On Thu, Oct 24, 2013 at 02:25:10PM +1100, Dave Chinner wrote:
> From: Dave Chinner <dchinner@xxxxxxxxxx>
>
> Page cache allocation doesn't always go through ->begin_write and
> hence we don't always get the opportunity to set the allocation
> context to GFP_NOFS. Failing to do this means we open up the direct
> relcaim stack to recurse into the filesystem and consume a
> significant amount of stack.
>
> On RHEL6.4 kernels we are seeing ra_submit() and
> generic_file_splice_read() from an nfsd context recursing into the
> filesystem via the inode cache shrinker and evicting inodes. This is
> causing truncation to be run (e.g EOF block freeing) and causing
> bmap btree block merges and free space btree block splits to occur.
> These btree manipulations are occurring with the call chain already
> 30 functions deep and hence there is not enough stack space to
> complete such operations.
It seems like we really should fix this in the VFS as it could affect
all non-trivial filesystems.
> To avoid these specific overruns, we need to prevent the page cache
> allocation from recursing via direct reclaim. We can do that because
> the allocation functions take the allocation context from that which
> is stored in the mapping for the inode. We don't set that right now,
> so the default is GFP_HIGHUSER_MOVABLE, which is effectively a
> GFP_KERNEL context. We need it to be the equivalent of GFP_NOFS, so
> when we initialise an inode, set the mapping gfp mask appropriately.
>
> Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
> ---
> fs/xfs/xfs_iops.c | 9 +++++++++
> 1 file changed, 9 insertions(+)
>
> diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c
> index c4cd6d4..27e0e54 100644
> --- a/fs/xfs/xfs_iops.c
> +++ b/fs/xfs/xfs_iops.c
> @@ -1168,6 +1168,7 @@ xfs_setup_inode(
> struct xfs_inode *ip)
> {
> struct inode *inode = &ip->i_vnode;
> + gfp_t gfp_mask;
>
> inode->i_ino = ip->i_ino;
> inode->i_state = I_NEW;
> @@ -1230,6 +1231,14 @@ xfs_setup_inode(
> }
>
> /*
> + * Ensure all page cache allocations are done from GFP_NOFS context to
> + * prevent direct reclaim recursion back into the filesystem and blowing
> + * stacks or deadlocking.
> + */
> + gfp_mask = mapping_gfp_mask(inode->i_mapping);
> + mapping_set_gfp_mask(inode->i_mapping, (gfp_mask & ~(__GFP_FS)));
> +
> + /*
> * If there is no attribute fork no ACL can exist on this inode,
> * and it can't have any file capabilities attached to it either.
> */
> --
> 1.8.4.rc3
>
> _______________________________________________
> xfs mailing list
> xfs@xxxxxxxxxxx
> http://oss.sgi.com/mailman/listinfo/xfs
---end quoted text---
|