xfs
[Top] [All Lists]

Re: [PATCH 1/3] XFS: Print error when xfs_ialloc_ag_select fails to find

To: raghu.prabhu13@xxxxxxxxx
Subject: Re: [PATCH 1/3] XFS: Print error when xfs_ialloc_ag_select fails to find continuous free space.
From: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Wed, 6 Jun 2012 00:00:21 +1000
Cc: xfs@xxxxxxxxxxx, Raghavendra D Prabhu <rprabhu@xxxxxxxxxxx>
In-reply-to: <2dabcd19de5dafab66154800d4985dff58343dc0.1338721614.git.rprabhu@xxxxxxxxxxx>
References: <cover.1338721614.git.rprabhu@xxxxxxxxxxx> <2dabcd19de5dafab66154800d4985dff58343dc0.1338721614.git.rprabhu@xxxxxxxxxxx>
User-agent: Mutt/1.5.21 (2010-09-15)
On Sun, Jun 03, 2012 at 04:42:47PM +0530, raghu.prabhu13@xxxxxxxxx wrote:
> From: Raghavendra D Prabhu <rprabhu@xxxxxxxxxxx>
> 
> When xfs_ialloc_ag_select fails to find any AG with continuous free blocks for
> needed inode allocation, printk error about it once.
> 
> Signed-off-by: Raghavendra D Prabhu <rprabhu@xxxxxxxxxxx>
> ---
>  fs/xfs/xfs_ialloc.c |    5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)
> 
> diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/xfs_ialloc.c
> index 177a21a..a02180a 100644
> --- a/fs/xfs/xfs_ialloc.c
> +++ b/fs/xfs/xfs_ialloc.c
> @@ -543,8 +543,11 @@ nextag:
>               if (agno >= agcount)
>                       agno = 0;
>               if (agno == pagno) {
> -                     if (flags == 0)
> +                     if (flags == 0) {
> +                             pr_err_once("XFS (%s): Out of continuous free 
> blocks for inode allocation",
> +                                             mp->m_fsname);

Couple of things for all 3 patches. Firstly - 80 columns. We tend to keep the
pformat string on a single line so it is easy to grep for like so:

                                pr_err_once(mp,
                "Insufficient contiguous free space for inode allocation");

Secondly, I'm not sure "print once and once only" is the right
behaviour here.  This will only warn on one filesystem and only
once for the uptime of the system. So if you hit this 5 minutes
after boot, then clean up, you'll never get it emitted again even
when the condition reappears 6 months later. hence I think this
should be a rate-limited print, not a print once.

And finally, you should add a xfs_err_ratelimited wrapper, not not
use pr_err_ratelimited() directly and open code the identifier
prefixes.

i.e. something like:

#define xfs_printk_ratelimited(func, mp, fmt, ...)              \
({                                                              \
        static DEFINE_RATELIMIT_STATE(_rs,                      \
                                DEFAULT_RATELIMIT_INTERVAL,     \
                                DEFAULT_RATELIMIT_BURST);       \
                                                                \
        if (__ratelimit(&_rs))                                  \
                (func)((mp), fmt, ##__VA_ARGS__);               \
})

#define xfs_err_once(mp, fmt, ...)                      \
                xfs_printk_ratelimited(xfs_err, (mp), fmt, ##__VA_ARGS__); 

So we can easily extend it to other logging levels as needed.


Cheers,

Dave.
-- 
Dave Chinner
david@xxxxxxxxxxxxx

<Prev in Thread] Current Thread [Next in Thread>