[Intel-gfx] [PATCH 06/12] drm/i915: Populate ctx ID for periodic OA reports

[Lionel Landwerlin]

Hi Sagar,

I'm curious to what happens if 2 contexts submit requests which a time 
period smaller than the sampling OA period on Gen7.5.
My understanding is that with this change you'll only retain the last 
submission and then the ctx_id reported in the SAMPLE_CTX_ID field will 
be incorrect for the first workload.

Am I missing something?

-
Lionel

On 31/07/17 08:59, Sagar Arun Kamble wrote:
> From: Sourab Gupta <sourab.gupta at intel.com>
>
> This adds support for populating the ctx id for the periodic OA reports
> when requested through the corresponding property.
>
> For Gen8, the OA reports itself have the ctx ID and it is the one
> programmed into HW while submitting workloads. Thus it's retrieved from
> reports itself.
> For Gen7, the OA reports don't have any such field, and we can populate
> this field with the last seen ctx ID while sending CS reports.
>
> Signed-off-by: Sourab Gupta <sourab.gupta at intel.com>
> Signed-off-by: Sagar Arun Kamble <sagar.a.kamble at intel.com>
> ---
>   drivers/gpu/drm/i915/i915_drv.h  |  8 ++++++
>   drivers/gpu/drm/i915/i915_perf.c | 58 +++++++++++++++++++++++++++++++---------
>   2 files changed, 54 insertions(+), 12 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
> index fb81315..6c011f3 100644
> --- a/drivers/gpu/drm/i915/i915_drv.h
> +++ b/drivers/gpu/drm/i915/i915_drv.h
> @@ -2077,6 +2077,8 @@ struct i915_perf_stream {
>   
>   	wait_queue_head_t poll_wq;
>   	bool pollin;
> +
> +	u32 last_ctx_id;
>   };
>   
>   /**
> @@ -2151,6 +2153,12 @@ struct i915_oa_ops {
>   	 * generations.
>   	 */
>   	u32 (*oa_hw_tail_read)(struct drm_i915_private *dev_priv);
> +
> +	/**
> +	 * @get_ctx_id: Retrieve the ctx_id associated with the (periodic) OA
> +	 * report.
> +	 */
> +	u32 (*get_ctx_id)(struct i915_perf_stream *stream, const u8 *report);
>   };
>   
>   /*
> diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c
> index 905c5bb..1f5ebdb 100644
> --- a/drivers/gpu/drm/i915/i915_perf.c
> +++ b/drivers/gpu/drm/i915/i915_perf.c
> @@ -790,6 +790,45 @@ static u32 oa_buffer_num_reports_unlocked(
>   	return aged_tail == INVALID_TAIL_PTR ? 0 : num_reports;
>   }
>   
> +static u32 gen7_oa_buffer_get_ctx_id(struct i915_perf_stream *stream,
> +				    const u8 *report)
> +{
> +	if (!stream->cs_mode)
> +		WARN_ONCE(1,
> +			"CTX ID can't be retrieved if command stream mode not enabled");
> +
> +	/*
> +	 * OA reports generated in Gen7 don't have the ctx ID information.
> +	 * Therefore, just rely on the ctx ID information from the last CS
> +	 * sample forwarded
> +	 */
> +	return stream->last_ctx_id;
> +}
> +
> +static u32 gen8_oa_buffer_get_ctx_id(struct i915_perf_stream *stream,
> +				    const u8 *report)
> +{
> +	u32 ctx_id;
> +
> +	/* The ctx ID present in the OA reports have intel_context::hw_id
> +	 * present, since this is programmed into the ELSP in execlist mode.
> +	 * In non-execlist mode, fall back to retrieving the ctx ID from the
> +	 * last saved ctx ID from command stream mode.
> +	 */
> +	if (i915.enable_execlists) {
> +		u32 *report32 = (void *)report;
> +
> +		ctx_id = report32[2] & 0x1fffff;
> +	} else {
> +		if (!stream->cs_mode)
> +			WARN_ONCE(1,
> +				"CTX ID can't be retrieved if command stream mode not enabled");
> +
> +		ctx_id = stream->last_ctx_id;
> +	}
> +	return ctx_id;
> +}
> +
>   /**
>    * append_oa_status - Appends a status record to a userspace read() buffer.
>    * @stream: An i915-perf stream opened for OA metrics
> @@ -914,22 +953,12 @@ static int append_oa_buffer_sample(struct i915_perf_stream *stream,
>   	struct drm_i915_private *dev_priv = stream->dev_priv;
>   	u32 sample_flags = stream->sample_flags;
>   	struct i915_perf_sample_data data = { 0 };
> -	u32 *report32 = (u32 *)report;
>   
>   	if (sample_flags & SAMPLE_OA_SOURCE)
>   		data.source = I915_PERF_SAMPLE_OA_SOURCE_OABUFFER;
>   
>   	if (sample_flags & SAMPLE_CTX_ID) {
> -		if (INTEL_INFO(dev_priv)->gen < 8)
> -			data.ctx_id = 0;
> -		else {
> -			/*
> -			 * XXX: Just keep the lower 21 bits for now since I'm
> -			 * not entirely sure if the HW touches any of the higher
> -			 * bits in this field
> -			 */
> -			data.ctx_id = report32[2] & 0x1fffff;
> -		}
> +		data.ctx_id = dev_priv->perf.oa.ops.get_ctx_id(stream, report);
>   	}
>   
>   	if (sample_flags & SAMPLE_OA_REPORT)
> @@ -1524,8 +1553,10 @@ static int append_cs_buffer_sample(struct i915_perf_stream *stream,
>   	if (sample_flags & SAMPLE_OA_SOURCE)
>   		data.source = I915_PERF_SAMPLE_OA_SOURCE_CS;
>   
> -	if (sample_flags & SAMPLE_CTX_ID)
> +	if (sample_flags & SAMPLE_CTX_ID) {
>   		data.ctx_id = node->ctx_id;
> +		stream->last_ctx_id = data.ctx_id;
> +	}
>   
>   	return append_perf_sample(stream, buf, count, offset, &data);
>   }
> @@ -3838,6 +3869,7 @@ void i915_perf_init(struct drm_i915_private *dev_priv)
>   		dev_priv->perf.oa.ops.read = gen7_oa_read;
>   		dev_priv->perf.oa.ops.oa_hw_tail_read =
>   			gen7_oa_hw_tail_read;
> +		dev_priv->perf.oa.ops.get_ctx_id = gen7_oa_buffer_get_ctx_id;
>   
>   		dev_priv->perf.oa.timestamp_frequency = 12500000;
>   
> @@ -3933,6 +3965,8 @@ void i915_perf_init(struct drm_i915_private *dev_priv)
>   			dev_priv->perf.oa.ops.read = gen8_oa_read;
>   			dev_priv->perf.oa.ops.oa_hw_tail_read =
>   				gen8_oa_hw_tail_read;
> +			dev_priv->perf.oa.ops.get_ctx_id =
> +				gen8_oa_buffer_get_ctx_id;
>   
>   			dev_priv->perf.oa.oa_formats = gen8_plus_oa_formats;
>   		}