summaryrefslogtreecommitdiffstats
path: root/tests/bugs
diff options
context:
space:
mode:
authorSoumya Koduri <skoduri@redhat.com>2016-04-13 12:50:49 +0530
committerJeff Darcy <jdarcy@redhat.com>2016-04-26 04:57:42 -0700
commitedeb67a5344389f88a6912460fb0cf2e76a32df3 (patch)
treebc7e20203b67de85ea2a7fd54d9cf47657e1097d /tests/bugs
parent8590c1cf3c27468177c425c920cab01f52b251e5 (diff)
inode: Always fetch first entry from the inode lists during inode_table_destroy
In inode_table_destroy, we iterate through lru and active lists to move the entries to purge list so that they can be destroyed during inode_table_prune. But if used "list_for_each_entry" or "list_for_each_entry_safe" to iterate, we could end up accessing the entries which may have got moved to different(purge) lists in the process and can result in either infinite loop or crash. The safe approach seems to fetch the first entry of the list in each iteration till it gets empty. Change-Id: I24a18881833bd9419c2d8e5e8807bc71ec396479 BUG: 1326627 Signed-off-by: Soumya Koduri <skoduri@redhat.com> Reviewed-on: http://review.gluster.org/13987 Smoke: Gluster Build System <jenkins@build.gluster.com> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Kaleb KEITHLEY <kkeithle@redhat.com> Reviewed-by: Niels de Vos <ndevos@redhat.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
Diffstat (limited to 'tests/bugs')
0 files changed, 0 insertions, 0 deletions