Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753421Ab3FICUP (ORCPT ); Sat, 8 Jun 2013 22:20:15 -0400 Received: from mail-pb0-f47.google.com ([209.85.160.47]:58811 "EHLO mail-pb0-f47.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753189Ab3FICUG (ORCPT ); Sat, 8 Jun 2013 22:20:06 -0400 From: Kent Overstreet To: axboe@kernel.dk, tytso@mit.edu, linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org Cc: Kent Overstreet , "Martin K. Petersen" , Alasdair Kergon , dm-devel@redhat.com Subject: [PATCH 20/26] block: Don't save/copy bvec array anymore, share when cloning Date: Sat, 8 Jun 2013 19:19:02 -0700 Message-Id: <1370744348-15407-21-git-send-email-koverstreet@google.com> X-Mailer: git-send-email 1.8.3.rc1 In-Reply-To: <1370744348-15407-1-git-send-email-koverstreet@google.com> References: <1370744348-15407-1-git-send-email-koverstreet@google.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 12241 Lines: 437 Now that drivers have been converted to the bvec_iter primitives, they shouldn't be modifying the biovec anymore and thus saving it is unnecessary - code that was previously making a backup of the bvec array can now just save bio->bi_iter. Also, when cloning bios we can usually just reuse the original bio's bvec array. For code that does need to modify the clone's biovec (the bounce buffer code, mainly), add bio_clone_biovec(). Signed-off-by: Kent Overstreet Cc: Jens Axboe Cc: "Martin K. Petersen" Cc: Alasdair Kergon Cc: dm-devel@redhat.com --- drivers/md/bcache/request.c | 2 - drivers/md/bcache/request.h | 1 - drivers/md/dm-bio-record.h | 25 -------- fs/bio-integrity.c | 12 +--- fs/bio.c | 152 +++++++++++++++++++------------------------- include/linux/bio.h | 1 + mm/bounce.c | 1 + 7 files changed, 71 insertions(+), 123 deletions(-) diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c index ca513d4..3c95b0e 100644 --- a/drivers/md/bcache/request.c +++ b/drivers/md/bcache/request.c @@ -677,8 +677,6 @@ static void do_bio_hook(struct search *s) struct bio *bio = &s->bio.bio; bio_init(bio); - bio->bi_io_vec = s->bv; - bio->bi_max_vecs = BIO_MAX_PAGES; __bio_clone(bio, s->orig_bio); bio->bi_end_io = request_endio; bio->bi_private = &s->cl; diff --git a/drivers/md/bcache/request.h b/drivers/md/bcache/request.h index 7873029b..5c1ece2 100644 --- a/drivers/md/bcache/request.h +++ b/drivers/md/bcache/request.h @@ -26,7 +26,6 @@ struct search { /* Anything past op->keys won't get zeroed in do_bio_hook */ struct btree_op op; - struct bio_vec bv[BIO_MAX_PAGES]; }; void bch_cache_read_endio(struct bio *, int); diff --git a/drivers/md/dm-bio-record.h b/drivers/md/dm-bio-record.h index 4f46e8e..dd36461 100644 --- a/drivers/md/dm-bio-record.h +++ b/drivers/md/dm-bio-record.h @@ -17,49 +17,24 @@ * original bio state. */ -struct dm_bio_vec_details { -#if PAGE_SIZE < 65536 - __u16 bv_len; - __u16 bv_offset; -#else - unsigned bv_len; - unsigned bv_offset; -#endif -}; - struct dm_bio_details { struct block_device *bi_bdev; unsigned long bi_flags; struct bvec_iter bi_iter; - struct dm_bio_vec_details bi_io_vec[BIO_MAX_PAGES]; }; static inline void dm_bio_record(struct dm_bio_details *bd, struct bio *bio) { - unsigned i; - bd->bi_bdev = bio->bi_bdev; bd->bi_flags = bio->bi_flags; bd->bi_iter = bio->bi_iter; - - for (i = 0; i < bio->bi_vcnt; i++) { - bd->bi_io_vec[i].bv_len = bio->bi_io_vec[i].bv_len; - bd->bi_io_vec[i].bv_offset = bio->bi_io_vec[i].bv_offset; - } } static inline void dm_bio_restore(struct dm_bio_details *bd, struct bio *bio) { - unsigned i; - bio->bi_bdev = bd->bi_bdev; bio->bi_flags = bd->bi_flags; bio->bi_iter = bd->bi_iter; - - for (i = 0; i < bio->bi_vcnt; i++) { - bio->bi_io_vec[i].bv_len = bd->bi_io_vec[i].bv_len; - bio->bi_io_vec[i].bv_offset = bd->bi_io_vec[i].bv_offset; - } } #endif diff --git a/fs/bio-integrity.c b/fs/bio-integrity.c index 72fa942..0c466e6 100644 --- a/fs/bio-integrity.c +++ b/fs/bio-integrity.c @@ -594,17 +594,11 @@ int bio_integrity_clone(struct bio *bio, struct bio *bio_src, struct bio_integrity_payload *bip_src = bio_src->bi_integrity; struct bio_integrity_payload *bip; - BUG_ON(bip_src == NULL); - - bip = bio_integrity_alloc(bio, gfp_mask, bip_src->bip_vcnt); - + bip = bio_integrity_alloc(bio, gfp_mask, 0); if (bip == NULL) - return -EIO; - - memcpy(bip->bip_vec, bip_src->bip_vec, - bip_src->bip_vcnt * sizeof(struct bio_vec)); + return -ENOMEM; - bip->bip_vcnt = bip_src->bip_vcnt; + bip->bip_vec = bip_src->bip_vec; bip->bip_iter = bip_src->bip_iter; return 0; diff --git a/fs/bio.c b/fs/bio.c index ee033a7..10d71cf 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -544,8 +544,7 @@ EXPORT_SYMBOL(bio_phys_segments); */ void __bio_clone(struct bio *bio, struct bio *bio_src) { - memcpy(bio->bi_io_vec, bio_src->bi_io_vec, - bio_src->bi_max_vecs * sizeof(struct bio_vec)); + BUG_ON(bio->bi_pool && BIO_POOL_IDX(bio) != BIO_POOL_NONE); /* * most users will be overriding ->bi_bdev with a new target, @@ -554,8 +553,8 @@ void __bio_clone(struct bio *bio, struct bio *bio_src) bio->bi_bdev = bio_src->bi_bdev; bio->bi_flags |= 1 << BIO_CLONED; bio->bi_rw = bio_src->bi_rw; - bio->bi_vcnt = bio_src->bi_vcnt; bio->bi_iter = bio_src->bi_iter; + bio->bi_io_vec = bio_src->bi_io_vec; } EXPORT_SYMBOL(__bio_clone); @@ -572,7 +571,7 @@ struct bio *bio_clone_bioset(struct bio *bio, gfp_t gfp_mask, { struct bio *b; - b = bio_alloc_bioset(gfp_mask, bio->bi_max_vecs, bs); + b = bio_alloc_bioset(gfp_mask, 0, bs); if (!b) return NULL; @@ -594,6 +593,50 @@ struct bio *bio_clone_bioset(struct bio *bio, gfp_t gfp_mask, EXPORT_SYMBOL(bio_clone_bioset); /** + * bio_clone_biovec: Given a cloned bio, give the clone its own copy of the + * biovec + * @bio: cloned bio + * + * @bio must have been allocated from a bioset - i.e. returned from + * bio_clone_bioset() + */ +int bio_clone_biovec(struct bio *bio, gfp_t gfp_mask) +{ + unsigned long idx = BIO_POOL_NONE; + unsigned nr_iovecs = 0; + struct bio_vec bv, *bvl = NULL; + struct bvec_iter iter; + + BUG_ON(!bio->bi_pool); + BUG_ON(BIO_POOL_IDX(bio) != BIO_POOL_NONE); + + bio_for_each_segment(bv, bio, iter) + nr_iovecs++; + + if (nr_iovecs > BIO_INLINE_VECS) { + bvl = bvec_alloc(gfp_mask, nr_iovecs, &idx, + bio->bi_pool->bvec_pool); + if (!bvl) + return -ENOMEM; + } else if (nr_iovecs) { + bvl = bio->bi_inline_vecs; + } + + bio_for_each_segment(bv, bio, iter) + bvl[bio->bi_vcnt++] = bv; + + bio->bi_io_vec = bvl; + bio->bi_iter.bi_idx = 0; + bio->bi_iter.bi_bvec_done = 0; + + bio->bi_flags &= BIO_POOL_MASK - 1; + bio->bi_flags |= idx << BIO_POOL_OFFSET; + + return 0; +} +EXPORT_SYMBOL(bio_clone_biovec); + +/** * bio_get_nr_vecs - return approx number of vecs * @bdev: I/O target * @@ -918,60 +961,33 @@ void bio_copy_data(struct bio *dst, struct bio *src) EXPORT_SYMBOL(bio_copy_data); struct bio_map_data { - struct bio_vec *iovecs; - struct sg_iovec *sgvecs; int nr_sgvecs; int is_our_pages; + struct sg_iovec sgvecs[]; }; 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, unsigned int iov_count, gfp_t gfp_mask) { - struct bio_map_data *bmd; - if (iov_count > UIO_MAXIOV) return NULL; - bmd = kmalloc(sizeof(*bmd), gfp_mask); - if (!bmd) - return NULL; - - 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; + return kmalloc(sizeof(struct bio_map_data) + + sizeof(struct sg_iovec) * iov_count, gfp_mask); } -static int __bio_copy_iov(struct bio *bio, struct bio_vec *iovecs, - struct sg_iovec *iov, int iov_count, +static int __bio_copy_iov(struct bio *bio, struct sg_iovec *iov, int iov_count, int to_user, int from_user, int do_free_page) { int ret = 0, i; @@ -981,7 +997,7 @@ static int __bio_copy_iov(struct bio *bio, struct bio_vec *iovecs, bio_for_each_segment_all(bvec, bio, i) { char *bv_addr = page_address(bvec->bv_page); - unsigned int bv_len = iovecs[i].bv_len; + unsigned int bv_len = bvec->bv_len; while (bv_len && iov_idx < iov_count) { unsigned int bytes; @@ -1035,10 +1051,10 @@ int bio_uncopy_user(struct bio *bio) int ret = 0; if (!bio_flagged(bio, BIO_NULL_MAPPED)) - ret = __bio_copy_iov(bio, bmd->iovecs, bmd->sgvecs, - bmd->nr_sgvecs, bio_data_dir(bio) == READ, + ret = __bio_copy_iov(bio, bmd->sgvecs, bmd->nr_sgvecs, + bio_data_dir(bio) == READ, 0, bmd->is_our_pages); - bio_free_map_data(bmd); + kfree(bmd); bio_put(bio); return ret; } @@ -1152,7 +1168,7 @@ struct bio *bio_copy_user_iov(struct request_queue *q, */ 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); + ret = __bio_copy_iov(bio, iov, iov_count, 0, 1, 0); if (ret) goto cleanup; } @@ -1166,7 +1182,7 @@ cleanup: bio_put(bio); out_bmd: - bio_free_map_data(bmd); + kfree(bmd); return ERR_PTR(ret); } @@ -1483,16 +1499,15 @@ static void bio_copy_kern_endio(struct bio *bio, int err) bio_for_each_segment_all(bvec, bio, i) { char *addr = page_address(bvec->bv_page); - int len = bmd->iovecs[i].bv_len; if (read) - memcpy(p, addr, len); + memcpy(p, addr, bvec->bv_len); __free_page(bvec->bv_page); - p += len; + p += bvec->bv_len; } - bio_free_map_data(bmd); + kfree(bmd); bio_put(bio); } @@ -1720,60 +1735,25 @@ EXPORT_SYMBOL(bio_endio); * Allocates and returns a new bio which represents @sectors from the start of * @bio, and updates @bio to represent the remaining sectors. * - * If bio_sectors(@bio) was less than or equal to @sectors, returns @bio - * unchanged. + * The newly allocated bio will point to @bio's bi_io_vec; it is the caller's + * responsibility to ensure that @bio is not freed before the split. */ struct bio *bio_split(struct bio *bio, int sectors, gfp_t gfp, struct bio_set *bs) { - unsigned vcnt = 0, nbytes = sectors << 9; - struct bio_vec bv; - struct bvec_iter iter; struct bio *split = NULL; BUG_ON(sectors <= 0); BUG_ON(sectors >= bio_sectors(bio)); - if (bio->bi_rw & REQ_DISCARD) { - split = bio_alloc_bioset(gfp, 1, bs); - goto out; - } - - bio_for_each_segment(bv, bio, iter) { - vcnt++; - - if (nbytes <= bv.bv_len) - break; - - nbytes -= bv.bv_len; - } - - split = bio_alloc_bioset(gfp, vcnt, bs); + split = bio_clone_bioset(bio, gfp, bs); if (!split) return NULL; - bio_for_each_segment(bv, bio, iter) { - split->bi_io_vec[split->bi_vcnt++] = bv; - - if (split->bi_vcnt == vcnt) - break; - } + split->bi_iter.bi_size = sectors << 9; - split->bi_io_vec[split->bi_vcnt - 1].bv_len = nbytes; -out: - split->bi_bdev = bio->bi_bdev; - split->bi_iter.bi_sector = bio->bi_iter.bi_sector; - split->bi_iter.bi_size = sectors << 9; - split->bi_rw = bio->bi_rw; - - if (bio_integrity(bio)) { - if (bio_integrity_clone(split, bio, gfp)) { - bio_put(split); - return NULL; - } - - bio_integrity_trim(split, 0, bio_sectors(split)); - } + if (bio_integrity(split)) + bio_integrity_trim(split, 0, sectors); bio_advance(bio, split->bi_iter.bi_size); diff --git a/include/linux/bio.h b/include/linux/bio.h index aa015ee..444cc91 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -293,6 +293,7 @@ extern void bio_put(struct bio *); extern void __bio_clone(struct bio *, struct bio *); extern struct bio *bio_clone_bioset(struct bio *, gfp_t, struct bio_set *bs); +extern int bio_clone_biovec(struct bio *bio, gfp_t gfp_mask); extern struct bio_set *fs_bio_set; diff --git a/mm/bounce.c b/mm/bounce.c index 4525e3d..985fe23 100644 --- a/mm/bounce.c +++ b/mm/bounce.c @@ -209,6 +209,7 @@ static void __blk_queue_bounce(struct request_queue *q, struct bio **bio_orig, return; bounce: bio = bio_clone_bioset(*bio_orig, GFP_NOIO, fs_bio_set); + bio_clone_biovec(bio, GFP_NOIO); bio_for_each_segment_all(to, bio, i) { struct page *page = to->bv_page; -- 1.8.3.rc1 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/