linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] kexec: do not call kmsg_dump in crashdump
@ 2011-11-25  8:19 Dave Young
  2011-11-25  9:32 ` Cong Wang
  2011-11-28 22:27 ` Eric W. Biederman
  0 siblings, 2 replies; 5+ messages in thread
From: Dave Young @ 2011-11-25  8:19 UTC (permalink / raw)
  To: ebiederm, kexec, linux-kernel

mtd/ram oops dumper will make more functions envolved. In case crash kexec, 
we should avoid kmsg dumper to ensure crashdump success.

kill kmsg_dump callback while crash dumping 

Signed-off-by: Dave Young <dyoung@redhat.com>
---
 kernel/kexec.c |    2 --
 1 file changed, 2 deletions(-)

--- linux-2.6.orig/kernel/kexec.c	2011-11-18 16:34:18.000000000 +0800
+++ linux-2.6/kernel/kexec.c	2011-11-24 16:05:08.138660500 +0800
@@ -1094,8 +1094,6 @@ void crash_kexec(struct pt_regs *regs)
 		if (kexec_crash_image) {
 			struct pt_regs fixed_regs;
 
-			kmsg_dump(KMSG_DUMP_KEXEC);
-
 			crash_setup_regs(&fixed_regs, regs);
 			crash_save_vmcoreinfo();
 			machine_crash_shutdown(&fixed_regs);

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH] kexec: do not call kmsg_dump in crashdump
  2011-11-25  8:19 [PATCH] kexec: do not call kmsg_dump in crashdump Dave Young
@ 2011-11-25  9:32 ` Cong Wang
  2011-11-25 10:03   ` Dave Young
  2011-11-28 22:27 ` Eric W. Biederman
  1 sibling, 1 reply; 5+ messages in thread
From: Cong Wang @ 2011-11-25  9:32 UTC (permalink / raw)
  To: Dave Young; +Cc: ebiederm, kexec, linux-kernel

On Fri, Nov 25, 2011 at 4:19 PM, Dave Young <dyoung@redhat.com> wrote:
> mtd/ram oops dumper will make more functions envolved. In case crash kexec,
> we should avoid kmsg dumper to ensure crashdump success.
>
> kill kmsg_dump callback while crash dumping

https://lkml.org/lkml/2011/2/1/33

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH] kexec: do not call kmsg_dump in crashdump
  2011-11-25  9:32 ` Cong Wang
@ 2011-11-25 10:03   ` Dave Young
  0 siblings, 0 replies; 5+ messages in thread
From: Dave Young @ 2011-11-25 10:03 UTC (permalink / raw)
  To: Cong Wang; +Cc: ebiederm, kexec, linux-kernel

On 11/25/2011 05:32 PM, Cong Wang wrote:

> On Fri, Nov 25, 2011 at 4:19 PM, Dave Young <dyoung@redhat.com> wrote:
>> mtd/ram oops dumper will make more functions envolved. In case crash kexec,
>> we should avoid kmsg dumper to ensure crashdump success.
>>
>> kill kmsg_dump callback while crash dumping
> 
> https://lkml.org/lkml/2011/2/1/33


Thanks for give the link. So it's intend to be that.

I'm still confused with the KMSG_DUMP_KEXEC.

crash_kexec()
	kmsg_dump(KMSG_DUMP_KEXEC)
kmsg_dump(KMSG_DUMP_PANIC)

Looks like it's equal to original logic, isn't it?

kmsg_dump(anything);
crash_kexec()

So why intruduce KMSG_DUMP_KEXEC? why not just kmsg_dump before crash_kexec?

thoughts?

-- 
Thanks
Dave

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH] kexec: do not call kmsg_dump in crashdump
  2011-11-25  8:19 [PATCH] kexec: do not call kmsg_dump in crashdump Dave Young
  2011-11-25  9:32 ` Cong Wang
@ 2011-11-28 22:27 ` Eric W. Biederman
  2011-11-28 22:34   ` Andrew Morton
  1 sibling, 1 reply; 5+ messages in thread
From: Eric W. Biederman @ 2011-11-28 22:27 UTC (permalink / raw)
  To: Dave Young; +Cc: kexec, linux-kernel, Andrew Morton

Dave Young <dyoung@redhat.com> writes:

> mtd/ram oops dumper will make more functions envolved. In case crash kexec, 
> we should avoid kmsg dumper to ensure crashdump success.
>
> kill kmsg_dump callback while crash dumping 

Agreed.

There is a similar?  Or possibly the same patch already queued in
Andrews tree.

Eric

>
> Signed-off-by: Dave Young <dyoung@redhat.com>
> ---
>  kernel/kexec.c |    2 --
>  1 file changed, 2 deletions(-)
>
> --- linux-2.6.orig/kernel/kexec.c	2011-11-18 16:34:18.000000000 +0800
> +++ linux-2.6/kernel/kexec.c	2011-11-24 16:05:08.138660500 +0800
> @@ -1094,8 +1094,6 @@ void crash_kexec(struct pt_regs *regs)
>  		if (kexec_crash_image) {
>  			struct pt_regs fixed_regs;
>  
> -			kmsg_dump(KMSG_DUMP_KEXEC);
> -
>  			crash_setup_regs(&fixed_regs, regs);
>  			crash_save_vmcoreinfo();
>  			machine_crash_shutdown(&fixed_regs);

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH] kexec: do not call kmsg_dump in crashdump
  2011-11-28 22:27 ` Eric W. Biederman
@ 2011-11-28 22:34   ` Andrew Morton
  0 siblings, 0 replies; 5+ messages in thread
From: Andrew Morton @ 2011-11-28 22:34 UTC (permalink / raw)
  To: Eric W. Biederman; +Cc: Dave Young, kexec, linux-kernel

On Mon, 28 Nov 2011 14:27:02 -0800
ebiederm@xmission.com (Eric W. Biederman) wrote:

> Dave Young <dyoung@redhat.com> writes:
> 
> > mtd/ram oops dumper will make more functions envolved. In case crash kexec, 
> > we should avoid kmsg dumper to ensure crashdump success.
> >
> > kill kmsg_dump callback while crash dumping 
> 
> Agreed.
> 
> There is a similar?  Or possibly the same patch already queued in
> Andrews tree.

Below.

I've been sitting on this for ages because we couldn't seem to decide
whether it was desirable.  I think we decided a couple of weeks ago to
merge it so I have it planned for 3.3 unless we re-re-re-decide again.


From: WANG Cong <xiyou.wangcong@gmail.com>
Subject: kexec: remove KMSG_DUMP_KEXEC

KMSG_DUMP_KEXEC is useless because we already save kernel messages inside
/proc/vmcore, and it is unsafe to allow modules to do other stuffs in a
crash dump scenario.

[akpm@linux-foundation.org: fix powerpc build]
Signed-off-by: WANG Cong <xiyou.wangcong@gmail.com>
Reported-by: Vivek Goyal <vgoyal@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Acked-by: Jarod Wilson <jarod@redhat.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 arch/powerpc/platforms/pseries/nvram.c |    1 -
 drivers/char/ramoops.c                 |    3 +--
 drivers/mtd/mtdoops.c                  |    3 +--
 include/linux/kmsg_dump.h              |    1 -
 kernel/kexec.c                         |    3 ---
 5 files changed, 2 insertions(+), 9 deletions(-)

diff -puN arch/powerpc/platforms/pseries/nvram.c~kexec-remove-kmsg_dump_kexec arch/powerpc/platforms/pseries/nvram.c
--- a/arch/powerpc/platforms/pseries/nvram.c~kexec-remove-kmsg_dump_kexec
+++ a/arch/powerpc/platforms/pseries/nvram.c
@@ -636,7 +636,6 @@ static void oops_to_nvram(struct kmsg_du
 		/* These are almost always orderly shutdowns. */
 		return;
 	case KMSG_DUMP_OOPS:
-	case KMSG_DUMP_KEXEC:
 		break;
 	case KMSG_DUMP_PANIC:
 		panicking = true;
diff -puN drivers/char/ramoops.c~kexec-remove-kmsg_dump_kexec drivers/char/ramoops.c
--- a/drivers/char/ramoops.c~kexec-remove-kmsg_dump_kexec
+++ a/drivers/char/ramoops.c
@@ -83,8 +83,7 @@ static void ramoops_do_dump(struct kmsg_
 	struct timeval timestamp;
 
 	if (reason != KMSG_DUMP_OOPS &&
-	    reason != KMSG_DUMP_PANIC &&
-	    reason != KMSG_DUMP_KEXEC)
+	    reason != KMSG_DUMP_PANIC)
 		return;
 
 	/* Only dump oopses if dump_oops is set */
diff -puN drivers/mtd/mtdoops.c~kexec-remove-kmsg_dump_kexec drivers/mtd/mtdoops.c
--- a/drivers/mtd/mtdoops.c~kexec-remove-kmsg_dump_kexec
+++ a/drivers/mtd/mtdoops.c
@@ -308,8 +308,7 @@ static void mtdoops_do_dump(struct kmsg_
 	char *dst;
 
 	if (reason != KMSG_DUMP_OOPS &&
-	    reason != KMSG_DUMP_PANIC &&
-	    reason != KMSG_DUMP_KEXEC)
+	    reason != KMSG_DUMP_PANIC)
 		return;
 
 	/* Only dump oopses if dump_oops is set */
diff -puN include/linux/kmsg_dump.h~kexec-remove-kmsg_dump_kexec include/linux/kmsg_dump.h
--- a/include/linux/kmsg_dump.h~kexec-remove-kmsg_dump_kexec
+++ a/include/linux/kmsg_dump.h
@@ -18,7 +18,6 @@
 enum kmsg_dump_reason {
 	KMSG_DUMP_OOPS,
 	KMSG_DUMP_PANIC,
-	KMSG_DUMP_KEXEC,
 	KMSG_DUMP_RESTART,
 	KMSG_DUMP_HALT,
 	KMSG_DUMP_POWEROFF,
diff -puN kernel/kexec.c~kexec-remove-kmsg_dump_kexec kernel/kexec.c
--- a/kernel/kexec.c~kexec-remove-kmsg_dump_kexec
+++ a/kernel/kexec.c
@@ -32,7 +32,6 @@
 #include <linux/console.h>
 #include <linux/vmalloc.h>
 #include <linux/swap.h>
-#include <linux/kmsg_dump.h>
 #include <linux/syscore_ops.h>
 
 #include <asm/page.h>
@@ -1094,8 +1093,6 @@ void crash_kexec(struct pt_regs *regs)
 		if (kexec_crash_image) {
 			struct pt_regs fixed_regs;
 
-			kmsg_dump(KMSG_DUMP_KEXEC);
-
 			crash_setup_regs(&fixed_regs, regs);
 			crash_save_vmcoreinfo();
 			machine_crash_shutdown(&fixed_regs);
_


^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2011-11-28 22:34 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2011-11-25  8:19 [PATCH] kexec: do not call kmsg_dump in crashdump Dave Young
2011-11-25  9:32 ` Cong Wang
2011-11-25 10:03   ` Dave Young
2011-11-28 22:27 ` Eric W. Biederman
2011-11-28 22:34   ` Andrew Morton

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 NNTP newsgroup(s).