Merge lp:~serge-hallyn/ubuntu/natty/multipath-tools/free_waiter-race into lp:ubuntu/natty/multipath-tools

[Serge Hallyn]

=== modified file 'debian/changelog'
--- debian/changelog	2011-02-09 12:45:34 +0000
+++ debian/changelog	2011-02-09 21:53:14 +0000
@@ -1,3 +1,10 @@
+multipath-tools (0.4.8-14ubuntu9) natty; urgency=low
+
+  * Fix segv caused by race condition with free_waiter threads during
+    shutdown, using backport from upstream.  (LP: #713237)
+
+ -- dann frazier <[e-mail]>  Thu, 03 Feb 2011 16:37:10 -0700
+
 multipath-tools (0.4.8-14ubuntu8) natty; urgency=low
 
   * Fix segv on shutdown when log buffer is empty, using patch cherry-picked

=== added file 'debian/patches/1004--race-condition-fix-with-free_waiter-threads-during-shutdown.patch'
--- debian/patches/1004--race-condition-fix-with-free_waiter-threads-during-shutdown.patch	1970-01-01 00:00:00 +0000
+++ debian/patches/1004--race-condition-fix-with-free_waiter-threads-during-shutdown.patch	2011-02-09 21:53:14 +0000
@@ -0,0 +1,219 @@
+Description: Race-condition fix with free_waiter threads during shutdown.
+Author: Konrad Rzeszutek <[e-mail]>
+Origin: upstream, http://www.spinics.net/lists/dm-devel/msg08911.html, commit a403f57b991f3be7c7ea6250d95fad1554c9d6ca
+Bug-Ubuntu: https://bugs.launchpad.net/ubuntu/+source/multipath-tools/+bug/713237
+
+Index: fix-segv/libmultipath/lock.c
+===================================================================
+--- fix-segv.orig/libmultipath/lock.c	2011-02-03 16:36:19.803282771 -0700
++++ fix-segv/libmultipath/lock.c	2011-02-03 16:58:16.483280080 -0700
+@@ -1,8 +1,8 @@
+ #include <pthread.h>
+ #include "lock.h"
+-
++#include <stdio.h>
+ void cleanup_lock (void * data)
+ {
+-	unlock((pthread_mutex_t *)data);
++	unlock ((*(struct mutex_lock *)data));
+ }
+ 
+Index: fix-segv/libmultipath/lock.h
+===================================================================
+--- fix-segv.orig/libmultipath/lock.h	2011-02-03 16:36:19.791279834 -0700
++++ fix-segv/libmultipath/lock.h	2011-02-03 16:58:38.403579621 -0700
+@@ -1,19 +1,30 @@
+ #ifndef _LOCK_H
+ #define _LOCK_H
+ 
++#include <pthread.h>
++
++/*
++ * Wrapper for the mutex. Includes a ref-count to keep
++ * track of how many there are out-standing threads blocking
++ * on a mutex. */
++struct mutex_lock {
++	pthread_mutex_t *mutex;
++	int depth;
++};
++
+ #ifdef LCKDBG
+ #define lock(a) \
+-	        fprintf(stderr, "%s:%s(%i) lock %p\n", __FILE__, __FUNCTION__, __LINE__, a); \
+-        pthread_mutex_lock(a)
++		fprintf(stderr, "%s:%s(%i) lock %p depth: %d (%ld)\n", __FILE__, __FUNCTION__, __LINE__, a.mutex, a.depth, pthread_self()); \
++		a.depth++; pthread_mutex_lock(a.mutex)
+ #define unlock(a) \
+-	        fprintf(stderr, "%s:%s(%i) unlock %p\n", __FILE__, __FUNCTION__, __LINE__, a); \
+-        pthread_mutex_unlock(a)
++		fprintf(stderr, "%s:%s(%i) unlock %p depth: %d (%ld)\n", __FILE__, __FUNCTION__, __LINE__, a.mutex, a.depth, pthread_self()); \
++	a.depth--; pthread_mutex_unlock(a.mutex)
+ #define lock_cleanup_pop(a) \
+-	        fprintf(stderr, "%s:%s(%i) unlock %p\n", __FILE__, __FUNCTION__, __LINE__, a); \
+-        pthread_cleanup_pop(1);
++		fprintf(stderr, "%s:%s(%i) unlock %p depth: %d (%ld)\n", __FILE__, __FUNCTION__, __LINE__, a.mutex, a.depth, pthread_self()); \
++	pthread_cleanup_pop(1);
+ #else
+-#define lock(a) pthread_mutex_lock(a)
+-#define unlock(a) pthread_mutex_unlock(a)
++#define lock(a) a.depth++; pthread_mutex_lock(a.mutex)
++#define unlock(a) a.depth--; pthread_mutex_unlock(a.mutex)
+ #define lock_cleanup_pop(a) pthread_cleanup_pop(1);
+ #endif
+ 
+Index: fix-segv/libmultipath/structs_vec.h
+===================================================================
+--- fix-segv.orig/libmultipath/structs_vec.h	2011-02-03 16:36:19.787281481 -0700
++++ fix-segv/libmultipath/structs_vec.h	2011-02-03 16:58:16.487280599 -0700
+@@ -1,9 +1,15 @@
+ #ifndef _STRUCTS_VEC_H
+ #define _STRUCTS_VEC_H
+ 
++#include "lock.h"
++/*
++struct mutex_lock {
++	pthread_mutex_t *mutex;
++	int depth;
++}; */
+ struct vectors {
+ #if DAEMON
+-	pthread_mutex_t *lock;
++	struct mutex_lock lock; /* defined in lock.h */
+ #endif
+ 	vector pathvec;
+ 	vector mpvec;
+Index: fix-segv/libmultipath/waiter.c
+===================================================================
+--- fix-segv.orig/libmultipath/waiter.c	2011-02-03 16:36:19.795279169 -0700
++++ fix-segv/libmultipath/waiter.c	2011-02-03 16:58:16.487280599 -0700
+@@ -45,7 +45,10 @@
+ 		 */
+ 		wp->mpp->waiter = NULL;
+ 	else
+-		condlog(3, "free_waiter, mpp freed before wp=%p,", wp);
++		/*
++		* This is OK condition during shutdown.
++		*/
++		condlog(3, "free_waiter, mpp freed before wp=%p (%s).", wp, wp->mapname);
+ 
+ 	unlock(wp->vecs->lock);
+ 
+@@ -58,13 +61,16 @@
+ void stop_waiter_thread (struct multipath *mpp, struct vectors *vecs)
+ {
+ 	struct event_thread *wp = (struct event_thread *)mpp->waiter;
++	pthread_t thread;
+ 	
+ 	if (!wp) {
+ 		condlog(3, "%s: no waiter thread", mpp->alias);
+ 		return;
+ 	}
+-	condlog(2, "%s: stop event checker thread", wp->mapname);
+-	pthread_kill((pthread_t)wp->thread, SIGUSR1);
++	thread = wp->thread;
++	condlog(2, "%s: stop event checker thread (%lu)", wp->mapname, thread);
++
++	pthread_kill(thread, SIGUSR1);
+ }
+ 
+ static sigset_t unblock_signals(void)
+@@ -148,7 +154,7 @@
+ 		 * 4) a path reinstate : nothing to do
+ 		 * 5) a switch group : nothing to do
+ 		 */
+-		pthread_cleanup_push(cleanup_lock, waiter->vecs->lock);
++		pthread_cleanup_push(cleanup_lock, &waiter->vecs->lock);
+ 		lock(waiter->vecs->lock);
+ 		r = update_multipath(waiter->vecs, waiter->mapname);
+ 		lock_cleanup_pop(waiter->vecs->lock);
+Index: fix-segv/multipath/main.c
+===================================================================
+--- fix-segv.orig/multipath/main.c	2011-02-03 16:36:19.819283822 -0700
++++ fix-segv/multipath/main.c	2011-02-03 16:58:16.495278814 -0700
+@@ -419,6 +419,7 @@
+ 		condlog(3, "restart multipath configuration process");
+ 	
+ out:
++
+ 	sysfs_cleanup();
+ 	free_config(conf);
+ 	dm_lib_release();
+Index: fix-segv/multipathd/main.c
+===================================================================
+--- fix-segv.orig/multipathd/main.c	2011-02-03 16:58:16.463280045 -0700
++++ fix-segv/multipathd/main.c	2011-02-03 16:58:16.499281808 -0700
+@@ -591,7 +591,7 @@
+ 	*len = 0;
+ 	vecs = (struct vectors *)trigger_data;
+ 
+-	pthread_cleanup_push(cleanup_lock, vecs->lock);
++	pthread_cleanup_push(cleanup_lock, &vecs->lock);
+ 	lock(vecs->lock);
+ 
+ 	r = parse_cmd(str, reply, len, vecs);
+@@ -744,9 +744,9 @@
+ 	condlog(3, "unlink pidfile");
+ 	unlink(DEFAULT_PIDFILE);
+ 
+-	lock(&exit_mutex);
++	pthread_mutex_lock(&exit_mutex);
+ 	pthread_cond_signal(&exit_cond);
+-	unlock(&exit_mutex);
++	pthread_mutex_unlock(&exit_mutex);
+ 
+ 	return status;
+ }
+@@ -879,7 +879,7 @@
+ 	}
+ 
+ 	while (1) {
+-		pthread_cleanup_push(cleanup_lock, vecs->lock);
++		pthread_cleanup_push(cleanup_lock, &vecs->lock);
+ 		lock(vecs->lock);
+ 		condlog(4, "tick");
+ 
+@@ -1153,13 +1153,14 @@
+ 	if (!vecs)
+ 		return NULL;
+ 
+-	vecs->lock =
++	vecs->lock.mutex =
+ 		(pthread_mutex_t *)MALLOC(sizeof(pthread_mutex_t));
+ 
+-	if (!vecs->lock)
++	if (!vecs->lock.mutex)
+ 		goto out;
+ 
+-	pthread_mutex_init(vecs->lock, NULL);
++	pthread_mutex_init(vecs->lock.mutex, NULL);
++	vecs->lock.depth = 0;
+ 
+ 	return vecs;
+ 
+@@ -1307,7 +1308,6 @@
+ 		condlog(0, "failure during configuration");
+ 		exit(1);
+ 	}
+-
+ 	/*
+ 	 * start threads
+ 	 */
+@@ -1341,9 +1341,15 @@
+ 	free_polls();
+ 
+ 	unlock(vecs->lock);
+-	pthread_mutex_destroy(vecs->lock);
+-	FREE(vecs->lock);
+-	vecs->lock = NULL;
++	/* Now all the waitevent threads will start rushing in. */
++	while (vecs->lock.depth > 0) {
++		sleep (1); /* This is weak. */
++		condlog(3,"Have %d wait event checkers threads to de-alloc, waiting..\n", vecs->lock.depth);
++	}
++	pthread_mutex_destroy(vecs->lock.mutex);
++	FREE(vecs->lock.mutex);
++	vecs->lock.depth = 0;
++	vecs->lock.mutex = NULL;
+ 	FREE(vecs);
+ 	vecs = NULL;
+ 	free_config(conf);

=== modified file 'debian/patches/series'
--- debian/patches/series	2011-02-01 21:45:14 +0000
+++ debian/patches/series	2011-02-09 21:53:14 +0000
@@ -14,3 +14,4 @@
 1001--intel-mpath-prio-alua.patch
 1002--Fix-for-uevent-devpath-handling.patch
 1003--multipathd-crash-on-shutdown.patch
+1004--race-condition-fix-with-free_waiter-threads-during-shutdown.patch