public inbox for lvm2-cvs@sourceware.org
help / color / mirror / Atom feed
From: mbroz@sourceware.org
To: lvm-devel@redhat.com, lvm2-cvs@sourceware.org
Subject: LVM2 ./WHATS_NEW daemons/clvmd/lvm-functions.c
Date: Wed, 09 Dec 2009 18:55:00 -0000	[thread overview]
Message-ID: <20091209185554.18558.qmail@sourceware.org> (raw)

CVSROOT:	/cvs/lvm2
Module name:	LVM2
Changes by:	mbroz@sourceware.org	2009-12-09 18:55:53

Modified files:
	.              : WHATS_NEW 
	daemons/clvmd  : lvm-functions.c 

Log message:
	Allow implicit lock conversion for pre/post callbacks.
	
	This is unnoticed regression from commit 31672ff60e405795cad70d6d7888ac011f5373ce
	
	The pre/post callback need to convert lock always, local node
	is going to modify metadata in this case, it it fails conversion,
	the call is ignored.
	
	Also it fixes bug when the lock is not yet held, we cannot set LKF_CONVERT
	in this case, it will fail because this lock do not exist.
	
	Note that the automatic conversion is still disabled in activate
	call, so the original fix (reactivation of exlusive LV) should
	be still in place.

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/LVM2/WHATS_NEW.diff?cvsroot=lvm2&r1=1.1345&r2=1.1346
http://sourceware.org/cgi-bin/cvsweb.cgi/LVM2/daemons/clvmd/lvm-functions.c.diff?cvsroot=lvm2&r1=1.74&r2=1.75

--- LVM2/WHATS_NEW	2009/12/09 18:45:12	1.1345
+++ LVM2/WHATS_NEW	2009/12/09 18:55:53	1.1346
@@ -1,5 +1,6 @@
 Version 2.02.57 -
 ====================================
+  Allow implicit lock conversion in pre/post callbacks on local node.
   Allow implicit convert to the same cluster lock mode.
   Replace magic masks in cluster locking code by defined masks.
   Remove log volume from metadata if initial deactivation fails.
--- LVM2/daemons/clvmd/lvm-functions.c	2009/12/09 18:45:13	1.74
+++ LVM2/daemons/clvmd/lvm-functions.c	2009/12/09 18:55:53	1.75
@@ -274,7 +274,7 @@
 			return -1;
 
 		lvi->lock_mode = mode;
-		status = sync_lock(resource, mode, flags, &lvi->lock_id);
+		status = sync_lock(resource, mode, flags & ~LKF_CONVERT, &lvi->lock_id);
 		saved_errno = errno;
 		if (status) {
 			free(lvi);
@@ -348,7 +348,11 @@
 		mode = LKM_EXMODE;
 	}
 
-	/* Try to get the lock if it's a clustered volume group */
+	/*
+	 * Try to get the lock if it's a clustered volume group.
+	 * Use lock conversion only if requested, to prevent implicit conversion
+	 * of exclusive lock to shared one during activation.
+	 */
 	if (lock_flags & LCK_CLUSTER_VG) {
 		status = hold_lock(resource, mode, LKF_NOQUEUE | (lock_flags & LCK_CONVERT?LKF_CONVERT:0));
 		if (status) {
@@ -546,12 +550,13 @@
 	/* Nearly all the stuff happens cluster-wide. Apart from SUSPEND. Here we get the
 	   lock out on this node (because we are the node modifying the metadata)
 	   before suspending cluster-wide.
+	   LKF_CONVERT is used always, local node is going to modify metadata
 	 */
 	if ((command & (LCK_SCOPE_MASK | LCK_TYPE_MASK)) == LCK_LV_SUSPEND) {
 		DEBUGLOG("pre_lock_lv: resource '%s', cmd = %s, flags = %s\n",
 			 resource, decode_locking_cmd(command), decode_flags(lock_flags));
 
-		if (hold_lock(resource, LKM_PWMODE, LKF_NOQUEUE| (lock_flags & LCK_CONVERT?LKF_CONVERT:0)))
+		if (hold_lock(resource, LKM_PWMODE, LKF_NOQUEUE | LKF_CONVERT))
 			return errno;
 	}
 	return 0;
@@ -583,7 +588,7 @@
 				return EIO;
 
 			if (lvi.exists) {
-				if (hold_lock(resource, LKM_CRMODE, lock_flags & LCK_CONVERT?LKF_CONVERT:0))
+				if (hold_lock(resource, LKM_CRMODE, LKF_CONVERT))
 					return errno;
 			} else {
 				if (hold_unlock(resource))


             reply	other threads:[~2009-12-09 18:55 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-12-09 18:55 mbroz [this message]
  -- strict thread matches above, loose matches on Subject: below --
2012-03-01  9:54 zkabelac
2012-02-23 22:50 zkabelac
2011-08-10 11:00 mbroz
2011-03-02 23:38 mbroz
2010-07-01 21:23 mbroz
2010-03-24 22:25 snitzer
2010-01-26  8:00 mbroz
2009-12-09 19:01 mbroz
2009-12-09 19:00 mbroz
2009-12-09 18:45 mbroz
2009-10-22 17:45 agk
2009-10-01 14:14 ccaulfield
2009-08-05 14:18 ccaulfield
2009-03-05 16:25 mbroz
2008-04-08 13:03 ccaulfield
2008-04-04  8:53 ccaulfield
2007-08-24  8:29 pcaulfield
2007-08-23 12:19 pcaulfield
2007-04-24 15:13 pcaulfield
2005-02-14  9:07 pcaulfield
2005-02-02 11:42 pcaulfield
2004-10-04  9:23 pcaulfield

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=20091209185554.18558.qmail@sourceware.org \
    --to=mbroz@sourceware.org \
    --cc=lvm-devel@redhat.com \
    --cc=lvm2-cvs@sourceware.org \
    /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 read-only IMAP folder(s) and NNTP newsgroup(s).