[Intel-gfx] [PATCH 06/23] drm/i915/kbl: Add WaEnableGapsTsvCreditFix
Arun Siluvery
arun.siluvery at linux.intel.com
Fri May 27 08:14:13 UTC 2016
On 26/05/2016 20:59, Mika Kuoppala wrote:
> We need this crucial workaround from skl also to all kbl revisions.
> Lack of it was causing system hangs on skl enabling so this is
> a must have.
>
> References: HSD#2126660
> Signed-off-by: Mika Kuoppala <mika.kuoppala at intel.com>
> ---
> drivers/gpu/drm/i915/intel_ringbuffer.c | 13 +++++++------
> 1 file changed, 7 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
> index 91d5d093f3cb..3902700d37ef 100644
> --- a/drivers/gpu/drm/i915/intel_ringbuffer.c
> +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
> @@ -1000,6 +1000,13 @@ static int gen9_init_workarounds(struct intel_engine_cs *engine)
> WA_SET_BIT_MASKED(HALF_SLICE_CHICKEN3,
> GEN8_SAMPLER_POWER_BYPASS_DIS);
>
> + /* WaEnableGapsTsvCreditFix:skl,kbl */
> + if (IS_SKL_REVID(dev_priv, SKL_REVID_C0, REVID_FOREVER) ||
> + IS_KABYLAKE(dev_priv)) {
> + I915_WRITE(GEN8_GARBCNTL, (I915_READ(GEN8_GARBCNTL) |
> + GEN9_GAPS_TSV_CREDIT_DISABLE));
> + }
Based on the feedback that I got earlier, if a WA is applicable to
different platforms and specific revisions then it was suggested to keep
them in individual functions so that we can avoid revid checks in the
common function.
regards
Arun
> +
> /* WaDisableSTUnitPowerOptimization:skl,bxt,kbl */
> WA_SET_BIT_MASKED(HALF_SLICE_CHICKEN2, GEN8_ST_PO_DISABLE);
>
> @@ -1094,12 +1101,6 @@ static int skl_init_workarounds(struct intel_engine_cs *engine)
> I915_WRITE(GEN8_L3SQCREG4, I915_READ(GEN8_L3SQCREG4) |
> GEN8_LQSC_RO_PERF_DIS);
>
> - /* WaEnableGapsTsvCreditFix:skl */
> - if (IS_SKL_REVID(dev_priv, SKL_REVID_C0, REVID_FOREVER)) {
> - I915_WRITE(GEN8_GARBCNTL, (I915_READ(GEN8_GARBCNTL) |
> - GEN9_GAPS_TSV_CREDIT_DISABLE));
> - }
> -
> /* WaDisablePowerCompilerClockGating:skl */
> if (IS_SKL_REVID(dev_priv, SKL_REVID_B0, SKL_REVID_B0))
> WA_SET_BIT_MASKED(HIZ_CHICKEN,
>
More information about the Intel-gfx
mailing list