[PATCHSET] The GFS2 filesystem (for review)

From: Steven Whitehouse
Date: Thu Aug 31 2006 - 09:20:16 EST


Following on from this message are the 16 patches of GFS2 which we are
again posting for review. Since the last posting we have, I hope,
addressed all the issues raised as well as fixing a number of bugs. In
particular, we have now only one new exported symbol (see patch 16 of
the series).

It is now possible to mount the gfs2meta filesystem for a particular
gfs2 filesystem at the same time as the gfs2 filesystem is mounted. In
fact we enforce that rule at least for the time being due to
dependencies in mounting the two filesystems.

Christoph requested that we find a different way to deal with the
problem of the VFS checking the file size in the direct i/o read path
before handing control to GFS2 (in other words at a point when the size
may change). We have done this with a two line patch to filemap.c (also
in patch 16 of the series) so that its logic is now very similar to that
followed in the direct i/o write path. This has eliminated the need for
us to duplicate parts of the core VFS code inside GFS2.

As usual the git tree is at:


It is based off a fairly recent pull from Linus' tree and includes the
DLM which is not included in the patch set being posted here.


To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/