summaryrefslogtreecommitdiffstats
path: root/fs/gfs2/sys.c
diff options
context:
space:
mode:
authorSteven Whitehouse <swhiteho@redhat.com>2009-05-19 10:01:18 +0100
committerSteven Whitehouse <swhiteho@redhat.com>2009-05-19 10:01:18 +0100
commitfe64d517df0970a68417184a12fcd4ba0589cc28 (patch)
treed977f214fdf6ba96254cfbf6683e8583ecebe504 /fs/gfs2/sys.c
parent9582d41135c0d362f04ed6bf3dc8d693a7eafee2 (diff)
GFS2: Umount recovery race fix
This patch fixes a race condition where we can receive recovery requests part way through processing a umount. This was causing problems since the recovery thread had already gone away. Looking in more detail at the recovery code, it was really trying to implement a slight variation on a work queue, and that happens to align nicely with the recently introduced slow-work subsystem. As a result I've updated the code to use slow-work, rather than its own home grown variety of work queue. When using the wait_on_bit() function, I noticed that the wait function that was supplied as an argument was appearing in the WCHAN field, so I've updated the function names in order to produce more meaningful output. Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Diffstat (limited to 'fs/gfs2/sys.c')
-rw-r--r--fs/gfs2/sys.c53
1 files changed, 26 insertions, 27 deletions
diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c
index 894bf773ec9..9f6d48b75fd 100644
--- a/fs/gfs2/sys.c
+++ b/fs/gfs2/sys.c
@@ -356,34 +356,33 @@ static ssize_t first_done_show(struct gfs2_sbd *sdp, char *buf)
return sprintf(buf, "%d\n", ls->ls_first_done);
}
-static ssize_t recover_show(struct gfs2_sbd *sdp, char *buf)
-{
- struct lm_lockstruct *ls = &sdp->sd_lockstruct;
- return sprintf(buf, "%d\n", ls->ls_recover_jid);
-}
-
-static void gfs2_jdesc_make_dirty(struct gfs2_sbd *sdp, unsigned int jid)
+static ssize_t recover_store(struct gfs2_sbd *sdp, const char *buf, size_t len)
{
+ unsigned jid;
struct gfs2_jdesc *jd;
+ int rv;
+
+ rv = sscanf(buf, "%u", &jid);
+ if (rv != 1)
+ return -EINVAL;
+ rv = -ESHUTDOWN;
spin_lock(&sdp->sd_jindex_spin);
+ if (test_bit(SDF_NORECOVERY, &sdp->sd_flags))
+ goto out;
+ rv = -EBUSY;
+ if (sdp->sd_jdesc->jd_jid == jid)
+ goto out;
+ rv = -ENOENT;
list_for_each_entry(jd, &sdp->sd_jindex_list, jd_list) {
if (jd->jd_jid != jid)
continue;
- jd->jd_dirty = 1;
+ rv = slow_work_enqueue(&jd->jd_work);
break;
}
+out:
spin_unlock(&sdp->sd_jindex_spin);
-}
-
-static ssize_t recover_store(struct gfs2_sbd *sdp, const char *buf, size_t len)
-{
- struct lm_lockstruct *ls = &sdp->sd_lockstruct;
- ls->ls_recover_jid = simple_strtol(buf, NULL, 0);
- gfs2_jdesc_make_dirty(sdp, ls->ls_recover_jid);
- if (sdp->sd_recoverd_process)
- wake_up_process(sdp->sd_recoverd_process);
- return len;
+ return rv ? rv : len;
}
static ssize_t recover_done_show(struct gfs2_sbd *sdp, char *buf)
@@ -401,15 +400,15 @@ static ssize_t recover_status_show(struct gfs2_sbd *sdp, char *buf)
#define GDLM_ATTR(_name,_mode,_show,_store) \
static struct gfs2_attr gdlm_attr_##_name = __ATTR(_name,_mode,_show,_store)
-GDLM_ATTR(proto_name, 0444, proto_name_show, NULL);
-GDLM_ATTR(block, 0644, block_show, block_store);
-GDLM_ATTR(withdraw, 0644, withdraw_show, withdraw_store);
-GDLM_ATTR(id, 0444, lkid_show, NULL);
-GDLM_ATTR(first, 0444, lkfirst_show, NULL);
-GDLM_ATTR(first_done, 0444, first_done_show, NULL);
-GDLM_ATTR(recover, 0644, recover_show, recover_store);
-GDLM_ATTR(recover_done, 0444, recover_done_show, NULL);
-GDLM_ATTR(recover_status, 0444, recover_status_show, NULL);
+GDLM_ATTR(proto_name, 0444, proto_name_show, NULL);
+GDLM_ATTR(block, 0644, block_show, block_store);
+GDLM_ATTR(withdraw, 0644, withdraw_show, withdraw_store);
+GDLM_ATTR(id, 0444, lkid_show, NULL);
+GDLM_ATTR(first, 0444, lkfirst_show, NULL);
+GDLM_ATTR(first_done, 0444, first_done_show, NULL);
+GDLM_ATTR(recover, 0200, NULL, recover_store);
+GDLM_ATTR(recover_done, 0444, recover_done_show, NULL);
+GDLM_ATTR(recover_status, 0444, recover_status_show, NULL);
static struct attribute *lock_module_attrs[] = {
&gdlm_attr_proto_name.attr,