From b1999477ed91c3c33891acfe0e18a4457e5c4915 Mon Sep 17 00:00:00 2001 From: Guo Zeng Date: Tue, 14 Apr 2015 11:55:55 +0000 Subject: ARM: prima2: move to use REGMAP APIs for rtciobrg all devices behind rtciobrg needs a special way to access. currently they are using a platform-specific API. this patch moves to REGMAP, then clients can use regmap APIs to read/write. for the moment, old APIs are still kept, once all clients move to regmap, old APIs will be dropped. this patch also does minor clean for comments, authors statement. Signed-off-by: Guo Zeng Signed-off-by: Barry Song --- include/linux/rtc/sirfsoc_rtciobrg.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/linux') diff --git a/include/linux/rtc/sirfsoc_rtciobrg.h b/include/linux/rtc/sirfsoc_rtciobrg.h index 2c92e1c8e055..aefd997262e4 100644 --- a/include/linux/rtc/sirfsoc_rtciobrg.h +++ b/include/linux/rtc/sirfsoc_rtciobrg.h @@ -9,10 +9,14 @@ #ifndef _SIRFSOC_RTC_IOBRG_H_ #define _SIRFSOC_RTC_IOBRG_H_ +struct regmap_config; + extern void sirfsoc_rtc_iobrg_besyncing(void); extern u32 sirfsoc_rtc_iobrg_readl(u32 addr); extern void sirfsoc_rtc_iobrg_writel(u32 val, u32 addr); +struct regmap *devm_regmap_init_iobg(struct device *dev, + const struct regmap_config *config); #endif -- cgit v1.2.3 From 31f02455455d405320e2f749696bef4e02903b35 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Tue, 30 Jun 2015 12:07:17 -0400 Subject: sparse: fix misplaced __pmem definition Move the definition of __pmem outside of CONFIG_SPARSE_RCU_POINTER to fix: drivers/nvdimm/pmem.c:198:17: sparse: too many arguments for function __builtin_expect drivers/nvdimm/pmem.c:36:33: sparse: expected ; at end of declaration drivers/nvdimm/pmem.c:48:21: sparse: void declaration ...due to __pmem failing to be defined in some configurations when CONFIG_SPARSE_RCU_POINTER=y. Reported-by: kbuild test robot Reported-by: Dan Carpenter Signed-off-by: Dan Williams --- include/linux/compiler.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/compiler.h b/include/linux/compiler.h index 26fc8bc77f85..d8fbd500330e 100644 --- a/include/linux/compiler.h +++ b/include/linux/compiler.h @@ -17,11 +17,11 @@ # define __release(x) __context__(x,-1) # define __cond_lock(x,c) ((c) ? ({ __acquire(x); 1; }) : 0) # define __percpu __attribute__((noderef, address_space(3))) +# define __pmem __attribute__((noderef, address_space(5))) #ifdef CONFIG_SPARSE_RCU_POINTER # define __rcu __attribute__((noderef, address_space(4))) #else # define __rcu -# define __pmem __attribute__((noderef, address_space(5))) #endif extern void __chk_user_ptr(const volatile void __user *); extern void __chk_io_ptr(const volatile void __iomem *); -- cgit v1.2.3 From bd7ade3cd9b0850264306f5c2b79024a417b6396 Mon Sep 17 00:00:00 2001 From: Nikolay Borisov Date: Thu, 2 Jul 2015 01:32:44 -0400 Subject: bufferhead: Add _gfp version for sb_getblk() sb_getblk() is used during ext4 (and possibly other FSes) writeback paths. Sometimes such path require allocating memory and guaranteeing that such allocation won't block. Currently, however, there is no way to provide user flags for sb_getblk which could lead to deadlocks. This patch implements a sb_getblk_gfp with the only difference it can accept user-provided GFP flags. Signed-off-by: Nikolay Borisov Signed-off-by: Theodore Ts'o Cc: stable@vger.kernel.org --- include/linux/buffer_head.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'include/linux') diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h index 73b45225a7ca..e6797ded700e 100644 --- a/include/linux/buffer_head.h +++ b/include/linux/buffer_head.h @@ -317,6 +317,13 @@ sb_getblk(struct super_block *sb, sector_t block) return __getblk_gfp(sb->s_bdev, block, sb->s_blocksize, __GFP_MOVABLE); } + +static inline struct buffer_head * +sb_getblk_gfp(struct super_block *sb, sector_t block, gfp_t gfp) +{ + return __getblk_gfp(sb->s_bdev, block, sb->s_blocksize, gfp); +} + static inline struct buffer_head * sb_find_get_block(struct super_block *sb, sector_t block) { -- cgit v1.2.3 From 0294112ee3135fbd15eaa70015af8283642dd970 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 4 Jul 2015 03:09:03 +0200 Subject: ACPI / PNP: Reserve ACPI resources at the fs_initcall_sync stage This effectively reverts the following three commits: 7bc10388ccdd ACPI / resources: free memory on error in add_region_before() 0f1b414d1907 ACPI / PNP: Avoid conflicting resource reservations b9a5e5e18fbf ACPI / init: Fix the ordering of acpi_reserve_resources() (commit b9a5e5e18fbf introduced regressions some of which, but not all, were addressed by commit 0f1b414d1907 and commit 7bc10388ccdd was a fixup on top of the latter) and causes ACPI fixed hardware resources to be reserved at the fs_initcall_sync stage of system initialization. The story is as follows. First, a boot regression was reported due to an apparent resource reservation ordering change after a commit that shouldn't lead to such changes. Investigation led to the conclusion that the problem happened because acpi_reserve_resources() was executed at the device_initcall() stage of system initialization which wasn't strictly ordered with respect to driver initialization (and with respect to the initialization of the pcieport driver in particular), so a random change causing the device initcalls to be run in a different order might break things. The response to that was to attempt to run acpi_reserve_resources() as soon as we knew that ACPI would be in use (commit b9a5e5e18fbf). However, that turned out to be too early, because it caused resource reservations made by the PNP system driver to fail on at least one system and that failure was addressed by commit 0f1b414d1907. That fix still turned out to be insufficient, though, because calling acpi_reserve_resources() before the fs_initcall stage of system initialization caused a boot regression to happen on the eCAFE EC-800-H20G/S netbook. That meant that we only could call acpi_reserve_resources() at the fs_initcall initialization stage or later, but then we might just as well call it after the PNP initalization in which case commit 0f1b414d1907 wouldn't be necessary any more. For this reason, the changes made by commit 0f1b414d1907 are reverted (along with a memory leak fixup on top of that commit), the changes made by commit b9a5e5e18fbf that went too far are reverted too and acpi_reserve_resources() is changed into fs_initcall_sync, which will cause it to be executed after the PNP subsystem initialization (which is an fs_initcall) and before device initcalls (including the pcieport driver initialization) which should avoid the initial issue. Link: https://bugzilla.kernel.org/show_bug.cgi?id=100581 Link: http://marc.info/?t=143092384600002&r=1&w=2 Link: https://bugzilla.kernel.org/show_bug.cgi?id=99831 Link: http://marc.info/?t=143389402600001&r=1&w=2 Fixes: b9a5e5e18fbf "ACPI / init: Fix the ordering of acpi_reserve_resources()" Reported-by: Roland Dreier Cc: All applicable Signed-off-by: Rafael J. Wysocki --- drivers/acpi/osl.c | 12 +++- drivers/acpi/resource.c | 162 ------------------------------------------------ drivers/pnp/system.c | 35 +++-------- include/linux/acpi.h | 10 --- 4 files changed, 18 insertions(+), 201 deletions(-) (limited to 'include/linux') diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c index c262e4acd68d..3b8963f21b36 100644 --- a/drivers/acpi/osl.c +++ b/drivers/acpi/osl.c @@ -175,10 +175,14 @@ static void __init acpi_request_region (struct acpi_generic_address *gas, if (!addr || !length) return; - acpi_reserve_region(addr, length, gas->space_id, 0, desc); + /* Resources are never freed */ + if (gas->space_id == ACPI_ADR_SPACE_SYSTEM_IO) + request_region(addr, length, desc); + else if (gas->space_id == ACPI_ADR_SPACE_SYSTEM_MEMORY) + request_mem_region(addr, length, desc); } -static void __init acpi_reserve_resources(void) +static int __init acpi_reserve_resources(void) { acpi_request_region(&acpi_gbl_FADT.xpm1a_event_block, acpi_gbl_FADT.pm1_event_length, "ACPI PM1a_EVT_BLK"); @@ -207,7 +211,10 @@ static void __init acpi_reserve_resources(void) if (!(acpi_gbl_FADT.gpe1_block_length & 0x1)) acpi_request_region(&acpi_gbl_FADT.xgpe1_block, acpi_gbl_FADT.gpe1_block_length, "ACPI GPE1_BLK"); + + return 0; } +fs_initcall_sync(acpi_reserve_resources); void acpi_os_printf(const char *fmt, ...) { @@ -1862,7 +1869,6 @@ acpi_status __init acpi_os_initialize(void) acpi_status __init acpi_os_initialize1(void) { - acpi_reserve_resources(); kacpid_wq = alloc_workqueue("kacpid", 0, 1); kacpi_notify_wq = alloc_workqueue("kacpi_notify", 0, 1); kacpi_hotplug_wq = alloc_ordered_workqueue("kacpi_hotplug", 0); diff --git a/drivers/acpi/resource.c b/drivers/acpi/resource.c index 10561ce16ed1..8244f013f210 100644 --- a/drivers/acpi/resource.c +++ b/drivers/acpi/resource.c @@ -26,7 +26,6 @@ #include #include #include -#include #include #ifdef CONFIG_X86 @@ -622,164 +621,3 @@ int acpi_dev_filter_resource_type(struct acpi_resource *ares, return (type & types) ? 0 : 1; } EXPORT_SYMBOL_GPL(acpi_dev_filter_resource_type); - -struct reserved_region { - struct list_head node; - u64 start; - u64 end; -}; - -static LIST_HEAD(reserved_io_regions); -static LIST_HEAD(reserved_mem_regions); - -static int request_range(u64 start, u64 end, u8 space_id, unsigned long flags, - char *desc) -{ - unsigned int length = end - start + 1; - struct resource *res; - - res = space_id == ACPI_ADR_SPACE_SYSTEM_IO ? - request_region(start, length, desc) : - request_mem_region(start, length, desc); - if (!res) - return -EIO; - - res->flags &= ~flags; - return 0; -} - -static int add_region_before(u64 start, u64 end, u8 space_id, - unsigned long flags, char *desc, - struct list_head *head) -{ - struct reserved_region *reg; - int error; - - reg = kmalloc(sizeof(*reg), GFP_KERNEL); - if (!reg) - return -ENOMEM; - - error = request_range(start, end, space_id, flags, desc); - if (error) { - kfree(reg); - return error; - } - - reg->start = start; - reg->end = end; - list_add_tail(®->node, head); - return 0; -} - -/** - * acpi_reserve_region - Reserve an I/O or memory region as a system resource. - * @start: Starting address of the region. - * @length: Length of the region. - * @space_id: Identifier of address space to reserve the region from. - * @flags: Resource flags to clear for the region after requesting it. - * @desc: Region description (for messages). - * - * Reserve an I/O or memory region as a system resource to prevent others from - * using it. If the new region overlaps with one of the regions (in the given - * address space) already reserved by this routine, only the non-overlapping - * parts of it will be reserved. - * - * Returned is either 0 (success) or a negative error code indicating a resource - * reservation problem. It is the code of the first encountered error, but the - * routine doesn't abort until it has attempted to request all of the parts of - * the new region that don't overlap with other regions reserved previously. - * - * The resources requested by this routine are never released. - */ -int acpi_reserve_region(u64 start, unsigned int length, u8 space_id, - unsigned long flags, char *desc) -{ - struct list_head *regions; - struct reserved_region *reg; - u64 end = start + length - 1; - int ret = 0, error = 0; - - if (space_id == ACPI_ADR_SPACE_SYSTEM_IO) - regions = &reserved_io_regions; - else if (space_id == ACPI_ADR_SPACE_SYSTEM_MEMORY) - regions = &reserved_mem_regions; - else - return -EINVAL; - - if (list_empty(regions)) - return add_region_before(start, end, space_id, flags, desc, regions); - - list_for_each_entry(reg, regions, node) - if (reg->start == end + 1) { - /* The new region can be prepended to this one. */ - ret = request_range(start, end, space_id, flags, desc); - if (!ret) - reg->start = start; - - return ret; - } else if (reg->start > end) { - /* No overlap. Add the new region here and get out. */ - return add_region_before(start, end, space_id, flags, - desc, ®->node); - } else if (reg->end == start - 1) { - goto combine; - } else if (reg->end >= start) { - goto overlap; - } - - /* The new region goes after the last existing one. */ - return add_region_before(start, end, space_id, flags, desc, regions); - - overlap: - /* - * The new region overlaps an existing one. - * - * The head part of the new region immediately preceding the existing - * overlapping one can be combined with it right away. - */ - if (reg->start > start) { - error = request_range(start, reg->start - 1, space_id, flags, desc); - if (error) - ret = error; - else - reg->start = start; - } - - combine: - /* - * The new region is adjacent to an existing one. If it extends beyond - * that region all the way to the next one, it is possible to combine - * all three of them. - */ - while (reg->end < end) { - struct reserved_region *next = NULL; - u64 a = reg->end + 1, b = end; - - if (!list_is_last(®->node, regions)) { - next = list_next_entry(reg, node); - if (next->start <= end) - b = next->start - 1; - } - error = request_range(a, b, space_id, flags, desc); - if (!error) { - if (next && next->start == b + 1) { - reg->end = next->end; - list_del(&next->node); - kfree(next); - } else { - reg->end = end; - break; - } - } else if (next) { - if (!ret) - ret = error; - - reg = next; - } else { - break; - } - } - - return ret ? ret : error; -} -EXPORT_SYMBOL_GPL(acpi_reserve_region); diff --git a/drivers/pnp/system.c b/drivers/pnp/system.c index 515f33882ab8..49c1720df59a 100644 --- a/drivers/pnp/system.c +++ b/drivers/pnp/system.c @@ -7,7 +7,6 @@ * Bjorn Helgaas */ -#include #include #include #include @@ -23,41 +22,25 @@ static const struct pnp_device_id pnp_dev_table[] = { {"", 0} }; -#ifdef CONFIG_ACPI -static bool __reserve_range(u64 start, unsigned int length, bool io, char *desc) -{ - u8 space_id = io ? ACPI_ADR_SPACE_SYSTEM_IO : ACPI_ADR_SPACE_SYSTEM_MEMORY; - return !acpi_reserve_region(start, length, space_id, IORESOURCE_BUSY, desc); -} -#else -static bool __reserve_range(u64 start, unsigned int length, bool io, char *desc) -{ - struct resource *res; - - res = io ? request_region(start, length, desc) : - request_mem_region(start, length, desc); - if (res) { - res->flags &= ~IORESOURCE_BUSY; - return true; - } - return false; -} -#endif - static void reserve_range(struct pnp_dev *dev, struct resource *r, int port) { char *regionid; const char *pnpid = dev_name(&dev->dev); resource_size_t start = r->start, end = r->end; - bool reserved; + struct resource *res; regionid = kmalloc(16, GFP_KERNEL); if (!regionid) return; snprintf(regionid, 16, "pnp %s", pnpid); - reserved = __reserve_range(start, end - start + 1, !!port, regionid); - if (!reserved) + if (port) + res = request_region(start, end - start + 1, regionid); + else + res = request_mem_region(start, end - start + 1, regionid); + if (res) + res->flags &= ~IORESOURCE_BUSY; + else kfree(regionid); /* @@ -66,7 +49,7 @@ static void reserve_range(struct pnp_dev *dev, struct resource *r, int port) * have double reservations. */ dev_info(&dev->dev, "%pR %s reserved\n", r, - reserved ? "has been" : "could not be"); + res ? "has been" : "could not be"); } static void reserve_resources_of_dev(struct pnp_dev *dev) diff --git a/include/linux/acpi.h b/include/linux/acpi.h index c471dfc93b71..fedfd1bea3bf 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -309,9 +309,6 @@ int acpi_check_region(resource_size_t start, resource_size_t n, int acpi_resources_are_enforced(void); -int acpi_reserve_region(u64 start, unsigned int length, u8 space_id, - unsigned long flags, char *desc); - #ifdef CONFIG_HIBERNATION void __init acpi_no_s4_hw_signature(void); #endif @@ -507,13 +504,6 @@ static inline int acpi_check_region(resource_size_t start, resource_size_t n, return 0; } -static inline int acpi_reserve_region(u64 start, unsigned int length, - u8 space_id, unsigned long flags, - char *desc) -{ - return -ENXIO; -} - struct acpi_table_header; static inline int acpi_table_parse(char *id, int (*handler)(struct acpi_table_header *)) -- cgit v1.2.3 From 26095a01d359827eeccec5459c28ddd976183179 Mon Sep 17 00:00:00 2001 From: "Suthikulpanit, Suravee" Date: Tue, 7 Jul 2015 01:55:20 +0200 Subject: ACPI / scan: Add support for ACPI _CLS device matching Device drivers typically use ACPI _HIDs/_CIDs listed in struct device_driver acpi_match_table to match devices. However, for generic drivers, we do not want to list _HID for all supported devices. Also, certain classes of devices do not have _CID (e.g. SATA, USB). Instead, we can leverage ACPI _CLS, which specifies PCI-defined class code (i.e. base-class, subclass and programming interface). This patch adds support for matching ACPI devices using the _CLS method. To support loadable module, current design uses _HID or _CID to match device's modalias. With the new way of matching with _CLS this would requires modification to the current ACPI modalias key to include _CLS. This patch appends PCI-defined class-code to the existing ACPI modalias as following. acpi::::..::: E.g: # cat /sys/devices/platform/AMDI0600:00/modalias acpi:AMDI0600:010601: where bb is th base-class code, ss is te sub-class code, and pp is the programming interface code Since there would not be _HID/_CID in the ACPI matching table of the driver, this patch adds a field to acpi_device_id to specify the matching _CLS. static const struct acpi_device_id ahci_acpi_match[] = { { ACPI_DEVICE_CLASS(PCI_CLASS_STORAGE_SATA_AHCI, 0xffffff) }, {}, }; In this case, the corresponded entry in modules.alias file would be: alias acpi*:010601:* ahci_platform Acked-by: Mika Westerberg Reviewed-by: Hanjun Guo Signed-off-by: Suravee Suthikulpanit Signed-off-by: Rafael J. Wysocki --- drivers/acpi/scan.c | 32 ++++++++++++++++++++++++++++++-- include/linux/acpi.h | 14 ++++++++++++++ include/linux/mod_devicetable.h | 2 ++ scripts/mod/devicetable-offsets.c | 2 ++ scripts/mod/file2alias.c | 32 ++++++++++++++++++++++++++++++-- 5 files changed, 78 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c index 2649a068671d..ec256352f423 100644 --- a/drivers/acpi/scan.c +++ b/drivers/acpi/scan.c @@ -1019,6 +1019,29 @@ static bool acpi_of_match_device(struct acpi_device *adev, return false; } +static bool __acpi_match_device_cls(const struct acpi_device_id *id, + struct acpi_hardware_id *hwid) +{ + int i, msk, byte_shift; + char buf[3]; + + if (!id->cls) + return false; + + /* Apply class-code bitmask, before checking each class-code byte */ + for (i = 1; i <= 3; i++) { + byte_shift = 8 * (3 - i); + msk = (id->cls_msk >> byte_shift) & 0xFF; + if (!msk) + continue; + + sprintf(buf, "%02x", (id->cls >> byte_shift) & msk); + if (strncmp(buf, &hwid->id[(i - 1) * 2], 2)) + return false; + } + return true; +} + static const struct acpi_device_id *__acpi_match_device( struct acpi_device *device, const struct acpi_device_id *ids, @@ -1036,9 +1059,12 @@ static const struct acpi_device_id *__acpi_match_device( list_for_each_entry(hwid, &device->pnp.ids, list) { /* First, check the ACPI/PNP IDs provided by the caller. */ - for (id = ids; id->id[0]; id++) - if (!strcmp((char *) id->id, hwid->id)) + for (id = ids; id->id[0] || id->cls; id++) { + if (id->id[0] && !strcmp((char *) id->id, hwid->id)) return id; + else if (id->cls && __acpi_match_device_cls(id, hwid)) + return id; + } /* * Next, check ACPI_DT_NAMESPACE_HID and try to match the @@ -2101,6 +2127,8 @@ static void acpi_set_pnp_ids(acpi_handle handle, struct acpi_device_pnp *pnp, if (info->valid & ACPI_VALID_UID) pnp->unique_id = kstrdup(info->unique_id.string, GFP_KERNEL); + if (info->valid & ACPI_VALID_CLS) + acpi_add_id(pnp, info->class_code.string); kfree(info); diff --git a/include/linux/acpi.h b/include/linux/acpi.h index c471dfc93b71..bdfdb9f65c23 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -58,6 +58,19 @@ static inline acpi_handle acpi_device_handle(struct acpi_device *adev) acpi_fwnode_handle(adev) : NULL) #define ACPI_HANDLE(dev) acpi_device_handle(ACPI_COMPANION(dev)) +/** + * ACPI_DEVICE_CLASS - macro used to describe an ACPI device with + * the PCI-defined class-code information + * + * @_cls : the class, subclass, prog-if triple for this device + * @_msk : the class mask for this device + * + * This macro is used to create a struct acpi_device_id that matches a + * specific PCI class. The .id and .driver_data fields will be left + * initialized with the default value. + */ +#define ACPI_DEVICE_CLASS(_cls, _msk) .cls = (_cls), .cls_msk = (_msk), + static inline bool has_acpi_companion(struct device *dev) { return is_acpi_node(dev->fwnode); @@ -446,6 +459,7 @@ struct platform_device *acpi_create_platform_device(struct acpi_device *); #define ACPI_COMPANION(dev) (NULL) #define ACPI_COMPANION_SET(dev, adev) do { } while (0) #define ACPI_HANDLE(dev) (NULL) +#define ACPI_DEVICE_CLASS(_cls, _msk) .cls = (0), .cls_msk = (0), struct fwnode_handle; diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h index 8183d6640ca7..34f25b7bf642 100644 --- a/include/linux/mod_devicetable.h +++ b/include/linux/mod_devicetable.h @@ -189,6 +189,8 @@ struct css_device_id { struct acpi_device_id { __u8 id[ACPI_ID_LEN]; kernel_ulong_t driver_data; + __u32 cls; + __u32 cls_msk; }; #define PNP_ID_LEN 8 diff --git a/scripts/mod/devicetable-offsets.c b/scripts/mod/devicetable-offsets.c index eff7de1fc82e..e70fcd12eeeb 100644 --- a/scripts/mod/devicetable-offsets.c +++ b/scripts/mod/devicetable-offsets.c @@ -63,6 +63,8 @@ int main(void) DEVID(acpi_device_id); DEVID_FIELD(acpi_device_id, id); + DEVID_FIELD(acpi_device_id, cls); + DEVID_FIELD(acpi_device_id, cls_msk); DEVID(pnp_device_id); DEVID_FIELD(pnp_device_id, id); diff --git a/scripts/mod/file2alias.c b/scripts/mod/file2alias.c index 84c86f3cd6cd..5f2088209132 100644 --- a/scripts/mod/file2alias.c +++ b/scripts/mod/file2alias.c @@ -523,12 +523,40 @@ static int do_serio_entry(const char *filename, } ADD_TO_DEVTABLE("serio", serio_device_id, do_serio_entry); -/* looks like: "acpi:ACPI0003 or acpi:PNP0C0B" or "acpi:LNXVIDEO" */ +/* looks like: "acpi:ACPI0003" or "acpi:PNP0C0B" or "acpi:LNXVIDEO" or + * "acpi:bbsspp" (bb=base-class, ss=sub-class, pp=prog-if) + * + * NOTE: Each driver should use one of the following : _HID, _CIDs + * or _CLS. Also, bb, ss, and pp can be substituted with ?? + * as don't care byte. + */ static int do_acpi_entry(const char *filename, void *symval, char *alias) { DEF_FIELD_ADDR(symval, acpi_device_id, id); - sprintf(alias, "acpi*:%s:*", *id); + DEF_FIELD_ADDR(symval, acpi_device_id, cls); + DEF_FIELD_ADDR(symval, acpi_device_id, cls_msk); + + if (id && strlen((const char *)*id)) + sprintf(alias, "acpi*:%s:*", *id); + else if (cls) { + int i, byte_shift, cnt = 0; + unsigned int msk; + + sprintf(&alias[cnt], "acpi*:"); + cnt = 6; + for (i = 1; i <= 3; i++) { + byte_shift = 8 * (3-i); + msk = (*cls_msk >> byte_shift) & 0xFF; + if (msk) + sprintf(&alias[cnt], "%02x", + (*cls >> byte_shift) & 0xFF); + else + sprintf(&alias[cnt], "??"); + cnt += 2; + } + sprintf(&alias[cnt], ":*"); + } return 1; } ADD_TO_DEVTABLE("acpi", acpi_device_id, do_acpi_entry); -- cgit v1.2.3 From f32dd117051185da6e923b35491a44d7debeeea5 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 7 Jul 2015 16:29:38 +0200 Subject: tick/broadcast: Make idle check independent from mode and config Currently the broadcast busy check, which prevents the idle code from going into deep idle, works only in one shot mode. If NOHZ and HIGHRES are off (config or command line) there is no sanity check at all, so under certain conditions cpus are allowed to go into deep idle, where the local timer stops, and are not woken up again because there is no broadcast timer installed or a hrtimer based broadcast device is not evaluated. Move tick_broadcast_oneshot_control() into the common code and provide proper subfunctions for the various config combinations. The common check in tick_broadcast_oneshot_control() is for the C3STOP misfeature flag of the local clock event device. If its not set, idle can proceed. If set, further checks are necessary. Provide checks for the trivial cases: - If broadcast is disabled in the config, then return busy - If oneshot mode (NOHZ/HIGHES) is disabled in the config, return busy if the broadcast device is hrtimer based. - If oneshot mode is enabled in the config call the original tick_broadcast_oneshot_control() function. That function needs extra checks which will be implemented in seperate patches. [ Split out from a larger combo patch ] Reported-and-tested-by: Sudeep Holla Signed-off-by: Thomas Gleixner Cc: Suzuki Poulose Cc: Lorenzo Pieralisi Cc: Catalin Marinas Cc: Rafael J. Wysocki Cc: Peter Zijlstra Cc: Preeti U Murthy Cc: Ingo Molnar Link: http://lkml.kernel.org/r/alpine.DEB.2.11.1507070929360.3916@nanos --- include/linux/tick.h | 4 ---- kernel/time/tick-broadcast.c | 26 +++++++++++--------------- kernel/time/tick-common.c | 21 +++++++++++++++++++++ kernel/time/tick-sched.h | 10 ++++++++++ 4 files changed, 42 insertions(+), 19 deletions(-) (limited to 'include/linux') diff --git a/include/linux/tick.h b/include/linux/tick.h index 3741ba1a652c..6916dcb61857 100644 --- a/include/linux/tick.h +++ b/include/linux/tick.h @@ -67,11 +67,7 @@ extern void tick_broadcast_control(enum tick_broadcast_mode mode); static inline void tick_broadcast_control(enum tick_broadcast_mode mode) { } #endif /* BROADCAST */ -#if defined(CONFIG_GENERIC_CLOCKEVENTS_BROADCAST) && defined(CONFIG_TICK_ONESHOT) extern int tick_broadcast_oneshot_control(enum tick_broadcast_state state); -#else -static inline int tick_broadcast_oneshot_control(enum tick_broadcast_state state) { return 0; } -#endif static inline void tick_broadcast_enable(void) { diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c index 9877d0b0aefc..ef77b16ad5df 100644 --- a/kernel/time/tick-broadcast.c +++ b/kernel/time/tick-broadcast.c @@ -685,18 +685,7 @@ static void broadcast_shutdown_local(struct clock_event_device *bc, clockevents_switch_state(dev, CLOCK_EVT_STATE_SHUTDOWN); } -/** - * tick_broadcast_oneshot_control - Enter/exit broadcast oneshot mode - * @state: The target state (enter/exit) - * - * The system enters/leaves a state, where affected devices might stop - * Returns 0 on success, -EBUSY if the cpu is used to broadcast wakeups. - * - * Called with interrupts disabled, so clockevents_lock is not - * required here because the local clock event device cannot go away - * under us. - */ -int tick_broadcast_oneshot_control(enum tick_broadcast_state state) +int __tick_broadcast_oneshot_control(enum tick_broadcast_state state) { struct clock_event_device *bc, *dev; struct tick_device *td; @@ -717,9 +706,6 @@ int tick_broadcast_oneshot_control(enum tick_broadcast_state state) td = this_cpu_ptr(&tick_cpu_device); dev = td->evtdev; - if (!(dev->features & CLOCK_EVT_FEAT_C3STOP)) - return 0; - raw_spin_lock(&tick_broadcast_lock); bc = tick_broadcast_device.evtdev; cpu = smp_processor_id(); @@ -961,6 +947,16 @@ bool tick_broadcast_oneshot_available(void) return bc ? bc->features & CLOCK_EVT_FEAT_ONESHOT : false; } +#else +int __tick_broadcast_oneshot_control(enum tick_broadcast_state state) +{ + struct clock_event_device *bc = tick_broadcast_device.evtdev; + + if (!bc || (bc->features & CLOCK_EVT_FEAT_HRTIMER)) + return -EBUSY; + + return 0; +} #endif void __init tick_broadcast_init(void) diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c index 76446cb5dfe1..55e13efff1ab 100644 --- a/kernel/time/tick-common.c +++ b/kernel/time/tick-common.c @@ -343,6 +343,27 @@ out_bc: tick_install_broadcast_device(newdev); } +/** + * tick_broadcast_oneshot_control - Enter/exit broadcast oneshot mode + * @state: The target state (enter/exit) + * + * The system enters/leaves a state, where affected devices might stop + * Returns 0 on success, -EBUSY if the cpu is used to broadcast wakeups. + * + * Called with interrupts disabled, so clockevents_lock is not + * required here because the local clock event device cannot go away + * under us. + */ +int tick_broadcast_oneshot_control(enum tick_broadcast_state state) +{ + struct tick_device *td = this_cpu_ptr(&tick_cpu_device); + + if (!(td->evtdev->features & CLOCK_EVT_FEAT_C3STOP)) + return 0; + + return __tick_broadcast_oneshot_control(state); +} + #ifdef CONFIG_HOTPLUG_CPU /* * Transfer the do_timer job away from a dying cpu. diff --git a/kernel/time/tick-sched.h b/kernel/time/tick-sched.h index 42fdf4958bcc..a4a8d4e9baa1 100644 --- a/kernel/time/tick-sched.h +++ b/kernel/time/tick-sched.h @@ -71,4 +71,14 @@ extern void tick_cancel_sched_timer(int cpu); static inline void tick_cancel_sched_timer(int cpu) { } #endif +#ifdef CONFIG_GENERIC_CLOCKEVENTS_BROADCAST +extern int __tick_broadcast_oneshot_control(enum tick_broadcast_state state); +#else +static inline int +__tick_broadcast_oneshot_control(enum tick_broadcast_state state) +{ + return -EBUSY; +} +#endif + #endif -- cgit v1.2.3 From 37b64a42067a04a22468c4e52c12af00d72e462b Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 7 Jul 2015 21:56:34 +0200 Subject: tick/broadcast: Unbreak CONFIG_GENERIC_CLOCKEVENTS=n build Making tick_broadcast_oneshot_control() independent from CONFIG_GENERIC_CLOCKEVENTS_BROADCAST broke the build for CONFIG_GENERIC_CLOCKEVENTS=n because the function is not defined there. Provide a proper stub inline. Fixes: f32dd1170511 'tick/broadcast: Make idle check independent from mode and config' Reported-by: kbuild test robot Signed-off-by: Thomas Gleixner --- include/linux/tick.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'include/linux') diff --git a/include/linux/tick.h b/include/linux/tick.h index 6916dcb61857..edbfc9a5293e 100644 --- a/include/linux/tick.h +++ b/include/linux/tick.h @@ -67,7 +67,14 @@ extern void tick_broadcast_control(enum tick_broadcast_mode mode); static inline void tick_broadcast_control(enum tick_broadcast_mode mode) { } #endif /* BROADCAST */ +#ifdef CONFIG_GENERIC_CLOCKEVENTS extern int tick_broadcast_oneshot_control(enum tick_broadcast_state state); +#else +static inline int tick_broadcast_oneshot_control(enum tick_broadcast_state state) +{ + return 0; +} +#endif static inline void tick_broadcast_enable(void) { -- cgit v1.2.3 From a899418167264c7bac574b1a0f1b2c26c5b0995a Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sun, 5 Jul 2015 17:12:30 +0000 Subject: hotplug: Prevent alloc/free of irq descriptors during cpu up/down When a cpu goes up some architectures (e.g. x86) have to walk the irq space to set up the vector space for the cpu. While this needs extra protection at the architecture level we can avoid a few race conditions by preventing the concurrent allocation/free of irq descriptors and the associated data. When a cpu goes down it moves the interrupts which are targeted to this cpu away by reassigning the affinities. While this happens interrupts can be allocated and freed, which opens a can of race conditions in the code which reassignes the affinities because interrupt descriptors might be freed underneath. Example: CPU1 CPU2 cpu_up/down irq_desc = irq_to_desc(irq); remove_from_radix_tree(desc); raw_spin_lock(&desc->lock); free(desc); We could protect the irq descriptors with RCU, but that would require a full tree change of all accesses to interrupt descriptors. But fortunately these kind of race conditions are rather limited to a few things like cpu hotplug. The normal setup/teardown is very well serialized. So the simpler and obvious solution is: Prevent allocation and freeing of interrupt descriptors accross cpu hotplug. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Cc: xiao jin Cc: Joerg Roedel Cc: Borislav Petkov Cc: Yanmin Zhang Link: http://lkml.kernel.org/r/20150705171102.063519515@linutronix.de --- include/linux/irqdesc.h | 7 ++++++- kernel/cpu.c | 22 +++++++++++++++++++++- kernel/irq/internals.h | 4 ---- 3 files changed, 27 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/irqdesc.h b/include/linux/irqdesc.h index 624a668e61f1..fcea4e48e21f 100644 --- a/include/linux/irqdesc.h +++ b/include/linux/irqdesc.h @@ -87,7 +87,12 @@ struct irq_desc { const char *name; } ____cacheline_internodealigned_in_smp; -#ifndef CONFIG_SPARSE_IRQ +#ifdef CONFIG_SPARSE_IRQ +extern void irq_lock_sparse(void); +extern void irq_unlock_sparse(void); +#else +static inline void irq_lock_sparse(void) { } +static inline void irq_unlock_sparse(void) { } extern struct irq_desc irq_desc[NR_IRQS]; #endif diff --git a/kernel/cpu.c b/kernel/cpu.c index 9c9c9fab16cc..6a374544d495 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -21,6 +21,7 @@ #include #include #include +#include #include #include "smpboot.h" @@ -392,13 +393,19 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen) smpboot_park_threads(cpu); /* - * So now all preempt/rcu users must observe !cpu_active(). + * Prevent irq alloc/free while the dying cpu reorganizes the + * interrupt affinities. */ + irq_lock_sparse(); + /* + * So now all preempt/rcu users must observe !cpu_active(). + */ err = __stop_machine(take_cpu_down, &tcd_param, cpumask_of(cpu)); if (err) { /* CPU didn't die: tell everyone. Can't complain. */ cpu_notify_nofail(CPU_DOWN_FAILED | mod, hcpu); + irq_unlock_sparse(); goto out_release; } BUG_ON(cpu_online(cpu)); @@ -415,6 +422,9 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen) smp_mb(); /* Read from cpu_dead_idle before __cpu_die(). */ per_cpu(cpu_dead_idle, cpu) = false; + /* Interrupts are moved away from the dying cpu, reenable alloc/free */ + irq_unlock_sparse(); + hotplug_cpu__broadcast_tick_pull(cpu); /* This actually kills the CPU. */ __cpu_die(cpu); @@ -517,8 +527,18 @@ static int _cpu_up(unsigned int cpu, int tasks_frozen) goto out_notify; } + /* + * Some architectures have to walk the irq descriptors to + * setup the vector space for the cpu which comes online. + * Prevent irq alloc/free across the bringup. + */ + irq_lock_sparse(); + /* Arch-specific enabling code. */ ret = __cpu_up(cpu, idle); + + irq_unlock_sparse(); + if (ret != 0) goto out_notify; BUG_ON(!cpu_online(cpu)); diff --git a/kernel/irq/internals.h b/kernel/irq/internals.h index 4834ee828c41..61008b8433ab 100644 --- a/kernel/irq/internals.h +++ b/kernel/irq/internals.h @@ -76,12 +76,8 @@ extern void unmask_threaded_irq(struct irq_desc *desc); #ifdef CONFIG_SPARSE_IRQ static inline void irq_mark_irq(unsigned int irq) { } -extern void irq_lock_sparse(void); -extern void irq_unlock_sparse(void); #else extern void irq_mark_irq(unsigned int irq); -static inline void irq_lock_sparse(void) { } -static inline void irq_unlock_sparse(void) { } #endif extern void init_kstat_irqs(struct irq_desc *desc, int node, int nr); -- cgit v1.2.3 From 1f6823faa8c563431a94e614d2b63ce16bb6f658 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 9 Jul 2015 10:51:46 +0200 Subject: time: Get rid of do_posix_clock_monotonic_gettime All users gone. Remove it before we get another one. Signed-off-by: Thomas Gleixner --- include/linux/timekeeping.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/timekeeping.h b/include/linux/timekeeping.h index 3aa72e648650..6e191e4e6ab6 100644 --- a/include/linux/timekeeping.h +++ b/include/linux/timekeeping.h @@ -145,7 +145,6 @@ static inline void getboottime(struct timespec *ts) } #endif -#define do_posix_clock_monotonic_gettime(ts) ktime_get_ts(ts) #define ktime_get_real_ts64(ts) getnstimeofday64(ts) /* -- cgit v1.2.3 From 757856d2b9568a701df9ea6a4be68effbb9d6f44 Mon Sep 17 00:00:00 2001 From: Ilya Dryomov Date: Thu, 25 Jun 2015 17:47:45 +0300 Subject: libceph: enable ceph in a non-default network namespace Grab a reference on a network namespace of the 'rbd map' (in case of rbd) or 'mount' (in case of ceph) process and use that to open sockets instead of always using init_net and bailing if network namespace is anything but init_net. Be careful to not share struct ceph_client instances between different namespaces and don't add any code in the !CONFIG_NET_NS case. This is based on a patch from Hong Zhiguo . Signed-off-by: Ilya Dryomov Reviewed-by: Sage Weil --- include/linux/ceph/messenger.h | 3 +++ net/ceph/ceph_common.c | 16 ++++++++++------ net/ceph/messenger.c | 10 +++++++++- 3 files changed, 22 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/ceph/messenger.h b/include/linux/ceph/messenger.h index e15499422fdc..37753278987a 100644 --- a/include/linux/ceph/messenger.h +++ b/include/linux/ceph/messenger.h @@ -8,6 +8,7 @@ #include #include #include +#include #include #include @@ -56,6 +57,7 @@ struct ceph_messenger { struct ceph_entity_addr my_enc_addr; atomic_t stopping; + possible_net_t net; bool nocrc; bool tcp_nodelay; @@ -267,6 +269,7 @@ extern void ceph_messenger_init(struct ceph_messenger *msgr, u64 required_features, bool nocrc, bool tcp_nodelay); +extern void ceph_messenger_fini(struct ceph_messenger *msgr); extern void ceph_con_init(struct ceph_connection *con, void *private, const struct ceph_connection_operations *ops, diff --git a/net/ceph/ceph_common.c b/net/ceph/ceph_common.c index cb7db320dd27..f30329f72641 100644 --- a/net/ceph/ceph_common.c +++ b/net/ceph/ceph_common.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include #include @@ -16,8 +17,6 @@ #include #include #include -#include -#include #include @@ -131,6 +130,13 @@ int ceph_compare_options(struct ceph_options *new_opt, int i; int ret; + /* + * Don't bother comparing options if network namespaces don't + * match. + */ + if (!net_eq(current->nsproxy->net_ns, read_pnet(&client->msgr.net))) + return -1; + ret = memcmp(opt1, opt2, ofs); if (ret) return ret; @@ -335,9 +341,6 @@ ceph_parse_options(char *options, const char *dev_name, int err = -ENOMEM; substring_t argstr[MAX_OPT_ARGS]; - if (current->nsproxy->net_ns != &init_net) - return ERR_PTR(-EINVAL); - opt = kzalloc(sizeof(*opt), GFP_KERNEL); if (!opt) return ERR_PTR(-ENOMEM); @@ -608,6 +611,7 @@ struct ceph_client *ceph_create_client(struct ceph_options *opt, void *private, fail_monc: ceph_monc_stop(&client->monc); fail: + ceph_messenger_fini(&client->msgr); kfree(client); return ERR_PTR(err); } @@ -621,8 +625,8 @@ void ceph_destroy_client(struct ceph_client *client) /* unmount */ ceph_osdc_stop(&client->osdc); - ceph_monc_stop(&client->monc); + ceph_messenger_fini(&client->msgr); ceph_debugfs_client_cleanup(client); diff --git a/net/ceph/messenger.c b/net/ceph/messenger.c index 1679f47280e2..5c1f98ea6741 100644 --- a/net/ceph/messenger.c +++ b/net/ceph/messenger.c @@ -6,6 +6,7 @@ #include #include #include +#include #include #include #include @@ -479,7 +480,7 @@ static int ceph_tcp_connect(struct ceph_connection *con) int ret; BUG_ON(con->sock); - ret = sock_create_kern(&init_net, con->peer_addr.in_addr.ss_family, + ret = sock_create_kern(read_pnet(&con->msgr->net), paddr->ss_family, SOCK_STREAM, IPPROTO_TCP, &sock); if (ret) return ret; @@ -2944,11 +2945,18 @@ void ceph_messenger_init(struct ceph_messenger *msgr, msgr->tcp_nodelay = tcp_nodelay; atomic_set(&msgr->stopping, 0); + write_pnet(&msgr->net, get_net(current->nsproxy->net_ns)); dout("%s %p\n", __func__, msgr); } EXPORT_SYMBOL(ceph_messenger_init); +void ceph_messenger_fini(struct ceph_messenger *msgr) +{ + put_net(read_pnet(&msgr->net)); +} +EXPORT_SYMBOL(ceph_messenger_fini); + static void clear_standby(struct ceph_connection *con) { /* come back from STANDBY? */ -- cgit v1.2.3