From: Steven Rostedt <rostedt@goodmis.org>
To: linux-kernel@vger.kernel.org
Cc: Andrew Morton <akpm@linux-foundation.org>,
Ingo Molnar <mingo@elte.hu>, Thomas Gleixner <tglx@linutronix.de>,
"H. Peter Anvin" <hpa@zytor.com>
Subject: [patch 01/11] x86/irqs: Do not trace arch_local_{*,irq_*} functions
Date: Fri, 01 Jul 2011 23:04:36 -0400 [thread overview]
Message-ID: <20110702033852.733414762@goodmis.org> (raw)
In-Reply-To: 20110702030435.308336775@goodmis.org
[-- Attachment #1: 0001-x86-irqs-Do-not-trace-arch_local_-irq_-functions.patch --]
[-- Type: text/plain, Size: 3159 bytes --]
From: Steven Rostedt <srostedt@redhat.com>
I triggered a triple fault with gcc 4.5.1 because it did not honor
the inline annotation to arch_local_save_flags() function and that
function was added to the pool of functions traced by the function tracer.
When preempt_schedule() called arch_local_save_flags() (called by
irqs_disabled()), it was traced, but the first thing the function tracer
does is disable preemption. When it enables preemption, the NEED_RESCHED
flag will not have been cleared and the preemption check will trigger
the call to preempt_schedule() again.
Although the dynamic function tracer crashed immediately, the static
version of the function tracer (CONFIG_DYNAMIC_FTRACE is not set) actually
was able to show where the problem was.
swapper-1 3.N.. 103885us : arch_local_save_flags <-preempt_schedule
swapper-1 3.N.. 103886us : arch_local_save_flags <-preempt_schedule
swapper-1 3.N.. 103886us : arch_local_save_flags <-preempt_schedule
swapper-1 3.N.. 103887us : arch_local_save_flags <-preempt_schedule
swapper-1 3.N.. 103887us : arch_local_save_flags <-preempt_schedule
swapper-1 3.N.. 103888us : arch_local_save_flags <-preempt_schedule
swapper-1 3.N.. 103888us : arch_local_save_flags <-preempt_schedule
It went on for a while before it triple faulted with a corrupted stack.
The arch_local_save_flags and arch_local_irq_* functions should not be
traced. Even though they are marked as inline, gcc may still make
them a function and enable tracing of them.
The simple solution is to just mark them as notrace. I had to add the
<linux/types.h> for this file to include the notrace tag.
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
arch/x86/include/asm/irqflags.h | 11 ++++++-----
1 files changed, 6 insertions(+), 5 deletions(-)
diff --git a/arch/x86/include/asm/irqflags.h b/arch/x86/include/asm/irqflags.h
index 5745ce8..bba3cf8 100644
--- a/arch/x86/include/asm/irqflags.h
+++ b/arch/x86/include/asm/irqflags.h
@@ -60,23 +60,24 @@ static inline void native_halt(void)
#include <asm/paravirt.h>
#else
#ifndef __ASSEMBLY__
+#include <linux/types.h>
-static inline unsigned long arch_local_save_flags(void)
+static inline notrace unsigned long arch_local_save_flags(void)
{
return native_save_fl();
}
-static inline void arch_local_irq_restore(unsigned long flags)
+static inline notrace void arch_local_irq_restore(unsigned long flags)
{
native_restore_fl(flags);
}
-static inline void arch_local_irq_disable(void)
+static inline notrace void arch_local_irq_disable(void)
{
native_irq_disable();
}
-static inline void arch_local_irq_enable(void)
+static inline notrace void arch_local_irq_enable(void)
{
native_irq_enable();
}
@@ -102,7 +103,7 @@ static inline void halt(void)
/*
* For spinlocks, etc:
*/
-static inline unsigned long arch_local_irq_save(void)
+static inline notrace unsigned long arch_local_irq_save(void)
{
unsigned long flags = arch_local_save_flags();
arch_local_irq_disable();
--
1.7.5.4
next parent reply other threads:[~2011-07-02 3:45 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20110702030435.308336775@goodmis.org>
2011-07-02 3:04 ` Steven Rostedt [this message]
2011-07-07 0:51 ` [patch 01/11] x86/irqs: Do not trace arch_local_{*,irq_*} functions Steven Rostedt
2011-07-07 21:43 ` [tip:perf/core] tracing, x86/irq: Do not trace arch_local_{*,irq_*}() functions tip-bot for Steven Rostedt
2011-07-02 3:04 ` [patch 02/11] arm/irqs: Do not trace arch_local_{*,irq_*} functions Steven Rostedt
2011-07-02 3:04 ` [patch 03/11] blackfin/irqs: " Steven Rostedt
2011-07-02 6:50 ` Mike Frysinger
2011-07-02 16:11 ` Steven Rostedt
2011-07-02 16:39 ` Mike Frysinger
2011-07-02 16:40 ` Mike Frysinger
2011-07-05 12:51 ` Steven Rostedt
2011-07-05 16:20 ` Mike Frysinger
2011-07-05 19:07 ` Mike Frysinger
2011-07-05 19:13 ` Steven Rostedt
2011-07-05 19:25 ` Mike Frysinger
2011-07-02 3:04 ` [patch 04/11] ia64/irqs: " Steven Rostedt
2011-07-02 3:04 ` [patch 05/11] microblaze/irqs: " Steven Rostedt
2011-07-05 10:43 ` Michal Simek
2011-07-05 12:49 ` Steven Rostedt
2011-07-02 3:04 ` [patch 06/11] mips/irqs: " Steven Rostedt
2011-07-02 3:04 ` [patch 07/11] parisc/irqs: " Steven Rostedt
2011-07-02 3:04 ` [patch 08/11] powerpc/irqs: " Steven Rostedt
2011-07-02 3:04 ` [patch 09/11] s390/irqs: " Steven Rostedt
2011-07-04 8:10 ` Martin Schwidefsky
2011-07-05 12:49 ` Steven Rostedt
2011-07-02 3:04 ` [patch 10/11] sh/unicore32/irqs: " Steven Rostedt
2011-07-04 1:56 ` Guan Xuetao
2011-07-05 12:50 ` Steven Rostedt
2011-07-11 3:34 ` Paul Mundt
2011-07-13 1:36 ` Steven Rostedt
2011-07-02 3:04 ` [patch 11/11] sparc/irqs: " Steven Rostedt
2011-07-02 4:11 ` David Miller
2011-07-05 12:49 ` Steven Rostedt
2011-07-06 6:43 ` David Miller
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=20110702033852.733414762@goodmis.org \
--to=rostedt@goodmis.org \
--cc=akpm@linux-foundation.org \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=tglx@linutronix.de \
/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