From patchwork Mon Aug 31 07:14:39 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: John Hubbard X-Patchwork-Id: 11745599 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id C249E1709 for ; Mon, 31 Aug 2020 07:15:12 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 9D8B9208DB for ; Mon, 31 Aug 2020 07:15:12 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (2048-bit key) header.d=nvidia.com header.i=@nvidia.com header.b="biJEmAt8" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1728001AbgHaHOx (ORCPT ); Mon, 31 Aug 2020 03:14:53 -0400 Received: from hqnvemgate26.nvidia.com ([216.228.121.65]:9943 "EHLO hqnvemgate26.nvidia.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1725978AbgHaHOs (ORCPT ); Mon, 31 Aug 2020 03:14:48 -0400 Received: from hqpgpgate102.nvidia.com (Not Verified[216.228.121.13]) by hqnvemgate26.nvidia.com (using TLS: TLSv1.2, DES-CBC3-SHA) id ; Mon, 31 Aug 2020 00:14:33 -0700 Received: from hqmail.nvidia.com ([172.20.161.6]) by hqpgpgate102.nvidia.com (PGP Universal service); Mon, 31 Aug 2020 00:14:47 -0700 X-PGP-Universal: processed; by hqpgpgate102.nvidia.com on Mon, 31 Aug 2020 00:14:47 -0700 Received: from HQMAIL109.nvidia.com (172.20.187.15) by HQMAIL101.nvidia.com (172.20.187.10) with Microsoft SMTP Server (TLS) id 15.0.1473.3; Mon, 31 Aug 2020 07:14:46 +0000 Received: from rnnvemgw01.nvidia.com (10.128.109.123) by HQMAIL109.nvidia.com (172.20.187.15) with Microsoft SMTP Server (TLS) id 15.0.1473.3 via Frontend Transport; Mon, 31 Aug 2020 07:14:46 +0000 Received: from sandstorm.nvidia.com (Not Verified[10.2.61.194]) by rnnvemgw01.nvidia.com with Trustwave SEG (v7,5,8,10121) id ; Mon, 31 Aug 2020 00:14:46 -0700 From: John Hubbard To: Andrew Morton CC: Alexander Viro , Christoph Hellwig , Ilya Dryomov , Jens Axboe , , , , , LKML , John Hubbard Subject: [PATCH v3 3/3] bio: convert get_user_pages_fast() --> pin_user_pages_fast() Date: Mon, 31 Aug 2020 00:14:39 -0700 Message-ID: <20200831071439.1014766-4-jhubbard@nvidia.com> X-Mailer: git-send-email 2.28.0 In-Reply-To: <20200831071439.1014766-1-jhubbard@nvidia.com> References: <20200831071439.1014766-1-jhubbard@nvidia.com> MIME-Version: 1.0 X-NVConfidentiality: public DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=nvidia.com; s=n1; t=1598858073; bh=Mgmhx7MxlormERSy5m5hA+CT8AL2TlMSJ/1FsF2TodE=; h=X-PGP-Universal:From:To:CC:Subject:Date:Message-ID:X-Mailer: In-Reply-To:References:MIME-Version:X-NVConfidentiality: Content-Transfer-Encoding:Content-Type; b=biJEmAt8Fu1372bMrTva0n0kV/jdU2U+h6V3BFWjbxkkXC3u0EhOS9Cs7IqImnOyp RyiXOdB2mjs8+MKyD70Q7RTjgysLpH0T6ePrXVhb+ijWxcJG9drwtSD7rIbQ5A4FLP ivcLNunNfh10/Oez+AU8BEVv1kIBOv8FKZwv3/4Pv9IvkQvZKuC3HwiD/TTg2KxTyp 64PGWqnU/PG4L3niPjVPNzmR1pw+6k6/z9ay8M+qZG+sV2+SJOYpwjjuDH6VjbKbRE 08/AJHB/QfakVELyqZJrnC5vfw75HR2jEgY8uR7HdZ2DabIMTnpuoQ5w6Qb7Q97Zvg /J5Z/YHhe05Pw== Sender: linux-xfs-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-xfs@vger.kernel.org Change generic block/bio Direct IO routines, to acquire FOLL_PIN user pages via the recently added routines: iov_iter_pin_pages() iov_iter_pin_pages_alloc() pin_page() This effectively converts several file systems (ext4, for example) that use the common Direct IO routines. Change the corresponding page release calls from put_page() to unpin_user_page(). Change bio_release_pages() to handle FOLL_PIN pages. In fact, after this patch, that is the *only* type of pages that bio_release_pages() handles. Design notes ============ Quite a few approaches have been considered over the years. This one is inspired by Christoph Hellwig's July, 2019 observation that there are only 5 ITER_ types, and we can simplify handling of them for Direct IO [1]. Accordingly, this patch implements the following pseudocode: Direct IO behavior: ITER_IOVEC: pin_user_pages_fast(); break; ITER_PIPE: for each page: pin_page(); break; ITER_KVEC: // already elevated page refcount, leave alone ITER_BVEC: // already elevated page refcount, leave alone ITER_DISCARD: // discard return -EFAULT or -ENVALID; ...which works for callers that already have sorted out which case they are in. Such as, Direct IO in the block/bio layers. Note that this does leave ITER_KVEC and ITER_BVEC unconverted, for now. Page acquisition: The iov_iter_get_pages*() routines above are at just the right level in the call stack: the callers already know which system to use, and so it's a small change to just drop in the replacement routines. And it's a fan-in/fan-out point: block/bio call sites for Direct IO funnel their page acquisitions through the iov_iter_get_pages*() routines, and there are many other callers of those. And we can't convert all of the callers at once--too many subsystems are involved, and it would be a too large and too risky patch. Page release: there are already separate release routines: put_page() vs. unpin_user_page(), so it's already done there. [1] https://lore.kernel.org/kvm/20190724061750.GA19397@infradead.org/ [2] "Explicit pinning of user-space pages": https://lwn.net/Articles/807108/ Cc: Christoph Hellwig Cc: Al Viro Signed-off-by: John Hubbard --- block/bio.c | 24 ++++++++++++------------ block/blk-map.c | 6 +++--- fs/direct-io.c | 28 ++++++++++++++-------------- fs/iomap/direct-io.c | 2 +- 4 files changed, 30 insertions(+), 30 deletions(-) diff --git a/block/bio.c b/block/bio.c index a9931f23d933..76c5843f6050 100644 --- a/block/bio.c +++ b/block/bio.c @@ -955,7 +955,7 @@ 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); + unpin_user_page(bvec->bv_page); } } EXPORT_SYMBOL_GPL(bio_release_pages); @@ -986,9 +986,9 @@ static int __bio_iov_bvec_add_pages(struct bio *bio, struct iov_iter *iter) * @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. + * pages will have to be released using put_page() or unpin_user_page() when + * done. 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) { @@ -1009,7 +1009,7 @@ static int __bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter) BUILD_BUG_ON(PAGE_PTRS_PER_BVEC < 2); pages += entries_left * (PAGE_PTRS_PER_BVEC - 1); - size = iov_iter_get_pages(iter, pages, LONG_MAX, nr_pages, &offset); + size = iov_iter_pin_pages(iter, pages, LONG_MAX, nr_pages, &offset); if (unlikely(size <= 0)) return size ? size : -EFAULT; @@ -1020,7 +1020,7 @@ static int __bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter) if (__bio_try_merge_page(bio, page, len, offset, &same_page)) { if (same_page) - put_page(page); + unpin_user_page(page); } else { if (WARN_ON_ONCE(bio_full(bio, len))) return -EINVAL; @@ -1056,7 +1056,7 @@ static int __bio_iov_append_get_pages(struct bio *bio, struct iov_iter *iter) BUILD_BUG_ON(PAGE_PTRS_PER_BVEC < 2); pages += entries_left * (PAGE_PTRS_PER_BVEC - 1); - size = iov_iter_get_pages(iter, pages, LONG_MAX, nr_pages, &offset); + size = iov_iter_pin_pages(iter, pages, LONG_MAX, nr_pages, &offset); if (unlikely(size <= 0)) return size ? size : -EFAULT; @@ -1069,7 +1069,7 @@ static int __bio_iov_append_get_pages(struct bio *bio, struct iov_iter *iter) max_append_sectors, &same_page) != len) return -EINVAL; if (same_page) - put_page(page); + unpin_user_page(page); offset = 0; } @@ -1113,8 +1113,8 @@ int bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter) } else { if (is_bvec) ret = __bio_iov_bvec_add_pages(bio, iter); - else - ret = __bio_iov_iter_get_pages(bio, iter); + else + ret = __bio_iov_iter_get_pages(bio, iter); } } while (!ret && iov_iter_count(iter) && !bio_full(bio, 0)); @@ -1326,8 +1326,8 @@ 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 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/block/blk-map.c b/block/blk-map.c index 6e804892d5ec..2057a8e5b4bc 100644 --- a/block/blk-map.c +++ b/block/blk-map.c @@ -275,7 +275,7 @@ static struct bio *bio_map_user_iov(struct request_queue *q, size_t offs, added = 0; int npages; - bytes = iov_iter_get_pages_alloc(iter, &pages, LONG_MAX, &offs); + bytes = iov_iter_pin_pages_alloc(iter, &pages, LONG_MAX, &offs); if (unlikely(bytes <= 0)) { ret = bytes ? bytes : -EFAULT; goto out_unmap; @@ -298,7 +298,7 @@ static struct bio *bio_map_user_iov(struct request_queue *q, if (!bio_add_hw_page(q, bio, page, n, offs, max_sectors, &same_page)) { if (same_page) - put_page(page); + unpin_user_page(page); break; } @@ -312,7 +312,7 @@ static struct bio *bio_map_user_iov(struct request_queue *q, * release the pages we didn't map into the bio, if any */ while (j < npages) - put_page(pages[j++]); + unpin_user_page(pages[j++]); kvfree(pages); /* couldn't stuff something into bio? */ if (bytes) diff --git a/fs/direct-io.c b/fs/direct-io.c index 183299892465..5130ba32ae91 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -170,7 +170,7 @@ static inline int dio_refill_pages(struct dio *dio, struct dio_submit *sdio) { ssize_t ret; - ret = iov_iter_get_pages(sdio->iter, dio->pages, LONG_MAX, DIO_PAGES, + ret = iov_iter_pin_pages(sdio->iter, dio->pages, LONG_MAX, DIO_PAGES, &sdio->from); if (ret < 0 && sdio->blocks_available && (dio->op == REQ_OP_WRITE)) { @@ -182,7 +182,7 @@ static inline int dio_refill_pages(struct dio *dio, struct dio_submit *sdio) */ if (dio->page_errors == 0) dio->page_errors = ret; - get_page(page); + pin_page(page); dio->pages[0] = page; sdio->head = 0; sdio->tail = 1; @@ -472,7 +472,7 @@ static inline void dio_bio_submit(struct dio *dio, struct dio_submit *sdio) static inline void dio_cleanup(struct dio *dio, struct dio_submit *sdio) { while (sdio->head < sdio->tail) - put_page(dio->pages[sdio->head++]); + unpin_user_page(dio->pages[sdio->head++]); } /* @@ -739,7 +739,7 @@ static inline int dio_bio_add_page(struct dio_submit *sdio) */ if ((sdio->cur_page_len + sdio->cur_page_offset) == PAGE_SIZE) sdio->pages_in_io--; - get_page(sdio->cur_page); + pin_page(sdio->cur_page); sdio->final_block_in_bio = sdio->cur_page_block + (sdio->cur_page_len >> sdio->blkbits); ret = 0; @@ -853,13 +853,13 @@ submit_page_section(struct dio *dio, struct dio_submit *sdio, struct page *page, */ if (sdio->cur_page) { ret = dio_send_cur_page(dio, sdio, map_bh); - put_page(sdio->cur_page); + unpin_user_page(sdio->cur_page); sdio->cur_page = NULL; if (ret) return ret; } - get_page(page); /* It is in dio */ + pin_page(page); /* It is in dio */ sdio->cur_page = page; sdio->cur_page_offset = offset; sdio->cur_page_len = len; @@ -874,7 +874,7 @@ submit_page_section(struct dio *dio, struct dio_submit *sdio, struct page *page, ret = dio_send_cur_page(dio, sdio, map_bh); if (sdio->bio) dio_bio_submit(dio, sdio); - put_page(sdio->cur_page); + unpin_user_page(sdio->cur_page); sdio->cur_page = NULL; } return ret; @@ -974,7 +974,7 @@ static int do_direct_IO(struct dio *dio, struct dio_submit *sdio, ret = get_more_blocks(dio, sdio, map_bh); if (ret) { - put_page(page); + unpin_user_page(page); goto out; } if (!buffer_mapped(map_bh)) @@ -1019,7 +1019,7 @@ static int do_direct_IO(struct dio *dio, struct dio_submit *sdio, /* AKPM: eargh, -ENOTBLK is a hack */ if (dio->op == REQ_OP_WRITE) { - put_page(page); + unpin_user_page(page); return -ENOTBLK; } @@ -1032,7 +1032,7 @@ static int do_direct_IO(struct dio *dio, struct dio_submit *sdio, if (sdio->block_in_file >= i_size_aligned >> blkbits) { /* We hit eof */ - put_page(page); + unpin_user_page(page); goto out; } zero_user(page, from, 1 << blkbits); @@ -1072,7 +1072,7 @@ static int do_direct_IO(struct dio *dio, struct dio_submit *sdio, sdio->next_block_for_io, map_bh); if (ret) { - put_page(page); + unpin_user_page(page); goto out; } sdio->next_block_for_io += this_chunk_blocks; @@ -1087,8 +1087,8 @@ static int do_direct_IO(struct dio *dio, struct dio_submit *sdio, break; } - /* Drop the ref which was taken in get_user_pages() */ - put_page(page); + /* Drop the ref which was taken in pin_user_pages() */ + unpin_user_page(page); } out: return ret; @@ -1327,7 +1327,7 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode, ret2 = dio_send_cur_page(dio, &sdio, &map_bh); if (retval == 0) retval = ret2; - put_page(sdio.cur_page); + unpin_user_page(sdio.cur_page); sdio.cur_page = NULL; } if (sdio.bio) diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c index c1aafb2ab990..ea29f0892a8c 100644 --- a/fs/iomap/direct-io.c +++ b/fs/iomap/direct-io.c @@ -194,7 +194,7 @@ iomap_dio_zero(struct iomap_dio *dio, struct iomap *iomap, loff_t pos, bio->bi_private = dio; bio->bi_end_io = iomap_dio_bio_end_io; - get_page(page); + pin_page(page); __bio_add_page(bio, page, len, 0); bio_set_op_attrs(bio, REQ_OP_WRITE, flags); iomap_dio_submit_bio(dio, iomap, bio, pos);