linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Darrick J. Wong" <darrick.wong@oracle.com>
To: "Theodore Ts'o" <tytso@mit.edu>
Cc: linux-ext4 <linux-ext4@vger.kernel.org>,
	linux-fsdevel <linux-fsdevel@vger.kernel.org>
Subject: [RFC PATCH 2/2] mm: Gate stable page writes on the bdi flag
Date: Fri, 26 Oct 2012 18:36:08 -0700	[thread overview]
Message-ID: <20121027013608.GB19591@blackbox.djwong.org> (raw)
In-Reply-To: <20121026101909.GB19617@blackbox.djwong.org>

Create a helper function to decide if a particular address space is backed by a
device that requires stable page writes.  For each wait_on_page_writeback call
that was inserted in the original stable page write patchset, change it to wait
only if the helper says it's necessary.  This should eliminate unnecessary
waiting for devices that don't require page contents to be immutable during
writeout.

Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---

 fs/buffer.c         |    3 ++-
 fs/ext4/inode.c     |    4 ++--
 include/linux/fs.h  |    2 ++
 mm/filemap.c        |    3 ++-
 mm/page-writeback.c |    7 +++++++
 5 files changed, 15 insertions(+), 4 deletions(-)

diff --git a/fs/buffer.c b/fs/buffer.c
index b5f0442..f508ece 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -2334,7 +2334,8 @@ int __block_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf,
 	if (unlikely(ret < 0))
 		goto out_unlock;
 	set_page_dirty(page);
-	wait_on_page_writeback(page);
+	if (page->mapping && mapping_needs_stable_writes(page->mapping))
+		wait_on_page_writeback(page);
 	return 0;
 out_unlock:
 	unlock_page(page);
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index b3c243b..82908b8 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -4813,8 +4813,8 @@ int ext4_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
 	if (page_has_buffers(page)) {
 		if (!walk_page_buffers(NULL, page_buffers(page), 0, len, NULL,
 					ext4_bh_unmapped)) {
-			/* Wait so that we don't change page under IO */
-			wait_on_page_writeback(page);
+			if (mapping_needs_stable_writes(mapping))
+				wait_on_page_writeback(page);
 			ret = VM_FAULT_LOCKED;
 			goto out;
 		}
diff --git a/include/linux/fs.h b/include/linux/fs.h
index b33cfc9..d99db0e 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -466,6 +466,8 @@ struct block_device {
 	struct percpu_rw_semaphore	bd_block_size_semaphore;
 };
 
+int mapping_needs_stable_writes(struct address_space *mapping);
+
 /*
  * Radix-tree tags, for tagging dirty and writeback pages within the pagecache
  * radix trees
diff --git a/mm/filemap.c b/mm/filemap.c
index 83efee7..aedc6bd 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -2274,7 +2274,8 @@ repeat:
 		return NULL;
 	}
 found:
-	wait_on_page_writeback(page);
+	if (mapping_needs_stable_writes(mapping))
+		wait_on_page_writeback(page);
 	return page;
 }
 EXPORT_SYMBOL(grab_cache_page_write_begin);
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 830893b..d0f042c 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -2275,3 +2275,10 @@ int mapping_tagged(struct address_space *mapping, int tag)
 	return radix_tree_tagged(&mapping->page_tree, tag);
 }
 EXPORT_SYMBOL(mapping_tagged);
+
+int mapping_needs_stable_writes(struct address_space *mapping)
+{
+	return mapping->backing_dev_info &&
+	       bdi_cap_stable_writes(mapping->backing_dev_info);
+}
+EXPORT_SYMBOL_GPL(mapping_needs_stable_writes);

  parent reply	other threads:[~2012-10-27  1:36 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-10-26 10:19 semi-stable page writes Darrick J. Wong
2012-10-27  1:35 ` [RFC PATCH 1/2] bdi: Create a flag to indicate that a backing device needs stable " Darrick J. Wong
2012-10-29 18:13   ` Jan Kara
2012-10-29 18:30     ` Jan Kara
2012-10-29 23:48       ` NeilBrown
2012-10-30  0:10         ` Jan Kara
2012-10-30  0:34           ` NeilBrown
2012-10-30 13:38             ` Jan Kara
2012-10-30 21:49               ` NeilBrown
2012-10-30  4:10   ` Martin K. Petersen
2012-10-30  4:48     ` NeilBrown
2012-10-30 12:19       ` Martin K. Petersen
2012-10-30 20:14         ` Darrick J. Wong
2012-10-30 22:14           ` NeilBrown
2012-10-30 23:58             ` Boaz Harrosh
2012-10-31  8:56             ` Darrick J. Wong
2012-10-31 11:56               ` Jan Kara
2012-10-31 19:36                 ` Darrick J. Wong
2012-10-31 23:12                   ` Boaz Harrosh
2012-11-01  5:51                     ` Darrick J. Wong
2012-11-01  6:25                       ` Boaz Harrosh
2012-11-01  8:59                   ` Jan Kara
2012-11-01 17:24                     ` Boaz Harrosh
2012-11-01 22:42                       ` Jan Kara
2012-10-30 22:40   ` Boaz Harrosh
2012-10-27  1:36 ` Darrick J. Wong [this message]
2012-10-29 18:28   ` [RFC PATCH 2/2] mm: Gate stable page writes on the bdi flag Jan Kara
2012-10-31  8:58     ` Darrick J. Wong
2012-10-29 22:01 ` semi-stable page writes Dave Chinner
2012-10-30  1:00   ` Theodore Ts'o
2012-10-30 23:30     ` Dave Chinner
2012-10-31 11:45       ` Jan Kara
2012-10-30 20:40   ` Darrick J. Wong
2012-10-30 23:43     ` Dave Chinner
2012-10-31  9:05       ` Darrick J. Wong

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=20121027013608.GB19591@blackbox.djwong.org \
    --to=darrick.wong@oracle.com \
    --cc=linux-ext4@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=tytso@mit.edu \
    /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: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).