[Intel-gfx] [PATCH v3 7/8] drm/i915: Create vGPU specific write MMIO to reduce traps

[Tvrtko Ursulin]

On 11/13/2014 12:02 PM, Yu Zhang wrote:
> In the virtualized environment, forcewake operations are not
> necessory for the driver, because mmio accesses will be trapped

necessary

> and emulated by the host side, and real forcewake operations are
> also done in the host. New mmio write handlers are added to directly
> call the __raw_i915_write, therefore will reduce many traps and
> increase the overall performance for drivers running in the VM
> with Intel GVT-g enhancement.
>
> v2:
> take Chris' comments:
>          - register the mmio hooks in intel_uncore_init()
> v3:
> take Daniel's comments:
>          - use macros to assign mmio write functions for vGPU
>
> Signed-off-by: Yu Zhang <yu.c.zhang at linux.intel.com>
> Signed-off-by: Jike Song <jike.song at intel.com>
> Signed-off-by: Kevin Tian <kevin.tian at intel.com>
> ---
>   drivers/gpu/drm/i915/intel_uncore.c | 16 ++++++++++++++++
>   1 file changed, 16 insertions(+)
>
> diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
> index cae27bb..b76c21d 100644
> --- a/drivers/gpu/drm/i915/intel_uncore.c
> +++ b/drivers/gpu/drm/i915/intel_uncore.c
> @@ -727,6 +727,14 @@ hsw_write##x(struct drm_i915_private *dev_priv, off_t reg, u##x val, bool trace)
>   	REG_WRITE_FOOTER; \
>   }
>
> +#define __vgpu_write(x) \
> +static void \
> +vgpu_write##x(struct drm_i915_private *dev_priv, off_t reg, u##x val, bool trace) { \
> +	REG_WRITE_HEADER; \
> +	__raw_i915_write##x(dev_priv, reg, val); \
> +	REG_WRITE_FOOTER; \
> +}
> +
>   static const u32 gen8_shadowed_regs[] = {
>   	FORCEWAKE_MT,
>   	GEN6_RPNSWREQ,
> @@ -821,6 +829,10 @@ __gen4_write(8)
>   __gen4_write(16)
>   __gen4_write(32)
>   __gen4_write(64)
> +__vgpu_write(8)
> +__vgpu_write(16)
> +__vgpu_write(32)
> +__vgpu_write(64)
>
>   #undef __chv_write
>   #undef __gen8_write
> @@ -828,6 +840,7 @@ __gen4_write(64)
>   #undef __gen6_write
>   #undef __gen5_write
>   #undef __gen4_write
> +#undef __vgpu_write
>   #undef REG_WRITE_FOOTER
>   #undef REG_WRITE_HEADER
>
> @@ -939,6 +952,9 @@ void intel_uncore_init(struct drm_device *dev)
>   		break;
>   	}
>
> +	if (intel_vgpu_active(dev))
> +		ASSIGN_WRITE_MMIO_VFUNCS(vgpu);
> +
>   	i915_check_and_clear_faults(dev);
>   }
>   #undef ASSIGN_WRITE_MMIO_VFUNCS
>

Maybe I am missing something obvious, but why are read variants not needed?

Regards,

Tvrtko