Commit 7b5747f4 authored May 31, 2018 by Andreas Gruenbacher Committed by Bob Peterson Jun 04, 2018

GFS2: Fix allocation error bug with recursive rgrp glocking



Before this patch function gfs2_write_begin, upon discovering an
error, called gfs2_trim_blocks while the rgrp glock was still held.
That's because gfs2_inplace_release is not called until later.
This patch reorganizes the logic a bit so gfs2_inplace_release
is called to release the lock prior to the call to gfs2_trim_blocks,
thus preventing the glock recursion.

Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Signed-off-by: Bob Peterson <rpeterso@redhat.com>

parent 07e23d68

fs/gfs2/aops.c

+8 −5

Original line number	Diff line number	Diff line
		@@ -747,18 +747,21 @@ out:
		put_page(page);

		gfs2_trans_end(sdp);
		if (alloc_required) {
		gfs2_inplace_release(ip);
		if (pos + len > ip->i_inode.i_size)
		gfs2_trim_blocks(&ip->i_inode);
		goto out_trans_fail;
		}
		goto out_qunlock;

		out_endtrans:
		gfs2_trans_end(sdp);
		out_trans_fail:
		if (alloc_required) {
		if (alloc_required)
		gfs2_inplace_release(ip);
		out_qunlock:
		if (alloc_required)
		gfs2_quota_unlock(ip);
		}
		out_unlock:
		if (&ip->i_inode == sdp->sd_rindex) {
		gfs2_glock_dq(&m_ip->i_gh);

Admin message