From: Jan Kara <jack@suse.cz> To: Jens Axboe <axboe@kernel.dk> Cc: Christoph Hellwig <hch@infradead.org>, linux-fsdevel@vger.kernel.org, linux-ext4@vger.kernel.org, ocfs2-devel@oss.oracle.com, Evgeniy Dushistov <dushistov@mail.ru>, linux-ntfs-dev@lists.sourceforge.net, Jan Kara <jack@suse.cz> Subject: [PATCH 2/6] direct-io: Use clean_bdev_aliases() instead of handmade iteration Date: Fri, 4 Nov 2016 18:08:12 +0100 [thread overview] Message-ID: <1478279296-6575-3-git-send-email-jack@suse.cz> (raw) In-Reply-To: <1478279296-6575-1-git-send-email-jack@suse.cz> Use new provided function instead of an iteration through all allocated blocks. Signed-off-by: Jan Kara <jack@suse.cz> --- fs/direct-io.c | 28 +++++++--------------------- 1 file changed, 7 insertions(+), 21 deletions(-) diff --git a/fs/direct-io.c b/fs/direct-io.c index fb9aa16a7727..12ac532a444a 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -843,24 +843,6 @@ submit_page_section(struct dio *dio, struct dio_submit *sdio, struct page *page, } /* - * Clean any dirty buffers in the blockdev mapping which alias newly-created - * file blocks. Only called for S_ISREG files - blockdevs do not set - * buffer_new - */ -static void clean_blockdev_aliases(struct dio *dio, struct buffer_head *map_bh) -{ - unsigned i; - unsigned nblocks; - - nblocks = map_bh->b_size >> dio->inode->i_blkbits; - - for (i = 0; i < nblocks; i++) { - unmap_underlying_metadata(map_bh->b_bdev, - map_bh->b_blocknr + i); - } -} - -/* * If we are not writing the entire block and get_block() allocated * the block for us, we need to fill-in the unused portion of the * block with zeros. This happens only if user-buffer, fileoffset or @@ -960,11 +942,15 @@ static int do_direct_IO(struct dio *dio, struct dio_submit *sdio, goto do_holes; sdio->blocks_available = - map_bh->b_size >> sdio->blkbits; + map_bh->b_size >> blkbits; sdio->next_block_for_io = map_bh->b_blocknr << sdio->blkfactor; - if (buffer_new(map_bh)) - clean_blockdev_aliases(dio, map_bh); + if (buffer_new(map_bh)) { + clean_bdev_aliases( + map_bh->b_bdev, + map_bh->b_blocknr, + map_bh->b_size >> blkbits); + } if (!sdio->blkfactor) goto do_holes; -- 2.6.6
WARNING: multiple messages have this Message-ID (diff)
From: Jan Kara <jack@suse.cz> To: Jens Axboe <axboe@kernel.dk> Cc: Christoph Hellwig <hch@infradead.org>, linux-fsdevel@vger.kernel.org, linux-ext4@vger.kernel.org, ocfs2-devel@oss.oracle.com, Evgeniy Dushistov <dushistov@mail.ru>, linux-ntfs-dev@lists.sourceforge.net, Jan Kara <jack@suse.cz> Subject: [Ocfs2-devel] [PATCH 2/6] direct-io: Use clean_bdev_aliases() instead of handmade iteration Date: Fri, 4 Nov 2016 18:08:12 +0100 [thread overview] Message-ID: <1478279296-6575-3-git-send-email-jack@suse.cz> (raw) In-Reply-To: <1478279296-6575-1-git-send-email-jack@suse.cz> Use new provided function instead of an iteration through all allocated blocks. Signed-off-by: Jan Kara <jack@suse.cz> --- fs/direct-io.c | 28 +++++++--------------------- 1 file changed, 7 insertions(+), 21 deletions(-) diff --git a/fs/direct-io.c b/fs/direct-io.c index fb9aa16a7727..12ac532a444a 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -843,24 +843,6 @@ submit_page_section(struct dio *dio, struct dio_submit *sdio, struct page *page, } /* - * Clean any dirty buffers in the blockdev mapping which alias newly-created - * file blocks. Only called for S_ISREG files - blockdevs do not set - * buffer_new - */ -static void clean_blockdev_aliases(struct dio *dio, struct buffer_head *map_bh) -{ - unsigned i; - unsigned nblocks; - - nblocks = map_bh->b_size >> dio->inode->i_blkbits; - - for (i = 0; i < nblocks; i++) { - unmap_underlying_metadata(map_bh->b_bdev, - map_bh->b_blocknr + i); - } -} - -/* * If we are not writing the entire block and get_block() allocated * the block for us, we need to fill-in the unused portion of the * block with zeros. This happens only if user-buffer, fileoffset or @@ -960,11 +942,15 @@ static int do_direct_IO(struct dio *dio, struct dio_submit *sdio, goto do_holes; sdio->blocks_available = - map_bh->b_size >> sdio->blkbits; + map_bh->b_size >> blkbits; sdio->next_block_for_io = map_bh->b_blocknr << sdio->blkfactor; - if (buffer_new(map_bh)) - clean_blockdev_aliases(dio, map_bh); + if (buffer_new(map_bh)) { + clean_bdev_aliases( + map_bh->b_bdev, + map_bh->b_blocknr, + map_bh->b_size >> blkbits); + } if (!sdio->blkfactor) goto do_holes; -- 2.6.6
next prev parent reply other threads:[~2016-11-04 17:08 UTC|newest] Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top 2016-11-04 17:08 [PATCH 0/6 v2] fs: Provide function to unmap metadata for a range of blocks Jan Kara 2016-11-04 17:08 ` [Ocfs2-devel] " Jan Kara 2016-11-04 17:08 ` [PATCH 1/6] " Jan Kara 2016-11-04 17:08 ` [Ocfs2-devel] " Jan Kara 2016-11-04 17:08 ` Jan Kara [this message] 2016-11-04 17:08 ` [Ocfs2-devel] [PATCH 2/6] direct-io: Use clean_bdev_aliases() instead of handmade iteration Jan Kara 2016-11-04 17:08 ` [PATCH 3/6] ext4: Use clean_bdev_aliases() instead of iteration Jan Kara 2016-11-04 17:08 ` [Ocfs2-devel] " Jan Kara 2016-11-04 17:08 ` Jan Kara 2016-11-07 2:40 ` [lkp] [ext4] adad5aa544: fio.write_bw_MBps +4074.4% improvement kernel test robot 2016-11-07 2:40 ` kernel test robot 2016-11-04 17:08 ` [PATCH 4/6] ext2: Use clean_bdev_aliases() instead of iteration Jan Kara 2016-11-04 17:08 ` [Ocfs2-devel] " Jan Kara 2016-11-04 17:08 ` Jan Kara 2016-11-04 17:08 ` [PATCH 5/6] fs: Add helper to clean bdev aliases under a bh and use it Jan Kara 2016-11-04 17:08 ` [Ocfs2-devel] " Jan Kara 2016-11-11 1:17 ` [fs] 85b4a9366d: [No primary change] fio.latency_10us% 124.5% increase kernel test robot 2016-11-04 17:08 ` [PATCH 6/6] fs: Remove unmap_underlying_metadata Jan Kara 2016-11-04 17:08 ` [Ocfs2-devel] " Jan Kara 2016-11-04 17:08 ` Jan Kara 2016-11-04 20:32 ` [PATCH 0/6 v2] fs: Provide function to unmap metadata for a range of blocks Jens Axboe
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=1478279296-6575-3-git-send-email-jack@suse.cz \ --to=jack@suse.cz \ --cc=axboe@kernel.dk \ --cc=dushistov@mail.ru \ --cc=hch@infradead.org \ --cc=linux-ext4@vger.kernel.org \ --cc=linux-fsdevel@vger.kernel.org \ --cc=linux-ntfs-dev@lists.sourceforge.net \ --cc=ocfs2-devel@oss.oracle.com \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.