * - * * -\*****************************************************************************/ - -static int __init ricoh_mmc_drv_init(void) -{ - printk(KERN_INFO DRIVER_NAME - ": Ricoh MMC Controller disabling driver\n"); - printk(KERN_INFO DRIVER_NAME ": Copyright(c) Philip Langdale\n"); - - return pci_register_driver(&ricoh_mmc_driver); -} - -static void __exit ricoh_mmc_drv_exit(void) -{ - pci_unregister_driver(&ricoh_mmc_driver); -} - -module_init(ricoh_mmc_drv_init); -module_exit(ricoh_mmc_drv_exit); - -MODULE_AUTHOR("Philip Langdale <philipl@alumni.utexas.net>"); -MODULE_DESCRIPTION("Ricoh MMC Controller disabling driver"); -MODULE_LICENSE("GPL"); - diff --git a/drivers/mmc/host/s3cmci.c b/drivers/mmc/host/s3cmci.c index d96e1abf2d64..2fdf7689ae6c 100644 --- a/drivers/mmc/host/s3cmci.c +++ b/drivers/mmc/host/s3cmci.c @@ -1179,7 +1179,7 @@ static int s3cmci_card_present(struct mmc_host *mmc) struct s3c24xx_mci_pdata *pdata = host->pdata; int ret; - if (pdata->gpio_detect == 0) + if (pdata->no_detect) return -ENOSYS; ret = gpio_get_value(pdata->gpio_detect) ? 0 : 1; @@ -1360,6 +1360,8 @@ static struct mmc_host_ops s3cmci_ops = { static struct s3c24xx_mci_pdata s3cmci_def_pdata = { /* This is currently here to avoid a number of if (host->pdata) * checks. Any zero fields to ensure reasonable defaults are picked. */ + .no_wprotect = 1, + .no_detect = 1, }; #ifdef CONFIG_CPU_FREQ diff --git a/drivers/mmc/host/sdhci-pci.c b/drivers/mmc/host/sdhci-pci.c index 5c3a1767770a..8e1020cf73f4 100644 --- a/drivers/mmc/host/sdhci-pci.c +++ b/drivers/mmc/host/sdhci-pci.c @@ -80,9 +80,6 @@ struct sdhci_pci_chip { static int ricoh_probe(struct sdhci_pci_chip *chip) { - if (chip->pdev->subsystem_vendor == PCI_VENDOR_ID_IBM) - chip->quirks |= SDHCI_QUIRK_CLOCK_BEFORE_RESET; - if (chip->pdev->subsystem_vendor == PCI_VENDOR_ID_SAMSUNG || chip->pdev->subsystem_vendor == PCI_VENDOR_ID_SONY) chip->quirks |= SDHCI_QUIRK_NO_CARD_NO_RESET; @@ -92,7 +89,9 @@ static int ricoh_probe(struct sdhci_pci_chip *chip) static const struct sdhci_pci_fixes sdhci_ricoh = { .probe = ricoh_probe, - .quirks = SDHCI_QUIRK_32BIT_DMA_ADDR, + .quirks = SDHCI_QUIRK_32BIT_DMA_ADDR | + SDHCI_QUIRK_FORCE_DMA | + SDHCI_QUIRK_CLOCK_BEFORE_RESET, }; static const struct sdhci_pci_fixes sdhci_ene_712 = { @@ -501,6 +500,7 @@ static int sdhci_pci_suspend (struct pci_dev *pdev, pm_message_t state) { struct sdhci_pci_chip *chip; struct sdhci_pci_slot *slot; + mmc_pm_flag_t pm_flags = 0; int i, ret; chip = pci_get_drvdata(pdev); @@ -519,6 +519,8 @@ static int sdhci_pci_suspend (struct pci_dev *pdev, pm_message_t state) sdhci_resume_host(chip->slots[i]->host); return ret; } + + pm_flags |= slot->host->mmc->pm_flags; } if (chip->fixes && chip->fixes->suspend) { @@ -531,9 +533,15 @@ static int sdhci_pci_suspend (struct pci_dev *pdev, pm_message_t state) } pci_save_state(pdev); - pci_enable_wake(pdev, pci_choose_state(pdev, state), 0); - pci_disable_device(pdev); - pci_set_power_state(pdev, pci_choose_state(pdev, state)); + if (pm_flags & MMC_PM_KEEP_POWER) { + if (pm_flags & MMC_PM_WAKE_SDIO_IRQ) + pci_enable_wake(pdev, PCI_D3hot, 1); + pci_set_power_state(pdev, PCI_D3hot); + } else { + pci_enable_wake(pdev, pci_choose_state(pdev, state), 0); + pci_disable_device(pdev); + pci_set_power_state(pdev, pci_choose_state(pdev, state)); + } return 0; } @@ -653,6 +661,8 @@ static struct sdhci_pci_slot * __devinit sdhci_pci_probe_slot( goto unmap; } + host->mmc->pm_caps = MMC_PM_KEEP_POWER | MMC_PM_WAKE_SDIO_IRQ; + ret = sdhci_add_host(host); if (ret) goto remove; diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c index c279fbc4c2e5..d6ab62d539fb 100644 --- a/drivers/mmc/host/sdhci.c +++ b/drivers/mmc/host/sdhci.c @@ -174,20 +174,31 @@ static void sdhci_reset(struct sdhci_host *host, u8 mask) sdhci_clear_set_irqs(host, SDHCI_INT_ALL_MASK, ier); } -static void sdhci_init(struct sdhci_host *host) +static void sdhci_set_ios(struct mmc_host *mmc, struct mmc_ios *ios); + +static void sdhci_init(struct sdhci_host *host, int soft) { - sdhci_reset(host, SDHCI_RESET_ALL); + if (soft) + sdhci_reset(host, SDHCI_RESET_CMD|SDHCI_RESET_DATA); + else + sdhci_reset(host, SDHCI_RESET_ALL); sdhci_clear_set_irqs(host, SDHCI_INT_ALL_MASK, SDHCI_INT_BUS_POWER | SDHCI_INT_DATA_END_BIT | SDHCI_INT_DATA_CRC | SDHCI_INT_DATA_TIMEOUT | SDHCI_INT_INDEX | SDHCI_INT_END_BIT | SDHCI_INT_CRC | SDHCI_INT_TIMEOUT | SDHCI_INT_DATA_END | SDHCI_INT_RESPONSE); + + if (soft) { + /* force clock reconfiguration */ + host->clock = 0; + sdhci_set_ios(host->mmc, &host->mmc->ios); + } } static void sdhci_reinit(struct sdhci_host *host) { - sdhci_init(host); + sdhci_init(host, 0); sdhci_enable_card_detection(host); } @@ -376,6 +387,20 @@ static void sdhci_kunmap_atomic(void *buffer, unsigned long *flags) local_irq_restore(*flags); } +static void sdhci_set_adma_desc(u8 *desc, u32 addr, int len, unsigned cmd) +{ + __le32 *dataddr = (__le32 __force *)(desc + 4); + __le16 *cmdlen = (__le16 __force *)desc; + + /* SDHCI specification says ADMA descriptors should be 4 byte + * aligned, so using 16 or 32bit operations should be safe. */ + + cmdlen[0] = cpu_to_le16(cmd); + cmdlen[1] = cpu_to_le16(len); + + dataddr[0] = cpu_to_le32(addr); +} + static int sdhci_adma_table_pre(struct sdhci_host *host, struct mmc_data *data) { @@ -443,19 +468,11 @@ static int sdhci_adma_table_pre(struct sdhci_host *host, sdhci_kunmap_atomic(buffer, &flags); } - desc[7] = (align_addr >> 24) & 0xff; - desc[6] = (align_addr >> 16) & 0xff; - desc[5] = (align_addr >> 8) & 0xff; - desc[4] = (align_addr >> 0) & 0xff; + /* tran, valid */ + sdhci_set_adma_desc(desc, align_addr, offset, 0x21); BUG_ON(offset > 65536); - desc[3] = (offset >> 8) & 0xff; - desc[2] = (offset >> 0) & 0xff; - - desc[1] = 0x00; - desc[0] = 0x21; /* tran, valid */ - align += 4; align_addr += 4; @@ -465,19 +482,10 @@ static int sdhci_adma_table_pre(struct sdhci_host *host, len -= offset; } - desc[7] = (addr >> 24) & 0xff; - desc[6] = (addr >> 16) & 0xff; - desc[5] = (addr >> 8) & 0xff; - desc[4] = (addr >> 0) & 0xff; - BUG_ON(len > 65536); - desc[3] = (len >> 8) & 0xff; - desc[2] = (len >> 0) & 0xff; - - desc[1] = 0x00; - desc[0] = 0x21; /* tran, valid */ - + /* tran, valid */ + sdhci_set_adma_desc(desc, addr, len, 0x21); desc += 8; /* @@ -490,16 +498,9 @@ static int sdhci_adma_table_pre(struct sdhci_host *host, /* * Add a terminating entry. */ - desc[7] = 0; - desc[6] = 0; - desc[5] = 0; - desc[4] = 0; - desc[3] = 0; - desc[2] = 0; - - desc[1] = 0x00; - desc[0] = 0x03; /* nop, end, valid */ + /* nop, end, valid */ + sdhci_set_adma_desc(desc, 0, 0, 0x3); /* * Resync align buffer as we might have changed it. @@ -1610,16 +1611,13 @@ int sdhci_resume_host(struct sdhci_host *host) if (ret) return ret; - sdhci_init(host); + sdhci_init(host, (host->mmc->pm_flags & MMC_PM_KEEP_POWER)); mmiowb(); ret = mmc_resume_host(host->mmc); - if (ret) - return ret; - sdhci_enable_card_detection(host); - return 0; + return ret; } EXPORT_SYMBOL_GPL(sdhci_resume_host); @@ -1874,7 +1872,7 @@ int sdhci_add_host(struct sdhci_host *host) if (ret) goto untasklet; - sdhci_init(host); + sdhci_init(host, 0); #ifdef CONFIG_MMC_DEBUG sdhci_dumpregs(host); diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c index 14cec04c34f9..bc45ef9af17d 100644 --- a/drivers/mtd/ubi/build.c +++ b/drivers/mtd/ubi/build.c @@ -37,6 +37,7 @@ #include <linux/module.h> #include <linux/moduleparam.h> #include <linux/stringify.h> +#include <linux/namei.h> #include <linux/stat.h> #include <linux/miscdevice.h> #include <linux/log2.h> @@ -50,7 +51,8 @@ /** * struct mtd_dev_param - MTD device parameter description data structure. - * @name: MTD device name or number string + * @name: MTD character device node path, MTD device name, or MTD device number + * string * @vid_hdr_offs: VID header offset */ struct mtd_dev_param { @@ -59,10 +61,10 @@ struct mtd_dev_param { }; /* Numbers of elements set in the @mtd_dev_param array */ -static int mtd_devs; +static int __initdata mtd_devs; /* MTD devices specification parameters */ -static struct mtd_dev_param mtd_dev_param[UBI_MAX_DEVICES]; +static struct mtd_dev_param __initdata mtd_dev_param[UBI_MAX_DEVICES]; /* Root UBI "class" object (corresponds to '/<sysfs>/class/ubi/') */ struct class *ubi_class; @@ -363,11 +365,13 @@ static void dev_release(struct device *dev) /** * ubi_sysfs_init - initialize sysfs for an UBI device. * @ubi: UBI device description object + * @ref: set to %1 on exit in case of failure if a reference to @ubi->dev was + * taken * * This function returns zero in case of success and a negative error code in * case of failure. */ -static int ubi_sysfs_init(struct ubi_device *ubi) +static int ubi_sysfs_init(struct ubi_device *ubi, int *ref) { int err; @@ -379,6 +383,7 @@ static int ubi_sysfs_init(struct ubi_device *ubi) if (err) return err; + *ref = 1; err = device_create_file(&ubi->dev, &dev_eraseblock_size); if (err) return err; @@ -434,7 +439,7 @@ static void ubi_sysfs_close(struct ubi_device *ubi) } /** - * kill_volumes - destroy all volumes. + * kill_volumes - destroy all user volumes. * @ubi: UBI device description object */ static void kill_volumes(struct ubi_device *ubi) @@ -447,36 +452,29 @@ static void kill_volumes(struct ubi_device *ubi) } /** - * free_user_volumes - free all user volumes. - * @ubi: UBI device description object - * - * Normally the volumes are freed at the release function of the volume device - * objects. However, on error paths the volumes have to be freed before the - * device objects have been initialized. - */ -static void free_user_volumes(struct ubi_device *ubi) -{ - int i; - - for (i = 0; i < ubi->vtbl_slots; i++) - if (ubi->volumes[i]) { - kfree(ubi->volumes[i]->eba_tbl); - kfree(ubi->volumes[i]); - } -} - -/** * uif_init - initialize user interfaces for an UBI device. * @ubi: UBI device description object + * @ref: set to %1 on exit in case of failure if a reference to @ubi->dev was + * taken, otherwise set to %0 + * + * This function initializes various user interfaces for an UBI device. If the + * initialization fails at an early stage, this function frees all the + * resources it allocated, returns an error, and @ref is set to %0. However, + * if the initialization fails after the UBI device was registered in the + * driver core subsystem, this function takes a reference to @ubi->dev, because + * otherwise the release function ('dev_release()') would free whole @ubi + * object. The @ref argument is set to %1 in this case. The caller has to put + * this reference. * * This function returns zero in case of success and a negative error code in - * case of failure. Note, this function destroys all volumes if it fails. + * case of failure. */ -static int uif_init(struct ubi_device *ubi) +static int uif_init(struct ubi_device *ubi, int *ref) { int i, err; dev_t dev; + *ref = 0; sprintf(ubi->ubi_name, UBI_NAME_STR "%d", ubi->ubi_num); /* @@ -504,7 +502,7 @@ static int uif_init(struct ubi_device *ubi) goto out_unreg; } - err = ubi_sysfs_init(ubi); + err = ubi_sysfs_init(ubi, ref); if (err) goto out_sysfs; @@ -522,6 +520,8 @@ static int uif_init(struct ubi_device *ubi) out_volumes: kill_volumes(ubi); out_sysfs: + if (*ref) + get_device(&ubi->dev); ubi_sysfs_close(ubi); cdev_del(&ubi->cdev); out_unreg: @@ -875,7 +875,7 @@ static int ubi_reboot_notifier(struct notifier_block *n, unsigned long state, int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset) { struct ubi_device *ubi; - int i, err, do_free = 1; + int i, err, ref = 0; /* * Check if we already have the same MTD device attached. @@ -975,9 +975,9 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset) goto out_detach; } - err = uif_init(ubi); + err = uif_init(ubi, &ref); if (err) - goto out_nofree; + goto out_detach; ubi->bgt_thread = kthread_create(ubi_thread, ubi, ubi->bgt_name); if (IS_ERR(ubi->bgt_thread)) { @@ -1025,12 +1025,8 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset) out_uif: uif_close(ubi); -out_nofree: - do_free = 0; out_detach: ubi_wl_close(ubi); - if (do_free) - free_user_volumes(ubi); free_internal_volumes(ubi); vfree(ubi->vtbl); out_free: @@ -1039,7 +1035,10 @@ out_free: #ifdef CONFIG_MTD_UBI_DEBUG_PARANOID vfree(ubi->dbg_peb_buf); #endif - kfree(ubi); + if (ref) + put_device(&ubi->dev); + else + kfree(ubi); return err; } @@ -1096,7 +1095,7 @@ int ubi_detach_mtd_dev(int ubi_num, int anyway) /* * Get a reference to the device in order to prevent 'dev_release()' - * from freeing @ubi object. + * from freeing the @ubi object. */ get_device(&ubi->dev); @@ -1116,13 +1115,50 @@ int ubi_detach_mtd_dev(int ubi_num, int anyway) } /** - * find_mtd_device - open an MTD device by its name or number. - * @mtd_dev: name or number of the device + * open_mtd_by_chdev - open an MTD device by its character device node path. + * @mtd_dev: MTD character device node path + * + * This helper function opens an MTD device by its character node device path. + * Returns MTD device description object in case of success and a negative + * error code in case of failure. + */ +static struct mtd_info * __init open_mtd_by_chdev(const char *mtd_dev) +{ + int err, major, minor, mode; + struct path path; + + /* Probably this is an MTD character device node path */ + err = kern_path(mtd_dev, LOOKUP_FOLLOW, &path); + if (err) + return ERR_PTR(err); + + /* MTD device number is defined by the major / minor numbers */ + major = imajor(path.dentry->d_inode); + minor = iminor(path.dentry->d_inode); + mode = path.dentry->d_inode->i_mode; + path_put(&path); + if (major != MTD_CHAR_MAJOR || !S_ISCHR(mode)) + return ERR_PTR(-EINVAL); + + if (minor & 1) + /* + * Just do not think the "/dev/mtdrX" devices support is need, + * so do not support them to avoid doing extra work. + */ + return ERR_PTR(-EINVAL); + + return get_mtd_device(NULL, minor / 2); +} + +/** + * open_mtd_device - open MTD device by name, character device path, or number. + * @mtd_dev: name, character device node path, or MTD device device number * * This function tries to open and MTD device described by @mtd_dev string, - * which is first treated as an ASCII number, and if it is not true, it is - * treated as MTD device name. Returns MTD device description object in case of - * success and a negative error code in case of failure. + * which is first treated as ASCII MTD device number, and if it is not true, it + * is treated as MTD device name, and if that is also not true, it is treated + * as MTD character device node path. Returns MTD device description object in + * case of success and a negative error code in case of failure. */ static struct mtd_info * __init open_mtd_device(const char *mtd_dev) { @@ -1137,6 +1173,9 @@ static struct mtd_info * __init open_mtd_device(const char *mtd_dev) * MTD device name. */ mtd = get_mtd_device_nm(mtd_dev); + if (IS_ERR(mtd) && PTR_ERR(mtd) == -ENODEV) + /* Probably this is an MTD character device node path */ + mtd = open_mtd_by_chdev(mtd_dev); } else mtd = get_mtd_device(NULL, mtd_num); @@ -1352,13 +1391,15 @@ static int __init ubi_mtd_param_parse(const char *val, struct kernel_param *kp) module_param_call(mtd, ubi_mtd_param_parse, NULL, NULL, 000); MODULE_PARM_DESC(mtd, "MTD devices to attach. Parameter format: " - "mtd=<name|num>[,<vid_hdr_offs>].\n" + "mtd=<name|num|path>[,<vid_hdr_offs>].\n" "Multiple \"mtd\" parameters may be specified.\n" - "MTD devices may be specified by their number or name.\n" + "MTD devices may be specified by their number, name, or " + "path to the MTD character device node.\n" "Optional \"vid_hdr_offs\" parameter specifies UBI VID " - "header position and data starting position to be used " - "by UBI.\n" - "Example: mtd=content,1984 mtd=4 - attach MTD device" + "header position to be used by UBI.\n" + "Example 1: mtd=/dev/mtd0 - attach MTD device " + "/dev/mtd0.\n" + "Example 2: mtd=content,1984 mtd=4 - attach MTD device " "with name \"content\" using VID header offset 1984, and " "MTD device number 4 with default VID header offset."); diff --git a/drivers/mtd/ubi/debug.h b/drivers/mtd/ubi/debug.h index f30bcb372c05..17a107129726 100644 --- a/drivers/mtd/ubi/debug.h +++ b/drivers/mtd/ubi/debug.h @@ -96,8 +96,11 @@ void ubi_dbg_dump_flash(struct ubi_device *ubi, int pnum, int offset, int len); #ifdef CONFIG_MTD_UBI_DEBUG_PARANOID int ubi_dbg_check_all_ff(struct ubi_device *ubi, int pnum, int offset, int len); +int ubi_dbg_check_write(struct ubi_device *ubi, const void *buf, int pnum, + int offset, int len); #else #define ubi_dbg_check_all_ff(ubi, pnum, offset, len) 0 +#define ubi_dbg_check_write(ubi, buf, pnum, offset, len) 0 #endif #ifdef CONFIG_MTD_UBI_DEBUG_DISABLE_BGT @@ -176,6 +179,7 @@ static inline int ubi_dbg_is_erase_failure(void) #define ubi_dbg_is_write_failure() 0 #define ubi_dbg_is_erase_failure() 0 #define ubi_dbg_check_all_ff(ubi, pnum, offset, len) 0 +#define ubi_dbg_check_write(ubi, buf, pnum, offset, len) 0 #endif /* !CONFIG_MTD_UBI_DEBUG */ #endif /* !__UBI_DEBUG_H__ */ diff --git a/drivers/mtd/ubi/io.c b/drivers/mtd/ubi/io.c index 8aa51e7a6a7d..b4ecc84c7549 100644 --- a/drivers/mtd/ubi/io.c +++ b/drivers/mtd/ubi/io.c @@ -143,7 +143,7 @@ int ubi_io_read(const struct ubi_device *ubi, void *buf, int pnum, int offset, err = paranoid_check_not_bad(ubi, pnum); if (err) - return err > 0 ? -EINVAL : err; + return err; addr = (loff_t)pnum * ubi->peb_size + offset; retry: @@ -236,12 +236,12 @@ int ubi_io_write(struct ubi_device *ubi, const void *buf, int pnum, int offset, err = paranoid_check_not_bad(ubi, pnum); if (err) - return err > 0 ? -EINVAL : err; + return err; /* The area we are writing to has to contain all 0xFF bytes */ err = ubi_dbg_check_all_ff(ubi, pnum, offset, len); if (err) - return err > 0 ? -EINVAL : err; + return err; if (offset >= ubi->leb_start) { /* @@ -250,10 +250,10 @@ int ubi_io_write(struct ubi_device *ubi, const void *buf, int pnum, int offset, */ err = paranoid_check_peb_ec_hdr(ubi, pnum); if (err) - return err > 0 ? -EINVAL : err; + return err; err = paranoid_check_peb_vid_hdr(ubi, pnum); if (err) - return err > 0 ? -EINVAL : err; + return err; } if (ubi_dbg_is_write_failure()) { @@ -273,6 +273,21 @@ int ubi_io_write(struct ubi_device *ubi, const void *buf, int pnum, int offset, } else ubi_assert(written == len); + if (!err) { + err = ubi_dbg_check_write(ubi, buf, pnum, offset, len); + if (err) + return err; + + /* + * Since we always write sequentially, the rest of the PEB has + * to contain only 0xFF bytes. + */ + offset += len; + len = ubi->peb_size - offset; + if (len) + err = ubi_dbg_check_all_ff(ubi, pnum, offset, len); + } + return err; } @@ -348,7 +363,7 @@ retry: err = ubi_dbg_check_all_ff(ubi, pnum, 0, ubi->peb_size); if (err) - return err > 0 ? -EINVAL : err; + return err; if (ubi_dbg_is_erase_failure() && !err) { dbg_err("cannot erase PEB %d (emulated)", pnum); @@ -542,7 +557,7 @@ int ubi_io_sync_erase(struct ubi_device *ubi, int pnum, int torture) err = paranoid_check_not_bad(ubi, pnum); if (err != 0) - return err > 0 ? -EINVAL : err; + return err; if (ubi->ro_mode) { ubi_err("read-only mode"); @@ -819,7 +834,7 @@ int ubi_io_write_ec_hdr(struct ubi_device *ubi, int pnum, err = paranoid_check_ec_hdr(ubi, pnum, ec_hdr); if (err) - return -EINVAL; + return err; err = ubi_io_write(ubi, ec_hdr, pnum, 0, ubi->ec_hdr_alsize); return err; @@ -1083,7 +1098,7 @@ int ubi_io_write_vid_hdr(struct ubi_device *ubi, int pnum, err = paranoid_check_peb_ec_hdr(ubi, pnum); if (err) - return err > 0 ? -EINVAL : err; + return err; vid_hdr->magic = cpu_to_be32(UBI_VID_HDR_MAGIC); vid_hdr->version = UBI_VERSION; @@ -1092,7 +1107,7 @@ int ubi_io_write_vid_hdr(struct ubi_device *ubi, int pnum, err = paranoid_check_vid_hdr(ubi, pnum, vid_hdr); if (err) - return -EINVAL; + return err; p = (char *)vid_hdr - ubi->vid_hdr_shift; err = ubi_io_write(ubi, p, pnum, ubi->vid_hdr_aloffset, @@ -1107,8 +1122,8 @@ int ubi_io_write_vid_hdr(struct ubi_device *ubi, int pnum, * @ubi: UBI device description object * @pnum: physical eraseblock number to check * - * This function returns zero if the physical eraseblock is good, a positive - * number if it is bad and a negative error code if an error occurred. + * This function returns zero if the physical eraseblock is good, %-EINVAL if + * it is bad and a negative error code if an error occurred. */ static int paranoid_check_not_bad(const struct ubi_device *ubi, int pnum) { @@ -1120,7 +1135,7 @@ static int paranoid_check_not_bad(const struct ubi_device *ubi, int pnum) ubi_err("paranoid check failed for PEB %d", pnum); ubi_dbg_dump_stack(); - return err; + return err > 0 ? -EINVAL : err; } /** @@ -1130,7 +1145,7 @@ static int paranoid_check_not_bad(const struct ubi_device *ubi, int pnum) * @ec_hdr: the erase counter header to check * * This function returns zero if the erase counter header contains valid - * values, and %1 if not. + * values, and %-EINVAL if not. */ static int paranoid_check_ec_hdr(const struct ubi_device *ubi, int pnum, const struct ubi_ec_hdr *ec_hdr) @@ -1156,7 +1171,7 @@ static int paranoid_check_ec_hdr(const struct ubi_device *ubi, int pnum, fail: ubi_dbg_dump_ec_hdr(ec_hdr); ubi_dbg_dump_stack(); - return 1; + return -EINVAL; } /** @@ -1164,8 +1179,8 @@ fail: * @ubi: UBI device description object * @pnum: the physical eraseblock number to check * - * This function returns zero if the erase counter header is all right, %1 if - * not, and a negative error code if an error occurred. + * This function returns zero if the erase counter header is all right and and + * a negative error code if not or if an error occurred. */ static int paranoid_check_peb_ec_hdr(const struct ubi_device *ubi, int pnum) { @@ -1188,7 +1203,7 @@ static int paranoid_check_peb_ec_hdr(const struct ubi_device *ubi, int pnum) ubi_err("paranoid check failed for PEB %d", pnum); ubi_dbg_dump_ec_hdr(ec_hdr); ubi_dbg_dump_stack(); - err = 1; + err = -EINVAL; goto exit; } @@ -1206,7 +1221,7 @@ exit: * @vid_hdr: the volume identifier header to check * * This function returns zero if the volume identifier header is all right, and - * %1 if not. + * %-EINVAL if not. */ static int paranoid_check_vid_hdr(const struct ubi_device *ubi, int pnum, const struct ubi_vid_hdr *vid_hdr) @@ -1233,7 +1248,7 @@ fail: ubi_err("paranoid check failed for PEB %d", pnum); ubi_dbg_dump_vid_hdr(vid_hdr); ubi_dbg_dump_stack(); - return 1; + return -EINVAL; } @@ -1243,7 +1258,7 @@ fail: * @pnum: the physical eraseblock number to check * * This function returns zero if the volume identifier header is all right, - * %1 if not, and a negative error code if an error occurred. + * and a negative error code if not or if an error occurred. */ static int paranoid_check_peb_vid_hdr(const struct ubi_device *ubi, int pnum) { @@ -1270,7 +1285,7 @@ static int paranoid_check_peb_vid_hdr(const struct ubi_device *ubi, int pnum) ubi_err("paranoid check failed for PEB %d", pnum); ubi_dbg_dump_vid_hdr(vid_hdr); ubi_dbg_dump_stack(); - err = 1; + err = -EINVAL; goto exit; } @@ -1282,6 +1297,61 @@ exit: } /** + * ubi_dbg_check_write - make sure write succeeded. + * @ubi: UBI device description object + * @buf: buffer with data which were written + * @pnum: physical eraseblock number the data were written to + * @offset: offset within the physical eraseblock the data were written to + * @len: how many bytes were written + * + * This functions reads data which were recently written and compares it with + * the original data buffer - the data have to match. Returns zero if the data + * match and a negative error code if not or in case of failure. + */ +int ubi_dbg_check_write(struct ubi_device *ubi, const void *buf, int pnum, + int offset, int len) +{ + int err, i; + + mutex_lock(&ubi->dbg_buf_mutex); + err = ubi_io_read(ubi, ubi->dbg_peb_buf, pnum, offset, len); + if (err) + goto out_unlock; + + for (i = 0; i < len; i++) { + uint8_t c = ((uint8_t *)buf)[i]; + uint8_t c1 = ((uint8_t *)ubi->dbg_peb_buf)[i]; + int dump_len; + + if (c == c1) + continue; + + ubi_err("paranoid check failed for PEB %d:%d, len %d", + pnum, offset, len); + ubi_msg("data differ at position %d", i); + dump_len = max_t(int, 128, len - i); + ubi_msg("hex dump of the original buffer from %d to %d", + i, i + dump_len); + print_hex_dump(KERN_DEBUG, "", DUMP_PREFIX_OFFSET, 32, 1, + buf + i, dump_len, 1); + ubi_msg("hex dump of the read buffer from %d to %d", + i, i + dump_len); + print_hex_dump(KERN_DEBUG, "", DUMP_PREFIX_OFFSET, 32, 1, + ubi->dbg_peb_buf + i, dump_len, 1); + ubi_dbg_dump_stack(); + err = -EINVAL; + goto out_unlock; + } + mutex_unlock(&ubi->dbg_buf_mutex); + + return 0; + +out_unlock: + mutex_unlock(&ubi->dbg_buf_mutex); + return err; +} + +/** * ubi_dbg_check_all_ff - check that a region of flash is empty. * @ubi: UBI device description object * @pnum: the physical eraseblock number to check @@ -1289,8 +1359,8 @@ exit: * @len: the length of the region to check * * This function returns zero if only 0xFF bytes are present at offset - * @offset of the physical eraseblock @pnum, %1 if not, and a negative error - * code if an error occurred. + * @offset of the physical eraseblock @pnum, and a negative error code if not + * or if an error occurred. */ int ubi_dbg_check_all_ff(struct ubi_device *ubi, int pnum, int offset, int len) { @@ -1321,7 +1391,7 @@ fail: ubi_msg("hex dump of the %d-%d region", offset, offset + len); print_hex_dump(KERN_DEBUG, "", DUMP_PREFIX_OFFSET, 32, 1, ubi->dbg_peb_buf, len, 1); - err = 1; + err = -EINVAL; error: ubi_dbg_dump_stack(); mutex_unlock(&ubi->dbg_buf_mutex); diff --git a/drivers/mtd/ubi/scan.c b/drivers/mtd/ubi/scan.c index 90af61a2c3e4..594184bbd56a 100644 --- a/drivers/mtd/ubi/scan.c +++ b/drivers/mtd/ubi/scan.c @@ -974,11 +974,8 @@ struct ubi_scan_info *ubi_scan(struct ubi_device *ubi) seb->ec = si->mean_ec; err = paranoid_check_si(ubi, si); - if (err) { - if (err > 0) - err = -EINVAL; + if (err) goto out_vidh; - } ubi_free_vid_hdr(ubi, vidh); kfree(ech); @@ -1086,8 +1083,8 @@ void ubi_scan_destroy_si(struct ubi_scan_info *si) * @ubi: UBI device description object * @si: scanning information * - * This function returns zero if the scanning information is all right, %1 if - * not and a negative error code if an error occurred. + * This function returns zero if the scanning information is all right, and a + * negative error code if not or if an error occurred. */ static int paranoid_check_si(struct ubi_device *ubi, struct ubi_scan_info *si) { @@ -1346,7 +1343,7 @@ bad_vid_hdr: out: ubi_dbg_dump_stack(); - return 1; + return -EINVAL; } #endif /* CONFIG_MTD_UBI_DEBUG_PARANOID */ diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c index 600c7229d5cf..f64ddabd4ac8 100644 --- a/drivers/mtd/ubi/wl.c +++ b/drivers/mtd/ubi/wl.c @@ -464,7 +464,7 @@ retry: ubi->peb_size - ubi->vid_hdr_aloffset); if (err) { ubi_err("new PEB %d does not contain all 0xFF bytes", e->pnum); - return err > 0 ? -EINVAL : err; + return err; } return e->pnum; @@ -513,7 +513,7 @@ static int sync_erase(struct ubi_device *ubi, struct ubi_wl_entry *e, dbg_wl("erase PEB %d, old EC %llu", e->pnum, ec); err = paranoid_check_ec(ubi, e->pnum, e->ec); - if (err > 0) + if (err) return -EINVAL; ec_hdr = kzalloc(ubi->ec_hdr_alsize, GFP_NOFS); @@ -1572,8 +1572,7 @@ void ubi_wl_close(struct ubi_device *ubi) * @ec: the erase counter to check * * This function returns zero if the erase counter of physical eraseblock @pnum - * is equivalent to @ec, %1 if not, and a negative error code if an error - * occurred. + * is equivalent to @ec, and a negative error code if not or if an error occurred. */ static int paranoid_check_ec(struct ubi_device *ubi, int pnum, int ec) { @@ -1611,8 +1610,8 @@ out_free: * @e: the wear-leveling entry to check * @root: the root of the tree * - * This function returns zero if @e is in the @root RB-tree and %1 if it is - * not. + * This function returns zero if @e is in the @root RB-tree and %-EINVAL if it + * is not. */ static int paranoid_check_in_wl_tree(struct ubi_wl_entry *e, struct rb_root *root) @@ -1623,7 +1622,7 @@ static int paranoid_check_in_wl_tree(struct ubi_wl_entry *e, ubi_err("paranoid check failed for PEB %d, EC %d, RB-tree %p ", e->pnum, e->ec, root); ubi_dbg_dump_stack(); - return 1; + return -EINVAL; } /** @@ -1632,7 +1631,7 @@ static int paranoid_check_in_wl_tree(struct ubi_wl_entry *e, * @ubi: UBI device description object * @e: the wear-leveling entry to check * - * This function returns zero if @e is in @ubi->pq and %1 if it is not. + * This function returns zero if @e is in @ubi->pq and %-EINVAL if it is not. */ static int paranoid_check_in_pq(struct ubi_device *ubi, struct ubi_wl_entry *e) { @@ -1647,6 +1646,6 @@ static int paranoid_check_in_pq(struct ubi_device *ubi, struct ubi_wl_entry *e) ubi_err("paranoid check failed for PEB %d, EC %d, Protect queue", e->pnum, e->ec); ubi_dbg_dump_stack(); - return 1; + return -EINVAL; } #endif /* CONFIG_MTD_UBI_DEBUG_PARANOID */ diff --git a/drivers/net/gianfar.c b/drivers/net/gianfar.c index 6aa526ee9096..61a7b4351e78 100644 --- a/drivers/net/gianfar.c +++ b/drivers/net/gianfar.c @@ -998,7 +998,7 @@ static int gfar_probe(struct of_device *ofdev, } /* Need to reverse the bit maps as bit_map's MSB is q0 - * but, for_each_bit parses from right to left, which + * but, for_each_set_bit parses from right to left, which * basically reverses the queue numbers */ for (i = 0; i< priv->num_grps; i++) { priv->gfargrp[i].tx_bit_map = reverse_bitmap( @@ -1011,7 +1011,7 @@ static int gfar_probe(struct of_device *ofdev, * also assign queues to groups */ for (grp_idx = 0; grp_idx < priv->num_grps; grp_idx++) { priv->gfargrp[grp_idx].num_rx_queues = 0x0; - for_each_bit(i, &priv->gfargrp[grp_idx].rx_bit_map, + for_each_set_bit(i, &priv->gfargrp[grp_idx].rx_bit_map, priv->num_rx_queues) { priv->gfargrp[grp_idx].num_rx_queues++; priv->rx_queue[i]->grp = &priv->gfargrp[grp_idx]; @@ -1019,7 +1019,7 @@ static int gfar_probe(struct of_device *ofdev, rqueue = rqueue | ((RQUEUE_EN0 | RQUEUE_EX0) >> i); } priv->gfargrp[grp_idx].num_tx_queues = 0x0; - for_each_bit (i, &priv->gfargrp[grp_idx].tx_bit_map, + for_each_set_bit(i, &priv->gfargrp[grp_idx].tx_bit_map, priv->num_tx_queues) { priv->gfargrp[grp_idx].num_tx_queues++; priv->tx_queue[i]->grp = &priv->gfargrp[grp_idx]; @@ -1709,7 +1709,7 @@ void gfar_configure_coalescing(struct gfar_private *priv, if (priv->mode == MQ_MG_MODE) { baddr = ®s->txic0; - for_each_bit (i, &tx_mask, priv->num_tx_queues) { + for_each_set_bit(i, &tx_mask, priv->num_tx_queues) { if (likely(priv->tx_queue[i]->txcoalescing)) { gfar_write(baddr + i, 0); gfar_write(baddr + i, priv->tx_queue[i]->txic); @@ -1717,7 +1717,7 @@ void gfar_configure_coalescing(struct gfar_private *priv, } baddr = ®s->rxic0; - for_each_bit (i, &rx_mask, priv->num_rx_queues) { + for_each_set_bit(i, &rx_mask, priv->num_rx_queues) { if (likely(priv->rx_queue[i]->rxcoalescing)) { gfar_write(baddr + i, 0); gfar_write(baddr + i, priv->rx_queue[i]->rxic); @@ -2607,7 +2607,7 @@ static int gfar_poll(struct napi_struct *napi, int budget) budget_per_queue = left_over_budget/num_queues; left_over_budget = 0; - for_each_bit(i, &gfargrp->rx_bit_map, priv->num_rx_queues) { + for_each_set_bit(i, &gfargrp->rx_bit_map, priv->num_rx_queues) { if (test_bit(i, &serviced_queues)) continue; rx_queue = priv->rx_queue[i]; diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c index 45e3532b166f..684af371462d 100644 --- a/drivers/net/ixgbe/ixgbe_main.c +++ b/drivers/net/ixgbe/ixgbe_main.c @@ -1050,7 +1050,7 @@ static void ixgbe_configure_msix(struct ixgbe_adapter *adapter) */ for (v_idx = 0; v_idx < q_vectors; v_idx++) { q_vector = adapter->q_vector[v_idx]; - /* XXX for_each_bit(...) */ + /* XXX for_each_set_bit(...) */ r_idx = find_first_bit(q_vector->rxr_idx, adapter->num_rx_queues); diff --git a/drivers/net/ixgbevf/ixgbevf_main.c b/drivers/net/ixgbevf/ixgbevf_main.c index 235b5fd4b8d4..ca653c49b765 100644 --- a/drivers/net/ixgbevf/ixgbevf_main.c +++ b/drivers/net/ixgbevf/ixgbevf_main.c @@ -751,7 +751,7 @@ static void ixgbevf_configure_msix(struct ixgbevf_adapter *adapter) */ for (v_idx = 0; v_idx < q_vectors; v_idx++) { q_vector = adapter->q_vector[v_idx]; - /* XXX for_each_bit(...) */ + /* XXX for_each_set_bit(...) */ r_idx = find_first_bit(q_vector->rxr_idx, adapter->num_rx_queues); diff --git a/drivers/net/wireless/ath/ar9170/main.c b/drivers/net/wireless/ath/ar9170/main.c index 8a964f130367..a6452af9c6c5 100644 --- a/drivers/net/wireless/ath/ar9170/main.c +++ b/drivers/net/wireless/ath/ar9170/main.c @@ -394,7 +394,7 @@ static void ar9170_tx_fake_ampdu_status(struct ar9170 *ar) ieee80211_tx_status_irqsafe(ar->hw, skb); } - for_each_bit(i, &queue_bitmap, BITS_PER_BYTE) { + for_each_set_bit(i, &queue_bitmap, BITS_PER_BYTE) { #ifdef AR9170_QUEUE_STOP_DEBUG printk(KERN_DEBUG "%s: wake queue %d\n", wiphy_name(ar->hw->wiphy), i); diff --git a/drivers/net/wireless/iwmc3200wifi/debugfs.c b/drivers/net/wireless/iwmc3200wifi/debugfs.c index be992ca41cf1..c29c994de0e2 100644 --- a/drivers/net/wireless/iwmc3200wifi/debugfs.c +++ b/drivers/net/wireless/iwmc3200wifi/debugfs.c @@ -89,7 +89,7 @@ static int iwm_debugfs_dbg_modules_write(void *data, u64 val) for (i = 0; i < __IWM_DM_NR; i++) iwm->dbg.dbg_module[i] = 0; - for_each_bit(bit, &iwm->dbg.dbg_modules, __IWM_DM_NR) + for_each_set_bit(bit, &iwm->dbg.dbg_modules, __IWM_DM_NR) iwm->dbg.dbg_module[bit] = iwm->dbg.dbg_level; return 0; diff --git a/drivers/net/wireless/iwmc3200wifi/rx.c b/drivers/net/wireless/iwmc3200wifi/rx.c index ad8f7eabb5aa..8456b4dbd146 100644 --- a/drivers/net/wireless/iwmc3200wifi/rx.c +++ b/drivers/net/wireless/iwmc3200wifi/rx.c @@ -1116,7 +1116,7 @@ static int iwm_ntf_stop_resume_tx(struct iwm_priv *iwm, u8 *buf, return -EINVAL; } - for_each_bit(bit, (unsigned long *)&tid_msk, IWM_UMAC_TID_NR) { + for_each_set_bit(bit, (unsigned long *)&tid_msk, IWM_UMAC_TID_NR) { tid_info = &sta_info->tid_info[bit]; mutex_lock(&tid_info->mutex); diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 039e87b71442..81d19d5683ac 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -2533,6 +2533,91 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x1518, quirk_i82576_sriov); #endif /* CONFIG_PCI_IOV */ +/* + * This is a quirk for the Ricoh MMC controller found as a part of + * some mulifunction chips. + + * This is very similiar and based on the ricoh_mmc driver written by + * Philip Langdale. Thank you for these magic sequences. + * + * These chips implement the four main memory card controllers (SD, MMC, MS, xD) + * and one or both of cardbus or firewire. + * + * It happens that they implement SD and MMC + * support as separate controllers (and PCI functions). The linux SDHCI + * driver supports MMC cards but the chip detects MMC cards in hardware + * and directs them to the MMC controller - so the SDHCI driver never sees + * them. + * + * To get around this, we must disable the useless MMC controller. + * At that point, the SDHCI controller will start seeing them + * It seems to be the case that the relevant PCI registers to deactivate the + * MMC controller live on PCI function 0, which might be the cardbus controller + * or the firewire controller, depending on the particular chip in question + * + * This has to be done early, because as soon as we disable the MMC controller + * other pci functions shift up one level, e.g. function #2 becomes function + * #1, and this will confuse the pci core. + */ + +#ifdef CONFIG_MMC_RICOH_MMC +static void ricoh_mmc_fixup_rl5c476(struct pci_dev *dev) +{ + /* disable via cardbus interface */ + u8 write_enable; + u8 write_target; + u8 disable; + + /* disable must be done via function #0 */ + if (PCI_FUNC(dev->devfn)) + return; + + pci_read_config_byte(dev, 0xB7, &disable); + if (disable & 0x02) + return; + + pci_read_config_byte(dev, 0x8E, &write_enable); + pci_write_config_byte(dev, 0x8E, 0xAA); + pci_read_config_byte(dev, 0x8D, &write_target); + pci_write_config_byte(dev, 0x8D, 0xB7); + pci_write_config_byte(dev, 0xB7, disable | 0x02); + pci_write_config_byte(dev, 0x8E, write_enable); + pci_write_config_byte(dev, 0x8D, write_target); + + dev_notice(&dev->dev, "proprietary Ricoh MMC controller disabled (via cardbus function)\n"); + dev_notice(&dev->dev, "MMC cards are now supported by standard SDHCI controller\n"); +} +DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_RICOH, PCI_DEVICE_ID_RICOH_RL5C476, ricoh_mmc_fixup_rl5c476); +DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_RICOH, PCI_DEVICE_ID_RICOH_RL5C476, ricoh_mmc_fixup_rl5c476); + +static void ricoh_mmc_fixup_r5c832(struct pci_dev *dev) +{ + /* disable via firewire interface */ + u8 write_enable; + u8 disable; + + /* disable must be done via function #0 */ + if (PCI_FUNC(dev->devfn)) + return; + + pci_read_config_byte(dev, 0xCB, &disable); + + if (disable & 0x02) + return; + + pci_read_config_byte(dev, 0xCA, &write_enable); + pci_write_config_byte(dev, 0xCA, 0x57); + pci_write_config_byte(dev, 0xCB, disable | 0x02); + pci_write_config_byte(dev, 0xCA, write_enable); + + dev_notice(&dev->dev, "proprietary Ricoh MMC controller disabled (via firewire function)\n"); + dev_notice(&dev->dev, "MMC cards are now supported by standard SDHCI controller\n"); +} +DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_RICOH, PCI_DEVICE_ID_RICOH_R5C832, ricoh_mmc_fixup_r5c832); +DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_RICOH, PCI_DEVICE_ID_RICOH_R5C832, ricoh_mmc_fixup_r5c832); +#endif /*CONFIG_MMC_RICOH_MMC*/ + + static void pci_do_fixups(struct pci_dev *dev, struct pci_fixup *f, struct pci_fixup *end) { diff --git a/drivers/rtc/class.c b/drivers/rtc/class.c index be5a6b73e601..40845c7e9322 100644 --- a/drivers/rtc/class.c +++ b/drivers/rtc/class.c @@ -226,6 +226,7 @@ static void __exit rtc_exit(void) { rtc_dev_exit(); class_destroy(rtc_class); + idr_destroy(&rtc_idr); } subsys_initcall(rtc_init); diff --git a/drivers/rtc/rtc-at91sam9.c b/drivers/rtc/rtc-at91sam9.c index 86c61f143515..78a018b5c941 100644 --- a/drivers/rtc/rtc-at91sam9.c +++ b/drivers/rtc/rtc-at91sam9.c @@ -161,7 +161,7 @@ static int at91_rtc_readalarm(struct device *dev, struct rtc_wkalrm *alrm) if (offset == 0) return -EILSEQ; - memset(alrm, 0, sizeof(alrm)); + memset(alrm, 0, sizeof(*alrm)); if (alarm != ALARM_DISABLED && offset != 0) { rtc_time_to_tm(offset + alarm, tm); diff --git a/drivers/rtc/rtc-coh901331.c b/drivers/rtc/rtc-coh901331.c index 03ea530981d1..44c4399ee714 100644 --- a/drivers/rtc/rtc-coh901331.c +++ b/drivers/rtc/rtc-coh901331.c @@ -271,12 +271,13 @@ static int coh901331_resume(struct platform_device *pdev) { struct coh901331_port *rtap = dev_get_drvdata(&pdev->dev); - if (device_may_wakeup(&pdev->dev)) + if (device_may_wakeup(&pdev->dev)) { disable_irq_wake(rtap->irq); - else + } else { clk_enable(rtap->clk); writel(rtap->irqmaskstore, rtap->virtbase + COH901331_IRQ_MASK); clk_disable(rtap->clk); + } return 0; } #else diff --git a/drivers/rtc/rtc-ep93xx.c b/drivers/rtc/rtc-ep93xx.c index 9da02d108b73..91bde976bc0f 100644 --- a/drivers/rtc/rtc-ep93xx.c +++ b/drivers/rtc/rtc-ep93xx.c @@ -115,6 +115,15 @@ static ssize_t ep93xx_rtc_show_comp_delete(struct device *dev, } static DEVICE_ATTR(comp_delete, S_IRUGO, ep93xx_rtc_show_comp_delete, NULL); +static struct attribute *ep93xx_rtc_attrs[] = { + &dev_attr_comp_preload.attr, + &dev_attr_comp_delete.attr, + NULL +}; + +static const struct attribute_group ep93xx_rtc_sysfs_files = { + .attrs = ep93xx_rtc_attrs, +}; static int __init ep93xx_rtc_probe(struct platform_device *pdev) { @@ -123,27 +132,22 @@ static int __init ep93xx_rtc_probe(struct platform_device *pdev) struct rtc_device *rtc; int err; - ep93xx_rtc = kzalloc(sizeof(struct ep93xx_rtc), GFP_KERNEL); - if (ep93xx_rtc == NULL) + ep93xx_rtc = devm_kzalloc(&pdev->dev, sizeof(*ep93xx_rtc), GFP_KERNEL); + if (!ep93xx_rtc) return -ENOMEM; res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - if (res == NULL) { - err = -ENXIO; - goto fail_free; - } + if (!res) + return -ENXIO; - res = request_mem_region(res->start, resource_size(res), pdev->name); - if (res == NULL) { - err = -EBUSY; - goto fail_free; - } + if (!devm_request_mem_region(&pdev->dev, res->start, + resource_size(res), pdev->name)) + return -EBUSY; - ep93xx_rtc->mmio_base = ioremap(res->start, resource_size(res)); - if (ep93xx_rtc->mmio_base == NULL) { - err = -ENXIO; - goto fail; - } + ep93xx_rtc->mmio_base = devm_ioremap(&pdev->dev, res->start, + resource_size(res)); + if (!ep93xx_rtc->mmio_base) + return -ENXIO; pdev->dev.platform_data = ep93xx_rtc; @@ -151,53 +155,34 @@ static int __init ep93xx_rtc_probe(struct platform_device *pdev) &pdev->dev, &ep93xx_rtc_ops, THIS_MODULE); if (IS_ERR(rtc)) { err = PTR_ERR(rtc); - goto fail; + goto exit; } platform_set_drvdata(pdev, rtc); - err = device_create_file(&pdev->dev, &dev_attr_comp_preload); + err = sysfs_create_group(&pdev->dev.kobj, &ep93xx_rtc_sysfs_files); if (err) goto fail; - err = device_create_file(&pdev->dev, &dev_attr_comp_delete); - if (err) { - device_remove_file(&pdev->dev, &dev_attr_comp_preload); - goto fail; - } return 0; fail: - if (ep93xx_rtc->mmio_base) { - iounmap(ep93xx_rtc->mmio_base); - pdev->dev.platform_data = NULL; - } - release_mem_region(res->start, resource_size(res)); -fail_free: - kfree(ep93xx_rtc); + platform_set_drvdata(pdev, NULL); + rtc_device_unregister(rtc); +exit: + pdev->dev.platform_data = NULL; return err; } static int __exit ep93xx_rtc_remove(struct platform_device *pdev) { struct rtc_device *rtc = platform_get_drvdata(pdev); - struct ep93xx_rtc *ep93xx_rtc = pdev->dev.platform_data; - struct resource *res; - - /* cleanup sysfs */ - device_remove_file(&pdev->dev, &dev_attr_comp_delete); - device_remove_file(&pdev->dev, &dev_attr_comp_preload); + sysfs_remove_group(&pdev->dev.kobj, &ep93xx_rtc_sysfs_files); + platform_set_drvdata(pdev, NULL); rtc_device_unregister(rtc); - - iounmap(ep93xx_rtc->mmio_base); pdev->dev.platform_data = NULL; - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - release_mem_region(res->start, resource_size(res)); - - platform_set_drvdata(pdev, NULL); - return 0; } diff --git a/drivers/rtc/rtc-mc13783.c b/drivers/rtc/rtc-mc13783.c index 850f983c039c..d60c81b7b693 100644 --- a/drivers/rtc/rtc-mc13783.c +++ b/drivers/rtc/rtc-mc13783.c @@ -28,6 +28,34 @@ struct mc13783_rtc { int valid; }; +static int mc13783_rtc_irq_enable_unlocked(struct device *dev, + unsigned int enabled, int irq) +{ + struct mc13783_rtc *priv = dev_get_drvdata(dev); + int (*func)(struct mc13783 *mc13783, int irq); + + if (!priv->valid) + return -ENODATA; + + func = enabled ? mc13783_irq_unmask : mc13783_irq_mask; + return func(priv->mc13783, irq); +} + +static int mc13783_rtc_irq_enable(struct device *dev, + unsigned int enabled, int irq) +{ + struct mc13783_rtc *priv = dev_get_drvdata(dev); + int ret; + + mc13783_lock(priv->mc13783); + + ret = mc13783_rtc_irq_enable_unlocked(dev, enabled, irq); + + mc13783_unlock(priv->mc13783); + + return ret; +} + static int mc13783_rtc_read_time(struct device *dev, struct rtc_time *tm) { struct mc13783_rtc *priv = dev_get_drvdata(dev); @@ -78,6 +106,7 @@ static int mc13783_rtc_set_mmss(struct device *dev, unsigned long secs) { struct mc13783_rtc *priv = dev_get_drvdata(dev); unsigned int seconds, days; + unsigned int alarmseconds; int ret; seconds = secs % 86400; @@ -86,7 +115,22 @@ static int mc13783_rtc_set_mmss(struct device *dev, unsigned long secs) mc13783_lock(priv->mc13783); /* - * first write seconds=0 to prevent a day switch between writing days + * temporarily invalidate alarm to prevent triggering it when the day is + * already updated while the time isn't yet. + */ + ret = mc13783_reg_read(priv->mc13783, MC13783_RTCTODA, &alarmseconds); + if (unlikely(ret)) + goto out; + + if (alarmseconds < 86400) { + ret = mc13783_reg_write(priv->mc13783, + MC13783_RTCTODA, 0x1ffff); + if (unlikely(ret)) + goto out; + } + + /* + * write seconds=0 to prevent a day switch between writing days * and seconds below */ ret = mc13783_reg_write(priv->mc13783, MC13783_RTCTOD, 0); @@ -101,11 +145,19 @@ static int mc13783_rtc_set_mmss(struct device *dev, unsigned long secs) if (unlikely(ret)) goto out; - ret = mc13783_ackirq(priv->mc13783, MC13783_IRQ_RTCRST); + /* restore alarm */ + if (alarmseconds < 86400) { + ret = mc13783_reg_write(priv->mc13783, + MC13783_RTCTODA, alarmseconds); + if (unlikely(ret)) + goto out; + } + + ret = mc13783_irq_ack(priv->mc13783, MC13783_IRQ_RTCRST); if (unlikely(ret)) goto out; - ret = mc13783_unmask(priv->mc13783, MC13783_IRQ_RTCRST); + ret = mc13783_irq_unmask(priv->mc13783, MC13783_IRQ_RTCRST); out: priv->valid = !ret; @@ -114,41 +166,139 @@ out: return ret; } -static irqreturn_t mc13783_rtc_update_handler(int irq, void *dev) +static int mc13783_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alarm) { - struct mc13783_rtc *priv = dev; - struct mc13783 *mc13783 = priv->mc13783; + struct mc13783_rtc *priv = dev_get_drvdata(dev); + unsigned seconds, days; + unsigned long s1970; + int enabled, pending; + int ret; - dev_dbg(&priv->rtc->dev, "1HZ\n"); + mc13783_lock(priv->mc13783); - rtc_update_irq(priv->rtc, 1, RTC_IRQF | RTC_UF); + ret = mc13783_reg_read(priv->mc13783, MC13783_RTCTODA, &seconds); + if (unlikely(ret)) + goto out; + if (seconds >= 86400) { + ret = -ENODATA; + goto out; + } + + ret = mc13783_reg_read(priv->mc13783, MC13783_RTCDAY, &days); + if (unlikely(ret)) + goto out; - mc13783_ackirq(mc13783, irq); + ret = mc13783_irq_status(priv->mc13783, MC13783_IRQ_TODA, + &enabled, &pending); - return IRQ_HANDLED; +out: + mc13783_unlock(priv->mc13783); + + if (ret) + return ret; + + alarm->enabled = enabled; + alarm->pending = pending; + + s1970 = days * 86400 + seconds; + + rtc_time_to_tm(s1970, &alarm->time); + dev_dbg(dev, "%s: %lu\n", __func__, s1970); + + return 0; } -static int mc13783_rtc_update_irq_enable(struct device *dev, - unsigned int enabled) +static int mc13783_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alarm) { struct mc13783_rtc *priv = dev_get_drvdata(dev); - int ret = -ENODATA; + unsigned long s1970; + unsigned seconds, days; + int ret; mc13783_lock(priv->mc13783); - if (!priv->valid) + + /* disable alarm to prevent false triggering */ + ret = mc13783_reg_write(priv->mc13783, MC13783_RTCTODA, 0x1ffff); + if (unlikely(ret)) goto out; - ret = (enabled ? mc13783_unmask : mc13783_mask)(priv->mc13783, - MC13783_IRQ_1HZ); + ret = mc13783_irq_ack(priv->mc13783, MC13783_IRQ_TODA); + if (unlikely(ret)) + goto out; + + ret = rtc_tm_to_time(&alarm->time, &s1970); + if (unlikely(ret)) + goto out; + + dev_dbg(dev, "%s: o%2.s %lu\n", __func__, alarm->enabled ? "n" : "ff", + s1970); + + ret = mc13783_rtc_irq_enable_unlocked(dev, alarm->enabled, + MC13783_IRQ_TODA); + if (unlikely(ret)) + goto out; + + seconds = s1970 % 86400; + days = s1970 / 86400; + + ret = mc13783_reg_write(priv->mc13783, MC13783_RTCDAYA, days); + if (unlikely(ret)) + goto out; + + ret = mc13783_reg_write(priv->mc13783, MC13783_RTCTODA, seconds); + out: mc13783_unlock(priv->mc13783); return ret; } +static irqreturn_t mc13783_rtc_alarm_handler(int irq, void *dev) +{ + struct mc13783_rtc *priv = dev; + struct mc13783 *mc13783 = priv->mc13783; + + dev_dbg(&priv->rtc->dev, "Alarm\n"); + + rtc_update_irq(priv->rtc, 1, RTC_IRQF | RTC_AF); + + mc13783_irq_ack(mc13783, irq); + + return IRQ_HANDLED; +} + +static irqreturn_t mc13783_rtc_update_handler(int irq, void *dev) +{ + struct mc13783_rtc *priv = dev; + struct mc13783 *mc13783 = priv->mc13783; + + dev_dbg(&priv->rtc->dev, "1HZ\n"); + + rtc_update_irq(priv->rtc, 1, RTC_IRQF | RTC_UF); + + mc13783_irq_ack(mc13783, irq); + + return IRQ_HANDLED; +} + +static int mc13783_rtc_update_irq_enable(struct device *dev, + unsigned int enabled) +{ + return mc13783_rtc_irq_enable(dev, enabled, MC13783_IRQ_1HZ); +} + +static int mc13783_rtc_alarm_irq_enable(struct device *dev, + unsigned int enabled) +{ + return mc13783_rtc_irq_enable(dev, enabled, MC13783_IRQ_TODA); +} + static const struct rtc_class_ops mc13783_rtc_ops = { .read_time = mc13783_rtc_read_time, .set_mmss = mc13783_rtc_set_mmss, + .read_alarm = mc13783_rtc_read_alarm, + .set_alarm = mc13783_rtc_set_alarm, + .alarm_irq_enable = mc13783_rtc_alarm_irq_enable, .update_irq_enable = mc13783_rtc_update_irq_enable, }; @@ -160,7 +310,7 @@ static irqreturn_t mc13783_rtc_reset_handler(int irq, void *dev) dev_dbg(&priv->rtc->dev, "RTCRST\n"); priv->valid = 0; - mc13783_mask(mc13783, irq); + mc13783_irq_mask(mc13783, irq); return IRQ_HANDLED; } @@ -169,6 +319,7 @@ static int __devinit mc13783_rtc_probe(struct platform_device *pdev) { int ret; struct mc13783_rtc *priv; + int rtcrst_pending; priv = kzalloc(sizeof(*priv), GFP_KERNEL); if (!priv) @@ -177,8 +328,6 @@ static int __devinit mc13783_rtc_probe(struct platform_device *pdev) priv->mc13783 = dev_get_drvdata(pdev->dev.parent); platform_set_drvdata(pdev, priv); - priv->valid = 1; - mc13783_lock(priv->mc13783); ret = mc13783_irq_request(priv->mc13783, MC13783_IRQ_RTCRST, @@ -186,33 +335,45 @@ static int __devinit mc13783_rtc_probe(struct platform_device *pdev) if (ret) goto err_reset_irq_request; + ret = mc13783_irq_status(priv->mc13783, MC13783_IRQ_RTCRST, + NULL, &rtcrst_pending); + if (ret) + goto err_reset_irq_status; + + priv->valid = !rtcrst_pending; + ret = mc13783_irq_request_nounmask(priv->mc13783, MC13783_IRQ_1HZ, mc13783_rtc_update_handler, DRIVER_NAME, priv); if (ret) goto err_update_irq_request; - mc13783_unlock(priv->mc13783); + ret = mc13783_irq_request_nounmask(priv->mc13783, MC13783_IRQ_TODA, + mc13783_rtc_alarm_handler, DRIVER_NAME, priv); + if (ret) + goto err_alarm_irq_request; priv->rtc = rtc_device_register(pdev->name, &pdev->dev, &mc13783_rtc_ops, THIS_MODULE); - if (IS_ERR(priv->rtc)) { ret = PTR_ERR(priv->rtc); - mc13783_lock(priv->mc13783); + mc13783_irq_free(priv->mc13783, MC13783_IRQ_TODA, priv); +err_alarm_irq_request: mc13783_irq_free(priv->mc13783, MC13783_IRQ_1HZ, priv); err_update_irq_request: +err_reset_irq_status: + mc13783_irq_free(priv->mc13783, MC13783_IRQ_RTCRST, priv); err_reset_irq_request: - mc13783_unlock(priv->mc13783); - platform_set_drvdata(pdev, NULL); kfree(priv); } + mc13783_unlock(priv->mc13783); + return ret; } @@ -220,10 +381,11 @@ static int __devexit mc13783_rtc_remove(struct platform_device *pdev) { struct mc13783_rtc *priv = platform_get_drvdata(pdev); - rtc_device_unregister(priv->rtc); - mc13783_lock(priv->mc13783); + rtc_device_unregister(priv->rtc); + + mc13783_irq_free(priv->mc13783, MC13783_IRQ_TODA, priv); mc13783_irq_free(priv->mc13783, MC13783_IRQ_1HZ, priv); mc13783_irq_free(priv->mc13783, MC13783_IRQ_RTCRST, priv); diff --git a/drivers/rtc/rtc-mxc.c b/drivers/rtc/rtc-mxc.c index 6bd5072d4eb7..8710f9415d98 100644 --- a/drivers/rtc/rtc-mxc.c +++ b/drivers/rtc/rtc-mxc.c @@ -396,8 +396,11 @@ static int __init mxc_rtc_probe(struct platform_device *pdev) pdata->ioaddr = ioremap(res->start, resource_size(res)); clk = clk_get(&pdev->dev, "ckil"); - if (IS_ERR(clk)) - return PTR_ERR(clk); + if (IS_ERR(clk)) { + iounmap(pdata->ioaddr); + ret = PTR_ERR(clk); + goto exit_free_pdata; + } rate = clk_get_rate(clk); clk_put(clk); diff --git a/drivers/rtc/rtc-pcf2123.c b/drivers/rtc/rtc-pcf2123.c index e75df9d50e27..2ceb365533b2 100644 --- a/drivers/rtc/rtc-pcf2123.c +++ b/drivers/rtc/rtc-pcf2123.c @@ -315,7 +315,7 @@ kfree_exit: return ret; } -static int pcf2123_remove(struct spi_device *spi) +static int __devexit pcf2123_remove(struct spi_device *spi) { struct pcf2123_plat_data *pdata = spi->dev.platform_data; int i; diff --git a/drivers/rtc/rtc-twl.c b/drivers/rtc/rtc-twl.c index c6a83a2a722c..ed1b86828124 100644 --- a/drivers/rtc/rtc-twl.c +++ b/drivers/rtc/rtc-twl.c @@ -57,7 +57,7 @@ enum { REG_RTC_COMP_LSB_REG, REG_RTC_COMP_MSB_REG, }; -const static u8 twl4030_rtc_reg_map[] = { +static const u8 twl4030_rtc_reg_map[] = { [REG_SECONDS_REG] = 0x00, [REG_MINUTES_REG] = 0x01, [REG_HOURS_REG] = 0x02, @@ -80,7 +80,7 @@ const static u8 twl4030_rtc_reg_map[] = { [REG_RTC_COMP_LSB_REG] = 0x10, [REG_RTC_COMP_MSB_REG] = 0x11, }; -const static u8 twl6030_rtc_reg_map[] = { +static const u8 twl6030_rtc_reg_map[] = { [REG_SECONDS_REG] = 0x00, [REG_MINUTES_REG] = 0x01, [REG_HOURS_REG] = 0x02, diff --git a/drivers/usb/core/devices.c b/drivers/usb/core/devices.c index c83c975152a6..d41811bfef2a 100644 --- a/drivers/usb/core/devices.c +++ b/drivers/usb/core/devices.c @@ -117,13 +117,20 @@ static const char *format_endpt = * However, these will come from functions that return ptrs to each of them. */ -static DECLARE_WAIT_QUEUE_HEAD(deviceconndiscwq); -/* guarded by usbfs_mutex */ -static unsigned int conndiscevcnt; - -/* this struct stores the poll state for <mountpoint>/devices pollers */ -struct usb_device_status { - unsigned int lastev; +/* + * Wait for an connect/disconnect event to happen. We initialize + * the event counter with an odd number, and each event will increment + * the event counter by two, so it will always _stay_ odd. That means + * that it will never be zero, so "event 0" will never match a current + * event, and thus 'poll' will always trigger as readable for the first + * time it gets called. + */ +static struct device_connect_event { + atomic_t count; + wait_queue_head_t wait; +} device_event = { + .count = ATOMIC_INIT(1), + .wait = __WAIT_QUEUE_HEAD_INITIALIZER(device_event.wait) }; struct class_info { @@ -157,10 +164,8 @@ static const struct class_info clas_info[] = void usbfs_conn_disc_event(void) { - mutex_lock(&usbfs_mutex); - conndiscevcnt++; - mutex_unlock(&usbfs_mutex); - wake_up(&deviceconndiscwq); + atomic_add(2, &device_event.count); + wake_up(&device_event.wait); } static const char *class_decode(const int class) @@ -632,42 +637,16 @@ static ssize_t usb_device_read(struct file *file, char __user *buf, static unsigned int usb_device_poll(struct file *file, struct poll_table_struct *wait) { - struct usb_device_status *st; - unsigned int mask = 0; - - mutex_lock(&usbfs_mutex); - st = file->private_data; - if (!st) { - st = kmalloc(sizeof(struct usb_device_status), GFP_KERNEL); - if (!st) { - mutex_unlock(&usbfs_mutex); - return POLLIN; - } - - st->lastev = conndiscevcnt; - file->private_data = st; - mask = POLLIN; - } + unsigned int event_count; - if (file->f_mode & FMODE_READ) - poll_wait(file, &deviceconndiscwq, wait); - if (st->lastev != conndiscevcnt) - mask |= POLLIN; - st->lastev = conndiscevcnt; - mutex_unlock(&usbfs_mutex); - return mask; -} + poll_wait(file, &device_event.wait, wait); -static int usb_device_open(struct inode *inode, struct file *file) -{ - file->private_data = NULL; - return 0; -} + event_count = atomic_read(&device_event.count); + if (file->f_version != event_count) { + file->f_version = event_count; + return POLLIN | POLLRDNORM; + } -static int usb_device_release(struct inode *inode, struct file *file) -{ - kfree(file->private_data); - file->private_data = NULL; return 0; } @@ -699,6 +678,4 @@ const struct file_operations usbfs_devices_fops = { .llseek = usb_device_lseek, .read = usb_device_read, .poll = usb_device_poll, - .open = usb_device_open, - .release = usb_device_release, }; diff --git a/drivers/video/console/Kconfig b/drivers/video/console/Kconfig index fc7d9bbb548c..8e8f18d29d7a 100644 --- a/drivers/video/console/Kconfig +++ b/drivers/video/console/Kconfig @@ -37,6 +37,7 @@ config VGACON_SOFT_SCROLLBACK config VGACON_SOFT_SCROLLBACK_SIZE int "Scrollback Buffer Size (in KB)" depends on VGACON_SOFT_SCROLLBACK + range 1 1024 default "64" help Enter the amount of System RAM to allocate for the scrollback diff --git a/drivers/video/console/vgacon.c b/drivers/video/console/vgacon.c index cc4bbbe44aca..182dd6f8aadd 100644 --- a/drivers/video/console/vgacon.c +++ b/drivers/video/console/vgacon.c @@ -112,7 +112,7 @@ static int vga_video_font_height; static int vga_scan_lines __read_mostly; static unsigned int vga_rolled_over; -int vgacon_text_mode_force = 0; +static int vgacon_text_mode_force; bool vgacon_text_force(void) { diff --git a/drivers/virtio/virtio_pci.c b/drivers/virtio/virtio_pci.c index 1b6573216998..625447f645d9 100644 --- a/drivers/virtio/virtio_pci.c +++ b/drivers/virtio/virtio_pci.c @@ -649,6 +649,7 @@ static int __devinit virtio_pci_probe(struct pci_dev *pci_dev, goto out_req_regions; pci_set_drvdata(pci_dev, vp_dev); + pci_set_master(pci_dev); /* we use the subsystem vendor/device id as the virtio vendor/device * id. this allows us to use the same PCI vendor/device id for all diff --git a/drivers/xen/Kconfig b/drivers/xen/Kconfig index cab100acf983..fad3df2c1276 100644 --- a/drivers/xen/Kconfig +++ b/drivers/xen/Kconfig @@ -1,6 +1,8 @@ +menu "Xen driver support" + depends on XEN + config XEN_BALLOON bool "Xen memory balloon driver" - depends on XEN default y help The balloon driver allows the Xen domain to request more memory from @@ -20,7 +22,6 @@ config XEN_SCRUB_PAGES config XEN_DEV_EVTCHN tristate "Xen /dev/xen/evtchn device" - depends on XEN default y help The evtchn driver allows a userspace process to triger event @@ -30,7 +31,6 @@ config XEN_DEV_EVTCHN config XENFS tristate "Xen filesystem" - depends on XEN default y help The xen filesystem provides a way for domains to share @@ -53,11 +53,13 @@ config XEN_COMPAT_XENFS config XEN_SYS_HYPERVISOR bool "Create xen entries under /sys/hypervisor" - depends on XEN && SYSFS + depends on SYSFS select SYS_HYPERVISOR default y help Create entries under /sys/hypervisor describing the Xen hypervisor environment. When running native or in another virtual environment, /sys/hypervisor will still be present, - but will have no xen contents.
\ No newline at end of file + but will have no xen contents. + +endmenu diff --git a/fs/Kconfig b/fs/Kconfig index 64d44efad7a5..7405f071be67 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -177,6 +177,7 @@ source "fs/efs/Kconfig" source "fs/jffs2/Kconfig" # UBIFS File system configuration source "fs/ubifs/Kconfig" +source "fs/logfs/Kconfig" source "fs/cramfs/Kconfig" source "fs/squashfs/Kconfig" source "fs/freevxfs/Kconfig" diff --git a/fs/Makefile b/fs/Makefile index af6d04700d9c..c3633aa46911 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -99,6 +99,7 @@ obj-$(CONFIG_NTFS_FS) += ntfs/ obj-$(CONFIG_UFS_FS) += ufs/ obj-$(CONFIG_EFS_FS) += efs/ obj-$(CONFIG_JFFS2_FS) += jffs2/ +obj-$(CONFIG_LOGFS) += logfs/ obj-$(CONFIG_UBIFS_FS) += ubifs/ obj-$(CONFIG_AFFS_FS) += affs/ obj-$(CONFIG_ROMFS_FS) += romfs/ diff --git a/fs/attr.c b/fs/attr.c index 0a6ea54cde7d..0815e93bb487 100644 --- a/fs/attr.c +++ b/fs/attr.c @@ -81,7 +81,7 @@ int inode_newsize_ok(const struct inode *inode, loff_t offset) if (inode->i_size < offset) { unsigned long limit; - limit = current->signal->rlim[RLIMIT_FSIZE].rlim_cur; + limit = rlimit(RLIMIT_FSIZE); if (limit != RLIM_INFINITY && offset > limit) goto out_sig; if (offset > inode->i_sb->s_maxbytes) diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c index fdd397099172..15d80bb35d6f 100644 --- a/fs/binfmt_aout.c +++ b/fs/binfmt_aout.c @@ -24,6 +24,7 @@ #include <linux/binfmts.h> #include <linux/personality.h> #include <linux/init.h> +#include <linux/coredump.h> #include <asm/system.h> #include <asm/uaccess.h> @@ -60,26 +61,6 @@ static int set_brk(unsigned long start, unsigned long end) } /* - * These are the only things you should do on a core-file: use only these - * macros to write out all the necessary info. - */ - -static int dump_write(struct file *file, const void *addr, int nr) -{ - return file->f_op->write(file, addr, nr, &file->f_pos) == nr; -} - -#define DUMP_WRITE(addr, nr) \ - if (!dump_write(file, (void *)(addr), (nr))) \ - goto end_coredump; - -#define DUMP_SEEK(offset) \ -if (file->f_op->llseek) { \ - if (file->f_op->llseek(file,(offset),0) != (offset)) \ - goto end_coredump; \ -} else file->f_pos = (offset) - -/* * Routine writes a core dump image in the current directory. * Currently only a stub-function. * @@ -130,26 +111,31 @@ static int aout_core_dump(struct coredump_params *cprm) set_fs(KERNEL_DS); /* struct user */ - DUMP_WRITE(&dump,sizeof(dump)); + if (!dump_write(file, &dump, sizeof(dump))) + goto end_coredump; /* Now dump all of the user data. Include malloced stuff as well */ - DUMP_SEEK(PAGE_SIZE); + if (!dump_seek(cprm->file, PAGE_SIZE - sizeof(dump))) + goto end_coredump; /* now we start writing out the user space info */ set_fs(USER_DS); /* Dump the data area */ if (dump.u_dsize != 0) { dump_start = START_DATA(dump); dump_size = dump.u_dsize << PAGE_SHIFT; - DUMP_WRITE(dump_start,dump_size); + if (!dump_write(file, dump_start, dump_size)) + goto end_coredump; } /* Now prepare to dump the stack area */ if (dump.u_ssize != 0) { dump_start = START_STACK(dump); dump_size = dump.u_ssize << PAGE_SHIFT; - DUMP_WRITE(dump_start,dump_size); + if (!dump_write(file, dump_start, dump_size)) + goto end_coredump; } /* Finally dump the task struct. Not be used by gdb, but could be useful */ set_fs(KERNEL_DS); - DUMP_WRITE(current,sizeof(*current)); + if (!dump_write(file, current, sizeof(*current))) + goto end_coredump; end_coredump: set_fs(fs); return has_dumped; @@ -247,7 +233,7 @@ static int load_aout_binary(struct linux_binprm * bprm, struct pt_regs * regs) * size limits imposed on them by creating programs with large * arrays in the data or bss. */ - rlim = current->signal->rlim[RLIMIT_DATA].rlim_cur; + rlim = rlimit(RLIMIT_DATA); if (rlim >= RLIM_INFINITY) rlim = ~0; if (ex.a_data + ex.a_bss > rlim) diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index fd5b2ea5d299..535e763ab1a6 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -31,6 +31,7 @@ #include <linux/random.h> #include <linux/elf.h> #include <linux/utsname.h> +#include <linux/coredump.h> #include <asm/uaccess.h> #include <asm/param.h> #include <asm/page.h> @@ -1085,36 +1086,6 @@ out: * Modelled on fs/exec.c:aout_core_dump() * Jeremy Fitzhardinge <jeremy@sw.oz.au> */ -/* - * These are the only things you should do on a core-file: use only these - * functions to write out all the necessary info. - */ -static int dump_write(struct file *file, const void *addr, int nr) -{ - return file->f_op->write(file, addr, nr, &file->f_pos) == nr; -} - -static int dump_seek(struct file *file, loff_t off) -{ - if (file->f_op->llseek && file->f_op->llseek != no_llseek) { - if (file->f_op->llseek(file, off, SEEK_CUR) < 0) - return 0; - } else { - char *buf = (char *)get_zeroed_page(GFP_KERNEL); - if (!buf) - return 0; - while (off > 0) { - unsigned long n = off; - if (n > PAGE_SIZE) - n = PAGE_SIZE; - if (!dump_write(file, buf, n)) - return 0; - off -= n; - } - free_page((unsigned long)buf); - } - return 1; -} /* * Decide what to dump of a segment, part, all or none. @@ -1249,11 +1220,6 @@ static int writenote(struct memelfnote *men, struct file *file, } #undef DUMP_WRITE -#define DUMP_WRITE(addr, nr) \ - if ((size += (nr)) > cprm->limit || \ - !dump_write(cprm->file, (addr), (nr))) \ - goto end_coredump; - static void fill_elf_header(struct elfhdr *elf, int segs, u16 machine, u32 flags, u8 osabi) { @@ -1872,6 +1838,34 @@ static struct vm_area_struct *next_vma(struct vm_area_struct *this_vma, return gate_vma; } +static void fill_extnum_info(struct elfhdr *elf, struct elf_shdr *shdr4extnum, + elf_addr_t e_shoff, int segs) +{ + elf->e_shoff = e_shoff; + elf->e_shentsize = sizeof(*shdr4extnum); + elf->e_shnum = 1; + elf->e_shstrndx = SHN_UNDEF; + + memset(shdr4extnum, 0, sizeof(*shdr4extnum)); + + shdr4extnum->sh_type = SHT_NULL; + shdr4extnum->sh_size = elf->e_shnum; + shdr4extnum->sh_link = elf->e_shstrndx; + shdr4extnum->sh_info = segs; +} + +static size_t elf_core_vma_data_size(struct vm_area_struct *gate_vma, + unsigned long mm_flags) +{ + struct vm_area_struct *vma; + size_t size = 0; + + for (vma = first_vma(current, gate_vma); vma != NULL; + vma = next_vma(vma, gate_vma)) + size += vma_dump_size(vma, mm_flags); + return size; +} + /* * Actual dumper * @@ -1888,8 +1882,11 @@ static int elf_core_dump(struct coredump_params *cprm) struct vm_area_struct *vma, *gate_vma; struct elfhdr *elf = NULL; loff_t offset = 0, dataoff, foffset; - unsigned long mm_flags; struct elf_note_info info; + struct elf_phdr *phdr4note = NULL; + struct elf_shdr *shdr4extnum = NULL; + Elf_Half e_phnum; + elf_addr_t e_shoff; /* * We no longer stop all VM operations. @@ -1912,20 +1909,25 @@ static int elf_core_dump(struct coredump_params *cprm) * Please check DEFAULT_MAX_MAP_COUNT definition when you modify here. */ segs = current->mm->map_count; -#ifdef ELF_CORE_EXTRA_PHDRS - segs += ELF_CORE_EXTRA_PHDRS; -#endif + segs += elf_core_extra_phdrs(); gate_vma = get_gate_vma(current); if (gate_vma != NULL) segs++; + /* for notes section */ + segs++; + + /* If segs > PN_XNUM(0xffff), then e_phnum overflows. To avoid + * this, kernel supports extended numbering. Have a look at + * include/linux/elf.h for further information. */ + e_phnum = segs > PN_XNUM ? PN_XNUM : segs; + /* * Collect all the non-memory information about the process for the * notes. This also sets up the file header. */ - if (!fill_note_info(elf, segs + 1, /* including notes section */ - &info, cprm->signr, cprm->regs)) + if (!fill_note_info(elf, e_phnum, &info, cprm->signr, cprm->regs)) goto cleanup; has_dumped = 1; @@ -1934,31 +1936,47 @@ static int elf_core_dump(struct coredump_params *cprm) fs = get_fs(); set_fs(KERNEL_DS); - DUMP_WRITE(elf, sizeof(*elf)); offset += sizeof(*elf); /* Elf header */ - offset += (segs + 1) * sizeof(struct elf_phdr); /* Program headers */ + offset += segs * sizeof(struct elf_phdr); /* Program headers */ foffset = offset; /* Write notes phdr entry */ { - struct elf_phdr phdr; size_t sz = get_note_info_size(&info); sz += elf_coredump_extra_notes_size(); - fill_elf_note_phdr(&phdr, sz, offset); + phdr4note = kmalloc(sizeof(*phdr4note), GFP_KERNEL); + if (!phdr4note) + goto end_coredump; + + fill_elf_note_phdr(phdr4note, sz, offset); offset += sz; - DUMP_WRITE(&phdr, sizeof(phdr)); } dataoff = offset = roundup(offset, ELF_EXEC_PAGESIZE); - /* - * We must use the same mm->flags while dumping core to avoid - * inconsistency between the program headers and bodies, otherwise an - * unusable core file can be generated. - */ - mm_flags = current->mm->flags; + offset += elf_core_vma_data_size(gate_vma, cprm->mm_flags); + offset += elf_core_extra_data_size(); + e_shoff = offset; + + if (e_phnum == PN_XNUM) { + shdr4extnum = kmalloc(sizeof(*shdr4extnum), GFP_KERNEL); + if (!shdr4extnum) + goto end_coredump; + fill_extnum_info(elf, shdr4extnum, e_shoff, segs); + } + + offset = dataoff; + + size += sizeof(*elf); + if (size > cprm->limit || !dump_write(cprm->file, elf, sizeof(*elf))) + goto end_coredump; + + size += sizeof(*phdr4note); + if (size > cprm->limit + || !dump_write(cprm->file, phdr4note, sizeof(*phdr4note))) + goto end_coredump; /* Write program headers for segments dump */ for (vma = first_vma(current, gate_vma); vma != NULL; @@ -1969,7 +1987,7 @@ static int elf_core_dump(struct coredump_params *cprm) phdr.p_offset = offset; phdr.p_vaddr = vma->vm_start; phdr.p_paddr = 0; - phdr.p_filesz = vma_dump_size(vma, mm_flags); + phdr.p_filesz = vma_dump_size(vma, cprm->mm_flags); phdr.p_memsz = vma->vm_end - vma->vm_start; offset += phdr.p_filesz; phdr.p_flags = vma->vm_flags & VM_READ ? PF_R : 0; @@ -1979,12 +1997,14 @@ static int elf_core_dump(struct coredump_params *cprm) phdr.p_flags |= PF_X; phdr.p_align = ELF_EXEC_PAGESIZE; - DUMP_WRITE(&phdr, sizeof(phdr)); + size += sizeof(phdr); + if (size > cprm->limit + || !dump_write(cprm->file, &phdr, sizeof(phdr))) + goto end_coredump; } -#ifdef ELF_CORE_WRITE_EXTRA_PHDRS - ELF_CORE_WRITE_EXTRA_PHDRS; -#endif + if (!elf_core_write_extra_phdrs(cprm->file, offset, &size, cprm->limit)) + goto end_coredump; /* write out the notes section */ if (!write_note_info(&info, cprm->file, &foffset)) @@ -2002,7 +2022,7 @@ static int elf_core_dump(struct coredump_params *cprm) unsigned long addr; unsigned long end; - end = vma->vm_start + vma_dump_size(vma, mm_flags); + end = vma->vm_start + vma_dump_size(vma, cprm->mm_flags); for (addr = vma->vm_start; addr < end; addr += PAGE_SIZE) { struct page *page; @@ -2023,15 +2043,24 @@ static int elf_core_dump(struct coredump_params *cprm) } } -#ifdef ELF_CORE_WRITE_EXTRA_DATA - ELF_CORE_WRITE_EXTRA_DATA; -#endif + if (!elf_core_write_extra_data(cprm->file, &size, cprm->limit)) + goto end_coredump; + + if (e_phnum == PN_XNUM) { + size += sizeof(*shdr4extnum); + if (size > cprm->limit + || !dump_write(cprm->file, shdr4extnum, + sizeof(*shdr4extnum))) + goto end_coredump; + } end_coredump: set_fs(fs); cleanup: free_note_info(&info); + kfree(shdr4extnum); + kfree(phdr4note); kfree(elf); out: return has_dumped; diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index 18d77297ccc8..6d6a16c5e9bb 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -34,6 +34,7 @@ #include <linux/elf.h> #include <linux/elf-fdpic.h> #include <linux/elfcore.h> +#include <linux/coredump.h> #include <asm/uaccess.h> #include <asm/param.h> @@ -1216,26 +1217,6 @@ static int elf_fdpic_map_file_by_direct_mmap(struct elf_fdpic_params *params, #ifdef CONFIG_ELF_CORE /* - * These are the only things you should do on a core-file: use only these - * functions to write out all the necessary info. - */ -static int dump_write(struct file *file, const void *addr, int nr) -{ - return file->f_op->write(file, addr, nr, &file->f_pos) == nr; -} - -static int dump_seek(struct file *file, loff_t off) -{ - if (file->f_op->llseek) { - if (file->f_op->llseek(file, off, SEEK_SET) != off) - return 0; - } else { - file->f_pos = off; - } - return 1; -} - -/* * Decide whether a segment is worth dumping; default is yes to be * sure (missing info is worse than too much; etc). * Personally I'd include everything, and use the coredump limit... @@ -1313,35 +1294,35 @@ static int notesize(struct memelfnote *en) /* #define DEBUG */ -#define DUMP_WRITE(addr, nr) \ - do { if (!dump_write(file, (addr), (nr))) return 0; } while(0) -#define DUMP_SEEK(off) \ - do { if (!dump_seek(file, (off))) return 0; } while(0) +#define DUMP_WRITE(addr, nr, foffset) \ + do { if (!dump_write(file, (addr), (nr))) return 0; *foffset += (nr); } while(0) -static int writenote(struct memelfnote *men, struct file *file) +static int alignfile(struct file *file, loff_t *foffset) { - struct elf_note en; + static const char buf[4] = { 0, }; + DUMP_WRITE(buf, roundup(*foffset, 4) - *foffset, foffset); + return 1; +} +static int writenote(struct memelfnote *men, struct file *file, + loff_t *foffset) +{ + struct elf_note en; en.n_namesz = strlen(men->name) + 1; en.n_descsz = men->datasz; en.n_type = men->type; - DUMP_WRITE(&en, sizeof(en)); - DUMP_WRITE(men->name, en.n_namesz); - /* XXX - cast from long long to long to avoid need for libgcc.a */ - DUMP_SEEK(roundup((unsigned long)file->f_pos, 4)); /* XXX */ - DUMP_WRITE(men->data, men->datasz); - DUMP_SEEK(roundup((unsigned long)file->f_pos, 4)); /* XXX */ + DUMP_WRITE(&en, sizeof(en), foffset); + DUMP_WRITE(men->name, en.n_namesz, foffset); + if (!alignfile(file, foffset)) + return 0; + DUMP_WRITE(men->data, men->datasz, foffset); + if (!alignfile(file, foffset)) + return 0; return 1; } #undef DUMP_WRITE -#undef DUMP_SEEK - -#define DUMP_WRITE(addr, nr) \ - if ((size += (nr)) > cprm->limit || \ - !dump_write(cprm->file, (addr), (nr))) \ - goto end_coredump; static inline void fill_elf_fdpic_header(struct elfhdr *elf, int segs) { @@ -1524,6 +1505,22 @@ static int elf_dump_thread_status(long signr, struct elf_thread_status *t) return sz; } +static void fill_extnum_info(struct elfhdr *elf, struct elf_shdr *shdr4extnum, + elf_addr_t e_shoff, int segs) +{ + elf->e_shoff = e_shoff; + elf->e_shentsize = sizeof(*shdr4extnum); + elf->e_shnum = 1; + elf->e_shstrndx = SHN_UNDEF; + + memset(shdr4extnum, 0, sizeof(*shdr4extnum)); + + shdr4extnum->sh_type = SHT_NULL; + shdr4extnum->sh_size = elf->e_shnum; + shdr4extnum->sh_link = elf->e_shstrndx; + shdr4extnum->sh_info = segs; +} + /* * dump the segments for an MMU process */ @@ -1552,7 +1549,7 @@ static int elf_fdpic_dump_segments(struct file *file, size_t *size, err = -EIO; kunmap(page); page_cache_release(page); - } else if (!dump_seek(file, file->f_pos + PAGE_SIZE)) + } else if (!dump_seek(file, PAGE_SIZE)) err = -EFBIG; if (err) goto out; @@ -1588,6 +1585,17 @@ static int elf_fdpic_dump_segments(struct file *file, size_t *size, } #endif +static size_t elf_core_vma_data_size(unsigned long mm_flags) +{ + struct vm_area_struct *vma; + size_t size = 0; + + for (vma = current->mm->mmap; vma; vma->vm_next) + if (maydump(vma, mm_flags)) + size += vma->vm_end - vma->vm_start; + return size; +} + /* * Actual dumper * @@ -1605,7 +1613,7 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm) int i; struct vm_area_struct *vma; struct elfhdr *elf = NULL; - loff_t offset = 0, dataoff; + loff_t offset = 0, dataoff, foffset; int numnote; struct memelfnote *notes = NULL; struct elf_prstatus *prstatus = NULL; /* NT_PRSTATUS */ @@ -1618,7 +1626,10 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm) #endif int thread_status_size = 0; elf_addr_t *auxv; - unsigned long mm_flags; + struct elf_phdr *phdr4note = NULL; + struct elf_shdr *shdr4extnum = NULL; + Elf_Half e_phnum; + elf_addr_t e_shoff; /* * We no longer stop all VM operations. @@ -1683,12 +1694,18 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm) elf_core_copy_regs(&prstatus->pr_reg, cprm->regs); segs = current->mm->map_count; -#ifdef ELF_CORE_EXTRA_PHDRS - segs += ELF_CORE_EXTRA_PHDRS; -#endif + segs += elf_core_extra_phdrs(); + + /* for notes section */ + segs++; + + /* If segs > PN_XNUM(0xffff), then e_phnum overflows. To avoid + * this, kernel supports extended numbering. Have a look at + * include/linux/elf.h for further information. */ + e_phnum = segs > PN_XNUM ? PN_XNUM : segs; /* Set up header */ - fill_elf_fdpic_header(elf, segs + 1); /* including notes section */ + fill_elf_fdpic_header(elf, e_phnum); has_dumped = 1; current->flags |= PF_DUMPCORE; @@ -1727,13 +1744,12 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm) fs = get_fs(); set_fs(KERNEL_DS); - DUMP_WRITE(elf, sizeof(*elf)); offset += sizeof(*elf); /* Elf header */ - offset += (segs+1) * sizeof(struct elf_phdr); /* Program headers */ + offset += segs * sizeof(struct elf_phdr); /* Program headers */ + foffset = offset; /* Write notes phdr entry */ { - struct elf_phdr phdr; int sz = 0; for (i = 0; i < numnote; i++) @@ -1741,20 +1757,38 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm) sz += thread_status_size; - fill_elf_note_phdr(&phdr, sz, offset); + phdr4note = kmalloc(sizeof(*phdr4note), GFP_KERNEL); + if (!phdr4note) + goto end_coredump; + + fill_elf_note_phdr(phdr4note, sz, offset); offset += sz; - DUMP_WRITE(&phdr, sizeof(phdr)); } /* Page-align dumped data */ dataoff = offset = roundup(offset, ELF_EXEC_PAGESIZE); - /* - * We must use the same mm->flags while dumping core to avoid - * inconsistency between the program headers and bodies, otherwise an - * unusable core file can be generated. - */ - mm_flags = current->mm->flags; + offset += elf_core_vma_data_size(cprm->mm_flags); + offset += elf_core_extra_data_size(); + e_shoff = offset; + + if (e_phnum == PN_XNUM) { + shdr4extnum = kmalloc(sizeof(*shdr4extnum), GFP_KERNEL); + if (!shdr4extnum) + goto end_coredump; + fill_extnum_info(elf, shdr4extnum, e_shoff, segs); + } + + offset = dataoff; + + size += sizeof(*elf); + if (size > cprm->limit || !dump_write(cprm->file, elf, sizeof(*elf))) + goto end_coredump; + + size += sizeof(*phdr4note); + if (size > cprm->limit + || !dump_write(cprm->file, phdr4note, sizeof(*phdr4note))) + goto end_coredump; /* write program headers for segments dump */ for (vma = current->mm->mmap; vma; vma = vma->vm_next) { @@ -1767,7 +1801,7 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm) phdr.p_offset = offset; phdr.p_vaddr = vma->vm_start; phdr.p_paddr = 0; - phdr.p_filesz = maydump(vma, mm_flags) ? sz : 0; + phdr.p_filesz = maydump(vma, cprm->mm_flags) ? sz : 0; phdr.p_memsz = sz; offset += phdr.p_filesz; phdr.p_flags = vma->vm_flags & VM_READ ? PF_R : 0; @@ -1777,16 +1811,18 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm) phdr.p_flags |= PF_X; phdr.p_align = ELF_EXEC_PAGESIZE; - DUMP_WRITE(&phdr, sizeof(phdr)); + size += sizeof(phdr); + if (size > cprm->limit + || !dump_write(cprm->file, &phdr, sizeof(phdr))) + goto end_coredump; } -#ifdef ELF_CORE_WRITE_EXTRA_PHDRS - ELF_CORE_WRITE_EXTRA_PHDRS; -#endif + if (!elf_core_write_extra_phdrs(cprm->file, offset, &size, cprm->limit)) + goto end_coredump; /* write out the notes section */ for (i = 0; i < numnote; i++) - if (!writenote(notes + i, cprm->file)) + if (!writenote(notes + i, cprm->file, &foffset)) goto end_coredump; /* write out the thread status notes section */ @@ -1795,20 +1831,27 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm) list_entry(t, struct elf_thread_status, list); for (i = 0; i < tmp->num_notes; i++) - if (!writenote(&tmp->notes[i], cprm->file)) + if (!writenote(&tmp->notes[i], cprm->file, &foffset)) goto end_coredump; } - if (!dump_seek(cprm->file, dataoff)) + if (!dump_seek(cprm->file, dataoff - foffset)) goto end_coredump; if (elf_fdpic_dump_segments(cprm->file, &size, &cprm->limit, - mm_flags) < 0) + cprm->mm_flags) < 0) goto end_coredump; -#ifdef ELF_CORE_WRITE_EXTRA_DATA - ELF_CORE_WRITE_EXTRA_DATA; -#endif + if (!elf_core_write_extra_data(cprm->file, &size, cprm->limit)) + goto end_coredump; + + if (e_phnum == PN_XNUM) { + size += sizeof(*shdr4extnum); + if (size > cprm->limit + || !dump_write(cprm->file, shdr4extnum, + sizeof(*shdr4extnum))) + goto end_coredump; + } if (cprm->file->f_pos != offset) { /* Sanity check */ @@ -1826,7 +1869,7 @@ cleanup: list_del(tmp); kfree(list_entry(tmp, struct elf_thread_status, list)); } - + kfree(phdr4note); kfree(elf); kfree(prstatus); kfree(psinfo); diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c index 42c6b4a54445..e0e769bdca59 100644 --- a/fs/binfmt_flat.c +++ b/fs/binfmt_flat.c @@ -501,7 +501,7 @@ static int load_flat_file(struct linux_binprm * bprm, * size limits imposed on them by creating programs with large * arrays in the data or bss. */ - rlim = current->signal->rlim[RLIMIT_DATA].rlim_cur; + rlim = rlimit(RLIMIT_DATA); if (rlim >= RLIM_INFINITY) rlim = ~0; if (data_len + bss_len > rlim) { diff --git a/fs/compat_binfmt_elf.c b/fs/compat_binfmt_elf.c index 0adced2f296f..112e45a17e99 100644 --- a/fs/compat_binfmt_elf.c +++ b/fs/compat_binfmt_elf.c @@ -28,10 +28,12 @@ #undef elfhdr #undef elf_phdr +#undef elf_shdr #undef elf_note #undef elf_addr_t #define elfhdr elf32_hdr #define elf_phdr elf32_phdr +#define elf_shdr elf32_shdr #define elf_note elf32_note #define elf_addr_t Elf32_Addr diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index 0ca9ec4a79c3..6d55b61bfa79 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -545,7 +545,7 @@ static int mt_ioctl_trans(unsigned int fd, unsigned int cmd, void __user *argp) kcmd = MTIOCPOS; karg = &pos; break; - case MTIOCGET32: + default: /* MTIOCGET32 */ kcmd = MTIOCGET; karg = &get; break; @@ -663,7 +663,7 @@ static int raw_ioctl(unsigned fd, unsigned cmd, switch (cmd) { case RAW_SETBIND: - case RAW_GETBIND: { + default: { /* RAW_GETBIND */ struct raw_config_request req; mm_segment_t oldfs = get_fs(); diff --git a/fs/exec.c b/fs/exec.c index cce6bbdbdbb1..49cdaa19e5b9 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -195,7 +195,7 @@ static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, * to work from. */ rlim = current->signal->rlim; - if (size > rlim[RLIMIT_STACK].rlim_cur / 4) { + if (size > ACCESS_ONCE(rlim[RLIMIT_STACK].rlim_cur) / 4) { put_page(page); return NULL; } @@ -246,6 +246,7 @@ static int __bprm_mm_init(struct linux_binprm *bprm) vma->vm_start = vma->vm_end - PAGE_SIZE; vma->vm_flags = VM_STACK_FLAGS; vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); + INIT_LIST_HEAD(&vma->anon_vma_chain); err = insert_vm_struct(mm, vma); if (err) goto err; @@ -516,7 +517,8 @@ static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift) /* * cover the whole range: [new_start, old_end) */ - vma_adjust(vma, new_start, old_end, vma->vm_pgoff, NULL); + if (vma_adjust(vma, new_start, old_end, vma->vm_pgoff, NULL)) + return -ENOMEM; /* * move the page tables downwards, on failure we rely on @@ -547,15 +549,13 @@ static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift) tlb_finish_mmu(tlb, new_end, old_end); /* - * shrink the vma to just the new range. + * Shrink the vma to just the new range. Always succeeds. */ vma_adjust(vma, new_start, new_end, vma->vm_pgoff, NULL); return 0; } -#define EXTRA_STACK_VM_PAGES 20 /* random */ - /* * Finalizes the stack vm_area_struct. The flags and permissions are updated, * the stack is optionally relocated, and some extra space is added. @@ -577,7 +577,7 @@ int setup_arg_pages(struct linux_binprm *bprm, #ifdef CONFIG_STACK_GROWSUP /* Limit stack size to 1GB */ - stack_base = current->signal->rlim[RLIMIT_STACK].rlim_max; + stack_base = rlimit_max(RLIMIT_STACK); if (stack_base > (1 << 30)) stack_base = 1 << 30; @@ -630,7 +630,7 @@ int setup_arg_pages(struct linux_binprm *bprm, goto out_unlock; } - stack_expand = EXTRA_STACK_VM_PAGES * PAGE_SIZE; + stack_expand = 131072UL; /* randomly 32*4k (or 2*64k) pages */ stack_size = vma->vm_end - vma->vm_start; /* * Align this down to a page boundary as expand_stack @@ -718,6 +718,7 @@ static int exec_mmap(struct mm_struct *mm) /* Notify parent that we're no longer interested in the old VM */ tsk = current; old_mm = current->mm; + sync_mm_rss(tsk, old_mm); mm_release(tsk, old_mm); if (old_mm) { @@ -1532,7 +1533,7 @@ static int format_corename(char *corename, long signr) /* core limit size */ case 'c': rc = snprintf(out_ptr, out_end - out_ptr, - "%lu", current->signal->rlim[RLIMIT_CORE].rlim_cur); + "%lu", rlimit(RLIMIT_CORE)); if (rc > out_end - out_ptr) goto out; out_ptr += rc; @@ -1560,12 +1561,13 @@ out: return ispipe; } -static int zap_process(struct task_struct *start) +static int zap_process(struct task_struct *start, int exit_code) { struct task_struct *t; int nr = 0; start->signal->flags = SIGNAL_GROUP_EXIT; + start->signal->group_exit_code = exit_code; start->signal->group_stop_count = 0; t = start; @@ -1590,8 +1592,7 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm, spin_lock_irq(&tsk->sighand->siglock); if (!signal_group_exit(tsk->signal)) { mm->core_state = core_state; - tsk->signal->group_exit_code = exit_code; - nr = zap_process(tsk); + nr = zap_process(tsk, exit_code); } spin_unlock_irq(&tsk->sighand->siglock); if (unlikely(nr < 0)) @@ -1640,7 +1641,7 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm, if (p->mm) { if (unlikely(p->mm == mm)) { lock_task_sighand(p, &flags); - nr += zap_process(p); + nr += zap_process(p, exit_code); unlock_task_sighand(p, &flags); } break; @@ -1747,14 +1748,19 @@ void set_dumpable(struct mm_struct *mm, int value) } } -int get_dumpable(struct mm_struct *mm) +static int __get_dumpable(unsigned long mm_flags) { int ret; - ret = mm->flags & 0x3; + ret = mm_flags & MMF_DUMPABLE_MASK; return (ret >= 2) ? 2 : ret; } +int get_dumpable(struct mm_struct *mm) +{ + return __get_dumpable(mm->flags); +} + static void wait_for_dump_helpers(struct file *file) { struct pipe_inode_info *pipe; @@ -1797,7 +1803,13 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs) struct coredump_params cprm = { .signr = signr, .regs = regs, - .limit = current->signal->rlim[RLIMIT_CORE].rlim_cur, + .limit = rlimit(RLIMIT_CORE), + /* + * We must use the same mm->flags while dumping core to avoid + * inconsistency of bit flags, since this flag is not protected + * by any locks. + */ + .mm_flags = mm->flags, }; audit_core_dumps(signr); @@ -1816,7 +1828,7 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs) /* * If another thread got here first, or we are not dumpable, bail out. */ - if (mm->core_state || !get_dumpable(mm)) { + if (mm->core_state || !__get_dumpable(cprm.mm_flags)) { up_write(&mm->mmap_sem); put_cred(cred); goto fail; @@ -1827,7 +1839,8 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs) * process nor do we know its entire history. We only know it * was tainted so we dump it as root in mode 2. */ - if (get_dumpable(mm) == 2) { /* Setuid core dump mode */ + if (__get_dumpable(cprm.mm_flags) == 2) { + /* Setuid core dump mode */ flag = O_EXCL; /* Stop rewrite attacks */ cred->fsuid = 0; /* Dump root private */ } @@ -1923,8 +1936,9 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs) /* * Dont allow local users get cute and trick others to coredump * into their pre-created files: + * Note, this is not relevant for pipes */ - if (inode->i_uid != current_fsuid()) + if (!ispipe && (inode->i_uid != current_fsuid())) goto close_fail; if (!cprm.file->f_op) goto close_fail; diff --git a/fs/fcntl.c b/fs/fcntl.c index 97e01dc0d95f..452d02f9075e 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -344,7 +344,7 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg, switch (cmd) { case F_DUPFD: case F_DUPFD_CLOEXEC: - if (arg >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur) + if (arg >= rlimit(RLIMIT_NOFILE)) break; err = alloc_fd(arg, cmd == F_DUPFD_CLOEXEC ? O_CLOEXEC : 0); if (err >= 0) { diff --git a/fs/file.c b/fs/file.c index 38039af67663..34bb7f71d994 100644 --- a/fs/file.c +++ b/fs/file.c @@ -257,7 +257,7 @@ int expand_files(struct files_struct *files, int nr) * N.B. For clone tasks sharing a files structure, this test * will limit the total number of files that can be opened. */ - if (nr >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur) + if (nr >= rlimit(RLIMIT_NOFILE)) return -EMFILE; /* Do we need to expand? */ diff --git a/fs/file_table.c b/fs/file_table.c index b98404b54383..32d12b78bac8 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -393,7 +393,9 @@ retry: continue; if (!(f->f_mode & FMODE_WRITE)) continue; + spin_lock(&f->f_lock); f->f_mode &= ~FMODE_WRITE; + spin_unlock(&f->f_lock); if (file_check_writeable(f) != 0) continue; file_release_write(f); diff --git a/fs/lockd/host.c b/fs/lockd/host.c index 4600c2037b8b..bb464d12104c 100644 --- a/fs/lockd/host.c +++ b/fs/lockd/host.c @@ -479,8 +479,8 @@ again: mutex_lock(&nlm_host_mutex); } } } - mutex_unlock(&nlm_host_mutex); + nsm_release(nsm); } /* diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c index f956651d0f65..fefa4df3f005 100644 --- a/fs/lockd/mon.c +++ b/fs/lockd/mon.c @@ -349,9 +349,9 @@ retry: * nsm_reboot_lookup - match NLMPROC_SM_NOTIFY arguments to an nsm_handle * @info: pointer to NLMPROC_SM_NOTIFY arguments * - * Returns a matching nsm_handle if found in the nsm cache; the returned - * nsm_handle's reference count is bumped and sm_monitored is cleared. - * Otherwise returns NULL if some error occurred. + * Returns a matching nsm_handle if found in the nsm cache. The returned + * nsm_handle's reference count is bumped. Otherwise returns NULL if some + * error occurred. */ struct nsm_handle *nsm_reboot_lookup(const struct nlm_reboot *info) { @@ -370,12 +370,6 @@ struct nsm_handle *nsm_reboot_lookup(const struct nlm_reboot *info) atomic_inc(&cached->sm_count); spin_unlock(&nsm_lock); - /* - * During subsequent lock activity, force a fresh - * notification to be set up for this host. - */ - cached->sm_monitored = 0; - dprintk("lockd: host %s (%s) rebooted, cnt %d\n", cached->sm_name, cached->sm_addrbuf, atomic_read(&cached->sm_count)); diff --git a/fs/lockd/svc.c b/fs/lockd/svc.c index e50cfa3d9654..7d150517ddf0 100644 --- a/fs/lockd/svc.c +++ b/fs/lockd/svc.c @@ -243,11 +243,9 @@ static int make_socks(struct svc_serv *serv) if (err < 0) goto out_err; -#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) err = create_lockd_family(serv, PF_INET6); if (err < 0 && err != -EAFNOSUPPORT) goto out_err; -#endif /* CONFIG_IPV6 || CONFIG_IPV6_MODULE */ warned = 0; return 0; diff --git a/fs/logfs/Kconfig b/fs/logfs/Kconfig new file mode 100644 index 000000000000..daf9a9b32dd3 --- /dev/null +++ b/fs/logfs/Kconfig @@ -0,0 +1,17 @@ +config LOGFS + tristate "LogFS file system (EXPERIMENTAL)" + depends on (MTD || BLOCK) && EXPERIMENTAL + select ZLIB_INFLATE + select ZLIB_DEFLATE + select CRC32 + select BTREE + help + Flash filesystem aimed to scale efficiently to large devices. + In comparison to JFFS2 it offers significantly faster mount + times and potentially less RAM usage, although the latter has + not been measured yet. + + In its current state it is still very experimental and should + not be used for other than testing purposes. + + If unsure, say N. diff --git a/fs/logfs/Makefile b/fs/logfs/Makefile new file mode 100644 index 000000000000..4820027787ee --- /dev/null +++ b/fs/logfs/Makefile @@ -0,0 +1,13 @@ +obj-$(CONFIG_LOGFS) += logfs.o + +logfs-y += compr.o +logfs-y += dir.o +logfs-y += file.o +logfs-y += gc.o +logfs-y += inode.o +logfs-y += journal.o +logfs-y += readwrite.o +logfs-y += segment.o +logfs-y += super.o +logfs-$(CONFIG_BLOCK) += dev_bdev.o +logfs-$(CONFIG_MTD) += dev_mtd.o diff --git a/fs/logfs/compr.c b/fs/logfs/compr.c new file mode 100644 index 000000000000..44bbfd249abc --- /dev/null +++ b/fs/logfs/compr.c @@ -0,0 +1,95 @@ +/* + * fs/logfs/compr.c - compression routines + * + * As should be obvious for Linux kernel code, license is GPLv2 + * + * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> + */ +#include "logfs.h" +#include <linux/vmalloc.h> +#include <linux/zlib.h> + +#define COMPR_LEVEL 3 + +static DEFINE_MUTEX(compr_mutex); +static struct z_stream_s stream; + +int logfs_compress(void *in, void *out, size_t inlen, size_t outlen) +{ + int err, ret; + + ret = -EIO; + mutex_lock(&compr_mutex); + err = zlib_deflateInit(&stream, COMPR_LEVEL); + if (err != Z_OK) + goto error; + + stream.next_in = in; + stream.avail_in = inlen; + stream.total_in = 0; + stream.next_out = out; + stream.avail_out = outlen; + stream.total_out = 0; + + err = zlib_deflate(&stream, Z_FINISH); + if (err != Z_STREAM_END) + goto error; + + err = zlib_deflateEnd(&stream); + if (err != Z_OK) + goto error; + + if (stream.total_out >= stream.total_in) + goto error; + + ret = stream.total_out; +error: + mutex_unlock(&compr_mutex); + return ret; +} + +int logfs_uncompress(void *in, void *out, size_t inlen, size_t outlen) +{ + int err, ret; + + ret = -EIO; + mutex_lock(&compr_mutex); + err = zlib_inflateInit(&stream); + if (err != Z_OK) + goto error; + + stream.next_in = in; + stream.avail_in = inlen; + stream.total_in = 0; + stream.next_out = out; + stream.avail_out = outlen; + stream.total_out = 0; + + err = zlib_inflate(&stream, Z_FINISH); + if (err != Z_STREAM_END) + goto error; + + err = zlib_inflateEnd(&stream); + if (err != Z_OK) + goto error; + + ret = 0; +error: + mutex_unlock(&compr_mutex); + return ret; +} + +int __init logfs_compr_init(void) +{ + size_t size = max(zlib_deflate_workspacesize(), + zlib_inflate_workspacesize()); + stream.workspace = vmalloc(size); + if (!stream.workspace) + return -ENOMEM; + return 0; +} + +void logfs_compr_exit(void) +{ + vfree(stream.workspace); +} diff --git a/fs/logfs/dev_bdev.c b/fs/logfs/dev_bdev.c new file mode 100644 index 000000000000..9718c22f186d --- /dev/null +++ b/fs/logfs/dev_bdev.c @@ -0,0 +1,327 @@ +/* + * fs/logfs/dev_bdev.c - Device access methods for block devices + * + * As should be obvious for Linux kernel code, license is GPLv2 + * + * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> + */ +#include "logfs.h" +#include <linux/bio.h> +#include <linux/blkdev.h> +#include <linux/buffer_head.h> + +#define PAGE_OFS(ofs) ((ofs) & (PAGE_SIZE-1)) + +static void request_complete(struct bio *bio, int err) +{ + complete((struct completion *)bio->bi_private); +} + +static int sync_request(struct page *page, struct block_device *bdev, int rw) +{ + struct bio bio; + struct bio_vec bio_vec; + struct completion complete; + + bio_init(&bio); + bio.bi_io_vec = &bio_vec; + bio_vec.bv_page = page; + bio_vec.bv_len = PAGE_SIZE; + bio_vec.bv_offset = 0; + bio.bi_vcnt = 1; + bio.bi_idx = 0; + bio.bi_size = PAGE_SIZE; + bio.bi_bdev = bdev; + bio.bi_sector = page->index * (PAGE_SIZE >> 9); + init_completion(&complete); + bio.bi_private = &complete; + bio.bi_end_io = request_complete; + + submit_bio(rw, &bio); + generic_unplug_device(bdev_get_queue(bdev)); + wait_for_completion(&complete); + return test_bit(BIO_UPTODATE, &bio.bi_flags) ? 0 : -EIO; +} + +static int bdev_readpage(void *_sb, struct page *page) +{ + struct super_block *sb = _sb; + struct block_device *bdev = logfs_super(sb)->s_bdev; + int err; + + err = sync_request(page, bdev, READ); + if (err) { + ClearPageUptodate(page); + SetPageError(page); + } else { + SetPageUptodate(page); + ClearPageError(page); + } + unlock_page(page); + return err; +} + +static DECLARE_WAIT_QUEUE_HEAD(wq); + +static void writeseg_end_io(struct bio *bio, int err) +{ + const int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags); + struct bio_vec *bvec = bio->bi_io_vec + bio->bi_vcnt - 1; + struct super_block *sb = bio->bi_private; + struct logfs_super *super = logfs_super(sb); + struct page *page; + + BUG_ON(!uptodate); /* FIXME: Retry io or write elsewhere */ + BUG_ON(err); + BUG_ON(bio->bi_vcnt == 0); + do { + page = bvec->bv_page; + if (--bvec >= bio->bi_io_vec) + prefetchw(&bvec->bv_page->flags); + + end_page_writeback(page); + } while (bvec >= bio->bi_io_vec); + bio_put(bio); + if (atomic_dec_and_test(&super->s_pending_writes)) + wake_up(&wq); +} + +static int __bdev_writeseg(struct super_block *sb, u64 ofs, pgoff_t index, + size_t nr_pages) +{ + struct logfs_super *super = logfs_super(sb); + struct address_space *mapping = super->s_mapping_inode->i_mapping; + struct bio *bio; + struct page *page; + struct request_queue *q = bdev_get_queue(sb->s_bdev); + unsigned int max_pages = queue_max_hw_sectors(q) >> (PAGE_SHIFT - 9); + int i; + + bio = bio_alloc(GFP_NOFS, max_pages); + BUG_ON(!bio); /* FIXME: handle this */ + + for (i = 0; i < nr_pages; i++) { + if (i >= max_pages) { + /* Block layer cannot split bios :( */ + bio->bi_vcnt = i; + bio->bi_idx = 0; + bio->bi_size = i * PAGE_SIZE; + bio->bi_bdev = super->s_bdev; + bio->bi_sector = ofs >> 9; + bio->bi_private = sb; + bio->bi_end_io = writeseg_end_io; + atomic_inc(&super->s_pending_writes); + submit_bio(WRITE, bio); + + ofs += i * PAGE_SIZE; + index += i; + nr_pages -= i; + i = 0; + + bio = bio_alloc(GFP_NOFS, max_pages); + BUG_ON(!bio); + } + page = find_lock_page(mapping, index + i); + BUG_ON(!page); + bio->bi_io_vec[i].bv_page = page; + bio->bi_io_vec[i].bv_len = PAGE_SIZE; + bio->bi_io_vec[i].bv_offset = 0; + + BUG_ON(PageWriteback(page)); + set_page_writeback(page); + unlock_page(page); + } + bio->bi_vcnt = nr_pages; + bio->bi_idx = 0; + bio->bi_size = nr_pages * PAGE_SIZE; + bio->bi_bdev = super->s_bdev; + bio->bi_sector = ofs >> 9; + bio->bi_private = sb; + bio->bi_end_io = writeseg_end_io; + atomic_inc(&super->s_pending_writes); + submit_bio(WRITE, bio); + return 0; +} + +static void bdev_writeseg(struct super_block *sb, u64 ofs, size_t len) +{ + struct logfs_super *super = logfs_super(sb); + int head; + + BUG_ON(super->s_flags & LOGFS_SB_FLAG_RO); + + if (len == 0) { + /* This can happen when the object fit perfectly into a + * segment, the segment gets written per sync and subsequently + * closed. + */ + return; + } + head = ofs & (PAGE_SIZE - 1); + if (head) { + ofs -= head; + len += head; + } + len = PAGE_ALIGN(len); + __bdev_writeseg(sb, ofs, ofs >> PAGE_SHIFT, len >> PAGE_SHIFT); + generic_unplug_device(bdev_get_queue(logfs_super(sb)->s_bdev)); +} + + +static void erase_end_io(struct bio *bio, int err) +{ + const int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags); + struct super_block *sb = bio->bi_private; + struct logfs_super *super = logfs_super(sb); + + BUG_ON(!uptodate); /* FIXME: Retry io or write elsewhere */ + BUG_ON(err); + BUG_ON(bio->bi_vcnt == 0); + bio_put(bio); + if (atomic_dec_and_test(&super->s_pending_writes)) + wake_up(&wq); +} + +static int do_erase(struct super_block *sb, u64 ofs, pgoff_t index, + size_t nr_pages) +{ + struct logfs_super *super = logfs_super(sb); + struct bio *bio; + struct request_queue *q = bdev_get_queue(sb->s_bdev); + unsigned int max_pages = queue_max_hw_sectors(q) >> (PAGE_SHIFT - 9); + int i; + + bio = bio_alloc(GFP_NOFS, max_pages); + BUG_ON(!bio); /* FIXME: handle this */ + + for (i = 0; i < nr_pages; i++) { + if (i >= max_pages) { + /* Block layer cannot split bios :( */ + bio->bi_vcnt = i; + bio->bi_idx = 0; + bio->bi_size = i * PAGE_SIZE; + bio->bi_bdev = super->s_bdev; + bio->bi_sector = ofs >> 9; + bio->bi_private = sb; + bio->bi_end_io = erase_end_io; + atomic_inc(&super->s_pending_writes); + submit_bio(WRITE, bio); + + ofs += i * PAGE_SIZE; + index += i; + nr_pages -= i; + i = 0; + + bio = bio_alloc(GFP_NOFS, max_pages); + BUG_ON(!bio); + } + bio->bi_io_vec[i].bv_page = super->s_erase_page; + bio->bi_io_vec[i].bv_len = PAGE_SIZE; + bio->bi_io_vec[i].bv_offset = 0; + } + bio->bi_vcnt = nr_pages; + bio->bi_idx = 0; + bio->bi_size = nr_pages * PAGE_SIZE; + bio->bi_bdev = super->s_bdev; + bio->bi_sector = ofs >> 9; + bio->bi_private = sb; + bio->bi_end_io = erase_end_io; + atomic_inc(&super->s_pending_writes); + submit_bio(WRITE, bio); + return 0; +} + +static int bdev_erase(struct super_block *sb, loff_t to, size_t len, + int ensure_write) +{ + struct logfs_super *super = logfs_super(sb); + + BUG_ON(to & (PAGE_SIZE - 1)); + BUG_ON(len & (PAGE_SIZE - 1)); + + if (super->s_flags & LOGFS_SB_FLAG_RO) + return -EROFS; + + if (ensure_write) { + /* + * Object store doesn't care whether erases happen or not. + * But for the journal they are required. Otherwise a scan + * can find an old commit entry and assume it is the current + * one, travelling back in time. + */ + do_erase(sb, to, to >> PAGE_SHIFT, len >> PAGE_SHIFT); + } + + return 0; +} + +static void bdev_sync(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + + wait_event(wq, atomic_read(&super->s_pending_writes) == 0); +} + +static struct page *bdev_find_first_sb(struct super_block *sb, u64 *ofs) +{ + struct logfs_super *super = logfs_super(sb); + struct address_space *mapping = super->s_mapping_inode->i_mapping; + filler_t *filler = bdev_readpage; + + *ofs = 0; + return read_cache_page(mapping, 0, filler, sb); +} + +static struct page *bdev_find_last_sb(struct super_block *sb, u64 *ofs) +{ + struct logfs_super *super = logfs_super(sb); + struct address_space *mapping = super->s_mapping_inode->i_mapping; + filler_t *filler = bdev_readpage; + u64 pos = (super->s_bdev->bd_inode->i_size & ~0xfffULL) - 0x1000; + pgoff_t index = pos >> PAGE_SHIFT; + + *ofs = pos; + return read_cache_page(mapping, index, filler, sb); +} + +static int bdev_write_sb(struct super_block *sb, struct page *page) +{ + struct block_device *bdev = logfs_super(sb)->s_bdev; + + /* Nothing special to do for block devices. */ + return sync_request(page, bdev, WRITE); +} + +static void bdev_put_device(struct super_block *sb) +{ + close_bdev_exclusive(logfs_super(sb)->s_bdev, FMODE_READ|FMODE_WRITE); +} + +static const struct logfs_device_ops bd_devops = { + .find_first_sb = bdev_find_first_sb, + .find_last_sb = bdev_find_last_sb, + .write_sb = bdev_write_sb, + .readpage = bdev_readpage, + .writeseg = bdev_writeseg, + .erase = bdev_erase, + .sync = bdev_sync, + .put_device = bdev_put_device, +}; + +int logfs_get_sb_bdev(struct file_system_type *type, int flags, + const char *devname, struct vfsmount *mnt) +{ + struct block_device *bdev; + + bdev = open_bdev_exclusive(devname, FMODE_READ|FMODE_WRITE, type); + if (IS_ERR(bdev)) + return PTR_ERR(bdev); + + if (MAJOR(bdev->bd_dev) == MTD_BLOCK_MAJOR) { + int mtdnr = MINOR(bdev->bd_dev); + close_bdev_exclusive(bdev, FMODE_READ|FMODE_WRITE); + return logfs_get_sb_mtd(type, flags, mtdnr, mnt); + } + + return logfs_get_sb_device(type, flags, NULL, bdev, &bd_devops, mnt); +} diff --git a/fs/logfs/dev_mtd.c b/fs/logfs/dev_mtd.c new file mode 100644 index 000000000000..cafb6ef2e05b --- /dev/null +++ b/fs/logfs/dev_mtd.c @@ -0,0 +1,254 @@ +/* + * fs/logfs/dev_mtd.c - Device access methods for MTD + * + * As should be obvious for Linux kernel code, license is GPLv2 + * + * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> + */ +#include "logfs.h" +#include <linux/completion.h> +#include <linux/mount.h> +#include <linux/sched.h> + +#define PAGE_OFS(ofs) ((ofs) & (PAGE_SIZE-1)) + +static int mtd_read(struct super_block *sb, loff_t ofs, size_t len, void *buf) +{ + struct mtd_info *mtd = logfs_super(sb)->s_mtd; + size_t retlen; + int ret; + + ret = mtd->read(mtd, ofs, len, &retlen, buf); + BUG_ON(ret == -EINVAL); + if (ret) + return ret; + + /* Not sure if we should loop instead. */ + if (retlen != len) + return -EIO; + + return 0; +} + +static int mtd_write(struct super_block *sb, loff_t ofs, size_t len, void *buf) +{ + struct logfs_super *super = logfs_super(sb); + struct mtd_info *mtd = super->s_mtd; + size_t retlen; + loff_t page_start, page_end; + int ret; + + if (super->s_flags & LOGFS_SB_FLAG_RO) + return -EROFS; + + BUG_ON((ofs >= mtd->size) || (len > mtd->size - ofs)); + BUG_ON(ofs != (ofs >> super->s_writeshift) << super->s_writeshift); + BUG_ON(len > PAGE_CACHE_SIZE); + page_start = ofs & PAGE_CACHE_MASK; + page_end = PAGE_CACHE_ALIGN(ofs + len) - 1; + ret = mtd->write(mtd, ofs, len, &retlen, buf); + if (ret || (retlen != len)) + return -EIO; + + return 0; +} + +/* + * For as long as I can remember (since about 2001) mtd->erase has been an + * asynchronous interface lacking the first driver to actually use the + * asynchronous properties. So just to prevent the first implementor of such + * a thing from breaking logfs in 2350, we do the usual pointless dance to + * declare a completion variable and wait for completion before returning + * from mtd_erase(). What an excercise in futility! + */ +static void logfs_erase_callback(struct erase_info *ei) +{ + complete((struct completion *)ei->priv); +} + +static int mtd_erase_mapping(struct super_block *sb, loff_t ofs, size_t len) +{ + struct logfs_super *super = logfs_super(sb); + struct address_space *mapping = super->s_mapping_inode->i_mapping; + struct page *page; + pgoff_t index = ofs >> PAGE_SHIFT; + + for (index = ofs >> PAGE_SHIFT; index < (ofs + len) >> PAGE_SHIFT; index++) { + page = find_get_page(mapping, index); + if (!page) + continue; + memset(page_address(page), 0xFF, PAGE_SIZE); + page_cache_release(page); + } + return 0; +} + +static int mtd_erase(struct super_block *sb, loff_t ofs, size_t len, + int ensure_write) +{ + struct mtd_info *mtd = logfs_super(sb)->s_mtd; + struct erase_info ei; + DECLARE_COMPLETION_ONSTACK(complete); + int ret; + + BUG_ON(len % mtd->erasesize); + if (logfs_super(sb)->s_flags & LOGFS_SB_FLAG_RO) + return -EROFS; + + memset(&ei, 0, sizeof(ei)); + ei.mtd = mtd; + ei.addr = ofs; + ei.len = len; + ei.callback = logfs_erase_callback; + ei.priv = (long)&complete; + ret = mtd->erase(mtd, &ei); + if (ret) + return -EIO; + + wait_for_completion(&complete); + if (ei.state != MTD_ERASE_DONE) + return -EIO; + return mtd_erase_mapping(sb, ofs, len); +} + +static void mtd_sync(struct super_block *sb) +{ + struct mtd_info *mtd = logfs_super(sb)->s_mtd; + + if (mtd->sync) + mtd->sync(mtd); +} + +static int mtd_readpage(void *_sb, struct page *page) +{ + struct super_block *sb = _sb; + int err; + + err = mtd_read(sb, page->index << PAGE_SHIFT, PAGE_SIZE, + page_address(page)); + if (err == -EUCLEAN) { + err = 0; + /* FIXME: force GC this segment */ + } + if (err) { + ClearPageUptodate(page); + SetPageError(page); + } else { + SetPageUptodate(page); + ClearPageError(page); + } + unlock_page(page); + return err; +} + +static struct page *mtd_find_first_sb(struct super_block *sb, u64 *ofs) +{ + struct logfs_super *super = logfs_super(sb); + struct address_space *mapping = super->s_mapping_inode->i_mapping; + filler_t *filler = mtd_readpage; + struct mtd_info *mtd = super->s_mtd; + + if (!mtd->block_isbad) + return NULL; + + *ofs = 0; + while (mtd->block_isbad(mtd, *ofs)) { + *ofs += mtd->erasesize; + if (*ofs >= mtd->size) + return NULL; + } + BUG_ON(*ofs & ~PAGE_MASK); + return read_cache_page(mapping, *ofs >> PAGE_SHIFT, filler, sb); +} + +static struct page *mtd_find_last_sb(struct super_block *sb, u64 *ofs) +{ + struct logfs_super *super = logfs_super(sb); + struct address_space *mapping = super->s_mapping_inode->i_mapping; + filler_t *filler = mtd_readpage; + struct mtd_info *mtd = super->s_mtd; + + if (!mtd->block_isbad) + return NULL; + + *ofs = mtd->size - mtd->erasesize; + while (mtd->block_isbad(mtd, *ofs)) { + *ofs -= mtd->erasesize; + if (*ofs <= 0) + return NULL; + } + *ofs = *ofs + mtd->erasesize - 0x1000; + BUG_ON(*ofs & ~PAGE_MASK); + return read_cache_page(mapping, *ofs >> PAGE_SHIFT, filler, sb); +} + +static int __mtd_writeseg(struct super_block *sb, u64 ofs, pgoff_t index, + size_t nr_pages) +{ + struct logfs_super *super = logfs_super(sb); + struct address_space *mapping = super->s_mapping_inode->i_mapping; + struct page *page; + int i, err; + + for (i = 0; i < nr_pages; i++) { + page = find_lock_page(mapping, index + i); + BUG_ON(!page); + + err = mtd_write(sb, page->index << PAGE_SHIFT, PAGE_SIZE, + page_address(page)); + unlock_page(page); + page_cache_release(page); + if (err) + return err; + } + return 0; +} + +static void mtd_writeseg(struct super_block *sb, u64 ofs, size_t len) +{ + struct logfs_super *super = logfs_super(sb); + int head; + + if (super->s_flags & LOGFS_SB_FLAG_RO) + return; + + if (len == 0) { + /* This can happen when the object fit perfectly into a + * segment, the segment gets written per sync and subsequently + * closed. + */ + return; + } + head = ofs & (PAGE_SIZE - 1); + if (head) { + ofs -= head; + len += head; + } + len = PAGE_ALIGN(len); + __mtd_writeseg(sb, ofs, ofs >> PAGE_SHIFT, len >> PAGE_SHIFT); +} + +static void mtd_put_device(struct super_block *sb) +{ + put_mtd_device(logfs_super(sb)->s_mtd); +} + +static const struct logfs_device_ops mtd_devops = { + .find_first_sb = mtd_find_first_sb, + .find_last_sb = mtd_find_last_sb, + .readpage = mtd_readpage, + .writeseg = mtd_writeseg, + .erase = mtd_erase, + .sync = mtd_sync, + .put_device = mtd_put_device, +}; + +int logfs_get_sb_mtd(struct file_system_type *type, int flags, + int mtdnr, struct vfsmount *mnt) +{ + struct mtd_info *mtd; + const struct logfs_device_ops *devops = &mtd_devops; + + mtd = get_mtd_device(NULL, mtdnr); + return logfs_get_sb_device(type, flags, mtd, NULL, devops, mnt); +} diff --git a/fs/logfs/dir.c b/fs/logfs/dir.c new file mode 100644 index 000000000000..56a8bfbb0120 --- /dev/null +++ b/fs/logfs/dir.c @@ -0,0 +1,827 @@ +/* + * fs/logfs/dir.c - directory-related code + * + * As should be obvious for Linux kernel code, license is GPLv2 + * + * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> + */ +#include "logfs.h" + + +/* + * Atomic dir operations + * + * Directory operations are by default not atomic. Dentries and Inodes are + * created/removed/altered in seperate operations. Therefore we need to do + * a small amount of journaling. + * + * Create, link, mkdir, mknod and symlink all share the same function to do + * the work: __logfs_create. This function works in two atomic steps: + * 1. allocate inode (remember in journal) + * 2. allocate dentry (clear journal) + * + * As we can only get interrupted between the two, when the inode we just + * created is simply stored in the anchor. On next mount, if we were + * interrupted, we delete the inode. From a users point of view the + * operation never happened. + * + * Unlink and rmdir also share the same function: unlink. Again, this + * function works in two atomic steps + * 1. remove dentry (remember inode in journal) + * 2. unlink inode (clear journal) + * + * And again, on the next mount, if we were interrupted, we delete the inode. + * From a users point of view the operation succeeded. + * + * Rename is the real pain to deal with, harder than all the other methods + * combined. Depending on the circumstances we can run into three cases. + * A "target rename" where the target dentry already existed, a "local + * rename" where both parent directories are identical or a "cross-directory + * rename" in the remaining case. + * + * Local rename is atomic, as the old dentry is simply rewritten with a new + * name. + * + * Cross-directory rename works in two steps, similar to __logfs_create and + * logfs_unlink: + * 1. Write new dentry (remember old dentry in journal) + * 2. Remove old dentry (clear journal) + * + * Here we remember a dentry instead of an inode. On next mount, if we were + * interrupted, we delete the dentry. From a users point of view, the + * operation succeeded. + * + * Target rename works in three atomic steps: + * 1. Attach old inode to new dentry (remember old dentry and new inode) + * 2. Remove old dentry (still remember the new inode) + * 3. Remove victim inode + * + * Here we remember both an inode an a dentry. If we get interrupted + * between steps 1 and 2, we delete both the dentry and the inode. If + * we get interrupted between steps 2 and 3, we delete just the inode. + * In either case, the remaining objects are deleted on next mount. From + * a users point of view, the operation succeeded. + */ + +static int write_dir(struct inode *dir, struct logfs_disk_dentry *dd, + loff_t pos) +{ + return logfs_inode_write(dir, dd, sizeof(*dd), pos, WF_LOCK, NULL); +} + +static int write_inode(struct inode *inode) +{ + return __logfs_write_inode(inode, WF_LOCK); +} + +static s64 dir_seek_data(struct inode *inode, s64 pos) +{ + s64 new_pos = logfs_seek_data(inode, pos); + + return max(pos, new_pos - 1); +} + +static int beyond_eof(struct inode *inode, loff_t bix) +{ + loff_t pos = bix << inode->i_sb->s_blocksize_bits; + return pos >= i_size_read(inode); +} + +/* + * Prime value was chosen to be roughly 256 + 26. r5 hash uses 11, + * so short names (len <= 9) don't even occupy the complete 32bit name + * space. A prime >256 ensures short names quickly spread the 32bit + * name space. Add about 26 for the estimated amount of information + * of each character and pick a prime nearby, preferrably a bit-sparse + * one. + */ +static u32 hash_32(const char *s, int len, u32 seed) +{ + u32 hash = seed; + int i; + + for (i = 0; i < len; i++) + hash = hash * 293 + s[i]; + return hash; +} + +/* + * We have to satisfy several conflicting requirements here. Small + * directories should stay fairly compact and not require too many + * indirect blocks. The number of possible locations for a given hash + * should be small to make lookup() fast. And we should try hard not + * to overflow the 32bit name space or nfs and 32bit host systems will + * be unhappy. + * + * So we use the following scheme. First we reduce the hash to 0..15 + * and try a direct block. If that is occupied we reduce the hash to + * 16..255 and try an indirect block. Same for 2x and 3x indirect + * blocks. Lastly we reduce the hash to 0x800_0000 .. 0xffff_ffff, + * but use buckets containing eight entries instead of a single one. + * + * Using 16 entries should allow for a reasonable amount of hash + * collisions, so the 32bit name space can be packed fairly tight + * before overflowing. Oh and currently we don't overflow but return + * and error. + * + * How likely are collisions? Doing the appropriate math is beyond me + * and the Bronstein textbook. But running a test program to brute + * force collisions for a couple of days showed that on average the + * first collision occurs after 598M entries, with 290M being the + * smallest result. Obviously 21 entries could already cause a + * collision if all entries are carefully chosen. + */ +static pgoff_t hash_index(u32 hash, int round) +{ + u32 i0_blocks = I0_BLOCKS; + u32 i1_blocks = I1_BLOCKS; + u32 i2_blocks = I2_BLOCKS; + u32 i3_blocks = I3_BLOCKS; + + switch (round) { + case 0: + return hash % i0_blocks; + case 1: + return i0_blocks + hash % (i1_blocks - i0_blocks); + case 2: + return i1_blocks + hash % (i2_blocks - i1_blocks); + case 3: + return i2_blocks + hash % (i3_blocks - i2_blocks); + case 4 ... 19: + return i3_blocks + 16 * (hash % (((1<<31) - i3_blocks) / 16)) + + round - 4; + } + BUG(); +} + +static struct page *logfs_get_dd_page(struct inode *dir, struct dentry *dentry) +{ + struct qstr *name = &dentry->d_name; + struct page *page; + struct logfs_disk_dentry *dd; + u32 hash = hash_32(name->name, name->len, 0); + pgoff_t index; + int round; + + if (name->len > LOGFS_MAX_NAMELEN) + return ERR_PTR(-ENAMETOOLONG); + + for (round = 0; round < 20; round++) { + index = hash_index(hash, round); + + if (beyond_eof(dir, index)) + return NULL; + if (!logfs_exist_block(dir, index)) + continue; + page = read_cache_page(dir->i_mapping, index, + (filler_t *)logfs_readpage, NULL); + if (IS_ERR(page)) + return page; + dd = kmap_atomic(page, KM_USER0); + BUG_ON(dd->namelen == 0); + + if (name->len != be16_to_cpu(dd->namelen) || + memcmp(name->name, dd->name, name->len)) { + kunmap_atomic(dd, KM_USER0); + page_cache_release(page); + continue; + } + + kunmap_atomic(dd, KM_USER0); + return page; + } + return NULL; +} + +static int logfs_remove_inode(struct inode *inode) +{ + int ret; + + inode->i_nlink--; + ret = write_inode(inode); + LOGFS_BUG_ON(ret, inode->i_sb); + return ret; +} + +static void abort_transaction(struct inode *inode, struct logfs_transaction *ta) +{ + if (logfs_inode(inode)->li_block) + logfs_inode(inode)->li_block->ta = NULL; + kfree(ta); +} + +static int logfs_unlink(struct inode *dir, struct dentry *dentry) +{ + struct logfs_super *super = logfs_super(dir->i_sb); + struct inode *inode = dentry->d_inode; + struct logfs_transaction *ta; + struct page *page; + pgoff_t index; + int ret; + + ta = kzalloc(sizeof(*ta), GFP_KERNEL); + if (!ta) + return -ENOMEM; + + ta->state = UNLINK_1; + ta->ino = inode->i_ino; + + inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME; + + page = logfs_get_dd_page(dir, dentry); + if (!page) { + kfree(ta); + return -ENOENT; + } + if (IS_ERR(page)) { + kfree(ta); + return PTR_ERR(page); + } + index = page->index; + page_cache_release(page); + + mutex_lock(&super->s_dirop_mutex); + logfs_add_transaction(dir, ta); + + ret = logfs_delete(dir, index, NULL); + if (!ret) + ret = write_inode(dir); + + if (ret) { + abort_transaction(dir, ta); + printk(KERN_ERR"LOGFS: unable to delete inode\n"); + goto out; + } + + ta->state = UNLINK_2; + logfs_add_transaction(inode, ta); + ret = logfs_remove_inode(inode); +out: + mutex_unlock(&super->s_dirop_mutex); + return ret; +} + +static inline int logfs_empty_dir(struct inode *dir) +{ + u64 data; + + data = logfs_seek_data(dir, 0) << dir->i_sb->s_blocksize_bits; + return data >= i_size_read(dir); +} + +static int logfs_rmdir(struct inode *dir, struct dentry *dentry) +{ + struct inode *inode = dentry->d_inode; + + if (!logfs_empty_dir(inode)) + return -ENOTEMPTY; + + return logfs_unlink(dir, dentry); +} + +/* FIXME: readdir currently has it's own dir_walk code. I don't see a good + * way to combine the two copies */ +#define IMPLICIT_NODES 2 +static int __logfs_readdir(struct file *file, void *buf, filldir_t filldir) +{ + struct inode *dir = file->f_dentry->d_inode; + loff_t pos = file->f_pos - IMPLICIT_NODES; + struct page *page; + struct logfs_disk_dentry *dd; + int full; + + BUG_ON(pos < 0); + for (;; pos++) { + if (beyond_eof(dir, pos)) + break; + if (!logfs_exist_block(dir, pos)) { + /* deleted dentry */ + pos = dir_seek_data(dir, pos); + continue; + } + page = read_cache_page(dir->i_mapping, pos, + (filler_t *)logfs_readpage, NULL); + if (IS_ERR(page)) + return PTR_ERR(page); + dd = kmap_atomic(page, KM_USER0); + BUG_ON(dd->namelen == 0); + + full = filldir(buf, (char *)dd->name, be16_to_cpu(dd->namelen), + pos, be64_to_cpu(dd->ino), dd->type); + kunmap_atomic(dd, KM_USER0); + page_cache_release(page); + if (full) + break; + } + + file->f_pos = pos + IMPLICIT_NODES; + return 0; +} + +static int logfs_readdir(struct file *file, void *buf, filldir_t filldir) +{ + struct inode *inode = file->f_dentry->d_inode; + ino_t pino = parent_ino(file->f_dentry); + int err; + + if (file->f_pos < 0) + return -EINVAL; + + if (file->f_pos == 0) { + if (filldir(buf, ".", 1, 1, inode->i_ino, DT_DIR) < 0) + return 0; + file->f_pos++; + } + if (file->f_pos == 1) { + if (filldir(buf, "..", 2, 2, pino, DT_DIR) < 0) + return 0; + file->f_pos++; + } + + err = __logfs_readdir(file, buf, filldir); + return err; +} + +static void logfs_set_name(struct logfs_disk_dentry *dd, struct qstr *name) +{ + dd->namelen = cpu_to_be16(name->len); + memcpy(dd->name, name->name, name->len); +} + +static struct dentry *logfs_lookup(struct inode *dir, struct dentry *dentry, + struct nameidata *nd) +{ + struct page *page; + struct logfs_disk_dentry *dd; + pgoff_t index; + u64 ino = 0; + struct inode *inode; + + page = logfs_get_dd_page(dir, dentry); + if (IS_ERR(page)) + return ERR_CAST(page); + if (!page) { + d_add(dentry, NULL); + return NULL; + } + index = page->index; + dd = kmap_atomic(page, KM_USER0); + ino = be64_to_cpu(dd->ino); + kunmap_atomic(dd, KM_USER0); + page_cache_release(page); + + inode = logfs_iget(dir->i_sb, ino); + if (IS_ERR(inode)) { + printk(KERN_ERR"LogFS: Cannot read inode #%llx for dentry (%lx, %lx)n", + ino, dir->i_ino, index); + return ERR_CAST(inode); + } + return d_splice_alias(inode, dentry); +} + +static void grow_dir(struct inode *dir, loff_t index) +{ + index = (index + 1) << dir->i_sb->s_blocksize_bits; + if (i_size_read(dir) < index) + i_size_write(dir, index); +} + +static int logfs_write_dir(struct inode *dir, struct dentry *dentry, + struct inode *inode) +{ + struct page *page; + struct logfs_disk_dentry *dd; + u32 hash = hash_32(dentry->d_name.name, dentry->d_name.len, 0); + pgoff_t index; + int round, err; + + for (round = 0; round < 20; round++) { + index = hash_index(hash, round); + + if (logfs_exist_block(dir, index)) + continue; + page = find_or_create_page(dir->i_mapping, index, GFP_KERNEL); + if (!page) + return -ENOMEM; + + dd = kmap_atomic(page, KM_USER0); + memset(dd, 0, sizeof(*dd)); + dd->ino = cpu_to_be64(inode->i_ino); + dd->type = logfs_type(inode); + logfs_set_name(dd, &dentry->d_name); + kunmap_atomic(dd, KM_USER0); + + err = logfs_write_buf(dir, page, WF_LOCK); + unlock_page(page); + page_cache_release(page); + if (!err) + grow_dir(dir, index); + return err; + } + /* FIXME: Is there a better return value? In most cases neither + * the filesystem nor the directory are full. But we have had + * too many collisions for this particular hash and no fallback. + */ + return -ENOSPC; +} + +static int __logfs_create(struct inode *dir, struct dentry *dentry, + struct inode *inode, const char *dest, long destlen) +{ + struct logfs_super *super = logfs_super(dir->i_sb); + struct logfs_inode *li = logfs_inode(inode); + struct logfs_transaction *ta; + int ret; + + ta = kzalloc(sizeof(*ta), GFP_KERNEL); + if (!ta) + return -ENOMEM; + + ta->state = CREATE_1; + ta->ino = inode->i_ino; + mutex_lock(&super->s_dirop_mutex); + logfs_add_transaction(inode, ta); + + if (dest) { + /* symlink */ + ret = logfs_inode_write(inode, dest, destlen, 0, WF_LOCK, NULL); + if (!ret) + ret = write_inode(inode); + } else { + /* creat/mkdir/mknod */ + ret = write_inode(inode); + } + if (ret) { + abort_transaction(inode, ta); + li->li_flags |= LOGFS_IF_STILLBORN; + /* FIXME: truncate symlink */ + inode->i_nlink--; + iput(inode); + goto out; + } + + ta->state = CREATE_2; + logfs_add_transaction(dir, ta); + ret = logfs_write_dir(dir, dentry, inode); + /* sync directory */ + if (!ret) + ret = write_inode(dir); + + if (ret) { + logfs_del_transaction(dir, ta); + ta->state = CREATE_2; + logfs_add_transaction(inode, ta); + logfs_remove_inode(inode); + iput(inode); + goto out; + } + d_instantiate(dentry, inode); +out: + mutex_unlock(&super->s_dirop_mutex); + return ret; +} + +static int logfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +{ + struct inode *inode; + + /* + * FIXME: why do we have to fill in S_IFDIR, while the mode is + * correct for mknod, creat, etc.? Smells like the vfs *should* + * do it for us but for some reason fails to do so. + */ + inode = logfs_new_inode(dir, S_IFDIR | mode); + if (IS_ERR(inode)) + return PTR_ERR(inode); + + inode->i_op = &logfs_dir_iops; + inode->i_fop = &logfs_dir_fops; + + return __logfs_create(dir, dentry, inode, NULL, 0); +} + +static int logfs_create(struct inode *dir, struct dentry *dentry, int mode, + struct nameidata *nd) +{ + struct inode *inode; + + inode = logfs_new_inode(dir, mode); + if (IS_ERR(inode)) + return PTR_ERR(inode); + + inode->i_op = &logfs_reg_iops; + inode->i_fop = &logfs_reg_fops; + inode->i_mapping->a_ops = &logfs_reg_aops; + + return __logfs_create(dir, dentry, inode, NULL, 0); +} + +static int logfs_mknod(struct inode *dir, struct dentry *dentry, int mode, + dev_t rdev) +{ + struct inode *inode; + + if (dentry->d_name.len > LOGFS_MAX_NAMELEN) + return -ENAMETOOLONG; + + inode = logfs_new_inode(dir, mode); + if (IS_ERR(inode)) + return PTR_ERR(inode); + + init_special_inode(inode, mode, rdev); + + return __logfs_create(dir, dentry, inode, NULL, 0); +} + +static int logfs_symlink(struct inode *dir, struct dentry *dentry, + const char *target) +{ + struct inode *inode; + size_t destlen = strlen(target) + 1; + + if (destlen > dir->i_sb->s_blocksize) + return -ENAMETOOLONG; + + inode = logfs_new_inode(dir, S_IFLNK | 0777); + if (IS_ERR(inode)) + return PTR_ERR(inode); + + inode->i_op = &logfs_symlink_iops; + inode->i_mapping->a_ops = &logfs_reg_aops; + + return __logfs_create(dir, dentry, inode, target, destlen); +} + +static int logfs_permission(struct inode *inode, int mask) +{ + return generic_permission(inode, mask, NULL); +} + +static int logfs_link(struct dentry *old_dentry, struct inode *dir, + struct dentry *dentry) +{ + struct inode *inode = old_dentry->d_inode; + + if (inode->i_nlink >= LOGFS_LINK_MAX) + return -EMLINK; + + inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME; + atomic_inc(&inode->i_count); + inode->i_nlink++; + mark_inode_dirty_sync(inode); + + return __logfs_create(dir, dentry, inode, NULL, 0); +} + +static int logfs_get_dd(struct inode *dir, struct dentry *dentry, + struct logfs_disk_dentry *dd, loff_t *pos) +{ + struct page *page; + void *map; + + page = logfs_get_dd_page(dir, dentry); + if (IS_ERR(page)) + return PTR_ERR(page); + *pos = page->index; + map = kmap_atomic(page, KM_USER0); + memcpy(dd, map, sizeof(*dd)); + kunmap_atomic(map, KM_USER0); + page_cache_release(page); + return 0; +} + +static int logfs_delete_dd(struct inode *dir, loff_t pos) +{ + /* + * Getting called with pos somewhere beyond eof is either a goofup + * within this file or means someone maliciously edited the + * (crc-protected) journal. + */ + BUG_ON(beyond_eof(dir, pos)); + dir->i_ctime = dir->i_mtime = CURRENT_TIME; + log_dir(" Delete dentry (%lx, %llx)\n", dir->i_ino, pos); + return logfs_delete(dir, pos, NULL); +} + +/* + * Cross-directory rename, target does not exist. Just a little nasty. + * Create a new dentry in the target dir, then remove the old dentry, + * all the while taking care to remember our operation in the journal. + */ +static int logfs_rename_cross(struct inode *old_dir, struct dentry *old_dentry, + struct inode *new_dir, struct dentry *new_dentry) +{ + struct logfs_super *super = logfs_super(old_dir->i_sb); + struct logfs_disk_dentry dd; + struct logfs_transaction *ta; + loff_t pos; + int err; + + /* 1. locate source dd */ + err = logfs_get_dd(old_dir, old_dentry, &dd, &pos); + if (err) + return err; + + ta = kzalloc(sizeof(*ta), GFP_KERNEL); + if (!ta) + return -ENOMEM; + + ta->state = CROSS_RENAME_1; + ta->dir = old_dir->i_ino; + ta->pos = pos; + + /* 2. write target dd */ + mutex_lock(&super->s_dirop_mutex); + logfs_add_transaction(new_dir, ta); + err = logfs_write_dir(new_dir, new_dentry, old_dentry->d_inode); + if (!err) + err = write_inode(new_dir); + + if (err) { + super->s_rename_dir = 0; + super->s_rename_pos = 0; + abort_transaction(new_dir, ta); + goto out; + } + + /* 3. remove source dd */ + ta->state = CROSS_RENAME_2; + logfs_add_transaction(old_dir, ta); + err = logfs_delete_dd(old_dir, pos); + if (!err) + err = write_inode(old_dir); + LOGFS_BUG_ON(err, old_dir->i_sb); +out: + mutex_unlock(&super->s_dirop_mutex); + return err; +} + +static int logfs_replace_inode(struct inode *dir, struct dentry *dentry, + struct logfs_disk_dentry *dd, struct inode *inode) +{ + loff_t pos; + int err; + + err = logfs_get_dd(dir, dentry, dd, &pos); + if (err) + return err; + dd->ino = cpu_to_be64(inode->i_ino); + dd->type = logfs_type(inode); + + err = write_dir(dir, dd, pos); + if (err) + return err; + log_dir("Replace dentry (%lx, %llx) %s -> %llx\n", dir->i_ino, pos, + dd->name, be64_to_cpu(dd->ino)); + return write_inode(dir); +} + +/* Target dentry exists - the worst case. We need to attach the source + * inode to the target dentry, then remove the orphaned target inode and + * source dentry. + */ +static int logfs_rename_target(struct inode *old_dir, struct dentry *old_dentry, + struct inode *new_dir, struct dentry *new_dentry) +{ + struct logfs_super *super = logfs_super(old_dir->i_sb); + struct inode *old_inode = old_dentry->d_inode; + struct inode *new_inode = new_dentry->d_inode; + int isdir = S_ISDIR(old_inode->i_mode); + struct logfs_disk_dentry dd; + struct logfs_transaction *ta; + loff_t pos; + int err; + + BUG_ON(isdir != S_ISDIR(new_inode->i_mode)); + if (isdir) { + if (!logfs_empty_dir(new_inode)) + return -ENOTEMPTY; + } + + /* 1. locate source dd */ + err = logfs_get_dd(old_dir, old_dentry, &dd, &pos); + if (err) + return err; + + ta = kzalloc(sizeof(*ta), GFP_KERNEL); + if (!ta) + return -ENOMEM; + + ta->state = TARGET_RENAME_1; + ta->dir = old_dir->i_ino; + ta->pos = pos; + ta->ino = new_inode->i_ino; + + /* 2. attach source inode to target dd */ + mutex_lock(&super->s_dirop_mutex); + logfs_add_transaction(new_dir, ta); + err = logfs_replace_inode(new_dir, new_dentry, &dd, old_inode); + if (err) { + super->s_rename_dir = 0; + super->s_rename_pos = 0; + super->s_victim_ino = 0; + abort_transaction(new_dir, ta); + goto out; + } + + /* 3. remove source dd */ + ta->state = TARGET_RENAME_2; + logfs_add_transaction(old_dir, ta); + err = logfs_delete_dd(old_dir, pos); + if (!err) + err = write_inode(old_dir); + LOGFS_BUG_ON(err, old_dir->i_sb); + + /* 4. remove target inode */ + ta->state = TARGET_RENAME_3; + logfs_add_transaction(new_inode, ta); + err = logfs_remove_inode(new_inode); + +out: + mutex_unlock(&super->s_dirop_mutex); + return err; +} + +static int logfs_rename(struct inode *old_dir, struct dentry *old_dentry, + struct inode *new_dir, struct dentry *new_dentry) +{ + if (new_dentry->d_inode) + return logfs_rename_target(old_dir, old_dentry, + new_dir, new_dentry); + return logfs_rename_cross(old_dir, old_dentry, new_dir, new_dentry); +} + +/* No locking done here, as this is called before .get_sb() returns. */ +int logfs_replay_journal(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + struct inode *inode; + u64 ino, pos; + int err; + + if (super->s_victim_ino) { + /* delete victim inode */ + ino = super->s_victim_ino; + printk(KERN_INFO"LogFS: delete unmapped inode #%llx\n", ino); + inode = logfs_iget(sb, ino); + if (IS_ERR(inode)) + goto fail; + + LOGFS_BUG_ON(i_size_read(inode) > 0, sb); + super->s_victim_ino = 0; + err = logfs_remove_inode(inode); + iput(inode); + if (err) { + super->s_victim_ino = ino; + goto fail; + } + } + if (super->s_rename_dir) { + /* delete old dd from rename */ + ino = super->s_rename_dir; + pos = super->s_rename_pos; + printk(KERN_INFO"LogFS: delete unbacked dentry (%llx, %llx)\n", + ino, pos); + inode = logfs_iget(sb, ino); + if (IS_ERR(inode)) + goto fail; + + super->s_rename_dir = 0; + super->s_rename_pos = 0; + err = logfs_delete_dd(inode, pos); + iput(inode); + if (err) { + super->s_rename_dir = ino; + super->s_rename_pos = pos; + goto fail; + } + } + return 0; +fail: + LOGFS_BUG(sb); + return -EIO; +} + +const struct inode_operations logfs_symlink_iops = { + .readlink = generic_readlink, + .follow_link = page_follow_link_light, +}; + +const struct inode_operations logfs_dir_iops = { + .create = logfs_create, + .link = logfs_link, + .lookup = logfs_lookup, + .mkdir = logfs_mkdir, + .mknod = logfs_mknod, + .rename = logfs_rename, + .rmdir = logfs_rmdir, + .permission = logfs_permission, + .symlink = logfs_symlink, + .unlink = logfs_unlink, +}; +const struct file_operations logfs_dir_fops = { + .fsync = logfs_fsync, + .ioctl = logfs_ioctl, + .readdir = logfs_readdir, + .read = generic_read_dir, +}; diff --git a/fs/logfs/file.c b/fs/logfs/file.c new file mode 100644 index 000000000000..370f367a933e --- /dev/null +++ b/fs/logfs/file.c @@ -0,0 +1,263 @@ +/* + * fs/logfs/file.c - prepare_write, commit_write and friends + * + * As should be obvious for Linux kernel code, license is GPLv2 + * + * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> + */ +#include "logfs.h" +#include <linux/sched.h> +#include <linux/writeback.h> + +static int logfs_write_begin(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned flags, + struct page **pagep, void **fsdata) +{ + struct inode *inode = mapping->host; + struct page *page; + pgoff_t index = pos >> PAGE_CACHE_SHIFT; + + page = grab_cache_page_write_begin(mapping, index, flags); + if (!page) + return -ENOMEM; + *pagep = page; + + if ((len == PAGE_CACHE_SIZE) || PageUptodate(page)) + return 0; + if ((pos & PAGE_CACHE_MASK) >= i_size_read(inode)) { + unsigned start = pos & (PAGE_CACHE_SIZE - 1); + unsigned end = start + len; + + /* Reading beyond i_size is simple: memset to zero */ + zero_user_segments(page, 0, start, end, PAGE_CACHE_SIZE); + return 0; + } + return logfs_readpage_nolock(page); +} + +static int logfs_write_end(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, struct page *page, + void *fsdata) +{ + struct inode *inode = mapping->host; + pgoff_t index = page->index; + unsigned start = pos & (PAGE_CACHE_SIZE - 1); + unsigned end = start + copied; + int ret = 0; + + BUG_ON(PAGE_CACHE_SIZE != inode->i_sb->s_blocksize); + BUG_ON(page->index > I3_BLOCKS); + + if (copied < len) { + /* + * Short write of a non-initialized paged. Just tell userspace + * to retry the entire page. + */ + if (!PageUptodate(page)) { + copied = 0; + goto out; + } + } + if (copied == 0) + goto out; /* FIXME: do we need to update inode? */ + + if (i_size_read(inode) < (index << PAGE_CACHE_SHIFT) + end) { + i_size_write(inode, (index << PAGE_CACHE_SHIFT) + end); + mark_inode_dirty_sync(inode); + } + + SetPageUptodate(page); + if (!PageDirty(page)) { + if (!get_page_reserve(inode, page)) + __set_page_dirty_nobuffers(page); + else + ret = logfs_write_buf(inode, page, WF_LOCK); + } +out: + unlock_page(page); + page_cache_release(page); + return ret ? ret : copied; +} + +int logfs_readpage(struct file *file, struct page *page) +{ + int ret; + + ret = logfs_readpage_nolock(page); + unlock_page(page); + return ret; +} + +/* Clear the page's dirty flag in the radix tree. */ +/* TODO: mucking with PageWriteback is silly. Add a generic function to clear + * the dirty bit from the radix tree for filesystems that don't have to wait + * for page writeback to finish (i.e. any compressing filesystem). + */ +static void clear_radix_tree_dirty(struct page *page) +{ + BUG_ON(PagePrivate(page) || page->private); + set_page_writeback(page); + end_page_writeback(page); +} + +static int __logfs_writepage(struct page *page) +{ + struct inode *inode = page->mapping->host; + int err; + + err = logfs_write_buf(inode, page, WF_LOCK); + if (err) + set_page_dirty(page); + else + clear_radix_tree_dirty(page); + unlock_page(page); + return err; +} + +static int logfs_writepage(struct page *page, struct writeback_control *wbc) +{ + struct inode *inode = page->mapping->host; + loff_t i_size = i_size_read(inode); + pgoff_t end_index = i_size >> PAGE_CACHE_SHIFT; + unsigned offset; + u64 bix; + level_t level; + + log_file("logfs_writepage(%lx, %lx, %p)\n", inode->i_ino, page->index, + page); + + logfs_unpack_index(page->index, &bix, &level); + + /* Indirect blocks are never truncated */ + if (level != 0) + return __logfs_writepage(page); + + /* + * TODO: everything below is a near-verbatim copy of nobh_writepage(). + * The relevant bits should be factored out after logfs is merged. + */ + + /* Is the page fully inside i_size? */ + if (bix < end_index) + return __logfs_writepage(page); + + /* Is the page fully outside i_size? (truncate in progress) */ + offset = i_size & (PAGE_CACHE_SIZE-1); + if (bix > end_index || offset == 0) { + unlock_page(page); + return 0; /* don't care */ + } + + /* + * The page straddles i_size. It must be zeroed out on each and every + * writepage invokation because it may be mmapped. "A file is mapped + * in multiples of the page size. For a file that is not a multiple of + * the page size, the remaining memory is zeroed when mapped, and + * writes to that region are not written out to the file." + */ + zero_user_segment(page, offset, PAGE_CACHE_SIZE); + return __logfs_writepage(page); +} + +static void logfs_invalidatepage(struct page *page, unsigned long offset) +{ + move_page_to_btree(page); + BUG_ON(PagePrivate(page) || page->private); +} + +static int logfs_releasepage(struct page *page, gfp_t only_xfs_uses_this) +{ + return 0; /* None of these are easy to release */ +} + + +int logfs_ioctl(struct inode *inode, struct file *file, unsigned int cmd, + unsigned long arg) +{ + struct logfs_inode *li = logfs_inode(inode); + unsigned int oldflags, flags; + int err; + + switch (cmd) { + case FS_IOC_GETFLAGS: + flags = li->li_flags & LOGFS_FL_USER_VISIBLE; + return put_user(flags, (int __user *)arg); + case FS_IOC_SETFLAGS: + if (IS_RDONLY(inode)) + return -EROFS; + + if (!is_owner_or_cap(inode)) + return -EACCES; + + err = get_user(flags, (int __user *)arg); + if (err) + return err; + + mutex_lock(&inode->i_mutex); + oldflags = li->li_flags; + flags &= LOGFS_FL_USER_MODIFIABLE; + flags |= oldflags & ~LOGFS_FL_USER_MODIFIABLE; + li->li_flags = flags; + mutex_unlock(&inode->i_mutex); + + inode->i_ctime = CURRENT_TIME; + mark_inode_dirty_sync(inode); + return 0; + + default: + return -ENOTTY; + } +} + +int logfs_fsync(struct file *file, struct dentry *dentry, int datasync) +{ + struct super_block *sb = dentry->d_inode->i_sb; + struct logfs_super *super = logfs_super(sb); + + /* FIXME: write anchor */ + super->s_devops->sync(sb); + return 0; +} + +static int logfs_setattr(struct dentry *dentry, struct iattr *attr) +{ + struct inode *inode = dentry->d_inode; + int err = 0; + + if (attr->ia_valid & ATTR_SIZE) + err = logfs_truncate(inode, attr->ia_size); + attr->ia_valid &= ~ATTR_SIZE; + + if (!err) + err = inode_change_ok(inode, attr); + if (!err) + err = inode_setattr(inode, attr); + return err; +} + +const struct inode_operations logfs_reg_iops = { + .setattr = logfs_setattr, +}; + +const struct file_operations logfs_reg_fops = { + .aio_read = generic_file_aio_read, + .aio_write = generic_file_aio_write, + .fsync = logfs_fsync, + .ioctl = logfs_ioctl, + .llseek = generic_file_llseek, + .mmap = generic_file_readonly_mmap, + .open = generic_file_open, + .read = do_sync_read, + .write = do_sync_write, +}; + +const struct address_space_operations logfs_reg_aops = { + .invalidatepage = logfs_invalidatepage, + .readpage = logfs_readpage, + .releasepage = logfs_releasepage, + .set_page_dirty = __set_page_dirty_nobuffers, + .writepage = logfs_writepage, + .writepages = generic_writepages, + .write_begin = logfs_write_begin, + .write_end = logfs_write_end, +}; diff --git a/fs/logfs/gc.c b/fs/logfs/gc.c new file mode 100644 index 000000000000..92949f95a901 --- /dev/null +++ b/fs/logfs/gc.c @@ -0,0 +1,730 @@ +/* + * fs/logfs/gc.c - garbage collection code + * + * As should be obvious for Linux kernel code, license is GPLv2 + * + * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> + */ +#include "logfs.h" +#include <linux/sched.h> + +/* + * Wear leveling needs to kick in when the difference between low erase + * counts and high erase counts gets too big. A good value for "too big" + * may be somewhat below 10% of maximum erase count for the device. + * Why not 397, to pick a nice round number with no specific meaning? :) + * + * WL_RATELIMIT is the minimum time between two wear level events. A huge + * number of segments may fulfil the requirements for wear leveling at the + * same time. If that happens we don't want to cause a latency from hell, + * but just gently pick one segment every so often and minimize overhead. + */ +#define WL_DELTA 397 +#define WL_RATELIMIT 100 +#define MAX_OBJ_ALIASES 2600 +#define SCAN_RATIO 512 /* number of scanned segments per gc'd segment */ +#define LIST_SIZE 64 /* base size of candidate lists */ +#define SCAN_ROUNDS 128 /* maximum number of complete medium scans */ +#define SCAN_ROUNDS_HIGH 4 /* maximum number of higher-level scans */ + +static int no_free_segments(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + + return super->s_free_list.count; +} + +/* journal has distance -1, top-most ifile layer distance 0 */ +static u8 root_distance(struct super_block *sb, gc_level_t __gc_level) +{ + struct logfs_super *super = logfs_super(sb); + u8 gc_level = (__force u8)__gc_level; + + switch (gc_level) { + case 0: /* fall through */ + case 1: /* fall through */ + case 2: /* fall through */ + case 3: + /* file data or indirect blocks */ + return super->s_ifile_levels + super->s_iblock_levels - gc_level; + case 6: /* fall through */ + case 7: /* fall through */ + case 8: /* fall through */ + case 9: + /* inode file data or indirect blocks */ + return super->s_ifile_levels - (gc_level - 6); + default: + printk(KERN_ERR"LOGFS: segment of unknown level %x found\n", + gc_level); + WARN_ON(1); + return super->s_ifile_levels + super->s_iblock_levels; + } +} + +static int segment_is_reserved(struct super_block *sb, u32 segno) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_area *area; + void *reserved; + int i; + + /* Some segments are reserved. Just pretend they were all valid */ + reserved = btree_lookup32(&super->s_reserved_segments, segno); + if (reserved) + return 1; + + /* Currently open segments */ + for_each_area(i) { + area = super->s_area[i]; + if (area->a_is_open && area->a_segno == segno) + return 1; + } + + return 0; +} + +static void logfs_mark_segment_bad(struct super_block *sb, u32 segno) +{ + BUG(); +} + +/* + * Returns the bytes consumed by valid objects in this segment. Object headers + * are counted, the segment header is not. + */ +static u32 logfs_valid_bytes(struct super_block *sb, u32 segno, u32 *ec, + gc_level_t *gc_level) +{ + struct logfs_segment_entry se; + u32 ec_level; + + logfs_get_segment_entry(sb, segno, &se); + if (se.ec_level == cpu_to_be32(BADSEG) || + se.valid == cpu_to_be32(RESERVED)) + return RESERVED; + + ec_level = be32_to_cpu(se.ec_level); + *ec = ec_level >> 4; + *gc_level = GC_LEVEL(ec_level & 0xf); + return be32_to_cpu(se.valid); +} + +static void logfs_cleanse_block(struct super_block *sb, u64 ofs, u64 ino, + u64 bix, gc_level_t gc_level) +{ + struct inode *inode; + int err, cookie; + + inode = logfs_safe_iget(sb, ino, &cookie); + err = logfs_rewrite_block(inode, bix, ofs, gc_level, 0); + BUG_ON(err); + logfs_safe_iput(inode, cookie); +} + +static u32 logfs_gc_segment(struct super_block *sb, u32 segno, u8 dist) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_segment_header sh; + struct logfs_object_header oh; + u64 ofs, ino, bix; + u32 seg_ofs, logical_segno, cleaned = 0; + int err, len, valid; + gc_level_t gc_level; + + LOGFS_BUG_ON(segment_is_reserved(sb, segno), sb); + + btree_insert32(&super->s_reserved_segments, segno, (void *)1, GFP_NOFS); + err = wbuf_read(sb, dev_ofs(sb, segno, 0), sizeof(sh), &sh); + BUG_ON(err); + gc_level = GC_LEVEL(sh.level); + logical_segno = be32_to_cpu(sh.segno); + if (sh.crc != logfs_crc32(&sh, sizeof(sh), 4)) { + logfs_mark_segment_bad(sb, segno); + cleaned = -1; + goto out; + } + + for (seg_ofs = LOGFS_SEGMENT_HEADERSIZE; + seg_ofs + sizeof(oh) < super->s_segsize; ) { + ofs = dev_ofs(sb, logical_segno, seg_ofs); + err = wbuf_read(sb, dev_ofs(sb, segno, seg_ofs), sizeof(oh), + &oh); + BUG_ON(err); + + if (!memchr_inv(&oh, 0xff, sizeof(oh))) + break; + + if (oh.crc != logfs_crc32(&oh, sizeof(oh) - 4, 4)) { + logfs_mark_segment_bad(sb, segno); + cleaned = super->s_segsize - 1; + goto out; + } + + ino = be64_to_cpu(oh.ino); + bix = be64_to_cpu(oh.bix); + len = sizeof(oh) + be16_to_cpu(oh.len); + valid = logfs_is_valid_block(sb, ofs, ino, bix, gc_level); + if (valid == 1) { + logfs_cleanse_block(sb, ofs, ino, bix, gc_level); + cleaned += len; + } else if (valid == 2) { + /* Will be invalid upon journal commit */ + cleaned += len; + } + seg_ofs += len; + } +out: + btree_remove32(&super->s_reserved_segments, segno); + return cleaned; +} + +static struct gc_candidate *add_list(struct gc_candidate *cand, + struct candidate_list *list) +{ + struct rb_node **p = &list->rb_tree.rb_node; + struct rb_node *parent = NULL; + struct gc_candidate *cur; + int comp; + + cand->list = list; + while (*p) { + parent = *p; + cur = rb_entry(parent, struct gc_candidate, rb_node); + + if (list->sort_by_ec) + comp = cand->erase_count < cur->erase_count; + else + comp = cand->valid < cur->valid; + + if (comp) + p = &parent->rb_left; + else + p = &parent->rb_right; + } + rb_link_node(&cand->rb_node, parent, p); + rb_insert_color(&cand->rb_node, &list->rb_tree); + + if (list->count <= list->maxcount) { + list->count++; + return NULL; + } + cand = rb_entry(rb_last(&list->rb_tree), struct gc_candidate, rb_node); + rb_erase(&cand->rb_node, &list->rb_tree); + cand->list = NULL; + return cand; +} + +static void remove_from_list(struct gc_candidate *cand) +{ + struct candidate_list *list = cand->list; + + rb_erase(&cand->rb_node, &list->rb_tree); + list->count--; +} + +static void free_candidate(struct super_block *sb, struct gc_candidate *cand) +{ + struct logfs_super *super = logfs_super(sb); + + btree_remove32(&super->s_cand_tree, cand->segno); + kfree(cand); +} + +u32 get_best_cand(struct super_block *sb, struct candidate_list *list, u32 *ec) +{ + struct gc_candidate *cand; + u32 segno; + + BUG_ON(list->count == 0); + + cand = rb_entry(rb_first(&list->rb_tree), struct gc_candidate, rb_node); + remove_from_list(cand); + segno = cand->segno; + if (ec) + *ec = cand->erase_count; + free_candidate(sb, cand); + return segno; +} + +/* + * We have several lists to manage segments with. The reserve_list is used to + * deal with bad blocks. We try to keep the best (lowest ec) segments on this + * list. + * The free_list contains free segments for normal usage. It usually gets the + * second pick after the reserve_list. But when the free_list is running short + * it is more important to keep the free_list full than to keep a reserve. + * + * Segments that are not free are put onto a per-level low_list. If we have + * to run garbage collection, we pick a candidate from there. All segments on + * those lists should have at least some free space so GC will make progress. + * + * And last we have the ec_list, which is used to pick segments for wear + * leveling. + * + * If all appropriate lists are full, we simply free the candidate and forget + * about that segment for a while. We have better candidates for each purpose. + */ +static void __add_candidate(struct super_block *sb, struct gc_candidate *cand) +{ + struct logfs_super *super = logfs_super(sb); + u32 full = super->s_segsize - LOGFS_SEGMENT_RESERVE; + + if (cand->valid == 0) { + /* 100% free segments */ + log_gc_noisy("add reserve segment %x (ec %x) at %llx\n", + cand->segno, cand->erase_count, + dev_ofs(sb, cand->segno, 0)); + cand = add_list(cand, &super->s_reserve_list); + if (cand) { + log_gc_noisy("add free segment %x (ec %x) at %llx\n", + cand->segno, cand->erase_count, + dev_ofs(sb, cand->segno, 0)); + cand = add_list(cand, &super->s_free_list); + } + } else { + /* good candidates for Garbage Collection */ + if (cand->valid < full) + cand = add_list(cand, &super->s_low_list[cand->dist]); + /* good candidates for wear leveling, + * segments that were recently written get ignored */ + if (cand) + cand = add_list(cand, &super->s_ec_list); + } + if (cand) + free_candidate(sb, cand); +} + +static int add_candidate(struct super_block *sb, u32 segno, u32 valid, u32 ec, + u8 dist) +{ + struct logfs_super *super = logfs_super(sb); + struct gc_candidate *cand; + + cand = kmalloc(sizeof(*cand), GFP_NOFS); + if (!cand) + return -ENOMEM; + + cand->segno = segno; + cand->valid = valid; + cand->erase_count = ec; + cand->dist = dist; + + btree_insert32(&super->s_cand_tree, segno, cand, GFP_NOFS); + __add_candidate(sb, cand); + return 0; +} + +static void remove_segment_from_lists(struct super_block *sb, u32 segno) +{ + struct logfs_super *super = logfs_super(sb); + struct gc_candidate *cand; + + cand = btree_lookup32(&super->s_cand_tree, segno); + if (cand) { + remove_from_list(cand); + free_candidate(sb, cand); + } +} + +static void scan_segment(struct super_block *sb, u32 segno) +{ + u32 valid, ec = 0; + gc_level_t gc_level = 0; + u8 dist; + + if (segment_is_reserved(sb, segno)) + return; + + remove_segment_from_lists(sb, segno); + valid = logfs_valid_bytes(sb, segno, &ec, &gc_level); + if (valid == RESERVED) + return; + + dist = root_distance(sb, gc_level); + add_candidate(sb, segno, valid, ec, dist); +} + +static struct gc_candidate *first_in_list(struct candidate_list *list) +{ + if (list->count == 0) + return NULL; + return rb_entry(rb_first(&list->rb_tree), struct gc_candidate, rb_node); +} + +/* + * Find the best segment for garbage collection. Main criterion is + * the segment requiring the least effort to clean. Secondary + * criterion is to GC on the lowest level available. + * + * So we search the least effort segment on the lowest level first, + * then move up and pick another segment iff is requires significantly + * less effort. Hence the LOGFS_MAX_OBJECTSIZE in the comparison. + */ +static struct gc_candidate *get_candidate(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + int i, max_dist; + struct gc_candidate *cand = NULL, *this; + + max_dist = min(no_free_segments(sb), LOGFS_NO_AREAS); + + for (i = max_dist; i >= 0; i--) { + this = first_in_list(&super->s_low_list[i]); + if (!this) + continue; + if (!cand) + cand = this; + if (this->valid + LOGFS_MAX_OBJECTSIZE <= cand->valid) + cand = this; + } + return cand; +} + +static int __logfs_gc_once(struct super_block *sb, struct gc_candidate *cand) +{ + struct logfs_super *super = logfs_super(sb); + gc_level_t gc_level; + u32 cleaned, valid, segno, ec; + u8 dist; + + if (!cand) { + log_gc("GC attempted, but no candidate found\n"); + return 0; + } + + segno = cand->segno; + dist = cand->dist; + valid = logfs_valid_bytes(sb, segno, &ec, &gc_level); + free_candidate(sb, cand); + log_gc("GC segment #%02x at %llx, %x required, %x free, %x valid, %llx free\n", + segno, (u64)segno << super->s_segshift, + dist, no_free_segments(sb), valid, + super->s_free_bytes); + cleaned = logfs_gc_segment(sb, segno, dist); + log_gc("GC segment #%02x complete - now %x valid\n", segno, + valid - cleaned); + BUG_ON(cleaned != valid); + return 1; +} + +static int logfs_gc_once(struct super_block *sb) +{ + struct gc_candidate *cand; + + cand = get_candidate(sb); + if (cand) + remove_from_list(cand); + return __logfs_gc_once(sb, cand); +} + +/* returns 1 if a wrap occurs, 0 otherwise */ +static int logfs_scan_some(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + u32 segno; + int i, ret = 0; + + segno = super->s_sweeper; + for (i = SCAN_RATIO; i > 0; i--) { + segno++; + if (segno >= super->s_no_segs) { + segno = 0; + ret = 1; + /* Break out of the loop. We want to read a single + * block from the segment size on next invocation if + * SCAN_RATIO is set to match block size + */ + break; + } + + scan_segment(sb, segno); + } + super->s_sweeper = segno; + return ret; +} + +/* + * In principle, this function should loop forever, looking for GC candidates + * and moving data. LogFS is designed in such a way that this loop is + * guaranteed to terminate. + * + * Limiting the loop to some iterations serves purely to catch cases when + * these guarantees have failed. An actual endless loop is an obvious bug + * and should be reported as such. + */ +static void __logfs_gc_pass(struct super_block *sb, int target) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_block *block; + int round, progress, last_progress = 0; + + if (no_free_segments(sb) >= target && + super->s_no_object_aliases < MAX_OBJ_ALIASES) + return; + + log_gc("__logfs_gc_pass(%x)\n", target); + for (round = 0; round < SCAN_ROUNDS; ) { + if (no_free_segments(sb) >= target) + goto write_alias; + + /* Sync in-memory state with on-medium state in case they + * diverged */ + logfs_write_anchor(sb); + round += logfs_scan_some(sb); + if (no_free_segments(sb) >= target) + goto write_alias; + progress = logfs_gc_once(sb); + if (progress) + last_progress = round; + else if (round - last_progress > 2) + break; + continue; + + /* + * The goto logic is nasty, I just don't know a better way to + * code it. GC is supposed to ensure two things: + * 1. Enough free segments are available. + * 2. The number of aliases is bounded. + * When 1. is achieved, we take a look at 2. and write back + * some alias-containing blocks, if necessary. However, after + * each such write we need to go back to 1., as writes can + * consume free segments. + */ +write_alias: + if (super->s_no_object_aliases < MAX_OBJ_ALIASES) + return; + if (list_empty(&super->s_object_alias)) { + /* All aliases are still in btree */ + return; + } + log_gc("Write back one alias\n"); + block = list_entry(super->s_object_alias.next, + struct logfs_block, alias_list); + block->ops->write_block(block); + /* + * To round off the nasty goto logic, we reset round here. It + * is a safety-net for GC not making any progress and limited + * to something reasonably small. If incremented it for every + * single alias, the loop could terminate rather quickly. + */ + round = 0; + } + LOGFS_BUG(sb); +} + +static int wl_ratelimit(struct super_block *sb, u64 *next_event) +{ + struct logfs_super *super = logfs_super(sb); + + if (*next_event < super->s_gec) { + *next_event = super->s_gec + WL_RATELIMIT; + return 0; + } + return 1; +} + +static void logfs_wl_pass(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + struct gc_candidate *wl_cand, *free_cand; + + if (wl_ratelimit(sb, &super->s_wl_gec_ostore)) + return; + + wl_cand = first_in_list(&super->s_ec_list); + if (!wl_cand) + return; + free_cand = first_in_list(&super->s_free_list); + if (!free_cand) + return; + + if (wl_cand->erase_count < free_cand->erase_count + WL_DELTA) { + remove_from_list(wl_cand); + __logfs_gc_once(sb, wl_cand); + } +} + +/* + * The journal needs wear leveling as well. But moving the journal is an + * expensive operation so we try to avoid it as much as possible. And if we + * have to do it, we move the whole journal, not individual segments. + * + * Ratelimiting is not strictly necessary here, it mainly serves to avoid the + * calculations. First we check whether moving the journal would be a + * significant improvement. That means that a) the current journal segments + * have more wear than the future journal segments and b) the current journal + * segments have more wear than normal ostore segments. + * Rationale for b) is that we don't have to move the journal if it is aging + * less than the ostore, even if the reserve segments age even less (they are + * excluded from wear leveling, after all). + * Next we check that the superblocks have less wear than the journal. Since + * moving the journal requires writing the superblocks, we have to protect the + * superblocks even more than the journal. + * + * Also we double the acceptable wear difference, compared to ostore wear + * leveling. Journal data is read and rewritten rapidly, comparatively. So + * soft errors have much less time to accumulate and we allow the journal to + * be a bit worse than the ostore. + */ +static void logfs_journal_wl_pass(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + struct gc_candidate *cand; + u32 min_journal_ec = -1, max_reserve_ec = 0; + int i; + + if (wl_ratelimit(sb, &super->s_wl_gec_journal)) + return; + + if (super->s_reserve_list.count < super->s_no_journal_segs) { + /* Reserve is not full enough to move complete journal */ + return; + } + + journal_for_each(i) + if (super->s_journal_seg[i]) + min_journal_ec = min(min_journal_ec, + super->s_journal_ec[i]); + cand = rb_entry(rb_first(&super->s_free_list.rb_tree), + struct gc_candidate, rb_node); + max_reserve_ec = cand->erase_count; + for (i = 0; i < 2; i++) { + struct logfs_segment_entry se; + u32 segno = seg_no(sb, super->s_sb_ofs[i]); + u32 ec; + + logfs_get_segment_entry(sb, segno, &se); + ec = be32_to_cpu(se.ec_level) >> 4; + max_reserve_ec = max(max_reserve_ec, ec); + } + + if (min_journal_ec > max_reserve_ec + 2 * WL_DELTA) { + do_logfs_journal_wl_pass(sb); + } +} + +void logfs_gc_pass(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + + //BUG_ON(mutex_trylock(&logfs_super(sb)->s_w_mutex)); + /* Write journal before free space is getting saturated with dirty + * objects. + */ + if (super->s_dirty_used_bytes + super->s_dirty_free_bytes + + LOGFS_MAX_OBJECTSIZE >= super->s_free_bytes) + logfs_write_anchor(sb); + __logfs_gc_pass(sb, super->s_total_levels); + logfs_wl_pass(sb); + logfs_journal_wl_pass(sb); +} + +static int check_area(struct super_block *sb, int i) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_area *area = super->s_area[i]; + struct logfs_object_header oh; + u32 segno = area->a_segno; + u32 ofs = area->a_used_bytes; + __be32 crc; + int err; + + if (!area->a_is_open) + return 0; + + for (ofs = area->a_used_bytes; + ofs <= super->s_segsize - sizeof(oh); + ofs += (u32)be16_to_cpu(oh.len) + sizeof(oh)) { + err = wbuf_read(sb, dev_ofs(sb, segno, ofs), sizeof(oh), &oh); + if (err) + return err; + + if (!memchr_inv(&oh, 0xff, sizeof(oh))) + break; + + crc = logfs_crc32(&oh, sizeof(oh) - 4, 4); + if (crc != oh.crc) { + printk(KERN_INFO "interrupted header at %llx\n", + dev_ofs(sb, segno, ofs)); + return 0; + } + } + if (ofs != area->a_used_bytes) { + printk(KERN_INFO "%x bytes unaccounted data found at %llx\n", + ofs - area->a_used_bytes, + dev_ofs(sb, segno, area->a_used_bytes)); + area->a_used_bytes = ofs; + } + return 0; +} + +int logfs_check_areas(struct super_block *sb) +{ + int i, err; + + for_each_area(i) { + err = check_area(sb, i); + if (err) + return err; + } + return 0; +} + +static void logfs_init_candlist(struct candidate_list *list, int maxcount, + int sort_by_ec) +{ + list->count = 0; + list->maxcount = maxcount; + list->sort_by_ec = sort_by_ec; + list->rb_tree = RB_ROOT; +} + +int logfs_init_gc(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + int i; + + btree_init_mempool32(&super->s_cand_tree, super->s_btree_pool); + logfs_init_candlist(&super->s_free_list, LIST_SIZE + SCAN_RATIO, 1); + logfs_init_candlist(&super->s_reserve_list, + super->s_bad_seg_reserve, 1); + for_each_area(i) + logfs_init_candlist(&super->s_low_list[i], LIST_SIZE, 0); + logfs_init_candlist(&super->s_ec_list, LIST_SIZE, 1); + return 0; +} + +static void logfs_cleanup_list(struct super_block *sb, + struct candidate_list *list) +{ + struct gc_candidate *cand; + + while (list->count) { + cand = rb_entry(list->rb_tree.rb_node, struct gc_candidate, + rb_node); + remove_from_list(cand); + free_candidate(sb, cand); + } + BUG_ON(list->rb_tree.rb_node); +} + +void logfs_cleanup_gc(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + int i; + + if (!super->s_free_list.count) + return; + + /* + * FIXME: The btree may still contain a single empty node. So we + * call the grim visitor to clean up that mess. Btree code should + * do it for us, really. + */ + btree_grim_visitor32(&super->s_cand_tree, 0, NULL); + logfs_cleanup_list(sb, &super->s_free_list); + logfs_cleanup_list(sb, &super->s_reserve_list); + for_each_area(i) + logfs_cleanup_list(sb, &super->s_low_list[i]); + logfs_cleanup_list(sb, &super->s_ec_list); +} diff --git a/fs/logfs/inode.c b/fs/logfs/inode.c new file mode 100644 index 000000000000..33ec1aeaeec4 --- /dev/null +++ b/fs/logfs/inode.c @@ -0,0 +1,417 @@ +/* + * fs/logfs/inode.c - inode handling code + * + * As should be obvious for Linux kernel code, license is GPLv2 + * + * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> + */ +#include "logfs.h" +#include <linux/writeback.h> +#include <linux/backing-dev.h> + +/* + * How soon to reuse old inode numbers? LogFS doesn't store deleted inodes + * on the medium. It therefore also lacks a method to store the previous + * generation number for deleted inodes. Instead a single generation number + * is stored which will be used for new inodes. Being just a 32bit counter, + * this can obvious wrap relatively quickly. So we only reuse inodes if we + * know that a fair number of inodes can be created before we have to increment + * the generation again - effectively adding some bits to the counter. + * But being too aggressive here means we keep a very large and very sparse + * inode file, wasting space on indirect blocks. + * So what is a good value? Beats me. 64k seems moderately bad on both + * fronts, so let's use that for now... + * + * NFS sucks, as everyone already knows. + */ +#define INOS_PER_WRAP (0x10000) + +/* + * Logfs' requirement to read inodes for garbage collection makes life a bit + * harder. GC may have to read inodes that are in I_FREEING state, when they + * are being written out - and waiting for GC to make progress, naturally. + * + * So we cannot just call iget() or some variant of it, but first have to check + * wether the inode in question might be in I_FREEING state. Therefore we + * maintain our own per-sb list of "almost deleted" inodes and check against + * that list first. Normally this should be at most 1-2 entries long. + * + * Also, inodes have logfs-specific reference counting on top of what the vfs + * does. When .destroy_inode is called, normally the reference count will drop + * to zero and the inode gets deleted. But if GC accessed the inode, its + * refcount will remain nonzero and final deletion will have to wait. + * + * As a result we have two sets of functions to get/put inodes: + * logfs_safe_iget/logfs_safe_iput - safe to call from GC context + * logfs_iget/iput - normal version + */ +static struct kmem_cache *logfs_inode_cache; + +static DEFINE_SPINLOCK(logfs_inode_lock); + +static void logfs_inode_setops(struct inode *inode) +{ + switch (inode->i_mode & S_IFMT) { + case S_IFDIR: + inode->i_op = &logfs_dir_iops; + inode->i_fop = &logfs_dir_fops; + inode->i_mapping->a_ops = &logfs_reg_aops; + break; + case S_IFREG: + inode->i_op = &logfs_reg_iops; + inode->i_fop = &logfs_reg_fops; + inode->i_mapping->a_ops = &logfs_reg_aops; + break; + case S_IFLNK: + inode->i_op = &logfs_symlink_iops; + inode->i_mapping->a_ops = &logfs_reg_aops; + break; + case S_IFSOCK: /* fall through */ + case S_IFBLK: /* fall through */ + case S_IFCHR: /* fall through */ + case S_IFIFO: + init_special_inode(inode, inode->i_mode, inode->i_rdev); + break; + default: + BUG(); + } +} + +static struct inode *__logfs_iget(struct super_block *sb, ino_t ino) +{ + struct inode *inode = iget_locked(sb, ino); + int err; + + if (!inode) + return ERR_PTR(-ENOMEM); + if (!(inode->i_state & I_NEW)) + return inode; + + err = logfs_read_inode(inode); + if (err || inode->i_nlink == 0) { + /* inode->i_nlink == 0 can be true when called from + * block validator */ + /* set i_nlink to 0 to prevent caching */ + inode->i_nlink = 0; + logfs_inode(inode)->li_flags |= LOGFS_IF_ZOMBIE; + iget_failed(inode); + if (!err) + err = -ENOENT; + return ERR_PTR(err); + } + + logfs_inode_setops(inode); + unlock_new_inode(inode); + return inode; +} + +struct inode *logfs_iget(struct super_block *sb, ino_t ino) +{ + BUG_ON(ino == LOGFS_INO_MASTER); + BUG_ON(ino == LOGFS_INO_SEGFILE); + return __logfs_iget(sb, ino); +} + +/* + * is_cached is set to 1 if we hand out a cached inode, 0 otherwise. + * this allows logfs_iput to do the right thing later + */ +struct inode *logfs_safe_iget(struct super_block *sb, ino_t ino, int *is_cached) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_inode *li; + + if (ino == LOGFS_INO_MASTER) + return super->s_master_inode; + if (ino == LOGFS_INO_SEGFILE) + return super->s_segfile_inode; + + spin_lock(&logfs_inode_lock); + list_for_each_entry(li, &super->s_freeing_list, li_freeing_list) + if (li->vfs_inode.i_ino == ino) { + li->li_refcount++; + spin_unlock(&logfs_inode_lock); + *is_cached = 1; + return &li->vfs_inode; + } + spin_unlock(&logfs_inode_lock); + + *is_cached = 0; + return __logfs_iget(sb, ino); +} + +static void __logfs_destroy_inode(struct inode *inode) +{ + struct logfs_inode *li = logfs_inode(inode); + + BUG_ON(li->li_block); + list_del(&li->li_freeing_list); + kmem_cache_free(logfs_inode_cache, li); +} + +static void logfs_destroy_inode(struct inode *inode) +{ + struct logfs_inode *li = logfs_inode(inode); + + BUG_ON(list_empty(&li->li_freeing_list)); + spin_lock(&logfs_inode_lock); + li->li_refcount--; + if (li->li_refcount == 0) + __logfs_destroy_inode(inode); + spin_unlock(&logfs_inode_lock); +} + +void logfs_safe_iput(struct inode *inode, int is_cached) +{ + if (inode->i_ino == LOGFS_INO_MASTER) + return; + if (inode->i_ino == LOGFS_INO_SEGFILE) + return; + + if (is_cached) { + logfs_destroy_inode(inode); + return; + } + + iput(inode); +} + +static void logfs_init_inode(struct super_block *sb, struct inode *inode) +{ + struct logfs_inode *li = logfs_inode(inode); + int i; + + li->li_flags = 0; + li->li_height = 0; + li->li_used_bytes = 0; + li->li_block = NULL; + inode->i_uid = 0; + inode->i_gid = 0; + inode->i_size = 0; + inode->i_blocks = 0; + inode->i_ctime = CURRENT_TIME; + inode->i_mtime = CURRENT_TIME; + inode->i_nlink = 1; + INIT_LIST_HEAD(&li->li_freeing_list); + + for (i = 0; i < LOGFS_EMBEDDED_FIELDS; i++) + li->li_data[i] = 0; + + return; +} + +static struct inode *logfs_alloc_inode(struct super_block *sb) +{ + struct logfs_inode *li; + + li = kmem_cache_alloc(logfs_inode_cache, GFP_NOFS); + if (!li) + return NULL; + logfs_init_inode(sb, &li->vfs_inode); + return &li->vfs_inode; +} + +/* + * In logfs inodes are written to an inode file. The inode file, like any + * other file, is managed with a inode. The inode file's inode, aka master + * inode, requires special handling in several respects. First, it cannot be + * written to the inode file, so it is stored in the journal instead. + * + * Secondly, this inode cannot be written back and destroyed before all other + * inodes have been written. The ordering is important. Linux' VFS is happily + * unaware of the ordering constraint and would ordinarily destroy the master + * inode at umount time while other inodes are still in use and dirty. Not + * good. + * + * So logfs makes sure the master inode is not written until all other inodes + * have been destroyed. Sadly, this method has another side-effect. The VFS + * will notice one remaining inode and print a frightening warning message. + * Worse, it is impossible to judge whether such a warning was caused by the + * master inode or any other inodes have leaked as well. + * + * Our attempt of solving this is with logfs_new_meta_inode() below. Its + * purpose is to create a new inode that will not trigger the warning if such + * an inode is still in use. An ugly hack, no doubt. Suggections for + * improvement are welcome. + */ +struct inode *logfs_new_meta_inode(struct super_block *sb, u64 ino) +{ + struct inode *inode; + + inode = logfs_alloc_inode(sb); + if (!inode) + return ERR_PTR(-ENOMEM); + + inode->i_mode = S_IFREG; + inode->i_ino = ino; + inode->i_sb = sb; + + /* This is a blatant copy of alloc_inode code. We'd need alloc_inode + * to be nonstatic, alas. */ + { + struct address_space * const mapping = &inode->i_data; + + mapping->a_ops = &logfs_reg_aops; + mapping->host = inode; + mapping->flags = 0; + mapping_set_gfp_mask(mapping, GFP_NOFS); + mapping->assoc_mapping = NULL; + mapping->backing_dev_info = &default_backing_dev_info; + inode->i_mapping = mapping; + inode->i_nlink = 1; + } + + return inode; +} + +struct inode *logfs_read_meta_inode(struct super_block *sb, u64 ino) +{ + struct inode *inode; + int err; + + inode = logfs_new_meta_inode(sb, ino); + if (IS_ERR(inode)) + return inode; + + err = logfs_read_inode(inode); + if (err) { + destroy_meta_inode(inode); + return ERR_PTR(err); + } + logfs_inode_setops(inode); + return inode; +} + +static int logfs_write_inode(struct inode *inode, struct writeback_control *wbc) +{ + int ret; + long flags = WF_LOCK; + + /* Can only happen if creat() failed. Safe to skip. */ + if (logfs_inode(inode)->li_flags & LOGFS_IF_STILLBORN) + return 0; + + ret = __logfs_write_inode(inode, flags); + LOGFS_BUG_ON(ret, inode->i_sb); + return ret; +} + +void destroy_meta_inode(struct inode *inode) +{ + if (inode) { + if (inode->i_data.nrpages) + truncate_inode_pages(&inode->i_data, 0); + logfs_clear_inode(inode); + kmem_cache_free(logfs_inode_cache, logfs_inode(inode)); + } +} + +/* called with inode_lock held */ +static void logfs_drop_inode(struct inode *inode) +{ + struct logfs_super *super = logfs_super(inode->i_sb); + struct logfs_inode *li = logfs_inode(inode); + + spin_lock(&logfs_inode_lock); + list_move(&li->li_freeing_list, &super->s_freeing_list); + spin_unlock(&logfs_inode_lock); + generic_drop_inode(inode); +} + +static void logfs_set_ino_generation(struct super_block *sb, + struct inode *inode) +{ + struct logfs_super *super = logfs_super(sb); + u64 ino; + + mutex_lock(&super->s_journal_mutex); + ino = logfs_seek_hole(super->s_master_inode, super->s_last_ino); + super->s_last_ino = ino; + super->s_inos_till_wrap--; + if (super->s_inos_till_wrap < 0) { + super->s_last_ino = LOGFS_RESERVED_INOS; + super->s_generation++; + super->s_inos_till_wrap = INOS_PER_WRAP; + } + inode->i_ino = ino; + inode->i_generation = super->s_generation; + mutex_unlock(&super->s_journal_mutex); +} + +struct inode *logfs_new_inode(struct inode *dir, int mode) +{ + struct super_block *sb = dir->i_sb; + struct inode *inode; + + inode = new_inode(sb); + if (!inode) + return ERR_PTR(-ENOMEM); + + logfs_init_inode(sb, inode); + + /* inherit parent flags */ + logfs_inode(inode)->li_flags |= + logfs_inode(dir)->li_flags & LOGFS_FL_INHERITED; + + inode->i_mode = mode; + logfs_set_ino_generation(sb, inode); + + inode->i_uid = current_fsuid(); + inode->i_gid = current_fsgid(); + if (dir->i_mode & S_ISGID) { + inode->i_gid = dir->i_gid; + if (S_ISDIR(mode)) + inode->i_mode |= S_ISGID; + } + + logfs_inode_setops(inode); + insert_inode_hash(inode); + + return inode; +} + +static void logfs_init_once(void *_li) +{ + struct logfs_inode *li = _li; + int i; + + li->li_flags = 0; + li->li_used_bytes = 0; + li->li_refcount = 1; + for (i = 0; i < LOGFS_EMBEDDED_FIELDS; i++) + li->li_data[i] = 0; + inode_init_once(&li->vfs_inode); +} + +static int logfs_sync_fs(struct super_block *sb, int wait) +{ + /* FIXME: write anchor */ + logfs_super(sb)->s_devops->sync(sb); + return 0; +} + +const struct super_operations logfs_super_operations = { + .alloc_inode = logfs_alloc_inode, + .clear_inode = logfs_clear_inode, + .delete_inode = logfs_delete_inode, + .destroy_inode = logfs_destroy_inode, + .drop_inode = logfs_drop_inode, + .write_inode = logfs_write_inode, + .statfs = logfs_statfs, + .sync_fs = logfs_sync_fs, +}; + +int logfs_init_inode_cache(void) +{ + logfs_inode_cache = kmem_cache_create("logfs_inode_cache", + sizeof(struct logfs_inode), 0, SLAB_RECLAIM_ACCOUNT, + logfs_init_once); + if (!logfs_inode_cache) + return -ENOMEM; + return 0; +} + +void logfs_destroy_inode_cache(void) +{ + kmem_cache_destroy(logfs_inode_cache); +} diff --git a/fs/logfs/journal.c b/fs/logfs/journal.c new file mode 100644 index 000000000000..6ad30a4c9052 --- /dev/null +++ b/fs/logfs/journal.c @@ -0,0 +1,883 @@ +/* + * fs/logfs/journal.c - journal handling code + * + * As should be obvious for Linux kernel code, license is GPLv2 + * + * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> + */ +#include "logfs.h" + +static void logfs_calc_free(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + u64 reserve, no_segs = super->s_no_segs; + s64 free; + int i; + + /* superblock segments */ + no_segs -= 2; + super->s_no_journal_segs = 0; + /* journal */ + journal_for_each(i) + if (super->s_journal_seg[i]) { + no_segs--; + super->s_no_journal_segs++; + } + + /* open segments plus one extra per level for GC */ + no_segs -= 2 * super->s_total_levels; + + free = no_segs * (super->s_segsize - LOGFS_SEGMENT_RESERVE); + free -= super->s_used_bytes; + /* just a bit extra */ + free -= super->s_total_levels * 4096; + + /* Bad blocks are 'paid' for with speed reserve - the filesystem + * simply gets slower as bad blocks accumulate. Until the bad blocks + * exceed the speed reserve - then the filesystem gets smaller. + */ + reserve = super->s_bad_segments + super->s_bad_seg_reserve; + reserve *= super->s_segsize - LOGFS_SEGMENT_RESERVE; + reserve = max(reserve, super->s_speed_reserve); + free -= reserve; + if (free < 0) + free = 0; + + super->s_free_bytes = free; +} + +static void reserve_sb_and_journal(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + struct btree_head32 *head = &super->s_reserved_segments; + int i, err; + + err = btree_insert32(head, seg_no(sb, super->s_sb_ofs[0]), (void *)1, + GFP_KERNEL); + BUG_ON(err); + + err = btree_insert32(head, seg_no(sb, super->s_sb_ofs[1]), (void *)1, + GFP_KERNEL); + BUG_ON(err); + + journal_for_each(i) { + if (!super->s_journal_seg[i]) + continue; + err = btree_insert32(head, super->s_journal_seg[i], (void *)1, + GFP_KERNEL); + BUG_ON(err); + } +} + +static void read_dynsb(struct super_block *sb, + struct logfs_je_dynsb *dynsb) +{ + struct logfs_super *super = logfs_super(sb); + + super->s_gec = be64_to_cpu(dynsb->ds_gec); + super->s_sweeper = be64_to_cpu(dynsb->ds_sweeper); + super->s_victim_ino = be64_to_cpu(dynsb->ds_victim_ino); + super->s_rename_dir = be64_to_cpu(dynsb->ds_rename_dir); + super->s_rename_pos = be64_to_cpu(dynsb->ds_rename_pos); + super->s_used_bytes = be64_to_cpu(dynsb->ds_used_bytes); + super->s_generation = be32_to_cpu(dynsb->ds_generation); +} + +static void read_anchor(struct super_block *sb, + struct logfs_je_anchor *da) +{ + struct logfs_super *super = logfs_super(sb); + struct inode *inode = super->s_master_inode; + struct logfs_inode *li = logfs_inode(inode); + int i; + + super->s_last_ino = be64_to_cpu(da->da_last_ino); + li->li_flags = 0; + li->li_height = da->da_height; + i_size_write(inode, be64_to_cpu(da->da_size)); + li->li_used_bytes = be64_to_cpu(da->da_used_bytes); + + for (i = 0; i < LOGFS_EMBEDDED_FIELDS; i++) + li->li_data[i] = be64_to_cpu(da->da_data[i]); +} + +static void read_erasecount(struct super_block *sb, + struct logfs_je_journal_ec *ec) +{ + struct logfs_super *super = logfs_super(sb); + int i; + + journal_for_each(i) + super->s_journal_ec[i] = be32_to_cpu(ec->ec[i]); +} + +static int read_area(struct super_block *sb, struct logfs_je_area *a) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_area *area = super->s_area[a->gc_level]; + u64 ofs; + u32 writemask = ~(super->s_writesize - 1); + + if (a->gc_level >= LOGFS_NO_AREAS) + return -EIO; + if (a->vim != VIM_DEFAULT) + return -EIO; /* TODO: close area and continue */ + + area->a_used_bytes = be32_to_cpu(a->used_bytes); + area->a_written_bytes = area->a_used_bytes & writemask; + area->a_segno = be32_to_cpu(a->segno); + if (area->a_segno) + area->a_is_open = 1; + + ofs = dev_ofs(sb, area->a_segno, area->a_written_bytes); + if (super->s_writesize > 1) + logfs_buf_recover(area, ofs, a + 1, super->s_writesize); + else + logfs_buf_recover(area, ofs, NULL, 0); + return 0; +} + +static void *unpack(void *from, void *to) +{ + struct logfs_journal_header *jh = from; + void *data = from + sizeof(struct logfs_journal_header); + int err; + size_t inlen, outlen; + + inlen = be16_to_cpu(jh->h_len); + outlen = be16_to_cpu(jh->h_datalen); + + if (jh->h_compr == COMPR_NONE) + memcpy(to, data, inlen); + else { + err = logfs_uncompress(data, to, inlen, outlen); + BUG_ON(err); + } + return to; +} + +static int __read_je_header(struct super_block *sb, u64 ofs, + struct logfs_journal_header *jh) +{ + struct logfs_super *super = logfs_super(sb); + size_t bufsize = max_t(size_t, sb->s_blocksize, super->s_writesize) + + MAX_JOURNAL_HEADER; + u16 type, len, datalen; + int err; + + /* read header only */ + err = wbuf_read(sb, ofs, sizeof(*jh), jh); + if (err) + return err; + type = be16_to_cpu(jh->h_type); + len = be16_to_cpu(jh->h_len); + datalen = be16_to_cpu(jh->h_datalen); + if (len > sb->s_blocksize) + return -EIO; + if ((type < JE_FIRST) || (type > JE_LAST)) + return -EIO; + if (datalen > bufsize) + return -EIO; + return 0; +} + +static int __read_je_payload(struct super_block *sb, u64 ofs, + struct logfs_journal_header *jh) +{ + u16 len; + int err; + + len = be16_to_cpu(jh->h_len); + err = wbuf_read(sb, ofs + sizeof(*jh), len, jh + 1); + if (err) + return err; + if (jh->h_crc != logfs_crc32(jh, len + sizeof(*jh), 4)) { + /* Old code was confused. It forgot about the header length + * and stopped calculating the crc 16 bytes before the end + * of data - ick! + * FIXME: Remove this hack once the old code is fixed. + */ + if (jh->h_crc == logfs_crc32(jh, len, 4)) + WARN_ON_ONCE(1); + else + return -EIO; + } + return 0; +} + +/* + * jh needs to be large enough to hold the complete entry, not just the header + */ +static int __read_je(struct super_block *sb, u64 ofs, + struct logfs_journal_header *jh) +{ + int err; + + err = __read_je_header(sb, ofs, jh); + if (err) + return err; + return __read_je_payload(sb, ofs, jh); +} + +static int read_je(struct super_block *sb, u64 ofs) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_journal_header *jh = super->s_compressed_je; + void *scratch = super->s_je; + u16 type, datalen; + int err; + + err = __read_je(sb, ofs, jh); + if (err) + return err; + type = be16_to_cpu(jh->h_type); + datalen = be16_to_cpu(jh->h_datalen); + + switch (type) { + case JE_DYNSB: + read_dynsb(sb, unpack(jh, scratch)); + break; + case JE_ANCHOR: + read_anchor(sb, unpack(jh, scratch)); + break; + case JE_ERASECOUNT: + read_erasecount(sb, unpack(jh, scratch)); + break; + case JE_AREA: + read_area(sb, unpack(jh, scratch)); + break; + case JE_OBJ_ALIAS: + err = logfs_load_object_aliases(sb, unpack(jh, scratch), + datalen); + break; + default: + WARN_ON_ONCE(1); + return -EIO; + } + return err; +} + +static int logfs_read_segment(struct super_block *sb, u32 segno) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_journal_header *jh = super->s_compressed_je; + u64 ofs, seg_ofs = dev_ofs(sb, segno, 0); + u32 h_ofs, last_ofs = 0; + u16 len, datalen, last_len = 0; + int i, err; + + /* search for most recent commit */ + for (h_ofs = 0; h_ofs < super->s_segsize; h_ofs += sizeof(*jh)) { + ofs = seg_ofs + h_ofs; + err = __read_je_header(sb, ofs, jh); + if (err) + continue; + if (jh->h_type != cpu_to_be16(JE_COMMIT)) + continue; + err = __read_je_payload(sb, ofs, jh); + if (err) + continue; + len = be16_to_cpu(jh->h_len); + datalen = be16_to_cpu(jh->h_datalen); + if ((datalen > sizeof(super->s_je_array)) || + (datalen % sizeof(__be64))) + continue; + last_ofs = h_ofs; + last_len = datalen; + h_ofs += ALIGN(len, sizeof(*jh)) - sizeof(*jh); + } + /* read commit */ + if (last_ofs == 0) + return -ENOENT; + ofs = seg_ofs + last_ofs; + log_journal("Read commit from %llx\n", ofs); + err = __read_je(sb, ofs, jh); + BUG_ON(err); /* We should have caught it in the scan loop already */ + if (err) + return err; + /* uncompress */ + unpack(jh, super->s_je_array); + super->s_no_je = last_len / sizeof(__be64); + /* iterate over array */ + for (i = 0; i < super->s_no_je; i++) { + err = read_je(sb, be64_to_cpu(super->s_je_array[i])); + if (err) + return err; + } + super->s_journal_area->a_segno = segno; + return 0; +} + +static u64 read_gec(struct super_block *sb, u32 segno) +{ + struct logfs_segment_header sh; + __be32 crc; + int err; + + if (!segno) + return 0; + err = wbuf_read(sb, dev_ofs(sb, segno, 0), sizeof(sh), &sh); + if (err) + return 0; + crc = logfs_crc32(&sh, sizeof(sh), 4); + if (crc != sh.crc) { + WARN_ON(sh.gec != cpu_to_be64(0xffffffffffffffffull)); + /* Most likely it was just erased */ + return 0; + } + return be64_to_cpu(sh.gec); +} + +static int logfs_read_journal(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + u64 gec[LOGFS_JOURNAL_SEGS], max; + u32 segno; + int i, max_i; + + max = 0; + max_i = -1; + journal_for_each(i) { + segno = super->s_journal_seg[i]; + gec[i] = read_gec(sb, super->s_journal_seg[i]); + if (gec[i] > max) { + max = gec[i]; + max_i = i; + } + } + if (max_i == -1) + return -EIO; + /* FIXME: Try older segments in case of error */ + return logfs_read_segment(sb, super->s_journal_seg[max_i]); +} + +/* + * First search the current segment (outer loop), then pick the next segment + * in the array, skipping any zero entries (inner loop). + */ +static void journal_get_free_segment(struct logfs_area *area) +{ + struct logfs_super *super = logfs_super(area->a_sb); + int i; + + journal_for_each(i) { + if (area->a_segno != super->s_journal_seg[i]) + continue; + + do { + i++; + if (i == LOGFS_JOURNAL_SEGS) + i = 0; + } while (!super->s_journal_seg[i]); + + area->a_segno = super->s_journal_seg[i]; + area->a_erase_count = ++(super->s_journal_ec[i]); + log_journal("Journal now at %x (ec %x)\n", area->a_segno, + area->a_erase_count); + return; + } + BUG(); +} + +static void journal_get_erase_count(struct logfs_area *area) +{ + /* erase count is stored globally and incremented in + * journal_get_free_segment() - nothing to do here */ +} + +static int journal_erase_segment(struct logfs_area *area) +{ + struct super_block *sb = area->a_sb; + struct logfs_segment_header sh; + u64 ofs; + int err; + + err = logfs_erase_segment(sb, area->a_segno, 1); + if (err) + return err; + + sh.pad = 0; + sh.type = SEG_JOURNAL; + sh.level = 0; + sh.segno = cpu_to_be32(area->a_segno); + sh.ec = cpu_to_be32(area->a_erase_count); + sh.gec = cpu_to_be64(logfs_super(sb)->s_gec); + sh.crc = logfs_crc32(&sh, sizeof(sh), 4); + + /* This causes a bug in segment.c. Not yet. */ + //logfs_set_segment_erased(sb, area->a_segno, area->a_erase_count, 0); + + ofs = dev_ofs(sb, area->a_segno, 0); + area->a_used_bytes = ALIGN(sizeof(sh), 16); + logfs_buf_write(area, ofs, &sh, sizeof(sh)); + return 0; +} + +static size_t __logfs_write_header(struct logfs_super *super, + struct logfs_journal_header *jh, size_t len, size_t datalen, + u16 type, u8 compr) +{ + jh->h_len = cpu_to_be16(len); + jh->h_type = cpu_to_be16(type); + jh->h_datalen = cpu_to_be16(datalen); + jh->h_compr = compr; + jh->h_pad[0] = 'H'; + jh->h_pad[1] = 'E'; + jh->h_pad[2] = 'A'; + jh->h_pad[3] = 'D'; + jh->h_pad[4] = 'R'; + jh->h_crc = logfs_crc32(jh, len + sizeof(*jh), 4); + return ALIGN(len, 16) + sizeof(*jh); +} + +static size_t logfs_write_header(struct logfs_super *super, + struct logfs_journal_header *jh, size_t datalen, u16 type) +{ + size_t len = datalen; + + return __logfs_write_header(super, jh, len, datalen, type, COMPR_NONE); +} + +static inline size_t logfs_journal_erasecount_size(struct logfs_super *super) +{ + return LOGFS_JOURNAL_SEGS * sizeof(__be32); +} + +static void *logfs_write_erasecount(struct super_block *sb, void *_ec, + u16 *type, size_t *len) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_je_journal_ec *ec = _ec; + int i; + + journal_for_each(i) + ec->ec[i] = cpu_to_be32(super->s_journal_ec[i]); + *type = JE_ERASECOUNT; + *len = logfs_journal_erasecount_size(super); + return ec; +} + +static void account_shadow(void *_shadow, unsigned long _sb, u64 ignore, + size_t ignore2) +{ + struct logfs_shadow *shadow = _shadow; + struct super_block *sb = (void *)_sb; + struct logfs_super *super = logfs_super(sb); + + /* consume new space */ + super->s_free_bytes -= shadow->new_len; + super->s_used_bytes += shadow->new_len; + super->s_dirty_used_bytes -= shadow->new_len; + + /* free up old space */ + super->s_free_bytes += shadow->old_len; + super->s_used_bytes -= shadow->old_len; + super->s_dirty_free_bytes -= shadow->old_len; + + logfs_set_segment_used(sb, shadow->old_ofs, -shadow->old_len); + logfs_set_segment_used(sb, shadow->new_ofs, shadow->new_len); + + log_journal("account_shadow(%llx, %llx, %x) %llx->%llx %x->%x\n", + shadow->ino, shadow->bix, shadow->gc_level, + shadow->old_ofs, shadow->new_ofs, + shadow->old_len, shadow->new_len); + mempool_free(shadow, super->s_shadow_pool); +} + +static void account_shadows(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + struct inode *inode = super->s_master_inode; + struct logfs_inode *li = logfs_inode(inode); + struct shadow_tree *tree = &super->s_shadow_tree; + + btree_grim_visitor64(&tree->new, (unsigned long)sb, account_shadow); + btree_grim_visitor64(&tree->old, (unsigned long)sb, account_shadow); + + if (li->li_block) { + /* + * We never actually use the structure, when attached to the + * master inode. But it is easier to always free it here than + * to have checks in several places elsewhere when allocating + * it. + */ + li->li_block->ops->free_block(sb, li->li_block); + } + BUG_ON((s64)li->li_used_bytes < 0); +} + +static void *__logfs_write_anchor(struct super_block *sb, void *_da, + u16 *type, size_t *len) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_je_anchor *da = _da; + struct inode *inode = super->s_master_inode; + struct logfs_inode *li = logfs_inode(inode); + int i; + + da->da_height = li->li_height; + da->da_last_ino = cpu_to_be64(super->s_last_ino); + da->da_size = cpu_to_be64(i_size_read(inode)); + da->da_used_bytes = cpu_to_be64(li->li_used_bytes); + for (i = 0; i < LOGFS_EMBEDDED_FIELDS; i++) + da->da_data[i] = cpu_to_be64(li->li_data[i]); + *type = JE_ANCHOR; + *len = sizeof(*da); + return da; +} + +static void *logfs_write_dynsb(struct super_block *sb, void *_dynsb, + u16 *type, size_t *len) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_je_dynsb *dynsb = _dynsb; + + dynsb->ds_gec = cpu_to_be64(super->s_gec); + dynsb->ds_sweeper = cpu_to_be64(super->s_sweeper); + dynsb->ds_victim_ino = cpu_to_be64(super->s_victim_ino); + dynsb->ds_rename_dir = cpu_to_be64(super->s_rename_dir); + dynsb->ds_rename_pos = cpu_to_be64(super->s_rename_pos); + dynsb->ds_used_bytes = cpu_to_be64(super->s_used_bytes); + dynsb->ds_generation = cpu_to_be32(super->s_generation); + *type = JE_DYNSB; + *len = sizeof(*dynsb); + return dynsb; +} + +static void write_wbuf(struct super_block *sb, struct logfs_area *area, + void *wbuf) +{ + struct logfs_super *super = logfs_super(sb); + struct address_space *mapping = super->s_mapping_inode->i_mapping; + u64 ofs; + pgoff_t index; + int page_ofs; + struct page *page; + + ofs = dev_ofs(sb, area->a_segno, + area->a_used_bytes & ~(super->s_writesize - 1)); + index = ofs >> PAGE_SHIFT; + page_ofs = ofs & (PAGE_SIZE - 1); + + page = find_lock_page(mapping, index); + BUG_ON(!page); + memcpy(wbuf, page_address(page) + page_ofs, super->s_writesize); + unlock_page(page); +} + +static void *logfs_write_area(struct super_block *sb, void *_a, + u16 *type, size_t *len) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_area *area = super->s_area[super->s_sum_index]; + struct logfs_je_area *a = _a; + + a->vim = VIM_DEFAULT; + a->gc_level = super->s_sum_index; + a->used_bytes = cpu_to_be32(area->a_used_bytes); + a->segno = cpu_to_be32(area->a_segno); + if (super->s_writesize > 1) + write_wbuf(sb, area, a + 1); + + *type = JE_AREA; + *len = sizeof(*a) + super->s_writesize; + return a; +} + +static void *logfs_write_commit(struct super_block *sb, void *h, + u16 *type, size_t *len) +{ + struct logfs_super *super = logfs_super(sb); + + *type = JE_COMMIT; + *len = super->s_no_je * sizeof(__be64); + return super->s_je_array; +} + +static size_t __logfs_write_je(struct super_block *sb, void *buf, u16 type, + size_t len) +{ + struct logfs_super *super = logfs_super(sb); + void *header = super->s_compressed_je; + void *data = header + sizeof(struct logfs_journal_header); + ssize_t compr_len, pad_len; + u8 compr = COMPR_ZLIB; + + if (len == 0) + return logfs_write_header(super, header, 0, type); + + compr_len = logfs_compress(buf, data, len, sb->s_blocksize); + if (compr_len < 0 || type == JE_ANCHOR) { + BUG_ON(len > sb->s_blocksize); + memcpy(data, buf, len); + compr_len = len; + compr = COMPR_NONE; + } + + pad_len = ALIGN(compr_len, 16); + memset(data + compr_len, 0, pad_len - compr_len); + + return __logfs_write_header(super, header, compr_len, len, type, compr); +} + +static s64 logfs_get_free_bytes(struct logfs_area *area, size_t *bytes, + int must_pad) +{ + u32 writesize = logfs_super(area->a_sb)->s_writesize; + s32 ofs; + int ret; + + ret = logfs_open_area(area, *bytes); + if (ret) + return -EAGAIN; + + ofs = area->a_used_bytes; + area->a_used_bytes += *bytes; + + if (must_pad) { + area->a_used_bytes = ALIGN(area->a_used_bytes, writesize); + *bytes = area->a_used_bytes - ofs; + } + + return dev_ofs(area->a_sb, area->a_segno, ofs); +} + +static int logfs_write_je_buf(struct super_block *sb, void *buf, u16 type, + size_t buf_len) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_area *area = super->s_journal_area; + struct logfs_journal_header *jh = super->s_compressed_je; + size_t len; + int must_pad = 0; + s64 ofs; + + len = __logfs_write_je(sb, buf, type, buf_len); + if (jh->h_type == cpu_to_be16(JE_COMMIT)) + must_pad = 1; + + ofs = logfs_get_free_bytes(area, &len, must_pad); + if (ofs < 0) + return ofs; + logfs_buf_write(area, ofs, super->s_compressed_je, len); + super->s_je_array[super->s_no_je++] = cpu_to_be64(ofs); + return 0; +} + +static int logfs_write_je(struct super_block *sb, + void* (*write)(struct super_block *sb, void *scratch, + u16 *type, size_t *len)) +{ + void *buf; + size_t len; + u16 type; + + buf = write(sb, logfs_super(sb)->s_je, &type, &len); + return logfs_write_je_buf(sb, buf, type, len); +} + +int write_alias_journal(struct super_block *sb, u64 ino, u64 bix, + level_t level, int child_no, __be64 val) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_obj_alias *oa = super->s_je; + int err = 0, fill = super->s_je_fill; + + log_aliases("logfs_write_obj_aliases #%x(%llx, %llx, %x, %x) %llx\n", + fill, ino, bix, level, child_no, be64_to_cpu(val)); + oa[fill].ino = cpu_to_be64(ino); + oa[fill].bix = cpu_to_be64(bix); + oa[fill].val = val; + oa[fill].level = (__force u8)level; + oa[fill].child_no = cpu_to_be16(child_no); + fill++; + if (fill >= sb->s_blocksize / sizeof(*oa)) { + err = logfs_write_je_buf(sb, oa, JE_OBJ_ALIAS, sb->s_blocksize); + fill = 0; + } + + super->s_je_fill = fill; + return err; +} + +static int logfs_write_obj_aliases(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + int err; + + log_journal("logfs_write_obj_aliases: %d aliases to write\n", + super->s_no_object_aliases); + super->s_je_fill = 0; + err = logfs_write_obj_aliases_pagecache(sb); + if (err) + return err; + + if (super->s_je_fill) + err = logfs_write_je_buf(sb, super->s_je, JE_OBJ_ALIAS, + super->s_je_fill + * sizeof(struct logfs_obj_alias)); + return err; +} + +/* + * Write all journal entries. The goto logic ensures that all journal entries + * are written whenever a new segment is used. It is ugly and potentially a + * bit wasteful, but robustness is more important. With this we can *always* + * erase all journal segments except the one containing the most recent commit. + */ +void logfs_write_anchor(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_area *area = super->s_journal_area; + int i, err; + + if (!(super->s_flags & LOGFS_SB_FLAG_DIRTY)) + return; + super->s_flags &= ~LOGFS_SB_FLAG_DIRTY; + + BUG_ON(super->s_flags & LOGFS_SB_FLAG_SHUTDOWN); + mutex_lock(&super->s_journal_mutex); + + /* Do this first or suffer corruption */ + logfs_sync_segments(sb); + account_shadows(sb); + +again: + super->s_no_je = 0; + for_each_area(i) { + if (!super->s_area[i]->a_is_open) + continue; + super->s_sum_index = i; + err = logfs_write_je(sb, logfs_write_area); + if (err) + goto again; + } + err = logfs_write_obj_aliases(sb); + if (err) + goto again; + err = logfs_write_je(sb, logfs_write_erasecount); + if (err) + goto again; + err = logfs_write_je(sb, __logfs_write_anchor); + if (err) + goto again; + err = logfs_write_je(sb, logfs_write_dynsb); + if (err) + goto again; + /* + * Order is imperative. First we sync all writes, including the + * non-committed journal writes. Then we write the final commit and + * sync the current journal segment. + * There is a theoretical bug here. Syncing the journal segment will + * write a number of journal entries and the final commit. All these + * are written in a single operation. If the device layer writes the + * data back-to-front, the commit will precede the other journal + * entries, leaving a race window. + * Two fixes are possible. Preferred is to fix the device layer to + * ensure writes happen front-to-back. Alternatively we can insert + * another logfs_sync_area() super->s_devops->sync() combo before + * writing the commit. + */ + /* + * On another subject, super->s_devops->sync is usually not necessary. + * Unless called from sys_sync or friends, a barrier would suffice. + */ + super->s_devops->sync(sb); + err = logfs_write_je(sb, logfs_write_commit); + if (err) + goto again; + log_journal("Write commit to %llx\n", + be64_to_cpu(super->s_je_array[super->s_no_je - 1])); + logfs_sync_area(area); + BUG_ON(area->a_used_bytes != area->a_written_bytes); + super->s_devops->sync(sb); + + mutex_unlock(&super->s_journal_mutex); + return; +} + +void do_logfs_journal_wl_pass(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_area *area = super->s_journal_area; + u32 segno, ec; + int i, err; + + log_journal("Journal requires wear-leveling.\n"); + /* Drop old segments */ + journal_for_each(i) + if (super->s_journal_seg[i]) { + logfs_set_segment_unreserved(sb, + super->s_journal_seg[i], + super->s_journal_ec[i]); + super->s_journal_seg[i] = 0; + super->s_journal_ec[i] = 0; + } + /* Get new segments */ + for (i = 0; i < super->s_no_journal_segs; i++) { + segno = get_best_cand(sb, &super->s_reserve_list, &ec); + super->s_journal_seg[i] = segno; + super->s_journal_ec[i] = ec; + logfs_set_segment_reserved(sb, segno); + } + /* Manually move journal_area */ + area->a_segno = super->s_journal_seg[0]; + area->a_is_open = 0; + area->a_used_bytes = 0; + /* Write journal */ + logfs_write_anchor(sb); + /* Write superblocks */ + err = logfs_write_sb(sb); + BUG_ON(err); +} + +static const struct logfs_area_ops journal_area_ops = { + .get_free_segment = journal_get_free_segment, + .get_erase_count = journal_get_erase_count, + .erase_segment = journal_erase_segment, +}; + +int logfs_init_journal(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + size_t bufsize = max_t(size_t, sb->s_blocksize, super->s_writesize) + + MAX_JOURNAL_HEADER; + int ret = -ENOMEM; + + mutex_init(&super->s_journal_mutex); + btree_init_mempool32(&super->s_reserved_segments, super->s_btree_pool); + + super->s_je = kzalloc(bufsize, GFP_KERNEL); + if (!super->s_je) + return ret; + + super->s_compressed_je = kzalloc(bufsize, GFP_KERNEL); + if (!super->s_compressed_je) + return ret; + + super->s_master_inode = logfs_new_meta_inode(sb, LOGFS_INO_MASTER); + if (IS_ERR(super->s_master_inode)) + return PTR_ERR(super->s_master_inode); + + ret = logfs_read_journal(sb); + if (ret) + return -EIO; + + reserve_sb_and_journal(sb); + logfs_calc_free(sb); + + super->s_journal_area->a_ops = &journal_area_ops; + return 0; +} + +void logfs_cleanup_journal(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + + btree_grim_visitor32(&super->s_reserved_segments, 0, NULL); + destroy_meta_inode(super->s_master_inode); + super->s_master_inode = NULL; + + kfree(super->s_compressed_je); + kfree(super->s_je); +} diff --git a/fs/logfs/logfs.h b/fs/logfs/logfs.h new file mode 100644 index 000000000000..129779431373 --- /dev/null +++ b/fs/logfs/logfs.h @@ -0,0 +1,724 @@ +/* + * fs/logfs/logfs.h + * + * As should be obvious for Linux kernel code, license is GPLv2 + * + * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> + * + * Private header for logfs. + */ +#ifndef FS_LOGFS_LOGFS_H +#define FS_LOGFS_LOGFS_H + +#undef __CHECK_ENDIAN__ +#define __CHECK_ENDIAN__ + +#include <linux/btree.h> +#include <linux/crc32.h> +#include <linux/fs.h> +#include <linux/kernel.h> +#include <linux/mempool.h> +#include <linux/pagemap.h> +#include <linux/mtd/mtd.h> +#include "logfs_abi.h" + +#define LOGFS_DEBUG_SUPER (0x0001) +#define LOGFS_DEBUG_SEGMENT (0x0002) +#define LOGFS_DEBUG_JOURNAL (0x0004) +#define LOGFS_DEBUG_DIR (0x0008) +#define LOGFS_DEBUG_FILE (0x0010) +#define LOGFS_DEBUG_INODE (0x0020) +#define LOGFS_DEBUG_READWRITE (0x0040) +#define LOGFS_DEBUG_GC (0x0080) +#define LOGFS_DEBUG_GC_NOISY (0x0100) +#define LOGFS_DEBUG_ALIASES (0x0200) +#define LOGFS_DEBUG_BLOCKMOVE (0x0400) +#define LOGFS_DEBUG_ALL (0xffffffff) + +#define LOGFS_DEBUG (0x01) +/* + * To enable specific log messages, simply define LOGFS_DEBUG to match any + * or all of the above. + */ +#ifndef LOGFS_DEBUG +#define LOGFS_DEBUG (0) +#endif + +#define log_cond(cond, fmt, arg...) do { \ + if (cond) \ + printk(KERN_DEBUG fmt, ##arg); \ +} while (0) + +#define log_super(fmt, arg...) \ + log_cond(LOGFS_DEBUG & LOGFS_DEBUG_SUPER, fmt, ##arg) +#define log_segment(fmt, arg...) \ + log_cond(LOGFS_DEBUG & LOGFS_DEBUG_SEGMENT, fmt, ##arg) +#define log_journal(fmt, arg...) \ + log_cond(LOGFS_DEBUG & LOGFS_DEBUG_JOURNAL, fmt, ##arg) +#define log_dir(fmt, arg...) \ + log_cond(LOGFS_DEBUG & LOGFS_DEBUG_DIR, fmt, ##arg) +#define log_file(fmt, arg...) \ + log_cond(LOGFS_DEBUG & LOGFS_DEBUG_FILE, fmt, ##arg) +#define log_inode(fmt, arg...) \ + log_cond(LOGFS_DEBUG & LOGFS_DEBUG_INODE, fmt, ##arg) +#define log_readwrite(fmt, arg...) \ + log_cond(LOGFS_DEBUG & LOGFS_DEBUG_READWRITE, fmt, ##arg) +#define log_gc(fmt, arg...) \ + log_cond(LOGFS_DEBUG & LOGFS_DEBUG_GC, fmt, ##arg) +#define log_gc_noisy(fmt, arg...) \ + log_cond(LOGFS_DEBUG & LOGFS_DEBUG_GC_NOISY, fmt, ##arg) +#define log_aliases(fmt, arg...) \ + log_cond(LOGFS_DEBUG & LOGFS_DEBUG_ALIASES, fmt, ##arg) +#define log_blockmove(fmt, arg...) \ + log_cond(LOGFS_DEBUG & LOGFS_DEBUG_BLOCKMOVE, fmt, ##arg) + +#define PG_pre_locked PG_owner_priv_1 +#define PagePreLocked(page) test_bit(PG_pre_locked, &(page)->flags) +#define SetPagePreLocked(page) set_bit(PG_pre_locked, &(page)->flags) +#define ClearPagePreLocked(page) clear_bit(PG_pre_locked, &(page)->flags) + +/* FIXME: This should really be somewhere in the 64bit area. */ +#define LOGFS_LINK_MAX (1<<30) + +/* Read-only filesystem */ +#define LOGFS_SB_FLAG_RO 0x0001 +#define LOGFS_SB_FLAG_DIRTY 0x0002 +#define LOGFS_SB_FLAG_OBJ_ALIAS 0x0004 +#define LOGFS_SB_FLAG_SHUTDOWN 0x0008 + +/* Write Control Flags */ +#define WF_LOCK 0x01 /* take write lock */ +#define WF_WRITE 0x02 /* write block */ +#define WF_DELETE 0x04 /* delete old block */ + +typedef u8 __bitwise level_t; +typedef u8 __bitwise gc_level_t; + +#define LEVEL(level) ((__force level_t)(level)) +#define GC_LEVEL(gc_level) ((__force gc_level_t)(gc_level)) + +#define SUBLEVEL(level) ( (void)((level) == LEVEL(1)), \ + (__force level_t)((__force u8)(level) - 1) ) + +/** + * struct logfs_area - area management information + * + * @a_sb: the superblock this area belongs to + * @a_is_open: 1 if the area is currently open, else 0 + * @a_segno: segment number of area + * @a_written_bytes: number of bytes already written back + * @a_used_bytes: number of used bytes + * @a_ops: area operations (either journal or ostore) + * @a_erase_count: erase count + * @a_level: GC level + */ +struct logfs_area { /* a segment open for writing */ + struct super_block *a_sb; + int a_is_open; + u32 a_segno; + u32 a_written_bytes; + u32 a_used_bytes; + const struct logfs_area_ops *a_ops; + u32 a_erase_count; + gc_level_t a_level; +}; + +/** + * struct logfs_area_ops - area operations + * + * @get_free_segment: fill area->ofs with the offset of a free segment + * @get_erase_count: fill area->erase_count (needs area->ofs) + * @erase_segment: erase and setup segment + */ +struct logfs_area_ops { + void (*get_free_segment)(struct logfs_area *area); + void (*get_erase_count)(struct logfs_area *area); + int (*erase_segment)(struct logfs_area *area); +}; + +/** + * struct logfs_device_ops - device access operations + * + * @readpage: read one page (mm page) + * @writeseg: write one segment. may be a partial segment + * @erase: erase one segment + * @read: read from the device + * @erase: erase part of the device + */ +struct logfs_device_ops { + struct page *(*find_first_sb)(struct super_block *sb, u64 *ofs); + struct page *(*find_last_sb)(struct super_block *sb, u64 *ofs); + int (*write_sb)(struct super_block *sb, struct page *page); + int (*readpage)(void *_sb, struct page *page); + void (*writeseg)(struct super_block *sb, u64 ofs, size_t len); + int (*erase)(struct super_block *sb, loff_t ofs, size_t len, + int ensure_write); + void (*sync)(struct super_block *sb); + void (*put_device)(struct super_block *sb); +}; + +/** + * struct candidate_list - list of similar candidates + */ +struct candidate_list { + struct rb_root rb_tree; + int count; + int maxcount; + int sort_by_ec; +}; + +/** + * struct gc_candidate - "candidate" segment to be garbage collected next + * + * @list: list (either free of low) + * @segno: segment number + * @valid: number of valid bytes + * @erase_count: erase count of segment + * @dist: distance from tree root + * + * Candidates can be on two lists. The free list contains electees rather + * than candidates - segments that no longer contain any valid data. The + * low list contains candidates to be picked for GC. It should be kept + * short. It is not required to always pick a perfect candidate. In the + * worst case GC will have to move more data than absolutely necessary. + */ +struct gc_candidate { + struct rb_node rb_node; + struct candidate_list *list; + u32 segno; + u32 valid; + u32 erase_count; + u8 dist; +}; + +/** + * struct logfs_journal_entry - temporary structure used during journal scan + * + * @used: + * @version: normalized version + * @len: length + * @offset: offset + */ +struct logfs_journal_entry { + int used; + s16 version; + u16 len; + u16 datalen; + u64 offset; +}; + +enum transaction_state { + CREATE_1 = 1, + CREATE_2, + UNLINK_1, + UNLINK_2, + CROSS_RENAME_1, + CROSS_RENAME_2, + TARGET_RENAME_1, + TARGET_RENAME_2, + TARGET_RENAME_3 +}; + +/** + * struct logfs_transaction - essential fields to support atomic dirops + * + * @ino: target inode + * @dir: inode of directory containing dentry + * @pos: pos of dentry in directory + */ +struct logfs_transaction { + enum transaction_state state; + u64 ino; + u64 dir; + u64 pos; +}; + +/** + * struct logfs_shadow - old block in the shadow of a not-yet-committed new one + * @old_ofs: offset of old block on medium + * @new_ofs: offset of new block on medium + * @ino: inode number + * @bix: block index + * @old_len: size of old block, including header + * @new_len: size of new block, including header + * @level: block level + */ +struct logfs_shadow { + u64 old_ofs; + u64 new_ofs; + u64 ino; + u64 bix; + int old_len; + int new_len; + gc_level_t gc_level; +}; + +/** + * struct shadow_tree + * @new: shadows where old_ofs==0, indexed by new_ofs + * @old: shadows where old_ofs!=0, indexed by old_ofs + */ +struct shadow_tree { + struct btree_head64 new; + struct btree_head64 old; +}; + +struct object_alias_item { + struct list_head list; + __be64 val; + int child_no; +}; + +/** + * struct logfs_block - contains any block state + * @type: indirect block or inode + * @full: number of fully populated children + * @partial: number of partially populated children + * + * Most blocks are directly represented by page cache pages. But when a block + * becomes dirty, is part of a transaction, contains aliases or is otherwise + * special, a struct logfs_block is allocated to track the additional state. + * Inodes are very similar to indirect blocks, so they can also get one of + * these structures added when appropriate. + */ +#define BLOCK_INDIRECT 1 /* Indirect block */ +#define BLOCK_INODE 2 /* Inode */ +struct logfs_block_ops; +struct logfs_block { + struct list_head alias_list; + struct list_head item_list; + struct super_block *sb; + u64 ino; + u64 bix; + level_t level; + struct page *page; + struct inode *inode; + struct logfs_transaction *ta; + unsigned long alias_map[LOGFS_BLOCK_FACTOR / BITS_PER_LONG]; + struct logfs_block_ops *ops; + int full; + int partial; + int reserved_bytes; +}; + +typedef int write_alias_t(struct super_block *sb, u64 ino, u64 bix, + level_t level, int child_no, __be64 val); +struct logfs_block_ops { + void (*write_block)(struct logfs_block *block); + gc_level_t (*block_level)(struct logfs_block *block); + void (*free_block)(struct super_block *sb, struct logfs_block*block); + int (*write_alias)(struct super_block *sb, + struct logfs_block *block, + write_alias_t *write_one_alias); +}; + +struct logfs_super { + struct mtd_info *s_mtd; /* underlying device */ + struct block_device *s_bdev; /* underlying device */ + const struct logfs_device_ops *s_devops;/* device access */ + struct inode *s_master_inode; /* inode file */ + struct inode *s_segfile_inode; /* segment file */ + struct inode *s_mapping_inode; /* device mapping */ + atomic_t s_pending_writes; /* outstanting bios */ + long s_flags; + mempool_t *s_btree_pool; /* for btree nodes */ + mempool_t *s_alias_pool; /* aliases in segment.c */ + u64 s_feature_incompat; + u64 s_feature_ro_compat; + u64 s_feature_compat; + u64 s_feature_flags; + u64 s_sb_ofs[2]; + struct page *s_erase_page; /* for dev_bdev.c */ + /* alias.c fields */ + struct btree_head32 s_segment_alias; /* remapped segments */ + int s_no_object_aliases; + struct list_head s_object_alias; /* remapped objects */ + struct btree_head128 s_object_alias_tree; /* remapped objects */ + struct mutex s_object_alias_mutex; + /* dir.c fields */ + struct mutex s_dirop_mutex; /* for creat/unlink/rename */ + u64 s_victim_ino; /* used for atomic dir-ops */ + u64 s_rename_dir; /* source directory ino */ + u64 s_rename_pos; /* position of source dd */ + /* gc.c fields */ + long s_segsize; /* size of a segment */ + int s_segshift; /* log2 of segment size */ + long s_segmask; /* 1 << s_segshift - 1 */ + long s_no_segs; /* segments on device */ + long s_no_journal_segs; /* segments used for journal */ + long s_no_blocks; /* blocks per segment */ + long s_writesize; /* minimum write size */ + int s_writeshift; /* log2 of write size */ + u64 s_size; /* filesystem size */ + struct logfs_area *s_area[LOGFS_NO_AREAS]; /* open segment array */ + u64 s_gec; /* global erase count */ + u64 s_wl_gec_ostore; /* time of last wl event */ + u64 s_wl_gec_journal; /* time of last wl event */ + u64 s_sweeper; /* current sweeper pos */ + u8 s_ifile_levels; /* max level of ifile */ + u8 s_iblock_levels; /* max level of regular files */ + u8 s_data_levels; /* # of segments to leaf block*/ + u8 s_total_levels; /* sum of above three */ + struct btree_head32 s_cand_tree; /* all candidates */ + struct candidate_list s_free_list; /* 100% free segments */ + struct candidate_list s_reserve_list; /* Bad segment reserve */ + struct candidate_list s_low_list[LOGFS_NO_AREAS];/* good candidates */ + struct candidate_list s_ec_list; /* wear level candidates */ + struct btree_head32 s_reserved_segments;/* sb, journal, bad, etc. */ + /* inode.c fields */ + u64 s_last_ino; /* highest ino used */ + long s_inos_till_wrap; + u32 s_generation; /* i_generation for new files */ + struct list_head s_freeing_list; /* inodes being freed */ + /* journal.c fields */ + struct mutex s_journal_mutex; + void *s_je; /* journal entry to compress */ + void *s_compressed_je; /* block to write to journal */ + u32 s_journal_seg[LOGFS_JOURNAL_SEGS]; /* journal segments */ + u32 s_journal_ec[LOGFS_JOURNAL_SEGS]; /* journal erasecounts */ + u64 s_last_version; + struct logfs_area *s_journal_area; /* open journal segment */ + __be64 s_je_array[64]; + int s_no_je; + + int s_sum_index; /* for the 12 summaries */ + struct shadow_tree s_shadow_tree; + int s_je_fill; /* index of current je */ + /* readwrite.c fields */ + struct mutex s_write_mutex; + int s_lock_count; + mempool_t *s_block_pool; /* struct logfs_block pool */ + mempool_t *s_shadow_pool; /* struct logfs_shadow pool */ + /* + * Space accounting: + * - s_used_bytes specifies space used to store valid data objects. + * - s_dirty_used_bytes is space used to store non-committed data + * objects. Those objects have already been written themselves, + * but they don't become valid until all indirect blocks up to the + * journal have been written as well. + * - s_dirty_free_bytes is space used to store the old copy of a + * replaced object, as long as the replacement is non-committed. + * In other words, it is the amount of space freed when all dirty + * blocks are written back. + * - s_free_bytes is the amount of free space available for any + * purpose. + * - s_root_reserve is the amount of free space available only to + * the root user. Non-privileged users can no longer write once + * this watermark has been reached. + * - s_speed_reserve is space which remains unused to speed up + * garbage collection performance. + * - s_dirty_pages is the space reserved for currently dirty pages. + * It is a pessimistic estimate, so some/most will get freed on + * page writeback. + * + * s_used_bytes + s_free_bytes + s_speed_reserve = total usable size + */ + u64 s_free_bytes; + u64 s_used_bytes; + u64 s_dirty_free_bytes; + u64 s_dirty_used_bytes; + u64 s_root_reserve; + u64 s_speed_reserve; + u64 s_dirty_pages; + /* Bad block handling: + * - s_bad_seg_reserve is a number of segments usually kept + * free. When encountering bad blocks, the affected segment's data + * is _temporarily_ moved to a reserved segment. + * - s_bad_segments is the number of known bad segments. + */ + u32 s_bad_seg_reserve; + u32 s_bad_segments; +}; + +/** + * struct logfs_inode - in-memory inode + * + * @vfs_inode: struct inode + * @li_data: data pointers + * @li_used_bytes: number of used bytes + * @li_freeing_list: used to track inodes currently being freed + * @li_flags: inode flags + * @li_refcount: number of internal (GC-induced) references + */ +struct logfs_inode { + struct inode vfs_inode; + u64 li_data[LOGFS_EMBEDDED_FIELDS]; + u64 li_used_bytes; + struct list_head li_freeing_list; + struct logfs_block *li_block; + u32 li_flags; + u8 li_height; + int li_refcount; +}; + +#define journal_for_each(__i) for (__i = 0; __i < LOGFS_JOURNAL_SEGS; __i++) +#define for_each_area(__i) for (__i = 0; __i < LOGFS_NO_AREAS; __i++) +#define for_each_area_down(__i) for (__i = LOGFS_NO_AREAS - 1; __i >= 0; __i--) + +/* compr.c */ +int logfs_compress(void *in, void *out, size_t inlen, size_t outlen); +int logfs_uncompress(void *in, void *out, size_t inlen, size_t outlen); +int __init logfs_compr_init(void); +void logfs_compr_exit(void); + +/* dev_bdev.c */ +#ifdef CONFIG_BLOCK +int logfs_get_sb_bdev(struct file_system_type *type, int flags, + const char *devname, struct vfsmount *mnt); +#else +static inline int logfs_get_sb_bdev(struct file_system_type *type, int flags, + const char *devname, struct vfsmount *mnt) +{ + return -ENODEV; +} +#endif + +/* dev_mtd.c */ +#ifdef CONFIG_MTD +int logfs_get_sb_mtd(struct file_system_type *type, int flags, + int mtdnr, struct vfsmount *mnt); +#else +static inline int logfs_get_sb_mtd(struct file_system_type *type, int flags, + int mtdnr, struct vfsmount *mnt) +{ + return -ENODEV; +} +#endif + +/* dir.c */ +extern const struct inode_operations logfs_symlink_iops; +extern const struct inode_operations logfs_dir_iops; +extern const struct file_operations logfs_dir_fops; +int logfs_replay_journal(struct super_block *sb); + +/* file.c */ +extern const struct inode_operations logfs_reg_iops; +extern const struct file_operations logfs_reg_fops; +extern const struct address_space_operations logfs_reg_aops; +int logfs_readpage(struct file *file, struct page *page); +int logfs_ioctl(struct inode *inode, struct file *file, unsigned int cmd, + unsigned long arg); +int logfs_fsync(struct file *file, struct dentry *dentry, int datasync); + +/* gc.c */ +u32 get_best_cand(struct super_block *sb, struct candidate_list *list, u32 *ec); +void logfs_gc_pass(struct super_block *sb); +int logfs_check_areas(struct super_block *sb); +int logfs_init_gc(struct super_block *sb); +void logfs_cleanup_gc(struct super_block *sb); + +/* inode.c */ +extern const struct super_operations logfs_super_operations; +struct inode *logfs_iget(struct super_block *sb, ino_t ino); +struct inode *logfs_safe_iget(struct super_block *sb, ino_t ino, int *cookie); +void logfs_safe_iput(struct inode *inode, int cookie); +struct inode *logfs_new_inode(struct inode *dir, int mode); +struct inode *logfs_new_meta_inode(struct super_block *sb, u64 ino); +struct inode *logfs_read_meta_inode(struct super_block *sb, u64 ino); +int logfs_init_inode_cache(void); +void logfs_destroy_inode_cache(void); +void destroy_meta_inode(struct inode *inode); +void logfs_set_blocks(struct inode *inode, u64 no); +/* these logically belong into inode.c but actually reside in readwrite.c */ +int logfs_read_inode(struct inode *inode); +int __logfs_write_inode(struct inode *inode, long flags); +void logfs_delete_inode(struct inode *inode); +void logfs_clear_inode(struct inode *inode); + +/* journal.c */ +void logfs_write_anchor(struct super_block *sb); +int logfs_init_journal(struct super_block *sb); +void logfs_cleanup_journal(struct super_block *sb); +int write_alias_journal(struct super_block *sb, u64 ino, u64 bix, + level_t level, int child_no, __be64 val); +void do_logfs_journal_wl_pass(struct super_block *sb); + +/* readwrite.c */ +pgoff_t logfs_pack_index(u64 bix, level_t level); +void logfs_unpack_index(pgoff_t index, u64 *bix, level_t *level); +int logfs_inode_write(struct inode *inode, const void *buf, size_t count, + loff_t bix, long flags, struct shadow_tree *shadow_tree); +int logfs_readpage_nolock(struct page *page); +int logfs_write_buf(struct inode *inode, struct page *page, long flags); +int logfs_delete(struct inode *inode, pgoff_t index, + struct shadow_tree *shadow_tree); +int logfs_rewrite_block(struct inode *inode, u64 bix, u64 ofs, + gc_level_t gc_level, long flags); +int logfs_is_valid_block(struct super_block *sb, u64 ofs, u64 ino, u64 bix, + gc_level_t gc_level); +int logfs_truncate(struct inode *inode, u64 size); +u64 logfs_seek_hole(struct inode *inode, u64 bix); +u64 logfs_seek_data(struct inode *inode, u64 bix); +int logfs_open_segfile(struct super_block *sb); +int logfs_init_rw(struct super_block *sb); +void logfs_cleanup_rw(struct super_block *sb); +void logfs_add_transaction(struct inode *inode, struct logfs_transaction *ta); +void logfs_del_transaction(struct inode *inode, struct logfs_transaction *ta); +void logfs_write_block(struct logfs_block *block, long flags); +int logfs_write_obj_aliases_pagecache(struct super_block *sb); +void logfs_get_segment_entry(struct super_block *sb, u32 segno, + struct logfs_segment_entry *se); +void logfs_set_segment_used(struct super_block *sb, u64 ofs, int increment); +void logfs_set_segment_erased(struct super_block *sb, u32 segno, u32 ec, + gc_level_t gc_level); +void logfs_set_segment_reserved(struct super_block *sb, u32 segno); +void logfs_set_segment_unreserved(struct super_block *sb, u32 segno, u32 ec); +struct logfs_block *__alloc_block(struct super_block *sb, + u64 ino, u64 bix, level_t level); +void __free_block(struct super_block *sb, struct logfs_block *block); +void btree_write_block(struct logfs_block *block); +void initialize_block_counters(struct page *page, struct logfs_block *block, + __be64 *array, int page_is_empty); +int logfs_exist_block(struct inode *inode, u64 bix); +int get_page_reserve(struct inode *inode, struct page *page); +extern struct logfs_block_ops indirect_block_ops; + +/* segment.c */ +int logfs_erase_segment(struct super_block *sb, u32 ofs, int ensure_erase); +int wbuf_read(struct super_block *sb, u64 ofs, size_t len, void *buf); +int logfs_segment_read(struct inode *inode, struct page *page, u64 ofs, u64 bix, + level_t level); +int logfs_segment_write(struct inode *inode, struct page *page, + struct logfs_shadow *shadow); +int logfs_segment_delete(struct inode *inode, struct logfs_shadow *shadow); +int logfs_load_object_aliases(struct super_block *sb, + struct logfs_obj_alias *oa, int count); +void move_page_to_btree(struct page *page); +int logfs_init_mapping(struct super_block *sb); +void logfs_sync_area(struct logfs_area *area); +void logfs_sync_segments(struct super_block *sb); + +/* area handling */ +int logfs_init_areas(struct super_block *sb); +void logfs_cleanup_areas(struct super_block *sb); +int logfs_open_area(struct logfs_area *area, size_t bytes); +void __logfs_buf_write(struct logfs_area *area, u64 ofs, void *buf, size_t len, + int use_filler); + +static inline void logfs_buf_write(struct logfs_area *area, u64 ofs, + void *buf, size_t len) +{ + __logfs_buf_write(area, ofs, buf, len, 0); +} + +static inline void logfs_buf_recover(struct logfs_area *area, u64 ofs, + void *buf, size_t len) +{ + __logfs_buf_write(area, ofs, buf, len, 1); +} + +/* super.c */ +struct page *emergency_read_begin(struct address_space *mapping, pgoff_t index); +void emergency_read_end(struct page *page); +void logfs_crash_dump(struct super_block *sb); +void *memchr_inv(const void *s, int c, size_t n); +int logfs_statfs(struct dentry *dentry, struct kstatfs *stats); +int logfs_get_sb_device(struct file_system_type *type, int flags, + struct mtd_info *mtd, struct block_device *bdev, + const struct logfs_device_ops *devops, struct vfsmount *mnt); +int logfs_check_ds(struct logfs_disk_super *ds); +int logfs_write_sb(struct super_block *sb); + +static inline struct logfs_super *logfs_super(struct super_block *sb) +{ + return sb->s_fs_info; +} + +static inline struct logfs_inode *logfs_inode(struct inode *inode) +{ + return container_of(inode, struct logfs_inode, vfs_inode); +} + +static inline void logfs_set_ro(struct super_block *sb) +{ + logfs_super(sb)->s_flags |= LOGFS_SB_FLAG_RO; +} + +#define LOGFS_BUG(sb) do { \ + struct super_block *__sb = sb; \ + logfs_crash_dump(__sb); \ + logfs_super(__sb)->s_flags |= LOGFS_SB_FLAG_RO; \ + BUG(); \ +} while (0) + +#define LOGFS_BUG_ON(condition, sb) \ + do { if (unlikely(condition)) LOGFS_BUG((sb)); } while (0) + +static inline __be32 logfs_crc32(void *data, size_t len, size_t skip) +{ + return cpu_to_be32(crc32(~0, data+skip, len-skip)); +} + +static inline u8 logfs_type(struct inode *inode) +{ + return (inode->i_mode >> 12) & 15; +} + +static inline pgoff_t logfs_index(struct super_block *sb, u64 pos) +{ + return pos >> sb->s_blocksize_bits; +} + +static inline u64 dev_ofs(struct super_block *sb, u32 segno, u32 ofs) +{ + return ((u64)segno << logfs_super(sb)->s_segshift) + ofs; +} + +static inline u32 seg_no(struct super_block *sb, u64 ofs) +{ + return ofs >> logfs_super(sb)->s_segshift; +} + +static inline u32 seg_ofs(struct super_block *sb, u64 ofs) +{ + return ofs & logfs_super(sb)->s_segmask; +} + +static inline u64 seg_align(struct super_block *sb, u64 ofs) +{ + return ofs & ~logfs_super(sb)->s_segmask; +} + +static inline struct logfs_block *logfs_block(struct page *page) +{ + return (void *)page->private; +} + +static inline level_t shrink_level(gc_level_t __level) +{ + u8 level = (__force u8)__level; + + if (level >= LOGFS_MAX_LEVELS) + level -= LOGFS_MAX_LEVELS; + return (__force level_t)level; +} + +static inline gc_level_t expand_level(u64 ino, level_t __level) +{ + u8 level = (__force u8)__level; + + if (ino == LOGFS_INO_MASTER) { + /* ifile has seperate areas */ + level += LOGFS_MAX_LEVELS; + } + return (__force gc_level_t)level; +} + +static inline int logfs_block_shift(struct super_block *sb, level_t level) +{ + level = shrink_level((__force gc_level_t)level); + return (__force int)level * (sb->s_blocksize_bits - 3); +} + +static inline u64 logfs_block_mask(struct super_block *sb, level_t level) +{ + return ~0ull << logfs_block_shift(sb, level); +} + +static inline struct logfs_area *get_area(struct super_block *sb, + gc_level_t gc_level) +{ + return logfs_super(sb)->s_area[(__force u8)gc_level]; +} + +#endif diff --git a/fs/logfs/logfs_abi.h b/fs/logfs/logfs_abi.h new file mode 100644 index 000000000000..f674725663fe --- /dev/null +++ b/fs/logfs/logfs_abi.h @@ -0,0 +1,629 @@ +/* + * fs/logfs/logfs_abi.h + * + * As should be obvious for Linux kernel code, license is GPLv2 + * + * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> + * + * Public header for logfs. + */ +#ifndef FS_LOGFS_LOGFS_ABI_H +#define FS_LOGFS_LOGFS_ABI_H + +/* For out-of-kernel compiles */ +#ifndef BUILD_BUG_ON +#define BUILD_BUG_ON(condition) /**/ +#endif + +#define SIZE_CHECK(type, size) \ +static inline void check_##type(void) \ +{ \ + BUILD_BUG_ON(sizeof(struct type) != (size)); \ +} + +/* + * Throughout the logfs code, we're constantly dealing with blocks at + * various positions or offsets. To remove confusion, we stricly + * distinguish between a "position" - the logical position within a + * file and an "offset" - the physical location within the device. + * + * Any usage of the term offset for a logical location or position for + * a physical one is a bug and should get fixed. + */ + +/* + * Block are allocated in one of several segments depending on their + * level. The following levels are used: + * 0 - regular data block + * 1 - i1 indirect blocks + * 2 - i2 indirect blocks + * 3 - i3 indirect blocks + * 4 - i4 indirect blocks + * 5 - i5 indirect blocks + * 6 - ifile data blocks + * 7 - ifile i1 indirect blocks + * 8 - ifile i2 indirect blocks + * 9 - ifile i3 indirect blocks + * 10 - ifile i4 indirect blocks + * 11 - ifile i5 indirect blocks + * Potential levels to be used in the future: + * 12 - gc recycled blocks, long-lived data + * 13 - replacement blocks, short-lived data + * + * Levels 1-11 are necessary for robust gc operations and help seperate + * short-lived metadata from longer-lived file data. In the future, + * file data should get seperated into several segments based on simple + * heuristics. Old data recycled during gc operation is expected to be + * long-lived. New data is of uncertain life expectancy. New data + * used to replace older blocks in existing files is expected to be + * short-lived. + */ + + +/* Magic numbers. 64bit for superblock, 32bit for statfs f_type */ +#define LOGFS_MAGIC 0x7a3a8e5cb9d5bf67ull +#define LOGFS_MAGIC_U32 0xc97e8168u + +/* + * Various blocksize related macros. Blocksize is currently fixed at 4KiB. + * Sooner or later that should become configurable and the macros replaced + * by something superblock-dependent. Pointers in indirect blocks are and + * will remain 64bit. + * + * LOGFS_BLOCKSIZE - self-explaining + * LOGFS_BLOCK_FACTOR - number of pointers per indirect block + * LOGFS_BLOCK_BITS - log2 of LOGFS_BLOCK_FACTOR, used for shifts + */ +#define LOGFS_BLOCKSIZE (4096ull) +#define LOGFS_BLOCK_FACTOR (LOGFS_BLOCKSIZE / sizeof(u64)) +#define LOGFS_BLOCK_BITS (9) + +/* + * Number of blocks at various levels of indirection. There are 16 direct + * block pointers plus a single indirect pointer. + */ +#define I0_BLOCKS (16) +#define I1_BLOCKS LOGFS_BLOCK_FACTOR +#define I2_BLOCKS (LOGFS_BLOCK_FACTOR * I1_BLOCKS) +#define I3_BLOCKS (LOGFS_BLOCK_FACTOR * I2_BLOCKS) +#define I4_BLOCKS (LOGFS_BLOCK_FACTOR * I3_BLOCKS) +#define I5_BLOCKS (LOGFS_BLOCK_FACTOR * I4_BLOCKS) + +#define INDIRECT_INDEX I0_BLOCKS +#define LOGFS_EMBEDDED_FIELDS (I0_BLOCKS + 1) + +/* + * Sizes at which files require another level of indirection. Files smaller + * than LOGFS_EMBEDDED_SIZE can be completely stored in the inode itself, + * similar like ext2 fast symlinks. + * + * Data at a position smaller than LOGFS_I0_SIZE is accessed through the + * direct pointers, else through the 1x indirect pointer and so forth. + */ +#define LOGFS_EMBEDDED_SIZE (LOGFS_EMBEDDED_FIELDS * sizeof(u64)) +#define LOGFS_I0_SIZE (I0_BLOCKS * LOGFS_BLOCKSIZE) +#define LOGFS_I1_SIZE (I1_BLOCKS * LOGFS_BLOCKSIZE) +#define LOGFS_I2_SIZE (I2_BLOCKS * LOGFS_BLOCKSIZE) +#define LOGFS_I3_SIZE (I3_BLOCKS * LOGFS_BLOCKSIZE) +#define LOGFS_I4_SIZE (I4_BLOCKS * LOGFS_BLOCKSIZE) +#define LOGFS_I5_SIZE (I5_BLOCKS * LOGFS_BLOCKSIZE) + +/* + * Each indirect block pointer must have this flag set, if all block pointers + * behind it are set, i.e. there is no hole hidden in the shadow of this + * indirect block pointer. + */ +#define LOGFS_FULLY_POPULATED (1ULL << 63) +#define pure_ofs(ofs) (ofs & ~LOGFS_FULLY_POPULATED) + +/* + * LogFS needs to seperate data into levels. Each level is defined as the + * maximal possible distance from the master inode (inode of the inode file). + * Data blocks reside on level 0, 1x indirect block on level 1, etc. + * Inodes reside on level 6, indirect blocks for the inode file on levels 7-11. + * This effort is necessary to guarantee garbage collection to always make + * progress. + * + * LOGFS_MAX_INDIRECT is the maximal indirection through indirect blocks, + * LOGFS_MAX_LEVELS is one more for the actual data level of a file. It is + * the maximal number of levels for one file. + * LOGFS_NO_AREAS is twice that, as the inode file and regular files are + * effectively stacked on top of each other. + */ +#define LOGFS_MAX_INDIRECT (5) +#define LOGFS_MAX_LEVELS (LOGFS_MAX_INDIRECT + 1) +#define LOGFS_NO_AREAS (2 * LOGFS_MAX_LEVELS) + +/* Maximum size of filenames */ +#define LOGFS_MAX_NAMELEN (255) + +/* Number of segments in the primary journal. */ +#define LOGFS_JOURNAL_SEGS (16) + +/* Maximum number of free/erased/etc. segments in journal entries */ +#define MAX_CACHED_SEGS (64) + + +/* + * LOGFS_OBJECT_HEADERSIZE is the size of a single header in the object store, + * LOGFS_MAX_OBJECTSIZE the size of the largest possible object, including + * its header, + * LOGFS_SEGMENT_RESERVE is the amount of space reserved for each segment for + * its segment header and the padded space at the end when no further objects + * fit. + */ +#define LOGFS_OBJECT_HEADERSIZE (0x1c) +#define LOGFS_SEGMENT_HEADERSIZE (0x18) +#define LOGFS_MAX_OBJECTSIZE (LOGFS_OBJECT_HEADERSIZE + LOGFS_BLOCKSIZE) +#define LOGFS_SEGMENT_RESERVE \ + (LOGFS_SEGMENT_HEADERSIZE + LOGFS_MAX_OBJECTSIZE - 1) + +/* + * Segment types: + * SEG_SUPER - Data or indirect block + * SEG_JOURNAL - Inode + * SEG_OSTORE - Dentry + */ +enum { + SEG_SUPER = 0x01, + SEG_JOURNAL = 0x02, + SEG_OSTORE = 0x03, +}; + +/** + * struct logfs_segment_header - per-segment header in the ostore + * + * @crc: crc32 of header (there is no data) + * @pad: unused, must be 0 + * @type: segment type, see above + * @level: GC level for all objects in this segment + * @segno: segment number + * @ec: erase count for this segment + * @gec: global erase count at time of writing + */ +struct logfs_segment_header { + __be32 crc; + __be16 pad; + __u8 type; + __u8 level; + __be32 segno; + __be32 ec; + __be64 gec; +}; + +SIZE_CHECK(logfs_segment_header, LOGFS_SEGMENT_HEADERSIZE); + +#define LOGFS_FEATURES_INCOMPAT (0ull) +#define LOGFS_FEATURES_RO_COMPAT (0ull) +#define LOGFS_FEATURES_COMPAT (0ull) + +/** + * struct logfs_disk_super - on-medium superblock + * + * @ds_magic: magic number, must equal LOGFS_MAGIC + * @ds_crc: crc32 of structure starting with the next field + * @ds_ifile_levels: maximum number of levels for ifile + * @ds_iblock_levels: maximum number of levels for regular files + * @ds_data_levels: number of seperate levels for data + * @pad0: reserved, must be 0 + * @ds_feature_incompat: incompatible filesystem features + * @ds_feature_ro_compat: read-only compatible filesystem features + * @ds_feature_compat: compatible filesystem features + * @ds_flags: flags + * @ds_segment_shift: log2 of segment size + * @ds_block_shift: log2 of block size + * @ds_write_shift: log2 of write size + * @pad1: reserved, must be 0 + * @ds_journal_seg: segments used by primary journal + * @ds_root_reserve: bytes reserved for the superuser + * @ds_speed_reserve: bytes reserved to speed up GC + * @ds_bad_seg_reserve: number of segments reserved to handle bad blocks + * @pad2: reserved, must be 0 + * @pad3: reserved, must be 0 + * + * Contains only read-only fields. Read-write fields like the amount of used + * space is tracked in the dynamic superblock, which is stored in the journal. + */ +struct logfs_disk_super { + struct logfs_segment_header ds_sh; + __be64 ds_magic; + + __be32 ds_crc; + __u8 ds_ifile_levels; + __u8 ds_iblock_levels; + __u8 ds_data_levels; + __u8 ds_segment_shift; + __u8 ds_block_shift; + __u8 ds_write_shift; + __u8 pad0[6]; + + __be64 ds_filesystem_size; + __be32 ds_segment_size; + __be32 ds_bad_seg_reserve; + + __be64 ds_feature_incompat; + __be64 ds_feature_ro_compat; + + __be64 ds_feature_compat; + __be64 ds_feature_flags; + + __be64 ds_root_reserve; + __be64 ds_speed_reserve; + + __be32 ds_journal_seg[LOGFS_JOURNAL_SEGS]; + + __be64 ds_super_ofs[2]; + __be64 pad3[8]; +}; + +SIZE_CHECK(logfs_disk_super, 256); + +/* + * Object types: + * OBJ_BLOCK - Data or indirect block + * OBJ_INODE - Inode + * OBJ_DENTRY - Dentry + */ +enum { + OBJ_BLOCK = 0x04, + OBJ_INODE = 0x05, + OBJ_DENTRY = 0x06, +}; + +/** + * struct logfs_object_header - per-object header in the ostore + * + * @crc: crc32 of header, excluding data_crc + * @len: length of data + * @type: object type, see above + * @compr: compression type + * @ino: inode number + * @bix: block index + * @data_crc: crc32 of payload + */ +struct logfs_object_header { + __be32 crc; + __be16 len; + __u8 type; + __u8 compr; + __be64 ino; + __be64 bix; + __be32 data_crc; +} __attribute__((packed)); + +SIZE_CHECK(logfs_object_header, LOGFS_OBJECT_HEADERSIZE); + +/* + * Reserved inode numbers: + * LOGFS_INO_MASTER - master inode (for inode file) + * LOGFS_INO_ROOT - root directory + * LOGFS_INO_SEGFILE - per-segment used bytes and erase count + */ +enum { + LOGFS_INO_MAPPING = 0x00, + LOGFS_INO_MASTER = 0x01, + LOGFS_INO_ROOT = 0x02, + LOGFS_INO_SEGFILE = 0x03, + LOGFS_RESERVED_INOS = 0x10, +}; + +/* + * Inode flags. High bits should never be written to the medium. They are + * reserved for in-memory usage. + * Low bits should either remain in sync with the corresponding FS_*_FL or + * reuse slots that obviously don't make sense for logfs. + * + * LOGFS_IF_DIRTY Inode must be written back + * LOGFS_IF_ZOMBIE Inode has been deleted + * LOGFS_IF_STILLBORN -ENOSPC happened when creating inode + */ +#define LOGFS_IF_COMPRESSED 0x00000004 /* == FS_COMPR_FL */ +#define LOGFS_IF_DIRTY 0x20000000 +#define LOGFS_IF_ZOMBIE 0x40000000 +#define LOGFS_IF_STILLBORN 0x80000000 + +/* Flags available to chattr */ +#define LOGFS_FL_USER_VISIBLE (LOGFS_IF_COMPRESSED) +#define LOGFS_FL_USER_MODIFIABLE (LOGFS_IF_COMPRESSED) +/* Flags inherited from parent directory on file/directory creation */ +#define LOGFS_FL_INHERITED (LOGFS_IF_COMPRESSED) + +/** + * struct logfs_disk_inode - on-medium inode + * + * @di_mode: file mode + * @di_pad: reserved, must be 0 + * @di_flags: inode flags, see above + * @di_uid: user id + * @di_gid: group id + * @di_ctime: change time + * @di_mtime: modify time + * @di_refcount: reference count (aka nlink or link count) + * @di_generation: inode generation, for nfs + * @di_used_bytes: number of bytes used + * @di_size: file size + * @di_data: data pointers + */ +struct logfs_disk_inode { + __be16 di_mode; + __u8 di_height; + __u8 di_pad; + __be32 di_flags; + __be32 di_uid; + __be32 di_gid; + + __be64 di_ctime; + __be64 di_mtime; + + __be64 di_atime; + __be32 di_refcount; + __be32 di_generation; + + __be64 di_used_bytes; + __be64 di_size; + + __be64 di_data[LOGFS_EMBEDDED_FIELDS]; +}; + +SIZE_CHECK(logfs_disk_inode, 200); + +#define INODE_POINTER_OFS \ + (offsetof(struct logfs_disk_inode, di_data) / sizeof(__be64)) +#define INODE_USED_OFS \ + (offsetof(struct logfs_disk_inode, di_used_bytes) / sizeof(__be64)) +#define INODE_SIZE_OFS \ + (offsetof(struct logfs_disk_inode, di_size) / sizeof(__be64)) +#define INODE_HEIGHT_OFS (0) + +/** + * struct logfs_disk_dentry - on-medium dentry structure + * + * @ino: inode number + * @namelen: length of file name + * @type: file type, identical to bits 12..15 of mode + * @name: file name + */ +/* FIXME: add 6 bytes of padding to remove the __packed */ +struct logfs_disk_dentry { + __be64 ino; + __be16 namelen; + __u8 type; + __u8 name[LOGFS_MAX_NAMELEN]; +} __attribute__((packed)); + +SIZE_CHECK(logfs_disk_dentry, 266); + +#define RESERVED 0xffffffff +#define BADSEG 0xffffffff +/** + * struct logfs_segment_entry - segment file entry + * + * @ec_level: erase count and level + * @valid: number of valid bytes + * + * Segment file contains one entry for every segment. ec_level contains the + * erasecount in the upper 28 bits and the level in the lower 4 bits. An + * ec_level of BADSEG (-1) identifies bad segments. valid contains the number + * of valid bytes or RESERVED (-1 again) if the segment is used for either the + * superblock or the journal, or when the segment is bad. + */ +struct logfs_segment_entry { + __be32 ec_level; + __be32 valid; +}; + +SIZE_CHECK(logfs_segment_entry, 8); + +/** + * struct logfs_journal_header - header for journal entries (JEs) + * + * @h_crc: crc32 of journal entry + * @h_len: length of compressed journal entry, + * not including header + * @h_datalen: length of uncompressed data + * @h_type: JE type + * @h_compr: compression type + * @h_pad: reserved + */ +struct logfs_journal_header { + __be32 h_crc; + __be16 h_len; + __be16 h_datalen; + __be16 h_type; + __u8 h_compr; + __u8 h_pad[5]; +}; + +SIZE_CHECK(logfs_journal_header, 16); + +/* + * Life expectency of data. + * VIM_DEFAULT - default vim + * VIM_SEGFILE - for segment file only - very short-living + * VIM_GC - GC'd data - likely long-living + */ +enum logfs_vim { + VIM_DEFAULT = 0, + VIM_SEGFILE = 1, +}; + +/** + * struct logfs_je_area - wbuf header + * + * @segno: segment number of area + * @used_bytes: number of bytes already used + * @gc_level: GC level + * @vim: life expectancy of data + * + * "Areas" are segments currently being used for writing. There is at least + * one area per GC level. Several may be used to seperate long-living from + * short-living data. If an area with unknown vim is encountered, it can + * simply be closed. + * The write buffer immediately follow this header. + */ +struct logfs_je_area { + __be32 segno; + __be32 used_bytes; + __u8 gc_level; + __u8 vim; +} __attribute__((packed)); + +SIZE_CHECK(logfs_je_area, 10); + +#define MAX_JOURNAL_HEADER \ + (sizeof(struct logfs_journal_header) + sizeof(struct logfs_je_area)) + +/** + * struct logfs_je_dynsb - dynamic superblock + * + * @ds_gec: global erase count + * @ds_sweeper: current position of GC "sweeper" + * @ds_rename_dir: source directory ino (see dir.c documentation) + * @ds_rename_pos: position of source dd (see dir.c documentation) + * @ds_victim_ino: victims of incomplete dir operation (see dir.c) + * @ds_victim_ino: parent inode of victim (see dir.c) + * @ds_used_bytes: number of used bytes + */ +struct logfs_je_dynsb { + __be64 ds_gec; + __be64 ds_sweeper; + + __be64 ds_rename_dir; + __be64 ds_rename_pos; + + __be64 ds_victim_ino; + __be64 ds_victim_parent; /* XXX */ + + __be64 ds_used_bytes; + __be32 ds_generation; + __be32 pad; +}; + +SIZE_CHECK(logfs_je_dynsb, 64); + +/** + * struct logfs_je_anchor - anchor of filesystem tree, aka master inode + * + * @da_size: size of inode file + * @da_last_ino: last created inode + * @da_used_bytes: number of bytes used + * @da_data: data pointers + */ +struct logfs_je_anchor { + __be64 da_size; + __be64 da_last_ino; + + __be64 da_used_bytes; + u8 da_height; + u8 pad[7]; + + __be64 da_data[LOGFS_EMBEDDED_FIELDS]; +}; + +SIZE_CHECK(logfs_je_anchor, 168); + +/** + * struct logfs_je_spillout - spillout entry (from 1st to 2nd journal) + * + * @so_segment: segments used for 2nd journal + * + * Length of the array is given by h_len field in the header. + */ +struct logfs_je_spillout { + __be64 so_segment[0]; +}; + +SIZE_CHECK(logfs_je_spillout, 0); + +/** + * struct logfs_je_journal_ec - erase counts for all journal segments + * + * @ec: erase count + * + * Length of the array is given by h_len field in the header. + */ +struct logfs_je_journal_ec { + __be32 ec[0]; +}; + +SIZE_CHECK(logfs_je_journal_ec, 0); + +/** + * struct logfs_je_free_segments - list of free segmetns with erase count + */ +struct logfs_je_free_segments { + __be32 segno; + __be32 ec; +}; + +SIZE_CHECK(logfs_je_free_segments, 8); + +/** + * struct logfs_seg_alias - list of segment aliases + */ +struct logfs_seg_alias { + __be32 old_segno; + __be32 new_segno; +}; + +SIZE_CHECK(logfs_seg_alias, 8); + +/** + * struct logfs_obj_alias - list of object aliases + */ +struct logfs_obj_alias { + __be64 ino; + __be64 bix; + __be64 val; + u8 level; + u8 pad[5]; + __be16 child_no; +}; + +SIZE_CHECK(logfs_obj_alias, 32); + +/** + * Compression types. + * + * COMPR_NONE - uncompressed + * COMPR_ZLIB - compressed with zlib + */ +enum { + COMPR_NONE = 0, + COMPR_ZLIB = 1, +}; + +/* + * Journal entries come in groups of 16. First group contains unique + * entries, next groups contain one entry per level + * + * JE_FIRST - smallest possible journal entry number + * + * JEG_BASE - base group, containing unique entries + * JE_COMMIT - commit entry, validates all previous entries + * JE_DYNSB - dynamic superblock, anything that ought to be in the + * superblock but cannot because it is read-write data + * JE_ANCHOR - anchor aka master inode aka inode file's inode + * JE_ERASECOUNT erasecounts for all journal segments + * JE_SPILLOUT - unused + * JE_SEG_ALIAS - aliases segments + * JE_AREA - area description + * + * JE_LAST - largest possible journal entry number + */ +enum { + JE_FIRST = 0x01, + + JEG_BASE = 0x00, + JE_COMMIT = 0x02, + JE_DYNSB = 0x03, + JE_ANCHOR = 0x04, + JE_ERASECOUNT = 0x05, + JE_SPILLOUT = 0x06, + JE_OBJ_ALIAS = 0x0d, + JE_AREA = 0x0e, + + JE_LAST = 0x0e, +}; + +#endif diff --git a/fs/logfs/readwrite.c b/fs/logfs/readwrite.c new file mode 100644 index 000000000000..7a23b3e7c0a7 --- /dev/null +++ b/fs/logfs/readwrite.c @@ -0,0 +1,2246 @@ +/* + * fs/logfs/readwrite.c + * + * As should be obvious for Linux kernel code, license is GPLv2 + * + * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> + * + * + * Actually contains five sets of very similar functions: + * read read blocks from a file + * seek_hole find next hole + * seek_data find next data block + * valid check whether a block still belongs to a file + * write write blocks to a file + * delete delete a block (for directories and ifile) + * rewrite move existing blocks of a file to a new location (gc helper) + * truncate truncate a file + */ +#include "logfs.h" +#include <linux/sched.h> + +static u64 adjust_bix(u64 bix, level_t level) +{ + switch (level) { + case 0: + return bix; + case LEVEL(1): + return max_t(u64, bix, I0_BLOCKS); + case LEVEL(2): + return max_t(u64, bix, I1_BLOCKS); + case LEVEL(3): + return max_t(u64, bix, I2_BLOCKS); + case LEVEL(4): + return max_t(u64, bix, I3_BLOCKS); + case LEVEL(5): + return max_t(u64, bix, I4_BLOCKS); + default: + WARN_ON(1); + return bix; + } +} + +static inline u64 maxbix(u8 height) +{ + return 1ULL << (LOGFS_BLOCK_BITS * height); +} + +/** + * The inode address space is cut in two halves. Lower half belongs to data + * pages, upper half to indirect blocks. If the high bit (INDIRECT_BIT) is + * set, the actual block index (bix) and level can be derived from the page + * index. + * + * The lowest three bits of the block index are set to 0 after packing and + * unpacking. Since the lowest n bits (9 for 4KiB blocksize) are ignored + * anyway this is harmless. + */ +#define ARCH_SHIFT (BITS_PER_LONG - 32) +#define INDIRECT_BIT (0x80000000UL << ARCH_SHIFT) +#define LEVEL_SHIFT (28 + ARCH_SHIFT) +static inline pgoff_t first_indirect_block(void) +{ + return INDIRECT_BIT | (1ULL << LEVEL_SHIFT); +} + +pgoff_t logfs_pack_index(u64 bix, level_t level) +{ + pgoff_t index; + + BUG_ON(bix >= INDIRECT_BIT); + if (level == 0) + return bix; + + index = INDIRECT_BIT; + index |= (__force long)level << LEVEL_SHIFT; + index |= bix >> ((__force u8)level * LOGFS_BLOCK_BITS); + return index; +} + +void logfs_unpack_index(pgoff_t index, u64 *bix, level_t *level) +{ + u8 __level; + + if (!(index & INDIRECT_BIT)) { + *bix = index; + *level = 0; + return; + } + + __level = (index & ~INDIRECT_BIT) >> LEVEL_SHIFT; + *level = LEVEL(__level); + *bix = (index << (__level * LOGFS_BLOCK_BITS)) & ~INDIRECT_BIT; + *bix = adjust_bix(*bix, *level); + return; +} +#undef ARCH_SHIFT +#undef INDIRECT_BIT +#undef LEVEL_SHIFT + +/* + * Time is stored as nanoseconds since the epoch. + */ +static struct timespec be64_to_timespec(__be64 betime) +{ + return ns_to_timespec(be64_to_cpu(betime)); +} + +static __be64 timespec_to_be64(struct timespec tsp) +{ + return cpu_to_be64((u64)tsp.tv_sec * NSEC_PER_SEC + tsp.tv_nsec); +} + +static void logfs_disk_to_inode(struct logfs_disk_inode *di, struct inode*inode) +{ + struct logfs_inode *li = logfs_inode(inode); + int i; + + inode->i_mode = be16_to_cpu(di->di_mode); + li->li_height = di->di_height; + li->li_flags = be32_to_cpu(di->di_flags); + inode->i_uid = be32_to_cpu(di->di_uid); + inode->i_gid = be32_to_cpu(di->di_gid); + inode->i_size = be64_to_cpu(di->di_size); + logfs_set_blocks(inode, be64_to_cpu(di->di_used_bytes)); + inode->i_atime = be64_to_timespec(di->di_atime); + inode->i_ctime = be64_to_timespec(di->di_ctime); + inode->i_mtime = be64_to_timespec(di->di_mtime); + inode->i_nlink = be32_to_cpu(di->di_refcount); + inode->i_generation = be32_to_cpu(di->di_generation); + + switch (inode->i_mode & S_IFMT) { + case S_IFSOCK: /* fall through */ + case S_IFBLK: /* fall through */ + case S_IFCHR: /* fall through */ + case S_IFIFO: + inode->i_rdev = be64_to_cpu(di->di_data[0]); + break; + case S_IFDIR: /* fall through */ + case S_IFREG: /* fall through */ + case S_IFLNK: + for (i = 0; i < LOGFS_EMBEDDED_FIELDS; i++) + li->li_data[i] = be64_to_cpu(di->di_data[i]); + break; + default: + BUG(); + } +} + +static void logfs_inode_to_disk(struct inode *inode, struct logfs_disk_inode*di) +{ + struct logfs_inode *li = logfs_inode(inode); + int i; + + di->di_mode = cpu_to_be16(inode->i_mode); + di->di_height = li->li_height; + di->di_pad = 0; + di->di_flags = cpu_to_be32(li->li_flags); + di->di_uid = cpu_to_be32(inode->i_uid); + di->di_gid = cpu_to_be32(inode->i_gid); + di->di_size = cpu_to_be64(i_size_read(inode)); + di->di_used_bytes = cpu_to_be64(li->li_used_bytes); + di->di_atime = timespec_to_be64(inode->i_atime); + di->di_ctime = timespec_to_be64(inode->i_ctime); + di->di_mtime = timespec_to_be64(inode->i_mtime); + di->di_refcount = cpu_to_be32(inode->i_nlink); + di->di_generation = cpu_to_be32(inode->i_generation); + + switch (inode->i_mode & S_IFMT) { + case S_IFSOCK: /* fall through */ + case S_IFBLK: /* fall through */ + case S_IFCHR: /* fall through */ + case S_IFIFO: + di->di_data[0] = cpu_to_be64(inode->i_rdev); + break; + case S_IFDIR: /* fall through */ + case S_IFREG: /* fall through */ + case S_IFLNK: + for (i = 0; i < LOGFS_EMBEDDED_FIELDS; i++) + di->di_data[i] = cpu_to_be64(li->li_data[i]); + break; + default: + BUG(); + } +} + +static void __logfs_set_blocks(struct inode *inode) +{ + struct super_block *sb = inode->i_sb; + struct logfs_inode *li = logfs_inode(inode); + + inode->i_blocks = ULONG_MAX; + if (li->li_used_bytes >> sb->s_blocksize_bits < ULONG_MAX) + inode->i_blocks = ALIGN(li->li_used_bytes, 512) >> 9; +} + +void logfs_set_blocks(struct inode *inode, u64 bytes) +{ + struct logfs_inode *li = logfs_inode(inode); + + li->li_used_bytes = bytes; + __logfs_set_blocks(inode); +} + +static void prelock_page(struct super_block *sb, struct page *page, int lock) +{ + struct logfs_super *super = logfs_super(sb); + + BUG_ON(!PageLocked(page)); + if (lock) { + BUG_ON(PagePreLocked(page)); + SetPagePreLocked(page); + } else { + /* We are in GC path. */ + if (PagePreLocked(page)) + super->s_lock_count++; + else + SetPagePreLocked(page); + } +} + +static void preunlock_page(struct super_block *sb, struct page *page, int lock) +{ + struct logfs_super *super = logfs_super(sb); + + BUG_ON(!PageLocked(page)); + if (lock) + ClearPagePreLocked(page); + else { + /* We are in GC path. */ + BUG_ON(!PagePreLocked(page)); + if (super->s_lock_count) + super->s_lock_count--; + else + ClearPagePreLocked(page); + } +} + +/* + * Logfs is prone to an AB-BA deadlock where one task tries to acquire + * s_write_mutex with a locked page and GC tries to get that page while holding + * s_write_mutex. + * To solve this issue logfs will ignore the page lock iff the page in question + * is waiting for s_write_mutex. We annotate this fact by setting PG_pre_locked + * in addition to PG_locked. + */ +static void logfs_get_wblocks(struct super_block *sb, struct page *page, + int lock) +{ + struct logfs_super *super = logfs_super(sb); + + if (page) + prelock_page(sb, page, lock); + + if (lock) { + mutex_lock(&super->s_write_mutex); + logfs_gc_pass(sb); + /* FIXME: We also have to check for shadowed space + * and mempool fill grade */ + } +} + +static void logfs_put_wblocks(struct super_block *sb, struct page *page, + int lock) +{ + struct logfs_super *super = logfs_super(sb); + + if (page) + preunlock_page(sb, page, lock); + /* Order matters - we must clear PG_pre_locked before releasing + * s_write_mutex or we could race against another task. */ + if (lock) + mutex_unlock(&super->s_write_mutex); +} + +static struct page *logfs_get_read_page(struct inode *inode, u64 bix, + level_t level) +{ + return find_or_create_page(inode->i_mapping, + logfs_pack_index(bix, level), GFP_NOFS); +} + +static void logfs_put_read_page(struct page *page) +{ + unlock_page(page); + page_cache_release(page); +} + +static void logfs_lock_write_page(struct page *page) +{ + int loop = 0; + + while (unlikely(!trylock_page(page))) { + if (loop++ > 0x1000) { + /* Has been observed once so far... */ + printk(KERN_ERR "stack at %p\n", &loop); + BUG(); + } + if (PagePreLocked(page)) { + /* Holder of page lock is waiting for us, it + * is safe to use this page. */ + break; + } + /* Some other process has this page locked and has + * nothing to do with us. Wait for it to finish. + */ + schedule(); + } + BUG_ON(!PageLocked(page)); +} + +static struct page *logfs_get_write_page(struct inode *inode, u64 bix, + level_t level) +{ + struct address_space *mapping = inode->i_mapping; + pgoff_t index = logfs_pack_index(bix, level); + struct page *page; + int err; + +repeat: + page = find_get_page(mapping, index); + if (!page) { + page = __page_cache_alloc(GFP_NOFS); + if (!page) + return NULL; + err = add_to_page_cache_lru(page, mapping, index, GFP_NOFS); + if (unlikely(err)) { + page_cache_release(page); + if (err == -EEXIST) + goto repeat; + return NULL; + } + } else logfs_lock_write_page(page); + BUG_ON(!PageLocked(page)); + return page; +} + +static void logfs_unlock_write_page(struct page *page) +{ + if (!PagePreLocked(page)) + unlock_page(page); +} + +static void logfs_put_write_page(struct page *page) +{ + logfs_unlock_write_page(page); + page_cache_release(page); +} + +static struct page *logfs_get_page(struct inode *inode, u64 bix, level_t level, + int rw) +{ + if (rw == READ) + return logfs_get_read_page(inode, bix, level); + else + return logfs_get_write_page(inode, bix, level); +} + +static void logfs_put_page(struct page *page, int rw) +{ + if (rw == READ) + logfs_put_read_page(page); + else + logfs_put_write_page(page); +} + +static unsigned long __get_bits(u64 val, int skip, int no) +{ + u64 ret = val; + + ret >>= skip * no; + ret <<= 64 - no; + ret >>= 64 - no; + return ret; +} + +static unsigned long get_bits(u64 val, level_t skip) +{ + return __get_bits(val, (__force int)skip, LOGFS_BLOCK_BITS); +} + +static inline void init_shadow_tree(struct super_block *sb, + struct shadow_tree *tree) +{ + struct logfs_super *super = logfs_super(sb); + + btree_init_mempool64(&tree->new, super->s_btree_pool); + btree_init_mempool64(&tree->old, super->s_btree_pool); +} + +static void indirect_write_block(struct logfs_block *block) +{ + struct page *page; + struct inode *inode; + int ret; + + page = block->page; + inode = page->mapping->host; + logfs_lock_write_page(page); + ret = logfs_write_buf(inode, page, 0); + logfs_unlock_write_page(page); + /* + * This needs some rework. Unless you want your filesystem to run + * completely synchronously (you don't), the filesystem will always + * report writes as 'successful' before the actual work has been + * done. The actual work gets done here and this is where any errors + * will show up. And there isn't much we can do about it, really. + * + * Some attempts to fix the errors (move from bad blocks, retry io,...) + * have already been done, so anything left should be either a broken + * device or a bug somewhere in logfs itself. Being relatively new, + * the odds currently favor a bug, so for now the line below isn't + * entirely tasteles. + */ + BUG_ON(ret); +} + +static void inode_write_block(struct logfs_block *block) +{ + struct inode *inode; + int ret; + + inode = block->inode; + if (inode->i_ino == LOGFS_INO_MASTER) + logfs_write_anchor(inode->i_sb); + else { + ret = __logfs_write_inode(inode, 0); + /* see indirect_write_block comment */ + BUG_ON(ret); + } +} + +static gc_level_t inode_block_level(struct logfs_block *block) +{ + BUG_ON(block->inode->i_ino == LOGFS_INO_MASTER); + return GC_LEVEL(LOGFS_MAX_LEVELS); +} + +static gc_level_t indirect_block_level(struct logfs_block *block) +{ + struct page *page; + struct inode *inode; + u64 bix; + level_t level; + + page = block->page; + inode = page->mapping->host; + logfs_unpack_index(page->index, &bix, &level); + return expand_level(inode->i_ino, level); +} + +/* + * This silences a false, yet annoying gcc warning. I hate it when my editor + * jumps into bitops.h each time I recompile this file. + * TODO: Complain to gcc folks about this and upgrade compiler. + */ +static unsigned long fnb(const unsigned long *addr, + unsigned long size, unsigned long offset) +{ + return find_next_bit(addr, size, offset); +} + +static __be64 inode_val0(struct inode *inode) +{ + struct logfs_inode *li = logfs_inode(inode); + u64 val; + + /* + * Explicit shifting generates good code, but must match the format + * of the structure. Add some paranoia just in case. + */ + BUILD_BUG_ON(offsetof(struct logfs_disk_inode, di_mode) != 0); + BUILD_BUG_ON(offsetof(struct logfs_disk_inode, di_height) != 2); + BUILD_BUG_ON(offsetof(struct logfs_disk_inode, di_flags) != 4); + + val = (u64)inode->i_mode << 48 | + (u64)li->li_height << 40 | + (u64)li->li_flags; + return cpu_to_be64(val); +} + +static int inode_write_alias(struct super_block *sb, + struct logfs_block *block, write_alias_t *write_one_alias) +{ + struct inode *inode = block->inode; + struct logfs_inode *li = logfs_inode(inode); + unsigned long pos; + u64 ino , bix; + __be64 val; + level_t level; + int err; + + for (pos = 0; ; pos++) { + pos = fnb(block->alias_map, LOGFS_BLOCK_FACTOR, pos); + if (pos >= LOGFS_EMBEDDED_FIELDS + INODE_POINTER_OFS) + return 0; + + switch (pos) { + case INODE_HEIGHT_OFS: + val = inode_val0(inode); + break; + case INODE_USED_OFS: + val = cpu_to_be64(li->li_used_bytes);; + break; + case INODE_SIZE_OFS: + val = cpu_to_be64(i_size_read(inode)); + break; + case INODE_POINTER_OFS ... INODE_POINTER_OFS + LOGFS_EMBEDDED_FIELDS - 1: + val = cpu_to_be64(li->li_data[pos - INODE_POINTER_OFS]); + break; + default: + BUG(); + } + + ino = LOGFS_INO_MASTER; + bix = inode->i_ino; + level = LEVEL(0); + err = write_one_alias(sb, ino, bix, level, pos, val); + if (err) + return err; + } +} + +static int indirect_write_alias(struct super_block *sb, + struct logfs_block *block, write_alias_t *write_one_alias) +{ + unsigned long pos; + struct page *page = block->page; + u64 ino , bix; + __be64 *child, val; + level_t level; + int err; + + for (pos = 0; ; pos++) { + pos = fnb(block->alias_map, LOGFS_BLOCK_FACTOR, pos); + if (pos >= LOGFS_BLOCK_FACTOR) + return 0; + + ino = page->mapping->host->i_ino; + logfs_unpack_index(page->index, &bix, &level); + child = kmap_atomic(page, KM_USER0); + val = child[pos]; + kunmap_atomic(child, KM_USER0); + err = write_one_alias(sb, ino, bix, level, pos, val); + if (err) + return err; + } +} + +int logfs_write_obj_aliases_pagecache(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_block *block; + int err; + + list_for_each_entry(block, &super->s_object_alias, alias_list) { + err = block->ops->write_alias(sb, block, write_alias_journal); + if (err) + return err; + } + return 0; +} + +void __free_block(struct super_block *sb, struct logfs_block *block) +{ + BUG_ON(!list_empty(&block->item_list)); + list_del(&block->alias_list); + mempool_free(block, logfs_super(sb)->s_block_pool); +} + +static void inode_free_block(struct super_block *sb, struct logfs_block *block) +{ + struct inode *inode = block->inode; + + logfs_inode(inode)->li_block = NULL; + __free_block(sb, block); +} + +static void indirect_free_block(struct super_block *sb, + struct logfs_block *block) +{ + ClearPagePrivate(block->page); + block->page->private = 0; + __free_block(sb, block); +} + + +static struct logfs_block_ops inode_block_ops = { + .write_block = inode_write_block, + .block_level = inode_block_level, + .free_block = inode_free_block, + .write_alias = inode_write_alias, +}; + +struct logfs_block_ops indirect_block_ops = { + .write_block = indirect_write_block, + .block_level = indirect_block_level, + .free_block = indirect_free_block, + .write_alias = indirect_write_alias, +}; + +struct logfs_block *__alloc_block(struct super_block *sb, + u64 ino, u64 bix, level_t level) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_block *block; + + block = mempool_alloc(super->s_block_pool, GFP_NOFS); + memset(block, 0, sizeof(*block)); + INIT_LIST_HEAD(&block->alias_list); + INIT_LIST_HEAD(&block->item_list); + block->sb = sb; + block->ino = ino; + block->bix = bix; + block->level = level; + return block; +} + +static void alloc_inode_block(struct inode *inode) +{ + struct logfs_inode *li = logfs_inode(inode); + struct logfs_block *block; + + if (li->li_block) + return; + + block = __alloc_block(inode->i_sb, LOGFS_INO_MASTER, inode->i_ino, 0); + block->inode = inode; + li->li_block = block; + block->ops = &inode_block_ops; +} + +void initialize_block_counters(struct page *page, struct logfs_block *block, + __be64 *array, int page_is_empty) +{ + u64 ptr; + int i, start; + + block->partial = 0; + block->full = 0; + start = 0; + if (page->index < first_indirect_block()) { + /* Counters are pointless on level 0 */ + return; + } + if (page->index == first_indirect_block()) { + /* Skip unused pointers */ + start = I0_BLOCKS; + block->full = I0_BLOCKS; + } + if (!page_is_empty) { + for (i = start; i < LOGFS_BLOCK_FACTOR; i++) { + ptr = be64_to_cpu(array[i]); + if (ptr) + block->partial++; + if (ptr & LOGFS_FULLY_POPULATED) + block->full++; + } + } +} + +static void alloc_data_block(struct inode *inode, struct page *page) +{ + struct logfs_block *block; + u64 bix; + level_t level; + + if (PagePrivate(page)) + return; + + logfs_unpack_index(page->index, &bix, &level); + block = __alloc_block(inode->i_sb, inode->i_ino, bix, level); + block->page = page; + SetPagePrivate(page); + page->private = (unsigned long)block; + block->ops = &indirect_block_ops; +} + +static void alloc_indirect_block(struct inode *inode, struct page *page, + int page_is_empty) +{ + struct logfs_block *block; + __be64 *array; + + if (PagePrivate(page)) + return; + + alloc_data_block(inode, page); + + block = logfs_block(page); + array = kmap_atomic(page, KM_USER0); + initialize_block_counters(page, block, array, page_is_empty); + kunmap_atomic(array, KM_USER0); +} + +static void block_set_pointer(struct page *page, int index, u64 ptr) +{ + struct logfs_block *block = logfs_block(page); + __be64 *array; + u64 oldptr; + + BUG_ON(!block); + array = kmap_atomic(page, KM_USER0); + oldptr = be64_to_cpu(array[index]); + array[index] = cpu_to_be64(ptr); + kunmap_atomic(array, KM_USER0); + SetPageUptodate(page); + + block->full += !!(ptr & LOGFS_FULLY_POPULATED) + - !!(oldptr & LOGFS_FULLY_POPULATED); + block->partial += !!ptr - !!oldptr; +} + +static u64 block_get_pointer(struct page *page, int index) +{ + __be64 *block; + u64 ptr; + + block = kmap_atomic(page, KM_USER0); + ptr = be64_to_cpu(block[index]); + kunmap_atomic(block, KM_USER0); + return ptr; +} + +static int logfs_read_empty(struct page *page) +{ + zero_user_segment(page, 0, PAGE_CACHE_SIZE); + return 0; +} + +static int logfs_read_direct(struct inode *inode, struct page *page) +{ + struct logfs_inode *li = logfs_inode(inode); + pgoff_t index = page->index; + u64 block; + + block = li->li_data[index]; + if (!block) + return logfs_read_empty(page); + + return logfs_segment_read(inode, page, block, index, 0); +} + +static int logfs_read_loop(struct inode *inode, struct page *page, + int rw_context) +{ + struct logfs_inode *li = logfs_inode(inode); + u64 bix, bofs = li->li_data[INDIRECT_INDEX]; + level_t level, target_level; + int ret; + struct page *ipage; + + logfs_unpack_index(page->index, &bix, &target_level); + if (!bofs) + return logfs_read_empty(page); + + if (bix >= maxbix(li->li_height)) + return logfs_read_empty(page); + + for (level = LEVEL(li->li_height); + (__force u8)level > (__force u8)target_level; + level = SUBLEVEL(level)){ + ipage = logfs_get_page(inode, bix, level, rw_context); + if (!ipage) + return -ENOMEM; + + ret = logfs_segment_read(inode, ipage, bofs, bix, level); + if (ret) { + logfs_put_read_page(ipage); + return ret; + } + + bofs = block_get_pointer(ipage, get_bits(bix, SUBLEVEL(level))); + logfs_put_page(ipage, rw_context); + if (!bofs) + return logfs_read_empty(page); + } + + return logfs_segment_read(inode, page, bofs, bix, 0); +} + +static int logfs_read_block(struct inode *inode, struct page *page, + int rw_context) +{ + pgoff_t index = page->index; + + if (index < I0_BLOCKS) + return logfs_read_direct(inode, page); + return logfs_read_loop(inode, page, rw_context); +} + +static int logfs_exist_loop(struct inode *inode, u64 bix) +{ + struct logfs_inode *li = logfs_inode(inode); + u64 bofs = li->li_data[INDIRECT_INDEX]; + level_t level; + int ret; + struct page *ipage; + + if (!bofs) + return 0; + if (bix >= maxbix(li->li_height)) + return 0; + + for (level = LEVEL(li->li_height); level != 0; level = SUBLEVEL(level)) { + ipage = logfs_get_read_page(inode, bix, level); + if (!ipage) + return -ENOMEM; + + ret = logfs_segment_read(inode, ipage, bofs, bix, level); + if (ret) { + logfs_put_read_page(ipage); + return ret; + } + + bofs = block_get_pointer(ipage, get_bits(bix, SUBLEVEL(level))); + logfs_put_read_page(ipage); + if (!bofs) + return 0; + } + + return 1; +} + +int logfs_exist_block(struct inode *inode, u64 bix) +{ + struct logfs_inode *li = logfs_inode(inode); + + if (bix < I0_BLOCKS) + return !!li->li_data[bix]; + return logfs_exist_loop(inode, bix); +} + +static u64 seek_holedata_direct(struct inode *inode, u64 bix, int data) +{ + struct logfs_inode *li = logfs_inode(inode); + + for (; bix < I0_BLOCKS; bix++) + if (data ^ (li->li_data[bix] == 0)) + return bix; + return I0_BLOCKS; +} + +static u64 seek_holedata_loop(struct inode *inode, u64 bix, int data) +{ + struct logfs_inode *li = logfs_inode(inode); + __be64 *rblock; + u64 increment, bofs = li->li_data[INDIRECT_INDEX]; + level_t level; + int ret, slot; + struct page *page; + + BUG_ON(!bofs); + + for (level = LEVEL(li->li_height); level != 0; level = SUBLEVEL(level)) { + increment = 1 << (LOGFS_BLOCK_BITS * ((__force u8)level-1)); + page = logfs_get_read_page(inode, bix, level); + if (!page) + return bix; + + ret = logfs_segment_read(inode, page, bofs, bix, level); + if (ret) { + logfs_put_read_page(page); + return bix; + } + + slot = get_bits(bix, SUBLEVEL(level)); + rblock = kmap_atomic(page, KM_USER0); + while (slot < LOGFS_BLOCK_FACTOR) { + if (data && (rblock[slot] != 0)) + break; + if (!data && !(be64_to_cpu(rblock[slot]) & LOGFS_FULLY_POPULATED)) + break; + slot++; + bix += increment; + bix &= ~(increment - 1); + } + if (slot >= LOGFS_BLOCK_FACTOR) { + kunmap_atomic(rblock, KM_USER0); + logfs_put_read_page(page); + return bix; + } + bofs = be64_to_cpu(rblock[slot]); + kunmap_atomic(rblock, KM_USER0); + logfs_put_read_page(page); + if (!bofs) { + BUG_ON(data); + return bix; + } + } + return bix; +} + +/** + * logfs_seek_hole - find next hole starting at a given block index + * @inode: inode to search in + * @bix: block index to start searching + * + * Returns next hole. If the file doesn't contain any further holes, the + * block address next to eof is returned instead. + */ +u64 logfs_seek_hole(struct inode *inode, u64 bix) +{ + struct logfs_inode *li = logfs_inode(inode); + + if (bix < I0_BLOCKS) { + bix = seek_holedata_direct(inode, bix, 0); + if (bix < I0_BLOCKS) + return bix; + } + + if (!li->li_data[INDIRECT_INDEX]) + return bix; + else if (li->li_data[INDIRECT_INDEX] & LOGFS_FULLY_POPULATED) + bix = maxbix(li->li_height); + else { + bix = seek_holedata_loop(inode, bix, 0); + if (bix < maxbix(li->li_height)) + return bix; + /* Should not happen anymore. But if some port writes semi- + * corrupt images (as this one used to) we might run into it. + */ + WARN_ON_ONCE(bix == maxbix(li->li_height)); + } + + return bix; +} + +static u64 __logfs_seek_data(struct inode *inode, u64 bix) +{ + struct logfs_inode *li = logfs_inode(inode); + + if (bix < I0_BLOCKS) { + bix = seek_holedata_direct(inode, bix, 1); + if (bix < I0_BLOCKS) + return bix; + } + + if (bix < maxbix(li->li_height)) { + if (!li->li_data[INDIRECT_INDEX]) + bix = maxbix(li->li_height); + else + return seek_holedata_loop(inode, bix, 1); + } + + return bix; +} + +/** + * logfs_seek_data - find next data block after a given block index + * @inode: inode to search in + * @bix: block index to start searching + * + * Returns next data block. If the file doesn't contain any further data + * blocks, the last block in the file is returned instead. + */ +u64 logfs_seek_data(struct inode *inode, u64 bix) +{ + struct super_block *sb = inode->i_sb; + u64 ret, end; + + ret = __logfs_seek_data(inode, bix); + end = i_size_read(inode) >> sb->s_blocksize_bits; + if (ret >= end) + ret = max(bix, end); + return ret; +} + +static int logfs_is_valid_direct(struct logfs_inode *li, u64 bix, u64 ofs) +{ + return pure_ofs(li->li_data[bix]) == ofs; +} + +static int __logfs_is_valid_loop(struct inode *inode, u64 bix, + u64 ofs, u64 bofs) +{ + struct logfs_inode *li = logfs_inode(inode); + level_t level; + int ret; + struct page *page; + + for (level = LEVEL(li->li_height); level != 0; level = SUBLEVEL(level)){ + page = logfs_get_write_page(inode, bix, level); + BUG_ON(!page); + + ret = logfs_segment_read(inode, page, bofs, bix, level); + if (ret) { + logfs_put_write_page(page); + return 0; + } + + bofs = block_get_pointer(page, get_bits(bix, SUBLEVEL(level))); + logfs_put_write_page(page); + if (!bofs) + return 0; + + if (pure_ofs(bofs) == ofs) + return 1; + } + return 0; +} + +static int logfs_is_valid_loop(struct inode *inode, u64 bix, u64 ofs) +{ + struct logfs_inode *li = logfs_inode(inode); + u64 bofs = li->li_data[INDIRECT_INDEX]; + + if (!bofs) + return 0; + + if (bix >= maxbix(li->li_height)) + return 0; + + if (pure_ofs(bofs) == ofs) + return 1; + + return __logfs_is_valid_loop(inode, bix, ofs, bofs); +} + +static int __logfs_is_valid_block(struct inode *inode, u64 bix, u64 ofs) +{ + struct logfs_inode *li = logfs_inode(inode); + + if ((inode->i_nlink == 0) && atomic_read(&inode->i_count) == 1) + return 0; + + if (bix < I0_BLOCKS) + return logfs_is_valid_direct(li, bix, ofs); + return logfs_is_valid_loop(inode, bix, ofs); +} + +/** + * logfs_is_valid_block - check whether this block is still valid + * + * @sb - superblock + * @ofs - block physical offset + * @ino - block inode number + * @bix - block index + * @level - block level + * + * Returns 0 if the block is invalid, 1 if it is valid and 2 if it will + * become invalid once the journal is written. + */ +int logfs_is_valid_block(struct super_block *sb, u64 ofs, u64 ino, u64 bix, + gc_level_t gc_level) +{ + struct logfs_super *super = logfs_super(sb); + struct inode *inode; + int ret, cookie; + + /* Umount closes a segment with free blocks remaining. Those + * blocks are by definition invalid. */ + if (ino == -1) + return 0; + + LOGFS_BUG_ON((u64)(u_long)ino != ino, sb); + + inode = logfs_safe_iget(sb, ino, &cookie); + if (IS_ERR(inode)) + goto invalid; + + ret = __logfs_is_valid_block(inode, bix, ofs); + logfs_safe_iput(inode, cookie); + if (ret) + return ret; + +invalid: + /* Block is nominally invalid, but may still sit in the shadow tree, + * waiting for a journal commit. + */ + if (btree_lookup64(&super->s_shadow_tree.old, ofs)) + return 2; + return 0; +} + +int logfs_readpage_nolock(struct page *page) +{ + struct inode *inode = page->mapping->host; + int ret = -EIO; + + ret = logfs_read_block(inode, page, READ); + + if (ret) { + ClearPageUptodate(page); + SetPageError(page); + } else { + SetPageUptodate(page); + ClearPageError(page); + } + flush_dcache_page(page); + + return ret; +} + +static int logfs_reserve_bytes(struct inode *inode, int bytes) +{ + struct logfs_super *super = logfs_super(inode->i_sb); + u64 available = super->s_free_bytes + super->s_dirty_free_bytes + - super->s_dirty_used_bytes - super->s_dirty_pages; + + if (!bytes) + return 0; + + if (available < bytes) + return -ENOSPC; + + if (available < bytes + super->s_root_reserve && + !capable(CAP_SYS_RESOURCE)) + return -ENOSPC; + + return 0; +} + +int get_page_reserve(struct inode *inode, struct page *page) +{ + struct logfs_super *super = logfs_super(inode->i_sb); + int ret; + + if (logfs_block(page) && logfs_block(page)->reserved_bytes) + return 0; + + logfs_get_wblocks(inode->i_sb, page, WF_LOCK); + ret = logfs_reserve_bytes(inode, 6 * LOGFS_MAX_OBJECTSIZE); + if (!ret) { + alloc_data_block(inode, page); + logfs_block(page)->reserved_bytes += 6 * LOGFS_MAX_OBJECTSIZE; + super->s_dirty_pages += 6 * LOGFS_MAX_OBJECTSIZE; + } + logfs_put_wblocks(inode->i_sb, page, WF_LOCK); + return ret; +} + +/* + * We are protected by write lock. Push victims up to superblock level + * and release transaction when appropriate. + */ +/* FIXME: This is currently called from the wrong spots. */ +static void logfs_handle_transaction(struct inode *inode, + struct logfs_transaction *ta) +{ + struct logfs_super *super = logfs_super(inode->i_sb); + + if (!ta) + return; + logfs_inode(inode)->li_block->ta = NULL; + + if (inode->i_ino != LOGFS_INO_MASTER) { + BUG(); /* FIXME: Yes, this needs more thought */ + /* just remember the transaction until inode is written */ + //BUG_ON(logfs_inode(inode)->li_transaction); + //logfs_inode(inode)->li_transaction = ta; + return; + } + + switch (ta->state) { + case CREATE_1: /* fall through */ + case UNLINK_1: + BUG_ON(super->s_victim_ino); + super->s_victim_ino = ta->ino; + break; + case CREATE_2: /* fall through */ + case UNLINK_2: + BUG_ON(super->s_victim_ino != ta->ino); + super->s_victim_ino = 0; + /* transaction ends here - free it */ + kfree(ta); + break; + case CROSS_RENAME_1: + BUG_ON(super->s_rename_dir); + BUG_ON(super->s_rename_pos); + super->s_rename_dir = ta->dir; + super->s_rename_pos = ta->pos; + break; + case CROSS_RENAME_2: + BUG_ON(super->s_rename_dir != ta->dir); + BUG_ON(super->s_rename_pos != ta->pos); + super->s_rename_dir = 0; + super->s_rename_pos = 0; + kfree(ta); + break; + case TARGET_RENAME_1: + BUG_ON(super->s_rename_dir); + BUG_ON(super->s_rename_pos); + BUG_ON(super->s_victim_ino); + super->s_rename_dir = ta->dir; + super->s_rename_pos = ta->pos; + super->s_victim_ino = ta->ino; + break; + case TARGET_RENAME_2: + BUG_ON(super->s_rename_dir != ta->dir); + BUG_ON(super->s_rename_pos != ta->pos); + BUG_ON(super->s_victim_ino != ta->ino); + super->s_rename_dir = 0; + super->s_rename_pos = 0; + break; + case TARGET_RENAME_3: + BUG_ON(super->s_rename_dir); + BUG_ON(super->s_rename_pos); + BUG_ON(super->s_victim_ino != ta->ino); + super->s_victim_ino = 0; + kfree(ta); + break; + default: + BUG(); + } +} + +/* + * Not strictly a reservation, but rather a check that we still have enough + * space to satisfy the write. + */ +static int logfs_reserve_blocks(struct inode *inode, int blocks) +{ + return logfs_reserve_bytes(inode, blocks * LOGFS_MAX_OBJECTSIZE); +} + +struct write_control { + u64 ofs; + long flags; +}; + +static struct logfs_shadow *alloc_shadow(struct inode *inode, u64 bix, + level_t level, u64 old_ofs) +{ + struct logfs_super *super = logfs_super(inode->i_sb); + struct logfs_shadow *shadow; + + shadow = mempool_alloc(super->s_shadow_pool, GFP_NOFS); + memset(shadow, 0, sizeof(*shadow)); + shadow->ino = inode->i_ino; + shadow->bix = bix; + shadow->gc_level = expand_level(inode->i_ino, level); + shadow->old_ofs = old_ofs & ~LOGFS_FULLY_POPULATED; + return shadow; +} + +static void free_shadow(struct inode *inode, struct logfs_shadow *shadow) +{ + struct logfs_super *super = logfs_super(inode->i_sb); + + mempool_free(shadow, super->s_shadow_pool); +} + +/** + * fill_shadow_tree - Propagate shadow tree changes due to a write + * @inode: Inode owning the page + * @page: Struct page that was written + * @shadow: Shadow for the current write + * + * Writes in logfs can result in two semi-valid objects. The old object + * is still valid as long as it can be reached by following pointers on + * the medium. Only when writes propagate all the way up to the journal + * has the new object safely replaced the old one. + * + * To handle this problem, a struct logfs_shadow is used to represent + * every single write. It is attached to the indirect block, which is + * marked dirty. When the indirect block is written, its shadows are + * handed up to the next indirect block (or inode). Untimately they + * will reach the master inode and be freed upon journal commit. + * + * This function handles a single step in the propagation. It adds the + * shadow for the current write to the tree, along with any shadows in + * the page's tree, in case it was an indirect block. If a page is + * written, the inode parameter is left NULL, if an inode is written, + * the page parameter is left NULL. + */ +static void fill_shadow_tree(struct inode *inode, struct page *page, + struct logfs_shadow *shadow) +{ + struct logfs_super *super = logfs_super(inode->i_sb); + struct logfs_block *block = logfs_block(page); + struct shadow_tree *tree = &super->s_shadow_tree; + + if (PagePrivate(page)) { + if (block->alias_map) + super->s_no_object_aliases -= bitmap_weight( + block->alias_map, LOGFS_BLOCK_FACTOR); + logfs_handle_transaction(inode, block->ta); + block->ops->free_block(inode->i_sb, block); + } + if (shadow) { + if (shadow->old_ofs) + btree_insert64(&tree->old, shadow->old_ofs, shadow, + GFP_NOFS); + else + btree_insert64(&tree->new, shadow->new_ofs, shadow, + GFP_NOFS); + + super->s_dirty_used_bytes += shadow->new_len; + super->s_dirty_free_bytes += shadow->old_len; + } +} + +static void logfs_set_alias(struct super_block *sb, struct logfs_block *block, + long child_no) +{ + struct logfs_super *super = logfs_super(sb); + + if (block->inode && block->inode->i_ino == LOGFS_INO_MASTER) { + /* Aliases in the master inode are pointless. */ + return; + } + + if (!test_bit(child_no, block->alias_map)) { + set_bit(child_no, block->alias_map); + super->s_no_object_aliases++; + } + list_move_tail(&block->alias_list, &super->s_object_alias); +} + +/* + * Object aliases can and often do change the size and occupied space of a + * file. So not only do we have to change the pointers, we also have to + * change inode->i_size and li->li_used_bytes. Which is done by setting + * another two object aliases for the inode itself. + */ +static void set_iused(struct inode *inode, struct logfs_shadow *shadow) +{ + struct logfs_inode *li = logfs_inode(inode); + + if (shadow->new_len == shadow->old_len) + return; + + alloc_inode_block(inode); + li->li_used_bytes += shadow->new_len - shadow->old_len; + __logfs_set_blocks(inode); + logfs_set_alias(inode->i_sb, li->li_block, INODE_USED_OFS); + logfs_set_alias(inode->i_sb, li->li_block, INODE_SIZE_OFS); +} + +static int logfs_write_i0(struct inode *inode, struct page *page, + struct write_control *wc) +{ + struct logfs_shadow *shadow; + u64 bix; + level_t level; + int full, err = 0; + + logfs_unpack_index(page->index, &bix, &level); + if (wc->ofs == 0) + if (logfs_reserve_blocks(inode, 1)) + return -ENOSPC; + + shadow = alloc_shadow(inode, bix, level, wc->ofs); + if (wc->flags & WF_WRITE) + err = logfs_segment_write(inode, page, shadow); + if (wc->flags & WF_DELETE) + logfs_segment_delete(inode, shadow); + if (err) { + free_shadow(inode, shadow); + return err; + } + + set_iused(inode, shadow); + full = 1; + if (level != 0) { + alloc_indirect_block(inode, page, 0); + full = logfs_block(page)->full == LOGFS_BLOCK_FACTOR; + } + fill_shadow_tree(inode, page, shadow); + wc->ofs = shadow->new_ofs; + if (wc->ofs && full) + wc->ofs |= LOGFS_FULLY_POPULATED; + return 0; +} + +static int logfs_write_direct(struct inode *inode, struct page *page, + long flags) +{ + struct logfs_inode *li = logfs_inode(inode); + struct write_control wc = { + .ofs = li->li_data[page->index], + .flags = flags, + }; + int err; + + alloc_inode_block(inode); + + err = logfs_write_i0(inode, page, &wc); + if (err) + return err; + + li->li_data[page->index] = wc.ofs; + logfs_set_alias(inode->i_sb, li->li_block, + page->index + INODE_POINTER_OFS); + return 0; +} + +static int ptr_change(u64 ofs, struct page *page) +{ + struct logfs_block *block = logfs_block(page); + int empty0, empty1, full0, full1; + + empty0 = ofs == 0; + empty1 = block->partial == 0; + if (empty0 != empty1) + return 1; + + /* The !! is necessary to shrink result to int */ + full0 = !!(ofs & LOGFS_FULLY_POPULATED); + full1 = block->full == LOGFS_BLOCK_FACTOR; + if (full0 != full1) + return 1; + return 0; +} + +static int __logfs_write_rec(struct inode *inode, struct page *page, + struct write_control *this_wc, + pgoff_t bix, level_t target_level, level_t level) +{ + int ret, page_empty = 0; + int child_no = get_bits(bix, SUBLEVEL(level)); + struct page *ipage; + struct write_control child_wc = { + .flags = this_wc->flags, + }; + + ipage = logfs_get_write_page(inode, bix, level); + if (!ipage) + return -ENOMEM; + + if (this_wc->ofs) { + ret = logfs_segment_read(inode, ipage, this_wc->ofs, bix, level); + if (ret) + goto out; + } else if (!PageUptodate(ipage)) { + page_empty = 1; + logfs_read_empty(ipage); + } + + child_wc.ofs = block_get_pointer(ipage, child_no); + + if ((__force u8)level-1 > (__force u8)target_level) + ret = __logfs_write_rec(inode, page, &child_wc, bix, + target_level, SUBLEVEL(level)); + else + ret = logfs_write_i0(inode, page, &child_wc); + + if (ret) + goto out; + + alloc_indirect_block(inode, ipage, page_empty); + block_set_pointer(ipage, child_no, child_wc.ofs); + /* FIXME: first condition seems superfluous */ + if (child_wc.ofs || logfs_block(ipage)->partial) + this_wc->flags |= WF_WRITE; + /* the condition on this_wc->ofs ensures that we won't consume extra + * space for indirect blocks in the future, which we cannot reserve */ + if (!this_wc->ofs || ptr_change(this_wc->ofs, ipage)) + ret = logfs_write_i0(inode, ipage, this_wc); + else + logfs_set_alias(inode->i_sb, logfs_block(ipage), child_no); +out: + logfs_put_write_page(ipage); + return ret; +} + +static int logfs_write_rec(struct inode *inode, struct page *page, + pgoff_t bix, level_t target_level, long flags) +{ + struct logfs_inode *li = logfs_inode(inode); + struct write_control wc = { + .ofs = li->li_data[INDIRECT_INDEX], + .flags = flags, + }; + int ret; + + alloc_inode_block(inode); + + if (li->li_height > (__force u8)target_level) + ret = __logfs_write_rec(inode, page, &wc, bix, target_level, + LEVEL(li->li_height)); + else + ret = logfs_write_i0(inode, page, &wc); + if (ret) + return ret; + + if (li->li_data[INDIRECT_INDEX] != wc.ofs) { + li->li_data[INDIRECT_INDEX] = wc.ofs; + logfs_set_alias(inode->i_sb, li->li_block, + INDIRECT_INDEX + INODE_POINTER_OFS); + } + return ret; +} + +void logfs_add_transaction(struct inode *inode, struct logfs_transaction *ta) +{ + alloc_inode_block(inode); + logfs_inode(inode)->li_block->ta = ta; +} + +void logfs_del_transaction(struct inode *inode, struct logfs_transaction *ta) +{ + struct logfs_block *block = logfs_inode(inode)->li_block; + + if (block && block->ta) + block->ta = NULL; +} + +static int grow_inode(struct inode *inode, u64 bix, level_t level) +{ + struct logfs_inode *li = logfs_inode(inode); + u8 height = (__force u8)level; + struct page *page; + struct write_control wc = { + .flags = WF_WRITE, + }; + int err; + + BUG_ON(height > 5 || li->li_height > 5); + while (height > li->li_height || bix >= maxbix(li->li_height)) { + page = logfs_get_write_page(inode, I0_BLOCKS + 1, + LEVEL(li->li_height + 1)); + if (!page) + return -ENOMEM; + logfs_read_empty(page); + alloc_indirect_block(inode, page, 1); + block_set_pointer(page, 0, li->li_data[INDIRECT_INDEX]); + err = logfs_write_i0(inode, page, &wc); + logfs_put_write_page(page); + if (err) + return err; + li->li_data[INDIRECT_INDEX] = wc.ofs; + wc.ofs = 0; + li->li_height++; + logfs_set_alias(inode->i_sb, li->li_block, INODE_HEIGHT_OFS); + } + return 0; +} + +static int __logfs_write_buf(struct inode *inode, struct page *page, long flags) +{ + struct logfs_super *super = logfs_super(inode->i_sb); + pgoff_t index = page->index; + u64 bix; + level_t level; + int err; + + flags |= WF_WRITE | WF_DELETE; + inode->i_ctime = inode->i_mtime = CURRENT_TIME; + + logfs_unpack_index(index, &bix, &level); + if (logfs_block(page) && logfs_block(page)->reserved_bytes) + super->s_dirty_pages -= logfs_block(page)->reserved_bytes; + + if (index < I0_BLOCKS) + return logfs_write_direct(inode, page, flags); + + bix = adjust_bix(bix, level); + err = grow_inode(inode, bix, level); + if (err) + return err; + return logfs_write_rec(inode, page, bix, level, flags); +} + +int logfs_write_buf(struct inode *inode, struct page *page, long flags) +{ + struct super_block *sb = inode->i_sb; + int ret; + + logfs_get_wblocks(sb, page, flags & WF_LOCK); + ret = __logfs_write_buf(inode, page, flags); + logfs_put_wblocks(sb, page, flags & WF_LOCK); + return ret; +} + +static int __logfs_delete(struct inode *inode, struct page *page) +{ + long flags = WF_DELETE; + + inode->i_ctime = inode->i_mtime = CURRENT_TIME; + + if (page->index < I0_BLOCKS) + return logfs_write_direct(inode, page, flags); + return logfs_write_rec(inode, page, page->index, 0, flags); +} + +int logfs_delete(struct inode *inode, pgoff_t index, + struct shadow_tree *shadow_tree) +{ + struct super_block *sb = inode->i_sb; + struct page *page; + int ret; + + page = logfs_get_read_page(inode, index, 0); + if (!page) + return -ENOMEM; + + logfs_get_wblocks(sb, page, 1); + ret = __logfs_delete(inode, page); + logfs_put_wblocks(sb, page, 1); + + logfs_put_read_page(page); + + return ret; +} + +/* Rewrite cannot mark the inode dirty but has to write it immediatly. */ +int logfs_rewrite_block(struct inode *inode, u64 bix, u64 ofs, + gc_level_t gc_level, long flags) +{ + level_t level = shrink_level(gc_level); + struct page *page; + int err; + + page = logfs_get_write_page(inode, bix, level); + if (!page) + return -ENOMEM; + + err = logfs_segment_read(inode, page, ofs, bix, level); + if (!err) { + if (level != 0) + alloc_indirect_block(inode, page, 0); + err = logfs_write_buf(inode, page, flags); + } + logfs_put_write_page(page); + return err; +} + +static int truncate_data_block(struct inode *inode, struct page *page, + u64 ofs, struct logfs_shadow *shadow, u64 size) +{ + loff_t pageofs = page->index << inode->i_sb->s_blocksize_bits; + u64 bix; + level_t level; + int err; + + /* Does truncation happen within this page? */ + if (size <= pageofs || size - pageofs >= PAGE_SIZE) + return 0; + + logfs_unpack_index(page->index, &bix, &level); + BUG_ON(level != 0); + + err = logfs_segment_read(inode, page, ofs, bix, level); + if (err) + return err; + + zero_user_segment(page, size - pageofs, PAGE_CACHE_SIZE); + return logfs_segment_write(inode, page, shadow); +} + +static int logfs_truncate_i0(struct inode *inode, struct page *page, + struct write_control *wc, u64 size) +{ + struct logfs_shadow *shadow; + u64 bix; + level_t level; + int err = 0; + + logfs_unpack_index(page->index, &bix, &level); + BUG_ON(level != 0); + shadow = alloc_shadow(inode, bix, level, wc->ofs); + + err = truncate_data_block(inode, page, wc->ofs, shadow, size); + if (err) { + free_shadow(inode, shadow); + return err; + } + + logfs_segment_delete(inode, shadow); + set_iused(inode, shadow); + fill_shadow_tree(inode, page, shadow); + wc->ofs = shadow->new_ofs; + return 0; +} + +static int logfs_truncate_direct(struct inode *inode, u64 size) +{ + struct logfs_inode *li = logfs_inode(inode); + struct write_control wc; + struct page *page; + int e; + int err; + + alloc_inode_block(inode); + + for (e = I0_BLOCKS - 1; e >= 0; e--) { + if (size > (e+1) * LOGFS_BLOCKSIZE) + break; + + wc.ofs = li->li_data[e]; + if (!wc.ofs) + continue; + + page = logfs_get_write_page(inode, e, 0); + if (!page) + return -ENOMEM; + err = logfs_segment_read(inode, page, wc.ofs, e, 0); + if (err) { + logfs_put_write_page(page); + return err; + } + err = logfs_truncate_i0(inode, page, &wc, size); + logfs_put_write_page(page); + if (err) + return err; + + li->li_data[e] = wc.ofs; + } + return 0; +} + +/* FIXME: these need to become per-sb once we support different blocksizes */ +static u64 __logfs_step[] = { + 1, + I1_BLOCKS, + I2_BLOCKS, + I3_BLOCKS, +}; + +static u64 __logfs_start_index[] = { + I0_BLOCKS, + I1_BLOCKS, + I2_BLOCKS, + I3_BLOCKS +}; + +static inline u64 logfs_step(level_t level) +{ + return __logfs_step[(__force u8)level]; +} + +static inline u64 logfs_factor(u8 level) +{ + return __logfs_step[level] * LOGFS_BLOCKSIZE; +} + +static inline u64 logfs_start_index(level_t level) +{ + return __logfs_start_index[(__force u8)level]; +} + +static void logfs_unpack_raw_index(pgoff_t index, u64 *bix, level_t *level) +{ + logfs_unpack_index(index, bix, level); + if (*bix <= logfs_start_index(SUBLEVEL(*level))) + *bix = 0; +} + +static int __logfs_truncate_rec(struct inode *inode, struct page *ipage, + struct write_control *this_wc, u64 size) +{ + int truncate_happened = 0; + int e, err = 0; + u64 bix, child_bix, next_bix; + level_t level; + struct page *page; + struct write_control child_wc = { /* FIXME: flags */ }; + + logfs_unpack_raw_index(ipage->index, &bix, &level); + err = logfs_segment_read(inode, ipage, this_wc->ofs, bix, level); + if (err) + return err; + + for (e = LOGFS_BLOCK_FACTOR - 1; e >= 0; e--) { + child_bix = bix + e * logfs_step(SUBLEVEL(level)); + next_bix = child_bix + logfs_step(SUBLEVEL(level)); + if (size > next_bix * LOGFS_BLOCKSIZE) + break; + + child_wc.ofs = pure_ofs(block_get_pointer(ipage, e)); + if (!child_wc.ofs) + continue; + + page = logfs_get_write_page(inode, child_bix, SUBLEVEL(level)); + if (!page) + return -ENOMEM; + + if ((__force u8)level > 1) + err = __logfs_truncate_rec(inode, page, &child_wc, size); + else + err = logfs_truncate_i0(inode, page, &child_wc, size); + logfs_put_write_page(page); + if (err) + return err; + + truncate_happened = 1; + alloc_indirect_block(inode, ipage, 0); + block_set_pointer(ipage, e, child_wc.ofs); + } + + if (!truncate_happened) { + printk("ineffectual truncate (%lx, %lx, %llx)\n", inode->i_ino, ipage->index, size); + return 0; + } + + this_wc->flags = WF_DELETE; + if (logfs_block(ipage)->partial) + this_wc->flags |= WF_WRITE; + + return logfs_write_i0(inode, ipage, this_wc); +} + +static int logfs_truncate_rec(struct inode *inode, u64 size) +{ + struct logfs_inode *li = logfs_inode(inode); + struct write_control wc = { + .ofs = li->li_data[INDIRECT_INDEX], + }; + struct page *page; + int err; + + alloc_inode_block(inode); + + if (!wc.ofs) + return 0; + + page = logfs_get_write_page(inode, 0, LEVEL(li->li_height)); + if (!page) + return -ENOMEM; + + err = __logfs_truncate_rec(inode, page, &wc, size); + logfs_put_write_page(page); + if (err) + return err; + + if (li->li_data[INDIRECT_INDEX] != wc.ofs) + li->li_data[INDIRECT_INDEX] = wc.ofs; + return 0; +} + +static int __logfs_truncate(struct inode *inode, u64 size) +{ + int ret; + + if (size >= logfs_factor(logfs_inode(inode)->li_height)) + return 0; + + ret = logfs_truncate_rec(inode, size); + if (ret) + return ret; + + return logfs_truncate_direct(inode, size); +} + +int logfs_truncate(struct inode *inode, u64 size) +{ + struct super_block *sb = inode->i_sb; + int err; + + logfs_get_wblocks(sb, NULL, 1); + err = __logfs_truncate(inode, size); + if (!err) + err = __logfs_write_inode(inode, 0); + logfs_put_wblocks(sb, NULL, 1); + + if (!err) + err = vmtruncate(inode, size); + + /* I don't trust error recovery yet. */ + WARN_ON(err); + return err; +} + +static void move_page_to_inode(struct inode *inode, struct page *page) +{ + struct logfs_inode *li = logfs_inode(inode); + struct logfs_block *block = logfs_block(page); + + if (!block) + return; + + log_blockmove("move_page_to_inode(%llx, %llx, %x)\n", + block->ino, block->bix, block->level); + BUG_ON(li->li_block); + block->ops = &inode_block_ops; + block->inode = inode; + li->li_block = block; + + block->page = NULL; + page->private = 0; + ClearPagePrivate(page); +} + +static void move_inode_to_page(struct page *page, struct inode *inode) +{ + struct logfs_inode *li = logfs_inode(inode); + struct logfs_block *block = li->li_block; + + if (!block) + return; + + log_blockmove("move_inode_to_page(%llx, %llx, %x)\n", + block->ino, block->bix, block->level); + BUG_ON(PagePrivate(page)); + block->ops = &indirect_block_ops; + block->page = page; + page->private = (unsigned long)block; + SetPagePrivate(page); + + block->inode = NULL; + li->li_block = NULL; +} + +int logfs_read_inode(struct inode *inode) +{ + struct super_block *sb = inode->i_sb; + struct logfs_super *super = logfs_super(sb); + struct inode *master_inode = super->s_master_inode; + struct page *page; + struct logfs_disk_inode *di; + u64 ino = inode->i_ino; + + if (ino << sb->s_blocksize_bits > i_size_read(master_inode)) + return -ENODATA; + if (!logfs_exist_block(master_inode, ino)) + return -ENODATA; + + page = read_cache_page(master_inode->i_mapping, ino, + (filler_t *)logfs_readpage, NULL); + if (IS_ERR(page)) + return PTR_ERR(page); + + di = kmap_atomic(page, KM_USER0); + logfs_disk_to_inode(di, inode); + kunmap_atomic(di, KM_USER0); + move_page_to_inode(inode, page); + page_cache_release(page); + return 0; +} + +/* Caller must logfs_put_write_page(page); */ +static struct page *inode_to_page(struct inode *inode) +{ + struct inode *master_inode = logfs_super(inode->i_sb)->s_master_inode; + struct logfs_disk_inode *di; + struct page *page; + + BUG_ON(inode->i_ino == LOGFS_INO_MASTER); + + page = logfs_get_write_page(master_inode, inode->i_ino, 0); + if (!page) + return NULL; + + di = kmap_atomic(page, KM_USER0); + logfs_inode_to_disk(inode, di); + kunmap_atomic(di, KM_USER0); + move_inode_to_page(page, inode); + return page; +} + +/* Cheaper version of write_inode. All changes are concealed in + * aliases, which are moved back. No write to the medium happens. + */ +void logfs_clear_inode(struct inode *inode) +{ + struct super_block *sb = inode->i_sb; + struct logfs_inode *li = logfs_inode(inode); + struct logfs_block *block = li->li_block; + struct page *page; + + /* Only deleted files may be dirty at this point */ + BUG_ON(inode->i_state & I_DIRTY && inode->i_nlink); + if (!block) + return; + if ((logfs_super(sb)->s_flags & LOGFS_SB_FLAG_SHUTDOWN)) { + block->ops->free_block(inode->i_sb, block); + return; + } + + BUG_ON(inode->i_ino < LOGFS_RESERVED_INOS); + page = inode_to_page(inode); + BUG_ON(!page); /* FIXME: Use emergency page */ + logfs_put_write_page(page); +} + +static int do_write_inode(struct inode *inode) +{ + struct super_block *sb = inode->i_sb; + struct inode *master_inode = logfs_super(sb)->s_master_inode; + loff_t size = (inode->i_ino + 1) << inode->i_sb->s_blocksize_bits; + struct page *page; + int err; + + BUG_ON(inode->i_ino == LOGFS_INO_MASTER); + /* FIXME: lock inode */ + + if (i_size_read(master_inode) < size) + i_size_write(master_inode, size); + + /* TODO: Tell vfs this inode is clean now */ + + page = inode_to_page(inode); + if (!page) + return -ENOMEM; + + /* FIXME: transaction is part of logfs_block now. Is that enough? */ + err = logfs_write_buf(master_inode, page, 0); + logfs_put_write_page(page); + return err; +} + +static void logfs_mod_segment_entry(struct super_block *sb, u32 segno, + int write, + void (*change_se)(struct logfs_segment_entry *, long), + long arg) +{ + struct logfs_super *super = logfs_super(sb); + struct inode *inode; + struct page *page; + struct logfs_segment_entry *se; + pgoff_t page_no; + int child_no; + + page_no = segno >> (sb->s_blocksize_bits - 3); + child_no = segno & ((sb->s_blocksize >> 3) - 1); + + inode = super->s_segfile_inode; + page = logfs_get_write_page(inode, page_no, 0); + BUG_ON(!page); /* FIXME: We need some reserve page for this case */ + if (!PageUptodate(page)) + logfs_read_block(inode, page, WRITE); + + if (write) + alloc_indirect_block(inode, page, 0); + se = kmap_atomic(page, KM_USER0); + change_se(se + child_no, arg); + if (write) { + logfs_set_alias(sb, logfs_block(page), child_no); + BUG_ON((int)be32_to_cpu(se[child_no].valid) > super->s_segsize); + } + kunmap_atomic(se, KM_USER0); + + logfs_put_write_page(page); +} + +static void __get_segment_entry(struct logfs_segment_entry *se, long _target) +{ + struct logfs_segment_entry *target = (void *)_target; + + *target = *se; +} + +void logfs_get_segment_entry(struct super_block *sb, u32 segno, + struct logfs_segment_entry *se) +{ + logfs_mod_segment_entry(sb, segno, 0, __get_segment_entry, (long)se); +} + +static void __set_segment_used(struct logfs_segment_entry *se, long increment) +{ + u32 valid; + + valid = be32_to_cpu(se->valid); + valid += increment; + se->valid = cpu_to_be32(valid); +} + +void logfs_set_segment_used(struct super_block *sb, u64 ofs, int increment) +{ + struct logfs_super *super = logfs_super(sb); + u32 segno = ofs >> super->s_segshift; + + if (!increment) + return; + + logfs_mod_segment_entry(sb, segno, 1, __set_segment_used, increment); +} + +static void __set_segment_erased(struct logfs_segment_entry *se, long ec_level) +{ + se->ec_level = cpu_to_be32(ec_level); +} + +void logfs_set_segment_erased(struct super_block *sb, u32 segno, u32 ec, + gc_level_t gc_level) +{ + u32 ec_level = ec << 4 | (__force u8)gc_level; + + logfs_mod_segment_entry(sb, segno, 1, __set_segment_erased, ec_level); +} + +static void __set_segment_reserved(struct logfs_segment_entry *se, long ignore) +{ + se->valid = cpu_to_be32(RESERVED); +} + +void logfs_set_segment_reserved(struct super_block *sb, u32 segno) +{ + logfs_mod_segment_entry(sb, segno, 1, __set_segment_reserved, 0); +} + +static void __set_segment_unreserved(struct logfs_segment_entry *se, + long ec_level) +{ + se->valid = 0; + se->ec_level = cpu_to_be32(ec_level); +} + +void logfs_set_segment_unreserved(struct super_block *sb, u32 segno, u32 ec) +{ + u32 ec_level = ec << 4; + + logfs_mod_segment_entry(sb, segno, 1, __set_segment_unreserved, + ec_level); +} + +int __logfs_write_inode(struct inode *inode, long flags) +{ + struct super_block *sb = inode->i_sb; + int ret; + + logfs_get_wblocks(sb, NULL, flags & WF_LOCK); + ret = do_write_inode(inode); + logfs_put_wblocks(sb, NULL, flags & WF_LOCK); + return ret; +} + +static int do_delete_inode(struct inode *inode) +{ + struct super_block *sb = inode->i_sb; + struct inode *master_inode = logfs_super(sb)->s_master_inode; + struct page *page; + int ret; + + page = logfs_get_write_page(master_inode, inode->i_ino, 0); + if (!page) + return -ENOMEM; + + move_inode_to_page(page, inode); + + logfs_get_wblocks(sb, page, 1); + ret = __logfs_delete(master_inode, page); + logfs_put_wblocks(sb, page, 1); + + logfs_put_write_page(page); + return ret; +} + +/* + * ZOMBIE inodes have already been deleted before and should remain dead, + * if it weren't for valid checking. No need to kill them again here. + */ +void logfs_delete_inode(struct inode *inode) +{ + struct logfs_inode *li = logfs_inode(inode); + + if (!(li->li_flags & LOGFS_IF_ZOMBIE)) { + li->li_flags |= LOGFS_IF_ZOMBIE; + if (i_size_read(inode) > 0) + logfs_truncate(inode, 0); + do_delete_inode(inode); + } + truncate_inode_pages(&inode->i_data, 0); + clear_inode(inode); +} + +void btree_write_block(struct logfs_block *block) +{ + struct inode *inode; + struct page *page; + int err, cookie; + + inode = logfs_safe_iget(block->sb, block->ino, &cookie); + page = logfs_get_write_page(inode, block->bix, block->level); + + err = logfs_readpage_nolock(page); + BUG_ON(err); + BUG_ON(!PagePrivate(page)); + BUG_ON(logfs_block(page) != block); + err = __logfs_write_buf(inode, page, 0); + BUG_ON(err); + BUG_ON(PagePrivate(page) || page->private); + + logfs_put_write_page(page); + logfs_safe_iput(inode, cookie); +} + +/** + * logfs_inode_write - write inode or dentry objects + * + * @inode: parent inode (ifile or directory) + * @buf: object to write (inode or dentry) + * @n: object size + * @_pos: object number (file position in blocks/objects) + * @flags: write flags + * @lock: 0 if write lock is already taken, 1 otherwise + * @shadow_tree: shadow below this inode + * + * FIXME: All caller of this put a 200-300 byte variable on the stack, + * only to call here and do a memcpy from that stack variable. A good + * example of wasted performance and stack space. + */ +int logfs_inode_write(struct inode *inode, const void *buf, size_t count, + loff_t bix, long flags, struct shadow_tree *shadow_tree) +{ + loff_t pos = bix << inode->i_sb->s_blocksize_bits; + int err; + struct page *page; + void *pagebuf; + + BUG_ON(pos & (LOGFS_BLOCKSIZE-1)); + BUG_ON(count > LOGFS_BLOCKSIZE); + page = logfs_get_write_page(inode, bix, 0); + if (!page) + return -ENOMEM; + + pagebuf = kmap_atomic(page, KM_USER0); + memcpy(pagebuf, buf, count); + flush_dcache_page(page); + kunmap_atomic(pagebuf, KM_USER0); + + if (i_size_read(inode) < pos + LOGFS_BLOCKSIZE) + i_size_write(inode, pos + LOGFS_BLOCKSIZE); + + err = logfs_write_buf(inode, page, flags); + logfs_put_write_page(page); + return err; +} + +int logfs_open_segfile(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + struct inode *inode; + + inode = logfs_read_meta_inode(sb, LOGFS_INO_SEGFILE); + if (IS_ERR(inode)) + return PTR_ERR(inode); + super->s_segfile_inode = inode; + return 0; +} + +int logfs_init_rw(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + int min_fill = 3 * super->s_no_blocks; + + INIT_LIST_HEAD(&super->s_object_alias); + mutex_init(&super->s_write_mutex); + super->s_block_pool = mempool_create_kmalloc_pool(min_fill, + sizeof(struct logfs_block)); + super->s_shadow_pool = mempool_create_kmalloc_pool(min_fill, + sizeof(struct logfs_shadow)); + return 0; +} + +void logfs_cleanup_rw(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + + destroy_meta_inode(super->s_segfile_inode); + if (super->s_block_pool) + mempool_destroy(super->s_block_pool); + if (super->s_shadow_pool) + mempool_destroy(super->s_shadow_pool); +} diff --git a/fs/logfs/segment.c b/fs/logfs/segment.c new file mode 100644 index 000000000000..1a14f9910d55 --- /dev/null +++ b/fs/logfs/segment.c @@ -0,0 +1,927 @@ +/* + * fs/logfs/segment.c - Handling the Object Store + * + * As should be obvious for Linux kernel code, license is GPLv2 + * + * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> + * + * Object store or ostore makes up the complete device with exception of + * the superblock and journal areas. Apart from its own metadata it stores + * three kinds of objects: inodes, dentries and blocks, both data and indirect. + */ +#include "logfs.h" + +static int logfs_mark_segment_bad(struct super_block *sb, u32 segno) +{ + struct logfs_super *super = logfs_super(sb); + struct btree_head32 *head = &super->s_reserved_segments; + int err; + + err = btree_insert32(head, segno, (void *)1, GFP_NOFS); + if (err) + return err; + logfs_super(sb)->s_bad_segments++; + /* FIXME: write to journal */ + return 0; +} + +int logfs_erase_segment(struct super_block *sb, u32 segno, int ensure_erase) +{ + struct logfs_super *super = logfs_super(sb); + + super->s_gec++; + + return super->s_devops->erase(sb, (u64)segno << super->s_segshift, + super->s_segsize, ensure_erase); +} + +static s64 logfs_get_free_bytes(struct logfs_area *area, size_t bytes) +{ + s32 ofs; + + logfs_open_area(area, bytes); + + ofs = area->a_used_bytes; + area->a_used_bytes += bytes; + BUG_ON(area->a_used_bytes >= logfs_super(area->a_sb)->s_segsize); + + return dev_ofs(area->a_sb, area->a_segno, ofs); +} + +static struct page *get_mapping_page(struct super_block *sb, pgoff_t index, + int use_filler) +{ + struct logfs_super *super = logfs_super(sb); + struct address_space *mapping = super->s_mapping_inode->i_mapping; + filler_t *filler = super->s_devops->readpage; + struct page *page; + + BUG_ON(mapping_gfp_mask(mapping) & __GFP_FS); + if (use_filler) + page = read_cache_page(mapping, index, filler, sb); + else { + page = find_or_create_page(mapping, index, GFP_NOFS); + unlock_page(page); + } + return page; +} + +void __logfs_buf_write(struct logfs_area *area, u64 ofs, void *buf, size_t len, + int use_filler) +{ + pgoff_t index = ofs >> PAGE_SHIFT; + struct page *page; + long offset = ofs & (PAGE_SIZE-1); + long copylen; + + /* Only logfs_wbuf_recover may use len==0 */ + BUG_ON(!len && !use_filler); + do { + copylen = min((ulong)len, PAGE_SIZE - offset); + + page = get_mapping_page(area->a_sb, index, use_filler); + SetPageUptodate(page); + BUG_ON(!page); /* FIXME: reserve a pool */ + memcpy(page_address(page) + offset, buf, copylen); + SetPagePrivate(page); + page_cache_release(page); + + buf += copylen; + len -= copylen; + offset = 0; + index++; + } while (len); +} + +/* + * bdev_writeseg will write full pages. Memset the tail to prevent data leaks. + */ +static void pad_wbuf(struct logfs_area *area, int final) +{ + struct super_block *sb = area->a_sb; + struct logfs_super *super = logfs_super(sb); + struct page *page; + u64 ofs = dev_ofs(sb, area->a_segno, area->a_used_bytes); + pgoff_t index = ofs >> PAGE_SHIFT; + long offset = ofs & (PAGE_SIZE-1); + u32 len = PAGE_SIZE - offset; + + if (len == PAGE_SIZE) { + /* The math in this function can surely use some love */ + len = 0; + } + if (len) { + BUG_ON(area->a_used_bytes >= super->s_segsize); + + page = get_mapping_page(area->a_sb, index, 0); + BUG_ON(!page); /* FIXME: reserve a pool */ + memset(page_address(page) + offset, 0xff, len); + SetPagePrivate(page); + page_cache_release(page); + } + + if (!final) + return; + + area->a_used_bytes += len; + for ( ; area->a_used_bytes < super->s_segsize; + area->a_used_bytes += PAGE_SIZE) { + /* Memset another page */ + index++; + page = get_mapping_page(area->a_sb, index, 0); + BUG_ON(!page); /* FIXME: reserve a pool */ + memset(page_address(page), 0xff, PAGE_SIZE); + SetPagePrivate(page); + page_cache_release(page); + } +} + +/* + * We have to be careful with the alias tree. Since lookup is done by bix, + * it needs to be normalized, so 14, 15, 16, etc. all match when dealing with + * indirect blocks. So always use it through accessor functions. + */ +static void *alias_tree_lookup(struct super_block *sb, u64 ino, u64 bix, + level_t level) +{ + struct btree_head128 *head = &logfs_super(sb)->s_object_alias_tree; + pgoff_t index = logfs_pack_index(bix, level); + + return btree_lookup128(head, ino, index); +} + +static int alias_tree_insert(struct super_block *sb, u64 ino, u64 bix, + level_t level, void *val) +{ + struct btree_head128 *head = &logfs_super(sb)->s_object_alias_tree; + pgoff_t index = logfs_pack_index(bix, level); + + return btree_insert128(head, ino, index, val, GFP_NOFS); +} + +static int btree_write_alias(struct super_block *sb, struct logfs_block *block, + write_alias_t *write_one_alias) +{ + struct object_alias_item *item; + int err; + + list_for_each_entry(item, &block->item_list, list) { + err = write_alias_journal(sb, block->ino, block->bix, + block->level, item->child_no, item->val); + if (err) + return err; + } + return 0; +} + +static gc_level_t btree_block_level(struct logfs_block *block) +{ + return expand_level(block->ino, block->level); +} + +static struct logfs_block_ops btree_block_ops = { + .write_block = btree_write_block, + .block_level = btree_block_level, + .free_block = __free_block, + .write_alias = btree_write_alias, +}; + +int logfs_load_object_aliases(struct super_block *sb, + struct logfs_obj_alias *oa, int count) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_block *block; + struct object_alias_item *item; + u64 ino, bix; + level_t level; + int i, err; + + super->s_flags |= LOGFS_SB_FLAG_OBJ_ALIAS; + count /= sizeof(*oa); + for (i = 0; i < count; i++) { + item = mempool_alloc(super->s_alias_pool, GFP_NOFS); + if (!item) + return -ENOMEM; + memset(item, 0, sizeof(*item)); + + super->s_no_object_aliases++; + item->val = oa[i].val; + item->child_no = be16_to_cpu(oa[i].child_no); + + ino = be64_to_cpu(oa[i].ino); + bix = be64_to_cpu(oa[i].bix); + level = LEVEL(oa[i].level); + + log_aliases("logfs_load_object_aliases(%llx, %llx, %x, %x) %llx\n", + ino, bix, level, item->child_no, + be64_to_cpu(item->val)); + block = alias_tree_lookup(sb, ino, bix, level); + if (!block) { + block = __alloc_block(sb, ino, bix, level); + block->ops = &btree_block_ops; + err = alias_tree_insert(sb, ino, bix, level, block); + BUG_ON(err); /* mempool empty */ + } + if (test_and_set_bit(item->child_no, block->alias_map)) { + printk(KERN_ERR"LogFS: Alias collision detected\n"); + return -EIO; + } + list_move_tail(&block->alias_list, &super->s_object_alias); + list_add(&item->list, &block->item_list); + } + return 0; +} + +static void kill_alias(void *_block, unsigned long ignore0, + u64 ignore1, u64 ignore2, size_t ignore3) +{ + struct logfs_block *block = _block; + struct super_block *sb = block->sb; + struct logfs_super *super = logfs_super(sb); + struct object_alias_item *item; + + while (!list_empty(&block->item_list)) { + item = list_entry(block->item_list.next, typeof(*item), list); + list_del(&item->list); + mempool_free(item, super->s_alias_pool); + } + block->ops->free_block(sb, block); +} + +static int obj_type(struct inode *inode, level_t level) +{ + if (level == 0) { + if (S_ISDIR(inode->i_mode)) + return OBJ_DENTRY; + if (inode->i_ino == LOGFS_INO_MASTER) + return OBJ_INODE; + } + return OBJ_BLOCK; +} + +static int obj_len(struct super_block *sb, int obj_type) +{ + switch (obj_type) { + case OBJ_DENTRY: + return sizeof(struct logfs_disk_dentry); + case OBJ_INODE: + return sizeof(struct logfs_disk_inode); + case OBJ_BLOCK: + return sb->s_blocksize; + default: + BUG(); + } +} + +static int __logfs_segment_write(struct inode *inode, void *buf, + struct logfs_shadow *shadow, int type, int len, int compr) +{ + struct logfs_area *area; + struct super_block *sb = inode->i_sb; + s64 ofs; + struct logfs_object_header h; + int acc_len; + + if (shadow->gc_level == 0) + acc_len = len; + else + acc_len = obj_len(sb, type); + + area = get_area(sb, shadow->gc_level); + ofs = logfs_get_free_bytes(area, len + LOGFS_OBJECT_HEADERSIZE); + LOGFS_BUG_ON(ofs <= 0, sb); + /* + * Order is important. logfs_get_free_bytes(), by modifying the + * segment file, may modify the content of the very page we're about + * to write now. Which is fine, as long as the calculated crc and + * written data still match. So do the modifications _before_ + * calculating the crc. + */ + + h.len = cpu_to_be16(len); + h.type = type; + h.compr = compr; + h.ino = cpu_to_be64(inode->i_ino); + h.bix = cpu_to_be64(shadow->bix); + h.crc = logfs_crc32(&h, sizeof(h) - 4, 4); + h.data_crc = logfs_crc32(buf, len, 0); + + logfs_buf_write(area, ofs, &h, sizeof(h)); + logfs_buf_write(area, ofs + LOGFS_OBJECT_HEADERSIZE, buf, len); + + shadow->new_ofs = ofs; + shadow->new_len = acc_len + LOGFS_OBJECT_HEADERSIZE; + + return 0; +} + +static s64 logfs_segment_write_compress(struct inode *inode, void *buf, + struct logfs_shadow *shadow, int type, int len) +{ + struct super_block *sb = inode->i_sb; + void *compressor_buf = logfs_super(sb)->s_compressed_je; + ssize_t compr_len; + int ret; + + mutex_lock(&logfs_super(sb)->s_journal_mutex); + compr_len = logfs_compress(buf, compressor_buf, len, len); + + if (compr_len >= 0) { + ret = __logfs_segment_write(inode, compressor_buf, shadow, + type, compr_len, COMPR_ZLIB); + } else { + ret = __logfs_segment_write(inode, buf, shadow, type, len, + COMPR_NONE); + } + mutex_unlock(&logfs_super(sb)->s_journal_mutex); + return ret; +} + +/** + * logfs_segment_write - write data block to object store + * @inode: inode containing data + * + * Returns an errno or zero. + */ +int logfs_segment_write(struct inode *inode, struct page *page, + struct logfs_shadow *shadow) +{ + struct super_block *sb = inode->i_sb; + struct logfs_super *super = logfs_super(sb); + int do_compress, type, len; + int ret; + void *buf; + + super->s_flags |= LOGFS_SB_FLAG_DIRTY; + BUG_ON(super->s_flags & LOGFS_SB_FLAG_SHUTDOWN); + do_compress = logfs_inode(inode)->li_flags & LOGFS_IF_COMPRESSED; + if (shadow->gc_level != 0) { + /* temporarily disable compression for indirect blocks */ + do_compress = 0; + } + + type = obj_type(inode, shrink_level(shadow->gc_level)); + len = obj_len(sb, type); + buf = kmap(page); + if (do_compress) + ret = logfs_segment_write_compress(inode, buf, shadow, type, + len); + else + ret = __logfs_segment_write(inode, buf, shadow, type, len, + COMPR_NONE); + kunmap(page); + + log_segment("logfs_segment_write(%llx, %llx, %x) %llx->%llx %x->%x\n", + shadow->ino, shadow->bix, shadow->gc_level, + shadow->old_ofs, shadow->new_ofs, + shadow->old_len, shadow->new_len); + /* this BUG_ON did catch a locking bug. useful */ + BUG_ON(!(shadow->new_ofs & (super->s_segsize - 1))); + return ret; +} + +int wbuf_read(struct super_block *sb, u64 ofs, size_t len, void *buf) +{ + pgoff_t index = ofs >> PAGE_SHIFT; + struct page *page; + long offset = ofs & (PAGE_SIZE-1); + long copylen; + + while (len) { + copylen = min((ulong)len, PAGE_SIZE - offset); + + page = get_mapping_page(sb, index, 1); + if (IS_ERR(page)) + return PTR_ERR(page); + memcpy(buf, page_address(page) + offset, copylen); + page_cache_release(page); + + buf += copylen; + len -= copylen; + offset = 0; + index++; + } + return 0; +} + +/* + * The "position" of indirect blocks is ambiguous. It can be the position + * of any data block somewhere behind this indirect block. So we need to + * normalize the positions through logfs_block_mask() before comparing. + */ +static int check_pos(struct super_block *sb, u64 pos1, u64 pos2, level_t level) +{ + return (pos1 & logfs_block_mask(sb, level)) != + (pos2 & logfs_block_mask(sb, level)); +} + +#if 0 +static int read_seg_header(struct super_block *sb, u64 ofs, + struct logfs_segment_header *sh) +{ + __be32 crc; + int err; + + err = wbuf_read(sb, ofs, sizeof(*sh), sh); + if (err) + return err; + crc = logfs_crc32(sh, sizeof(*sh), 4); + if (crc != sh->crc) { + printk(KERN_ERR"LOGFS: header crc error at %llx: expected %x, " + "got %x\n", ofs, be32_to_cpu(sh->crc), + be32_to_cpu(crc)); + return -EIO; + } + return 0; +} +#endif + +static int read_obj_header(struct super_block *sb, u64 ofs, + struct logfs_object_header *oh) +{ + __be32 crc; + int err; + + err = wbuf_read(sb, ofs, sizeof(*oh), oh); + if (err) + return err; + crc = logfs_crc32(oh, sizeof(*oh) - 4, 4); + if (crc != oh->crc) { + printk(KERN_ERR"LOGFS: header crc error at %llx: expected %x, " + "got %x\n", ofs, be32_to_cpu(oh->crc), + be32_to_cpu(crc)); + return -EIO; + } + return 0; +} + +static void move_btree_to_page(struct inode *inode, struct page *page, + __be64 *data) +{ + struct super_block *sb = inode->i_sb; + struct logfs_super *super = logfs_super(sb); + struct btree_head128 *head = &super->s_object_alias_tree; + struct logfs_block *block; + struct object_alias_item *item, *next; + + if (!(super->s_flags & LOGFS_SB_FLAG_OBJ_ALIAS)) + return; + + block = btree_remove128(head, inode->i_ino, page->index); + if (!block) + return; + + log_blockmove("move_btree_to_page(%llx, %llx, %x)\n", + block->ino, block->bix, block->level); + list_for_each_entry_safe(item, next, &block->item_list, list) { + data[item->child_no] = item->val; + list_del(&item->list); + mempool_free(item, super->s_alias_pool); + } + block->page = page; + SetPagePrivate(page); + page->private = (unsigned long)block; + block->ops = &indirect_block_ops; + initialize_block_counters(page, block, data, 0); +} + +/* + * This silences a false, yet annoying gcc warning. I hate it when my editor + * jumps into bitops.h each time I recompile this file. + * TODO: Complain to gcc folks about this and upgrade compiler. + */ +static unsigned long fnb(const unsigned long *addr, + unsigned long size, unsigned long offset) +{ + return find_next_bit(addr, size, offset); +} + +void move_page_to_btree(struct page *page) +{ + struct logfs_block *block = logfs_block(page); + struct super_block *sb = block->sb; + struct logfs_super *super = logfs_super(sb); + struct object_alias_item *item; + unsigned long pos; + __be64 *child; + int err; + + if (super->s_flags & LOGFS_SB_FLAG_SHUTDOWN) { + block->ops->free_block(sb, block); + return; + } + log_blockmove("move_page_to_btree(%llx, %llx, %x)\n", + block->ino, block->bix, block->level); + super->s_flags |= LOGFS_SB_FLAG_OBJ_ALIAS; + + for (pos = 0; ; pos++) { + pos = fnb(block->alias_map, LOGFS_BLOCK_FACTOR, pos); + if (pos >= LOGFS_BLOCK_FACTOR) + break; + + item = mempool_alloc(super->s_alias_pool, GFP_NOFS); + BUG_ON(!item); /* mempool empty */ + memset(item, 0, sizeof(*item)); + + child = kmap_atomic(page, KM_USER0); + item->val = child[pos]; + kunmap_atomic(child, KM_USER0); + item->child_no = pos; + list_add(&item->list, &block->item_list); + } + block->page = NULL; + ClearPagePrivate(page); + page->private = 0; + block->ops = &btree_block_ops; + err = alias_tree_insert(block->sb, block->ino, block->bix, block->level, + block); + BUG_ON(err); /* mempool empty */ + ClearPageUptodate(page); +} + +static int __logfs_segment_read(struct inode *inode, void *buf, + u64 ofs, u64 bix, level_t level) +{ + struct super_block *sb = inode->i_sb; + void *compressor_buf = logfs_super(sb)->s_compressed_je; + struct logfs_object_header oh; + __be32 crc; + u16 len; + int err, block_len; + + block_len = obj_len(sb, obj_type(inode, level)); + err = read_obj_header(sb, ofs, &oh); + if (err) + goto out_err; + + err = -EIO; + if (be64_to_cpu(oh.ino) != inode->i_ino + || check_pos(sb, be64_to_cpu(oh.bix), bix, level)) { + printk(KERN_ERR"LOGFS: (ino, bix) don't match at %llx: " + "expected (%lx, %llx), got (%llx, %llx)\n", + ofs, inode->i_ino, bix, + be64_to_cpu(oh.ino), be64_to_cpu(oh.bix)); + goto out_err; + } + + len = be16_to_cpu(oh.len); + + switch (oh.compr) { + case COMPR_NONE: + err = wbuf_read(sb, ofs + LOGFS_OBJECT_HEADERSIZE, len, buf); + if (err) + goto out_err; + crc = logfs_crc32(buf, len, 0); + if (crc != oh.data_crc) { + printk(KERN_ERR"LOGFS: uncompressed data crc error at " + "%llx: expected %x, got %x\n", ofs, + be32_to_cpu(oh.data_crc), + be32_to_cpu(crc)); + goto out_err; + } + break; + case COMPR_ZLIB: + mutex_lock(&logfs_super(sb)->s_journal_mutex); + err = wbuf_read(sb, ofs + LOGFS_OBJECT_HEADERSIZE, len, + compressor_buf); + if (err) { + mutex_unlock(&logfs_super(sb)->s_journal_mutex); + goto out_err; + } + crc = logfs_crc32(compressor_buf, len, 0); + if (crc != oh.data_crc) { + printk(KERN_ERR"LOGFS: compressed data crc error at " + "%llx: expected %x, got %x\n", ofs, + be32_to_cpu(oh.data_crc), + be32_to_cpu(crc)); + mutex_unlock(&logfs_super(sb)->s_journal_mutex); + goto out_err; + } + err = logfs_uncompress(compressor_buf, buf, len, block_len); + mutex_unlock(&logfs_super(sb)->s_journal_mutex); + if (err) { + printk(KERN_ERR"LOGFS: uncompress error at %llx\n", ofs); + goto out_err; + } + break; + default: + LOGFS_BUG(sb); + err = -EIO; + goto out_err; + } + return 0; + +out_err: + logfs_set_ro(sb); + printk(KERN_ERR"LOGFS: device is read-only now\n"); + LOGFS_BUG(sb); + return err; +} + +/** + * logfs_segment_read - read data block from object store + * @inode: inode containing data + * @buf: data buffer + * @ofs: physical data offset + * @bix: block index + * @level: block level + * + * Returns 0 on success or a negative errno. + */ +int logfs_segment_read(struct inode *inode, struct page *page, + u64 ofs, u64 bix, level_t level) +{ + int err; + void *buf; + + if (PageUptodate(page)) + return 0; + + ofs &= ~LOGFS_FULLY_POPULATED; + + buf = kmap(page); + err = __logfs_segment_read(inode, buf, ofs, bix, level); + if (!err) { + move_btree_to_page(inode, page, buf); + SetPageUptodate(page); + } + kunmap(page); + log_segment("logfs_segment_read(%lx, %llx, %x) %llx (%d)\n", + inode->i_ino, bix, level, ofs, err); + return err; +} + +int logfs_segment_delete(struct inode *inode, struct logfs_shadow *shadow) +{ + struct super_block *sb = inode->i_sb; + struct logfs_super *super = logfs_super(sb); + struct logfs_object_header h; + u16 len; + int err; + + super->s_flags |= LOGFS_SB_FLAG_DIRTY; + BUG_ON(super->s_flags & LOGFS_SB_FLAG_SHUTDOWN); + BUG_ON(shadow->old_ofs & LOGFS_FULLY_POPULATED); + if (!shadow->old_ofs) + return 0; + + log_segment("logfs_segment_delete(%llx, %llx, %x) %llx->%llx %x->%x\n", + shadow->ino, shadow->bix, shadow->gc_level, + shadow->old_ofs, shadow->new_ofs, + shadow->old_len, shadow->new_len); + err = read_obj_header(sb, shadow->old_ofs, &h); + LOGFS_BUG_ON(err, sb); + LOGFS_BUG_ON(be64_to_cpu(h.ino) != inode->i_ino, sb); + LOGFS_BUG_ON(check_pos(sb, shadow->bix, be64_to_cpu(h.bix), + shrink_level(shadow->gc_level)), sb); + + if (shadow->gc_level == 0) + len = be16_to_cpu(h.len); + else + len = obj_len(sb, h.type); + shadow->old_len = len + sizeof(h); + return 0; +} + +static void freeseg(struct super_block *sb, u32 segno) +{ + struct logfs_super *super = logfs_super(sb); + struct address_space *mapping = super->s_mapping_inode->i_mapping; + struct page *page; + u64 ofs, start, end; + + start = dev_ofs(sb, segno, 0); + end = dev_ofs(sb, segno + 1, 0); + for (ofs = start; ofs < end; ofs += PAGE_SIZE) { + page = find_get_page(mapping, ofs >> PAGE_SHIFT); + if (!page) + continue; + ClearPagePrivate(page); + page_cache_release(page); + } +} + +int logfs_open_area(struct logfs_area *area, size_t bytes) +{ + struct super_block *sb = area->a_sb; + struct logfs_super *super = logfs_super(sb); + int err, closed = 0; + + if (area->a_is_open && area->a_used_bytes + bytes <= super->s_segsize) + return 0; + + if (area->a_is_open) { + u64 ofs = dev_ofs(sb, area->a_segno, area->a_written_bytes); + u32 len = super->s_segsize - area->a_written_bytes; + + log_gc("logfs_close_area(%x)\n", area->a_segno); + pad_wbuf(area, 1); + super->s_devops->writeseg(area->a_sb, ofs, len); + freeseg(sb, area->a_segno); + closed = 1; + } + + area->a_used_bytes = 0; + area->a_written_bytes = 0; +again: + area->a_ops->get_free_segment(area); + area->a_ops->get_erase_count(area); + + log_gc("logfs_open_area(%x, %x)\n", area->a_segno, area->a_level); + err = area->a_ops->erase_segment(area); + if (err) { + printk(KERN_WARNING "LogFS: Error erasing segment %x\n", + area->a_segno); + logfs_mark_segment_bad(sb, area->a_segno); + goto again; + } + area->a_is_open = 1; + return closed; +} + +void logfs_sync_area(struct logfs_area *area) +{ + struct super_block *sb = area->a_sb; + struct logfs_super *super = logfs_super(sb); + u64 ofs = dev_ofs(sb, area->a_segno, area->a_written_bytes); + u32 len = (area->a_used_bytes - area->a_written_bytes); + + if (super->s_writesize) + len &= ~(super->s_writesize - 1); + if (len == 0) + return; + pad_wbuf(area, 0); + super->s_devops->writeseg(sb, ofs, len); + area->a_written_bytes += len; +} + +void logfs_sync_segments(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + int i; + + for_each_area(i) + logfs_sync_area(super->s_area[i]); +} + +/* + * Pick a free segment to be used for this area. Effectively takes a + * candidate from the free list (not really a candidate anymore). + */ +static void ostore_get_free_segment(struct logfs_area *area) +{ + struct super_block *sb = area->a_sb; + struct logfs_super *super = logfs_super(sb); + + if (super->s_free_list.count == 0) { + printk(KERN_ERR"LOGFS: ran out of free segments\n"); + LOGFS_BUG(sb); + } + + area->a_segno = get_best_cand(sb, &super->s_free_list, NULL); +} + +static void ostore_get_erase_count(struct logfs_area *area) +{ + struct logfs_segment_entry se; + u32 ec_level; + + logfs_get_segment_entry(area->a_sb, area->a_segno, &se); + BUG_ON(se.ec_level == cpu_to_be32(BADSEG) || + se.valid == cpu_to_be32(RESERVED)); + + ec_level = be32_to_cpu(se.ec_level); + area->a_erase_count = (ec_level >> 4) + 1; +} + +static int ostore_erase_segment(struct logfs_area *area) +{ + struct super_block *sb = area->a_sb; + struct logfs_segment_header sh; + u64 ofs; + int err; + + err = logfs_erase_segment(sb, area->a_segno, 0); + if (err) + return err; + + sh.pad = 0; + sh.type = SEG_OSTORE; + sh.level = (__force u8)area->a_level; + sh.segno = cpu_to_be32(area->a_segno); + sh.ec = cpu_to_be32(area->a_erase_count); + sh.gec = cpu_to_be64(logfs_super(sb)->s_gec); + sh.crc = logfs_crc32(&sh, sizeof(sh), 4); + + logfs_set_segment_erased(sb, area->a_segno, area->a_erase_count, + area->a_level); + + ofs = dev_ofs(sb, area->a_segno, 0); + area->a_used_bytes = sizeof(sh); + logfs_buf_write(area, ofs, &sh, sizeof(sh)); + return 0; +} + +static const struct logfs_area_ops ostore_area_ops = { + .get_free_segment = ostore_get_free_segment, + .get_erase_count = ostore_get_erase_count, + .erase_segment = ostore_erase_segment, +}; + +static void free_area(struct logfs_area *area) +{ + if (area) + freeseg(area->a_sb, area->a_segno); + kfree(area); +} + +static struct logfs_area *alloc_area(struct super_block *sb) +{ + struct logfs_area *area; + + area = kzalloc(sizeof(*area), GFP_KERNEL); + if (!area) + return NULL; + + area->a_sb = sb; + return area; +} + +static void map_invalidatepage(struct page *page, unsigned long l) +{ + BUG(); +} + +static int map_releasepage(struct page *page, gfp_t g) +{ + /* Don't release these pages */ + return 0; +} + +static const struct address_space_operations mapping_aops = { + .invalidatepage = map_invalidatepage, + .releasepage = map_releasepage, + .set_page_dirty = __set_page_dirty_nobuffers, +}; + +int logfs_init_mapping(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + struct address_space *mapping; + struct inode *inode; + + inode = logfs_new_meta_inode(sb, LOGFS_INO_MAPPING); + if (IS_ERR(inode)) + return PTR_ERR(inode); + super->s_mapping_inode = inode; + mapping = inode->i_mapping; + mapping->a_ops = &mapping_aops; + /* Would it be possible to use __GFP_HIGHMEM as well? */ + mapping_set_gfp_mask(mapping, GFP_NOFS); + return 0; +} + +int logfs_init_areas(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + int i = -1; + + super->s_alias_pool = mempool_create_kmalloc_pool(600, + sizeof(struct object_alias_item)); + if (!super->s_alias_pool) + return -ENOMEM; + + super->s_journal_area = alloc_area(sb); + if (!super->s_journal_area) + goto err; + + for_each_area(i) { + super->s_area[i] = alloc_area(sb); + if (!super->s_area[i]) + goto err; + super->s_area[i]->a_level = GC_LEVEL(i); + super->s_area[i]->a_ops = &ostore_area_ops; + } + btree_init_mempool128(&super->s_object_alias_tree, + super->s_btree_pool); + return 0; + +err: + for (i--; i >= 0; i--) + free_area(super->s_area[i]); + free_area(super->s_journal_area); + mempool_destroy(super->s_alias_pool); + return -ENOMEM; +} + +void logfs_cleanup_areas(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + int i; + + btree_grim_visitor128(&super->s_object_alias_tree, 0, kill_alias); + for_each_area(i) + free_area(super->s_area[i]); + free_area(super->s_journal_area); + destroy_meta_inode(super->s_mapping_inode); +} diff --git a/fs/logfs/super.c b/fs/logfs/super.c new file mode 100644 index 000000000000..c66beab78dee --- /dev/null +++ b/fs/logfs/super.c @@ -0,0 +1,650 @@ +/* + * fs/logfs/super.c + * + * As should be obvious for Linux kernel code, license is GPLv2 + * + * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org> + * + * Generally contains mount/umount code and also serves as a dump area for + * any functions that don't fit elsewhere and neither justify a file of their + * own. + */ +#include "logfs.h" +#include <linux/bio.h> +#include <linux/mtd/mtd.h> +#include <linux/statfs.h> +#include <linux/buffer_head.h> + +static DEFINE_MUTEX(emergency_mutex); +static struct page *emergency_page; + +struct page *emergency_read_begin(struct address_space *mapping, pgoff_t index) +{ + filler_t *filler = (filler_t *)mapping->a_ops->readpage; + struct page *page; + int err; + + page = read_cache_page(mapping, index, filler, NULL); + if (page) + return page; + + /* No more pages available, switch to emergency page */ + printk(KERN_INFO"Logfs: Using emergency page\n"); + mutex_lock(&emergency_mutex); + err = filler(NULL, emergency_page); + if (err) { + mutex_unlock(&emergency_mutex); + printk(KERN_EMERG"Logfs: Error reading emergency page\n"); + return ERR_PTR(err); + } + return emergency_page; +} + +void emergency_read_end(struct page *page) +{ + if (page == emergency_page) + mutex_unlock(&emergency_mutex); + else + page_cache_release(page); +} + +static void dump_segfile(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_segment_entry se; + u32 segno; + + for (segno = 0; segno < super->s_no_segs; segno++) { + logfs_get_segment_entry(sb, segno, &se); + printk("%3x: %6x %8x", segno, be32_to_cpu(se.ec_level), + be32_to_cpu(se.valid)); + if (++segno < super->s_no_segs) { + logfs_get_segment_entry(sb, segno, &se); + printk(" %6x %8x", be32_to_cpu(se.ec_level), + be32_to_cpu(se.valid)); + } + if (++segno < super->s_no_segs) { + logfs_get_segment_entry(sb, segno, &se); + printk(" %6x %8x", be32_to_cpu(se.ec_level), + be32_to_cpu(se.valid)); + } + if (++segno < super->s_no_segs) { + logfs_get_segment_entry(sb, segno, &se); + printk(" %6x %8x", be32_to_cpu(se.ec_level), + be32_to_cpu(se.valid)); + } + printk("\n"); + } +} + +/* + * logfs_crash_dump - dump debug information to device + * + * The LogFS superblock only occupies part of a segment. This function will + * write as much debug information as it can gather into the spare space. + */ +void logfs_crash_dump(struct super_block *sb) +{ + dump_segfile(sb); +} + +/* + * TODO: move to lib/string.c + */ +/** + * memchr_inv - Find a character in an area of memory. + * @s: The memory area + * @c: The byte to search for + * @n: The size of the area. + * + * returns the address of the first character other than @c, or %NULL + * if the whole buffer contains just @c. + */ +void *memchr_inv(const void *s, int c, size_t n) +{ + const unsigned char *p = s; + while (n-- != 0) + if ((unsigned char)c != *p++) + return (void *)(p - 1); + + return NULL; +} + +/* + * FIXME: There should be a reserve for root, similar to ext2. + */ +int logfs_statfs(struct dentry *dentry, struct kstatfs *stats) +{ + struct super_block *sb = dentry->d_sb; + struct logfs_super *super = logfs_super(sb); + + stats->f_type = LOGFS_MAGIC_U32; + stats->f_bsize = sb->s_blocksize; + stats->f_blocks = super->s_size >> LOGFS_BLOCK_BITS >> 3; + stats->f_bfree = super->s_free_bytes >> sb->s_blocksize_bits; + stats->f_bavail = super->s_free_bytes >> sb->s_blocksize_bits; + stats->f_files = 0; + stats->f_ffree = 0; + stats->f_namelen = LOGFS_MAX_NAMELEN; + return 0; +} + +static int logfs_sb_set(struct super_block *sb, void *_super) +{ + struct logfs_super *super = _super; + + sb->s_fs_info = super; + sb->s_mtd = super->s_mtd; + sb->s_bdev = super->s_bdev; + return 0; +} + +static int logfs_sb_test(struct super_block *sb, void *_super) +{ + struct logfs_super *super = _super; + struct mtd_info *mtd = super->s_mtd; + + if (mtd && sb->s_mtd == mtd) + return 1; + if (super->s_bdev && sb->s_bdev == super->s_bdev) + return 1; + return 0; +} + +static void set_segment_header(struct logfs_segment_header *sh, u8 type, + u8 level, u32 segno, u32 ec) +{ + sh->pad = 0; + sh->type = type; + sh->level = level; + sh->segno = cpu_to_be32(segno); + sh->ec = cpu_to_be32(ec); + sh->gec = cpu_to_be64(segno); + sh->crc = logfs_crc32(sh, LOGFS_SEGMENT_HEADERSIZE, 4); +} + +static void logfs_write_ds(struct super_block *sb, struct logfs_disk_super *ds, + u32 segno, u32 ec) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_segment_header *sh = &ds->ds_sh; + int i; + + memset(ds, 0, sizeof(*ds)); + set_segment_header(sh, SEG_SUPER, 0, segno, ec); + + ds->ds_ifile_levels = super->s_ifile_levels; + ds->ds_iblock_levels = super->s_iblock_levels; + ds->ds_data_levels = super->s_data_levels; /* XXX: Remove */ + ds->ds_segment_shift = super->s_segshift; + ds->ds_block_shift = sb->s_blocksize_bits; + ds->ds_write_shift = super->s_writeshift; + ds->ds_filesystem_size = cpu_to_be64(super->s_size); + ds->ds_segment_size = cpu_to_be32(super->s_segsize); + ds->ds_bad_seg_reserve = cpu_to_be32(super->s_bad_seg_reserve); + ds->ds_feature_incompat = cpu_to_be64(super->s_feature_incompat); + ds->ds_feature_ro_compat= cpu_to_be64(super->s_feature_ro_compat); + ds->ds_feature_compat = cpu_to_be64(super->s_feature_compat); + ds->ds_feature_flags = cpu_to_be64(super->s_feature_flags); + ds->ds_root_reserve = cpu_to_be64(super->s_root_reserve); + ds->ds_speed_reserve = cpu_to_be64(super->s_speed_reserve); + journal_for_each(i) + ds->ds_journal_seg[i] = cpu_to_be32(super->s_journal_seg[i]); + ds->ds_magic = cpu_to_be64(LOGFS_MAGIC); + ds->ds_crc = logfs_crc32(ds, sizeof(*ds), + LOGFS_SEGMENT_HEADERSIZE + 12); +} + +static int write_one_sb(struct super_block *sb, + struct page *(*find_sb)(struct super_block *sb, u64 *ofs)) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_disk_super *ds; + struct logfs_segment_entry se; + struct page *page; + u64 ofs; + u32 ec, segno; + int err; + + page = find_sb(sb, &ofs); + if (!page) + return -EIO; + ds = page_address(page); + segno = seg_no(sb, ofs); + logfs_get_segment_entry(sb, segno, &se); + ec = be32_to_cpu(se.ec_level) >> 4; + ec++; + logfs_set_segment_erased(sb, segno, ec, 0); + logfs_write_ds(sb, ds, segno, ec); + err = super->s_devops->write_sb(sb, page); + page_cache_release(page); + return err; +} + +int logfs_write_sb(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + int err; + + /* First superblock */ + err = write_one_sb(sb, super->s_devops->find_first_sb); + if (err) + return err; + + /* Last superblock */ + err = write_one_sb(sb, super->s_devops->find_last_sb); + if (err) + return err; + return 0; +} + +static int ds_cmp(const void *ds0, const void *ds1) +{ + size_t len = sizeof(struct logfs_disk_super); + + /* We know the segment headers differ, so ignore them */ + len -= LOGFS_SEGMENT_HEADERSIZE; + ds0 += LOGFS_SEGMENT_HEADERSIZE; + ds1 += LOGFS_SEGMENT_HEADERSIZE; + return memcmp(ds0, ds1, len); +} + +static int logfs_recover_sb(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + struct logfs_disk_super _ds0, *ds0 = &_ds0; + struct logfs_disk_super _ds1, *ds1 = &_ds1; + int err, valid0, valid1; + + /* read first superblock */ + err = wbuf_read(sb, super->s_sb_ofs[0], sizeof(*ds0), ds0); + if (err) + return err; + /* read last superblock */ + err = wbuf_read(sb, super->s_sb_ofs[1], sizeof(*ds1), ds1); + if (err) + return err; + valid0 = logfs_check_ds(ds0) == 0; + valid1 = logfs_check_ds(ds1) == 0; + + if (!valid0 && valid1) { + printk(KERN_INFO"First superblock is invalid - fixing.\n"); + return write_one_sb(sb, super->s_devops->find_first_sb); + } + if (valid0 && !valid1) { + printk(KERN_INFO"Last superblock is invalid - fixing.\n"); + return write_one_sb(sb, super->s_devops->find_last_sb); + } + if (valid0 && valid1 && ds_cmp(ds0, ds1)) { + printk(KERN_INFO"Superblocks don't match - fixing.\n"); + return write_one_sb(sb, super->s_devops->find_last_sb); + } + /* If neither is valid now, something's wrong. Didn't we properly + * check them before?!? */ + BUG_ON(!valid0 && !valid1); + return 0; +} + +static int logfs_make_writeable(struct super_block *sb) +{ + int err; + + /* Repair any broken superblock copies */ + err = logfs_recover_sb(sb); + if (err) + return err; + + /* Check areas for trailing unaccounted data */ + err = logfs_check_areas(sb); + if (err) + return err; + + err = logfs_open_segfile(sb); + if (err) + return err; + + /* Do one GC pass before any data gets dirtied */ + logfs_gc_pass(sb); + + /* after all initializations are done, replay the journal + * for rw-mounts, if necessary */ + err = logfs_replay_journal(sb); + if (err) + return err; + + return 0; +} + +static int logfs_get_sb_final(struct super_block *sb, struct vfsmount *mnt) +{ + struct logfs_super *super = logfs_super(sb); + struct inode *rootdir; + int err; + + /* root dir */ + rootdir = logfs_iget(sb, LOGFS_INO_ROOT); + if (IS_ERR(rootdir)) + goto fail; + + sb->s_root = d_alloc_root(rootdir); + if (!sb->s_root) + goto fail; + + super->s_erase_page = alloc_pages(GFP_KERNEL, 0); + if (!super->s_erase_page) + goto fail2; + memset(page_address(super->s_erase_page), 0xFF, PAGE_SIZE); + + /* FIXME: check for read-only mounts */ + err = logfs_make_writeable(sb); + if (err) + goto fail3; + + log_super("LogFS: Finished mounting\n"); + simple_set_mnt(mnt, sb); + return 0; + +fail3: + __free_page(super->s_erase_page); +fail2: + iput(rootdir); +fail: + iput(logfs_super(sb)->s_master_inode); + return -EIO; +} + +int logfs_check_ds(struct logfs_disk_super *ds) +{ + struct logfs_segment_header *sh = &ds->ds_sh; + + if (ds->ds_magic != cpu_to_be64(LOGFS_MAGIC)) + return -EINVAL; + if (sh->crc != logfs_crc32(sh, LOGFS_SEGMENT_HEADERSIZE, 4)) + return -EINVAL; + if (ds->ds_crc != logfs_crc32(ds, sizeof(*ds), + LOGFS_SEGMENT_HEADERSIZE + 12)) + return -EINVAL; + return 0; +} + +static struct page *find_super_block(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + struct page *first, *last; + + first = super->s_devops->find_first_sb(sb, &super->s_sb_ofs[0]); + if (!first || IS_ERR(first)) + return NULL; + last = super->s_devops->find_last_sb(sb, &super->s_sb_ofs[1]); + if (!last || IS_ERR(first)) { + page_cache_release(first); + return NULL; + } + + if (!logfs_check_ds(page_address(first))) { + page_cache_release(last); + return first; + } + + /* First one didn't work, try the second superblock */ + if (!logfs_check_ds(page_address(last))) { + page_cache_release(first); + return last; + } + + /* Neither worked, sorry folks */ + page_cache_release(first); + page_cache_release(last); + return NULL; +} + +static int __logfs_read_sb(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + struct page *page; + struct logfs_disk_super *ds; + int i; + + page = find_super_block(sb); + if (!page) + return -EIO; + + ds = page_address(page); + super->s_size = be64_to_cpu(ds->ds_filesystem_size); + super->s_root_reserve = be64_to_cpu(ds->ds_root_reserve); + super->s_speed_reserve = be64_to_cpu(ds->ds_speed_reserve); + super->s_bad_seg_reserve = be32_to_cpu(ds->ds_bad_seg_reserve); + super->s_segsize = 1 << ds->ds_segment_shift; + super->s_segmask = (1 << ds->ds_segment_shift) - 1; + super->s_segshift = ds->ds_segment_shift; + sb->s_blocksize = 1 << ds->ds_block_shift; + sb->s_blocksize_bits = ds->ds_block_shift; + super->s_writesize = 1 << ds->ds_write_shift; + super->s_writeshift = ds->ds_write_shift; + super->s_no_segs = super->s_size >> super->s_segshift; + super->s_no_blocks = super->s_segsize >> sb->s_blocksize_bits; + super->s_feature_incompat = be64_to_cpu(ds->ds_feature_incompat); + super->s_feature_ro_compat = be64_to_cpu(ds->ds_feature_ro_compat); + super->s_feature_compat = be64_to_cpu(ds->ds_feature_compat); + super->s_feature_flags = be64_to_cpu(ds->ds_feature_flags); + + journal_for_each(i) + super->s_journal_seg[i] = be32_to_cpu(ds->ds_journal_seg[i]); + + super->s_ifile_levels = ds->ds_ifile_levels; + super->s_iblock_levels = ds->ds_iblock_levels; + super->s_data_levels = ds->ds_data_levels; + super->s_total_levels = super->s_ifile_levels + super->s_iblock_levels + + super->s_data_levels; + page_cache_release(page); + return 0; +} + +static int logfs_read_sb(struct super_block *sb, int read_only) +{ + struct logfs_super *super = logfs_super(sb); + int ret; + + super->s_btree_pool = mempool_create(32, btree_alloc, btree_free, NULL); + if (!super->s_btree_pool) + return -ENOMEM; + + btree_init_mempool64(&super->s_shadow_tree.new, super->s_btree_pool); + btree_init_mempool64(&super->s_shadow_tree.old, super->s_btree_pool); + + ret = logfs_init_mapping(sb); + if (ret) + return ret; + + ret = __logfs_read_sb(sb); + if (ret) + return ret; + + if (super->s_feature_incompat & ~LOGFS_FEATURES_INCOMPAT) + return -EIO; + if ((super->s_feature_ro_compat & ~LOGFS_FEATURES_RO_COMPAT) && + !read_only) + return -EIO; + + mutex_init(&super->s_dirop_mutex); + mutex_init(&super->s_object_alias_mutex); + INIT_LIST_HEAD(&super->s_freeing_list); + + ret = logfs_init_rw(sb); + if (ret) + return ret; + + ret = logfs_init_areas(sb); + if (ret) + return ret; + + ret = logfs_init_gc(sb); + if (ret) + return ret; + + ret = logfs_init_journal(sb); + if (ret) + return ret; + + return 0; +} + +static void logfs_kill_sb(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + + log_super("LogFS: Start unmounting\n"); + /* Alias entries slow down mount, so evict as many as possible */ + sync_filesystem(sb); + logfs_write_anchor(sb); + + /* + * From this point on alias entries are simply dropped - and any + * writes to the object store are considered bugs. + */ + super->s_flags |= LOGFS_SB_FLAG_SHUTDOWN; + log_super("LogFS: Now in shutdown\n"); + generic_shutdown_super(sb); + + BUG_ON(super->s_dirty_used_bytes || super->s_dirty_free_bytes); + + logfs_cleanup_gc(sb); + logfs_cleanup_journal(sb); + logfs_cleanup_areas(sb); + logfs_cleanup_rw(sb); + if (super->s_erase_page) + __free_page(super->s_erase_page); + super->s_devops->put_device(sb); + mempool_destroy(super->s_btree_pool); + mempool_destroy(super->s_alias_pool); + kfree(super); + log_super("LogFS: Finished unmounting\n"); +} + +int logfs_get_sb_device(struct file_system_type *type, int flags, + struct mtd_info *mtd, struct block_device *bdev, + const struct logfs_device_ops *devops, struct vfsmount *mnt) +{ + struct logfs_super *super; + struct super_block *sb; + int err = -ENOMEM; + static int mount_count; + + log_super("LogFS: Start mount %x\n", mount_count++); + super = kzalloc(sizeof(*super), GFP_KERNEL); + if (!super) + goto err0; + + super->s_mtd = mtd; + super->s_bdev = bdev; + err = -EINVAL; + sb = sget(type, logfs_sb_test, logfs_sb_set, super); + if (IS_ERR(sb)) + goto err0; + + if (sb->s_root) { + /* Device is already in use */ + err = 0; + simple_set_mnt(mnt, sb); + goto err0; + } + + super->s_devops = devops; + + /* + * sb->s_maxbytes is limited to 8TB. On 32bit systems, the page cache + * only covers 16TB and the upper 8TB are used for indirect blocks. + * On 64bit system we could bump up the limit, but that would make + * the filesystem incompatible with 32bit systems. + */ + sb->s_maxbytes = (1ull << 43) - 1; + sb->s_op = &logfs_super_operations; + sb->s_flags = flags | MS_NOATIME; + + err = logfs_read_sb(sb, sb->s_flags & MS_RDONLY); + if (err) + goto err1; + + sb->s_flags |= MS_ACTIVE; + err = logfs_get_sb_final(sb, mnt); + if (err) + goto err1; + return 0; + +err1: + up_write(&sb->s_umount); + deactivate_super(sb); + return err; +err0: + kfree(super); + //devops->put_device(sb); + return err; +} + +static int logfs_get_sb(struct file_system_type *type, int flags, + const char *devname, void *data, struct vfsmount *mnt) +{ + ulong mtdnr; + + if (!devname) + return logfs_get_sb_bdev(type, flags, devname, mnt); + if (strncmp(devname, "mtd", 3)) + return logfs_get_sb_bdev(type, flags, devname, mnt); + + { + char *garbage; + mtdnr = simple_strtoul(devname+3, &garbage, 0); + if (*garbage) + return -EINVAL; + } + + return logfs_get_sb_mtd(type, flags, mtdnr, mnt); +} + +static struct file_system_type logfs_fs_type = { + .owner = THIS_MODULE, + .name = "logfs", + .get_sb = logfs_get_sb, + .kill_sb = logfs_kill_sb, + .fs_flags = FS_REQUIRES_DEV, + +}; + +static int __init logfs_init(void) +{ + int ret; + + emergency_page = alloc_pages(GFP_KERNEL, 0); + if (!emergency_page) + return -ENOMEM; + + ret = logfs_compr_init(); + if (ret) + goto out1; + + ret = logfs_init_inode_cache(); + if (ret) + goto out2; + + return register_filesystem(&logfs_fs_type); +out2: + logfs_compr_exit(); +out1: + __free_pages(emergency_page, 0); + return ret; +} + +static void __exit logfs_exit(void) +{ + unregister_filesystem(&logfs_fs_type); + logfs_destroy_inode_cache(); + logfs_compr_exit(); + __free_pages(emergency_page, 0); +} + +module_init(logfs_init); +module_exit(logfs_exit); + +MODULE_LICENSE("GPL v2"); +MODULE_AUTHOR("Joern Engel <joern@logfs.org>"); +MODULE_DESCRIPTION("scalable flash filesystem"); diff --git a/fs/namei.c b/fs/namei.c index 3d9d2f965f84..48e60a187325 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1656,7 +1656,7 @@ static struct file *do_last(struct nameidata *nd, struct path *path, if (path->dentry->d_inode->i_op->follow_link) return NULL; error = -ENOTDIR; - if (*want_dir & !path->dentry->d_inode->i_op->lookup) + if (*want_dir && !path->dentry->d_inode->i_op->lookup) goto exit_dput; path_to_nameidata(path, nd); audit_inode(pathname, nd->path.dentry); diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c index 73ab220354df..36dfdae95123 100644 --- a/fs/nfs/callback.c +++ b/fs/nfs/callback.c @@ -118,7 +118,6 @@ nfs4_callback_up(struct svc_serv *serv) dprintk("NFS: Callback listener port = %u (af %u)\n", nfs_callback_tcpport, PF_INET); -#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) ret = svc_create_xprt(serv, "tcp", PF_INET6, nfs_callback_set_tcpport, SVC_SOCK_ANONYMOUS); if (ret > 0) { @@ -129,7 +128,6 @@ nfs4_callback_up(struct svc_serv *serv) ret = 0; else goto out_err; -#endif /* defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) */ return svc_prepare_thread(serv, &serv->sv_pools[0]); diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index c6eed2a3b093..4bc22c763de7 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -525,6 +525,8 @@ static struct rpc_cred *callback_cred; int set_callback_cred(void) { + if (callback_cred) + return 0; callback_cred = rpc_lookup_machine_cred(); if (!callback_cred) return -ENOMEM; @@ -542,7 +544,8 @@ void do_probe_callback(struct nfs4_client *clp) }; int status; - status = rpc_call_async(cb->cb_client, &msg, RPC_TASK_SOFT, + status = rpc_call_async(cb->cb_client, &msg, + RPC_TASK_SOFT | RPC_TASK_SOFTCONN, &nfsd4_cb_probe_ops, (void *)clp); if (status) { warn_no_callback_path(clp, status); diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index 5a754f7b71ed..98fb98e330b4 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -119,9 +119,7 @@ out_no_tfm: static void nfsd4_sync_rec_dir(void) { - mutex_lock(&rec_dir.dentry->d_inode->i_mutex); - nfsd_sync_dir(rec_dir.dentry); - mutex_unlock(&rec_dir.dentry->d_inode->i_mutex); + vfs_fsync(NULL, rec_dir.dentry, 0); } int diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index f19ed866c95f..c97fddbd17db 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -1998,7 +1998,9 @@ nfs4_file_downgrade(struct file *filp, unsigned int share_access) { if (share_access & NFS4_SHARE_ACCESS_WRITE) { drop_file_write_access(filp); + spin_lock(&filp->f_lock); filp->f_mode = (filp->f_mode | FMODE_READ) & ~FMODE_WRITE; + spin_unlock(&filp->f_lock); } } @@ -2480,8 +2482,10 @@ nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nf } memcpy(&open->op_stateid, &stp->st_stateid, sizeof(stateid_t)); - if (nfsd4_has_session(&resp->cstate)) + if (nfsd4_has_session(&resp->cstate)) { open->op_stateowner->so_confirmed = 1; + nfsd4_create_clid_dir(open->op_stateowner->so_client); + } /* * Attempt to hand out a delegation. No error return, because the diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index bbf72d8f9fc0..78c7e24e5129 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -1434,7 +1434,7 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp) } op->opnum = ntohl(*argp->p++); - if (op->opnum >= OP_ACCESS && op->opnum < ops->nops) + if (op->opnum >= FIRST_NFS4_OP && op->opnum <= LAST_NFS4_OP) op->status = ops->decoders[op->opnum](argp, &op->u); else { op->opnum = OP_ILLEGAL; diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 2604c3e70ea5..0f0e77f2012f 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -988,6 +988,7 @@ static ssize_t __write_ports_delfd(char *buf) static ssize_t __write_ports_addxprt(char *buf) { char transport[16]; + struct svc_xprt *xprt; int port, err; if (sscanf(buf, "%15s %4u", transport, &port) != 2) @@ -1002,13 +1003,24 @@ static ssize_t __write_ports_addxprt(char *buf) err = svc_create_xprt(nfsd_serv, transport, PF_INET, port, SVC_SOCK_ANONYMOUS); - if (err < 0) { - /* Give a reasonable perror msg for bad transport string */ - if (err == -ENOENT) - err = -EPROTONOSUPPORT; - return err; - } + if (err < 0) + goto out_err; + + err = svc_create_xprt(nfsd_serv, transport, + PF_INET6, port, SVC_SOCK_ANONYMOUS); + if (err < 0 && err != -EAFNOSUPPORT) + goto out_close; return 0; +out_close: + xprt = svc_find_xprt(nfsd_serv, transport, PF_INET, port); + if (xprt != NULL) { + svc_close_xprt(xprt); + svc_xprt_put(xprt); + } +out_err: + /* Decrease the count, but don't shut down the service */ + nfsd_serv->sv_nrthreads--; + return err; } /* diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 8eca17df4f63..a11b0e8678ee 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -26,6 +26,8 @@ #include <linux/jhash.h> #include <linux/ima.h> #include <asm/uaccess.h> +#include <linux/exportfs.h> +#include <linux/writeback.h> #ifdef CONFIG_NFSD_V3 #include "xdr3.h" @@ -270,6 +272,32 @@ out: return err; } +/* + * Commit metadata changes to stable storage. + */ +static int +commit_metadata(struct svc_fh *fhp) +{ + struct inode *inode = fhp->fh_dentry->d_inode; + const struct export_operations *export_ops = inode->i_sb->s_export_op; + int error = 0; + + if (!EX_ISSYNC(fhp->fh_export)) + return 0; + + if (export_ops->commit_metadata) { + error = export_ops->commit_metadata(inode); + } else { + struct writeback_control wbc = { + .sync_mode = WB_SYNC_ALL, + .nr_to_write = 0, /* metadata only */ + }; + + error = sync_inode(inode, &wbc); + } + + return error; +} /* * Set various file attributes. @@ -767,43 +795,6 @@ nfsd_close(struct file *filp) } /* - * Sync a file - * As this calls fsync (not fdatasync) there is no need for a write_inode - * after it. - */ -static inline int nfsd_dosync(struct file *filp, struct dentry *dp, - const struct file_operations *fop) -{ - struct inode *inode = dp->d_inode; - int (*fsync) (struct file *, struct dentry *, int); - int err; - - err = filemap_write_and_wait(inode->i_mapping); - if (err == 0 && fop && (fsync = fop->fsync)) - err = fsync(filp, dp, 0); - return err; -} - -static int -nfsd_sync(struct file *filp) -{ - int err; - struct inode *inode = filp->f_path.dentry->d_inode; - dprintk("nfsd: sync file %s\n", filp->f_path.dentry->d_name.name); - mutex_lock(&inode->i_mutex); - err=nfsd_dosync(filp, filp->f_path.dentry, filp->f_op); - mutex_unlock(&inode->i_mutex); - - return err; -} - -int -nfsd_sync_dir(struct dentry *dp) -{ - return nfsd_dosync(NULL, dp, dp->d_inode->i_fop); -} - -/* * Obtain the readahead parameters for the file * specified by (dev, ino). */ @@ -1006,7 +997,7 @@ static int wait_for_concurrent_writes(struct file *file) if (inode->i_state & I_DIRTY) { dprintk("nfsd: write sync %d\n", task_pid_nr(current)); - err = nfsd_sync(file); + err = vfs_fsync(file, file->f_path.dentry, 0); } last_ino = inode->i_ino; last_dev = inode->i_sb->s_dev; @@ -1154,8 +1145,9 @@ out: #ifdef CONFIG_NFSD_V3 /* * Commit all pending writes to stable storage. - * Strictly speaking, we could sync just the indicated file region here, - * but there's currently no way we can ask the VFS to do so. + * + * Note: we only guarantee that data that lies within the range specified + * by the 'offset' and 'count' parameters will be synced. * * Unfortunately we cannot lock the file to make sure we return full WCC * data to the client, as locking happens lower down in the filesystem. @@ -1165,23 +1157,32 @@ nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t offset, unsigned long count) { struct file *file; - __be32 err; + loff_t end = LLONG_MAX; + __be32 err = nfserr_inval; - if ((u64)count > ~(u64)offset) - return nfserr_inval; + if (offset < 0) + goto out; + if (count != 0) { + end = offset + (loff_t)count - 1; + if (end < offset) + goto out; + } err = nfsd_open(rqstp, fhp, S_IFREG, NFSD_MAY_WRITE, &file); if (err) - return err; + goto out; if (EX_ISSYNC(fhp->fh_export)) { - if (file->f_op && file->f_op->fsync) { - err = nfserrno(nfsd_sync(file)); - } else { + int err2 = vfs_fsync_range(file, file->f_path.dentry, + offset, end, 0); + + if (err2 != -EINVAL) + err = nfserrno(err2); + else err = nfserr_notsupp; - } } nfsd_close(file); +out: return err; } #endif /* CONFIG_NFSD_V3 */ @@ -1334,12 +1335,14 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, goto out_nfserr; } - if (EX_ISSYNC(fhp->fh_export)) { - err = nfserrno(nfsd_sync_dir(dentry)); - write_inode_now(dchild->d_inode, 1); - } + err = nfsd_create_setattr(rqstp, resfhp, iap); - err2 = nfsd_create_setattr(rqstp, resfhp, iap); + /* + * nfsd_setattr already committed the child. Transactional filesystems + * had a chance to commit changes for both parent and child + * simultaneously making the following commit_metadata a noop. + */ + err2 = nfserrno(commit_metadata(fhp)); if (err2) err = err2; mnt_drop_write(fhp->fh_export->ex_path.mnt); @@ -1371,7 +1374,6 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, struct dentry *dentry, *dchild = NULL; struct inode *dirp; __be32 err; - __be32 err2; int host_err; __u32 v_mtime=0, v_atime=0; @@ -1466,11 +1468,6 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, if (created) *created = 1; - if (EX_ISSYNC(fhp->fh_export)) { - err = nfserrno(nfsd_sync_dir(dentry)); - /* setattr will sync the child (or not) */ - } - nfsd_check_ignore_resizing(iap); if (createmode == NFS3_CREATE_EXCLUSIVE) { @@ -1485,9 +1482,13 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, } set_attr: - err2 = nfsd_create_setattr(rqstp, resfhp, iap); - if (err2) - err = err2; + err = nfsd_create_setattr(rqstp, resfhp, iap); + + /* + * nfsd_setattr already committed the child (and possibly also the parent). + */ + if (!err) + err = nfserrno(commit_metadata(fhp)); mnt_drop_write(fhp->fh_export->ex_path.mnt); /* @@ -1602,12 +1603,9 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, } } else host_err = vfs_symlink(dentry->d_inode, dnew, path); - - if (!host_err) { - if (EX_ISSYNC(fhp->fh_export)) - host_err = nfsd_sync_dir(dentry); - } err = nfserrno(host_err); + if (!err) + err = nfserrno(commit_metadata(fhp)); fh_unlock(fhp); mnt_drop_write(fhp->fh_export->ex_path.mnt); @@ -1669,11 +1667,9 @@ nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp, } host_err = vfs_link(dold, dirp, dnew); if (!host_err) { - if (EX_ISSYNC(ffhp->fh_export)) { - err = nfserrno(nfsd_sync_dir(ddir)); - write_inode_now(dest, 1); - } - err = 0; + err = nfserrno(commit_metadata(ffhp)); + if (!err) + err = nfserrno(commit_metadata(tfhp)); } else { if (host_err == -EXDEV && rqstp->rq_vers == 2) err = nfserr_acces; @@ -1769,10 +1765,10 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, goto out_dput_new; host_err = vfs_rename(fdir, odentry, tdir, ndentry); - if (!host_err && EX_ISSYNC(tfhp->fh_export)) { - host_err = nfsd_sync_dir(tdentry); + if (!host_err) { + host_err = commit_metadata(tfhp); if (!host_err) - host_err = nfsd_sync_dir(fdentry); + host_err = commit_metadata(ffhp); } mnt_drop_write(ffhp->fh_export->ex_path.mnt); @@ -1853,12 +1849,9 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, dput(rdentry); - if (host_err) - goto out_drop; - if (EX_ISSYNC(fhp->fh_export)) - host_err = nfsd_sync_dir(dentry); + if (!host_err) + host_err = commit_metadata(fhp); -out_drop: mnt_drop_write(fhp->fh_export->ex_path.mnt); out_nfserr: err = nfserrno(host_err); diff --git a/fs/ocfs2/quota_local.c b/fs/ocfs2/quota_local.c index 21f9e71223ca..a6467f3d262e 100644 --- a/fs/ocfs2/quota_local.c +++ b/fs/ocfs2/quota_local.c @@ -457,7 +457,7 @@ static int ocfs2_recover_local_quota_file(struct inode *lqinode, break; } dchunk = (struct ocfs2_local_disk_chunk *)hbh->b_data; - for_each_bit(bit, rchunk->rc_bitmap, ol_chunk_entries(sb)) { + for_each_set_bit(bit, rchunk->rc_bitmap, ol_chunk_entries(sb)) { qbh = NULL; status = ocfs2_read_quota_block(lqinode, ol_dqblk_block(sb, chunk, bit), diff --git a/fs/proc/array.c b/fs/proc/array.c index 18e20feee251..aa8637b81028 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -273,7 +273,7 @@ static inline void task_sig(struct seq_file *m, struct task_struct *p) rcu_read_lock(); /* FIXME: is this correct? */ qsize = atomic_read(&__task_cred(p)->user->sigpending); rcu_read_unlock(); - qlim = p->signal->rlim[RLIMIT_SIGPENDING].rlim_cur; + qlim = task_rlimit(p, RLIMIT_SIGPENDING); unlock_task_sighand(p, &flags); } @@ -420,7 +420,7 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, cutime = sig->cutime; cstime = sig->cstime; cgtime = sig->cgtime; - rsslim = sig->rlim[RLIMIT_RSS].rlim_cur; + rsslim = ACCESS_ONCE(sig->rlim[RLIMIT_RSS].rlim_cur); /* add up live thread stats at the group level */ if (whole) { diff --git a/fs/proc/generic.c b/fs/proc/generic.c index 9580abeadeb3..08f4d71dacd7 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -291,19 +291,17 @@ static const struct inode_operations proc_file_inode_operations = { * returns the struct proc_dir_entry for "/proc/tty/driver", and * returns "serial" in residual. */ -static int xlate_proc_name(const char *name, - struct proc_dir_entry **ret, const char **residual) +static int __xlate_proc_name(const char *name, struct proc_dir_entry **ret, + const char **residual) { const char *cp = name, *next; struct proc_dir_entry *de; int len; - int rtn = 0; de = *ret; if (!de) de = &proc_root; - spin_lock(&proc_subdir_lock); while (1) { next = strchr(cp, '/'); if (!next) @@ -315,16 +313,25 @@ static int xlate_proc_name(const char *name, break; } if (!de) { - rtn = -ENOENT; - goto out; + WARN(1, "name '%s'\n", name); + return -ENOENT; } cp += len + 1; } *residual = cp; *ret = de; -out: + return 0; +} + +static int xlate_proc_name(const char *name, struct proc_dir_entry **ret, + const char **residual) +{ + int rv; + + spin_lock(&proc_subdir_lock); + rv = __xlate_proc_name(name, ret, residual); spin_unlock(&proc_subdir_lock); - return rtn; + return rv; } static DEFINE_IDA(proc_inum_ida); @@ -797,11 +804,13 @@ void remove_proc_entry(const char *name, struct proc_dir_entry *parent) const char *fn = name; int len; - if (xlate_proc_name(name, &parent, &fn) != 0) + spin_lock(&proc_subdir_lock); + if (__xlate_proc_name(name, &parent, &fn) != 0) { + spin_unlock(&proc_subdir_lock); return; + } len = strlen(fn); - spin_lock(&proc_subdir_lock); for (p = &parent->subdir; *p; p=&(*p)->next ) { if (proc_match(len, fn, *p)) { de = *p; @@ -811,8 +820,10 @@ void remove_proc_entry(const char *name, struct proc_dir_entry *parent) } } spin_unlock(&proc_subdir_lock); - if (!de) + if (!de) { + WARN(1, "name '%s'\n", name); return; + } spin_lock(&de->pde_unload_lock); /* diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index f277c4a111cb..183f8ff5f400 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -16,7 +16,7 @@ void task_mem(struct seq_file *m, struct mm_struct *mm) { - unsigned long data, text, lib; + unsigned long data, text, lib, swap; unsigned long hiwater_vm, total_vm, hiwater_rss, total_rss; /* @@ -36,6 +36,7 @@ void task_mem(struct seq_file *m, struct mm_struct *mm) data = mm->total_vm - mm->shared_vm - mm->stack_vm; text = (PAGE_ALIGN(mm->end_code) - (mm->start_code & PAGE_MASK)) >> 10; lib = (mm->exec_vm << (PAGE_SHIFT-10)) - text; + swap = get_mm_counter(mm, MM_SWAPENTS); seq_printf(m, "VmPeak:\t%8lu kB\n" "VmSize:\t%8lu kB\n" @@ -46,7 +47,8 @@ void task_mem(struct seq_file *m, struct mm_struct *mm) "VmStk:\t%8lu kB\n" "VmExe:\t%8lu kB\n" "VmLib:\t%8lu kB\n" - "VmPTE:\t%8lu kB\n", + "VmPTE:\t%8lu kB\n" + "VmSwap:\t%8lu kB\n", hiwater_vm << (PAGE_SHIFT-10), (total_vm - mm->reserved_vm) << (PAGE_SHIFT-10), mm->locked_vm << (PAGE_SHIFT-10), @@ -54,7 +56,8 @@ void task_mem(struct seq_file *m, struct mm_struct *mm) total_rss << (PAGE_SHIFT-10), data << (PAGE_SHIFT-10), mm->stack_vm << (PAGE_SHIFT-10), text, lib, - (PTRS_PER_PTE*sizeof(pte_t)*mm->nr_ptes) >> 10); + (PTRS_PER_PTE*sizeof(pte_t)*mm->nr_ptes) >> 10, + swap << (PAGE_SHIFT-10)); } unsigned long task_vsize(struct mm_struct *mm) @@ -65,11 +68,11 @@ unsigned long task_vsize(struct mm_struct *mm) int task_statm(struct mm_struct *mm, int *shared, int *text, int *data, int *resident) { - *shared = get_mm_counter(mm, file_rss); + *shared = get_mm_counter(mm, MM_FILEPAGES); *text = (PAGE_ALIGN(mm->end_code) - (mm->start_code & PAGE_MASK)) >> PAGE_SHIFT; *data = mm->total_vm - mm->shared_vm; - *resident = *shared + get_mm_counter(mm, anon_rss); + *resident = *shared + get_mm_counter(mm, MM_ANONPAGES); return mm->total_vm; } diff --git a/fs/select.c b/fs/select.c index fd38ce2e32e3..73715e90030f 100644 --- a/fs/select.c +++ b/fs/select.c @@ -821,7 +821,7 @@ int do_sys_poll(struct pollfd __user *ufds, unsigned int nfds, struct poll_list *walk = head; unsigned long todo = nfds; - if (nfds > current->signal->rlim[RLIMIT_NOFILE].rlim_cur) + if (nfds > rlimit(RLIMIT_NOFILE)) return -EINVAL; len = min_t(unsigned int, nfds, N_STACK_PPS); diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 5c5a366aa332..b4769e40e8bc 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -105,7 +105,6 @@ xfs-y += $(addprefix $(XFS_LINUX)/, \ xfs_globals.o \ xfs_ioctl.o \ xfs_iops.o \ - xfs_lrw.o \ xfs_super.o \ xfs_sync.o \ xfs_xattr.o) diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index 66abe36c1213..9083357f9e44 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -39,6 +39,7 @@ #include "xfs_iomap.h" #include "xfs_vnodeops.h" #include "xfs_trace.h" +#include "xfs_bmap.h" #include <linux/mpage.h> #include <linux/pagevec.h> #include <linux/writeback.h> @@ -163,14 +164,17 @@ xfs_ioend_new_eof( } /* - * Update on-disk file size now that data has been written to disk. - * The current in-memory file size is i_size. If a write is beyond - * eof i_new_size will be the intended file size until i_size is - * updated. If this write does not extend all the way to the valid - * file size then restrict this update to the end of the write. + * Update on-disk file size now that data has been written to disk. The + * current in-memory file size is i_size. If a write is beyond eof i_new_size + * will be the intended file size until i_size is updated. If this write does + * not extend all the way to the valid file size then restrict this update to + * the end of the write. + * + * This function does not block as blocking on the inode lock in IO completion + * can lead to IO completion order dependency deadlocks.. If it can't get the + * inode ilock it will return EAGAIN. Callers must handle this. */ - -STATIC void +STATIC int xfs_setfilesize( xfs_ioend_t *ioend) { @@ -181,16 +185,40 @@ xfs_setfilesize( ASSERT(ioend->io_type != IOMAP_READ); if (unlikely(ioend->io_error)) - return; + return 0; + + if (!xfs_ilock_nowait(ip, XFS_ILOCK_EXCL)) + return EAGAIN; - xfs_ilock(ip, XFS_ILOCK_EXCL); isize = xfs_ioend_new_eof(ioend); if (isize) { ip->i_d.di_size = isize; - xfs_mark_inode_dirty_sync(ip); + xfs_mark_inode_dirty(ip); } xfs_iunlock(ip, XFS_ILOCK_EXCL); + return 0; +} + +/* + * Schedule IO completion handling on a xfsdatad if this was + * the final hold on this ioend. If we are asked to wait, + * flush the workqueue. + */ +STATIC void +xfs_finish_ioend( + xfs_ioend_t *ioend, + int wait) +{ + if (atomic_dec_and_test(&ioend->io_remaining)) { + struct workqueue_struct *wq; + + wq = (ioend->io_type == IOMAP_UNWRITTEN) ? + xfsconvertd_workqueue : xfsdatad_workqueue; + queue_work(wq, &ioend->io_work); + if (wait) + flush_workqueue(wq); + } } /* @@ -198,11 +226,11 @@ xfs_setfilesize( */ STATIC void xfs_end_io( - struct work_struct *work) + struct work_struct *work) { - xfs_ioend_t *ioend = - container_of(work, xfs_ioend_t, io_work); - struct xfs_inode *ip = XFS_I(ioend->io_inode); + xfs_ioend_t *ioend = container_of(work, xfs_ioend_t, io_work); + struct xfs_inode *ip = XFS_I(ioend->io_inode); + int error = 0; /* * For unwritten extents we need to issue transactions to convert a @@ -210,7 +238,6 @@ xfs_end_io( */ if (ioend->io_type == IOMAP_UNWRITTEN && likely(!ioend->io_error && !XFS_FORCED_SHUTDOWN(ip->i_mount))) { - int error; error = xfs_iomap_write_unwritten(ip, ioend->io_offset, ioend->io_size); @@ -222,30 +249,23 @@ xfs_end_io( * We might have to update the on-disk file size after extending * writes. */ - if (ioend->io_type != IOMAP_READ) - xfs_setfilesize(ioend); - xfs_destroy_ioend(ioend); -} - -/* - * Schedule IO completion handling on a xfsdatad if this was - * the final hold on this ioend. If we are asked to wait, - * flush the workqueue. - */ -STATIC void -xfs_finish_ioend( - xfs_ioend_t *ioend, - int wait) -{ - if (atomic_dec_and_test(&ioend->io_remaining)) { - struct workqueue_struct *wq; - - wq = (ioend->io_type == IOMAP_UNWRITTEN) ? - xfsconvertd_workqueue : xfsdatad_workqueue; - queue_work(wq, &ioend->io_work); - if (wait) - flush_workqueue(wq); + if (ioend->io_type != IOMAP_READ) { + error = xfs_setfilesize(ioend); + ASSERT(!error || error == EAGAIN); } + + /* + * If we didn't complete processing of the ioend, requeue it to the + * tail of the workqueue for another attempt later. Otherwise destroy + * it. + */ + if (error == EAGAIN) { + atomic_inc(&ioend->io_remaining); + xfs_finish_ioend(ioend, 0); + /* ensure we don't spin on blocked ioends */ + delay(1); + } else + xfs_destroy_ioend(ioend); } /* @@ -341,7 +361,7 @@ xfs_submit_ioend_bio( * but don't update the inode size until I/O completion. */ if (xfs_ioend_new_eof(ioend)) - xfs_mark_inode_dirty_sync(XFS_I(ioend->io_inode)); + xfs_mark_inode_dirty(XFS_I(ioend->io_inode)); submit_bio(wbc->sync_mode == WB_SYNC_ALL ? WRITE_SYNC_PLUG : WRITE, bio); @@ -874,6 +894,118 @@ xfs_cluster_write( } } +STATIC void +xfs_vm_invalidatepage( + struct page *page, + unsigned long offset) +{ + trace_xfs_invalidatepage(page->mapping->host, page, offset); + block_invalidatepage(page, offset); +} + +/* + * If the page has delalloc buffers on it, we need to punch them out before we + * invalidate the page. If we don't, we leave a stale delalloc mapping on the + * inode that can trip a BUG() in xfs_get_blocks() later on if a direct IO read + * is done on that same region - the delalloc extent is returned when none is + * supposed to be there. + * + * We prevent this by truncating away the delalloc regions on the page before + * invalidating it. Because they are delalloc, we can do this without needing a + * transaction. Indeed - if we get ENOSPC errors, we have to be able to do this + * truncation without a transaction as there is no space left for block + * reservation (typically why we see a ENOSPC in writeback). + * + * This is not a performance critical path, so for now just do the punching a + * buffer head at a time. + */ +STATIC void +xfs_aops_discard_page( + struct page *page) +{ + struct inode *inode = page->mapping->host; + struct xfs_inode *ip = XFS_I(inode); + struct buffer_head *bh, *head; + loff_t offset = page_offset(page); + ssize_t len = 1 << inode->i_blkbits; + + if (!xfs_is_delayed_page(page, IOMAP_DELAY)) + goto out_invalidate; + + xfs_fs_cmn_err(CE_ALERT, ip->i_mount, + "page discard on page %p, inode 0x%llx, offset %llu.", + page, ip->i_ino, offset); + + xfs_ilock(ip, XFS_ILOCK_EXCL); + bh = head = page_buffers(page); + do { + int done; + xfs_fileoff_t offset_fsb; + xfs_bmbt_irec_t imap; + int nimaps = 1; + int error; + xfs_fsblock_t firstblock; + xfs_bmap_free_t flist; + + if (!buffer_delay(bh)) + goto next_buffer; + + offset_fsb = XFS_B_TO_FSBT(ip->i_mount, offset); + + /* + * Map the range first and check that it is a delalloc extent + * before trying to unmap the range. Otherwise we will be + * trying to remove a real extent (which requires a + * transaction) or a hole, which is probably a bad idea... + */ + error = xfs_bmapi(NULL, ip, offset_fsb, 1, + XFS_BMAPI_ENTIRE, NULL, 0, &imap, + &nimaps, NULL, NULL); + + if (error) { + /* something screwed, just bail */ + xfs_fs_cmn_err(CE_ALERT, ip->i_mount, + "page discard failed delalloc mapping lookup."); + break; + } + if (!nimaps) { + /* nothing there */ + goto next_buffer; + } + if (imap.br_startblock != DELAYSTARTBLOCK) { + /* been converted, ignore */ + goto next_buffer; + } + WARN_ON(imap.br_blockcount == 0); + + /* + * Note: while we initialise the firstblock/flist pair, they + * should never be used because blocks should never be + * allocated or freed for a delalloc extent and hence we need + * don't cancel or finish them after the xfs_bunmapi() call. + */ + xfs_bmap_init(&flist, &firstblock); + error = xfs_bunmapi(NULL, ip, offset_fsb, 1, 0, 1, &firstblock, + &flist, NULL, &done); + + ASSERT(!flist.xbf_count && !flist.xbf_first); + if (error) { + /* something screwed, just bail */ + xfs_fs_cmn_err(CE_ALERT, ip->i_mount, + "page discard unable to remove delalloc mapping."); + break; + } +next_buffer: + offset += len; + + } while ((bh = bh->b_this_page) != head); + + xfs_iunlock(ip, XFS_ILOCK_EXCL); +out_invalidate: + xfs_vm_invalidatepage(page, 0); + return; +} + /* * Calling this without startio set means we are being asked to make a dirty * page ready for freeing it's buffers. When called with startio set then @@ -1125,7 +1257,7 @@ error: */ if (err != -EAGAIN) { if (!unmapped) - block_invalidatepage(page, 0); + xfs_aops_discard_page(page); ClearPageUptodate(page); } return err; @@ -1535,15 +1667,6 @@ xfs_vm_readpages( return mpage_readpages(mapping, pages, nr_pages, xfs_get_blocks); } -STATIC void -xfs_vm_invalidatepage( - struct page *page, - unsigned long offset) -{ - trace_xfs_invalidatepage(page->mapping->host, page, offset); - block_invalidatepage(page, offset); -} - const struct address_space_operations xfs_address_space_operations = { .readpage = xfs_vm_readpage, .readpages = xfs_vm_readpages, diff --git a/fs/xfs/linux-2.6/xfs_export.c b/fs/xfs/linux-2.6/xfs_export.c index 87b8cbd23d4b..846b75aeb2ab 100644 --- a/fs/xfs/linux-2.6/xfs_export.c +++ b/fs/xfs/linux-2.6/xfs_export.c @@ -29,6 +29,7 @@ #include "xfs_vnodeops.h" #include "xfs_bmap_btree.h" #include "xfs_inode.h" +#include "xfs_inode_item.h" /* * Note that we only accept fileids which are long enough rather than allow @@ -215,9 +216,28 @@ xfs_fs_get_parent( return d_obtain_alias(VFS_I(cip)); } +STATIC int +xfs_fs_nfs_commit_metadata( + struct inode *inode) +{ + struct xfs_inode *ip = XFS_I(inode); + struct xfs_mount *mp = ip->i_mount; + int error = 0; + + xfs_ilock(ip, XFS_ILOCK_SHARED); + if (xfs_ipincount(ip)) { + error = _xfs_log_force_lsn(mp, ip->i_itemp->ili_last_lsn, + XFS_LOG_SYNC, NULL); + } + xfs_iunlock(ip, XFS_ILOCK_SHARED); + + return error; +} + const struct export_operations xfs_export_operations = { .encode_fh = xfs_fs_encode_fh, .fh_to_dentry = xfs_fs_fh_to_dentry, .fh_to_parent = xfs_fs_fh_to_parent, .get_parent = xfs_fs_get_parent, + .commit_metadata = xfs_fs_nfs_commit_metadata, }; diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c index e4caeb28ce2e..42dd3bcfba6b 100644 --- a/fs/xfs/linux-2.6/xfs_file.c +++ b/fs/xfs/linux-2.6/xfs_file.c @@ -16,6 +16,7 @@ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #include "xfs.h" +#include "xfs_fs.h" #include "xfs_bit.h" #include "xfs_log.h" #include "xfs_inum.h" @@ -34,52 +35,279 @@ #include "xfs_dir2_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" +#include "xfs_inode_item.h" +#include "xfs_bmap.h" #include "xfs_error.h" #include "xfs_rw.h" #include "xfs_vnodeops.h" #include "xfs_da_btree.h" #include "xfs_ioctl.h" +#include "xfs_trace.h" #include <linux/dcache.h> static const struct vm_operations_struct xfs_file_vm_ops; -STATIC ssize_t -xfs_file_aio_read( - struct kiocb *iocb, - const struct iovec *iov, - unsigned long nr_segs, - loff_t pos) +/* + * xfs_iozero + * + * xfs_iozero clears the specified range of buffer supplied, + * and marks all the affected blocks as valid and modified. If + * an affected block is not allocated, it will be allocated. If + * an affected block is not completely overwritten, and is not + * valid before the operation, it will be read from disk before + * being partially zeroed. + */ +STATIC int +xfs_iozero( + struct xfs_inode *ip, /* inode */ + loff_t pos, /* offset in file */ + size_t count) /* size of data to zero */ { - struct file *file = iocb->ki_filp; - int ioflags = 0; + struct page *page; + struct address_space *mapping; + int status; - BUG_ON(iocb->ki_pos != pos); - if (unlikely(file->f_flags & O_DIRECT)) - ioflags |= IO_ISDIRECT; - if (file->f_mode & FMODE_NOCMTIME) - ioflags |= IO_INVIS; - return xfs_read(XFS_I(file->f_path.dentry->d_inode), iocb, iov, - nr_segs, &iocb->ki_pos, ioflags); + mapping = VFS_I(ip)->i_mapping; + do { + unsigned offset, bytes; + void *fsdata; + + offset = (pos & (PAGE_CACHE_SIZE -1)); /* Within page */ + bytes = PAGE_CACHE_SIZE - offset; + if (bytes > count) + bytes = count; + + status = pagecache_write_begin(NULL, mapping, pos, bytes, + AOP_FLAG_UNINTERRUPTIBLE, + &page, &fsdata); + if (status) + break; + + zero_user(page, offset, bytes); + + status = pagecache_write_end(NULL, mapping, pos, bytes, bytes, + page, fsdata); + WARN_ON(status <= 0); /* can't return less than zero! */ + pos += bytes; + count -= bytes; + status = 0; + } while (count); + + return (-status); +} + +STATIC int +xfs_file_fsync( + struct file *file, + struct dentry *dentry, + int datasync) +{ + struct xfs_inode *ip = XFS_I(dentry->d_inode); + struct xfs_trans *tp; + int error = 0; + int log_flushed = 0; + + xfs_itrace_entry(ip); + + if (XFS_FORCED_SHUTDOWN(ip->i_mount)) + return -XFS_ERROR(EIO); + + xfs_iflags_clear(ip, XFS_ITRUNCATED); + + /* + * We always need to make sure that the required inode state is safe on + * disk. The inode might be clean but we still might need to force the + * log because of committed transactions that haven't hit the disk yet. + * Likewise, there could be unflushed non-transactional changes to the + * inode core that have to go to disk and this requires us to issue + * a synchronous transaction to capture these changes correctly. + * + * This code relies on the assumption that if the i_update_core field + * of the inode is clear and the inode is unpinned then it is clean + * and no action is required. + */ + xfs_ilock(ip, XFS_ILOCK_SHARED); + + /* + * First check if the VFS inode is marked dirty. All the dirtying + * of non-transactional updates no goes through mark_inode_dirty*, + * which allows us to distinguish beteeen pure timestamp updates + * and i_size updates which need to be caught for fdatasync. + * After that also theck for the dirty state in the XFS inode, which + * might gets cleared when the inode gets written out via the AIL + * or xfs_iflush_cluster. + */ + if (((dentry->d_inode->i_state & I_DIRTY_DATASYNC) || + ((dentry->d_inode->i_state & I_DIRTY_SYNC) && !datasync)) && + ip->i_update_core) { + /* + * Kick off a transaction to log the inode core to get the + * updates. The sync transaction will also force the log. + */ + xfs_iunlock(ip, XFS_ILOCK_SHARED); + tp = xfs_trans_alloc(ip->i_mount, XFS_TRANS_FSYNC_TS); + error = xfs_trans_reserve(tp, 0, + XFS_FSYNC_TS_LOG_RES(ip->i_mount), 0, 0, 0); + if (error) { + xfs_trans_cancel(tp, 0); + return -error; + } + xfs_ilock(ip, XFS_ILOCK_EXCL); + + /* + * Note - it's possible that we might have pushed ourselves out + * of the way during trans_reserve which would flush the inode. + * But there's no guarantee that the inode buffer has actually + * gone out yet (it's delwri). Plus the buffer could be pinned + * anyway if it's part of an inode in another recent + * transaction. So we play it safe and fire off the + * transaction anyway. + */ + xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); + xfs_trans_ihold(tp, ip); + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + xfs_trans_set_sync(tp); + error = _xfs_trans_commit(tp, 0, &log_flushed); + + xfs_iunlock(ip, XFS_ILOCK_EXCL); + } else { + /* + * Timestamps/size haven't changed since last inode flush or + * inode transaction commit. That means either nothing got + * written or a transaction committed which caught the updates. + * If the latter happened and the transaction hasn't hit the + * disk yet, the inode will be still be pinned. If it is, + * force the log. + */ + if (xfs_ipincount(ip)) { + error = _xfs_log_force_lsn(ip->i_mount, + ip->i_itemp->ili_last_lsn, + XFS_LOG_SYNC, &log_flushed); + } + xfs_iunlock(ip, XFS_ILOCK_SHARED); + } + + if (ip->i_mount->m_flags & XFS_MOUNT_BARRIER) { + /* + * If the log write didn't issue an ordered tag we need + * to flush the disk cache for the data device now. + */ + if (!log_flushed) + xfs_blkdev_issue_flush(ip->i_mount->m_ddev_targp); + + /* + * If this inode is on the RT dev we need to flush that + * cache as well. + */ + if (XFS_IS_REALTIME_INODE(ip)) + xfs_blkdev_issue_flush(ip->i_mount->m_rtdev_targp); + } + + return -error; } STATIC ssize_t -xfs_file_aio_write( +xfs_file_aio_read( struct kiocb *iocb, - const struct iovec *iov, + const struct iovec *iovp, unsigned long nr_segs, loff_t pos) { struct file *file = iocb->ki_filp; + struct inode *inode = file->f_mapping->host; + struct xfs_inode *ip = XFS_I(inode); + struct xfs_mount *mp = ip->i_mount; + size_t size = 0; + ssize_t ret = 0; int ioflags = 0; + xfs_fsize_t n; + unsigned long seg; + + XFS_STATS_INC(xs_read_calls); BUG_ON(iocb->ki_pos != pos); + if (unlikely(file->f_flags & O_DIRECT)) ioflags |= IO_ISDIRECT; if (file->f_mode & FMODE_NOCMTIME) ioflags |= IO_INVIS; - return xfs_write(XFS_I(file->f_mapping->host), iocb, iov, nr_segs, - &iocb->ki_pos, ioflags); + + /* START copy & waste from filemap.c */ + for (seg = 0; seg < nr_segs; seg++) { + const struct iovec *iv = &iovp[seg]; + + /* + * If any segment has a negative length, or the cumulative + * length ever wraps negative then return -EINVAL. + */ + size += iv->iov_len; + if (unlikely((ssize_t)(size|iv->iov_len) < 0)) + return XFS_ERROR(-EINVAL); + } + /* END copy & waste from filemap.c */ + + if (unlikely(ioflags & IO_ISDIRECT)) { + xfs_buftarg_t *target = + XFS_IS_REALTIME_INODE(ip) ? + mp->m_rtdev_targp : mp->m_ddev_targp; + if ((iocb->ki_pos & target->bt_smask) || + (size & target->bt_smask)) { + if (iocb->ki_pos == ip->i_size) + return 0; + return -XFS_ERROR(EINVAL); + } + } + + n = XFS_MAXIOFFSET(mp) - iocb->ki_pos; + if (n <= 0 || size == 0) + return 0; + + if (n < size) + size = n; + + if (XFS_FORCED_SHUTDOWN(mp)) + return -EIO; + + if (unlikely(ioflags & IO_ISDIRECT)) + mutex_lock(&inode->i_mutex); + xfs_ilock(ip, XFS_IOLOCK_SHARED); + + if (DM_EVENT_ENABLED(ip, DM_EVENT_READ) && !(ioflags & IO_INVIS)) { + int dmflags = FILP_DELAY_FLAG(file) | DM_SEM_FLAG_RD(ioflags); + int iolock = XFS_IOLOCK_SHARED; + + ret = -XFS_SEND_DATA(mp, DM_EVENT_READ, ip, iocb->ki_pos, size, + dmflags, &iolock); + if (ret) { + xfs_iunlock(ip, XFS_IOLOCK_SHARED); + if (unlikely(ioflags & IO_ISDIRECT)) + mutex_unlock(&inode->i_mutex); + return ret; + } + } + + if (unlikely(ioflags & IO_ISDIRECT)) { + if (inode->i_mapping->nrpages) { + ret = -xfs_flushinval_pages(ip, + (iocb->ki_pos & PAGE_CACHE_MASK), + -1, FI_REMAPF_LOCKED); + } + mutex_unlock(&inode->i_mutex); + if (ret) { + xfs_iunlock(ip, XFS_IOLOCK_SHARED); + return ret; + } + } + + trace_xfs_file_read(ip, size, iocb->ki_pos, ioflags); + + ret = generic_file_aio_read(iocb, iovp, nr_segs, iocb->ki_pos); + if (ret > 0) + XFS_STATS_ADD(xs_read_bytes, ret); + + xfs_iunlock(ip, XFS_IOLOCK_SHARED); + return ret; } STATIC ssize_t @@ -87,16 +315,44 @@ xfs_file_splice_read( struct file *infilp, loff_t *ppos, struct pipe_inode_info *pipe, - size_t len, + size_t count, unsigned int flags) { + struct xfs_inode *ip = XFS_I(infilp->f_mapping->host); + struct xfs_mount *mp = ip->i_mount; int ioflags = 0; + ssize_t ret; + + XFS_STATS_INC(xs_read_calls); if (infilp->f_mode & FMODE_NOCMTIME) ioflags |= IO_INVIS; - return xfs_splice_read(XFS_I(infilp->f_path.dentry->d_inode), - infilp, ppos, pipe, len, flags, ioflags); + if (XFS_FORCED_SHUTDOWN(ip->i_mount)) + return -EIO; + + xfs_ilock(ip, XFS_IOLOCK_SHARED); + + if (DM_EVENT_ENABLED(ip, DM_EVENT_READ) && !(ioflags & IO_INVIS)) { + int iolock = XFS_IOLOCK_SHARED; + int error; + + error = XFS_SEND_DATA(mp, DM_EVENT_READ, ip, *ppos, count, + FILP_DELAY_FLAG(infilp), &iolock); + if (error) { + xfs_iunlock(ip, XFS_IOLOCK_SHARED); + return -error; + } + } + + trace_xfs_file_splice_read(ip, count, *ppos, ioflags); + + ret = generic_file_splice_read(infilp, ppos, pipe, count, flags); + if (ret > 0) + XFS_STATS_ADD(xs_read_bytes, ret); + + xfs_iunlock(ip, XFS_IOLOCK_SHARED); + return ret; } STATIC ssize_t @@ -104,16 +360,538 @@ xfs_file_splice_write( struct pipe_inode_info *pipe, struct file *outfilp, loff_t *ppos, - size_t len, + size_t count, unsigned int flags) { + struct inode *inode = outfilp->f_mapping->host; + struct xfs_inode *ip = XFS_I(inode); + struct xfs_mount *mp = ip->i_mount; + xfs_fsize_t isize, new_size; int ioflags = 0; + ssize_t ret; + + XFS_STATS_INC(xs_write_calls); if (outfilp->f_mode & FMODE_NOCMTIME) ioflags |= IO_INVIS; - return xfs_splice_write(XFS_I(outfilp->f_path.dentry->d_inode), - pipe, outfilp, ppos, len, flags, ioflags); + if (XFS_FORCED_SHUTDOWN(ip->i_mount)) + return -EIO; + + xfs_ilock(ip, XFS_IOLOCK_EXCL); + + if (DM_EVENT_ENABLED(ip, DM_EVENT_WRITE) && !(ioflags & IO_INVIS)) { + int iolock = XFS_IOLOCK_EXCL; + int error; + + error = XFS_SEND_DATA(mp, DM_EVENT_WRITE, ip, *ppos, count, + FILP_DELAY_FLAG(outfilp), &iolock); + if (error) { + xfs_iunlock(ip, XFS_IOLOCK_EXCL); + return -error; + } + } + + new_size = *ppos + count; + + xfs_ilock(ip, XFS_ILOCK_EXCL); + if (new_size > ip->i_size) + ip->i_new_size = new_size; + xfs_iunlock(ip, XFS_ILOCK_EXCL); + + trace_xfs_file_splice_write(ip, count, *ppos, ioflags); + + ret = generic_file_splice_write(pipe, outfilp, ppos, count, flags); + if (ret > 0) + XFS_STATS_ADD(xs_write_bytes, ret); + + isize = i_size_read(inode); + if (unlikely(ret < 0 && ret != -EFAULT && *ppos > isize)) + *ppos = isize; + + if (*ppos > ip->i_size) { + xfs_ilock(ip, XFS_ILOCK_EXCL); + if (*ppos > ip->i_size) + ip->i_size = *ppos; + xfs_iunlock(ip, XFS_ILOCK_EXCL); + } + + if (ip->i_new_size) { + xfs_ilock(ip, XFS_ILOCK_EXCL); + ip->i_new_size = 0; + if (ip->i_d.di_size > ip->i_size) + ip->i_d.di_size = ip->i_size; + xfs_iunlock(ip, XFS_ILOCK_EXCL); + } + xfs_iunlock(ip, XFS_IOLOCK_EXCL); + return ret; +} + +/* + * This routine is called to handle zeroing any space in the last + * block of the file that is beyond the EOF. We do this since the + * size is being increased without writing anything to that block + * and we don't want anyone to read the garbage on the disk. + */ +STATIC int /* error (positive) */ +xfs_zero_last_block( + xfs_inode_t *ip, + xfs_fsize_t offset, + xfs_fsize_t isize) +{ + xfs_fileoff_t last_fsb; + xfs_mount_t *mp = ip->i_mount; + int nimaps; + int zero_offset; + int zero_len; + int error = 0; + xfs_bmbt_irec_t imap; + + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); + + zero_offset = XFS_B_FSB_OFFSET(mp, isize); + if (zero_offset == 0) { + /* + * There are no extra bytes in the last block on disk to + * zero, so return. + */ + return 0; + } + + last_fsb = XFS_B_TO_FSBT(mp, isize); + nimaps = 1; + error = xfs_bmapi(NULL, ip, last_fsb, 1, 0, NULL, 0, &imap, + &nimaps, NULL, NULL); + if (error) { + return error; + } + ASSERT(nimaps > 0); + /* + * If the block underlying isize is just a hole, then there + * is nothing to zero. + */ + if (imap.br_startblock == HOLESTARTBLOCK) { + return 0; + } + /* + * Zero the part of the last block beyond the EOF, and write it + * out sync. We need to drop the ilock while we do this so we + * don't deadlock when the buffer cache calls back to us. + */ + xfs_iunlock(ip, XFS_ILOCK_EXCL); + + zero_len = mp->m_sb.sb_blocksize - zero_offset; + if (isize + zero_len > offset) + zero_len = offset - isize; + error = xfs_iozero(ip, isize, zero_len); + + xfs_ilock(ip, XFS_ILOCK_EXCL); + ASSERT(error >= 0); + return error; +} + +/* + * Zero any on disk space between the current EOF and the new, + * larger EOF. This handles the normal case of zeroing the remainder + * of the last block in the file and the unusual case of zeroing blocks + * out beyond the size of the file. This second case only happens + * with fixed size extents and when the system crashes before the inode + * size was updated but after blocks were allocated. If fill is set, + * then any holes in the range are filled and zeroed. If not, the holes + * are left alone as holes. + */ + +int /* error (positive) */ +xfs_zero_eof( + xfs_inode_t *ip, + xfs_off_t offset, /* starting I/O offset */ + xfs_fsize_t isize) /* current inode size */ +{ + xfs_mount_t *mp = ip->i_mount; + xfs_fileoff_t start_zero_fsb; + xfs_fileoff_t end_zero_fsb; + xfs_fileoff_t zero_count_fsb; + xfs_fileoff_t last_fsb; + xfs_fileoff_t zero_off; + xfs_fsize_t zero_len; + int nimaps; + int error = 0; + xfs_bmbt_irec_t imap; + + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL)); + ASSERT(offset > isize); + + /* + * First handle zeroing the block on which isize resides. + * We only zero a part of that block so it is handled specially. + */ + error = xfs_zero_last_block(ip, offset, isize); + if (error) { + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL)); + return error; + } + + /* + * Calculate the range between the new size and the old + * where blocks needing to be zeroed may exist. To get the + * block where the last byte in the file currently resides, + * we need to subtract one from the size and truncate back + * to a block boundary. We subtract 1 in case the size is + * exactly on a block boundary. + */ + last_fsb = isize ? XFS_B_TO_FSBT(mp, isize - 1) : (xfs_fileoff_t)-1; + start_zero_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)isize); + end_zero_fsb = XFS_B_TO_FSBT(mp, offset - 1); + ASSERT((xfs_sfiloff_t)last_fsb < (xfs_sfiloff_t)start_zero_fsb); + if (last_fsb == end_zero_fsb) { + /* + * The size was only incremented on its last block. + * We took care of that above, so just return. + */ + return 0; + } + + ASSERT(start_zero_fsb <= end_zero_fsb); + while (start_zero_fsb <= end_zero_fsb) { + nimaps = 1; + zero_count_fsb = end_zero_fsb - start_zero_fsb + 1; + error = xfs_bmapi(NULL, ip, start_zero_fsb, zero_count_fsb, + 0, NULL, 0, &imap, &nimaps, NULL, NULL); + if (error) { + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL)); + return error; + } + ASSERT(nimaps > 0); + + if (imap.br_state == XFS_EXT_UNWRITTEN || + imap.br_startblock == HOLESTARTBLOCK) { + /* + * This loop handles initializing pages that were + * partially initialized by the code below this + * loop. It basically zeroes the part of the page + * that sits on a hole and sets the page as P_HOLE + * and calls remapf if it is a mapped file. + */ + start_zero_fsb = imap.br_startoff + imap.br_blockcount; + ASSERT(start_zero_fsb <= (end_zero_fsb + 1)); + continue; + } + + /* + * There are blocks we need to zero. + * Drop the inode lock while we're doing the I/O. + * We'll still have the iolock to protect us. + */ + xfs_iunlock(ip, XFS_ILOCK_EXCL); + + zero_off = XFS_FSB_TO_B(mp, start_zero_fsb); + zero_len = XFS_FSB_TO_B(mp, imap.br_blockcount); + + if ((zero_off + zero_len) > offset) + zero_len = offset - zero_off; + + error = xfs_iozero(ip, zero_off, zero_len); + if (error) { + goto out_lock; + } + + start_zero_fsb = imap.br_startoff + imap.br_blockcount; + ASSERT(start_zero_fsb <= (end_zero_fsb + 1)); + + xfs_ilock(ip, XFS_ILOCK_EXCL); + } + + return 0; + +out_lock: + xfs_ilock(ip, XFS_ILOCK_EXCL); + ASSERT(error >= 0); + return error; +} + +STATIC ssize_t +xfs_file_aio_write( + struct kiocb *iocb, + const struct iovec *iovp, + unsigned long nr_segs, + loff_t pos) +{ + struct file *file = iocb->ki_filp; + struct address_space *mapping = file->f_mapping; + struct inode *inode = mapping->host; + struct xfs_inode *ip = XFS_I(inode); + struct xfs_mount *mp = ip->i_mount; + ssize_t ret = 0, error = 0; + int ioflags = 0; + xfs_fsize_t isize, new_size; + int iolock; + int eventsent = 0; + size_t ocount = 0, count; + int need_i_mutex; + + XFS_STATS_INC(xs_write_calls); + + BUG_ON(iocb->ki_pos != pos); + + if (unlikely(file->f_flags & O_DIRECT)) + ioflags |= IO_ISDIRECT; + if (file->f_mode & FMODE_NOCMTIME) + ioflags |= IO_INVIS; + + error = generic_segment_checks(iovp, &nr_segs, &ocount, VERIFY_READ); + if (error) + return error; + + count = ocount; + if (count == 0) + return 0; + + xfs_wait_for_freeze(mp, SB_FREEZE_WRITE); + + if (XFS_FORCED_SHUTDOWN(mp)) + return -EIO; + +relock: + if (ioflags & IO_ISDIRECT) { + iolock = XFS_IOLOCK_SHARED; + need_i_mutex = 0; + } else { + iolock = XFS_IOLOCK_EXCL; + need_i_mutex = 1; + mutex_lock(&inode->i_mutex); + } + + xfs_ilock(ip, XFS_ILOCK_EXCL|iolock); + +start: + error = -generic_write_checks(file, &pos, &count, + S_ISBLK(inode->i_mode)); + if (error) { + xfs_iunlock(ip, XFS_ILOCK_EXCL|iolock); + goto out_unlock_mutex; + } + + if ((DM_EVENT_ENABLED(ip, DM_EVENT_WRITE) && + !(ioflags & IO_INVIS) && !eventsent)) { + int dmflags = FILP_DELAY_FLAG(file); + + if (need_i_mutex) + dmflags |= DM_FLAGS_IMUX; + + xfs_iunlock(ip, XFS_ILOCK_EXCL); + error = XFS_SEND_DATA(ip->i_mount, DM_EVENT_WRITE, ip, + pos, count, dmflags, &iolock); + if (error) { + goto out_unlock_internal; + } + xfs_ilock(ip, XFS_ILOCK_EXCL); + eventsent = 1; + + /* + * The iolock was dropped and reacquired in XFS_SEND_DATA + * so we have to recheck the size when appending. + * We will only "goto start;" once, since having sent the + * event prevents another call to XFS_SEND_DATA, which is + * what allows the size to change in the first place. + */ + if ((file->f_flags & O_APPEND) && pos != ip->i_size) + goto start; + } + + if (ioflags & IO_ISDIRECT) { + xfs_buftarg_t *target = + XFS_IS_REALTIME_INODE(ip) ? + mp->m_rtdev_targp : mp->m_ddev_targp; + + if ((pos & target->bt_smask) || (count & target->bt_smask)) { + xfs_iunlock(ip, XFS_ILOCK_EXCL|iolock); + return XFS_ERROR(-EINVAL); + } + + if (!need_i_mutex && (mapping->nrpages || pos > ip->i_size)) { + xfs_iunlock(ip, XFS_ILOCK_EXCL|iolock); + iolock = XFS_IOLOCK_EXCL; + need_i_mutex = 1; + mutex_lock(&inode->i_mutex); + xfs_ilock(ip, XFS_ILOCK_EXCL|iolock); + goto start; + } + } + + new_size = pos + count; + if (new_size > ip->i_size) + ip->i_new_size = new_size; + + if (likely(!(ioflags & IO_INVIS))) + file_update_time(file); + + /* + * If the offset is beyond the size of the file, we have a couple + * of things to do. First, if there is already space allocated + * we need to either create holes or zero the disk or ... + * + * If there is a page where the previous size lands, we need + * to zero it out up to the new size. + */ + + if (pos > ip->i_size) { + error = xfs_zero_eof(ip, pos, ip->i_size); + if (error) { + xfs_iunlock(ip, XFS_ILOCK_EXCL); + goto out_unlock_internal; + } + } + xfs_iunlock(ip, XFS_ILOCK_EXCL); + + /* + * If we're writing the file then make sure to clear the + * setuid and setgid bits if the process is not being run + * by root. This keeps people from modifying setuid and + * setgid binaries. + */ + error = -file_remove_suid(file); + if (unlikely(error)) + goto out_unlock_internal; + + /* We can write back this queue in page reclaim */ + current->backing_dev_info = mapping->backing_dev_info; + + if ((ioflags & IO_ISDIRECT)) { + if (mapping->nrpages) { + WARN_ON(need_i_mutex == 0); + error = xfs_flushinval_pages(ip, + (pos & PAGE_CACHE_MASK), + -1, FI_REMAPF_LOCKED); + if (error) + goto out_unlock_internal; + } + + if (need_i_mutex) { + /* demote the lock now the cached pages are gone */ + xfs_ilock_demote(ip, XFS_IOLOCK_EXCL); + mutex_unlock(&inode->i_mutex); + + iolock = XFS_IOLOCK_SHARED; + need_i_mutex = 0; + } + + trace_xfs_file_direct_write(ip, count, iocb->ki_pos, ioflags); + ret = generic_file_direct_write(iocb, iovp, + &nr_segs, pos, &iocb->ki_pos, count, ocount); + + /* + * direct-io write to a hole: fall through to buffered I/O + * for completing the rest of the request. + */ + if (ret >= 0 && ret != count) { + XFS_STATS_ADD(xs_write_bytes, ret); + + pos += ret; + count -= ret; + + ioflags &= ~IO_ISDIRECT; + xfs_iunlock(ip, iolock); + goto relock; + } + } else { + int enospc = 0; + ssize_t ret2 = 0; + +write_retry: + trace_xfs_file_buffered_write(ip, count, iocb->ki_pos, ioflags); + ret2 = generic_file_buffered_write(iocb, iovp, nr_segs, + pos, &iocb->ki_pos, count, ret); + /* + * if we just got an ENOSPC, flush the inode now we + * aren't holding any page locks and retry *once* + */ + if (ret2 == -ENOSPC && !enospc) { + error = xfs_flush_pages(ip, 0, -1, 0, FI_NONE); + if (error) + goto out_unlock_internal; + enospc = 1; + goto write_retry; + } + ret = ret2; + } + + current->backing_dev_info = NULL; + + isize = i_size_read(inode); + if (unlikely(ret < 0 && ret != -EFAULT && iocb->ki_pos > isize)) + iocb->ki_pos = isize; + + if (iocb->ki_pos > ip->i_size) { + xfs_ilock(ip, XFS_ILOCK_EXCL); + if (iocb->ki_pos > ip->i_size) + ip->i_size = iocb->ki_pos; + xfs_iunlock(ip, XFS_ILOCK_EXCL); + } + + if (ret == -ENOSPC && + DM_EVENT_ENABLED(ip, DM_EVENT_NOSPACE) && !(ioflags & IO_INVIS)) { + xfs_iunlock(ip, iolock); + if (need_i_mutex) + mutex_unlock(&inode->i_mutex); + error = XFS_SEND_NAMESP(ip->i_mount, DM_EVENT_NOSPACE, ip, + DM_RIGHT_NULL, ip, DM_RIGHT_NULL, NULL, NULL, + 0, 0, 0); /* Delay flag intentionally unused */ + if (need_i_mutex) + mutex_lock(&inode->i_mutex); + xfs_ilock(ip, iolock); + if (error) + goto out_unlock_internal; + goto start; + } + + error = -ret; + if (ret <= 0) + goto out_unlock_internal; + + XFS_STATS_ADD(xs_write_bytes, ret); + + /* Handle various SYNC-type writes */ + if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) { + loff_t end = pos + ret - 1; + int error2; + + xfs_iunlock(ip, iolock); + if (need_i_mutex) + mutex_unlock(&inode->i_mutex); + + error2 = filemap_write_and_wait_range(mapping, pos, end); + if (!error) + error = error2; + if (need_i_mutex) + mutex_lock(&inode->i_mutex); + xfs_ilock(ip, iolock); + + error2 = -xfs_file_fsync(file, file->f_path.dentry, + (file->f_flags & __O_SYNC) ? 0 : 1); + if (!error) + error = error2; + } + + out_unlock_internal: + if (ip->i_new_size) { + xfs_ilock(ip, XFS_ILOCK_EXCL); + ip->i_new_size = 0; + /* + * If this was a direct or synchronous I/O that failed (such + * as ENOSPC) then part of the I/O may have been written to + * disk before the error occured. In this case the on-disk + * file size may have been adjusted beyond the in-memory file + * size and now needs to be truncated back. + */ + if (ip->i_d.di_size > ip->i_size) + ip->i_d.di_size = ip->i_size; + xfs_iunlock(ip, XFS_ILOCK_EXCL); + } + xfs_iunlock(ip, iolock); + out_unlock_mutex: + if (need_i_mutex) + mutex_unlock(&inode->i_mutex); + return -error; } STATIC int @@ -160,28 +938,6 @@ xfs_file_release( return -xfs_release(XFS_I(inode)); } -/* - * We ignore the datasync flag here because a datasync is effectively - * identical to an fsync. That is, datasync implies that we need to write - * only the metadata needed to be able to access the data that is written - * if we crash after the call completes. Hence if we are writing beyond - * EOF we have to log the inode size change as well, which makes it a - * full fsync. If we don't write beyond EOF, the inode core will be - * clean in memory and so we don't need to log the inode, just like - * fsync. - */ -STATIC int -xfs_file_fsync( - struct file *file, - struct dentry *dentry, - int datasync) -{ - struct xfs_inode *ip = XFS_I(dentry->d_inode); - - xfs_iflags_clear(ip, XFS_ITRUNCATED); - return -xfs_fsync(ip); -} - STATIC int xfs_file_readdir( struct file *filp, @@ -203,9 +959,9 @@ xfs_file_readdir( * * Try to give it an estimate that's good enough, maybe at some * point we can change the ->readdir prototype to include the - * buffer size. + * buffer size. For now we use the current glibc buffer size. */ - bufsize = (size_t)min_t(loff_t, PAGE_SIZE, ip->i_d.di_size); + bufsize = (size_t)min_t(loff_t, 32768, ip->i_d.di_size); error = xfs_readdir(ip, dirent, bufsize, (xfs_off_t *)&filp->f_pos, filldir); diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c index e8566bbf0f00..61a99608731e 100644 --- a/fs/xfs/linux-2.6/xfs_iops.c +++ b/fs/xfs/linux-2.6/xfs_iops.c @@ -91,6 +91,16 @@ xfs_mark_inode_dirty_sync( mark_inode_dirty_sync(inode); } +void +xfs_mark_inode_dirty( + xfs_inode_t *ip) +{ + struct inode *inode = VFS_I(ip); + + if (!(inode->i_state & (I_WILL_FREE|I_FREEING|I_CLEAR))) + mark_inode_dirty(inode); +} + /* * Change the requested timestamp in the given inode. * We don't lock across timestamp updates, and we don't log them but diff --git a/fs/xfs/linux-2.6/xfs_linux.h b/fs/xfs/linux-2.6/xfs_linux.h index 5af0c81ca1ae..facfb323a706 100644 --- a/fs/xfs/linux-2.6/xfs_linux.h +++ b/fs/xfs/linux-2.6/xfs_linux.h @@ -88,7 +88,6 @@ #include <xfs_super.h> #include <xfs_globals.h> #include <xfs_fs_subr.h> -#include <xfs_lrw.h> #include <xfs_buf.h> /* diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c deleted file mode 100644 index eac6f80d786d..000000000000 --- a/fs/xfs/linux-2.6/xfs_lrw.c +++ /dev/null @@ -1,796 +0,0 @@ -/* - * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_fs.h" -#include "xfs_bit.h" -#include "xfs_log.h" -#include "xfs_inum.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_dir2.h" -#include "xfs_alloc.h" -#include "xfs_dmapi.h" -#include "xfs_quota.h" -#include "xfs_mount.h" -#include "xfs_bmap_btree.h" -#include "xfs_alloc_btree.h" -#include "xfs_ialloc_btree.h" -#include "xfs_dir2_sf.h" -#include "xfs_attr_sf.h" -#include "xfs_dinode.h" -#include "xfs_inode.h" -#include "xfs_bmap.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_rtalloc.h" -#include "xfs_error.h" -#include "xfs_itable.h" -#include "xfs_rw.h" -#include "xfs_attr.h" -#include "xfs_inode_item.h" -#include "xfs_buf_item.h" -#include "xfs_utils.h" -#include "xfs_iomap.h" -#include "xfs_vnodeops.h" -#include "xfs_trace.h" - -#include <linux/capability.h> -#include <linux/writeback.h> - - -/* - * xfs_iozero - * - * xfs_iozero clears the specified range of buffer supplied, - * and marks all the affected blocks as valid and modified. If - * an affected block is not allocated, it will be allocated. If - * an affected block is not completely overwritten, and is not - * valid before the operation, it will be read from disk before - * being partially zeroed. - */ -STATIC int -xfs_iozero( - struct xfs_inode *ip, /* inode */ - loff_t pos, /* offset in file */ - size_t count) /* size of data to zero */ -{ - struct page *page; - struct address_space *mapping; - int status; - - mapping = VFS_I(ip)->i_mapping; - do { - unsigned offset, bytes; - void *fsdata; - - offset = (pos & (PAGE_CACHE_SIZE -1)); /* Within page */ - bytes = PAGE_CACHE_SIZE - offset; - if (bytes > count) - bytes = count; - - status = pagecache_write_begin(NULL, mapping, pos, bytes, - AOP_FLAG_UNINTERRUPTIBLE, - &page, &fsdata); - if (status) - break; - - zero_user(page, offset, bytes); - - status = pagecache_write_end(NULL, mapping, pos, bytes, bytes, - page, fsdata); - WARN_ON(status <= 0); /* can't return less than zero! */ - pos += bytes; - count -= bytes; - status = 0; - } while (count); - - return (-status); -} - -ssize_t /* bytes read, or (-) error */ -xfs_read( - xfs_inode_t *ip, - struct kiocb *iocb, - const struct iovec *iovp, - unsigned int segs, - loff_t *offset, - int ioflags) -{ - struct file *file = iocb->ki_filp; - struct inode *inode = file->f_mapping->host; - xfs_mount_t *mp = ip->i_mount; - size_t size = 0; - ssize_t ret = 0; - xfs_fsize_t n; - unsigned long seg; - - - XFS_STATS_INC(xs_read_calls); - - /* START copy & waste from filemap.c */ - for (seg = 0; seg < segs; seg++) { - const struct iovec *iv = &iovp[seg]; - - /* - * If any segment has a negative length, or the cumulative - * length ever wraps negative then return -EINVAL. - */ - size += iv->iov_len; - if (unlikely((ssize_t)(size|iv->iov_len) < 0)) - return XFS_ERROR(-EINVAL); - } - /* END copy & waste from filemap.c */ - - if (unlikely(ioflags & IO_ISDIRECT)) { - xfs_buftarg_t *target = - XFS_IS_REALTIME_INODE(ip) ? - mp->m_rtdev_targp : mp->m_ddev_targp; - if ((*offset & target->bt_smask) || - (size & target->bt_smask)) { - if (*offset == ip->i_size) { - return (0); - } - return -XFS_ERROR(EINVAL); - } - } - - n = XFS_MAXIOFFSET(mp) - *offset; - if ((n <= 0) || (size == 0)) - return 0; - - if (n < size) - size = n; - - if (XFS_FORCED_SHUTDOWN(mp)) - return -EIO; - - if (unlikely(ioflags & IO_ISDIRECT)) - mutex_lock(&inode->i_mutex); - xfs_ilock(ip, XFS_IOLOCK_SHARED); - - if (DM_EVENT_ENABLED(ip, DM_EVENT_READ) && !(ioflags & IO_INVIS)) { - int dmflags = FILP_DELAY_FLAG(file) | DM_SEM_FLAG_RD(ioflags); - int iolock = XFS_IOLOCK_SHARED; - - ret = -XFS_SEND_DATA(mp, DM_EVENT_READ, ip, *offset, size, - dmflags, &iolock); - if (ret) { - xfs_iunlock(ip, XFS_IOLOCK_SHARED); - if (unlikely(ioflags & IO_ISDIRECT)) - mutex_unlock(&inode->i_mutex); - return ret; - } - } - - if (unlikely(ioflags & IO_ISDIRECT)) { - if (inode->i_mapping->nrpages) - ret = -xfs_flushinval_pages(ip, (*offset & PAGE_CACHE_MASK), - -1, FI_REMAPF_LOCKED); - mutex_unlock(&inode->i_mutex); - if (ret) { - xfs_iunlock(ip, XFS_IOLOCK_SHARED); - return ret; - } - } - - trace_xfs_file_read(ip, size, *offset, ioflags); - - iocb->ki_pos = *offset; - ret = generic_file_aio_read(iocb, iovp, segs, *offset); - if (ret > 0) - XFS_STATS_ADD(xs_read_bytes, ret); - - xfs_iunlock(ip, XFS_IOLOCK_SHARED); - return ret; -} - -ssize_t -xfs_splice_read( - xfs_inode_t *ip, - struct file *infilp, - loff_t *ppos, - struct pipe_inode_info *pipe, - size_t count, - int flags, - int ioflags) -{ - xfs_mount_t *mp = ip->i_mount; - ssize_t ret; - - XFS_STATS_INC(xs_read_calls); - if (XFS_FORCED_SHUTDOWN(ip->i_mount)) - return -EIO; - - xfs_ilock(ip, XFS_IOLOCK_SHARED); - - if (DM_EVENT_ENABLED(ip, DM_EVENT_READ) && !(ioflags & IO_INVIS)) { - int iolock = XFS_IOLOCK_SHARED; - int error; - - error = XFS_SEND_DATA(mp, DM_EVENT_READ, ip, *ppos, count, - FILP_DELAY_FLAG(infilp), &iolock); - if (error) { - xfs_iunlock(ip, XFS_IOLOCK_SHARED); - return -error; - } - } - - trace_xfs_file_splice_read(ip, count, *ppos, ioflags); - - ret = generic_file_splice_read(infilp, ppos, pipe, count, flags); - if (ret > 0) - XFS_STATS_ADD(xs_read_bytes, ret); - - xfs_iunlock(ip, XFS_IOLOCK_SHARED); - return ret; -} - -ssize_t -xfs_splice_write( - xfs_inode_t *ip, - struct pipe_inode_info *pipe, - struct file *outfilp, - loff_t *ppos, - size_t count, - int flags, - int ioflags) -{ - xfs_mount_t *mp = ip->i_mount; - ssize_t ret; - struct inode *inode = outfilp->f_mapping->host; - xfs_fsize_t isize, new_size; - - XFS_STATS_INC(xs_write_calls); - if (XFS_FORCED_SHUTDOWN(ip->i_mount)) - return -EIO; - - xfs_ilock(ip, XFS_IOLOCK_EXCL); - - if (DM_EVENT_ENABLED(ip, DM_EVENT_WRITE) && !(ioflags & IO_INVIS)) { - int iolock = XFS_IOLOCK_EXCL; - int error; - - error = XFS_SEND_DATA(mp, DM_EVENT_WRITE, ip, *ppos, count, - FILP_DELAY_FLAG(outfilp), &iolock); - if (error) { - xfs_iunlock(ip, XFS_IOLOCK_EXCL); - return -error; - } - } - - new_size = *ppos + count; - - xfs_ilock(ip, XFS_ILOCK_EXCL); - if (new_size > ip->i_size) - ip->i_new_size = new_size; - xfs_iunlock(ip, XFS_ILOCK_EXCL); - - trace_xfs_file_splice_write(ip, count, *ppos, ioflags); - - ret = generic_file_splice_write(pipe, outfilp, ppos, count, flags); - if (ret > 0) - XFS_STATS_ADD(xs_write_bytes, ret); - - isize = i_size_read(inode); - if (unlikely(ret < 0 && ret != -EFAULT && *ppos > isize)) - *ppos = isize; - - if (*ppos > ip->i_size) { - xfs_ilock(ip, XFS_ILOCK_EXCL); - if (*ppos > ip->i_size) - ip->i_size = *ppos; - xfs_iunlock(ip, XFS_ILOCK_EXCL); - } - - if (ip->i_new_size) { - xfs_ilock(ip, XFS_ILOCK_EXCL); - ip->i_new_size = 0; - if (ip->i_d.di_size > ip->i_size) - ip->i_d.di_size = ip->i_size; - xfs_iunlock(ip, XFS_ILOCK_EXCL); - } - xfs_iunlock(ip, XFS_IOLOCK_EXCL); - return ret; -} - -/* - * This routine is called to handle zeroing any space in the last - * block of the file that is beyond the EOF. We do this since the - * size is being increased without writing anything to that block - * and we don't want anyone to read the garbage on the disk. - */ -STATIC int /* error (positive) */ -xfs_zero_last_block( - xfs_inode_t *ip, - xfs_fsize_t offset, - xfs_fsize_t isize) -{ - xfs_fileoff_t last_fsb; - xfs_mount_t *mp = ip->i_mount; - int nimaps; - int zero_offset; - int zero_len; - int error = 0; - xfs_bmbt_irec_t imap; - - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - - zero_offset = XFS_B_FSB_OFFSET(mp, isize); - if (zero_offset == 0) { - /* - * There are no extra bytes in the last block on disk to - * zero, so return. - */ - return 0; - } - - last_fsb = XFS_B_TO_FSBT(mp, isize); - nimaps = 1; - error = xfs_bmapi(NULL, ip, last_fsb, 1, 0, NULL, 0, &imap, - &nimaps, NULL, NULL); - if (error) { - return error; - } - ASSERT(nimaps > 0); - /* - * If the block underlying isize is just a hole, then there - * is nothing to zero. - */ - if (imap.br_startblock == HOLESTARTBLOCK) { - return 0; - } - /* - * Zero the part of the last block beyond the EOF, and write it - * out sync. We need to drop the ilock while we do this so we - * don't deadlock when the buffer cache calls back to us. - */ - xfs_iunlock(ip, XFS_ILOCK_EXCL); - - zero_len = mp->m_sb.sb_blocksize - zero_offset; - if (isize + zero_len > offset) - zero_len = offset - isize; - error = xfs_iozero(ip, isize, zero_len); - - xfs_ilock(ip, XFS_ILOCK_EXCL); - ASSERT(error >= 0); - return error; -} - -/* - * Zero any on disk space between the current EOF and the new, - * larger EOF. This handles the normal case of zeroing the remainder - * of the last block in the file and the unusual case of zeroing blocks - * out beyond the size of the file. This second case only happens - * with fixed size extents and when the system crashes before the inode - * size was updated but after blocks were allocated. If fill is set, - * then any holes in the range are filled and zeroed. If not, the holes - * are left alone as holes. - */ - -int /* error (positive) */ -xfs_zero_eof( - xfs_inode_t *ip, - xfs_off_t offset, /* starting I/O offset */ - xfs_fsize_t isize) /* current inode size */ -{ - xfs_mount_t *mp = ip->i_mount; - xfs_fileoff_t start_zero_fsb; - xfs_fileoff_t end_zero_fsb; - xfs_fileoff_t zero_count_fsb; - xfs_fileoff_t last_fsb; - xfs_fileoff_t zero_off; - xfs_fsize_t zero_len; - int nimaps; - int error = 0; - xfs_bmbt_irec_t imap; - - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL)); - ASSERT(offset > isize); - - /* - * First handle zeroing the block on which isize resides. - * We only zero a part of that block so it is handled specially. - */ - error = xfs_zero_last_block(ip, offset, isize); - if (error) { - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL)); - return error; - } - - /* - * Calculate the range between the new size and the old - * where blocks needing to be zeroed may exist. To get the - * block where the last byte in the file currently resides, - * we need to subtract one from the size and truncate back - * to a block boundary. We subtract 1 in case the size is - * exactly on a block boundary. - */ - last_fsb = isize ? XFS_B_TO_FSBT(mp, isize - 1) : (xfs_fileoff_t)-1; - start_zero_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)isize); - end_zero_fsb = XFS_B_TO_FSBT(mp, offset - 1); - ASSERT((xfs_sfiloff_t)last_fsb < (xfs_sfiloff_t)start_zero_fsb); - if (last_fsb == end_zero_fsb) { - /* - * The size was only incremented on its last block. - * We took care of that above, so just return. - */ - return 0; - } - - ASSERT(start_zero_fsb <= end_zero_fsb); - while (start_zero_fsb <= end_zero_fsb) { - nimaps = 1; - zero_count_fsb = end_zero_fsb - start_zero_fsb + 1; - error = xfs_bmapi(NULL, ip, start_zero_fsb, zero_count_fsb, - 0, NULL, 0, &imap, &nimaps, NULL, NULL); - if (error) { - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL)); - return error; - } - ASSERT(nimaps > 0); - - if (imap.br_state == XFS_EXT_UNWRITTEN || - imap.br_startblock == HOLESTARTBLOCK) { - /* - * This loop handles initializing pages that were - * partially initialized by the code below this - * loop. It basically zeroes the part of the page - * that sits on a hole and sets the page as P_HOLE - * and calls remapf if it is a mapped file. - */ - start_zero_fsb = imap.br_startoff + imap.br_blockcount; - ASSERT(start_zero_fsb <= (end_zero_fsb + 1)); - continue; - } - - /* - * There are blocks we need to zero. - * Drop the inode lock while we're doing the I/O. - * We'll still have the iolock to protect us. - */ - xfs_iunlock(ip, XFS_ILOCK_EXCL); - - zero_off = XFS_FSB_TO_B(mp, start_zero_fsb); - zero_len = XFS_FSB_TO_B(mp, imap.br_blockcount); - - if ((zero_off + zero_len) > offset) - zero_len = offset - zero_off; - - error = xfs_iozero(ip, zero_off, zero_len); - if (error) { - goto out_lock; - } - - start_zero_fsb = imap.br_startoff + imap.br_blockcount; - ASSERT(start_zero_fsb <= (end_zero_fsb + 1)); - - xfs_ilock(ip, XFS_ILOCK_EXCL); - } - - return 0; - -out_lock: - xfs_ilock(ip, XFS_ILOCK_EXCL); - ASSERT(error >= 0); - return error; -} - -ssize_t /* bytes written, or (-) error */ -xfs_write( - struct xfs_inode *xip, - struct kiocb *iocb, - const struct iovec *iovp, - unsigned int nsegs, - loff_t *offset, - int ioflags) -{ - struct file *file = iocb->ki_filp; - struct address_space *mapping = file->f_mapping; - struct inode *inode = mapping->host; - unsigned long segs = nsegs; - xfs_mount_t *mp; - ssize_t ret = 0, error = 0; - xfs_fsize_t isize, new_size; - int iolock; - int eventsent = 0; - size_t ocount = 0, count; - loff_t pos; - int need_i_mutex; - - XFS_STATS_INC(xs_write_calls); - - error = generic_segment_checks(iovp, &segs, &ocount, VERIFY_READ); - if (error) - return error; - - count = ocount; - pos = *offset; - - if (count == 0) - return 0; - - mp = xip->i_mount; - - xfs_wait_for_freeze(mp, SB_FREEZE_WRITE); - - if (XFS_FORCED_SHUTDOWN(mp)) - return -EIO; - -relock: - if (ioflags & IO_ISDIRECT) { - iolock = XFS_IOLOCK_SHARED; - need_i_mutex = 0; - } else { - iolock = XFS_IOLOCK_EXCL; - need_i_mutex = 1; - mutex_lock(&inode->i_mutex); - } - - xfs_ilock(xip, XFS_ILOCK_EXCL|iolock); - -start: - error = -generic_write_checks(file, &pos, &count, - S_ISBLK(inode->i_mode)); - if (error) { - xfs_iunlock(xip, XFS_ILOCK_EXCL|iolock); - goto out_unlock_mutex; - } - - if ((DM_EVENT_ENABLED(xip, DM_EVENT_WRITE) && - !(ioflags & IO_INVIS) && !eventsent)) { - int dmflags = FILP_DELAY_FLAG(file); - - if (need_i_mutex) - dmflags |= DM_FLAGS_IMUX; - - xfs_iunlock(xip, XFS_ILOCK_EXCL); - error = XFS_SEND_DATA(xip->i_mount, DM_EVENT_WRITE, xip, - pos, count, dmflags, &iolock); - if (error) { - goto out_unlock_internal; - } - xfs_ilock(xip, XFS_ILOCK_EXCL); - eventsent = 1; - - /* - * The iolock was dropped and reacquired in XFS_SEND_DATA - * so we have to recheck the size when appending. - * We will only "goto start;" once, since having sent the - * event prevents another call to XFS_SEND_DATA, which is - * what allows the size to change in the first place. - */ - if ((file->f_flags & O_APPEND) && pos != xip->i_size) - goto start; - } - - if (ioflags & IO_ISDIRECT) { - xfs_buftarg_t *target = - XFS_IS_REALTIME_INODE(xip) ? - mp->m_rtdev_targp : mp->m_ddev_targp; - - if ((pos & target->bt_smask) || (count & target->bt_smask)) { - xfs_iunlock(xip, XFS_ILOCK_EXCL|iolock); - return XFS_ERROR(-EINVAL); - } - - if (!need_i_mutex && (mapping->nrpages || pos > xip->i_size)) { - xfs_iunlock(xip, XFS_ILOCK_EXCL|iolock); - iolock = XFS_IOLOCK_EXCL; - need_i_mutex = 1; - mutex_lock(&inode->i_mutex); - xfs_ilock(xip, XFS_ILOCK_EXCL|iolock); - goto start; - } - } - - new_size = pos + count; - if (new_size > xip->i_size) - xip->i_new_size = new_size; - - if (likely(!(ioflags & IO_INVIS))) - file_update_time(file); - - /* - * If the offset is beyond the size of the file, we have a couple - * of things to do. First, if there is already space allocated - * we need to either create holes or zero the disk or ... - * - * If there is a page where the previous size lands, we need - * to zero it out up to the new size. - */ - - if (pos > xip->i_size) { - error = xfs_zero_eof(xip, pos, xip->i_size); - if (error) { - xfs_iunlock(xip, XFS_ILOCK_EXCL); - goto out_unlock_internal; - } - } - xfs_iunlock(xip, XFS_ILOCK_EXCL); - - /* - * If we're writing the file then make sure to clear the - * setuid and setgid bits if the process is not being run - * by root. This keeps people from modifying setuid and - * setgid binaries. - */ - error = -file_remove_suid(file); - if (unlikely(error)) - goto out_unlock_internal; - - /* We can write back this queue in page reclaim */ - current->backing_dev_info = mapping->backing_dev_info; - - if ((ioflags & IO_ISDIRECT)) { - if (mapping->nrpages) { - WARN_ON(need_i_mutex == 0); - error = xfs_flushinval_pages(xip, - (pos & PAGE_CACHE_MASK), - -1, FI_REMAPF_LOCKED); - if (error) - goto out_unlock_internal; - } - - if (need_i_mutex) { - /* demote the lock now the cached pages are gone */ - xfs_ilock_demote(xip, XFS_IOLOCK_EXCL); - mutex_unlock(&inode->i_mutex); - - iolock = XFS_IOLOCK_SHARED; - need_i_mutex = 0; - } - - trace_xfs_file_direct_write(xip, count, *offset, ioflags); - ret = generic_file_direct_write(iocb, iovp, - &segs, pos, offset, count, ocount); - - /* - * direct-io write to a hole: fall through to buffered I/O - * for completing the rest of the request. - */ - if (ret >= 0 && ret != count) { - XFS_STATS_ADD(xs_write_bytes, ret); - - pos += ret; - count -= ret; - - ioflags &= ~IO_ISDIRECT; - xfs_iunlock(xip, iolock); - goto relock; - } - } else { - int enospc = 0; - ssize_t ret2 = 0; - -write_retry: - trace_xfs_file_buffered_write(xip, count, *offset, ioflags); - ret2 = generic_file_buffered_write(iocb, iovp, segs, - pos, offset, count, ret); - /* - * if we just got an ENOSPC, flush the inode now we - * aren't holding any page locks and retry *once* - */ - if (ret2 == -ENOSPC && !enospc) { - error = xfs_flush_pages(xip, 0, -1, 0, FI_NONE); - if (error) - goto out_unlock_internal; - enospc = 1; - goto write_retry; - } - ret = ret2; - } - - current->backing_dev_info = NULL; - - isize = i_size_read(inode); - if (unlikely(ret < 0 && ret != -EFAULT && *offset > isize)) - *offset = isize; - - if (*offset > xip->i_size) { - xfs_ilock(xip, XFS_ILOCK_EXCL); - if (*offset > xip->i_size) - xip->i_size = *offset; - xfs_iunlock(xip, XFS_ILOCK_EXCL); - } - - if (ret == -ENOSPC && - DM_EVENT_ENABLED(xip, DM_EVENT_NOSPACE) && !(ioflags & IO_INVIS)) { - xfs_iunlock(xip, iolock); - if (need_i_mutex) - mutex_unlock(&inode->i_mutex); - error = XFS_SEND_NAMESP(xip->i_mount, DM_EVENT_NOSPACE, xip, - DM_RIGHT_NULL, xip, DM_RIGHT_NULL, NULL, NULL, - 0, 0, 0); /* Delay flag intentionally unused */ - if (need_i_mutex) - mutex_lock(&inode->i_mutex); - xfs_ilock(xip, iolock); - if (error) - goto out_unlock_internal; - goto start; - } - - error = -ret; - if (ret <= 0) - goto out_unlock_internal; - - XFS_STATS_ADD(xs_write_bytes, ret); - - /* Handle various SYNC-type writes */ - if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) { - loff_t end = pos + ret - 1; - int error2; - - xfs_iunlock(xip, iolock); - if (need_i_mutex) - mutex_unlock(&inode->i_mutex); - - error2 = filemap_write_and_wait_range(mapping, pos, end); - if (!error) - error = error2; - if (need_i_mutex) - mutex_lock(&inode->i_mutex); - xfs_ilock(xip, iolock); - - error2 = xfs_fsync(xip); - if (!error) - error = error2; - } - - out_unlock_internal: - if (xip->i_new_size) { - xfs_ilock(xip, XFS_ILOCK_EXCL); - xip->i_new_size = 0; - /* - * If this was a direct or synchronous I/O that failed (such - * as ENOSPC) then part of the I/O may have been written to - * disk before the error occured. In this case the on-disk - * file size may have been adjusted beyond the in-memory file - * size and now needs to be truncated back. - */ - if (xip->i_d.di_size > xip->i_size) - xip->i_d.di_size = xip->i_size; - xfs_iunlock(xip, XFS_ILOCK_EXCL); - } - xfs_iunlock(xip, iolock); - out_unlock_mutex: - if (need_i_mutex) - mutex_unlock(&inode->i_mutex); - return -error; -} - -/* - * If the underlying (data/log/rt) device is readonly, there are some - * operations that cannot proceed. - */ -int -xfs_dev_is_read_only( - xfs_mount_t *mp, - char *message) -{ - if (xfs_readonly_buftarg(mp->m_ddev_targp) || - xfs_readonly_buftarg(mp->m_logdev_targp) || - (mp->m_rtdev_targp && xfs_readonly_buftarg(mp->m_rtdev_targp))) { - cmn_err(CE_NOTE, - "XFS: %s required on read-only device.", message); - cmn_err(CE_NOTE, - "XFS: write access unavailable, cannot proceed."); - return EROFS; - } - return 0; -} diff --git a/fs/xfs/linux-2.6/xfs_lrw.h b/fs/xfs/linux-2.6/xfs_lrw.h deleted file mode 100644 index 342ae8c0d011..000000000000 --- a/fs/xfs/linux-2.6/xfs_lrw.h +++ /dev/null @@ -1,29 +0,0 @@ -/* - * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_LRW_H__ -#define __XFS_LRW_H__ - -struct xfs_mount; -struct xfs_inode; -struct xfs_buf; - -extern int xfs_dev_is_read_only(struct xfs_mount *, char *); - -extern int xfs_zero_eof(struct xfs_inode *, xfs_off_t, xfs_fsize_t); - -#endif /* __XFS_LRW_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c index a9f6d20aff41..05cd85317f6f 100644 --- a/fs/xfs/linux-2.6/xfs_sync.c +++ b/fs/xfs/linux-2.6/xfs_sync.c @@ -607,7 +607,8 @@ xfssyncd( set_freezable(); timeleft = xfs_syncd_centisecs * msecs_to_jiffies(10); for (;;) { - timeleft = schedule_timeout_interruptible(timeleft); + if (list_empty(&mp->m_sync_list)) + timeleft = schedule_timeout_interruptible(timeleft); /* swsusp */ try_to_freeze(); if (kthread_should_stop() && list_empty(&mp->m_sync_list)) @@ -627,8 +628,7 @@ xfssyncd( list_add_tail(&mp->m_sync_work.w_list, &mp->m_sync_list); } - list_for_each_entry_safe(work, n, &mp->m_sync_list, w_list) - list_move(&work->w_list, &tmp); + list_splice_init(&mp->m_sync_list, &tmp); spin_unlock(&mp->m_sync_lock); list_for_each_entry_safe(work, n, &tmp, w_list) { @@ -688,12 +688,12 @@ xfs_inode_set_reclaim_tag( struct xfs_perag *pag; pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ip->i_ino)); - read_lock(&pag->pag_ici_lock); + write_lock(&pag->pag_ici_lock); spin_lock(&ip->i_flags_lock); __xfs_inode_set_reclaim_tag(pag, ip); __xfs_iflags_set(ip, XFS_IRECLAIMABLE); spin_unlock(&ip->i_flags_lock); - read_unlock(&pag->pag_ici_lock); + write_unlock(&pag->pag_ici_lock); xfs_perag_put(pag); } diff --git a/fs/xfs/linux-2.6/xfs_trace.c b/fs/xfs/linux-2.6/xfs_trace.c index 856eb3c8d605..5a107601e969 100644 --- a/fs/xfs/linux-2.6/xfs_trace.c +++ b/fs/xfs/linux-2.6/xfs_trace.c @@ -52,22 +52,6 @@ #include "quota/xfs_dquot.h" /* - * Format fsblock number into a static buffer & return it. - */ -STATIC char *xfs_fmtfsblock(xfs_fsblock_t bno) -{ - static char rval[50]; - - if (bno == NULLFSBLOCK) - sprintf(rval, "NULLFSBLOCK"); - else if (isnullstartblock(bno)) - sprintf(rval, "NULLSTARTBLOCK(%lld)", startblockval(bno)); - else - sprintf(rval, "%lld", (xfs_dfsbno_t)bno); - return rval; -} - -/* * We include this last to have the helpers above available for the trace * event implementations. */ diff --git a/fs/xfs/linux-2.6/xfs_trace.h b/fs/xfs/linux-2.6/xfs_trace.h index a4574dcf5065..fcaa62f0799e 100644 --- a/fs/xfs/linux-2.6/xfs_trace.h +++ b/fs/xfs/linux-2.6/xfs_trace.h @@ -197,13 +197,13 @@ TRACE_EVENT(xfs_iext_insert, __entry->caller_ip = caller_ip; ), TP_printk("dev %d:%d ino 0x%llx state %s idx %ld " - "offset %lld block %s count %lld flag %d caller %pf", + "offset %lld block %lld count %lld flag %d caller %pf", MAJOR(__entry->dev), MINOR(__entry->dev), __entry->ino, __print_flags(__entry->bmap_state, "|", XFS_BMAP_EXT_FLAGS), (long)__entry->idx, __entry->startoff, - xfs_fmtfsblock(__entry->startblock), + (__int64_t)__entry->startblock, __entry->blockcount, __entry->state, (char *)__entry->caller_ip) @@ -241,13 +241,13 @@ DECLARE_EVENT_CLASS(xfs_bmap_class, __entry->caller_ip = caller_ip; ), TP_printk("dev %d:%d ino 0x%llx state %s idx %ld " - "offset %lld block %s count %lld flag %d caller %pf", + "offset %lld block %lld count %lld flag %d caller %pf", MAJOR(__entry->dev), MINOR(__entry->dev), __entry->ino, __print_flags(__entry->bmap_state, "|", XFS_BMAP_EXT_FLAGS), (long)__entry->idx, __entry->startoff, - xfs_fmtfsblock(__entry->startblock), + (__int64_t)__entry->startblock, __entry->blockcount, __entry->state, (char *)__entry->caller_ip) @@ -593,7 +593,7 @@ DECLARE_EVENT_CLASS(xfs_dquot_class, TP_ARGS(dqp), TP_STRUCT__entry( __field(dev_t, dev) - __field(__be32, id) + __field(u32, id) __field(unsigned, flags) __field(unsigned, nrefs) __field(unsigned long long, res_bcount) @@ -606,7 +606,7 @@ DECLARE_EVENT_CLASS(xfs_dquot_class, ), \ TP_fast_assign( __entry->dev = dqp->q_mount->m_super->s_dev; - __entry->id = dqp->q_core.d_id; + __entry->id = be32_to_cpu(dqp->q_core.d_id); __entry->flags = dqp->dq_flags; __entry->nrefs = dqp->q_nrefs; __entry->res_bcount = dqp->q_res_bcount; @@ -622,10 +622,10 @@ DECLARE_EVENT_CLASS(xfs_dquot_class, be64_to_cpu(dqp->q_core.d_ino_softlimit); ), TP_printk("dev %d:%d id 0x%x flags %s nrefs %u res_bc 0x%llx " - "bcnt 0x%llx [hard 0x%llx | soft 0x%llx] " - "icnt 0x%llx [hard 0x%llx | soft 0x%llx]", + "bcnt 0x%llx bhardlimit 0x%llx bsoftlimit 0x%llx " + "icnt 0x%llx ihardlimit 0x%llx isoftlimit 0x%llx]", MAJOR(__entry->dev), MINOR(__entry->dev), - be32_to_cpu(__entry->id), + __entry->id, __print_flags(__entry->flags, "|", XFS_DQ_FLAGS), __entry->nrefs, __entry->res_bcount, @@ -881,7 +881,7 @@ TRACE_EVENT(name, \ ), \ TP_printk("dev %d:%d ino 0x%llx size 0x%llx new_size 0x%llx " \ "offset 0x%llx count %zd flags %s " \ - "startoff 0x%llx startblock %s blockcount 0x%llx", \ + "startoff 0x%llx startblock %lld blockcount 0x%llx", \ MAJOR(__entry->dev), MINOR(__entry->dev), \ __entry->ino, \ __entry->size, \ @@ -890,7 +890,7 @@ TRACE_EVENT(name, \ __entry->count, \ __print_flags(__entry->flags, "|", BMAPI_FLAGS), \ __entry->startoff, \ - xfs_fmtfsblock(__entry->startblock), \ + (__int64_t)__entry->startblock, \ __entry->blockcount) \ ) DEFINE_IOMAP_EVENT(xfs_iomap_enter); diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 1869fb973819..5c11e4d17010 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -2550,22 +2550,134 @@ xfs_bmap_rtalloc( } STATIC int +xfs_bmap_btalloc_nullfb( + struct xfs_bmalloca *ap, + struct xfs_alloc_arg *args, + xfs_extlen_t *blen) +{ + struct xfs_mount *mp = ap->ip->i_mount; + struct xfs_perag *pag; + xfs_agnumber_t ag, startag; + int notinit = 0; + int error; + + if (ap->userdata && xfs_inode_is_filestream(ap->ip)) + args->type = XFS_ALLOCTYPE_NEAR_BNO; + else + args->type = XFS_ALLOCTYPE_START_BNO; + args->total = ap->total; + + /* + * Search for an allocation group with a single extent large enough + * for the request. If one isn't found, then adjust the minimum + * allocation size to the largest space found. + */ + startag = ag = XFS_FSB_TO_AGNO(mp, args->fsbno); + if (startag == NULLAGNUMBER) + startag = ag = 0; + + pag = xfs_perag_get(mp, ag); + while (*blen < ap->alen) { + if (!pag->pagf_init) { + error = xfs_alloc_pagf_init(mp, args->tp, ag, + XFS_ALLOC_FLAG_TRYLOCK); + if (error) { + xfs_perag_put(pag); + return error; + } + } + + /* + * See xfs_alloc_fix_freelist... + */ + if (pag->pagf_init) { + xfs_extlen_t longest; + longest = xfs_alloc_longest_free_extent(mp, pag); + if (*blen < longest) + *blen = longest; + } else + notinit = 1; + + if (xfs_inode_is_filestream(ap->ip)) { + if (*blen >= ap->alen) + break; + + if (ap->userdata) { + /* + * If startag is an invalid AG, we've + * come here once before and + * xfs_filestream_new_ag picked the + * best currently available. + * + * Don't continue looping, since we + * could loop forever. + */ + if (startag == NULLAGNUMBER) + break; + + error = xfs_filestream_new_ag(ap, &ag); + xfs_perag_put(pag); + if (error) + return error; + + /* loop again to set 'blen'*/ + startag = NULLAGNUMBER; + pag = xfs_perag_get(mp, ag); + continue; + } + } + if (++ag == mp->m_sb.sb_agcount) + ag = 0; + if (ag == startag) + break; + xfs_perag_put(pag); + pag = xfs_perag_get(mp, ag); + } + xfs_perag_put(pag); + + /* + * Since the above loop did a BUF_TRYLOCK, it is + * possible that there is space for this request. + */ + if (notinit || *blen < ap->minlen) + args->minlen = ap->minlen; + /* + * If the best seen length is less than the request + * length, use the best as the minimum. + */ + else if (*blen < ap->alen) + args->minlen = *blen; + /* + * Otherwise we've seen an extent as big as alen, + * use that as the minimum. + */ + else + args->minlen = ap->alen; + + /* + * set the failure fallback case to look in the selected + * AG as the stream may have moved. + */ + if (xfs_inode_is_filestream(ap->ip)) + ap->rval = args->fsbno = XFS_AGB_TO_FSB(mp, ag, 0); + + return 0; +} + +STATIC int xfs_bmap_btalloc( xfs_bmalloca_t *ap) /* bmap alloc argument struct */ { xfs_mount_t *mp; /* mount point structure */ xfs_alloctype_t atype = 0; /* type for allocation routines */ xfs_extlen_t align; /* minimum allocation alignment */ - xfs_agnumber_t ag; xfs_agnumber_t fb_agno; /* ag number of ap->firstblock */ - xfs_agnumber_t startag; + xfs_agnumber_t ag; xfs_alloc_arg_t args; xfs_extlen_t blen; xfs_extlen_t nextminlen = 0; - xfs_perag_t *pag; int nullfb; /* true if ap->firstblock isn't set */ int isaligned; - int notinit; int tryagain; int error; @@ -2612,103 +2724,9 @@ xfs_bmap_btalloc( args.firstblock = ap->firstblock; blen = 0; if (nullfb) { - if (ap->userdata && xfs_inode_is_filestream(ap->ip)) - args.type = XFS_ALLOCTYPE_NEAR_BNO; - else - args.type = XFS_ALLOCTYPE_START_BNO; - args.total = ap->total; - - /* - * Search for an allocation group with a single extent - * large enough for the request. - * - * If one isn't found, then adjust the minimum allocation - * size to the largest space found. - */ - startag = ag = XFS_FSB_TO_AGNO(mp, args.fsbno); - if (startag == NULLAGNUMBER) - startag = ag = 0; - notinit = 0; - pag = xfs_perag_get(mp, ag); - while (blen < ap->alen) { - if (!pag->pagf_init && - (error = xfs_alloc_pagf_init(mp, args.tp, - ag, XFS_ALLOC_FLAG_TRYLOCK))) { - xfs_perag_put(pag); - return error; - } - /* - * See xfs_alloc_fix_freelist... - */ - if (pag->pagf_init) { - xfs_extlen_t longest; - longest = xfs_alloc_longest_free_extent(mp, pag); - if (blen < longest) - blen = longest; - } else - notinit = 1; - - if (xfs_inode_is_filestream(ap->ip)) { - if (blen >= ap->alen) - break; - - if (ap->userdata) { - /* - * If startag is an invalid AG, we've - * come here once before and - * xfs_filestream_new_ag picked the - * best currently available. - * - * Don't continue looping, since we - * could loop forever. - */ - if (startag == NULLAGNUMBER) - break; - - error = xfs_filestream_new_ag(ap, &ag); - xfs_perag_put(pag); - if (error) - return error; - - /* loop again to set 'blen'*/ - startag = NULLAGNUMBER; - pag = xfs_perag_get(mp, ag); - continue; - } - } - if (++ag == mp->m_sb.sb_agcount) - ag = 0; - if (ag == startag) - break; - xfs_perag_put(pag); - pag = xfs_perag_get(mp, ag); - } - xfs_perag_put(pag); - /* - * Since the above loop did a BUF_TRYLOCK, it is - * possible that there is space for this request. - */ - if (notinit || blen < ap->minlen) - args.minlen = ap->minlen; - /* - * If the best seen length is less than the request - * length, use the best as the minimum. - */ - else if (blen < ap->alen) - args.minlen = blen; - /* - * Otherwise we've seen an extent as big as alen, - * use that as the minimum. - */ - else - args.minlen = ap->alen; - - /* - * set the failure fallback case to look in the selected - * AG as the stream may have moved. - */ - if (xfs_inode_is_filestream(ap->ip)) - ap->rval = args.fsbno = XFS_AGB_TO_FSB(mp, ag, 0); + error = xfs_bmap_btalloc_nullfb(ap, &args, &blen); + if (error) + return error; } else if (ap->low) { if (xfs_inode_is_filestream(ap->ip)) args.type = XFS_ALLOCTYPE_FIRST_AG; diff --git a/fs/xfs/xfs_fs.h b/fs/xfs/xfs_fs.h index f52ac276277e..7cf7220e7d5f 100644 --- a/fs/xfs/xfs_fs.h +++ b/fs/xfs/xfs_fs.h @@ -292,7 +292,8 @@ typedef struct xfs_bstat { __s32 bs_extents; /* number of extents */ __u32 bs_gen; /* generation count */ __u16 bs_projid; /* project id */ - unsigned char bs_pad[14]; /* pad space, unused */ + __u16 bs_forkoff; /* inode fork offset in bytes */ + unsigned char bs_pad[12]; /* pad space, unused */ __u32 bs_dmevmask; /* DMIG event mask */ __u16 bs_dmstate; /* DMIG state info */ __u16 bs_aextents; /* attribute number of extents */ diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c index e281eb4a1c49..6845db90818f 100644 --- a/fs/xfs/xfs_iget.c +++ b/fs/xfs/xfs_iget.c @@ -190,13 +190,12 @@ xfs_iget_cache_hit( trace_xfs_iget_reclaim(ip); /* - * We need to set XFS_INEW atomically with clearing the - * reclaimable tag so that we do have an indicator of the - * inode still being initialized. + * We need to set XFS_IRECLAIM to prevent xfs_reclaim_inode + * from stomping over us while we recycle the inode. We can't + * clear the radix tree reclaimable tag yet as it requires + * pag_ici_lock to be held exclusive. */ - ip->i_flags |= XFS_INEW; - ip->i_flags &= ~XFS_IRECLAIMABLE; - __xfs_inode_clear_reclaim_tag(mp, pag, ip); + ip->i_flags |= XFS_IRECLAIM; spin_unlock(&ip->i_flags_lock); read_unlock(&pag->pag_ici_lock); @@ -216,7 +215,15 @@ xfs_iget_cache_hit( trace_xfs_iget_reclaim(ip); goto out_error; } + + write_lock(&pag->pag_ici_lock); + spin_lock(&ip->i_flags_lock); + ip->i_flags &= ~(XFS_IRECLAIMABLE | XFS_IRECLAIM); + ip->i_flags |= XFS_INEW; + __xfs_inode_clear_reclaim_tag(mp, pag, ip); inode->i_state = I_NEW; + spin_unlock(&ip->i_flags_lock); + write_unlock(&pag->pag_ici_lock); } else { /* If the VFS inode is being torn down, pause and try again. */ if (!igrab(inode)) { diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index fa31360046d4..0ffd56447045 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -2439,75 +2439,31 @@ xfs_idestroy_fork( } /* - * Increment the pin count of the given buffer. - * This value is protected by ipinlock spinlock in the mount structure. + * This is called to unpin an inode. The caller must have the inode locked + * in at least shared mode so that the buffer cannot be subsequently pinned + * once someone is waiting for it to be unpinned. */ -void -xfs_ipin( - xfs_inode_t *ip) -{ - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - - atomic_inc(&ip->i_pincount); -} - -/* - * Decrement the pin count of the given inode, and wake up - * anyone in xfs_iwait_unpin() if the count goes to 0. The - * inode must have been previously pinned with a call to xfs_ipin(). - */ -void -xfs_iunpin( - xfs_inode_t *ip) -{ - ASSERT(atomic_read(&ip->i_pincount) > 0); - - if (atomic_dec_and_test(&ip->i_pincount)) - wake_up(&ip->i_ipin_wait); -} - -/* - * This is called to unpin an inode. It can be directed to wait or to return - * immediately without waiting for the inode to be unpinned. The caller must - * have the inode locked in at least shared mode so that the buffer cannot be - * subsequently pinned once someone is waiting for it to be unpinned. - */ -STATIC void -__xfs_iunpin_wait( - xfs_inode_t *ip, - int wait) +static void +xfs_iunpin_nowait( + struct xfs_inode *ip) { - xfs_inode_log_item_t *iip = ip->i_itemp; - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED)); - if (atomic_read(&ip->i_pincount) == 0) - return; /* Give the log a push to start the unpinning I/O */ - if (iip && iip->ili_last_lsn) - xfs_log_force_lsn(ip->i_mount, iip->ili_last_lsn, 0); - else - xfs_log_force(ip->i_mount, 0); + xfs_log_force_lsn(ip->i_mount, ip->i_itemp->ili_last_lsn, 0); - if (wait) - wait_event(ip->i_ipin_wait, (atomic_read(&ip->i_pincount) == 0)); } void xfs_iunpin_wait( - xfs_inode_t *ip) + struct xfs_inode *ip) { - __xfs_iunpin_wait(ip, 1); -} - -static inline void -xfs_iunpin_nowait( - xfs_inode_t *ip) -{ - __xfs_iunpin_wait(ip, 0); + if (xfs_ipincount(ip)) { + xfs_iunpin_nowait(ip); + wait_event(ip->i_ipin_wait, (xfs_ipincount(ip) == 0)); + } } - /* * xfs_iextents_copy() * diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index 6c912b027596..9965e40a4615 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -471,8 +471,6 @@ int xfs_itruncate_finish(struct xfs_trans **, xfs_inode_t *, int xfs_iunlink(struct xfs_trans *, xfs_inode_t *); void xfs_iext_realloc(xfs_inode_t *, int, int); -void xfs_ipin(xfs_inode_t *); -void xfs_iunpin(xfs_inode_t *); void xfs_iunpin_wait(xfs_inode_t *); int xfs_iflush(xfs_inode_t *, uint); void xfs_ichgtime(xfs_inode_t *, int); @@ -480,6 +478,7 @@ void xfs_lock_inodes(xfs_inode_t **, int, uint); void xfs_lock_two_inodes(xfs_inode_t *, xfs_inode_t *, uint); void xfs_synchronize_times(xfs_inode_t *); +void xfs_mark_inode_dirty(xfs_inode_t *); void xfs_mark_inode_dirty_sync(xfs_inode_t *); #define IHOLD(ip) \ diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c index d4dc063111f8..7bfea8540159 100644 --- a/fs/xfs/xfs_inode_item.c +++ b/fs/xfs/xfs_inode_item.c @@ -535,23 +535,23 @@ xfs_inode_item_format( /* * This is called to pin the inode associated with the inode log - * item in memory so it cannot be written out. Do this by calling - * xfs_ipin() to bump the pin count in the inode while holding the - * inode pin lock. + * item in memory so it cannot be written out. */ STATIC void xfs_inode_item_pin( xfs_inode_log_item_t *iip) { ASSERT(xfs_isilocked(iip->ili_inode, XFS_ILOCK_EXCL)); - xfs_ipin(iip->ili_inode); + + atomic_inc(&iip->ili_inode->i_pincount); } /* * This is called to unpin the inode associated with the inode log * item which was previously pinned with a call to xfs_inode_item_pin(). - * Just call xfs_iunpin() on the inode to do this. + * + * Also wake up anyone in xfs_iunpin_wait() if the count goes to 0. */ /* ARGSUSED */ STATIC void @@ -559,7 +559,11 @@ xfs_inode_item_unpin( xfs_inode_log_item_t *iip, int stale) { - xfs_iunpin(iip->ili_inode); + struct xfs_inode *ip = iip->ili_inode; + + ASSERT(atomic_read(&ip->i_pincount) > 0); + if (atomic_dec_and_test(&ip->i_pincount)) + wake_up(&ip->i_ipin_wait); } /* ARGSUSED */ @@ -568,7 +572,7 @@ xfs_inode_item_unpin_remove( xfs_inode_log_item_t *iip, xfs_trans_t *tp) { - xfs_iunpin(iip->ili_inode); + xfs_inode_item_unpin(iip, 0); } /* diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c index 3af02314c605..b1b801e4a28e 100644 --- a/fs/xfs/xfs_itable.c +++ b/fs/xfs/xfs_itable.c @@ -106,6 +106,7 @@ xfs_bulkstat_one_iget( buf->bs_dmevmask = dic->di_dmevmask; buf->bs_dmstate = dic->di_dmstate; buf->bs_aextents = dic->di_anextents; + buf->bs_forkoff = XFS_IFORK_BOFF(ip); switch (dic->di_format) { case XFS_DINODE_FMT_DEV: @@ -176,6 +177,7 @@ xfs_bulkstat_one_dinode( buf->bs_dmevmask = be32_to_cpu(dic->di_dmevmask); buf->bs_dmstate = be16_to_cpu(dic->di_dmstate); buf->bs_aextents = be16_to_cpu(dic->di_anextents); + buf->bs_forkoff = XFS_DFORK_BOFF(dic); switch (dic->di_format) { case XFS_DINODE_FMT_DEV: diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index 4f16be4b6ee5..e8fba92d7cd9 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -60,7 +60,7 @@ STATIC int xlog_space_left(xlog_t *log, int cycle, int bytes); STATIC int xlog_sync(xlog_t *log, xlog_in_core_t *iclog); STATIC void xlog_dealloc_log(xlog_t *log); STATIC int xlog_write(xfs_mount_t *mp, xfs_log_iovec_t region[], - int nentries, xfs_log_ticket_t tic, + int nentries, struct xlog_ticket *tic, xfs_lsn_t *start_lsn, xlog_in_core_t **commit_iclog, uint flags); @@ -243,14 +243,14 @@ xlog_tic_add_region(xlog_ticket_t *tic, uint len, uint type) * out when the next write occurs. */ xfs_lsn_t -xfs_log_done(xfs_mount_t *mp, - xfs_log_ticket_t xtic, - void **iclog, - uint flags) +xfs_log_done( + struct xfs_mount *mp, + struct xlog_ticket *ticket, + struct xlog_in_core **iclog, + uint flags) { - xlog_t *log = mp->m_log; - xlog_ticket_t *ticket = (xfs_log_ticket_t) xtic; - xfs_lsn_t lsn = 0; + struct log *log = mp->m_log; + xfs_lsn_t lsn = 0; if (XLOG_FORCED_SHUTDOWN(log) || /* @@ -258,8 +258,7 @@ xfs_log_done(xfs_mount_t *mp, * If we get an error, just continue and give back the log ticket. */ (((ticket->t_flags & XLOG_TIC_INITED) == 0) && - (xlog_commit_record(mp, ticket, - (xlog_in_core_t **)iclog, &lsn)))) { + (xlog_commit_record(mp, ticket, iclog, &lsn)))) { lsn = (xfs_lsn_t) -1; if (ticket->t_flags & XLOG_TIC_PERM_RESERV) { flags |= XFS_LOG_REL_PERM_RESERV; @@ -289,7 +288,7 @@ xfs_log_done(xfs_mount_t *mp, } return lsn; -} /* xfs_log_done */ +} /* * Attaches a new iclog I/O completion callback routine during @@ -298,11 +297,11 @@ xfs_log_done(xfs_mount_t *mp, * executing the callback at an appropriate time. */ int -xfs_log_notify(xfs_mount_t *mp, /* mount of partition */ - void *iclog_hndl, /* iclog to hang callback off */ - xfs_log_callback_t *cb) +xfs_log_notify( + struct xfs_mount *mp, + struct xlog_in_core *iclog, + xfs_log_callback_t *cb) { - xlog_in_core_t *iclog = (xlog_in_core_t *)iclog_hndl; int abortflg; spin_lock(&iclog->ic_callback_lock); @@ -316,16 +315,14 @@ xfs_log_notify(xfs_mount_t *mp, /* mount of partition */ } spin_unlock(&iclog->ic_callback_lock); return abortflg; -} /* xfs_log_notify */ +} int -xfs_log_release_iclog(xfs_mount_t *mp, - void *iclog_hndl) +xfs_log_release_iclog( + struct xfs_mount *mp, + struct xlog_in_core *iclog) { - xlog_t *log = mp->m_log; - xlog_in_core_t *iclog = (xlog_in_core_t *)iclog_hndl; - - if (xlog_state_release_iclog(log, iclog)) { + if (xlog_state_release_iclog(mp->m_log, iclog)) { xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR); return EIO; } @@ -344,17 +341,18 @@ xfs_log_release_iclog(xfs_mount_t *mp, * reservation, we prevent over allocation problems. */ int -xfs_log_reserve(xfs_mount_t *mp, - int unit_bytes, - int cnt, - xfs_log_ticket_t *ticket, - __uint8_t client, - uint flags, - uint t_type) +xfs_log_reserve( + struct xfs_mount *mp, + int unit_bytes, + int cnt, + struct xlog_ticket **ticket, + __uint8_t client, + uint flags, + uint t_type) { - xlog_t *log = mp->m_log; - xlog_ticket_t *internal_ticket; - int retval = 0; + struct log *log = mp->m_log; + struct xlog_ticket *internal_ticket; + int retval = 0; ASSERT(client == XFS_TRANSACTION || client == XFS_LOG); ASSERT((flags & XFS_LOG_NOSLEEP) == 0); @@ -367,7 +365,7 @@ xfs_log_reserve(xfs_mount_t *mp, if (*ticket != NULL) { ASSERT(flags & XFS_LOG_PERM_RESERV); - internal_ticket = (xlog_ticket_t *)*ticket; + internal_ticket = *ticket; trace_xfs_log_reserve(log, internal_ticket); @@ -519,7 +517,7 @@ xfs_log_unmount_write(xfs_mount_t *mp) xlog_in_core_t *first_iclog; #endif xfs_log_iovec_t reg[1]; - xfs_log_ticket_t tic = NULL; + xlog_ticket_t *tic = NULL; xfs_lsn_t lsn; int error; @@ -656,24 +654,24 @@ xfs_log_unmount(xfs_mount_t *mp) * transaction occur with one call to xfs_log_write(). */ int -xfs_log_write(xfs_mount_t * mp, - xfs_log_iovec_t reg[], - int nentries, - xfs_log_ticket_t tic, - xfs_lsn_t *start_lsn) +xfs_log_write( + struct xfs_mount *mp, + struct xfs_log_iovec reg[], + int nentries, + struct xlog_ticket *tic, + xfs_lsn_t *start_lsn) { - int error; - xlog_t *log = mp->m_log; + struct log *log = mp->m_log; + int error; if (XLOG_FORCED_SHUTDOWN(log)) return XFS_ERROR(EIO); - if ((error = xlog_write(mp, reg, nentries, tic, start_lsn, NULL, 0))) { + error = xlog_write(mp, reg, nentries, tic, start_lsn, NULL, 0); + if (error) xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR); - } return error; -} /* xfs_log_write */ - +} void xfs_log_move_tail(xfs_mount_t *mp, @@ -1642,16 +1640,16 @@ xlog_print_tic_res(xfs_mount_t *mp, xlog_ticket_t *ticket) * bytes have been written out. */ STATIC int -xlog_write(xfs_mount_t * mp, - xfs_log_iovec_t reg[], - int nentries, - xfs_log_ticket_t tic, - xfs_lsn_t *start_lsn, - xlog_in_core_t **commit_iclog, - uint flags) +xlog_write( + struct xfs_mount *mp, + struct xfs_log_iovec reg[], + int nentries, + struct xlog_ticket *ticket, + xfs_lsn_t *start_lsn, + struct xlog_in_core **commit_iclog, + uint flags) { xlog_t *log = mp->m_log; - xlog_ticket_t *ticket = (xlog_ticket_t *)tic; xlog_in_core_t *iclog = NULL; /* ptr to current in-core log */ xlog_op_header_t *logop_head; /* ptr to log operation header */ __psint_t ptr; /* copy address into data region */ @@ -1765,7 +1763,7 @@ xlog_write(xfs_mount_t * mp, default: xfs_fs_cmn_err(CE_WARN, mp, "Bad XFS transaction clientid 0x%x in ticket 0x%p", - logop_head->oh_clientid, tic); + logop_head->oh_clientid, ticket); return XFS_ERROR(EIO); } diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h index 7074be9d13e9..97a24c7795a4 100644 --- a/fs/xfs/xfs_log.h +++ b/fs/xfs/xfs_log.h @@ -110,8 +110,6 @@ typedef struct xfs_log_iovec { uint i_type; /* type of region */ } xfs_log_iovec_t; -typedef void* xfs_log_ticket_t; - /* * Structure used to pass callback function and the function's argument * to the log manager. @@ -126,10 +124,12 @@ typedef struct xfs_log_callback { #ifdef __KERNEL__ /* Log manager interfaces */ struct xfs_mount; +struct xlog_in_core; struct xlog_ticket; + xfs_lsn_t xfs_log_done(struct xfs_mount *mp, - xfs_log_ticket_t ticket, - void **iclog, + struct xlog_ticket *ticket, + struct xlog_in_core **iclog, uint flags); int _xfs_log_force(struct xfs_mount *mp, uint flags, @@ -151,21 +151,21 @@ int xfs_log_mount_finish(struct xfs_mount *mp); void xfs_log_move_tail(struct xfs_mount *mp, xfs_lsn_t tail_lsn); int xfs_log_notify(struct xfs_mount *mp, - void *iclog, + struct xlog_in_core *iclog, xfs_log_callback_t *callback_entry); int xfs_log_release_iclog(struct xfs_mount *mp, - void *iclog_hndl); + struct xlog_in_core *iclog); int xfs_log_reserve(struct xfs_mount *mp, int length, int count, - xfs_log_ticket_t *ticket, + struct xlog_ticket **ticket, __uint8_t clientid, uint flags, uint t_type); int xfs_log_write(struct xfs_mount *mp, xfs_log_iovec_t region[], int nentries, - xfs_log_ticket_t ticket, + struct xlog_ticket *ticket, xfs_lsn_t *start_lsn); int xfs_log_unmount_write(struct xfs_mount *mp); void xfs_log_unmount(struct xfs_mount *mp); diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 6afaaeb2950a..e79b56b4bca6 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -1097,13 +1097,15 @@ xfs_default_resblks(xfs_mount_t *mp) __uint64_t resblks; /* - * We default to 5% or 1024 fsbs of space reserved, whichever is smaller. - * This may drive us straight to ENOSPC on mount, but that implies - * we were already there on the last unmount. Warn if this occurs. + * We default to 5% or 8192 fsbs of space reserved, whichever is + * smaller. This is intended to cover concurrent allocation + * transactions when we initially hit enospc. These each require a 4 + * block reservation. Hence by default we cover roughly 2000 concurrent + * allocation reservations. */ resblks = mp->m_sb.sb_dblocks; do_div(resblks, 20); - resblks = min_t(__uint64_t, resblks, 1024); + resblks = min_t(__uint64_t, resblks, 8192); return resblks; } @@ -1417,6 +1419,9 @@ xfs_mountfs( * when at ENOSPC. This is needed for operations like create with * attr, unwritten extent conversion at ENOSPC, etc. Data allocations * are not allowed to use this reserved space. + * + * This may drive us straight to ENOSPC on mount, but that implies + * we were already there on the last unmount. Warn if this occurs. */ if (!(mp->m_flags & XFS_MOUNT_RDONLY)) { resblks = xfs_default_resblks(mp); @@ -1725,26 +1730,30 @@ xfs_mod_incore_sb_unlocked( lcounter += rem; } } else { /* Taking blocks away */ - lcounter += delta; + if (lcounter >= 0) { + mp->m_sb.sb_fdblocks = lcounter + + XFS_ALLOC_SET_ASIDE(mp); + return 0; + } - /* - * If were out of blocks, use any available reserved blocks if - * were allowed to. - */ + /* + * We are out of blocks, use any available reserved + * blocks if were allowed to. + */ + if (!rsvd) + return XFS_ERROR(ENOSPC); - if (lcounter < 0) { - if (rsvd) { - lcounter = (long long)mp->m_resblks_avail + delta; - if (lcounter < 0) { - return XFS_ERROR(ENOSPC); - } - mp->m_resblks_avail = lcounter; - return 0; - } else { /* not reserved */ - return XFS_ERROR(ENOSPC); - } + lcounter = (long long)mp->m_resblks_avail + delta; + if (lcounter >= 0) { + mp->m_resblks_avail = lcounter; + return 0; } + printk_once(KERN_WARNING + "Filesystem \"%s\": reserve blocks depleted! " + "Consider increasing reserve pool size.", + mp->m_fsname); + return XFS_ERROR(ENOSPC); } mp->m_sb.sb_fdblocks = lcounter + XFS_ALLOC_SET_ASIDE(mp); @@ -2052,6 +2061,26 @@ xfs_mount_log_sb( return error; } +/* + * If the underlying (data/log/rt) device is readonly, there are some + * operations that cannot proceed. + */ +int +xfs_dev_is_read_only( + struct xfs_mount *mp, + char *message) +{ + if (xfs_readonly_buftarg(mp->m_ddev_targp) || + xfs_readonly_buftarg(mp->m_logdev_targp) || + (mp->m_rtdev_targp && xfs_readonly_buftarg(mp->m_rtdev_targp))) { + cmn_err(CE_NOTE, + "XFS: %s required on read-only device.", message); + cmn_err(CE_NOTE, + "XFS: write access unavailable, cannot proceed."); + return EROFS; + } + return 0; +} #ifdef HAVE_PERCPU_SB /* diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 14dafd608230..4fa0bc7b983e 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -436,6 +436,8 @@ extern void xfs_freesb(xfs_mount_t *); extern int xfs_fs_writable(xfs_mount_t *); extern int xfs_sb_validate_fsb_count(struct xfs_sb *, __uint64_t); +extern int xfs_dev_is_read_only(struct xfs_mount *, char *); + extern int xfs_dmops_get(struct xfs_mount *); extern void xfs_dmops_put(struct xfs_mount *); diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c index be942d4e3324..f73e358bae8d 100644 --- a/fs/xfs/xfs_trans.c +++ b/fs/xfs/xfs_trans.c @@ -796,7 +796,7 @@ _xfs_trans_commit( int sync; #define XFS_TRANS_LOGVEC_COUNT 16 xfs_log_iovec_t log_vector_fast[XFS_TRANS_LOGVEC_COUNT]; - void *commit_iclog; + struct xlog_in_core *commit_iclog; int shutdown; commit_lsn = -1; diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h index c93e3a102857..79c8bab9dfff 100644 --- a/fs/xfs/xfs_trans.h +++ b/fs/xfs/xfs_trans.h @@ -910,7 +910,7 @@ typedef struct xfs_trans { unsigned int t_blk_res_used; /* # of resvd blocks used */ unsigned int t_rtx_res; /* # of rt extents resvd */ unsigned int t_rtx_res_used; /* # of resvd rt extents used */ - xfs_log_ticket_t t_ticket; /* log mgr ticket */ + struct xlog_ticket *t_ticket; /* log mgr ticket */ xfs_lsn_t t_lsn; /* log seq num of start of * transaction. */ xfs_lsn_t t_commit_lsn; /* log seq num of end of diff --git a/fs/xfs/xfs_trans_buf.c b/fs/xfs/xfs_trans_buf.c index 5ffd544434eb..fb586360d1c9 100644 --- a/fs/xfs/xfs_trans_buf.c +++ b/fs/xfs/xfs_trans_buf.c @@ -46,6 +46,65 @@ STATIC xfs_buf_t *xfs_trans_buf_item_match(xfs_trans_t *, xfs_buftarg_t *, STATIC xfs_buf_t *xfs_trans_buf_item_match_all(xfs_trans_t *, xfs_buftarg_t *, xfs_daddr_t, int); +/* + * Add the locked buffer to the transaction. + * + * The buffer must be locked, and it cannot be associated with any + * transaction. + * + * If the buffer does not yet have a buf log item associated with it, + * then allocate one for it. Then add the buf item to the transaction. + */ +STATIC void +_xfs_trans_bjoin( + struct xfs_trans *tp, + struct xfs_buf *bp, + int reset_recur) +{ + struct xfs_buf_log_item *bip; + + ASSERT(XFS_BUF_ISBUSY(bp)); + ASSERT(XFS_BUF_FSPRIVATE2(bp, void *) == NULL); + + /* + * The xfs_buf_log_item pointer is stored in b_fsprivate. If + * it doesn't have one yet, then allocate one and initialize it. + * The checks to see if one is there are in xfs_buf_item_init(). + */ + xfs_buf_item_init(bp, tp->t_mountp); + bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t *); + ASSERT(!(bip->bli_flags & XFS_BLI_STALE)); + ASSERT(!(bip->bli_format.blf_flags & XFS_BLI_CANCEL)); + ASSERT(!(bip->bli_flags & XFS_BLI_LOGGED)); + if (reset_recur) + bip->bli_recur = 0; + + /* + * Take a reference for this transaction on the buf item. + */ + atomic_inc(&bip->bli_refcount); + + /* + * Get a log_item_desc to point at the new item. + */ + (void) xfs_trans_add_item(tp, (xfs_log_item_t *)bip); + + /* + * Initialize b_fsprivate2 so we can find it with incore_match() + * in xfs_trans_get_buf() and friends above. + */ + XFS_BUF_SET_FSPRIVATE2(bp, tp); + +} + +void +xfs_trans_bjoin( + struct xfs_trans *tp, + struct xfs_buf *bp) +{ + _xfs_trans_bjoin(tp, bp, 0); + trace_xfs_trans_bjoin(bp->b_fspriv); +} /* * Get and lock the buffer for the caller if it is not already @@ -132,40 +191,8 @@ xfs_trans_get_buf(xfs_trans_t *tp, ASSERT(!XFS_BUF_GETERROR(bp)); - /* - * The xfs_buf_log_item pointer is stored in b_fsprivate. If - * it doesn't have one yet, then allocate one and initialize it. - * The checks to see if one is there are in xfs_buf_item_init(). - */ - xfs_buf_item_init(bp, tp->t_mountp); - - /* - * Set the recursion count for the buffer within this transaction - * to 0. - */ - bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t*); - ASSERT(!(bip->bli_flags & XFS_BLI_STALE)); - ASSERT(!(bip->bli_format.blf_flags & XFS_BLI_CANCEL)); - ASSERT(!(bip->bli_flags & XFS_BLI_LOGGED)); - bip->bli_recur = 0; - - /* - * Take a reference for this transaction on the buf item. - */ - atomic_inc(&bip->bli_refcount); - - /* - * Get a log_item_desc to point at the new item. - */ - (void) xfs_trans_add_item(tp, (xfs_log_item_t*)bip); - - /* - * Initialize b_fsprivate2 so we can find it with incore_match() - * above. - */ - XFS_BUF_SET_FSPRIVATE2(bp, tp); - - trace_xfs_trans_get_buf(bip); + _xfs_trans_bjoin(tp, bp, 1); + trace_xfs_trans_get_buf(bp->b_fspriv); return (bp); } @@ -210,44 +237,11 @@ xfs_trans_getsb(xfs_trans_t *tp, } bp = xfs_getsb(mp, flags); - if (bp == NULL) { + if (bp == NULL) return NULL; - } - - /* - * The xfs_buf_log_item pointer is stored in b_fsprivate. If - * it doesn't have one yet, then allocate one and initialize it. - * The checks to see if one is there are in xfs_buf_item_init(). - */ - xfs_buf_item_init(bp, mp); - - /* - * Set the recursion count for the buffer within this transaction - * to 0. - */ - bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t*); - ASSERT(!(bip->bli_flags & XFS_BLI_STALE)); - ASSERT(!(bip->bli_format.blf_flags & XFS_BLI_CANCEL)); - ASSERT(!(bip->bli_flags & XFS_BLI_LOGGED)); - bip->bli_recur = 0; - - /* - * Take a reference for this transaction on the buf item. - */ - atomic_inc(&bip->bli_refcount); - - /* - * Get a log_item_desc to point at the new item. - */ - (void) xfs_trans_add_item(tp, (xfs_log_item_t*)bip); - - /* - * Initialize b_fsprivate2 so we can find it with incore_match() - * above. - */ - XFS_BUF_SET_FSPRIVATE2(bp, tp); - trace_xfs_trans_getsb(bip); + _xfs_trans_bjoin(tp, bp, 1); + trace_xfs_trans_getsb(bp->b_fspriv); return (bp); } @@ -425,40 +419,9 @@ xfs_trans_read_buf( if (XFS_FORCED_SHUTDOWN(mp)) goto shutdown_abort; - /* - * The xfs_buf_log_item pointer is stored in b_fsprivate. If - * it doesn't have one yet, then allocate one and initialize it. - * The checks to see if one is there are in xfs_buf_item_init(). - */ - xfs_buf_item_init(bp, tp->t_mountp); + _xfs_trans_bjoin(tp, bp, 1); + trace_xfs_trans_read_buf(bp->b_fspriv); - /* - * Set the recursion count for the buffer within this transaction - * to 0. - */ - bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t*); - ASSERT(!(bip->bli_flags & XFS_BLI_STALE)); - ASSERT(!(bip->bli_format.blf_flags & XFS_BLI_CANCEL)); - ASSERT(!(bip->bli_flags & XFS_BLI_LOGGED)); - bip->bli_recur = 0; - - /* - * Take a reference for this transaction on the buf item. - */ - atomic_inc(&bip->bli_refcount); - - /* - * Get a log_item_desc to point at the new item. - */ - (void) xfs_trans_add_item(tp, (xfs_log_item_t*)bip); - - /* - * Initialize b_fsprivate2 so we can find it with incore_match() - * above. - */ - XFS_BUF_SET_FSPRIVATE2(bp, tp); - - trace_xfs_trans_read_buf(bip); *bpp = bp; return 0; @@ -623,53 +586,6 @@ xfs_trans_brelse(xfs_trans_t *tp, } /* - * Add the locked buffer to the transaction. - * The buffer must be locked, and it cannot be associated with any - * transaction. - * - * If the buffer does not yet have a buf log item associated with it, - * then allocate one for it. Then add the buf item to the transaction. - */ -void -xfs_trans_bjoin(xfs_trans_t *tp, - xfs_buf_t *bp) -{ - xfs_buf_log_item_t *bip; - - ASSERT(XFS_BUF_ISBUSY(bp)); - ASSERT(XFS_BUF_FSPRIVATE2(bp, void *) == NULL); - - /* - * The xfs_buf_log_item pointer is stored in b_fsprivate. If - * it doesn't have one yet, then allocate one and initialize it. - * The checks to see if one is there are in xfs_buf_item_init(). - */ - xfs_buf_item_init(bp, tp->t_mountp); - bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t *); - ASSERT(!(bip->bli_flags & XFS_BLI_STALE)); - ASSERT(!(bip->bli_format.blf_flags & XFS_BLI_CANCEL)); - ASSERT(!(bip->bli_flags & XFS_BLI_LOGGED)); - - /* - * Take a reference for this transaction on the buf item. - */ - atomic_inc(&bip->bli_refcount); - - /* - * Get a log_item_desc to point at the new item. - */ - (void) xfs_trans_add_item(tp, (xfs_log_item_t *)bip); - - /* - * Initialize b_fsprivate2 so we can find it with incore_match() - * in xfs_trans_get_buf() and friends above. - */ - XFS_BUF_SET_FSPRIVATE2(bp, tp); - - trace_xfs_trans_bjoin(bip); -} - -/* * Mark the buffer as not needing to be unlocked when the buf item's * IOP_UNLOCK() routine is called. The buffer must already be locked * and associated with the given transaction. diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index ddd2c5d1b854..9d376be0ea38 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -584,113 +584,6 @@ xfs_readlink( } /* - * xfs_fsync - * - * This is called to sync the inode and its data out to disk. We need to hold - * the I/O lock while flushing the data, and the inode lock while flushing the - * inode. The inode lock CANNOT be held while flushing the data, so acquire - * after we're done with that. - */ -int -xfs_fsync( - xfs_inode_t *ip) -{ - xfs_trans_t *tp; - int error = 0; - int log_flushed = 0; - - xfs_itrace_entry(ip); - - if (XFS_FORCED_SHUTDOWN(ip->i_mount)) - return XFS_ERROR(EIO); - - /* - * We always need to make sure that the required inode state is safe on - * disk. The inode might be clean but we still might need to force the - * log because of committed transactions that haven't hit the disk yet. - * Likewise, there could be unflushed non-transactional changes to the - * inode core that have to go to disk and this requires us to issue - * a synchronous transaction to capture these changes correctly. - * - * This code relies on the assumption that if the update_* fields - * of the inode are clear and the inode is unpinned then it is clean - * and no action is required. - */ - xfs_ilock(ip, XFS_ILOCK_SHARED); - - if (!ip->i_update_core) { - /* - * Timestamps/size haven't changed since last inode flush or - * inode transaction commit. That means either nothing got - * written or a transaction committed which caught the updates. - * If the latter happened and the transaction hasn't hit the - * disk yet, the inode will be still be pinned. If it is, - * force the log. - */ - xfs_iunlock(ip, XFS_ILOCK_SHARED); - if (xfs_ipincount(ip)) { - if (ip->i_itemp->ili_last_lsn) { - error = _xfs_log_force_lsn(ip->i_mount, - ip->i_itemp->ili_last_lsn, - XFS_LOG_SYNC, &log_flushed); - } else { - error = _xfs_log_force(ip->i_mount, - XFS_LOG_SYNC, &log_flushed); - } - } - } else { - /* - * Kick off a transaction to log the inode core to get the - * updates. The sync transaction will also force the log. - */ - xfs_iunlock(ip, XFS_ILOCK_SHARED); - tp = xfs_trans_alloc(ip->i_mount, XFS_TRANS_FSYNC_TS); - error = xfs_trans_reserve(tp, 0, - XFS_FSYNC_TS_LOG_RES(ip->i_mount), 0, 0, 0); - if (error) { - xfs_trans_cancel(tp, 0); - return error; - } - xfs_ilock(ip, XFS_ILOCK_EXCL); - - /* - * Note - it's possible that we might have pushed ourselves out - * of the way during trans_reserve which would flush the inode. - * But there's no guarantee that the inode buffer has actually - * gone out yet (it's delwri). Plus the buffer could be pinned - * anyway if it's part of an inode in another recent - * transaction. So we play it safe and fire off the - * transaction anyway. - */ - xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); - xfs_trans_ihold(tp, ip); - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - xfs_trans_set_sync(tp); - error = _xfs_trans_commit(tp, 0, &log_flushed); - - xfs_iunlock(ip, XFS_ILOCK_EXCL); - } - - if (ip->i_mount->m_flags & XFS_MOUNT_BARRIER) { - /* - * If the log write didn't issue an ordered tag we need - * to flush the disk cache for the data device now. - */ - if (!log_flushed) - xfs_blkdev_issue_flush(ip->i_mount->m_ddev_targp); - - /* - * If this inode is on the RT dev we need to flush that - * cache as well. - */ - if (XFS_IS_REALTIME_INODE(ip)) - xfs_blkdev_issue_flush(ip->i_mount->m_rtdev_targp); - } - - return error; -} - -/* * Flags for xfs_free_eofblocks */ #define XFS_FREE_EOF_TRYLOCK (1<<0) diff --git a/fs/xfs/xfs_vnodeops.h b/fs/xfs/xfs_vnodeops.h index 774f40729ca1..d8dfa8d0dadd 100644 --- a/fs/xfs/xfs_vnodeops.h +++ b/fs/xfs/xfs_vnodeops.h @@ -21,7 +21,6 @@ int xfs_setattr(struct xfs_inode *ip, struct iattr *vap, int flags); #define XFS_ATTR_NOACL 0x08 /* Don't call xfs_acl_chmod */ int xfs_readlink(struct xfs_inode *ip, char *link); -int xfs_fsync(struct xfs_inode *ip); int xfs_release(struct xfs_inode *ip); int xfs_inactive(struct xfs_inode *ip); int xfs_lookup(struct xfs_inode *dp, struct xfs_name *name, @@ -50,18 +49,6 @@ int xfs_attr_set(struct xfs_inode *dp, const unsigned char *name, int xfs_attr_remove(struct xfs_inode *dp, const unsigned char *name, int flags); int xfs_attr_list(struct xfs_inode *dp, char *buffer, int bufsize, int flags, struct attrlist_cursor_kern *cursor); -ssize_t xfs_read(struct xfs_inode *ip, struct kiocb *iocb, - const struct iovec *iovp, unsigned int segs, - loff_t *offset, int ioflags); -ssize_t xfs_splice_read(struct xfs_inode *ip, struct file *infilp, - loff_t *ppos, struct pipe_inode_info *pipe, size_t count, - int flags, int ioflags); -ssize_t xfs_splice_write(struct xfs_inode *ip, - struct pipe_inode_info *pipe, struct file *outfilp, - loff_t *ppos, size_t count, int flags, int ioflags); -ssize_t xfs_write(struct xfs_inode *xip, struct kiocb *iocb, - const struct iovec *iovp, unsigned int nsegs, - loff_t *offset, int ioflags); int xfs_bmap(struct xfs_inode *ip, xfs_off_t offset, ssize_t count, int flags, struct xfs_iomap *iomapp, int *niomaps); void xfs_tosspages(struct xfs_inode *inode, xfs_off_t first, @@ -72,4 +59,6 @@ int xfs_flush_pages(struct xfs_inode *ip, xfs_off_t first, xfs_off_t last, uint64_t flags, int fiopt); int xfs_wait_on_pages(struct xfs_inode *ip, xfs_off_t first, xfs_off_t last); +int xfs_zero_eof(struct xfs_inode *, xfs_off_t, xfs_fsize_t); + #endif /* _XFS_VNODEOPS_H */ diff --git a/include/asm-generic/gpio.h b/include/asm-generic/gpio.h index 485eeb6c4ef3..979c6a57f2f1 100644 --- a/include/asm-generic/gpio.h +++ b/include/asm-generic/gpio.h @@ -136,6 +136,32 @@ extern int __gpio_cansleep(unsigned gpio); extern int __gpio_to_irq(unsigned gpio); +#define GPIOF_DIR_OUT (0 << 0) +#define GPIOF_DIR_IN (1 << 0) + +#define GPIOF_INIT_LOW (0 << 1) +#define GPIOF_INIT_HIGH (1 << 1) + +#define GPIOF_IN (GPIOF_DIR_IN) +#define GPIOF_OUT_INIT_LOW (GPIOF_DIR_OUT | GPIOF_INIT_LOW) +#define GPIOF_OUT_INIT_HIGH (GPIOF_DIR_OUT | GPIOF_INIT_HIGH) + +/** + * struct gpio - a structure describing a GPIO with configuration + * @gpio: the GPIO number + * @flags: GPIO configuration as specified by GPIOF_* + * @label: a literal description string of this GPIO + */ +struct gpio { + unsigned gpio; + unsigned long flags; + const char *label; +}; + +extern int gpio_request_one(unsigned gpio, unsigned long flags, const char *label); +extern int gpio_request_array(struct gpio *array, size_t num); +extern void gpio_free_array(struct gpio *array, size_t num); + #ifdef CONFIG_GPIO_SYSFS /* diff --git a/include/linux/binfmts.h b/include/linux/binfmts.h index 89c6249fc561..c809e286d213 100644 --- a/include/linux/binfmts.h +++ b/include/linux/binfmts.h @@ -74,6 +74,7 @@ struct coredump_params { struct pt_regs *regs; struct file *file; unsigned long limit; + unsigned long mm_flags; }; /* diff --git a/include/linux/bitops.h b/include/linux/bitops.h index 25b8b2f33ae9..b79389879238 100644 --- a/include/linux/bitops.h +++ b/include/linux/bitops.h @@ -16,11 +16,13 @@ */ #include <asm/bitops.h> -#define for_each_bit(bit, addr, size) \ +#define for_each_set_bit(bit, addr, size) \ for ((bit) = find_first_bit((addr), (size)); \ (bit) < (size); \ (bit) = find_next_bit((addr), (size), (bit) + 1)) +/* Temporary */ +#define for_each_bit(bit, addr, size) for_each_set_bit(bit, addr, size) static __inline__ int get_bitmask_order(unsigned int count) { diff --git a/include/linux/btree-128.h b/include/linux/btree-128.h new file mode 100644 index 000000000000..0b3414c4c928 --- /dev/null +++ b/include/linux/btree-128.h @@ -0,0 +1,109 @@ +extern struct btree_geo btree_geo128; + +struct btree_head128 { struct btree_head h; }; + +static inline void btree_init_mempool128(struct btree_head128 *head, + mempool_t *mempool) +{ + btree_init_mempool(&head->h, mempool); +} + +static inline int btree_init128(struct btree_head128 *head) +{ + return btree_init(&head->h); +} + +static inline void btree_destroy128(struct btree_head128 *head) +{ + btree_destroy(&head->h); +} + +static inline void *btree_lookup128(struct btree_head128 *head, u64 k1, u64 k2) +{ + u64 key[2] = {k1, k2}; + return btree_lookup(&head->h, &btree_geo128, (unsigned long *)&key); +} + +static inline void *btree_get_prev128(struct btree_head128 *head, + u64 *k1, u64 *k2) +{ + u64 key[2] = {*k1, *k2}; + void *val; + + val = btree_get_prev(&head->h, &btree_geo128, + (unsigned long *)&key); + *k1 = key[0]; + *k2 = key[1]; + return val; +} + +static inline int btree_insert128(struct btree_head128 *head, u64 k1, u64 k2, + void *val, gfp_t gfp) +{ + u64 key[2] = {k1, k2}; + return btree_insert(&head->h, &btree_geo128, + (unsigned long *)&key, val, gfp); +} + +static inline int btree_update128(struct btree_head128 *head, u64 k1, u64 k2, + void *val) +{ + u64 key[2] = {k1, k2}; + return btree_update(&head->h, &btree_geo128, + (unsigned long *)&key, val); +} + +static inline void *btree_remove128(struct btree_head128 *head, u64 k1, u64 k2) +{ + u64 key[2] = {k1, k2}; + return btree_remove(&head->h, &btree_geo128, (unsigned long *)&key); +} + +static inline void *btree_last128(struct btree_head128 *head, u64 *k1, u64 *k2) +{ + u64 key[2]; + void *val; + + val = btree_last(&head->h, &btree_geo128, (unsigned long *)&key[0]); + if (val) { + *k1 = key[0]; + *k2 = key[1]; + } + + return val; +} + +static inline int btree_merge128(struct btree_head128 *target, + struct btree_head128 *victim, + gfp_t gfp) +{ + return btree_merge(&target->h, &victim->h, &btree_geo128, gfp); +} + +void visitor128(void *elem, unsigned long opaque, unsigned long *__key, + size_t index, void *__func); + +typedef void (*visitor128_t)(void *elem, unsigned long opaque, + u64 key1, u64 key2, size_t index); + +static inline size_t btree_visitor128(struct btree_head128 *head, + unsigned long opaque, + visitor128_t func2) +{ + return btree_visitor(&head->h, &btree_geo128, opaque, + visitor128, func2); +} + +static inline size_t btree_grim_visitor128(struct btree_head128 *head, + unsigned long opaque, + visitor128_t func2) +{ + return btree_grim_visitor(&head->h, &btree_geo128, opaque, + visitor128, func2); +} + +#define btree_for_each_safe128(head, k1, k2, val) \ + for (val = btree_last128(head, &k1, &k2); \ + val; \ + val = btree_get_prev128(head, &k1, &k2)) + diff --git a/include/linux/btree-type.h b/include/linux/btree-type.h new file mode 100644 index 000000000000..9a1147ef8563 --- /dev/null +++ b/include/linux/btree-type.h @@ -0,0 +1,147 @@ +#define __BTREE_TP(pfx, type, sfx) pfx ## type ## sfx +#define _BTREE_TP(pfx, type, sfx) __BTREE_TP(pfx, type, sfx) +#define BTREE_TP(pfx) _BTREE_TP(pfx, BTREE_TYPE_SUFFIX,) +#define BTREE_FN(name) BTREE_TP(btree_ ## name) +#define BTREE_TYPE_HEAD BTREE_TP(struct btree_head) +#define VISITOR_FN BTREE_TP(visitor) +#define VISITOR_FN_T _BTREE_TP(visitor, BTREE_TYPE_SUFFIX, _t) + +BTREE_TYPE_HEAD { + struct btree_head h; +}; + +static inline void BTREE_FN(init_mempool)(BTREE_TYPE_HEAD *head, + mempool_t *mempool) +{ + btree_init_mempool(&head->h, mempool); +} + +static inline int BTREE_FN(init)(BTREE_TYPE_HEAD *head) +{ + return btree_init(&head->h); +} + +static inline void BTREE_FN(destroy)(BTREE_TYPE_HEAD *head) +{ + btree_destroy(&head->h); +} + +static inline int BTREE_FN(merge)(BTREE_TYPE_HEAD *target, + BTREE_TYPE_HEAD *victim, + gfp_t gfp) +{ + return btree_merge(&target->h, &victim->h, BTREE_TYPE_GEO, gfp); +} + +#if (BITS_PER_LONG > BTREE_TYPE_BITS) +static inline void *BTREE_FN(lookup)(BTREE_TYPE_HEAD *head, BTREE_KEYTYPE key) +{ + unsigned long _key = key; + return btree_lookup(&head->h, BTREE_TYPE_GEO, &_key); +} + +static inline int BTREE_FN(insert)(BTREE_TYPE_HEAD *head, BTREE_KEYTYPE key, + void *val, gfp_t gfp) +{ + unsigned long _key = key; + return btree_insert(&head->h, BTREE_TYPE_GEO, &_key, val, gfp); +} + +static inline int BTREE_FN(update)(BTREE_TYPE_HEAD *head, BTREE_KEYTYPE key, + void *val) +{ + unsigned long _key = key; + return btree_update(&head->h, BTREE_TYPE_GEO, &_key, val); +} + +static inline void *BTREE_FN(remove)(BTREE_TYPE_HEAD *head, BTREE_KEYTYPE key) +{ + unsigned long _key = key; + return btree_remove(&head->h, BTREE_TYPE_GEO, &_key); +} + +static inline void *BTREE_FN(last)(BTREE_TYPE_HEAD *head, BTREE_KEYTYPE *key) +{ + unsigned long _key; + void *val = btree_last(&head->h, BTREE_TYPE_GEO, &_key); + if (val) + *key = _key; + return val; +} + +static inline void *BTREE_FN(get_prev)(BTREE_TYPE_HEAD *head, BTREE_KEYTYPE *key) +{ + unsigned long _key = *key; + void *val = btree_get_prev(&head->h, BTREE_TYPE_GEO, &_key); + if (val) + *key = _key; + return val; +} +#else +static inline void *BTREE_FN(lookup)(BTREE_TYPE_HEAD *head, BTREE_KEYTYPE key) +{ + return btree_lookup(&head->h, BTREE_TYPE_GEO, (unsigned long *)&key); +} + +static inline int BTREE_FN(insert)(BTREE_TYPE_HEAD *head, BTREE_KEYTYPE key, + void *val, gfp_t gfp) +{ + return btree_insert(&head->h, BTREE_TYPE_GEO, (unsigned long *)&key, + val, gfp); +} + +static inline int BTREE_FN(update)(BTREE_TYPE_HEAD *head, BTREE_KEYTYPE key, + void *val) +{ + return btree_update(&head->h, BTREE_TYPE_GEO, (unsigned long *)&key, val); +} + +static inline void *BTREE_FN(remove)(BTREE_TYPE_HEAD *head, BTREE_KEYTYPE key) +{ + return btree_remove(&head->h, BTREE_TYPE_GEO, (unsigned long *)&key); +} + +static inline void *BTREE_FN(last)(BTREE_TYPE_HEAD *head, BTREE_KEYTYPE *key) +{ + return btree_last(&head->h, BTREE_TYPE_GEO, (unsigned long *)key); +} + +static inline void *BTREE_FN(get_prev)(BTREE_TYPE_HEAD *head, BTREE_KEYTYPE *key) +{ + return btree_get_prev(&head->h, BTREE_TYPE_GEO, (unsigned long *)key); +} +#endif + +void VISITOR_FN(void *elem, unsigned long opaque, unsigned long *key, + size_t index, void *__func); + +typedef void (*VISITOR_FN_T)(void *elem, unsigned long opaque, + BTREE_KEYTYPE key, size_t index); + +static inline size_t BTREE_FN(visitor)(BTREE_TYPE_HEAD *head, + unsigned long opaque, + VISITOR_FN_T func2) +{ + return btree_visitor(&head->h, BTREE_TYPE_GEO, opaque, + visitorl, func2); +} + +static inline size_t BTREE_FN(grim_visitor)(BTREE_TYPE_HEAD *head, + unsigned long opaque, + VISITOR_FN_T func2) +{ + return btree_grim_visitor(&head->h, BTREE_TYPE_GEO, opaque, + visitorl, func2); +} + +#undef VISITOR_FN +#undef VISITOR_FN_T +#undef __BTREE_TP +#undef _BTREE_TP +#undef BTREE_TP +#undef BTREE_FN +#undef BTREE_TYPE_HEAD +#undef BTREE_TYPE_SUFFIX +#undef BTREE_TYPE_GEO +#undef BTREE_KEYTYPE +#undef BTREE_TYPE_BITS diff --git a/include/linux/btree.h b/include/linux/btree.h new file mode 100644 index 000000000000..65b5bb058324 --- /dev/null +++ b/include/linux/btree.h @@ -0,0 +1,243 @@ +#ifndef BTREE_H +#define BTREE_H + +#include <linux/kernel.h> +#include <linux/mempool.h> + +/** + * DOC: B+Tree basics + * + * A B+Tree is a data structure for looking up arbitrary (currently allowing + * unsigned long, u32, u64 and 2 * u64) keys into pointers. The data structure + * is described at http://en.wikipedia.org/wiki/B-tree, we currently do not + * use binary search to find the key on lookups. + * + * Each B+Tree consists of a head, that contains bookkeeping information and + * a variable number (starting with zero) nodes. Each node contains the keys + * and pointers to sub-nodes, or, for leaf nodes, the keys and values for the + * tree entries. + * + * Each node in this implementation has the following layout: + * [key1, key2, ..., keyN] [val1, val2, ..., valN] + * + * Each key here is an array of unsigned longs, geo->no_longs in total. The + * number of keys and values (N) is geo->no_pairs. + */ + +/** + * struct btree_head - btree head + * + * @node: the first node in the tree + * @mempool: mempool used for node allocations + * @height: current of the tree + */ +struct btree_head { + unsigned long *node; + mempool_t *mempool; + int height; +}; + +/* btree geometry */ +struct btree_geo; + +/** + * btree_alloc - allocate function for the mempool + * @gfp_mask: gfp mask for the allocation + * @pool_data: unused + */ +void *btree_alloc(gfp_t gfp_mask, void *pool_data); + +/** + * btree_free - free function for the mempool + * @element: the element to free + * @pool_data: unused + */ +void btree_free(void *element, void *pool_data); + +/** + * btree_init_mempool - initialise a btree with given mempool + * + * @head: the btree head to initialise + * @mempool: the mempool to use + * + * When this function is used, there is no need to destroy + * the mempool. + */ +void btree_init_mempool(struct btree_head *head, mempool_t *mempool); + +/** + * btree_init - initialise a btree + * + * @head: the btree head to initialise + * + * This function allocates the memory pool that the + * btree needs. Returns zero or a negative error code + * (-%ENOMEM) when memory allocation fails. + * + */ +int __must_check btree_init(struct btree_head *head); + +/** + * btree_destroy - destroy mempool + * + * @head: the btree head to destroy + * + * This function destroys the internal memory pool, use only + * when using btree_init(), not with btree_init_mempool(). + */ +void btree_destroy(struct btree_head *head); + +/** + * btree_lookup - look up a key in the btree + * + * @head: the btree to look in + * @geo: the btree geometry + * @key: the key to look up + * + * This function returns the value for the given key, or %NULL. + */ +void *btree_lookup(struct btree_head *head, struct btree_geo *geo, + unsigned long *key); + +/** + * btree_insert - insert an entry into the btree + * + * @head: the btree to add to + * @geo: the btree geometry + * @key: the key to add (must not already be present) + * @val: the value to add (must not be %NULL) + * @gfp: allocation flags for node allocations + * + * This function returns 0 if the item could be added, or an + * error code if it failed (may fail due to memory pressure). + */ +int __must_check btree_insert(struct btree_head *head, struct btree_geo *geo, + unsigned long *key, void *val, gfp_t gfp); +/** + * btree_update - update an entry in the btree + * + * @head: the btree to update + * @geo: the btree geometry + * @key: the key to update + * @val: the value to change it to (must not be %NULL) + * + * This function returns 0 if the update was successful, or + * -%ENOENT if the key could not be found. + */ +int btree_update(struct btree_head *head, struct btree_geo *geo, + unsigned long *key, void *val); +/** + * btree_remove - remove an entry from the btree + * + * @head: the btree to update + * @geo: the btree geometry + * @key: the key to remove + * + * This function returns the removed entry, or %NULL if the key + * could not be found. + */ +void *btree_remove(struct btree_head *head, struct btree_geo *geo, + unsigned long *key); + +/** + * btree_merge - merge two btrees + * + * @target: the tree that gets all the entries + * @victim: the tree that gets merged into @target + * @geo: the btree geometry + * @gfp: allocation flags + * + * The two trees @target and @victim may not contain the same keys, + * that is a bug and triggers a BUG(). This function returns zero + * if the trees were merged successfully, and may return a failure + * when memory allocation fails, in which case both trees might have + * been partially merged, i.e. some entries have been moved from + * @victim to @target. + */ +int btree_merge(struct btree_head *target, struct btree_head *victim, + struct btree_geo *geo, gfp_t gfp); + +/** + * btree_last - get last entry in btree + * + * @head: btree head + * @geo: btree geometry + * @key: last key + * + * Returns the last entry in the btree, and sets @key to the key + * of that entry; returns NULL if the tree is empty, in that case + * key is not changed. + */ +void *btree_last(struct btree_head *head, struct btree_geo *geo, + unsigned long *key); + +/** + * btree_get_prev - get previous entry + * + * @head: btree head + * @geo: btree geometry + * @key: pointer to key + * + * The function returns the next item right before the value pointed to by + * @key, and updates @key with its key, or returns %NULL when there is no + * entry with a key smaller than the given key. + */ +void *btree_get_prev(struct btree_head *head, struct btree_geo *geo, + unsigned long *key); + + +/* internal use, use btree_visitor{l,32,64,128} */ +size_t btree_visitor(struct btree_head *head, struct btree_geo *geo, + unsigned long opaque, + void (*func)(void *elem, unsigned long opaque, + unsigned long *key, size_t index, + void *func2), + void *func2); + +/* internal use, use btree_grim_visitor{l,32,64,128} */ +size_t btree_grim_visitor(struct btree_head *head, struct btree_geo *geo, + unsigned long opaque, + void (*func)(void *elem, unsigned long opaque, + unsigned long *key, + size_t index, void *func2), + void *func2); + + +#include <linux/btree-128.h> + +extern struct btree_geo btree_geo32; +#define BTREE_TYPE_SUFFIX l +#define BTREE_TYPE_BITS BITS_PER_LONG +#define BTREE_TYPE_GEO &btree_geo32 +#define BTREE_KEYTYPE unsigned long +#include <linux/btree-type.h> + +#define btree_for_each_safel(head, key, val) \ + for (val = btree_lastl(head, &key); \ + val; \ + val = btree_get_prevl(head, &key)) + +#define BTREE_TYPE_SUFFIX 32 +#define BTREE_TYPE_BITS 32 +#define BTREE_TYPE_GEO &btree_geo32 +#define BTREE_KEYTYPE u32 +#include <linux/btree-type.h> + +#define btree_for_each_safe32(head, key, val) \ + for (val = btree_last32(head, &key); \ + val; \ + val = btree_get_prev32(head, &key)) + +extern struct btree_geo btree_geo64; +#define BTREE_TYPE_SUFFIX 64 +#define BTREE_TYPE_BITS 64 +#define BTREE_TYPE_GEO &btree_geo64 +#define BTREE_KEYTYPE u64 +#include <linux/btree-type.h> + +#define btree_for_each_safe64(head, key, val) \ + for (val = btree_last64(head, &key); \ + val; \ + val = btree_get_prev64(head, &key)) + +#endif diff --git a/include/linux/coredump.h b/include/linux/coredump.h new file mode 100644 index 000000000000..b3c91d7cede4 --- /dev/null +++ b/include/linux/coredump.h @@ -0,0 +1,41 @@ +#ifndef _LINUX_COREDUMP_H +#define _LINUX_COREDUMP_H + +#include <linux/types.h> +#include <linux/mm.h> +#include <linux/fs.h> + +/* + * These are the only things you should do on a core-file: use only these + * functions to write out all the necessary info. + */ +static inline int dump_write(struct file *file, const void *addr, int nr) +{ + return file->f_op->write(file, addr, nr, &file->f_pos) == nr; +} + +static inline int dump_seek(struct file *file, loff_t off) +{ + if (file->f_op->llseek && file->f_op->llseek != no_llseek) { + if (file->f_op->llseek(file, off, SEEK_CUR) < 0) + return 0; + } else { + char *buf = (char *)get_zeroed_page(GFP_KERNEL); + + if (!buf) + return 0; + while (off > 0) { + unsigned long n = off; + + if (n > PAGE_SIZE) + n = PAGE_SIZE; + if (!dump_write(file, buf, n)) + return 0; + off -= n; + } + free_page((unsigned long)buf); + } + return 1; +} + +#endif /* _LINUX_COREDUMP_H */ diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h index dbcee7647d9a..bae6fe24d1f9 100644 --- a/include/linux/cpumask.h +++ b/include/linux/cpumask.h @@ -90,10 +90,10 @@ extern const struct cpumask *const cpu_active_mask; #define cpu_present(cpu) cpumask_test_cpu((cpu), cpu_present_mask) #define cpu_active(cpu) cpumask_test_cpu((cpu), cpu_active_mask) #else -#define num_online_cpus() 1 -#define num_possible_cpus() 1 -#define num_present_cpus() 1 -#define num_active_cpus() 1 +#define num_online_cpus() 1U +#define num_possible_cpus() 1U +#define num_present_cpus() 1U +#define num_active_cpus() 1U #define cpu_online(cpu) ((cpu) == 0) #define cpu_possible(cpu) ((cpu) == 0) #define cpu_present(cpu) ((cpu) == 0) diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h index d4c9c0b88adc..1381cd97b4ed 100644 --- a/include/linux/device-mapper.h +++ b/include/linux/device-mapper.h @@ -118,10 +118,9 @@ struct dm_dev { /* * Constructors should call these functions to ensure destination devices * are opened/closed correctly. - * FIXME: too many arguments. */ -int dm_get_device(struct dm_target *ti, const char *path, sector_t start, - sector_t len, fmode_t mode, struct dm_dev **result); +int dm_get_device(struct dm_target *ti, const char *path, fmode_t mode, + struct dm_dev **result); void dm_put_device(struct dm_target *ti, struct dm_dev *d); /* diff --git a/include/linux/dm-io.h b/include/linux/dm-io.h index b6bf17ee2f61..5c9186b93fff 100644 --- a/include/linux/dm-io.h +++ b/include/linux/dm-io.h @@ -37,14 +37,14 @@ enum dm_io_mem_type { struct dm_io_memory { enum dm_io_mem_type type; + unsigned offset; + union { struct page_list *pl; struct bio_vec *bvec; void *vma; void *addr; } ptr; - - unsigned offset; }; struct dm_io_notify { diff --git a/include/linux/dm-ioctl.h b/include/linux/dm-ioctl.h index aa95508d2f95..2c445e113790 100644 --- a/include/linux/dm-ioctl.h +++ b/include/linux/dm-ioctl.h @@ -266,9 +266,9 @@ enum { #define DM_DEV_SET_GEOMETRY _IOWR(DM_IOCTL, DM_DEV_SET_GEOMETRY_CMD, struct dm_ioctl) #define DM_VERSION_MAJOR 4 -#define DM_VERSION_MINOR 16 +#define DM_VERSION_MINOR 17 #define DM_VERSION_PATCHLEVEL 0 -#define DM_VERSION_EXTRA "-ioctl (2009-11-05)" +#define DM_VERSION_EXTRA "-ioctl (2010-03-05)" /* Status bits */ #define DM_READONLY_FLAG (1 << 0) /* In/Out */ @@ -316,4 +316,9 @@ enum { */ #define DM_QUERY_INACTIVE_TABLE_FLAG (1 << 12) /* In */ +/* + * If set, a uevent was generated for which the caller may need to wait. + */ +#define DM_UEVENT_GENERATED_FLAG (1 << 13) /* Out */ + #endif /* _LINUX_DM_IOCTL_H */ diff --git a/include/linux/elf.h b/include/linux/elf.h index ad990c5f63f6..597858418051 100644 --- a/include/linux/elf.h +++ b/include/linux/elf.h @@ -50,6 +50,28 @@ typedef __s64 Elf64_Sxword; #define PT_GNU_STACK (PT_LOOS + 0x474e551) +/* + * Extended Numbering + * + * If the real number of program header table entries is larger than + * or equal to PN_XNUM(0xffff), it is set to sh_info field of the + * section header at index 0, and PN_XNUM is set to e_phnum + * field. Otherwise, the section header at index 0 is zero + * initialized, if it exists. + * + * Specifications are available in: + * + * - Sun microsystems: Linker and Libraries. + * Part No: 817-1984-17, September 2008. + * URL: http://docs.sun.com/app/docs/doc/817-1984 + * + * - System V ABI AMD64 Architecture Processor Supplement + * Draft Version 0.99., + * May 11, 2009. + * URL: http://www.x86-64.org/ + */ +#define PN_XNUM 0xffff + /* These constants define the different elf file types */ #define ET_NONE 0 #define ET_REL 1 @@ -286,7 +308,7 @@ typedef struct elf64_phdr { #define SHN_COMMON 0xfff2 #define SHN_HIRESERVE 0xffff -typedef struct { +typedef struct elf32_shdr { Elf32_Word sh_name; Elf32_Word sh_type; Elf32_Word sh_flags; @@ -394,16 +416,20 @@ typedef struct elf64_note { extern Elf32_Dyn _DYNAMIC []; #define elfhdr elf32_hdr #define elf_phdr elf32_phdr +#define elf_shdr elf32_shdr #define elf_note elf32_note #define elf_addr_t Elf32_Off +#define Elf_Half Elf32_Half #else extern Elf64_Dyn _DYNAMIC []; #define elfhdr elf64_hdr #define elf_phdr elf64_phdr +#define elf_shdr elf64_shdr #define elf_note elf64_note #define elf_addr_t Elf64_Off +#define Elf_Half Elf64_Half #endif diff --git a/include/linux/elfcore.h b/include/linux/elfcore.h index 00d6a68d0421..e687bc3ba4da 100644 --- a/include/linux/elfcore.h +++ b/include/linux/elfcore.h @@ -8,6 +8,8 @@ #include <linux/user.h> #endif #include <linux/ptrace.h> +#include <linux/elf.h> +#include <linux/fs.h> struct elf_siginfo { @@ -150,5 +152,20 @@ static inline int elf_core_copy_task_xfpregs(struct task_struct *t, elf_fpxregse #endif /* __KERNEL__ */ +/* + * These functions parameterize elf_core_dump in fs/binfmt_elf.c to write out + * extra segments containing the gate DSO contents. Dumping its + * contents makes post-mortem fully interpretable later without matching up + * the same kernel and hardware config to see what PC values meant. + * Dumping its extra ELF program headers includes all the other information + * a debugger needs to easily find how the gate DSO was being used. + */ +extern Elf_Half elf_core_extra_phdrs(void); +extern int +elf_core_write_extra_phdrs(struct file *file, loff_t offset, size_t *size, + unsigned long limit); +extern int +elf_core_write_extra_data(struct file *file, size_t *size, unsigned long limit); +extern size_t elf_core_extra_data_size(void); #endif /* _LINUX_ELFCORE_H */ diff --git a/include/linux/exportfs.h b/include/linux/exportfs.h index dc12f416a49f..a9cd507f8cd2 100644 --- a/include/linux/exportfs.h +++ b/include/linux/exportfs.h @@ -96,6 +96,7 @@ struct fid { * @fh_to_parent: find the implied object's parent and get a dentry for it * @get_name: find the name for a given inode in a given directory * @get_parent: find the parent of a given directory + * @commit_metadata: commit metadata changes to stable storage * * See Documentation/filesystems/nfs/Exporting for details on how to use * this interface correctly. @@ -137,6 +138,9 @@ struct fid { * is also a directory. In the event that it cannot be found, or storage * space cannot be allocated, a %ERR_PTR should be returned. * + * commit_metadata: + * @commit_metadata should commit metadata changes to stable storage. + * * Locking rules: * get_parent is called with child->d_inode->i_mutex down * get_name is not (which is possibly inconsistent) @@ -152,6 +156,7 @@ struct export_operations { int (*get_name)(struct dentry *parent, char *name, struct dentry *child); struct dentry * (*get_parent)(struct dentry *child); + int (*commit_metadata)(struct inode *inode); }; extern int exportfs_encode_fh(struct dentry *dentry, struct fid *fid, diff --git a/include/linux/firmware-map.h b/include/linux/firmware-map.h index 875451f1373a..c6dcc1dfe781 100644 --- a/include/linux/firmware-map.h +++ b/include/linux/firmware-map.h @@ -24,17 +24,17 @@ */ #ifdef CONFIG_FIRMWARE_MEMMAP -int firmware_map_add(u64 start, u64 end, const char *type); int firmware_map_add_early(u64 start, u64 end, const char *type); +int firmware_map_add_hotplug(u64 start, u64 end, const char *type); #else /* CONFIG_FIRMWARE_MEMMAP */ -static inline int firmware_map_add(u64 start, u64 end, const char *type) +static inline int firmware_map_add_early(u64 start, u64 end, const char *type) { return 0; } -static inline int firmware_map_add_early(u64 start, u64 end, const char *type) +static inline int firmware_map_add_hotplug(u64 start, u64 end, const char *type) { return 0; } diff --git a/include/linux/fs.h b/include/linux/fs.h index 45689621a851..10b8dedcd18b 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -60,24 +60,24 @@ struct inodes_stat_t { */ /* file is open for reading */ -#define FMODE_READ ((__force fmode_t)1) +#define FMODE_READ ((__force fmode_t)0x1) /* file is open for writing */ -#define FMODE_WRITE ((__force fmode_t)2) +#define FMODE_WRITE ((__force fmode_t)0x2) /* file is seekable */ -#define FMODE_LSEEK ((__force fmode_t)4) +#define FMODE_LSEEK ((__force fmode_t)0x4) /* file can be accessed using pread */ -#define FMODE_PREAD ((__force fmode_t)8) +#define FMODE_PREAD ((__force fmode_t)0x8) /* file can be accessed using pwrite */ -#define FMODE_PWRITE ((__force fmode_t)16) +#define FMODE_PWRITE ((__force fmode_t)0x10) /* File is opened for execution with sys_execve / sys_uselib */ -#define FMODE_EXEC ((__force fmode_t)32) +#define FMODE_EXEC ((__force fmode_t)0x20) /* File is opened with O_NDELAY (only set for block devices) */ -#define FMODE_NDELAY ((__force fmode_t)64) +#define FMODE_NDELAY ((__force fmode_t)0x40) /* File is opened with O_EXCL (only set for block devices) */ -#define FMODE_EXCL ((__force fmode_t)128) +#define FMODE_EXCL ((__force fmode_t)0x80) /* File is opened using open(.., 3, ..) and is writeable only for ioctls (specialy hack for floppy.c) */ -#define FMODE_WRITE_IOCTL ((__force fmode_t)256) +#define FMODE_WRITE_IOCTL ((__force fmode_t)0x100) /* * Don't update ctime and mtime. @@ -85,7 +85,10 @@ struct inodes_stat_t { * Currently a special hack for the XFS open_by_handle ioctl, but we'll * hopefully graduate it to a proper O_CMTIME flag supported by open(2) soon. */ -#define FMODE_NOCMTIME ((__force fmode_t)2048) +#define FMODE_NOCMTIME ((__force fmode_t)0x800) + +/* Expect random access pattern */ +#define FMODE_RANDOM ((__force fmode_t)0x1000) /* * The below are the various read and write types that we support. Some of diff --git a/include/linux/gfp.h b/include/linux/gfp.h index 557bdad320b6..4c6d41333f98 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -30,7 +30,8 @@ struct vm_area_struct; * _might_ fail. This depends upon the particular VM implementation. * * __GFP_NOFAIL: The VM implementation _must_ retry infinitely: the caller - * cannot handle allocation failures. + * cannot handle allocation failures. This modifier is deprecated and no new + * users should be added. * * __GFP_NORETRY: The VM implementation must not retry indefinitely. * @@ -83,6 +84,7 @@ struct vm_area_struct; #define GFP_HIGHUSER_MOVABLE (__GFP_WAIT | __GFP_IO | __GFP_FS | \ __GFP_HARDWALL | __GFP_HIGHMEM | \ __GFP_MOVABLE) +#define GFP_IOFS (__GFP_IO | __GFP_FS) #ifdef CONFIG_NUMA #define GFP_THISNODE (__GFP_THISNODE | __GFP_NOWARN | __GFP_NORETRY) @@ -325,7 +327,7 @@ void free_pages_exact(void *virt, size_t size); extern void __free_pages(struct page *page, unsigned int order); extern void free_pages(unsigned long addr, unsigned int order); -extern void free_hot_page(struct page *page); +extern void free_hot_cold_page(struct page *page, int cold); #define __free_page(page) __free_pages((page), 0) #define free_page(addr) free_pages((addr),0) @@ -337,9 +339,7 @@ void drain_local_pages(void *dummy); extern gfp_t gfp_allowed_mask; -static inline void set_gfp_allowed_mask(gfp_t mask) -{ - gfp_allowed_mask = mask; -} +extern void set_gfp_allowed_mask(gfp_t mask); +extern gfp_t clear_gfp_allowed_mask(gfp_t mask); #endif /* __LINUX_GFP_H */ diff --git a/include/linux/i2c/pca953x.h b/include/linux/i2c/pca953x.h index 81736d6a8db7..d5c5a60c8a0b 100644 --- a/include/linux/i2c/pca953x.h +++ b/include/linux/i2c/pca953x.h @@ -1,3 +1,9 @@ +#ifndef _LINUX_PCA953X_H +#define _LINUX_PCA953X_H + +#include <linux/types.h> +#include <linux/i2c.h> + /* platform data for the PCA9539 16-bit I/O expander driver */ struct pca953x_platform_data { @@ -7,6 +13,9 @@ struct pca953x_platform_data { /* initial polarity inversion setting */ uint16_t invert; + /* interrupt base */ + int irq_base; + void *context; /* param to setup/teardown */ int (*setup)(struct i2c_client *client, @@ -17,3 +26,5 @@ struct pca953x_platform_data { void *context); char **names; }; + +#endif /* _LINUX_PCA953X_H */ diff --git a/include/linux/list.h b/include/linux/list.h index 5d9c6558e8ab..8392884a2977 100644 --- a/include/linux/list.h +++ b/include/linux/list.h @@ -498,7 +498,7 @@ static inline void list_splice_tail_init(struct list_head *list, pos = n, n = list_entry(n->member.next, typeof(*n), member)) /** - * list_for_each_entry_safe_continue + * list_for_each_entry_safe_continue - continue list iteration safe against removal * @pos: the type * to use as a loop cursor. * @n: another type * to use as temporary storage * @head: the head for your list. @@ -514,7 +514,7 @@ static inline void list_splice_tail_init(struct list_head *list, pos = n, n = list_entry(n->member.next, typeof(*n), member)) /** - * list_for_each_entry_safe_from + * list_for_each_entry_safe_from - iterate over list from current point safe against removal * @pos: the type * to use as a loop cursor. * @n: another type * to use as temporary storage * @head: the head for your list. @@ -529,7 +529,7 @@ static inline void list_splice_tail_init(struct list_head *list, pos = n, n = list_entry(n->member.next, typeof(*n), member)) /** - * list_for_each_entry_safe_reverse + * list_for_each_entry_safe_reverse - iterate backwards over list safe against removal * @pos: the type * to use as a loop cursor. * @n: another type * to use as temporary storage * @head: the head for your list. diff --git a/include/linux/mfd/mc13783.h b/include/linux/mfd/mc13783.h index 94cb51a64037..8895d9d8879c 100644 --- a/include/linux/mfd/mc13783.h +++ b/include/linux/mfd/mc13783.h @@ -26,10 +26,30 @@ int mc13783_irq_request(struct mc13783 *mc13783, int irq, int mc13783_irq_request_nounmask(struct mc13783 *mc13783, int irq, irq_handler_t handler, const char *name, void *dev); int mc13783_irq_free(struct mc13783 *mc13783, int irq, void *dev); -int mc13783_ackirq(struct mc13783 *mc13783, int irq); -int mc13783_mask(struct mc13783 *mc13783, int irq); -int mc13783_unmask(struct mc13783 *mc13783, int irq); +int mc13783_irq_mask(struct mc13783 *mc13783, int irq); +int mc13783_irq_unmask(struct mc13783 *mc13783, int irq); +int mc13783_irq_status(struct mc13783 *mc13783, int irq, + int *enabled, int *pending); +int mc13783_irq_ack(struct mc13783 *mc13783, int irq); + +static inline int mc13783_mask(struct mc13783 *mc13783, int irq) __deprecated; +static inline int mc13783_mask(struct mc13783 *mc13783, int irq) +{ + return mc13783_irq_mask(mc13783, irq); +} + +static inline int mc13783_unmask(struct mc13783 *mc13783, int irq) __deprecated; +static inline int mc13783_unmask(struct mc13783 *mc13783, int irq) +{ + return mc13783_irq_unmask(mc13783, irq); +} + +static inline int mc13783_ackirq(struct mc13783 *mc13783, int irq) __deprecated; +static inline int mc13783_ackirq(struct mc13783 *mc13783, int irq) +{ + return mc13783_irq_ack(mc13783, irq); +} #define MC13783_ADC0 43 #define MC13783_ADC0_ADREFEN (1 << 10) diff --git a/include/linux/mm.h b/include/linux/mm.h index 90957f14195c..3899395a03de 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -870,6 +870,108 @@ extern int mprotect_fixup(struct vm_area_struct *vma, */ int __get_user_pages_fast(unsigned long start, int nr_pages, int write, struct page **pages); +/* + * per-process(per-mm_struct) statistics. + */ +#if defined(SPLIT_RSS_COUNTING) +/* + * The mm counters are not protected by its page_table_lock, + * so must be incremented atomically. + */ +static inline void set_mm_counter(struct mm_struct *mm, int member, long value) +{ + atomic_long_set(&mm->rss_stat.count[member], value); +} + +unsigned long get_mm_counter(struct mm_struct *mm, int member); + +static inline void add_mm_counter(struct mm_struct *mm, int member, long value) +{ + atomic_long_add(value, &mm->rss_stat.count[member]); +} + +static inline void inc_mm_counter(struct mm_struct *mm, int member) +{ + atomic_long_inc(&mm->rss_stat.count[member]); +} + +static inline void dec_mm_counter(struct mm_struct *mm, int member) +{ + atomic_long_dec(&mm->rss_stat.count[member]); +} + +#else /* !USE_SPLIT_PTLOCKS */ +/* + * The mm counters are protected by its page_table_lock, + * so can be incremented directly. + */ +static inline void set_mm_counter(struct mm_struct *mm, int member, long value) +{ + mm->rss_stat.count[member] = value; +} + +static inline unsigned long get_mm_counter(struct mm_struct *mm, int member) +{ + return mm->rss_stat.count[member]; +} + +static inline void add_mm_counter(struct mm_struct *mm, int member, long value) +{ + mm->rss_stat.count[member] += value; +} + +static inline void inc_mm_counter(struct mm_struct *mm, int member) +{ + mm->rss_stat.count[member]++; +} + +static inline void dec_mm_counter(struct mm_struct *mm, int member) +{ + mm->rss_stat.count[member]--; +} + +#endif /* !USE_SPLIT_PTLOCKS */ + +static inline unsigned long get_mm_rss(struct mm_struct *mm) +{ + return get_mm_counter(mm, MM_FILEPAGES) + + get_mm_counter(mm, MM_ANONPAGES); +} + +static inline unsigned long get_mm_hiwater_rss(struct mm_struct *mm) +{ + return max(mm->hiwater_rss, get_mm_rss(mm)); +} + +static inline unsigned long get_mm_hiwater_vm(struct mm_struct *mm) +{ + return max(mm->hiwater_vm, mm->total_vm); +} + +static inline void update_hiwater_rss(struct mm_struct *mm) +{ + unsigned long _rss = get_mm_rss(mm); + + if ((mm)->hiwater_rss < _rss) + (mm)->hiwater_rss = _rss; +} + +static inline void update_hiwater_vm(struct mm_struct *mm) +{ + if (mm->hiwater_vm < mm->total_vm) + mm->hiwater_vm = mm->total_vm; +} + +static inline void setmax_mm_hiwater_rss(unsigned long *maxrss, + struct mm_struct *mm) +{ + unsigned long hiwater_rss = get_mm_hiwater_rss(mm); + + if (*maxrss < hiwater_rss) + *maxrss = hiwater_rss; +} + +void sync_mm_rss(struct task_struct *task, struct mm_struct *mm); /* * A callback you can register to apply pressure to ageable caches. @@ -1114,7 +1216,7 @@ static inline void vma_nonlinear_insert(struct vm_area_struct *vma, /* mmap.c */ extern int __vm_enough_memory(struct mm_struct *mm, long pages, int cap_sys_admin); -extern void vma_adjust(struct vm_area_struct *vma, unsigned long start, +extern int vma_adjust(struct vm_area_struct *vma, unsigned long start, unsigned long end, pgoff_t pgoff, struct vm_area_struct *insert); extern struct vm_area_struct *vma_merge(struct mm_struct *, struct vm_area_struct *prev, unsigned long addr, unsigned long end, diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index 36f96271306c..048b46270aa5 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -24,12 +24,6 @@ struct address_space; #define USE_SPLIT_PTLOCKS (NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS) -#if USE_SPLIT_PTLOCKS -typedef atomic_long_t mm_counter_t; -#else /* !USE_SPLIT_PTLOCKS */ -typedef unsigned long mm_counter_t; -#endif /* !USE_SPLIT_PTLOCKS */ - /* * Each physical page in the system has a struct page associated with * it to keep track of whatever it is we are using the page for at the @@ -169,7 +163,8 @@ struct vm_area_struct { * can only be in the i_mmap tree. An anonymous MAP_PRIVATE, stack * or brk vma (with NULL file) can only be in an anon_vma list. */ - struct list_head anon_vma_node; /* Serialized by anon_vma->lock */ + struct list_head anon_vma_chain; /* Serialized by mmap_sem & + * page_table_lock */ struct anon_vma *anon_vma; /* Serialized by page_table_lock */ /* Function pointers to deal with this struct. */ @@ -201,6 +196,29 @@ struct core_state { struct completion startup; }; +enum { + MM_FILEPAGES, + MM_ANONPAGES, + MM_SWAPENTS, + NR_MM_COUNTERS +}; + +#if USE_SPLIT_PTLOCKS +#define SPLIT_RSS_COUNTING +struct mm_rss_stat { + atomic_long_t count[NR_MM_COUNTERS]; +}; +/* per-thread cached information, */ +struct task_rss_stat { + int events; /* for synchronization threshold */ + int count[NR_MM_COUNTERS]; +}; +#else /* !USE_SPLIT_PTLOCKS */ +struct mm_rss_stat { + unsigned long count[NR_MM_COUNTERS]; +}; +#endif /* !USE_SPLIT_PTLOCKS */ + struct mm_struct { struct vm_area_struct * mmap; /* list of VMAs */ struct rb_root mm_rb; @@ -227,11 +245,6 @@ struct mm_struct { * by mmlist_lock */ - /* Special counters, in some configurations protected by the - * page_table_lock, in other configurations by being atomic. - */ - mm_counter_t _file_rss; - mm_counter_t _anon_rss; unsigned long hiwater_rss; /* High-watermark of RSS usage */ unsigned long hiwater_vm; /* High-water virtual memory usage */ @@ -244,6 +257,12 @@ struct mm_struct { unsigned long saved_auxv[AT_VECTOR_SIZE]; /* for /proc/PID/auxv */ + /* + * Special counters, in some configurations protected by the + * page_table_lock, in other configurations by being atomic. + */ + struct mm_rss_stat rss_stat; + struct linux_binfmt *binfmt; cpumask_t cpu_vm_mask; diff --git a/include/linux/mmc/card.h b/include/linux/mmc/card.h index 2ee22e8af110..d02d2c6e0cfe 100644 --- a/include/linux/mmc/card.h +++ b/include/linux/mmc/card.h @@ -99,6 +99,8 @@ struct mmc_card { #define MMC_STATE_BLOCKADDR (1<<3) /* card uses block-addressing */ unsigned int quirks; /* card quirks */ #define MMC_QUIRK_LENIENT_FN0 (1<<0) /* allow SDIO FN0 writes outside of the VS CCCR range */ +#define MMC_QUIRK_BLKSZ_FOR_BYTE_MODE (1<<1) /* use func->cur_blksize */ + /* for byte mode */ u32 raw_cid[4]; /* raw card CID */ u32 raw_csd[4]; /* raw card CSD */ @@ -139,6 +141,11 @@ static inline int mmc_card_lenient_fn0(const struct mmc_card *c) return c->quirks & MMC_QUIRK_LENIENT_FN0; } +static inline int mmc_blksz_for_byte_mode(const struct mmc_card *c) +{ + return c->quirks & MMC_QUIRK_BLKSZ_FOR_BYTE_MODE; +} + #define mmc_card_name(c) ((c)->cid.prod_name) #define mmc_card_id(c) (dev_name(&(c)->dev)) diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h index eaf36364b7d4..43eaf5ca5848 100644 --- a/include/linux/mmc/host.h +++ b/include/linux/mmc/host.h @@ -14,6 +14,7 @@ #include <linux/sched.h> #include <linux/mmc/core.h> +#include <linux/mmc/pm.h> struct mmc_ios { unsigned int clock; /* clock rate */ @@ -152,6 +153,8 @@ struct mmc_host { #define MMC_CAP_NONREMOVABLE (1 << 8) /* Nonremovable e.g. eMMC */ #define MMC_CAP_WAIT_WHILE_BUSY (1 << 9) /* Waits while card is busy */ + mmc_pm_flag_t pm_caps; /* supported pm features */ + /* host specific block data */ unsigned int max_seg_size; /* see blk_queue_max_segment_size */ unsigned short max_hw_segs; /* see blk_queue_max_hw_segments */ @@ -197,6 +200,8 @@ struct mmc_host { struct task_struct *sdio_irq_thread; atomic_t sdio_irq_thread_abort; + mmc_pm_flag_t pm_flags; /* requested pm features */ + #ifdef CONFIG_LEDS_TRIGGERS struct led_trigger *led; /* activity led */ #endif diff --git a/include/linux/mmc/pm.h b/include/linux/mmc/pm.h new file mode 100644 index 000000000000..d37aac49cf9a --- /dev/null +++ b/include/linux/mmc/pm.h @@ -0,0 +1,30 @@ +/* + * linux/include/linux/mmc/pm.h + * + * Author: Nicolas Pitre + * Copyright: (C) 2009 Marvell Technology Group Ltd. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#ifndef LINUX_MMC_PM_H +#define LINUX_MMC_PM_H + +/* + * These flags are used to describe power management features that + * some cards (typically SDIO cards) might wish to benefit from when + * the host system is being suspended. There are several layers of + * abstractions involved, from the host controller driver, to the MMC core + * code, to the SDIO core code, to finally get to the actual SDIO function + * driver. This file is therefore used for common definitions shared across + * all those layers. + */ + +typedef unsigned int mmc_pm_flag_t; + +#define MMC_PM_KEEP_POWER (1 << 0) /* preserve card power during suspend */ +#define MMC_PM_WAKE_SDIO_IRQ (1 << 1) /* wake up host system on SDIO IRQ assertion */ + +#endif diff --git a/include/linux/mmc/sdio.h b/include/linux/mmc/sdio.h index 47ba464f5170..0ebaef577ff5 100644 --- a/include/linux/mmc/sdio.h +++ b/include/linux/mmc/sdio.h @@ -95,6 +95,8 @@ #define SDIO_BUS_WIDTH_1BIT 0x00 #define SDIO_BUS_WIDTH_4BIT 0x02 +#define SDIO_BUS_ASYNC_INT 0x20 + #define SDIO_BUS_CD_DISABLE 0x80 /* disable pull-up on DAT3 (pin 1) */ #define SDIO_CCCR_CAPS 0x08 diff --git a/include/linux/mmc/sdio_func.h b/include/linux/mmc/sdio_func.h index ac3ab683fec6..c6c0cceba5fe 100644 --- a/include/linux/mmc/sdio_func.h +++ b/include/linux/mmc/sdio_func.h @@ -15,6 +15,8 @@ #include <linux/device.h> #include <linux/mod_devicetable.h> +#include <linux/mmc/pm.h> + struct mmc_card; struct sdio_func; @@ -153,5 +155,8 @@ extern unsigned char sdio_f0_readb(struct sdio_func *func, extern void sdio_f0_writeb(struct sdio_func *func, unsigned char b, unsigned int addr, int *err_ret); +extern mmc_pm_flag_t sdio_get_host_pm_caps(struct sdio_func *func); +extern int sdio_set_host_pm_flags(struct sdio_func *func, mmc_pm_flag_t flags); + #endif diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index a01a103341bd..bc209d8b7b5c 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -306,6 +306,7 @@ struct zone { * free areas of different sizes */ spinlock_t lock; + int all_unreclaimable; /* All pages pinned */ #ifdef CONFIG_MEMORY_HOTPLUG /* see spanned/present_pages for more description */ seqlock_t span_seqlock; @@ -417,7 +418,6 @@ struct zone { } ____cacheline_internodealigned_in_smp; typedef enum { - ZONE_ALL_UNRECLAIMABLE, /* all pages pinned */ ZONE_RECLAIM_LOCKED, /* prevents concurrent reclaim */ ZONE_OOM_LOCKED, /* zone is in OOM killer zonelist */ } zone_flags_t; @@ -437,11 +437,6 @@ static inline void zone_clear_flag(struct zone *zone, zone_flags_t flag) clear_bit(flag, &zone->flags); } -static inline int zone_is_all_unreclaimable(const struct zone *zone) -{ - return test_bit(ZONE_ALL_UNRECLAIMABLE, &zone->flags); -} - static inline int zone_is_reclaim_locked(const struct zone *zone) { return test_bit(ZONE_RECLAIM_LOCKED, &zone->flags); diff --git a/include/linux/nodemask.h b/include/linux/nodemask.h index 454997cccbd8..c4fa64b585ff 100644 --- a/include/linux/nodemask.h +++ b/include/linux/nodemask.h @@ -69,8 +69,6 @@ * int node_online(node) Is some node online? * int node_possible(node) Is some node possible? * - * int any_online_node(mask) First online node in mask - * * node_set_online(node) set bit 'node' in node_online_map * node_set_offline(node) clear bit 'node' in node_online_map * @@ -467,15 +465,6 @@ static inline int num_node_state(enum node_states state) #define node_online_map node_states[N_ONLINE] #define node_possible_map node_states[N_POSSIBLE] -#define any_online_node(mask) \ -({ \ - int node; \ - for_each_node_mask(node, (mask)) \ - if (node_online(node)) \ - break; \ - node; \ -}) - #define num_online_nodes() num_node_state(N_ONLINE) #define num_possible_nodes() num_node_state(N_POSSIBLE) #define node_online(node) node_state((node), N_ONLINE) diff --git a/include/linux/rmap.h b/include/linux/rmap.h index b019ae64e2ab..d25bd224d370 100644 --- a/include/linux/rmap.h +++ b/include/linux/rmap.h @@ -37,7 +37,27 @@ struct anon_vma { * is serialized by a system wide lock only visible to * mm_take_all_locks() (mm_all_locks_mutex). */ - struct list_head head; /* List of private "related" vmas */ + struct list_head head; /* Chain of private "related" vmas */ +}; + +/* + * The copy-on-write semantics of fork mean that an anon_vma + * can become associated with multiple processes. Furthermore, + * each child process will have its own anon_vma, where new + * pages for that process are instantiated. + * + * This structure allows us to find the anon_vmas associated + * with a VMA, or the VMAs associated with an anon_vma. + * The "same_vma" list contains the anon_vma_chains linking + * all the anon_vmas associated with this VMA. + * The "same_anon_vma" list contains the anon_vma_chains + * which link all the VMAs associated with this anon_vma. + */ +struct anon_vma_chain { + struct vm_area_struct *vma; + struct anon_vma *anon_vma; + struct list_head same_vma; /* locked by mmap_sem & page_table_lock */ + struct list_head same_anon_vma; /* locked by anon_vma->lock */ }; #ifdef CONFIG_MMU @@ -89,15 +109,23 @@ static inline void anon_vma_unlock(struct vm_area_struct *vma) */ void anon_vma_init(void); /* create anon_vma_cachep */ int anon_vma_prepare(struct vm_area_struct *); -void __anon_vma_merge(struct vm_area_struct *, struct vm_area_struct *); -void anon_vma_unlink(struct vm_area_struct *); -void anon_vma_link(struct vm_area_struct *); +void unlink_anon_vmas(struct vm_area_struct *); +int anon_vma_clone(struct vm_area_struct *, struct vm_area_struct *); +int anon_vma_fork(struct vm_area_struct *, struct vm_area_struct *); void __anon_vma_link(struct vm_area_struct *); void anon_vma_free(struct anon_vma *); +static inline void anon_vma_merge(struct vm_area_struct *vma, + struct vm_area_struct *next) +{ + VM_BUG_ON(vma->anon_vma != next->anon_vma); + unlink_anon_vmas(next); +} + /* * rmap interfaces called when adding or removing pte of page */ +void page_move_anon_rmap(struct page *, struct vm_area_struct *, unsigned long); void page_add_anon_rmap(struct page *, struct vm_area_struct *, unsigned long); void page_add_new_anon_rmap(struct page *, struct vm_area_struct *, unsigned long); void page_add_file_rmap(struct page *); @@ -181,7 +209,7 @@ static inline int page_referenced(struct page *page, int is_locked, unsigned long *vm_flags) { *vm_flags = 0; - return TestClearPageReferenced(page); + return 0; } #define try_to_unmap(page, refs) SWAP_FAIL diff --git a/include/linux/sched.h b/include/linux/sched.h index 4b1753f7e48e..46c6f8d5dc06 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -396,60 +396,6 @@ extern void arch_unmap_area_topdown(struct mm_struct *, unsigned long); static inline void arch_pick_mmap_layout(struct mm_struct *mm) {} #endif -#if USE_SPLIT_PTLOCKS -/* - * The mm counters are not protected by its page_table_lock, - * so must be incremented atomically. - */ -#define set_mm_counter(mm, member, value) atomic_long_set(&(mm)->_##member, value) -#define get_mm_counter(mm, member) ((unsigned long)atomic_long_read(&(mm)->_##member)) -#define add_mm_counter(mm, member, value) atomic_long_add(value, &(mm)->_##member) -#define inc_mm_counter(mm, member) atomic_long_inc(&(mm)->_##member) -#define dec_mm_counter(mm, member) atomic_long_dec(&(mm)->_##member) - -#else /* !USE_SPLIT_PTLOCKS */ -/* - * The mm counters are protected by its page_table_lock, - * so can be incremented directly. - */ -#define set_mm_counter(mm, member, value) (mm)->_##member = (value) -#define get_mm_counter(mm, member) ((mm)->_##member) -#define add_mm_counter(mm, member, value) (mm)->_##member += (value) -#define inc_mm_counter(mm, member) (mm)->_##member++ -#define dec_mm_counter(mm, member) (mm)->_##member-- - -#endif /* !USE_SPLIT_PTLOCKS */ - -#define get_mm_rss(mm) \ - (get_mm_counter(mm, file_rss) + get_mm_counter(mm, anon_rss)) -#define update_hiwater_rss(mm) do { \ - unsigned long _rss = get_mm_rss(mm); \ - if ((mm)->hiwater_rss < _rss) \ - (mm)->hiwater_rss = _rss; \ -} while (0) -#define update_hiwater_vm(mm) do { \ - if ((mm)->hiwater_vm < (mm)->total_vm) \ - (mm)->hiwater_vm = (mm)->total_vm; \ -} while (0) - -static inline unsigned long get_mm_hiwater_rss(struct mm_struct *mm) -{ - return max(mm->hiwater_rss, get_mm_rss(mm)); -} - -static inline void setmax_mm_hiwater_rss(unsigned long *maxrss, - struct mm_struct *mm) -{ - unsigned long hiwater_rss = get_mm_hiwater_rss(mm); - - if (*maxrss < hiwater_rss) - *maxrss = hiwater_rss; -} - -static inline unsigned long get_mm_hiwater_vm(struct mm_struct *mm) -{ - return max(mm->hiwater_vm, mm->total_vm); -} extern void set_dumpable(struct mm_struct *mm, int value); extern int get_dumpable(struct mm_struct *mm); @@ -1274,7 +1220,9 @@ struct task_struct { struct plist_node pushable_tasks; struct mm_struct *mm, *active_mm; - +#if defined(SPLIT_RSS_COUNTING) + struct task_rss_stat rss_stat; +#endif /* task state */ int exit_state; int exit_code, exit_signal; diff --git a/include/linux/smp.h b/include/linux/smp.h index 7a0570e6a596..cfa2d20e35f1 100644 --- a/include/linux/smp.h +++ b/include/linux/smp.h @@ -154,7 +154,7 @@ smp_call_function_any(const struct cpumask *mask, void (*func)(void *info), /* * smp_processor_id(): get the current CPU ID. * - * if DEBUG_PREEMPT is enabled the we check whether it is + * if DEBUG_PREEMPT is enabled then we check whether it is * used in a preemption-safe way. (smp_processor_id() is safe * if it's used in a preemption-off critical section, or in * a thread that is bound to the current CPU.) diff --git a/include/linux/spi/max7301.h b/include/linux/spi/max7301.h index 6dfd83f19b4b..34af0a3477bf 100644 --- a/include/linux/spi/max7301.h +++ b/include/linux/spi/max7301.h @@ -1,9 +1,27 @@ #ifndef LINUX_SPI_MAX7301_H #define LINUX_SPI_MAX7301_H +#include <linux/gpio.h> + +/* + * Some registers must be read back to modify. + * To save time we cache them here in memory + */ +struct max7301 { + struct mutex lock; + u8 port_config[8]; /* field 0 is unused */ + u32 out_level; /* cached output levels */ + struct gpio_chip chip; + struct device *dev; + int (*write)(struct device *dev, unsigned int reg, unsigned int val); + int (*read)(struct device *dev, unsigned int reg); +}; + struct max7301_platform_data { /* number assigned to the first GPIO */ unsigned base; }; +extern int __max730x_remove(struct device *dev); +extern int __max730x_probe(struct max7301 *ts); #endif diff --git a/init/initramfs.c b/init/initramfs.c index b37d34beb90b..37d3859b1b32 100644 --- a/init/initramfs.c +++ b/init/initramfs.c @@ -525,7 +525,7 @@ static void __init clean_rootfs(void) int fd; void *buf; struct linux_dirent64 *dirp; - int count; + int num; fd = sys_open("/", O_RDONLY, 0); WARN_ON(fd < 0); @@ -539,9 +539,9 @@ static void __init clean_rootfs(void) } dirp = buf; - count = sys_getdents64(fd, dirp, BUF_SIZE); - while (count > 0) { - while (count > 0) { + num = sys_getdents64(fd, dirp, BUF_SIZE); + while (num > 0) { + while (num > 0) { struct stat st; int ret; @@ -554,12 +554,12 @@ static void __init clean_rootfs(void) sys_unlink(dirp->d_name); } - count -= dirp->d_reclen; + num -= dirp->d_reclen; dirp = (void *)dirp + dirp->d_reclen; } dirp = buf; memset(buf, 0, BUF_SIZE); - count = sys_getdents64(fd, dirp, BUF_SIZE); + num = sys_getdents64(fd, dirp, BUF_SIZE); } sys_close(fd); diff --git a/init/main.c b/init/main.c index 40aaa020cd68..a1ab78ceb4b6 100644 --- a/init/main.c +++ b/init/main.c @@ -174,7 +174,7 @@ static int __init maxcpus(char *str) early_param("maxcpus", maxcpus); #else -const unsigned int setup_max_cpus = NR_CPUS; +static const unsigned int setup_max_cpus = NR_CPUS; #endif /* @@ -618,7 +618,7 @@ asmlinkage void __init start_kernel(void) local_irq_enable(); /* Interrupts are enabled now so all GFP allocations are safe. */ - set_gfp_allowed_mask(__GFP_BITS_MASK); + gfp_allowed_mask = __GFP_BITS_MASK; kmem_cache_init_late(); @@ -847,7 +847,8 @@ static noinline int init_post(void) run_init_process("/bin/init"); run_init_process("/bin/sh"); - panic("No init found. Try passing init= option to kernel."); + panic("No init found. Try passing init= option to kernel. " + "See Linux Documentation/init.txt for guidance."); } static int __init kernel_init(void * unused) diff --git a/kernel/Makefile b/kernel/Makefile index 7b974699f8c2..a987aa1676b5 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -91,6 +91,9 @@ obj-$(CONFIG_TASK_DELAY_ACCT) += delayacct.o obj-$(CONFIG_TASKSTATS) += taskstats.o tsacct.o obj-$(CONFIG_TRACEPOINTS) += tracepoint.o obj-$(CONFIG_LATENCYTOP) += latencytop.o +obj-$(CONFIG_BINFMT_ELF) += elfcore.o +obj-$(CONFIG_COMPAT_BINFMT_ELF) += elfcore.o +obj-$(CONFIG_BINFMT_ELF_FDPIC) += elfcore.o obj-$(CONFIG_FUNCTION_TRACER) += trace/ obj-$(CONFIG_TRACING) += trace/ obj-$(CONFIG_X86_DS) += trace/ diff --git a/kernel/cpu.c b/kernel/cpu.c index 677f25376a38..f8cced2692b3 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -338,7 +338,7 @@ int __cpuinit cpu_up(unsigned int cpu) if (!cpu_possible(cpu)) { printk(KERN_ERR "can't online cpu %d because it is not " "configured as may-hotadd at boot time\n", cpu); -#if defined(CONFIG_IA64) || defined(CONFIG_X86_64) +#if defined(CONFIG_IA64) printk(KERN_ERR "please check additional_cpus= boot " "parameter\n"); #endif diff --git a/kernel/elfcore.c b/kernel/elfcore.c new file mode 100644 index 000000000000..ff915efef66d --- /dev/null +++ b/kernel/elfcore.c @@ -0,0 +1,28 @@ +#include <linux/elf.h> +#include <linux/fs.h> +#include <linux/mm.h> + +#include <asm/elf.h> + + +Elf_Half __weak elf_core_extra_phdrs(void) +{ + return 0; +} + +int __weak elf_core_write_extra_phdrs(struct file *file, loff_t offset, size_t *size, + unsigned long limit) +{ + return 1; +} + +int __weak elf_core_write_extra_data(struct file *file, size_t *size, + unsigned long limit) +{ + return 1; +} + +size_t __weak elf_core_extra_data_size(void) +{ + return 0; +} diff --git a/kernel/exit.c b/kernel/exit.c index 45ed043b8bf5..ce1e48c2d93d 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -952,7 +952,8 @@ NORET_TYPE void do_exit(long code) preempt_count()); acct_update_integrals(tsk); - + /* sync mm's RSS info before statistics gathering */ + sync_mm_rss(tsk, tsk->mm); group_dead = atomic_dec_and_test(&tsk->signal->live); if (group_dead) { hrtimer_cancel(&tsk->signal->real_timer); @@ -1188,7 +1189,7 @@ static int wait_task_zombie(struct wait_opts *wo, struct task_struct *p) if (unlikely(wo->wo_flags & WNOWAIT)) { int exit_code = p->exit_code; - int why, status; + int why; get_task_struct(p); read_unlock(&tasklist_lock); diff --git a/kernel/fork.c b/kernel/fork.c index 17bbf093356d..b0ec34abc0bb 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -329,15 +329,17 @@ static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm) if (!tmp) goto fail_nomem; *tmp = *mpnt; + INIT_LIST_HEAD(&tmp->anon_vma_chain); pol = mpol_dup(vma_policy(mpnt)); retval = PTR_ERR(pol); if (IS_ERR(pol)) goto fail_nomem_policy; vma_set_policy(tmp, pol); + if (anon_vma_fork(tmp, mpnt)) + goto fail_nomem_anon_vma_fork; tmp->vm_flags &= ~VM_LOCKED; tmp->vm_mm = mm; tmp->vm_next = NULL; - anon_vma_link(tmp); file = tmp->vm_file; if (file) { struct inode *inode = file->f_path.dentry->d_inode; @@ -392,6 +394,8 @@ out: flush_tlb_mm(oldmm); up_write(&oldmm->mmap_sem); return retval; +fail_nomem_anon_vma_fork: + mpol_put(pol); fail_nomem_policy: kmem_cache_free(vm_area_cachep, tmp); fail_nomem: @@ -455,8 +459,7 @@ static struct mm_struct * mm_init(struct mm_struct * mm, struct task_struct *p) (current->mm->flags & MMF_INIT_MASK) : default_dump_filter; mm->core_state = NULL; mm->nr_ptes = 0; - set_mm_counter(mm, file_rss, 0); - set_mm_counter(mm, anon_rss, 0); + memset(&mm->rss_stat, 0, sizeof(mm->rss_stat)); spin_lock_init(&mm->page_table_lock); mm->free_area_cache = TASK_UNMAPPED_BASE; mm->cached_hole_size = ~0UL; @@ -825,6 +828,8 @@ void __cleanup_sighand(struct sighand_struct *sighand) */ static void posix_cpu_timers_init_group(struct signal_struct *sig) { + unsigned long cpu_limit; + /* Thread group counters. */ thread_group_cputime_init(sig); @@ -839,9 +844,9 @@ static void posix_cpu_timers_init_group(struct signal_struct *sig) sig->cputime_expires.virt_exp = cputime_zero; sig->cputime_expires.sched_exp = 0; - if (sig->rlim[RLIMIT_CPU].rlim_cur != RLIM_INFINITY) { - sig->cputime_expires.prof_exp = - secs_to_cputime(sig->rlim[RLIMIT_CPU].rlim_cur); + cpu_limit = ACCESS_ONCE(sig->rlim[RLIMIT_CPU].rlim_cur); + if (cpu_limit != RLIM_INFINITY) { + sig->cputime_expires.prof_exp = secs_to_cputime(cpu_limit); sig->cputimer.running = 1; } @@ -1034,7 +1039,7 @@ static struct task_struct *copy_process(unsigned long clone_flags, #endif retval = -EAGAIN; if (atomic_read(&p->real_cred->user->processes) >= - p->signal->rlim[RLIMIT_NPROC].rlim_cur) { + task_rlimit(p, RLIMIT_NPROC)) { if (!capable(CAP_SYS_ADMIN) && !capable(CAP_SYS_RESOURCE) && p->real_cred->user != INIT_USER) goto bad_fork_free; diff --git a/kernel/panic.c b/kernel/panic.c index c787333282b8..13d966b4c14a 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -36,15 +36,36 @@ ATOMIC_NOTIFIER_HEAD(panic_notifier_list); EXPORT_SYMBOL(panic_notifier_list); -static long no_blink(long time) -{ - return 0; -} - /* Returns how long it waited in ms */ long (*panic_blink)(long time); EXPORT_SYMBOL(panic_blink); +static void panic_blink_one_second(void) +{ + static long i = 0, end; + + if (panic_blink) { + end = i + MSEC_PER_SEC; + + while (i < end) { + i += panic_blink(i); + mdelay(1); + i++; + } + } else { + /* + * When running under a hypervisor a small mdelay may get + * rounded up to the hypervisor timeslice. For example, with + * a 1ms in 10ms hypervisor timeslice we might inflate a + * mdelay(1) loop by 10x. + * + * If we have nothing to blink, spin on 1 second calls to + * mdelay to avoid this. + */ + mdelay(MSEC_PER_SEC); + } +} + /** * panic - halt the system * @fmt: The text string to print @@ -95,9 +116,6 @@ NORET_TYPE void panic(const char * fmt, ...) bust_spinlocks(0); - if (!panic_blink) - panic_blink = no_blink; - if (panic_timeout > 0) { /* * Delay timeout seconds before rebooting the machine. @@ -105,11 +123,9 @@ NORET_TYPE void panic(const char * fmt, ...) */ printk(KERN_EMERG "Rebooting in %d seconds..", panic_timeout); - for (i = 0; i < panic_timeout*1000; ) { + for (i = 0; i < panic_timeout; i++) { touch_nmi_watchdog(); - i += panic_blink(i); - mdelay(1); - i++; + panic_blink_one_second(); } /* * This will not be a clean reboot, with everything @@ -135,11 +151,9 @@ NORET_TYPE void panic(const char * fmt, ...) } #endif local_irq_enable(); - for (i = 0; ; ) { + while (1) { touch_softlockup_watchdog(); - i += panic_blink(i); - mdelay(1); - i++; + panic_blink_one_second(); } } diff --git a/kernel/params.c b/kernel/params.c index cf1b69183127..8d95f5451b22 100644 --- a/kernel/params.c +++ b/kernel/params.c @@ -24,7 +24,6 @@ #include <linux/err.h> #include <linux/slab.h> #include <linux/ctype.h> -#include <linux/string.h> #if 0 #define DEBUGP printk diff --git a/kernel/perf_event.c b/kernel/perf_event.c index a661e7991865..8e352c756ba7 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -2610,7 +2610,7 @@ static int perf_mmap(struct file *file, struct vm_area_struct *vma) if (user_locked > user_lock_limit) extra = user_locked - user_lock_limit; - lock_limit = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur; + lock_limit = rlimit(RLIMIT_MEMLOCK); lock_limit >>= PAGE_SHIFT; locked = vma->vm_mm->locked_vm + extra; diff --git a/kernel/pid.c b/kernel/pid.c index b08e697cd83f..86b296943e5f 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -376,7 +376,7 @@ struct task_struct *pid_task(struct pid *pid, enum pid_type type) EXPORT_SYMBOL(pid_task); /* - * Must be called under rcu_read_lock() or with tasklist_lock read-held. + * Must be called under rcu_read_lock(). */ struct task_struct *find_task_by_pid_ns(pid_t nr, struct pid_namespace *ns) { diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index 438ff4523513..1a22dfd42df9 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -982,6 +982,7 @@ static void check_thread_timers(struct task_struct *tsk, int maxfire; struct list_head *timers = tsk->cpu_timers; struct signal_struct *const sig = tsk->signal; + unsigned long soft; maxfire = 20; tsk->cputime_expires.prof_exp = cputime_zero; @@ -1030,9 +1031,10 @@ static void check_thread_timers(struct task_struct *tsk, /* * Check for the special case thread timers. */ - if (sig->rlim[RLIMIT_RTTIME].rlim_cur != RLIM_INFINITY) { - unsigned long hard = sig->rlim[RLIMIT_RTTIME].rlim_max; - unsigned long *soft = &sig->rlim[RLIMIT_RTTIME].rlim_cur; + soft = ACCESS_ONCE(sig->rlim[RLIMIT_RTTIME].rlim_cur); + if (soft != RLIM_INFINITY) { + unsigned long hard = + ACCESS_ONCE(sig->rlim[RLIMIT_RTTIME].rlim_max); if (hard != RLIM_INFINITY && tsk->rt.timeout > DIV_ROUND_UP(hard, USEC_PER_SEC/HZ)) { @@ -1043,14 +1045,13 @@ static void check_thread_timers(struct task_struct *tsk, __group_send_sig_info(SIGKILL, SEND_SIG_PRIV, tsk); return; } - if (tsk->rt.timeout > DIV_ROUND_UP(*soft, USEC_PER_SEC/HZ)) { + if (tsk->rt.timeout > DIV_ROUND_UP(soft, USEC_PER_SEC/HZ)) { /* * At the soft limit, send a SIGXCPU every second. */ - if (sig->rlim[RLIMIT_RTTIME].rlim_cur - < sig->rlim[RLIMIT_RTTIME].rlim_max) { - sig->rlim[RLIMIT_RTTIME].rlim_cur += - USEC_PER_SEC; + if (soft < hard) { + soft += USEC_PER_SEC; + sig->rlim[RLIMIT_RTTIME].rlim_cur = soft; } printk(KERN_INFO "RT Watchdog Timeout: %s[%d]\n", @@ -1121,6 +1122,7 @@ static void check_process_timers(struct task_struct *tsk, unsigned long long sum_sched_runtime, sched_expires; struct list_head *timers = sig->cpu_timers; struct task_cputime cputime; + unsigned long soft; /* * Don't sample the current process CPU clocks if there are no timers. @@ -1193,11 +1195,13 @@ static void check_process_timers(struct task_struct *tsk, SIGPROF); check_cpu_itimer(tsk, &sig->it[CPUCLOCK_VIRT], &virt_expires, utime, SIGVTALRM); - - if (sig->rlim[RLIMIT_CPU].rlim_cur != RLIM_INFINITY) { + soft = ACCESS_ONCE(sig->rlim[RLIMIT_CPU].rlim_cur); + if (soft != RLIM_INFINITY) { unsigned long psecs = cputime_to_secs(ptime); + unsigned long hard = + ACCESS_ONCE(sig->rlim[RLIMIT_CPU].rlim_max); cputime_t x; - if (psecs >= sig->rlim[RLIMIT_CPU].rlim_max) { + if (psecs >= hard) { /* * At the hard limit, we just die. * No need to calculate anything else now. @@ -1205,17 +1209,17 @@ static void check_process_timers(struct task_struct *tsk, __group_send_sig_info(SIGKILL, SEND_SIG_PRIV, tsk); return; } - if (psecs >= sig->rlim[RLIMIT_CPU].rlim_cur) { + if (psecs >= soft) { /* * At the soft limit, send a SIGXCPU every second. */ __group_send_sig_info(SIGXCPU, SEND_SIG_PRIV, tsk); - if (sig->rlim[RLIMIT_CPU].rlim_cur - < sig->rlim[RLIMIT_CPU].rlim_max) { - sig->rlim[RLIMIT_CPU].rlim_cur++; + if (soft < hard) { + soft++; + sig->rlim[RLIMIT_CPU].rlim_cur = soft; } } - x = secs_to_cputime(sig->rlim[RLIMIT_CPU].rlim_cur); + x = secs_to_cputime(soft); if (cputime_eq(prof_expires, cputime_zero) || cputime_lt(x, prof_expires)) { prof_expires = x; diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index bbfe472d7524..da5288ec2392 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -323,6 +323,7 @@ static int create_image(int platform_mode) int hibernation_snapshot(int platform_mode) { int error; + gfp_t saved_mask; error = platform_begin(platform_mode); if (error) @@ -334,6 +335,7 @@ int hibernation_snapshot(int platform_mode) goto Close; suspend_console(); + saved_mask = clear_gfp_allowed_mask(GFP_IOFS); error = dpm_suspend_start(PMSG_FREEZE); if (error) goto Recover_platform; @@ -351,6 +353,7 @@ int hibernation_snapshot(int platform_mode) dpm_resume_end(in_suspend ? (error ? PMSG_RECOVER : PMSG_THAW) : PMSG_RESTORE); + set_gfp_allowed_mask(saved_mask); resume_console(); Close: platform_end(platform_mode); @@ -445,14 +448,17 @@ static int resume_target_kernel(bool platform_mode) int hibernation_restore(int platform_mode) { int error; + gfp_t saved_mask; pm_prepare_console(); suspend_console(); + saved_mask = clear_gfp_allowed_mask(GFP_IOFS); error = dpm_suspend_start(PMSG_QUIESCE); if (!error) { error = resume_target_kernel(platform_mode); dpm_resume_end(PMSG_RECOVER); } + set_gfp_allowed_mask(saved_mask); resume_console(); pm_restore_console(); return error; @@ -466,6 +472,7 @@ int hibernation_restore(int platform_mode) int hibernation_platform_enter(void) { int error; + gfp_t saved_mask; if (!hibernation_ops) return -ENOSYS; @@ -481,6 +488,7 @@ int hibernation_platform_enter(void) entering_platform_hibernation = true; suspend_console(); + saved_mask = clear_gfp_allowed_mask(GFP_IOFS); error = dpm_suspend_start(PMSG_HIBERNATE); if (error) { if (hibernation_ops->recover) @@ -518,6 +526,7 @@ int hibernation_platform_enter(void) Resume_devices: entering_platform_hibernation = false; dpm_resume_end(PMSG_RESTORE); + set_gfp_allowed_mask(saved_mask); resume_console(); Close: diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index 6f10dfc2d3e9..44cce10b582d 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -189,6 +189,7 @@ static int suspend_enter(suspend_state_t state) int suspend_devices_and_enter(suspend_state_t state) { int error; + gfp_t saved_mask; if (!suspend_ops) return -ENOSYS; @@ -199,6 +200,7 @@ int suspend_devices_and_enter(suspend_state_t state) goto Close; } suspend_console(); + saved_mask = clear_gfp_allowed_mask(GFP_IOFS); suspend_test_start(); error = dpm_suspend_start(PMSG_SUSPEND); if (error) { @@ -215,6 +217,7 @@ int suspend_devices_and_enter(suspend_state_t state) suspend_test_start(); dpm_resume_end(PMSG_RESUME); suspend_test_finish("resume devices"); + set_gfp_allowed_mask(saved_mask); resume_console(); Close: if (suspend_ops->end) diff --git a/kernel/printk.c b/kernel/printk.c index 40674122ecf2..75077ad0b537 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -70,8 +70,6 @@ int console_printk[4] = { DEFAULT_CONSOLE_LOGLEVEL, /* default_console_loglevel */ }; -static int saved_console_loglevel = -1; - /* * Low level drivers may need that to know if they can schedule in * their unblank() callback or not. So let's export it. @@ -146,6 +144,7 @@ static char __log_buf[__LOG_BUF_LEN]; static char *log_buf = __log_buf; static int log_buf_len = __LOG_BUF_LEN; static unsigned logged_chars; /* Number of chars produced since last read+clear operation */ +static int saved_console_loglevel = -1; #ifdef CONFIG_KEXEC /* diff --git a/kernel/relay.c b/kernel/relay.c index c705a41b4ba3..3d97f2821611 100644 --- a/kernel/relay.c +++ b/kernel/relay.c @@ -1215,14 +1215,14 @@ static void relay_page_release(struct splice_pipe_desc *spd, unsigned int i) /* * subbuf_splice_actor - splice up to one subbuf's worth of data */ -static int subbuf_splice_actor(struct file *in, +static ssize_t subbuf_splice_actor(struct file *in, loff_t *ppos, struct pipe_inode_info *pipe, size_t len, unsigned int flags, int *nonpad_ret) { - unsigned int pidx, poff, total_len, subbuf_pages, nr_pages, ret; + unsigned int pidx, poff, total_len, subbuf_pages, nr_pages; struct rchan_buf *rbuf = in->private_data; unsigned int subbuf_size = rbuf->chan->subbuf_size; uint64_t pos = (uint64_t) *ppos; @@ -1241,6 +1241,7 @@ static int subbuf_splice_actor(struct file *in, .ops = &relay_pipe_buf_ops, .spd_release = relay_page_release, }; + ssize_t ret; if (rbuf->subbufs_produced == rbuf->subbufs_consumed) return 0; diff --git a/kernel/sched.c b/kernel/sched.c index abb36b16b93b..b47ceeec1a91 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -4353,7 +4353,7 @@ int can_nice(const struct task_struct *p, const int nice) /* convert nice value [19,-20] to rlimit style value [1,40] */ int nice_rlim = 20 - nice; - return (nice_rlim <= p->signal->rlim[RLIMIT_NICE].rlim_cur || + return (nice_rlim <= task_rlimit(p, RLIMIT_NICE) || capable(CAP_SYS_NICE)); } @@ -4530,7 +4530,7 @@ recheck: if (!lock_task_sighand(p, &flags)) return -ESRCH; - rlim_rtprio = p->signal->rlim[RLIMIT_RTPRIO].rlim_cur; + rlim_rtprio = task_rlimit(p, RLIMIT_RTPRIO); unlock_task_sighand(p, &flags); /* can't set/change the rt policy */ diff --git a/kernel/sched_cpupri.c b/kernel/sched_cpupri.c index eeb3506c4834..82095bf2099f 100644 --- a/kernel/sched_cpupri.c +++ b/kernel/sched_cpupri.c @@ -47,7 +47,7 @@ static int convert_prio(int prio) } #define for_each_cpupri_active(array, idx) \ - for_each_bit(idx, array, CPUPRI_NR_PRIORITIES) + for_each_set_bit(idx, array, CPUPRI_NR_PRIORITIES) /** * cpupri_find - find the best (lowest-pri) CPU in the system diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index bf3e38fdbe6d..5a6ed1f0990a 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -1662,8 +1662,9 @@ static void watchdog(struct rq *rq, struct task_struct *p) if (!p->signal) return; - soft = p->signal->rlim[RLIMIT_RTTIME].rlim_cur; - hard = p->signal->rlim[RLIMIT_RTTIME].rlim_max; + /* max may change after cur was read, this will be fixed next tick */ + soft = task_rlimit(p, RLIMIT_RTTIME); + hard = task_rlimit_max(p, RLIMIT_RTTIME); if (soft != RLIM_INFINITY) { unsigned long next; diff --git a/kernel/signal.c b/kernel/signal.c index 5bb9baffa4f1..dbd7fe073c55 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -245,7 +245,7 @@ __sigqueue_alloc(int sig, struct task_struct *t, gfp_t flags, int override_rlimi if (override_rlimit || atomic_read(&user->sigpending) <= - t->signal->rlim[RLIMIT_SIGPENDING].rlim_cur) { + task_rlimit(t, RLIMIT_SIGPENDING)) { q = kmem_cache_alloc(sigqueue_cachep, flags); } else { print_dropped_signal(sig); diff --git a/kernel/sys.c b/kernel/sys.c index 877fe4f8e05e..9814e43fb23b 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -571,8 +571,7 @@ static int set_user(struct cred *new) if (!new_user) return -EAGAIN; - if (atomic_read(&new_user->processes) >= - current->signal->rlim[RLIMIT_NPROC].rlim_cur && + if (atomic_read(&new_user->processes) >= rlimit(RLIMIT_NPROC) && new_user != INIT_USER) { free_uid(new_user); return -EAGAIN; diff --git a/kernel/tsacct.c b/kernel/tsacct.c index 00d59d048edf..0a67e041edf8 100644 --- a/kernel/tsacct.c +++ b/kernel/tsacct.c @@ -21,6 +21,7 @@ #include <linux/tsacct_kern.h> #include <linux/acct.h> #include <linux/jiffies.h> +#include <linux/mm.h> /* * fill in basic accounting fields diff --git a/lib/Kconfig b/lib/Kconfig index 97b136ff117e..170d8ca901d8 100644 --- a/lib/Kconfig +++ b/lib/Kconfig @@ -160,6 +160,9 @@ config TEXTSEARCH_BM config TEXTSEARCH_FSM tristate +config BTREE + boolean + config HAS_IOMEM boolean depends on !NO_IOMEM diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 5e3407d997b2..b520ec1f33c5 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -864,8 +864,7 @@ config DEBUG_FORCE_WEAK_PER_CPU config LKDTM tristate "Linux Kernel Dump Test Tool Module" - depends on DEBUG_KERNEL - depends on KPROBES + depends on DEBUG_FS depends on BLOCK default n help @@ -876,7 +875,7 @@ config LKDTM called lkdtm. Documentation on how to use the module can be found in - drivers/misc/lkdtm.c + Documentation/fault-injection/provoke-crashes.txt config FAULT_INJECTION bool "Fault-injection framework" diff --git a/lib/Makefile b/lib/Makefile index 3b0b4a696db9..2e152aed7198 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -41,6 +41,7 @@ lib-$(CONFIG_GENERIC_FIND_NEXT_BIT) += find_next_bit.o obj-$(CONFIG_GENERIC_FIND_LAST_BIT) += find_last_bit.o obj-$(CONFIG_GENERIC_HWEIGHT) += hweight.o obj-$(CONFIG_LOCK_KERNEL) += kernel_lock.o +obj-$(CONFIG_BTREE) += btree.o obj-$(CONFIG_DEBUG_PREEMPT) += smp_processor_id.o obj-$(CONFIG_DEBUG_LIST) += list_debug.o obj-$(CONFIG_DEBUG_OBJECTS) += debugobjects.o diff --git a/lib/bitmap.c b/lib/bitmap.c index 11bf49750583..ffb78c916ccd 100644 --- a/lib/bitmap.c +++ b/lib/bitmap.c @@ -487,7 +487,7 @@ int __bitmap_parse(const char *buf, unsigned int buflen, EXPORT_SYMBOL(__bitmap_parse); /** - * bitmap_parse_user() + * bitmap_parse_user - convert an ASCII hex string in a user buffer into a bitmap * * @ubuf: pointer to user buffer containing string. * @ulen: buffer size in bytes. If string is smaller than this @@ -619,7 +619,7 @@ int bitmap_parselist(const char *bp, unsigned long *maskp, int nmaskbits) EXPORT_SYMBOL(bitmap_parselist); /** - * bitmap_pos_to_ord(buf, pos, bits) + * bitmap_pos_to_ord - find ordinal of set bit at given position in bitmap * @buf: pointer to a bitmap * @pos: a bit position in @buf (0 <= @pos < @bits) * @bits: number of valid bit positions in @buf @@ -655,7 +655,7 @@ static int bitmap_pos_to_ord(const unsigned long *buf, int pos, int bits) } /** - * bitmap_ord_to_pos(buf, ord, bits) + * bitmap_ord_to_pos - find position of n-th set bit in bitmap * @buf: pointer to bitmap * @ord: ordinal bit position (n-th set bit, n >= 0) * @bits: number of valid bit positions in @buf @@ -733,10 +733,9 @@ void bitmap_remap(unsigned long *dst, const unsigned long *src, bitmap_zero(dst, bits); w = bitmap_weight(new, bits); - for (oldbit = find_first_bit(src, bits); - oldbit < bits; - oldbit = find_next_bit(src, bits, oldbit + 1)) { + for_each_set_bit(oldbit, src, bits) { int n = bitmap_pos_to_ord(old, oldbit, bits); + if (n < 0 || w == 0) set_bit(oldbit, dst); /* identity map */ else @@ -903,9 +902,7 @@ void bitmap_onto(unsigned long *dst, const unsigned long *orig, */ m = 0; - for (n = find_first_bit(relmap, bits); - n < bits; - n = find_next_bit(relmap, bits, n + 1)) { + for_each_set_bit(n, relmap, bits) { /* m == bitmap_pos_to_ord(relmap, n, bits) */ if (test_bit(m, orig)) set_bit(n, dst); @@ -934,9 +931,7 @@ void bitmap_fold(unsigned long *dst, const unsigned long *orig, return; bitmap_zero(dst, bits); - for (oldbit = find_first_bit(orig, bits); - oldbit < bits; - oldbit = find_next_bit(orig, bits, oldbit + 1)) + for_each_set_bit(oldbit, orig, bits) set_bit(oldbit % sz, dst); } EXPORT_SYMBOL(bitmap_fold); diff --git a/lib/btree.c b/lib/btree.c new file mode 100644 index 000000000000..41859a820218 --- /dev/null +++ b/lib/btree.c @@ -0,0 +1,797 @@ +/* + * lib/btree.c - Simple In-memory B+Tree + * + * As should be obvious for Linux kernel code, license is GPLv2 + * + * Copyright (c) 2007-2008 Joern Engel <joern@logfs.org> + * Bits and pieces stolen from Peter Zijlstra's code, which is + * Copyright 2007, Red Hat Inc. Peter Zijlstra <pzijlstr@redhat.com> + * GPLv2 + * + * see http://programming.kicks-ass.net/kernel-patches/vma_lookup/btree.patch + * + * A relatively simple B+Tree implementation. I have written it as a learning + * excercise to understand how B+Trees work. Turned out to be useful as well. + * + * B+Trees can be used similar to Linux radix trees (which don't have anything + * in common with textbook radix trees, beware). Prerequisite for them working + * well is that access to a random tree node is much faster than a large number + * of operations within each node. + * + * Disks have fulfilled the prerequisite for a long time. More recently DRAM + * has gained similar properties, as memory access times, when measured in cpu + * cycles, have increased. Cacheline sizes have increased as well, which also + * helps B+Trees. + * + * Compared to radix trees, B+Trees are more efficient when dealing with a + * sparsely populated address space. Between 25% and 50% of the memory is + * occupied with valid pointers. When densely populated, radix trees contain + * ~98% pointers - hard to beat. Very sparse radix trees contain only ~2% + * pointers. + * + * This particular implementation stores pointers identified by a long value. + * Storing NULL pointers is illegal, lookup will return NULL when no entry + * was found. + * + * A tricks was used that is not commonly found in textbooks. The lowest + * values are to the right, not to the left. All used slots within a node + * are on the left, all unused slots contain NUL values. Most operations + * simply loop once over all slots and terminate on the first NUL. + */ + +#include <linux/btree.h> +#include <linux/cache.h> +#include <linux/kernel.h> +#include <linux/slab.h> +#include <linux/module.h> + +#define MAX(a, b) ((a) > (b) ? (a) : (b)) +#define NODESIZE MAX(L1_CACHE_BYTES, 128) + +struct btree_geo { + int keylen; + int no_pairs; + int no_longs; +}; + +struct btree_geo btree_geo32 = { + .keylen = 1, + .no_pairs = NODESIZE / sizeof(long) / 2, + .no_longs = NODESIZE / sizeof(long) / 2, +}; +EXPORT_SYMBOL_GPL(btree_geo32); + +#define LONG_PER_U64 (64 / BITS_PER_LONG) +struct btree_geo btree_geo64 = { + .keylen = LONG_PER_U64, + .no_pairs = NODESIZE / sizeof(long) / (1 + LONG_PER_U64), + .no_longs = LONG_PER_U64 * (NODESIZE / sizeof(long) / (1 + LONG_PER_U64)), +}; +EXPORT_SYMBOL_GPL(btree_geo64); + +struct btree_geo btree_geo128 = { + .keylen = 2 * LONG_PER_U64, + .no_pairs = NODESIZE / sizeof(long) / (1 + 2 * LONG_PER_U64), + .no_longs = 2 * LONG_PER_U64 * (NODESIZE / sizeof(long) / (1 + 2 * LONG_PER_U64)), +}; +EXPORT_SYMBOL_GPL(btree_geo128); + +static struct kmem_cache *btree_cachep; + +void *btree_alloc(gfp_t gfp_mask, void *pool_data) +{ + return kmem_cache_alloc(btree_cachep, gfp_mask); +} +EXPORT_SYMBOL_GPL(btree_alloc); + +void btree_free(void *element, void *pool_data) +{ + kmem_cache_free(btree_cachep, element); +} +EXPORT_SYMBOL_GPL(btree_free); + +static unsigned long *btree_node_alloc(struct btree_head *head, gfp_t gfp) +{ + unsigned long *node; + + node = mempool_alloc(head->mempool, gfp); + memset(node, 0, NODESIZE); + return node; +} + +static int longcmp(const unsigned long *l1, const unsigned long *l2, size_t n) +{ + size_t i; + + for (i = 0; i < n; i++) { + if (l1[i] < l2[i]) + return -1; + if (l1[i] > l2[i]) + return 1; + } + return 0; +} + +static unsigned long *longcpy(unsigned long *dest, const unsigned long *src, + size_t n) +{ + size_t i; + + for (i = 0; i < n; i++) + dest[i] = src[i]; + return dest; +} + +static unsigned long *longset(unsigned long *s, unsigned long c, size_t n) +{ + size_t i; + + for (i = 0; i < n; i++) + s[i] = c; + return s; +} + +static void dec_key(struct btree_geo *geo, unsigned long *key) +{ + unsigned long val; + int i; + + for (i = geo->keylen - 1; i >= 0; i--) { + val = key[i]; + key[i] = val - 1; + if (val) + break; + } +} + +static unsigned long *bkey(struct btree_geo *geo, unsigned long *node, int n) +{ + return &node[n * geo->keylen]; +} + +static void *bval(struct btree_geo *geo, unsigned long *node, int n) +{ + return (void *)node[geo->no_longs + n]; +} + +static void setkey(struct btree_geo *geo, unsigned long *node, int n, + unsigned long *key) +{ + longcpy(bkey(geo, node, n), key, geo->keylen); +} + +static void setval(struct btree_geo *geo, unsigned long *node, int n, + void *val) +{ + node[geo->no_longs + n] = (unsigned long) val; +} + +static void clearpair(struct btree_geo *geo, unsigned long *node, int n) +{ + longset(bkey(geo, node, n), 0, geo->keylen); + node[geo->no_longs + n] = 0; +} + +static inline void __btree_init(struct btree_head *head) +{ + head->node = NULL; + head->height = 0; +} + +void btree_init_mempool(struct btree_head *head, mempool_t *mempool) +{ + __btree_init(head); + head->mempool = mempool; +} +EXPORT_SYMBOL_GPL(btree_init_mempool); + +int btree_init(struct btree_head *head) +{ + __btree_init(head); + head->mempool = mempool_create(0, btree_alloc, btree_free, NULL); + if (!head->mempool) + return -ENOMEM; + return 0; +} +EXPORT_SYMBOL_GPL(btree_init); + +void btree_destroy(struct btree_head *head) +{ + mempool_destroy(head->mempool); + head->mempool = NULL; +} +EXPORT_SYMBOL_GPL(btree_destroy); + +void *btree_last(struct btree_head *head, struct btree_geo *geo, + unsigned long *key) +{ + int height = head->height; + unsigned long *node = head->node; + + if (height == 0) + return NULL; + + for ( ; height > 1; height--) + node = bval(geo, node, 0); + + longcpy(key, bkey(geo, node, 0), geo->keylen); + return bval(geo, node, 0); +} +EXPORT_SYMBOL_GPL(btree_last); + +static int keycmp(struct btree_geo *geo, unsigned long *node, int pos, + unsigned long *key) +{ + return longcmp(bkey(geo, node, pos), key, geo->keylen); +} + +static int keyzero(struct btree_geo *geo, unsigned long *key) +{ + int i; + + for (i = 0; i < geo->keylen; i++) + if (key[i]) + return 0; + + return 1; +} + +void *btree_lookup(struct btree_head *head, struct btree_geo *geo, + unsigned long *key) +{ + int i, height = head->height; + unsigned long *node = head->node; + + if (height == 0) + return NULL; + + for ( ; height > 1; height--) { + for (i = 0; i < geo->no_pairs; i++) + if (keycmp(geo, node, i, key) <= 0) + break; + if (i == geo->no_pairs) + return NULL; + node = bval(geo, node, i); + if (!node) + return NULL; + } + + if (!node) + return NULL; + + for (i = 0; i < geo->no_pairs; i++) + if (keycmp(geo, node, i, key) == 0) + return bval(geo, node, i); + return NULL; +} +EXPORT_SYMBOL_GPL(btree_lookup); + +int btree_update(struct btree_head *head, struct btree_geo *geo, + unsigned long *key, void *val) +{ + int i, height = head->height; + unsigned long *node = head->node; + + if (height == 0) + return -ENOENT; + + for ( ; height > 1; height--) { + for (i = 0; i < geo->no_pairs; i++) + if (keycmp(geo, node, i, key) <= 0) + break; + if (i == geo->no_pairs) + return -ENOENT; + node = bval(geo, node, i); + if (!node) + return -ENOENT; + } + + if (!node) + return -ENOENT; + + for (i = 0; i < geo->no_pairs; i++) + if (keycmp(geo, node, i, key) == 0) { + setval(geo, node, i, val); + return 0; + } + return -ENOENT; +} +EXPORT_SYMBOL_GPL(btree_update); + +/* + * Usually this function is quite similar to normal lookup. But the key of + * a parent node may be smaller than the smallest key of all its siblings. + * In such a case we cannot just return NULL, as we have only proven that no + * key smaller than __key, but larger than this parent key exists. + * So we set __key to the parent key and retry. We have to use the smallest + * such parent key, which is the last parent key we encountered. + */ +void *btree_get_prev(struct btree_head *head, struct btree_geo *geo, + unsigned long *__key) +{ + int i, height; + unsigned long *node, *oldnode; + unsigned long *retry_key = NULL, key[geo->keylen]; + + if (keyzero(geo, __key)) + return NULL; + + if (head->height == 0) + return NULL; +retry: + longcpy(key, __key, geo->keylen); + dec_key(geo, key); + + node = head->node; + for (height = head->height ; height > 1; height--) { + for (i = 0; i < geo->no_pairs; i++) + if (keycmp(geo, node, i, key) <= 0) + break; + if (i == geo->no_pairs) + goto miss; + oldnode = node; + node = bval(geo, node, i); + if (!node) + goto miss; + retry_key = bkey(geo, oldnode, i); + } + + if (!node) + goto miss; + + for (i = 0; i < geo->no_pairs; i++) { + if (keycmp(geo, node, i, key) <= 0) { + if (bval(geo, node, i)) { + longcpy(__key, bkey(geo, node, i), geo->keylen); + return bval(geo, node, i); + } else + goto miss; + } + } +miss: + if (retry_key) { + __key = retry_key; + retry_key = NULL; + goto retry; + } + return NULL; +} + +static int getpos(struct btree_geo *geo, unsigned long *node, + unsigned long *key) +{ + int i; + + for (i = 0; i < geo->no_pairs; i++) { + if (keycmp(geo, node, i, key) <= 0) + break; + } + return i; +} + +static int getfill(struct btree_geo *geo, unsigned long *node, int start) +{ + int i; + + for (i = start; i < geo->no_pairs; i++) + if (!bval(geo, node, i)) + break; + return i; +} + +/* + * locate the correct leaf node in the btree + */ +static unsigned long *find_level(struct btree_head *head, struct btree_geo *geo, + unsigned long *key, int level) +{ + unsigned long *node = head->node; + int i, height; + + for (height = head->height; height > level; height--) { + for (i = 0; i < geo->no_pairs; i++) + if (keycmp(geo, node, i, key) <= 0) + break; + + if ((i == geo->no_pairs) || !bval(geo, node, i)) { + /* right-most key is too large, update it */ + /* FIXME: If the right-most key on higher levels is + * always zero, this wouldn't be necessary. */ + i--; + setkey(geo, node, i, key); + } + BUG_ON(i < 0); + node = bval(geo, node, i); + } + BUG_ON(!node); + return node; +} + +static int btree_grow(struct btree_head *head, struct btree_geo *geo, + gfp_t gfp) +{ + unsigned long *node; + int fill; + + node = btree_node_alloc(head, gfp); + if (!node) + return -ENOMEM; + if (head->node) { + fill = getfill(geo, head->node, 0); + setkey(geo, node, 0, bkey(geo, head->node, fill - 1)); + setval(geo, node, 0, head->node); + } + head->node = node; + head->height++; + return 0; +} + +static void btree_shrink(struct btree_head *head, struct btree_geo *geo) +{ + unsigned long *node; + int fill; + + if (head->height <= 1) + return; + + node = head->node; + fill = getfill(geo, node, 0); + BUG_ON(fill > 1); + head->node = bval(geo, node, 0); + head->height--; + mempool_free(node, head->mempool); +} + +static int btree_insert_level(struct btree_head *head, struct btree_geo *geo, + unsigned long *key, void *val, int level, + gfp_t gfp) +{ + unsigned long *node; + int i, pos, fill, err; + + BUG_ON(!val); + if (head->height < level) { + err = btree_grow(head, geo, gfp); + if (err) + return err; + } + +retry: + node = find_level(head, geo, key, level); + pos = getpos(geo, node, key); + fill = getfill(geo, node, pos); + /* two identical keys are not allowed */ + BUG_ON(pos < fill && keycmp(geo, node, pos, key) == 0); + + if (fill == geo->no_pairs) { + /* need to split node */ + unsigned long *new; + + new = btree_node_alloc(head, gfp); + if (!new) + return -ENOMEM; + err = btree_insert_level(head, geo, + bkey(geo, node, fill / 2 - 1), + new, level + 1, gfp); + if (err) { + mempool_free(new, head->mempool); + return err; + } + for (i = 0; i < fill / 2; i++) { + setkey(geo, new, i, bkey(geo, node, i)); + setval(geo, new, i, bval(geo, node, i)); + setkey(geo, node, i, bkey(geo, node, i + fill / 2)); + setval(geo, node, i, bval(geo, node, i + fill / 2)); + clearpair(geo, node, i + fill / 2); + } + if (fill & 1) { + setkey(geo, node, i, bkey(geo, node, fill - 1)); + setval(geo, node, i, bval(geo, node, fill - 1)); + clearpair(geo, node, fill - 1); + } + goto retry; + } + BUG_ON(fill >= geo->no_pairs); + + /* shift and insert */ + for (i = fill; i > pos; i--) { + setkey(geo, node, i, bkey(geo, node, i - 1)); + setval(geo, node, i, bval(geo, node, i - 1)); + } + setkey(geo, node, pos, key); + setval(geo, node, pos, val); + + return 0; +} + +int btree_insert(struct btree_head *head, struct btree_geo *geo, + unsigned long *key, void *val, gfp_t gfp) +{ + return btree_insert_level(head, geo, key, val, 1, gfp); +} +EXPORT_SYMBOL_GPL(btree_insert); + +static void *btree_remove_level(struct btree_head *head, struct btree_geo *geo, + unsigned long *key, int level); +static void merge(struct btree_head *head, struct btree_geo *geo, int level, + unsigned long *left, int lfill, + unsigned long *right, int rfill, + unsigned long *parent, int lpos) +{ + int i; + + for (i = 0; i < rfill; i++) { + /* Move all keys to the left */ + setkey(geo, left, lfill + i, bkey(geo, right, i)); + setval(geo, left, lfill + i, bval(geo, right, i)); + } + /* Exchange left and right child in parent */ + setval(geo, parent, lpos, right); + setval(geo, parent, lpos + 1, left); + /* Remove left (formerly right) child from parent */ + btree_remove_level(head, geo, bkey(geo, parent, lpos), level + 1); + mempool_free(right, head->mempool); +} + +static void rebalance(struct btree_head *head, struct btree_geo *geo, + unsigned long *key, int level, unsigned long *child, int fill) +{ + unsigned long *parent, *left = NULL, *right = NULL; + int i, no_left, no_right; + + if (fill == 0) { + /* Because we don't steal entries from a neigbour, this case + * can happen. Parent node contains a single child, this + * node, so merging with a sibling never happens. + */ + btree_remove_level(head, geo, key, level + 1); + mempool_free(child, head->mempool); + return; + } + + parent = find_level(head, geo, key, level + 1); + i = getpos(geo, parent, key); + BUG_ON(bval(geo, parent, i) != child); + + if (i > 0) { + left = bval(geo, parent, i - 1); + no_left = getfill(geo, left, 0); + if (fill + no_left <= geo->no_pairs) { + merge(head, geo, level, + left, no_left, + child, fill, + parent, i - 1); + return; + } + } + if (i + 1 < getfill(geo, parent, i)) { + right = bval(geo, parent, i + 1); + no_right = getfill(geo, right, 0); + if (fill + no_right <= geo->no_pairs) { + merge(head, geo, level, + child, fill, + right, no_right, + parent, i); + return; + } + } + /* + * We could also try to steal one entry from the left or right + * neighbor. By not doing so we changed the invariant from + * "all nodes are at least half full" to "no two neighboring + * nodes can be merged". Which means that the average fill of + * all nodes is still half or better. + */ +} + +static void *btree_remove_level(struct btree_head *head, struct btree_geo *geo, + unsigned long *key, int level) +{ + unsigned long *node; + int i, pos, fill; + void *ret; + + if (level > head->height) { + /* we recursed all the way up */ + head->height = 0; + head->node = NULL; + return NULL; + } + + node = find_level(head, geo, key, level); + pos = getpos(geo, node, key); + fill = getfill(geo, node, pos); + if ((level == 1) && (keycmp(geo, node, pos, key) != 0)) + return NULL; + ret = bval(geo, node, pos); + + /* remove and shift */ + for (i = pos; i < fill - 1; i++) { + setkey(geo, node, i, bkey(geo, node, i + 1)); + setval(geo, node, i, bval(geo, node, i + 1)); + } + clearpair(geo, node, fill - 1); + + if (fill - 1 < geo->no_pairs / 2) { + if (level < head->height) + rebalance(head, geo, key, level, node, fill - 1); + else if (fill - 1 == 1) + btree_shrink(head, geo); + } + + return ret; +} + +void *btree_remove(struct btree_head *head, struct btree_geo *geo, + unsigned long *key) +{ + if (head->height == 0) + return NULL; + + return btree_remove_level(head, geo, key, 1); +} +EXPORT_SYMBOL_GPL(btree_remove); + +int btree_merge(struct btree_head *target, struct btree_head *victim, + struct btree_geo *geo, gfp_t gfp) +{ + unsigned long key[geo->keylen]; + unsigned long dup[geo->keylen]; + void *val; + int err; + + BUG_ON(target == victim); + + if (!(target->node)) { + /* target is empty, just copy fields over */ + target->node = victim->node; + target->height = victim->height; + __btree_init(victim); + return 0; + } + + /* TODO: This needs some optimizations. Currently we do three tree + * walks to remove a single object from the victim. + */ + for (;;) { + if (!btree_last(victim, geo, key)) + break; + val = btree_lookup(victim, geo, key); + err = btree_insert(target, geo, key, val, gfp); + if (err) + return err; + /* We must make a copy of the key, as the original will get + * mangled inside btree_remove. */ + longcpy(dup, key, geo->keylen); + btree_remove(victim, geo, dup); + } + return 0; +} +EXPORT_SYMBOL_GPL(btree_merge); + +static size_t __btree_for_each(struct btree_head *head, struct btree_geo *geo, + unsigned long *node, unsigned long opaque, + void (*func)(void *elem, unsigned long opaque, + unsigned long *key, size_t index, + void *func2), + void *func2, int reap, int height, size_t count) +{ + int i; + unsigned long *child; + + for (i = 0; i < geo->no_pairs; i++) { + child = bval(geo, node, i); + if (!child) + break; + if (height > 1) + count = __btree_for_each(head, geo, child, opaque, + func, func2, reap, height - 1, count); + else + func(child, opaque, bkey(geo, node, i), count++, + func2); + } + if (reap) + mempool_free(node, head->mempool); + return count; +} + +static void empty(void *elem, unsigned long opaque, unsigned long *key, + size_t index, void *func2) +{ +} + +void visitorl(void *elem, unsigned long opaque, unsigned long *key, + size_t index, void *__func) +{ + visitorl_t func = __func; + + func(elem, opaque, *key, index); +} +EXPORT_SYMBOL_GPL(visitorl); + +void visitor32(void *elem, unsigned long opaque, unsigned long *__key, + size_t index, void *__func) +{ + visitor32_t func = __func; + u32 *key = (void *)__key; + + func(elem, opaque, *key, index); +} +EXPORT_SYMBOL_GPL(visitor32); + +void visitor64(void *elem, unsigned long opaque, unsigned long *__key, + size_t index, void *__func) +{ + visitor64_t func = __func; + u64 *key = (void *)__key; + + func(elem, opaque, *key, index); +} +EXPORT_SYMBOL_GPL(visitor64); + +void visitor128(void *elem, unsigned long opaque, unsigned long *__key, + size_t index, void *__func) +{ + visitor128_t func = __func; + u64 *key = (void *)__key; + + func(elem, opaque, key[0], key[1], index); +} +EXPORT_SYMBOL_GPL(visitor128); + +size_t btree_visitor(struct btree_head *head, struct btree_geo *geo, + unsigned long opaque, + void (*func)(void *elem, unsigned long opaque, + unsigned long *key, + size_t index, void *func2), + void *func2) +{ + size_t count = 0; + + if (!func2) + func = empty; + if (head->node) + count = __btree_for_each(head, geo, head->node, opaque, func, + func2, 0, head->height, 0); + return count; +} +EXPORT_SYMBOL_GPL(btree_visitor); + +size_t btree_grim_visitor(struct btree_head *head, struct btree_geo *geo, + unsigned long opaque, + void (*func)(void *elem, unsigned long opaque, + unsigned long *key, + size_t index, void *func2), + void *func2) +{ + size_t count = 0; + + if (!func2) + func = empty; + if (head->node) + count = __btree_for_each(head, geo, head->node, opaque, func, + func2, 1, head->height, 0); + __btree_init(head); + return count; +} +EXPORT_SYMBOL_GPL(btree_grim_visitor); + +static int __init btree_module_init(void) +{ + btree_cachep = kmem_cache_create("btree_node", NODESIZE, 0, + SLAB_HWCACHE_ALIGN, NULL); + return 0; +} + +static void __exit btree_module_exit(void) +{ + kmem_cache_destroy(btree_cachep); +} + +/* If core code starts using btree, initialization should happen even earlier */ +module_init(btree_module_init); +module_exit(btree_module_exit); + +MODULE_AUTHOR("Joern Engel <joern@logfs.org>"); +MODULE_AUTHOR("Johannes Berg <johannes@sipsolutions.net>"); +MODULE_LICENSE("GPL"); diff --git a/lib/crc32.c b/lib/crc32.c index 02e3b31b3a79..0f45fbff34cb 100644 --- a/lib/crc32.c +++ b/lib/crc32.c @@ -30,11 +30,15 @@ #include <asm/atomic.h> #include "crc32defs.h" #if CRC_LE_BITS == 8 -#define tole(x) __constant_cpu_to_le32(x) -#define tobe(x) __constant_cpu_to_be32(x) +# define tole(x) __constant_cpu_to_le32(x) #else -#define tole(x) (x) -#define tobe(x) (x) +# define tole(x) (x) +#endif + +#if CRC_BE_BITS == 8 +# define tobe(x) __constant_cpu_to_be32(x) +#else +# define tobe(x) (x) #endif #include "crc32table.h" @@ -52,20 +56,19 @@ crc32_body(u32 crc, unsigned char const *buf, size_t len, const u32 *tab) # else # define DO_CRC(x) crc = tab[((crc >> 24) ^ (x)) & 255] ^ (crc << 8) # endif - const u32 *b = (const u32 *)buf; + const u32 *b; size_t rem_len; /* Align it */ - if (unlikely((long)b & 3 && len)) { - u8 *p = (u8 *)b; + if (unlikely((long)buf & 3 && len)) { do { - DO_CRC(*p++); - } while ((--len) && ((long)p)&3); - b = (u32 *)p; + DO_CRC(*buf++); + } while ((--len) && ((long)buf)&3); } rem_len = len & 3; /* load data 32 bits wide, xor data 32 bits wide. */ len = len >> 2; + b = (const u32 *)buf; for (--b; len; --len) { crc ^= *++b; /* use pre increment for speed */ DO_CRC(0); @@ -82,6 +85,7 @@ crc32_body(u32 crc, unsigned char const *buf, size_t len, const u32 *tab) } while (--len); } return crc; +#undef DO_CRC } #endif /** @@ -119,9 +123,6 @@ u32 __pure crc32_le(u32 crc, unsigned char const *p, size_t len) crc = __cpu_to_le32(crc); crc = crc32_body(crc, p, len, tab); return __le32_to_cpu(crc); -#undef ENDIAN_SHIFT -#undef DO_CRC - # elif CRC_LE_BITS == 4 while (len--) { crc ^= *p++; @@ -179,9 +180,6 @@ u32 __pure crc32_be(u32 crc, unsigned char const *p, size_t len) crc = __cpu_to_be32(crc); crc = crc32_body(crc, p, len, tab); return __be32_to_cpu(crc); -#undef ENDIAN_SHIFT -#undef DO_CRC - # elif CRC_BE_BITS == 4 while (len--) { crc ^= *p++ << 24; diff --git a/lib/list_sort.c b/lib/list_sort.c index 19d11e0bb958..4b5cb794c38b 100644 --- a/lib/list_sort.c +++ b/lib/list_sort.c @@ -4,99 +4,214 @@ #include <linux/slab.h> #include <linux/list.h> +#define MAX_LIST_LENGTH_BITS 20 + +/* + * Returns a list organized in an intermediate format suited + * to chaining of merge() calls: null-terminated, no reserved or + * sentinel head node, "prev" links not maintained. + */ +static struct list_head *merge(void *priv, + int (*cmp)(void *priv, struct list_head *a, + struct list_head *b), + struct list_head *a, struct list_head *b) +{ + struct list_head head, *tail = &head; + + while (a && b) { + /* if equal, take 'a' -- important for sort stability */ + if ((*cmp)(priv, a, b) <= 0) { + tail->next = a; + a = a->next; + } else { + tail->next = b; + b = b->next; + } + tail = tail->next; + } + tail->next = a?:b; + return head.next; +} + +/* + * Combine final list merge with restoration of standard doubly-linked + * list structure. This approach duplicates code from merge(), but + * runs faster than the tidier alternatives of either a separate final + * prev-link restoration pass, or maintaining the prev links + * throughout. + */ +static void merge_and_restore_back_links(void *priv, + int (*cmp)(void *priv, struct list_head *a, + struct list_head *b), + struct list_head *head, + struct list_head *a, struct list_head *b) +{ + struct list_head *tail = head; + + while (a && b) { + /* if equal, take 'a' -- important for sort stability */ + if ((*cmp)(priv, a, b) <= 0) { + tail->next = a; + a->prev = tail; + a = a->next; + } else { + tail->next = b; + b->prev = tail; + b = b->next; + } + tail = tail->next; + } + tail->next = a ? : b; + + do { + /* + * In worst cases this loop may run many iterations. + * Continue callbacks to the client even though no + * element comparison is needed, so the client's cmp() + * routine can invoke cond_resched() periodically. + */ + (*cmp)(priv, tail, tail); + + tail->next->prev = tail; + tail = tail->next; + } while (tail->next); + + tail->next = head; + head->prev = tail; +} + /** - * list_sort - sort a list. - * @priv: private data, passed to @cmp + * list_sort - sort a list + * @priv: private data, opaque to list_sort(), passed to @cmp * @head: the list to sort * @cmp: the elements comparison function * - * This function has been implemented by Mark J Roberts <mjr@znex.org>. It - * implements "merge sort" which has O(nlog(n)) complexity. The list is sorted - * in ascending order. + * This function implements "merge sort", which has O(nlog(n)) + * complexity. * - * The comparison function @cmp is supposed to return a negative value if @a is - * less than @b, and a positive value if @a is greater than @b. If @a and @b - * are equivalent, then it does not matter what this function returns. + * The comparison function @cmp must return a negative value if @a + * should sort before @b, and a positive value if @a should sort after + * @b. If @a and @b are equivalent, and their original relative + * ordering is to be preserved, @cmp must return 0. */ void list_sort(void *priv, struct list_head *head, - int (*cmp)(void *priv, struct list_head *a, - struct list_head *b)) + int (*cmp)(void *priv, struct list_head *a, + struct list_head *b)) { - struct list_head *p, *q, *e, *list, *tail, *oldhead; - int insize, nmerges, psize, qsize, i; + struct list_head *part[MAX_LIST_LENGTH_BITS+1]; /* sorted partial lists + -- last slot is a sentinel */ + int lev; /* index into part[] */ + int max_lev = 0; + struct list_head *list; if (list_empty(head)) return; + memset(part, 0, sizeof(part)); + + head->prev->next = NULL; list = head->next; - list_del(head); - insize = 1; - for (;;) { - p = oldhead = list; - list = tail = NULL; - nmerges = 0; - - while (p) { - nmerges++; - q = p; - psize = 0; - for (i = 0; i < insize; i++) { - psize++; - q = q->next == oldhead ? NULL : q->next; - if (!q) - break; - } - qsize = insize; - while (psize > 0 || (qsize > 0 && q)) { - if (!psize) { - e = q; - q = q->next; - qsize--; - if (q == oldhead) - q = NULL; - } else if (!qsize || !q) { - e = p; - p = p->next; - psize--; - if (p == oldhead) - p = NULL; - } else if (cmp(priv, p, q) <= 0) { - e = p; - p = p->next; - psize--; - if (p == oldhead) - p = NULL; - } else { - e = q; - q = q->next; - qsize--; - if (q == oldhead) - q = NULL; - } - if (tail) - tail->next = e; - else - list = e; - e->prev = tail; - tail = e; + while (list) { + struct list_head *cur = list; + list = list->next; + cur->next = NULL; + + for (lev = 0; part[lev]; lev++) { + cur = merge(priv, cmp, part[lev], cur); + part[lev] = NULL; + } + if (lev > max_lev) { + if (unlikely(lev >= ARRAY_SIZE(part)-1)) { + printk_once(KERN_DEBUG "list passed to" + " list_sort() too long for" + " efficiency\n"); + lev--; } - p = q; + max_lev = lev; } + part[lev] = cur; + } - tail->next = list; - list->prev = tail; + for (lev = 0; lev < max_lev; lev++) + if (part[lev]) + list = merge(priv, cmp, part[lev], list); - if (nmerges <= 1) - break; + merge_and_restore_back_links(priv, cmp, head, part[max_lev], list); +} +EXPORT_SYMBOL(list_sort); - insize *= 2; - } +#ifdef DEBUG_LIST_SORT +struct debug_el { + struct list_head l_h; + int value; + unsigned serial; +}; - head->next = list; - head->prev = list->prev; - list->prev->next = head; - list->prev = head; +static int cmp(void *priv, struct list_head *a, struct list_head *b) +{ + return container_of(a, struct debug_el, l_h)->value + - container_of(b, struct debug_el, l_h)->value; } -EXPORT_SYMBOL(list_sort); +/* + * The pattern of set bits in the list length determines which cases + * are hit in list_sort(). + */ +#define LIST_SORT_TEST_LENGTH (512+128+2) /* not including head */ + +static int __init list_sort_test(void) +{ + int i, r = 1, count; + struct list_head *head = kmalloc(sizeof(*head), GFP_KERNEL); + struct list_head *cur; + + printk(KERN_WARNING "testing list_sort()\n"); + + cur = head; + for (i = 0; i < LIST_SORT_TEST_LENGTH; i++) { + struct debug_el *el = kmalloc(sizeof(*el), GFP_KERNEL); + BUG_ON(!el); + /* force some equivalencies */ + el->value = (r = (r * 725861) % 6599) % (LIST_SORT_TEST_LENGTH/3); + el->serial = i; + + el->l_h.prev = cur; + cur->next = &el->l_h; + cur = cur->next; + } + head->prev = cur; + + list_sort(NULL, head, cmp); + + count = 1; + for (cur = head->next; cur->next != head; cur = cur->next) { + struct debug_el *el = container_of(cur, struct debug_el, l_h); + int cmp_result = cmp(NULL, cur, cur->next); + if (cur->next->prev != cur) { + printk(KERN_EMERG "list_sort() returned " + "a corrupted list!\n"); + return 1; + } else if (cmp_result > 0) { + printk(KERN_EMERG "list_sort() failed to sort!\n"); + return 1; + } else if (cmp_result == 0 && + el->serial >= container_of(cur->next, + struct debug_el, l_h)->serial) { + printk(KERN_EMERG "list_sort() failed to preserve order" + " of equivalent elements!\n"); + return 1; + } + kfree(cur->prev); + count++; + } + kfree(cur); + if (count != LIST_SORT_TEST_LENGTH) { + printk(KERN_EMERG "list_sort() returned list of" + "different length!\n"); + return 1; + } + return 0; +} +module_init(list_sort_test); +#endif diff --git a/lib/show_mem.c b/lib/show_mem.c index 238e72a18ce1..fdc77c82f922 100644 --- a/lib/show_mem.c +++ b/lib/show_mem.c @@ -15,7 +15,7 @@ void show_mem(void) unsigned long total = 0, reserved = 0, shared = 0, nonshared = 0, highmem = 0; - printk(KERN_INFO "Mem-Info:\n"); + printk("Mem-Info:\n"); show_free_areas(); for_each_online_pgdat(pgdat) { @@ -49,15 +49,15 @@ void show_mem(void) pgdat_resize_unlock(pgdat, &flags); } - printk(KERN_INFO "%lu pages RAM\n", total); + printk("%lu pages RAM\n", total); #ifdef CONFIG_HIGHMEM - printk(KERN_INFO "%lu pages HighMem\n", highmem); + printk("%lu pages HighMem\n", highmem); #endif - printk(KERN_INFO "%lu pages reserved\n", reserved); - printk(KERN_INFO "%lu pages shared\n", shared); - printk(KERN_INFO "%lu pages non-shared\n", nonshared); + printk("%lu pages reserved\n", reserved); + printk("%lu pages shared\n", shared); + printk("%lu pages non-shared\n", nonshared); #ifdef CONFIG_QUICKLIST - printk(KERN_INFO "%lu pages in pagetable cache\n", + printk("%lu pages in pagetable cache\n", quicklist_total_size()); #endif } diff --git a/lib/string.c b/lib/string.c index a1cdcfcc42d0..f71bead1be3e 100644 --- a/lib/string.c +++ b/lib/string.c @@ -36,25 +36,21 @@ int strnicmp(const char *s1, const char *s2, size_t len) /* Yes, Virginia, it had better be unsigned */ unsigned char c1, c2; - c1 = c2 = 0; - if (len) { - do { - c1 = *s1; - c2 = *s2; - s1++; - s2++; - if (!c1) - break; - if (!c2) - break; - if (c1 == c2) - continue; - c1 = tolower(c1); - c2 = tolower(c2); - if (c1 != c2) - break; - } while (--len); - } + if (!len) + return 0; + + do { + c1 = *s1++; + c2 = *s2++; + if (!c1 || !c2) + break; + if (c1 == c2) + continue; + c1 = tolower(c1); + c2 = tolower(c2); + if (c1 != c2) + break; + } while (--len); return (int)c1 - (int)c2; } EXPORT_SYMBOL(strnicmp); @@ -693,13 +689,13 @@ EXPORT_SYMBOL(strstr); */ char *strnstr(const char *s1, const char *s2, size_t len) { - size_t l1 = len, l2; + size_t l2; l2 = strlen(s2); if (!l2) return (char *)s1; - while (l1 >= l2) { - l1--; + while (len >= l2) { + len--; if (!memcmp(s1, s2, l2)) return (char *)s1; s1++; diff --git a/lib/vsprintf.c b/lib/vsprintf.c index af4aaa6c36f3..0d461c7c14db 100644 --- a/lib/vsprintf.c +++ b/lib/vsprintf.c @@ -381,8 +381,8 @@ static noinline char *put_dec(char *buf, unsigned long long num) #define PLUS 4 /* show plus */ #define SPACE 8 /* space if plus */ #define LEFT 16 /* left justified */ -#define SMALL 32 /* Must be 32 == 0x20 */ -#define SPECIAL 64 /* 0x */ +#define SMALL 32 /* use lowercase in hex (must be 32 == 0x20) */ +#define SPECIAL 64 /* prefix hex with "0x", octal with "0" */ enum format_type { FORMAT_TYPE_NONE, /* Just a string part */ @@ -408,12 +408,12 @@ enum format_type { }; struct printf_spec { - enum format_type type; - int flags; /* flags to number() */ - int field_width; /* width of output field */ - int base; - int precision; /* # of digits/chars */ - int qualifier; + u16 type; + s16 field_width; /* width of output field */ + u8 flags; /* flags to number() */ + u8 base; + s8 precision; /* # of digits/chars */ + u8 qualifier; }; static char *number(char *buf, char *end, unsigned long long num, @@ -597,22 +597,29 @@ static char *resource_string(char *buf, char *end, struct resource *res, #ifndef MEM_RSRC_PRINTK_SIZE #define MEM_RSRC_PRINTK_SIZE 10 #endif - struct printf_spec hex_spec = { + static const struct printf_spec io_spec = { .base = 16, + .field_width = IO_RSRC_PRINTK_SIZE, .precision = -1, .flags = SPECIAL | SMALL | ZEROPAD, }; - struct printf_spec dec_spec = { + static const struct printf_spec mem_spec = { + .base = 16, + .field_width = MEM_RSRC_PRINTK_SIZE, + .precision = -1, + .flags = SPECIAL | SMALL | ZEROPAD, + }; + static const struct printf_spec dec_spec = { .base = 10, .precision = -1, .flags = 0, }; - struct printf_spec str_spec = { + static const struct printf_spec str_spec = { .field_width = -1, .precision = 10, .flags = LEFT, }; - struct printf_spec flag_spec = { + static const struct printf_spec flag_spec = { .base = 16, .precision = -1, .flags = SPECIAL | SMALL, @@ -628,35 +635,31 @@ static char *resource_string(char *buf, char *end, struct resource *res, 2*RSRC_BUF_SIZE + FLAG_BUF_SIZE + RAW_BUF_SIZE)]; char *p = sym, *pend = sym + sizeof(sym); - int size = -1, addr = 0; int decode = (fmt[0] == 'R') ? 1 : 0; - - if (res->flags & IORESOURCE_IO) { - size = IO_RSRC_PRINTK_SIZE; - addr = 1; - } else if (res->flags & IORESOURCE_MEM) { - size = MEM_RSRC_PRINTK_SIZE; - addr = 1; - } + const struct printf_spec *specp; *p++ = '['; - if (res->flags & IORESOURCE_IO) + if (res->flags & IORESOURCE_IO) { p = string(p, pend, "io ", str_spec); - else if (res->flags & IORESOURCE_MEM) + specp = &io_spec; + } else if (res->flags & IORESOURCE_MEM) { p = string(p, pend, "mem ", str_spec); - else if (res->flags & IORESOURCE_IRQ) + specp = &mem_spec; + } else if (res->flags & IORESOURCE_IRQ) { p = string(p, pend, "irq ", str_spec); - else if (res->flags & IORESOURCE_DMA) + specp = &dec_spec; + } else if (res->flags & IORESOURCE_DMA) { p = string(p, pend, "dma ", str_spec); - else { + specp = &dec_spec; + } else { p = string(p, pend, "??? ", str_spec); + specp = &mem_spec; decode = 0; } - hex_spec.field_width = size; - p = number(p, pend, res->start, addr ? hex_spec : dec_spec); + p = number(p, pend, res->start, *specp); if (res->start != res->end) { *p++ = '-'; - p = number(p, pend, res->end, addr ? hex_spec : dec_spec); + p = number(p, pend, res->end, *specp); } if (decode) { if (res->flags & IORESOURCE_MEM_64) @@ -1333,7 +1336,7 @@ int vsnprintf(char *buf, size_t size, const char *fmt, va_list args) break; case FORMAT_TYPE_NRCHARS: { - int qualifier = spec.qualifier; + u8 qualifier = spec.qualifier; if (qualifier == 'l') { long *ip = va_arg(args, long *); @@ -1619,7 +1622,7 @@ do { \ case FORMAT_TYPE_NRCHARS: { /* skip %n 's argument */ - int qualifier = spec.qualifier; + u8 qualifier = spec.qualifier; void *skip_arg; if (qualifier == 'l') skip_arg = va_arg(args, long *); @@ -1885,7 +1888,9 @@ int vsscanf(const char *buf, const char *fmt, va_list args) char *next; char digit; int num = 0; - int qualifier, base, field_width; + u8 qualifier; + u8 base; + s16 field_width; bool is_sign; while (*fmt && *str) { @@ -1963,7 +1968,7 @@ int vsscanf(const char *buf, const char *fmt, va_list args) { char *s = (char *)va_arg(args, char *); if (field_width == -1) - field_width = INT_MAX; + field_width = SHORT_MAX; /* first, skip leading white space in buffer */ str = skip_spaces(str); diff --git a/mm/fadvise.c b/mm/fadvise.c index e43359214f6f..8d723c9e8b75 100644 --- a/mm/fadvise.c +++ b/mm/fadvise.c @@ -77,12 +77,20 @@ SYSCALL_DEFINE(fadvise64_64)(int fd, loff_t offset, loff_t len, int advice) switch (advice) { case POSIX_FADV_NORMAL: file->f_ra.ra_pages = bdi->ra_pages; + spin_lock(&file->f_lock); + file->f_mode &= ~FMODE_RANDOM; + spin_unlock(&file->f_lock); break; case POSIX_FADV_RANDOM: - file->f_ra.ra_pages = 0; + spin_lock(&file->f_lock); + file->f_mode |= FMODE_RANDOM; + spin_unlock(&file->f_lock); break; case POSIX_FADV_SEQUENTIAL: file->f_ra.ra_pages = bdi->ra_pages * 2; + spin_lock(&file->f_lock); + file->f_mode &= ~FMODE_RANDOM; + spin_unlock(&file->f_lock); break; case POSIX_FADV_WILLNEED: if (!mapping->a_ops->readpage) { diff --git a/mm/filemap.c b/mm/filemap.c index 148b52a5bb7e..045b31c37653 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1986,7 +1986,7 @@ EXPORT_SYMBOL(iov_iter_single_seg_count); inline int generic_write_checks(struct file *file, loff_t *pos, size_t *count, int isblk) { struct inode *inode = file->f_mapping->host; - unsigned long limit = current->signal->rlim[RLIMIT_FSIZE].rlim_cur; + unsigned long limit = rlimit(RLIMIT_FSIZE); if (unlikely(*pos < 0)) return -EINVAL; diff --git a/mm/filemap_xip.c b/mm/filemap_xip.c index 1888b2d71bb8..78b94f0b6d5d 100644 --- a/mm/filemap_xip.c +++ b/mm/filemap_xip.c @@ -194,7 +194,7 @@ retry: flush_cache_page(vma, address, pte_pfn(*pte)); pteval = ptep_clear_flush_notify(vma, address, pte); page_remove_rmap(page); - dec_mm_counter(mm, file_rss); + dec_mm_counter(mm, MM_FILEPAGES); BUG_ON(pte_dirty(pteval)); pte_unmap_unlock(pte, ptl); page_cache_release(page); diff --git a/mm/fremap.c b/mm/fremap.c index b6ec85abbb39..46f5dacf90a2 100644 --- a/mm/fremap.c +++ b/mm/fremap.c @@ -40,7 +40,7 @@ static void zap_pte(struct mm_struct *mm, struct vm_area_struct *vma, page_remove_rmap(page); page_cache_release(page); update_hiwater_rss(mm); - dec_mm_counter(mm, file_rss); + dec_mm_counter(mm, MM_FILEPAGES); } } else { if (!pte_file(pte)) @@ -1563,10 +1563,12 @@ int page_referenced_ksm(struct page *page, struct mem_cgroup *memcg, again: hlist_for_each_entry(rmap_item, hlist, &stable_node->hlist, hlist) { struct anon_vma *anon_vma = rmap_item->anon_vma; + struct anon_vma_chain *vmac; struct vm_area_struct *vma; spin_lock(&anon_vma->lock); - list_for_each_entry(vma, &anon_vma->head, anon_vma_node) { + list_for_each_entry(vmac, &anon_vma->head, same_anon_vma) { + vma = vmac->vma; if (rmap_item->address < vma->vm_start || rmap_item->address >= vma->vm_end) continue; @@ -1614,10 +1616,12 @@ int try_to_unmap_ksm(struct page *page, enum ttu_flags flags) again: hlist_for_each_entry(rmap_item, hlist, &stable_node->hlist, hlist) { struct anon_vma *anon_vma = rmap_item->anon_vma; + struct anon_vma_chain *vmac; struct vm_area_struct *vma; spin_lock(&anon_vma->lock); - list_for_each_entry(vma, &anon_vma->head, anon_vma_node) { + list_for_each_entry(vmac, &anon_vma->head, same_anon_vma) { + vma = vmac->vma; if (rmap_item->address < vma->vm_start || rmap_item->address >= vma->vm_end) continue; @@ -1664,10 +1668,12 @@ int rmap_walk_ksm(struct page *page, int (*rmap_one)(struct page *, again: hlist_for_each_entry(rmap_item, hlist, &stable_node->hlist, hlist) { struct anon_vma *anon_vma = rmap_item->anon_vma; + struct anon_vma_chain *vmac; struct vm_area_struct *vma; spin_lock(&anon_vma->lock); - list_for_each_entry(vma, &anon_vma->head, anon_vma_node) { + list_for_each_entry(vmac, &anon_vma->head, same_anon_vma) { + vma = vmac->vma; if (rmap_item->address < vma->vm_start || rmap_item->address >= vma->vm_end) continue; diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 954032b80bed..d813823ab08f 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -2545,7 +2545,7 @@ static int mem_cgroup_force_empty_list(struct mem_cgroup *mem, pc = list_entry(list->prev, struct page_cgroup, lru); if (busy == pc) { list_move(&pc->lru, list); - busy = 0; + busy = NULL; spin_unlock_irqrestore(&zone->lru_lock, flags); continue; } diff --git a/mm/memory-failure.c b/mm/memory-failure.c index 17299fd4577c..d1f335162976 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -383,9 +383,12 @@ static void collect_procs_anon(struct page *page, struct list_head *to_kill, if (av == NULL) /* Not actually mapped anymore */ goto out; for_each_process (tsk) { + struct anon_vma_chain *vmac; + if (!task_early_kill(tsk)) continue; - list_for_each_entry (vma, &av->head, anon_vma_node) { + list_for_each_entry(vmac, &av->head, same_anon_vma) { + vma = vmac->vma; if (!page_mapped_in_vma(page, vma)) continue; if (vma->vm_mm == tsk->mm) diff --git a/mm/memory.c b/mm/memory.c index 72fb5f39bccc..d1153e37e9ba 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -121,6 +121,80 @@ static int __init init_zero_pfn(void) } core_initcall(init_zero_pfn); + +#if defined(SPLIT_RSS_COUNTING) + +void __sync_task_rss_stat(struct task_struct *task, struct mm_struct *mm) +{ + int i; + + for (i = 0; i < NR_MM_COUNTERS; i++) { + if (task->rss_stat.count[i]) { + add_mm_counter(mm, i, task->rss_stat.count[i]); + task->rss_stat.count[i] = 0; + } + } + task->rss_stat.events = 0; +} + +static void add_mm_counter_fast(struct mm_struct *mm, int member, int val) +{ + struct task_struct *task = current; + + if (likely(task->mm == mm)) + task->rss_stat.count[member] += val; + else + add_mm_counter(mm, member, val); +} +#define inc_mm_counter_fast(mm, member) add_mm_counter_fast(mm, member, 1) +#define dec_mm_counter_fast(mm, member) add_mm_counter_fast(mm, member, -1) + +/* sync counter once per 64 page faults */ +#define TASK_RSS_EVENTS_THRESH (64) +static void check_sync_rss_stat(struct task_struct *task) +{ + if (unlikely(task != current)) + return; + if (unlikely(task->rss_stat.events++ > TASK_RSS_EVENTS_THRESH)) + __sync_task_rss_stat(task, task->mm); +} + +unsigned long get_mm_counter(struct mm_struct *mm, int member) +{ + long val = 0; + + /* + * Don't use task->mm here...for avoiding to use task_get_mm().. + * The caller must guarantee task->mm is not invalid. + */ + val = atomic_long_read(&mm->rss_stat.count[member]); + /* + * counter is updated in asynchronous manner and may go to minus. + * But it's never be expected number for users. + */ + if (val < 0) + return 0; + return (unsigned long)val; +} + +void sync_mm_rss(struct task_struct *task, struct mm_struct *mm) +{ + __sync_task_rss_stat(task, mm); +} +#else + +#define inc_mm_counter_fast(mm, member) inc_mm_counter(mm, member) +#define dec_mm_counter_fast(mm, member) dec_mm_counter(mm, member) + +static void check_sync_rss_stat(struct task_struct *task) +{ +} + +void sync_mm_rss(struct task_struct *task, struct mm_struct *mm) +{ +} +#endif + /* * If a p?d_bad entry is found while walking page tables, report * the error, before resetting entry to p?d_none. Usually (but @@ -300,7 +374,7 @@ void free_pgtables(struct mmu_gather *tlb, struct vm_area_struct *vma, * Hide vma from rmap and truncate_pagecache before freeing * pgtables */ - anon_vma_unlink(vma); + unlink_anon_vmas(vma); unlink_file_vma(vma); if (is_vm_hugetlb_page(vma)) { @@ -314,7 +388,7 @@ void free_pgtables(struct mmu_gather *tlb, struct vm_area_struct *vma, && !is_vm_hugetlb_page(next)) { vma = next; next = vma->vm_next; - anon_vma_unlink(vma); + unlink_anon_vmas(vma); unlink_file_vma(vma); } free_pgd_range(tlb, addr, vma->vm_end, @@ -376,12 +450,20 @@ int __pte_alloc_kernel(pmd_t *pmd, unsigned long address) return 0; } -static inline void add_mm_rss(struct mm_struct *mm, int file_rss, int anon_rss) +static inline void init_rss_vec(int *rss) { - if (file_rss) - add_mm_counter(mm, file_rss, file_rss); - if (anon_rss) - add_mm_counter(mm, anon_rss, anon_rss); + memset(rss, 0, sizeof(int) * NR_MM_COUNTERS); +} + +static inline void add_mm_rss_vec(struct mm_struct *mm, int *rss) +{ + int i; + + if (current->mm == mm) + sync_mm_rss(current, mm); + for (i = 0; i < NR_MM_COUNTERS; i++) + if (rss[i]) + add_mm_counter(mm, i, rss[i]); } /* @@ -597,7 +679,9 @@ copy_one_pte(struct mm_struct *dst_mm, struct mm_struct *src_mm, &src_mm->mmlist); spin_unlock(&mmlist_lock); } - if (is_write_migration_entry(entry) && + if (likely(!non_swap_entry(entry))) + rss[MM_SWAPENTS]++; + else if (is_write_migration_entry(entry) && is_cow_mapping(vm_flags)) { /* * COW mappings require pages in both parent @@ -632,7 +716,10 @@ copy_one_pte(struct mm_struct *dst_mm, struct mm_struct *src_mm, if (page) { get_page(page); page_dup_rmap(page); - rss[PageAnon(page)]++; + if (PageAnon(page)) + rss[MM_ANONPAGES]++; + else + rss[MM_FILEPAGES]++; } out_set_pte: @@ -648,11 +735,12 @@ static int copy_pte_range(struct mm_struct *dst_mm, struct mm_struct *src_mm, pte_t *src_pte, *dst_pte; spinlock_t *src_ptl, *dst_ptl; int progress = 0; - int rss[2]; + int rss[NR_MM_COUNTERS]; swp_entry_t entry = (swp_entry_t){0}; again: - rss[1] = rss[0] = 0; + init_rss_vec(rss); + dst_pte = pte_alloc_map_lock(dst_mm, dst_pmd, addr, &dst_ptl); if (!dst_pte) return -ENOMEM; @@ -688,7 +776,7 @@ again: arch_leave_lazy_mmu_mode(); spin_unlock(src_ptl); pte_unmap_nested(orig_src_pte); - add_mm_rss(dst_mm, rss[0], rss[1]); + add_mm_rss_vec(dst_mm, rss); pte_unmap_unlock(orig_dst_pte, dst_ptl); cond_resched(); @@ -816,8 +904,9 @@ static unsigned long zap_pte_range(struct mmu_gather *tlb, struct mm_struct *mm = tlb->mm; pte_t *pte; spinlock_t *ptl; - int file_rss = 0; - int anon_rss = 0; + int rss[NR_MM_COUNTERS]; + + init_rss_vec(rss); pte = pte_offset_map_lock(mm, pmd, addr, &ptl); arch_enter_lazy_mmu_mode(); @@ -863,14 +952,14 @@ static unsigned long zap_pte_range(struct mmu_gather *tlb, set_pte_at(mm, addr, pte, pgoff_to_pte(page->index)); if (PageAnon(page)) - anon_rss--; + rss[MM_ANONPAGES]--; else { if (pte_dirty(ptent)) set_page_dirty(page); if (pte_young(ptent) && likely(!VM_SequentialReadHint(vma))) mark_page_accessed(page); - file_rss--; + rss[MM_FILEPAGES]--; } page_remove_rmap(page); if (unlikely(page_mapcount(page) < 0)) @@ -887,13 +976,18 @@ static unsigned long zap_pte_range(struct mmu_gather *tlb, if (pte_file(ptent)) { if (unlikely(!(vma->vm_flags & VM_NONLINEAR))) print_bad_pte(vma, addr, ptent, NULL); - } else if - (unlikely(!free_swap_and_cache(pte_to_swp_entry(ptent)))) - print_bad_pte(vma, addr, ptent, NULL); + } else { + swp_entry_t entry = pte_to_swp_entry(ptent); + + if (!non_swap_entry(entry)) + rss[MM_SWAPENTS]--; + if (unlikely(!free_swap_and_cache(entry))) + print_bad_pte(vma, addr, ptent, NULL); + } pte_clear_not_present_full(mm, addr, pte, tlb->fullmm); } while (pte++, addr += PAGE_SIZE, (addr != end && *zap_work > 0)); - add_mm_rss(mm, file_rss, anon_rss); + add_mm_rss_vec(mm, rss); arch_leave_lazy_mmu_mode(); pte_unmap_unlock(pte - 1, ptl); @@ -1527,7 +1621,7 @@ static int insert_page(struct vm_area_struct *vma, unsigned long addr, /* Ok, finally just insert the thing.. */ get_page(page); - inc_mm_counter(mm, file_rss); + inc_mm_counter_fast(mm, MM_FILEPAGES); page_add_file_rmap(page); set_pte_at(mm, addr, pte, mk_pte(page, prot)); @@ -2044,6 +2138,13 @@ static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma, page_cache_release(old_page); } reuse = reuse_swap_page(old_page); + if (reuse) + /* + * The page is all ours. Move it to our anon_vma so + * the rmap code will not search our parent or siblings. + * Protected against the rmap code by the page lock. + */ + page_move_anon_rmap(old_page, vma, address); unlock_page(old_page); } else if (unlikely((vma->vm_flags & (VM_WRITE|VM_SHARED)) == (VM_WRITE|VM_SHARED))) { @@ -2163,11 +2264,11 @@ gotten: if (likely(pte_same(*page_table, orig_pte))) { if (old_page) { if (!PageAnon(old_page)) { - dec_mm_counter(mm, file_rss); - inc_mm_counter(mm, anon_rss); + dec_mm_counter_fast(mm, MM_FILEPAGES); + inc_mm_counter_fast(mm, MM_ANONPAGES); } } else - inc_mm_counter(mm, anon_rss); + inc_mm_counter_fast(mm, MM_ANONPAGES); flush_cache_page(vma, address, pte_pfn(orig_pte)); entry = mk_pte(new_page, vma->vm_page_prot); entry = maybe_mkwrite(pte_mkdirty(entry), vma); @@ -2604,7 +2705,8 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma, * discarded at swap_free(). */ - inc_mm_counter(mm, anon_rss); + inc_mm_counter_fast(mm, MM_ANONPAGES); + dec_mm_counter_fast(mm, MM_SWAPENTS); pte = mk_pte(page, vma->vm_page_prot); if ((flags & FAULT_FLAG_WRITE) && reuse_swap_page(page)) { pte = maybe_mkwrite(pte_mkdirty(pte), vma); @@ -2688,7 +2790,7 @@ static int do_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma, if (!pte_none(*page_table)) goto release; - inc_mm_counter(mm, anon_rss); + inc_mm_counter_fast(mm, MM_ANONPAGES); page_add_new_anon_rmap(page, vma, address); setpte: set_pte_at(mm, address, page_table, entry); @@ -2842,10 +2944,10 @@ static int __do_fault(struct mm_struct *mm, struct vm_area_struct *vma, if (flags & FAULT_FLAG_WRITE) entry = maybe_mkwrite(pte_mkdirty(entry), vma); if (anon) { - inc_mm_counter(mm, anon_rss); + inc_mm_counter_fast(mm, MM_ANONPAGES); page_add_new_anon_rmap(page, vma, address); } else { - inc_mm_counter(mm, file_rss); + inc_mm_counter_fast(mm, MM_FILEPAGES); page_add_file_rmap(page); if (flags & FAULT_FLAG_WRITE) { dirty_page = page; @@ -3023,6 +3125,9 @@ int handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma, count_vm_event(PGFAULT); + /* do counter updates before entering really critical section. */ + check_sync_rss_stat(current); + if (unlikely(is_vm_hugetlb_page(vma))) return hugetlb_fault(mm, vma, address, flags); diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index 030ce8a5bb0e..78e34e63c7b8 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -28,6 +28,7 @@ #include <linux/pfn.h> #include <linux/suspend.h> #include <linux/mm_inline.h> +#include <linux/firmware-map.h> #include <asm/tlbflush.h> @@ -523,6 +524,9 @@ int __ref add_memory(int nid, u64 start, u64 size) BUG_ON(ret); } + /* create new memmap entry */ + firmware_map_add_hotplug(start, start + size, "System RAM"); + goto out; error: diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 290fb5bf0440..bda230e52acd 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -563,24 +563,50 @@ static int policy_vma(struct vm_area_struct *vma, struct mempolicy *new) } /* Step 2: apply policy to a range and do splits. */ -static int mbind_range(struct vm_area_struct *vma, unsigned long start, - unsigned long end, struct mempolicy *new) +static int mbind_range(struct mm_struct *mm, unsigned long start, + unsigned long end, struct mempolicy *new_pol) { struct vm_area_struct *next; - int err; + struct vm_area_struct *prev; + struct vm_area_struct *vma; + int err = 0; + pgoff_t pgoff; + unsigned long vmstart; + unsigned long vmend; - err = 0; - for (; vma && vma->vm_start < end; vma = next) { + vma = find_vma_prev(mm, start, &prev); + if (!vma || vma->vm_start > start) + return -EFAULT; + + for (; vma && vma->vm_start < end; prev = vma, vma = next) { next = vma->vm_next; - if (vma->vm_start < start) - err = split_vma(vma->vm_mm, vma, start, 1); - if (!err && vma->vm_end > end) - err = split_vma(vma->vm_mm, vma, end, 0); - if (!err) - err = policy_vma(vma, new); + vmstart = max(start, vma->vm_start); + vmend = min(end, vma->vm_end); + + pgoff = vma->vm_pgoff + ((start - vma->vm_start) >> PAGE_SHIFT); + prev = vma_merge(mm, prev, vmstart, vmend, vma->vm_flags, + vma->anon_vma, vma->vm_file, pgoff, new_pol); + if (prev) { + vma = prev; + next = vma->vm_next; + continue; + } + if (vma->vm_start != vmstart) { + err = split_vma(vma->vm_mm, vma, vmstart, 1); + if (err) + goto out; + } + if (vma->vm_end != vmend) { + err = split_vma(vma->vm_mm, vma, vmend, 0); + if (err) + goto out; + } + err = policy_vma(vma, new_pol); if (err) - break; + goto out; } + + out: return err; } @@ -862,36 +888,36 @@ int do_migrate_pages(struct mm_struct *mm, if (err) goto out; -/* - * Find a 'source' bit set in 'tmp' whose corresponding 'dest' - * bit in 'to' is not also set in 'tmp'. Clear the found 'source' - * bit in 'tmp', and return that <source, dest> pair for migration. - * The pair of nodemasks 'to' and 'from' define the map. - * - * If no pair of bits is found that way, fallback to picking some - * pair of 'source' and 'dest' bits that are not the same. If the - * 'source' and 'dest' bits are the same, this represents a node - * that will be migrating to itself, so no pages need move. - * - * If no bits are left in 'tmp', or if all remaining bits left - * in 'tmp' correspond to the same bit in 'to', return false - * (nothing left to migrate). - * - * This lets us pick a pair of nodes to migrate between, such that - * if possible the dest node is not already occupied by some other - * source node, minimizing the risk of overloading the memory on a - * node that would happen if we migrated incoming memory to a node - * before migrating outgoing memory source that same node. - * - * A single scan of tmp is sufficient. As we go, we remember the - * most recent <s, d> pair that moved (s != d). If we find a pair - * that not only moved, but what's better, moved to an empty slot - * (d is not set in tmp), then we break out then, with that pair. - * Otherwise when we finish scannng from_tmp, we at least have the - * most recent <s, d> pair that moved. If we get all the way through - * the scan of tmp without finding any node that moved, much less - * moved to an empty node, then there is nothing left worth migrating. - */ + /* + * Find a 'source' bit set in 'tmp' whose corresponding 'dest' + * bit in 'to' is not also set in 'tmp'. Clear the found 'source' + * bit in 'tmp', and return that <source, dest> pair for migration. + * The pair of nodemasks 'to' and 'from' define the map. + * + * If no pair of bits is found that way, fallback to picking some + * pair of 'source' and 'dest' bits that are not the same. If the + * 'source' and 'dest' bits are the same, this represents a node + * that will be migrating to itself, so no pages need move. + * + * If no bits are left in 'tmp', or if all remaining bits left + * in 'tmp' correspond to the same bit in 'to', return false + * (nothing left to migrate). + * + * This lets us pick a pair of nodes to migrate between, such that + * if possible the dest node is not already occupied by some other + * source node, minimizing the risk of overloading the memory on a + * node that would happen if we migrated incoming memory to a node + * before migrating outgoing memory source that same node. + * + * A single scan of tmp is sufficient. As we go, we remember the + * most recent <s, d> pair that moved (s != d). If we find a pair + * that not only moved, but what's better, moved to an empty slot + * (d is not set in tmp), then we break out then, with that pair. + * Otherwise when we finish scannng from_tmp, we at least have the + * most recent <s, d> pair that moved. If we get all the way through + * the scan of tmp without finding any node that moved, much less + * moved to an empty node, then there is nothing left worth migrating. + */ tmp = *from_nodes; while (!nodes_empty(tmp)) { @@ -1047,7 +1073,7 @@ static long do_mbind(unsigned long start, unsigned long len, if (!IS_ERR(vma)) { int nr_failed = 0; - err = mbind_range(vma, start, end, new); + err = mbind_range(mm, start, end, new); if (!list_empty(&pagelist)) nr_failed = migrate_pages(&pagelist, new_vma_page, diff --git a/mm/migrate.c b/mm/migrate.c index edb6101ed774..88000b89fc9a 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -275,8 +275,6 @@ static int migrate_page_move_mapping(struct address_space *mapping, */ static void migrate_page_copy(struct page *newpage, struct page *page) { - int anon; - copy_highpage(newpage, page); if (PageError(page)) @@ -313,8 +311,6 @@ static void migrate_page_copy(struct page *newpage, struct page *page) ClearPageSwapCache(page); ClearPagePrivate(page); set_page_private(page, 0); - /* page->mapping contains a flag for PageAnon() */ - anon = PageAnon(page); page->mapping = NULL; /* diff --git a/mm/mlock.c b/mm/mlock.c index 2b8335a89400..8f4e2dfceec1 100644 --- a/mm/mlock.c +++ b/mm/mlock.c @@ -25,7 +25,7 @@ int can_do_mlock(void) { if (capable(CAP_IPC_LOCK)) return 1; - if (current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur != 0) + if (rlimit(RLIMIT_MEMLOCK) != 0) return 1; return 0; } @@ -487,7 +487,7 @@ SYSCALL_DEFINE2(mlock, unsigned long, start, size_t, len) locked = len >> PAGE_SHIFT; locked += current->mm->locked_vm; - lock_limit = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur; + lock_limit = rlimit(RLIMIT_MEMLOCK); lock_limit >>= PAGE_SHIFT; /* check against resource limits */ @@ -550,7 +550,7 @@ SYSCALL_DEFINE1(mlockall, int, flags) down_write(¤t->mm->mmap_sem); - lock_limit = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur; + lock_limit = rlimit(RLIMIT_MEMLOCK); lock_limit >>= PAGE_SHIFT; ret = -ENOMEM; @@ -584,7 +584,7 @@ int user_shm_lock(size_t size, struct user_struct *user) int allowed = 0; locked = (size + PAGE_SIZE - 1) >> PAGE_SHIFT; - lock_limit = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur; + lock_limit = rlimit(RLIMIT_MEMLOCK); if (lock_limit == RLIM_INFINITY) allowed = 1; lock_limit >>= PAGE_SHIFT; @@ -618,12 +618,12 @@ int account_locked_memory(struct mm_struct *mm, struct rlimit *rlim, down_write(&mm->mmap_sem); - lim = rlim[RLIMIT_AS].rlim_cur >> PAGE_SHIFT; + lim = ACCESS_ONCE(rlim[RLIMIT_AS].rlim_cur) >> PAGE_SHIFT; vm = mm->total_vm + pgsz; if (lim < vm) goto out; - lim = rlim[RLIMIT_MEMLOCK].rlim_cur >> PAGE_SHIFT; + lim = ACCESS_ONCE(rlim[RLIMIT_MEMLOCK].rlim_cur) >> PAGE_SHIFT; vm = mm->locked_vm + pgsz; if (lim < vm) goto out; diff --git a/mm/mmap.c b/mm/mmap.c index ee2298936fe6..f1b4448626bf 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -265,7 +265,7 @@ SYSCALL_DEFINE1(brk, unsigned long, brk) * segment grow beyond its set limit the in case where the limit is * not page aligned -Ram Gupta */ - rlim = current->signal->rlim[RLIMIT_DATA].rlim_cur; + rlim = rlimit(RLIMIT_DATA); if (rlim < RLIM_INFINITY && (brk - mm->start_brk) + (mm->end_data - mm->start_data) > rlim) goto out; @@ -437,7 +437,6 @@ __vma_link(struct mm_struct *mm, struct vm_area_struct *vma, { __vma_link_list(mm, vma, prev, rb_parent); __vma_link_rb(mm, vma, rb_link, rb_parent); - __anon_vma_link(vma); } static void vma_link(struct mm_struct *mm, struct vm_area_struct *vma, @@ -499,7 +498,7 @@ __vma_unlink(struct mm_struct *mm, struct vm_area_struct *vma, * are necessary. The "insert" vma (if any) is to be inserted * before we drop the necessary locks. */ -void vma_adjust(struct vm_area_struct *vma, unsigned long start, +int vma_adjust(struct vm_area_struct *vma, unsigned long start, unsigned long end, pgoff_t pgoff, struct vm_area_struct *insert) { struct mm_struct *mm = vma->vm_mm; @@ -542,6 +541,26 @@ again: remove_next = 1 + (end > next->vm_end); } } + /* + * When changing only vma->vm_end, we don't really need anon_vma lock. + */ + if (vma->anon_vma && (insert || importer || start != vma->vm_start)) + anon_vma = vma->anon_vma; + if (anon_vma) { + /* + * Easily overlooked: when mprotect shifts the boundary, + * make sure the expanding vma has anon_vma set if the + * shrinking vma had, to cover any anon pages imported. + */ + if (importer && !importer->anon_vma) { + /* Block reverse map lookups until things are set up. */ + if (anon_vma_clone(importer, vma)) { + return -ENOMEM; + } + importer->anon_vma = anon_vma; + } + } + if (file) { mapping = file->f_mapping; if (!(vma->vm_flags & VM_NONLINEAR)) @@ -567,25 +586,6 @@ again: remove_next = 1 + (end > next->vm_end); } } - /* - * When changing only vma->vm_end, we don't really need - * anon_vma lock. - */ - if (vma->anon_vma && (insert || importer || start != vma->vm_start)) - anon_vma = vma->anon_vma; - if (anon_vma) { - spin_lock(&anon_vma->lock); - /* - * Easily overlooked: when mprotect shifts the boundary, - * make sure the expanding vma has anon_vma set if the - * shrinking vma had, to cover any anon pages imported. - */ - if (importer && !importer->anon_vma) { - importer->anon_vma = anon_vma; - __anon_vma_link(importer); - } - } - if (root) { flush_dcache_mmap_lock(mapping); vma_prio_tree_remove(vma, root); @@ -616,8 +616,6 @@ again: remove_next = 1 + (end > next->vm_end); __vma_unlink(mm, next, vma); if (file) __remove_shared_vm_struct(next, file, mapping); - if (next->anon_vma) - __anon_vma_merge(vma, next); } else if (insert) { /* * split_vma has split insert from vma, and needs @@ -627,8 +625,6 @@ again: remove_next = 1 + (end > next->vm_end); __insert_vm_struct(mm, insert); } - if (anon_vma) - spin_unlock(&anon_vma->lock); if (mapping) spin_unlock(&mapping->i_mmap_lock); @@ -638,6 +634,8 @@ again: remove_next = 1 + (end > next->vm_end); if (next->vm_flags & VM_EXECUTABLE) removed_exe_file_vma(mm); } + if (next->anon_vma) + anon_vma_merge(vma, next); mm->map_count--; mpol_put(vma_policy(next)); kmem_cache_free(vm_area_cachep, next); @@ -653,6 +651,8 @@ again: remove_next = 1 + (end > next->vm_end); } validate_mm(mm); + + return 0; } /* @@ -759,6 +759,7 @@ struct vm_area_struct *vma_merge(struct mm_struct *mm, { pgoff_t pglen = (end - addr) >> PAGE_SHIFT; struct vm_area_struct *area, *next; + int err; /* * We later require that vma->vm_flags == vm_flags, @@ -792,11 +793,13 @@ struct vm_area_struct *vma_merge(struct mm_struct *mm, is_mergeable_anon_vma(prev->anon_vma, next->anon_vma)) { /* cases 1, 6 */ - vma_adjust(prev, prev->vm_start, + err = vma_adjust(prev, prev->vm_start, next->vm_end, prev->vm_pgoff, NULL); } else /* cases 2, 5, 7 */ - vma_adjust(prev, prev->vm_start, + err = vma_adjust(prev, prev->vm_start, end, prev->vm_pgoff, NULL); + if (err) + return NULL; return prev; } @@ -808,11 +811,13 @@ struct vm_area_struct *vma_merge(struct mm_struct *mm, can_vma_merge_before(next, vm_flags, anon_vma, file, pgoff+pglen)) { if (prev && addr < prev->vm_end) /* case 4 */ - vma_adjust(prev, prev->vm_start, + err = vma_adjust(prev, prev->vm_start, addr, prev->vm_pgoff, NULL); else /* cases 3, 8 */ - vma_adjust(area, addr, next->vm_end, + err = vma_adjust(area, addr, next->vm_end, next->vm_pgoff - pglen, NULL); + if (err) + return NULL; return area; } @@ -967,7 +972,7 @@ unsigned long do_mmap_pgoff(struct file *file, unsigned long addr, unsigned long locked, lock_limit; locked = len >> PAGE_SHIFT; locked += mm->locked_vm; - lock_limit = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur; + lock_limit = rlimit(RLIMIT_MEMLOCK); lock_limit >>= PAGE_SHIFT; if (locked > lock_limit && !capable(CAP_IPC_LOCK)) return -EAGAIN; @@ -1205,6 +1210,7 @@ munmap_back: vma->vm_flags = vm_flags; vma->vm_page_prot = vm_get_page_prot(vm_flags); vma->vm_pgoff = pgoff; + INIT_LIST_HEAD(&vma->anon_vma_chain); if (file) { error = -EINVAL; @@ -1265,13 +1271,8 @@ out: mm->total_vm += len >> PAGE_SHIFT; vm_stat_account(mm, vm_flags, file, len >> PAGE_SHIFT); if (vm_flags & VM_LOCKED) { - /* - * makes pages present; downgrades, drops, reacquires mmap_sem - */ - long nr_pages = mlock_vma_pages_range(vma, addr, addr + len); - if (nr_pages < 0) - return nr_pages; /* vma gone! */ - mm->locked_vm += (len >> PAGE_SHIFT) - nr_pages; + if (!mlock_vma_pages_range(vma, addr, addr + len)) + mm->locked_vm += (len >> PAGE_SHIFT); } else if ((flags & MAP_POPULATE) && !(flags & MAP_NONBLOCK)) make_pages_present(addr, addr + len); return addr; @@ -1599,7 +1600,7 @@ static int acct_stack_growth(struct vm_area_struct *vma, unsigned long size, uns return -ENOMEM; /* Stack limit test */ - if (size > rlim[RLIMIT_STACK].rlim_cur) + if (size > ACCESS_ONCE(rlim[RLIMIT_STACK].rlim_cur)) return -ENOMEM; /* mlock limit tests */ @@ -1607,7 +1608,8 @@ static int acct_stack_growth(struct vm_area_struct *vma, unsigned long size, uns unsigned long locked; unsigned long limit; locked = mm->locked_vm + grow; - limit = rlim[RLIMIT_MEMLOCK].rlim_cur >> PAGE_SHIFT; + limit = ACCESS_ONCE(rlim[RLIMIT_MEMLOCK].rlim_cur); + limit >>= PAGE_SHIFT; if (locked > limit && !capable(CAP_IPC_LOCK)) return -ENOMEM; } @@ -1754,8 +1756,7 @@ find_extend_vma(struct mm_struct *mm, unsigned long addr) if (!prev || expand_stack(prev, addr)) return NULL; if (prev->vm_flags & VM_LOCKED) { - if (mlock_vma_pages_range(prev, addr, prev->vm_end) < 0) - return NULL; /* vma gone! */ + mlock_vma_pages_range(prev, addr, prev->vm_end); } return prev; } @@ -1783,8 +1784,7 @@ find_extend_vma(struct mm_struct * mm, unsigned long addr) if (expand_stack(vma, addr)) return NULL; if (vma->vm_flags & VM_LOCKED) { - if (mlock_vma_pages_range(vma, addr, start) < 0) - return NULL; /* vma gone! */ + mlock_vma_pages_range(vma, addr, start); } return vma; } @@ -1871,6 +1871,7 @@ static int __split_vma(struct mm_struct * mm, struct vm_area_struct * vma, { struct mempolicy *pol; struct vm_area_struct *new; + int err = -ENOMEM; if (is_vm_hugetlb_page(vma) && (addr & ~(huge_page_mask(hstate_vma(vma))))) @@ -1878,11 +1879,13 @@ static int __split_vma(struct mm_struct * mm, struct vm_area_struct * vma, new = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (!new) - return -ENOMEM; + goto out_err; /* most fields are the same, copy all, and then fixup */ *new = *vma; + INIT_LIST_HEAD(&new->anon_vma_chain); + if (new_below) new->vm_end = addr; else { @@ -1892,11 +1895,14 @@ static int __split_vma(struct mm_struct * mm, struct vm_area_struct * vma, pol = mpol_dup(vma_policy(vma)); if (IS_ERR(pol)) { - kmem_cache_free(vm_area_cachep, new); - return PTR_ERR(pol); + err = PTR_ERR(pol); + goto out_free_vma; } vma_set_policy(new, pol); + if (anon_vma_clone(new, vma)) + goto out_free_mpol; + if (new->vm_file) { get_file(new->vm_file); if (vma->vm_flags & VM_EXECUTABLE) @@ -1907,12 +1913,28 @@ static int __split_vma(struct mm_struct * mm, struct vm_area_struct * vma, new->vm_ops->open(new); if (new_below) - vma_adjust(vma, addr, vma->vm_end, vma->vm_pgoff + + err = vma_adjust(vma, addr, vma->vm_end, vma->vm_pgoff + ((addr - new->vm_start) >> PAGE_SHIFT), new); else - vma_adjust(vma, vma->vm_start, addr, vma->vm_pgoff, new); + err = vma_adjust(vma, vma->vm_start, addr, vma->vm_pgoff, new); - return 0; + /* Success. */ + if (!err) + return 0; + + /* Clean everything up if vma_adjust failed. */ + new->vm_ops->close(new); + if (new->vm_file) { + if (vma->vm_flags & VM_EXECUTABLE) + removed_exe_file_vma(mm); + fput(new->vm_file); + } + out_free_mpol: + mpol_put(pol); + out_free_vma: + kmem_cache_free(vm_area_cachep, new); + out_err: + return err; } /* @@ -2074,7 +2096,7 @@ unsigned long do_brk(unsigned long addr, unsigned long len) unsigned long locked, lock_limit; locked = len >> PAGE_SHIFT; locked += mm->locked_vm; - lock_limit = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur; + lock_limit = rlimit(RLIMIT_MEMLOCK); lock_limit >>= PAGE_SHIFT; if (locked > lock_limit && !capable(CAP_IPC_LOCK)) return -EAGAIN; @@ -2122,6 +2144,7 @@ unsigned long do_brk(unsigned long addr, unsigned long len) return -ENOMEM; } + INIT_LIST_HEAD(&vma->anon_vma_chain); vma->vm_mm = mm; vma->vm_start = addr; vma->vm_end = addr + len; @@ -2258,10 +2281,11 @@ struct vm_area_struct *copy_vma(struct vm_area_struct **vmap, if (new_vma) { *new_vma = *vma; pol = mpol_dup(vma_policy(vma)); - if (IS_ERR(pol)) { - kmem_cache_free(vm_area_cachep, new_vma); - return NULL; - } + if (IS_ERR(pol)) + goto out_free_vma; + INIT_LIST_HEAD(&new_vma->anon_vma_chain); + if (anon_vma_clone(new_vma, vma)) + goto out_free_mempol; vma_set_policy(new_vma, pol); new_vma->vm_start = addr; new_vma->vm_end = addr + len; @@ -2277,6 +2301,12 @@ struct vm_area_struct *copy_vma(struct vm_area_struct **vmap, } } return new_vma; + + out_free_mempol: + mpol_put(pol); + out_free_vma: + kmem_cache_free(vm_area_cachep, new_vma); + return NULL; } /* @@ -2288,7 +2318,7 @@ int may_expand_vm(struct mm_struct *mm, unsigned long npages) unsigned long cur = mm->total_vm; /* pages */ unsigned long lim; - lim = current->signal->rlim[RLIMIT_AS].rlim_cur >> PAGE_SHIFT; + lim = rlimit(RLIMIT_AS) >> PAGE_SHIFT; if (cur + npages > lim) return 0; @@ -2354,6 +2384,7 @@ int install_special_mapping(struct mm_struct *mm, if (unlikely(vma == NULL)) return -ENOMEM; + INIT_LIST_HEAD(&vma->anon_vma_chain); vma->vm_mm = mm; vma->vm_start = addr; vma->vm_end = addr + len; @@ -2454,6 +2485,7 @@ static void vm_lock_mapping(struct mm_struct *mm, struct address_space *mapping) int mm_take_all_locks(struct mm_struct *mm) { struct vm_area_struct *vma; + struct anon_vma_chain *avc; int ret = -EINTR; BUG_ON(down_read_trylock(&mm->mmap_sem)); @@ -2471,7 +2503,8 @@ int mm_take_all_locks(struct mm_struct *mm) if (signal_pending(current)) goto out_unlock; if (vma->anon_vma) - vm_lock_anon_vma(mm, vma->anon_vma); + list_for_each_entry(avc, &vma->anon_vma_chain, same_vma) + vm_lock_anon_vma(mm, avc->anon_vma); } ret = 0; @@ -2526,13 +2559,15 @@ static void vm_unlock_mapping(struct address_space *mapping) void mm_drop_all_locks(struct mm_struct *mm) { struct vm_area_struct *vma; + struct anon_vma_chain *avc; BUG_ON(down_read_trylock(&mm->mmap_sem)); BUG_ON(!mutex_is_locked(&mm_all_locks_mutex)); for (vma = mm->mmap; vma; vma = vma->vm_next) { if (vma->anon_vma) - vm_unlock_anon_vma(vma->anon_vma); + list_for_each_entry(avc, &vma->anon_vma_chain, same_vma) + vm_unlock_anon_vma(avc->anon_vma); if (vma->vm_file && vma->vm_file->f_mapping) vm_unlock_mapping(vma->vm_file->f_mapping); } diff --git a/mm/mremap.c b/mm/mremap.c index 845190898d59..e9c75efce609 100644 --- a/mm/mremap.c +++ b/mm/mremap.c @@ -285,7 +285,7 @@ static struct vm_area_struct *vma_to_resize(unsigned long addr, if (vma->vm_flags & VM_LOCKED) { unsigned long locked, lock_limit; locked = mm->locked_vm << PAGE_SHIFT; - lock_limit = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur; + lock_limit = rlimit(RLIMIT_MEMLOCK); locked += new_len - old_len; if (locked > lock_limit && !capable(CAP_IPC_LOCK)) goto Eagain; @@ -460,8 +460,11 @@ unsigned long do_mremap(unsigned long addr, if (vma_expandable(vma, new_len - old_len)) { int pages = (new_len - old_len) >> PAGE_SHIFT; - vma_adjust(vma, vma->vm_start, - addr + new_len, vma->vm_pgoff, NULL); + if (vma_adjust(vma, vma->vm_start, addr + new_len, + vma->vm_pgoff, NULL)) { + ret = -ENOMEM; + goto out; + } mm->total_vm += pages; vm_stat_account(mm, vma->vm_flags, vma->vm_file, pages); diff --git a/mm/nommu.c b/mm/nommu.c index 48a2ecfaf059..b9b5cceb1b68 100644 --- a/mm/nommu.c +++ b/mm/nommu.c @@ -146,7 +146,7 @@ int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm, (VM_MAYREAD | VM_MAYWRITE) : (VM_READ | VM_WRITE); for (i = 0; i < nr_pages; i++) { - vma = find_vma(mm, start); + vma = find_extend_vma(mm, start); if (!vma) goto finish_or_fault; @@ -764,7 +764,7 @@ EXPORT_SYMBOL(find_vma); */ struct vm_area_struct *find_extend_vma(struct mm_struct *mm, unsigned long addr) { - return find_vma(mm, addr); + return find_vma(mm, addr & PAGE_MASK); } /* @@ -1209,7 +1209,7 @@ unsigned long do_mmap_pgoff(struct file *file, region->vm_flags = vm_flags; region->vm_pgoff = pgoff; - INIT_LIST_HEAD(&vma->anon_vma_node); + INIT_LIST_HEAD(&vma->anon_vma_chain); vma->vm_flags = vm_flags; vma->vm_pgoff = pgoff; diff --git a/mm/oom_kill.c b/mm/oom_kill.c index 237050478f28..35755a4156d6 100644 --- a/mm/oom_kill.c +++ b/mm/oom_kill.c @@ -401,8 +401,8 @@ static void __oom_kill_task(struct task_struct *p, int verbose) "vsz:%lukB, anon-rss:%lukB, file-rss:%lukB\n", task_pid_nr(p), p->comm, K(p->mm->total_vm), - K(get_mm_counter(p->mm, anon_rss)), - K(get_mm_counter(p->mm, file_rss))); + K(get_mm_counter(p->mm, MM_ANONPAGES)), + K(get_mm_counter(p->mm, MM_FILEPAGES))); task_unlock(p); /* diff --git a/mm/page_alloc.c b/mm/page_alloc.c index a6b17aa4740b..a8182c89de59 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -76,6 +76,31 @@ unsigned long totalreserve_pages __read_mostly; int percpu_pagelist_fraction; gfp_t gfp_allowed_mask __read_mostly = GFP_BOOT_MASK; +#ifdef CONFIG_PM_SLEEP +/* + * The following functions are used by the suspend/hibernate code to temporarily + * change gfp_allowed_mask in order to avoid using I/O during memory allocations + * while devices are suspended. To avoid races with the suspend/hibernate code, + * they should always be called with pm_mutex held (gfp_allowed_mask also should + * only be modified with pm_mutex held, unless the suspend/hibernate code is + * guaranteed not to run in parallel with that modification). + */ +void set_gfp_allowed_mask(gfp_t mask) +{ + WARN_ON(!mutex_is_locked(&pm_mutex)); + gfp_allowed_mask = mask; +} + +gfp_t clear_gfp_allowed_mask(gfp_t mask) +{ + gfp_t ret = gfp_allowed_mask; + + WARN_ON(!mutex_is_locked(&pm_mutex)); + gfp_allowed_mask &= ~mask; + return ret; +} +#endif /* CONFIG_PM_SLEEP */ + #ifdef CONFIG_HUGETLB_PAGE_SIZE_VARIABLE int pageblock_order __read_mostly; #endif @@ -530,7 +555,7 @@ static void free_pcppages_bulk(struct zone *zone, int count, int batch_free = 0; spin_lock(&zone->lock); - zone_clear_flag(zone, ZONE_ALL_UNRECLAIMABLE); + zone->all_unreclaimable = 0; zone->pages_scanned = 0; __mod_zone_page_state(zone, NR_FREE_PAGES, count); @@ -568,7 +593,7 @@ static void free_one_page(struct zone *zone, struct page *page, int order, int migratetype) { spin_lock(&zone->lock); - zone_clear_flag(zone, ZONE_ALL_UNRECLAIMABLE); + zone->all_unreclaimable = 0; zone->pages_scanned = 0; __mod_zone_page_state(zone, NR_FREE_PAGES, 1 << order); @@ -583,6 +608,7 @@ static void __free_pages_ok(struct page *page, unsigned int order) int bad = 0; int wasMlocked = __TestClearPageMlocked(page); + trace_mm_page_free_direct(page, order); kmemcheck_free_shadow(page, order); for (i = 0 ; i < (1 << order) ; ++i) @@ -1073,8 +1099,9 @@ void mark_free_pages(struct zone *zone) /* * Free a 0-order page + * cold == 1 ? free a cold page : free a hot page */ -static void free_hot_cold_page(struct page *page, int cold) +void free_hot_cold_page(struct page *page, int cold) { struct zone *zone = page_zone(page); struct per_cpu_pages *pcp; @@ -1082,6 +1109,7 @@ static void free_hot_cold_page(struct page *page, int cold) int migratetype; int wasMlocked = __TestClearPageMlocked(page); + trace_mm_page_free_direct(page, 0); kmemcheck_free_shadow(page, 0); if (PageAnon(page)) @@ -1133,12 +1161,6 @@ out: local_irq_restore(flags); } -void free_hot_page(struct page *page) -{ - trace_mm_page_free_direct(page, 0); - free_hot_cold_page(page, 0); -} - /* * split_page takes a non-compound higher-order page, and splits it into * n (1<<order) sub-pages: page[0..n] @@ -2008,9 +2030,8 @@ void __pagevec_free(struct pagevec *pvec) void __free_pages(struct page *page, unsigned int order) { if (put_page_testzero(page)) { - trace_mm_page_free_direct(page, order); if (order == 0) - free_hot_page(page); + free_hot_cold_page(page, 0); else __free_pages_ok(page, order); } @@ -2266,7 +2287,7 @@ void show_free_areas(void) K(zone_page_state(zone, NR_BOUNCE)), K(zone_page_state(zone, NR_WRITEBACK_TEMP)), zone->pages_scanned, - (zone_is_all_unreclaimable(zone) ? "yes" : "no") + (zone->all_unreclaimable ? "yes" : "no") ); printk("lowmem_reserve[]:"); for (i = 0; i < MAX_NR_ZONES; i++) @@ -4371,8 +4392,12 @@ void __init free_area_init_nodes(unsigned long *max_zone_pfn) for (i = 0; i < MAX_NR_ZONES; i++) { if (i == ZONE_MOVABLE) continue; - printk(" %-8s %0#10lx -> %0#10lx\n", - zone_names[i], + printk(" %-8s ", zone_names[i]); + if (arch_zone_lowest_possible_pfn[i] == + arch_zone_highest_possible_pfn[i]) + printk("empty\n"); + else + printk("%0#10lx -> %0#10lx\n", arch_zone_lowest_possible_pfn[i], arch_zone_highest_possible_pfn[i]); } diff --git a/mm/readahead.c b/mm/readahead.c index 033bc135a41f..337b20e946f6 100644 --- a/mm/readahead.c +++ b/mm/readahead.c @@ -501,6 +501,12 @@ void page_cache_sync_readahead(struct address_space *mapping, if (!ra->ra_pages) return; + /* be dumb */ + if (filp->f_mode & FMODE_RANDOM) { + force_page_cache_readahead(mapping, filp, offset, req_size); + return; + } + /* do read-ahead */ ondemand_readahead(mapping, ra, filp, false, offset, req_size); } diff --git a/mm/rmap.c b/mm/rmap.c index 278cd277bdec..fcd593c9c997 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -62,6 +62,7 @@ #include "internal.h" static struct kmem_cache *anon_vma_cachep; +static struct kmem_cache *anon_vma_chain_cachep; static inline struct anon_vma *anon_vma_alloc(void) { @@ -73,6 +74,16 @@ void anon_vma_free(struct anon_vma *anon_vma) kmem_cache_free(anon_vma_cachep, anon_vma); } +static inline struct anon_vma_chain *anon_vma_chain_alloc(void) +{ + return kmem_cache_alloc(anon_vma_chain_cachep, GFP_KERNEL); +} + +void anon_vma_chain_free(struct anon_vma_chain *anon_vma_chain) +{ + kmem_cache_free(anon_vma_chain_cachep, anon_vma_chain); +} + /** * anon_vma_prepare - attach an anon_vma to a memory region * @vma: the memory region in question @@ -103,18 +114,23 @@ void anon_vma_free(struct anon_vma *anon_vma) int anon_vma_prepare(struct vm_area_struct *vma) { struct anon_vma *anon_vma = vma->anon_vma; + struct anon_vma_chain *avc; might_sleep(); if (unlikely(!anon_vma)) { struct mm_struct *mm = vma->vm_mm; struct anon_vma *allocated; + avc = anon_vma_chain_alloc(); + if (!avc) + goto out_enomem; + anon_vma = find_mergeable_anon_vma(vma); allocated = NULL; if (!anon_vma) { anon_vma = anon_vma_alloc(); if (unlikely(!anon_vma)) - return -ENOMEM; + goto out_enomem_free_avc; allocated = anon_vma; } spin_lock(&anon_vma->lock); @@ -123,53 +139,113 @@ int anon_vma_prepare(struct vm_area_struct *vma) spin_lock(&mm->page_table_lock); if (likely(!vma->anon_vma)) { vma->anon_vma = anon_vma; - list_add_tail(&vma->anon_vma_node, &anon_vma->head); + avc->anon_vma = anon_vma; + avc->vma = vma; + list_add(&avc->same_vma, &vma->anon_vma_chain); + list_add(&avc->same_anon_vma, &anon_vma->head); allocated = NULL; } spin_unlock(&mm->page_table_lock); spin_unlock(&anon_vma->lock); - if (unlikely(allocated)) + if (unlikely(allocated)) { anon_vma_free(allocated); + anon_vma_chain_free(avc); + } } return 0; + + out_enomem_free_avc: + anon_vma_chain_free(avc); + out_enomem: + return -ENOMEM; } -void __anon_vma_merge(struct vm_area_struct *vma, struct vm_area_struct *next) +static void anon_vma_chain_link(struct vm_area_struct *vma, + struct anon_vma_chain *avc, + struct anon_vma *anon_vma) { - BUG_ON(vma->anon_vma != next->anon_vma); - list_del(&next->anon_vma_node); + avc->vma = vma; + avc->anon_vma = anon_vma; + list_add(&avc->same_vma, &vma->anon_vma_chain); + + spin_lock(&anon_vma->lock); + list_add_tail(&avc->same_anon_vma, &anon_vma->head); + spin_unlock(&anon_vma->lock); } -void __anon_vma_link(struct vm_area_struct *vma) +/* + * Attach the anon_vmas from src to dst. + * Returns 0 on success, -ENOMEM on failure. + */ +int anon_vma_clone(struct vm_area_struct *dst, struct vm_area_struct *src) { - struct anon_vma *anon_vma = vma->anon_vma; + struct anon_vma_chain *avc, *pavc; - if (anon_vma) - list_add_tail(&vma->anon_vma_node, &anon_vma->head); + list_for_each_entry(pavc, &src->anon_vma_chain, same_vma) { + avc = anon_vma_chain_alloc(); + if (!avc) + goto enomem_failure; + anon_vma_chain_link(dst, avc, pavc->anon_vma); + } + return 0; + + enomem_failure: + unlink_anon_vmas(dst); + return -ENOMEM; } -void anon_vma_link(struct vm_area_struct *vma) +/* + * Attach vma to its own anon_vma, as well as to the anon_vmas that + * the corresponding VMA in the parent process is attached to. + * Returns 0 on success, non-zero on failure. + */ +int anon_vma_fork(struct vm_area_struct *vma, struct vm_area_struct *pvma) { - struct anon_vma *anon_vma = vma->anon_vma; + struct anon_vma_chain *avc; + struct anon_vma *anon_vma; - if (anon_vma) { - spin_lock(&anon_vma->lock); - list_add_tail(&vma->anon_vma_node, &anon_vma->head); - spin_unlock(&anon_vma->lock); - } + /* Don't bother if the parent process has no anon_vma here. */ + if (!pvma->anon_vma) + return 0; + + /* + * First, attach the new VMA to the parent VMA's anon_vmas, + * so rmap can find non-COWed pages in child processes. + */ + if (anon_vma_clone(vma, pvma)) + return -ENOMEM; + + /* Then add our own anon_vma. */ + anon_vma = anon_vma_alloc(); + if (!anon_vma) + goto out_error; + avc = anon_vma_chain_alloc(); + if (!avc) + goto out_error_free_anon_vma; + anon_vma_chain_link(vma, avc, anon_vma); + /* Mark this anon_vma as the one where our new (COWed) pages go. */ + vma->anon_vma = anon_vma; + + return 0; + + out_error_free_anon_vma: + anon_vma_free(anon_vma); + out_error: + return -ENOMEM; } -void anon_vma_unlink(struct vm_area_struct *vma) +static void anon_vma_unlink(struct anon_vma_chain *anon_vma_chain) { - struct anon_vma *anon_vma = vma->anon_vma; + struct anon_vma *anon_vma = anon_vma_chain->anon_vma; int empty; + /* If anon_vma_fork fails, we can get an empty anon_vma_chain. */ if (!anon_vma) return; spin_lock(&anon_vma->lock); - list_del(&vma->anon_vma_node); + list_del(&anon_vma_chain->same_anon_vma); /* We must garbage collect the anon_vma if it's empty */ empty = list_empty(&anon_vma->head) && !ksm_refcount(anon_vma); @@ -179,6 +255,18 @@ void anon_vma_unlink(struct vm_area_struct *vma) anon_vma_free(anon_vma); } +void unlink_anon_vmas(struct vm_area_struct *vma) +{ + struct anon_vma_chain *avc, *next; + + /* Unlink each anon_vma chained to the VMA. */ + list_for_each_entry_safe(avc, next, &vma->anon_vma_chain, same_vma) { + anon_vma_unlink(avc); + list_del(&avc->same_vma); + anon_vma_chain_free(avc); + } +} + static void anon_vma_ctor(void *data) { struct anon_vma *anon_vma = data; @@ -192,6 +280,7 @@ void __init anon_vma_init(void) { anon_vma_cachep = kmem_cache_create("anon_vma", sizeof(struct anon_vma), 0, SLAB_DESTROY_BY_RCU|SLAB_PANIC, anon_vma_ctor); + anon_vma_chain_cachep = KMEM_CACHE(anon_vma_chain, SLAB_PANIC); } /* @@ -396,7 +485,7 @@ static int page_referenced_anon(struct page *page, { unsigned int mapcount; struct anon_vma *anon_vma; - struct vm_area_struct *vma; + struct anon_vma_chain *avc; int referenced = 0; anon_vma = page_lock_anon_vma(page); @@ -404,7 +493,8 @@ static int page_referenced_anon(struct page *page, return referenced; mapcount = page_mapcount(page); - list_for_each_entry(vma, &anon_vma->head, anon_vma_node) { + list_for_each_entry(avc, &anon_vma->head, same_anon_vma) { + struct vm_area_struct *vma = avc->vma; unsigned long address = vma_address(page, vma); if (address == -EFAULT) continue; @@ -511,9 +601,6 @@ int page_referenced(struct page *page, int referenced = 0; int we_locked = 0; - if (TestClearPageReferenced(page)) - referenced++; - *vm_flags = 0; if (page_mapped(page) && page_rmapping(page)) { if (!is_locked && (!PageAnon(page) || PageKsm(page))) { @@ -614,6 +701,30 @@ int page_mkclean(struct page *page) EXPORT_SYMBOL_GPL(page_mkclean); /** + * page_move_anon_rmap - move a page to our anon_vma + * @page: the page to move to our anon_vma + * @vma: the vma the page belongs to + * @address: the user virtual address mapped + * + * When a page belongs exclusively to one process after a COW event, + * that page can be moved into the anon_vma that belongs to just that + * process, so the rmap code will not search the parent or sibling + * processes. + */ +void page_move_anon_rmap(struct page *page, + struct vm_area_struct *vma, unsigned long address) +{ + struct anon_vma *anon_vma = vma->anon_vma; + + VM_BUG_ON(!PageLocked(page)); + VM_BUG_ON(!anon_vma); + VM_BUG_ON(page->index != linear_page_index(vma, address)); + + anon_vma = (void *) anon_vma + PAGE_MAPPING_ANON; + page->mapping = (struct address_space *) anon_vma; +} + +/** * __page_set_anon_rmap - setup new anonymous rmap * @page: the page to add the mapping to * @vma: the vm area in which the mapping is added @@ -652,9 +763,6 @@ static void __page_check_anon_rmap(struct page *page, * are initially only visible via the pagetables, and the pte is locked * over the call to page_add_new_anon_rmap. */ - struct anon_vma *anon_vma = vma->anon_vma; - anon_vma = (void *) anon_vma + PAGE_MAPPING_ANON; - BUG_ON(page->mapping != (struct address_space *)anon_vma); BUG_ON(page->index != linear_page_index(vma, address)); #endif } @@ -815,9 +923,9 @@ int try_to_unmap_one(struct page *page, struct vm_area_struct *vma, if (PageHWPoison(page) && !(flags & TTU_IGNORE_HWPOISON)) { if (PageAnon(page)) - dec_mm_counter(mm, anon_rss); + dec_mm_counter(mm, MM_ANONPAGES); else - dec_mm_counter(mm, file_rss); + dec_mm_counter(mm, MM_FILEPAGES); set_pte_at(mm, address, pte, swp_entry_to_pte(make_hwpoison_entry(page))); } else if (PageAnon(page)) { @@ -839,7 +947,8 @@ int try_to_unmap_one(struct page *page, struct vm_area_struct *vma, list_add(&mm->mmlist, &init_mm.mmlist); spin_unlock(&mmlist_lock); } - dec_mm_counter(mm, anon_rss); + dec_mm_counter(mm, MM_ANONPAGES); + inc_mm_counter(mm, MM_SWAPENTS); } else if (PAGE_MIGRATION) { /* * Store the pfn of the page in a special migration @@ -857,7 +966,7 @@ int try_to_unmap_one(struct page *page, struct vm_area_struct *vma, entry = make_migration_entry(page, pte_write(pteval)); set_pte_at(mm, address, pte, swp_entry_to_pte(entry)); } else - dec_mm_counter(mm, file_rss); + dec_mm_counter(mm, MM_FILEPAGES); page_remove_rmap(page); page_cache_release(page); @@ -996,7 +1105,7 @@ static int try_to_unmap_cluster(unsigned long cursor, unsigned int *mapcount, page_remove_rmap(page); page_cache_release(page); - dec_mm_counter(mm, file_rss); + dec_mm_counter(mm, MM_FILEPAGES); (*mapcount)--; } pte_unmap_unlock(pte - 1, ptl); @@ -1024,14 +1133,15 @@ static int try_to_unmap_cluster(unsigned long cursor, unsigned int *mapcount, static int try_to_unmap_anon(struct page *page, enum ttu_flags flags) { struct anon_vma *anon_vma; - struct vm_area_struct *vma; + struct anon_vma_chain *avc; int ret = SWAP_AGAIN; anon_vma = page_lock_anon_vma(page); if (!anon_vma) return ret; - list_for_each_entry(vma, &anon_vma->head, anon_vma_node) { + list_for_each_entry(avc, &anon_vma->head, same_anon_vma) { + struct vm_area_struct *vma = avc->vma; unsigned long address = vma_address(page, vma); if (address == -EFAULT) continue; @@ -1222,7 +1332,7 @@ static int rmap_walk_anon(struct page *page, int (*rmap_one)(struct page *, struct vm_area_struct *, unsigned long, void *), void *arg) { struct anon_vma *anon_vma; - struct vm_area_struct *vma; + struct anon_vma_chain *avc; int ret = SWAP_AGAIN; /* @@ -1237,7 +1347,8 @@ static int rmap_walk_anon(struct page *page, int (*rmap_one)(struct page *, if (!anon_vma) return ret; spin_lock(&anon_vma->lock); - list_for_each_entry(vma, &anon_vma->head, anon_vma_node) { + list_for_each_entry(avc, &anon_vma->head, same_anon_vma) { + struct vm_area_struct *vma = avc->vma; unsigned long address = vma_address(page, vma); if (address == -EFAULT) continue; diff --git a/mm/swap.c b/mm/swap.c index 308e57d8d7ed..9036b89813ac 100644 --- a/mm/swap.c +++ b/mm/swap.c @@ -55,7 +55,7 @@ static void __page_cache_release(struct page *page) del_page_from_lru(zone, page); spin_unlock_irqrestore(&zone->lru_lock, flags); } - free_hot_page(page); + free_hot_cold_page(page, 0); } static void put_compound_page(struct page *page) diff --git a/mm/swapfile.c b/mm/swapfile.c index 6c0585b16418..84374d8cf814 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -840,7 +840,8 @@ static int unuse_pte(struct vm_area_struct *vma, pmd_t *pmd, goto out; } - inc_mm_counter(vma->vm_mm, anon_rss); + dec_mm_counter(vma->vm_mm, MM_SWAPENTS); + inc_mm_counter(vma->vm_mm, MM_ANONPAGES); get_page(page); set_pte_at(vma->vm_mm, addr, pte, pte_mkold(mk_pte(page, vma->vm_page_prot))); @@ -1759,11 +1760,11 @@ SYSCALL_DEFINE2(swapon, const char __user *, specialfile, int, swap_flags) unsigned int type; int i, prev; int error; - union swap_header *swap_header = NULL; - unsigned int nr_good_pages = 0; + union swap_header *swap_header; + unsigned int nr_good_pages; int nr_extents = 0; sector_t span; - unsigned long maxpages = 1; + unsigned long maxpages; unsigned long swapfilepages; unsigned char *swap_map = NULL; struct page *page = NULL; @@ -1922,9 +1923,13 @@ SYSCALL_DEFINE2(swapon, const char __user *, specialfile, int, swap_flags) * swap pte. */ maxpages = swp_offset(pte_to_swp_entry( - swp_entry_to_pte(swp_entry(0, ~0UL)))) - 1; - if (maxpages > swap_header->info.last_page) - maxpages = swap_header->info.last_page; + swp_entry_to_pte(swp_entry(0, ~0UL)))) + 1; + if (maxpages > swap_header->info.last_page) { + maxpages = swap_header->info.last_page + 1; + /* p->max is an unsigned int: don't overflow it */ + if ((unsigned int)maxpages == 0) + maxpages = UINT_MAX; + } p->highest_bit = maxpages - 1; error = -EINVAL; @@ -1948,23 +1953,24 @@ SYSCALL_DEFINE2(swapon, const char __user *, specialfile, int, swap_flags) } memset(swap_map, 0, maxpages); + nr_good_pages = maxpages - 1; /* omit header page */ + for (i = 0; i < swap_header->info.nr_badpages; i++) { - int page_nr = swap_header->info.badpages[i]; - if (page_nr <= 0 || page_nr >= swap_header->info.last_page) { + unsigned int page_nr = swap_header->info.badpages[i]; + if (page_nr == 0 || page_nr > swap_header->info.last_page) { error = -EINVAL; goto bad_swap; } - swap_map[page_nr] = SWAP_MAP_BAD; + if (page_nr < maxpages) { + swap_map[page_nr] = SWAP_MAP_BAD; + nr_good_pages--; + } } error = swap_cgroup_swapon(type, maxpages); if (error) goto bad_swap; - nr_good_pages = swap_header->info.last_page - - swap_header->info.nr_badpages - - 1 /* header page */; - if (nr_good_pages) { swap_map[0] = SWAP_MAP_BAD; p->max = maxpages; @@ -2155,7 +2161,11 @@ void swap_shmem_alloc(swp_entry_t entry) } /* - * increase reference count of swap entry by 1. + * Increase reference count of swap entry by 1. + * Returns 0 for success, or -ENOMEM if a swap_count_continuation is required + * but could not be atomically allocated. Returns 0, just as if it succeeded, + * if __swap_duplicate() fails for another reason (-EINVAL or -ENOENT), which + * might occur if a page table entry has got corrupted. */ int swap_duplicate(swp_entry_t entry) { diff --git a/mm/vmscan.c b/mm/vmscan.c index c26986c85ce0..79c809895fba 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -262,27 +262,6 @@ unsigned long shrink_slab(unsigned long scanned, gfp_t gfp_mask, return ret; } -/* Called without lock on whether page is mapped, so answer is unstable */ -static inline int page_mapping_inuse(struct page *page) -{ - struct address_space *mapping; - - /* Page is in somebody's page tables. */ - if (page_mapped(page)) - return 1; - - /* Be more reluctant to reclaim swapcache than pagecache */ - if (PageSwapCache(page)) - return 1; - - mapping = page_mapping(page); - if (!mapping) - return 0; - - /* File is mmap'd by somebody? */ - return mapping_mapped(mapping); -} - static inline int is_page_cache_freeable(struct page *page) { /* @@ -579,6 +558,65 @@ redo: put_page(page); /* drop ref from isolate */ } +enum page_references { + PAGEREF_RECLAIM, + PAGEREF_RECLAIM_CLEAN, + PAGEREF_KEEP, + PAGEREF_ACTIVATE, +}; + +static enum page_references page_check_references(struct page *page, + struct scan_control *sc) +{ + int referenced_ptes, referenced_page; + unsigned long vm_flags; + + referenced_ptes = page_referenced(page, 1, sc->mem_cgroup, &vm_flags); + referenced_page = TestClearPageReferenced(page); + + /* Lumpy reclaim - ignore references */ + if (sc->order > PAGE_ALLOC_COSTLY_ORDER) + return PAGEREF_RECLAIM; + + /* + * Mlock lost the isolation race with us. Let try_to_unmap() + * move the page to the unevictable list. + */ + if (vm_flags & VM_LOCKED) + return PAGEREF_RECLAIM; + + if (referenced_ptes) { + if (PageAnon(page)) + return PAGEREF_ACTIVATE; + /* + * All mapped pages start out with page table + * references from the instantiating fault, so we need + * to look twice if a mapped file page is used more + * than once. + * + * Mark it and spare it for another trip around the + * inactive list. Another page table reference will + * lead to its activation. + * + * Note: the mark is set for activated pages as well + * so that recently deactivated but used pages are + * quickly recovered. + */ + SetPageReferenced(page); + + if (referenced_page) + return PAGEREF_ACTIVATE; + + return PAGEREF_KEEP; + } + + /* Reclaim if clean, defer dirty pages to writeback */ + if (referenced_page) + return PAGEREF_RECLAIM_CLEAN; + + return PAGEREF_RECLAIM; +} + /* * shrink_page_list() returns the number of reclaimed pages */ @@ -590,16 +628,15 @@ static unsigned long shrink_page_list(struct list_head *page_list, struct pagevec freed_pvec; int pgactivate = 0; unsigned long nr_reclaimed = 0; - unsigned long vm_flags; cond_resched(); pagevec_init(&freed_pvec, 1); while (!list_empty(page_list)) { + enum page_references references; struct address_space *mapping; struct page *page; int may_enter_fs; - int referenced; cond_resched(); @@ -641,17 +678,16 @@ static unsigned long shrink_page_list(struct list_head *page_list, goto keep_locked; } - referenced = page_referenced(page, 1, - sc->mem_cgroup, &vm_flags); - /* - * In active use or really unfreeable? Activate it. - * If page which have PG_mlocked lost isoltation race, - * try_to_unmap moves it to unevictable list - */ - if (sc->order <= PAGE_ALLOC_COSTLY_ORDER && - referenced && page_mapping_inuse(page) - && !(vm_flags & VM_LOCKED)) + references = page_check_references(page, sc); + switch (references) { + case PAGEREF_ACTIVATE: goto activate_locked; + case PAGEREF_KEEP: + goto keep_locked; + case PAGEREF_RECLAIM: + case PAGEREF_RECLAIM_CLEAN: + ; /* try to reclaim the page below */ + } /* * Anonymous process memory has backing store? @@ -685,7 +721,7 @@ static unsigned long shrink_page_list(struct list_head *page_list, } if (PageDirty(page)) { - if (sc->order <= PAGE_ALLOC_COSTLY_ORDER && referenced) + if (references == PAGEREF_RECLAIM_CLEAN) goto keep_locked; if (!may_enter_fs) goto keep_locked; @@ -1350,9 +1386,7 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone, continue; } - /* page_referenced clears PageReferenced */ - if (page_mapping_inuse(page) && - page_referenced(page, 0, sc->mem_cgroup, &vm_flags)) { + if (page_referenced(page, 0, sc->mem_cgroup, &vm_flags)) { nr_rotated++; /* * Identify referenced, file-backed active pages and @@ -1501,6 +1535,13 @@ static void get_scan_ratio(struct zone *zone, struct scan_control *sc, unsigned long ap, fp; struct zone_reclaim_stat *reclaim_stat = get_reclaim_stat(zone, sc); + /* If we have no swap space, do not bother scanning anon pages. */ + if (!sc->may_swap || (nr_swap_pages <= 0)) { + percent[0] = 0; + percent[1] = 100; + return; + } + anon = zone_nr_lru_pages(zone, sc, LRU_ACTIVE_ANON) + zone_nr_lru_pages(zone, sc, LRU_INACTIVE_ANON); file = zone_nr_lru_pages(zone, sc, LRU_ACTIVE_FILE) + @@ -1598,22 +1639,20 @@ static void shrink_zone(int priority, struct zone *zone, unsigned long nr_reclaimed = sc->nr_reclaimed; unsigned long nr_to_reclaim = sc->nr_to_reclaim; struct zone_reclaim_stat *reclaim_stat = get_reclaim_stat(zone, sc); - int noswap = 0; - /* If we have no swap space, do not bother scanning anon pages. */ - if (!sc->may_swap || (nr_swap_pages <= 0)) { - noswap = 1; - percent[0] = 0; - percent[1] = 100; - } else - get_scan_ratio(zone, sc, percent); + get_scan_ratio(zone, sc, percent); for_each_evictable_lru(l) { int file = is_file_lru(l); unsigned long scan; + if (percent[file] == 0) { + nr[l] = 0; + continue; + } + scan = zone_nr_lru_pages(zone, sc, l); - if (priority || noswap) { + if (priority) { scan >>= priority; scan = (scan * percent[file]) / 100; } @@ -1694,8 +1733,7 @@ static void shrink_zones(int priority, struct zonelist *zonelist, continue; note_zone_scanning_priority(zone, priority); - if (zone_is_all_unreclaimable(zone) && - priority != DEF_PRIORITY) + if (zone->all_unreclaimable && priority != DEF_PRIORITY) continue; /* Let kswapd poll it */ sc->all_unreclaimable = 0; } else { @@ -1922,7 +1960,7 @@ static int sleeping_prematurely(pg_data_t *pgdat, int order, long remaining) if (!populated_zone(zone)) continue; - if (zone_is_all_unreclaimable(zone)) + if (zone->all_unreclaimable) continue; if (!zone_watermark_ok(zone, order, high_wmark_pages(zone), @@ -2012,8 +2050,7 @@ loop_again: if (!populated_zone(zone)) continue; - if (zone_is_all_unreclaimable(zone) && - priority != DEF_PRIORITY) + if (zone->all_unreclaimable && priority != DEF_PRIORITY) continue; /* @@ -2056,13 +2093,9 @@ loop_again: if (!populated_zone(zone)) continue; - if (zone_is_all_unreclaimable(zone) && - priority != DEF_PRIORITY) + if (zone->all_unreclaimable && priority != DEF_PRIORITY) continue; - if (!zone_watermark_ok(zone, order, - high_wmark_pages(zone), end_zone, 0)) - all_zones_ok = 0; temp_priority[i] = priority; sc.nr_scanned = 0; note_zone_scanning_priority(zone, priority); @@ -2087,12 +2120,11 @@ loop_again: lru_pages); sc.nr_reclaimed += reclaim_state->reclaimed_slab; total_scanned += sc.nr_scanned; - if (zone_is_all_unreclaimable(zone)) + if (zone->all_unreclaimable) continue; - if (nr_slab == 0 && zone->pages_scanned >= - (zone_reclaimable_pages(zone) * 6)) - zone_set_flag(zone, - ZONE_ALL_UNRECLAIMABLE); + if (nr_slab == 0 && + zone->pages_scanned >= (zone_reclaimable_pages(zone) * 6)) + zone->all_unreclaimable = 1; /* * If we've done a decent amount of scanning and * the reclaim ratio is low, start doing writepage @@ -2102,13 +2134,18 @@ loop_again: total_scanned > sc.nr_reclaimed + sc.nr_reclaimed / 2) sc.may_writepage = 1; - /* - * We are still under min water mark. it mean we have - * GFP_ATOMIC allocation failure risk. Hurry up! - */ - if (!zone_watermark_ok(zone, order, min_wmark_pages(zone), - end_zone, 0)) - has_under_min_watermark_zone = 1; + if (!zone_watermark_ok(zone, order, + high_wmark_pages(zone), end_zone, 0)) { + all_zones_ok = 0; + /* + * We are still under min water mark. This + * means that we have a GFP_ATOMIC allocation + * failure risk. Hurry up! + */ + if (!zone_watermark_ok(zone, order, + min_wmark_pages(zone), end_zone, 0)) + has_under_min_watermark_zone = 1; + } } if (all_zones_ok) @@ -2550,6 +2587,7 @@ static int __zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order) * and RECLAIM_SWAP. */ p->flags |= PF_MEMALLOC | PF_SWAPWRITE; + lockdep_set_current_reclaim_state(gfp_mask); reclaim_state.reclaimed_slab = 0; p->reclaim_state = &reclaim_state; @@ -2593,6 +2631,7 @@ static int __zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order) p->reclaim_state = NULL; current->flags &= ~(PF_MEMALLOC | PF_SWAPWRITE); + lockdep_clear_current_reclaim_state(); return sc.nr_reclaimed >= nr_pages; } @@ -2615,7 +2654,7 @@ int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order) zone_page_state(zone, NR_SLAB_RECLAIMABLE) <= zone->min_slab_pages) return ZONE_RECLAIM_FULL; - if (zone_is_all_unreclaimable(zone)) + if (zone->all_unreclaimable) return ZONE_RECLAIM_FULL; /* diff --git a/mm/vmstat.c b/mm/vmstat.c index fc5aa183bc45..7f760cbc73f3 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -763,7 +763,7 @@ static void zoneinfo_show_print(struct seq_file *m, pg_data_t *pgdat, "\n prev_priority: %i" "\n start_pfn: %lu" "\n inactive_ratio: %u", - zone_is_all_unreclaimable(zone), + zone->all_unreclaimable, zone->prev_priority, zone->zone_start_pfn, zone->inactive_ratio); diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 6dcf8c9c784c..8420a4205b76 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -133,7 +133,7 @@ svc_pool_map_choose_mode(void) return SVC_POOL_PERNODE; } - node = any_online_node(node_online_map); + node = first_online_node; if (nr_cpus_node(node) > 2) { /* * Non-trivial SMP, or CONFIG_NUMA on diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index 7d1f9e928f69..8f0f1fb3dc52 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -173,11 +173,13 @@ static struct svc_xprt *__svc_xpo_create(struct svc_xprt_class *xcl, .sin_addr.s_addr = htonl(INADDR_ANY), .sin_port = htons(port), }; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) struct sockaddr_in6 sin6 = { .sin6_family = AF_INET6, .sin6_addr = IN6ADDR_ANY_INIT, .sin6_port = htons(port), }; +#endif /* defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) */ struct sockaddr *sap; size_t len; @@ -186,10 +188,12 @@ static struct svc_xprt *__svc_xpo_create(struct svc_xprt_class *xcl, sap = (struct sockaddr *)&sin; len = sizeof(sin); break; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) case PF_INET6: sap = (struct sockaddr *)&sin6; len = sizeof(sin6); break; +#endif /* defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) */ default: return ERR_PTR(-EAFNOSUPPORT); } @@ -231,7 +235,10 @@ int svc_create_xprt(struct svc_serv *serv, const char *xprt_name, err: spin_unlock(&svc_xprt_class_lock); dprintk("svc: transport %s not found\n", xprt_name); - return -ENOENT; + + /* This errno is exposed to user space. Provide a reasonable + * perror msg for a bad transport. */ + return -EPROTONOSUPPORT; } EXPORT_SYMBOL_GPL(svc_create_xprt); @@ -699,8 +706,10 @@ int svc_recv(struct svc_rqst *rqstp, long timeout) spin_unlock_bh(&pool->sp_lock); len = 0; - if (test_bit(XPT_LISTENER, &xprt->xpt_flags) && - !test_bit(XPT_CLOSE, &xprt->xpt_flags)) { + if (test_bit(XPT_CLOSE, &xprt->xpt_flags)) { + dprintk("svc_recv: found XPT_CLOSE\n"); + svc_delete_xprt(xprt); + } else if (test_bit(XPT_LISTENER, &xprt->xpt_flags)) { struct svc_xprt *newxpt; newxpt = xprt->xpt_ops->xpo_accept(xprt); if (newxpt) { @@ -726,7 +735,7 @@ int svc_recv(struct svc_rqst *rqstp, long timeout) svc_xprt_received(newxpt); } svc_xprt_received(xprt); - } else if (!test_bit(XPT_CLOSE, &xprt->xpt_flags)) { + } else { dprintk("svc: server %p, pool %u, transport %p, inuse=%d\n", rqstp, pool->sp_id, xprt, atomic_read(&xprt->xpt_ref.refcount)); @@ -739,11 +748,6 @@ int svc_recv(struct svc_rqst *rqstp, long timeout) dprintk("svc: got len=%d\n", len); } - if (test_bit(XPT_CLOSE, &xprt->xpt_flags)) { - dprintk("svc_recv: found XPT_CLOSE\n"); - svc_delete_xprt(xprt); - } - /* No data, incomplete (TCP) read, or accept() */ if (len == 0 || len == -EAGAIN) { rqstp->rq_res.len = 0; @@ -889,11 +893,8 @@ void svc_delete_xprt(struct svc_xprt *xprt) if (test_bit(XPT_TEMP, &xprt->xpt_flags)) serv->sv_tmpcnt--; - for (dr = svc_deferred_dequeue(xprt); dr; - dr = svc_deferred_dequeue(xprt)) { - svc_xprt_put(xprt); + while ((dr = svc_deferred_dequeue(xprt)) != NULL) kfree(dr); - } svc_xprt_put(xprt); spin_unlock_bh(&serv->sv_lock); diff --git a/net/sunrpc/svcauth_unix.c b/net/sunrpc/svcauth_unix.c index d8c041114497..afdcb0459a83 100644 --- a/net/sunrpc/svcauth_unix.c +++ b/net/sunrpc/svcauth_unix.c @@ -15,6 +15,7 @@ #include <linux/kernel.h> #define RPCDBG_FACILITY RPCDBG_AUTH +#include <linux/sunrpc/clnt.h> /* * AUTHUNIX and AUTHNULL credentials are both handled here. @@ -187,10 +188,13 @@ static int ip_map_parse(struct cache_detail *cd, * for scratch: */ char *buf = mesg; int len; - int b1, b2, b3, b4, b5, b6, b7, b8; - char c; char class[8]; - struct in6_addr addr; + union { + struct sockaddr sa; + struct sockaddr_in s4; + struct sockaddr_in6 s6; + } address; + struct sockaddr_in6 sin6; int err; struct ip_map *ipmp; @@ -209,24 +213,24 @@ static int ip_map_parse(struct cache_detail *cd, len = qword_get(&mesg, buf, mlen); if (len <= 0) return -EINVAL; - if (sscanf(buf, "%u.%u.%u.%u%c", &b1, &b2, &b3, &b4, &c) == 4) { - addr.s6_addr32[0] = 0; - addr.s6_addr32[1] = 0; - addr.s6_addr32[2] = htonl(0xffff); - addr.s6_addr32[3] = - htonl((((((b1<<8)|b2)<<8)|b3)<<8)|b4); - } else if (sscanf(buf, "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x%c", - &b1, &b2, &b3, &b4, &b5, &b6, &b7, &b8, &c) == 8) { - addr.s6_addr16[0] = htons(b1); - addr.s6_addr16[1] = htons(b2); - addr.s6_addr16[2] = htons(b3); - addr.s6_addr16[3] = htons(b4); - addr.s6_addr16[4] = htons(b5); - addr.s6_addr16[5] = htons(b6); - addr.s6_addr16[6] = htons(b7); - addr.s6_addr16[7] = htons(b8); - } else + if (rpc_pton(buf, len, &address.sa, sizeof(address)) == 0) return -EINVAL; + switch (address.sa.sa_family) { + case AF_INET: + /* Form a mapped IPv4 address in sin6 */ + memset(&sin6, 0, sizeof(sin6)); + sin6.sin6_family = AF_INET6; + sin6.sin6_addr.s6_addr32[2] = htonl(0xffff); + sin6.sin6_addr.s6_addr32[3] = address.s4.sin_addr.s_addr; + break; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + case AF_INET6: + memcpy(&sin6, &address.s6, sizeof(sin6)); + break; +#endif + default: + return -EINVAL; + } expiry = get_expiry(&mesg); if (expiry ==0) @@ -243,7 +247,8 @@ static int ip_map_parse(struct cache_detail *cd, } else dom = NULL; - ipmp = ip_map_lookup(class, &addr); + /* IPv6 scope IDs are ignored for now */ + ipmp = ip_map_lookup(class, &sin6.sin6_addr); if (ipmp) { err = ip_map_update(ipmp, container_of(dom, struct unix_domain, h), @@ -619,7 +624,7 @@ static int unix_gid_show(struct seq_file *m, else glen = 0; - seq_printf(m, "%d %d:", ug->uid, glen); + seq_printf(m, "%u %d:", ug->uid, glen); for (i = 0; i < glen; i++) seq_printf(m, " %d", GROUP_AT(ug->gi, i)); seq_printf(m, "\n"); diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 870929e08e5d..a29f259204e6 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -968,6 +968,7 @@ static int svc_tcp_recv_record(struct svc_sock *svsk, struct svc_rqst *rqstp) return len; err_delete: set_bit(XPT_CLOSE, &svsk->sk_xprt.xpt_flags); + svc_xprt_received(&svsk->sk_xprt); err_again: return -EAGAIN; } @@ -1357,7 +1358,7 @@ int svc_addsock(struct svc_serv *serv, const int fd, char *name_return, if (!so) return err; - if (so->sk->sk_family != AF_INET) + if ((so->sk->sk_family != PF_INET) && (so->sk->sk_family != PF_INET6)) err = -EAFNOSUPPORT; else if (so->sk->sk_protocol != IPPROTO_TCP && so->sk->sk_protocol != IPPROTO_UDP) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 3257d3d96767..a4d74344d805 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -145,11 +145,14 @@ our $Sparse = qr{ __kprobes| __ref }x; + +# Notes to $Attribute: +# We need \b after 'init' otherwise 'initconst' will cause a false positive in a check our $Attribute = qr{ const| __read_mostly| __kprobes| - __(?:mem|cpu|dev|)(?:initdata|init)| + __(?:mem|cpu|dev|)(?:initdata|initconst|init\b)| ____cacheline_aligned| ____cacheline_aligned_in_smp| ____cacheline_internodealigned_in_smp| @@ -189,6 +192,14 @@ our $typeTypedefs = qr{(?x: atomic_t )}; +our $logFunctions = qr{(?x: + printk| + pr_(debug|dbg|vdbg|devel|info|warning|err|notice|alert|crit|emerg|cont)| + dev_(printk|dbg|vdbg|info|warn|err|notice|alert|crit|emerg|WARN)| + WARN| + panic +)}; + our @typeList = ( qr{void}, qr{(?:unsigned\s+)?char}, @@ -1377,12 +1388,17 @@ sub process { #80 column limit if ($line =~ /^\+/ && $prevrawline !~ /\/\*\*/ && $rawline !~ /^.\s*\*\s*\@$Ident\s/ && - $line !~ /^\+\s*printk\s*\(\s*(?:KERN_\S+\s*)?"[X\t]*"\s*(?:,|\)\s*;)\s*$/ && + $line !~ /^\+\s*$logFunctions\s*\(\s*(?:KERN_\S+\s*)?"[X\t]*"\s*(?:,|\)\s*;)\s*$/ && $length > 80) { WARN("line over 80 characters\n" . $herecurr); } +# check for spaces before a quoted newline + if ($rawline =~ /^.*\".*\s\\n/) { + WARN("unnecessary whitespace before a quoted newline\n" . $herecurr); + } + # check for adding lines without a newline. if ($line =~ /^\+/ && defined $lines[$linenr] && $lines[$linenr] =~ /^\\ No newline at end of file/) { WARN("adding a line without newline at end of file\n" . $herecurr); @@ -1411,6 +1427,12 @@ sub process { ERROR("code indent should use tabs where possible\n" . $herevet); } +# check for space before tabs. + if ($rawline =~ /^\+/ && $rawline =~ / \t/) { + my $herevet = "$here\n" . cat_vet($rawline) . "\n"; + WARN("please, no space before tabs\n" . $herevet); + } + # check we are in a valid C source file if not then ignore this hunk next if ($realfile !~ /\.(h|c)$/); @@ -2182,8 +2204,10 @@ sub process { # Find out how long the conditional actually is. my @newlines = ($c =~ /\n/gs); my $cond_lines = 1 + $#newlines; + my $stat_real = ''; - my $stat_real = raw_line($linenr, $cond_lines); + $stat_real = raw_line($linenr, $cond_lines) + . "\n" if ($cond_lines); if (defined($stat_real) && $cond_lines > 1) { $stat_real = "[...]\n$stat_real"; } @@ -2348,6 +2372,8 @@ sub process { DECLARE_PER_CPU| DEFINE_PER_CPU| __typeof__\(| + union| + struct| \.$Ident\s*=\s*| ^\"|\"$ }x; @@ -2572,6 +2598,11 @@ sub process { WARN("plain inline is preferred over $1\n" . $herecurr); } +# check for sizeof(&) + if ($line =~ /\bsizeof\s*\(\s*\&/) { + WARN("sizeof(& should be avoided\n" . $herecurr); + } + # check for new externs in .c files. if ($realfile =~ /\.c$/ && defined $stat && $stat =~ /^.\s*(?:extern\s+)?$Type\s+($Ident)(\s*)\(/s) @@ -2634,9 +2665,46 @@ sub process { if ($line =~ /^.\s*__initcall\s*\(/) { WARN("please use device_initcall() instead of __initcall()\n" . $herecurr); } -# check for struct file_operations, ensure they are const. +# check for various ops structs, ensure they are const. + my $struct_ops = qr{acpi_dock_ops| + address_space_operations| + backlight_ops| + block_device_operations| + dentry_operations| + dev_pm_ops| + dma_map_ops| + extent_io_ops| + file_lock_operations| + file_operations| + hv_ops| + ide_dma_ops| + intel_dvo_dev_ops| + item_operations| + iwl_ops| + kgdb_arch| + kgdb_io| + kset_uevent_ops| + lock_manager_operations| + microcode_ops| + mtrr_ops| + neigh_ops| + nlmsvc_binding| + pci_raw_ops| + pipe_buf_operations| + platform_hibernation_ops| + platform_suspend_ops| + proto_ops| + rpc_pipe_ops| + seq_operations| + snd_ac97_build_ops| + soc_pcmcia_socket_ops| + stacktrace_ops| + sysfs_ops| + tty_operations| + usb_mon_operations| + wd_ops}x; if ($line !~ /\bconst\b/ && - $line =~ /\bstruct\s+(file_operations|seq_operations)\b/) { + $line =~ /\bstruct\s+($struct_ops)\b/) { WARN("struct $1 should normally be const\n" . $herecurr); } diff --git a/scripts/get_maintainer.pl b/scripts/get_maintainer.pl index 2f3230db7ffb..f76f3d13276d 100755 --- a/scripts/get_maintainer.pl +++ b/scripts/get_maintainer.pl @@ -41,6 +41,8 @@ my $web = 0; my $subsystem = 0; my $status = 0; my $keywords = 1; +my $sections = 0; +my $file_emails = 0; my $from_filename = 0; my $pattern_depth = 0; my $version = 0; @@ -120,9 +122,11 @@ if (!GetOptions( 'web!' => \$web, 'pattern-depth=i' => \$pattern_depth, 'k|keywords!' => \$keywords, + 'sections!' => \$sections, + 'fe|file-emails!' => \$file_emails, 'f|file' => \$from_filename, 'v|version' => \$version, - 'h|help' => \$help, + 'h|help|usage' => \$help, )) { die "$P: invalid argument - use --help if necessary\n"; } @@ -137,9 +141,9 @@ if ($version != 0) { exit 0; } -if ($#ARGV < 0) { - usage(); - die "$P: argument missing: patchfile or -f file please\n"; +if (-t STDIN && !@ARGV) { + # We're talking to a terminal, but have no command line arguments. + die "$P: missing patchfile or -f file - use --help if necessary\n"; } if ($output_separator ne ", ") { @@ -150,16 +154,24 @@ if ($output_rolestats) { $output_roles = 1; } -my $selections = $email + $scm + $status + $subsystem + $web; -if ($selections == 0) { - usage(); - die "$P: Missing required option: email, scm, status, subsystem or web\n"; +if ($sections) { + $email = 0; + $email_list = 0; + $scm = 0; + $status = 0; + $subsystem = 0; + $web = 0; + $keywords = 0; +} else { + my $selections = $email + $scm + $status + $subsystem + $web; + if ($selections == 0) { + die "$P: Missing required option: email, scm, status, subsystem or web\n"; + } } if ($email && ($email_maintainer + $email_list + $email_subscriber_list + $email_git + $email_git_penguin_chiefs + $email_git_blame) == 0) { - usage(); die "$P: Please select at least 1 email option\n"; } @@ -173,8 +185,9 @@ if (!top_of_kernel_tree($lk_path)) { my @typevalue = (); my %keyword_hash; -open(MAINT, "<${lk_path}MAINTAINERS") || die "$P: Can't open MAINTAINERS\n"; -while (<MAINT>) { +open (my $maint, '<', "${lk_path}MAINTAINERS") + or die "$P: Can't open MAINTAINERS: $!\n"; +while (<$maint>) { my $line = $_; if ($line =~ m/^(\C):\s*(.*)/) { @@ -199,13 +212,14 @@ while (<MAINT>) { push(@typevalue, $line); } } -close(MAINT); +close($maint); my %mailmap; if ($email_remove_duplicates) { - open(MAILMAP, "<${lk_path}.mailmap") || warn "$P: Can't open .mailmap\n"; - while (<MAILMAP>) { + open(my $mailmap, '<', "${lk_path}.mailmap") + or warn "$P: Can't open .mailmap: $!\n"; + while (<$mailmap>) { my $line = $_; next if ($line =~ m/^\s*#/); @@ -224,7 +238,7 @@ if ($email_remove_duplicates) { $mailmap{$name} = \@arr; } } - close(MAILMAP); + close($mailmap); } ## use the filenames on the command line or find the filenames in the patchfiles @@ -232,31 +246,47 @@ if ($email_remove_duplicates) { my @files = (); my @range = (); my @keyword_tvi = (); +my @file_emails = (); + +if (!@ARGV) { + push(@ARGV, "&STDIN"); +} foreach my $file (@ARGV) { - ##if $file is a directory and it lacks a trailing slash, add one - if ((-d $file)) { - $file =~ s@([^/])$@$1/@; - } elsif (!(-f $file)) { - die "$P: file '${file}' not found\n"; + if ($file ne "&STDIN") { + ##if $file is a directory and it lacks a trailing slash, add one + if ((-d $file)) { + $file =~ s@([^/])$@$1/@; + } elsif (!(-f $file)) { + die "$P: file '${file}' not found\n"; + } } if ($from_filename) { push(@files, $file); - if (-f $file && $keywords) { - open(FILE, "<$file") or die "$P: Can't open ${file}\n"; - my $text = do { local($/) ; <FILE> }; - foreach my $line (keys %keyword_hash) { - if ($text =~ m/$keyword_hash{$line}/x) { - push(@keyword_tvi, $line); + if (-f $file && ($keywords || $file_emails)) { + open(my $f, '<', $file) + or die "$P: Can't open $file: $!\n"; + my $text = do { local($/) ; <$f> }; + close($f); + if ($keywords) { + foreach my $line (keys %keyword_hash) { + if ($text =~ m/$keyword_hash{$line}/x) { + push(@keyword_tvi, $line); + } } } - close(FILE); + if ($file_emails) { + my @poss_addr = $text =~ m$[A-Za-zÀ-ÿ\"\' \,\.\+-]*\s*[\,]*\s*[\(\<\{]{0,1}[A-Za-z0-9_\.\+-]+\@[A-Za-z0-9\.-]+\.[A-Za-z0-9]+[\)\>\}]{0,1}$g; + push(@file_emails, clean_file_emails(@poss_addr)); + } } } else { my $file_cnt = @files; my $lastfile; - open(PATCH, "<$file") or die "$P: Can't open ${file}\n"; - while (<PATCH>) { + + open(my $patch, '<', $file) + or die "$P: Can't open $file: $!\n"; + while (<$patch>) { my $patch_line = $_; if (m/^\+\+\+\s+(\S+)/) { my $filename = $1; @@ -276,7 +306,8 @@ foreach my $file (@ARGV) { } } } - close(PATCH); + close($patch); + if ($file_cnt == @files) { warn "$P: file '${file}' doesn't appear to be a patch. " . "Add -f to options?\n"; @@ -285,6 +316,8 @@ foreach my $file (@ARGV) { } } +@file_emails = uniq(@file_emails); + my @email_to = (); my @list_to = (); my @scm = (); @@ -314,6 +347,7 @@ foreach my $file (@files) { if ($type eq 'X') { if (file_match_pattern($file, $value)) { $exclude = 1; + last; } } } @@ -340,12 +374,28 @@ foreach my $file (@files) { } } - $tvi += ($end - $start); - + $tvi = $end + 1; } foreach my $line (sort {$hash{$b} <=> $hash{$a}} keys %hash) { add_categories($line); + if ($sections) { + my $i; + my $start = find_starting_index($line); + my $end = find_ending_index($line); + for ($i = $start; $i < $end; $i++) { + my $line = $typevalue[$i]; + if ($line =~ /^[FX]:/) { ##Restore file patterns + $line =~ s/([^\\])\.([^\*])/$1\?$2/g; + $line =~ s/([^\\])\.$/$1\?/g; ##Convert . back to ? + $line =~ s/\\\./\./g; ##Convert \. to . + $line =~ s/\.\*/\*/g; ##Convert .* to * + } + $line =~ s/^([A-Z]):/$1:\t/g; + print("$line\n"); + } + print("\n"); + } } if ($email && $email_git) { @@ -377,6 +427,14 @@ if ($email) { } } } + + foreach my $email (@file_emails) { + my ($name, $address) = parse_email($email); + + my $tmp_email = format_email($name, $address, $email_usename); + push_email_address($tmp_email, ''); + add_role($tmp_email, 'in file'); + } } if ($email || $email_list) { @@ -453,6 +511,7 @@ MAINTAINER field selection options: --remove-duplicates => minimize duplicate email names/addresses --roles => show roles (status:subsystem, git-signer, list, etc...) --rolestats => show roles and statistics (commits/total_commits, %) + --file-emails => add email addresses found in -f file (default: 0 (off)) --scm => print SCM tree(s) if any --status => print status if any --subsystem => print subsystem name if any @@ -466,6 +525,7 @@ Output type options: Other options: --pattern-depth => Number of pattern directory traversals (default: 0 (all)) --keywords => scan patch for keywords (default: 1 (on)) + --sections => print the entire subsystem sections with pattern matches --version => show version --help => show this help information @@ -545,7 +605,7 @@ sub parse_email { $name =~ s/^\"|\"$//g; $address =~ s/^\s+|\s+$//g; - if ($name =~ /[^a-z0-9 \.\-]/i) { ##has "must quote" chars + if ($name =~ /[^\w \-]/i) { ##has "must quote" chars $name =~ s/(?<!\\)"/\\"/g; ##escape quotes $name = "\"$name\""; } @@ -562,7 +622,7 @@ sub format_email { $name =~ s/^\"|\"$//g; $address =~ s/^\s+|\s+$//g; - if ($name =~ /[^a-z0-9 \.\-]/i) { ##has "must quote" chars + if ($name =~ /[^\w \-]/i) { ##has "must quote" chars $name =~ s/(?<!\\)"/\\"/g; ##escape quotes $name = "\"$name\""; } @@ -811,7 +871,9 @@ sub add_role { foreach my $entry (@email_to) { if ($email_remove_duplicates) { my ($entry_name, $entry_address) = parse_email($entry->[0]); - if ($name eq $entry_name || $address eq $entry_address) { + if (($name eq $entry_name || $address eq $entry_address) + && ($role eq "" || !($entry->[1] =~ m/$role/)) + ) { if ($entry->[1] eq "") { $entry->[1] = "$role"; } else { @@ -819,7 +881,9 @@ sub add_role { } } } else { - if ($email eq $entry->[0]) { + if ($email eq $entry->[0] + && ($role eq "" || !($entry->[1] =~ m/$role/)) + ) { if ($entry->[1] eq "") { $entry->[1] = "$role"; } else { @@ -1099,6 +1163,51 @@ sub sort_and_uniq { return @parms; } +sub clean_file_emails { + my (@file_emails) = @_; + my @fmt_emails = (); + + foreach my $email (@file_emails) { + $email =~ s/[\(\<\{]{0,1}([A-Za-z0-9_\.\+-]+\@[A-Za-z0-9\.-]+)[\)\>\}]{0,1}/\<$1\>/g; + my ($name, $address) = parse_email($email); + if ($name eq '"[,\.]"') { + $name = ""; + } + + my @nw = split(/[^A-Za-zÀ-ÿ\'\,\.\+-]/, $name); + if (@nw > 2) { + my $first = $nw[@nw - 3]; + my $middle = $nw[@nw - 2]; + my $last = $nw[@nw - 1]; + + if (((length($first) == 1 && $first =~ m/[A-Za-z]/) || + (length($first) == 2 && substr($first, -1) eq ".")) || + (length($middle) == 1 || + (length($middle) == 2 && substr($middle, -1) eq "."))) { + $name = "$first $middle $last"; + } else { + $name = "$middle $last"; + } + } + + if (substr($name, -1) =~ /[,\.]/) { + $name = substr($name, 0, length($name) - 1); + } elsif (substr($name, -2) =~ /[,\.]"/) { + $name = substr($name, 0, length($name) - 2) . '"'; + } + + if (substr($name, 0, 1) =~ /[,\.]/) { + $name = substr($name, 1, length($name) - 1); + } elsif (substr($name, 0, 2) =~ /"[,\.]/) { + $name = '"' . substr($name, 2, length($name) - 2); + } + + my $fmt_email = format_email($name, $address, $email_usename); + push(@fmt_emails, $fmt_email); + } + return @fmt_emails; +} + sub merge_email { my @lines; my %saw; @@ -1183,7 +1292,7 @@ sub rfc822_strip_comments { # valid: returns true if the parameter is an RFC822 valid address # -sub rfc822_valid ($) { +sub rfc822_valid { my $s = rfc822_strip_comments(shift); if (!$rfc822re) { @@ -1203,7 +1312,7 @@ sub rfc822_valid ($) { # from success with no addresses found, because an empty string is # a valid list. -sub rfc822_validlist ($) { +sub rfc822_validlist { my $s = rfc822_strip_comments(shift); if (!$rfc822re) { diff --git a/sound/soc/codecs/uda1380.c b/sound/soc/codecs/uda1380.c index a2763c2e7348..9cd0a66b7663 100644 --- a/sound/soc/codecs/uda1380.c +++ b/sound/soc/codecs/uda1380.c @@ -137,7 +137,7 @@ static void uda1380_flush_work(struct work_struct *work) { int bit, reg; - for_each_bit(bit, &uda1380_cache_dirty, UDA1380_CACHEREGNUM - 0x10) { + for_each_set_bit(bit, &uda1380_cache_dirty, UDA1380_CACHEREGNUM - 0x10) { reg = 0x10 + bit; pr_debug("uda1380: flush reg %x val %x:\n", reg, uda1380_read_reg_cache(uda1380_codec, reg)); |