From nobody Thu Nov 14 04:30:40 2024 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 84B8CC61DA4 for ; Tue, 14 Feb 2023 17:14:31 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S229816AbjBNROa (ORCPT ); Tue, 14 Feb 2023 12:14:30 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41014 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229845AbjBNRO1 (ORCPT ); Tue, 14 Feb 2023 12:14:27 -0500 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 056831115E for ; Tue, 14 Feb 2023 09:13:45 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1676394825; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=zSLOPyPBTyGpP+6+bR7yKU/VhAXOTSaAngOTYdMp3ic=; b=Ab+l4l18HMer0UkCotZtzJxCXsKRuUDnXLPb/4rHVHLifcBrFqsKXqxySP6cswV7sLkDck 2hs+qnVxlyLU3QJYtubGRRlmT0wBuJw1Op7reV82gEP/TjU2F2KTWj4BgNJQMVCacziW49 7+DxLuXi3QXLbzqB9TP0UUlm7mMqXQU= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-237-r6zz-Qt7OtqamK4CCCrB6g-1; Tue, 14 Feb 2023 12:13:41 -0500 X-MC-Unique: r6zz-Qt7OtqamK4CCCrB6g-1 Received: from smtp.corp.redhat.com (int-mx09.intmail.prod.int.rdu2.redhat.com [10.11.54.9]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 095C2101A521; Tue, 14 Feb 2023 17:13:38 +0000 (UTC) Received: from warthog.procyon.org.uk (unknown [10.33.36.24]) by smtp.corp.redhat.com (Postfix) with ESMTP id 0D5A4492B03; Tue, 14 Feb 2023 17:13:35 +0000 (UTC) From: David Howells To: Jens Axboe , Al Viro , Christoph Hellwig Cc: David Howells , Matthew Wilcox , Jan Kara , Jeff Layton , David Hildenbrand , Jason Gunthorpe , Logan Gunthorpe , Hillf Danton , linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, Christoph Hellwig , John Hubbard Subject: [PATCH v14 01/17] mm: Pass info, not iter, into filemap_get_pages() Date: Tue, 14 Feb 2023 17:13:14 +0000 Message-Id: <20230214171330.2722188-2-dhowells@redhat.com> In-Reply-To: <20230214171330.2722188-1-dhowells@redhat.com> References: <20230214171330.2722188-1-dhowells@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.9 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" filemap_get_pages() and a number of functions that it calls take an iterator to provide two things: the number of bytes to be got from the file specified and whether partially uptodate pages are allowed. Change these functions so that this information is passed in directly. This allows it to be called without having an iterator to hand. Signed-off-by: David Howells Reviewed-by: Christoph Hellwig cc: Jens Axboe cc: Christoph Hellwig cc: Matthew Wilcox cc: Al Viro cc: David Hildenbrand cc: John Hubbard cc: linux-mm@kvack.org cc: linux-block@vger.kernel.org cc: linux-fsdevel@vger.kernel.org --- mm/filemap.c | 27 ++++++++++++++------------- 1 file changed, 14 insertions(+), 13 deletions(-) diff --git a/mm/filemap.c b/mm/filemap.c index c4d4ace9cc70..876e77278d2a 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -2440,21 +2440,19 @@ static int filemap_read_folio(struct file *file, fi= ller_t filler, } =20 static bool filemap_range_uptodate(struct address_space *mapping, - loff_t pos, struct iov_iter *iter, struct folio *folio) + loff_t pos, size_t count, struct folio *folio, + bool need_uptodate) { - int count; - if (folio_test_uptodate(folio)) return true; /* pipes can't handle partially uptodate pages */ - if (iov_iter_is_pipe(iter)) + if (need_uptodate) return false; if (!mapping->a_ops->is_partially_uptodate) return false; if (mapping->host->i_blkbits >=3D folio_shift(folio)) return false; =20 - count =3D iter->count; if (folio_pos(folio) > pos) { count -=3D folio_pos(folio) - pos; pos =3D 0; @@ -2466,8 +2464,8 @@ static bool filemap_range_uptodate(struct address_spa= ce *mapping, } =20 static int filemap_update_page(struct kiocb *iocb, - struct address_space *mapping, struct iov_iter *iter, - struct folio *folio) + struct address_space *mapping, size_t count, + struct folio *folio, bool need_uptodate) { int error; =20 @@ -2501,7 +2499,8 @@ static int filemap_update_page(struct kiocb *iocb, goto unlock; =20 error =3D 0; - if (filemap_range_uptodate(mapping, iocb->ki_pos, iter, folio)) + if (filemap_range_uptodate(mapping, iocb->ki_pos, count, folio, + need_uptodate)) goto unlock; =20 error =3D -EAGAIN; @@ -2577,8 +2576,8 @@ static int filemap_readahead(struct kiocb *iocb, stru= ct file *file, return 0; } =20 -static int filemap_get_pages(struct kiocb *iocb, struct iov_iter *iter, - struct folio_batch *fbatch) +static int filemap_get_pages(struct kiocb *iocb, size_t count, + struct folio_batch *fbatch, bool need_uptodate) { struct file *filp =3D iocb->ki_filp; struct address_space *mapping =3D filp->f_mapping; @@ -2588,7 +2587,7 @@ static int filemap_get_pages(struct kiocb *iocb, stru= ct iov_iter *iter, struct folio *folio; int err =3D 0; =20 - last_index =3D DIV_ROUND_UP(iocb->ki_pos + iter->count, PAGE_SIZE); + last_index =3D DIV_ROUND_UP(iocb->ki_pos + count, PAGE_SIZE); retry: if (fatal_signal_pending(current)) return -EINTR; @@ -2621,7 +2620,8 @@ static int filemap_get_pages(struct kiocb *iocb, stru= ct iov_iter *iter, if ((iocb->ki_flags & IOCB_WAITQ) && folio_batch_count(fbatch) > 1) iocb->ki_flags |=3D IOCB_NOWAIT; - err =3D filemap_update_page(iocb, mapping, iter, folio); + err =3D filemap_update_page(iocb, mapping, count, folio, + need_uptodate); if (err) goto err; } @@ -2691,7 +2691,8 @@ ssize_t filemap_read(struct kiocb *iocb, struct iov_i= ter *iter, if (unlikely(iocb->ki_pos >=3D i_size_read(inode))) break; =20 - error =3D filemap_get_pages(iocb, iter, &fbatch); + error =3D filemap_get_pages(iocb, iter->count, &fbatch, + iov_iter_is_pipe(iter)); if (error < 0) break; From nobody Thu Nov 14 04:30:40 2024 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 73C27C64EC7 for ; Tue, 14 Feb 2023 17:14:47 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231604AbjBNROq (ORCPT ); Tue, 14 Feb 2023 12:14:46 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41016 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231859AbjBNROg (ORCPT ); Tue, 14 Feb 2023 12:14:36 -0500 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id C15A61024A for ; Tue, 14 Feb 2023 09:13:48 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1676394828; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=H0CTUnwfVEWUZBHu5fA2fJ8y0quxYsONFQfvaH+Rj5U=; b=e1LwX8vNBS4aNbTfO8iuq/aGJcCDlyuBsqb0XdX9+4IahwwHlVb8szygS/mhf5toLyk4+a hwEZd0Mmo2qr+9eZDlY67iqIYlgs2XDbDav4J91lc3eXZpI8RAJIJI5Xj1URVBBcf6Zw3j HkV2XA2gPw0h2vwS7WRZn6lT5cC+B/M= Received: from mimecast-mx02.redhat.com (mx3-rdu2.redhat.com [66.187.233.73]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-49-fW_ij3QZO56kIwrqyILfmQ-1; Tue, 14 Feb 2023 12:13:42 -0500 X-MC-Unique: fW_ij3QZO56kIwrqyILfmQ-1 Received: from smtp.corp.redhat.com (int-mx07.intmail.prod.int.rdu2.redhat.com [10.11.54.7]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id A9D5F1C08796; Tue, 14 Feb 2023 17:13:40 +0000 (UTC) Received: from warthog.procyon.org.uk (unknown [10.33.36.24]) by smtp.corp.redhat.com (Postfix) with ESMTP id B1699140EBF6; Tue, 14 Feb 2023 17:13:38 +0000 (UTC) From: David Howells To: Jens Axboe , Al Viro , Christoph Hellwig Cc: David Howells , Matthew Wilcox , Jan Kara , Jeff Layton , David Hildenbrand , Jason Gunthorpe , Logan Gunthorpe , Hillf Danton , linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, Christoph Hellwig , John Hubbard Subject: [PATCH v14 02/17] splice: Add a func to do a splice from a buffered file without ITER_PIPE Date: Tue, 14 Feb 2023 17:13:15 +0000 Message-Id: <20230214171330.2722188-3-dhowells@redhat.com> In-Reply-To: <20230214171330.2722188-1-dhowells@redhat.com> References: <20230214171330.2722188-1-dhowells@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.7 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Provide a function to do splice read from a buffered file, pulling the folios out of the pagecache directly by calling filemap_get_pages() to do any required reading and then pasting the returned folios into the pipe. A helper function is provided to do the actual folio pasting and will handle multipage folios by splicing as many of the relevant subpages as will fit into the pipe. The code is loosely based on filemap_read() and might belong in mm/filemap.c with that as it needs to use filemap_get_pages(). Signed-off-by: David Howells cc: Jens Axboe cc: Christoph Hellwig cc: Al Viro cc: David Hildenbrand cc: John Hubbard cc: linux-mm@kvack.org cc: linux-block@vger.kernel.org cc: linux-fsdevel@vger.kernel.org --- Notes: ver #14) - Rename to filemap_splice_read(). - Create a helper, pipe_head_buf(), to get the head buffer. - Use init_sync_kiocb(). - Move to mm/filemap.c. - Split the implementation of filemap_splice_read() from the patch to make generic_file_splice_read() use it and direct_splice_read(). include/linux/fs.h | 3 ++ mm/filemap.c | 128 +++++++++++++++++++++++++++++++++++++++++++++ mm/internal.h | 6 +++ 3 files changed, 137 insertions(+) diff --git a/include/linux/fs.h b/include/linux/fs.h index c1769a2c5d70..28743e38df91 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -3163,6 +3163,9 @@ ssize_t vfs_iocb_iter_write(struct file *file, struct= kiocb *iocb, struct iov_iter *iter); =20 /* fs/splice.c */ +ssize_t filemap_splice_read(struct file *in, loff_t *ppos, + struct pipe_inode_info *pipe, + size_t len, unsigned int flags); extern ssize_t generic_file_splice_read(struct file *, loff_t *, struct pipe_inode_info *, size_t, unsigned int); extern ssize_t iter_file_splice_write(struct pipe_inode_info *, diff --git a/mm/filemap.c b/mm/filemap.c index 876e77278d2a..8c7b135c8e23 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -42,6 +42,8 @@ #include #include #include +#include +#include #include #include #include "internal.h" @@ -2842,6 +2844,132 @@ generic_file_read_iter(struct kiocb *iocb, struct i= ov_iter *iter) } EXPORT_SYMBOL(generic_file_read_iter); =20 +/* + * Splice subpages from a folio into a pipe. + */ +size_t splice_folio_into_pipe(struct pipe_inode_info *pipe, + struct folio *folio, loff_t fpos, size_t size) +{ + struct page *page; + size_t spliced =3D 0, offset =3D offset_in_folio(folio, fpos); + + page =3D folio_page(folio, offset / PAGE_SIZE); + size =3D min(size, folio_size(folio) - offset); + offset %=3D PAGE_SIZE; + + while (spliced < size && + !pipe_full(pipe->head, pipe->tail, pipe->max_usage)) { + struct pipe_buffer *buf =3D pipe_head_buf(pipe); + size_t part =3D min_t(size_t, PAGE_SIZE - offset, size - spliced); + + *buf =3D (struct pipe_buffer) { + .ops =3D &page_cache_pipe_buf_ops, + .page =3D page, + .offset =3D offset, + .len =3D part, + }; + folio_get(folio); + pipe->head++; + page++; + spliced +=3D part; + offset =3D 0; + } + + return spliced; +} + +/* + * Splice folios from the pagecache of a buffered (ie. non-O_DIRECT) file = into + * a pipe. + */ +ssize_t filemap_splice_read(struct file *in, loff_t *ppos, + struct pipe_inode_info *pipe, + size_t len, unsigned int flags) +{ + struct folio_batch fbatch; + struct kiocb iocb; + size_t total_spliced =3D 0, used, npages; + loff_t isize, end_offset; + bool writably_mapped; + int i, error =3D 0; + + init_sync_kiocb(&iocb, in); + iocb.ki_pos =3D *ppos; + + /* Work out how much data we can actually add into the pipe */ + used =3D pipe_occupancy(pipe->head, pipe->tail); + npages =3D max_t(ssize_t, pipe->max_usage - used, 0); + len =3D min_t(size_t, len, npages * PAGE_SIZE); + + folio_batch_init(&fbatch); + + do { + cond_resched(); + + if (*ppos >=3D i_size_read(file_inode(in))) + break; + + iocb.ki_pos =3D *ppos; + error =3D filemap_get_pages(&iocb, len, &fbatch, true); + if (error < 0) + break; + + /* + * i_size must be checked after we know the pages are Uptodate. + * + * Checking i_size after the check allows us to calculate + * the correct value for "nr", which means the zero-filled + * part of the page is not copied back to userspace (unless + * another truncate extends the file - this is desired though). + */ + isize =3D i_size_read(file_inode(in)); + if (unlikely(*ppos >=3D isize)) + break; + end_offset =3D min_t(loff_t, isize, *ppos + len); + + /* + * Once we start copying data, we don't want to be touching any + * cachelines that might be contended: + */ + writably_mapped =3D mapping_writably_mapped(in->f_mapping); + + for (i =3D 0; i < folio_batch_count(&fbatch); i++) { + struct folio *folio =3D fbatch.folios[i]; + size_t n; + + if (folio_pos(folio) >=3D end_offset) + goto out; + folio_mark_accessed(folio); + + /* + * If users can be writing to this folio using arbitrary + * virtual addresses, take care of potential aliasing + * before reading the folio on the kernel side. + */ + if (writably_mapped) + flush_dcache_folio(folio); + + n =3D splice_folio_into_pipe(pipe, folio, *ppos, len); + if (!n) + goto out; + len -=3D n; + total_spliced +=3D n; + *ppos +=3D n; + in->f_ra.prev_pos =3D *ppos; + if (pipe_full(pipe->head, pipe->tail, pipe->max_usage)) + goto out; + } + + folio_batch_release(&fbatch); + } while (len); + +out: + folio_batch_release(&fbatch); + file_accessed(in); + + return total_spliced ? total_spliced : error; +} + static inline loff_t folio_seek_hole_data(struct xa_state *xas, struct address_space *mapping, struct folio *folio, loff_t start, loff_t end, bool seek_data) diff --git a/mm/internal.h b/mm/internal.h index bcf75a8b032d..6d4ca98f3844 100644 --- a/mm/internal.h +++ b/mm/internal.h @@ -794,6 +794,12 @@ struct migration_target_control { gfp_t gfp_mask; }; =20 +/* + * mm/filemap.c + */ +size_t splice_folio_into_pipe(struct pipe_inode_info *pipe, + struct folio *folio, loff_t fpos, size_t size); + /* * mm/vmalloc.c */ From nobody Thu Nov 14 04:30:40 2024 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 1DB8BC61DA4 for ; Tue, 14 Feb 2023 17:15:10 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S232411AbjBNROt (ORCPT ); Tue, 14 Feb 2023 12:14:49 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41144 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231966AbjBNROg (ORCPT ); Tue, 14 Feb 2023 12:14:36 -0500 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 89B5B5B98 for ; Tue, 14 Feb 2023 09:13:50 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1676394829; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=Wrfy5DSUofzooKuydOZLN29ccYXYPtSMXDanySrmvRI=; b=ECxwjEg72Aygmn9QqzZA8toC1d6o9uldlZEsTn4ThgDYTO+eQLPgsyqeNTS4s3J0h6tURC mOTcphw7r0NvOEu69u+JhFjJt1qYFuqdSInzlJalRL8NxWN1k0RRN+UtD0+olnv5ZVajsv S5522d92MnFjjDKCKaKhZlc3cMtCryU= Received: from mimecast-mx02.redhat.com (mx3-rdu2.redhat.com [66.187.233.73]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-593-llfpyhjxOKyRI8fnUf8UoA-1; Tue, 14 Feb 2023 12:13:44 -0500 X-MC-Unique: llfpyhjxOKyRI8fnUf8UoA-1 Received: from smtp.corp.redhat.com (int-mx07.intmail.prod.int.rdu2.redhat.com [10.11.54.7]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 6EDA11C08795; Tue, 14 Feb 2023 17:13:43 +0000 (UTC) Received: from warthog.procyon.org.uk (unknown [10.33.36.24]) by smtp.corp.redhat.com (Postfix) with ESMTP id 53202140EBF6; Tue, 14 Feb 2023 17:13:41 +0000 (UTC) From: David Howells To: Jens Axboe , Al Viro , Christoph Hellwig Cc: David Howells , Matthew Wilcox , Jan Kara , Jeff Layton , David Hildenbrand , Jason Gunthorpe , Logan Gunthorpe , Hillf Danton , linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, syzbot+a440341a59e3b7142895@syzkaller.appspotmail.com, Christoph Hellwig , John Hubbard Subject: [PATCH v14 03/17] splice: Add a func to do a splice from an O_DIRECT file without ITER_PIPE Date: Tue, 14 Feb 2023 17:13:16 +0000 Message-Id: <20230214171330.2722188-4-dhowells@redhat.com> In-Reply-To: <20230214171330.2722188-1-dhowells@redhat.com> References: <20230214171330.2722188-1-dhowells@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.7 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Implement a function, direct_file_splice(), that deals with this by using an ITER_BVEC iterator instead of an ITER_PIPE iterator as the former won't free its buffers when reverted. The function bulk allocates all the buffers it thinks it is going to use in advance, does the read synchronously and only then trims the buffer down. The pages we did use get pushed into the pipe. This fixes a problem with the upcoming iov_iter_extract_pages() function, whereby pages extracted from a non-user-backed iterator such as ITER_PIPE aren't pinned. __iomap_dio_rw(), however, calls iov_iter_revert() to shorten the iterator to just the bufferage it is going to use - which has the side-effect of freeing the excess pipe buffers, even though they're attached to a bio and may get written to by DMA (thanks to Hillf Danton for spotting this[1]). This then causes memory corruption that is particularly noticable when the syzbot test[2] is run. The test boils down to: out =3D creat(argv[1], 0666); ftruncate(out, 0x800); lseek(out, 0x200, SEEK_SET); in =3D open(argv[1], O_RDONLY | O_DIRECT | O_NOFOLLOW); sendfile(out, in, NULL, 0x1dd00); run repeatedly in parallel. What I think is happening is that ftruncate() occasionally shortens the DIO read that's about to be made by sendfile's splice core by reducing i_size. This should be more efficient for DIO read by virtue of doing a bulk page allocation, but slightly less efficient by ignoring any partial page in the pipe. Reported-by: syzbot+a440341a59e3b7142895@syzkaller.appspotmail.com Signed-off-by: David Howells cc: Jens Axboe cc: Christoph Hellwig cc: Al Viro cc: David Hildenbrand cc: John Hubbard cc: linux-mm@kvack.org cc: linux-block@vger.kernel.org cc: linux-fsdevel@vger.kernel.org Link: https://lore.kernel.org/r/20230207094731.1390-1-hdanton@sina.com/ [1] Link: https://lore.kernel.org/r/000000000000b0b3c005f3a09383@google.com/ [2] --- Notes: ver #14) - Use alloc_pages_bulk_array() rather than alloc_pages_bulk_list(). - Use release_pages() rather than a loop calling __free_page(). - Rename to direct_splice_read(). - Don't call from generic_file_splice_read() yet. =20 ver #13) - Don't completely replace generic_file_splice_read(), but rather only= use this if we're doing a splicing from an O_DIRECT file fd. fs/splice.c | 92 +++++++++++++++++++++++++++++++++++++++ include/linux/fs.h | 3 ++ include/linux/pipe_fs_i.h | 20 +++++++++ lib/iov_iter.c | 6 --- 4 files changed, 115 insertions(+), 6 deletions(-) diff --git a/fs/splice.c b/fs/splice.c index 5969b7a1d353..4c6332854b63 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -282,6 +282,98 @@ void splice_shrink_spd(struct splice_pipe_desc *spd) kfree(spd->partial); } =20 +/* + * Splice data from an O_DIRECT file into pages and then add them to the o= utput + * pipe. + */ +ssize_t direct_splice_read(struct file *in, loff_t *ppos, + struct pipe_inode_info *pipe, + size_t len, unsigned int flags) +{ + struct iov_iter to; + struct bio_vec *bv; + struct kiocb kiocb; + struct page **pages; + ssize_t ret; + size_t used, npages, chunk, remain, reclaim; + int i; + + /* Work out how much data we can actually add into the pipe */ + used =3D pipe_occupancy(pipe->head, pipe->tail); + npages =3D max_t(ssize_t, pipe->max_usage - used, 0); + len =3D min_t(size_t, len, npages * PAGE_SIZE); + npages =3D DIV_ROUND_UP(len, PAGE_SIZE); + + bv =3D kzalloc(array_size(npages, sizeof(bv[0])) + + array_size(npages, sizeof(struct page *)), GFP_KERNEL); + if (!bv) + return -ENOMEM; + + pages =3D (void *)(bv + npages); + npages =3D alloc_pages_bulk_array(GFP_USER, npages, pages); + if (!npages) { + kfree(bv); + return -ENOMEM; + } + + remain =3D len =3D min_t(size_t, len, npages * PAGE_SIZE); + + for (i =3D 0; i < npages; i++) { + chunk =3D min_t(size_t, PAGE_SIZE, remain); + bv[i].bv_page =3D pages[i]; + bv[i].bv_offset =3D 0; + bv[i].bv_len =3D chunk; + remain -=3D chunk; + } + + /* Do the I/O */ + iov_iter_bvec(&to, ITER_DEST, bv, npages, len); + init_sync_kiocb(&kiocb, in); + kiocb.ki_pos =3D *ppos; + ret =3D call_read_iter(in, &kiocb, &to); + + reclaim =3D npages * PAGE_SIZE; + remain =3D 0; + if (ret > 0) { + reclaim -=3D ret; + remain =3D ret; + *ppos =3D kiocb.ki_pos; + file_accessed(in); + } else if (ret < 0) { + /* + * callers of ->splice_read() expect -EAGAIN on + * "can't put anything in there", rather than -EFAULT. + */ + if (ret =3D=3D -EFAULT) + ret =3D -EAGAIN; + } + + /* Free any pages that didn't get touched at all. */ + reclaim /=3D PAGE_SIZE; + if (reclaim) { + npages -=3D reclaim; + release_pages(pages + npages, reclaim); + } + + /* Push the remaining pages into the pipe. */ + for (i =3D 0; i < npages; i++) { + struct pipe_buffer *buf =3D pipe_head_buf(pipe); + + chunk =3D min_t(size_t, remain, PAGE_SIZE); + *buf =3D (struct pipe_buffer) { + .ops =3D &default_pipe_buf_ops, + .page =3D bv[i].bv_page, + .offset =3D 0, + .len =3D chunk, + }; + pipe->head++; + remain -=3D chunk; + } + + kfree(bv); + return ret; +} + /** * generic_file_splice_read - splice data from file to a pipe * @in: file to splice from diff --git a/include/linux/fs.h b/include/linux/fs.h index 28743e38df91..551c9403f9b3 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -3166,6 +3166,9 @@ ssize_t vfs_iocb_iter_write(struct file *file, struct= kiocb *iocb, ssize_t filemap_splice_read(struct file *in, loff_t *ppos, struct pipe_inode_info *pipe, size_t len, unsigned int flags); +ssize_t direct_splice_read(struct file *in, loff_t *ppos, + struct pipe_inode_info *pipe, + size_t len, unsigned int flags); extern ssize_t generic_file_splice_read(struct file *, loff_t *, struct pipe_inode_info *, size_t, unsigned int); extern ssize_t iter_file_splice_write(struct pipe_inode_info *, diff --git a/include/linux/pipe_fs_i.h b/include/linux/pipe_fs_i.h index 6cb65df3e3ba..d2c3f16cf6b1 100644 --- a/include/linux/pipe_fs_i.h +++ b/include/linux/pipe_fs_i.h @@ -156,6 +156,26 @@ static inline bool pipe_full(unsigned int head, unsign= ed int tail, return pipe_occupancy(head, tail) >=3D limit; } =20 +/** + * pipe_buf - Return the pipe buffer for the specified slot in the pipe ri= ng + * @pipe: The pipe to access + * @slot: The slot of interest + */ +static inline struct pipe_buffer *pipe_buf(const struct pipe_inode_info *p= ipe, + unsigned int slot) +{ + return &pipe->bufs[slot & (pipe->ring_size - 1)]; +} + +/** + * pipe_head_buf - Return the pipe buffer at the head of the pipe ring + * @pipe: The pipe to access + */ +static inline struct pipe_buffer *pipe_head_buf(const struct pipe_inode_in= fo *pipe) +{ + return pipe_buf(pipe, pipe->head); +} + /** * pipe_buf_get - get a reference to a pipe_buffer * @pipe: the pipe that the buffer belongs to diff --git a/lib/iov_iter.c b/lib/iov_iter.c index f9a3ff37ecd1..47c484551c59 100644 --- a/lib/iov_iter.c +++ b/lib/iov_iter.c @@ -186,12 +186,6 @@ static int copyin(void *to, const void __user *from, s= ize_t n) return res; } =20 -static inline struct pipe_buffer *pipe_buf(const struct pipe_inode_info *p= ipe, - unsigned int slot) -{ - return &pipe->bufs[slot & (pipe->ring_size - 1)]; -} - #ifdef PIPE_PARANOIA static bool sanity(const struct iov_iter *i) { From nobody Thu Nov 14 04:30:40 2024 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 69BD8C6379F for ; Tue, 14 Feb 2023 17:15:10 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S232797AbjBNROv (ORCPT ); Tue, 14 Feb 2023 12:14:51 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41142 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S232355AbjBNROm (ORCPT ); Tue, 14 Feb 2023 12:14:42 -0500 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 287FEC159 for ; Tue, 14 Feb 2023 09:13:51 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1676394830; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=oNTxwJtD7ZehgoJ1QPpdyGLg4+saLDUsiyn+vcqXfIw=; b=Z7NBFiZYnwRIo5/XnxqO/6HZ++SdURXmvZyTpEKp+Y4kXHaETwPtMVYjDkqjVfAotF/IXH tYUAv089QFZP+aQNOB9DP3ixd/Pk2YfNyczWprQVEt6/A2qB8wz1jxE39aGeztRHUSqPZx wTBC3ucZUFvBd0auxGgbCZIFfP0HcBk= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-344-D5m4tvNWOIeGlA5oelilFg-1; Tue, 14 Feb 2023 12:13:47 -0500 X-MC-Unique: D5m4tvNWOIeGlA5oelilFg-1 Received: from smtp.corp.redhat.com (int-mx06.intmail.prod.int.rdu2.redhat.com [10.11.54.6]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 69FDE800050; Tue, 14 Feb 2023 17:13:46 +0000 (UTC) Received: from warthog.procyon.org.uk (unknown [10.33.36.24]) by smtp.corp.redhat.com (Postfix) with ESMTP id 2684F2166B26; Tue, 14 Feb 2023 17:13:44 +0000 (UTC) From: David Howells To: Jens Axboe , Al Viro , Christoph Hellwig Cc: David Howells , Matthew Wilcox , Jan Kara , Jeff Layton , David Hildenbrand , Jason Gunthorpe , Logan Gunthorpe , Hillf Danton , linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, Daniel Golle , Guenter Roeck , Christoph Hellwig , John Hubbard , Hugh Dickins Subject: [PATCH v14 04/17] shmem: Implement splice-read Date: Tue, 14 Feb 2023 17:13:17 +0000 Message-Id: <20230214171330.2722188-5-dhowells@redhat.com> In-Reply-To: <20230214171330.2722188-1-dhowells@redhat.com> References: <20230214171330.2722188-1-dhowells@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.6 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" The new filemap_splice_read() has an implicit expectation via filemap_get_pages() that ->read_folio() exists if ->readahead() doesn't fully populate the pagecache of the file it is reading from[1], potentially leading to a jump to NULL if this doesn't exist. shmem, however, (and by extension, tmpfs, ramfs and rootfs), doesn't have ->read_folio(), Work around this by equipping shmem with its own splice-read implementation, based on filemap_splice_read(), but able to paste in zero_page when there's a page missing. Signed-off-by: David Howells cc: Daniel Golle cc: Guenter Roeck cc: Christoph Hellwig cc: Jens Axboe cc: Al Viro cc: John Hubbard cc: David Hildenbrand cc: Matthew Wilcox cc: Hugh Dickins cc: linux-block@vger.kernel.org cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org Link: https://lore.kernel.org/r/Y+pdHFFTk1TTEBsO@makrotopia.org/ [1] --- mm/shmem.c | 124 ++++++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 123 insertions(+), 1 deletion(-) diff --git a/mm/shmem.c b/mm/shmem.c index 0005ab2c29af..7145a5345f4d 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -2711,6 +2711,128 @@ static ssize_t shmem_file_read_iter(struct kiocb *i= ocb, struct iov_iter *to) return retval ? retval : error; } =20 +static bool zero_pipe_buf_try_steal(struct pipe_inode_info *pipe, + struct pipe_buffer *buf) +{ + return false; +} + +static const struct pipe_buf_operations zero_pipe_buf_ops =3D { + .release =3D generic_pipe_buf_release, + .try_steal =3D zero_pipe_buf_try_steal, + .get =3D generic_pipe_buf_get, +}; + +static size_t splice_zeropage_into_pipe(struct pipe_inode_info *pipe, + loff_t fpos, size_t size) +{ + size_t offset =3D fpos & ~PAGE_MASK; + + size =3D min_t(size_t, size, PAGE_SIZE - offset); + + if (!pipe_full(pipe->head, pipe->tail, pipe->max_usage)) { + struct pipe_buffer *buf =3D pipe_head_buf(pipe); + + *buf =3D (struct pipe_buffer) { + .ops =3D &zero_pipe_buf_ops, + .page =3D ZERO_PAGE(0), + .offset =3D offset, + .len =3D size, + }; + get_page(buf->page); + pipe->head++; + } + + return size; +} + +static ssize_t shmem_file_splice_read(struct file *in, loff_t *ppos, + struct pipe_inode_info *pipe, + size_t len, unsigned int flags) +{ + struct inode *inode =3D file_inode(in); + struct address_space *mapping =3D inode->i_mapping; + struct folio *folio =3D NULL; + size_t total_spliced =3D 0, used, npages, n, part; + loff_t isize; + int error =3D 0; + + /* Work out how much data we can actually add into the pipe */ + used =3D pipe_occupancy(pipe->head, pipe->tail); + npages =3D max_t(ssize_t, pipe->max_usage - used, 0); + len =3D min_t(size_t, len, npages * PAGE_SIZE); + + do { + if (*ppos >=3D i_size_read(inode)) + break; + + error =3D shmem_get_folio(inode, *ppos / PAGE_SIZE, &folio, SGP_READ); + if (error) { + if (error =3D=3D -EINVAL) + error =3D 0; + break; + } + if (folio) { + folio_unlock(folio); + + if (folio_test_hwpoison(folio)) { + error =3D -EIO; + break; + } + } + + /* + * i_size must be checked after we know the pages are Uptodate. + * + * Checking i_size after the check allows us to calculate + * the correct value for "nr", which means the zero-filled + * part of the page is not copied back to userspace (unless + * another truncate extends the file - this is desired though). + */ + isize =3D i_size_read(inode); + if (unlikely(*ppos >=3D isize)) + break; + part =3D min_t(loff_t, isize - *ppos, len); + + if (folio) { + /* + * If users can be writing to this page using arbitrary + * virtual addresses, take care about potential aliasing + * before reading the page on the kernel side. + */ + if (mapping_writably_mapped(mapping)) + flush_dcache_folio(folio); + folio_mark_accessed(folio); + /* + * Ok, we have the page, and it's up-to-date, so we can + * now splice it into the pipe. + */ + n =3D splice_folio_into_pipe(pipe, folio, *ppos, part); + folio_put(folio); + folio =3D NULL; + } else { + n =3D splice_zeropage_into_pipe(pipe, *ppos, len); + } + + if (!n) + break; + len -=3D n; + total_spliced +=3D n; + *ppos +=3D n; + in->f_ra.prev_pos =3D *ppos; + if (pipe_full(pipe->head, pipe->tail, pipe->max_usage)) + break; + + cond_resched(); + } while (len); + + if (folio) + folio_put(folio); + + file_accessed(in); + return total_spliced ? total_spliced : error; +} + static loff_t shmem_file_llseek(struct file *file, loff_t offset, int when= ce) { struct address_space *mapping =3D file->f_mapping; @@ -3929,7 +4051,7 @@ static const struct file_operations shmem_file_operat= ions =3D { .read_iter =3D shmem_file_read_iter, .write_iter =3D generic_file_write_iter, .fsync =3D noop_fsync, - .splice_read =3D generic_file_splice_read, + .splice_read =3D shmem_file_splice_read, .splice_write =3D iter_file_splice_write, .fallocate =3D shmem_fallocate, #endif From nobody Thu Nov 14 04:30:40 2024 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 08462C05027 for ; Tue, 14 Feb 2023 17:15:21 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S232463AbjBNRPT (ORCPT ); Tue, 14 Feb 2023 12:15:19 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41306 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S232822AbjBNROw (ORCPT ); Tue, 14 Feb 2023 12:14:52 -0500 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id A044F5266 for ; Tue, 14 Feb 2023 09:14:04 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1676394843; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=LVjIujk8d1N0k6RGniY3UP+0o4vFHY30HdwwNIbIhak=; b=MA2Tp0y4F6kIgwUR6MJz0HS9ffds8A1m3jEaav9Jj+wqi2zJd/v719q2TeR3tm8nKY50Ut HhvvM/2V3IkPs0zUVv4+Q17Ev0X1hJefoF8cCFLzO4MOKRShGH59LtNBlRTErsH8ZlBdCA e/X3QsjeoH7S1U0rjAZD9p+SOljZF9A= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-608-VwTllVNdOty5IfHcJbDwmw-1; Tue, 14 Feb 2023 12:14:00 -0500 X-MC-Unique: VwTllVNdOty5IfHcJbDwmw-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.rdu2.redhat.com [10.11.54.2]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 4BBF285CCE0; Tue, 14 Feb 2023 17:13:59 +0000 (UTC) Received: from warthog.procyon.org.uk (unknown [10.33.36.24]) by smtp.corp.redhat.com (Postfix) with ESMTP id 0784C40C10FA; Tue, 14 Feb 2023 17:13:46 +0000 (UTC) From: David Howells To: Jens Axboe , Al Viro , Christoph Hellwig Cc: David Howells , Matthew Wilcox , Jan Kara , Jeff Layton , David Hildenbrand , Jason Gunthorpe , Logan Gunthorpe , Hillf Danton , linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, Christoph Hellwig , John Hubbard , Miklos Szeredi , linux-unionfs@vger.kernel.org Subject: [PATCH v14 05/17] overlayfs: Implement splice-read Date: Tue, 14 Feb 2023 17:13:18 +0000 Message-Id: <20230214171330.2722188-6-dhowells@redhat.com> In-Reply-To: <20230214171330.2722188-1-dhowells@redhat.com> References: <20230214171330.2722188-1-dhowells@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.2 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Implement splice-read for overlayfs by passing the request down a layer rather than going through generic_file_splice_read() which is going to be changed to assume that ->read_folio() is present on buffered files. Signed-off-by: David Howells cc: Christoph Hellwig cc: Jens Axboe cc: Al Viro cc: John Hubbard cc: David Hildenbrand cc: Matthew Wilcox cc: Miklos Szeredi cc: linux-unionfs@vger.kernel.org cc: linux-block@vger.kernel.org cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- fs/overlayfs/file.c | 36 +++++++++++++++++++++++++++++++++++- 1 file changed, 35 insertions(+), 1 deletion(-) diff --git a/fs/overlayfs/file.c b/fs/overlayfs/file.c index c9d0c362c7ef..267b61df6fcd 100644 --- a/fs/overlayfs/file.c +++ b/fs/overlayfs/file.c @@ -419,6 +419,40 @@ static ssize_t ovl_write_iter(struct kiocb *iocb, stru= ct iov_iter *iter) return ret; } =20 +static ssize_t ovl_splice_read(struct file *in, loff_t *ppos, + struct pipe_inode_info *pipe, size_t len, + unsigned int flags) +{ + const struct cred *old_cred; + struct fd real; + ssize_t ret; + + ret =3D ovl_real_fdget(in, &real); + if (ret) + return ret; + + ret =3D -EINVAL; + if (in->f_flags & O_DIRECT && + !(real.file->f_mode & FMODE_CAN_ODIRECT)) + goto out_fdput; + if (!real.file->f_op->splice_read) + goto out_fdput; + + ret =3D rw_verify_area(READ, in, ppos, len); + if (unlikely(ret < 0)) + return ret; + + old_cred =3D ovl_override_creds(file_inode(in)->i_sb); + ret =3D real.file->f_op->splice_read(real.file, ppos, pipe, len, flags); + + revert_creds(old_cred); + ovl_file_accessed(in); +out_fdput: + fdput(real); + + return ret; +} + /* * Calling iter_file_splice_write() directly from overlay's f_op may deadl= ock * due to lock order inversion between pipe->mutex in iter_file_splice_wri= te() @@ -695,7 +729,7 @@ const struct file_operations ovl_file_operations =3D { .fallocate =3D ovl_fallocate, .fadvise =3D ovl_fadvise, .flush =3D ovl_flush, - .splice_read =3D generic_file_splice_read, + .splice_read =3D ovl_splice_read, .splice_write =3D ovl_splice_write, =20 .copy_file_range =3D ovl_copy_file_range, From nobody Thu Nov 14 04:30:40 2024 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id DEDB6C05027 for ; Tue, 14 Feb 2023 17:15:39 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S232958AbjBNRPi (ORCPT ); Tue, 14 Feb 2023 12:15:38 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41460 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S232985AbjBNRPD (ORCPT ); Tue, 14 Feb 2023 12:15:03 -0500 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id BBE3323658 for ; Tue, 14 Feb 2023 09:14:15 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1676394854; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=9CqUl4uoQ+LRGKUoO3X6p5TYca6H7tIaOYxOpOq6akk=; b=KseqYForNka+to6QSiiet1EMcoSNgAlLXSF6ZLGOKBxyzBKaE+aRvcX8wV8/u0v6brFfH7 qqp2ptl3fP3F6H25IyjmL1C0Od3Ig2IFB5x4rp9d22pcP/kSnmWcpUHLV/xaU4b693M6ZV W37jMx87hNhWhkbkO79BJw8s/AFXZjk= Received: from mimecast-mx02.redhat.com (mx3-rdu2.redhat.com [66.187.233.73]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-253-cDT_luHMOIyAlH60JlEtxw-1; Tue, 14 Feb 2023 12:14:04 -0500 X-MC-Unique: cDT_luHMOIyAlH60JlEtxw-1 Received: from smtp.corp.redhat.com (int-mx10.intmail.prod.int.rdu2.redhat.com [10.11.54.10]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id A98F52817248; Tue, 14 Feb 2023 17:14:02 +0000 (UTC) Received: from warthog.procyon.org.uk (unknown [10.33.36.24]) by smtp.corp.redhat.com (Postfix) with ESMTP id 0D058492B15; Tue, 14 Feb 2023 17:13:59 +0000 (UTC) From: David Howells To: Jens Axboe , Al Viro , Christoph Hellwig Cc: David Howells , Matthew Wilcox , Jan Kara , Jeff Layton , David Hildenbrand , Jason Gunthorpe , Logan Gunthorpe , Hillf Danton , linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, Christoph Hellwig , John Hubbard , Jan Harkes , coda@cs.cmu.edu, codalist@coda.cs.cmu.edu, linux-unionfs@vger.kernel.org Subject: [PATCH v14 06/17] coda: Implement splice-read Date: Tue, 14 Feb 2023 17:13:19 +0000 Message-Id: <20230214171330.2722188-7-dhowells@redhat.com> In-Reply-To: <20230214171330.2722188-1-dhowells@redhat.com> References: <20230214171330.2722188-1-dhowells@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.10 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Implement splice-read for coda by passing the request down a layer rather than going through generic_file_splice_read() which is going to be changed to assume that ->read_folio() is present on buffered files. Signed-off-by: David Howells cc: Christoph Hellwig cc: Jens Axboe cc: Al Viro cc: John Hubbard cc: David Hildenbrand cc: Matthew Wilcox cc: Jan Harkes cc: coda@cs.cmu.edu cc: codalist@coda.cs.cmu.edu cc: linux-unionfs@vger.kernel.org cc: linux-block@vger.kernel.org cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org Acked-by: Jan Harkes --- fs/coda/file.c | 36 +++++++++++++++++++++++++++++++++++- 1 file changed, 35 insertions(+), 1 deletion(-) diff --git a/fs/coda/file.c b/fs/coda/file.c index 3f3c81e6b1ab..33cd7880d30e 100644 --- a/fs/coda/file.c +++ b/fs/coda/file.c @@ -23,6 +23,7 @@ #include #include #include +#include =20 #include #include "coda_psdev.h" @@ -94,6 +95,39 @@ coda_file_write_iter(struct kiocb *iocb, struct iov_iter= *to) return ret; } =20 +static ssize_t +coda_file_splice_read(struct file *coda_file, loff_t *ppos, + struct pipe_inode_info *pipe, + size_t len, unsigned int flags) +{ + struct inode *coda_inode =3D file_inode(coda_file); + struct coda_file_info *cfi =3D coda_ftoc(coda_file); + struct file *in =3D cfi->cfi_container; + loff_t ki_pos =3D *ppos; + ssize_t ret; + + if (!in->f_op->splice_read) + return -EINVAL; + + ret =3D rw_verify_area(READ, in, ppos, len); + if (unlikely(ret < 0)) + return ret; + + ret =3D venus_access_intent(coda_inode->i_sb, coda_i2f(coda_inode), + &cfi->cfi_access_intent, + len, ki_pos, CODA_ACCESS_TYPE_READ); + if (ret) + goto finish_read; + + ret =3D in->f_op->splice_read(in, ppos, pipe, len, flags); + +finish_read: + venus_access_intent(coda_inode->i_sb, coda_i2f(coda_inode), + &cfi->cfi_access_intent, + len, ki_pos, CODA_ACCESS_TYPE_READ_FINISH); + return ret; +} + static void coda_vm_open(struct vm_area_struct *vma) { @@ -302,5 +336,5 @@ const struct file_operations coda_file_operations =3D { .open =3D coda_open, .release =3D coda_release, .fsync =3D coda_fsync, - .splice_read =3D generic_file_splice_read, + .splice_read =3D coda_file_splice_read, }; From nobody Thu Nov 14 04:30:40 2024 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 5F31FC6379F for ; Tue, 14 Feb 2023 17:15:28 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S232635AbjBNRP0 (ORCPT ); Tue, 14 Feb 2023 12:15:26 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41422 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S232951AbjBNRPA (ORCPT ); Tue, 14 Feb 2023 12:15:00 -0500 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 804C3D52C for ; Tue, 14 Feb 2023 09:14:12 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1676394851; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=tsT35xW/A1CGBEVLxMBn7T6yW2S9XrBAkcV2v0SNR0w=; b=DO5Tjuhh87CtwwC4zDiqALyTxFxtcHBdCGyt6ClW28NN8iSH9ddOdLZO2QbrLLXhIT6gXp ggX4VlTV1Ac9Rl2ZpEeKQyHhRJbrqD9XYADCJVi2jwo8Z+YeUOcMaVY97/Fpp7UXl0sNh4 idGUyj9ehjAqlvSUyRJYOv3seSd3iwI= Received: from mimecast-mx02.redhat.com (mx3-rdu2.redhat.com [66.187.233.73]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-619--tEiHNGSOsmK7bn8sDqt9Q-1; Tue, 14 Feb 2023 12:14:08 -0500 X-MC-Unique: -tEiHNGSOsmK7bn8sDqt9Q-1 Received: from smtp.corp.redhat.com (int-mx06.intmail.prod.int.rdu2.redhat.com [10.11.54.6]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 8CC982817223; Tue, 14 Feb 2023 17:14:05 +0000 (UTC) Received: from warthog.procyon.org.uk (unknown [10.33.36.24]) by smtp.corp.redhat.com (Postfix) with ESMTP id 4B91B2166B29; Tue, 14 Feb 2023 17:14:03 +0000 (UTC) From: David Howells To: Jens Axboe , Al Viro , Christoph Hellwig Cc: David Howells , Matthew Wilcox , Jan Kara , Jeff Layton , David Hildenbrand , Jason Gunthorpe , Logan Gunthorpe , Hillf Danton , linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, Christoph Hellwig , John Hubbard , Miklos Szeredi , Arnd Bergmann , Greg Kroah-Hartman Subject: [PATCH v14 07/17] tty, proc, kernfs, random: Use direct_splice_read() Date: Tue, 14 Feb 2023 17:13:20 +0000 Message-Id: <20230214171330.2722188-8-dhowells@redhat.com> In-Reply-To: <20230214171330.2722188-1-dhowells@redhat.com> References: <20230214171330.2722188-1-dhowells@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.6 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Use direct_splice_read() for tty, procfs, kernfs and random files rather than going through generic_file_splice_read() as they just copy the file into the output buffer and don't splice pages. This avoids the need for them to have a ->read_folio() to satisfy filemap_splice_read(). Signed-off-by: David Howells cc: Christoph Hellwig cc: Jens Axboe cc: Al Viro cc: John Hubbard cc: David Hildenbrand cc: Matthew Wilcox cc: Miklos Szeredi cc: Arnd Bergmann cc: Greg Kroah-Hartman cc: linux-block@vger.kernel.org cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- drivers/char/random.c | 4 ++-- drivers/tty/tty_io.c | 4 ++-- fs/kernfs/file.c | 2 +- fs/proc/inode.c | 4 ++-- fs/proc/proc_sysctl.c | 2 +- fs/proc_namespace.c | 6 +++--- 6 files changed, 11 insertions(+), 11 deletions(-) diff --git a/drivers/char/random.c b/drivers/char/random.c index ce3ccd172cc8..792713616ba8 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -1546,7 +1546,7 @@ const struct file_operations random_fops =3D { .compat_ioctl =3D compat_ptr_ioctl, .fasync =3D random_fasync, .llseek =3D noop_llseek, - .splice_read =3D generic_file_splice_read, + .splice_read =3D direct_splice_read, .splice_write =3D iter_file_splice_write, }; =20 @@ -1557,7 +1557,7 @@ const struct file_operations urandom_fops =3D { .compat_ioctl =3D compat_ptr_ioctl, .fasync =3D random_fasync, .llseek =3D noop_llseek, - .splice_read =3D generic_file_splice_read, + .splice_read =3D direct_splice_read, .splice_write =3D iter_file_splice_write, }; =20 diff --git a/drivers/tty/tty_io.c b/drivers/tty/tty_io.c index 3149114bf130..495678e9b95e 100644 --- a/drivers/tty/tty_io.c +++ b/drivers/tty/tty_io.c @@ -466,7 +466,7 @@ static const struct file_operations tty_fops =3D { .llseek =3D no_llseek, .read_iter =3D tty_read, .write_iter =3D tty_write, - .splice_read =3D generic_file_splice_read, + .splice_read =3D direct_splice_read, .splice_write =3D iter_file_splice_write, .poll =3D tty_poll, .unlocked_ioctl =3D tty_ioctl, @@ -481,7 +481,7 @@ static const struct file_operations console_fops =3D { .llseek =3D no_llseek, .read_iter =3D tty_read, .write_iter =3D redirected_tty_write, - .splice_read =3D generic_file_splice_read, + .splice_read =3D direct_splice_read, .splice_write =3D iter_file_splice_write, .poll =3D tty_poll, .unlocked_ioctl =3D tty_ioctl, diff --git a/fs/kernfs/file.c b/fs/kernfs/file.c index e4a50e4ff0d2..9d23b8141db7 100644 --- a/fs/kernfs/file.c +++ b/fs/kernfs/file.c @@ -1011,7 +1011,7 @@ const struct file_operations kernfs_file_fops =3D { .release =3D kernfs_fop_release, .poll =3D kernfs_fop_poll, .fsync =3D noop_fsync, - .splice_read =3D generic_file_splice_read, + .splice_read =3D direct_splice_read, .splice_write =3D iter_file_splice_write, }; =20 diff --git a/fs/proc/inode.c b/fs/proc/inode.c index f495fdb39151..711f12706469 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -591,7 +591,7 @@ static const struct file_operations proc_iter_file_ops = =3D { .llseek =3D proc_reg_llseek, .read_iter =3D proc_reg_read_iter, .write =3D proc_reg_write, - .splice_read =3D generic_file_splice_read, + .splice_read =3D direct_splice_read, .poll =3D proc_reg_poll, .unlocked_ioctl =3D proc_reg_unlocked_ioctl, .mmap =3D proc_reg_mmap, @@ -617,7 +617,7 @@ static const struct file_operations proc_reg_file_ops_c= ompat =3D { static const struct file_operations proc_iter_file_ops_compat =3D { .llseek =3D proc_reg_llseek, .read_iter =3D proc_reg_read_iter, - .splice_read =3D generic_file_splice_read, + .splice_read =3D direct_splice_read, .write =3D proc_reg_write, .poll =3D proc_reg_poll, .unlocked_ioctl =3D proc_reg_unlocked_ioctl, diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index 48f2d60bd78a..92533bd0e67b 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -869,7 +869,7 @@ static const struct file_operations proc_sys_file_opera= tions =3D { .poll =3D proc_sys_poll, .read_iter =3D proc_sys_read, .write_iter =3D proc_sys_write, - .splice_read =3D generic_file_splice_read, + .splice_read =3D direct_splice_read, .splice_write =3D iter_file_splice_write, .llseek =3D default_llseek, }; diff --git a/fs/proc_namespace.c b/fs/proc_namespace.c index 846f9455ae22..492abbbeff5e 100644 --- a/fs/proc_namespace.c +++ b/fs/proc_namespace.c @@ -324,7 +324,7 @@ static int mountstats_open(struct inode *inode, struct = file *file) const struct file_operations proc_mounts_operations =3D { .open =3D mounts_open, .read_iter =3D seq_read_iter, - .splice_read =3D generic_file_splice_read, + .splice_read =3D direct_splice_read, .llseek =3D seq_lseek, .release =3D mounts_release, .poll =3D mounts_poll, @@ -333,7 +333,7 @@ const struct file_operations proc_mounts_operations =3D= { const struct file_operations proc_mountinfo_operations =3D { .open =3D mountinfo_open, .read_iter =3D seq_read_iter, - .splice_read =3D generic_file_splice_read, + .splice_read =3D direct_splice_read, .llseek =3D seq_lseek, .release =3D mounts_release, .poll =3D mounts_poll, @@ -342,7 +342,7 @@ const struct file_operations proc_mountinfo_operations = =3D { const struct file_operations proc_mountstats_operations =3D { .open =3D mountstats_open, .read_iter =3D seq_read_iter, - .splice_read =3D generic_file_splice_read, + .splice_read =3D direct_splice_read, .llseek =3D seq_lseek, .release =3D mounts_release, }; From nobody Thu Nov 14 04:30:40 2024 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 3B933C64ED6 for ; Tue, 14 Feb 2023 17:15:32 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S232608AbjBNRPa (ORCPT ); Tue, 14 Feb 2023 12:15:30 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41442 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S233010AbjBNRPF (ORCPT ); Tue, 14 Feb 2023 12:15:05 -0500 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 74A102194E for ; Tue, 14 Feb 2023 09:14:14 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1676394853; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=J4i4bwQTS34x6oFo2ZgYGZTE2NYetAQq5iydlM1fsxY=; b=Zvz8HpPRRRLMRsGpzm1QXdbqqPBrWifV8wfrwlRSUrrbKcVqOU2ZxpESf/dGN9nBTPysLz jUpFUVpvVKUnvuTxuhvZrubv08v9sIzOzkqKtEKN7ZUAC8R/g3qOT3VIo5RWnw3FJEk0dR nMUnRLNv1oyVcbSZadc0IPFFnxVVCT8= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-52-ZogOeZ8jO0-C97KX6CrYIA-1; Tue, 14 Feb 2023 12:14:10 -0500 X-MC-Unique: ZogOeZ8jO0-C97KX6CrYIA-1 Received: from smtp.corp.redhat.com (int-mx07.intmail.prod.int.rdu2.redhat.com [10.11.54.7]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 4219F100F907; Tue, 14 Feb 2023 17:14:08 +0000 (UTC) Received: from warthog.procyon.org.uk (unknown [10.33.36.24]) by smtp.corp.redhat.com (Postfix) with ESMTP id 4767E1415108; Tue, 14 Feb 2023 17:14:06 +0000 (UTC) From: David Howells To: Jens Axboe , Al Viro , Christoph Hellwig Cc: David Howells , Matthew Wilcox , Jan Kara , Jeff Layton , David Hildenbrand , Jason Gunthorpe , Logan Gunthorpe , Hillf Danton , linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, Christoph Hellwig , John Hubbard Subject: [PATCH v14 08/17] splice: Do splice read from a file without using ITER_PIPE Date: Tue, 14 Feb 2023 17:13:21 +0000 Message-Id: <20230214171330.2722188-9-dhowells@redhat.com> In-Reply-To: <20230214171330.2722188-1-dhowells@redhat.com> References: <20230214171330.2722188-1-dhowells@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.7 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Make generic_file_splice_read() use filemap_splice_read() and direct_splice_read() rather than using an ITER_PIPE and call_read_iter(). With this, ITER_PIPE is no longer used. Signed-off-by: David Howells cc: Jens Axboe cc: Christoph Hellwig cc: Al Viro cc: David Hildenbrand cc: John Hubbard cc: linux-mm@kvack.org cc: linux-block@vger.kernel.org cc: linux-fsdevel@vger.kernel.org --- Notes: ver #14) - Split out filemap_splice_read() into a separate patch. fs/splice.c | 30 +++++++----------------------- 1 file changed, 7 insertions(+), 23 deletions(-) diff --git a/fs/splice.c b/fs/splice.c index 4c6332854b63..a93478338cec 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -391,29 +391,13 @@ ssize_t generic_file_splice_read(struct file *in, lof= f_t *ppos, struct pipe_inode_info *pipe, size_t len, unsigned int flags) { - struct iov_iter to; - struct kiocb kiocb; - int ret; - - iov_iter_pipe(&to, ITER_DEST, pipe, len); - init_sync_kiocb(&kiocb, in); - kiocb.ki_pos =3D *ppos; - ret =3D call_read_iter(in, &kiocb, &to); - if (ret > 0) { - *ppos =3D kiocb.ki_pos; - file_accessed(in); - } else if (ret < 0) { - /* free what was emitted */ - pipe_discard_from(pipe, to.start_head); - /* - * callers of ->splice_read() expect -EAGAIN on - * "can't put anything in there", rather than -EFAULT. - */ - if (ret =3D=3D -EFAULT) - ret =3D -EAGAIN; - } - - return ret; + if (unlikely(*ppos >=3D file_inode(in)->i_sb->s_maxbytes)) + return 0; + if (unlikely(!len)) + return 0; + if (in->f_flags & O_DIRECT) + return direct_splice_read(in, ppos, pipe, len, flags); + return filemap_splice_read(in, ppos, pipe, len, flags); } EXPORT_SYMBOL(generic_file_splice_read); From nobody Thu Nov 14 04:30:40 2024 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 2E2F4C61DA4 for ; Tue, 14 Feb 2023 17:15:37 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S232395AbjBNRPf (ORCPT ); Tue, 14 Feb 2023 12:15:35 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41444 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S233038AbjBNRPG (ORCPT ); Tue, 14 Feb 2023 12:15:06 -0500 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 44EC711142 for ; Tue, 14 Feb 2023 09:14:14 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1676394853; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=q/WvJQSOUXoHuE1z1Cu24KZfrF9JVSOUntkfW3f52zA=; b=BfNcWIj+35TA+1F7WXk8XHnb43kB+/EQTnoCCNA15tT8ymQ+jzl8WjILg+Fy0ggX/oL8nO 9Kim804Agox4K0Xeg1Wmq0avpok6g9fE5kvhaDosymu/5Vj1g0RTy6zYLhNy2Qo7NZD+gN 5WtdfDYydpgeqo2qPfaycSosPukFN2U= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-298-LwGEXxIWPwy9ik3zHqW1GQ-1; Tue, 14 Feb 2023 12:14:11 -0500 X-MC-Unique: LwGEXxIWPwy9ik3zHqW1GQ-1 Received: from smtp.corp.redhat.com (int-mx03.intmail.prod.int.rdu2.redhat.com [10.11.54.3]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id E8B6380D0E0; Tue, 14 Feb 2023 17:14:10 +0000 (UTC) Received: from warthog.procyon.org.uk (unknown [10.33.36.24]) by smtp.corp.redhat.com (Postfix) with ESMTP id DA42D1121318; Tue, 14 Feb 2023 17:14:08 +0000 (UTC) From: David Howells To: Jens Axboe , Al Viro , Christoph Hellwig Cc: David Howells , Matthew Wilcox , Jan Kara , Jeff Layton , David Hildenbrand , Jason Gunthorpe , Logan Gunthorpe , Hillf Danton , linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, Christoph Hellwig , John Hubbard Subject: [PATCH v14 09/17] iov_iter: Kill ITER_PIPE Date: Tue, 14 Feb 2023 17:13:22 +0000 Message-Id: <20230214171330.2722188-10-dhowells@redhat.com> In-Reply-To: <20230214171330.2722188-1-dhowells@redhat.com> References: <20230214171330.2722188-1-dhowells@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.3 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" The ITER_PIPE-type iterator was only used for generic_file_splice_read(), but that has now been switched to either pull pages directly from the pagecache for buffered file splice-reads or to use ITER_BVEC instead for O_DIRECT file splice-reads. This leaves ITER_PIPE unused - so remove it. Signed-off-by: David Howells Reviewed-by: Christoph Hellwig cc: Jens Axboe cc: Al Viro cc: David Hildenbrand cc: John Hubbard cc: linux-mm@kvack.org cc: linux-block@vger.kernel.org cc: linux-fsdevel@vger.kernel.org --- fs/cifs/file.c | 8 +- include/linux/uio.h | 14 -- lib/iov_iter.c | 429 +------------------------------------------- mm/filemap.c | 3 +- 4 files changed, 5 insertions(+), 449 deletions(-) diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 22dfc1f8b4f1..57ca4eea69dd 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -3806,13 +3806,7 @@ cifs_readdata_to_iov(struct cifs_readdata *rdata, st= ruct iov_iter *iter) size_t copy =3D min_t(size_t, remaining, PAGE_SIZE); size_t written; =20 - if (unlikely(iov_iter_is_pipe(iter))) { - void *addr =3D kmap_atomic(page); - - written =3D copy_to_iter(addr, copy, iter); - kunmap_atomic(addr); - } else - written =3D copy_page_to_iter(page, 0, copy, iter); + written =3D copy_page_to_iter(page, 0, copy, iter); remaining -=3D written; if (written < copy && iov_iter_count(iter) > 0) break; diff --git a/include/linux/uio.h b/include/linux/uio.h index 9f158238edba..dcc0ca5ef491 100644 --- a/include/linux/uio.h +++ b/include/linux/uio.h @@ -11,7 +11,6 @@ #include =20 struct page; -struct pipe_inode_info; =20 struct kvec { void *iov_base; /* and that should *never* hold a userland pointer */ @@ -23,7 +22,6 @@ enum iter_type { ITER_IOVEC, ITER_KVEC, ITER_BVEC, - ITER_PIPE, ITER_XARRAY, ITER_DISCARD, ITER_UBUF, @@ -53,15 +51,10 @@ struct iov_iter { const struct kvec *kvec; const struct bio_vec *bvec; struct xarray *xarray; - struct pipe_inode_info *pipe; void __user *ubuf; }; union { unsigned long nr_segs; - struct { - unsigned int head; - unsigned int start_head; - }; loff_t xarray_start; }; }; @@ -99,11 +92,6 @@ static inline bool iov_iter_is_bvec(const struct iov_ite= r *i) return iov_iter_type(i) =3D=3D ITER_BVEC; } =20 -static inline bool iov_iter_is_pipe(const struct iov_iter *i) -{ - return iov_iter_type(i) =3D=3D ITER_PIPE; -} - static inline bool iov_iter_is_discard(const struct iov_iter *i) { return iov_iter_type(i) =3D=3D ITER_DISCARD; @@ -245,8 +233,6 @@ void iov_iter_kvec(struct iov_iter *i, unsigned int dir= ection, const struct kvec unsigned long nr_segs, size_t count); void iov_iter_bvec(struct iov_iter *i, unsigned int direction, const struc= t bio_vec *bvec, unsigned long nr_segs, size_t count); -void iov_iter_pipe(struct iov_iter *i, unsigned int direction, struct pipe= _inode_info *pipe, - size_t count); void iov_iter_discard(struct iov_iter *i, unsigned int direction, size_t c= ount); void iov_iter_xarray(struct iov_iter *i, unsigned int direction, struct xa= rray *xarray, loff_t start, size_t count); diff --git a/lib/iov_iter.c b/lib/iov_iter.c index 47c484551c59..adc5e8aa8ae8 100644 --- a/lib/iov_iter.c +++ b/lib/iov_iter.c @@ -14,8 +14,6 @@ #include #include =20 -#define PIPE_PARANOIA /* for now */ - /* covers ubuf and kbuf alike */ #define iterate_buf(i, n, base, len, off, __p, STEP) { \ size_t __maybe_unused off =3D 0; \ @@ -186,150 +184,6 @@ static int copyin(void *to, const void __user *from, = size_t n) return res; } =20 -#ifdef PIPE_PARANOIA -static bool sanity(const struct iov_iter *i) -{ - struct pipe_inode_info *pipe =3D i->pipe; - unsigned int p_head =3D pipe->head; - unsigned int p_tail =3D pipe->tail; - unsigned int p_occupancy =3D pipe_occupancy(p_head, p_tail); - unsigned int i_head =3D i->head; - unsigned int idx; - - if (i->last_offset) { - struct pipe_buffer *p; - if (unlikely(p_occupancy =3D=3D 0)) - goto Bad; // pipe must be non-empty - if (unlikely(i_head !=3D p_head - 1)) - goto Bad; // must be at the last buffer... - - p =3D pipe_buf(pipe, i_head); - if (unlikely(p->offset + p->len !=3D abs(i->last_offset))) - goto Bad; // ... at the end of segment - } else { - if (i_head !=3D p_head) - goto Bad; // must be right after the last buffer - } - return true; -Bad: - printk(KERN_ERR "idx =3D %d, offset =3D %d\n", i_head, i->last_offset); - printk(KERN_ERR "head =3D %d, tail =3D %d, buffers =3D %d\n", - p_head, p_tail, pipe->ring_size); - for (idx =3D 0; idx < pipe->ring_size; idx++) - printk(KERN_ERR "[%p %p %d %d]\n", - pipe->bufs[idx].ops, - pipe->bufs[idx].page, - pipe->bufs[idx].offset, - pipe->bufs[idx].len); - WARN_ON(1); - return false; -} -#else -#define sanity(i) true -#endif - -static struct page *push_anon(struct pipe_inode_info *pipe, unsigned size) -{ - struct page *page =3D alloc_page(GFP_USER); - if (page) { - struct pipe_buffer *buf =3D pipe_buf(pipe, pipe->head++); - *buf =3D (struct pipe_buffer) { - .ops =3D &default_pipe_buf_ops, - .page =3D page, - .offset =3D 0, - .len =3D size - }; - } - return page; -} - -static void push_page(struct pipe_inode_info *pipe, struct page *page, - unsigned int offset, unsigned int size) -{ - struct pipe_buffer *buf =3D pipe_buf(pipe, pipe->head++); - *buf =3D (struct pipe_buffer) { - .ops =3D &page_cache_pipe_buf_ops, - .page =3D page, - .offset =3D offset, - .len =3D size - }; - get_page(page); -} - -static inline int last_offset(const struct pipe_buffer *buf) -{ - if (buf->ops =3D=3D &default_pipe_buf_ops) - return buf->len; // buf->offset is 0 for those - else - return -(buf->offset + buf->len); -} - -static struct page *append_pipe(struct iov_iter *i, size_t size, - unsigned int *off) -{ - struct pipe_inode_info *pipe =3D i->pipe; - int offset =3D i->last_offset; - struct pipe_buffer *buf; - struct page *page; - - if (offset > 0 && offset < PAGE_SIZE) { - // some space in the last buffer; add to it - buf =3D pipe_buf(pipe, pipe->head - 1); - size =3D min_t(size_t, size, PAGE_SIZE - offset); - buf->len +=3D size; - i->last_offset +=3D size; - i->count -=3D size; - *off =3D offset; - return buf->page; - } - // OK, we need a new buffer - *off =3D 0; - size =3D min_t(size_t, size, PAGE_SIZE); - if (pipe_full(pipe->head, pipe->tail, pipe->max_usage)) - return NULL; - page =3D push_anon(pipe, size); - if (!page) - return NULL; - i->head =3D pipe->head - 1; - i->last_offset =3D size; - i->count -=3D size; - return page; -} - -static size_t copy_page_to_iter_pipe(struct page *page, size_t offset, siz= e_t bytes, - struct iov_iter *i) -{ - struct pipe_inode_info *pipe =3D i->pipe; - unsigned int head =3D pipe->head; - - if (unlikely(bytes > i->count)) - bytes =3D i->count; - - if (unlikely(!bytes)) - return 0; - - if (!sanity(i)) - return 0; - - if (offset && i->last_offset =3D=3D -offset) { // could we merge it? - struct pipe_buffer *buf =3D pipe_buf(pipe, head - 1); - if (buf->page =3D=3D page) { - buf->len +=3D bytes; - i->last_offset -=3D bytes; - i->count -=3D bytes; - return bytes; - } - } - if (pipe_full(pipe->head, pipe->tail, pipe->max_usage)) - return 0; - - push_page(pipe, page, offset, bytes); - i->last_offset =3D -(offset + bytes); - i->head =3D head; - i->count -=3D bytes; - return bytes; -} - /* * fault_in_iov_iter_readable - fault in iov iterator for reading * @i: iterator @@ -433,46 +287,6 @@ void iov_iter_init(struct iov_iter *i, unsigned int di= rection, } EXPORT_SYMBOL(iov_iter_init); =20 -// returns the offset in partial buffer (if any) -static inline unsigned int pipe_npages(const struct iov_iter *i, int *npag= es) -{ - struct pipe_inode_info *pipe =3D i->pipe; - int used =3D pipe->head - pipe->tail; - int off =3D i->last_offset; - - *npages =3D max((int)pipe->max_usage - used, 0); - - if (off > 0 && off < PAGE_SIZE) { // anon and not full - (*npages)++; - return off; - } - return 0; -} - -static size_t copy_pipe_to_iter(const void *addr, size_t bytes, - struct iov_iter *i) -{ - unsigned int off, chunk; - - if (unlikely(bytes > i->count)) - bytes =3D i->count; - if (unlikely(!bytes)) - return 0; - - if (!sanity(i)) - return 0; - - for (size_t n =3D bytes; n; n -=3D chunk) { - struct page *page =3D append_pipe(i, n, &off); - chunk =3D min_t(size_t, n, PAGE_SIZE - off); - if (!page) - return bytes - n; - memcpy_to_page(page, off, addr, chunk); - addr +=3D chunk; - } - return bytes; -} - static __wsum csum_and_memcpy(void *to, const void *from, size_t len, __wsum sum, size_t off) { @@ -480,44 +294,10 @@ static __wsum csum_and_memcpy(void *to, const void *f= rom, size_t len, return csum_block_add(sum, next, off); } =20 -static size_t csum_and_copy_to_pipe_iter(const void *addr, size_t bytes, - struct iov_iter *i, __wsum *sump) -{ - __wsum sum =3D *sump; - size_t off =3D 0; - unsigned int chunk, r; - - if (unlikely(bytes > i->count)) - bytes =3D i->count; - if (unlikely(!bytes)) - return 0; - - if (!sanity(i)) - return 0; - - while (bytes) { - struct page *page =3D append_pipe(i, bytes, &r); - char *p; - - if (!page) - break; - chunk =3D min_t(size_t, bytes, PAGE_SIZE - r); - p =3D kmap_local_page(page); - sum =3D csum_and_memcpy(p + r, addr + off, chunk, sum, off); - kunmap_local(p); - off +=3D chunk; - bytes -=3D chunk; - } - *sump =3D sum; - return off; -} - size_t _copy_to_iter(const void *addr, size_t bytes, struct iov_iter *i) { if (WARN_ON_ONCE(i->data_source)) return 0; - if (unlikely(iov_iter_is_pipe(i))) - return copy_pipe_to_iter(addr, bytes, i); if (user_backed_iter(i)) might_fault(); iterate_and_advance(i, bytes, base, len, off, @@ -539,42 +319,6 @@ static int copyout_mc(void __user *to, const void *fro= m, size_t n) return n; } =20 -static size_t copy_mc_pipe_to_iter(const void *addr, size_t bytes, - struct iov_iter *i) -{ - size_t xfer =3D 0; - unsigned int off, chunk; - - if (unlikely(bytes > i->count)) - bytes =3D i->count; - if (unlikely(!bytes)) - return 0; - - if (!sanity(i)) - return 0; - - while (bytes) { - struct page *page =3D append_pipe(i, bytes, &off); - unsigned long rem; - char *p; - - if (!page) - break; - chunk =3D min_t(size_t, bytes, PAGE_SIZE - off); - p =3D kmap_local_page(page); - rem =3D copy_mc_to_kernel(p + off, addr + xfer, chunk); - chunk -=3D rem; - kunmap_local(p); - xfer +=3D chunk; - bytes -=3D chunk; - if (rem) { - iov_iter_revert(i, rem); - break; - } - } - return xfer; -} - /** * _copy_mc_to_iter - copy to iter with source memory error exception hand= ling * @addr: source kernel address @@ -594,9 +338,8 @@ static size_t copy_mc_pipe_to_iter(const void *addr, si= ze_t bytes, * alignment and poison alignment assumptions to avoid re-triggering * hardware exceptions. * - * * ITER_KVEC, ITER_PIPE, and ITER_BVEC can return short copies. - * Compare to copy_to_iter() where only ITER_IOVEC attempts might return - * a short copy. + * * ITER_KVEC and ITER_BVEC can return short copies. Compare to + * copy_to_iter() where only ITER_IOVEC attempts might return a short co= py. * * Return: number of bytes copied (may be %0) */ @@ -604,8 +347,6 @@ size_t _copy_mc_to_iter(const void *addr, size_t bytes,= struct iov_iter *i) { if (WARN_ON_ONCE(i->data_source)) return 0; - if (unlikely(iov_iter_is_pipe(i))) - return copy_mc_pipe_to_iter(addr, bytes, i); if (user_backed_iter(i)) might_fault(); __iterate_and_advance(i, bytes, base, len, off, @@ -711,8 +452,6 @@ size_t copy_page_to_iter(struct page *page, size_t offs= et, size_t bytes, return 0; if (WARN_ON_ONCE(i->data_source)) return 0; - if (unlikely(iov_iter_is_pipe(i))) - return copy_page_to_iter_pipe(page, offset, bytes, i); page +=3D offset / PAGE_SIZE; // first subpage offset %=3D PAGE_SIZE; while (1) { @@ -761,36 +500,8 @@ size_t copy_page_from_iter(struct page *page, size_t o= ffset, size_t bytes, } EXPORT_SYMBOL(copy_page_from_iter); =20 -static size_t pipe_zero(size_t bytes, struct iov_iter *i) -{ - unsigned int chunk, off; - - if (unlikely(bytes > i->count)) - bytes =3D i->count; - if (unlikely(!bytes)) - return 0; - - if (!sanity(i)) - return 0; - - for (size_t n =3D bytes; n; n -=3D chunk) { - struct page *page =3D append_pipe(i, n, &off); - char *p; - - if (!page) - return bytes - n; - chunk =3D min_t(size_t, n, PAGE_SIZE - off); - p =3D kmap_local_page(page); - memset(p + off, 0, chunk); - kunmap_local(p); - } - return bytes; -} - size_t iov_iter_zero(size_t bytes, struct iov_iter *i) { - if (unlikely(iov_iter_is_pipe(i))) - return pipe_zero(bytes, i); iterate_and_advance(i, bytes, base, len, count, clear_user(base, len), memset(base, 0, len) @@ -821,32 +532,6 @@ size_t copy_page_from_iter_atomic(struct page *page, u= nsigned offset, size_t byt } EXPORT_SYMBOL(copy_page_from_iter_atomic); =20 -static void pipe_advance(struct iov_iter *i, size_t size) -{ - struct pipe_inode_info *pipe =3D i->pipe; - int off =3D i->last_offset; - - if (!off && !size) { - pipe_discard_from(pipe, i->start_head); // discard everything - return; - } - i->count -=3D size; - while (1) { - struct pipe_buffer *buf =3D pipe_buf(pipe, i->head); - if (off) /* make it relative to the beginning of buffer */ - size +=3D abs(off) - buf->offset; - if (size <=3D buf->len) { - buf->len =3D size; - i->last_offset =3D last_offset(buf); - break; - } - size -=3D buf->len; - i->head++; - off =3D 0; - } - pipe_discard_from(pipe, i->head + 1); // discard everything past this one -} - static void iov_iter_bvec_advance(struct iov_iter *i, size_t size) { const struct bio_vec *bvec, *end; @@ -898,8 +583,6 @@ void iov_iter_advance(struct iov_iter *i, size_t size) iov_iter_iovec_advance(i, size); } else if (iov_iter_is_bvec(i)) { iov_iter_bvec_advance(i, size); - } else if (iov_iter_is_pipe(i)) { - pipe_advance(i, size); } else if (iov_iter_is_discard(i)) { i->count -=3D size; } @@ -913,26 +596,6 @@ void iov_iter_revert(struct iov_iter *i, size_t unroll) if (WARN_ON(unroll > MAX_RW_COUNT)) return; i->count +=3D unroll; - if (unlikely(iov_iter_is_pipe(i))) { - struct pipe_inode_info *pipe =3D i->pipe; - unsigned int head =3D pipe->head; - - while (head > i->start_head) { - struct pipe_buffer *b =3D pipe_buf(pipe, --head); - if (unroll < b->len) { - b->len -=3D unroll; - i->last_offset =3D last_offset(b); - i->head =3D head; - return; - } - unroll -=3D b->len; - pipe_buf_release(pipe, b); - pipe->head--; - } - i->last_offset =3D 0; - i->head =3D head; - return; - } if (unlikely(iov_iter_is_discard(i))) return; if (unroll <=3D i->iov_offset) { @@ -1020,24 +683,6 @@ void iov_iter_bvec(struct iov_iter *i, unsigned int d= irection, } EXPORT_SYMBOL(iov_iter_bvec); =20 -void iov_iter_pipe(struct iov_iter *i, unsigned int direction, - struct pipe_inode_info *pipe, - size_t count) -{ - BUG_ON(direction !=3D READ); - WARN_ON(pipe_full(pipe->head, pipe->tail, pipe->ring_size)); - *i =3D (struct iov_iter){ - .iter_type =3D ITER_PIPE, - .data_source =3D false, - .pipe =3D pipe, - .head =3D pipe->head, - .start_head =3D pipe->head, - .last_offset =3D 0, - .count =3D count - }; -} -EXPORT_SYMBOL(iov_iter_pipe); - /** * iov_iter_xarray - Initialise an I/O iterator to use the pages in an xar= ray * @i: The iterator to initialise. @@ -1162,19 +807,6 @@ bool iov_iter_is_aligned(const struct iov_iter *i, un= signed addr_mask, if (iov_iter_is_bvec(i)) return iov_iter_aligned_bvec(i, addr_mask, len_mask); =20 - if (iov_iter_is_pipe(i)) { - size_t size =3D i->count; - - if (size & len_mask) - return false; - if (size && i->last_offset > 0) { - if (i->last_offset & addr_mask) - return false; - } - - return true; - } - if (iov_iter_is_xarray(i)) { if (i->count & len_mask) return false; @@ -1244,14 +876,6 @@ unsigned long iov_iter_alignment(const struct iov_ite= r *i) if (iov_iter_is_bvec(i)) return iov_iter_alignment_bvec(i); =20 - if (iov_iter_is_pipe(i)) { - size_t size =3D i->count; - - if (size && i->last_offset > 0) - return size | i->last_offset; - return size; - } - if (iov_iter_is_xarray(i)) return (i->xarray_start + i->iov_offset) | i->count; =20 @@ -1303,36 +927,6 @@ static int want_pages_array(struct page ***res, size_= t size, return count; } =20 -static ssize_t pipe_get_pages(struct iov_iter *i, - struct page ***pages, size_t maxsize, unsigned maxpages, - size_t *start) -{ - unsigned int npages, count, off, chunk; - struct page **p; - size_t left; - - if (!sanity(i)) - return -EFAULT; - - *start =3D off =3D pipe_npages(i, &npages); - if (!npages) - return -EFAULT; - count =3D want_pages_array(pages, maxsize, off, min(npages, maxpages)); - if (!count) - return -ENOMEM; - p =3D *pages; - for (npages =3D 0, left =3D maxsize ; npages < count; npages++, left -=3D= chunk) { - struct page *page =3D append_pipe(i, left, &off); - if (!page) - break; - chunk =3D min_t(size_t, left, PAGE_SIZE - off); - get_page(*p++ =3D page); - } - if (!npages) - return -EFAULT; - return maxsize - left; -} - static ssize_t iter_xarray_populate_pages(struct page **pages, struct xarr= ay *xa, pgoff_t index, unsigned int nr_pages) { @@ -1480,8 +1074,6 @@ static ssize_t __iov_iter_get_pages_alloc(struct iov_= iter *i, } return maxsize; } - if (iov_iter_is_pipe(i)) - return pipe_get_pages(i, pages, maxsize, maxpages, start); if (iov_iter_is_xarray(i)) return iter_xarray_get_pages(i, pages, maxsize, maxpages, start); return -EFAULT; @@ -1571,9 +1163,7 @@ size_t csum_and_copy_to_iter(const void *addr, size_t= bytes, void *_csstate, } =20 sum =3D csum_shift(csstate->csum, csstate->off); - if (unlikely(iov_iter_is_pipe(i))) - bytes =3D csum_and_copy_to_pipe_iter(addr, bytes, i, &sum); - else iterate_and_advance(i, bytes, base, len, off, ({ + iterate_and_advance(i, bytes, base, len, off, ({ next =3D csum_and_copy_to_user(addr + off, base, len); sum =3D csum_block_add(sum, next, off); next ? 0 : len; @@ -1658,15 +1248,6 @@ int iov_iter_npages(const struct iov_iter *i, int ma= xpages) return iov_npages(i, maxpages); if (iov_iter_is_bvec(i)) return bvec_npages(i, maxpages); - if (iov_iter_is_pipe(i)) { - int npages; - - if (!sanity(i)) - return 0; - - pipe_npages(i, &npages); - return min(npages, maxpages); - } if (iov_iter_is_xarray(i)) { unsigned offset =3D (i->xarray_start + i->iov_offset) % PAGE_SIZE; int npages =3D DIV_ROUND_UP(offset + i->count, PAGE_SIZE); @@ -1679,10 +1260,6 @@ EXPORT_SYMBOL(iov_iter_npages); const void *dup_iter(struct iov_iter *new, struct iov_iter *old, gfp_t fla= gs) { *new =3D *old; - if (unlikely(iov_iter_is_pipe(new))) { - WARN_ON(1); - return NULL; - } if (iov_iter_is_bvec(new)) return new->bvec =3D kmemdup(new->bvec, new->nr_segs * sizeof(struct bio_vec), diff --git a/mm/filemap.c b/mm/filemap.c index 8c7b135c8e23..c01bbcb9fa92 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -2693,8 +2693,7 @@ ssize_t filemap_read(struct kiocb *iocb, struct iov_i= ter *iter, if (unlikely(iocb->ki_pos >=3D i_size_read(inode))) break; =20 - error =3D filemap_get_pages(iocb, iter->count, &fbatch, - iov_iter_is_pipe(iter)); + error =3D filemap_get_pages(iocb, iter->count, &fbatch, false); if (error < 0) break; From nobody Thu Nov 14 04:30:40 2024 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 094DAC05027 for ; Tue, 14 Feb 2023 17:16:14 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S232014AbjBNRQM (ORCPT ); Tue, 14 Feb 2023 12:16:12 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41588 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S232106AbjBNRPP (ORCPT ); Tue, 14 Feb 2023 12:15:15 -0500 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id D108B274BD for ; Tue, 14 Feb 2023 09:14:20 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1676394860; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=Q5bJAUy0Yf/xIkHbL0gYjIax658ix+8tiANnbR9foj4=; b=Tui+ndrQCX+jhiLls7YdprYH7VbpNEBjKwA8mHPuK+nNt1lxWBm7myryx5yYeaRlVxI+1s 6QZebQ5Q4hpivYVNo3szCKWTHgpvPSnk3xNHf1jdBbhhsc7DV5uUAq8luGLH1IaaZcbZog ENhYI9gSYzzXdl5WcqWOgYzC2SVC5PA= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-411-XEGHYl5VN7igzLHfz7WV0g-1; Tue, 14 Feb 2023 12:14:14 -0500 X-MC-Unique: XEGHYl5VN7igzLHfz7WV0g-1 Received: from smtp.corp.redhat.com (int-mx04.intmail.prod.int.rdu2.redhat.com [10.11.54.4]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 9EDA5800050; Tue, 14 Feb 2023 17:14:13 +0000 (UTC) Received: from warthog.procyon.org.uk (unknown [10.33.36.24]) by smtp.corp.redhat.com (Postfix) with ESMTP id A9D902026D4B; Tue, 14 Feb 2023 17:14:11 +0000 (UTC) From: David Howells To: Jens Axboe , Al Viro , Christoph Hellwig Cc: David Howells , Matthew Wilcox , Jan Kara , Jeff Layton , David Hildenbrand , Jason Gunthorpe , Logan Gunthorpe , Hillf Danton , linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, Christoph Hellwig , John Hubbard Subject: [PATCH v14 10/17] iov_iter: Define flags to qualify page extraction. Date: Tue, 14 Feb 2023 17:13:23 +0000 Message-Id: <20230214171330.2722188-11-dhowells@redhat.com> In-Reply-To: <20230214171330.2722188-1-dhowells@redhat.com> References: <20230214171330.2722188-1-dhowells@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.4 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Define flags to qualify page extraction to pass into iov_iter_*_pages*() rather than passing in FOLL_* flags. For now only a flag to allow peer-to-peer DMA is supported. Signed-off-by: David Howells Reviewed-by: Christoph Hellwig Reviewed-by: John Hubbard cc: Al Viro cc: Jens Axboe cc: Logan Gunthorpe cc: linux-fsdevel@vger.kernel.org cc: linux-block@vger.kernel.org --- Notes: ver #12) - Use __bitwise for the extraction flags typedef. =20 ver #11) - Use __bitwise for the extraction flags. =20 ver #9) - Change extract_flags to extraction_flags. =20 ver #7) - Don't use FOLL_* as a parameter, but rather define constants specifically to use with iov_iter_*_pages*(). - Drop the I/O direction constants for now. block/bio.c | 6 +++--- block/blk-map.c | 8 ++++---- include/linux/uio.h | 10 ++++++++-- lib/iov_iter.c | 14 ++++++++------ 4 files changed, 23 insertions(+), 15 deletions(-) diff --git a/block/bio.c b/block/bio.c index ab59a491a883..b97f3991c904 100644 --- a/block/bio.c +++ b/block/bio.c @@ -1245,11 +1245,11 @@ static int bio_iov_add_zone_append_page(struct bio = *bio, struct page *page, */ static int __bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter) { + iov_iter_extraction_t extraction_flags =3D 0; unsigned short nr_pages =3D bio->bi_max_vecs - bio->bi_vcnt; unsigned short entries_left =3D bio->bi_max_vecs - bio->bi_vcnt; struct bio_vec *bv =3D bio->bi_io_vec + bio->bi_vcnt; struct page **pages =3D (struct page **)bv; - unsigned int gup_flags =3D 0; ssize_t size, left; unsigned len, i =3D 0; size_t offset, trim; @@ -1264,7 +1264,7 @@ static int __bio_iov_iter_get_pages(struct bio *bio, = struct iov_iter *iter) pages +=3D entries_left * (PAGE_PTRS_PER_BVEC - 1); =20 if (bio->bi_bdev && blk_queue_pci_p2pdma(bio->bi_bdev->bd_disk->queue)) - gup_flags |=3D FOLL_PCI_P2PDMA; + extraction_flags |=3D ITER_ALLOW_P2PDMA; =20 /* * Each segment in the iov is required to be a block size multiple. @@ -1275,7 +1275,7 @@ static int __bio_iov_iter_get_pages(struct bio *bio, = struct iov_iter *iter) */ size =3D iov_iter_get_pages(iter, pages, UINT_MAX - bio->bi_iter.bi_size, - nr_pages, &offset, gup_flags); + nr_pages, &offset, extraction_flags); if (unlikely(size <=3D 0)) return size ? size : -EFAULT; =20 diff --git a/block/blk-map.c b/block/blk-map.c index 19940c978c73..080dd60485be 100644 --- a/block/blk-map.c +++ b/block/blk-map.c @@ -265,9 +265,9 @@ static struct bio *blk_rq_map_bio_alloc(struct request = *rq, static int bio_map_user_iov(struct request *rq, struct iov_iter *iter, gfp_t gfp_mask) { + iov_iter_extraction_t extraction_flags =3D 0; unsigned int max_sectors =3D queue_max_hw_sectors(rq->q); unsigned int nr_vecs =3D iov_iter_npages(iter, BIO_MAX_VECS); - unsigned int gup_flags =3D 0; struct bio *bio; int ret; int j; @@ -280,7 +280,7 @@ static int bio_map_user_iov(struct request *rq, struct = iov_iter *iter, return -ENOMEM; =20 if (blk_queue_pci_p2pdma(rq->q)) - gup_flags |=3D FOLL_PCI_P2PDMA; + extraction_flags |=3D ITER_ALLOW_P2PDMA; =20 while (iov_iter_count(iter)) { struct page **pages, *stack_pages[UIO_FASTIOV]; @@ -291,10 +291,10 @@ static int bio_map_user_iov(struct request *rq, struc= t iov_iter *iter, if (nr_vecs <=3D ARRAY_SIZE(stack_pages)) { pages =3D stack_pages; bytes =3D iov_iter_get_pages(iter, pages, LONG_MAX, - nr_vecs, &offs, gup_flags); + nr_vecs, &offs, extraction_flags); } else { bytes =3D iov_iter_get_pages_alloc(iter, &pages, - LONG_MAX, &offs, gup_flags); + LONG_MAX, &offs, extraction_flags); } if (unlikely(bytes <=3D 0)) { ret =3D bytes ? bytes : -EFAULT; diff --git a/include/linux/uio.h b/include/linux/uio.h index dcc0ca5ef491..af70e4c9ea27 100644 --- a/include/linux/uio.h +++ b/include/linux/uio.h @@ -12,6 +12,8 @@ =20 struct page; =20 +typedef unsigned int __bitwise iov_iter_extraction_t; + struct kvec { void *iov_base; /* and that should *never* hold a userland pointer */ size_t iov_len; @@ -238,12 +240,12 @@ void iov_iter_xarray(struct iov_iter *i, unsigned int= direction, struct xarray * loff_t start, size_t count); ssize_t iov_iter_get_pages(struct iov_iter *i, struct page **pages, size_t maxsize, unsigned maxpages, size_t *start, - unsigned gup_flags); + iov_iter_extraction_t extraction_flags); ssize_t iov_iter_get_pages2(struct iov_iter *i, struct page **pages, size_t maxsize, unsigned maxpages, size_t *start); ssize_t iov_iter_get_pages_alloc(struct iov_iter *i, struct page ***pages, size_t maxsize, size_t *start, - unsigned gup_flags); + iov_iter_extraction_t extraction_flags); ssize_t iov_iter_get_pages_alloc2(struct iov_iter *i, struct page ***pages, size_t maxsize, size_t *start); int iov_iter_npages(const struct iov_iter *i, int maxpages); @@ -346,4 +348,8 @@ static inline void iov_iter_ubuf(struct iov_iter *i, un= signed int direction, }; } =20 +/* Flags for iov_iter_get/extract_pages*() */ +/* Allow P2PDMA on the extracted pages */ +#define ITER_ALLOW_P2PDMA ((__force iov_iter_extraction_t)0x01) + #endif diff --git a/lib/iov_iter.c b/lib/iov_iter.c index adc5e8aa8ae8..34ee3764d0fa 100644 --- a/lib/iov_iter.c +++ b/lib/iov_iter.c @@ -1020,9 +1020,9 @@ static struct page *first_bvec_segment(const struct i= ov_iter *i, static ssize_t __iov_iter_get_pages_alloc(struct iov_iter *i, struct page ***pages, size_t maxsize, unsigned int maxpages, size_t *start, - unsigned int gup_flags) + iov_iter_extraction_t extraction_flags) { - unsigned int n; + unsigned int n, gup_flags =3D 0; =20 if (maxsize > i->count) maxsize =3D i->count; @@ -1030,6 +1030,8 @@ static ssize_t __iov_iter_get_pages_alloc(struct iov_= iter *i, return 0; if (maxsize > MAX_RW_COUNT) maxsize =3D MAX_RW_COUNT; + if (extraction_flags & ITER_ALLOW_P2PDMA) + gup_flags |=3D FOLL_PCI_P2PDMA; =20 if (likely(user_backed_iter(i))) { unsigned long addr; @@ -1081,14 +1083,14 @@ static ssize_t __iov_iter_get_pages_alloc(struct io= v_iter *i, =20 ssize_t iov_iter_get_pages(struct iov_iter *i, struct page **pages, size_t maxsize, unsigned maxpages, - size_t *start, unsigned gup_flags) + size_t *start, iov_iter_extraction_t extraction_flags) { if (!maxpages) return 0; BUG_ON(!pages); =20 return __iov_iter_get_pages_alloc(i, &pages, maxsize, maxpages, - start, gup_flags); + start, extraction_flags); } EXPORT_SYMBOL_GPL(iov_iter_get_pages); =20 @@ -1101,14 +1103,14 @@ EXPORT_SYMBOL(iov_iter_get_pages2); =20 ssize_t iov_iter_get_pages_alloc(struct iov_iter *i, struct page ***pages, size_t maxsize, - size_t *start, unsigned gup_flags) + size_t *start, iov_iter_extraction_t extraction_flags) { ssize_t len; =20 *pages =3D NULL; =20 len =3D __iov_iter_get_pages_alloc(i, pages, maxsize, ~0U, start, - gup_flags); + extraction_flags); if (len <=3D 0) { kvfree(*pages); *pages =3D NULL; From nobody Thu Nov 14 04:30:40 2024 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 951BBC64ED8 for ; Tue, 14 Feb 2023 17:16:04 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S232222AbjBNRQC (ORCPT ); Tue, 14 Feb 2023 12:16:02 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41590 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S232094AbjBNRPP (ORCPT ); Tue, 14 Feb 2023 12:15:15 -0500 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 7725828221 for ; Tue, 14 Feb 2023 09:14:23 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1676394862; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=G1Az3EBV9SOKUryI5J/Szq/Ryazp7FGIfAGTSAZ/Hdw=; b=Dls3xRBDeI9dUydHRP0ag0H7WI/23bBTe6AJrffnP1Tk6EciZZRrYcudzWOFLxJ8i+6qzJ hHnOoMjilAQkATOkQn4Jpmcxqxd1DQ1vuv40PLw0l2JyTLtuNZEIhVi1Dchv9hvh6Mldg1 SwUUHpF/vOnS+BHOjXnKvcMlr4ILaKI= Received: from mimecast-mx02.redhat.com (mx3-rdu2.redhat.com [66.187.233.73]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-564-UFOQF0M0P1K7pE4T4FEjsg-1; Tue, 14 Feb 2023 12:14:17 -0500 X-MC-Unique: UFOQF0M0P1K7pE4T4FEjsg-1 Received: from smtp.corp.redhat.com (int-mx03.intmail.prod.int.rdu2.redhat.com [10.11.54.3]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 601132817236; Tue, 14 Feb 2023 17:14:16 +0000 (UTC) Received: from warthog.procyon.org.uk (unknown [10.33.36.24]) by smtp.corp.redhat.com (Postfix) with ESMTP id 42A961121318; Tue, 14 Feb 2023 17:14:14 +0000 (UTC) From: David Howells To: Jens Axboe , Al Viro , Christoph Hellwig Cc: David Howells , Matthew Wilcox , Jan Kara , Jeff Layton , David Hildenbrand , Jason Gunthorpe , Logan Gunthorpe , Hillf Danton , linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, Christoph Hellwig , John Hubbard Subject: [PATCH v14 11/17] iov_iter: Add a function to extract a page list from an iterator Date: Tue, 14 Feb 2023 17:13:24 +0000 Message-Id: <20230214171330.2722188-12-dhowells@redhat.com> In-Reply-To: <20230214171330.2722188-1-dhowells@redhat.com> References: <20230214171330.2722188-1-dhowells@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.3 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Add a function, iov_iter_extract_pages(), to extract a list of pages from an iterator. The pages may be returned with a pin added or nothing, depending on the type of iterator. Add a second function, iov_iter_extract_will_pin(), to determine how the cleanup should be done. There are two cases: (1) ITER_IOVEC or ITER_UBUF iterator. Extracted pages will have pins (FOLL_PIN) obtained on them so that a concurrent fork() will forcibly copy the page so that DMA is done to/from the parent's buffer and is unavailable to/unaffected by the child process. iov_iter_extract_will_pin() will return true for this case. The caller should use something like unpin_user_page() to dispose of the page. (2) Any other sort of iterator. No refs or pins are obtained on the page, the assumption is made that the caller will manage page retention. iov_iter_extract_will_pin() will return false. The pages don't need additional disposal. Signed-off-by: David Howells Reviewed-by: Christoph Hellwig cc: Al Viro cc: John Hubbard cc: David Hildenbrand cc: Matthew Wilcox cc: linux-fsdevel@vger.kernel.org cc: linux-mm@kvack.org --- Notes: ver #12) - ITER_PIPE is gone, so drop related bits. - Don't specify FOLL_PIN as that's implied by pin_user_pages_fast(). =20 ver #11) - Fix iov_iter_extract_kvec_pages() to include the offset into the pag= e in the returned starting offset. - Use __bitwise for the extraction flags =20 ver #10) - Fix use of i->kvec in iov_iter_extract_bvec_pages() to be i->bvec. =20 ver #9) - Rename iov_iter_extract_mode() to iov_iter_extract_will_pin() and ma= ke it return true/false not FOLL_PIN/0 as FOLL_PIN is going to be made private to mm/. - Change extract_flags to extraction_flags. =20 ver #8) - It seems that all DIO is supposed to be done under FOLL_PIN now, and= not FOLL_GET, so switch to only using pin_user_pages() for user-backed iters. - Wrap an argument in brackets in the iov_iter_extract_mode() macro. - Drop the extract_flags argument to iov_iter_extract_mode() for now [hch]. =20 ver #7) - Switch to passing in iter-specific flags rather than FOLL_* flags. - Drop the direction flags for now. - Use ITER_ALLOW_P2PDMA to request FOLL_PCI_P2PDMA. - Disallow use of ITER_ALLOW_P2PDMA with non-user-backed iter. - Add support for extraction from KVEC-type iters. - Use iov_iter_advance() rather than open-coding it. - Make BVEC- and KVEC-type skip over initial empty vectors. =20 ver #6) - Add back the function to indicate the cleanup mode. - Drop the cleanup_mode return arg to iov_iter_extract_pages(). - Pass FOLL_SOURCE/DEST_BUF in gup_flags. Check this against the iter data_source. =20 ver #4) - Use ITER_SOURCE/DEST instead of WRITE/READ. - Allow additional FOLL_* flags, such as FOLL_PCI_P2PDMA to be passed = in. =20 ver #3) - Switch to using EXPORT_SYMBOL_GPL to prevent indirect 3rd-party acce= ss to get/pin_user_pages_fast()[1]. include/linux/uio.h | 27 ++++- lib/iov_iter.c | 264 ++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 290 insertions(+), 1 deletion(-) diff --git a/include/linux/uio.h b/include/linux/uio.h index af70e4c9ea27..cf6658066736 100644 --- a/include/linux/uio.h +++ b/include/linux/uio.h @@ -347,9 +347,34 @@ static inline void iov_iter_ubuf(struct iov_iter *i, u= nsigned int direction, .count =3D count }; } - /* Flags for iov_iter_get/extract_pages*() */ /* Allow P2PDMA on the extracted pages */ #define ITER_ALLOW_P2PDMA ((__force iov_iter_extraction_t)0x01) =20 +ssize_t iov_iter_extract_pages(struct iov_iter *i, struct page ***pages, + size_t maxsize, unsigned int maxpages, + iov_iter_extraction_t extraction_flags, + size_t *offset0); + +/** + * iov_iter_extract_will_pin - Indicate how pages from the iterator will b= e retained + * @iter: The iterator + * + * Examine the iterator and indicate by returning true or false as to how,= if + * at all, pages extracted from the iterator will be retained by the extra= ction + * function. + * + * %true indicates that the pages will have a pin placed in them that the + * caller must unpin. This is must be done for DMA/async DIO to force for= k() + * to forcibly copy a page for the child (the parent must retain the origi= nal + * page). + * + * %false indicates that no measures are taken and that it's up to the cal= ler + * to retain the pages. + */ +static inline bool iov_iter_extract_will_pin(const struct iov_iter *iter) +{ + return user_backed_iter(iter); +} + #endif diff --git a/lib/iov_iter.c b/lib/iov_iter.c index 34ee3764d0fa..8d34b6552179 100644 --- a/lib/iov_iter.c +++ b/lib/iov_iter.c @@ -1487,3 +1487,267 @@ void iov_iter_restore(struct iov_iter *i, struct io= v_iter_state *state) i->iov -=3D state->nr_segs - i->nr_segs; i->nr_segs =3D state->nr_segs; } + +/* + * Extract a list of contiguous pages from an ITER_XARRAY iterator. This = does not + * get references on the pages, nor does it get a pin on them. + */ +static ssize_t iov_iter_extract_xarray_pages(struct iov_iter *i, + struct page ***pages, size_t maxsize, + unsigned int maxpages, + iov_iter_extraction_t extraction_flags, + size_t *offset0) +{ + struct page *page, **p; + unsigned int nr =3D 0, offset; + loff_t pos =3D i->xarray_start + i->iov_offset; + pgoff_t index =3D pos >> PAGE_SHIFT; + XA_STATE(xas, i->xarray, index); + + offset =3D pos & ~PAGE_MASK; + *offset0 =3D offset; + + maxpages =3D want_pages_array(pages, maxsize, offset, maxpages); + if (!maxpages) + return -ENOMEM; + p =3D *pages; + + rcu_read_lock(); + for (page =3D xas_load(&xas); page; page =3D xas_next(&xas)) { + if (xas_retry(&xas, page)) + continue; + + /* Has the page moved or been split? */ + if (unlikely(page !=3D xas_reload(&xas))) { + xas_reset(&xas); + continue; + } + + p[nr++] =3D find_subpage(page, xas.xa_index); + if (nr =3D=3D maxpages) + break; + } + rcu_read_unlock(); + + maxsize =3D min_t(size_t, nr * PAGE_SIZE - offset, maxsize); + iov_iter_advance(i, maxsize); + return maxsize; +} + +/* + * Extract a list of contiguous pages from an ITER_BVEC iterator. This do= es + * not get references on the pages, nor does it get a pin on them. + */ +static ssize_t iov_iter_extract_bvec_pages(struct iov_iter *i, + struct page ***pages, size_t maxsize, + unsigned int maxpages, + iov_iter_extraction_t extraction_flags, + size_t *offset0) +{ + struct page **p, *page; + size_t skip =3D i->iov_offset, offset; + int k; + + for (;;) { + if (i->nr_segs =3D=3D 0) + return 0; + maxsize =3D min(maxsize, i->bvec->bv_len - skip); + if (maxsize) + break; + i->iov_offset =3D 0; + i->nr_segs--; + i->bvec++; + skip =3D 0; + } + + skip +=3D i->bvec->bv_offset; + page =3D i->bvec->bv_page + skip / PAGE_SIZE; + offset =3D skip % PAGE_SIZE; + *offset0 =3D offset; + + maxpages =3D want_pages_array(pages, maxsize, offset, maxpages); + if (!maxpages) + return -ENOMEM; + p =3D *pages; + for (k =3D 0; k < maxpages; k++) + p[k] =3D page + k; + + maxsize =3D min_t(size_t, maxsize, maxpages * PAGE_SIZE - offset); + iov_iter_advance(i, maxsize); + return maxsize; +} + +/* + * Extract a list of virtually contiguous pages from an ITER_KVEC iterator. + * This does not get references on the pages, nor does it get a pin on the= m. + */ +static ssize_t iov_iter_extract_kvec_pages(struct iov_iter *i, + struct page ***pages, size_t maxsize, + unsigned int maxpages, + iov_iter_extraction_t extraction_flags, + size_t *offset0) +{ + struct page **p, *page; + const void *kaddr; + size_t skip =3D i->iov_offset, offset, len; + int k; + + for (;;) { + if (i->nr_segs =3D=3D 0) + return 0; + maxsize =3D min(maxsize, i->kvec->iov_len - skip); + if (maxsize) + break; + i->iov_offset =3D 0; + i->nr_segs--; + i->kvec++; + skip =3D 0; + } + + kaddr =3D i->kvec->iov_base + skip; + offset =3D (unsigned long)kaddr & ~PAGE_MASK; + *offset0 =3D offset; + + maxpages =3D want_pages_array(pages, maxsize, offset, maxpages); + if (!maxpages) + return -ENOMEM; + p =3D *pages; + + kaddr -=3D offset; + len =3D offset + maxsize; + for (k =3D 0; k < maxpages; k++) { + size_t seg =3D min_t(size_t, len, PAGE_SIZE); + + if (is_vmalloc_or_module_addr(kaddr)) + page =3D vmalloc_to_page(kaddr); + else + page =3D virt_to_page(kaddr); + + p[k] =3D page; + len -=3D seg; + kaddr +=3D PAGE_SIZE; + } + + maxsize =3D min_t(size_t, maxsize, maxpages * PAGE_SIZE - offset); + iov_iter_advance(i, maxsize); + return maxsize; +} + +/* + * Extract a list of contiguous pages from a user iterator and get a pin on + * each of them. This should only be used if the iterator is user-backed + * (IOBUF/UBUF). + * + * It does not get refs on the pages, but the pages must be unpinned by the + * caller once the transfer is complete. + * + * This is safe to be used where background IO/DMA *is* going to be modify= ing + * the buffer; using a pin rather than a ref makes forces fork() to give t= he + * child a copy of the page. + */ +static ssize_t iov_iter_extract_user_pages(struct iov_iter *i, + struct page ***pages, + size_t maxsize, + unsigned int maxpages, + iov_iter_extraction_t extraction_flags, + size_t *offset0) +{ + unsigned long addr; + unsigned int gup_flags =3D 0; + size_t offset; + int res; + + if (i->data_source =3D=3D ITER_DEST) + gup_flags |=3D FOLL_WRITE; + if (extraction_flags & ITER_ALLOW_P2PDMA) + gup_flags |=3D FOLL_PCI_P2PDMA; + if (i->nofault) + gup_flags |=3D FOLL_NOFAULT; + + addr =3D first_iovec_segment(i, &maxsize); + *offset0 =3D offset =3D addr % PAGE_SIZE; + addr &=3D PAGE_MASK; + maxpages =3D want_pages_array(pages, maxsize, offset, maxpages); + if (!maxpages) + return -ENOMEM; + res =3D pin_user_pages_fast(addr, maxpages, gup_flags, *pages); + if (unlikely(res <=3D 0)) + return res; + maxsize =3D min_t(size_t, maxsize, res * PAGE_SIZE - offset); + iov_iter_advance(i, maxsize); + return maxsize; +} + +/** + * iov_iter_extract_pages - Extract a list of contiguous pages from an ite= rator + * @i: The iterator to extract from + * @pages: Where to return the list of pages + * @maxsize: The maximum amount of iterator to extract + * @maxpages: The maximum size of the list of pages + * @extraction_flags: Flags to qualify request + * @offset0: Where to return the starting offset into (*@pages)[0] + * + * Extract a list of contiguous pages from the current point of the iterat= or, + * advancing the iterator. The maximum number of pages and the maximum am= ount + * of page contents can be set. + * + * If *@pages is NULL, a page list will be allocated to the required size = and + * *@pages will be set to its base. If *@pages is not NULL, it will be as= sumed + * that the caller allocated a page list at least @maxpages in size and th= is + * will be filled in. + * + * @extraction_flags can have ITER_ALLOW_P2PDMA set to request peer-to-pee= r DMA + * be allowed on the pages extracted. + * + * The iov_iter_extract_will_pin() function can be used to query how clean= up + * should be performed. + * + * Extra refs or pins on the pages may be obtained as follows: + * + * (*) If the iterator is user-backed (ITER_IOVEC/ITER_UBUF), pins will be + * added to the pages, but refs will not be taken. + * iov_iter_extract_will_pin() will return true. + * + * (*) If the iterator is ITER_KVEC, ITER_BVEC or ITER_XARRAY, the pages = are + * merely listed; no extra refs or pins are obtained. + * iov_iter_extract_will_pin() will return 0. + * + * Note also: + * + * (*) Use with ITER_DISCARD is not supported as that has no content. + * + * On success, the function sets *@pages to the new pagelist, if allocated= , and + * sets *offset0 to the offset into the first page. + * + * It may also return -ENOMEM and -EFAULT. + */ +ssize_t iov_iter_extract_pages(struct iov_iter *i, + struct page ***pages, + size_t maxsize, + unsigned int maxpages, + iov_iter_extraction_t extraction_flags, + size_t *offset0) +{ + maxsize =3D min_t(size_t, min_t(size_t, maxsize, i->count), MAX_RW_COUNT); + if (!maxsize) + return 0; + + if (likely(user_backed_iter(i))) + return iov_iter_extract_user_pages(i, pages, maxsize, + maxpages, extraction_flags, + offset0); + if (iov_iter_is_kvec(i)) + return iov_iter_extract_kvec_pages(i, pages, maxsize, + maxpages, extraction_flags, + offset0); + if (iov_iter_is_bvec(i)) + return iov_iter_extract_bvec_pages(i, pages, maxsize, + maxpages, extraction_flags, + offset0); + if (iov_iter_is_xarray(i)) + return iov_iter_extract_xarray_pages(i, pages, maxsize, + maxpages, extraction_flags, + offset0); + return -EFAULT; +} +EXPORT_SYMBOL_GPL(iov_iter_extract_pages); From nobody Thu Nov 14 04:30:40 2024 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 0330FC6379F for ; Tue, 14 Feb 2023 17:16:17 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233149AbjBNRQQ (ORCPT ); Tue, 14 Feb 2023 12:16:16 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41592 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S232221AbjBNRPR (ORCPT ); Tue, 14 Feb 2023 12:15:17 -0500 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id ADEA12BF23 for ; Tue, 14 Feb 2023 09:14:25 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1676394864; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=AA3SDxqAw6tV+jeEhwhhXCCPQx7iw6NiGu/cAPT0aQk=; b=BCSrFSm9m4vCFf6Z47aB/J5gIS1hhLR9Jora5YI3jbQWFJxTF6Dk9i7xJS2VU7WmjNZILq HHtLpHmi6534FdaMK1rjGeqt2Yd+EDp+u5sbZehD45CDaUGyULi/wh6owDjNebyE37s8a4 WeQMySuKgnsoqnNVSu4XO97LQGV267A= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-515-DQOy9zFmOZSICpQlPKxSDw-1; Tue, 14 Feb 2023 12:14:20 -0500 X-MC-Unique: DQOy9zFmOZSICpQlPKxSDw-1 Received: from smtp.corp.redhat.com (int-mx04.intmail.prod.int.rdu2.redhat.com [10.11.54.4]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id EE7998027EB; Tue, 14 Feb 2023 17:14:18 +0000 (UTC) Received: from warthog.procyon.org.uk (unknown [10.33.36.24]) by smtp.corp.redhat.com (Postfix) with ESMTP id 179BC2026D4B; Tue, 14 Feb 2023 17:14:16 +0000 (UTC) From: David Howells To: Jens Axboe , Al Viro , Christoph Hellwig Cc: David Howells , Matthew Wilcox , Jan Kara , Jeff Layton , David Hildenbrand , Jason Gunthorpe , Logan Gunthorpe , Hillf Danton , linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, John Hubbard Subject: [PATCH v14 12/17] iomap: Don't get an reference on ZERO_PAGE for direct I/O block zeroing Date: Tue, 14 Feb 2023 17:13:25 +0000 Message-Id: <20230214171330.2722188-13-dhowells@redhat.com> In-Reply-To: <20230214171330.2722188-1-dhowells@redhat.com> References: <20230214171330.2722188-1-dhowells@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.4 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" ZERO_PAGE can't go away, no need to hold an extra reference. Signed-off-by: David Howells Reviewed-by: David Hildenbrand Reviewed-by: John Hubbard cc: Al Viro cc: David Hildenbrand cc: linux-fsdevel@vger.kernel.org --- fs/iomap/direct-io.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c index 9804714b1751..47db4ead1e74 100644 --- a/fs/iomap/direct-io.c +++ b/fs/iomap/direct-io.c @@ -202,7 +202,7 @@ static void iomap_dio_zero(const struct iomap_iter *ite= r, struct iomap_dio *dio, bio->bi_private =3D dio; bio->bi_end_io =3D iomap_dio_bio_end_io; =20 - get_page(page); + bio_set_flag(bio, BIO_NO_PAGE_REF); __bio_add_page(bio, page, len, 0); iomap_dio_submit_bio(iter, dio, bio, pos); } From nobody Thu Nov 14 04:30:40 2024 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 4CB14C05027 for ; Tue, 14 Feb 2023 17:16:10 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S232633AbjBNRQI (ORCPT ); Tue, 14 Feb 2023 12:16:08 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41600 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S232123AbjBNRPQ (ORCPT ); Tue, 14 Feb 2023 12:15:16 -0500 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 9D35DCC08 for ; Tue, 14 Feb 2023 09:14:28 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1676394867; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=9PNg3WPaFWzxoLjSof2qU8psox3R+qpqOMDOMyjiLi8=; b=OqGm8++iNY9gpd7HVvl4dXkXgTPrOeS+T2u6jJH/TeoTy74o4FxCaR/ZhTDwoWhmHjVkkZ tMDI5U3SxsAHrD9Sr8A7T9bexi1F9uJ51+0UF7M/u47Vk5IvUxEqYBZftd8mWMMQgsVHtj hQ+7PfjlecsLUff+fb1FzDQx9wwgk6c= Received: from mimecast-mx02.redhat.com (mx3-rdu2.redhat.com [66.187.233.73]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-88-Gp-NasbONamSrWK0LXoPrQ-1; Tue, 14 Feb 2023 12:14:22 -0500 X-MC-Unique: Gp-NasbONamSrWK0LXoPrQ-1 Received: from smtp.corp.redhat.com (int-mx08.intmail.prod.int.rdu2.redhat.com [10.11.54.8]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 8698B1C08798; Tue, 14 Feb 2023 17:14:21 +0000 (UTC) Received: from warthog.procyon.org.uk (unknown [10.33.36.24]) by smtp.corp.redhat.com (Postfix) with ESMTP id 9817DC15BAD; Tue, 14 Feb 2023 17:14:19 +0000 (UTC) From: David Howells To: Jens Axboe , Al Viro , Christoph Hellwig Cc: David Howells , Matthew Wilcox , Jan Kara , Jeff Layton , David Hildenbrand , Jason Gunthorpe , Logan Gunthorpe , Hillf Danton , linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, Christoph Hellwig , John Hubbard Subject: [PATCH v14 13/17] block: Fix bio_flagged() so that gcc can better optimise it Date: Tue, 14 Feb 2023 17:13:26 +0000 Message-Id: <20230214171330.2722188-14-dhowells@redhat.com> In-Reply-To: <20230214171330.2722188-1-dhowells@redhat.com> References: <20230214171330.2722188-1-dhowells@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.8 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Fix bio_flagged() so that multiple instances of it, such as: if (bio_flagged(bio, BIO_PAGE_REFFED) || bio_flagged(bio, BIO_PAGE_PINNED)) can be combined by the gcc optimiser into a single test in assembly (arguably, this is a compiler optimisation issue[1]). The missed optimisation stems from bio_flagged() comparing the result of the bitwise-AND to zero. This results in an out-of-line bio_release_page() being compiled to something like: <+0>: mov 0x14(%rdi),%eax <+3>: test $0x1,%al <+5>: jne 0xffffffff816dac53 <+7>: test $0x2,%al <+9>: je 0xffffffff816dac5c <+11>: movzbl %sil,%esi <+15>: jmp 0xffffffff816daba1 <__bio_release_pages> <+20>: jmp 0xffffffff81d0b800 <__x86_return_thunk> However, the test is superfluous as the return type is bool. Removing it results in: <+0>: testb $0x3,0x14(%rdi) <+4>: je 0xffffffff816e4af4 <+6>: movzbl %sil,%esi <+10>: jmp 0xffffffff816dab7c <__bio_release_pages> <+15>: jmp 0xffffffff81d0b7c0 <__x86_return_thunk> instead. Also, the MOVZBL instruction looks unnecessary[2] - I think it's just 're-booling' the mark_dirty parameter. Signed-off-by: David Howells Reviewed-by: Christoph Hellwig Reviewed-by: John Hubbard cc: Jens Axboe cc: linux-block@vger.kernel.org Link: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D108370 [1] Link: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D108371 [2] Link: https://lore.kernel.org/r/167391056756.2311931.356007731815807265.stg= it@warthog.procyon.org.uk/ # v6 --- include/linux/bio.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/bio.h b/include/linux/bio.h index c1da63f6c808..10366b8bdb13 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -227,7 +227,7 @@ static inline void bio_cnt_set(struct bio *bio, unsigne= d int count) =20 static inline bool bio_flagged(struct bio *bio, unsigned int bit) { - return (bio->bi_flags & (1U << bit)) !=3D 0; + return bio->bi_flags & (1U << bit); } =20 static inline void bio_set_flag(struct bio *bio, unsigned int bit) From nobody Thu Nov 14 04:30:40 2024 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id B1DC3C61DA4 for ; Tue, 14 Feb 2023 17:16:22 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233163AbjBNRQU (ORCPT ); Tue, 14 Feb 2023 12:16:20 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41636 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S232822AbjBNRP3 (ORCPT ); Tue, 14 Feb 2023 12:15:29 -0500 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 4EADE12045 for ; Tue, 14 Feb 2023 09:14:29 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1676394868; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=dqniJi1fcm9DuVb2jGUzwy7ATEE/wnT//KjRosSsKEA=; b=LlnHaPK0NvrQVvkv3Nv1y3lmLSlNb5eVRmfB+/SvEj+zIep3SM6WP+tT5TE1FCGQi0RVEW 4moWoM8qvSXjPtvhBeCEHTeCqSCgWSoET5rdBMDl+rXFdhE8QbkZ590M9ietY8M888RgKp 0RkoH6Hi/YhDjyRlMPncgtf+yf7mLYc= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-376-CS34Hu8FOKSxyZVTcNhLCw-1; Tue, 14 Feb 2023 12:14:25 -0500 X-MC-Unique: CS34Hu8FOKSxyZVTcNhLCw-1 Received: from smtp.corp.redhat.com (int-mx06.intmail.prod.int.rdu2.redhat.com [10.11.54.6]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 4250A85A588; Tue, 14 Feb 2023 17:14:24 +0000 (UTC) Received: from warthog.procyon.org.uk (unknown [10.33.36.24]) by smtp.corp.redhat.com (Postfix) with ESMTP id 4367C2166B26; Tue, 14 Feb 2023 17:14:22 +0000 (UTC) From: David Howells To: Jens Axboe , Al Viro , Christoph Hellwig Cc: David Howells , Matthew Wilcox , Jan Kara , Jeff Layton , David Hildenbrand , Jason Gunthorpe , Logan Gunthorpe , Hillf Danton , linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, Christoph Hellwig , John Hubbard Subject: [PATCH v14 14/17] block: Replace BIO_NO_PAGE_REF with BIO_PAGE_REFFED with inverted logic Date: Tue, 14 Feb 2023 17:13:27 +0000 Message-Id: <20230214171330.2722188-15-dhowells@redhat.com> In-Reply-To: <20230214171330.2722188-1-dhowells@redhat.com> References: <20230214171330.2722188-1-dhowells@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.6 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" From: Christoph Hellwig Replace BIO_NO_PAGE_REF with a BIO_PAGE_REFFED flag that has the inverted meaning is only set when a page reference has been acquired that needs to be released by bio_release_pages(). Signed-off-by: Christoph Hellwig Signed-off-by: David Howells Reviewed-by: John Hubbard cc: Al Viro cc: Jens Axboe cc: Jan Kara cc: Matthew Wilcox cc: Logan Gunthorpe cc: linux-block@vger.kernel.org --- Notes: ver #8) - Split out from another patch [hch]. - Don't default to BIO_PAGE_REFFED [hch]. =20 ver #5) - Split from patch that uses iov_iter_extract_pages(). block/bio.c | 2 +- block/blk-map.c | 1 + fs/direct-io.c | 2 ++ fs/iomap/direct-io.c | 1 - include/linux/bio.h | 2 +- include/linux/blk_types.h | 2 +- 6 files changed, 6 insertions(+), 4 deletions(-) diff --git a/block/bio.c b/block/bio.c index b97f3991c904..bf9bf53232be 100644 --- a/block/bio.c +++ b/block/bio.c @@ -1198,7 +1198,6 @@ void bio_iov_bvec_set(struct bio *bio, struct iov_ite= r *iter) bio->bi_io_vec =3D (struct bio_vec *)iter->bvec; bio->bi_iter.bi_bvec_done =3D iter->iov_offset; bio->bi_iter.bi_size =3D size; - bio_set_flag(bio, BIO_NO_PAGE_REF); bio_set_flag(bio, BIO_CLONED); } =20 @@ -1343,6 +1342,7 @@ int bio_iov_iter_get_pages(struct bio *bio, struct io= v_iter *iter) return 0; } =20 + bio_set_flag(bio, BIO_PAGE_REFFED); do { ret =3D __bio_iov_iter_get_pages(bio, iter); } while (!ret && iov_iter_count(iter) && !bio_full(bio, 0)); diff --git a/block/blk-map.c b/block/blk-map.c index 080dd60485be..f1f70b50388d 100644 --- a/block/blk-map.c +++ b/block/blk-map.c @@ -282,6 +282,7 @@ static int bio_map_user_iov(struct request *rq, struct = iov_iter *iter, if (blk_queue_pci_p2pdma(rq->q)) extraction_flags |=3D ITER_ALLOW_P2PDMA; =20 + bio_set_flag(bio, BIO_PAGE_REFFED); while (iov_iter_count(iter)) { struct page **pages, *stack_pages[UIO_FASTIOV]; ssize_t bytes; diff --git a/fs/direct-io.c b/fs/direct-io.c index 03d381377ae1..07810465fc9d 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -403,6 +403,8 @@ dio_bio_alloc(struct dio *dio, struct dio_submit *sdio, bio->bi_end_io =3D dio_bio_end_aio; else bio->bi_end_io =3D dio_bio_end_io; + /* for now require references for all pages */ + bio_set_flag(bio, BIO_PAGE_REFFED); sdio->bio =3D bio; sdio->logical_offset_in_bio =3D sdio->cur_page_fs_offset; } diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c index 47db4ead1e74..c0e75900e754 100644 --- a/fs/iomap/direct-io.c +++ b/fs/iomap/direct-io.c @@ -202,7 +202,6 @@ static void iomap_dio_zero(const struct iomap_iter *ite= r, struct iomap_dio *dio, bio->bi_private =3D dio; bio->bi_end_io =3D iomap_dio_bio_end_io; =20 - bio_set_flag(bio, BIO_NO_PAGE_REF); __bio_add_page(bio, page, len, 0); iomap_dio_submit_bio(iter, dio, bio, pos); } diff --git a/include/linux/bio.h b/include/linux/bio.h index 10366b8bdb13..805957c99147 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -484,7 +484,7 @@ void zero_fill_bio(struct bio *bio); =20 static inline void bio_release_pages(struct bio *bio, bool mark_dirty) { - if (!bio_flagged(bio, BIO_NO_PAGE_REF)) + if (bio_flagged(bio, BIO_PAGE_REFFED)) __bio_release_pages(bio, mark_dirty); } =20 diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h index 99be590f952f..7daa261f4f98 100644 --- a/include/linux/blk_types.h +++ b/include/linux/blk_types.h @@ -318,7 +318,7 @@ struct bio { * bio flags */ enum { - BIO_NO_PAGE_REF, /* don't put release vec pages */ + BIO_PAGE_REFFED, /* put pages in bio_release_pages() */ BIO_CLONED, /* doesn't own data */ BIO_BOUNCED, /* bio is a bounce bio */ BIO_QUIET, /* Make BIO Quiet */ From nobody Thu Nov 14 04:30:40 2024 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 0D6F3C6379F for ; Tue, 14 Feb 2023 17:16:36 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233183AbjBNRQe (ORCPT ); Tue, 14 Feb 2023 12:16:34 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:42078 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S232897AbjBNRPc (ORCPT ); Tue, 14 Feb 2023 12:15:32 -0500 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id A94232ED45 for ; Tue, 14 Feb 2023 09:14:34 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1676394873; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=GZNnQ4PFOAio/M36ha2lSqPmGocJT5Xn1bbdC3s23DY=; b=A86jUU/+zSRUwcxRGnBnJ/JiAmh/TzTeIQi3BSUhLO+DyIOBw+XSuvUWkaevc8LNKhzvS4 GInrRiQd7BA3K4SlWON/GqOCO8BRTm2WL6h4mqXAtF9HO2vziEMVbI7TnrlPXLTOZ33bOm CYrAD6Yx70ups078mdAPH9rKZ1oHROQ= Received: from mimecast-mx02.redhat.com (mx3-rdu2.redhat.com [66.187.233.73]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-466-lS0UXfJPMyu2AbeIcCgt5A-1; Tue, 14 Feb 2023 12:14:28 -0500 X-MC-Unique: lS0UXfJPMyu2AbeIcCgt5A-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.rdu2.redhat.com [10.11.54.2]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 5E66A38123AE; Tue, 14 Feb 2023 17:14:27 +0000 (UTC) Received: from warthog.procyon.org.uk (unknown [10.33.36.24]) by smtp.corp.redhat.com (Postfix) with ESMTP id D969340C1106; Tue, 14 Feb 2023 17:14:24 +0000 (UTC) From: David Howells To: Jens Axboe , Al Viro , Christoph Hellwig Cc: David Howells , Matthew Wilcox , Jan Kara , Jeff Layton , David Hildenbrand , Jason Gunthorpe , Logan Gunthorpe , Hillf Danton , linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, Christoph Hellwig , John Hubbard Subject: [PATCH v14 15/17] block: Add BIO_PAGE_PINNED and associated infrastructure Date: Tue, 14 Feb 2023 17:13:28 +0000 Message-Id: <20230214171330.2722188-16-dhowells@redhat.com> In-Reply-To: <20230214171330.2722188-1-dhowells@redhat.com> References: <20230214171330.2722188-1-dhowells@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.2 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Add BIO_PAGE_PINNED to indicate that the pages in a bio are pinned (FOLL_PIN) and that the pin will need removing. Signed-off-by: David Howells Reviewed-by: Christoph Hellwig Reviewed-by: John Hubbard cc: Al Viro cc: Jens Axboe cc: Jan Kara cc: Matthew Wilcox cc: Logan Gunthorpe cc: linux-block@vger.kernel.org --- Notes: ver #10) - Drop bio_set_cleanup_mode(), open coding it instead. =20 ver #9) - Only consider pinning in bio_set_cleanup_mode(). Ref'ing pages in struct bio is going away. - page_put_unpin() is removed; call unpin_user_page() and put_page() directly. - Use bio_release_page() in __bio_release_pages(). - BIO_PAGE_PINNED and BIO_PAGE_REFFED can't both be set, so use if-else when testing both of them. =20 ver #8) - Move the infrastructure to clean up pinned pages to this patch [hch]. - Put BIO_PAGE_PINNED before BIO_PAGE_REFFED as the latter should probably be removed at some point. FOLL_PIN can then be renumbered first. block/bio.c | 6 +++--- block/blk.h | 12 ++++++++++++ include/linux/bio.h | 3 ++- include/linux/blk_types.h | 1 + 4 files changed, 18 insertions(+), 4 deletions(-) diff --git a/block/bio.c b/block/bio.c index bf9bf53232be..547e38883934 100644 --- a/block/bio.c +++ b/block/bio.c @@ -1176,7 +1176,7 @@ void __bio_release_pages(struct bio *bio, bool mark_d= irty) 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); @@ -1496,8 +1496,8 @@ 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 unpin each page and will run one bio_put() against the + * BIO. */ =20 static void bio_dirty_fn(struct work_struct *work); diff --git a/block/blk.h b/block/blk.h index 4c3b3325219a..f02381405311 100644 --- a/block/blk.h +++ b/block/blk.h @@ -425,6 +425,18 @@ int bio_add_hw_page(struct request_queue *q, struct bi= o *bio, struct page *page, unsigned int len, unsigned int offset, unsigned int max_sectors, bool *same_page); =20 +/* + * Clean up a page appropriately, where the page may be pinned, may have a + * ref taken on it or neither. + */ +static inline void bio_release_page(struct bio *bio, struct page *page) +{ + if (bio_flagged(bio, BIO_PAGE_PINNED)) + unpin_user_page(page); + else if (bio_flagged(bio, BIO_PAGE_REFFED)) + put_page(page); +} + struct request_queue *blk_alloc_queue(int node_id); =20 int disk_scan_partitions(struct gendisk *disk, fmode_t mode, void *owner); diff --git a/include/linux/bio.h b/include/linux/bio.h index 805957c99147..b2c09997d79c 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -484,7 +484,8 @@ void zero_fill_bio(struct bio *bio); =20 static inline void bio_release_pages(struct bio *bio, bool mark_dirty) { - if (bio_flagged(bio, BIO_PAGE_REFFED)) + if (bio_flagged(bio, BIO_PAGE_REFFED) || + bio_flagged(bio, BIO_PAGE_PINNED)) __bio_release_pages(bio, mark_dirty); } =20 diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h index 7daa261f4f98..a0e339ff3d09 100644 --- a/include/linux/blk_types.h +++ b/include/linux/blk_types.h @@ -318,6 +318,7 @@ struct bio { * bio flags */ enum { + BIO_PAGE_PINNED, /* Unpin pages in bio_release_pages() */ BIO_PAGE_REFFED, /* put pages in bio_release_pages() */ BIO_CLONED, /* doesn't own data */ BIO_BOUNCED, /* bio is a bounce bio */ From nobody Thu Nov 14 04:30:40 2024 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id CA1ECC05027 for ; Tue, 14 Feb 2023 17:16:38 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233188AbjBNRQh (ORCPT ); Tue, 14 Feb 2023 12:16:37 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:42118 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S232951AbjBNRPd (ORCPT ); Tue, 14 Feb 2023 12:15:33 -0500 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 27A3429E36 for ; Tue, 14 Feb 2023 09:14:37 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1676394876; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=YX3eOMnEnHmNO2fsUm7QGJ9rpDZ86GbDPShQEUjns7k=; b=EKrrMRfQKTxOP5rEaXzjQ6FEpn9XK8ZSktVPhas91VKMrkTZo497EEZGg9KQnT11OmkA/4 YOP5IyPJb4Q0WQtHl8BKqLeDVaQcYbwEbQNhkR7r3YAh2vJe/f2ViAge+5ZEgRZiv0KEFe dfuMOnOFD1MTvTBGGTgHss0k9KvWsIE= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-52-XUarQZNaNTu_e2ZfEJgYPg-1; Tue, 14 Feb 2023 12:14:31 -0500 X-MC-Unique: XUarQZNaNTu_e2ZfEJgYPg-1 Received: from smtp.corp.redhat.com (int-mx03.intmail.prod.int.rdu2.redhat.com [10.11.54.3]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 3BA24100F90C; Tue, 14 Feb 2023 17:14:30 +0000 (UTC) Received: from warthog.procyon.org.uk (unknown [10.33.36.24]) by smtp.corp.redhat.com (Postfix) with ESMTP id 14FD41121318; Tue, 14 Feb 2023 17:14:27 +0000 (UTC) From: David Howells To: Jens Axboe , Al Viro , Christoph Hellwig Cc: David Howells , Matthew Wilcox , Jan Kara , Jeff Layton , David Hildenbrand , Jason Gunthorpe , Logan Gunthorpe , Hillf Danton , linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, Christoph Hellwig , John Hubbard Subject: [PATCH v14 16/17] block: Convert bio_iov_iter_get_pages to use iov_iter_extract_pages Date: Tue, 14 Feb 2023 17:13:29 +0000 Message-Id: <20230214171330.2722188-17-dhowells@redhat.com> In-Reply-To: <20230214171330.2722188-1-dhowells@redhat.com> References: <20230214171330.2722188-1-dhowells@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.3 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" This will pin pages or leave them unaltered rather than getting a ref on them as appropriate to the iterator. The pages need to be pinned for DIO rather than having refs taken on them to prevent VM copy-on-write from malfunctioning during a concurrent fork() (the result of the I/O could otherwise end up being affected by/visible to the child process). Signed-off-by: David Howells Reviewed-by: Christoph Hellwig Reviewed-by: John Hubbard cc: Al Viro cc: Jens Axboe cc: Jan Kara cc: Matthew Wilcox cc: Logan Gunthorpe cc: linux-block@vger.kernel.org --- Notes: ver #10) - Drop bio_set_cleanup_mode(), open coding it instead. =20 ver #8) - Split the patch up a bit [hch]. - We should only be using pinned/non-pinned pages and not ref'd pages, so adjust the comments appropriately. =20 ver #7) - Don't treat BIO_PAGE_REFFED/PINNED as being the same as FOLL_GET/PIN. =20 ver #5) - Transcribe the FOLL_* flags returned by iov_iter_extract_pages() to BIO_* flags and got rid of bi_cleanup_mode. - Replaced BIO_NO_PAGE_REF to BIO_PAGE_REFFED in the preceding patch. block/bio.c | 23 ++++++++++++----------- 1 file changed, 12 insertions(+), 11 deletions(-) diff --git a/block/bio.c b/block/bio.c index 547e38883934..fc57f0aa098e 100644 --- a/block/bio.c +++ b/block/bio.c @@ -1212,7 +1212,7 @@ static int bio_iov_add_page(struct bio *bio, struct p= age *page, } =20 if (same_page) - put_page(page); + bio_release_page(bio, page); return 0; } =20 @@ -1226,7 +1226,7 @@ static int bio_iov_add_zone_append_page(struct bio *b= io, struct page *page, queue_max_zone_append_sectors(q), &same_page) !=3D len) return -EINVAL; if (same_page) - put_page(page); + bio_release_page(bio, page); return 0; } =20 @@ -1237,10 +1237,10 @@ 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. + * Extracts pages from *iter and appends them to @bio's bvec array. The p= ages + * will have to be cleaned up in the way indicated by the BIO_PAGE_PINNED = flag. + * For a 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) { @@ -1272,9 +1272,9 @@ 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 =3D iov_iter_get_pages(iter, pages, - UINT_MAX - bio->bi_iter.bi_size, - nr_pages, &offset, extraction_flags); + size =3D iov_iter_extract_pages(iter, &pages, + UINT_MAX - bio->bi_iter.bi_size, + nr_pages, extraction_flags, &offset); if (unlikely(size <=3D 0)) return size ? size : -EFAULT; =20 @@ -1307,7 +1307,7 @@ 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++]); =20 return ret; } @@ -1342,7 +1342,8 @@ int bio_iov_iter_get_pages(struct bio *bio, struct io= v_iter *iter) return 0; } =20 - bio_set_flag(bio, BIO_PAGE_REFFED); + if (iov_iter_extract_will_pin(iter)) + bio_set_flag(bio, BIO_PAGE_PINNED); do { ret =3D __bio_iov_iter_get_pages(bio, iter); } while (!ret && iov_iter_count(iter) && !bio_full(bio, 0)); From nobody Thu Nov 14 04:30:40 2024 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id D138AC05027 for ; Tue, 14 Feb 2023 17:16:25 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233170AbjBNRQX (ORCPT ); Tue, 14 Feb 2023 12:16:23 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:42142 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231877AbjBNRPa (ORCPT ); Tue, 14 Feb 2023 12:15:30 -0500 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id BF3492ED50 for ; Tue, 14 Feb 2023 09:14:37 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1676394877; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=7RYncx3d2QCFno76YxQ26A3gARIW2WsBk/g5n5z32ew=; b=MHzV3ri0QY4HLx0TnzO7rM++fakyjk5QuQ43/dvi82B2kjvKbS2jUrA7E9wdXo7/eu9An5 i4YImwlJnQyIMk8LNFbtwwZvoYekfRSWzNgMA8EYdChZ2GRPBQfkTJmySdAKb3x8dhVUb+ Se37SU7jqbtlTOTMzBvYiFyewCklTcI= Received: from mimecast-mx02.redhat.com (mx3-rdu2.redhat.com [66.187.233.73]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-528-53yWEtX-Pm2chQYNZopvZA-1; Tue, 14 Feb 2023 12:14:33 -0500 X-MC-Unique: 53yWEtX-Pm2chQYNZopvZA-1 Received: from smtp.corp.redhat.com (int-mx05.intmail.prod.int.rdu2.redhat.com [10.11.54.5]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id C77ED38123AE; Tue, 14 Feb 2023 17:14:32 +0000 (UTC) Received: from warthog.procyon.org.uk (unknown [10.33.36.24]) by smtp.corp.redhat.com (Postfix) with ESMTP id D47AE18EC2; Tue, 14 Feb 2023 17:14:30 +0000 (UTC) From: David Howells To: Jens Axboe , Al Viro , Christoph Hellwig Cc: David Howells , Matthew Wilcox , Jan Kara , Jeff Layton , David Hildenbrand , Jason Gunthorpe , Logan Gunthorpe , Hillf Danton , linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, Christoph Hellwig , John Hubbard Subject: [PATCH v14 17/17] block: convert bio_map_user_iov to use iov_iter_extract_pages Date: Tue, 14 Feb 2023 17:13:30 +0000 Message-Id: <20230214171330.2722188-18-dhowells@redhat.com> In-Reply-To: <20230214171330.2722188-1-dhowells@redhat.com> References: <20230214171330.2722188-1-dhowells@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.5 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" This will pin pages or leave them unaltered rather than getting a ref on them as appropriate to the iterator. The pages need to be pinned for DIO rather than having refs taken on them to prevent VM copy-on-write from malfunctioning during a concurrent fork() (the result of the I/O could otherwise end up being visible to/affected by the child process). Signed-off-by: David Howells Reviewed-by: Christoph Hellwig Reviewed-by: John Hubbard cc: Al Viro cc: Jens Axboe cc: Jan Kara cc: Matthew Wilcox cc: Logan Gunthorpe cc: linux-block@vger.kernel.org --- Notes: ver #10) - Drop bio_set_cleanup_mode(), open coding it instead. =20 ver #8) - Split the patch up a bit [hch]. - We should only be using pinned/non-pinned pages and not ref'd pages, so adjust the comments appropriately. =20 ver #7) - Don't treat BIO_PAGE_REFFED/PINNED as being the same as FOLL_GET/PIN. =20 ver #5) - Transcribe the FOLL_* flags returned by iov_iter_extract_pages() to BIO_* flags and got rid of bi_cleanup_mode. - Replaced BIO_NO_PAGE_REF to BIO_PAGE_REFFED in the preceding patch. block/blk-map.c | 23 +++++++++++------------ 1 file changed, 11 insertions(+), 12 deletions(-) diff --git a/block/blk-map.c b/block/blk-map.c index f1f70b50388d..0f1593e144da 100644 --- a/block/blk-map.c +++ b/block/blk-map.c @@ -281,22 +281,21 @@ static int bio_map_user_iov(struct request *rq, struc= t iov_iter *iter, =20 if (blk_queue_pci_p2pdma(rq->q)) extraction_flags |=3D ITER_ALLOW_P2PDMA; + if (iov_iter_extract_will_pin(iter)) + bio_set_flag(bio, BIO_PAGE_PINNED); =20 - bio_set_flag(bio, BIO_PAGE_REFFED); while (iov_iter_count(iter)) { - struct page **pages, *stack_pages[UIO_FASTIOV]; + struct page *stack_pages[UIO_FASTIOV]; + struct page **pages =3D stack_pages; ssize_t bytes; size_t offs; int npages; =20 - if (nr_vecs <=3D ARRAY_SIZE(stack_pages)) { - pages =3D stack_pages; - bytes =3D iov_iter_get_pages(iter, pages, LONG_MAX, - nr_vecs, &offs, extraction_flags); - } else { - bytes =3D iov_iter_get_pages_alloc(iter, &pages, - LONG_MAX, &offs, extraction_flags); - } + if (nr_vecs > ARRAY_SIZE(stack_pages)) + pages =3D NULL; + + bytes =3D iov_iter_extract_pages(iter, &pages, LONG_MAX, + nr_vecs, extraction_flags, &offs); if (unlikely(bytes <=3D 0)) { ret =3D bytes ? bytes : -EFAULT; goto out_unmap; @@ -318,7 +317,7 @@ static int bio_map_user_iov(struct request *rq, struct = iov_iter *iter, if (!bio_add_hw_page(rq->q, bio, page, n, offs, max_sectors, &same_page)) { if (same_page) - put_page(page); + bio_release_page(bio, page); break; } =20 @@ -330,7 +329,7 @@ static int bio_map_user_iov(struct request *rq, struct = iov_iter *iter, * release the pages we didn't map into the bio, if any */ while (j < npages) - put_page(pages[j++]); + bio_release_page(bio, pages[j++]); if (pages !=3D stack_pages) kvfree(pages); /* couldn't stuff something into bio? */