From cf14c2e987ba0a09a7b09be2ecd55af0bc9c17b4 Mon Sep 17 00:00:00 2001 From: Philipp Reisner Date: Tue, 2 Feb 2010 21:03:50 +0100 Subject: drbd: --dry-run option for drbdsetup net ( drbdadm -- --dry-run connect ) Signed-off-by: Philipp Reisner Signed-off-by: Lars Ellenberg --- include/linux/drbd.h | 2 +- include/linux/drbd_nl.h | 1 + 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/drbd.h b/include/linux/drbd.h index 78962272338a..4341b1a97a34 100644 --- a/include/linux/drbd.h +++ b/include/linux/drbd.h @@ -56,7 +56,7 @@ extern const char *drbd_buildtag(void); #define REL_VERSION "8.3.7" #define API_VERSION 88 #define PRO_VERSION_MIN 86 -#define PRO_VERSION_MAX 91 +#define PRO_VERSION_MAX 92 enum drbd_io_error_p { diff --git a/include/linux/drbd_nl.h b/include/linux/drbd_nl.h index a4d82f895994..b41050e8cc2f 100644 --- a/include/linux/drbd_nl.h +++ b/include/linux/drbd_nl.h @@ -63,6 +63,7 @@ NL_PACKET(net_conf, 5, NL_BIT( 41, T_MAY_IGNORE, always_asbp) NL_BIT( 61, T_MAY_IGNORE, no_cork) NL_BIT( 62, T_MANDATORY, auto_sndbuf_size) + NL_BIT( 70, T_MANDATORY, dry_run) ) NL_PACKET(disconnect, 6, ) -- cgit v1.2.3 From 1f55243024087b56aef0b1e6d9c0ea89c76f0a6b Mon Sep 17 00:00:00 2001 From: Philipp Reisner Date: Thu, 4 Mar 2010 15:51:01 +0100 Subject: drbd: Renamed overwrite_peer to primary_force Signed-off-by: Philipp Reisner Signed-off-by: Lars Ellenberg --- drivers/block/drbd/drbd_nl.c | 2 +- include/linux/drbd_nl.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/block/drbd/drbd_nl.c b/drivers/block/drbd/drbd_nl.c index 6492e321ec00..6429d2b19e06 100644 --- a/drivers/block/drbd/drbd_nl.c +++ b/drivers/block/drbd/drbd_nl.c @@ -407,7 +407,7 @@ static int drbd_nl_primary(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp, } reply->ret_code = - drbd_set_role(mdev, R_PRIMARY, primary_args.overwrite_peer); + drbd_set_role(mdev, R_PRIMARY, primary_args.primary_force); return 0; } diff --git a/include/linux/drbd_nl.h b/include/linux/drbd_nl.h index b41050e8cc2f..f7431a4ca608 100644 --- a/include/linux/drbd_nl.h +++ b/include/linux/drbd_nl.h @@ -12,7 +12,7 @@ #endif NL_PACKET(primary, 1, - NL_BIT( 1, T_MAY_IGNORE, overwrite_peer) + NL_BIT( 1, T_MAY_IGNORE, primary_force) ) NL_PACKET(secondary, 2, ) -- cgit v1.2.3 From f11c9c5c259cb2c3d698548dc3936f773ab1f5b9 Mon Sep 17 00:00:00 2001 From: Edward Shishkin Date: Thu, 11 Mar 2010 14:09:47 -0800 Subject: vfs: improve writeback_inodes_wb() Do not pin/unpin superblock for every inode in writeback_inodes_wb(), pin it for the whole group of inodes which belong to the same superblock and call writeback_sb_inodes() handler for them. Signed-off-by: Edward Shishkin Cc: Jens Axboe Cc: Wu Fengguang Signed-off-by: Andrew Morton Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 133 +++++++++++++++++++++++++--------------------- include/linux/writeback.h | 3 ++ 2 files changed, 76 insertions(+), 60 deletions(-) (limited to 'include/linux') diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 76fc4d594acb..6841effa47ca 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -553,108 +553,85 @@ select_queue: return ret; } -static void unpin_sb_for_writeback(struct super_block **psb) +static void unpin_sb_for_writeback(struct super_block *sb) { - struct super_block *sb = *psb; - - if (sb) { - up_read(&sb->s_umount); - put_super(sb); - *psb = NULL; - } + up_read(&sb->s_umount); + put_super(sb); } +enum sb_pin_state { + SB_PINNED, + SB_NOT_PINNED, + SB_PIN_FAILED +}; + /* * For WB_SYNC_NONE writeback, the caller does not have the sb pinned * before calling writeback. So make sure that we do pin it, so it doesn't * go away while we are writing inodes from it. - * - * Returns 0 if the super was successfully pinned (or pinning wasn't needed), - * 1 if we failed. */ -static int pin_sb_for_writeback(struct writeback_control *wbc, - struct inode *inode, struct super_block **psb) +static enum sb_pin_state pin_sb_for_writeback(struct writeback_control *wbc, + struct super_block *sb) { - struct super_block *sb = inode->i_sb; - - /* - * If this sb is already pinned, nothing more to do. If not and - * *psb is non-NULL, unpin the old one first - */ - if (sb == *psb) - return 0; - else if (*psb) - unpin_sb_for_writeback(psb); - /* * Caller must already hold the ref for this */ if (wbc->sync_mode == WB_SYNC_ALL) { WARN_ON(!rwsem_is_locked(&sb->s_umount)); - return 0; + return SB_NOT_PINNED; } - spin_lock(&sb_lock); sb->s_count++; if (down_read_trylock(&sb->s_umount)) { if (sb->s_root) { spin_unlock(&sb_lock); - goto pinned; + return SB_PINNED; } /* * umounted, drop rwsem again and fall through to failure */ up_read(&sb->s_umount); } - sb->s_count--; spin_unlock(&sb_lock); - return 1; -pinned: - *psb = sb; - return 0; + return SB_PIN_FAILED; } -static void writeback_inodes_wb(struct bdi_writeback *wb, - struct writeback_control *wbc) +/* + * Write a portion of b_io inodes which belong to @sb. + * If @wbc->sb != NULL, then find and write all such + * inodes. Otherwise write only ones which go sequentially + * in reverse order. + * Return 1, if the caller writeback routine should be + * interrupted. Otherwise return 0. + */ +static int writeback_sb_inodes(struct super_block *sb, + struct bdi_writeback *wb, + struct writeback_control *wbc) { - struct super_block *sb = wbc->sb, *pin_sb = NULL; - const unsigned long start = jiffies; /* livelock avoidance */ - - spin_lock(&inode_lock); - - if (!wbc->for_kupdate || list_empty(&wb->b_io)) - queue_io(wb, wbc->older_than_this); - while (!list_empty(&wb->b_io)) { - struct inode *inode = list_entry(wb->b_io.prev, - struct inode, i_list); long pages_skipped; - - /* - * super block given and doesn't match, skip this inode - */ - if (sb && sb != inode->i_sb) { + struct inode *inode = list_entry(wb->b_io.prev, + struct inode, i_list); + if (wbc->sb && sb != inode->i_sb) { + /* super block given and doesn't + match, skip this inode */ redirty_tail(inode); continue; } - + if (sb != inode->i_sb) + /* finish with this superblock */ + return 0; if (inode->i_state & (I_NEW | I_WILL_FREE)) { requeue_io(inode); continue; } - /* * Was this inode dirtied after sync_sb_inodes was called? * This keeps sync from extra jobs and livelock. */ - if (inode_dirtied_after(inode, start)) - break; - - if (pin_sb_for_writeback(wbc, inode, &pin_sb)) { - requeue_io(inode); - continue; - } + if (inode_dirtied_after(inode, wbc->wb_start)) + return 1; BUG_ON(inode->i_state & (I_FREEING | I_CLEAR)); __iget(inode); @@ -673,14 +650,50 @@ static void writeback_inodes_wb(struct bdi_writeback *wb, spin_lock(&inode_lock); if (wbc->nr_to_write <= 0) { wbc->more_io = 1; - break; + return 1; } if (!list_empty(&wb->b_more_io)) wbc->more_io = 1; } + /* b_io is empty */ + return 1; +} + +static void writeback_inodes_wb(struct bdi_writeback *wb, + struct writeback_control *wbc) +{ + int ret = 0; - unpin_sb_for_writeback(&pin_sb); + wbc->wb_start = jiffies; /* livelock avoidance */ + spin_lock(&inode_lock); + if (!wbc->for_kupdate || list_empty(&wb->b_io)) + queue_io(wb, wbc->older_than_this); + + while (!list_empty(&wb->b_io)) { + struct inode *inode = list_entry(wb->b_io.prev, + struct inode, i_list); + struct super_block *sb = inode->i_sb; + enum sb_pin_state state; + + if (wbc->sb && sb != wbc->sb) { + /* super block given and doesn't + match, skip this inode */ + redirty_tail(inode); + continue; + } + state = pin_sb_for_writeback(wbc, sb); + + if (state == SB_PIN_FAILED) { + requeue_io(inode); + continue; + } + ret = writeback_sb_inodes(sb, wb, wbc); + if (state == SB_PINNED) + unpin_sb_for_writeback(sb); + if (ret) + break; + } spin_unlock(&inode_lock); /* Leave any unwritten inodes on b_io */ } diff --git a/include/linux/writeback.h b/include/linux/writeback.h index 76e8903cd204..36520ded3e06 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -34,6 +34,9 @@ struct writeback_control { enum writeback_sync_modes sync_mode; unsigned long *older_than_this; /* If !NULL, only write back inodes older than this */ + unsigned long wb_start; /* Time writeback_inodes_wb was + called. This is needed to avoid + extra jobs and livelock */ long nr_to_write; /* Write this many pages, and decrement this for each page written */ long pages_skipped; /* Pages which were not written */ -- cgit v1.2.3 From 2cda2728aa1c8c006418a24f867b25e5eb7a32e2 Mon Sep 17 00:00:00 2001 From: "Martin K. Petersen" Date: Mon, 15 Mar 2010 12:46:51 +0100 Subject: block: Fix overrun in lcm() and move it to lib lcm() was defined to take integer-sized arguments. The supplied arguments are multiplied, however, causing us to overflow given sufficiently large input. That in turn led to incorrect optimal I/O size reporting in some cases (RAID over RAID). Switch lcm() over to unsigned long similar to gcd() and move the function from blk-settings.c to lib. Signed-off-by: Martin K. Petersen Signed-off-by: Jens Axboe --- block/blk-settings.c | 11 +---------- include/linux/lcm.h | 8 ++++++++ lib/Makefile | 2 +- lib/lcm.c | 15 +++++++++++++++ 4 files changed, 25 insertions(+), 11 deletions(-) create mode 100644 include/linux/lcm.h create mode 100644 lib/lcm.c (limited to 'include/linux') diff --git a/block/blk-settings.c b/block/blk-settings.c index 31e7a9375c13..4c4700dca56a 100644 --- a/block/blk-settings.c +++ b/block/blk-settings.c @@ -8,6 +8,7 @@ #include #include /* for max_pfn/max_low_pfn */ #include +#include #include #include "blk.h" @@ -461,16 +462,6 @@ void blk_queue_stack_limits(struct request_queue *t, struct request_queue *b) } EXPORT_SYMBOL(blk_queue_stack_limits); -static unsigned int lcm(unsigned int a, unsigned int b) -{ - if (a && b) - return (a * b) / gcd(a, b); - else if (b) - return b; - - return a; -} - /** * blk_stack_limits - adjust queue_limits for stacked devices * @t: the stacking driver limits (top device) diff --git a/include/linux/lcm.h b/include/linux/lcm.h new file mode 100644 index 000000000000..7bf01d779b45 --- /dev/null +++ b/include/linux/lcm.h @@ -0,0 +1,8 @@ +#ifndef _LCM_H +#define _LCM_H + +#include + +unsigned long lcm(unsigned long a, unsigned long b) __attribute_const__; + +#endif /* _LCM_H */ diff --git a/lib/Makefile b/lib/Makefile index 2e152aed7198..0d4015205c64 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -21,7 +21,7 @@ lib-y += kobject.o kref.o klist.o obj-y += bcd.o div64.o sort.o parser.o halfmd4.o debug_locks.o random32.o \ bust_spinlocks.o hexdump.o kasprintf.o bitmap.o scatterlist.o \ - string_helpers.o gcd.o list_sort.o + string_helpers.o gcd.o lcm.o list_sort.o ifeq ($(CONFIG_DEBUG_KOBJECT),y) CFLAGS_kobject.o += -DDEBUG diff --git a/lib/lcm.c b/lib/lcm.c new file mode 100644 index 000000000000..157cd88a6ffc --- /dev/null +++ b/lib/lcm.c @@ -0,0 +1,15 @@ +#include +#include +#include + +/* Lowest common multiple */ +unsigned long lcm(unsigned long a, unsigned long b) +{ + if (a && b) + return (a * b) / gcd(a, b); + else if (b) + return b; + + return a; +} +EXPORT_SYMBOL_GPL(lcm); -- cgit v1.2.3 From ee714f2dd33e726346e34f5cda12543162f4753e Mon Sep 17 00:00:00 2001 From: "Martin K. Petersen" Date: Wed, 10 Mar 2010 00:48:32 -0500 Subject: block: Finalize conversion of block limits functions Remove compatibility wrappers and update remaining drivers. Signed-off-by: Martin K. Petersen Signed-off-by: Jens Axboe --- drivers/block/DAC960.c | 1 - drivers/block/virtio_blk.c | 5 ++--- include/linux/blkdev.h | 24 ------------------------ 3 files changed, 2 insertions(+), 28 deletions(-) (limited to 'include/linux') diff --git a/drivers/block/DAC960.c b/drivers/block/DAC960.c index 459f1bc25a7b..c5f22bb0a48e 100644 --- a/drivers/block/DAC960.c +++ b/drivers/block/DAC960.c @@ -2533,7 +2533,6 @@ static bool DAC960_RegisterBlockDevice(DAC960_Controller_T *Controller) Controller->RequestQueue[n] = RequestQueue; blk_queue_bounce_limit(RequestQueue, Controller->BounceBufferLimit); RequestQueue->queuedata = Controller; - blk_queue_max_hw_segments(RequestQueue, Controller->DriverScatterGatherLimit); blk_queue_max_segments(RequestQueue, Controller->DriverScatterGatherLimit); blk_queue_max_hw_sectors(RequestQueue, Controller->MaxBlocksPerCommand); disk->queue = RequestQueue; diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c index 3c64af05fa82..653817ceeedd 100644 --- a/drivers/block/virtio_blk.c +++ b/drivers/block/virtio_blk.c @@ -347,14 +347,13 @@ static int __devinit virtblk_probe(struct virtio_device *vdev) set_capacity(vblk->disk, cap); /* We can handle whatever the host told us to handle. */ - blk_queue_max_phys_segments(q, vblk->sg_elems-2); - blk_queue_max_hw_segments(q, vblk->sg_elems-2); + blk_queue_max_segments(q, vblk->sg_elems-2); /* No need to bounce any requests */ blk_queue_bounce_limit(q, BLK_BOUNCE_ANY); /* No real sector limit. */ - blk_queue_max_sectors(q, -1U); + blk_queue_max_hw_sectors(q, -1U); /* Host can optionally specify maximum segment size and number of * segments. */ diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index ebd22dbed861..41551c9341b6 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -921,26 +921,7 @@ extern void blk_cleanup_queue(struct request_queue *); extern void blk_queue_make_request(struct request_queue *, make_request_fn *); extern void blk_queue_bounce_limit(struct request_queue *, u64); extern void blk_queue_max_hw_sectors(struct request_queue *, unsigned int); - -/* Temporary compatibility wrapper */ -static inline void blk_queue_max_sectors(struct request_queue *q, unsigned int max) -{ - blk_queue_max_hw_sectors(q, max); -} - extern void blk_queue_max_segments(struct request_queue *, unsigned short); - -static inline void blk_queue_max_phys_segments(struct request_queue *q, unsigned short max) -{ - blk_queue_max_segments(q, max); -} - -static inline void blk_queue_max_hw_segments(struct request_queue *q, unsigned short max) -{ - blk_queue_max_segments(q, max); -} - - extern void blk_queue_max_segment_size(struct request_queue *, unsigned int); extern void blk_queue_max_discard_sectors(struct request_queue *q, unsigned int max_discard_sectors); @@ -1030,11 +1011,6 @@ static inline int sb_issue_discard(struct super_block *sb, extern int blk_verify_command(unsigned char *cmd, fmode_t has_write_perm); -#define MAX_PHYS_SEGMENTS 128 -#define MAX_HW_SEGMENTS 128 -#define SAFE_MAX_SECTORS 255 -#define MAX_SEGMENT_SIZE 65536 - enum blk_default_limits { BLK_MAX_SEGMENTS = 128, BLK_SAFE_MAX_SECTORS = 255, -- cgit v1.2.3 From 97fedbbe1046b3118f49df249840ca21041eefe4 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 16 Mar 2010 08:55:32 +0100 Subject: Remove GENHD_FL_DRIVERFS This flag is not used, so best discarded. Signed-off-by: NeilBrown -- Hi Jens, I came across this recently - these are the only two occurances of "GENHD_FL_DRIVERFS" in the kernel, so it cannot be needed. NeilBrown Signed-off-by: Jens Axboe --- drivers/scsi/sd.c | 2 +- include/linux/genhd.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c index 1dd4d8407694..466fae8ef770 100644 --- a/drivers/scsi/sd.c +++ b/drivers/scsi/sd.c @@ -2185,7 +2185,7 @@ static void sd_probe_async(void *data, async_cookie_t cookie) blk_queue_prep_rq(sdp->request_queue, sd_prep_fn); gd->driverfs_dev = &sdp->sdev_gendev; - gd->flags = GENHD_FL_EXT_DEVT | GENHD_FL_DRIVERFS; + gd->flags = GENHD_FL_EXT_DEVT; if (sdp->removable) gd->flags |= GENHD_FL_REMOVABLE; diff --git a/include/linux/genhd.h b/include/linux/genhd.h index 56b50514ab25..5f2f4c4d8fb0 100644 --- a/include/linux/genhd.h +++ b/include/linux/genhd.h @@ -109,7 +109,7 @@ struct hd_struct { }; #define GENHD_FL_REMOVABLE 1 -#define GENHD_FL_DRIVERFS 2 +/* 2 is unused */ #define GENHD_FL_MEDIA_CHANGE_NOTIFY 4 #define GENHD_FL_CD 8 #define GENHD_FL_UP 16 -- cgit v1.2.3 From 181fdde3b4268cb7b4af76ba6337e7ec8accbb36 Mon Sep 17 00:00:00 2001 From: Richard Kennedy Date: Fri, 19 Mar 2010 08:58:16 +0100 Subject: block: remove 16 bytes of padding from struct request on 64bits Remove alignment padding to shrink struct request from 336 to 320 bytes so needing one fewer cacheline and therefore removing 48 bytes from struct request_queue. Signed-off-by: Richard Kennedy Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 41551c9341b6..6690e8bae7bb 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -158,7 +158,6 @@ enum rq_flag_bits { struct request { struct list_head queuelist; struct call_single_data csd; - int cpu; struct request_queue *q; @@ -166,9 +165,11 @@ struct request { enum rq_cmd_type_bits cmd_type; unsigned long atomic_flags; + int cpu; + /* the following two fields are internal, NEVER access directly */ - sector_t __sector; /* sector cursor */ unsigned int __data_len; /* total data len */ + sector_t __sector; /* sector cursor */ struct bio *bio; struct bio *biotail; @@ -201,20 +202,20 @@ struct request { unsigned short ioprio; + int ref_count; + void *special; /* opaque pointer available for LLD use */ char *buffer; /* kaddr of the current segment if available */ int tag; int errors; - int ref_count; - /* * when request is used as a packet command carrier */ - unsigned short cmd_len; unsigned char __cmd[BLK_MAX_CDB]; unsigned char *cmd; + unsigned short cmd_len; unsigned int extra_len; /* length of alignment and padding */ unsigned int sense_len; -- cgit v1.2.3 From 8a6d9b149f105f8bdfa8e42dd9753e45a1887a16 Mon Sep 17 00:00:00 2001 From: Ferenc Wagner Date: Wed, 24 Mar 2010 08:20:03 +0100 Subject: i2o: Remove the dangerous kobj_to_i2o_device macro This macro worked only when applied to variables named 'kobj'. While this could have been fixed by simply renaming the macro argument, a more type-safe replacement by an inline function would be preferred. However, nobody uses this macro, so it's simpler to just remove it. Signed-off-by: Ferenc Wagner Signed-off-by: Jens Axboe --- include/linux/i2o.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/i2o.h b/include/linux/i2o.h index 87018dc5527d..9e7a12d6385d 100644 --- a/include/linux/i2o.h +++ b/include/linux/i2o.h @@ -782,7 +782,6 @@ extern int i2o_exec_lct_get(struct i2o_controller *); #define to_i2o_driver(drv) container_of(drv,struct i2o_driver, driver) #define to_i2o_device(dev) container_of(dev, struct i2o_device, device) #define to_i2o_controller(dev) container_of(dev, struct i2o_controller, device) -#define kobj_to_i2o_device(kobj) to_i2o_device(container_of(kobj, struct device, kobj)) /** * i2o_out_to_virt - Turn an I2O message to a virtual address -- cgit v1.2.3 From 6072f7491f5ef391a575e18a1165e72a3eef1601 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Wed, 31 Mar 2010 20:11:59 +0000 Subject: ide: Requeue request after DMA timeout I noticed that my KVM virtual machines were experiencing IDE issues resulting in processes stuck on waiting for buffers to complete. The root cause is of course race conditions in the ancient qemu backend that I'm using. However, the fact that the guest isn't recovering is a bug. I've tracked it down to the change made last year to dequeue requests at the start rather than at the end in the IDE layer. commit 8f6205cd572fece673da0255d74843680f67f879 Author: Tejun Heo Date: Fri May 8 11:53:59 2009 +0900 ide: dequeue in-flight request The problem is that the function ide_dma_timeout_retry does not requeue the current request, causing one request to be lost for each DMA timeout. This patch fixes this by requeueing the request. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- drivers/ide/ide-dma.c | 1 + drivers/ide/ide-io.c | 2 +- include/linux/ide.h | 1 + 3 files changed, 3 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c index ee58c88dee5a..fd40a8116574 100644 --- a/drivers/ide/ide-dma.c +++ b/drivers/ide/ide-dma.c @@ -492,6 +492,7 @@ ide_startstop_t ide_dma_timeout_retry(ide_drive_t *drive, int error) if (rq) { hwif->rq = NULL; rq->errors = 0; + ide_requeue_and_plug(drive, rq); } return ret; } diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c index db96138fefcd..172ac9218154 100644 --- a/drivers/ide/ide-io.c +++ b/drivers/ide/ide-io.c @@ -566,7 +566,7 @@ plug_device_2: blk_plug_device(q); } -static void ide_requeue_and_plug(ide_drive_t *drive, struct request *rq) +void ide_requeue_and_plug(ide_drive_t *drive, struct request *rq) { struct request_queue *q = drive->queue; unsigned long flags; diff --git a/include/linux/ide.h b/include/linux/ide.h index 97e6ab435184..3239d1c10acb 100644 --- a/include/linux/ide.h +++ b/include/linux/ide.h @@ -1169,6 +1169,7 @@ extern void ide_stall_queue(ide_drive_t *drive, unsigned long timeout); extern void ide_timer_expiry(unsigned long); extern irqreturn_t ide_intr(int irq, void *dev_id); extern void do_ide_request(struct request_queue *); +extern void ide_requeue_and_plug(ide_drive_t *drive, struct request *rq); void ide_init_disk(struct gendisk *, ide_drive_t *); -- cgit v1.2.3 From 042be38e6106ed70b42d096ab4a1ed4187e510e6 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Thu, 1 Apr 2010 14:32:43 -0700 Subject: ibft, x86: Change reserve_ibft_region() to find_ibft_region() This allows arch code could decide the way to reserve the ibft. And we should reserve ibft as early as possible, instead of BOOTMEM stage, in case the table is in RAM range and is not reserved by BIOS (this will often be the case.) Move to just after find_smp_config(). Also when CONFIG_NO_BOOTMEM=y, We will not have reserve_bootmem() anymore. -v2: fix typo about ibft pointed by Konrad Rzeszutek Wilk Signed-off-by: Yinghai Lu LKML-Reference: <4BB510FB.80601@kernel.org> Cc: Pekka Enberg Cc: Peter Jones Cc: Konrad Rzeszutek Wilk CC: Jan Beulich Signed-off-by: H. Peter Anvin --- arch/x86/kernel/setup.c | 14 ++++++++++++-- drivers/firmware/iscsi_ibft_find.c | 11 ++++++++--- include/linux/iscsi_ibft.h | 8 ++++++-- 3 files changed, 26 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index d76e18570c60..580e6b3dbdb8 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -608,6 +608,16 @@ static int __init setup_elfcorehdr(char *arg) early_param("elfcorehdr", setup_elfcorehdr); #endif +static __init void reserve_ibft_region(void) +{ + unsigned long addr, size = 0; + + addr = find_ibft_region(&size); + + if (size) + reserve_early_overlap_ok(addr, addr + size, "ibft"); +} + #ifdef CONFIG_X86_RESERVE_LOW_64K static int __init dmi_low_memory_corruption(const struct dmi_system_id *d) { @@ -910,6 +920,8 @@ void __init setup_arch(char **cmdline_p) */ find_smp_config(); + reserve_ibft_region(); + reserve_trampoline_memory(); #ifdef CONFIG_ACPI_SLEEP @@ -977,8 +989,6 @@ void __init setup_arch(char **cmdline_p) dma32_reserve_bootmem(); - reserve_ibft_region(); - #ifdef CONFIG_KVM_CLOCK kvmclock_init(); #endif diff --git a/drivers/firmware/iscsi_ibft_find.c b/drivers/firmware/iscsi_ibft_find.c index dfb15c06c88f..8f5d9e258829 100644 --- a/drivers/firmware/iscsi_ibft_find.c +++ b/drivers/firmware/iscsi_ibft_find.c @@ -52,7 +52,7 @@ EXPORT_SYMBOL_GPL(ibft_addr); * Routine used to find the iSCSI Boot Format Table. The logical * kernel address is set in the ibft_addr global variable. */ -void __init reserve_ibft_region(void) +unsigned long __init find_ibft_region(unsigned long *sizep) { unsigned long pos; unsigned int len = 0; @@ -78,6 +78,11 @@ void __init reserve_ibft_region(void) } } } - if (ibft_addr) - reserve_bootmem(pos, PAGE_ALIGN(len), BOOTMEM_DEFAULT); + if (ibft_addr) { + *sizep = PAGE_ALIGN(len); + return pos; + } + + *sizep = 0; + return 0; } diff --git a/include/linux/iscsi_ibft.h b/include/linux/iscsi_ibft.h index 6092487e2950..d2e4042f8f5e 100644 --- a/include/linux/iscsi_ibft.h +++ b/include/linux/iscsi_ibft.h @@ -42,9 +42,13 @@ extern struct ibft_table_header *ibft_addr; * mapped address is set in the ibft_addr variable. */ #ifdef CONFIG_ISCSI_IBFT_FIND -extern void __init reserve_ibft_region(void); +unsigned long find_ibft_region(unsigned long *sizep); #else -static inline void reserve_ibft_region(void) { } +static inline unsigned long find_ibft_region(unsigned long *sizep) +{ + *sizep = 0; + return 0; +} #endif #endif /* ISCSI_IBFT_H */ -- cgit v1.2.3 From 9d32c30542f9ecdb4b96a1a960924c9f403e3562 Mon Sep 17 00:00:00 2001 From: H Hartley Sweeten Date: Mon, 5 Apr 2010 22:29:09 -0700 Subject: Input: matrix_keypad - allow platform to disable key autorepeat In an embedded system the matrix_keypad driver might be used to interface with an external control panel and not an actual keyboard. On the control panel some of the keys could be used to turn on/off various functions. If key autorepeat is enabled this causes the function to quickly toggle between the on and off states and makes operation difficult. Add an option in the platform-specific data to disable the key autorepeat. Signed-off-by: H Hartley Sweeten Signed-off-by: Dmitry Torokhov --- drivers/input/keyboard/matrix_keypad.c | 4 +++- include/linux/input/matrix_keypad.h | 2 ++ 2 files changed, 5 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/input/keyboard/matrix_keypad.c b/drivers/input/keyboard/matrix_keypad.c index d3c8b61a941d..f9d86cfb0db0 100644 --- a/drivers/input/keyboard/matrix_keypad.c +++ b/drivers/input/keyboard/matrix_keypad.c @@ -373,7 +373,9 @@ static int __devinit matrix_keypad_probe(struct platform_device *pdev) input_dev->name = pdev->name; input_dev->id.bustype = BUS_HOST; input_dev->dev.parent = &pdev->dev; - input_dev->evbit[0] = BIT_MASK(EV_KEY) | BIT_MASK(EV_REP); + input_dev->evbit[0] = BIT_MASK(EV_KEY); + if (!pdata->no_autorepeat) + input_dev->evbit[0] |= BIT_MASK(EV_REP); input_dev->open = matrix_keypad_start; input_dev->close = matrix_keypad_stop; diff --git a/include/linux/input/matrix_keypad.h b/include/linux/input/matrix_keypad.h index 3bd018baae20..c964cd7f436a 100644 --- a/include/linux/input/matrix_keypad.h +++ b/include/linux/input/matrix_keypad.h @@ -44,6 +44,7 @@ struct matrix_keymap_data { * @active_low: gpio polarity * @wakeup: controls whether the device should be set up as wakeup * source + * @no_autorepeat: disable key autorepeat * * This structure represents platform-specific data that use used by * matrix_keypad driver to perform proper initialization. @@ -64,6 +65,7 @@ struct matrix_keypad_platform_data { bool active_low; bool wakeup; + bool no_autorepeat; }; /** -- cgit v1.2.3 From b01d0942c2b7a3026d2b7d38b5773d3d00420e06 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 6 Apr 2010 14:34:41 -0700 Subject: bitops: remove temporary for_each_bit() Migration has been completed so remove this now. There's one straggler in linux-next's drivers/mtd/sm_ftl.c. A patch has been sent. Cc: Akinobu Mita Cc: Stephen Rothwell Cc: David Woodhouse Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/bitops.h | 3 --- 1 file changed, 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/bitops.h b/include/linux/bitops.h index b79389879238..b796eab5ca75 100644 --- a/include/linux/bitops.h +++ b/include/linux/bitops.h @@ -21,9 +21,6 @@ (bit) < (size); \ (bit) = find_next_bit((addr), (size), (bit) + 1)) -/* Temporary */ -#define for_each_bit(bit, addr, size) for_each_set_bit(bit, addr, size) - static __inline__ int get_bitmask_order(unsigned int count) { int order; -- cgit v1.2.3 From 530cd330dc3865e3107304a6e84fdc332aa72f7d Mon Sep 17 00:00:00 2001 From: David Härdeman Date: Tue, 6 Apr 2010 14:34:43 -0700 Subject: include/linux/kfifo.h: fix INIT_KFIFO() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit DECLARE_KFIFO creates a union with a struct kfifo and a buffer array with size [size + sizeof(struct kfifo)]. INIT_KFIFO then sets the buffer pointer in struct kfifo to point to the beginning of the buffer array which means that the first call to kfifo_in will overwrite members of the struct kfifo. Signed-off-by: David Härdeman Acked-by: Stefani Seibold Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kfifo.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/kfifo.h b/include/linux/kfifo.h index ece0b1c33816..e117b1aee69c 100644 --- a/include/linux/kfifo.h +++ b/include/linux/kfifo.h @@ -86,7 +86,8 @@ union { \ */ #define INIT_KFIFO(name) \ name = __kfifo_initializer(sizeof(name##kfifo_buffer) - \ - sizeof(struct kfifo), name##kfifo_buffer) + sizeof(struct kfifo), \ + name##kfifo_buffer + sizeof(struct kfifo)) /** * DEFINE_KFIFO - macro to define and initialize a kfifo -- cgit v1.2.3 From 55ab3a1ff843e3f0e24d2da44e71bffa5d853010 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Tue, 6 Apr 2010 14:34:58 -0700 Subject: raw: fsync method is now required Commit 148f948ba877f4d3cdef036b1ff6d9f68986706a (vfs: Introduce new helpers for syncing after writing to O_SYNC file or IS_SYNC inode) broke the raw driver. We now call through generic_file_aio_write -> generic_write_sync -> vfs_fsync_range. vfs_fsync_range has: if (!fop || !fop->fsync) { ret = -EINVAL; goto out; } But drivers/char/raw.c doesn't set an fsync method. We have two options: fix it or remove the raw driver completely. I'm happy to do either, the fact this has been broken for so long suggests it is rarely used. The patch below adds an fsync method to the raw driver. My knowledge of the block layer is pretty sketchy so this could do with a once over. If we instead decide to remove the raw driver, this patch might still be useful as a backport to 2.6.33 and 2.6.32. Signed-off-by: Anton Blanchard Reviewed-by: Jan Kara Cc: Christoph Hellwig Cc: Alexander Viro Cc: Jens Axboe Reviewed-by: Jeff Moyer Tested-by: Jeff Moyer Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/raw.c | 1 + fs/block_dev.c | 3 ++- include/linux/fs.h | 1 + 3 files changed, 4 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/char/raw.c b/drivers/char/raw.c index d331c59b571c..f5a32317b751 100644 --- a/drivers/char/raw.c +++ b/drivers/char/raw.c @@ -248,6 +248,7 @@ static const struct file_operations raw_fops = { .aio_read = generic_file_aio_read, .write = do_sync_write, .aio_write = blkdev_aio_write, + .fsync = block_fsync, .open = raw_open, .release= raw_release, .ioctl = raw_ioctl, diff --git a/fs/block_dev.c b/fs/block_dev.c index d11d0289f3d2..8db62b2b6df8 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -404,7 +404,7 @@ static loff_t block_llseek(struct file *file, loff_t offset, int origin) * NULL first argument is nfsd_sync_dir() and that's not a directory. */ -static int block_fsync(struct file *filp, struct dentry *dentry, int datasync) +int block_fsync(struct file *filp, struct dentry *dentry, int datasync) { struct block_device *bdev = I_BDEV(filp->f_mapping->host); int error; @@ -418,6 +418,7 @@ static int block_fsync(struct file *filp, struct dentry *dentry, int datasync) error = 0; return error; } +EXPORT_SYMBOL(block_fsync); /* * pseudo-fs diff --git a/include/linux/fs.h b/include/linux/fs.h index 10b8dedcd18b..5d9c7e27c5a4 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2212,6 +2212,7 @@ extern int generic_segment_checks(const struct iovec *iov, /* fs/block_dev.c */ extern ssize_t blkdev_aio_write(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos); +extern int block_fsync(struct file *filp, struct dentry *dentry, int datasync); /* fs/splice.c */ extern ssize_t generic_file_splice_read(struct file *, loff_t *, -- cgit v1.2.3 From b1dd3b2843b3b73b7fc2ee47d96310cd1c051371 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 6 Apr 2010 14:35:00 -0700 Subject: vfs: rename block_fsync() to blkdev_fsync() Requested by hch, for consistency now it is exported. Cc: Alexander Viro Cc: Anton Blanchard Cc: Christoph Hellwig Cc: Jan Kara Cc: Jeff Moyer Cc: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/raw.c | 2 +- fs/block_dev.c | 6 +++--- include/linux/fs.h | 2 +- 3 files changed, 5 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/drivers/char/raw.c b/drivers/char/raw.c index f5a32317b751..8756ab0daa8b 100644 --- a/drivers/char/raw.c +++ b/drivers/char/raw.c @@ -248,7 +248,7 @@ static const struct file_operations raw_fops = { .aio_read = generic_file_aio_read, .write = do_sync_write, .aio_write = blkdev_aio_write, - .fsync = block_fsync, + .fsync = blkdev_fsync, .open = raw_open, .release= raw_release, .ioctl = raw_ioctl, diff --git a/fs/block_dev.c b/fs/block_dev.c index 8db62b2b6df8..2a6d0193f139 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -404,7 +404,7 @@ static loff_t block_llseek(struct file *file, loff_t offset, int origin) * NULL first argument is nfsd_sync_dir() and that's not a directory. */ -int block_fsync(struct file *filp, struct dentry *dentry, int datasync) +int blkdev_fsync(struct file *filp, struct dentry *dentry, int datasync) { struct block_device *bdev = I_BDEV(filp->f_mapping->host); int error; @@ -418,7 +418,7 @@ int block_fsync(struct file *filp, struct dentry *dentry, int datasync) error = 0; return error; } -EXPORT_SYMBOL(block_fsync); +EXPORT_SYMBOL(blkdev_fsync); /* * pseudo-fs @@ -1482,7 +1482,7 @@ const struct file_operations def_blk_fops = { .aio_read = generic_file_aio_read, .aio_write = blkdev_aio_write, .mmap = generic_file_mmap, - .fsync = block_fsync, + .fsync = blkdev_fsync, .unlocked_ioctl = block_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = compat_blkdev_ioctl, diff --git a/include/linux/fs.h b/include/linux/fs.h index 5d9c7e27c5a4..39d57bc6cc71 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2212,7 +2212,7 @@ extern int generic_segment_checks(const struct iovec *iov, /* fs/block_dev.c */ extern ssize_t blkdev_aio_write(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos); -extern int block_fsync(struct file *filp, struct dentry *dentry, int datasync); +extern int blkdev_fsync(struct file *filp, struct dentry *dentry, int datasync); /* fs/splice.c */ extern ssize_t generic_file_splice_read(struct file *, loff_t *, -- cgit v1.2.3 From bb1dc0bacb8ddd7ba6a5906c678a5a5a110cf695 Mon Sep 17 00:00:00 2001 From: Yong Zhang Date: Tue, 6 Apr 2010 14:35:02 -0700 Subject: kernel.h: fix wrong usage of __ratelimit() When __ratelimit() returns 1 this means that we can go ahead. Signed-off-by: Yong Zhang Cc: Ingo Molnar Cc: Joe Perches Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kernel.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 7f0707463360..9365227dbaf6 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -426,7 +426,7 @@ static inline char *pack_hex_byte(char *buf, u8 byte) .burst = DEFAULT_RATELIMIT_BURST, \ }; \ \ - if (!__ratelimit(&_rs)) \ + if (__ratelimit(&_rs)) \ printk(fmt, ##__VA_ARGS__); \ }) #else -- cgit v1.2.3 From 116354d177ba2da37e91cf884e3d11e67f825efd Mon Sep 17 00:00:00 2001 From: Naoya Horiguchi Date: Tue, 6 Apr 2010 14:35:04 -0700 Subject: pagemap: fix pfn calculation for hugepage When we look into pagemap using page-types with option -p, the value of pfn for hugepages looks wrong (see below.) This is because pte was evaluated only once for one vma although it should be updated for each hugepage. This patch fixes it. $ page-types -p 3277 -Nl -b huge voffset offset len flags 7f21e8a00 11e400 1 ___U___________H_G________________ 7f21e8a01 11e401 1ff ________________TG________________ ^^^ 7f21e8c00 11e400 1 ___U___________H_G________________ 7f21e8c01 11e401 1ff ________________TG________________ ^^^ One hugepage contains 1 head page and 511 tail pages in x86_64 and each two lines represent each hugepage. Voffset and offset mean virtual address and physical address in the page unit, respectively. The different hugepages should not have the same offset value. With this patch applied: $ page-types -p 3386 -Nl -b huge voffset offset len flags 7fec7a600 112c00 1 ___UD__________H_G________________ 7fec7a601 112c01 1ff ________________TG________________ ^^^ 7fec7a800 113200 1 ___UD__________H_G________________ 7fec7a801 113201 1ff ________________TG________________ ^^^ OK More info: - This patch modifies walk_page_range()'s hugepage walker. But the change only affects pagemap_read(), which is the only caller of hugepage callback. - Without this patch, hugetlb_entry() callback is called per vma, that doesn't match the natural expectation from its name. - With this patch, hugetlb_entry() is called per hugepte entry and the callback can become much simpler. Signed-off-by: Naoya Horiguchi Signed-off-by: KAMEZAWA Hiroyuki Acked-by: Matt Mackall Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/task_mmu.c | 27 +++++++-------------------- include/linux/mm.h | 4 ++-- mm/pagewalk.c | 47 +++++++++++++++++++++++++++++++++++++---------- 3 files changed, 46 insertions(+), 32 deletions(-) (limited to 'include/linux') diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index a05a669510a4..070553427dd5 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -662,31 +662,18 @@ static u64 huge_pte_to_pagemap_entry(pte_t pte, int offset) return pme; } -static int pagemap_hugetlb_range(pte_t *pte, unsigned long addr, - unsigned long end, struct mm_walk *walk) +/* This function walks within one hugetlb entry in the single call */ +static int pagemap_hugetlb_range(pte_t *pte, unsigned long hmask, + unsigned long addr, unsigned long end, + struct mm_walk *walk) { - struct vm_area_struct *vma; struct pagemapread *pm = walk->private; - struct hstate *hs = NULL; int err = 0; + u64 pfn; - vma = find_vma(walk->mm, addr); - if (vma) - hs = hstate_vma(vma); for (; addr != end; addr += PAGE_SIZE) { - u64 pfn = PM_NOT_PRESENT; - - if (vma && (addr >= vma->vm_end)) { - vma = find_vma(walk->mm, addr); - if (vma) - hs = hstate_vma(vma); - } - - if (vma && (vma->vm_start <= addr) && is_vm_hugetlb_page(vma)) { - /* calculate pfn of the "raw" page in the hugepage. */ - int offset = (addr & ~huge_page_mask(hs)) >> PAGE_SHIFT; - pfn = huge_pte_to_pagemap_entry(*pte, offset); - } + int offset = (addr & ~hmask) >> PAGE_SHIFT; + pfn = huge_pte_to_pagemap_entry(*pte, offset); err = add_to_pagemap(addr, pfn, pm); if (err) return err; diff --git a/include/linux/mm.h b/include/linux/mm.h index e70f21beb4b4..462acaf36f3a 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -783,8 +783,8 @@ struct mm_walk { int (*pmd_entry)(pmd_t *, unsigned long, unsigned long, struct mm_walk *); int (*pte_entry)(pte_t *, unsigned long, unsigned long, struct mm_walk *); int (*pte_hole)(unsigned long, unsigned long, struct mm_walk *); - int (*hugetlb_entry)(pte_t *, unsigned long, unsigned long, - struct mm_walk *); + int (*hugetlb_entry)(pte_t *, unsigned long, + unsigned long, unsigned long, struct mm_walk *); struct mm_struct *mm; void *private; }; diff --git a/mm/pagewalk.c b/mm/pagewalk.c index 7b47a57b6646..8b1a2ce21ee5 100644 --- a/mm/pagewalk.c +++ b/mm/pagewalk.c @@ -80,6 +80,37 @@ static int walk_pud_range(pgd_t *pgd, unsigned long addr, unsigned long end, return err; } +#ifdef CONFIG_HUGETLB_PAGE +static unsigned long hugetlb_entry_end(struct hstate *h, unsigned long addr, + unsigned long end) +{ + unsigned long boundary = (addr & huge_page_mask(h)) + huge_page_size(h); + return boundary < end ? boundary : end; +} + +static int walk_hugetlb_range(struct vm_area_struct *vma, + unsigned long addr, unsigned long end, + struct mm_walk *walk) +{ + struct hstate *h = hstate_vma(vma); + unsigned long next; + unsigned long hmask = huge_page_mask(h); + pte_t *pte; + int err = 0; + + do { + next = hugetlb_entry_end(h, addr, end); + pte = huge_pte_offset(walk->mm, addr & hmask); + if (pte && walk->hugetlb_entry) + err = walk->hugetlb_entry(pte, hmask, addr, next, walk); + if (err) + return err; + } while (addr = next, addr != end); + + return 0; +} +#endif + /** * walk_page_range - walk a memory map's page tables with a callback * @mm: memory map to walk @@ -128,20 +159,16 @@ int walk_page_range(unsigned long addr, unsigned long end, vma = find_vma(walk->mm, addr); #ifdef CONFIG_HUGETLB_PAGE if (vma && is_vm_hugetlb_page(vma)) { - pte_t *pte; - struct hstate *hs; - if (vma->vm_end < next) next = vma->vm_end; - hs = hstate_vma(vma); - pte = huge_pte_offset(walk->mm, - addr & huge_page_mask(hs)); - if (pte && !huge_pte_none(huge_ptep_get(pte)) - && walk->hugetlb_entry) - err = walk->hugetlb_entry(pte, addr, - next, walk); + /* + * Hugepage is very tightly coupled with vma, so + * walk through hugetlb entries within a given vma. + */ + err = walk_hugetlb_range(vma, addr, next, walk); if (err) break; + pgd = pgd_offset(walk->mm, next); continue; } #endif -- cgit v1.2.3 From 8725d5416213a145ccc9c236dbd26830ba409e00 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Tue, 6 Apr 2010 14:35:05 -0700 Subject: memcg: fix race in file_mapped accounting Presently, memcg's FILE_MAPPED accounting has following race with move_account (happens at rmdir()). increment page->mapcount (rmap.c) mem_cgroup_update_file_mapped() move_account() lock_page_cgroup() check page_mapped() if page_mapped(page)>1 { FILE_MAPPED -1 from old memcg FILE_MAPPED +1 to old memcg } ..... overwrite pc->mem_cgroup unlock_page_cgroup() lock_page_cgroup() FILE_MAPPED + 1 to pc->mem_cgroup unlock_page_cgroup() Then, old memcg (-1 file mapped) new memcg (+2 file mapped) This happens because move_account see page_mapped() which is not guarded by lock_page_cgroup(). This patch adds FILE_MAPPED flag to page_cgroup and move account information based on it. Now, all checks are synchronous with lock_page_cgroup(). Signed-off-by: KAMEZAWA Hiroyuki Reviewed-by: Balbir Singh Reviewed-by: Daisuke Nishimura Cc: Andrea Righi Cc: Andrea Arcangeli Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/page_cgroup.h | 6 ++++++ mm/memcontrol.c | 18 +++++++++--------- 2 files changed, 15 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/include/linux/page_cgroup.h b/include/linux/page_cgroup.h index 30b08136fdf3..aef22ae2af47 100644 --- a/include/linux/page_cgroup.h +++ b/include/linux/page_cgroup.h @@ -39,6 +39,7 @@ enum { PCG_CACHE, /* charged as cache */ PCG_USED, /* this object is in use. */ PCG_ACCT_LRU, /* page has been accounted for */ + PCG_FILE_MAPPED, /* page is accounted as "mapped" */ }; #define TESTPCGFLAG(uname, lname) \ @@ -73,6 +74,11 @@ CLEARPCGFLAG(AcctLRU, ACCT_LRU) TESTPCGFLAG(AcctLRU, ACCT_LRU) TESTCLEARPCGFLAG(AcctLRU, ACCT_LRU) + +SETPCGFLAG(FileMapped, FILE_MAPPED) +CLEARPCGFLAG(FileMapped, FILE_MAPPED) +TESTPCGFLAG(FileMapped, FILE_MAPPED) + static inline int page_cgroup_nid(struct page_cgroup *pc) { return page_to_nid(pc->page); diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 9ed760dc7448..f4ede99c8b9b 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1359,16 +1359,19 @@ void mem_cgroup_update_file_mapped(struct page *page, int val) lock_page_cgroup(pc); mem = pc->mem_cgroup; - if (!mem) - goto done; - - if (!PageCgroupUsed(pc)) + if (!mem || !PageCgroupUsed(pc)) goto done; /* * Preemption is already disabled. We can use __this_cpu_xxx */ - __this_cpu_add(mem->stat->count[MEM_CGROUP_STAT_FILE_MAPPED], val); + if (val > 0) { + __this_cpu_inc(mem->stat->count[MEM_CGROUP_STAT_FILE_MAPPED]); + SetPageCgroupFileMapped(pc); + } else { + __this_cpu_dec(mem->stat->count[MEM_CGROUP_STAT_FILE_MAPPED]); + ClearPageCgroupFileMapped(pc); + } done: unlock_page_cgroup(pc); @@ -1801,16 +1804,13 @@ static void __mem_cgroup_commit_charge(struct mem_cgroup *mem, static void __mem_cgroup_move_account(struct page_cgroup *pc, struct mem_cgroup *from, struct mem_cgroup *to, bool uncharge) { - struct page *page; - VM_BUG_ON(from == to); VM_BUG_ON(PageLRU(pc->page)); VM_BUG_ON(!PageCgroupLocked(pc)); VM_BUG_ON(!PageCgroupUsed(pc)); VM_BUG_ON(pc->mem_cgroup != from); - page = pc->page; - if (page_mapped(page) && !PageAnon(page)) { + if (PageCgroupFileMapped(pc)) { /* Update mapped_file data for mem_cgroup */ preempt_disable(); __this_cpu_dec(from->stat->count[MEM_CGROUP_STAT_FILE_MAPPED]); -- cgit v1.2.3 From b7a413015d2986edf020fba765c906cc9cbcbfc9 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Wed, 31 Mar 2010 21:56:42 +0300 Subject: virtio: disable multiport console support. Move MULTIPORT feature and related config changes out of exported headers, and disable the feature at runtime. At this point, it seems less risky to keep code around until we can enable it than rip it out completely. Signed-off-by: Michael S. Tsirkin Signed-off-by: Rusty Russell --- drivers/char/virtio_console.c | 49 +++++++++++++++++++++++++++++++++++------- include/linux/virtio_console.h | 23 -------------------- 2 files changed, 41 insertions(+), 31 deletions(-) (limited to 'include/linux') diff --git a/drivers/char/virtio_console.c b/drivers/char/virtio_console.c index 86e9011325dc..196428c2287a 100644 --- a/drivers/char/virtio_console.c +++ b/drivers/char/virtio_console.c @@ -33,6 +33,35 @@ #include #include "hvc_console.h" +/* Moved here from .h file in order to disable MULTIPORT. */ +#define VIRTIO_CONSOLE_F_MULTIPORT 1 /* Does host provide multiple ports? */ + +struct virtio_console_multiport_conf { + struct virtio_console_config config; + /* max. number of ports this device can hold */ + __u32 max_nr_ports; + /* number of ports added so far */ + __u32 nr_ports; +} __attribute__((packed)); + +/* + * A message that's passed between the Host and the Guest for a + * particular port. + */ +struct virtio_console_control { + __u32 id; /* Port number */ + __u16 event; /* The kind of control event (see below) */ + __u16 value; /* Extra information for the key */ +}; + +/* Some events for control messages */ +#define VIRTIO_CONSOLE_PORT_READY 0 +#define VIRTIO_CONSOLE_CONSOLE_PORT 1 +#define VIRTIO_CONSOLE_RESIZE 2 +#define VIRTIO_CONSOLE_PORT_OPEN 3 +#define VIRTIO_CONSOLE_PORT_NAME 4 +#define VIRTIO_CONSOLE_PORT_REMOVE 5 + /* * This is a global struct for storing common data for all the devices * this driver handles. @@ -121,7 +150,7 @@ struct ports_device { spinlock_t cvq_lock; /* The current config space is stored here */ - struct virtio_console_config config; + struct virtio_console_multiport_conf config; /* The virtio device we're associated with */ struct virtio_device *vdev; @@ -1214,7 +1243,7 @@ fail: */ static void config_work_handler(struct work_struct *work) { - struct virtio_console_config virtconconf; + struct virtio_console_multiport_conf virtconconf; struct ports_device *portdev; struct virtio_device *vdev; int err; @@ -1223,7 +1252,8 @@ static void config_work_handler(struct work_struct *work) vdev = portdev->vdev; vdev->config->get(vdev, - offsetof(struct virtio_console_config, nr_ports), + offsetof(struct virtio_console_multiport_conf, + nr_ports), &virtconconf.nr_ports, sizeof(virtconconf.nr_ports)); @@ -1415,16 +1445,19 @@ static int __devinit virtcons_probe(struct virtio_device *vdev) multiport = false; portdev->config.nr_ports = 1; portdev->config.max_nr_ports = 1; +#if 0 /* Multiport is not quite ready yet --RR */ if (virtio_has_feature(vdev, VIRTIO_CONSOLE_F_MULTIPORT)) { multiport = true; vdev->features[0] |= 1 << VIRTIO_CONSOLE_F_MULTIPORT; - vdev->config->get(vdev, offsetof(struct virtio_console_config, - nr_ports), + vdev->config->get(vdev, + offsetof(struct virtio_console_multiport_conf, + nr_ports), &portdev->config.nr_ports, sizeof(portdev->config.nr_ports)); - vdev->config->get(vdev, offsetof(struct virtio_console_config, - max_nr_ports), + vdev->config->get(vdev, + offsetof(struct virtio_console_multiport_conf, + max_nr_ports), &portdev->config.max_nr_ports, sizeof(portdev->config.max_nr_ports)); if (portdev->config.nr_ports > portdev->config.max_nr_ports) { @@ -1440,6 +1473,7 @@ static int __devinit virtcons_probe(struct virtio_device *vdev) /* Let the Host know we support multiple ports.*/ vdev->config->finalize_features(vdev); +#endif err = init_vqs(portdev); if (err < 0) { @@ -1522,7 +1556,6 @@ static struct virtio_device_id id_table[] = { static unsigned int features[] = { VIRTIO_CONSOLE_F_SIZE, - VIRTIO_CONSOLE_F_MULTIPORT, }; static struct virtio_driver virtio_console = { diff --git a/include/linux/virtio_console.h b/include/linux/virtio_console.h index ae4f039515b4..92228a8fbcbc 100644 --- a/include/linux/virtio_console.h +++ b/include/linux/virtio_console.h @@ -12,37 +12,14 @@ /* Feature bits */ #define VIRTIO_CONSOLE_F_SIZE 0 /* Does host provide console size? */ -#define VIRTIO_CONSOLE_F_MULTIPORT 1 /* Does host provide multiple ports? */ struct virtio_console_config { /* colums of the screens */ __u16 cols; /* rows of the screens */ __u16 rows; - /* max. number of ports this device can hold */ - __u32 max_nr_ports; - /* number of ports added so far */ - __u32 nr_ports; } __attribute__((packed)); -/* - * A message that's passed between the Host and the Guest for a - * particular port. - */ -struct virtio_console_control { - __u32 id; /* Port number */ - __u16 event; /* The kind of control event (see below) */ - __u16 value; /* Extra information for the key */ -}; - -/* Some events for control messages */ -#define VIRTIO_CONSOLE_PORT_READY 0 -#define VIRTIO_CONSOLE_CONSOLE_PORT 1 -#define VIRTIO_CONSOLE_RESIZE 2 -#define VIRTIO_CONSOLE_PORT_OPEN 3 -#define VIRTIO_CONSOLE_PORT_NAME 4 -#define VIRTIO_CONSOLE_PORT_REMOVE 5 - #ifdef __KERNEL__ int __init virtio_cons_early_init(int (*put_chars)(u32, const char *, int)); #endif /* __KERNEL__ */ -- cgit v1.2.3 From 45c4d015a92f72ec47acd0c7557abdc0c8a6499d Mon Sep 17 00:00:00 2001 From: Mark Lord Date: Wed, 7 Apr 2010 13:52:08 -0400 Subject: libata: Fix accesses at LBA28 boundary (old bug, but nasty) (v2) Most drives from Seagate, Hitachi, and possibly other brands, do not allow LBA28 access to sector number 0x0fffffff (2^28 - 1). So instead use LBA48 for such accesses. This bug could bite a lot of systems, especially when the user has taken care to align partitions to 4KB boundaries. On misaligned systems, it is less likely to be encountered, since a 4KB read would end at 0x10000000 rather than at 0x0fffffff. Signed-off-by: Mark Lord Signed-off-by: Jeff Garzik --- include/linux/ata.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/ata.h b/include/linux/ata.h index b4c85e2adef5..700c5b9b3583 100644 --- a/include/linux/ata.h +++ b/include/linux/ata.h @@ -1025,8 +1025,8 @@ static inline int ata_ok(u8 status) static inline int lba_28_ok(u64 block, u32 n_block) { - /* check the ending block number */ - return ((block + n_block) < ((u64)1 << 28)) && (n_block <= 256); + /* check the ending block number: must be LESS THAN 0x0fffffff */ + return ((block + n_block) < ((1 << 28) - 1)) && (n_block <= 256); } static inline int lba_48_ok(u64 block, u32 n_block) -- cgit v1.2.3 From fc1c183353a113c71675fecd0485e5aa0fe68d72 Mon Sep 17 00:00:00 2001 From: Pekka Enberg Date: Wed, 7 Apr 2010 19:23:40 +0300 Subject: slab: Generify kernel pointer validation As suggested by Linus, introduce a kern_ptr_validate() helper that does some sanity checks to make sure a pointer is a valid kernel pointer. This is a preparational step for fixing SLUB kmem_ptr_validate(). Cc: Andrew Morton Cc: Christoph Lameter Cc: David Rientjes Cc: Ingo Molnar Cc: Matt Mackall Cc: Nick Piggin Signed-off-by: Pekka Enberg Signed-off-by: Linus Torvalds --- include/linux/slab.h | 1 + mm/slab.c | 13 +------------ mm/util.c | 21 +++++++++++++++++++++ 3 files changed, 23 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/slab.h b/include/linux/slab.h index 488446289cab..49d1247cd6d9 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -106,6 +106,7 @@ int kmem_cache_shrink(struct kmem_cache *); void kmem_cache_free(struct kmem_cache *, void *); unsigned int kmem_cache_size(struct kmem_cache *); const char *kmem_cache_name(struct kmem_cache *); +int kern_ptr_validate(const void *ptr, unsigned long size); int kmem_ptr_validate(struct kmem_cache *cachep, const void *ptr); /* diff --git a/mm/slab.c b/mm/slab.c index a9f325b28bed..bac0f4fcc216 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -3602,21 +3602,10 @@ EXPORT_SYMBOL(kmem_cache_alloc_notrace); */ int kmem_ptr_validate(struct kmem_cache *cachep, const void *ptr) { - unsigned long addr = (unsigned long)ptr; - unsigned long min_addr = PAGE_OFFSET; - unsigned long align_mask = BYTES_PER_WORD - 1; unsigned long size = cachep->buffer_size; struct page *page; - if (unlikely(addr < min_addr)) - goto out; - if (unlikely(addr > (unsigned long)high_memory - size)) - goto out; - if (unlikely(addr & align_mask)) - goto out; - if (unlikely(!kern_addr_valid(addr))) - goto out; - if (unlikely(!kern_addr_valid(addr + size - 1))) + if (unlikely(!kern_ptr_validate(ptr, size))) goto out; page = virt_to_page(ptr); if (unlikely(!PageSlab(page))) diff --git a/mm/util.c b/mm/util.c index 834db7be240f..f5712e8964be 100644 --- a/mm/util.c +++ b/mm/util.c @@ -186,6 +186,27 @@ void kzfree(const void *p) } EXPORT_SYMBOL(kzfree); +int kern_ptr_validate(const void *ptr, unsigned long size) +{ + unsigned long addr = (unsigned long)ptr; + unsigned long min_addr = PAGE_OFFSET; + unsigned long align_mask = sizeof(void *) - 1; + + if (unlikely(addr < min_addr)) + goto out; + if (unlikely(addr > (unsigned long)high_memory - size)) + goto out; + if (unlikely(addr & align_mask)) + goto out; + if (unlikely(!kern_addr_valid(addr))) + goto out; + if (unlikely(!kern_addr_valid(addr + size - 1))) + goto out; + return 1; +out: + return 0; +} + /* * strndup_user - duplicate an existing string from user space * @s: The string to duplicate -- cgit v1.2.3 From ce82653d6cfcc95ba88c25908664878459fb1b8d Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 6 Apr 2010 22:36:20 +0100 Subject: radix_tree_tag_get() is not as safe as the docs make out [ver #2] radix_tree_tag_get() is not safe to use concurrently with radix_tree_tag_set() or radix_tree_tag_clear(). The problem is that the double tag_get() in radix_tree_tag_get(): if (!tag_get(node, tag, offset)) saw_unset_tag = 1; if (height == 1) { int ret = tag_get(node, tag, offset); may see the value change due to the action of set/clear. RCU is no protection against this as no pointers are being changed, no nodes are being replaced according to a COW protocol - set/clear alter the node directly. The documentation in linux/radix-tree.h, however, says that radix_tree_tag_get() is an exception to the rule that "any function modifying the tree or tags (...) must exclude other modifications, and exclude any functions reading the tree". The problem is that the next statement in radix_tree_tag_get() checks that the tag doesn't vary over time: BUG_ON(ret && saw_unset_tag); This has been seen happening in FS-Cache: https://www.redhat.com/archives/linux-cachefs/2010-April/msg00013.html To this end, remove the BUG_ON() from radix_tree_tag_get() and note in various comments that the value of the tag may change whilst the RCU read lock is held, and thus that the return value of radix_tree_tag_get() may not be relied upon unless radix_tree_tag_set/clear() and radix_tree_delete() are excluded from running concurrently with it. Reported-by: Romain DEGEZ Signed-off-by: David Howells Acked-by: Nick Piggin Signed-off-by: Linus Torvalds --- include/linux/radix-tree.h | 7 +++++++ lib/radix-tree.c | 12 ++++++------ 2 files changed, 13 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/radix-tree.h b/include/linux/radix-tree.h index c5da74918096..55ca73cf25e5 100644 --- a/include/linux/radix-tree.h +++ b/include/linux/radix-tree.h @@ -121,6 +121,13 @@ do { \ * (Note, rcu_assign_pointer and rcu_dereference are not needed to control * access to data items when inserting into or looking up from the radix tree) * + * Note that the value returned by radix_tree_tag_get() may not be relied upon + * if only the RCU read lock is held. Functions to set/clear tags and to + * delete nodes running concurrently with it may affect its result such that + * two consecutive reads in the same locked section may return different + * values. If reliability is required, modification functions must also be + * excluded from concurrency. + * * radix_tree_tagged is able to be called without locking or RCU. */ diff --git a/lib/radix-tree.c b/lib/radix-tree.c index 0871582aa29d..2a087e0f9863 100644 --- a/lib/radix-tree.c +++ b/lib/radix-tree.c @@ -555,6 +555,10 @@ EXPORT_SYMBOL(radix_tree_tag_clear); * * 0: tag not present or not set * 1: tag set + * + * Note that the return value of this function may not be relied on, even if + * the RCU lock is held, unless tag modification and node deletion are excluded + * from concurrency. */ int radix_tree_tag_get(struct radix_tree_root *root, unsigned long index, unsigned int tag) @@ -595,12 +599,8 @@ int radix_tree_tag_get(struct radix_tree_root *root, */ if (!tag_get(node, tag, offset)) saw_unset_tag = 1; - if (height == 1) { - int ret = tag_get(node, tag, offset); - - BUG_ON(ret && saw_unset_tag); - return !!ret; - } + if (height == 1) + return !!tag_get(node, tag, offset); node = rcu_dereference_raw(node->slots[offset]); shift -= RADIX_TREE_MAP_SHIFT; height--; -- cgit v1.2.3 From aa6fec3cdeb14ecc916eb78c4cd9ed79e4f7fe8d Mon Sep 17 00:00:00 2001 From: Clemens Ladisch Date: Wed, 31 Mar 2010 16:26:52 +0200 Subject: firewire: cdev: iso packet documentation Add the missing documentation for iso packets. Signed-off-by: Clemens Ladisch Signed-off-by: Stefan Richter --- include/linux/firewire-cdev.h | 39 +++++++++++++++++++++++++++++++++------ 1 file changed, 33 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/firewire-cdev.h b/include/linux/firewire-cdev.h index 40b11013408e..011fdf1eaec5 100644 --- a/include/linux/firewire-cdev.h +++ b/include/linux/firewire-cdev.h @@ -438,7 +438,7 @@ struct fw_cdev_remove_descriptor { * @type: %FW_CDEV_ISO_CONTEXT_TRANSMIT or %FW_CDEV_ISO_CONTEXT_RECEIVE * @header_size: Header size to strip for receive contexts * @channel: Channel to bind to - * @speed: Speed to transmit at + * @speed: Speed for transmit contexts * @closure: To be returned in &fw_cdev_event_iso_interrupt * @handle: Handle to context, written back by kernel * @@ -451,6 +451,9 @@ struct fw_cdev_remove_descriptor { * If a context was successfully created, the kernel writes back a handle to the * context, which must be passed in for subsequent operations on that context. * + * For receive contexts, @header_size must be at least 4 and must be a multiple + * of 4. + * * Note that the effect of a @header_size > 4 depends on * &fw_cdev_get_info.version, as documented at &fw_cdev_event_iso_interrupt. */ @@ -481,10 +484,34 @@ struct fw_cdev_create_iso_context { * * &struct fw_cdev_iso_packet is used to describe isochronous packet queues. * - * Use the FW_CDEV_ISO_ macros to fill in @control. The sy and tag fields are - * specified by IEEE 1394a and IEC 61883. - * - * FIXME - finish this documentation + * Use the FW_CDEV_ISO_ macros to fill in @control. + * + * For transmit packets, the header length must be a multiple of 4 and specifies + * the numbers of bytes in @header that will be prepended to the packet's + * payload; these bytes are copied into the kernel and will not be accessed + * after the ioctl has returned. The sy and tag fields are copied to the iso + * packet header (these fields are specified by IEEE 1394a and IEC 61883-1). + * The skip flag specifies that no packet is to be sent in a frame; when using + * this, all other fields except the interrupt flag must be zero. + * + * For receive packets, the header length must be a multiple of the context's + * header size; if the header length is larger than the context's header size, + * multiple packets are queued for this entry. The sy and tag fields are + * ignored. If the sync flag is set, the context drops all packets until + * a packet with a matching sy field is received (the sync value to wait for is + * specified in the &fw_cdev_start_iso structure). The payload length defines + * how many payload bytes can be received for one packet (in addition to payload + * quadlets that have been defined as headers and are stripped and returned in + * the &fw_cdev_event_iso_interrupt structure). If more bytes are received, the + * additional bytes are dropped. If less bytes are received, the remaining + * bytes in this part of the payload buffer will not be written to, not even by + * the next packet, i.e., packets received in consecutive frames will not + * necessarily be consecutive in memory. If an entry has queued multiple + * packets, the payload length is divided equally among them. + * + * When a packet with the interrupt flag set has been completed, the + * &fw_cdev_event_iso_interrupt event will be sent. An entry that has queued + * multiple receive packets is completed when its last packet is completed. */ struct fw_cdev_iso_packet { __u32 control; @@ -501,7 +528,7 @@ struct fw_cdev_iso_packet { * Queue a number of isochronous packets for reception or transmission. * This ioctl takes a pointer to an array of &fw_cdev_iso_packet structs, * which describe how to transmit from or receive into a contiguous region - * of a mmap()'ed payload buffer. As part of the packet descriptors, + * of a mmap()'ed payload buffer. As part of transmit packet descriptors, * a series of headers can be supplied, which will be prepended to the * payload during DMA. * -- cgit v1.2.3 From ca658b1e29d6be939207532e337fb640eb697f71 Mon Sep 17 00:00:00 2001 From: Stefan Richter Date: Sat, 10 Apr 2010 12:23:09 +0200 Subject: firewire: cdev: comment fixlet Signed-off-by: Stefan Richter --- include/linux/firewire-cdev.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/firewire-cdev.h b/include/linux/firewire-cdev.h index 011fdf1eaec5..6ffb24a1f2f2 100644 --- a/include/linux/firewire-cdev.h +++ b/include/linux/firewire-cdev.h @@ -647,8 +647,8 @@ struct fw_cdev_get_cycle_timer2 { * instead of allocated. * An %FW_CDEV_EVENT_ISO_RESOURCE_DEALLOCATED event concludes this operation. * - * To summarize, %FW_CDEV_IOC_DEALLOCATE_ISO_RESOURCE allocates iso resources - * for the lifetime of the fd or handle. + * To summarize, %FW_CDEV_IOC_ALLOCATE_ISO_RESOURCE allocates iso resources + * for the lifetime of the fd or @handle. * In contrast, %FW_CDEV_IOC_ALLOCATE_ISO_RESOURCE_ONCE allocates iso resources * for the duration of a bus generation. * -- cgit v1.2.3 From 0df5dd4aae211edeeeb84f7f84f6d093406d7c22 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 11 Apr 2010 16:48:44 -0400 Subject: NFSv4: fix delegated locking Arnaud Giersch reports that NFSv4 locking is broken when we hold a delegation since commit 8e469ebd6dc32cbaf620e134d79f740bf0ebab79 (NFSv4: Don't allow posix locking against servers that don't support it). According to Arnaud, the lock succeeds the first time he opens the file (since we cannot do a delegated open) but then fails after we start using delegated opens. The following patch fixes it by ensuring that locking behaviour is governed by a per-filesystem capability flag that is initially set, but gets cleared if the server ever returns an OPEN without the NFS4_OPEN_RESULT_LOCKTYPE_POSIX flag being set. Reported-by: Arnaud Giersch Signed-off-by: Trond Myklebust Cc: stable@kernel.org --- fs/nfs/client.c | 3 ++- fs/nfs/nfs4proc.c | 4 +++- include/linux/nfs_fs_sb.h | 1 + 3 files changed, 6 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 2a3d352c0bff..a8766c4ef2e0 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -1294,7 +1294,8 @@ static int nfs4_init_server(struct nfs_server *server, /* Initialise the client representation from the mount data */ server->flags = data->flags; - server->caps |= NFS_CAP_ATOMIC_OPEN|NFS_CAP_CHANGE_ATTR; + server->caps |= NFS_CAP_ATOMIC_OPEN|NFS_CAP_CHANGE_ATTR| + NFS_CAP_POSIX_LOCK; server->options = data->options; /* Get a client record */ diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index fe0cd9eb1d4d..638067007c65 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -1523,6 +1523,8 @@ static int _nfs4_proc_open(struct nfs4_opendata *data) nfs_post_op_update_inode(dir, o_res->dir_attr); } else nfs_refresh_inode(dir, o_res->dir_attr); + if ((o_res->rflags & NFS4_OPEN_RESULT_LOCKTYPE_POSIX) == 0) + server->caps &= ~NFS_CAP_POSIX_LOCK; if(o_res->rflags & NFS4_OPEN_RESULT_CONFIRM) { status = _nfs4_proc_open_confirm(data); if (status != 0) @@ -1664,7 +1666,7 @@ static int _nfs4_do_open(struct inode *dir, struct path *path, fmode_t fmode, in status = PTR_ERR(state); if (IS_ERR(state)) goto err_opendata_put; - if ((opendata->o_res.rflags & NFS4_OPEN_RESULT_LOCKTYPE_POSIX) != 0) + if (server->caps & NFS_CAP_POSIX_LOCK) set_bit(NFS_STATE_POSIX_LOCKS, &state->flags); nfs4_opendata_put(opendata); nfs4_put_state_owner(sp); diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h index 717a5e54eb1d..e82957acea56 100644 --- a/include/linux/nfs_fs_sb.h +++ b/include/linux/nfs_fs_sb.h @@ -176,6 +176,7 @@ struct nfs_server { #define NFS_CAP_ATIME (1U << 11) #define NFS_CAP_CTIME (1U << 12) #define NFS_CAP_MTIME (1U << 13) +#define NFS_CAP_POSIX_LOCK (1U << 14) /* maximum number of slots to use */ -- cgit v1.2.3 From b62730baea32f86fe91a7930e4b7ee8d82778b79 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 9 Apr 2010 15:39:10 -0700 Subject: rcu: Add rcu_access_pointer and rcu_dereference_protected This patch adds variants of rcu_dereference() that handle situations where the RCU-protected data structure cannot change, perhaps due to our holding the update-side lock, or where the RCU-protected pointer is only to be fetched, not dereferenced. These are needed due to some performance concerns with using rcu_dereference() where it is not required, aside from the need for lockdep/sparse checking. The new rcu_access_pointer() primitive is for the case where the pointer is be fetch and not dereferenced. This primitive may be used without protection, RCU or otherwise, due to the fact that it uses ACCESS_ONCE(). The new rcu_dereference_protected() primitive is for the case where updates are prevented, for example, due to holding the update-side lock. This primitive does neither ACCESS_ONCE() nor smp_read_barrier_depends(), so can only be used when updates are somehow prevented. Suggested-by: David Howells Signed-off-by: Paul E. McKenney Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com Cc: eric.dumazet@gmail.com LKML-Reference: <1270852752-25278-1-git-send-email-paulmck@linux.vnet.ibm.com> Signed-off-by: Ingo Molnar --- include/linux/rcupdate.h | 32 ++++++++++++++++++++++++++++++++ 1 file changed, 32 insertions(+) (limited to 'include/linux') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 872a98e13d6a..8fe86609441f 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -209,12 +209,44 @@ static inline int rcu_read_lock_sched_held(void) rcu_dereference_raw(p); \ }) +/** + * rcu_dereference_protected - fetch RCU pointer when updates prevented + * + * Return the value of the specified RCU-protected pointer, but omit + * both the smp_read_barrier_depends() and the ACCESS_ONCE(). This + * is useful in cases where update-side locks prevent the value of the + * pointer from changing. Please note that this primitive does -not- + * prevent the compiler from repeating this reference or combining it + * with other references, so it should not be used without protection + * of appropriate locks. + */ +#define rcu_dereference_protected(p, c) \ + ({ \ + if (debug_lockdep_rcu_enabled() && !(c)) \ + lockdep_rcu_dereference(__FILE__, __LINE__); \ + (p); \ + }) + #else /* #ifdef CONFIG_PROVE_RCU */ #define rcu_dereference_check(p, c) rcu_dereference_raw(p) +#define rcu_dereference_protected(p, c) (p) #endif /* #else #ifdef CONFIG_PROVE_RCU */ +/** + * rcu_access_pointer - fetch RCU pointer with no dereferencing + * + * Return the value of the specified RCU-protected pointer, but omit the + * smp_read_barrier_depends() and keep the ACCESS_ONCE(). This is useful + * when the value of this pointer is accessed, but the pointer is not + * dereferenced, for example, when testing an RCU-protected pointer against + * NULL. This may also be used in cases where update-side locks prevent + * the value of the pointer from changing, but rcu_dereference_protected() + * is a lighter-weight primitive for this use case. + */ +#define rcu_access_pointer(p) ACCESS_ONCE(p) + /** * rcu_read_lock - mark the beginning of an RCU read-side critical section. * -- cgit v1.2.3 From c08c68dd76bd6b776bc0eb45a5e8f354ed772cdf Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 9 Apr 2010 15:39:11 -0700 Subject: rcu: Better explain the condition parameter of rcu_dereference_check() Better explain the condition parameter of rcu_dereference_check() that describes the conditions under which the dereference is permitted to take place (and incorporate Yong Zhang's suggestion). This condition is only checked under lockdep proving. Signed-off-by: David Howells Signed-off-by: Paul E. McKenney Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: eric.dumazet@gmail.com LKML-Reference: <1270852752-25278-2-git-send-email-paulmck@linux.vnet.ibm.com> Signed-off-by: Ingo Molnar --- include/linux/rcupdate.h | 28 +++++++++++++++++++++++----- 1 file changed, 23 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 8fe86609441f..9f1ddfef84b5 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -195,12 +195,30 @@ static inline int rcu_read_lock_sched_held(void) /** * rcu_dereference_check - rcu_dereference with debug checking + * @p: The pointer to read, prior to dereferencing + * @c: The conditions under which the dereference will take place * - * Do an rcu_dereference(), but check that the context is correct. - * For example, rcu_dereference_check(gp, rcu_read_lock_held()) to - * ensure that the rcu_dereference_check() executes within an RCU - * read-side critical section. It is also possible to check for - * locks being held, for example, by using lockdep_is_held(). + * Do an rcu_dereference(), but check that the conditions under which the + * dereference will take place are correct. Typically the conditions indicate + * the various locking conditions that should be held at that point. The check + * should return true if the conditions are satisfied. + * + * For example: + * + * bar = rcu_dereference_check(foo->bar, rcu_read_lock_held() || + * lockdep_is_held(&foo->lock)); + * + * could be used to indicate to lockdep that foo->bar may only be dereferenced + * if either the RCU read lock is held, or that the lock required to replace + * the bar struct at foo->bar is held. + * + * Note that the list of conditions may also include indications of when a lock + * need not be held, for example during initialisation or destruction of the + * target struct: + * + * bar = rcu_dereference_check(foo->bar, rcu_read_lock_held() || + * lockdep_is_held(&foo->lock) || + * atomic_read(&foo->usage) == 0); */ #define rcu_dereference_check(p, c) \ ({ \ -- cgit v1.2.3 From 19b3eecc21b65a24b0aae2684ca0c8e1b99ef802 Mon Sep 17 00:00:00 2001 From: Stefan Richter Date: Sun, 11 Apr 2010 11:52:12 +0200 Subject: firewire: cdev: change license of exported header files to MIT license MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Among else, this allows projects like libdc1394 to carry copies of the ABI related header files without them or distributors having to worry about effects on the project's overall license terms. Switch to MIT license as suggested by Kristian. Also update the year in the copyright statement according to source history. Cc: Jay Fenlason Acked-by: Clemens Ladisch Signed-off-by: Stefan Richter Signed-off-by: Kristian Høgsberg --- include/linux/firewire-cdev.h | 35 ++++++++++++++++++++--------------- include/linux/firewire-constants.h | 29 +++++++++++++++++++++++++++-- 2 files changed, 47 insertions(+), 17 deletions(-) (limited to 'include/linux') diff --git a/include/linux/firewire-cdev.h b/include/linux/firewire-cdev.h index 6ffb24a1f2f2..81f3b14d5d76 100644 --- a/include/linux/firewire-cdev.h +++ b/include/linux/firewire-cdev.h @@ -1,21 +1,26 @@ /* * Char device interface. * - * Copyright (C) 2005-2006 Kristian Hoegsberg - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software Foundation, - * Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * Copyright (C) 2005-2007 Kristian Hoegsberg + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * PRECISION INSIGHT AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER + * DEALINGS IN THE SOFTWARE. */ #ifndef _LINUX_FIREWIRE_CDEV_H diff --git a/include/linux/firewire-constants.h b/include/linux/firewire-constants.h index b316770a43fd..9c63f06e67f2 100644 --- a/include/linux/firewire-constants.h +++ b/include/linux/firewire-constants.h @@ -1,3 +1,28 @@ +/* + * IEEE 1394 constants. + * + * Copyright (C) 2005-2007 Kristian Hoegsberg + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * PRECISION INSIGHT AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER + * DEALINGS IN THE SOFTWARE. + */ + #ifndef _LINUX_FIREWIRE_CONSTANTS_H #define _LINUX_FIREWIRE_CONSTANTS_H @@ -21,7 +46,7 @@ #define EXTCODE_WRAP_ADD 0x6 #define EXTCODE_VENDOR_DEPENDENT 0x7 -/* Juju specific tcodes */ +/* Linux firewire-core (Juju) specific tcodes */ #define TCODE_LOCK_MASK_SWAP (0x10 | EXTCODE_MASK_SWAP) #define TCODE_LOCK_COMPARE_SWAP (0x10 | EXTCODE_COMPARE_SWAP) #define TCODE_LOCK_FETCH_ADD (0x10 | EXTCODE_FETCH_ADD) @@ -36,7 +61,7 @@ #define RCODE_TYPE_ERROR 0x6 #define RCODE_ADDRESS_ERROR 0x7 -/* Juju specific rcodes */ +/* Linux firewire-core (Juju) specific rcodes */ #define RCODE_SEND_ERROR 0x10 #define RCODE_CANCELLED 0x11 #define RCODE_BUSY 0x12 -- cgit v1.2.3 From a2612cb16d4d8447793609cbdd2a2f4f156c0020 Mon Sep 17 00:00:00 2001 From: Stefan Richter Date: Thu, 15 Apr 2010 22:16:04 +0200 Subject: firewire: cdev: fix cut+paste mistake in disclaimer This was supposed to be generic "authors or copyright holders"; I mistakenly picked up text from a wrong file. Reported-by: Daniel K. Signed-off-by: Stefan Richter --- include/linux/firewire-cdev.h | 2 +- include/linux/firewire-constants.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/firewire-cdev.h b/include/linux/firewire-cdev.h index 81f3b14d5d76..68f883b30a53 100644 --- a/include/linux/firewire-cdev.h +++ b/include/linux/firewire-cdev.h @@ -17,7 +17,7 @@ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * PRECISION INSIGHT AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, DAMAGES OR + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER * DEALINGS IN THE SOFTWARE. diff --git a/include/linux/firewire-constants.h b/include/linux/firewire-constants.h index 9c63f06e67f2..9b4bb5fbba4b 100644 --- a/include/linux/firewire-constants.h +++ b/include/linux/firewire-constants.h @@ -17,7 +17,7 @@ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * PRECISION INSIGHT AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, DAMAGES OR + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER * DEALINGS IN THE SOFTWARE. -- cgit v1.2.3 From bc293d62b26ec590afc90a9e0a31c45d355b7bd8 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 15 Apr 2010 12:50:39 -0700 Subject: rcu: Make RCU lockdep check the lockdep_recursion variable The lockdep facility temporarily disables lockdep checking by incrementing the current->lockdep_recursion variable. Such disabling happens in NMIs and in other situations where lockdep might expect to recurse on itself. This patch therefore checks current->lockdep_recursion, disabling RCU lockdep splats when this variable is non-zero. In addition, this patch removes the "likely()", as suggested by Lai Jiangshan. Reported-by: Frederic Weisbecker Reported-by: David Miller Tested-by: Frederic Weisbecker Signed-off-by: Paul E. McKenney Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com Cc: eric.dumazet@gmail.com LKML-Reference: <20100415195039.GA22623@linux.vnet.ibm.com> Signed-off-by: Ingo Molnar --- include/linux/rcupdate.h | 5 +---- kernel/rcupdate.c | 7 +++++++ 2 files changed, 8 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 9f1ddfef84b5..07db2feb8572 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -101,10 +101,7 @@ extern struct lockdep_map rcu_sched_lock_map; # define rcu_read_release_sched() \ lock_release(&rcu_sched_lock_map, 1, _THIS_IP_) -static inline int debug_lockdep_rcu_enabled(void) -{ - return likely(rcu_scheduler_active && debug_locks); -} +extern int debug_lockdep_rcu_enabled(void); /** * rcu_read_lock_held - might we be in RCU read-side critical section? diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c index 63fe25433980..03a7ea1579f6 100644 --- a/kernel/rcupdate.c +++ b/kernel/rcupdate.c @@ -69,6 +69,13 @@ EXPORT_SYMBOL_GPL(rcu_scheduler_active); #ifdef CONFIG_DEBUG_LOCK_ALLOC +int debug_lockdep_rcu_enabled(void) +{ + return rcu_scheduler_active && debug_locks && + current->lockdep_recursion == 0; +} +EXPORT_SYMBOL_GPL(debug_lockdep_rcu_enabled); + /** * rcu_read_lock_bh_held - might we be in RCU-bh read-side critical section? * -- cgit v1.2.3 From be1a50d4eba4cdb3ebf9d97a0a8693c153436775 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Sat, 3 Apr 2010 17:37:45 +0200 Subject: regulator: Let drivers know when they use the stub API Have the stub variant of regulator_get() return NULL, so that drivers can (but still don't have to) handle this case specifically. Signed-off-by: Jean Delvare Cc: Mark Brown Cc: Jerome Oufella Acked-by: Mark Brown Signed-off-by: Liam Girdwood --- include/linux/regulator/consumer.h | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/regulator/consumer.h b/include/linux/regulator/consumer.h index 28c9fd020d39..ebd747265294 100644 --- a/include/linux/regulator/consumer.h +++ b/include/linux/regulator/consumer.h @@ -183,9 +183,13 @@ static inline struct regulator *__must_check regulator_get(struct device *dev, { /* Nothing except the stubbed out regulator API should be * looking at the value except to check if it is an error - * value so the actual return value doesn't matter. + * value. Drivers are free to handle NULL specifically by + * skipping all regulator API calls, but they don't have to. + * Drivers which don't, should make sure they properly handle + * corner cases of the API, such as regulator_get_voltage() + * returning 0. */ - return (struct regulator *)id; + return NULL; } static inline void regulator_put(struct regulator *regulator) { -- cgit v1.2.3 From 87bf6e7de1134f48681fd2ce4b7c1ec45458cb6d Mon Sep 17 00:00:00 2001 From: Takuya Yoshikawa Date: Mon, 12 Apr 2010 19:35:35 +0900 Subject: KVM: fix the handling of dirty bitmaps to avoid overflows Int is not long enough to store the size of a dirty bitmap. This patch fixes this problem with the introduction of a wrapper function to calculate the sizes of dirty bitmaps. Note: in mark_page_dirty(), we have to consider the fact that __set_bit() takes the offset as int, not long. Signed-off-by: Takuya Yoshikawa Signed-off-by: Marcelo Tosatti --- arch/ia64/kvm/kvm-ia64.c | 9 +++++---- arch/powerpc/kvm/book3s.c | 5 +++-- arch/x86/kvm/x86.c | 5 +++-- include/linux/kvm_host.h | 5 +++++ virt/kvm/kvm_main.c | 13 ++++++++----- 5 files changed, 24 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c index 73c5c2b05f64..7f3c0a2e60cd 100644 --- a/arch/ia64/kvm/kvm-ia64.c +++ b/arch/ia64/kvm/kvm-ia64.c @@ -1802,7 +1802,8 @@ static int kvm_ia64_sync_dirty_log(struct kvm *kvm, { struct kvm_memory_slot *memslot; int r, i; - long n, base; + long base; + unsigned long n; unsigned long *dirty_bitmap = (unsigned long *)(kvm->arch.vm_base + offsetof(struct kvm_vm_data, kvm_mem_dirty_log)); @@ -1815,7 +1816,7 @@ static int kvm_ia64_sync_dirty_log(struct kvm *kvm, if (!memslot->dirty_bitmap) goto out; - n = ALIGN(memslot->npages, BITS_PER_LONG) / 8; + n = kvm_dirty_bitmap_bytes(memslot); base = memslot->base_gfn / BITS_PER_LONG; for (i = 0; i < n/sizeof(long); ++i) { @@ -1831,7 +1832,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, struct kvm_dirty_log *log) { int r; - int n; + unsigned long n; struct kvm_memory_slot *memslot; int is_dirty = 0; @@ -1850,7 +1851,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, if (is_dirty) { kvm_flush_remote_tlbs(kvm); memslot = &kvm->memslots->memslots[log->slot]; - n = ALIGN(memslot->npages, BITS_PER_LONG) / 8; + n = kvm_dirty_bitmap_bytes(memslot); memset(memslot->dirty_bitmap, 0, n); } r = 0; diff --git a/arch/powerpc/kvm/book3s.c b/arch/powerpc/kvm/book3s.c index 25da07fd9f77..604af29b71ed 100644 --- a/arch/powerpc/kvm/book3s.c +++ b/arch/powerpc/kvm/book3s.c @@ -1004,7 +1004,8 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, struct kvm_vcpu *vcpu; ulong ga, ga_end; int is_dirty = 0; - int r, n; + int r; + unsigned long n; mutex_lock(&kvm->slots_lock); @@ -1022,7 +1023,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, kvm_for_each_vcpu(n, vcpu, kvm) kvmppc_mmu_pte_pflush(vcpu, ga, ga_end); - n = ALIGN(memslot->npages, BITS_PER_LONG) / 8; + n = kvm_dirty_bitmap_bytes(memslot); memset(memslot->dirty_bitmap, 0, n); } diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 9ad3d064c781..45aa90f8cc57 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -2612,8 +2612,9 @@ static int kvm_vm_ioctl_reinject(struct kvm *kvm, int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, struct kvm_dirty_log *log) { - int r, n, i; + int r, i; struct kvm_memory_slot *memslot; + unsigned long n; unsigned long is_dirty = 0; unsigned long *dirty_bitmap = NULL; @@ -2628,7 +2629,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, if (!memslot->dirty_bitmap) goto out; - n = ALIGN(memslot->npages, BITS_PER_LONG) / 8; + n = kvm_dirty_bitmap_bytes(memslot); r = -ENOMEM; dirty_bitmap = vmalloc(n); diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index a3fd0f91d943..9ad825e1c79b 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -119,6 +119,11 @@ struct kvm_memory_slot { int user_alloc; }; +static inline unsigned long kvm_dirty_bitmap_bytes(struct kvm_memory_slot *memslot) +{ + return ALIGN(memslot->npages, BITS_PER_LONG) / 8; +} + struct kvm_kernel_irq_routing_entry { u32 gsi; u32 type; diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 5a0cd194dce0..364daacafb58 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -648,7 +648,7 @@ skip_lpage: /* Allocate page dirty bitmap if needed */ if ((new.flags & KVM_MEM_LOG_DIRTY_PAGES) && !new.dirty_bitmap) { - unsigned dirty_bytes = ALIGN(npages, BITS_PER_LONG) / 8; + unsigned long dirty_bytes = kvm_dirty_bitmap_bytes(&new); new.dirty_bitmap = vmalloc(dirty_bytes); if (!new.dirty_bitmap) @@ -768,7 +768,7 @@ int kvm_get_dirty_log(struct kvm *kvm, { struct kvm_memory_slot *memslot; int r, i; - int n; + unsigned long n; unsigned long any = 0; r = -EINVAL; @@ -780,7 +780,7 @@ int kvm_get_dirty_log(struct kvm *kvm, if (!memslot->dirty_bitmap) goto out; - n = ALIGN(memslot->npages, BITS_PER_LONG) / 8; + n = kvm_dirty_bitmap_bytes(memslot); for (i = 0; !any && i < n/sizeof(long); ++i) any = memslot->dirty_bitmap[i]; @@ -1186,10 +1186,13 @@ void mark_page_dirty(struct kvm *kvm, gfn_t gfn) memslot = gfn_to_memslot_unaliased(kvm, gfn); if (memslot && memslot->dirty_bitmap) { unsigned long rel_gfn = gfn - memslot->base_gfn; + unsigned long *p = memslot->dirty_bitmap + + rel_gfn / BITS_PER_LONG; + int offset = rel_gfn % BITS_PER_LONG; /* avoid RMW */ - if (!generic_test_le_bit(rel_gfn, memslot->dirty_bitmap)) - generic___set_le_bit(rel_gfn, memslot->dirty_bitmap); + if (!generic_test_le_bit(offset, p)) + generic___set_le_bit(offset, p); } } -- cgit v1.2.3 From e80e2a60ff7914dae691345a976c80bbbff3ec74 Mon Sep 17 00:00:00 2001 From: Sridhar Samudrala Date: Tue, 30 Mar 2010 16:48:25 -0700 Subject: KVM: Increase NR_IOBUS_DEVS limit to 200 This patch increases the current hardcoded limit of NR_IOBUS_DEVS from 6 to 200. We are hitting this limit when creating a guest with more than 1 virtio-net device using vhost-net backend. Each virtio-net device requires 2 such devices to service notifications from rx/tx queues. Signed-off-by: Sridhar Samudrala Signed-off-by: Avi Kivity --- include/linux/kvm_host.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 9ad825e1c79b..169d07758ee5 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -54,7 +54,7 @@ extern struct kmem_cache *kvm_vcpu_cache; */ struct kvm_io_bus { int dev_count; -#define NR_IOBUS_DEVS 6 +#define NR_IOBUS_DEVS 200 struct kvm_io_device *devs[NR_IOBUS_DEVS]; }; -- cgit v1.2.3 From c3c532061e46156e8aab1268f38d66cfb63aeb2d Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 22 Apr 2010 11:37:01 +0200 Subject: bdi: add helper function for doing init and register of a bdi for a file system Pretty trivial helper, just sets up the bdi and registers it. An atomic sequence count is used to ensure that the registered sysfs names are unique. Signed-off-by: Jens Axboe --- include/linux/backing-dev.h | 1 + mm/backing-dev.c | 29 +++++++++++++++++++++++++++++ 2 files changed, 30 insertions(+) (limited to 'include/linux') diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index fcbc26af00e4..e19c677f219c 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -101,6 +101,7 @@ int bdi_register(struct backing_dev_info *bdi, struct device *parent, const char *fmt, ...); int bdi_register_dev(struct backing_dev_info *bdi, dev_t dev); void bdi_unregister(struct backing_dev_info *bdi); +int bdi_setup_and_register(struct backing_dev_info *, char *, unsigned int); void bdi_start_writeback(struct backing_dev_info *bdi, struct super_block *sb, long nr_pages); int bdi_writeback_task(struct bdi_writeback *wb); diff --git a/mm/backing-dev.c b/mm/backing-dev.c index f13e067e1467..dbda4707f593 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -11,6 +11,8 @@ #include #include +static atomic_long_t bdi_seq = ATOMIC_LONG_INIT(0); + void default_unplug_io_fn(struct backing_dev_info *bdi, struct page *page) { } @@ -715,6 +717,33 @@ void bdi_destroy(struct backing_dev_info *bdi) } EXPORT_SYMBOL(bdi_destroy); +/* + * For use from filesystems to quickly init and register a bdi associated + * with dirty writeback + */ +int bdi_setup_and_register(struct backing_dev_info *bdi, char *name, + unsigned int cap) +{ + char tmp[32]; + int err; + + bdi->name = name; + bdi->capabilities = cap; + err = bdi_init(bdi); + if (err) + return err; + + sprintf(tmp, "%.28s%s", name, "-%d"); + err = bdi_register(bdi, NULL, tmp, atomic_long_inc_return(&bdi_seq)); + if (err) { + bdi_destroy(bdi); + return err; + } + + return 0; +} +EXPORT_SYMBOL(bdi_setup_and_register); + static wait_queue_head_t congestion_wqh[2] = { __WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[0]), __WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[1]) -- cgit v1.2.3 From 5163d90076729413cb882d3dd5c3d3cfb5b9f035 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 22 Apr 2010 12:12:40 +0200 Subject: coda: add bdi backing to mount session This ensures that dirty data gets flushed properly. Signed-off-by: Jens Axboe --- fs/coda/inode.c | 8 ++++++++ include/linux/coda_psdev.h | 2 ++ 2 files changed, 10 insertions(+) (limited to 'include/linux') diff --git a/fs/coda/inode.c b/fs/coda/inode.c index a1695dcadd99..d97f9935a028 100644 --- a/fs/coda/inode.c +++ b/fs/coda/inode.c @@ -167,6 +167,10 @@ static int coda_fill_super(struct super_block *sb, void *data, int silent) return -EBUSY; } + error = bdi_setup_and_register(&vc->bdi, "coda", BDI_CAP_MAP_COPY); + if (error) + goto bdi_err; + vc->vc_sb = sb; sb->s_fs_info = vc; @@ -175,6 +179,7 @@ static int coda_fill_super(struct super_block *sb, void *data, int silent) sb->s_blocksize_bits = 12; sb->s_magic = CODA_SUPER_MAGIC; sb->s_op = &coda_super_operations; + sb->s_bdi = &vc->bdi; /* get root fid from Venus: this needs the root inode */ error = venus_rootfid(sb, &fid); @@ -200,6 +205,8 @@ static int coda_fill_super(struct super_block *sb, void *data, int silent) return 0; error: + bdi_destroy(&vc->bdi); + bdi_err: if (root) iput(root); if (vc) @@ -210,6 +217,7 @@ static int coda_fill_super(struct super_block *sb, void *data, int silent) static void coda_put_super(struct super_block *sb) { + bdi_destroy(&coda_vcp(sb)->bdi); coda_vcp(sb)->vc_sb = NULL; sb->s_fs_info = NULL; diff --git a/include/linux/coda_psdev.h b/include/linux/coda_psdev.h index 5b5d4731f956..644062e8d857 100644 --- a/include/linux/coda_psdev.h +++ b/include/linux/coda_psdev.h @@ -1,6 +1,7 @@ #ifndef __CODA_PSDEV_H #define __CODA_PSDEV_H +#include #include #define CODA_PSDEV_MAJOR 67 @@ -17,6 +18,7 @@ struct venus_comm { struct list_head vc_processing; int vc_inuse; struct super_block *vc_sb; + struct backing_dev_info bdi; }; -- cgit v1.2.3 From f1970c73cbb6b884152207e4dfe90639f5029905 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 22 Apr 2010 12:31:11 +0200 Subject: ncpfs: add bdi backing to mount session This ensures that dirty data gets flushed properly. Signed-off-by: Jens Axboe --- fs/ncpfs/inode.c | 8 ++++++++ include/linux/ncp_fs_sb.h | 2 ++ 2 files changed, 10 insertions(+) (limited to 'include/linux') diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index cf98da1be23e..fa3385154023 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -526,10 +526,15 @@ static int ncp_fill_super(struct super_block *sb, void *raw_data, int silent) sb->s_blocksize_bits = 10; sb->s_magic = NCP_SUPER_MAGIC; sb->s_op = &ncp_sops; + sb->s_bdi = &server->bdi; server = NCP_SBP(sb); memset(server, 0, sizeof(*server)); + error = bdi_setup_and_register(&server->bdi, "ncpfs", BDI_CAP_MAP_COPY); + if (error) + goto out_bdi; + server->ncp_filp = ncp_filp; server->ncp_sock = sock; @@ -719,6 +724,8 @@ out_fput2: if (server->info_filp) fput(server->info_filp); out_fput: + bdi_destroy(&server->bdi); +out_bdi: /* 23/12/1998 Marcin Dalecki : * * The previously used put_filp(ncp_filp); was bogous, since @@ -756,6 +763,7 @@ static void ncp_put_super(struct super_block *sb) kill_pid(server->m.wdog_pid, SIGTERM, 1); put_pid(server->m.wdog_pid); + bdi_destroy(&server->bdi); kfree(server->priv.data); kfree(server->auth.object_name); vfree(server->rxbuf); diff --git a/include/linux/ncp_fs_sb.h b/include/linux/ncp_fs_sb.h index 6330fc76b00f..5ec9ca671687 100644 --- a/include/linux/ncp_fs_sb.h +++ b/include/linux/ncp_fs_sb.h @@ -12,6 +12,7 @@ #include #include #include +#include #ifdef __KERNEL__ @@ -127,6 +128,7 @@ struct ncp_server { size_t len; __u8 data[128]; } unexpected_packet; + struct backing_dev_info bdi; }; extern void ncp_tcp_rcv_proc(struct work_struct *work); -- cgit v1.2.3 From 424264b7b220e8eee165dc3080ae48692af73dec Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 22 Apr 2010 12:37:07 +0200 Subject: smbfs: add bdi backing to mount session This ensures that dirty data gets flushed properly. Signed-off-by: Jens Axboe --- fs/smbfs/inode.c | 8 ++++++++ include/linux/smb_fs_sb.h | 3 +++ 2 files changed, 11 insertions(+) (limited to 'include/linux') diff --git a/fs/smbfs/inode.c b/fs/smbfs/inode.c index 1c4c8f089970..dfa1d67f8fca 100644 --- a/fs/smbfs/inode.c +++ b/fs/smbfs/inode.c @@ -479,6 +479,7 @@ smb_put_super(struct super_block *sb) if (server->conn_pid) kill_pid(server->conn_pid, SIGTERM, 1); + bdi_destroy(&server->bdi); kfree(server->ops); smb_unload_nls(server); sb->s_fs_info = NULL; @@ -525,6 +526,11 @@ static int smb_fill_super(struct super_block *sb, void *raw_data, int silent) if (!server) goto out_no_server; sb->s_fs_info = server; + + if (bdi_setup_and_register(&server->bdi, "smbfs", BDI_CAP_MAP_COPY)) + goto out_bdi; + + sb->s_bdi = &server->bdi; server->super_block = sb; server->mnt = NULL; @@ -624,6 +630,8 @@ out_no_smbiod: out_bad_option: kfree(mem); out_no_mem: + bdi_destroy(&server->bdi); +out_bdi: if (!server->mnt) printk(KERN_ERR "smb_fill_super: allocation failure\n"); sb->s_fs_info = NULL; diff --git a/include/linux/smb_fs_sb.h b/include/linux/smb_fs_sb.h index 8a060a7040d8..bb947dd1fba9 100644 --- a/include/linux/smb_fs_sb.h +++ b/include/linux/smb_fs_sb.h @@ -10,6 +10,7 @@ #define _SMB_FS_SB #include +#include #include /* @@ -74,6 +75,8 @@ struct smb_sb_info { struct smb_ops *ops; struct super_block *super_block; + + struct backing_dev_info bdi; }; static inline int -- cgit v1.2.3 From 71d0a6112a363e703e383ae5b12c492485c39701 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 22 Apr 2010 15:35:57 -0400 Subject: NFS: Fix an unstable write data integrity race Commit 2c61be0a9478258f77b66208a0c4b1f5f8161c3c (NFS: Ensure that the WRITE and COMMIT RPC calls are always uninterruptible) exposed a race on file close. In order to ensure correct close-to-open behaviour, we want to wait for all outstanding background commit operations to complete. This patch adds an inode flag that indicates if a commit operation is under way, and provides a mechanism to allow ->write_inode() to wait for its completion if this is a data integrity flush. Signed-off-by: Trond Myklebust --- fs/nfs/write.c | 36 ++++++++++++++++++++++++++++++++---- include/linux/nfs_fs.h | 1 + 2 files changed, 33 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/write.c b/fs/nfs/write.c index de38d63aa920..ccde2aeb3fec 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -1201,6 +1201,25 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data) #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4) +static int nfs_commit_set_lock(struct nfs_inode *nfsi, int may_wait) +{ + if (!test_and_set_bit(NFS_INO_COMMIT, &nfsi->flags)) + return 1; + if (may_wait && !out_of_line_wait_on_bit_lock(&nfsi->flags, + NFS_INO_COMMIT, nfs_wait_bit_killable, + TASK_KILLABLE)) + return 1; + return 0; +} + +static void nfs_commit_clear_lock(struct nfs_inode *nfsi) +{ + clear_bit(NFS_INO_COMMIT, &nfsi->flags); + smp_mb__after_clear_bit(); + wake_up_bit(&nfsi->flags, NFS_INO_COMMIT); +} + + static void nfs_commitdata_release(void *data) { struct nfs_write_data *wdata = data; @@ -1262,8 +1281,6 @@ static int nfs_commit_rpcsetup(struct list_head *head, task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) return PTR_ERR(task); - if (how & FLUSH_SYNC) - rpc_wait_for_completion_task(task); rpc_put_task(task); return 0; } @@ -1294,6 +1311,7 @@ nfs_commit_list(struct inode *inode, struct list_head *head, int how) BDI_RECLAIMABLE); nfs_clear_page_tag_locked(req); } + nfs_commit_clear_lock(NFS_I(inode)); return -ENOMEM; } @@ -1349,6 +1367,7 @@ static void nfs_commit_release(void *calldata) next: nfs_clear_page_tag_locked(req); } + nfs_commit_clear_lock(NFS_I(data->inode)); nfs_commitdata_release(calldata); } @@ -1363,8 +1382,11 @@ static const struct rpc_call_ops nfs_commit_ops = { static int nfs_commit_inode(struct inode *inode, int how) { LIST_HEAD(head); - int res; + int may_wait = how & FLUSH_SYNC; + int res = 0; + if (!nfs_commit_set_lock(NFS_I(inode), may_wait)) + goto out; spin_lock(&inode->i_lock); res = nfs_scan_commit(inode, &head, 0, 0); spin_unlock(&inode->i_lock); @@ -1372,7 +1394,13 @@ static int nfs_commit_inode(struct inode *inode, int how) int error = nfs_commit_list(inode, &head, how); if (error < 0) return error; - } + if (may_wait) + wait_on_bit(&NFS_I(inode)->flags, NFS_INO_COMMIT, + nfs_wait_bit_killable, + TASK_KILLABLE); + } else + nfs_commit_clear_lock(NFS_I(inode)); +out: return res; } diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 1a0b85aa151e..07ce4609fe50 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -209,6 +209,7 @@ struct nfs_inode { #define NFS_INO_FLUSHING (4) /* inode is flushing out data */ #define NFS_INO_FSCACHE (5) /* inode can be cached by FS-Cache */ #define NFS_INO_FSCACHE_LOCK (6) /* FS-Cache cookie management lock */ +#define NFS_INO_COMMIT (7) /* inode is committing unstable writes */ static inline struct nfs_inode *NFS_I(const struct inode *inode) { -- cgit v1.2.3 From 3a3076f4d6e2fa31338a0b007df42a3b32f079e0 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 23 Apr 2010 12:17:17 -0400 Subject: Cleanup generic block based fiemap This cleans up a few of the complaints of __generic_block_fiemap. I've fixed all the typing stuff, used inline functions instead of macros, gotten rid of a couple of variables, and made sure the size and block requests are all block aligned. It also fixes a problem where sometimes FIEMAP_EXTENT_LAST wasn't being set properly. Signed-off-by: Josef Bacik Signed-off-by: Linus Torvalds --- fs/ioctl.c | 92 +++++++++++++++++++++++++++++++----------------------- include/linux/fs.h | 5 +-- 2 files changed, 56 insertions(+), 41 deletions(-) (limited to 'include/linux') diff --git a/fs/ioctl.c b/fs/ioctl.c index 6c751106c2e5..7faefb4da939 100644 --- a/fs/ioctl.c +++ b/fs/ioctl.c @@ -228,14 +228,23 @@ static int ioctl_fiemap(struct file *filp, unsigned long arg) #ifdef CONFIG_BLOCK -#define blk_to_logical(inode, blk) (blk << (inode)->i_blkbits) -#define logical_to_blk(inode, offset) (offset >> (inode)->i_blkbits); +static inline sector_t logical_to_blk(struct inode *inode, loff_t offset) +{ + return (offset >> inode->i_blkbits); +} + +static inline loff_t blk_to_logical(struct inode *inode, sector_t blk) +{ + return (blk << inode->i_blkbits); +} /** * __generic_block_fiemap - FIEMAP for block based inodes (no locking) - * @inode - the inode to map - * @arg - the pointer to userspace where we copy everything to - * @get_block - the fs's get_block function + * @inode: the inode to map + * @fieinfo: the fiemap info struct that will be passed back to userspace + * @start: where to start mapping in the inode + * @len: how much space to map + * @get_block: the fs's get_block function * * This does FIEMAP for block based inodes. Basically it will just loop * through get_block until we hit the number of extents we want to map, or we @@ -250,58 +259,63 @@ static int ioctl_fiemap(struct file *filp, unsigned long arg) */ int __generic_block_fiemap(struct inode *inode, - struct fiemap_extent_info *fieinfo, u64 start, - u64 len, get_block_t *get_block) + struct fiemap_extent_info *fieinfo, loff_t start, + loff_t len, get_block_t *get_block) { - struct buffer_head tmp; - unsigned long long start_blk; - long long length = 0, map_len = 0; + struct buffer_head map_bh; + sector_t start_blk, last_blk; + loff_t isize = i_size_read(inode); u64 logical = 0, phys = 0, size = 0; u32 flags = FIEMAP_EXTENT_MERGED; - int ret = 0, past_eof = 0, whole_file = 0; + bool past_eof = false, whole_file = false; + int ret = 0; - if ((ret = fiemap_check_flags(fieinfo, FIEMAP_FLAG_SYNC))) + ret = fiemap_check_flags(fieinfo, FIEMAP_FLAG_SYNC); + if (ret) return ret; - start_blk = logical_to_blk(inode, start); - - length = (long long)min_t(u64, len, i_size_read(inode)); - if (length < len) - whole_file = 1; + /* + * Either the i_mutex or other appropriate locking needs to be held + * since we expect isize to not change at all through the duration of + * this call. + */ + if (len >= isize) { + whole_file = true; + len = isize; + } - map_len = length; + start_blk = logical_to_blk(inode, start); + last_blk = logical_to_blk(inode, start + len - 1); do { /* * we set b_size to the total size we want so it will map as * many contiguous blocks as possible at once */ - memset(&tmp, 0, sizeof(struct buffer_head)); - tmp.b_size = map_len; + memset(&map_bh, 0, sizeof(struct buffer_head)); + map_bh.b_size = len; - ret = get_block(inode, start_blk, &tmp, 0); + ret = get_block(inode, start_blk, &map_bh, 0); if (ret) break; /* HOLE */ - if (!buffer_mapped(&tmp)) { - length -= blk_to_logical(inode, 1); + if (!buffer_mapped(&map_bh)) { start_blk++; /* - * we want to handle the case where there is an + * We want to handle the case where there is an * allocated block at the front of the file, and then * nothing but holes up to the end of the file properly, * to make sure that extent at the front gets properly * marked with FIEMAP_EXTENT_LAST */ if (!past_eof && - blk_to_logical(inode, start_blk) >= - blk_to_logical(inode, 0)+i_size_read(inode)) + blk_to_logical(inode, start_blk) >= isize) past_eof = 1; /* - * first hole after going past the EOF, this is our + * First hole after going past the EOF, this is our * last extent */ if (past_eof && size) { @@ -309,15 +323,18 @@ int __generic_block_fiemap(struct inode *inode, ret = fiemap_fill_next_extent(fieinfo, logical, phys, size, flags); - break; + } else if (size) { + ret = fiemap_fill_next_extent(fieinfo, logical, + phys, size, flags); + size = 0; } /* if we have holes up to/past EOF then we're done */ - if (length <= 0 || past_eof) + if (start_blk > last_blk || past_eof || ret) break; } else { /* - * we have gone over the length of what we wanted to + * We have gone over the length of what we wanted to * map, and it wasn't the entire file, so add the extent * we got last time and exit. * @@ -331,7 +348,7 @@ int __generic_block_fiemap(struct inode *inode, * are good to go, just add the extent to the fieinfo * and break */ - if (length <= 0 && !whole_file) { + if (start_blk > last_blk && !whole_file) { ret = fiemap_fill_next_extent(fieinfo, logical, phys, size, flags); @@ -351,11 +368,10 @@ int __generic_block_fiemap(struct inode *inode, } logical = blk_to_logical(inode, start_blk); - phys = blk_to_logical(inode, tmp.b_blocknr); - size = tmp.b_size; + phys = blk_to_logical(inode, map_bh.b_blocknr); + size = map_bh.b_size; flags = FIEMAP_EXTENT_MERGED; - length -= tmp.b_size; start_blk += logical_to_blk(inode, size); /* @@ -363,15 +379,13 @@ int __generic_block_fiemap(struct inode *inode, * soon as we find a hole that the last extent we found * is marked with FIEMAP_EXTENT_LAST */ - if (!past_eof && - logical+size >= - blk_to_logical(inode, 0)+i_size_read(inode)) - past_eof = 1; + if (!past_eof && logical + size >= isize) + past_eof = true; } cond_resched(); } while (1); - /* if ret is 1 then we just hit the end of the extent array */ + /* If ret is 1 then we just hit the end of the extent array */ if (ret == 1) ret = 0; diff --git a/include/linux/fs.h b/include/linux/fs.h index 39d57bc6cc71..44f35aea2f1f 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2315,8 +2315,9 @@ extern int vfs_fstatat(int , char __user *, struct kstat *, int); extern int do_vfs_ioctl(struct file *filp, unsigned int fd, unsigned int cmd, unsigned long arg); extern int __generic_block_fiemap(struct inode *inode, - struct fiemap_extent_info *fieinfo, u64 start, - u64 len, get_block_t *get_block); + struct fiemap_extent_info *fieinfo, + loff_t start, loff_t len, + get_block_t *get_block); extern int generic_block_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, u64 start, u64 len, get_block_t *get_block); -- cgit v1.2.3 From 23be7468e8802a2ac1de6ee3eecb3ec7f14dc703 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Fri, 23 Apr 2010 13:17:56 -0400 Subject: hugetlb: fix infinite loop in get_futex_key() when backed by huge pages If a futex key happens to be located within a huge page mapped MAP_PRIVATE, get_futex_key() can go into an infinite loop waiting for a page->mapping that will never exist. See https://bugzilla.redhat.com/show_bug.cgi?id=552257 for more details about the problem. This patch makes page->mapping a poisoned value that includes PAGE_MAPPING_ANON mapped MAP_PRIVATE. This is enough for futex to continue but because of PAGE_MAPPING_ANON, the poisoned value is not dereferenced or used by futex. No other part of the VM should be dereferencing the page->mapping of a hugetlbfs page as its page cache is not on the LRU. This patch fixes the problem with the test case described in the bugzilla. [akpm@linux-foundation.org: mel cant spel] Signed-off-by: Mel Gorman Acked-by: Peter Zijlstra Acked-by: Darren Hart Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/poison.h | 9 +++++++++ mm/hugetlb.c | 5 ++++- 2 files changed, 13 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/poison.h b/include/linux/poison.h index 2110a81c5e2a..34066ffd893d 100644 --- a/include/linux/poison.h +++ b/include/linux/poison.h @@ -48,6 +48,15 @@ #define POISON_FREE 0x6b /* for use-after-free poisoning */ #define POISON_END 0xa5 /* end-byte of poisoning */ +/********** mm/hugetlb.c **********/ +/* + * Private mappings of hugetlb pages use this poisoned value for + * page->mapping. The core VM should not be doing anything with this mapping + * but futex requires the existence of some page->mapping value even though it + * is unused if PAGE_MAPPING_ANON is set. + */ +#define HUGETLB_POISON ((void *)(0x00300300 + POISON_POINTER_DELTA + PAGE_MAPPING_ANON)) + /********** arch/$ARCH/mm/init.c **********/ #define POISON_FREE_INITMEM 0xcc diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 6034dc9e9796..ffbdfc86aedf 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -546,6 +546,7 @@ static void free_huge_page(struct page *page) mapping = (struct address_space *) page_private(page); set_page_private(page, 0); + page->mapping = NULL; BUG_ON(page_count(page)); INIT_LIST_HEAD(&page->lru); @@ -2447,8 +2448,10 @@ retry: spin_lock(&inode->i_lock); inode->i_blocks += blocks_per_huge_page(h); spin_unlock(&inode->i_lock); - } else + } else { lock_page(page); + page->mapping = HUGETLB_POISON; + } } /* -- cgit v1.2.3 From 5129a469a91a91427334c40e29e64c6d0ab68caf Mon Sep 17 00:00:00 2001 From: Jörn Engel Date: Sun, 25 Apr 2010 08:54:42 +0200 Subject: Catch filesystems lacking s_bdi noop_backing_dev_info is used only as a flag to mark filesystems that don't have any backing store, like tmpfs, procfs, spufs, etc. Signed-off-by: Joern Engel Changed the BUG_ON() to a WARN_ON(). Note that adding dirty inodes to the noop_backing_dev_info is not legal and will not result in them being flushed, but we already catch this condition in __mark_inode_dirty() when checking for a registered bdi. Signed-off-by: Jens Axboe --- fs/super.c | 8 +++++--- fs/sync.c | 3 ++- include/linux/backing-dev.h | 1 + mm/backing-dev.c | 5 +++++ 4 files changed, 13 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/fs/super.c b/fs/super.c index f35ac6022109..dc72491a19f9 100644 --- a/fs/super.c +++ b/fs/super.c @@ -693,6 +693,7 @@ int set_anon_super(struct super_block *s, void *data) return -EMFILE; } s->s_dev = MKDEV(0, dev & MINORMASK); + s->s_bdi = &noop_backing_dev_info; return 0; } @@ -954,10 +955,11 @@ vfs_kern_mount(struct file_system_type *type, int flags, const char *name, void if (error < 0) goto out_free_secdata; BUG_ON(!mnt->mnt_sb); + WARN_ON(!mnt->mnt_sb->s_bdi); - error = security_sb_kern_mount(mnt->mnt_sb, flags, secdata); - if (error) - goto out_sb; + error = security_sb_kern_mount(mnt->mnt_sb, flags, secdata); + if (error) + goto out_sb; /* * filesystems should never set s_maxbytes larger than MAX_LFS_FILESIZE diff --git a/fs/sync.c b/fs/sync.c index fc5c3d75cf3c..92b228176f7c 100644 --- a/fs/sync.c +++ b/fs/sync.c @@ -14,6 +14,7 @@ #include #include #include +#include #include "internal.h" #define VALID_FLAGS (SYNC_FILE_RANGE_WAIT_BEFORE|SYNC_FILE_RANGE_WRITE| \ @@ -32,7 +33,7 @@ static int __sync_filesystem(struct super_block *sb, int wait) * This should be safe, as we require bdi backing to actually * write out data in the first place */ - if (!sb->s_bdi) + if (!sb->s_bdi || sb->s_bdi == &noop_backing_dev_info) return 0; if (sb->s_qcop && sb->s_qcop->quota_sync) diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index e19c677f219c..bd0e3c6f323f 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -247,6 +247,7 @@ int bdi_set_max_ratio(struct backing_dev_info *bdi, unsigned int max_ratio); #endif extern struct backing_dev_info default_backing_dev_info; +extern struct backing_dev_info noop_backing_dev_info; void default_unplug_io_fn(struct backing_dev_info *bdi, struct page *page); int writeback_in_progress(struct backing_dev_info *bdi); diff --git a/mm/backing-dev.c b/mm/backing-dev.c index dbda4707f593..707d0dc6da0f 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -27,6 +27,11 @@ struct backing_dev_info default_backing_dev_info = { }; EXPORT_SYMBOL_GPL(default_backing_dev_info); +struct backing_dev_info noop_backing_dev_info = { + .name = "noop", +}; +EXPORT_SYMBOL_GPL(noop_backing_dev_info); + static struct class *bdi_class; /* -- cgit v1.2.3 From 33f60e9640b2f60dde6735293d4aa5ecc5b1d5d5 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Wed, 28 Apr 2010 09:20:33 +0200 Subject: coda: move backing-dev.h kernel include inside __KERNEL__ Otherwise we must export backing-dev.h as well, which doesn't make any sense. Signed-off-by: Jens Axboe --- include/linux/coda_psdev.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/coda_psdev.h b/include/linux/coda_psdev.h index 644062e8d857..8859e2ede9fe 100644 --- a/include/linux/coda_psdev.h +++ b/include/linux/coda_psdev.h @@ -1,13 +1,14 @@ #ifndef __CODA_PSDEV_H #define __CODA_PSDEV_H -#include #include #define CODA_PSDEV_MAJOR 67 #define MAX_CODADEVS 5 /* how many do we allow */ #ifdef __KERNEL__ +#include + struct kstatfs; /* communication pending/processing queues */ -- cgit v1.2.3 From 073900a28d95c75a706bf40ebf092ea048c7b236 Mon Sep 17 00:00:00 2001 From: Daniel Mack Date: Mon, 12 Apr 2010 13:17:25 +0200 Subject: USB: rename usb_buffer_alloc() and usb_buffer_free() For more clearance what the functions actually do, usb_buffer_alloc() is renamed to usb_alloc_coherent() usb_buffer_free() is renamed to usb_free_coherent() They should only be used in code which really needs DMA coherency. [added compatibility macros so we can convert things easier - gregkh] Signed-off-by: Daniel Mack Cc: Alan Stern Cc: Pedro Ribeiro Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/usb.c | 20 ++++++++++---------- include/linux/usb.h | 18 +++++++++++++++--- 2 files changed, 25 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c index 1297e9b16a51..0561430f2ede 100644 --- a/drivers/usb/core/usb.c +++ b/drivers/usb/core/usb.c @@ -718,7 +718,7 @@ int __usb_get_extra_descriptor(char *buffer, unsigned size, EXPORT_SYMBOL_GPL(__usb_get_extra_descriptor); /** - * usb_buffer_alloc - allocate dma-consistent buffer for URB_NO_xxx_DMA_MAP + * usb_alloc_coherent - allocate dma-consistent buffer for URB_NO_xxx_DMA_MAP * @dev: device the buffer will be used with * @size: requested buffer size * @mem_flags: affect whether allocation may block @@ -737,30 +737,30 @@ EXPORT_SYMBOL_GPL(__usb_get_extra_descriptor); * architectures where CPU caches are not DMA-coherent. On systems without * bus-snooping caches, these buffers are uncached. * - * When the buffer is no longer used, free it with usb_buffer_free(). + * When the buffer is no longer used, free it with usb_free_coherent(). */ -void *usb_buffer_alloc(struct usb_device *dev, size_t size, gfp_t mem_flags, - dma_addr_t *dma) +void *usb_alloc_coherent(struct usb_device *dev, size_t size, gfp_t mem_flags, + dma_addr_t *dma) { if (!dev || !dev->bus) return NULL; return hcd_buffer_alloc(dev->bus, size, mem_flags, dma); } -EXPORT_SYMBOL_GPL(usb_buffer_alloc); +EXPORT_SYMBOL_GPL(usb_alloc_coherent); /** - * usb_buffer_free - free memory allocated with usb_buffer_alloc() + * usb_free_coherent - free memory allocated with usb_alloc_coherent() * @dev: device the buffer was used with * @size: requested buffer size * @addr: CPU address of buffer * @dma: DMA address of buffer * * This reclaims an I/O buffer, letting it be reused. The memory must have - * been allocated using usb_buffer_alloc(), and the parameters must match + * been allocated using usb_alloc_coherent(), and the parameters must match * those provided in that allocation request. */ -void usb_buffer_free(struct usb_device *dev, size_t size, void *addr, - dma_addr_t dma) +void usb_free_coherent(struct usb_device *dev, size_t size, void *addr, + dma_addr_t dma) { if (!dev || !dev->bus) return; @@ -768,7 +768,7 @@ void usb_buffer_free(struct usb_device *dev, size_t size, void *addr, return; hcd_buffer_free(dev->bus, size, addr, dma); } -EXPORT_SYMBOL_GPL(usb_buffer_free); +EXPORT_SYMBOL_GPL(usb_free_coherent); /** * usb_buffer_map - create DMA mapping(s) for an urb diff --git a/include/linux/usb.h b/include/linux/usb.h index ce1323c4e47c..739f1fd1cc15 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -1085,7 +1085,7 @@ typedef void (*usb_complete_t)(struct urb *); * Alternatively, drivers may pass the URB_NO_xxx_DMA_MAP transfer flags, * which tell the host controller driver that no such mapping is needed since * the device driver is DMA-aware. For example, a device driver might - * allocate a DMA buffer with usb_buffer_alloc() or call usb_buffer_map(). + * allocate a DMA buffer with usb_alloc_coherent() or call usb_buffer_map(). * When these transfer flags are provided, host controller drivers will * attempt to use the dma addresses found in the transfer_dma and/or * setup_dma fields rather than determining a dma address themselves. @@ -1366,11 +1366,23 @@ static inline int usb_urb_dir_out(struct urb *urb) return (urb->transfer_flags & URB_DIR_MASK) == URB_DIR_OUT; } -void *usb_buffer_alloc(struct usb_device *dev, size_t size, +void *usb_alloc_coherent(struct usb_device *dev, size_t size, gfp_t mem_flags, dma_addr_t *dma); -void usb_buffer_free(struct usb_device *dev, size_t size, +void usb_free_coherent(struct usb_device *dev, size_t size, void *addr, dma_addr_t dma); +/* Compatible macros while we switch over */ +static inline void *usb_buffer_alloc(struct usb_device *dev, size_t size, + gfp_t mem_flags, dma_addr_t *dma) +{ + return usb_alloc_coherent(dev, size, mem_flags, dma); +} +static inline void usb_buffer_free(struct usb_device *dev, size_t size, + void *addr, dma_addr_t dma) +{ + return usb_free_coherent(dev, size, addr, dma); +} + #if 0 struct urb *usb_buffer_map(struct urb *urb); void usb_buffer_dmasync(struct urb *urb); -- cgit v1.2.3 From 6629dcff19470a894ce294d0adb9cbab94ee1fb9 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Tue, 4 May 2010 11:09:28 +0200 Subject: i2c-core: Use per-adapter userspace device lists Using a single list for all userspace devices leads to a dead lock on multiplexed buses in some circumstances (mux chip instantiated from userspace). This is solved by using a separate list for each bus segment. Signed-off-by: Jean Delvare Acked-by: Michael Lawnick --- drivers/i2c/i2c-core.c | 34 ++++++++++++++++++---------------- include/linux/i2c.h | 2 ++ 2 files changed, 20 insertions(+), 16 deletions(-) (limited to 'include/linux') diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c index 5105126225c3..c2258a51fe0c 100644 --- a/drivers/i2c/i2c-core.c +++ b/drivers/i2c/i2c-core.c @@ -40,12 +40,11 @@ #include "i2c-core.h" -/* core_lock protects i2c_adapter_idr, userspace_devices, and guarantees +/* core_lock protects i2c_adapter_idr, and guarantees that device detection, deletion of detected devices, and attach_adapter and detach_adapter calls are serialized */ static DEFINE_MUTEX(core_lock); static DEFINE_IDR(i2c_adapter_idr); -static LIST_HEAD(userspace_devices); static struct device_type i2c_client_type; static int i2c_check_addr(struct i2c_adapter *adapter, int addr); @@ -542,9 +541,9 @@ i2c_sysfs_new_device(struct device *dev, struct device_attribute *attr, return -EEXIST; /* Keep track of the added device */ - mutex_lock(&core_lock); - list_add_tail(&client->detected, &userspace_devices); - mutex_unlock(&core_lock); + i2c_lock_adapter(adap); + list_add_tail(&client->detected, &adap->userspace_clients); + i2c_unlock_adapter(adap); dev_info(dev, "%s: Instantiated device %s at 0x%02hx\n", "new_device", info.type, info.addr); @@ -583,9 +582,10 @@ i2c_sysfs_delete_device(struct device *dev, struct device_attribute *attr, /* Make sure the device was added through sysfs */ res = -ENOENT; - mutex_lock(&core_lock); - list_for_each_entry_safe(client, next, &userspace_devices, detected) { - if (client->addr == addr && client->adapter == adap) { + i2c_lock_adapter(adap); + list_for_each_entry_safe(client, next, &adap->userspace_clients, + detected) { + if (client->addr == addr) { dev_info(dev, "%s: Deleting device %s at 0x%02hx\n", "delete_device", client->name, client->addr); @@ -595,7 +595,7 @@ i2c_sysfs_delete_device(struct device *dev, struct device_attribute *attr, break; } } - mutex_unlock(&core_lock); + i2c_unlock_adapter(adap); if (res < 0) dev_err(dev, "%s: Can't find device in list\n", @@ -677,6 +677,7 @@ static int i2c_register_adapter(struct i2c_adapter *adap) } rt_mutex_init(&adap->bus_lock); + INIT_LIST_HEAD(&adap->userspace_clients); /* Set default timeout to 1 second if not already set */ if (adap->timeout == 0) @@ -879,14 +880,15 @@ int i2c_del_adapter(struct i2c_adapter *adap) return res; /* Remove devices instantiated from sysfs */ - list_for_each_entry_safe(client, next, &userspace_devices, detected) { - if (client->adapter == adap) { - dev_dbg(&adap->dev, "Removing %s at 0x%x\n", - client->name, client->addr); - list_del(&client->detected); - i2c_unregister_device(client); - } + i2c_lock_adapter(adap); + list_for_each_entry_safe(client, next, &adap->userspace_clients, + detected) { + dev_dbg(&adap->dev, "Removing %s at 0x%x\n", client->name, + client->addr); + list_del(&client->detected); + i2c_unregister_device(client); } + i2c_unlock_adapter(adap); /* Detach any active clients. This can't fail, thus we do not checking the returned value. */ diff --git a/include/linux/i2c.h b/include/linux/i2c.h index 0a5da639b327..6ed1d59bfb1e 100644 --- a/include/linux/i2c.h +++ b/include/linux/i2c.h @@ -355,6 +355,8 @@ struct i2c_adapter { int nr; char name[48]; struct completion dev_released; + + struct list_head userspace_clients; }; #define to_i2c_adapter(d) container_of(d, struct i2c_adapter, dev) -- cgit v1.2.3