public inbox for intel-gfx@lists.freedesktop.org
 help / color / mirror / Atom feed
From: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>, intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH 11/16] drm/i915/execlists: Request the kernel context be pinned high
Date: Fri, 9 Dec 2016 15:08:53 +0000	[thread overview]
Message-ID: <446eaa6b-8ebb-ba0e-da64-5e73fa47100c@linux.intel.com> (raw)
In-Reply-To: <20161207135833.32740-12-chris@chris-wilson.co.uk>


On 07/12/2016 13:58, Chris Wilson wrote:
> PIN_HIGH is an expensive operation (in comparison to allocating from the
> hole stack) unsuitable for frequent use (such as switching between
> contexts). However, the kernel context should be pinned just once for
> the lifetime of the driver, and here it is appropriate to keep it out of
> the mappable range (in order to maximise mappable space for users).
>
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> ---
>  drivers/gpu/drm/i915/intel_lrc.c | 8 ++++++--
>  1 file changed, 6 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
> index 58cea1f9ad27..22ded92d0346 100644
> --- a/drivers/gpu/drm/i915/intel_lrc.c
> +++ b/drivers/gpu/drm/i915/intel_lrc.c
> @@ -767,6 +767,7 @@ static int execlists_context_pin(struct intel_engine_cs *engine,
>  				 struct i915_gem_context *ctx)
>  {
>  	struct intel_context *ce = &ctx->engine[engine->id];
> +	unsigned int flags;
>  	void *vaddr;
>  	int ret;
>
> @@ -781,8 +782,11 @@ static int execlists_context_pin(struct intel_engine_cs *engine,
>  			goto err;
>  	}
>
> -	ret = i915_vma_pin(ce->state, 0, GEN8_LR_CONTEXT_ALIGN,
> -			   PIN_OFFSET_BIAS | GUC_WOPCM_TOP | PIN_GLOBAL);
> +	flags = PIN_OFFSET_BIAS | GUC_WOPCM_TOP | PIN_GLOBAL;
> +	if (ctx == ctx->i915->kernel_context)
> +		flags |= PIN_HIGH;
> +
> +	ret = i915_vma_pin(ce->state, 0, GEN8_LR_CONTEXT_ALIGN, flags);
>  	if (ret)
>  		goto err;
>
>

Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>

Regards,

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

  reply	other threads:[~2016-12-09 15:08 UTC|newest]

Thread overview: 62+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-12-07 13:58 [RFC] Smattering of selftests Chris Wilson
2016-12-07 13:58 ` [PATCH 01/16] drm/i915: Provide a hook for selftests Chris Wilson
2016-12-08 10:47   ` Tvrtko Ursulin
2016-12-08 11:15     ` Chris Wilson
2016-12-08 12:30       ` Tvrtko Ursulin
2016-12-08 12:40         ` Chris Wilson
2016-12-07 13:58 ` [PATCH 02/16] kselftests: Exercise hw-independent mock tests for i915.ko Chris Wilson
2016-12-07 14:09   ` Chris Wilson
2016-12-08 15:50     ` Shuah Khan
2016-12-08 16:01       ` Chris Wilson
2016-12-08 16:44         ` Shuah Khan
2016-12-07 13:58 ` [PATCH 03/16] drm/i915: Add unit tests for the breadcrumb rbtree, insert/remove Chris Wilson
2016-12-08 11:00   ` Tvrtko Ursulin
2016-12-07 13:58 ` [PATCH 04/16] drm/i915: Add unit tests for the breadcrumb rbtree, completion Chris Wilson
2016-12-08 11:47   ` Tvrtko Ursulin
2016-12-07 13:58 ` [PATCH 05/16] drm/i915: Add unit tests for the breadcrumb rbtree, wakeups Chris Wilson
2016-12-08 17:38   ` Tvrtko Ursulin
2016-12-08 21:04     ` Chris Wilson
2016-12-09  9:33       ` Tvrtko Ursulin
2016-12-07 13:58 ` [PATCH 06/16] drm/i915: Add a reminder that i915_vma_move_to_active() requires struct_mutex Chris Wilson
2016-12-08 17:40   ` Tvrtko Ursulin
2016-12-07 13:58 ` [PATCH 07/16] drm/i915: Move intel_lrc_context_pin() to avoid the forward declaration Chris Wilson
2016-12-08 17:45   ` Tvrtko Ursulin
2016-12-08 20:55     ` Chris Wilson
2016-12-07 13:58 ` [PATCH 08/16] drm/i915: Unify active context tracking between legacy/execlists/guc Chris Wilson
2016-12-09 11:48   ` Tvrtko Ursulin
2016-12-09 12:17     ` Chris Wilson
2016-12-07 13:58 ` [PATCH 09/16] drm/i915: Simplify releasing context reference Chris Wilson
2016-12-09 15:03   ` Tvrtko Ursulin
2016-12-07 13:58 ` [PATCH 10/16] drm/i915: Mark the shadow gvt context as closed Chris Wilson
2016-12-09 15:07   ` Tvrtko Ursulin
2016-12-07 13:58 ` [PATCH 11/16] drm/i915/execlists: Request the kernel context be pinned high Chris Wilson
2016-12-09 15:08   ` Tvrtko Ursulin [this message]
2016-12-07 13:58 ` [PATCH 12/16] drm/i915: Swap if(enable_execlists) in i915_gem_request_alloc for a vfunc Chris Wilson
2016-12-09 15:16   ` Tvrtko Ursulin
2016-12-09 15:25     ` Chris Wilson
2016-12-09 15:53       ` Tvrtko Ursulin
2016-12-07 13:58 ` [PATCH 13/16] drm/i915: Add selftests for i915_gem_request Chris Wilson
2016-12-09 15:51   ` Tvrtko Ursulin
2016-12-07 13:58 ` [PATCH 14/16] drm/i915: Add a simple request selftest for waiting Chris Wilson
2016-12-09 15:59   ` Tvrtko Ursulin
2016-12-07 13:58 ` [PATCH 15/16] drm/i915: Add a simple fence selftest to i915_gem_request Chris Wilson
2016-12-09 16:02   ` Tvrtko Ursulin
2016-12-07 13:58 ` [PATCH 16/16] drm/i915: Add selftests for object allocation, phys Chris Wilson
2016-12-13 17:10   ` Tvrtko Ursulin
2016-12-17 13:55     ` Chris Wilson
2016-12-07 14:04 ` [RFC] Smattering of selftests Chris Wilson
2016-12-07 15:45 ` ✓ Fi.CI.BAT: success for series starting with [01/16] drm/i915: Provide a hook for selftests Patchwork
2016-12-07 18:52 ` [PATCH 1/2] drm/i915: Move uncore selfchecks to late selftest infrastructure Chris Wilson
2016-12-07 18:52   ` [PATCH 2/2] drm/i915: Test all fw tables during mock selftests Chris Wilson
2016-12-08 12:14     ` Tvrtko Ursulin
2016-12-08 12:28       ` Chris Wilson
2016-12-08 13:11     ` Joonas Lahtinen
2016-12-08 16:52     ` Tvrtko Ursulin
2016-12-08 22:29       ` Chris Wilson
2016-12-09  9:41         ` Tvrtko Ursulin
2016-12-09 10:18           ` Chris Wilson
2016-12-09 10:35             ` Tvrtko Ursulin
2016-12-09 10:51               ` Chris Wilson
2016-12-08 11:58   ` [PATCH 1/2] drm/i915: Move uncore selfchecks to late selftest infrastructure Tvrtko Ursulin
2016-12-07 19:56 ` [RFC] Smattering of selftests Paulo Zanoni
2016-12-07 20:52 ` ✓ Fi.CI.BAT: success for series starting with [01/16] drm/i915: Provide a hook for selftests (rev2) 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=446eaa6b-8ebb-ba0e-da64-5e73fa47100c@linux.intel.com \
    --to=tvrtko.ursulin@linux.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