X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=fs%2Fbio.c;h=124b95c4d58286649d8d0f982edd9074af18ac82;hb=47be12e4eec84c1846f29af64fe25a396b57a026;hp=52de79c299424d9806dafe475b362246c14a5087;hpb=50be345560f1ffdcb15cc0e146416b80529a2ef2;p=safe%2Fjmp%2Flinux-2.6 diff --git a/fs/bio.c b/fs/bio.c index 52de79c..124b95c 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2001 Jens Axboe + * Copyright (C) 2001 Jens Axboe * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -26,94 +26,238 @@ #include #include #include +#include #include /* for struct sg_iovec */ -#define BIO_POOL_SIZE 256 - -static kmem_cache_t *bio_slab __read_mostly; - -#define BIOVEC_NR_POOLS 6 +DEFINE_TRACE(block_split); /* - * a small number of entries is fine, not going to be performance critical. - * basically we just need to survive + * Test patch to inline a certain number of bi_io_vec's inside the bio + * itself, to shrink a bio data allocation from two mempool calls to one */ -#define BIO_SPLIT_ENTRIES 8 -mempool_t *bio_split_pool __read_mostly; +#define BIO_INLINE_VECS 4 -struct biovec_slab { - int nr_vecs; - char *name; - kmem_cache_t *slab; -}; +static mempool_t *bio_split_pool __read_mostly; /* * if you change this list, also change bvec_alloc or things will * break badly! cannot be bigger than what you can fit into an * unsigned short */ - #define BV(x) { .nr_vecs = x, .name = "biovec-"__stringify(x) } -static struct biovec_slab bvec_slabs[BIOVEC_NR_POOLS] __read_mostly = { +struct biovec_slab bvec_slabs[BIOVEC_NR_POOLS] __read_mostly = { BV(1), BV(4), BV(16), BV(64), BV(128), BV(BIO_MAX_PAGES), }; #undef BV /* - * bio_set is used to allow other portions of the IO system to - * allocate their own private memory pools for bio and iovec structures. - * These memory pools in turn all allocate from the bio_slab - * and the bvec_slabs[]. + * fs_bio_set is the bio_set containing bio and iovec memory pools used by + * IO code that does not need private memory pools. */ -struct bio_set { - mempool_t *bio_pool; - mempool_t *bvec_pools[BIOVEC_NR_POOLS]; -}; +struct bio_set *fs_bio_set; /* - * fs_bio_set is the bio_set containing bio and iovec memory pools used by - * IO code that does not need private memory pools. + * Our slab pool management */ -static struct bio_set *fs_bio_set; +struct bio_slab { + struct kmem_cache *slab; + unsigned int slab_ref; + unsigned int slab_size; + char name[8]; +}; +static DEFINE_MUTEX(bio_slab_lock); +static struct bio_slab *bio_slabs; +static unsigned int bio_slab_nr, bio_slab_max; + +static struct kmem_cache *bio_find_or_create_slab(unsigned int extra_size) +{ + unsigned int sz = sizeof(struct bio) + extra_size; + struct kmem_cache *slab = NULL; + struct bio_slab *bslab; + unsigned int i, entry = -1; + + mutex_lock(&bio_slab_lock); + + i = 0; + while (i < bio_slab_nr) { + struct bio_slab *bslab = &bio_slabs[i]; + + if (!bslab->slab && entry == -1) + entry = i; + else if (bslab->slab_size == sz) { + slab = bslab->slab; + bslab->slab_ref++; + break; + } + i++; + } + + if (slab) + goto out_unlock; + + if (bio_slab_nr == bio_slab_max && entry == -1) { + bio_slab_max <<= 1; + bio_slabs = krealloc(bio_slabs, + bio_slab_max * sizeof(struct bio_slab), + GFP_KERNEL); + if (!bio_slabs) + goto out_unlock; + } + if (entry == -1) + entry = bio_slab_nr++; + + bslab = &bio_slabs[entry]; + + snprintf(bslab->name, sizeof(bslab->name), "bio-%d", entry); + slab = kmem_cache_create(bslab->name, sz, 0, SLAB_HWCACHE_ALIGN, NULL); + if (!slab) + goto out_unlock; + + printk("bio: create slab <%s> at %d\n", bslab->name, entry); + bslab->slab = slab; + bslab->slab_ref = 1; + bslab->slab_size = sz; +out_unlock: + mutex_unlock(&bio_slab_lock); + return slab; +} -static inline struct bio_vec *bvec_alloc_bs(gfp_t gfp_mask, int nr, unsigned long *idx, struct bio_set *bs) +static void bio_put_slab(struct bio_set *bs) +{ + struct bio_slab *bslab = NULL; + unsigned int i; + + mutex_lock(&bio_slab_lock); + + for (i = 0; i < bio_slab_nr; i++) { + if (bs->bio_slab == bio_slabs[i].slab) { + bslab = &bio_slabs[i]; + break; + } + } + + if (WARN(!bslab, KERN_ERR "bio: unable to find slab!\n")) + goto out; + + WARN_ON(!bslab->slab_ref); + + if (--bslab->slab_ref) + goto out; + + kmem_cache_destroy(bslab->slab); + bslab->slab = NULL; + +out: + mutex_unlock(&bio_slab_lock); +} + +unsigned int bvec_nr_vecs(unsigned short idx) +{ + return bvec_slabs[idx].nr_vecs; +} + +void bvec_free_bs(struct bio_set *bs, struct bio_vec *bv, unsigned int idx) +{ + BIO_BUG_ON(idx >= BIOVEC_NR_POOLS); + + if (idx == BIOVEC_MAX_IDX) + mempool_free(bv, bs->bvec_pool); + else { + struct biovec_slab *bvs = bvec_slabs + idx; + + kmem_cache_free(bvs->slab, bv); + } +} + +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; + + /* + * If 'bs' is given, lookup the pool and do the mempool alloc. + * If not, this is a bio_kmalloc() allocation and just do a + * kzalloc() for the exact number of vecs right away. + */ + if (!bs) + bvl = kmalloc(nr * sizeof(struct bio_vec), gfp_mask); /* * see comment near bvec_array define! */ switch (nr) { - case 1 : *idx = 0; break; - case 2 ... 4: *idx = 1; break; - case 5 ... 16: *idx = 2; break; - case 17 ... 64: *idx = 3; break; - case 65 ... 128: *idx = 4; break; - case 129 ... BIO_MAX_PAGES: *idx = 5; break; - default: - return NULL; + case 1: + *idx = 0; + break; + case 2 ... 4: + *idx = 1; + break; + case 5 ... 16: + *idx = 2; + break; + case 17 ... 64: + *idx = 3; + break; + case 65 ... 128: + *idx = 4; + break; + case 129 ... BIO_MAX_PAGES: + *idx = 5; + break; + default: + return NULL; } + /* - * idx now points to the pool we want to allocate from + * idx now points to the pool we want to allocate from. only the + * 1-vec entry pool is mempool backed. */ + if (*idx == BIOVEC_MAX_IDX) { +fallback: + bvl = mempool_alloc(bs->bvec_pool, gfp_mask); + } else { + struct biovec_slab *bvs = bvec_slabs + *idx; + gfp_t __gfp_mask = gfp_mask & ~(__GFP_WAIT | __GFP_IO); + + /* + * Make this allocation restricted and don't dump info on + * allocation failures, since we'll fallback to the mempool + * in case of failure. + */ + __gfp_mask |= __GFP_NOMEMALLOC | __GFP_NORETRY | __GFP_NOWARN; - bp = bvec_slabs + *idx; - bvl = mempool_alloc(bs->bvec_pools[*idx], gfp_mask); - if (bvl) - memset(bvl, 0, bp->nr_vecs * sizeof(struct bio_vec)); + /* + * Try a slab allocation. If this fails and __GFP_WAIT + * is set, retry with the 1-entry mempool + */ + bvl = kmem_cache_alloc(bvs->slab, __gfp_mask); + if (unlikely(!bvl && (gfp_mask & __GFP_WAIT))) { + *idx = BIOVEC_MAX_IDX; + goto fallback; + } + } return bvl; } -void bio_free(struct bio *bio, struct bio_set *bio_set) +void bio_free(struct bio *bio, struct bio_set *bs) { - const int pool_idx = BIO_POOL_IDX(bio); + void *p; + + if (bio_has_allocated_vec(bio)) + bvec_free_bs(bs, bio->bi_io_vec, BIO_POOL_IDX(bio)); - BIO_BUG_ON(pool_idx >= BIOVEC_NR_POOLS); + if (bio_integrity(bio)) + bio_integrity_free(bio, bs); - mempool_free(bio->bi_io_vec, bio_set->bvec_pools[pool_idx]); - mempool_free(bio, bio_set->bio_pool); + /* + * If we have front padding, adjust the bio pointer before freeing + */ + p = bio; + if (bs->front_pad) + p -= bs->front_pad; + + mempool_free(p, bs->bio_pool); } /* @@ -124,58 +268,76 @@ static void bio_fs_destructor(struct bio *bio) bio_free(bio, fs_bio_set); } +static void bio_kmalloc_destructor(struct bio *bio) +{ + if (bio_has_allocated_vec(bio)) + kfree(bio->bi_io_vec); + kfree(bio); +} + void bio_init(struct bio *bio) { - bio->bi_next = NULL; - bio->bi_bdev = NULL; + memset(bio, 0, sizeof(*bio)); bio->bi_flags = 1 << BIO_UPTODATE; - bio->bi_rw = 0; - bio->bi_vcnt = 0; - bio->bi_idx = 0; - bio->bi_phys_segments = 0; - bio->bi_hw_segments = 0; - bio->bi_hw_front_size = 0; - bio->bi_hw_back_size = 0; - bio->bi_size = 0; - bio->bi_max_vecs = 0; - bio->bi_end_io = NULL; + bio->bi_comp_cpu = -1; atomic_set(&bio->bi_cnt, 1); - bio->bi_private = NULL; } /** * bio_alloc_bioset - allocate a bio for I/O * @gfp_mask: the GFP_ mask given to the slab allocator * @nr_iovecs: number of iovecs to pre-allocate - * @bs: the bio_set to allocate from + * @bs: the bio_set to allocate from. If %NULL, just use kmalloc * * Description: - * bio_alloc_bioset will first try it's on mempool to satisfy the allocation. + * bio_alloc_bioset will first try its own mempool to satisfy the allocation. * If %__GFP_WAIT is set then we will block on the internal pool waiting - * for a &struct bio to become free. + * for a &struct bio to become free. If a %NULL @bs is passed in, we will + * fall back to just using @kmalloc to allocate the required memory. * - * allocate bio and iovecs from the memory pools specified by the - * bio_set structure. + * Note that the caller must set ->bi_destructor on succesful return + * of a bio, to do the appropriate freeing of the bio once the reference + * count drops to zero. **/ 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); + struct bio *bio = NULL; + void *uninitialized_var(p); + + if (bs) { + p = mempool_alloc(bs->bio_pool, gfp_mask); + + if (p) + bio = p + bs->front_pad; + } else + bio = kmalloc(sizeof(*bio), gfp_mask); if (likely(bio)) { struct bio_vec *bvl = NULL; bio_init(bio); if (likely(nr_iovecs)) { - unsigned long idx; - - bvl = bvec_alloc_bs(gfp_mask, nr_iovecs, &idx, bs); + unsigned long uninitialized_var(idx); + + if (nr_iovecs <= BIO_INLINE_VECS) { + idx = 0; + bvl = bio->bi_inline_vecs; + nr_iovecs = BIO_INLINE_VECS; + } else { + bvl = bvec_alloc_bs(gfp_mask, nr_iovecs, &idx, + bs); + nr_iovecs = bvec_nr_vecs(idx); + } if (unlikely(!bvl)) { - mempool_free(bio, bs->bio_pool); + if (bs) + mempool_free(p, bs->bio_pool); + else + kfree(bio); bio = NULL; goto out; } bio->bi_flags |= idx << BIO_POOL_OFFSET; - bio->bi_max_vecs = bvec_slabs[idx].nr_vecs; + bio->bi_max_vecs = nr_iovecs; } bio->bi_io_vec = bvl; } @@ -193,6 +355,23 @@ struct bio *bio_alloc(gfp_t gfp_mask, int nr_iovecs) return bio; } +/* + * Like bio_alloc(), but doesn't use a mempool backing. This means that + * it CAN fail, but while bio_alloc() can only be used for allocations + * that have a short (finite) life span, bio_kmalloc() should be used + * for more permanent bio allocations (like allocating some bio's for + * initalization or setup purposes). + */ +struct bio *bio_kmalloc(gfp_t gfp_mask, int nr_iovecs) +{ + struct bio *bio = bio_alloc_bioset(gfp_mask, nr_iovecs, NULL); + + if (bio) + bio->bi_destructor = bio_kmalloc_destructor; + + return bio; +} + void zero_fill_bio(struct bio *bio) { unsigned long flags; @@ -229,7 +408,7 @@ void bio_put(struct bio *bio) } } -inline int bio_phys_segments(request_queue_t *q, struct bio *bio) +inline int bio_phys_segments(struct request_queue *q, struct bio *bio) { if (unlikely(!bio_flagged(bio, BIO_SEG_VALID))) blk_recount_segments(q, bio); @@ -237,14 +416,6 @@ inline int bio_phys_segments(request_queue_t *q, struct bio *bio) return bio->bi_phys_segments; } -inline int bio_hw_segments(request_queue_t *q, struct bio *bio) -{ - if (unlikely(!bio_flagged(bio, BIO_SEG_VALID))) - blk_recount_segments(q, bio); - - return bio->bi_hw_segments; -} - /** * __bio_clone - clone a bio * @bio: destination bio @@ -256,11 +427,13 @@ inline int bio_hw_segments(request_queue_t *q, struct bio *bio) */ 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)); + /* + * most users will be overriding ->bi_bdev with a new target, + * so we don't set nor calculate new physical/hw segment counts here + */ bio->bi_sector = bio_src->bi_sector; bio->bi_bdev = bio_src->bi_bdev; bio->bi_flags |= 1 << BIO_CLONED; @@ -268,8 +441,6 @@ void __bio_clone(struct bio *bio, struct bio *bio_src) bio->bi_vcnt = bio_src->bi_vcnt; bio->bi_size = bio_src->bi_size; bio->bi_idx = bio_src->bi_idx; - bio_phys_segments(q, bio); - bio_hw_segments(q, bio); } /** @@ -283,9 +454,19 @@ 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); - if (b) { - b->bi_destructor = bio_fs_destructor; - __bio_clone(b, bio); + if (!b) + return NULL; + + b->bi_destructor = bio_fs_destructor; + __bio_clone(b, bio); + + if (bio_integrity(bio)) { + int ret; + + ret = bio_integrity_clone(b, bio, fs_bio_set); + + if (ret < 0) + return NULL; } return b; @@ -302,7 +483,7 @@ struct bio *bio_clone(struct bio *bio, gfp_t gfp_mask) */ int bio_get_nr_vecs(struct block_device *bdev) { - request_queue_t *q = bdev_get_queue(bdev); + struct request_queue *q = bdev_get_queue(bdev); int nr_pages; nr_pages = ((q->max_sectors << 9) + PAGE_SIZE - 1) >> PAGE_SHIFT; @@ -314,7 +495,7 @@ int bio_get_nr_vecs(struct block_device *bdev) return nr_pages; } -static int __bio_add_page(request_queue_t *q, struct bio *bio, struct page +static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page *page, unsigned int len, unsigned int offset, unsigned short max_sectors) { @@ -341,10 +522,19 @@ static int __bio_add_page(request_queue_t *q, struct bio *bio, struct page if (page == prev->bv_page && offset == prev->bv_offset + prev->bv_len) { prev->bv_len += len; - if (q->merge_bvec_fn && - q->merge_bvec_fn(q, bio, prev) < len) { - prev->bv_len -= len; - return 0; + + if (q->merge_bvec_fn) { + struct bvec_merge_data bvm = { + .bi_bdev = bio->bi_bdev, + .bi_sector = bio->bi_sector, + .bi_size = bio->bi_size, + .bi_rw = bio->bi_rw, + }; + + if (q->merge_bvec_fn(q, &bvm, prev) < len) { + prev->bv_len -= len; + return 0; + } } goto done; @@ -360,8 +550,7 @@ static int __bio_add_page(request_queue_t *q, struct bio *bio, struct page */ while (bio->bi_phys_segments >= q->max_phys_segments - || bio->bi_hw_segments >= q->max_hw_segments - || BIOVEC_VIRT_OVERSIZE(bio->bi_size)) { + || bio->bi_phys_segments >= q->max_hw_segments) { if (retried_segments) return 0; @@ -385,11 +574,18 @@ static int __bio_add_page(request_queue_t *q, struct bio *bio, struct page * queue to get further control */ if (q->merge_bvec_fn) { + struct bvec_merge_data bvm = { + .bi_bdev = bio->bi_bdev, + .bi_sector = bio->bi_sector, + .bi_size = bio->bi_size, + .bi_rw = bio->bi_rw, + }; + /* * merge_bvec_fn() returns number of bytes it can accept * at this offset */ - if (q->merge_bvec_fn(q, bio, bvec) < len) { + if (q->merge_bvec_fn(q, &bvm, bvec) < len) { bvec->bv_page = NULL; bvec->bv_len = 0; bvec->bv_offset = 0; @@ -398,13 +594,11 @@ static int __bio_add_page(request_queue_t *q, struct bio *bio, struct page } /* If we may be able to merge these biovecs, force a recount */ - if (bio->bi_vcnt && (BIOVEC_PHYS_MERGEABLE(bvec-1, bvec) || - BIOVEC_VIRT_MERGEABLE(bvec-1, bvec))) + if (bio->bi_vcnt && (BIOVEC_PHYS_MERGEABLE(bvec-1, bvec))) bio->bi_flags &= ~(1 << BIO_SEG_VALID); bio->bi_vcnt++; bio->bi_phys_segments++; - bio->bi_hw_segments++; done: bio->bi_size += len; return len; @@ -424,7 +618,7 @@ static int __bio_add_page(request_queue_t *q, struct bio *bio, struct page * smaller than PAGE_SIZE, so it is always possible to add a single * page to an empty bio. This should only be used by REQ_PC bios. */ -int bio_add_pc_page(request_queue_t *q, struct bio *bio, struct page *page, +int bio_add_pc_page(struct request_queue *q, struct bio *bio, struct page *page, unsigned int len, unsigned int offset) { return __bio_add_page(q, bio, page, len, offset, q->max_hw_sectors); @@ -452,36 +646,104 @@ int bio_add_page(struct bio *bio, struct page *page, unsigned int len, struct bio_map_data { struct bio_vec *iovecs; - void __user *userptr; + struct sg_iovec *sgvecs; + int nr_sgvecs; + int is_our_pages; }; -static void bio_set_map_data(struct bio_map_data *bmd, struct bio *bio) +static void bio_set_map_data(struct bio_map_data *bmd, struct bio *bio, + struct sg_iovec *iov, int iov_count, + int is_our_pages) { memcpy(bmd->iovecs, bio->bi_io_vec, sizeof(struct bio_vec) * bio->bi_vcnt); + memcpy(bmd->sgvecs, iov, sizeof(struct sg_iovec) * iov_count); + bmd->nr_sgvecs = iov_count; + bmd->is_our_pages = is_our_pages; bio->bi_private = bmd; } static void bio_free_map_data(struct bio_map_data *bmd) { kfree(bmd->iovecs); + kfree(bmd->sgvecs); kfree(bmd); } -static struct bio_map_data *bio_alloc_map_data(int nr_segs) +static struct bio_map_data *bio_alloc_map_data(int nr_segs, int iov_count, + gfp_t gfp_mask) { - struct bio_map_data *bmd = kmalloc(sizeof(*bmd), GFP_KERNEL); + struct bio_map_data *bmd = kmalloc(sizeof(*bmd), gfp_mask); if (!bmd) return NULL; - bmd->iovecs = kmalloc(sizeof(struct bio_vec) * nr_segs, GFP_KERNEL); - if (bmd->iovecs) + bmd->iovecs = kmalloc(sizeof(struct bio_vec) * nr_segs, gfp_mask); + if (!bmd->iovecs) { + kfree(bmd); + return NULL; + } + + bmd->sgvecs = kmalloc(sizeof(struct sg_iovec) * iov_count, gfp_mask); + if (bmd->sgvecs) return bmd; + kfree(bmd->iovecs); kfree(bmd); return NULL; } +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) +{ + 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); + unsigned int bv_len = iovecs[i].bv_len; + + while (bv_len && iov_idx < iov_count) { + unsigned int bytes; + char *iov_addr; + + bytes = min_t(unsigned int, + iov[iov_idx].iov_len - iov_off, bv_len); + 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) + ret = copy_to_user(iov_addr, bv_addr, + bytes); + + if (ret) + ret = -EFAULT; + } + + bv_len -= bytes; + bv_addr += bytes; + iov_addr += bytes; + iov_off += bytes; + + if (iov[iov_idx].iov_len == iov_off) { + iov_idx++; + iov_off = 0; + } + } + + if (do_free_page) + __free_page(bvec->bv_page); + } + + return ret; +} + /** * bio_uncopy_user - finish previously mapped bio * @bio: bio being terminated @@ -492,79 +754,104 @@ static struct bio_map_data *bio_alloc_map_data(int nr_segs) int bio_uncopy_user(struct bio *bio) { struct bio_map_data *bmd = bio->bi_private; - const int read = bio_data_dir(bio) == READ; - struct bio_vec *bvec; - int i, ret = 0; + int ret = 0; - __bio_for_each_segment(bvec, bio, i, 0) { - char *addr = page_address(bvec->bv_page); - unsigned int len = bmd->iovecs[i].bv_len; - - if (read && !ret && copy_to_user(bmd->userptr, addr, len)) - ret = -EFAULT; - - __free_page(bvec->bv_page); - bmd->userptr += len; - } + if (!bio_flagged(bio, BIO_NULL_MAPPED)) + ret = __bio_copy_iov(bio, bmd->iovecs, bmd->sgvecs, + bmd->nr_sgvecs, 1, bmd->is_our_pages); bio_free_map_data(bmd); bio_put(bio); return ret; } /** - * bio_copy_user - copy user data to bio + * bio_copy_user_iov - copy user data to bio * @q: destination block queue - * @uaddr: start of user address - * @len: length in bytes + * @map_data: pointer to the rq_map_data holding pages (if necessary) + * @iov: the iovec. + * @iov_count: number of elements in the iovec * @write_to_vm: bool indicating writing to pages or not + * @gfp_mask: memory allocation flags * * Prepares and returns a bio for indirect user io, bouncing data * to/from kernel pages as necessary. Must be paired with * call bio_uncopy_user() on io completion. */ -struct bio *bio_copy_user(request_queue_t *q, unsigned long uaddr, - unsigned int len, int write_to_vm) +struct bio *bio_copy_user_iov(struct request_queue *q, + struct rq_map_data *map_data, + struct sg_iovec *iov, int iov_count, + int write_to_vm, gfp_t gfp_mask) { - unsigned long end = (uaddr + len + PAGE_SIZE - 1) >> PAGE_SHIFT; - unsigned long start = uaddr >> PAGE_SHIFT; struct bio_map_data *bmd; struct bio_vec *bvec; struct page *page; struct bio *bio; int i, ret; + int nr_pages = 0; + unsigned int len = 0; + unsigned int offset = map_data ? map_data->offset & ~PAGE_MASK : 0; + + for (i = 0; i < iov_count; i++) { + unsigned long uaddr; + unsigned long end; + unsigned long start; + + uaddr = (unsigned long)iov[i].iov_base; + end = (uaddr + iov[i].iov_len + PAGE_SIZE - 1) >> PAGE_SHIFT; + start = uaddr >> PAGE_SHIFT; + + nr_pages += end - start; + len += iov[i].iov_len; + } - bmd = bio_alloc_map_data(end - start); + bmd = bio_alloc_map_data(nr_pages, iov_count, gfp_mask); if (!bmd) return ERR_PTR(-ENOMEM); - bmd->userptr = (void __user *) uaddr; - ret = -ENOMEM; - bio = bio_alloc(GFP_KERNEL, end - start); + bio = bio_alloc(gfp_mask, nr_pages); if (!bio) goto out_bmd; bio->bi_rw |= (!write_to_vm << BIO_RW); ret = 0; + + if (map_data) { + nr_pages = 1 << map_data->page_order; + i = map_data->offset / PAGE_SIZE; + } while (len) { unsigned int bytes = PAGE_SIZE; + bytes -= offset; + if (bytes > len) bytes = len; - page = alloc_page(q->bounce_gfp | GFP_KERNEL); - if (!page) { - ret = -ENOMEM; - break; + if (map_data) { + if (i == map_data->nr_entries * nr_pages) { + ret = -ENOMEM; + break; + } + + page = map_data->pages[i / nr_pages]; + page += (i % nr_pages); + + i++; + } else { + page = alloc_page(q->bounce_gfp | gfp_mask); + if (!page) { + ret = -ENOMEM; + break; + } } - if (bio_add_pc_page(q, bio, page, bytes, 0) < bytes) { - ret = -EINVAL; + if (bio_add_pc_page(q, bio, page, bytes, offset) < bytes) break; - } len -= bytes; + offset = 0; } if (ret) @@ -573,27 +860,18 @@ struct bio *bio_copy_user(request_queue_t *q, unsigned long uaddr, /* * success */ - if (!write_to_vm) { - char __user *p = (char __user *) uaddr; - - /* - * for a write, copy in data to kernel pages - */ - ret = -EFAULT; - bio_for_each_segment(bvec, bio, i) { - char *addr = page_address(bvec->bv_page); - - if (copy_from_user(addr, p, bvec->bv_len)) - goto cleanup; - p += bvec->bv_len; - } + 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 (ret) + goto cleanup; } - bio_set_map_data(bmd, bio); + bio_set_map_data(bmd, bio, iov, iov_count, map_data ? 0 : 1); return bio; cleanup: - bio_for_each_segment(bvec, bio, i) - __free_page(bvec->bv_page); + if (!map_data) + bio_for_each_segment(bvec, bio, i) + __free_page(bvec->bv_page); bio_put(bio); out_bmd: @@ -601,10 +879,35 @@ out_bmd: return ERR_PTR(ret); } -static struct bio *__bio_map_user_iov(request_queue_t *q, +/** + * bio_copy_user - copy user data to bio + * @q: destination block queue + * @map_data: pointer to the rq_map_data holding pages (if necessary) + * @uaddr: start of user address + * @len: length in bytes + * @write_to_vm: bool indicating writing to pages or not + * @gfp_mask: memory allocation flags + * + * Prepares and returns a bio for indirect user io, bouncing data + * to/from kernel pages as necessary. Must be paired with + * call bio_uncopy_user() on io completion. + */ +struct bio *bio_copy_user(struct request_queue *q, struct rq_map_data *map_data, + unsigned long uaddr, unsigned int len, + int write_to_vm, gfp_t gfp_mask) +{ + struct sg_iovec iov; + + iov.iov_base = (void __user *)uaddr; + iov.iov_len = len; + + return bio_copy_user_iov(q, map_data, &iov, 1, write_to_vm, gfp_mask); +} + +static struct bio *__bio_map_user_iov(struct request_queue *q, struct block_device *bdev, struct sg_iovec *iov, int iov_count, - int write_to_vm) + int write_to_vm, gfp_t gfp_mask) { int i, j; int nr_pages = 0; @@ -621,22 +924,21 @@ static struct bio *__bio_map_user_iov(request_queue_t *q, 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 + * buffer must be aligned to at least hardsector size for now */ - if ((uaddr & queue_dma_alignment(q)) || (len & queue_dma_alignment(q))) + if (uaddr & queue_dma_alignment(q)) return ERR_PTR(-EINVAL); } if (!nr_pages) return ERR_PTR(-EINVAL); - bio = bio_alloc(GFP_KERNEL, nr_pages); + bio = bio_alloc(gfp_mask, nr_pages); if (!bio) return ERR_PTR(-ENOMEM); ret = -ENOMEM; - pages = kcalloc(nr_pages, sizeof(struct page *), GFP_KERNEL); + pages = kcalloc(nr_pages, sizeof(struct page *), gfp_mask); if (!pages) goto out; @@ -648,12 +950,8 @@ static struct bio *__bio_map_user_iov(request_queue_t *q, 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); - + ret = get_user_pages_fast(uaddr, local_nr_pages, + write_to_vm, &pages[cur_page]); if (ret < local_nr_pages) { ret = -EFAULT; goto out_unmap; @@ -714,46 +1012,48 @@ static struct bio *__bio_map_user_iov(request_queue_t *q, /** * bio_map_user - map user address into bio - * @q: the request_queue_t for the bio + * @q: the struct request_queue for the bio * @bdev: destination block device * @uaddr: start of user address * @len: length in bytes * @write_to_vm: bool indicating writing to pages or not + * @gfp_mask: memory allocation flags * * 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(request_queue_t *q, struct block_device *bdev, - unsigned long uaddr, unsigned int len, int write_to_vm) +struct bio *bio_map_user(struct request_queue *q, struct block_device *bdev, + unsigned long uaddr, unsigned int len, int write_to_vm, + gfp_t gfp_mask) { struct sg_iovec iov; iov.iov_base = (void __user *)uaddr; iov.iov_len = len; - return bio_map_user_iov(q, bdev, &iov, 1, write_to_vm); + return bio_map_user_iov(q, bdev, &iov, 1, write_to_vm, gfp_mask); } /** * bio_map_user_iov - map user sg_iovec table into bio - * @q: the request_queue_t for the bio + * @q: the struct request_queue 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 + * @gfp_mask: memory allocation flags * * 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 bio *bio_map_user_iov(struct request_queue *q, struct block_device *bdev, struct sg_iovec *iov, int iov_count, - int write_to_vm) + int write_to_vm, gfp_t gfp_mask) { struct bio *bio; - int len = 0, i; - - bio = __bio_map_user_iov(q, bdev, iov, iov_count, write_to_vm); + bio = __bio_map_user_iov(q, bdev, iov, iov_count, write_to_vm, + gfp_mask); if (IS_ERR(bio)) return bio; @@ -765,18 +1065,7 @@ struct bio *bio_map_user_iov(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; - - /* - * don't support partial mappings - */ - bio_endio(bio, bio->bi_size, 0); - bio_unmap_user(bio); - return ERR_PTR(-EINVAL); + return bio; } static void __bio_unmap_user(struct bio *bio) @@ -812,17 +1101,13 @@ 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) +static void bio_map_kern_endio(struct bio *bio, 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, +static struct bio *__bio_map_kern(struct request_queue *q, void *data, unsigned int len, gfp_t gfp_mask) { unsigned long kaddr = (unsigned long)data; @@ -861,7 +1146,7 @@ static struct bio *__bio_map_kern(request_queue_t *q, void *data, /** * bio_map_kern - map kernel address into bio - * @q: the request_queue_t for the bio + * @q: the struct request_queue for the bio * @data: pointer to buffer to map * @len: length in bytes * @gfp_mask: allocation flags for bio allocation @@ -869,7 +1154,7 @@ static struct bio *__bio_map_kern(request_queue_t *q, void *data, * 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, +struct bio *bio_map_kern(struct request_queue *q, void *data, unsigned int len, gfp_t gfp_mask) { struct bio *bio; @@ -888,6 +1173,67 @@ struct bio *bio_map_kern(request_queue_t *q, void *data, unsigned int len, return ERR_PTR(-EINVAL); } +static void bio_copy_kern_endio(struct bio *bio, int err) +{ + struct bio_vec *bvec; + const int read = bio_data_dir(bio) == READ; + struct bio_map_data *bmd = bio->bi_private; + int i; + char *p = bmd->sgvecs[0].iov_base; + + __bio_for_each_segment(bvec, bio, i, 0) { + char *addr = page_address(bvec->bv_page); + int len = bmd->iovecs[i].bv_len; + + if (read && !err) + memcpy(p, addr, len); + + __free_page(bvec->bv_page); + p += len; + } + + bio_free_map_data(bmd); + bio_put(bio); +} + +/** + * bio_copy_kern - copy kernel address into bio + * @q: the struct request_queue for the bio + * @data: pointer to buffer to copy + * @len: length in bytes + * @gfp_mask: allocation flags for bio and page allocation + * @reading: data direction is READ + * + * copy the kernel address into a bio suitable for io to a block + * device. Returns an error pointer in case of error. + */ +struct bio *bio_copy_kern(struct request_queue *q, void *data, unsigned int len, + gfp_t gfp_mask, int reading) +{ + struct bio *bio; + struct bio_vec *bvec; + int i; + + bio = bio_copy_user(q, NULL, (unsigned long)data, len, 1, gfp_mask); + if (IS_ERR(bio)) + return bio; + + if (!reading) { + void *p = data; + + bio_for_each_segment(bvec, bio, i) { + char *addr = page_address(bvec->bv_page); + + memcpy(addr, p, bvec->bv_len); + p += bvec->bv_len; + } + } + + bio->bi_end_io = bio_copy_kern_endio; + + return bio; +} + /* * bio_set_pages_dirty() and bio_check_pages_dirty() are support functions * for performing direct-IO in BIOs. @@ -954,16 +1300,16 @@ static void bio_release_pages(struct bio *bio) * run one bio_put() against the BIO. */ -static void bio_dirty_fn(void *data); +static void bio_dirty_fn(struct work_struct *work); -static DECLARE_WORK(bio_dirty_work, bio_dirty_fn, NULL); +static DECLARE_WORK(bio_dirty_work, bio_dirty_fn); static DEFINE_SPINLOCK(bio_dirty_lock); static struct bio *bio_dirty_list; /* * This runs in process context */ -static void bio_dirty_fn(void *data) +static void bio_dirty_fn(struct work_struct *work) { unsigned long flags; struct bio *bio; @@ -1016,34 +1362,26 @@ void bio_check_pages_dirty(struct bio *bio) /** * bio_endio - end I/O on a bio * @bio: bio - * @bytes_done: number of bytes completed * @error: error, if any * * Description: - * bio_endio() will end I/O on @bytes_done number of bytes. This may be - * just a partial part of the bio, or it may be the whole bio. bio_endio() - * is the preferred way to end I/O on a bio, it takes care of decrementing - * bi_size and clearing BIO_UPTODATE on error. @error is 0 on success, and - * and one of the established -Exxxx (-EIO, for instance) error values in - * case something went wrong. Noone should call bi_end_io() directly on - * a bio unless they own it and thus know that it has an end_io function. + * bio_endio() will end I/O on the whole bio. bio_endio() is the + * preferred way to end I/O on a bio, it takes care of clearing + * BIO_UPTODATE on error. @error is 0 on success, and and one of the + * established -Exxxx (-EIO, for instance) error values in case + * something went wrong. Noone should call bi_end_io() directly on a + * bio unless they own it and thus know that it has an end_io + * function. **/ -void bio_endio(struct bio *bio, unsigned int bytes_done, int error) +void bio_endio(struct bio *bio, int error) { if (error) clear_bit(BIO_UPTODATE, &bio->bi_flags); - - if (unlikely(bytes_done > bio->bi_size)) { - printk("%s: want %u bytes done, only %u left\n", __FUNCTION__, - bytes_done, bio->bi_size); - bytes_done = bio->bi_size; - } - - bio->bi_size -= bytes_done; - bio->bi_sector += (bytes_done >> 9); + else if (!test_bit(BIO_UPTODATE, &bio->bi_flags)) + error = -EIO; if (bio->bi_end_io) - bio->bi_end_io(bio, bytes_done, error); + bio->bi_end_io(bio, error); } void bio_pair_release(struct bio_pair *bp) @@ -1051,51 +1389,43 @@ void bio_pair_release(struct bio_pair *bp) if (atomic_dec_and_test(&bp->cnt)) { struct bio *master = bp->bio1.bi_private; - bio_endio(master, master->bi_size, bp->error); + bio_endio(master, bp->error); mempool_free(bp, bp->bio2.bi_private); } } -static int bio_pair_end_1(struct bio * bi, unsigned int done, int err) +static void bio_pair_end_1(struct bio *bi, int err) { struct bio_pair *bp = container_of(bi, struct bio_pair, bio1); if (err) bp->error = err; - if (bi->bi_size) - return 1; - bio_pair_release(bp); - return 0; } -static int bio_pair_end_2(struct bio * bi, unsigned int done, int err) +static void bio_pair_end_2(struct bio *bi, int err) { struct bio_pair *bp = container_of(bi, struct bio_pair, bio2); if (err) bp->error = err; - if (bi->bi_size) - return 1; - bio_pair_release(bp); - return 0; } /* * split a bio - only worry about a bio with a single page * in it's iovec */ -struct bio_pair *bio_split(struct bio *bi, mempool_t *pool, int first_sectors) +struct bio_pair *bio_split(struct bio *bi, int first_sectors) { - struct bio_pair *bp = mempool_alloc(pool, GFP_NOIO); + struct bio_pair *bp = mempool_alloc(bio_split_pool, GFP_NOIO); if (!bp) return bp; - blk_add_trace_pdu_int(bdev_get_queue(bi->bi_bdev), BLK_TA_SPLIT, bi, + trace_block_split(bdev_get_queue(bi->bi_bdev), bi, bi->bi_sector + first_sectors); BUG_ON(bi->bi_vcnt != 1); @@ -1124,45 +1454,69 @@ struct bio_pair *bio_split(struct bio *bi, mempool_t *pool, int first_sectors) bp->bio2.bi_end_io = bio_pair_end_2; bp->bio1.bi_private = bi; - bp->bio2.bi_private = pool; + bp->bio2.bi_private = bio_split_pool; + + if (bio_integrity(bi)) + bio_integrity_split(bi, bp, first_sectors); return bp; } - -/* - * create memory pools for biovec's in a bio_set. - * use the global biovec slabs created for general use. +/** + * bio_sector_offset - Find hardware sector offset in bio + * @bio: bio to inspect + * @index: bio_vec index + * @offset: offset in bv_page + * + * Return the number of hardware sectors between beginning of bio + * and an end point indicated by a bio_vec index and an offset + * within that vector's page. */ -static int biovec_create_pools(struct bio_set *bs, int pool_entries, int scale) +sector_t bio_sector_offset(struct bio *bio, unsigned short index, + unsigned int offset) { + unsigned int sector_sz = queue_hardsect_size(bio->bi_bdev->bd_disk->queue); + struct bio_vec *bv; + sector_t sectors; int i; - for (i = 0; i < BIOVEC_NR_POOLS; i++) { - struct biovec_slab *bp = bvec_slabs + i; - mempool_t **bvp = bs->bvec_pools + i; + sectors = 0; - if (pool_entries > 1 && i >= scale) - pool_entries >>= 1; + if (index >= bio->bi_idx) + index = bio->bi_vcnt - 1; - *bvp = mempool_create_slab_pool(pool_entries, bp->slab); - if (!*bvp) - return -ENOMEM; + __bio_for_each_segment(bv, bio, i, 0) { + if (i == index) { + if (offset > bv->bv_offset) + sectors += (offset - bv->bv_offset) / sector_sz; + break; + } + + sectors += bv->bv_len / sector_sz; } - return 0; + + return sectors; } +EXPORT_SYMBOL(bio_sector_offset); -static void biovec_free_pools(struct bio_set *bs) +/* + * create memory pools for biovec's in a bio_set. + * use the global biovec slabs created for general use. + */ +static int biovec_create_pools(struct bio_set *bs, int pool_entries) { - int i; + struct biovec_slab *bp = bvec_slabs + BIOVEC_MAX_IDX; - for (i = 0; i < BIOVEC_NR_POOLS; i++) { - mempool_t *bvp = bs->bvec_pools[i]; + bs->bvec_pool = mempool_create_slab_pool(pool_entries, bp->slab); + if (!bs->bvec_pool) + return -ENOMEM; - if (bvp) - mempool_destroy(bvp); - } + return 0; +} +static void biovec_free_pools(struct bio_set *bs) +{ + mempool_destroy(bs->bvec_pool); } void bioset_free(struct bio_set *bs) @@ -1170,23 +1524,51 @@ void bioset_free(struct bio_set *bs) if (bs->bio_pool) mempool_destroy(bs->bio_pool); + bioset_integrity_free(bs); biovec_free_pools(bs); + bio_put_slab(bs); kfree(bs); } -struct bio_set *bioset_create(int bio_pool_size, int bvec_pool_size, int scale) +/** + * bioset_create - Create a bio_set + * @pool_size: Number of bio and bio_vecs to cache in the mempool + * @front_pad: Number of bytes to allocate in front of the returned bio + * + * Description: + * Set up a bio_set to be used with @bio_alloc_bioset. Allows the caller + * to ask for a number of bytes to be allocated in front of the bio. + * Front pad allocation is useful for embedding the bio inside + * another structure, to avoid allocating extra data to go with the bio. + * Note that the bio must be embedded at the END of that structure always, + * or things will break badly. + */ +struct bio_set *bioset_create(unsigned int pool_size, unsigned int front_pad) { - struct bio_set *bs = kzalloc(sizeof(*bs), GFP_KERNEL); + unsigned int back_pad = BIO_INLINE_VECS * sizeof(struct bio_vec); + struct bio_set *bs; + bs = kzalloc(sizeof(*bs), GFP_KERNEL); if (!bs) return NULL; - bs->bio_pool = mempool_create_slab_pool(bio_pool_size, bio_slab); + bs->front_pad = front_pad; + + bs->bio_slab = bio_find_or_create_slab(front_pad + back_pad); + if (!bs->bio_slab) { + kfree(bs); + return NULL; + } + + bs->bio_pool = mempool_create_slab_pool(pool_size, bs->bio_slab); if (!bs->bio_pool) goto bad; - if (!biovec_create_pools(bs, bvec_pool_size, scale)) + if (bioset_integrity_create(bs, pool_size)) + goto bad; + + if (!biovec_create_pools(bs, pool_size)) return bs; bad: @@ -1204,44 +1586,22 @@ static void __init biovec_init_slabs(void) size = bvs->nr_vecs * sizeof(struct bio_vec); bvs->slab = kmem_cache_create(bvs->name, size, 0, - SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL, NULL); + SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL); } } static int __init init_bio(void) { - int megabytes, bvec_pool_entries; - int scale = BIOVEC_NR_POOLS; - - bio_slab = kmem_cache_create("bio", sizeof(struct bio), 0, - SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL, NULL); + bio_slab_max = 2; + bio_slab_nr = 0; + bio_slabs = kzalloc(bio_slab_max * sizeof(struct bio_slab), GFP_KERNEL); + if (!bio_slabs) + panic("bio: can't allocate bios\n"); + bio_integrity_init_slab(); biovec_init_slabs(); - megabytes = nr_free_pages() >> (20 - PAGE_SHIFT); - - /* - * find out where to start scaling - */ - if (megabytes <= 16) - scale = 0; - else if (megabytes <= 32) - scale = 1; - else if (megabytes <= 64) - scale = 2; - else if (megabytes <= 96) - scale = 3; - else if (megabytes <= 128) - scale = 4; - - /* - * Limit number of entries reserved -- mempools are only used when - * the system is completely unable to allocate memory, so we only - * need enough to make progress. - */ - bvec_pool_entries = 1 + scale; - - fs_bio_set = bioset_create(BIO_POOL_SIZE, bvec_pool_entries, scale); + fs_bio_set = bioset_create(BIO_POOL_SIZE, 0); if (!fs_bio_set) panic("bio: can't allocate bios\n"); @@ -1256,6 +1616,7 @@ static int __init init_bio(void) subsys_initcall(init_bio); EXPORT_SYMBOL(bio_alloc); +EXPORT_SYMBOL(bio_kmalloc); EXPORT_SYMBOL(bio_put); EXPORT_SYMBOL(bio_free); EXPORT_SYMBOL(bio_endio); @@ -1263,16 +1624,15 @@ EXPORT_SYMBOL(bio_init); EXPORT_SYMBOL(__bio_clone); EXPORT_SYMBOL(bio_clone); EXPORT_SYMBOL(bio_phys_segments); -EXPORT_SYMBOL(bio_hw_segments); EXPORT_SYMBOL(bio_add_page); EXPORT_SYMBOL(bio_add_pc_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_copy_kern); EXPORT_SYMBOL(bio_pair_release); EXPORT_SYMBOL(bio_split); -EXPORT_SYMBOL(bio_split_pool); EXPORT_SYMBOL(bio_copy_user); EXPORT_SYMBOL(bio_uncopy_user); EXPORT_SYMBOL(bioset_create);