All of lore.kernel.org
 help / color / mirror / Atom feed
From: Dimitri Sivanich <sivanich@sgi.com>
To: linux-ia64@vger.kernel.org
Subject: [PATCH] IA64: eliminate race condition in smp_flush_tlb_mm
Date: Tue, 28 Dec 2010 19:34:42 +0000	[thread overview]
Message-ID: <20101228193442.GA28269@sgi.com> (raw)

A race condition exists within smp_call_function_many() when called from
smp_flush_tlb_mm().  On rare occasions the cpu_vm_mask can be cleared
while smp_call_function_many is executing, occasionally resulting in a
hung process.

Make a copy of the mask prior to calling smp_call_function_many().

Signed-off-by: Dimitri Sivanich <sivanich@sgi.com>

---
 arch/ia64/kernel/smp.c |   13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

Index: linux-2.6.36.2/arch/ia64/kernel/smp.c
=================================--- linux-2.6.36.2.orig/arch/ia64/kernel/smp.c	2010-12-28 13:13:16.261122111 -0600
+++ linux-2.6.36.2/arch/ia64/kernel/smp.c	2010-12-28 13:30:17.504590090 -0600
@@ -293,6 +293,7 @@ smp_flush_tlb_all (void)
 void
 smp_flush_tlb_mm (struct mm_struct *mm)
 {
+	cpumask_var_t cpus;
 	preempt_disable();
 	/* this happens for the common case of a single-threaded fork():  */
 	if (likely(mm = current->active_mm && atomic_read(&mm->mm_users) = 1))
@@ -301,9 +302,15 @@ smp_flush_tlb_mm (struct mm_struct *mm)
 		preempt_enable();
 		return;
 	}
-
-	smp_call_function_many(mm_cpumask(mm),
-		(void (*)(void *))local_finish_flush_tlb_mm, mm, 1);
+	if (!alloc_cpumask_var(&cpus, GFP_ATOMIC)) {
+		smp_call_function((void (*)(void *))local_finish_flush_tlb_mm,
+			mm, 1);
+	} else {
+		cpumask_copy(cpus, mm_cpumask(mm));
+		smp_call_function_many(cpus,
+			(void (*)(void *))local_finish_flush_tlb_mm, mm, 1);
+		free_cpumask_var(cpus);
+	}
 	local_irq_disable();
 	local_finish_flush_tlb_mm(mm);
 	local_irq_enable();

                 reply	other threads:[~2010-12-28 19:34 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=20101228193442.GA28269@sgi.com \
    --to=sivanich@sgi.com \
    --cc=linux-ia64@vger.kernel.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.