xfs
[Top] [All Lists]

Re: [PATCH] xfs: inode lockdep annotations broke non-lockdep build

To: Dave Chinner <david@xxxxxxxxxxxxx>
Subject: Re: [PATCH] xfs: inode lockdep annotations broke non-lockdep build
From: Brian Foster <bfoster@xxxxxxxxxx>
Date: Thu, 20 Aug 2015 07:32:00 -0400
Cc: xfs@xxxxxxxxxxx
Delivered-to: xfs@xxxxxxxxxxx
In-reply-to: <1439979814-18934-1-git-send-email-david@xxxxxxxxxxxxx>
References: <1439979814-18934-1-git-send-email-david@xxxxxxxxxxxxx>
User-agent: Mutt/1.5.23 (2014-03-12)
On Wed, Aug 19, 2015 at 08:23:34PM +1000, Dave Chinner wrote:
> From: Dave Chinner <dchinner@xxxxxxxxxx>
> 
> Fix CONFIG_LOCKDEP=n build, because asserts I put in to ensure we
> aren't overrunning lockdep subclasses in commit 0952c81 ("xfs:
> clean up inode lockdep annotations") use a define that doesn't
> exist when CONFIG_LOCKDEP=n
> 
> Only check the subclass limits when lockdep is actually enabled.
> 
> Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
> ---
>  fs/xfs/xfs_inode.c | 16 ++++++++++++++--
>  1 file changed, 14 insertions(+), 2 deletions(-)
> 
> diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
> index dd584da..30555f8 100644
> --- a/fs/xfs/xfs_inode.c
> +++ b/fs/xfs/xfs_inode.c
> @@ -362,6 +362,17 @@ int xfs_lots_retries;
>  int xfs_lock_delays;
>  #endif
>  
> +#ifdef CONFIG_LOCKDEP
> +static bool
> +xfs_lockdep_subclass_ok(
> +     int subclass)
> +{
> +     return subclass < MAX_LOCKDEP_SUBCLASSES;
> +}
> +#else
> +#define xfs_lockdep_subclass_ok(subclass)    (true)
> +#endif
> +

FYI, there's a compile warning with debug and verbose warnings disabled:

...
  CC [M]  fs/xfs//xfs_super.o
fs/xfs//xfs_inode.c:367:1: warning: âxfs_lockdep_subclass_okâ defined but not 
used [-Wunused-function]
 xfs_lockdep_subclass_ok(
 ^
...

Perhaps it's best to just use the #define in both cases?

Brian

>  /*
>   * Bump the subclass so xfs_lock_inodes() acquires each lock with a different
>   * value. This can be called for any type of inode lock combination, 
> including
> @@ -375,11 +386,12 @@ xfs_lock_inumorder(int lock_mode, int subclass)
>  
>       ASSERT(!(lock_mode & (XFS_ILOCK_PARENT | XFS_ILOCK_RTBITMAP |
>                             XFS_ILOCK_RTSUM)));
> +     ASSERT(xfs_lockdep_subclass_ok(subclass));
>  
>       if (lock_mode & (XFS_IOLOCK_SHARED|XFS_IOLOCK_EXCL)) {
>               ASSERT(subclass <= XFS_IOLOCK_MAX_SUBCLASS);
> -             ASSERT(subclass + XFS_IOLOCK_PARENT_VAL <
> -                                             MAX_LOCKDEP_SUBCLASSES);
> +             ASSERT(xfs_lockdep_subclass_ok(subclass +
> +                                             XFS_IOLOCK_PARENT_VAL));
>               class += subclass << XFS_IOLOCK_SHIFT;
>               if (lock_mode & XFS_IOLOCK_PARENT)
>                       class += XFS_IOLOCK_PARENT_VAL << XFS_IOLOCK_SHIFT;
> -- 
> 2.5.0
> 
> _______________________________________________
> xfs mailing list
> xfs@xxxxxxxxxxx
> http://oss.sgi.com/mailman/listinfo/xfs

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