[Intel-gfx] [PATCH 2/2] drm/i915: Skip the Wa_1604555607 verification

[Tvrtko Ursulin]

On 20/11/2019 17:31, Ramalingam C wrote:
> At TGL A0 stepping, FF_MODE2 register read back is broken, hence
> disabling the WA verification.
> 
> Helper function called wa_write_masked_or_no_verify is defined for the
> same purpose.
> 
> Signed-off-by: Ramalingam C <ramalingam.c at intel.com>
> cc: Tvrtko Ursulin <tvrtko.ursulin at linux.intel.com>
> ---
>   drivers/gpu/drm/i915/gt/intel_workarounds.c | 20 +++++++++++++++++++-
>   1 file changed, 19 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c
> index 93efefa205d6..1698330c6f23 100644
> --- a/drivers/gpu/drm/i915/gt/intel_workarounds.c
> +++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c
> @@ -160,6 +160,20 @@ wa_write_masked_or(struct i915_wa_list *wal, i915_reg_t reg, u32 mask,
>   	_wa_add(wal, &wa);
>   }
>   
> +static void
> +wa_write_masked_or_no_verify(struct i915_wa_list *wal, i915_reg_t reg, u32 mask,
> +		   u32 val)
> +{
> +	struct i915_wa wa = {
> +		.reg  = reg,
> +		.mask = mask,
> +		.val  = val,
> +		.read = 0,
> +	};
> +
> +	_wa_add(wal, &wa);
> +}
> +
>   static void
>   wa_masked_en(struct i915_wa_list *wal, i915_reg_t reg, u32 val)
>   {
> @@ -578,7 +592,11 @@ static void tgl_ctx_workarounds_init(struct intel_engine_cs *engine,
>   	val = intel_uncore_read(engine->uncore, FF_MODE2);
>   	val &= ~FF_MODE2_TDS_TIMER_MASK;
>   	val |= FF_MODE2_TDS_TIMER_128;
> -	wa_write_masked_or(wal, FF_MODE2, FF_MODE2_TDS_TIMER_MASK, val);
> +	if (IS_TGL_REVID(engine->uncore->i915, 0, TGL_REVID_A0))

There is engine->i915.

> +		wa_write_masked_or_no_verify(wal, FF_MODE2,
> +					     FF_MODE2_TDS_TIMER_MASK, val);
> +	else
> +		wa_write_masked_or(wal, FF_MODE2, FF_MODE2_TDS_TIMER_MASK, val);

You did not think suggested alternative where read mask is directly 
passed in would work better? It would read a bit more compact:

   __wa_write_masked_or(..., IS_TGL_REVID(..) ? 0 : val)

Up to you what you prefer, I guess the explicit _no_verify brings some 
self-documenting benefits.

Also, do you think there is value in having two patches instead of just 
squashing into one? I would be tempted to squash.

Regards,

Tvrtko

>   }
>   
>   static void
>