linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Corey Minyard <minyard@acm.org>
To: "河合英宏 / KAWAI,HIDEHIRO" <hidehiro.kawai.ez@hitachi.com>
Cc: "openipmi-developer@lists.sourceforge.net" 
	<openipmi-developer@lists.sourceforge.net>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 7/7] ipmi/kcs: Don't run the KCS state machine when it is KCS_IDLE
Date: Sat, 22 Aug 2015 12:45:39 -0500	[thread overview]
Message-ID: <55D8B543.4000805@acm.org> (raw)
In-Reply-To: <04EAB7311EE43145B2D3536183D1A84454938089@GSjpTKYDCembx31.service.hitachi.net>

On 08/17/2015 09:54 PM, 河合英宏 / KAWAI,HIDEHIRO wrote:
>> From: Corey Minyard [mailto:tcminyard@gmail.com] On Behalf Of Corey Minyard
>>
>> This patch will break ATN handling on the interfaces.  So we can't do this.
> I understand.  So how about doing like this:
>
> 	/* All states wait for ibf, so just do it here. */
> -	if (!check_ibf(kcs, status, time))
> +	if (kcs->state != KCS_IDLE && !check_ibf(kcs, status, time))
> 		return SI_SM_CALL_WITH_DELAY;
>
> I think it is not necessary to wait IBF when the state is IDLE.
> In this way, we can also handle the ATN case.

I think it would be more reliable to go up a level and add a timeout. 
One should
be there, anyway.  I thought they were all covered, but I may have missed
something.

-corey

>
> Regards,
>
> Hidehiro Kawai
> Hitachi, Ltd. Research & Development Group
>
>> It's going to be extremely hard to recover if the BMC is not working
>> correctly when a panic happens.  I'm not sure what can be done, but if
>> you can fix it another way it would be good.
>>
>> -corey
>>
>> On 07/27/2015 12:55 AM, Hidehiro Kawai wrote:
>>> If a BMC is unresponsive for some reason, it ends up completing
>>> the requested message as an error, then kcs_event() is called once
>>> to advance the state machine.  However, since the BMC is
>>> unresponsive now, the status of the KCS interface may not be
>>> idle.  As the result, the state machine can continue to run and
>>> comsume CPU time indefinitely even if there is no more request
>>> message.  Moreover, if this happens in run-to-completion mode
>>> (i.e. context of panic_event()), the kernel hangs up.
>>>
>>> To fix this problem, this patch ignores kcs_event() call if there
>>> is no request message to be processed.
>>>
>>> Signed-off-by: Hidehiro Kawai <hidehiro.kawai.ez@hitachi.com>
>>> ---
>>>  drivers/char/ipmi/ipmi_kcs_sm.c |    4 ++++
>>>  1 file changed, 4 insertions(+)
>>>
>>> diff --git a/drivers/char/ipmi/ipmi_kcs_sm.c b/drivers/char/ipmi/ipmi_kcs_sm.c
>>> index 8c25f59..0e187fb 100644
>>> --- a/drivers/char/ipmi/ipmi_kcs_sm.c
>>> +++ b/drivers/char/ipmi/ipmi_kcs_sm.c
>>> @@ -353,6 +353,10 @@ static enum si_sm_result kcs_event(struct si_sm_data *kcs, long time)
>>>  	if (kcs_debug & KCS_DEBUG_STATES)
>>>  		printk(KERN_DEBUG "KCS: State = %d, %x\n", kcs->state, status);
>>>
>>> +	/* We don't want to run the state machine when the state is IDLE */
>>> +	if (kcs->state == KCS_IDLE)
>>> +		return SI_SM_IDLE;
>>> +
>>>  	/* All states wait for ibf, so just do it here. */
>>>  	if (!check_ibf(kcs, status, time))
>>>  		return SI_SM_CALL_WITH_DELAY;
>>>
>>>


  reply	other threads:[~2015-08-23 23:15 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-07-27  5:55 [PATCH 0/7] ipmi: various fixes for panic notifier robustness Hidehiro Kawai
2015-07-27  5:55 ` [PATCH 5/7] ipmi: Don't call receive handler in the panic context Hidehiro Kawai
2015-07-27  5:55 ` [PATCH 6/7] ipmi: Handle queued messages more certainly on panic Hidehiro Kawai
2015-08-12  4:13   ` Corey Minyard
2015-08-18  1:59     ` 河合英宏 / KAWAI,HIDEHIRO
2015-08-22 17:39       ` Corey Minyard
2015-07-27  5:55 ` [PATCH 4/7] ipmi: Avoid touching possible corrupted lists in the panic context Hidehiro Kawai
2015-07-27  5:55 ` [PATCH 3/7] ipmi: Don't flush messages in sneder() in run-to-completion mode Hidehiro Kawai
2015-07-27  5:55 ` [PATCH 1/7] ipmi: Remove unneeded set_run_to_completion call Hidehiro Kawai
2015-07-27  5:55 ` [PATCH 2/7] ipmi: Factor out message flushing procedure Hidehiro Kawai
2015-07-27  5:55 ` [PATCH 7/7] ipmi/kcs: Don't run the KCS state machine when it is KCS_IDLE Hidehiro Kawai
2015-08-12  4:15   ` Corey Minyard
2015-08-18  2:54     ` 河合英宏 / KAWAI,HIDEHIRO
2015-08-22 17:45       ` Corey Minyard [this message]
2015-08-24  1:52         ` 河合英宏 / KAWAI,HIDEHIRO
2015-08-24 16:00           ` Corey Minyard
2015-08-25  3:53             ` 河合英宏 / KAWAI,HIDEHIRO
2015-08-26 20:27               ` Corey Minyard
2015-08-27  1:35                 ` 河合英宏 / KAWAI,HIDEHIRO

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=55D8B543.4000805@acm.org \
    --to=minyard@acm.org \
    --cc=hidehiro.kawai.ez@hitachi.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=openipmi-developer@lists.sourceforge.net \
    /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;
as well as URLs for NNTP newsgroup(s).