From: swhiteho@redhat.com <swhiteho@redhat.com>
To: cluster-devel.redhat.com
Subject: [Cluster-devel] [PATCH] [DLM] cancel in conversion deadlock [4/6]
Date: Mon, 9 Jul 2007 17:02:20 +0100 [thread overview]
Message-ID: <11839970182635-git-send-email-swhiteho@redhat.com> (raw)
In-Reply-To: <11839970162125-git-send-email-swhiteho@redhat.com>
From: David Teigland <teigland@redhat.com>
When conversion deadlock is detected, cancel the conversion and return
EDEADLK to the application. This is a new default behavior where before
the dlm would allow the deadlock to exist indefinately.
The DLM_LKF_NODLCKWT flag can now be used in a conversion to prevent the
dlm from performing conversion deadlock detection/cancelation on it.
The DLM_LKF_CONVDEADLK flag can continue to be used as before to tell the
dlm to demote the granted mode of the lock being converted if it gets into
a conversion deadlock.
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c
index ad3797a..3c4d570 100644
--- a/fs/dlm/lock.c
+++ b/fs/dlm/lock.c
@@ -1408,10 +1408,8 @@ static int queue_conflict(struct list_head *head, struct dlm_lkb *lkb)
* queue for one resource. The granted mode of each lock blocks the requested
* mode of the other lock."
*
- * Part 2: if the granted mode of lkb is preventing the first lkb in the
- * convert queue from being granted, then demote lkb (set grmode to NL).
- * This second form requires that we check for conv-deadlk even when
- * now == 0 in _can_be_granted().
+ * Part 2: if the granted mode of lkb is preventing an earlier lkb in the
+ * convert queue from being granted, then deadlk/demote lkb.
*
* Example:
* Granted Queue: empty
@@ -1420,41 +1418,52 @@ static int queue_conflict(struct list_head *head, struct dlm_lkb *lkb)
*
* The first lock can't be granted because of the granted mode of the second
* lock and the second lock can't be granted because it's not first in the
- * list. We demote the granted mode of the second lock (the lkb passed to this
- * function).
+ * list. We either cancel lkb's conversion (PR->EX) and return EDEADLK, or we
+ * demote the granted mode of lkb (from PR to NL) if it has the CONVDEADLK
+ * flag set and return DEMOTED in the lksb flags.
*
- * After the resolution, the "grant pending" function needs to go back and try
- * to grant locks on the convert queue again since the first lock can now be
- * granted.
+ * Originally, this function detected conv-deadlk in a more limited scope:
+ * - if !modes_compat(lkb1, lkb2) && !modes_compat(lkb2, lkb1), or
+ * - if lkb1 was the first entry in the queue (not just earlier), and was
+ * blocked by the granted mode of lkb2, and there was nothing on the
+ * granted queue preventing lkb1 from being granted immediately, i.e.
+ * lkb2 was the only thing preventing lkb1 from being granted.
+ *
+ * That second condition meant we'd only say there was conv-deadlk if
+ * resolving it (by demotion) would lead to the first lock on the convert
+ * queue being granted right away. It allowed conversion deadlocks to exist
+ * between locks on the convert queue while they couldn't be granted anyway.
+ *
+ * Now, we detect and take action on conversion deadlocks immediately when
+ * they're created, even if they may not be immediately consequential. If
+ * lkb1 exists anywhere in the convert queue and lkb2 comes in with a granted
+ * mode that would prevent lkb1's conversion from being granted, we do a
+ * deadlk/demote on lkb2 right away and don't let it onto the convert queue.
+ * I think this means that the lkb_is_ahead condition below should always
+ * be zero, i.e. there will never be conv-deadlk between two locks that are
+ * both already on the convert queue.
*/
-static int conversion_deadlock_detect(struct dlm_rsb *rsb, struct dlm_lkb *lkb)
+static int conversion_deadlock_detect(struct dlm_rsb *r, struct dlm_lkb *lkb2)
{
- struct dlm_lkb *this, *first = NULL, *self = NULL;
+ struct dlm_lkb *lkb1;
+ int lkb_is_ahead = 0;
- list_for_each_entry(this, &rsb->res_convertqueue, lkb_statequeue) {
- if (!first)
- first = this;
- if (this == lkb) {
- self = lkb;
+ list_for_each_entry(lkb1, &r->res_convertqueue, lkb_statequeue) {
+ if (lkb1 == lkb2) {
+ lkb_is_ahead = 1;
continue;
}
- if (!modes_compat(this, lkb) && !modes_compat(lkb, this))
- return 1;
- }
-
- /* if lkb is on the convert queue and is preventing the first
- from being granted, then there's deadlock and we demote lkb.
- multiple converting locks may need to do this before the first
- converting lock can be granted. */
-
- if (self && self != first) {
- if (!modes_compat(lkb, first) &&
- !queue_conflict(&rsb->res_grantqueue, first))
- return 1;
+ if (!lkb_is_ahead) {
+ if (!modes_compat(lkb2, lkb1))
+ return 1;
+ } else {
+ if (!modes_compat(lkb2, lkb1) &&
+ !modes_compat(lkb1, lkb2))
+ return 1;
+ }
}
-
return 0;
}
@@ -1583,42 +1592,57 @@ static int _can_be_granted(struct dlm_rsb *r, struct dlm_lkb *lkb, int now)
if (!now && !conv && list_empty(&r->res_convertqueue) &&
first_in_list(lkb, &r->res_waitqueue))
return 1;
-
out:
- /*
- * The following, enabled by CONVDEADLK, departs from VMS.
- */
-
- if (conv && (lkb->lkb_exflags & DLM_LKF_CONVDEADLK) &&
- conversion_deadlock_detect(r, lkb)) {
- lkb->lkb_grmode = DLM_LOCK_NL;
- lkb->lkb_sbflags |= DLM_SBF_DEMOTED;
- }
-
return 0;
}
-/*
- * The ALTPR and ALTCW flags aren't traditional lock manager flags, but are a
- * simple way to provide a big optimization to applications that can use them.
- */
-
-static int can_be_granted(struct dlm_rsb *r, struct dlm_lkb *lkb, int now)
+static int can_be_granted(struct dlm_rsb *r, struct dlm_lkb *lkb, int now,
+ int *err)
{
- uint32_t flags = lkb->lkb_exflags;
int rv;
int8_t alt = 0, rqmode = lkb->lkb_rqmode;
+ int8_t is_convert = (lkb->lkb_grmode != DLM_LOCK_IV);
+
+ if (err)
+ *err = 0;
rv = _can_be_granted(r, lkb, now);
if (rv)
goto out;
- if (lkb->lkb_sbflags & DLM_SBF_DEMOTED)
+ /*
+ * The CONVDEADLK flag is non-standard and tells the dlm to resolve
+ * conversion deadlocks by demoting grmode to NL, otherwise the dlm
+ * cancels one of the locks.
+ */
+
+ if (is_convert && can_be_queued(lkb) &&
+ conversion_deadlock_detect(r, lkb)) {
+ if (lkb->lkb_exflags & DLM_LKF_CONVDEADLK) {
+ lkb->lkb_grmode = DLM_LOCK_NL;
+ lkb->lkb_sbflags |= DLM_SBF_DEMOTED;
+ } else if (!(lkb->lkb_exflags & DLM_LKF_NODLCKWT)) {
+ if (err)
+ *err = -EDEADLK;
+ else {
+ log_print("can_be_granted deadlock %x now %d",
+ lkb->lkb_id, now);
+ dlm_dump_rsb(r);
+ }
+ }
goto out;
+ }
- if (rqmode != DLM_LOCK_PR && flags & DLM_LKF_ALTPR)
+ /*
+ * The ALTPR and ALTCW flags are non-standard and tell the dlm to try
+ * to grant a request in a mode other than the normal rqmode. It's a
+ * simple way to provide a big optimization to applications that can
+ * use them.
+ */
+
+ if (rqmode != DLM_LOCK_PR && (lkb->lkb_exflags & DLM_LKF_ALTPR))
alt = DLM_LOCK_PR;
- else if (rqmode != DLM_LOCK_CW && flags & DLM_LKF_ALTCW)
+ else if (rqmode != DLM_LOCK_CW && (lkb->lkb_exflags & DLM_LKF_ALTCW))
alt = DLM_LOCK_CW;
if (alt) {
@@ -1633,10 +1657,20 @@ static int can_be_granted(struct dlm_rsb *r, struct dlm_lkb *lkb, int now)
return rv;
}
+/* FIXME: I don't think that can_be_granted() can/will demote or find deadlock
+ for locks pending on the convert list. Once verified (watch for these
+ log_prints), we should be able to just call _can_be_granted() and not
+ bother with the demote/deadlk cases here (and there's no easy way to deal
+ with a deadlk here, we'd have to generate something like grant_lock with
+ the deadlk error.) */
+
+/* returns the highest requested mode of all blocked conversions */
+
static int grant_pending_convert(struct dlm_rsb *r, int high)
{
struct dlm_lkb *lkb, *s;
int hi, demoted, quit, grant_restart, demote_restart;
+ int deadlk;
quit = 0;
restart:
@@ -1646,14 +1680,29 @@ static int grant_pending_convert(struct dlm_rsb *r, int high)
list_for_each_entry_safe(lkb, s, &r->res_convertqueue, lkb_statequeue) {
demoted = is_demoted(lkb);
- if (can_be_granted(r, lkb, 0)) {
+ deadlk = 0;
+
+ if (can_be_granted(r, lkb, 0, &deadlk)) {
grant_lock_pending(r, lkb);
grant_restart = 1;
- } else {
- hi = max_t(int, lkb->lkb_rqmode, hi);
- if (!demoted && is_demoted(lkb))
- demote_restart = 1;
+ continue;
}
+
+ if (!demoted && is_demoted(lkb)) {
+ log_print("WARN: pending demoted %x node %d %s",
+ lkb->lkb_id, lkb->lkb_nodeid, r->res_name);
+ demote_restart = 1;
+ continue;
+ }
+
+ if (deadlk) {
+ log_print("WARN: pending deadlock %x node %d %s",
+ lkb->lkb_id, lkb->lkb_nodeid, r->res_name);
+ dlm_dump_rsb(r);
+ continue;
+ }
+
+ hi = max_t(int, lkb->lkb_rqmode, hi);
}
if (grant_restart)
@@ -1671,7 +1720,7 @@ static int grant_pending_wait(struct dlm_rsb *r, int high)
struct dlm_lkb *lkb, *s;
list_for_each_entry_safe(lkb, s, &r->res_waitqueue, lkb_statequeue) {
- if (can_be_granted(r, lkb, 0))
+ if (can_be_granted(r, lkb, 0, NULL))
grant_lock_pending(r, lkb);
else
high = max_t(int, lkb->lkb_rqmode, high);
@@ -2121,7 +2170,7 @@ static int do_request(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
int error = 0;
- if (can_be_granted(r, lkb, 1)) {
+ if (can_be_granted(r, lkb, 1, NULL)) {
grant_lock(r, lkb);
queue_cast(r, lkb, 0);
goto out;
@@ -2147,16 +2196,32 @@ static int do_request(struct dlm_rsb *r, struct dlm_lkb *lkb)
static int do_convert(struct dlm_rsb *r, struct dlm_lkb *lkb)
{
int error = 0;
+ int deadlk = 0;
/* changing an existing lock may allow others to be granted */
- if (can_be_granted(r, lkb, 1)) {
+ if (can_be_granted(r, lkb, 1, &deadlk)) {
grant_lock(r, lkb);
queue_cast(r, lkb, 0);
grant_pending_locks(r);
goto out;
}
+ /* can_be_granted() detected that this lock would block in a conversion
+ deadlock, so we leave it on the granted queue and return EDEADLK in
+ the ast for the convert. */
+
+ if (deadlk) {
+ /* it's left on the granted queue */
+ log_debug(r->res_ls, "deadlock %x node %d sts%d g%d r%d %s",
+ lkb->lkb_id, lkb->lkb_nodeid, lkb->lkb_status,
+ lkb->lkb_grmode, lkb->lkb_rqmode, r->res_name);
+ revert_lock(r, lkb);
+ queue_cast(r, lkb, -EDEADLK);
+ error = -EDEADLK;
+ goto out;
+ }
+
/* is_demoted() means the can_be_granted() above set the grmode
to NL, and left us on the granted queue. This auto-demotion
(due to CONVDEADLK) might mean other locks, and/or this lock, are
@@ -2438,7 +2503,7 @@ int dlm_lock(dlm_lockspace_t *lockspace,
out_put:
if (convert || error)
__put_lkb(ls, lkb);
- if (error == -EAGAIN)
+ if (error == -EAGAIN || error == -EDEADLK)
error = 0;
out:
dlm_unlock_recovery(ls);
@@ -3312,6 +3377,12 @@ static void __receive_convert_reply(struct dlm_rsb *r, struct dlm_lkb *lkb,
queue_cast(r, lkb, -EAGAIN);
break;
+ case -EDEADLK:
+ receive_flags_reply(lkb, ms);
+ revert_lock_pc(r, lkb);
+ queue_cast(r, lkb, -EDEADLK);
+ break;
+
case -EINPROGRESS:
/* convert was queued on remote master */
receive_flags_reply(lkb, ms);
@@ -4284,7 +4355,7 @@ int dlm_user_convert(struct dlm_ls *ls, struct dlm_user_args *ua_tmp,
error = convert_lock(ls, lkb, &args);
- if (error == -EINPROGRESS || error == -EAGAIN)
+ if (error == -EINPROGRESS || error == -EAGAIN || error == -EDEADLK)
error = 0;
out_put:
dlm_put_lkb(lkb);
diff --git a/include/linux/dlm.h b/include/linux/dlm.h
index 975f17d..5227a95 100644
--- a/include/linux/dlm.h
+++ b/include/linux/dlm.h
@@ -85,7 +85,11 @@
* Only relevant to locks originating in userspace. A persistent lock will not
* be removed if the process holding the lock exits.
*
- * DLM_LKF_NODLKWT
+ * DLM_LKF_NODLCKWT
+ *
+ * Do not cancel the lock if it gets into conversion deadlock.
+ * Exclude this lock from being monitored due to DLM_LSFL_TIMEWARN.
+ *
* DLM_LKF_NODLCKBLK
*
* net yet implemented
--
1.5.1.2
next prev parent reply other threads:[~2007-07-09 16:02 UTC|newest]
Thread overview: 62+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-07-09 16:02 [Cluster-devel] [GFS2/DLM] Pre-pull Patch Posting swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] flush the glock completely in inode_go_sync swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [DLM] fix a couple of races swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] kernel changes to support new gfs2_grow command swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Kernel changes to support new gfs2_grow command (part 2) swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] use zero_user_page swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Addendum patch 2 for gfs2_grow swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Reduce size of struct gdlm_lock swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Clean up inode number handling swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Quotas non-functional - fix bug swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [DLM] keep dlm from panicing when traversing rsb list in debugfs swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [DLM] block scand during recovery [1/6] swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [DLM] add lock timeouts and warnings [2/6] swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [DLM] dlm_device interface changes [3/6] swhiteho
2007-07-09 16:02 ` swhiteho [this message]
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [DLM] fix new_lockspace error exit [5/6] swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [DLM] wait for config check during join [6/6] swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [DLM] fix compile breakage swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] latest gfs2-nmw headers break userland build swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [DLM] Compile fix swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [DLM] timeout fixes swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [DLM] canceling deadlocked lock swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [DLM] dumping master locks swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [DLM] show default protocol swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Quotas non-functional - fix another bug swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Make the log reserved blocks depend on block size swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [DLM] fix socket shutdown swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] fix jdata issues swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Fix sign problem in quota/statfs and cleanup _host structures swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Add nanosecond timestamp feature swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [DLM] fix reference counting swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [DLM] variable allocation swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Fix typo in rename of directories swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Fix bug in error path of inode swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Can't mount GFS2 file system on AoE device swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Recovery for lost unlinked inodes swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] gfs2_lookupi() uninitialised var fix swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] set plock owner in GETLK info swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] return conflicts for GETLK swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Fix deallocation issues swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [DLM] don't require FS flag on all nodes swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Journaled file write/unstuff bug swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Remove bogus '\0' in rgrp.c swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Use zero_user_page() in stuffed_readpage() swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] assertion failure after writing to journaled file, umount swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Simplify multiple glock aquisition swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Addendum to the journaled file/unmount patch swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Fix gfs2_block_truncate_page err return swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [DLM] Telnet to port 21064 can stop all lockspaces swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] inode size inconsistency swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] remounting w/o acl option leaves acls enabled swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] System won't suspend with GFS2 file system mounted swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] git-gfs2-nmw-build-fix swhiteho
2007-07-09 16:02 ` [Cluster-devel] [PATCH] [GFS2] Obtaining no_formal_ino from directory entry swhiteho
2007-07-09 16:03 ` [Cluster-devel] [PATCH] [GFS2] Remove i_mode passing from NFS File Handle swhiteho
2007-07-09 16:03 ` [Cluster-devel] [PATCH] [DLM] dump more lock values swhiteho
2007-07-09 16:03 ` [Cluster-devel] [PATCH] [GFS2] Small fixes to logging code swhiteho
2007-07-10 9:06 ` [Cluster-devel] Re: [PATCH] [GFS2] Remove i_mode passing from NFS File Handle Christoph Hellwig
2007-07-10 10:01 ` Steven Whitehouse
2007-07-10 11:36 ` Christoph Hellwig
2007-07-10 11:47 ` Steven Whitehouse
2007-07-10 7:50 ` [Cluster-devel] [GFS2/DLM] Pull request Steven Whitehouse
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=11839970182635-git-send-email-swhiteho@redhat.com \
--to=swhiteho@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).