All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ross Zwisler <ross.zwisler@linux.intel.com>
To: Jan Kara <jack@suse.cz>
Cc: linux-fsdevel@vger.kernel.org, linux-mm@kvack.org,
	linux-nvdimm@lists.01.org
Subject: Re: [PATCH 0/3 v1] dax: Clear dirty bits after flushing caches
Date: Tue, 28 Jun 2016 15:41:59 -0600	[thread overview]
Message-ID: <20160628214159.GB15457@linux.intel.com> (raw)
In-Reply-To: <1466523915-14644-1-git-send-email-jack@suse.cz>

On Tue, Jun 21, 2016 at 05:45:12PM +0200, Jan Kara wrote:
> Hello,
> 
> currently we never clear dirty bits in the radix tree of a DAX inode. Thus
> fsync(2) or even periodical writeback flush all the dirty pfns again and
> again. This patches implement clearing of the dirty tag in the radix tree
> so that we issue flush only when needed.
> 
> The difficulty with clearing the dirty tag is that we have to protect against
> a concurrent page fault setting the dirty tag and writing new data into the
> page. So we need a lock serializing page fault and clearing of the dirty tag
> and write-protecting PTEs (so that we get another pagefault when pfn is written
> to again and we have to set the dirty tag again).
> 
> The effect of the patch set is easily visible:
> 
> Writing 1 GB of data via mmap, then fsync twice.
> 
> Before this patch set both fsyncs take ~205 ms on my test machine, after the
> patch set the first fsync takes ~283 ms (the additional cost of walking PTEs,
> clearing dirty bits etc. is very noticeable), the second fsync takes below
> 1 us.
> 
> As a bonus, these patches make filesystem freezing for DAX filesystems
> reliable because mappings are now properly writeprotected while freezing the
> fs.
> 
> Patches have passed xfstests for both xfs and ext4.
> 
> So far the patches don't work with PMD pages - that's next on my todo list.

Regarding the PMD work, I had a go at this a while ago.  You may (or may not)
find these patches useful:

mm: add follow_pte_pmd()
	https://patchwork.kernel.org/patch/7616241/
mm: add pmd_mkclean()
	https://patchwork.kernel.org/patch/7616261/
mm: add pgoff_mkclean()
	https://patchwork.kernel.org/patch/7616221/
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm

WARNING: multiple messages have this Message-ID (diff)
From: Ross Zwisler <ross.zwisler@linux.intel.com>
To: Jan Kara <jack@suse.cz>
Cc: linux-nvdimm@lists.01.org, linux-mm@kvack.org,
	linux-fsdevel@vger.kernel.org,
	Dan Williams <dan.j.williams@intel.com>,
	Ross Zwisler <ross.zwisler@linux.intel.com>
Subject: Re: [PATCH 0/3 v1] dax: Clear dirty bits after flushing caches
Date: Tue, 28 Jun 2016 15:41:59 -0600	[thread overview]
Message-ID: <20160628214159.GB15457@linux.intel.com> (raw)
In-Reply-To: <1466523915-14644-1-git-send-email-jack@suse.cz>

On Tue, Jun 21, 2016 at 05:45:12PM +0200, Jan Kara wrote:
> Hello,
> 
> currently we never clear dirty bits in the radix tree of a DAX inode. Thus
> fsync(2) or even periodical writeback flush all the dirty pfns again and
> again. This patches implement clearing of the dirty tag in the radix tree
> so that we issue flush only when needed.
> 
> The difficulty with clearing the dirty tag is that we have to protect against
> a concurrent page fault setting the dirty tag and writing new data into the
> page. So we need a lock serializing page fault and clearing of the dirty tag
> and write-protecting PTEs (so that we get another pagefault when pfn is written
> to again and we have to set the dirty tag again).
> 
> The effect of the patch set is easily visible:
> 
> Writing 1 GB of data via mmap, then fsync twice.
> 
> Before this patch set both fsyncs take ~205 ms on my test machine, after the
> patch set the first fsync takes ~283 ms (the additional cost of walking PTEs,
> clearing dirty bits etc. is very noticeable), the second fsync takes below
> 1 us.
> 
> As a bonus, these patches make filesystem freezing for DAX filesystems
> reliable because mappings are now properly writeprotected while freezing the
> fs.
> 
> Patches have passed xfstests for both xfs and ext4.
> 
> So far the patches don't work with PMD pages - that's next on my todo list.

Regarding the PMD work, I had a go at this a while ago.  You may (or may not)
find these patches useful:

mm: add follow_pte_pmd()
	https://patchwork.kernel.org/patch/7616241/
mm: add pmd_mkclean()
	https://patchwork.kernel.org/patch/7616261/
mm: add pgoff_mkclean()
	https://patchwork.kernel.org/patch/7616221/

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  parent reply	other threads:[~2016-06-28 21:42 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-06-21 15:45 [PATCH 0/3 v1] dax: Clear dirty bits after flushing caches Jan Kara
2016-06-21 15:45 ` Jan Kara
2016-06-21 15:45 ` [PATCH 1/3] dax: Make cache flushing protected by entry lock Jan Kara
2016-06-21 15:45   ` Jan Kara
2016-06-24 21:44   ` Ross Zwisler
2016-06-24 21:44     ` Ross Zwisler
2016-06-29 20:28     ` Jan Kara
2016-06-29 20:28       ` Jan Kara
2016-06-21 15:45 ` [PATCH 2/3] mm: Export follow_pte() Jan Kara
2016-06-21 15:45   ` Jan Kara
2016-06-24 21:55   ` Ross Zwisler
2016-06-24 21:55     ` Ross Zwisler
2016-06-29 20:29     ` Jan Kara
2016-06-29 20:29       ` Jan Kara
2016-06-21 15:45 ` [PATCH 3/3] dax: Clear dirty entry tags on cache flush Jan Kara
2016-06-21 15:45   ` Jan Kara
2016-06-21 17:31   ` kbuild test robot
2016-06-21 17:31     ` kbuild test robot
2016-06-21 17:31     ` kbuild test robot
2016-06-21 20:59   ` kbuild test robot
2016-06-21 20:59     ` kbuild test robot
2016-06-21 20:59     ` kbuild test robot
2016-06-23 10:47   ` Jan Kara
2016-06-23 10:47     ` Jan Kara
2016-06-23 10:47     ` Jan Kara
2016-06-28 21:38   ` Ross Zwisler
2016-06-28 21:38     ` Ross Zwisler
2016-06-29 20:47     ` Jan Kara
2016-06-29 20:47       ` Jan Kara
2016-06-28 21:41 ` Ross Zwisler [this message]
2016-06-28 21:41   ` [PATCH 0/3 v1] dax: Clear dirty bits after flushing caches Ross Zwisler

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=20160628214159.GB15457@linux.intel.com \
    --to=ross.zwisler@linux.intel.com \
    --cc=jack@suse.cz \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=linux-nvdimm@lists.01.org \
    /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 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.