Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Michel Thierry <michel.thierry@intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>, intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH 05/20] drm/i915: Cancel reset-engine if we couldn't find an active request
Date: Tue, 16 May 2017 17:13:58 -0700	[thread overview]
Message-ID: <a5bd86ed-988a-829e-6bb9-a415b482885f@intel.com> (raw)
In-Reply-To: <20170516075424.GN19912@nuc-i3427.alporthouse.com>

On 16/05/17 00:54, Chris Wilson wrote:
> On Mon, May 15, 2017 at 03:25:27PM -0700, Michel Thierry wrote:
>> On 5/15/2017 2:47 PM, Chris Wilson wrote:
>>> On Mon, May 15, 2017 at 10:31:58PM +0100, Chris Wilson wrote:
>>>> On Mon, May 15, 2017 at 02:20:01PM -0700, Michel Thierry wrote:
>>>>> @@ -2827,21 +2830,34 @@ int i915_gem_reset_prepare_engine(struct intel_engine_cs *engine)
>>>>>
>>>>> 	if (engine_stalled(engine)) {
>>>>> 		request = i915_gem_find_active_request(engine);
>>>>> -		if (request && request->fence.error == -EIO)
>>>>> -			err = -EIO; /* Previous reset failed! */
>>>>> +
>>>>> +		if (request) {
>>>>> +			if (request->fence.error == -EIO)
>>>>> +				return ERR_PTR(-EIO); /* Previous reset failed! */
>>>>> +
>>>>> +			if (__i915_gem_request_completed(request,
>>>>> +							 engine->hangcheck.seqno))
>>>>
>>>> This is not the seqno for the request, so this is incorrect. It will
>>>> judge that the request was preempted (as hangcheck.seqno must be less
>>>> thn request->global_seqno) and so conclude that the request was never
>>>> completed.
>>>>
>>>> You just want if (i915_gem_request_completed(request))
>>
>> Thanks, I'll change the function.
>>
>>>
>>> Also not here. This pardon check should be deferred to the caller just
>>> before commiting to thre reset. In the case of global reset, we want to
>>> gather up all the engines' active requests first, complete our
>>> preparations and then double check the engine was hung.
>>
>> i915_reset_engine calls this directly, but 'full reset' [from
>> i915_gem_reset_prepare()] would not be affected and it won't pardon
>> anything... i915_gem_reset_engine is doing the double check you
>> mention.
>
> Aye, but in the long run I was thinking of capturing this request in
> engine->hangcheck.active_request and then we reuse that info in the later
> phases.

Capture hangcheck.active_request during hangcheck_declare_hang? Or still 
here in reset_prepare?

Thanks
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

  reply	other threads:[~2017-05-17  0:13 UTC|newest]

Thread overview: 62+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-04-27 23:12 [PATCH v7 00/20] Gen8+ engine-reset Michel Thierry
2017-04-27 23:12 ` [PATCH v7 01/20] drm/i915: Update i915.reset to handle engine resets Michel Thierry
2017-04-27 23:12 ` [PATCH v7 02/20] drm/i915: Modify error handler for per engine hang recovery Michel Thierry
2017-04-29 14:19   ` Chris Wilson
2017-05-08 18:31     ` Michel Thierry
2017-05-12 20:55       ` Michel Thierry
2017-05-12 21:09         ` Chris Wilson
2017-05-12 21:23           ` Michel Thierry
2017-05-15 21:14   ` [PATCH " Michel Thierry
2017-04-27 23:12 ` [PATCH v7 03/20] drm/i915: Add support for per engine reset recovery Michel Thierry
2017-04-27 23:50   ` Chris Wilson
2017-04-28 21:59     ` Michel Thierry
2017-05-04  0:26     ` Michel Thierry
2017-05-15 21:18   ` [PATCH " Michel Thierry
2017-04-27 23:12 ` [PATCH v7 04/20] drm/i915: Skip reset request if there is one already Michel Thierry
2017-04-29 14:21   ` Chris Wilson
2017-05-01 21:15     ` Michel Thierry
2017-04-27 23:12 ` [PATCH v7 05/20] drm/i915: Cancel reset-engine if we couldn't find an active request Michel Thierry
2017-04-29 14:26   ` Chris Wilson
2017-05-15 21:20   ` [PATCH " Michel Thierry
2017-05-15 21:31     ` Chris Wilson
2017-05-15 21:47       ` Chris Wilson
2017-05-15 22:25         ` Michel Thierry
2017-05-16  7:54           ` Chris Wilson
2017-05-17  0:13             ` Michel Thierry [this message]
2017-05-17  7:19               ` Chris Wilson
2017-05-17 20:41   ` [PATCH] " Michel Thierry
2017-05-17 20:52     ` Chris Wilson
2017-05-18  1:11       ` Michel Thierry
2017-05-18  7:56         ` Chris Wilson
2017-05-18 17:19           ` Michel Thierry
2017-05-18 18:22   ` [PATCH] drm/i915: Look for active requests earlier in the reset path Michel Thierry
2017-05-18 18:26     ` Michel Thierry
2017-05-18 19:55     ` Chris Wilson
2017-05-18 21:11   ` Michel Thierry
2017-05-18 21:16     ` Chris Wilson
2017-05-18 21:34       ` Michel Thierry
2017-04-27 23:12 ` [PATCH v7 06/20] drm/i915: Add engine reset count to error state Michel Thierry
2017-04-27 23:12 ` [PATCH v7 07/20] drm/i915: Export per-engine reset count info to debugfs Michel Thierry
2017-04-27 23:12 ` [PATCH v7 08/20] drm/i915: Enable Engine reset and recovery support Michel Thierry
2017-04-27 23:12 ` [PATCH v7 09/20] drm/i915: Add engine reset count in get-reset-stats ioctl Michel Thierry
2017-04-27 23:12 ` [PATCH v7 10/20] drm/i915/selftests: reset engine self tests Michel Thierry
2017-04-27 23:12 ` [PATCH v7 11/20] drm/i915/guc: fix mmio whitelist mmio_start offset and add reminder Michel Thierry
2017-04-27 23:12 ` [PATCH v7 12/20] drm/i915/guc: Provide register list to be saved/restored during engine reset Michel Thierry
2017-04-27 23:58   ` Chris Wilson
2017-04-28 15:36     ` Michel Thierry
2017-04-27 23:12 ` [PATCH v7 13/20] drm/i915/guc: Rename the function that resets the GuC Michel Thierry
2017-04-28  7:40   ` Tvrtko Ursulin
2017-05-01 20:09     ` Michel Thierry
2017-04-27 23:12 ` [PATCH v7 14/20] drm/i915/guc: Add support for reset engine using GuC commands Michel Thierry
2017-04-27 23:12 ` [PATCH v7 15/20] drm/i915: Watchdog timeout: Pass GuC shared data structure during param load Michel Thierry
2017-04-27 23:12 ` [PATCH v7 16/20] drm/i915: Watchdog timeout: IRQ handler for gen8+ Michel Thierry
2017-04-27 23:12 ` [PATCH v7 17/20] drm/i915: Watchdog timeout: Ringbuffer command emission " Michel Thierry
2017-04-27 23:12 ` [PATCH v7 18/20] drm/i915: Watchdog timeout: DRM kernel interface to set the timeout Michel Thierry
2017-04-27 23:12 ` [PATCH v7 19/20] drm/i915: Watchdog timeout: Include threshold value in error state Michel Thierry
2017-04-27 23:13 ` [PATCH v7 20/20] drm/i915: Watchdog timeout: Export media reset count from GuC to debugfs Michel Thierry
2017-04-27 23:30 ` ✓ Fi.CI.BAT: success for Gen8+ engine-reset (rev4) Patchwork
2017-05-15 21:32 ` ✓ Fi.CI.BAT: success for Gen8+ engine-reset (rev5) Patchwork
2017-05-15 21:48 ` ✓ Fi.CI.BAT: success for Gen8+ engine-reset (rev7) Patchwork
2017-05-17 21:09 ` ✓ Fi.CI.BAT: success for Gen8+ engine-reset (rev8) Patchwork
2017-05-18 18:40 ` ✓ Fi.CI.BAT: success for Gen8+ engine-reset (rev9) Patchwork
2017-05-18 21:29 ` ✓ Fi.CI.BAT: success for Gen8+ engine-reset (rev10) Patchwork

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=a5bd86ed-988a-829e-6bb9-a415b482885f@intel.com \
    --to=michel.thierry@intel.com \
    --cc=chris@chris-wilson.co.uk \
    --cc=intel-gfx@lists.freedesktop.org \
    /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