From: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>, intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH 6/8] drm/i915/guc: Keep the engine awake until the tasklet is idle
Date: Mon, 12 Aug 2019 13:38:39 -0700 [thread overview]
Message-ID: <fea59a24-656f-4bd3-77d4-8e00f17b84af@intel.com> (raw)
In-Reply-To: <156560667015.2301.4966068073571425460@skylake-alporthouse-com>
On 8/12/19 3:44 AM, Chris Wilson wrote:
> Quoting Chris Wilson (2019-08-12 10:10:43)
>> For the guc, we need to keep the engine awake (and not parked) and not
>> just the gt. If we let the engine park, we disable the irq and stop
>> processing the tasklet, leaving state outstanding inside the tasklet.
>>
>> The downside is, of course, we now have to wait until the tasklet is run
>> before we consider the engine idle.
>
> Fwiw, because of this I think it may be preferable to keep to using GT
> pm for the tasklet; and apply Daniele's patch to keep
> NEEDS_BREADCRUMB_TASKLET set (which is the right thing to do anyway now
> that we stop switching between submission modes).
> -Chris
>
Given that the GuC submission code is about to undergo a rework I
believe it'd be better to keep the fix contained to the GuC side of
things for now and avoid impacting the more general request paths (i.e.
patch 4 in this series, unless you still want that for other reasons).
I'll clean up and send the other patch.
Daniele
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2019-08-12 20:38 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-08-12 9:10 [PATCH 1/8] drm/i915/execlists: Avoid sync calls during park Chris Wilson
2019-08-12 9:10 ` [PATCH 2/8] drm/i915/selftests: Prevent the timeslice expiring during suppression tests Chris Wilson
2019-08-12 9:39 ` Mika Kuoppala
2019-08-12 9:58 ` Chris Wilson
2019-08-12 10:28 ` Mika Kuoppala
2019-08-12 9:10 ` [PATCH 3/8] drm/i915/guc: Use a local cancel_port_requests Chris Wilson
2019-08-12 9:10 ` [PATCH 4/8] drm/i915: Push the wakeref->count deferral to the backend Chris Wilson
2019-08-12 9:10 ` [PATCH 5/8] drm/i915/gt: Save/restore interrupts around breadcrumb disable Chris Wilson
2019-08-12 9:10 ` [PATCH 6/8] drm/i915/guc: Keep the engine awake until the tasklet is idle Chris Wilson
2019-08-12 10:44 ` Chris Wilson
2019-08-12 20:38 ` Daniele Ceraolo Spurio [this message]
2019-08-12 20:42 ` Chris Wilson
2019-08-12 9:10 ` [PATCH 7/8] drm/i915/gt: Use the local engine wakeref when checking RING registers Chris Wilson
2019-08-12 12:16 ` Mika Kuoppala
2019-08-12 9:10 ` [PATCH 8/8] drm/i915/execlists: Lift process_csb() out of the irq-off spinlock Chris Wilson
2019-08-12 11:13 ` [PATCH] " Chris Wilson
2019-08-12 15:29 ` kbuild test robot
2019-08-12 9:27 ` [PATCH 1/8] drm/i915/execlists: Avoid sync calls during park Mika Kuoppala
2019-08-12 9:33 ` Chris Wilson
2019-08-12 9:40 ` Mika Kuoppala
2019-08-12 12:54 ` ✗ Fi.CI.CHECKPATCH: warning for series starting with [1/8] drm/i915/execlists: Avoid sync calls during park (rev2) Patchwork
2019-08-12 12:57 ` ✗ Fi.CI.SPARSE: " Patchwork
2019-08-12 13:22 ` ✓ Fi.CI.BAT: success " Patchwork
2019-08-12 19:21 ` ✓ Fi.CI.IGT: " 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=fea59a24-656f-4bd3-77d4-8e00f17b84af@intel.com \
--to=daniele.ceraolospurio@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