diff options
| author |   Trond Myklebust <Trond.Myklebust@netapp.com> | 2007-04-02 19:29:52 -0400 |
|---|---|---|
| committer | Trond Myklebust <Trond.Myklebust@netapp.com> | 2007-04-30 22:17:06 -0700 |
| commit | c63c7b051395368573779c8309aa5c990dcf2f96 (patch) | |
| tree | db54090eef99349d15b95fcd8c2620a2403d8db8 /include/linux/writeback.h | |
| parent | NFS: Cleanup for nfs_readpages() (diff) | |
| download | linux-dev-c63c7b051395368573779c8309aa5c990dcf2f96.tar.xz linux-dev-c63c7b051395368573779c8309aa5c990dcf2f96.zip | |
NFS: Fix a race when doing NFS write coalescing
Currently we do write coalescing in a very inefficient manner: one pass in
generic_writepages() in order to lock the pages for writing, then one pass
in nfs_flush_mapping() and/or nfs_sync_mapping_wait() in order to gather
the locked pages for coalescing into RPC requests of size "wsize".
In fact, it turns out there is actually a deadlock possible here since we
only start I/O on the second pass. If the user signals the process while
we're in nfs_sync_mapping_wait(), for instance, then we may exit before
starting I/O on all the requests that have been queued up.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Diffstat (limited to 'include/linux/writeback.h')
| -rw-r--r-- | include/linux/writeback.h | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/include/linux/writeback.h b/include/linux/writeback.h index 0c78f7f4a976..daa6c125f66e 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -59,6 +59,8 @@ struct writeback_control { unsigned for_reclaim:1; /* Invoked from the page allocator */ unsigned for_writepages:1; /* This is a writepages() call */ unsigned range_cyclic:1; /* range_start is cyclic */ + + void *fs_private; /* For use by ->writepages() */ }; /* |