Convert the block layer's bio code to use iov_iter_extract_pages() instead of iov_iter_get_pages(). This will pin pages or leave them unaltered rather than getting a ref on them as appropriate to the source iterator. A field, bi_cleanup_mode, is added to the bio struct that gets set by iov_iter_extract_pages() with FOLL_* flags indicating what cleanup is necessary. FOLL_GET -> put_page(), FOLL_PIN -> unpin_user_page(). Other flags could also be used in future. Newly allocated bio structs have bi_cleanup_mode set to FOLL_GET to indicate that attached pages are ref'd by default. Cloning sets it to 0. __bio_iov_iter_get_pages() overrides it to what iov_iter_extract_pages() indicates. [!] Note that this is tested a bit with ext4, but nothing else. Signed-off-by: David Howells <dhowells@xxxxxxxxxx> cc: Al Viro <viro@xxxxxxxxxxxxxxxxxx> cc: Jens Axboe <axboe@xxxxxxxxx> cc: Christoph Hellwig <hch@xxxxxx> cc: Matthew Wilcox <willy@xxxxxxxxxxxxx> cc: Logan Gunthorpe <logang@xxxxxxxxxxxx> --- block/bio.c | 47 +++++++++++++++++++++++++++++++++------------ include/linux/blk_types.h | 1 + 2 files changed, 35 insertions(+), 13 deletions(-) diff --git a/block/bio.c b/block/bio.c index 5f96fcae3f75..eafcbeba0bab 100644 --- a/block/bio.c +++ b/block/bio.c @@ -243,6 +243,11 @@ static void bio_free(struct bio *bio) * Users of this function have their own bio allocation. Subsequently, * they must remember to pair any call to bio_init() with bio_uninit() * when IO has completed, or when the bio is released. + * + * We set the initial assumption that pages attached to the bio will be + * released with put_page() by setting bi_cleanup_mode to FOLL_GET, but this + * should be set to FOLL_PIN if the page should be unpinned instead; if the + * pages should not be put or unpinned, this should be set to 0 */ void bio_init(struct bio *bio, struct block_device *bdev, struct bio_vec *table, unsigned short max_vecs, blk_opf_t opf) @@ -274,6 +279,7 @@ void bio_init(struct bio *bio, struct block_device *bdev, struct bio_vec *table, #ifdef CONFIG_BLK_DEV_INTEGRITY bio->bi_integrity = NULL; #endif + bio->bi_cleanup_mode = FOLL_GET; bio->bi_vcnt = 0; atomic_set(&bio->__bi_remaining, 1); @@ -302,6 +308,7 @@ void bio_reset(struct bio *bio, struct block_device *bdev, blk_opf_t opf) { bio_uninit(bio); memset(bio, 0, BIO_RESET_BYTES); + bio->bi_cleanup_mode = FOLL_GET; atomic_set(&bio->__bi_remaining, 1); bio->bi_bdev = bdev; if (bio->bi_bdev) @@ -814,6 +821,7 @@ static int __bio_clone(struct bio *bio, struct bio *bio_src, gfp_t gfp) bio_set_flag(bio, BIO_CLONED); bio->bi_ioprio = bio_src->bi_ioprio; bio->bi_iter = bio_src->bi_iter; + bio->bi_cleanup_mode = 0; if (bio->bi_bdev) { if (bio->bi_bdev == bio_src->bi_bdev && @@ -1168,6 +1176,18 @@ bool bio_add_folio(struct bio *bio, struct folio *folio, size_t len, return bio_add_page(bio, &folio->page, len, off) > 0; } +/* + * Clean up a page according to the mode indicated by iov_iter_extract_pages(), + * where the page is may be pinned or may have a ref taken on it. + */ +static void bio_release_page(struct bio *bio, struct page *page) +{ + if (bio->bi_cleanup_mode & FOLL_PIN) + unpin_user_page(page); + if (bio->bi_cleanup_mode & FOLL_GET) + put_page(page); +} + void __bio_release_pages(struct bio *bio, bool mark_dirty) { struct bvec_iter_all iter_all; @@ -1176,7 +1196,7 @@ void __bio_release_pages(struct bio *bio, bool mark_dirty) bio_for_each_segment_all(bvec, bio, iter_all) { if (mark_dirty && !PageCompound(bvec->bv_page)) set_page_dirty_lock(bvec->bv_page); - put_page(bvec->bv_page); + bio_release_page(bio, bvec->bv_page); } } EXPORT_SYMBOL_GPL(__bio_release_pages); @@ -1213,7 +1233,7 @@ static int bio_iov_add_page(struct bio *bio, struct page *page, } if (same_page) - put_page(page); + bio_release_page(bio, page); return 0; } @@ -1227,7 +1247,7 @@ static int bio_iov_add_zone_append_page(struct bio *bio, struct page *page, queue_max_zone_append_sectors(q), &same_page) != len) return -EINVAL; if (same_page) - put_page(page); + bio_release_page(bio, page); return 0; } @@ -1238,10 +1258,10 @@ static int bio_iov_add_zone_append_page(struct bio *bio, struct page *page, * @bio: bio to add pages to * @iter: iov iterator describing the region to be mapped * - * Pins pages from *iter and appends them to @bio's bvec array. The - * pages will have to be released using put_page() when done. - * For multi-segment *iter, this function only adds pages from the - * next non-empty segment of the iov iterator. + * Pins pages from *iter and appends them to @bio's bvec array. The pages will + * have to be released using put_page() or unpin_user_page() when done as + * according to bi_cleanup_mode. For multi-segment *iter, this function only + * adds pages from the next non-empty segment of the iov iterator. */ static int __bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter) { @@ -1273,9 +1293,10 @@ static int __bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter) * result to ensure the bio's total size is correct. The remainder of * the iov data will be picked up in the next bio iteration. */ - size = iov_iter_get_pages(iter, pages, - UINT_MAX - bio->bi_iter.bi_size, - nr_pages, &offset, gup_flags); + size = iov_iter_extract_pages(iter, &pages, + UINT_MAX - bio->bi_iter.bi_size, + nr_pages, gup_flags, + &offset, &bio->bi_cleanup_mode); if (unlikely(size <= 0)) return size ? size : -EFAULT; @@ -1308,7 +1329,7 @@ static int __bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter) iov_iter_revert(iter, left); out: while (i < nr_pages) - put_page(pages[i++]); + bio_release_page(bio, pages[i++]); return ret; } @@ -1489,8 +1510,8 @@ void bio_set_pages_dirty(struct bio *bio) * the BIO and re-dirty the pages in process context. * * It is expected that bio_check_pages_dirty() will wholly own the BIO from - * here on. It will run one put_page() against each page and will run one - * bio_put() against the BIO. + * here on. It will run one put_page() or unpin_user_page() against each page + * and will run one bio_put() against the BIO. */ static void bio_dirty_fn(struct work_struct *work); diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h index 99be590f952f..883f873a01ef 100644 --- a/include/linux/blk_types.h +++ b/include/linux/blk_types.h @@ -289,6 +289,7 @@ struct bio { #endif }; + unsigned int bi_cleanup_mode; /* How to clean up pages */ unsigned short bi_vcnt; /* how many bio_vec's */ /*