From df46b9a44ceb5af2ea2351ce8e28ae7bd840b00f Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Mon, 20 Jun 2005 14:04:44 +0200 Subject: [PATCH] Add blk_rq_map_kern() Add blk_rq_map_kern which takes a kernel buffer and maps it into a request and bio. This can be used by the dm hw_handlers, old sg_scsi_ioctl, and one day scsi special requests so all requests comming into scsi will have bios. All requests having bios should allow scsi to use scatter lists for all IO and allow it to use block layer functions. Signed-off-by: Jens Axboe --- fs/bio.c | 66 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 66 insertions(+) (limited to 'fs/bio.c') diff --git a/fs/bio.c b/fs/bio.c index 3a1472acc361..707b9af2dd01 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -701,6 +701,71 @@ void bio_unmap_user(struct bio *bio) bio_put(bio); } +static struct bio *__bio_map_kern(request_queue_t *q, void *data, + unsigned int len, unsigned int gfp_mask) +{ + unsigned long kaddr = (unsigned long)data; + unsigned long end = (kaddr + len + PAGE_SIZE - 1) >> PAGE_SHIFT; + unsigned long start = kaddr >> PAGE_SHIFT; + const int nr_pages = end - start; + int offset, i; + struct bio *bio; + + bio = bio_alloc(gfp_mask, nr_pages); + if (!bio) + return ERR_PTR(-ENOMEM); + + offset = offset_in_page(kaddr); + for (i = 0; i < nr_pages; i++) { + unsigned int bytes = PAGE_SIZE - offset; + + if (len <= 0) + break; + + if (bytes > len) + bytes = len; + + if (__bio_add_page(q, bio, virt_to_page(data), bytes, + offset) < bytes) + break; + + data += bytes; + len -= bytes; + offset = 0; + } + + return bio; +} + +/** + * bio_map_kern - map kernel address into bio + * @q: the request_queue_t for the bio + * @data: pointer to buffer to map + * @len: length in bytes + * @gfp_mask: allocation flags for bio allocation + * + * Map the kernel address into a bio suitable for io to a block + * device. Returns an error pointer in case of error. + */ +struct bio *bio_map_kern(request_queue_t *q, void *data, unsigned int len, + unsigned int gfp_mask) +{ + struct bio *bio; + + bio = __bio_map_kern(q, data, len, gfp_mask); + if (IS_ERR(bio)) + return bio; + + if (bio->bi_size == len) + return bio; + + /* + * Don't support partial mappings. + */ + bio_put(bio); + return ERR_PTR(-EINVAL); +} + /* * bio_set_pages_dirty() and bio_check_pages_dirty() are support functions * for performing direct-IO in BIOs. @@ -1088,6 +1153,7 @@ EXPORT_SYMBOL(bio_add_page); EXPORT_SYMBOL(bio_get_nr_vecs); EXPORT_SYMBOL(bio_map_user); EXPORT_SYMBOL(bio_unmap_user); +EXPORT_SYMBOL(bio_map_kern); EXPORT_SYMBOL(bio_pair_release); EXPORT_SYMBOL(bio_split); EXPORT_SYMBOL(bio_split_pool); -- cgit v1.2.2 From b823825e8e09aac6dc1ca362cd5639a87329d636 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 20 Jun 2005 14:05:27 +0200 Subject: [PATCH] Keep the bio end_io parts inside of bio.c for blk_rq_map_kern() Signed-off-by: Jens Axboe --- fs/bio.c | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'fs/bio.c') diff --git a/fs/bio.c b/fs/bio.c index 707b9af2dd01..c0d9140e470c 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -701,6 +701,16 @@ void bio_unmap_user(struct bio *bio) bio_put(bio); } +static int bio_map_kern_endio(struct bio *bio, unsigned int bytes_done, int err) +{ + if (bio->bi_size) + return 1; + + bio_put(bio); + return 0; +} + + static struct bio *__bio_map_kern(request_queue_t *q, void *data, unsigned int len, unsigned int gfp_mask) { @@ -734,6 +744,7 @@ static struct bio *__bio_map_kern(request_queue_t *q, void *data, offset = 0; } + bio->bi_end_io = bio_map_kern_endio; return bio; } -- cgit v1.2.2 From f1970baf6d74e03bd32072ab453f2fc01bc1b8d3 Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Mon, 20 Jun 2005 14:06:52 +0200 Subject: [PATCH] Add scatter-gather support for the block layer SG_IO Signed-off-by: Jens Axboe --- fs/bio.c | 150 ++++++++++++++++++++++++++++++++++++++++++++------------------- 1 file changed, 105 insertions(+), 45 deletions(-) (limited to 'fs/bio.c') diff --git a/fs/bio.c b/fs/bio.c index c0d9140e470c..24e4045788e2 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -25,6 +25,7 @@ #include #include #include +#include /* for struct sg_iovec */ #define BIO_POOL_SIZE 256 @@ -549,22 +550,34 @@ out_bmd: return ERR_PTR(ret); } -static struct bio *__bio_map_user(request_queue_t *q, struct block_device *bdev, - unsigned long uaddr, unsigned int len, - int write_to_vm) +static struct bio *__bio_map_user_iov(request_queue_t *q, + struct block_device *bdev, + struct sg_iovec *iov, int iov_count, + int write_to_vm) { - unsigned long end = (uaddr + len + PAGE_SIZE - 1) >> PAGE_SHIFT; - unsigned long start = uaddr >> PAGE_SHIFT; - const int nr_pages = end - start; - int ret, offset, i; + int i, j; + int nr_pages = 0; struct page **pages; struct bio *bio; + int cur_page = 0; + int ret, offset; - /* - * transfer and buffer must be aligned to at least hardsector - * size for now, in the future we can relax this restriction - */ - if ((uaddr & queue_dma_alignment(q)) || (len & queue_dma_alignment(q))) + for (i = 0; i < iov_count; i++) { + unsigned long uaddr = (unsigned long)iov[i].iov_base; + unsigned long len = iov[i].iov_len; + unsigned long end = (uaddr + len + PAGE_SIZE - 1) >> PAGE_SHIFT; + unsigned long start = uaddr >> PAGE_SHIFT; + + nr_pages += end - start; + /* + * transfer and buffer must be aligned to at least hardsector + * size for now, in the future we can relax this restriction + */ + if ((uaddr & queue_dma_alignment(q)) || (len & queue_dma_alignment(q))) + return ERR_PTR(-EINVAL); + } + + if (!nr_pages) return ERR_PTR(-EINVAL); bio = bio_alloc(GFP_KERNEL, nr_pages); @@ -576,42 +589,54 @@ static struct bio *__bio_map_user(request_queue_t *q, struct block_device *bdev, if (!pages) goto out; - down_read(¤t->mm->mmap_sem); - ret = get_user_pages(current, current->mm, uaddr, nr_pages, - write_to_vm, 0, pages, NULL); - up_read(¤t->mm->mmap_sem); - - if (ret < nr_pages) - goto out; - - bio->bi_bdev = bdev; - - offset = uaddr & ~PAGE_MASK; - for (i = 0; i < nr_pages; i++) { - unsigned int bytes = PAGE_SIZE - offset; - - if (len <= 0) - break; - - if (bytes > len) - bytes = len; + memset(pages, 0, nr_pages * sizeof(struct page *)); + + for (i = 0; i < iov_count; i++) { + unsigned long uaddr = (unsigned long)iov[i].iov_base; + unsigned long len = iov[i].iov_len; + unsigned long end = (uaddr + len + PAGE_SIZE - 1) >> PAGE_SHIFT; + unsigned long start = uaddr >> PAGE_SHIFT; + const int local_nr_pages = end - start; + const int page_limit = cur_page + local_nr_pages; + + down_read(¤t->mm->mmap_sem); + ret = get_user_pages(current, current->mm, uaddr, + local_nr_pages, + write_to_vm, 0, &pages[cur_page], NULL); + up_read(¤t->mm->mmap_sem); + + if (ret < local_nr_pages) + goto out_unmap; + + + offset = uaddr & ~PAGE_MASK; + for (j = cur_page; j < page_limit; j++) { + unsigned int bytes = PAGE_SIZE - offset; + + if (len <= 0) + break; + + if (bytes > len) + bytes = len; + + /* + * sorry... + */ + if (__bio_add_page(q, bio, pages[j], bytes, offset) < bytes) + break; + + len -= bytes; + offset = 0; + } + cur_page = j; /* - * sorry... + * release the pages we didn't map into the bio, if any */ - if (__bio_add_page(q, bio, pages[i], bytes, offset) < bytes) - break; - - len -= bytes; - offset = 0; + while (j < page_limit) + page_cache_release(pages[j++]); } - /* - * release the pages we didn't map into the bio, if any - */ - while (i < nr_pages) - page_cache_release(pages[i++]); - kfree(pages); /* @@ -620,9 +645,17 @@ static struct bio *__bio_map_user(request_queue_t *q, struct block_device *bdev, if (!write_to_vm) bio->bi_rw |= (1 << BIO_RW); + bio->bi_bdev = bdev; bio->bi_flags |= (1 << BIO_USER_MAPPED); return bio; -out: + + out_unmap: + for (i = 0; i < nr_pages; i++) { + if(!pages[i]) + break; + page_cache_release(pages[i]); + } + out: kfree(pages); bio_put(bio); return ERR_PTR(ret); @@ -641,10 +674,34 @@ out: */ struct bio *bio_map_user(request_queue_t *q, struct block_device *bdev, unsigned long uaddr, unsigned int len, int write_to_vm) +{ + struct sg_iovec iov; + + iov.iov_base = (__user void *)uaddr; + iov.iov_len = len; + + return bio_map_user_iov(q, bdev, &iov, 1, write_to_vm); +} + +/** + * bio_map_user_iov - map user sg_iovec table into bio + * @q: the request_queue_t for the bio + * @bdev: destination block device + * @iov: the iovec. + * @iov_count: number of elements in the iovec + * @write_to_vm: bool indicating writing to pages or not + * + * Map the user space address into a bio suitable for io to a block + * device. Returns an error pointer in case of error. + */ +struct bio *bio_map_user_iov(request_queue_t *q, struct block_device *bdev, + struct sg_iovec *iov, int iov_count, + int write_to_vm) { struct bio *bio; + int len = 0, i; - bio = __bio_map_user(q, bdev, uaddr, len, write_to_vm); + bio = __bio_map_user_iov(q, bdev, iov, iov_count, write_to_vm); if (IS_ERR(bio)) return bio; @@ -657,6 +714,9 @@ struct bio *bio_map_user(request_queue_t *q, struct block_device *bdev, */ bio_get(bio); + for (i = 0; i < iov_count; i++) + len += iov[i].iov_len; + if (bio->bi_size == len) return bio; -- cgit v1.2.2 From e525e153c7a99fc64af68e7f50a2660babd6752b Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Sun, 7 Aug 2005 09:42:12 -0700 Subject: [PATCH] __bio_clone() dead comment Remove a very wrong comment. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/bio.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) (limited to 'fs/bio.c') diff --git a/fs/bio.c b/fs/bio.c index 249dd6bb66c8..1f2d4649b188 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -248,17 +248,13 @@ inline void __bio_clone(struct bio *bio, struct bio *bio_src) { request_queue_t *q = bdev_get_queue(bio_src->bi_bdev); - memcpy(bio->bi_io_vec, bio_src->bi_io_vec, bio_src->bi_max_vecs * sizeof(struct bio_vec)); + memcpy(bio->bi_io_vec, bio_src->bi_io_vec, + bio_src->bi_max_vecs * sizeof(struct bio_vec)); bio->bi_sector = bio_src->bi_sector; bio->bi_bdev = bio_src->bi_bdev; bio->bi_flags |= 1 << BIO_CLONED; bio->bi_rw = bio_src->bi_rw; - - /* - * notes -- maybe just leave bi_idx alone. assume identical mapping - * for the clone - */ bio->bi_vcnt = bio_src->bi_vcnt; bio->bi_size = bio_src->bi_size; bio->bi_idx = bio_src->bi_idx; -- cgit v1.2.2 From 3676347a5e216a7fec7f8eedbbcf8bed6b9c4e40 Mon Sep 17 00:00:00 2001 From: Peter Osterlund Date: Tue, 6 Sep 2005 15:16:42 -0700 Subject: [PATCH] kill bio->bi_set Jens: ->bi_set is totally unnecessary bloat of struct bio. Just define a proper destructor for the bio and it already knows what bio_set it belongs too. Peter: Fixed the bugs. Signed-off-by: Jens Axboe Signed-off-by: Peter Osterlund Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/bio.c | 32 +++++++++++++++++++++----------- 1 file changed, 21 insertions(+), 11 deletions(-) (limited to 'fs/bio.c') diff --git a/fs/bio.c b/fs/bio.c index 1f2d4649b188..bf3ec9d2b54c 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -104,18 +104,22 @@ static inline struct bio_vec *bvec_alloc_bs(unsigned int __nocast gfp_mask, int return bvl; } -/* - * default destructor for a bio allocated with bio_alloc_bioset() - */ -static void bio_destructor(struct bio *bio) +void bio_free(struct bio *bio, struct bio_set *bio_set) { const int pool_idx = BIO_POOL_IDX(bio); - struct bio_set *bs = bio->bi_set; BIO_BUG_ON(pool_idx >= BIOVEC_NR_POOLS); - mempool_free(bio->bi_io_vec, bs->bvec_pools[pool_idx]); - mempool_free(bio, bs->bio_pool); + mempool_free(bio->bi_io_vec, bio_set->bvec_pools[pool_idx]); + mempool_free(bio, bio_set->bio_pool); +} + +/* + * default destructor for a bio allocated with bio_alloc_bioset() + */ +static void bio_fs_destructor(struct bio *bio) +{ + bio_free(bio, fs_bio_set); } inline void bio_init(struct bio *bio) @@ -171,8 +175,6 @@ struct bio *bio_alloc_bioset(unsigned int __nocast gfp_mask, int nr_iovecs, stru bio->bi_max_vecs = bvec_slabs[idx].nr_vecs; } bio->bi_io_vec = bvl; - bio->bi_destructor = bio_destructor; - bio->bi_set = bs; } out: return bio; @@ -180,7 +182,12 @@ out: struct bio *bio_alloc(unsigned int __nocast gfp_mask, int nr_iovecs) { - return bio_alloc_bioset(gfp_mask, nr_iovecs, fs_bio_set); + struct bio *bio = bio_alloc_bioset(gfp_mask, nr_iovecs, fs_bio_set); + + if (bio) + bio->bi_destructor = bio_fs_destructor; + + return bio; } void zero_fill_bio(struct bio *bio) @@ -273,8 +280,10 @@ struct bio *bio_clone(struct bio *bio, unsigned int __nocast gfp_mask) { struct bio *b = bio_alloc_bioset(gfp_mask, bio->bi_max_vecs, fs_bio_set); - if (b) + if (b) { + b->bi_destructor = bio_fs_destructor; __bio_clone(b, bio); + } return b; } @@ -1075,6 +1084,7 @@ subsys_initcall(init_bio); EXPORT_SYMBOL(bio_alloc); EXPORT_SYMBOL(bio_put); +EXPORT_SYMBOL(bio_free); EXPORT_SYMBOL(bio_endio); EXPORT_SYMBOL(bio_init); EXPORT_SYMBOL(__bio_clone); -- cgit v1.2.2 From 3f70353ea91ad77c83500e70507a239b2ab0c980 Mon Sep 17 00:00:00 2001 From: "viro@ZenIV.linux.org.uk" Date: Fri, 9 Sep 2005 16:53:56 +0100 Subject: [PATCH] bogus cast in bio.c void * is not the same as void *... Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- fs/bio.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/bio.c') diff --git a/fs/bio.c b/fs/bio.c index a7d4fd3a3299..83a349574567 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -683,7 +683,7 @@ struct bio *bio_map_user(request_queue_t *q, struct block_device *bdev, { struct sg_iovec iov; - iov.iov_base = (__user void *)uaddr; + iov.iov_base = (void __user *)uaddr; iov.iov_len = len; return bio_map_user_iov(q, bdev, &iov, 1, write_to_vm); -- cgit v1.2.2 From dd0fc66fb33cd610bc1a5db8a5e232d34879b4d7 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 7 Oct 2005 07:46:04 +0100 Subject: [PATCH] gfp flags annotations - part 1 - added typedef unsigned int __nocast gfp_t; - replaced __nocast uses for gfp flags with gfp_t - it gives exactly the same warnings as far as sparse is concerned, doesn't change generated code (from gcc point of view we replaced unsigned int with typedef) and documents what's going on far better. Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- fs/bio.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'fs/bio.c') diff --git a/fs/bio.c b/fs/bio.c index 83a349574567..7d81a93afd48 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -75,7 +75,7 @@ struct bio_set { */ static struct bio_set *fs_bio_set; -static inline struct bio_vec *bvec_alloc_bs(unsigned int __nocast gfp_mask, int nr, unsigned long *idx, struct bio_set *bs) +static inline struct bio_vec *bvec_alloc_bs(gfp_t gfp_mask, int nr, unsigned long *idx, struct bio_set *bs) { struct bio_vec *bvl; struct biovec_slab *bp; @@ -155,7 +155,7 @@ inline void bio_init(struct bio *bio) * allocate bio and iovecs from the memory pools specified by the * bio_set structure. **/ -struct bio *bio_alloc_bioset(unsigned int __nocast gfp_mask, int nr_iovecs, struct bio_set *bs) +struct bio *bio_alloc_bioset(gfp_t gfp_mask, int nr_iovecs, struct bio_set *bs) { struct bio *bio = mempool_alloc(bs->bio_pool, gfp_mask); @@ -181,7 +181,7 @@ out: return bio; } -struct bio *bio_alloc(unsigned int __nocast gfp_mask, int nr_iovecs) +struct bio *bio_alloc(gfp_t gfp_mask, int nr_iovecs) { struct bio *bio = bio_alloc_bioset(gfp_mask, nr_iovecs, fs_bio_set); @@ -277,7 +277,7 @@ inline void __bio_clone(struct bio *bio, struct bio *bio_src) * * Like __bio_clone, only also allocates the returned bio */ -struct bio *bio_clone(struct bio *bio, unsigned int __nocast gfp_mask) +struct bio *bio_clone(struct bio *bio, gfp_t gfp_mask) { struct bio *b = bio_alloc_bioset(gfp_mask, bio->bi_max_vecs, fs_bio_set); @@ -1078,7 +1078,7 @@ struct bio_pair *bio_split(struct bio *bi, mempool_t *pool, int first_sectors) return bp; } -static void *bio_pair_alloc(unsigned int __nocast gfp_flags, void *data) +static void *bio_pair_alloc(gfp_t gfp_flags, void *data) { return kmalloc(sizeof(struct bio_pair), gfp_flags); } -- cgit v1.2.2 From 27496a8c67bef4d789d8e3c8317ca35813a507ae Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 21 Oct 2005 03:20:48 -0400 Subject: [PATCH] gfp_t: fs/* - ->releasepage() annotated (s/int/gfp_t), instances updated - missing gfp_t in fs/* added - fixed misannotation from the original sweep caught by bitwise checks: XFS used __nocast both for gfp_t and for flags used by XFS allocator. The latter left with unsigned int __nocast; we might want to add a different type for those but for now let's leave them alone. That, BTW, is a case when __nocast use had been actively confusing - it had been used in the same code for two different and similar types, with no way to catch misuses. Switch of gfp_t to bitwise had caught that immediately... One tricky bit is left alone to be dealt with later - mapping->flags is a mix of gfp_t and error indications. Left alone for now. Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- fs/bio.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs/bio.c') diff --git a/fs/bio.c b/fs/bio.c index 7d81a93afd48..460554b07ff9 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -778,7 +778,7 @@ static int bio_map_kern_endio(struct bio *bio, unsigned int bytes_done, int err) static struct bio *__bio_map_kern(request_queue_t *q, void *data, - unsigned int len, unsigned int gfp_mask) + unsigned int len, gfp_t gfp_mask) { unsigned long kaddr = (unsigned long)data; unsigned long end = (kaddr + len + PAGE_SIZE - 1) >> PAGE_SHIFT; @@ -825,7 +825,7 @@ static struct bio *__bio_map_kern(request_queue_t *q, void *data, * device. Returns an error pointer in case of error. */ struct bio *bio_map_kern(request_queue_t *q, void *data, unsigned int len, - unsigned int gfp_mask) + gfp_t gfp_mask) { struct bio *bio; -- cgit v1.2.2