The patch titled block: don't take lock while changing ra_pages has been added to the -mm tree. Its filename is block-dont-take-lock-while-changing-ra_pages.patch Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/SubmitChecklist when testing your code *** See http://userweb.kernel.org/~akpm/stuff/added-to-mm.txt to find out what to do about this The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/ ------------------------------------------------------ Subject: block: don't take lock while changing ra_pages From: Wu Fengguang <fengguang.wu@xxxxxxxxx> There's no need to take queue_lock or kernel_lock when modifying bdi->ra_pages. So remove them. Also remove out of date comment for queue_max_sectors_store(). Signed-off-by: Wu Fengguang <wfg@xxxxxxxxxxxxxxx> Cc: Jens Axboe <jens.axboe@xxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- block/blk-sysfs.c | 7 +------ block/compat_ioctl.c | 2 -- block/ioctl.c | 2 -- 3 files changed, 1 insertion(+), 10 deletions(-) diff -puN block/blk-sysfs.c~block-dont-take-lock-while-changing-ra_pages block/blk-sysfs.c --- a/block/blk-sysfs.c~block-dont-take-lock-while-changing-ra_pages +++ a/block/blk-sysfs.c @@ -88,9 +88,7 @@ queue_ra_store(struct request_queue *q, unsigned long ra_kb; ssize_t ret = queue_var_store(&ra_kb, page, count); - spin_lock_irq(q->queue_lock); q->backing_dev_info.ra_pages = ra_kb >> (PAGE_CACHE_SHIFT - 10); - spin_unlock_irq(q->queue_lock); return ret; } @@ -117,10 +115,7 @@ queue_max_sectors_store(struct request_q if (max_sectors_kb > max_hw_sectors_kb || max_sectors_kb < page_kb) return -EINVAL; - /* - * Take the queue lock to update the readahead and max_sectors - * values synchronously: - */ + spin_lock_irq(q->queue_lock); q->max_sectors = max_sectors_kb << 1; spin_unlock_irq(q->queue_lock); diff -puN block/compat_ioctl.c~block-dont-take-lock-while-changing-ra_pages block/compat_ioctl.c --- a/block/compat_ioctl.c~block-dont-take-lock-while-changing-ra_pages +++ a/block/compat_ioctl.c @@ -745,9 +745,7 @@ long compat_blkdev_ioctl(struct file *fi bdi = blk_get_backing_dev_info(bdev); if (bdi == NULL) return -ENOTTY; - lock_kernel(); bdi->ra_pages = (arg * 512) / PAGE_CACHE_SIZE; - unlock_kernel(); return 0; case BLKGETSIZE: size = bdev->bd_inode->i_size; diff -puN block/ioctl.c~block-dont-take-lock-while-changing-ra_pages block/ioctl.c --- a/block/ioctl.c~block-dont-take-lock-while-changing-ra_pages +++ a/block/ioctl.c @@ -323,9 +323,7 @@ int blkdev_ioctl(struct block_device *bd bdi = blk_get_backing_dev_info(bdev); if (bdi == NULL) return -ENOTTY; - lock_kernel(); bdi->ra_pages = (arg * 512) / PAGE_CACHE_SIZE; - unlock_kernel(); return 0; case BLKBSZSET: /* set the logical block size */ _ Patches currently in -mm which might be from fengguang.wu@xxxxxxxxx are linux-next.patch block-dont-take-lock-while-changing-ra_pages.patch -- To unsubscribe from this list: send the line "unsubscribe mm-commits" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html