Re: [Intel-gfx] [PATCH 02/10] drm/i915/wm: move remaining watermark code out of intel_pm.c

[Jani Nikula <jani.nikula@intel.com>]

On Wed, 08 Feb 2023, Ville Syrjälä <ville.syrjala@linux.intel.com> wrote:
> On Wed, Feb 08, 2023 at 11:48:40AM +0200, Jani Nikula wrote:
>> diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
>> index ee8f8d2d2a66..649c4d222f79 100644
>> --- a/drivers/gpu/drm/i915/intel_pm.c
>> +++ b/drivers/gpu/drm/i915/intel_pm.c
>> -
>> +static void gen9_init_clock_gating(struct drm_i915_private *dev_priv)
>> +{
>> +	if (HAS_LLC(dev_priv)) {
>>  		/*
>> -		 * Always set WM_LP_SPRITE_EN when spr_val != 0, even if the
>> -		 * level is disabled. Doing otherwise could cause underruns.
>> +		 * WaCompressedResourceDisplayNewHashMode:skl,kbl
>> +		 * Display WA #0390: skl,kbl
>> +		 *
>> +		 * Must match Sampler, Pixel Back End, and Media. See
>> +		 * WaCompressedResourceSamplerPbeMediaNewHashMode.
>>  		 */
>> -		if (DISPLAY_VER(dev_priv) <= 6 && r->spr_val) {
>> -			drm_WARN_ON(&dev_priv->drm, wm_lp != 1);
>> -			results->wm_lp_spr[wm_lp - 1] |= WM_LP_SPRITE_ENABLE;
>> -		}
>>
>
> git diff screwed up was some of the untouched stuff
> got reordered? Maybe --patience helps?

Locally --patience seems to do a nicer job, will use that for v2.

BR,
Jani.

-- 
Jani Nikula, Intel Open Source Graphics Center