[Intel-gfx] [PATCH v3 11/22] drm/i915/cnl: Move GT and Display workarounds from init_clock_gating
Oscar Mateo
oscar.mateo at intel.com
Tue Oct 17 21:22:57 UTC 2017
On 10/17/2017 05:52 AM, Chris Wilson wrote:
> Quoting Oscar Mateo (2017-10-13 21:54:05)
>> To their rightful place inside intel_workarounds.c
>>
>> Signed-off-by: Oscar Mateo <oscar.mateo at intel.com>
>> Cc: Rodrigo Vivi <rodrigo.vivi at intel.com>
>> Cc: Chris Wilson <chris at chris-wilson.co.uk>
>> Cc: Mika Kuoppala <mika.kuoppala at linux.intel.com>
>> Cc: Ville Syrjälä <ville.syrjala at linux.intel.com>
>> ---
>> static int cnl_gt_workarounds_init(struct drm_i915_private *dev_priv)
>> {
>> + /* This is not an Wa. Enable for better image quality */
>> + GT_WA_SET_BIT_MASKED(_3D_CHICKEN3,
>> + _3D_CHICKEN3_AA_LINE_QUALITY_FIX_ENABLE);
> 3D_CHICKEN and not part of the context? I presume you checked!
>
I did, back when Rodrigo used it in Cannonlake. It seems to be an
example of masked register that does not live in the context... :S
>> +
>> /* WaDisableI2mCycleOnWRPort:cnl (pre-prod) */
>> if (IS_CNL_REVID(dev_priv, CNL_REVID_B0, CNL_REVID_B0))
>> GT_WA_SET_BIT(GAMT_CHKN_BIT_REG,
>> @@ -856,6 +860,27 @@ static int cfl_display_workarounds_init(struct drm_i915_private *dev_priv)
>>
>> static int cnl_display_workarounds_init(struct drm_i915_private *dev_priv)
>> {
>> + if (HAS_PCH_CNP(dev_priv)) {
>> + /* Wa #1181 */
>> + DISPLAY_WA_SET_BIT(SOUTH_DSPCLK_GATE_D,
>> + CNP_PWM_CGE_GATING_DISABLE);
>> + }
>> +
>> + /* WaEnableChickenDCPR:cnl */
>> + DISPLAY_WA_SET_BIT(GEN8_CHICKEN_DCPR_1, MASK_WAKEMEM);
>> +
>> + /* WaFbcWakeMemOn:cnl */
>> + DISPLAY_WA_SET_BIT(DISP_ARB_CTL, DISP_FBC_MEMORY_WAKE);
>> +
>> + /* WaSarbUnitClockGatingDisable:cnl (pre-prod) */
>> + if (IS_CNL_REVID(dev_priv, CNL_REVID_A0, CNL_REVID_B0))
>> + DISPLAY_WA_SET_BIT(SLICE_UNIT_LEVEL_CLKGATE,
>> + SARBUNIT_CLKGATE_DIS);
>> +
>> + /* Display WA #1133: WaFbcSkipSegments:cnl */
>> + DISPLAY_WA_SET_FIELD(ILK_DPFC_CHICKEN, GLK_SKIP_SEG_COUNT_MASK,
>> + GLK_SKIP_SEG_EN | GLK_SKIP_SEG_COUNT(1));
>> +
>> return 0;
>> }
> They all match up, so
> Reviewed-by: Chris Wilson <chris at chris-wilson.co.uk>
> -Chris
Thanks for the reviewes!
More information about the Intel-gfx
mailing list