Re: [Intel-gfx] [PATCH v2 6/9] drm/i915: expose command stream timestamp frequency to userspace

On 09/11/17 14:13, Lionel Landwerlin wrote: On 09/11/17 14:06, Lionel Landwerlin wrote: +    } else if (INTEL_GEN(dev_priv) <= 9) { +    u32 ctc_reg = I915_READ(GEN8_CTC_MODE); +    u64 freq = 0; + +    if ((ctc_reg & GEN8_CTC_SOURCE_PARAMETER_MASK) ==

Re: [Intel-gfx] [PATCH v2 6/9] drm/i915: expose command stream timestamp frequency to userspace

On 11/9/2017 7:36 PM, Lionel Landwerlin wrote: On 09/11/17 11:58, Sagar Arun Kamble wrote: On 11/2/2017 9:59 PM, Lionel Landwerlin wrote: We use to have this fixed per generation, but starting with CNL userspace cannot tell just off the PCI ID. Let's make this information available. This

Re: [Intel-gfx] [PATCH v2 6/9] drm/i915: expose command stream timestamp frequency to userspace

On 09/11/17 14:06, Lionel Landwerlin wrote: +    } else if (INTEL_GEN(dev_priv) <= 9) { +    u32 ctc_reg = I915_READ(GEN8_CTC_MODE); +    u64 freq = 0; + +    if ((ctc_reg & GEN8_CTC_SOURCE_PARAMETER_MASK) == GEN8_CTC_SOURCE_DIVIDE_LOGIC) +    freq =

Re: [Intel-gfx] [PATCH v2 6/9] drm/i915: expose command stream timestamp frequency to userspace

On 09/11/17 11:58, Sagar Arun Kamble wrote: On 11/2/2017 9:59 PM, Lionel Landwerlin wrote: We use to have this fixed per generation, but starting with CNL userspace cannot tell just off the PCI ID. Let's make this information available. This is particularly useful for performance monitoring

Re: [Intel-gfx] [PATCH v2 6/9] drm/i915: expose command stream timestamp frequency to userspace

On 11/2/2017 9:59 PM, Lionel Landwerlin wrote: We use to have this fixed per generation, but starting with CNL userspace cannot tell just off the PCI ID. Let's make this information available. This is particularly useful for performance monitoring where much of the normalization work is done

Re: [Intel-gfx] [PATCH v2 6/9] drm/i915: expose command stream timestamp frequency to userspace

On 11/8/2017 11:06 PM, Lionel Landwerlin wrote: Is there anyone with spare time to review this patch? I'm on it. It's kind of required for userspace to make sense of timestamps on CNL. Thanks a lot, - Lionel On 02/11/17 16:29, Lionel Landwerlin wrote: We use to have this fixed per

Re: [Intel-gfx] [PATCH v2 6/9] drm/i915: expose command stream timestamp frequency to userspace

Is there anyone with spare time to review this patch? It's kind of required for userspace to make sense of timestamps on CNL. Thanks a lot, - Lionel On 02/11/17 16:29, Lionel Landwerlin wrote: We use to have this fixed per generation, but starting with CNL userspace cannot tell just off the

Re: [Intel-gfx] [PATCH v2 6/9] drm/i915: expose command stream timestamp frequency to userspace

On 07/11/17 11:30, Ewelina Musial wrote: On Thu, Nov 02, 2017 at 04:29:46PM +, Lionel Landwerlin wrote: We use to have this fixed per generation, but starting with CNL userspace cannot tell just off the PCI ID. Let's make this information available. This is particularly useful for

Re: [Intel-gfx] [PATCH v2 6/9] drm/i915: expose command stream timestamp frequency to userspace

On Thu, Nov 02, 2017 at 04:29:46PM +, Lionel Landwerlin wrote: > We use to have this fixed per generation, but starting with CNL userspace > cannot tell just off the PCI ID. Let's make this information available. This > is particularly useful for performance monitoring where much of the >

Re: [Intel-gfx] [PATCH v2 6/9] drm/i915: expose command stream timestamp frequency to userspace

This patch, along with the respective ones for Mesa, does fix the gl timestamp query piglit failures on CNL. So it is Tested-by: Rafael Antognolli On Thu, Nov 02, 2017 at 04:29:46PM +, Lionel Landwerlin wrote: > We use to have this fixed per generation, but

[Intel-gfx] [PATCH v2 6/9] drm/i915: expose command stream timestamp frequency to userspace

We use to have this fixed per generation, but starting with CNL userspace cannot tell just off the PCI ID. Let's make this information available. This is particularly useful for performance monitoring where much of the normalization work is done using those timestamps (this include pipeline