Re: [RFC V2 PATCH 0/1] kexec: crash_kexec_post_notifiers boot option related fixes

From: Hidehiro Kawai <hidehiro.kawai.ez@hitachi.com>
To: "Eric W. Biederman" <ebiederm@xmission.com>,
	Vivek Goyal <vgoyal@redhat.com>
Cc: linux-mips@linux-mips.org, Baoquan He <bhe@redhat.com>,
	kexec@lists.infradead.org, linux-kernel@vger.kernel.org,
	HATAYAMA Daisuke <d.hatayama@jp.fujitsu.com>,
	Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>,
	Daniel Walker <dwalker@fifo99.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Ingo Molnar <mingo@kernel.org>
Subject: Re: [RFC V2 PATCH 0/1] kexec: crash_kexec_post_notifiers boot option related fixes
Date: Mon, 03 Aug 2015 20:06:07 +0900	[thread overview]
Message-ID: <55BF4B1F.9000602@hitachi.com> (raw)
In-Reply-To: <20150724011615.6834.79628.stgit@softrs>

Hello Eric and Vivek,

Do you have any comments?

(2015/07/24 10:16), Hidehiro Kawai wrote:
> This is a bugfix patch for crash_kexec_post_notifiers boot option
> which allows users to call panic notifiers and kmsg dumpers before
> kdump.
> 
> This fixes one of the problems reported by Daniel Walker
> (https://lkml.org/lkml/2015/6/24/44). 
> 
>  Problem 1:
>  If crash_kexec_post_notifiers boot option is specified, some
>  shutting down process which assume other cpus are still alive
>  don't work properly.
> 
>  Problem 2 (addressed by this patch):
>  If crash_kexec_post_notifiers boot option is specified, register
>  information of other cpus are not saved to crash dumps.
> 
> Following Vivek's opinion, this patch replaces smp_send_stop()
> in panic() with suitable version for crash_kexec which saves
> cpu states and other things like cleaning up VMX/SVM.  Since this
> needs architecture specific implementation and it's not so trivial,
> this version only support for x86.  So the problem 1, known to
> happen on MIPS/OCTEON, is not addressed now.
> 
> To keep the modification impact low, this patch doesn't change
> the logic basically if crash_kexec_post_notifiers is not specified.
> 
> Please note that crash_kexec() can be called directly without
> entering panic().  Stopping other cpus functionality is still
> needed in crash_kexec().
> 
> Changes in V2:
> - Replace smp_send_stop() call with crash_kexec version which
>   saves cpu states and does cleanups instead of changing execution
>   flow
> - Drop a fix for Problem 1
> - Drop other patches because they aren't needed anymore
> 
> V1: https://lkml.org/lkml/2015/7/10/316
> 
> ---
> 
> Hidehiro Kawai (1):
>       panic/x86: Replace smp_send_stop() with crash_kexec version
> 
> 
>  arch/x86/kernel/crash.c |   16 +++++++++++-----
>  kernel/panic.c          |   29 +++++++++++++++++++++++------
>  2 files changed, 34 insertions(+), 11 deletions(-)
> 
> 

-- 
Hidehiro Kawai
Hitachi, Ltd. Research & Development Group

_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec