From 8aa7e847d834ed937a9ad37a0f2ad5b8584c1ab0 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 9 Jul 2009 14:52:32 +0200 Subject: Fix congestion_wait() sync/async vs read/write confusion Commit 1faa16d22877f4839bd433547d770c676d1d964c accidentally broke the bdi congestion wait queue logic, causing us to wait on congestion for WRITE (== 1) when we really wanted BLK_RW_ASYNC (== 0) instead. Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 49ae07951d5..bb3d3997870 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -779,18 +779,18 @@ extern int sg_scsi_ioctl(struct request_queue *, struct gendisk *, fmode_t, * congested queues, and wake up anyone who was waiting for requests to be * put back. */ -static inline void blk_clear_queue_congested(struct request_queue *q, int rw) +static inline void blk_clear_queue_congested(struct request_queue *q, int sync) { - clear_bdi_congested(&q->backing_dev_info, rw); + clear_bdi_congested(&q->backing_dev_info, sync); } /* * A queue has just entered congestion. Flag that in the queue's VM-visible * state flags and increment the global gounter of congested queues. */ -static inline void blk_set_queue_congested(struct request_queue *q, int rw) +static inline void blk_set_queue_congested(struct request_queue *q, int sync) { - set_bdi_congested(&q->backing_dev_info, rw); + set_bdi_congested(&q->backing_dev_info, sync); } extern void blk_start_queue(struct request_queue *q); -- cgit v1.2.3-70-g09d2 From ecb554a846f8e9d2a58f6d6c118168a63ac065aa Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Thu, 9 Jul 2009 14:46:53 +0200 Subject: block: fix sg SG_DXFER_TO_FROM_DEV regression I overlooked SG_DXFER_TO_FROM_DEV support when I converted sg to use the block layer mapping API (2.6.28). Douglas Gilbert explained SG_DXFER_TO_FROM_DEV: http://www.spinics.net/lists/linux-scsi/msg37135.html = The semantics of SG_DXFER_TO_FROM_DEV were: - copy user space buffer to kernel (LLD) buffer - do SCSI command which is assumed to be of the DATA_IN (data from device) variety. This would overwrite some or all of the kernel buffer - copy kernel (LLD) buffer back to the user space. The idea was to detect short reads by filling the original user space buffer with some marker bytes ("0xec" it would seem in this report). The "resid" value is a better way of detecting short reads but that was only added this century and requires co-operation from the LLD. = This patch changes the block layer mapping API to support this semantics. This simply adds another field to struct rq_map_data and enables __bio_copy_iov() to copy data from user space even with READ requests. It's better to add the flags field and kills null_mapped and the new from_user fields in struct rq_map_data but that approach makes it difficult to send this patch to stable trees because st and osst drivers use struct rq_map_data (they were converted to use the block layer in 2.6.29 and 2.6.30). Well, I should clean up the block layer mapping API. zhou sf reported this regiression and tested this patch: http://www.spinics.net/lists/linux-scsi/msg37128.html http://www.spinics.net/lists/linux-scsi/msg37168.html Reported-by: zhou sf Tested-by: zhou sf Cc: stable@kernel.org Signed-off-by: FUJITA Tomonori Signed-off-by: Jens Axboe --- drivers/scsi/sg.c | 4 ++++ fs/bio.c | 22 ++++++++++++---------- include/linux/blkdev.h | 1 + 3 files changed, 17 insertions(+), 10 deletions(-) (limited to 'include/linux/blkdev.h') diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c index 4d6f2fe1cfe..9230402c45a 100644 --- a/drivers/scsi/sg.c +++ b/drivers/scsi/sg.c @@ -1656,6 +1656,10 @@ static int sg_start_req(Sg_request *srp, unsigned char *cmd) md->nr_entries = req_schp->k_use_sg; md->offset = 0; md->null_mapped = hp->dxferp ? 0 : 1; + if (dxfer_dir == SG_DXFER_TO_FROM_DEV) + md->from_user = 1; + else + md->from_user = 0; } if (iov_count) { diff --git a/fs/bio.c b/fs/bio.c index 1486b19fc43..76738005c8e 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -705,14 +705,13 @@ static struct bio_map_data *bio_alloc_map_data(int nr_segs, int iov_count, } static int __bio_copy_iov(struct bio *bio, struct bio_vec *iovecs, - struct sg_iovec *iov, int iov_count, int uncopy, - int do_free_page) + struct sg_iovec *iov, int iov_count, + int to_user, int from_user, int do_free_page) { int ret = 0, i; struct bio_vec *bvec; int iov_idx = 0; unsigned int iov_off = 0; - int read = bio_data_dir(bio) == READ; __bio_for_each_segment(bvec, bio, i, 0) { char *bv_addr = page_address(bvec->bv_page); @@ -727,13 +726,14 @@ static int __bio_copy_iov(struct bio *bio, struct bio_vec *iovecs, iov_addr = iov[iov_idx].iov_base + iov_off; if (!ret) { - if (!read && !uncopy) - ret = copy_from_user(bv_addr, iov_addr, - bytes); - if (read && uncopy) + if (to_user) ret = copy_to_user(iov_addr, bv_addr, bytes); + if (from_user) + ret = copy_from_user(bv_addr, iov_addr, + bytes); + if (ret) ret = -EFAULT; } @@ -770,7 +770,8 @@ int bio_uncopy_user(struct bio *bio) if (!bio_flagged(bio, BIO_NULL_MAPPED)) ret = __bio_copy_iov(bio, bmd->iovecs, bmd->sgvecs, - bmd->nr_sgvecs, 1, bmd->is_our_pages); + bmd->nr_sgvecs, bio_data_dir(bio) == READ, + 0, bmd->is_our_pages); bio_free_map_data(bmd); bio_put(bio); return ret; @@ -875,8 +876,9 @@ struct bio *bio_copy_user_iov(struct request_queue *q, /* * success */ - if (!write_to_vm && (!map_data || !map_data->null_mapped)) { - ret = __bio_copy_iov(bio, bio->bi_io_vec, iov, iov_count, 0, 0); + if ((!write_to_vm && (!map_data || !map_data->null_mapped)) || + (map_data && map_data->from_user)) { + ret = __bio_copy_iov(bio, bio->bi_io_vec, iov, iov_count, 0, 1, 0); if (ret) goto cleanup; } diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index bb3d3997870..0146e0fecf1 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -723,6 +723,7 @@ struct rq_map_data { int nr_entries; unsigned long offset; int null_mapped; + int from_user; }; struct req_iterator { -- cgit v1.2.3-70-g09d2 From 373c0a7ed3ea3b34efedb7c83ffb521adff7c894 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sat, 11 Jul 2009 10:06:54 -0400 Subject: Fix compile error due to congestion_wait() changes Move the definition of BLK_RW_ASYNC/BLK_RW_SYNC into linux/backing-dev.h so that it is available to all callers of set/clear_bdi_congested(). This replaces commit 097041e576ee3a50d92dd643ee8ca65bf6a62e21 ("fuse: Fix build error"), which will be reverted. Signed-off-by: Trond Myklebust Acked-by: Larry Finger Cc: Jens Axboe Cc: Miklos Szeredi Signed-off-by: Linus Torvalds --- include/linux/backing-dev.h | 5 +++++ include/linux/blkdev.h | 5 ----- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index 3a52a63c135..1d52425a611 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -229,6 +229,11 @@ static inline int bdi_rw_congested(struct backing_dev_info *bdi) (1 << BDI_async_congested)); } +enum { + BLK_RW_ASYNC = 0, + BLK_RW_SYNC = 1, +}; + void clear_bdi_congested(struct backing_dev_info *bdi, int sync); void set_bdi_congested(struct backing_dev_info *bdi, int sync); long congestion_wait(int sync, long timeout); diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 0146e0fecf1..e7cb5dbf6c2 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -70,11 +70,6 @@ enum rq_cmd_type_bits { REQ_TYPE_ATA_PC, }; -enum { - BLK_RW_ASYNC = 0, - BLK_RW_SYNC = 1, -}; - /* * For request of type REQ_TYPE_LINUX_BLOCK, rq->cmd[0] is the opcode being * sent down (similar to how REQ_TYPE_BLOCK_PC means that ->cmd[] holds a -- cgit v1.2.3-70-g09d2 From 7c958e32649e0c35801762878fb0b6da8c55a515 Mon Sep 17 00:00:00 2001 From: "Martin K. Petersen" Date: Fri, 31 Jul 2009 11:49:11 -0400 Subject: block: Add a wrapper for setting minimum request size without a queue Introduce blk_limits_io_min() and make blk_queue_io_min() call it. Signed-off-by: Mike Snitzer Signed-off-by: Martin K. Petersen Signed-off-by: Jens Axboe --- block/blk-settings.c | 31 ++++++++++++++++++++++++------- include/linux/blkdev.h | 1 + 2 files changed, 25 insertions(+), 7 deletions(-) (limited to 'include/linux/blkdev.h') diff --git a/block/blk-settings.c b/block/blk-settings.c index 8e86e2d2b14..1f719743416 100644 --- a/block/blk-settings.c +++ b/block/blk-settings.c @@ -383,6 +383,29 @@ void blk_queue_alignment_offset(struct request_queue *q, unsigned int offset) } EXPORT_SYMBOL(blk_queue_alignment_offset); +/** + * blk_limits_io_min - set minimum request size for a device + * @limits: the queue limits + * @min: smallest I/O size in bytes + * + * Description: + * Some devices have an internal block size bigger than the reported + * hardware sector size. This function can be used to signal the + * smallest I/O the device can perform without incurring a performance + * penalty. + */ +void blk_limits_io_min(struct queue_limits *limits, unsigned int min) +{ + limits->io_min = min; + + if (limits->io_min < limits->logical_block_size) + limits->io_min = limits->logical_block_size; + + if (limits->io_min < limits->physical_block_size) + limits->io_min = limits->physical_block_size; +} +EXPORT_SYMBOL(blk_limits_io_min); + /** * blk_queue_io_min - set minimum request size for the queue * @q: the request queue for the device @@ -396,13 +419,7 @@ EXPORT_SYMBOL(blk_queue_alignment_offset); */ void blk_queue_io_min(struct request_queue *q, unsigned int min) { - q->limits.io_min = min; - - if (q->limits.io_min < q->limits.logical_block_size) - q->limits.io_min = q->limits.logical_block_size; - - if (q->limits.io_min < q->limits.physical_block_size) - q->limits.io_min = q->limits.physical_block_size; + blk_limits_io_min(&q->limits, min); } EXPORT_SYMBOL(blk_queue_io_min); diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index e7cb5dbf6c2..69103e053c9 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -913,6 +913,7 @@ extern void blk_queue_logical_block_size(struct request_queue *, unsigned short) extern void blk_queue_physical_block_size(struct request_queue *, unsigned short); extern void blk_queue_alignment_offset(struct request_queue *q, unsigned int alignment); +extern void blk_limits_io_min(struct queue_limits *limits, unsigned int min); extern void blk_queue_io_min(struct request_queue *q, unsigned int min); extern void blk_queue_io_opt(struct request_queue *q, unsigned int opt); extern void blk_set_default_limits(struct queue_limits *lim); -- cgit v1.2.3-70-g09d2