From: Vivek Goyal <vgoyal@in.ibm.com>
To: Takenori Nagano <t-nagano@ah.jp.nec.com>,
Keith Owens <kaos@sgi.com>,
"Eric W. Biederman" <ebiederm@xmission.com>,
k-miyoshi@cb.jp.nec.com, kexec@lists.infradead.org,
linux-kernel@vger.kernel.org,
Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [patch] add kdump_after_notifier
Date: Tue, 14 Aug 2007 18:54:54 +0530 [thread overview]
Message-ID: <20070814132454.GA8293@in.ibm.com> (raw)
In-Reply-To: <20070814083710.GA14538@suse.de>
On Tue, Aug 14, 2007 at 10:37:10AM +0200, Bernhard Walle wrote:
> * Takenori Nagano <t-nagano@ah.jp.nec.com> [2007-08-14 10:34]:
> > Vivek Goyal wrote:
> > > On Fri, Aug 03, 2007 at 02:05:47PM +1000, Keith Owens wrote:
> > >
> > >
> > > To sum up, couple of options come to mind.
> > >
> > > - Register all the RAS tools on die notifier and panic
> > > notifier lists with fairly high priority. Export list
> > > of RAS tools to user space and allow users to decide the
> > > order of execution and priority of RAS tools.
> > >
> > > - Create a separate RAS tool notifier list (ras_tool_notifer_list).
> > > All the RAS tools register on this list. This list gets priority
> > > over die or panic notifier list. User decides the oder of execution
> > > of RAS tools.
> > >
> > > Here assumption is that above list will not be exported to modules.
> > > All the RAS tools will be in kernel and they always get a priority
> > > to inspect an event.
> > >
> > > What do others think?
> >
> > Very good idea. But there is a problem how to give default priority to RAS tools.
> >
> > How about priority changeable notifier_list? User can change list order
> > dynamically if they want. Of course, we have to give highest priority to kdump
> > by default. It is very useful for users who want to use some RAS tools.
>
> I think that was the idea of the first “-” (“export list of RAS tools
> to user space”).
>
So for the time being I think we can put RAS tools on die notifier list
and if it runs into issues we can always think of creating a separate list.
Few things come to mind.
- Why there is a separate panic_notifier_list? Can't it be merged with
die_chain? die_val already got one of the event type as PANIC. If there
are no specific reasons then we should merge the two lists. Registering
RAS tools on a single list is easier.
- Modify Kdump to register on die_chain list.
- Modify Kdb to register on die_chain list.
- Export all the registered members of die_chain through sysfs along with
their priorities. Priorities should be modifiable. Most likely one
shall have to introduce additional field in struct notifier_block. This
field will be a string as an identifier of the user registerd. e.g
"Kdump", "Kdb" etc.
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.
Any thoughts?
Thanks
Vivek
>
> Thanks,
> Bernhard
_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec
WARNING: multiple messages have this Message-ID (diff)
From: Vivek Goyal <vgoyal@in.ibm.com>
To: Takenori Nagano <t-nagano@ah.jp.nec.com>,
Keith Owens <kaos@sgi.com>,
"Eric W. Biederman" <ebiederm@xmission.com>,
k-miyoshi@cb.jp.nec.com, kexec@lists.infradead.org,
linux-kernel@vger.kernel.org,
Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [patch] add kdump_after_notifier
Date: Tue, 14 Aug 2007 18:54:54 +0530 [thread overview]
Message-ID: <20070814132454.GA8293@in.ibm.com> (raw)
In-Reply-To: <20070814083710.GA14538@suse.de>
On Tue, Aug 14, 2007 at 10:37:10AM +0200, Bernhard Walle wrote:
> * Takenori Nagano <t-nagano@ah.jp.nec.com> [2007-08-14 10:34]:
> > Vivek Goyal wrote:
> > > On Fri, Aug 03, 2007 at 02:05:47PM +1000, Keith Owens wrote:
> > >
> > >
> > > To sum up, couple of options come to mind.
> > >
> > > - Register all the RAS tools on die notifier and panic
> > > notifier lists with fairly high priority. Export list
> > > of RAS tools to user space and allow users to decide the
> > > order of execution and priority of RAS tools.
> > >
> > > - Create a separate RAS tool notifier list (ras_tool_notifer_list).
> > > All the RAS tools register on this list. This list gets priority
> > > over die or panic notifier list. User decides the oder of execution
> > > of RAS tools.
> > >
> > > Here assumption is that above list will not be exported to modules.
> > > All the RAS tools will be in kernel and they always get a priority
> > > to inspect an event.
> > >
> > > What do others think?
> >
> > Very good idea. But there is a problem how to give default priority to RAS tools.
> >
> > How about priority changeable notifier_list? User can change list order
> > dynamically if they want. Of course, we have to give highest priority to kdump
> > by default. It is very useful for users who want to use some RAS tools.
>
> I think that was the idea of the first “-” (“export list of RAS tools
> to user space”).
>
So for the time being I think we can put RAS tools on die notifier list
and if it runs into issues we can always think of creating a separate list.
Few things come to mind.
- Why there is a separate panic_notifier_list? Can't it be merged with
die_chain? die_val already got one of the event type as PANIC. If there
are no specific reasons then we should merge the two lists. Registering
RAS tools on a single list is easier.
- Modify Kdump to register on die_chain list.
- Modify Kdb to register on die_chain list.
- Export all the registered members of die_chain through sysfs along with
their priorities. Priorities should be modifiable. Most likely one
shall have to introduce additional field in struct notifier_block. This
field will be a string as an identifier of the user registerd. e.g
"Kdump", "Kdb" etc.
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.
Any thoughts?
Thanks
Vivek
>
> Thanks,
> Bernhard
next prev parent reply other threads:[~2007-08-14 13:25 UTC|newest]
Thread overview: 77+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-07-19 12:15 [patch] add kdump_after_notifier Takenori Nagano
2007-07-19 12:15 ` Takenori Nagano
2007-07-26 14:07 ` Bernhard Walle
2007-07-26 14:07 ` Bernhard Walle
2007-07-26 15:32 ` Vivek Goyal
2007-07-26 15:32 ` Vivek Goyal
2007-07-26 15:34 ` Bernhard Walle
2007-07-26 15:34 ` Bernhard Walle
2007-07-26 15:44 ` Vivek Goyal
2007-07-26 15:44 ` Vivek Goyal
2007-07-26 15:47 ` Bernhard Walle
2007-07-26 15:47 ` Bernhard Walle
2007-07-26 15:54 ` Vivek Goyal
2007-07-26 15:54 ` Vivek Goyal
2007-07-26 16:14 ` Bernhard Walle
2007-07-26 16:14 ` Bernhard Walle
2007-07-26 16:21 ` Bernhard Walle
2007-07-26 16:21 ` Bernhard Walle
2007-07-26 23:28 ` Takenori Nagano
2007-07-26 23:28 ` Takenori Nagano
2007-07-30 9:16 ` Vivek Goyal
2007-07-30 9:16 ` Vivek Goyal
2007-07-30 13:42 ` Eric W. Biederman
2007-07-30 13:42 ` Eric W. Biederman
2007-07-31 5:55 ` Takenori Nagano
2007-07-31 5:55 ` Takenori Nagano
2007-07-31 6:53 ` Eric W. Biederman
2007-07-31 6:53 ` Eric W. Biederman
2007-08-01 9:26 ` Takenori Nagano
2007-08-01 9:26 ` Takenori Nagano
2007-08-01 10:00 ` Eric W. Biederman
2007-08-01 10:00 ` Eric W. Biederman
2007-08-02 8:11 ` Takenori Nagano
2007-08-02 8:11 ` Takenori Nagano
2007-08-02 11:28 ` Vivek Goyal
2007-08-02 11:28 ` Vivek Goyal
2007-08-03 4:05 ` Keith Owens
2007-08-03 4:05 ` Keith Owens
2007-08-03 6:25 ` Andrew Morton
2007-08-03 6:25 ` Andrew Morton
2007-08-03 6:34 ` Keith Owens
2007-08-03 6:34 ` Keith Owens
2007-08-03 7:37 ` Andrew Morton
2007-08-03 7:37 ` Andrew Morton
2007-08-03 7:10 ` Eric W. Biederman
2007-08-03 7:10 ` Eric W. Biederman
2007-08-05 11:07 ` Vivek Goyal
2007-08-05 11:07 ` Vivek Goyal
2007-08-14 8:34 ` Takenori Nagano
2007-08-14 8:34 ` Takenori Nagano
2007-08-14 8:37 ` Bernhard Walle
2007-08-14 8:37 ` Bernhard Walle
2007-08-14 8:48 ` Takenori Nagano
2007-08-14 8:48 ` Takenori Nagano
2007-08-14 8:53 ` Bernhard Walle
2007-08-14 8:53 ` Bernhard Walle
2007-08-14 13:24 ` Vivek Goyal [this message]
2007-08-14 13:24 ` Vivek Goyal
2007-08-16 9:26 ` Takenori Nagano
2007-08-16 9:26 ` Takenori Nagano
2007-08-16 9:45 ` Bernhard Walle
2007-08-16 9:45 ` Bernhard Walle
2007-08-17 10:56 ` Vivek Goyal
2007-08-17 10:56 ` Vivek Goyal
2007-08-21 7:45 ` Takenori Nagano
2007-08-21 7:45 ` Takenori Nagano
2007-08-23 3:52 ` Vivek Goyal
2007-08-23 3:52 ` Vivek Goyal
2007-08-21 13:18 ` Jay Lan
2007-08-21 13:18 ` Jay Lan
2007-08-21 13:21 ` Bernhard Walle
2007-08-21 13:21 ` Bernhard Walle
2007-08-23 3:56 ` Vivek Goyal
2007-08-23 3:56 ` Vivek Goyal
2007-08-23 17:34 ` Jay Lan
2007-08-23 17:34 ` Jay Lan
2007-10-04 11:30 ` Takenori Nagano
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=20070814132454.GA8293@in.ibm.com \
--to=vgoyal@in.ibm.com \
--cc=akpm@linux-foundation.org \
--cc=ebiederm@xmission.com \
--cc=k-miyoshi@cb.jp.nec.com \
--cc=kaos@sgi.com \
--cc=kexec@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=t-nagano@ah.jp.nec.com \
/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.