From: Chris Wilson <chris@chris-wilson.co.uk>
To: John Harrison <John.C.Harrison@Intel.com>
Cc: intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH 08/13] drm/i915: Drive request submission through fence callbacks
Date: Fri, 26 Aug 2016 17:20:56 +0100 [thread overview]
Message-ID: <20160826162056.GK27196@nuc-i3427.alporthouse.com> (raw)
In-Reply-To: <8355cb68-bdb6-82c1-8862-de350a340cba@Intel.com>
On Fri, Aug 26, 2016 at 01:47:50PM +0100, John Harrison wrote:
> On 25/08/2016 10:08, Chris Wilson wrote:
> >diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
> >index 5e60519ede8d..babeaa8b1273 100644
> >--- a/drivers/gpu/drm/i915/intel_lrc.c
> >+++ b/drivers/gpu/drm/i915/intel_lrc.c
> >@@ -538,14 +538,15 @@ static void intel_lrc_irq_handler(unsigned long data)
> > static void execlists_submit_request(struct drm_i915_gem_request *request)
> Need to document that submit_request is now called from a callback
> and potentially from a tasklet or even an IRQ?
Yes, the signal notify is always in atomic context, because the parent is
holding an irq spinlock. We should avoid limiting when kfence_complete()
is legal (hardirq support should be a requirement), but different users
may know that their fences are nevery used from such contexts (though
this notify is always atomic due to the spinlock).
The release notify may also be in any context - when mixing with dma
fences it may indeed be from atomic context due to the fence tacking a
spinlock around its callbacks.
-Chris
--
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2016-08-26 16:21 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-08-25 9:08 Shortest path to EGL_ANDRIOD_native_sync Chris Wilson
2016-08-25 9:08 ` [PATCH 01/13] drm/i915: Add a sw fence for collecting up dma fences Chris Wilson
2016-08-25 10:42 ` John Harrison
2016-08-25 10:51 ` Chris Wilson
2016-08-25 11:49 ` Joonas Lahtinen
2016-08-25 12:25 ` Chris Wilson
2016-08-25 16:52 ` Chris Wilson
2016-08-25 9:08 ` [PATCH 02/13] drm/i915: Only queue requests during execlists submission Chris Wilson
2016-08-26 9:41 ` Mika Kuoppala
2016-08-26 9:51 ` Chris Wilson
2016-08-26 9:54 ` Chris Wilson
2016-08-26 10:43 ` Mika Kuoppala
2016-08-26 10:53 ` Chris Wilson
2016-08-25 9:08 ` [PATCH 03/13] drm/i915: Record the position of the workarounds in the tail of the request Chris Wilson
2016-08-25 9:35 ` Mika Kuoppala
2016-08-25 9:08 ` [PATCH 04/13] drm/i915: Compute the ELSP register location once Chris Wilson
2016-08-25 9:51 ` Mika Kuoppala
2016-08-25 10:37 ` Chris Wilson
2016-08-25 9:08 ` [PATCH 05/13] drm/i915: Reorder submitting the requests to ELSP Chris Wilson
2016-08-25 13:02 ` Mika Kuoppala
2016-08-25 9:08 ` [PATCH 06/13] drm/i915: Simplify ELSP queue request tracking Chris Wilson
2016-08-25 9:08 ` [PATCH 07/13] drm/i915: Update reset path to fix incomplete requests Chris Wilson
2016-08-25 9:08 ` [PATCH 08/13] drm/i915: Drive request submission through fence callbacks Chris Wilson
2016-08-25 12:08 ` Joonas Lahtinen
2016-08-25 12:35 ` Chris Wilson
2016-08-26 12:47 ` John Harrison
2016-08-26 16:20 ` Chris Wilson [this message]
2016-08-25 9:08 ` [PATCH 09/13] drm/i915: Move execbuf object synchronisation to i915_gem_execbuffer Chris Wilson
2016-08-26 13:29 ` John Harrison
2016-08-25 9:08 ` [PATCH 10/13] drm/i915: Nonblocking request submission Chris Wilson
2016-08-26 13:39 ` John Harrison
2016-08-26 16:14 ` Chris Wilson
2016-08-25 9:08 ` [PATCH 11/13] drm/i915: Serialise execbuf operation after a dma-buf reservation object Chris Wilson
2016-08-26 13:52 ` John Harrison
2016-08-25 9:08 ` [PATCH 12/13] drm/i915: Enable userspace to opt-out of implicit fencing Chris Wilson
2016-08-25 9:08 ` [PATCH 13/13] drm/i915: Support explicit fencing for execbuf Chris Wilson
2016-08-26 15:08 ` John Harrison
2016-08-26 15:29 ` John Harrison
2016-08-26 15:44 ` Chris Wilson
2016-08-25 9:08 ` [PATCH libdrm 14/15] intel: Allow the client to control implicit synchronisation Chris Wilson
2016-08-25 9:08 ` [PATCH libdrm 15/15] intel: Support passing of explicit fencing from execbuf Chris Wilson
2016-09-30 20:53 ` Rafael Antognolli
2016-10-01 8:36 ` Chris Wilson
2016-08-25 9:08 ` [PATCH 16/21] i965: Add explicit fence tracking to batch flush Chris Wilson
2016-08-25 9:08 ` [PATCH 17/21] i965: Split intel_syncobject into vfuncs Chris Wilson
2016-08-25 9:08 ` [PATCH 18/21] i965: Add fd-fence backend to intel_syncobject Chris Wilson
2016-08-25 9:08 ` [PATCH 19/21] rfc! i965: Add intel_screen::has_fence_fd Chris Wilson
2016-08-25 9:08 ` [PATCH 20/21] i965: Implement EGL_ANDROID_native_fence_sync support for DRI2_FENCE Chris Wilson
2016-08-25 9:08 ` [PATCH 21/21] i965: Disable implicit sync when using EGL_ANDROID_native_fence_sync Chris Wilson
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=20160826162056.GK27196@nuc-i3427.alporthouse.com \
--to=chris@chris-wilson.co.uk \
--cc=John.C.Harrison@Intel.com \
--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 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.