From 32e688b8c1afafa389223a4813b97e8c128a1636 Mon Sep 17 00:00:00 2001 From: Rob Landley Date: Mon, 15 Mar 2010 15:21:31 +0100 Subject: Documentation/filesystems/proc.txt typo fix. Typo fix for a filename in procfs documentation. Signed-off-by: Rob Landley Signed-off-by: Jiri Kosina --- Documentation/filesystems/proc.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'Documentation/filesystems') diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt index a4f30faa4f1f..770700317c2c 100644 --- a/Documentation/filesystems/proc.txt +++ b/Documentation/filesystems/proc.txt @@ -305,7 +305,7 @@ Table 1-4: Contents of the stat files (as of 2.6.30-rc7) cgtime guest time of the task children in jiffies .............................................................................. -The /proc/PID/map file containing the currently mapped memory regions and +The /proc/PID/maps file containing the currently mapped memory regions and their access permissions. The format is: -- cgit v1.2.3 From 92d6b71ab906be706f3679353b30a8d2c3831144 Mon Sep 17 00:00:00 2001 From: Dimitri Sivanich Date: Thu, 11 Mar 2010 14:08:56 -0800 Subject: genirq: Expose irq_desc->node in proc/irq Expose irq_desc->node as /proc/irq/*/node. This file provides device hardware locality information for apps desiring to include hardware locality in irq mapping decisions. Signed-off-by: Dimitri Sivanich Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Thomas Gleixner --- Documentation/filesystems/proc.txt | 4 ++++ kernel/irq/proc.c | 23 +++++++++++++++++++++++ 2 files changed, 27 insertions(+) (limited to 'Documentation/filesystems') diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt index a4f30faa4f1f..6507d2ae5236 100644 --- a/Documentation/filesystems/proc.txt +++ b/Documentation/filesystems/proc.txt @@ -566,6 +566,10 @@ The default_smp_affinity mask applies to all non-active IRQs, which are the IRQs which have not yet been allocated/activated, and hence which lack a /proc/irq/[0-9]* directory. +The node file on an SMP system shows the node to which the device using the IRQ +reports itself as being attached. This hardware locality information does not +include information about any possible driver locality preference. + prof_cpu_mask specifies which CPUs are to be profiled by the system wide profiler. Default value is ffffffff (all cpus). diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c index 6f50eccc79c0..e346e08f5c34 100644 --- a/kernel/irq/proc.c +++ b/kernel/irq/proc.c @@ -146,6 +146,26 @@ static const struct file_operations default_affinity_proc_fops = { .release = single_release, .write = default_affinity_write, }; + +static int irq_node_proc_show(struct seq_file *m, void *v) +{ + struct irq_desc *desc = irq_to_desc((long) m->private); + + seq_printf(m, "%d\n", desc->node); + return 0; +} + +static int irq_node_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, irq_node_proc_show, PDE(inode)->data); +} + +static const struct file_operations irq_node_proc_fops = { + .open = irq_node_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; #endif static int irq_spurious_proc_show(struct seq_file *m, void *v) @@ -230,6 +250,9 @@ void register_irq_proc(unsigned int irq, struct irq_desc *desc) /* create /proc/irq//smp_affinity */ proc_create_data("smp_affinity", 0600, desc->dir, &irq_affinity_proc_fops, (void *)(long)irq); + + proc_create_data("node", 0444, desc->dir, + &irq_node_proc_fops, (void *)(long)irq); #endif proc_create_data("spurious", 0444, desc->dir, -- cgit v1.2.3 From 4cb947b59c5835783fb96aad2f7d92b1e4250aff Mon Sep 17 00:00:00 2001 From: Andrea Gelmini Date: Thu, 25 Mar 2010 11:04:49 +0100 Subject: GFS2: docs update Now http://sources.redhat.com/cluster/ is redirected to http://sources.redhat.com/cluster/wiki/ Also fixed tabs in the end. Signed-off-by: Andrea Gelmini Signed-off-by: Steven Whitehouse --- Documentation/filesystems/gfs2.txt | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'Documentation/filesystems') diff --git a/Documentation/filesystems/gfs2.txt b/Documentation/filesystems/gfs2.txt index 5e3ab8f3beff..0b59c0200912 100644 --- a/Documentation/filesystems/gfs2.txt +++ b/Documentation/filesystems/gfs2.txt @@ -1,7 +1,7 @@ Global File System ------------------ -http://sources.redhat.com/cluster/ +http://sources.redhat.com/cluster/wiki/ GFS is a cluster file system. It allows a cluster of computers to simultaneously use a block device that is shared between them (with FC, @@ -36,11 +36,11 @@ GFS2 is not on-disk compatible with previous versions of GFS, but it is pretty close. The following man pages can be found at the URL above: - fsck.gfs2 to repair a filesystem - gfs2_grow to expand a filesystem online - gfs2_jadd to add journals to a filesystem online - gfs2_tool to manipulate, examine and tune a filesystem + fsck.gfs2 to repair a filesystem + gfs2_grow to expand a filesystem online + gfs2_jadd to add journals to a filesystem online + gfs2_tool to manipulate, examine and tune a filesystem gfs2_quota to examine and change quota values in a filesystem gfs2_convert to convert a gfs filesystem to gfs2 in-place mount.gfs2 to help mount(8) mount a filesystem - mkfs.gfs2 to make a filesystem + mkfs.gfs2 to make a filesystem -- cgit v1.2.3 From a33f32244d8550da8b4a26e277ce07d5c6d158b5 Mon Sep 17 00:00:00 2001 From: Francis Galiegue Date: Fri, 23 Apr 2010 00:08:02 +0200 Subject: Documentation/: it's -> its where appropriate Fix obvious cases of "it's" being used when "its" was meant. Signed-off-by: Francis Galiegue Acked-by: Randy Dunlap Signed-off-by: Jiri Kosina --- Documentation/ABI/testing/sysfs-devices-memory | 2 +- Documentation/DMA-API-HOWTO.txt | 2 +- Documentation/DocBook/libata.tmpl | 2 +- Documentation/PCI/pci-error-recovery.txt | 4 ++-- Documentation/Smack.txt | 2 +- Documentation/arm/SA1100/ADSBitsy | 2 +- Documentation/arm/Sharp-LH/ADC-LH7-Touchscreen | 2 +- Documentation/atomic_ops.txt | 2 +- Documentation/blackfin/bfin-gpio-notes.txt | 2 +- Documentation/cachetlb.txt | 6 +++--- Documentation/cgroups/memcg_test.txt | 2 +- Documentation/cgroups/memory.txt | 2 +- Documentation/connector/connector.txt | 2 +- Documentation/dvb/ci.txt | 2 +- Documentation/dvb/contributors.txt | 2 +- Documentation/filesystems/autofs4-mount-control.txt | 2 +- Documentation/filesystems/ceph.txt | 2 +- Documentation/filesystems/dlmfs.txt | 2 +- Documentation/filesystems/fiemap.txt | 12 ++++++------ Documentation/filesystems/fuse.txt | 4 ++-- Documentation/filesystems/hpfs.txt | 2 +- Documentation/filesystems/nfs/rpc-cache.txt | 2 +- Documentation/filesystems/proc.txt | 4 ++-- Documentation/filesystems/smbfs.txt | 2 +- Documentation/filesystems/vfs.txt | 2 +- Documentation/hwmon/lm85 | 2 +- Documentation/input/joystick.txt | 2 +- Documentation/intel_txt.txt | 2 +- Documentation/kbuild/kconfig-language.txt | 2 +- Documentation/kernel-docs.txt | 10 +++++----- Documentation/kprobes.txt | 2 +- Documentation/laptops/laptop-mode.txt | 2 +- Documentation/lguest/lguest.c | 2 +- Documentation/md.txt | 2 +- Documentation/netlabel/lsm_interface.txt | 2 +- Documentation/networking/ifenslave.c | 2 +- Documentation/networking/packet_mmap.txt | 4 ++-- Documentation/power/regulator/consumer.txt | 10 +++++----- Documentation/power/regulator/machine.txt | 2 +- Documentation/power/regulator/overview.txt | 6 +++--- Documentation/powerpc/booting-without-of.txt | 2 +- Documentation/powerpc/phyp-assisted-dump.txt | 2 +- Documentation/rt-mutex-design.txt | 2 +- Documentation/scsi/ChangeLog.lpfc | 4 ++-- Documentation/scsi/FlashPoint.txt | 2 +- Documentation/scsi/dtc3x80.txt | 2 +- Documentation/scsi/ncr53c8xx.txt | 2 +- Documentation/scsi/osst.txt | 2 +- Documentation/scsi/scsi_fc_transport.txt | 4 ++-- Documentation/scsi/sym53c8xx_2.txt | 2 +- Documentation/sound/alsa/soc/dapm.txt | 4 ++-- Documentation/sound/alsa/soc/machine.txt | 2 +- Documentation/sound/alsa/soc/overview.txt | 2 +- Documentation/usb/WUSB-Design-overview.txt | 2 +- Documentation/vm/numa_memory_policy.txt | 4 ++-- Documentation/w1/w1.generic | 2 +- 56 files changed, 81 insertions(+), 81 deletions(-) (limited to 'Documentation/filesystems') diff --git a/Documentation/ABI/testing/sysfs-devices-memory b/Documentation/ABI/testing/sysfs-devices-memory index bf1627b02a03..aba7d989208c 100644 --- a/Documentation/ABI/testing/sysfs-devices-memory +++ b/Documentation/ABI/testing/sysfs-devices-memory @@ -43,7 +43,7 @@ Date: September 2008 Contact: Badari Pulavarty Description: The file /sys/devices/system/memory/memoryX/state - is read-write. When read, it's contents show the + is read-write. When read, its contents show the online/offline state of the memory section. When written, root can toggle the the online/offline state of a removable memory section (see removable file description above) diff --git a/Documentation/DMA-API-HOWTO.txt b/Documentation/DMA-API-HOWTO.txt index 52618ab069ad..2e435adfbd6b 100644 --- a/Documentation/DMA-API-HOWTO.txt +++ b/Documentation/DMA-API-HOWTO.txt @@ -742,7 +742,7 @@ failure can be determined by: Closing -This document, and the API itself, would not be in it's current +This document, and the API itself, would not be in its current form without the feedback and suggestions from numerous individuals. We would like to specifically mention, in no particular order, the following people: diff --git a/Documentation/DocBook/libata.tmpl b/Documentation/DocBook/libata.tmpl index ba9975771503..261b57bc6f08 100644 --- a/Documentation/DocBook/libata.tmpl +++ b/Documentation/DocBook/libata.tmpl @@ -490,7 +490,7 @@ void (*host_stop) (struct ata_host_set *host_set); allocates space for a legacy IDE PRD table and returns. - ->port_stop() is called after ->host_stop(). It's sole function + ->port_stop() is called after ->host_stop(). Its sole function is to release DMA/memory resources, now that they are no longer actively being used. Many drivers also free driver-private data from port at this time. diff --git a/Documentation/PCI/pci-error-recovery.txt b/Documentation/PCI/pci-error-recovery.txt index e83f2ea76415..898ded24510d 100644 --- a/Documentation/PCI/pci-error-recovery.txt +++ b/Documentation/PCI/pci-error-recovery.txt @@ -216,7 +216,7 @@ The driver should return one of the following result codes: - PCI_ERS_RESULT_NEED_RESET Driver returns this if it thinks the device is not - recoverable in it's current state and it needs a slot + recoverable in its current state and it needs a slot reset to proceed. - PCI_ERS_RESULT_DISCONNECT @@ -241,7 +241,7 @@ in working condition. The driver is not supposed to restart normal driver I/O operations at this point. It should limit itself to "probing" the device to -check it's recoverability status. If all is right, then the platform +check its recoverability status. If all is right, then the platform will call resume() once all drivers have ack'd link_reset(). Result codes: diff --git a/Documentation/Smack.txt b/Documentation/Smack.txt index 34614b4c708e..e9dab41c0fe0 100644 --- a/Documentation/Smack.txt +++ b/Documentation/Smack.txt @@ -73,7 +73,7 @@ NOTE: Smack labels are limited to 23 characters. The attr command If you don't do anything special all users will get the floor ("_") label when they log in. If you do want to log in via the hacked ssh at other labels use the attr command to set the smack value on the -home directory and it's contents. +home directory and its contents. You can add access rules in /etc/smack/accesses. They take the form: diff --git a/Documentation/arm/SA1100/ADSBitsy b/Documentation/arm/SA1100/ADSBitsy index 7197a9e958ee..f9f62e8c0719 100644 --- a/Documentation/arm/SA1100/ADSBitsy +++ b/Documentation/arm/SA1100/ADSBitsy @@ -32,7 +32,7 @@ Notes: - The flash on board is divided into 3 partitions. You should be careful to use flash on board. - It's partition is different from GraphicsClient Plus and GraphicsMaster + Its partition is different from GraphicsClient Plus and GraphicsMaster - 16bpp mode requires a different cable than what ships with the board. Contact ADS or look through the manual to wire your own. Currently, diff --git a/Documentation/arm/Sharp-LH/ADC-LH7-Touchscreen b/Documentation/arm/Sharp-LH/ADC-LH7-Touchscreen index 1e6a23fdf2fc..dc460f055647 100644 --- a/Documentation/arm/Sharp-LH/ADC-LH7-Touchscreen +++ b/Documentation/arm/Sharp-LH/ADC-LH7-Touchscreen @@ -7,7 +7,7 @@ The driver only implements a four-wire touch panel protocol. The touchscreen driver is maintenance free except for the pen-down or touch threshold. Some resistive displays and board combinations may -require tuning of this threshold. The driver exposes some of it's +require tuning of this threshold. The driver exposes some of its internal state in the sys filesystem. If the kernel is configured with it, CONFIG_SYSFS, and sysfs is mounted at /sys, there will be a directory diff --git a/Documentation/atomic_ops.txt b/Documentation/atomic_ops.txt index 396bec3b74ed..ac4d47187122 100644 --- a/Documentation/atomic_ops.txt +++ b/Documentation/atomic_ops.txt @@ -320,7 +320,7 @@ counter decrement would not become globally visible until the obj->active update does. As a historical note, 32-bit Sparc used to only allow usage of -24-bits of it's atomic_t type. This was because it used 8 bits +24-bits of its atomic_t type. This was because it used 8 bits as a spinlock for SMP safety. Sparc32 lacked a "compare and swap" type instruction. However, 32-bit Sparc has since been moved over to a "hash table of spinlocks" scheme, that allows the full 32-bit diff --git a/Documentation/blackfin/bfin-gpio-notes.txt b/Documentation/blackfin/bfin-gpio-notes.txt index 9898c7ded7d3..f731c1e56475 100644 --- a/Documentation/blackfin/bfin-gpio-notes.txt +++ b/Documentation/blackfin/bfin-gpio-notes.txt @@ -43,7 +43,7 @@ void bfin_gpio_irq_free(unsigned gpio); The request functions will record the function state for a certain pin, - the free functions will clear it's function state. + the free functions will clear its function state. Once a pin is requested, it can't be requested again before it is freed by previous caller, otherwise kernel will dump stacks, and the request function fail. diff --git a/Documentation/cachetlb.txt b/Documentation/cachetlb.txt index 2b5f823abd03..9164ae3b83bc 100644 --- a/Documentation/cachetlb.txt +++ b/Documentation/cachetlb.txt @@ -5,7 +5,7 @@ This document describes the cache/tlb flushing interfaces called by the Linux VM subsystem. It enumerates over each interface, -describes it's intended purpose, and what side effect is expected +describes its intended purpose, and what side effect is expected after the interface is invoked. The side effects described below are stated for a uniprocessor @@ -231,7 +231,7 @@ require a whole different set of interfaces to handle properly. The biggest problem is that of virtual aliasing in the data cache of a processor. -Is your port susceptible to virtual aliasing in it's D-cache? +Is your port susceptible to virtual aliasing in its D-cache? Well, if your D-cache is virtually indexed, is larger in size than PAGE_SIZE, and does not prevent multiple cache lines for the same physical address from existing at once, you have this problem. @@ -249,7 +249,7 @@ one way to solve this (in particular SPARC_FLAG_MMAPSHARED). Next, you have to solve the D-cache aliasing issue for all other cases. Please keep in mind that fact that, for a given page mapped into some user address space, there is always at least one more -mapping, that of the kernel in it's linear mapping starting at +mapping, that of the kernel in its linear mapping starting at PAGE_OFFSET. So immediately, once the first user maps a given physical page into its address space, by implication the D-cache aliasing problem has the potential to exist since the kernel already diff --git a/Documentation/cgroups/memcg_test.txt b/Documentation/cgroups/memcg_test.txt index f7f68b2ac199..b7eececfb195 100644 --- a/Documentation/cgroups/memcg_test.txt +++ b/Documentation/cgroups/memcg_test.txt @@ -244,7 +244,7 @@ Under below explanation, we assume CONFIG_MEM_RES_CTRL_SWAP=y. we have to check if OLDPAGE/NEWPAGE is a valid page after commit(). 8. LRU - Each memcg has its own private LRU. Now, it's handling is under global + Each memcg has its own private LRU. Now, its handling is under global VM's control (means that it's handled under global zone->lru_lock). Almost all routines around memcg's LRU is called by global LRU's list management functions under zone->lru_lock(). diff --git a/Documentation/cgroups/memory.txt b/Documentation/cgroups/memory.txt index 3a6aecd078ba..6cab1f29da4c 100644 --- a/Documentation/cgroups/memory.txt +++ b/Documentation/cgroups/memory.txt @@ -263,7 +263,7 @@ some of the pages cached in the cgroup (page cache pages). 4.2 Task migration -When a task migrates from one cgroup to another, it's charge is not +When a task migrates from one cgroup to another, its charge is not carried forward by default. The pages allocated from the original cgroup still remain charged to it, the charge is dropped when the page is freed or reclaimed. diff --git a/Documentation/connector/connector.txt b/Documentation/connector/connector.txt index 78c9466a9aa8..e5c5f5e6ab70 100644 --- a/Documentation/connector/connector.txt +++ b/Documentation/connector/connector.txt @@ -88,7 +88,7 @@ int cn_netlink_send(struct cn_msg *msg, u32 __groups, int gfp_mask); int gfp_mask - GFP mask. Note: When registering new callback user, connector core assigns - netlink group to the user which is equal to it's id.idx. + netlink group to the user which is equal to its id.idx. /*****************************************/ Protocol description. diff --git a/Documentation/dvb/ci.txt b/Documentation/dvb/ci.txt index 2ecd834585e6..4a0c2b56e690 100644 --- a/Documentation/dvb/ci.txt +++ b/Documentation/dvb/ci.txt @@ -41,7 +41,7 @@ This application requires the following to function properly as of now. * Cards that fall in this category ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ -At present the cards that fall in this category are the Twinhan and it's +At present the cards that fall in this category are the Twinhan and its clones, these cards are available as VVMER, Tomato, Hercules, Orange and so on. diff --git a/Documentation/dvb/contributors.txt b/Documentation/dvb/contributors.txt index 4865addebe1c..47c30098dab6 100644 --- a/Documentation/dvb/contributors.txt +++ b/Documentation/dvb/contributors.txt @@ -1,7 +1,7 @@ Thanks go to the following people for patches and contributions: Michael Hunold - for the initial saa7146 driver and it's recent overhaul + for the initial saa7146 driver and its recent overhaul Christian Theiss for his work on the initial Linux DVB driver diff --git a/Documentation/filesystems/autofs4-mount-control.txt b/Documentation/filesystems/autofs4-mount-control.txt index 8f78ded4b648..51986bf08a4d 100644 --- a/Documentation/filesystems/autofs4-mount-control.txt +++ b/Documentation/filesystems/autofs4-mount-control.txt @@ -146,7 +146,7 @@ found to be inadequate, in this case. The Generic Netlink system was used for this as raw Netlink would lead to a significant increase in complexity. There's no question that the Generic Netlink system is an elegant solution for common case ioctl functions but it's not a complete -replacement probably because it's primary purpose in life is to be a +replacement probably because its primary purpose in life is to be a message bus implementation rather than specifically an ioctl replacement. While it would be possible to work around this there is one concern that lead to the decision to not use it. This is that the autofs diff --git a/Documentation/filesystems/ceph.txt b/Documentation/filesystems/ceph.txt index 0660c9f5deef..763d8ebbbebd 100644 --- a/Documentation/filesystems/ceph.txt +++ b/Documentation/filesystems/ceph.txt @@ -90,7 +90,7 @@ Mount Options Specify the IP and/or port the client should bind to locally. There is normally not much reason to do this. If the IP is not specified, the client's IP address is determined by looking at the - address it's connection to the monitor originates from. + address its connection to the monitor originates from. wsize=X Specify the maximum write size in bytes. By default there is no diff --git a/Documentation/filesystems/dlmfs.txt b/Documentation/filesystems/dlmfs.txt index c50bbb2d52b4..1b528b2ad809 100644 --- a/Documentation/filesystems/dlmfs.txt +++ b/Documentation/filesystems/dlmfs.txt @@ -47,7 +47,7 @@ You'll want to start heartbeating on a volume which all the nodes in your lockspace can access. The easiest way to do this is via ocfs2_hb_ctl (distributed with ocfs2-tools). Right now it requires that an OCFS2 file system be in place so that it can automatically -find it's heartbeat area, though it will eventually support heartbeat +find its heartbeat area, though it will eventually support heartbeat against raw disks. Please see the ocfs2_hb_ctl and mkfs.ocfs2 manual pages distributed diff --git a/Documentation/filesystems/fiemap.txt b/Documentation/filesystems/fiemap.txt index 606233cd4618..1b805a0efbb0 100644 --- a/Documentation/filesystems/fiemap.txt +++ b/Documentation/filesystems/fiemap.txt @@ -38,7 +38,7 @@ flags, it will return EBADR and the contents of fm_flags will contain the set of flags which caused the error. If the kernel is compatible with all flags passed, the contents of fm_flags will be unmodified. It is up to userspace to determine whether rejection of a particular -flag is fatal to it's operation. This scheme is intended to allow the +flag is fatal to its operation. This scheme is intended to allow the fiemap interface to grow in the future but without losing compatibility with old software. @@ -56,7 +56,7 @@ If this flag is set, the kernel will sync the file before mapping extents. * FIEMAP_FLAG_XATTR If this flag is set, the extents returned will describe the inodes -extended attribute lookup tree, instead of it's data tree. +extended attribute lookup tree, instead of its data tree. Extent Mapping @@ -89,7 +89,7 @@ struct fiemap_extent { }; All offsets and lengths are in bytes and mirror those on disk. It is valid -for an extents logical offset to start before the request or it's logical +for an extents logical offset to start before the request or its logical length to extend past the request. Unless FIEMAP_EXTENT_NOT_ALIGNED is returned, fe_logical, fe_physical, and fe_length will be aligned to the block size of the file system. With the exception of extents flagged as @@ -125,7 +125,7 @@ been allocated for the file yet. * FIEMAP_EXTENT_DELALLOC - This will also set FIEMAP_EXTENT_UNKNOWN. -Delayed allocation - while there is data for this extent, it's +Delayed allocation - while there is data for this extent, its physical location has not been allocated yet. * FIEMAP_EXTENT_ENCODED @@ -159,7 +159,7 @@ Data is located within a meta data block. Data is packed into a block with data from other files. * FIEMAP_EXTENT_UNWRITTEN -Unwritten extent - the extent is allocated but it's data has not been +Unwritten extent - the extent is allocated but its data has not been initialized. This indicates the extent's data will be all zero if read through the filesystem but the contents are undefined if read directly from the device. @@ -176,7 +176,7 @@ VFS -> File System Implementation File systems wishing to support fiemap must implement a ->fiemap callback on their inode_operations structure. The fs ->fiemap call is responsible for -defining it's set of supported fiemap flags, and calling a helper function on +defining its set of supported fiemap flags, and calling a helper function on each discovered extent: struct inode_operations { diff --git a/Documentation/filesystems/fuse.txt b/Documentation/filesystems/fuse.txt index 397a41adb4c3..13af4a49e7db 100644 --- a/Documentation/filesystems/fuse.txt +++ b/Documentation/filesystems/fuse.txt @@ -91,7 +91,7 @@ Mount options 'default_permissions' By default FUSE doesn't check file access permissions, the - filesystem is free to implement it's access policy or leave it to + filesystem is free to implement its access policy or leave it to the underlying file access mechanism (e.g. in case of network filesystems). This option enables permission checking, restricting access based on file mode. It is usually useful together with the @@ -171,7 +171,7 @@ or may honor them by sending a reply to the _original_ request, with the error set to EINTR. It is also possible that there's a race between processing the -original request and it's INTERRUPT request. There are two possibilities: +original request and its INTERRUPT request. There are two possibilities: 1) The INTERRUPT request is processed before the original request is processed diff --git a/Documentation/filesystems/hpfs.txt b/Documentation/filesystems/hpfs.txt index fa45c3baed98..74630bd504fb 100644 --- a/Documentation/filesystems/hpfs.txt +++ b/Documentation/filesystems/hpfs.txt @@ -103,7 +103,7 @@ to analyze or change OS2SYS.INI. Codepages HPFS can contain several uppercasing tables for several codepages and each -file has a pointer to codepage it's name is in. However OS/2 was created in +file has a pointer to codepage its name is in. However OS/2 was created in America where people don't care much about codepages and so multiple codepages support is quite buggy. I have Czech OS/2 working in codepage 852 on my disk. Once I booted English OS/2 working in cp 850 and I created a file on my 852 diff --git a/Documentation/filesystems/nfs/rpc-cache.txt b/Documentation/filesystems/nfs/rpc-cache.txt index 8a382bea6808..ebcaaee21616 100644 --- a/Documentation/filesystems/nfs/rpc-cache.txt +++ b/Documentation/filesystems/nfs/rpc-cache.txt @@ -185,7 +185,7 @@ failed lookup meant a definite 'no'. request/response format ----------------------- -While each cache is free to use it's own format for requests +While each cache is free to use its own format for requests and responses over channel, the following is recommended as appropriate and support routines are available to help: Each request or response record should be printable ASCII diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt index 770700317c2c..f6b1b5fca1df 100644 --- a/Documentation/filesystems/proc.txt +++ b/Documentation/filesystems/proc.txt @@ -965,7 +965,7 @@ your system and how much traffic was routed over those devices: ...] 1375103 17405 0 0 0 0 0 0 ...] 1703981 5535 0 0 0 3 0 0 -In addition, each Channel Bond interface has it's own directory. For +In addition, each Channel Bond interface has its own directory. For example, the bond0 device will have a directory called /proc/net/bond0/. It will contain information that is specific to that bond, such as the current slaves of the bond, the link status of the slaves, and how @@ -1362,7 +1362,7 @@ been accounted as having caused 1MB of write. In other words: The number of bytes which this process caused to not happen, by truncating pagecache. A task can cause "negative" IO too. If this task truncates some dirty pagecache, some IO which another task has been accounted -for (in it's write_bytes) will not be happening. We _could_ just subtract that +for (in its write_bytes) will not be happening. We _could_ just subtract that from the truncating task's write_bytes, but there is information loss in doing that. diff --git a/Documentation/filesystems/smbfs.txt b/Documentation/filesystems/smbfs.txt index f673ef0de0f7..194fb0decd2c 100644 --- a/Documentation/filesystems/smbfs.txt +++ b/Documentation/filesystems/smbfs.txt @@ -3,6 +3,6 @@ protocol used by Windows for Workgroups, Windows 95 and Windows NT. Smbfs was inspired by Samba, the program written by Andrew Tridgell that turns any Unix host into a file server for DOS or Windows clients. -Smbfs is a SMB client, but uses parts of samba for it's operation. For +Smbfs is a SMB client, but uses parts of samba for its operation. For more info on samba, including documentation, please go to http://www.samba.org/ and then on to your nearest mirror. diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt index 3de2f32edd90..b66858538df5 100644 --- a/Documentation/filesystems/vfs.txt +++ b/Documentation/filesystems/vfs.txt @@ -72,7 +72,7 @@ structure (this is the kernel-side implementation of file descriptors). The freshly allocated file structure is initialized with a pointer to the dentry and a set of file operation member functions. These are taken from the inode data. The open() file method is then -called so the specific filesystem implementation can do it's work. You +called so the specific filesystem implementation can do its work. You can see that this is another switch performed by the VFS. The file structure is placed into the file descriptor table for the process. diff --git a/Documentation/hwmon/lm85 b/Documentation/hwmon/lm85 index a13680871bc7..a76aefeeb68a 100644 --- a/Documentation/hwmon/lm85 +++ b/Documentation/hwmon/lm85 @@ -157,7 +157,7 @@ temperature configuration points: There are three PWM outputs. The LM85 datasheet suggests that the pwm3 output control both fan3 and fan4. Each PWM can be individually -configured and assigned to a zone for it's control value. Each PWM can be +configured and assigned to a zone for its control value. Each PWM can be configured individually according to the following options. * pwm#_auto_pwm_min - this specifies the PWM value for temp#_auto_temp_off diff --git a/Documentation/input/joystick.txt b/Documentation/input/joystick.txt index 154d767b2acb..8007b7ca87bf 100644 --- a/Documentation/input/joystick.txt +++ b/Documentation/input/joystick.txt @@ -402,7 +402,7 @@ for the port of the SoundFusion is supported by the cs461x.c module. ~~~~~~~~~~~~~~~~~~~~~~~~ The Live! has a special PCI gameport, which, although it doesn't provide any "Enhanced" stuff like 4DWave and friends, is quite a bit faster than -it's ISA counterparts. It also requires special support, hence the +its ISA counterparts. It also requires special support, hence the emu10k1-gp.c module for it instead of the normal ns558.c one. 3.15 SoundBlaster 64 and 128 - ES1370 and ES1371, ESS Solo1 and S3 SonicVibes diff --git a/Documentation/intel_txt.txt b/Documentation/intel_txt.txt index f40a1f030019..1423bcc7c507 100644 --- a/Documentation/intel_txt.txt +++ b/Documentation/intel_txt.txt @@ -126,7 +126,7 @@ o Tboot then applies an (optional) user-defined launch policy to o Tboot adjusts the e820 table provided by the bootloader to reserve its own location in memory as well as to reserve certain other TXT-related regions. -o As part of it's launch, tboot DMA protects all of RAM (using the +o As part of its launch, tboot DMA protects all of RAM (using the VT-d PMRs). Thus, the kernel must be booted with 'intel_iommu=on' in order to remove this blanket protection and use VT-d's page-level protection. diff --git a/Documentation/kbuild/kconfig-language.txt b/Documentation/kbuild/kconfig-language.txt index c412c245848f..b472e4e0ba67 100644 --- a/Documentation/kbuild/kconfig-language.txt +++ b/Documentation/kbuild/kconfig-language.txt @@ -181,7 +181,7 @@ Expressions are listed in decreasing order of precedence. (7) Returns the result of max(/expr/, /expr/). An expression can have a value of 'n', 'm' or 'y' (or 0, 1, 2 -respectively for calculations). A menu entry becomes visible when it's +respectively for calculations). A menu entry becomes visible when its expression evaluates to 'm' or 'y'. There are two types of symbols: constant and non-constant symbols. diff --git a/Documentation/kernel-docs.txt b/Documentation/kernel-docs.txt index 28cdc2af2131..ec8d31ee12e0 100644 --- a/Documentation/kernel-docs.txt +++ b/Documentation/kernel-docs.txt @@ -116,7 +116,7 @@ Author: Ingo Molnar, Gadi Oxman and Miguel de Icaza. URL: http://www.linuxjournal.com/article.php?sid=2391 Keywords: RAID, MD driver. - Description: Linux Journal Kernel Korner article. Here is it's + Description: Linux Journal Kernel Korner article. Here is its abstract: "A description of the implementation of the RAID-1, RAID-4 and RAID-5 personalities of the MD device driver in the Linux kernel, providing users with high performance and reliable, @@ -127,7 +127,7 @@ URL: http://www.linuxjournal.com/article.php?sid=1219 Keywords: device driver, module, loading/unloading modules, allocating resources. - Description: Linux Journal Kernel Korner article. Here is it's + Description: Linux Journal Kernel Korner article. Here is its abstract: "This is the first of a series of four articles co-authored by Alessandro Rubini and Georg Zezchwitz which present a practical approach to writing Linux device drivers as kernel @@ -141,7 +141,7 @@ Keywords: character driver, init_module, clean_up module, autodetection, mayor number, minor number, file operations, open(), close(). - Description: Linux Journal Kernel Korner article. Here is it's + Description: Linux Journal Kernel Korner article. Here is its abstract: "This article, the second of four, introduces part of the actual code to create custom module implementing a character device driver. It describes the code for module initialization and @@ -152,7 +152,7 @@ URL: http://www.linuxjournal.com/article.php?sid=1221 Keywords: read(), write(), select(), ioctl(), blocking/non blocking mode, interrupt handler. - Description: Linux Journal Kernel Korner article. Here is it's + Description: Linux Journal Kernel Korner article. Here is its abstract: "This article, the third of four on writing character device drivers, introduces concepts of reading, writing, and using ioctl-calls". @@ -161,7 +161,7 @@ Author: Alessandro Rubini and Georg v. Zezschwitz. URL: http://www.linuxjournal.com/article.php?sid=1222 Keywords: interrupts, irqs, DMA, bottom halves, task queues. - Description: Linux Journal Kernel Korner article. Here is it's + Description: Linux Journal Kernel Korner article. Here is its abstract: "This is the fourth in a series of articles about writing character device drivers as loadable kernel modules. This month, we further investigate the field of interrupt handling. diff --git a/Documentation/kprobes.txt b/Documentation/kprobes.txt index 2f9115c0ae62..51ec634ac04b 100644 --- a/Documentation/kprobes.txt +++ b/Documentation/kprobes.txt @@ -332,7 +332,7 @@ occurs during execution of kp->pre_handler or kp->post_handler, or during single-stepping of the probed instruction, Kprobes calls kp->fault_handler. Any or all handlers can be NULL. If kp->flags is set KPROBE_FLAG_DISABLED, that kp will be registered but disabled, -so, it's handlers aren't hit until calling enable_kprobe(kp). +so, its handlers aren't hit until calling enable_kprobe(kp). NOTE: 1. With the introduction of the "symbol_name" field to struct kprobe, diff --git a/Documentation/laptops/laptop-mode.txt b/Documentation/laptops/laptop-mode.txt index 2c3c35093023..0bf25eebce94 100644 --- a/Documentation/laptops/laptop-mode.txt +++ b/Documentation/laptops/laptop-mode.txt @@ -207,7 +207,7 @@ Tips & Tricks * Drew Scott Daniels observed: "I don't know why, but when I decrease the number of colours that my display uses it consumes less battery power. I've seen this on powerbooks too. I hope that this is a piece of information that - might be useful to the Laptop Mode patch or it's users." + might be useful to the Laptop Mode patch or its users." * In syslog.conf, you can prefix entries with a dash ``-'' to omit syncing the file after every logging. When you're using laptop-mode and your disk doesn't diff --git a/Documentation/lguest/lguest.c b/Documentation/lguest/lguest.c index 3119f5db75bd..e9ce3c554514 100644 --- a/Documentation/lguest/lguest.c +++ b/Documentation/lguest/lguest.c @@ -263,7 +263,7 @@ static u8 *get_feature_bits(struct device *dev) * Launcher virtual with an offset. * * This can be tough to get your head around, but usually it just means that we - * use these trivial conversion functions when the Guest gives us it's + * use these trivial conversion functions when the Guest gives us its * "physical" addresses: */ static void *from_guest_phys(unsigned long addr) diff --git a/Documentation/md.txt b/Documentation/md.txt index 188f4768f1d5..e4e893ef3e01 100644 --- a/Documentation/md.txt +++ b/Documentation/md.txt @@ -136,7 +136,7 @@ raid_disks != 0. Then uninitialized devices can be added with ADD_NEW_DISK. The structure passed to ADD_NEW_DISK must specify the state of the device -and it's role in the array. +and its role in the array. Once started with RUN_ARRAY, uninitialized spares can be added with HOT_ADD_DISK. diff --git a/Documentation/netlabel/lsm_interface.txt b/Documentation/netlabel/lsm_interface.txt index 98dd9f7430f2..638c74f7de7f 100644 --- a/Documentation/netlabel/lsm_interface.txt +++ b/Documentation/netlabel/lsm_interface.txt @@ -38,7 +38,7 @@ Depending on the exact configuration, translation between the network packet label and the internal LSM security identifier can be time consuming. The NetLabel label mapping cache is a caching mechanism which can be used to sidestep much of this overhead once a mapping has been established. Once the -LSM has received a packet, used NetLabel to decode it's security attributes, +LSM has received a packet, used NetLabel to decode its security attributes, and translated the security attributes into a LSM internal identifier the LSM can use the NetLabel caching functions to associate the LSM internal identifier with the network packet's label. This means that in the future diff --git a/Documentation/networking/ifenslave.c b/Documentation/networking/ifenslave.c index 1b96ccda3836..2bac9618c345 100644 --- a/Documentation/networking/ifenslave.c +++ b/Documentation/networking/ifenslave.c @@ -756,7 +756,7 @@ static int enslave(char *master_ifname, char *slave_ifname) */ if (abi_ver < 1) { /* For old ABI, the master needs to be - * down before setting it's hwaddr + * down before setting its hwaddr */ res = set_if_down(master_ifname, master_flags.ifr_flags); if (res) { diff --git a/Documentation/networking/packet_mmap.txt b/Documentation/networking/packet_mmap.txt index 09ab0d290326..98f71a5cef00 100644 --- a/Documentation/networking/packet_mmap.txt +++ b/Documentation/networking/packet_mmap.txt @@ -100,7 +100,7 @@ by the kernel. The destruction of the socket and all associated resources is done by a simple call to close(fd). -Next I will describe PACKET_MMAP settings and it's constraints, +Next I will describe PACKET_MMAP settings and its constraints, also the mapping of the circular buffer in the user process and the use of this buffer. @@ -432,7 +432,7 @@ TP_STATUS_LOSING : indicates there were packet drops from last time the PACKET_STATISTICS option. TP_STATUS_CSUMNOTREADY: currently it's used for outgoing IP packets which - it's checksum will be done in hardware. So while + its checksum will be done in hardware. So while reading the packet we should not try to check the checksum. diff --git a/Documentation/power/regulator/consumer.txt b/Documentation/power/regulator/consumer.txt index cdebb5145c25..55c4175d8099 100644 --- a/Documentation/power/regulator/consumer.txt +++ b/Documentation/power/regulator/consumer.txt @@ -8,11 +8,11 @@ Please see overview.txt for a description of the terms used in this text. 1. Consumer Regulator Access (static & dynamic drivers) ======================================================= -A consumer driver can get access to it's supply regulator by calling :- +A consumer driver can get access to its supply regulator by calling :- regulator = regulator_get(dev, "Vcc"); -The consumer passes in it's struct device pointer and power supply ID. The core +The consumer passes in its struct device pointer and power supply ID. The core then finds the correct regulator by consulting a machine specific lookup table. If the lookup is successful then this call will return a pointer to the struct regulator that supplies this consumer. @@ -34,7 +34,7 @@ usually be called in your device drivers probe() and remove() respectively. 2. Regulator Output Enable & Disable (static & dynamic drivers) ==================================================================== -A consumer can enable it's power supply by calling:- +A consumer can enable its power supply by calling:- int regulator_enable(regulator); @@ -49,7 +49,7 @@ int regulator_is_enabled(regulator); This will return > zero when the regulator is enabled. -A consumer can disable it's supply when no longer needed by calling :- +A consumer can disable its supply when no longer needed by calling :- int regulator_disable(regulator); @@ -140,7 +140,7 @@ by calling :- int regulator_set_optimum_mode(struct regulator *regulator, int load_uA); This will cause the core to recalculate the total load on the regulator (based -on all it's consumers) and change operating mode (if necessary and permitted) +on all its consumers) and change operating mode (if necessary and permitted) to best match the current operating load. The load_uA value can be determined from the consumers datasheet. e.g.most diff --git a/Documentation/power/regulator/machine.txt b/Documentation/power/regulator/machine.txt index 63728fed620b..bdec39b9bd75 100644 --- a/Documentation/power/regulator/machine.txt +++ b/Documentation/power/regulator/machine.txt @@ -52,7 +52,7 @@ static struct regulator_init_data regulator1_data = { }; Regulator-1 supplies power to Regulator-2. This relationship must be registered -with the core so that Regulator-1 is also enabled when Consumer A enables it's +with the core so that Regulator-1 is also enabled when Consumer A enables its supply (Regulator-2). The supply regulator is set by the supply_regulator_dev field below:- diff --git a/Documentation/power/regulator/overview.txt b/Documentation/power/regulator/overview.txt index ffd185bb6054..9363e056188a 100644 --- a/Documentation/power/regulator/overview.txt +++ b/Documentation/power/regulator/overview.txt @@ -35,16 +35,16 @@ Some terms used in this document:- o Consumer - Electronic device that is supplied power by a regulator. Consumers can be classified into two types:- - Static: consumer does not change it's supply voltage or + Static: consumer does not change its supply voltage or current limit. It only needs to enable or disable it's - power supply. It's supply voltage is set by the hardware, + power supply. Its supply voltage is set by the hardware, bootloader, firmware or kernel board initialisation code. Dynamic: consumer needs to change it's supply voltage or current limit to meet operation demands. - o Power Domain - Electronic circuit that is supplied it's input power by the + o Power Domain - Electronic circuit that is supplied its input power by the output power of a regulator, switch or by another power domain. diff --git a/Documentation/powerpc/booting-without-of.txt b/Documentation/powerpc/booting-without-of.txt index 79f533f38c61..46d22105aa07 100644 --- a/Documentation/powerpc/booting-without-of.txt +++ b/Documentation/powerpc/booting-without-of.txt @@ -1289,7 +1289,7 @@ link between a device node and its interrupt parent in the interrupt tree. The value of interrupt-parent is the phandle of the parent node. -If the interrupt-parent property is not defined for a node, it's +If the interrupt-parent property is not defined for a node, its interrupt parent is assumed to be an ancestor in the node's _device tree_ hierarchy. diff --git a/Documentation/powerpc/phyp-assisted-dump.txt b/Documentation/powerpc/phyp-assisted-dump.txt index c4682b982a2e..ad340205d96a 100644 --- a/Documentation/powerpc/phyp-assisted-dump.txt +++ b/Documentation/powerpc/phyp-assisted-dump.txt @@ -19,7 +19,7 @@ dump offers several strong, practical advantages: immediately available to the system for normal use. -- After the dump is completed, no further reboots are required; the system will be fully usable, and running - in it's normal, production mode on it normal kernel. + in its normal, production mode on its normal kernel. The above can only be accomplished by coordination with, and assistance from the hypervisor. The procedure is diff --git a/Documentation/rt-mutex-design.txt b/Documentation/rt-mutex-design.txt index 4b736d24da7a..8df0b782c4d7 100644 --- a/Documentation/rt-mutex-design.txt +++ b/Documentation/rt-mutex-design.txt @@ -657,7 +657,7 @@ here. The waiter structure has a "task" field that points to the task that is blocked on the mutex. This field can be NULL the first time it goes through the loop -or if the task is a pending owner and had it's mutex stolen. If the "task" +or if the task is a pending owner and had its mutex stolen. If the "task" field is NULL then we need to set up the accounting for it. Task blocks on mutex diff --git a/Documentation/scsi/ChangeLog.lpfc b/Documentation/scsi/ChangeLog.lpfc index 2ffc1148eb95..e759e92e286d 100644 --- a/Documentation/scsi/ChangeLog.lpfc +++ b/Documentation/scsi/ChangeLog.lpfc @@ -707,7 +707,7 @@ Changes from 20040920 to 20041018 * Integrate patches from Christoph Hellwig: two new helpers common to lpfc_sli_resume_iocb and lpfc_sli_issue_iocb - singificant cleanup of those two functions - the unused SLI_IOCB_USE_TXQ is - gone - lpfc_sli_issue_iocb_wait loses it's flags argument + gone - lpfc_sli_issue_iocb_wait loses its flags argument totally. * Fix in lpfc_sli.c: we can not store a 5 bit value in a 4-bit field. @@ -1028,7 +1028,7 @@ Changes from 20040614 to 20040709 * Remove the need for buf_tmo. * Changed ULP_BDE64 to struct ulp_bde64. * Changed ULP_BDE to struct ulp_bde. - * Cleanup lpfc_os_return_scsi_cmd() and it's call path. + * Cleanup lpfc_os_return_scsi_cmd() and its call path. * Removed lpfc_no_device_delay. * Consolidating lpfc_hba_put_event() into lpfc_put_event(). * Removed following attributes and their functionality: diff --git a/Documentation/scsi/FlashPoint.txt b/Documentation/scsi/FlashPoint.txt index d5acaa300a46..1540a92f6d2b 100644 --- a/Documentation/scsi/FlashPoint.txt +++ b/Documentation/scsi/FlashPoint.txt @@ -71,7 +71,7 @@ peters@mylex.com Ever since its introduction last October, the BusLogic FlashPoint LT has been problematic for members of the Linux community, in that no Linux -drivers have been available for this new Ultra SCSI product. Despite it's +drivers have been available for this new Ultra SCSI product. Despite its officially being positioned as a desktop workstation product, and not being particularly well suited for a high performance multitasking operating system like Linux, the FlashPoint LT has been touted by computer system diff --git a/Documentation/scsi/dtc3x80.txt b/Documentation/scsi/dtc3x80.txt index e8ae6230ab3e..1d7af9f9a8ed 100644 --- a/Documentation/scsi/dtc3x80.txt +++ b/Documentation/scsi/dtc3x80.txt @@ -12,7 +12,7 @@ The 3180 does not. Otherwise, they are identical. The DTC3x80 does not support DMA but it does have Pseudo-DMA which is supported by the driver. -It's DTC406 scsi chip is supposedly compatible with the NCR 53C400. +Its DTC406 scsi chip is supposedly compatible with the NCR 53C400. It is memory mapped, uses an IRQ, but no dma or io-port. There is internal DMA, between SCSI bus and an on-chip 128-byte buffer. Double buffering is done automagically by the chip. Data is transferred diff --git a/Documentation/scsi/ncr53c8xx.txt b/Documentation/scsi/ncr53c8xx.txt index 08e2b4d04aab..cda5f8fa2c66 100644 --- a/Documentation/scsi/ncr53c8xx.txt +++ b/Documentation/scsi/ncr53c8xx.txt @@ -1479,7 +1479,7 @@ Wide16 SCSI. Enabling serial NVRAM support enables detection of the serial NVRAM included on Symbios and some Symbios compatible host adaptors, and Tekram boards. The serial NVRAM is used by Symbios and Tekram to hold set up parameters for the -host adaptor and it's attached drives. +host adaptor and its attached drives. The Symbios NVRAM also holds data on the boot order of host adaptors in a system with more than one host adaptor. This enables the order of scanning diff --git a/Documentation/scsi/osst.txt b/Documentation/scsi/osst.txt index f536907e241d..2b21890bc983 100644 --- a/Documentation/scsi/osst.txt +++ b/Documentation/scsi/osst.txt @@ -40,7 +40,7 @@ behavior looks very much the same as st to the userspace applications. History ------- -In the first place, osst shared it's identity very much with st. That meant +In the first place, osst shared its identity very much with st. That meant that it used the same kernel structures and the same device node as st. So you could only have either of them being present in the kernel. This has been fixed by registering an own device, now. diff --git a/Documentation/scsi/scsi_fc_transport.txt b/Documentation/scsi/scsi_fc_transport.txt index aec6549ab097..e00192de4d1c 100644 --- a/Documentation/scsi/scsi_fc_transport.txt +++ b/Documentation/scsi/scsi_fc_transport.txt @@ -70,7 +70,7 @@ Overview: up to an administrative entity controlling the vport. For example, if vports are to be associated with virtual machines, a XEN mgmt utility would be responsible for creating wwpn/wwnn's for the vport, - using it's own naming authority and OUI. (Note: it already does this + using its own naming authority and OUI. (Note: it already does this for virtual MAC addresses). @@ -81,7 +81,7 @@ Device Trees and Vport Objects: with rports and scsi target objects underneath it. Currently the FC transport creates the vport object and places it under the scsi_host object corresponding to the physical adapter. The LLDD will allocate - a new scsi_host for the vport and link it's object under the vport. + a new scsi_host for the vport and link its object under the vport. The remainder of the tree under the vports scsi_host is the same as the non-NPIV case. The transport is written currently to easily allow the parent of the vport to be something other than the scsi_host. diff --git a/Documentation/scsi/sym53c8xx_2.txt b/Documentation/scsi/sym53c8xx_2.txt index eb9a7b905b64..6f63b7989679 100644 --- a/Documentation/scsi/sym53c8xx_2.txt +++ b/Documentation/scsi/sym53c8xx_2.txt @@ -687,7 +687,7 @@ maintain the driver code. Enabling serial NVRAM support enables detection of the serial NVRAM included on Symbios and some Symbios compatible host adaptors, and Tekram boards. The serial NVRAM is used by Symbios and Tekram to hold set up parameters for the -host adaptor and it's attached drives. +host adaptor and its attached drives. The Symbios NVRAM also holds data on the boot order of host adaptors in a system with more than one host adaptor. This information is no longer used diff --git a/Documentation/sound/alsa/soc/dapm.txt b/Documentation/sound/alsa/soc/dapm.txt index 9ac842be9b4f..05bf5a0eee41 100644 --- a/Documentation/sound/alsa/soc/dapm.txt +++ b/Documentation/sound/alsa/soc/dapm.txt @@ -188,8 +188,8 @@ The WM8731 output mixer has 3 inputs (sources) 3. Mic Sidetone Input Each input in this example has a kcontrol associated with it (defined in example -above) and is connected to the output mixer via it's kcontrol name. We can now -connect the destination widget (wrt audio signal) with it's source widgets. +above) and is connected to the output mixer via its kcontrol name. We can now +connect the destination widget (wrt audio signal) with its source widgets. /* output mixer */ {"Output Mixer", "Line Bypass Switch", "Line Input"}, diff --git a/Documentation/sound/alsa/soc/machine.txt b/Documentation/sound/alsa/soc/machine.txt index bab7711ce963..2524c75557df 100644 --- a/Documentation/sound/alsa/soc/machine.txt +++ b/Documentation/sound/alsa/soc/machine.txt @@ -67,7 +67,7 @@ static struct snd_soc_dai_link corgi_dai = { .ops = &corgi_ops, }; -struct snd_soc_card then sets up the machine with it's DAIs. e.g. +struct snd_soc_card then sets up the machine with its DAIs. e.g. /* corgi audio machine driver */ static struct snd_soc_card snd_soc_corgi = { diff --git a/Documentation/sound/alsa/soc/overview.txt b/Documentation/sound/alsa/soc/overview.txt index 1e4c6d3655f2..138ac88c1461 100644 --- a/Documentation/sound/alsa/soc/overview.txt +++ b/Documentation/sound/alsa/soc/overview.txt @@ -33,7 +33,7 @@ features :- and machines. * Easy I2S/PCM audio interface setup between codec and SoC. Each SoC - interface and codec registers it's audio interface capabilities with the + interface and codec registers its audio interface capabilities with the core and are subsequently matched and configured when the application hardware parameters are known. diff --git a/Documentation/usb/WUSB-Design-overview.txt b/Documentation/usb/WUSB-Design-overview.txt index c480e9c32dbd..4c5e37939344 100644 --- a/Documentation/usb/WUSB-Design-overview.txt +++ b/Documentation/usb/WUSB-Design-overview.txt @@ -381,7 +381,7 @@ descriptor that gives us the status of the transfer, its identification we issue another URB to read into the destination buffer the chunk of data coming out of the remote endpoint. Done, wait for the next guy. The callbacks for the URBs issued from here are the ones that will declare -the xfer complete at some point and call it's callback. +the xfer complete at some point and call its callback. Seems simple, but the implementation is not trivial. diff --git a/Documentation/vm/numa_memory_policy.txt b/Documentation/vm/numa_memory_policy.txt index be45dbb9d7f2..6690fc34ef6d 100644 --- a/Documentation/vm/numa_memory_policy.txt +++ b/Documentation/vm/numa_memory_policy.txt @@ -45,7 +45,7 @@ most general to most specific: to establish the task policy for a child task exec()'d from an executable image that has no awareness of memory policy. See the MEMORY POLICY APIS section, below, for an overview of the system call - that a task may use to set/change it's task/process policy. + that a task may use to set/change its task/process policy. In a multi-threaded task, task policies apply only to the thread [Linux kernel task] that installs the policy and any threads @@ -301,7 +301,7 @@ decrement this reference count, respectively. mpol_put() will only free the structure back to the mempolicy kmem cache when the reference count goes to zero. -When a new memory policy is allocated, it's reference count is initialized +When a new memory policy is allocated, its reference count is initialized to '1', representing the reference held by the task that is installing the new policy. When a pointer to a memory policy structure is stored in another structure, another reference is added, as the task's reference will be dropped diff --git a/Documentation/w1/w1.generic b/Documentation/w1/w1.generic index e3333eec4320..212f4ac31c01 100644 --- a/Documentation/w1/w1.generic +++ b/Documentation/w1/w1.generic @@ -25,7 +25,7 @@ When a w1 master driver registers with the w1 subsystem, the following occurs: - sysfs entries for that w1 master are created - the w1 bus is periodically searched for new slave devices -When a device is found on the bus, w1 core checks if driver for it's family is +When a device is found on the bus, w1 core checks if driver for its family is loaded. If so, the family driver is attached to the slave. If there is no driver for the family, default one is assigned, which allows to perform almost any kind of operations. Each logical operation is a transaction -- cgit v1.2.3 From d02f00cc057809d96c044cc72d5b9809d59f7d49 Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Mon, 7 Dec 2009 13:10:48 -0800 Subject: ocfs2: allocation reservations This patch improves Ocfs2 allocation policy by allowing an inode to reserve a portion of the local alloc bitmap for itself. The reserved portion (allocation window) is advisory in that other allocation windows might steal it if the local alloc bitmap becomes full. Otherwise, the reservations are honored and guaranteed to be free. When the local alloc window is moved to a different portion of the bitmap, existing reservations are discarded. Reservation windows are represented internally by a red-black tree. Within that tree, each node represents the reservation window of one inode. An LRU of active reservations is also maintained. When new data is written, we allocate it from the inodes window. When all bits in a window are exhausted, we allocate a new one as close to the previous one as possible. Should we not find free space, an existing reservation is pulled off the LRU and cannibalized. Signed-off-by: Mark Fasheh --- Documentation/filesystems/ocfs2.txt | 3 + fs/ocfs2/Makefile | 1 + fs/ocfs2/cluster/masklog.c | 1 + fs/ocfs2/cluster/masklog.h | 1 + fs/ocfs2/localalloc.c | 64 ++- fs/ocfs2/ocfs2.h | 5 + fs/ocfs2/reservations.c | 849 ++++++++++++++++++++++++++++++++++++ fs/ocfs2/reservations.h | 154 +++++++ fs/ocfs2/suballoc.h | 2 + fs/ocfs2/super.c | 25 ++ 10 files changed, 1094 insertions(+), 11 deletions(-) create mode 100644 fs/ocfs2/reservations.c create mode 100644 fs/ocfs2/reservations.h (limited to 'Documentation/filesystems') diff --git a/Documentation/filesystems/ocfs2.txt b/Documentation/filesystems/ocfs2.txt index c58b9f5ba002..412df9095937 100644 --- a/Documentation/filesystems/ocfs2.txt +++ b/Documentation/filesystems/ocfs2.txt @@ -80,3 +80,6 @@ user_xattr (*) Enables Extended User Attributes. nouser_xattr Disables Extended User Attributes. acl Enables POSIX Access Control Lists support. noacl (*) Disables POSIX Access Control Lists support. +resv_level=4 (*) Set how agressive allocation reservations will be. + Valid values are between 0 (reservations off) to 8 + (maximum space for reservations). diff --git a/fs/ocfs2/Makefile b/fs/ocfs2/Makefile index 791c0886c060..07d9fd854350 100644 --- a/fs/ocfs2/Makefile +++ b/fs/ocfs2/Makefile @@ -29,6 +29,7 @@ ocfs2-objs := \ mmap.o \ namei.o \ refcounttree.o \ + reservations.o \ resize.o \ slot_map.o \ suballoc.o \ diff --git a/fs/ocfs2/cluster/masklog.c b/fs/ocfs2/cluster/masklog.c index 3bb928a2bf7d..c7fba396392d 100644 --- a/fs/ocfs2/cluster/masklog.c +++ b/fs/ocfs2/cluster/masklog.c @@ -116,6 +116,7 @@ static struct mlog_attribute mlog_attrs[MLOG_MAX_BITS] = { define_mask(ERROR), define_mask(NOTICE), define_mask(KTHREAD), + define_mask(RESERVATIONS), }; static struct attribute *mlog_attr_ptrs[MLOG_MAX_BITS] = {NULL, }; diff --git a/fs/ocfs2/cluster/masklog.h b/fs/ocfs2/cluster/masklog.h index 3dfddbec32f2..fd96e2a2fa56 100644 --- a/fs/ocfs2/cluster/masklog.h +++ b/fs/ocfs2/cluster/masklog.h @@ -119,6 +119,7 @@ #define ML_ERROR 0x0000000100000000ULL /* sent to KERN_ERR */ #define ML_NOTICE 0x0000000200000000ULL /* setn to KERN_NOTICE */ #define ML_KTHREAD 0x0000000400000000ULL /* kernel thread activity */ +#define ML_RESERVATIONS 0x0000000800000000ULL /* ocfs2 alloc reservations */ #define MLOG_INITIAL_AND_MASK (ML_ERROR|ML_NOTICE) #define MLOG_INITIAL_NOT_MASK (ML_ENTRY|ML_EXIT) diff --git a/fs/ocfs2/localalloc.c b/fs/ocfs2/localalloc.c index 7e7dd65d97ef..7fe8149a0002 100644 --- a/fs/ocfs2/localalloc.c +++ b/fs/ocfs2/localalloc.c @@ -52,7 +52,8 @@ static u32 ocfs2_local_alloc_count_bits(struct ocfs2_dinode *alloc); static int ocfs2_local_alloc_find_clear_bits(struct ocfs2_super *osb, struct ocfs2_dinode *alloc, - u32 numbits); + u32 *numbits, + struct ocfs2_alloc_reservation *resv); static void ocfs2_clear_local_alloc(struct ocfs2_dinode *alloc); @@ -262,6 +263,8 @@ void ocfs2_shutdown_local_alloc(struct ocfs2_super *osb) osb->local_alloc_state = OCFS2_LA_DISABLED; + ocfs2_resmap_uninit(&osb->osb_la_resmap); + main_bm_inode = ocfs2_get_system_file_inode(osb, GLOBAL_BITMAP_SYSTEM_INODE, OCFS2_INVALID_SLOT); @@ -493,7 +496,7 @@ static int ocfs2_local_alloc_in_range(struct inode *inode, alloc = (struct ocfs2_dinode *) osb->local_alloc_bh->b_data; la = OCFS2_LOCAL_ALLOC(alloc); - start = ocfs2_local_alloc_find_clear_bits(osb, alloc, bits_wanted); + start = ocfs2_local_alloc_find_clear_bits(osb, alloc, &bits_wanted, NULL); if (start == -1) { mlog_errno(-ENOSPC); return 0; @@ -659,7 +662,8 @@ int ocfs2_claim_local_alloc_bits(struct ocfs2_super *osb, alloc = (struct ocfs2_dinode *) osb->local_alloc_bh->b_data; la = OCFS2_LOCAL_ALLOC(alloc); - start = ocfs2_local_alloc_find_clear_bits(osb, alloc, bits_wanted); + start = ocfs2_local_alloc_find_clear_bits(osb, alloc, &bits_wanted, + ac->ac_resv); if (start == -1) { /* TODO: Shouldn't we just BUG here? */ status = -ENOSPC; @@ -669,8 +673,6 @@ int ocfs2_claim_local_alloc_bits(struct ocfs2_super *osb, bitmap = la->la_bitmap; *bit_off = le32_to_cpu(la->la_bm_off) + start; - /* local alloc is always contiguous by nature -- we never - * delete bits from it! */ *num_bits = bits_wanted; status = ocfs2_journal_access_di(handle, @@ -682,6 +684,9 @@ int ocfs2_claim_local_alloc_bits(struct ocfs2_super *osb, goto bail; } + ocfs2_resmap_claimed_bits(&osb->osb_la_resmap, ac->ac_resv, start, + bits_wanted); + while(bits_wanted--) ocfs2_set_bit(start++, bitmap); @@ -711,13 +716,17 @@ static u32 ocfs2_local_alloc_count_bits(struct ocfs2_dinode *alloc) } static int ocfs2_local_alloc_find_clear_bits(struct ocfs2_super *osb, - struct ocfs2_dinode *alloc, - u32 numbits) + struct ocfs2_dinode *alloc, + u32 *numbits, + struct ocfs2_alloc_reservation *resv) { int numfound, bitoff, left, startoff, lastzero; + int local_resv = 0; + struct ocfs2_alloc_reservation r; void *bitmap = NULL; + struct ocfs2_reservation_map *resmap = &osb->osb_la_resmap; - mlog_entry("(numbits wanted = %u)\n", numbits); + mlog_entry("(numbits wanted = %u)\n", *numbits); if (!alloc->id1.bitmap1.i_total) { mlog(0, "No bits in my window!\n"); @@ -725,6 +734,30 @@ static int ocfs2_local_alloc_find_clear_bits(struct ocfs2_super *osb, goto bail; } + if (!resv) { + local_resv = 1; + ocfs2_resv_init_once(&r); + ocfs2_resv_set_type(&r, OCFS2_RESV_FLAG_TMP); + resv = &r; + } + + numfound = *numbits; + if (ocfs2_resmap_resv_bits(resmap, resv, &bitoff, &numfound) == 0) { + if (numfound < *numbits) + *numbits = numfound; + goto bail; + } + + /* + * Code error. While reservations are enabled, local + * allocation should _always_ go through them. + */ + BUG_ON(osb->osb_resv_level != 0); + + /* + * Reservations are disabled. Handle this the old way. + */ + bitmap = OCFS2_LOCAL_ALLOC(alloc)->la_bitmap; numfound = bitoff = startoff = 0; @@ -750,7 +783,7 @@ static int ocfs2_local_alloc_find_clear_bits(struct ocfs2_super *osb, startoff = bitoff+1; } /* we got everything we needed */ - if (numfound == numbits) { + if (numfound == *numbits) { /* mlog(0, "Found it all!\n"); */ break; } @@ -759,12 +792,18 @@ static int ocfs2_local_alloc_find_clear_bits(struct ocfs2_super *osb, mlog(0, "Exiting loop, bitoff = %d, numfound = %d\n", bitoff, numfound); - if (numfound == numbits) + if (numfound == *numbits) { bitoff = startoff - numfound; - else + *numbits = numfound; + } else { + numfound = 0; bitoff = -1; + } bail: + if (local_resv) + ocfs2_resv_discard(resmap, resv); + mlog_exit(bitoff); return bitoff; } @@ -1087,6 +1126,9 @@ retry_enospc: memset(OCFS2_LOCAL_ALLOC(alloc)->la_bitmap, 0, le16_to_cpu(la->la_size)); + ocfs2_resmap_restart(&osb->osb_la_resmap, cluster_count, + OCFS2_LOCAL_ALLOC(alloc)->la_bitmap); + mlog(0, "New window allocated:\n"); mlog(0, "window la_bm_off = %u\n", OCFS2_LOCAL_ALLOC(alloc)->la_bm_off); diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h index adf5e2ebc2c4..9552560df6cd 100644 --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -47,6 +47,7 @@ /* For struct ocfs2_blockcheck_stats */ #include "blockcheck.h" +#include "reservations.h" /* Caching of metadata buffers */ @@ -349,6 +350,10 @@ struct ocfs2_super u64 la_last_gd; + struct ocfs2_reservation_map osb_la_resmap; + + unsigned int osb_resv_level; + /* Next three fields are for local node slot recovery during * mount. */ int dirty; diff --git a/fs/ocfs2/reservations.c b/fs/ocfs2/reservations.c new file mode 100644 index 000000000000..79642d608210 --- /dev/null +++ b/fs/ocfs2/reservations.c @@ -0,0 +1,849 @@ +/* -*- mode: c; c-basic-offset: 8; -*- + * vim: noexpandtab sw=8 ts=8 sts=0: + * + * reservations.c + * + * Allocation reservations implementation + * + * Some code borrowed from fs/ext3/balloc.c and is: + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * The rest is copyright (C) 2010 Novell. 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 version 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. + */ + +#include +#include +#include +#include +#include +#include + +#define MLOG_MASK_PREFIX ML_RESERVATIONS +#include + +#include "ocfs2.h" + +#ifdef CONFIG_OCFS2_DEBUG_FS +#define OCFS2_CHECK_RESERVATIONS +#endif + +DEFINE_SPINLOCK(resv_lock); + +#define OCFS2_MIN_RESV_WINDOW_BITS 8 +#define OCFS2_MAX_RESV_WINDOW_BITS 1024 + +static unsigned int ocfs2_resv_window_bits(struct ocfs2_reservation_map *resmap, + struct ocfs2_alloc_reservation *resv) +{ + struct ocfs2_super *osb = resmap->m_osb; + unsigned int bits; + + /* 8, 16, 32, 64, 128, 256, 512, 1024 */ + bits = 4 << osb->osb_resv_level; + + return bits; +} + +static inline unsigned int ocfs2_resv_end(struct ocfs2_alloc_reservation *resv) +{ + if (resv->r_len) + return resv->r_start + resv->r_len - 1; + return resv->r_start; +} + +static inline int ocfs2_resv_empty(struct ocfs2_alloc_reservation *resv) +{ + return !!(resv->r_len == 0); +} + +static inline int ocfs2_resmap_disabled(struct ocfs2_reservation_map *resmap) +{ + if (resmap->m_osb->osb_resv_level == 0) + return 1; + return 0; +} + +static void ocfs2_dump_resv(struct ocfs2_reservation_map *resmap) +{ + struct ocfs2_super *osb = resmap->m_osb; + struct rb_node *node; + struct ocfs2_alloc_reservation *resv; + int i = 0; + + mlog(ML_NOTICE, "Dumping resmap for device %s. Bitmap length: %u\n", + osb->dev_str, resmap->m_bitmap_len); + + node = rb_first(&resmap->m_reservations); + while (node) { + resv = rb_entry(node, struct ocfs2_alloc_reservation, r_node); + + mlog(ML_NOTICE, "start: %u\tend: %u\tlen: %u\tlast_start: %u" + "\tlast_len: %u\n", resv->r_start, + ocfs2_resv_end(resv), resv->r_len, resv->r_last_start, + resv->r_last_len); + + node = rb_next(node); + i++; + } + + mlog(ML_NOTICE, "%d reservations found. LRU follows\n", i); + + i = 0; + list_for_each_entry(resv, &resmap->m_lru, r_lru) { + mlog(ML_NOTICE, "LRU(%d) start: %u\tend: %u\tlen: %u\t" + "last_start: %u\tlast_len: %u\n", i, resv->r_start, + ocfs2_resv_end(resv), resv->r_len, resv->r_last_start, + resv->r_last_len); + + i++; + } +} + +#ifdef OCFS2_CHECK_RESERVATIONS +static int ocfs2_validate_resmap_bits(struct ocfs2_reservation_map *resmap, + int i, + struct ocfs2_alloc_reservation *resv) +{ + char *disk_bitmap = resmap->m_disk_bitmap; + unsigned int start = resv->r_start; + unsigned int end = ocfs2_resv_end(resv); + + while (start <= end) { + if (ocfs2_test_bit(start, disk_bitmap)) { + mlog(ML_ERROR, + "reservation %d covers an allocated area " + "starting at bit %u!\n", i, start); + return 1; + } + + start++; + } + return 0; +} + +static void ocfs2_check_resmap(struct ocfs2_reservation_map *resmap) +{ + unsigned int off = 0; + int i = 0; + struct rb_node *node; + struct ocfs2_alloc_reservation *resv; + + node = rb_first(&resmap->m_reservations); + while (node) { + resv = rb_entry(node, struct ocfs2_alloc_reservation, r_node); + + if (i > 0 && resv->r_start <= off) { + mlog(ML_ERROR, "reservation %d has bad start off!\n", + i); + goto bad; + } + + if (resv->r_len == 0) { + mlog(ML_ERROR, "reservation %d has no length!\n", + i); + goto bad; + } + + if (resv->r_start > ocfs2_resv_end(resv)) { + mlog(ML_ERROR, "reservation %d has invalid range!\n", + i); + goto bad; + } + + if (ocfs2_resv_end(resv) >= resmap->m_bitmap_len) { + mlog(ML_ERROR, "reservation %d extends past bitmap!\n", + i); + goto bad; + } + + if (ocfs2_validate_resmap_bits(resmap, i, resv)) + goto bad; + + off = ocfs2_resv_end(resv); + node = rb_next(node); + + i++; + } + return; + +bad: + ocfs2_dump_resv(resmap); + BUG(); +} +#else +static inline void ocfs2_check_resmap(struct ocfs2_reservation_map *resmap) +{ + +} +#endif + +void ocfs2_resv_init_once(struct ocfs2_alloc_reservation *resv) +{ + memset(resv, 0, sizeof(*resv)); + INIT_LIST_HEAD(&resv->r_lru); +} + +void ocfs2_resv_set_type(struct ocfs2_alloc_reservation *resv, + unsigned int flags) +{ + BUG_ON(flags & ~OCFS2_RESV_TYPES); + + resv->r_flags |= flags; +} + +int ocfs2_resmap_init(struct ocfs2_super *osb, + struct ocfs2_reservation_map *resmap) +{ + memset(resmap, 0, sizeof(*resmap)); + + resmap->m_osb = osb; + resmap->m_reservations = RB_ROOT; + /* m_bitmap_len is initialized to zero by the above memset. */ + INIT_LIST_HEAD(&resmap->m_lru); + + return 0; +} + +static void ocfs2_resv_mark_lru(struct ocfs2_reservation_map *resmap, + struct ocfs2_alloc_reservation *resv) +{ + assert_spin_locked(&resv_lock); + + if (!list_empty(&resv->r_lru)) + list_del_init(&resv->r_lru); + + list_add_tail(&resv->r_lru, &resmap->m_lru); +} + +static void __ocfs2_resv_trunc(struct ocfs2_alloc_reservation *resv) +{ + resv->r_len = 0; + resv->r_start = 0; +} + +static void ocfs2_resv_remove(struct ocfs2_reservation_map *resmap, + struct ocfs2_alloc_reservation *resv) +{ + if (resv->r_flags & OCFS2_RESV_FLAG_INUSE) { + list_del_init(&resv->r_lru); + rb_erase(&resv->r_node, &resmap->m_reservations); + resv->r_flags &= ~OCFS2_RESV_FLAG_INUSE; + } +} + +static void __ocfs2_resv_discard(struct ocfs2_reservation_map *resmap, + struct ocfs2_alloc_reservation *resv) +{ + assert_spin_locked(&resv_lock); + + __ocfs2_resv_trunc(resv); + /* + * last_len and last_start no longer make sense if + * we're changing the range of our allocations. + */ + resv->r_last_len = resv->r_last_start = 0; + + ocfs2_resv_remove(resmap, resv); +} + +/* does nothing if 'resv' is null */ +void ocfs2_resv_discard(struct ocfs2_reservation_map *resmap, + struct ocfs2_alloc_reservation *resv) +{ + if (resv) { + spin_lock(&resv_lock); + __ocfs2_resv_discard(resmap, resv); + spin_unlock(&resv_lock); + } +} + +static void ocfs2_resmap_clear_all_resv(struct ocfs2_reservation_map *resmap) +{ + struct rb_node *node; + struct ocfs2_alloc_reservation *resv; + + assert_spin_locked(&resv_lock); + + while ((node = rb_last(&resmap->m_reservations)) != NULL) { + resv = rb_entry(node, struct ocfs2_alloc_reservation, r_node); + + __ocfs2_resv_discard(resmap, resv); + } +} + +void ocfs2_resmap_restart(struct ocfs2_reservation_map *resmap, + unsigned int clen, char *disk_bitmap) +{ + if (ocfs2_resmap_disabled(resmap)) + return; + + spin_lock(&resv_lock); + + ocfs2_resmap_clear_all_resv(resmap); + resmap->m_bitmap_len = clen; + resmap->m_disk_bitmap = disk_bitmap; + + spin_unlock(&resv_lock); +} + +void ocfs2_resmap_uninit(struct ocfs2_reservation_map *resmap) +{ + /* Does nothing for now. Keep this around for API symmetry */ +} + +static void ocfs2_resv_insert(struct ocfs2_reservation_map *resmap, + struct ocfs2_alloc_reservation *new) +{ + struct rb_root *root = &resmap->m_reservations; + struct rb_node *parent = NULL; + struct rb_node **p = &root->rb_node; + struct ocfs2_alloc_reservation *tmp; + + assert_spin_locked(&resv_lock); + + mlog(0, "Insert reservation start: %u len: %u\n", new->r_start, + new->r_len); + + while (*p) { + parent = *p; + + tmp = rb_entry(parent, struct ocfs2_alloc_reservation, r_node); + + if (new->r_start < tmp->r_start) { + p = &(*p)->rb_left; + + /* + * This is a good place to check for + * overlapping reservations. + */ + BUG_ON(ocfs2_resv_end(new) >= tmp->r_start); + } else if (new->r_start > ocfs2_resv_end(tmp)) { + p = &(*p)->rb_right; + } else { + /* This should never happen! */ + mlog(ML_ERROR, "Duplicate reservation window!\n"); + BUG(); + } + } + + rb_link_node(&new->r_node, parent, p); + rb_insert_color(&new->r_node, root); + new->r_flags |= OCFS2_RESV_FLAG_INUSE; + + ocfs2_resv_mark_lru(resmap, new); + + ocfs2_check_resmap(resmap); +} + +/** + * ocfs2_find_resv_lhs() - find the window which contains goal + * @resmap: reservation map to search + * @goal: which bit to search for + * + * If a window containing that goal is not found, we return the window + * which comes before goal. Returns NULL on empty rbtree or no window + * before goal. + */ +static struct ocfs2_alloc_reservation * +ocfs2_find_resv_lhs(struct ocfs2_reservation_map *resmap, unsigned int goal) +{ + struct ocfs2_alloc_reservation *resv = NULL; + struct ocfs2_alloc_reservation *prev_resv = NULL; + struct rb_node *node = resmap->m_reservations.rb_node; + struct rb_node *prev = NULL; + + assert_spin_locked(&resv_lock); + + if (!node) + return NULL; + + node = rb_first(&resmap->m_reservations); + while (node) { + resv = rb_entry(node, struct ocfs2_alloc_reservation, r_node); + + if (resv->r_start <= goal && ocfs2_resv_end(resv) >= goal) + break; + + /* Check if we overshot the reservation just before goal? */ + if (resv->r_start > goal) { + resv = prev_resv; + break; + } + + prev_resv = resv; + prev = node; + node = rb_next(node); + } + + return resv; +} + +/* + * We are given a range within the bitmap, which corresponds to a gap + * inside the reservations tree (search_start, search_len). The range + * can be anything from the whole bitmap, to a gap between + * reservations. + * + * The start value of *rstart is insignificant. + * + * This function searches the bitmap range starting at search_start + * with length csearch_len for a set of contiguous free bits. We try + * to find up to 'wanted' bits, but can sometimes return less. + * + * Returns the length of allocation, 0 if no free bits are found. + * + * *cstart and *clen will also be populated with the result. + */ +static int ocfs2_resmap_find_free_bits(struct ocfs2_reservation_map *resmap, + unsigned int wanted, + unsigned int search_start, + unsigned int search_len, + unsigned int *rstart, + unsigned int *rlen) +{ + void *bitmap = resmap->m_disk_bitmap; + unsigned int best_start, best_len = 0; + int offset, start, found; + + mlog(0, "Find %u bits within range (%u, len %u) resmap len: %u\n", + wanted, search_start, search_len, resmap->m_bitmap_len); + + found = best_start = best_len = 0; + + start = search_start; + while ((offset = ocfs2_find_next_zero_bit(bitmap, resmap->m_bitmap_len, + start)) != -1) { + /* Search reached end of the region */ + if (offset >= (search_start + search_len)) + break; + + if (offset == start) { + /* we found a zero */ + found++; + /* move start to the next bit to test */ + start++; + } else { + /* got a zero after some ones */ + found = 1; + start = offset + 1; + } + if (found > best_len) { + best_len = found; + best_start = start - found; + } + + if (found >= wanted) + break; + } + + if (best_len == 0) + return 0; + + if (best_len >= wanted) + best_len = wanted; + + *rlen = best_len; + *rstart = best_start; + + mlog(0, "Found start: %u len: %u\n", best_start, best_len); + + return *rlen; +} + +static void __ocfs2_resv_find_window(struct ocfs2_reservation_map *resmap, + struct ocfs2_alloc_reservation *resv, + unsigned int goal, unsigned int wanted) +{ + struct rb_root *root = &resmap->m_reservations; + unsigned int gap_start, gap_end, gap_len; + struct ocfs2_alloc_reservation *prev_resv, *next_resv; + struct rb_node *prev, *next; + unsigned int cstart, clen; + unsigned int best_start = 0, best_len = 0; + + /* + * Nasty cases to consider: + * + * - rbtree is empty + * - our window should be first in all reservations + * - our window should be last in all reservations + * - need to make sure we don't go past end of bitmap + */ + + mlog(0, "resv start: %u resv end: %u goal: %u wanted: %u\n", + resv->r_start, ocfs2_resv_end(resv), goal, wanted); + + assert_spin_locked(&resv_lock); + + if (RB_EMPTY_ROOT(root)) { + /* + * Easiest case - empty tree. We can just take + * whatever window of free bits we want. + */ + + mlog(0, "Empty root\n"); + + clen = ocfs2_resmap_find_free_bits(resmap, wanted, goal, + resmap->m_bitmap_len - goal, + &cstart, &clen); + + /* + * This should never happen - the local alloc window + * will always have free bits when we're called. + */ + BUG_ON(goal == 0 && clen == 0); + + if (clen == 0) + return; + + resv->r_start = cstart; + resv->r_len = clen; + + ocfs2_resv_insert(resmap, resv); + return; + } + + prev_resv = ocfs2_find_resv_lhs(resmap, goal); + + if (prev_resv == NULL) { + mlog(0, "Goal on LHS of leftmost window\n"); + + /* + * A NULL here means that the search code couldn't + * find a window that starts before goal. + * + * However, we can take the first window after goal, + * which is also by definition, the leftmost window in + * the entire tree. If we can find free bits in the + * gap between goal and the LHS window, then the + * reservation can safely be placed there. + * + * Otherwise we fall back to a linear search, checking + * the gaps in between windows for a place to + * allocate. + */ + + next = rb_first(root); + next_resv = rb_entry(next, struct ocfs2_alloc_reservation, + r_node); + + /* + * The search should never return such a window. (see + * comment above + */ + if (next_resv->r_start <= goal) { + mlog(ML_ERROR, "goal: %u next_resv: start %u len %u\n", + goal, next_resv->r_start, next_resv->r_len); + ocfs2_dump_resv(resmap); + BUG(); + } + + clen = ocfs2_resmap_find_free_bits(resmap, wanted, goal, + next_resv->r_start - goal, + &cstart, &clen); + if (clen) { + best_len = clen; + best_start = cstart; + if (best_len == wanted) + goto out_insert; + } + + prev_resv = next_resv; + next_resv = NULL; + } + + prev = &prev_resv->r_node; + + /* Now we do a linear search for a window, starting at 'prev_rsv' */ + while (1) { + next = rb_next(prev); + if (next) { + mlog(0, "One more resv found in linear search\n"); + next_resv = rb_entry(next, + struct ocfs2_alloc_reservation, + r_node); + + gap_start = ocfs2_resv_end(prev_resv) + 1; + gap_end = next_resv->r_start - 1; + gap_len = gap_end - gap_start + 1; + } else { + mlog(0, "No next node\n"); + /* + * We're at the rightmost edge of the + * tree. See if a reservation between this + * window and the end of the bitmap will work. + */ + gap_start = ocfs2_resv_end(prev_resv) + 1; + gap_len = resmap->m_bitmap_len - gap_start; + gap_end = resmap->m_bitmap_len - 1; + } + + /* + * No need to check this gap if we have already found + * a larger region of free bits. + */ + if (gap_len <= best_len) + goto next_resv; + + clen = ocfs2_resmap_find_free_bits(resmap, wanted, gap_start, + gap_len, &cstart, &clen); + if (clen == wanted) { + best_len = clen; + best_start = cstart; + goto out_insert; + } else if (clen > best_len) { + best_len = clen; + best_start = cstart; + } + +next_resv: + if (!next) + break; + + prev = next; + prev_resv = rb_entry(prev, struct ocfs2_alloc_reservation, + r_node); + } + +out_insert: + if (best_len) { + resv->r_start = best_start; + resv->r_len = best_len; + ocfs2_resv_insert(resmap, resv); + } +} + +static void ocfs2_cannibalize_resv(struct ocfs2_reservation_map *resmap, + struct ocfs2_alloc_reservation *resv, + unsigned int wanted) +{ + struct ocfs2_alloc_reservation *lru_resv; + int tmpwindow = !!(resv->r_flags & OCFS2_RESV_FLAG_TMP); + unsigned int min_bits; + + if (!tmpwindow) + min_bits = ocfs2_resv_window_bits(resmap, resv) >> 1; + else + min_bits = wanted; /* We at know the temp window will use all + * of these bits */ + + /* + * Take the first reservation off the LRU as our 'target'. We + * don't try to be smart about it. There might be a case for + * searching based on size but I don't have enough data to be + * sure. --Mark (3/16/2010) + */ + lru_resv = list_first_entry(&resmap->m_lru, + struct ocfs2_alloc_reservation, r_lru); + + mlog(0, "lru resv: start: %u len: %u end: %u\n", lru_resv->r_start, + lru_resv->r_len, ocfs2_resv_end(lru_resv)); + + /* + * Cannibalize (some or all) of the target reservation and + * feed it to the current window. + */ + if (lru_resv->r_len <= min_bits) { + /* + * Discard completely if size is less than or equal to a + * reasonable threshold - 50% of window bits for non temporary + * windows. + */ + resv->r_start = lru_resv->r_start; + resv->r_len = lru_resv->r_len; + + __ocfs2_resv_discard(resmap, lru_resv); + } else { + unsigned int shrink; + if (tmpwindow) + shrink = min_bits; + else + shrink = lru_resv->r_len / 2; + + lru_resv->r_len -= shrink; + + resv->r_start = ocfs2_resv_end(lru_resv) + 1; + resv->r_len = shrink; + } + + mlog(0, "Reservation now looks like: r_start: %u r_end: %u " + "r_len: %u r_last_start: %u r_last_len: %u\n", + resv->r_start, ocfs2_resv_end(resv), resv->r_len, + resv->r_last_start, resv->r_last_len); + + ocfs2_resv_insert(resmap, resv); +} + +static void ocfs2_resv_find_window(struct ocfs2_reservation_map *resmap, + struct ocfs2_alloc_reservation *resv, + unsigned int wanted) +{ + unsigned int goal = 0; + + BUG_ON(!ocfs2_resv_empty(resv)); + + /* + * Begin by trying to get a window as close to the previous + * one as possible. Using the most recent allocation as a + * start goal makes sense. + */ + if (resv->r_last_len) { + goal = resv->r_last_start + resv->r_last_len; + if (goal >= resmap->m_bitmap_len) + goal = 0; + } + + __ocfs2_resv_find_window(resmap, resv, goal, wanted); + + /* Search from last alloc didn't work, try once more from beginning. */ + if (ocfs2_resv_empty(resv) && goal != 0) + __ocfs2_resv_find_window(resmap, resv, 0, wanted); + + if (ocfs2_resv_empty(resv)) { + /* + * Still empty? Pull oldest one off the LRU, remove it from + * tree, put this one in it's place. + */ + ocfs2_cannibalize_resv(resmap, resv, wanted); + } + + BUG_ON(ocfs2_resv_empty(resv)); +} + +int ocfs2_resmap_resv_bits(struct ocfs2_reservation_map *resmap, + struct ocfs2_alloc_reservation *resv, + int *cstart, int *clen) +{ + unsigned int wanted = *clen; + + if (resv == NULL || ocfs2_resmap_disabled(resmap)) + return -ENOSPC; + + spin_lock(&resv_lock); + + /* + * We don't want to over-allocate for temporary + * windows. Otherwise, we run the risk of fragmenting the + * allocation space. + */ + wanted = ocfs2_resv_window_bits(resmap, resv); + if ((resv->r_flags & OCFS2_RESV_FLAG_TMP) || wanted < *clen) + wanted = *clen; + + if (ocfs2_resv_empty(resv)) { + mlog(0, "empty reservation, find new window\n"); + + /* + * Try to get a window here. If it works, we must fall + * through and test the bitmap . This avoids some + * ping-ponging of windows due to non-reserved space + * being allocation before we initialize a window for + * that inode. + */ + ocfs2_resv_find_window(resmap, resv, wanted); + } + + BUG_ON(ocfs2_resv_empty(resv)); + + *cstart = resv->r_start; + *clen = resv->r_len; + + spin_unlock(&resv_lock); + return 0; +} + +static void + ocfs2_adjust_resv_from_alloc(struct ocfs2_reservation_map *resmap, + struct ocfs2_alloc_reservation *resv, + unsigned int start, unsigned int end) +{ + unsigned int lhs = 0, rhs = 0; + + BUG_ON(start < resv->r_start); + + /* + * Completely used? We can remove it then. + */ + if (ocfs2_resv_end(resv) <= end && resv->r_start >= start) { + __ocfs2_resv_discard(resmap, resv); + return; + } + + if (end < ocfs2_resv_end(resv)) + rhs = end - ocfs2_resv_end(resv); + + if (start > resv->r_start) + lhs = start - resv->r_start; + + /* + * This should have been trapped above. At the very least, rhs + * should be non zero. + */ + BUG_ON(rhs == 0 && lhs == 0); + + if (rhs >= lhs) { + unsigned int old_end = ocfs2_resv_end(resv); + + resv->r_start = end + 1; + resv->r_len = old_end - resv->r_start + 1; + } else { + resv->r_len = start - resv->r_start; + } +} + +void ocfs2_resmap_claimed_bits(struct ocfs2_reservation_map *resmap, + struct ocfs2_alloc_reservation *resv, + u32 cstart, u32 clen) +{ + unsigned int cend = cstart + clen - 1; + + if (resmap == NULL || ocfs2_resmap_disabled(resmap)) + return; + + if (resv == NULL) + return; + + spin_lock(&resv_lock); + + mlog(0, "claim bits: cstart: %u cend: %u clen: %u r_start: %u " + "r_end: %u r_len: %u, r_last_start: %u r_last_len: %u\n", + cstart, cend, clen, resv->r_start, ocfs2_resv_end(resv), + resv->r_len, resv->r_last_start, resv->r_last_len); + + BUG_ON(cstart < resv->r_start); + BUG_ON(cstart > ocfs2_resv_end(resv)); + BUG_ON(cend > ocfs2_resv_end(resv)); + + ocfs2_adjust_resv_from_alloc(resmap, resv, cstart, cend); + resv->r_last_start = cstart; + resv->r_last_len = clen; + + /* + * May have been discarded above from + * ocfs2_adjust_resv_from_alloc(). + */ + if (!ocfs2_resv_empty(resv)) + ocfs2_resv_mark_lru(resmap, resv); + + mlog(0, "Reservation now looks like: r_start: %u r_end: %u " + "r_len: %u r_last_start: %u r_last_len: %u\n", + resv->r_start, ocfs2_resv_end(resv), resv->r_len, + resv->r_last_start, resv->r_last_len); + + ocfs2_check_resmap(resmap); + + spin_unlock(&resv_lock); +} diff --git a/fs/ocfs2/reservations.h b/fs/ocfs2/reservations.h new file mode 100644 index 000000000000..8341cd0ef855 --- /dev/null +++ b/fs/ocfs2/reservations.h @@ -0,0 +1,154 @@ +/* -*- mode: c; c-basic-offset: 8; -*- + * vim: noexpandtab sw=8 ts=8 sts=0: + * + * reservations.h + * + * Allocation reservations function prototypes and structures. + * + * Copyright (C) 2010 Novell. 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 version 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. + */ + +#ifndef OCFS2_RESERVATIONS_H +#define OCFS2_RESERVATIONS_H + +#include + +#define OCFS2_DEFAULT_RESV_LEVEL 4 +#define OCFS2_MAX_RESV_LEVEL 9 +#define OCFS2_MIN_RESV_LEVEL 0 + +struct ocfs2_alloc_reservation { + struct rb_node r_node; + + unsigned int r_start; /* Begining of current window */ + unsigned int r_len; /* Length of the window */ + + unsigned int r_last_len; /* Length of most recent alloc */ + unsigned int r_last_start; /* Start of most recent alloc */ + struct list_head r_lru; /* LRU list head */ + + unsigned int r_flags; +}; + +#define OCFS2_RESV_FLAG_INUSE 0x01 /* Set when r_node is part of a btree */ +#define OCFS2_RESV_FLAG_TMP 0x02 /* Temporary reservation, will be + * destroyed immedately after use */ + +struct ocfs2_reservation_map { + struct rb_root m_reservations; + char *m_disk_bitmap; + + struct ocfs2_super *m_osb; + + /* The following are not initialized to meaningful values until a disk + * bitmap is provided. */ + u32 m_bitmap_len; /* Number of valid + * bits available */ + + struct list_head m_lru; /* LRU of reservations + * structures. */ + +}; + +void ocfs2_resv_init_once(struct ocfs2_alloc_reservation *resv); + +#define OCFS2_RESV_TYPES (OCFS2_RESV_FLAG_TMP) +void ocfs2_resv_set_type(struct ocfs2_alloc_reservation *resv, + unsigned int flags); + +/** + * ocfs2_resv_discard() - truncate a reservation + * @resmap: + * @resv: the reservation to truncate. + * + * After this function is called, the reservation will be empty, and + * unlinked from the rbtree. + */ +void ocfs2_resv_discard(struct ocfs2_reservation_map *resmap, + struct ocfs2_alloc_reservation *resv); + + +/** + * ocfs2_resmap_init() - Initialize fields of a reservations bitmap + * @resmap: struct ocfs2_reservation_map to initialize + * @obj: unused for now + * @ops: unused for now + * @max_bitmap_bytes: Maximum size of the bitmap (typically blocksize) + * + * Only possible return value other than '0' is -ENOMEM for failure to + * allocation mirror bitmap. + */ +int ocfs2_resmap_init(struct ocfs2_super *osb, + struct ocfs2_reservation_map *resmap); + +/** + * ocfs2_resmap_restart() - "restart" a reservation bitmap + * @resmap: reservations bitmap + * @clen: Number of valid bits in the bitmap + * @disk_bitmap: the disk bitmap this resmap should refer to. + * + * Re-initialize the parameters of a reservation bitmap. This is + * useful for local alloc window slides. + * + * This function will call ocfs2_trunc_resv against all existing + * reservations. A future version will recalculate existing + * reservations based on the new bitmap. + */ +void ocfs2_resmap_restart(struct ocfs2_reservation_map *resmap, + unsigned int clen, char *disk_bitmap); + +/** + * ocfs2_resmap_uninit() - uninitialize a reservation bitmap structure + * @resmap: the struct ocfs2_reservation_map to uninitialize + */ +void ocfs2_resmap_uninit(struct ocfs2_reservation_map *resmap); + +/** + * ocfs2_resmap_resv_bits() - Return still-valid reservation bits + * @resmap: reservations bitmap + * @resv: reservation to base search from + * @cstart: start of proposed allocation + * @clen: length (in clusters) of proposed allocation + * + * Using the reservation data from resv, this function will compare + * resmap and resmap->m_disk_bitmap to determine what part (if any) of + * the reservation window is still clear to use. If resv is empty, + * this function will try to allocate a window for it. + * + * On success, zero is returned and the valid allocation area is set in cstart + * and clen. + * + * Returns -ENOSPC if reservations are disabled. + */ +int ocfs2_resmap_resv_bits(struct ocfs2_reservation_map *resmap, + struct ocfs2_alloc_reservation *resv, + int *cstart, int *clen); + +/** + * ocfs2_resmap_claimed_bits() - Tell the reservation code that bits were used. + * @resmap: reservations bitmap + * @resv: optional reservation to recalulate based on new bitmap + * @cstart: start of allocation in clusters + * @clen: end of allocation in clusters. + * + * Tell the reservation code that bits were used to fulfill allocation in + * resmap. The bits don't have to have been part of any existing + * reservation. But we must always call this function when bits are claimed. + * Internally, the reservations code will use this information to mark the + * reservations bitmap. If resv is passed, it's next allocation window will be + * calculated. + */ +void ocfs2_resmap_claimed_bits(struct ocfs2_reservation_map *resmap, + struct ocfs2_alloc_reservation *resv, + u32 cstart, u32 clen); + +#endif /* OCFS2_RESERVATIONS_H */ diff --git a/fs/ocfs2/suballoc.h b/fs/ocfs2/suballoc.h index e0f46df357e6..da2f29a55ec3 100644 --- a/fs/ocfs2/suballoc.h +++ b/fs/ocfs2/suballoc.h @@ -54,6 +54,8 @@ struct ocfs2_alloc_context { u64 ac_last_group; u64 ac_max_block; /* Highest block number to allocate. 0 is is the same as ~0 - unlimited */ + + struct ocfs2_alloc_reservation *ac_resv; }; void ocfs2_init_steal_slots(struct ocfs2_super *osb); diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index dee03197a494..cfe672e72b27 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -95,6 +95,7 @@ struct mount_options unsigned int atime_quantum; signed short slot; unsigned int localalloc_opt; + unsigned int resv_level; char cluster_stack[OCFS2_STACK_LABEL_LEN + 1]; }; @@ -176,6 +177,7 @@ enum { Opt_noacl, Opt_usrquota, Opt_grpquota, + Opt_resv_level, Opt_err, }; @@ -202,6 +204,7 @@ static const match_table_t tokens = { {Opt_noacl, "noacl"}, {Opt_usrquota, "usrquota"}, {Opt_grpquota, "grpquota"}, + {Opt_resv_level, "resv_level=%u"}, {Opt_err, NULL} }; @@ -1030,6 +1033,7 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent) osb->osb_commit_interval = parsed_options.commit_interval; osb->local_alloc_default_bits = ocfs2_megabytes_to_clusters(sb, parsed_options.localalloc_opt); osb->local_alloc_bits = osb->local_alloc_default_bits; + osb->osb_resv_level = parsed_options.resv_level; status = ocfs2_verify_userspace_stack(osb, &parsed_options); if (status) @@ -1290,6 +1294,7 @@ static int ocfs2_parse_options(struct super_block *sb, mopt->slot = OCFS2_INVALID_SLOT; mopt->localalloc_opt = OCFS2_DEFAULT_LOCAL_ALLOC_SIZE; mopt->cluster_stack[0] = '\0'; + mopt->resv_level = OCFS2_DEFAULT_RESV_LEVEL; if (!options) { status = 1; @@ -1433,6 +1438,17 @@ static int ocfs2_parse_options(struct super_block *sb, mopt->mount_opt |= OCFS2_MOUNT_NO_POSIX_ACL; mopt->mount_opt &= ~OCFS2_MOUNT_POSIX_ACL; break; + case Opt_resv_level: + if (is_remount) + break; + if (match_int(&args[0], &option)) { + status = 0; + goto bail; + } + if (option >= OCFS2_MIN_RESV_LEVEL && + option < OCFS2_MAX_RESV_LEVEL) + mopt->resv_level = option; + break; default: mlog(ML_ERROR, "Unrecognized mount option \"%s\" " @@ -1514,6 +1530,9 @@ static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt) else seq_printf(s, ",noacl"); + if (osb->osb_resv_level != OCFS2_DEFAULT_RESV_LEVEL) + seq_printf(s, ",resv_level=%d", osb->osb_resv_level); + return 0; } @@ -2042,6 +2061,12 @@ static int ocfs2_initialize_super(struct super_block *sb, init_waitqueue_head(&osb->osb_mount_event); + status = ocfs2_resmap_init(osb, &osb->osb_la_resmap); + if (status) { + mlog_errno(status); + goto bail; + } + osb->vol_label = kmalloc(OCFS2_MAX_VOL_LABEL_LEN, GFP_KERNEL); if (!osb->vol_label) { mlog(ML_ERROR, "unable to alloc vol label\n"); -- cgit v1.2.3 From b07f8f24dfe54da0f074b78949044842e8df881f Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Mon, 5 Apr 2010 18:17:15 -0700 Subject: ocfs2: change default reservation window sizes The default reservation size of 4 (32-bit windows) is a bit too ambitious. Scale it back to 16 bits (resv_level=2). I have been testing various sizes on a 4-node cluster which runs a mixed workload that is heavily threaded. With a 256MB local alloc, I get *roughly* the following levels of average file fragmentation: resv_level=0 70% resv_level=1 21% resv_level=2 23% resv_level=3 24% resv_level=4 60% resv_level=5 did not test resv_level=6 60% resv_level=2 seemed like a good compromise between not letting windows be too small, but not so big that heavier workloads will immediately suffer without tuning. This patch also change the behavior of directory reservations - they now track file reservations. The previous compromise of giving directory windows only 8 bits wound up fragmenting more at some window sizes because file allocations had smaller unused windows to poach from. Signed-off-by: Mark Fasheh Signed-off-by: Joel Becker --- Documentation/filesystems/ocfs2.txt | 2 +- fs/ocfs2/reservations.c | 7 ++++--- fs/ocfs2/reservations.h | 2 +- 3 files changed, 6 insertions(+), 5 deletions(-) (limited to 'Documentation/filesystems') diff --git a/Documentation/filesystems/ocfs2.txt b/Documentation/filesystems/ocfs2.txt index 412df9095937..32339e584a9a 100644 --- a/Documentation/filesystems/ocfs2.txt +++ b/Documentation/filesystems/ocfs2.txt @@ -80,6 +80,6 @@ user_xattr (*) Enables Extended User Attributes. nouser_xattr Disables Extended User Attributes. acl Enables POSIX Access Control Lists support. noacl (*) Disables POSIX Access Control Lists support. -resv_level=4 (*) Set how agressive allocation reservations will be. +resv_level=2 (*) Set how agressive allocation reservations will be. Valid values are between 0 (reservations off) to 8 (maximum space for reservations). diff --git a/fs/ocfs2/reservations.c b/fs/ocfs2/reservations.c index 7fc6cfee95f1..87fa35791f18 100644 --- a/fs/ocfs2/reservations.c +++ b/fs/ocfs2/reservations.c @@ -55,9 +55,10 @@ static unsigned int ocfs2_resv_window_bits(struct ocfs2_reservation_map *resmap, if (!(resv->r_flags & OCFS2_RESV_FLAG_DIR)) { /* 8, 16, 32, 64, 128, 256, 512, 1024 */ bits = 4 << osb->osb_resv_level; - } else - bits = OCFS2_RESV_DIR_WINDOW_BITS; - + } else { + /* For now, treat directories the same as files. */ + bits = 4 << osb->osb_resv_level; + } return bits; } diff --git a/fs/ocfs2/reservations.h b/fs/ocfs2/reservations.h index 34bb308375c5..022aff601e15 100644 --- a/fs/ocfs2/reservations.h +++ b/fs/ocfs2/reservations.h @@ -22,7 +22,7 @@ #include -#define OCFS2_DEFAULT_RESV_LEVEL 4 +#define OCFS2_DEFAULT_RESV_LEVEL 2 #define OCFS2_MAX_RESV_LEVEL 9 #define OCFS2_MIN_RESV_LEVEL 0 -- cgit v1.2.3 From 83f92318fa33cc084e14e64dc903e605f75884c1 Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Mon, 5 Apr 2010 18:17:16 -0700 Subject: ocfs2: Add dir_resv_level mount option The default behavior for directory reservations stays the same, but we add a mount option so people can tweak the size of directory reservations according to their workloads. Signed-off-by: Mark Fasheh Signed-off-by: Joel Becker --- Documentation/filesystems/ocfs2.txt | 4 ++++ fs/ocfs2/dir.c | 6 ++++-- fs/ocfs2/ocfs2.h | 1 + fs/ocfs2/reservations.c | 9 ++++++--- fs/ocfs2/reservations.h | 2 ++ fs/ocfs2/super.c | 23 +++++++++++++++++++++++ 6 files changed, 40 insertions(+), 5 deletions(-) (limited to 'Documentation/filesystems') diff --git a/Documentation/filesystems/ocfs2.txt b/Documentation/filesystems/ocfs2.txt index 32339e584a9a..1f7ae144f6d8 100644 --- a/Documentation/filesystems/ocfs2.txt +++ b/Documentation/filesystems/ocfs2.txt @@ -83,3 +83,7 @@ noacl (*) Disables POSIX Access Control Lists support. resv_level=2 (*) Set how agressive allocation reservations will be. Valid values are between 0 (reservations off) to 8 (maximum space for reservations). +dir_resv_level= (*) By default, directory reservations will scale with file + reservations - users should rarely need to change this + value. If allocation reservations are turned off, this + option will have no effect. diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c index 8563f97c58af..6c9a28a2d3ae 100644 --- a/fs/ocfs2/dir.c +++ b/fs/ocfs2/dir.c @@ -2977,7 +2977,8 @@ static int ocfs2_expand_inline_dir(struct inode *dir, struct buffer_head *di_bh, * if we only get one now, that's enough to continue. The rest * will be claimed after the conversion to extents. */ - data_ac->ac_resv = &oi->ip_la_data_resv; + if (ocfs2_dir_resv_allowed(osb)) + data_ac->ac_resv = &oi->ip_la_data_resv; ret = ocfs2_claim_clusters(osb, handle, data_ac, 1, &bit_off, &len); if (ret) { mlog_errno(ret); @@ -3348,7 +3349,8 @@ static int ocfs2_extend_dir(struct ocfs2_super *osb, goto bail; } - data_ac->ac_resv = &OCFS2_I(dir)->ip_la_data_resv; + if (ocfs2_dir_resv_allowed(osb)) + data_ac->ac_resv = &OCFS2_I(dir)->ip_la_data_resv; credits = ocfs2_calc_extend_credits(sb, el, 1); } else { diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h index 09d7aee3dabe..a388528f485c 100644 --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -356,6 +356,7 @@ struct ocfs2_super struct ocfs2_reservation_map osb_la_resmap; unsigned int osb_resv_level; + unsigned int osb_dir_resv_level; /* Next three fields are for local node slot recovery during * mount. */ diff --git a/fs/ocfs2/reservations.c b/fs/ocfs2/reservations.c index 87fa35791f18..6497bcc00fa5 100644 --- a/fs/ocfs2/reservations.c +++ b/fs/ocfs2/reservations.c @@ -44,7 +44,11 @@ DEFINE_SPINLOCK(resv_lock); #define OCFS2_MIN_RESV_WINDOW_BITS 8 #define OCFS2_MAX_RESV_WINDOW_BITS 1024 -#define OCFS2_RESV_DIR_WINDOW_BITS OCFS2_MIN_RESV_WINDOW_BITS + +int ocfs2_dir_resv_allowed(struct ocfs2_super *osb) +{ + return (osb->osb_resv_level && osb->osb_dir_resv_level); +} static unsigned int ocfs2_resv_window_bits(struct ocfs2_reservation_map *resmap, struct ocfs2_alloc_reservation *resv) @@ -56,8 +60,7 @@ static unsigned int ocfs2_resv_window_bits(struct ocfs2_reservation_map *resmap, /* 8, 16, 32, 64, 128, 256, 512, 1024 */ bits = 4 << osb->osb_resv_level; } else { - /* For now, treat directories the same as files. */ - bits = 4 << osb->osb_resv_level; + bits = 4 << osb->osb_dir_resv_level; } return bits; } diff --git a/fs/ocfs2/reservations.h b/fs/ocfs2/reservations.h index 022aff601e15..25b0c0e31e91 100644 --- a/fs/ocfs2/reservations.h +++ b/fs/ocfs2/reservations.h @@ -67,6 +67,8 @@ void ocfs2_resv_init_once(struct ocfs2_alloc_reservation *resv); void ocfs2_resv_set_type(struct ocfs2_alloc_reservation *resv, unsigned int flags); +int ocfs2_dir_resv_allowed(struct ocfs2_super *osb); + /** * ocfs2_resv_discard() - truncate a reservation * @resmap: diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 5745682eb1c0..79d7d4cf45b1 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -96,6 +96,7 @@ struct mount_options signed short slot; int localalloc_opt; unsigned int resv_level; + int dir_resv_level; char cluster_stack[OCFS2_STACK_LABEL_LEN + 1]; }; @@ -178,6 +179,7 @@ enum { Opt_usrquota, Opt_grpquota, Opt_resv_level, + Opt_dir_resv_level, Opt_err, }; @@ -205,6 +207,7 @@ static const match_table_t tokens = { {Opt_usrquota, "usrquota"}, {Opt_grpquota, "grpquota"}, {Opt_resv_level, "resv_level=%u"}, + {Opt_dir_resv_level, "dir_resv_level=%u"}, {Opt_err, NULL} }; @@ -1034,6 +1037,11 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent) ocfs2_la_set_sizes(osb, parsed_options.localalloc_opt); osb->osb_resv_level = parsed_options.resv_level; + osb->osb_dir_resv_level = parsed_options.resv_level; + if (parsed_options.dir_resv_level == -1) + osb->osb_dir_resv_level = parsed_options.resv_level; + else + osb->osb_dir_resv_level = parsed_options.dir_resv_level; status = ocfs2_verify_userspace_stack(osb, &parsed_options); if (status) @@ -1295,6 +1303,7 @@ static int ocfs2_parse_options(struct super_block *sb, mopt->localalloc_opt = -1; mopt->cluster_stack[0] = '\0'; mopt->resv_level = OCFS2_DEFAULT_RESV_LEVEL; + mopt->dir_resv_level = -1; if (!options) { status = 1; @@ -1449,6 +1458,17 @@ static int ocfs2_parse_options(struct super_block *sb, option < OCFS2_MAX_RESV_LEVEL) mopt->resv_level = option; break; + case Opt_dir_resv_level: + if (is_remount) + break; + if (match_int(&args[0], &option)) { + status = 0; + goto bail; + } + if (option >= OCFS2_MIN_RESV_LEVEL && + option < OCFS2_MAX_RESV_LEVEL) + mopt->dir_resv_level = option; + break; default: mlog(ML_ERROR, "Unrecognized mount option \"%s\" " @@ -1533,6 +1553,9 @@ static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt) if (osb->osb_resv_level != OCFS2_DEFAULT_RESV_LEVEL) seq_printf(s, ",resv_level=%d", osb->osb_resv_level); + if (osb->osb_dir_resv_level != osb->osb_resv_level) + seq_printf(s, ",dir_resv_level=%d", osb->osb_resv_level); + return 0; } -- cgit v1.2.3 From 277a6a34175dcb0ee98dceee619e0e3190347a25 Mon Sep 17 00:00:00 2001 From: Ryusuke Konishi Date: Fri, 2 Apr 2010 18:02:33 +0900 Subject: nilfs2: change default of 'errors' mount option to 'remount-ro' mode Like ext3, nilfs has 'errors' mount option to allow specifying desired behavior on severe errors. Currently, the default action is 'errors=continue' and has potential to advance filesystem corruption for severe errors. This will change the action to 'errors=remount-ro' to avoid the issue. Signed-off-by: Ryusuke Konishi --- Documentation/filesystems/nilfs2.txt | 4 ++-- fs/nilfs2/super.c | 6 +++--- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'Documentation/filesystems') diff --git a/Documentation/filesystems/nilfs2.txt b/Documentation/filesystems/nilfs2.txt index cf6d0d85ca82..d3e7673995eb 100644 --- a/Documentation/filesystems/nilfs2.txt +++ b/Documentation/filesystems/nilfs2.txt @@ -50,8 +50,8 @@ NILFS2 supports the following mount options: (*) == default nobarrier Disables barriers. -errors=continue(*) Keep going on a filesystem error. -errors=remount-ro Remount the filesystem read-only on an error. +errors=continue Keep going on a filesystem error. +errors=remount-ro(*) Remount the filesystem read-only on an error. errors=panic Panic and halt the machine if an error occurs. cp=n Specify the checkpoint-number of the snapshot to be mounted. Checkpoints and snapshots are listed by lscp diff --git a/fs/nilfs2/super.c b/fs/nilfs2/super.c index 48145f505a6a..0b1758bf0726 100644 --- a/fs/nilfs2/super.c +++ b/fs/nilfs2/super.c @@ -470,10 +470,10 @@ static int nilfs_show_options(struct seq_file *seq, struct vfsmount *vfs) if (nilfs_test_opt(sbi, SNAPSHOT)) seq_printf(seq, ",cp=%llu", (unsigned long long int)sbi->s_snapshot_cno); - if (nilfs_test_opt(sbi, ERRORS_RO)) - seq_printf(seq, ",errors=remount-ro"); if (nilfs_test_opt(sbi, ERRORS_PANIC)) seq_printf(seq, ",errors=panic"); + if (nilfs_test_opt(sbi, ERRORS_CONT)) + seq_printf(seq, ",errors=continue"); if (nilfs_test_opt(sbi, STRICT_ORDER)) seq_printf(seq, ",order=strict"); if (nilfs_test_opt(sbi, NORECOVERY)) @@ -631,7 +631,7 @@ nilfs_set_default_options(struct nilfs_sb_info *sbi, struct nilfs_super_block *sbp) { sbi->s_mount_opt = - NILFS_MOUNT_ERRORS_CONT | NILFS_MOUNT_BARRIER; + NILFS_MOUNT_ERRORS_RO | NILFS_MOUNT_BARRIER; } static int nilfs_setup_super(struct nilfs_sb_info *sbi) -- cgit v1.2.3 From a8cd4561ea176f51e9f4707873ca4eff8fd5ee70 Mon Sep 17 00:00:00 2001 From: Anand Gadiyar Date: Mon, 10 May 2010 14:51:19 +0530 Subject: fix "seperate" typos in comments s/seperate/separate Signed-off-by: Anand Gadiyar Signed-off-by: Jiri Kosina --- Documentation/filesystems/logfs.txt | 8 ++++---- arch/arm/mach-s3c2443/clock.c | 2 +- arch/arm/plat-samsung/include/plat/gpio-core.h | 2 +- drivers/usb/host/ehci-omap.c | 2 +- fs/logfs/dir.c | 2 +- fs/logfs/logfs.h | 2 +- fs/logfs/logfs_abi.h | 10 +++++----- 7 files changed, 14 insertions(+), 14 deletions(-) (limited to 'Documentation/filesystems') diff --git a/Documentation/filesystems/logfs.txt b/Documentation/filesystems/logfs.txt index e64c94ba401a..bca42c22a143 100644 --- a/Documentation/filesystems/logfs.txt +++ b/Documentation/filesystems/logfs.txt @@ -59,7 +59,7 @@ Levels ------ Garbage collection (GC) may fail if all data is written -indiscriminately. One requirement of GC is that data is seperated +indiscriminately. One requirement of GC is that data is separated 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, @@ -67,7 +67,7 @@ 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. +each level requires its own separate segment to be open for writing. Inode File ---------- @@ -106,9 +106,9 @@ Vim --- By cleverly predicting the life time of data, it is possible to -seperate long-living data from short-living data and thereby reduce +separate 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 +have a separate 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. diff --git a/arch/arm/mach-s3c2443/clock.c b/arch/arm/mach-s3c2443/clock.c index 62cd4eaee01b..8a55e78f7090 100644 --- a/arch/arm/mach-s3c2443/clock.c +++ b/arch/arm/mach-s3c2443/clock.c @@ -391,7 +391,7 @@ static struct clk clk_hsmmc = { /* i2s_eplldiv * - * This clock is the output from the I2S divisor of ESYSCLK, and is seperate + * This clock is the output from the I2S divisor of ESYSCLK, and is separate * from the mux that comes after it (cannot merge into one single clock) */ diff --git a/arch/arm/plat-samsung/include/plat/gpio-core.h b/arch/arm/plat-samsung/include/plat/gpio-core.h index 49ff406a7066..16643e3029fc 100644 --- a/arch/arm/plat-samsung/include/plat/gpio-core.h +++ b/arch/arm/plat-samsung/include/plat/gpio-core.h @@ -97,7 +97,7 @@ extern void s3c_gpiolib_add(struct s3c_gpio_chip *chip); * others = Special functions (dependant on bank) * * Note, since the code to deal with the case where there are two control - * registers instead of one, we do not have a seperate set of function + * registers instead of one, we do not have a separate set of function * (samsung_gpiolib_add_4bit2_chips)for each case. */ extern void samsung_gpiolib_add_4bit_chips(struct s3c_gpio_chip *chip, diff --git a/drivers/usb/host/ehci-omap.c b/drivers/usb/host/ehci-omap.c index a67a0030dd57..bed6de342ecc 100644 --- a/drivers/usb/host/ehci-omap.c +++ b/drivers/usb/host/ehci-omap.c @@ -181,7 +181,7 @@ struct ehci_hcd_omap { void __iomem *ehci_base; /* Regulators for USB PHYs. - * Each PHY can have a seperate regulator. + * Each PHY can have a separate regulator. */ struct regulator *regulator[OMAP3_HS_USB_PORTS]; }; diff --git a/fs/logfs/dir.c b/fs/logfs/dir.c index 2396a85c0f55..72d1893ddd36 100644 --- a/fs/logfs/dir.c +++ b/fs/logfs/dir.c @@ -12,7 +12,7 @@ * 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 + * created/removed/altered in separate operations. Therefore we need to do * a small amount of journaling. * * Create, link, mkdir, mknod and symlink all share the same function to do diff --git a/fs/logfs/logfs.h b/fs/logfs/logfs.h index 0a3df1a0c936..9028d7f83bbd 100644 --- a/fs/logfs/logfs.h +++ b/fs/logfs/logfs.h @@ -704,7 +704,7 @@ 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 */ + /* ifile has separate areas */ level += LOGFS_MAX_LEVELS; } return (__force gc_level_t)level; diff --git a/fs/logfs/logfs_abi.h b/fs/logfs/logfs_abi.h index f674725663fe..ae960519c54a 100644 --- a/fs/logfs/logfs_abi.h +++ b/fs/logfs/logfs_abi.h @@ -50,9 +50,9 @@ static inline void check_##type(void) \ * 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 + * Levels 1-11 are necessary for robust gc operations and help separate * short-lived metadata from longer-lived file data. In the future, - * file data should get seperated into several segments based on simple + * file data should get separated 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 @@ -117,7 +117,7 @@ static inline void check_##type(void) \ #define pure_ofs(ofs) (ofs & ~LOGFS_FULLY_POPULATED) /* - * LogFS needs to seperate data into levels. Each level is defined as the + * LogFS needs to separate 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. @@ -204,7 +204,7 @@ SIZE_CHECK(logfs_segment_header, LOGFS_SEGMENT_HEADERSIZE); * @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 + * @ds_data_levels: number of separate levels for data * @pad0: reserved, must be 0 * @ds_feature_incompat: incompatible filesystem features * @ds_feature_ro_compat: read-only compatible filesystem features @@ -456,7 +456,7 @@ enum logfs_vim { * @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 + * one area per GC level. Several may be used to separate 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. -- cgit v1.2.3 From ca0dbd86b12be9af7cda230890eb741d5cb8b624 Mon Sep 17 00:00:00 2001 From: Thadeu Lima de Souza Cascardo Date: Fri, 7 May 2010 16:52:26 -0300 Subject: doc: inode uses a mutex instead of a semaphore. Replace the introduced i_sem by an i_mutex in the filesystem locking documentation. This was introduced [1] after all occurrences were already replaced in the same text [2]. However, the term "inode semaphore" has not been replaced then, and it's replaced now. [1] afddba49d18f346e5cc2938b6ed7c512db18ca68 [2] a7bc02f4f47fd0e7860c6589f0ad000d1476f7a3 Signed-off-by: Thadeu Lima de Souza Cascardo Cc: Nick Piggin Cc: Artem Bityutskiy Cc: Randy Dunlap Cc: Christoph Hellwig Cc: Al Viro Signed-off-by: Jiri Kosina --- Documentation/filesystems/Locking | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'Documentation/filesystems') diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking index 06bbbed71206..af1608070cd5 100644 --- a/Documentation/filesystems/Locking +++ b/Documentation/filesystems/Locking @@ -178,7 +178,7 @@ prototypes: locking rules: All except set_page_dirty may block - BKL PageLocked(page) i_sem + BKL PageLocked(page) i_mutex writepage: no yes, unlocks (see below) readpage: no yes, unlocks sync_page: no maybe @@ -429,7 +429,7 @@ check_flags: no implementations. If your fs is not using generic_file_llseek, you need to acquire and release the appropriate locks in your ->llseek(). For many filesystems, it is probably safe to acquire the inode -semaphore. Note some filesystems (i.e. remote ones) provide no +mutex. Note some filesystems (i.e. remote ones) provide no protection for i_size so you will need to use the BKL. Note: ext2_release() was *the* source of contention on fs-intensive -- cgit v1.2.3 From 4dc6ec00f6347b72312fa41dfc587d5302b05544 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Mon, 19 Apr 2010 15:11:28 -0400 Subject: nfsd4: implement reclaim_complete This is a mandatory operation. Also, here (not in open) is where we should be committing the reboot recovery information. Signed-off-by: J. Bruce Fields --- Documentation/filesystems/nfs/nfs41-server.txt | 2 +- fs/nfsd/nfs4proc.c | 5 ++++ fs/nfsd/nfs4state.c | 33 +++++++++++++++++++++++--- fs/nfsd/nfs4xdr.c | 12 +++++++++- fs/nfsd/xdr4.h | 6 +++++ 5 files changed, 53 insertions(+), 5 deletions(-) (limited to 'Documentation/filesystems') diff --git a/Documentation/filesystems/nfs/nfs41-server.txt b/Documentation/filesystems/nfs/nfs41-server.txt index 6a53a84afc72..04884914a1c8 100644 --- a/Documentation/filesystems/nfs/nfs41-server.txt +++ b/Documentation/filesystems/nfs/nfs41-server.txt @@ -137,7 +137,7 @@ NS*| OPENATTR | OPT | | Section 18.17 | | READ | REQ | | Section 18.22 | | READDIR | REQ | | Section 18.23 | | READLINK | OPT | | Section 18.24 | -NS | RECLAIM_COMPLETE | REQ | | Section 18.51 | + | RECLAIM_COMPLETE | REQ | | Section 18.51 | | RELEASE_LOCKOWNER | MNI | | N/A | | REMOVE | REQ | | Section 18.25 | | RENAME | REQ | | Section 18.26 | diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index e2dc9608281b..59ec449b0c7f 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -1312,6 +1312,11 @@ static struct nfsd4_operation nfsd4_ops[] = { .op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP, .op_name = "OP_SEQUENCE", }, + [OP_RECLAIM_COMPLETE] = { + .op_func = (nfsd4op_func)nfsd4_reclaim_complete, + .op_flags = ALLOWED_WITHOUT_FH, + .op_name = "OP_RECLAIM_COMPLETE", + }, }; static const char *nfsd4_op_name(unsigned opnum) diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index ede9dde52fe8..84b0fe9a262a 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -1501,6 +1501,35 @@ out: return status; } +__be32 +nfsd4_reclaim_complete(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_reclaim_complete *rc) +{ + if (rc->rca_one_fs) { + if (!cstate->current_fh.fh_dentry) + return nfserr_nofilehandle; + /* + * We don't take advantage of the rca_one_fs case. + * That's OK, it's optional, we can safely ignore it. + */ + return nfs_ok; + } + nfs4_lock_state(); + if (is_client_expired(cstate->session->se_client)) { + nfs4_unlock_state(); + /* + * The following error isn't really legal. + * But we only get here if the client just explicitly + * destroyed the client. Surely it no longer cares what + * error it gets back on an operation for the dead + * client. + */ + return nfserr_stale_clientid; + } + nfsd4_create_clid_dir(cstate->session->se_client); + nfs4_unlock_state(); + return nfs_ok; +} + __be32 nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_setclientid *setclid) @@ -2510,10 +2539,8 @@ 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 126d0caabb3c..ac17a7080239 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -1234,6 +1234,16 @@ nfsd4_decode_sequence(struct nfsd4_compoundargs *argp, DECODE_TAIL; } +static __be32 nfsd4_decode_reclaim_complete(struct nfsd4_compoundargs *argp, struct nfsd4_reclaim_complete *rc) +{ + DECODE_HEAD; + + READ_BUF(4); + READ32(rc->rca_one_fs); + + DECODE_TAIL; +} + static __be32 nfsd4_decode_noop(struct nfsd4_compoundargs *argp, void *p) { @@ -1346,7 +1356,7 @@ static nfsd4_dec nfsd41_dec_ops[] = { [OP_TEST_STATEID] = (nfsd4_dec)nfsd4_decode_notsupp, [OP_WANT_DELEGATION] = (nfsd4_dec)nfsd4_decode_notsupp, [OP_DESTROY_CLIENTID] = (nfsd4_dec)nfsd4_decode_notsupp, - [OP_RECLAIM_COMPLETE] = (nfsd4_dec)nfsd4_decode_notsupp, + [OP_RECLAIM_COMPLETE] = (nfsd4_dec)nfsd4_decode_reclaim_complete, }; struct nfsd4_minorversion_ops { diff --git a/fs/nfsd/xdr4.h b/fs/nfsd/xdr4.h index c28958ec216c..4d476ff08ae6 100644 --- a/fs/nfsd/xdr4.h +++ b/fs/nfsd/xdr4.h @@ -381,6 +381,10 @@ struct nfsd4_destroy_session { struct nfs4_sessionid sessionid; }; +struct nfsd4_reclaim_complete { + u32 rca_one_fs; +}; + struct nfsd4_op { int opnum; __be32 status; @@ -421,6 +425,7 @@ struct nfsd4_op { struct nfsd4_create_session create_session; struct nfsd4_destroy_session destroy_session; struct nfsd4_sequence sequence; + struct nfsd4_reclaim_complete reclaim_complete; } u; struct nfs4_replay * replay; }; @@ -523,6 +528,7 @@ extern __be32 nfsd4_sequence(struct svc_rqst *, extern __be32 nfsd4_destroy_session(struct svc_rqst *, struct nfsd4_compound_state *, struct nfsd4_destroy_session *); +__be32 nfsd4_reclaim_complete(struct svc_rqst *, struct nfsd4_compound_state *, struct nfsd4_reclaim_complete *); extern __be32 nfsd4_process_open1(struct nfsd4_compound_state *, struct nfsd4_open *open); extern __be32 nfsd4_process_open2(struct svc_rqst *rqstp, -- cgit v1.2.3 From b9d8b45ee3c5f62cdbd34ee006f3dd2ac51f7018 Mon Sep 17 00:00:00 2001 From: "Serge E. Hallyn" Date: Tue, 4 May 2010 21:45:38 -0500 Subject: sysfs-namespaces: add a high-level Documentation file The first three paragraphs are almost verbatim taken from Eric's commit message on the patch introducing network ns tags. The next two paragraphs I wrote to be a brief high level overview. The last section is taken from the commit message on "Implement sysfs tagged directory support", but updated. Hopefully correctly. Signed-off-by: Serge E. Hallyn Cc: Eric W. Biederman Cc: Tejun Heo Signed-off-by: Greg Kroah-Hartman --- Documentation/filesystems/sysfs-tagging.txt | 42 +++++++++++++++++++++++++++++ 1 file changed, 42 insertions(+) create mode 100644 Documentation/filesystems/sysfs-tagging.txt (limited to 'Documentation/filesystems') diff --git a/Documentation/filesystems/sysfs-tagging.txt b/Documentation/filesystems/sysfs-tagging.txt new file mode 100644 index 000000000000..caaaf1266d8f --- /dev/null +++ b/Documentation/filesystems/sysfs-tagging.txt @@ -0,0 +1,42 @@ +Sysfs tagging +------------- + +(Taken almost verbatim from Eric Biederman's netns tagging patch +commit msg) + +The problem. Network devices show up in sysfs and with the network +namespace active multiple devices with the same name can show up in +the same directory, ouch! + +To avoid that problem and allow existing applications in network +namespaces to see the same interface that is currently presented in +sysfs, sysfs now has tagging directory support. + +By using the network namespace pointers as tags to separate out the +the sysfs directory entries we ensure that we don't have conflicts +in the directories and applications only see a limited set of +the network devices. + +Each sysfs directory entry may be tagged with zero or one +namespaces. A sysfs_dirent is augmented with a void *s_ns. If a +directory entry is tagged, then sysfs_dirent->s_flags will have a +flag between KOBJ_NS_TYPE_NONE and KOBJ_NS_TYPES, and s_ns will +point to the namespace to which it belongs. + +Each sysfs superblock's sysfs_super_info contains an array void +*ns[KOBJ_NS_TYPES]. When a a task in a tagging namespace +kobj_nstype first mounts sysfs, a new superblock is created. It +will be differentiated from other sysfs mounts by having its +s_fs_info->ns[kobj_nstype] set to the new namespace. Note that +through bind mounting and mounts propagation, a task can easily view +the contents of other namespaces' sysfs mounts. Therefore, when a +namespace exits, it will call kobj_ns_exit() to invalidate any +sysfs_dirent->s_ns pointers pointing to it. + +Users of this interface: +- define a type in the kobj_ns_type enumeration. +- call kobj_ns_type_register() with its kobj_ns_type_operations which has + - current_ns() which returns current's namespace + - netlink_ns() which returns a socket's namespace + - initial_ns() which returns the initial namesapce +- call kobj_ns_exit() when an individual tag is no longer valid -- cgit v1.2.3 From 0636c73ee7b129f77f577aaaefc8dde057be6d18 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Fri, 30 Apr 2010 11:09:34 -0500 Subject: ext3: make barrier options consistent with ext4 ext4 was updated to accept barrier/nobarrier mount options in addition to the older barrier=0/1. The barrier story is complex enough, we should help people by making the options the same at least, even if the defaults are different. This patch allows the barrier/nobarrier mount options for ext3, while keeping nobarrier the default. It also unconditionally displays barrier status in show_options, and prints a message at mount time if barriers are not enabled, just as ext4 does. Signed-off-by: Eric Sandeen Signed-off-by: Jan Kara --- Documentation/filesystems/ext3.txt | 15 +++++++++++++-- fs/ext3/super.c | 33 ++++++++++++++++++++++++++------- 2 files changed, 39 insertions(+), 9 deletions(-) (limited to 'Documentation/filesystems') diff --git a/Documentation/filesystems/ext3.txt b/Documentation/filesystems/ext3.txt index 867c5b50cb42..272f80d5f966 100644 --- a/Documentation/filesystems/ext3.txt +++ b/Documentation/filesystems/ext3.txt @@ -59,8 +59,19 @@ commit=nrsec (*) Ext3 can be told to sync all its data and metadata Setting it to very large values will improve performance. -barrier=1 This enables/disables barriers. barrier=0 disables - it, barrier=1 enables it. +barrier=<0(*)|1> This enables/disables the use of write barriers in +barrier the jbd code. barrier=0 disables, barrier=1 enables. +nobarrier (*) This also requires an IO stack which can support + barriers, and if jbd gets an error on a barrier + write, it will disable again with a warning. + Write barriers enforce proper on-disk ordering + of journal commits, making volatile disk write caches + safe to use, at some performance penalty. If + your disks are battery-backed in one way or another, + disabling barriers may safely improve performance. + The mount options "barrier" and "nobarrier" can + also be used to enable or disable barriers, for + consistency with other ext3 mount options. orlov (*) This enables the new Orlov block allocator. It is enabled by default. diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 6b6e49de0916..0fc1293d0e96 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -653,8 +653,12 @@ static int ext3_show_options(struct seq_file *seq, struct vfsmount *vfs) seq_printf(seq, ",commit=%u", (unsigned) (sbi->s_commit_interval / HZ)); } - if (test_opt(sb, BARRIER)) - seq_puts(seq, ",barrier=1"); + + /* + * Always display barrier state so it's clear what the status is. + */ + seq_puts(seq, ",barrier="); + seq_puts(seq, test_opt(sb, BARRIER) ? "1" : "0"); if (test_opt(sb, NOBH)) seq_puts(seq, ",nobh"); @@ -810,8 +814,8 @@ enum { Opt_data_err_abort, Opt_data_err_ignore, Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota, Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_jqfmt_vfsv1, Opt_quota, - Opt_noquota, Opt_ignore, Opt_barrier, Opt_err, Opt_resize, - Opt_usrquota, Opt_grpquota + Opt_noquota, Opt_ignore, Opt_barrier, Opt_nobarrier, Opt_err, + Opt_resize, Opt_usrquota, Opt_grpquota }; static const match_table_t tokens = { @@ -865,6 +869,8 @@ static const match_table_t tokens = { {Opt_quota, "quota"}, {Opt_usrquota, "usrquota"}, {Opt_barrier, "barrier=%u"}, + {Opt_barrier, "barrier"}, + {Opt_nobarrier, "nobarrier"}, {Opt_resize, "resize"}, {Opt_err, NULL}, }; @@ -967,7 +973,11 @@ static int parse_options (char *options, struct super_block *sb, int token; if (!*p) continue; - + /* + * Initialize args struct so we know whether arg was + * found; some options take optional arguments. + */ + args[0].to = args[0].from = 0; token = match_token(p, tokens, args); switch (token) { case Opt_bsd_df: @@ -1215,9 +1225,15 @@ set_qf_format: case Opt_abort: set_opt(sbi->s_mount_opt, ABORT); break; + case Opt_nobarrier: + clear_opt(sbi->s_mount_opt, BARRIER); + break; case Opt_barrier: - if (match_int(&args[0], &option)) - return 0; + if (args[0].from) { + if (match_int(&args[0], &option)) + return 0; + } else + option = 1; /* No argument, default to 1 */ if (option) set_opt(sbi->s_mount_opt, BARRIER); else @@ -2276,6 +2292,9 @@ static int ext3_load_journal(struct super_block *sb, return -EINVAL; } + if (!(journal->j_flags & JFS_BARRIER)) + printk(KERN_INFO "EXT3-fs: barriers not enabled\n"); + if (!really_read_only && test_opt(sb, UPDATE_JOURNAL)) { err = journal_update_format(journal); if (err) { -- cgit v1.2.3