From: Mika Kuoppala <mika.kuoppala@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>, intel-gfx@lists.freedesktop.org
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Subject: Re: [PATCH 35/44] drm/i915: Treat kernel context as initialised
Date: Thu, 16 Jun 2016 11:20:31 +0300 [thread overview]
Message-ID: <87porhwnn4.fsf@intel.com> (raw)
In-Reply-To: <1465993109-19523-36-git-send-email-chris@chris-wilson.co.uk>
Chris Wilson <chris@chris-wilson.co.uk> writes:
> The kernel context exists simply as a placeholder and should never be
> executed with a render context. It does not need the golden render
> state, as that will always be applied to a user context. By skipping the
> initialisation we can avoid issues in attempting to program the golden
> render context when trying to make the hardware idle.
>
> v2: Rebase
>
> Testcase: igt/drm_module_reload_basic #byt
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=95634
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
> ---
> drivers/gpu/drm/i915/i915_gem_context.c | 14 +++++++++-----
> drivers/gpu/drm/i915/intel_ringbuffer.c | 10 ++++++++++
> 2 files changed, 19 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
> index c620fe6c9d96..60967b935d72 100644
> --- a/drivers/gpu/drm/i915/i915_gem_context.c
> +++ b/drivers/gpu/drm/i915/i915_gem_context.c
> @@ -434,13 +434,17 @@ void i915_gem_context_lost(struct drm_i915_private *dev_priv)
> i915_gem_context_unpin(engine->last_context, engine);
> engine->last_context = NULL;
> }
> -
> - /* Force the GPU state to be reinitialised on enabling */
> - dev_priv->kernel_context->engine[engine->id].initialised =
> - engine->init_context == NULL;
> }
>
> - /* Force the GPU state to be reinitialised on enabling */
> + /* Force the GPU state to be restored on enabling */
> + if (!i915.enable_execlists) {
> + for_each_engine(engine, dev_priv) {
> + struct intel_context *kce =
> + &dev_priv->kernel_context->engine[engine->id];
> +
> + kce->initialised = true;
> + }
> + }
> dev_priv->kernel_context->remap_slice = ALL_L3_SLICES(dev_priv);
> }
>
> diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
> index 86b80b4727fb..04a2d141e690 100644
> --- a/drivers/gpu/drm/i915/intel_ringbuffer.c
> +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
> @@ -2338,6 +2338,16 @@ static int intel_ring_context_pin(struct i915_gem_context *ctx,
> goto error;
> }
>
> + /* The kernel context is only used as a placeholder for flushing the
> + * active context. It is never used for submitting user rendering and
> + * as such never requires the golden render context, and so we
> can skip
But the kernel context does need a golden state. As that has been the
guaranteed working context to go sleep (r6) with. Patch 42/44 will make
it so?
-Mika
> + * emitting it when we switch to the kernel context. This is required
> + * as during eviction we cannot allocate and pin the renderstate in
> + * order to initialise the context.
> + */
> + if (ctx == ctx->i915->kernel_context)
> + ce->initialised = true;
> +
> i915_gem_context_reference(ctx);
> return 0;
>
> --
> 2.8.1
>
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2016-06-16 8:20 UTC|newest]
Thread overview: 63+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-06-15 12:17 BAT bug 95634, take 4 Chris Wilson
2016-06-15 12:17 ` [PATCH 01/44] drm: Export drm_dev_init() for subclassing Chris Wilson
2016-06-15 12:17 ` [PATCH 02/44] drm: Add a callback from connector registering Chris Wilson
2016-06-17 7:41 ` Daniel Vetter
2016-06-15 12:17 ` [PATCH 03/44] drm: Make drm_connector_register() safe against multiple calls Chris Wilson
2016-06-15 12:17 ` [PATCH 04/44] drm: Automatically unregister the connector during cleanup Chris Wilson
2016-06-17 7:44 ` Daniel Vetter
2016-06-15 12:17 ` [PATCH 05/44] drm: Automatically register/unregister all connectors Chris Wilson
2016-06-15 12:17 ` Chris Wilson
2016-06-15 12:17 ` [PATCH 06/44] drm/arc: Remove redundant calls to drm_connector_register_all() Chris Wilson
2016-06-15 12:17 ` [PATCH 07/44] drm/atmel-hlcdc: " Chris Wilson
2016-06-15 12:17 ` [PATCH 08/44] drm/hisilicon: " Chris Wilson
2016-06-15 12:17 ` [PATCH 09/44] drm/mediatek: " Chris Wilson
2016-06-15 12:17 ` Chris Wilson
2016-06-15 12:17 ` [PATCH 10/44] drm/msm: " Chris Wilson
2016-06-15 12:17 ` [PATCH 11/44] drm/rcar-du: " Chris Wilson
2016-06-15 12:17 ` [PATCH 12/44] drm: Pass the drm_dp_aux->hw_mutex to i2c for its locking Chris Wilson
2016-06-15 12:17 ` [PATCH 13/44] drm: Minimally initialise drm_dp_aux Chris Wilson
2016-06-15 12:17 ` [PATCH 14/44] drm/i915: Perform async fbdev initialisation much later Chris Wilson
2016-06-15 12:18 ` [PATCH 15/44] drm/i915: Make panel/backlight safe to setup/cleanup multiple times Chris Wilson
2016-06-16 6:34 ` Jani Nikula
2016-06-16 9:00 ` Chris Wilson
2016-06-15 12:18 ` [PATCH 16/44] drm/i915: Move panel's backlight setup next to panel init Chris Wilson
2016-06-15 12:18 ` [PATCH 17/44] drm/i915: Move intel_connector->unregister to connector->early_unregister Chris Wilson
2016-06-15 12:18 ` [PATCH 18/44] drm/i915: Move backlight unregistration to connector unregistration Chris Wilson
2016-06-15 12:18 ` [PATCH 19/44] drm/i915: Move registration actions to connector->late_register Chris Wilson
2016-06-15 12:18 ` [PATCH 20/44] drm/i915/dp: Free the drm_dp_aux along with the encoder Chris Wilson
2016-06-15 12:18 ` [PATCH 21/44] drm/i915: Move backlight registration to connector registration Chris Wilson
2016-06-15 12:18 ` [PATCH 22/44] drm/i915: Move connector registration to driver registration Chris Wilson
2016-06-15 12:18 ` [PATCH 23/44] drm/i915: Register debugfs interface last Chris Wilson
2016-06-15 12:18 ` [PATCH 24/44] drm/i915: Demidlayer driver loading Chris Wilson
2016-06-15 12:18 ` [PATCH 25/44] drm/i915: Demidlayer driver unloading Chris Wilson
2016-06-15 12:18 ` [PATCH 26/44] drm/i915: Remove redundant drm_connector_register_all() Chris Wilson
2016-06-15 12:18 ` [PATCH 27/44] drm/i915: Start exploiting drm_device subclassing Chris Wilson
2016-06-15 12:18 ` [PATCH 28/44] drm/i915: Merge i915_dma.c into i915_drv.c Chris Wilson
2016-06-15 12:18 ` [PATCH 29/44] drm/i915: Remove user controllable DRM_ERROR for i915_getparam() Chris Wilson
2016-06-15 12:33 ` Tvrtko Ursulin
2016-06-15 12:18 ` [PATCH 30/44] drm/i915: Remove user controllable DRM_ERROR for intel_get_pipe_from_crtc_id() Chris Wilson
2016-06-15 12:34 ` Tvrtko Ursulin
2016-06-15 12:18 ` [PATCH 31/44] drm/i915: Split out the PCI driver interface to i915_pci.c Chris Wilson
2016-06-15 12:18 ` [PATCH 32/44] drm/i915: Move module init/exit " Chris Wilson
2016-07-12 10:45 ` Joonas Lahtinen
2016-06-15 12:18 ` [PATCH 33/44] drm/i915: Skip idling an idle engine Chris Wilson
2016-06-15 12:18 ` [PATCH 34/44] drm/i915: Move legacy kernel context pinning to intel_ringbuffer.c Chris Wilson
2016-06-15 12:18 ` [PATCH 35/44] drm/i915: Treat kernel context as initialised Chris Wilson
2016-06-16 8:20 ` Mika Kuoppala [this message]
2016-06-16 11:00 ` Chris Wilson
2016-06-15 12:18 ` [PATCH 36/44] drm/i915: Mark all default contexts as uninitialised after context loss Chris Wilson
2016-06-15 12:18 ` [PATCH 37/44] drm/i915: No need to wait for idle on L3 remap Chris Wilson
2016-06-15 12:18 ` [PATCH 38/44] drm/i915: Split idling from forcing context switch Chris Wilson
2016-06-16 8:51 ` Mika Kuoppala
2016-06-16 10:58 ` Chris Wilson
2016-06-15 12:18 ` [PATCH 39/44] drm/i915: Only switch to default context when evicting from GGTT Chris Wilson
2016-06-15 12:18 ` [PATCH 40/44] drm/i915: Preserve current RPS frequency Chris Wilson
2016-06-16 9:15 ` Mika Kuoppala
2016-06-16 10:56 ` Chris Wilson
2016-06-15 12:18 ` [PATCH 41/44] drm/i915: Remove superfluous powersave work flushing Chris Wilson
2016-06-21 15:15 ` Chris Wilson
2016-06-15 12:18 ` [PATCH 42/44] drm/i915: Defer enabling rc6 til after we submit the first batch/context Chris Wilson
2016-06-21 15:16 ` Chris Wilson
2016-06-15 12:18 ` [PATCH 43/44] drm/i915/fbdev: Limit the global async-domain synchronization Chris Wilson
2016-06-15 12:18 ` [PATCH 44/44] drm/i915/fbdev: Flush mode configuration before lastclose Chris Wilson
2016-06-15 16:09 ` ✗ Ro.CI.BAT: warning for series starting with [01/44] drm: Export drm_dev_init() for subclassing 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=87porhwnn4.fsf@intel.com \
--to=mika.kuoppala@linux.intel.com \
--cc=chris@chris-wilson.co.uk \
--cc=daniel.vetter@ffwll.ch \
--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.