Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Mika Kuoppala <mika.kuoppala@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>, intel-gfx@lists.freedesktop.org
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Subject: Re: [Intel-gfx] [PATCH 4/6] drm/i915/gt: Move snb GT workarounds from init_clock_gating to workarounds
Date: Thu, 11 Jun 2020 13:04:22 +0300	[thread overview]
Message-ID: <87mu599avt.fsf@gaia.fi.intel.com> (raw)
In-Reply-To: <20200611080140.30228-4-chris@chris-wilson.co.uk>

Chris Wilson <chris@chris-wilson.co.uk> writes:

> Rescue the GT workarounds from being buried inside init_clock_gating so
> that we remember to apply them after a GT reset, and that they are
> included in our verification that the workarounds are applied.
>
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> ---
>  drivers/gpu/drm/i915/gt/intel_workarounds.c | 41 +++++++++++++++++++++
>  drivers/gpu/drm/i915/intel_pm.c             | 33 -----------------
>  2 files changed, 41 insertions(+), 33 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c
> index 688ca25d79d0..7b4f3434eb6b 100644
> --- a/drivers/gpu/drm/i915/gt/intel_workarounds.c
> +++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c
> @@ -714,6 +714,45 @@ int intel_engine_emit_ctx_wa(struct i915_request *rq)
>  	return 0;
>  }
>  
> +static void
> +snb_gt_workarounds_init(struct drm_i915_private *i915, struct i915_wa_list *wal)
> +{
> +	/* WaDisableHiZPlanesWhenMSAAEnabled:snb */
> +	wa_masked_en(wal,
> +		     _3D_CHICKEN,
> +		     _3D_CHICKEN_HIZ_PLANE_DISABLE_MSAA_4X_SNB);
> +
> +	/* WaDisable_RenderCache_OperationalFlush:snb */
> +	wa_masked_dis(wal, CACHE_MODE_0, RC_OP_FLUSH_ENABLE);
> +
> +	/*
> +	 * BSpec recoomends 8x4 when MSAA is used,

recommends.

Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>

> +	 * however in practice 16x4 seems fastest.
> +	 *
> +	 * Note that PS/WM thread counts depend on the WIZ hashing
> +	 * disable bit, which we don't touch here, but it's good
> +	 * to keep in mind (see 3DSTATE_PS and 3DSTATE_WM).
> +	 */
> +	wa_add(wal,
> +	       GEN6_GT_MODE, 0,
> +	       _MASKED_FIELD(GEN6_WIZ_HASHING_MASK, GEN6_WIZ_HASHING_16x4),
> +	       GEN6_WIZ_HASHING_16x4);
> +
> +	wa_masked_dis(wal, CACHE_MODE_0, CM0_STC_EVICT_DISABLE_LRA_SNB);
> +
> +	wa_masked_en(wal,
> +		     _3D_CHICKEN3,
> +		     /* WaStripsFansDisableFastClipPerformanceFix:snb */
> +		     _3D_CHICKEN3_SF_DISABLE_FASTCLIP_CULL |
> +		     /*
> +		      * Bspec says:
> +		      * "This bit must be set if 3DSTATE_CLIP clip mode is set
> +		      * to normal and 3DSTATE_SF number of SF output attributes
> +		      * is more than 16."
> +		      */
> +		   _3D_CHICKEN3_SF_DISABLE_PIPELINED_ATTR_FETCH);
> +}
> +
>  static void
>  ivb_gt_workarounds_init(struct drm_i915_private *i915, struct i915_wa_list *wal)
>  {
> @@ -1154,6 +1193,8 @@ gt_init_workarounds(struct drm_i915_private *i915, struct i915_wa_list *wal)
>  		vlv_gt_workarounds_init(i915, wal);
>  	else if (IS_IVYBRIDGE(i915))
>  		ivb_gt_workarounds_init(i915, wal);
> +	else if (IS_GEN(i915, 6))
> +		snb_gt_workarounds_init(i915, wal);
>  	else if (INTEL_GEN(i915) <= 8)
>  		return;
>  	else
> diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
> index 29abde47e987..b4bea6451418 100644
> --- a/drivers/gpu/drm/i915/intel_pm.c
> +++ b/drivers/gpu/drm/i915/intel_pm.c
> @@ -6993,27 +6993,6 @@ static void gen6_init_clock_gating(struct drm_i915_private *dev_priv)
>  		   I915_READ(ILK_DISPLAY_CHICKEN2) |
>  		   ILK_ELPIN_409_SELECT);
>  
> -	/* WaDisableHiZPlanesWhenMSAAEnabled:snb */
> -	I915_WRITE(_3D_CHICKEN,
> -		   _MASKED_BIT_ENABLE(_3D_CHICKEN_HIZ_PLANE_DISABLE_MSAA_4X_SNB));
> -
> -	/* WaDisable_RenderCache_OperationalFlush:snb */
> -	I915_WRITE(CACHE_MODE_0, _MASKED_BIT_DISABLE(RC_OP_FLUSH_ENABLE));
> -
> -	/*
> -	 * BSpec recoomends 8x4 when MSAA is used,
> -	 * however in practice 16x4 seems fastest.
> -	 *
> -	 * Note that PS/WM thread counts depend on the WIZ hashing
> -	 * disable bit, which we don't touch here, but it's good
> -	 * to keep in mind (see 3DSTATE_PS and 3DSTATE_WM).
> -	 */
> -	I915_WRITE(GEN6_GT_MODE,
> -		   _MASKED_FIELD(GEN6_WIZ_HASHING_MASK, GEN6_WIZ_HASHING_16x4));
> -
> -	I915_WRITE(CACHE_MODE_0,
> -		   _MASKED_BIT_DISABLE(CM0_STC_EVICT_DISABLE_LRA_SNB));
> -
>  	I915_WRITE(GEN6_UCGCTL1,
>  		   I915_READ(GEN6_UCGCTL1) |
>  		   GEN6_BLBUNIT_CLOCK_GATE_DISABLE |
> @@ -7036,18 +7015,6 @@ static void gen6_init_clock_gating(struct drm_i915_private *dev_priv)
>  		   GEN6_RCPBUNIT_CLOCK_GATE_DISABLE |
>  		   GEN6_RCCUNIT_CLOCK_GATE_DISABLE);
>  
> -	/* WaStripsFansDisableFastClipPerformanceFix:snb */
> -	I915_WRITE(_3D_CHICKEN3,
> -		   _MASKED_BIT_ENABLE(_3D_CHICKEN3_SF_DISABLE_FASTCLIP_CULL));
> -
> -	/*
> -	 * Bspec says:
> -	 * "This bit must be set if 3DSTATE_CLIP clip mode is set to normal and
> -	 * 3DSTATE_SF number of SF output attributes is more than 16."
> -	 */
> -	I915_WRITE(_3D_CHICKEN3,
> -		   _MASKED_BIT_ENABLE(_3D_CHICKEN3_SF_DISABLE_PIPELINED_ATTR_FETCH));
> -
>  	/*
>  	 * According to the spec the following bits should be
>  	 * set in order to enable memory self-refresh and fbc:
> -- 
> 2.20.1
>
> _______________________________________________
> 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

  reply	other threads:[~2020-06-11 10:07 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-06-11  8:01 [Intel-gfx] [PATCH 1/6] drm/i915/gt: Move hsw GT workarounds from init_clock_gating to workarounds Chris Wilson
2020-06-11  8:01 ` [Intel-gfx] [PATCH 2/6] drm/i915/gt: Move ivb " Chris Wilson
2020-06-11  9:59   ` Mika Kuoppala
2020-06-11 10:00     ` Mika Kuoppala
2020-06-11 10:41   ` Chris Wilson
2020-06-11  8:01 ` [Intel-gfx] [PATCH 3/6] drm/i915/gt: Move vlv " Chris Wilson
2020-06-11 10:02   ` Mika Kuoppala
2020-06-11  8:01 ` [Intel-gfx] [PATCH 4/6] drm/i915/gt: Move snb " Chris Wilson
2020-06-11 10:04   ` Mika Kuoppala [this message]
2020-06-11  8:01 ` [Intel-gfx] [PATCH 5/6] drm/i915/gt: Move ilk " Chris Wilson
2020-06-11 10:05   ` Mika Kuoppala
2020-06-11  8:01 ` [Intel-gfx] [PATCH 6/6] drm/i915/gt: Move gen4 " Chris Wilson
2020-06-11 10:07   ` Mika Kuoppala
2020-06-11  9:06 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for series starting with [1/6] drm/i915/gt: Move hsw " Patchwork
2020-06-11  9:07 ` [Intel-gfx] ✗ Fi.CI.SPARSE: " Patchwork
2020-06-11  9:25 ` [Intel-gfx] [PATCH 1/6] " Mika Kuoppala
2020-06-11  9:28 ` [Intel-gfx] ✓ Fi.CI.BAT: success for series starting with [1/6] " Patchwork
2020-06-11  9:30 ` [Intel-gfx] [PATCH] " Chris Wilson
2020-06-11  9:52   ` Mika Kuoppala
2020-06-11  9:56 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for series starting with drm/i915/gt: Move hsw GT workarounds from init_clock_gating to workarounds (rev2) Patchwork
2020-06-11  9:58 ` [Intel-gfx] ✗ Fi.CI.SPARSE: " Patchwork
2020-06-11 10:21 ` [Intel-gfx] ✓ Fi.CI.BAT: success " Patchwork
2020-06-11 13:43 ` [Intel-gfx] ✗ Fi.CI.IGT: failure " 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=87mu599avt.fsf@gaia.fi.intel.com \
    --to=mika.kuoppala@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