From: Gautham R Shenoy <ego@in.ibm.com>
To: Linus Torvalds <torvalds@linux-foundation.org>,
Andrew Morton <akpm@linux-foundation.org>
Cc: linux-kernel@vger.kernel.org,
Srivatsa Vaddagiri <vatsa@in.ibm.com>,
Rusty Russel <rusty@rustcorp.com.au>,
Dipankar Sarma <dipankar@in.ibm.com>,
Oleg Nesterov <oleg@tv-sign.ru>, Ingo Molnar <mingo@elte.hu>,
Paul E McKenney <paulmck@us.ibm.com>
Subject: [RFC PATCH 4/4] Remove CPU_DEAD/CPU_UP_CANCELLED handling from workqueue.c
Date: Tue, 16 Oct 2007 16:07:21 +0530 [thread overview]
Message-ID: <20071016103721.GD16570@in.ibm.com> (raw)
In-Reply-To: <20071016103308.GA9907@in.ibm.com>
cleanup_workqueue_thread() in the CPU_DEAD and CPU_UP_CANCELLED path
will cause a deadlock if the worker thread is executing a work item
which is blocked on get_online_cpus(). This will lead to a irrecoverable
hang.
Solution is not to cleanup the worker thread. Instead let it remain
even after the cpu goes offline. Since no one can queue any work
on an offlined cpu, this thread will be forever sleeping, untill
someone onlines the cpu.
Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
---
kernel/workqueue.c | 13 ++++++-------
1 file changed, 6 insertions(+), 7 deletions(-)
Index: linux-2.6.23/kernel/workqueue.c
===================================================================
--- linux-2.6.23.orig/kernel/workqueue.c
+++ linux-2.6.23/kernel/workqueue.c
@@ -30,6 +30,7 @@
#include <linux/hardirq.h>
#include <linux/mempolicy.h>
#include <linux/freezer.h>
+#include <linux/cpumask.h>
#include <linux/kallsyms.h>
#include <linux/debug_locks.h>
#include <linux/lockdep.h>
@@ -679,6 +680,7 @@ init_cpu_workqueue(struct workqueue_stru
spin_lock_init(&cwq->lock);
INIT_LIST_HEAD(&cwq->worklist);
init_waitqueue_head(&cwq->more_work);
+ cwq->thread = NULL;
return cwq;
}
@@ -712,7 +714,7 @@ static void start_workqueue_thread(struc
if (p != NULL) {
if (cpu >= 0)
- kthread_bind(p, cpu);
+ set_cpus_allowed(p, cpumask_of_cpu(cpu));
wake_up_process(p);
}
}
@@ -848,6 +850,9 @@ static int __devinit workqueue_cpu_callb
switch (action) {
case CPU_UP_PREPARE:
+ if (likely(cwq->thread != NULL &&
+ !IS_ERR(cwq->thread)))
+ break;
if (!create_workqueue_thread(cwq, cpu))
break;
printk(KERN_ERR "workqueue [%s] for %i failed\n",
@@ -858,12 +863,6 @@ static int __devinit workqueue_cpu_callb
case CPU_ONLINE:
start_workqueue_thread(cwq, cpu);
break;
-
- case CPU_UP_CANCELED:
- start_workqueue_thread(cwq, -1);
- case CPU_DEAD:
- cleanup_workqueue_thread(cwq, cpu);
- break;
}
}
--
Gautham R Shenoy
Linux Technology Center
IBM India.
"Freedom comes with a price tag of responsibility, which is still a bargain,
because Freedom is priceless!"
next prev parent reply other threads:[~2007-10-16 10:37 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-10-16 10:33 [RFC PATCH 0/4] Refcount Based Cpu-Hotplug Revisit Gautham R Shenoy
2007-10-16 10:34 ` [RFC PATCH 1/4] Refcount Based Cpu-Hotplug Implementation Gautham R Shenoy
2007-10-17 0:47 ` Rusty Russell
2007-10-17 5:37 ` Gautham R Shenoy
2007-10-17 6:29 ` Rusty Russell
2007-10-18 6:29 ` Gautham R Shenoy
2007-10-21 12:47 ` Oleg Nesterov
2007-10-17 10:53 ` Paul Jackson
2007-10-17 11:27 ` Paul Jackson
2007-10-17 11:50 ` Gautham R Shenoy
2007-10-17 12:04 ` Paul Jackson
2007-10-16 10:35 ` [RFC PATCH 2/4] Rename lock_cpu_hotplug to get_online_cpus Gautham R Shenoy
2007-10-17 16:13 ` Nathan Lynch
2007-10-18 7:57 ` Gautham R Shenoy
2007-10-18 8:22 ` Nathan Lynch
2007-10-18 8:59 ` Gautham R Shenoy
2007-10-18 17:30 ` Nathan Lynch
2007-10-19 5:04 ` Gautham R Shenoy
2007-10-22 0:43 ` Nathan Lynch
2007-10-22 4:51 ` Gautham R Shenoy
2007-10-16 10:36 ` [RFC PATCH 3/4] Replace per-subsystem mutexes with get_online_cpus Gautham R Shenoy
2007-10-21 11:39 ` Oleg Nesterov
2007-10-22 4:58 ` Gautham R Shenoy
2007-10-16 10:37 ` Gautham R Shenoy [this message]
2007-10-17 11:57 ` [RFC PATCH 4/4] Remove CPU_DEAD/CPU_UP_CANCELLED handling from workqueue.c Oleg Nesterov
2007-10-16 17:20 ` [RFC PATCH 0/4] Refcount Based Cpu-Hotplug Revisit Linus Torvalds
2007-10-17 2:11 ` Dipankar Sarma
2007-10-17 2:23 ` Linus Torvalds
2007-10-17 4:17 ` Gautham R Shenoy
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=20071016103721.GD16570@in.ibm.com \
--to=ego@in.ibm.com \
--cc=akpm@linux-foundation.org \
--cc=dipankar@in.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=oleg@tv-sign.ru \
--cc=paulmck@us.ibm.com \
--cc=rusty@rustcorp.com.au \
--cc=torvalds@linux-foundation.org \
--cc=vatsa@in.ibm.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