All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jan Kara <jack@suse.cz>
To: Dave Chinner <david@fromorbit.com>
Cc: xfs@oss.sgi.com, linux-fsdevel@vger.kernel.org,
	willy@linux.intel.com, jack@suse.cz
Subject: Re: [PATCH 1/8] xfs: mmap lock needs to be inside freeze protection
Date: Wed, 1 Apr 2015 16:34:23 +0200	[thread overview]
Message-ID: <20150401143423.GO26339@quack.suse.cz> (raw)
In-Reply-To: <1427194266-2885-2-git-send-email-david@fromorbit.com>

On Tue 24-03-15 21:50:59, Dave Chinner wrote:
> From: Dave Chinner <dchinner@redhat.com>
> 
> Lock ordering for the new mmap lock needs to be:
> 
> mmap_sem
>   sb_start_pagefault
>     i_mmap_lock
>       page lock
>         <fault processsing>
> 
> Right now xfs_vm_page_mkwrite gets this the wrong way around,
> While technically it cannot deadlock due to the current freeze
> ordering, it's still a landmine that might explode if we change
> anything in future. Hence we need to nest the locks correctly.
  Looks good to me. You can add:
Reviewed-by: Jan Kara <jack@suse.cz>

								Honza

> 
> Signed-off-by: Dave Chinner <dchinner@redhat.com>
> ---
>  fs/xfs/xfs_file.c | 11 ++++++++---
>  1 file changed, 8 insertions(+), 3 deletions(-)
> 
> diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c
> index dc5f609..a4c882e 100644
> --- a/fs/xfs/xfs_file.c
> +++ b/fs/xfs/xfs_file.c
> @@ -1449,15 +1449,20 @@ xfs_filemap_page_mkwrite(
>  	struct vm_fault		*vmf)
>  {
>  	struct xfs_inode	*ip = XFS_I(vma->vm_file->f_mapping->host);
> -	int			error;
> +	int			ret;
>  
>  	trace_xfs_filemap_page_mkwrite(ip);
>  
> +	sb_start_pagefault(VFS_I(ip)->i_sb);
> +	file_update_time(vma->vm_file);
>  	xfs_ilock(ip, XFS_MMAPLOCK_SHARED);
> -	error = block_page_mkwrite(vma, vmf, xfs_get_blocks);
> +
> +	ret = __block_page_mkwrite(vma, vmf, xfs_get_blocks);
> +
>  	xfs_iunlock(ip, XFS_MMAPLOCK_SHARED);
> +	sb_end_pagefault(VFS_I(ip)->i_sb);
>  
> -	return error;
> +	return block_page_mkwrite_return(ret);
>  }
>  
>  const struct file_operations xfs_file_operations = {
> -- 
> 2.0.0
> 
-- 
Jan Kara <jack@suse.cz>
SUSE Labs, CR

WARNING: multiple messages have this Message-ID (diff)
From: Jan Kara <jack@suse.cz>
To: Dave Chinner <david@fromorbit.com>
Cc: linux-fsdevel@vger.kernel.org, willy@linux.intel.com,
	jack@suse.cz, xfs@oss.sgi.com
Subject: Re: [PATCH 1/8] xfs: mmap lock needs to be inside freeze protection
Date: Wed, 1 Apr 2015 16:34:23 +0200	[thread overview]
Message-ID: <20150401143423.GO26339@quack.suse.cz> (raw)
In-Reply-To: <1427194266-2885-2-git-send-email-david@fromorbit.com>

On Tue 24-03-15 21:50:59, Dave Chinner wrote:
> From: Dave Chinner <dchinner@redhat.com>
> 
> Lock ordering for the new mmap lock needs to be:
> 
> mmap_sem
>   sb_start_pagefault
>     i_mmap_lock
>       page lock
>         <fault processsing>
> 
> Right now xfs_vm_page_mkwrite gets this the wrong way around,
> While technically it cannot deadlock due to the current freeze
> ordering, it's still a landmine that might explode if we change
> anything in future. Hence we need to nest the locks correctly.
  Looks good to me. You can add:
Reviewed-by: Jan Kara <jack@suse.cz>

								Honza

> 
> Signed-off-by: Dave Chinner <dchinner@redhat.com>
> ---
>  fs/xfs/xfs_file.c | 11 ++++++++---
>  1 file changed, 8 insertions(+), 3 deletions(-)
> 
> diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c
> index dc5f609..a4c882e 100644
> --- a/fs/xfs/xfs_file.c
> +++ b/fs/xfs/xfs_file.c
> @@ -1449,15 +1449,20 @@ xfs_filemap_page_mkwrite(
>  	struct vm_fault		*vmf)
>  {
>  	struct xfs_inode	*ip = XFS_I(vma->vm_file->f_mapping->host);
> -	int			error;
> +	int			ret;
>  
>  	trace_xfs_filemap_page_mkwrite(ip);
>  
> +	sb_start_pagefault(VFS_I(ip)->i_sb);
> +	file_update_time(vma->vm_file);
>  	xfs_ilock(ip, XFS_MMAPLOCK_SHARED);
> -	error = block_page_mkwrite(vma, vmf, xfs_get_blocks);
> +
> +	ret = __block_page_mkwrite(vma, vmf, xfs_get_blocks);
> +
>  	xfs_iunlock(ip, XFS_MMAPLOCK_SHARED);
> +	sb_end_pagefault(VFS_I(ip)->i_sb);
>  
> -	return error;
> +	return block_page_mkwrite_return(ret);
>  }
>  
>  const struct file_operations xfs_file_operations = {
> -- 
> 2.0.0
> 
-- 
Jan Kara <jack@suse.cz>
SUSE Labs, CR

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

  reply	other threads:[~2015-04-01 14:34 UTC|newest]

Thread overview: 58+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-03-24 10:50 [PATCH 0/8 v2] xfs: DAX support Dave Chinner
2015-03-24 10:50 ` Dave Chinner
2015-03-24 10:50 ` [PATCH 1/8] xfs: mmap lock needs to be inside freeze protection Dave Chinner
2015-04-01 14:34   ` Jan Kara [this message]
2015-04-01 14:34     ` Jan Kara
2015-04-06 17:48   ` Brian Foster
2015-04-06 17:48     ` Brian Foster
2015-03-24 10:51 ` [PATCH 2/8] dax: don't abuse get_block mapping for endio callbacks Dave Chinner
2015-03-24 10:51   ` Dave Chinner
2015-04-01 14:53   ` Jan Kara
2015-04-01 14:53     ` Jan Kara
2015-03-24 10:51 ` [PATCH 3/8] dax: expose __dax_fault for filesystems with locking constraints Dave Chinner
2015-03-24 10:51   ` Dave Chinner
2015-04-01 15:07   ` Jan Kara
2015-04-01 15:07     ` Jan Kara
2015-03-24 10:51 ` [PATCH 4/8] xfs: add DAX block zeroing support Dave Chinner
2015-03-24 10:51   ` Dave Chinner
2015-04-06 17:48   ` Brian Foster
2015-04-06 17:48     ` Brian Foster
2015-03-24 10:51 ` [PATCH 5/8] xfs: add DAX file operations support Dave Chinner
2015-03-24 10:51   ` Dave Chinner
2015-03-24 12:08   ` Boaz Harrosh
2015-03-24 12:08     ` Boaz Harrosh
2015-03-24 12:24     ` Boaz Harrosh
2015-03-24 12:24       ` Boaz Harrosh
2015-03-24 21:17     ` Dave Chinner
2015-03-24 21:17       ` Dave Chinner
2015-03-25  8:47       ` Boaz Harrosh
2015-03-25  8:47         ` Boaz Harrosh
2015-04-06 17:49   ` Brian Foster
2015-04-06 17:49     ` Brian Foster
2015-04-16  8:29     ` Dave Chinner
2015-04-16  8:29       ` Dave Chinner
2015-04-16  9:33   ` Boaz Harrosh
2015-04-16  9:33     ` Boaz Harrosh
2015-04-16 11:47     ` Dave Chinner
2015-04-16 11:47       ` Dave Chinner
2015-03-24 10:51 ` [PATCH 6/8] xfs: add DAX truncate support Dave Chinner
2015-03-24 10:51   ` Dave Chinner
2015-04-06 17:49   ` Brian Foster
2015-04-06 17:49     ` Brian Foster
2015-03-24 10:51 ` [PATCH 7/8] xfs: add DAX IO path support Dave Chinner
2015-03-24 10:51   ` Dave Chinner
2015-04-06 17:49   ` Brian Foster
2015-04-06 17:49     ` Brian Foster
2015-04-16  8:54     ` Dave Chinner
2015-04-16  8:54       ` Dave Chinner
2015-03-24 10:51 ` [PATCH 8/8] xfs: add initial DAX support Dave Chinner
2015-03-24 10:51   ` Dave Chinner
2015-03-24 12:52   ` Boaz Harrosh
2015-03-24 12:52     ` Boaz Harrosh
2015-03-24 21:25     ` Dave Chinner
2015-03-24 21:25       ` Dave Chinner
2015-03-25  9:14       ` Boaz Harrosh
2015-03-25  9:14         ` Boaz Harrosh
2015-04-06 19:00   ` Brian Foster
2015-04-06 19:00     ` Brian Foster
2015-05-28 23:45 [PATCH 0/8 v3] xfs: " Dave Chinner
2015-05-28 23:45 ` [PATCH 1/8] xfs: mmap lock needs to be inside freeze protection Dave Chinner

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=20150401143423.GO26339@quack.suse.cz \
    --to=jack@suse.cz \
    --cc=david@fromorbit.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=willy@linux.intel.com \
    --cc=xfs@oss.sgi.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: 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.