summaryrefslogtreecommitdiffstats
path: root/fs/ceph/inode.c
diff options
context:
space:
mode:
authorHenry C Chang <henry.cy.chang@gmail.com>2011-05-11 10:29:54 +0000
committerSage Weil <sage@newdream.net>2011-05-11 10:44:48 -0700
commitd3d0720d4a7a46e93e055e5b0f1a8bd612743ed6 (patch)
tree39d657139336012f05d7573116a0ba2405e85de7 /fs/ceph/inode.c
parenta26a185d27b49e1656b335ef8ad1a32f7a0e7d7f (diff)
downloadop-kernel-dev-d3d0720d4a7a46e93e055e5b0f1a8bd612743ed6.zip
op-kernel-dev-d3d0720d4a7a46e93e055e5b0f1a8bd612743ed6.tar.gz
ceph: do not use i_wrbuffer_ref as refcount for Fb cap
We increments i_wrbuffer_ref when taking the Fb cap. This breaks the dirty page accounting and causes looping in __ceph_do_pending_vmtruncate, and ceph client hangs. This bug can be reproduced occasionally by running blogbench. Add a new field i_wb_ref to inode and dedicate it to Fb reference counting. Signed-off-by: Henry C Chang <henry.cy.chang@gmail.com> Signed-off-by: Sage Weil <sage@newdream.net>
Diffstat (limited to 'fs/ceph/inode.c')
-rw-r--r--fs/ceph/inode.c1
1 files changed, 1 insertions, 0 deletions
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index 03d6daf..70b6a48 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -355,6 +355,7 @@ struct inode *ceph_alloc_inode(struct super_block *sb)
ci->i_rd_ref = 0;
ci->i_rdcache_ref = 0;
ci->i_wr_ref = 0;
+ ci->i_wb_ref = 0;
ci->i_wrbuffer_ref = 0;
ci->i_wrbuffer_ref_head = 0;
ci->i_shared_gen = 0;
OpenPOWER on IntegriCloud