Re: nfs MAP_SHARED corruption fix

[Trond Myklebust <trond.myklebust@fys.uio.no>]

>>>>> " " == Andrea Arcangeli <andrea@suse.de> writes:

     > This fixes corruption with MAP_SHARED on top of nfs filesystem
     > in 2.4:
     > --- 2.4.5pre1aa2/fs/nfs/write.c.~1~ Tue May 1 19:35:29 2001
     > +++ 2.4.5pre1aa2/fs/nfs/write.c Tue May 8 02:04:15 2001
     > @@ -1533,6 +1533,7 @@
     >  	if (!inode && file)
     >  		inode = file->f_dentry->d_inode;
 
     > + filemap_fdatasync(inode->i_mapping);
     >  	do {
     >  		error = 0; if (wait)

Yech! Apart from the fact that this means you do a full fdatasync()
even when you are simply trying to flush out single pages,
nfs_sync_file() gets called all over the place including in areas
where we know we're already holding a page lock.


AFAICs this fix will clearly deadlock...

Could you instead detail exactly which corruption problem you are
trying to fix?

Cheers,
  Trond