[Intel-gfx] [PATCH 1/2] drm/i915/gt: Disable preparser around xcs invalidations on tgl

[Mika Kuoppala]

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

> Unlike rcs where we have conclusive evidence from our selftesting that
> disabling the preparser before performing the TLB invalidate and
> relocations does impact upon the GPU execution, the evidence for the
> same requirement on xcs is much more circumstantial. Let's apply the
> preparser disable between batches as we invalidate the TLB as a dose of
> healthy paranoia, just in case.
>
> References: https://gitlab.freedesktop.org/drm/intel/-/issues/2169
> Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> Cc: Mika Kuoppala <mika.kuoppala at linux.intel.com>

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

> ---
>  drivers/gpu/drm/i915/gt/intel_lrc.c | 15 +++++++++++++--
>  1 file changed, 13 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c
> index 29c0fde8b4df..353b1717fe84 100644
> --- a/drivers/gpu/drm/i915/gt/intel_lrc.c
> +++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
> @@ -4764,14 +4764,21 @@ static int gen12_emit_flush(struct i915_request *request, u32 mode)
>  	intel_engine_mask_t aux_inv = 0;
>  	u32 cmd, *cs;
>  
> +	cmd = 4;
> +	if (mode & EMIT_INVALIDATE)
> +		cmd += 2;
>  	if (mode & EMIT_INVALIDATE)
>  		aux_inv = request->engine->mask & ~BIT(BCS0);
> +	if (aux_inv)
> +		cmd += 2 * hweight8(aux_inv) + 2;
>  
> -	cs = intel_ring_begin(request,
> -			      4 + (aux_inv ? 2 * hweight8(aux_inv) + 2 : 0));
> +	cs = intel_ring_begin(request, cmd);
>  	if (IS_ERR(cs))
>  		return PTR_ERR(cs);
>  
> +	if (mode & EMIT_INVALIDATE)
> +		*cs++ = preparser_disable(true);
> +
>  	cmd = MI_FLUSH_DW + 1;
>  
>  	/* We always require a command barrier so that subsequent
> @@ -4804,6 +4811,10 @@ static int gen12_emit_flush(struct i915_request *request, u32 mode)
>  		}
>  		*cs++ = MI_NOOP;
>  	}
> +
> +	if (mode & EMIT_INVALIDATE)
> +		*cs++ = preparser_disable(false);
> +
>  	intel_ring_advance(request, cs);
>  
>  	return 0;
> -- 
> 2.20.1