Re: [patch] add kdump_after_notifier

From: Jay Lan <jlan@sgi.com>
To: vgoyal@in.ibm.com
Cc: k-miyoshi@cb.jp.nec.com, Bernhard Walle <bwalle@suse.de>,
	kexec@lists.infradead.org,
	Takenori Nagano <t-nagano@ah.jp.nec.com>,
	linux-kernel@vger.kernel.org,
	"Eric W. Biederman" <ebiederm@xmission.com>,
	Keith Owens <kaos@ocs.com.au>,
	Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [patch] add kdump_after_notifier
Date: Thu, 23 Aug 2007 10:34:21 -0700	[thread overview]
Message-ID: <46CDC51D.5070206@sgi.com> (raw)
In-Reply-To: <20070823035629.GB365@in.ibm.com>

Vivek Goyal wrote:
> On Tue, Aug 21, 2007 at 06:18:31AM -0700, Jay Lan wrote:
> [..]
>>>>> Now user will be able to view all the die_chain users through sysfs and
>>>>> be able to modify the order in which these should run by modifying their
>>>>> priority. Hence all the RAS tools can co-exist.
>>>> This is my image of your proposal.
>>>>
>>>> - Print current order
>>>>
>>>> # cat /sys/class/misc/debug/panic_notifier_list
>>>> priority   name
>>>> 1          IPMI
>>>> 2          watchdog
>>>> 3          Kdb
>>>> 4          Kdump
>>>>
>>> I think Bernhard's suggestion looks better here. I noticed that 
>>> /sys/kernel/debug is already present. So how about following.
>>>
>>> /sys/kernel/debug/kdump/priority
>>> /sys/kernel/debug/kdb/priority
>>> /sys/kernel/debug/IPMI/priority
>> Why separate priority files is better than a central file?
>> At least i think you get a grand picture of priority being
>> defined for all parties with a central file?
>>
> 
> I thought of couple of reasons.
> - A very different syntax to modify the priority.
> - Separate directories allow easy future extensions in terms of more
>   files. For example, putting a small "description" file in each dir
>   where each registered user can specify what does it do.

The first can be easily resolved by providing a comment section in the
file with real examples. Users can simply uncomment a line to activate.
But future expansion is certainly is a good reason for this layout.

> 
> But I agree that a single file is good for consolidated view. As bernhard
> suggested, may be we should also implement a read only file where one
> will get a consolidated view.

Yep, this will help!

> 
>> What do we decide priority if more than one component has
>> the same priority value?
>>
> 
> I think first come first serve would be appropriate in this case instead of
> returning -EINVAL.

How does the kernel process the configuration files? By alphabetic order
of the filename? Either way, i think a clear failure/warning dmesg is
very important.

Thanks,
 - jay

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

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