i965: implement ARB_pipeline_statistics_query

NOTE: The implementation was initially one patch, this. All the history is kept
here, even though all the core mesa changes were moved to the parent of this
patch.

This patch implements ARB_pipeline_statistics_query. This addition to GL does
not add a new API. Instead, it adds new tokens to the existing query APIs. The
work to hook up the new tokens is trivial due to it's similarity to the previous
work done for the query APIs. I've implemented all the new tokens to some
degree, but have stubbed out the untested ones at the entry point for Begin().
Doing this should allow the remainder of the code to be left in.

The new tokens give GL clients a way to obtain stats about the GL pipeline.
Generally, you get the number of things going in, invocations, and number of
things coming out, primitives, of the various stages. There are two immediate
uses for this, performance information, and debugging various types of
misrendering. I doubt one can use these for debugging very complex applications,
but for piglit tests, it should be quite useful.

Tessellation shaders, and compute shaders are not addressed in this patch
because there is no upstream implementation. I've implemented how I believe
tessellation shader stats will work for Intel hardware (though there is a bit of
ambiguity). Compute shaders are a bit more interesting though, and I don't yet
know what we'll do there.

For the lazy, here is a link to the relevant part of the spec:
https://www.opengl.org/registry/specs/ARB/pipeline_statistics_query.txt

Running the piglit tests
http://lists.freedesktop.org/archives/piglit/2014-November/013321.html
(http://cgit.freedesktop.org/~bwidawsk/piglit/log/?h=pipe_stats)
yield the following results:

> piglit-run.py -t stats tests/all.py output/pipeline_stats
> [5/5] pass: 5 Running Test(s): 5

v2:
- Don't allow pipeline_stats to be per stream (Ilia). This may (not sure) be
  needed for AMD_transform_feedback4, which we do not support.
   > If AMD_transform_feedback4 is supported then GEOMETRY_SHADER_PRIMITIVES_-
   > EMITTED_ARB counts primitives emitted to any of the vertex streams for
   > which STREAM_RASTERIZATION_AMD is enabled.
- Remove comment from GL3.txt because it is only used for extensions that are
  part of required versions (Ilia)
- Move the new tokens to a new XML doc instead of using the main GL4x.xml (Ilia)
- Add a fallthrough comment (Ilia)
- Only divide PS invocations by 4 on HSW+ (Ben)

v3:
- Add ARB_pipeline_statistics_query to relnotes.html
- Add ARB_pipeline_statistics_query.xml to the Makefile.am, and master XML (Ilia)
- Correct extension number (Ilia)
- Add link to xml in the main GL API xml (Ilia)
- remove special GS case from gen6_end_query (Ian)
- Make lookup table static so gcc doesn't initialized it on every call (Ian)
- Use if (_mesa_has_geometry_shaders(ctx)) instead of explicit checks (Ian)
- Core mesa parts moved into a prep patch (Ilia)

v4:
- Change to 10.6 relnotes since we missed 10.5 window
- Moved compute shader stuff into the switch statement (Jordan)
- Jordan: Add compute shader support

v5:
- Fixed relnote style (Ilia)

v6:
- Rebased on master which beat me to adding the first relnotes - essentially
  this undoes v5 (which had a typo anyway)
- Some code style fixes (Ken)
- Remove some excess comments (Ken)
- Unify tessellation failure style - unreachable (Ken)
- Fix workaround comment for PS invocations (Ken)

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

1 files changed, 104 insertions, 0 deletions

diff --git a/src/mesa/drivers/dri/i965/gen6_queryobj.c b/src/mesa/drivers/dri/i965/gen6_queryobj.c
index de71bb5..09b66ca 100644
--- a/src/mesa/drivers/dri/i965/gen6_queryobj.c
+++ b/src/mesa/drivers/dri/i965/gen6_queryobj.c
@@ -109,6 +109,55 @@ write_xfb_primitives_written(struct brw_context *brw,
    }
 }
 
+static inline const int
+pipeline_target_to_index(int target)
+{
+   if (target == GL_GEOMETRY_SHADER_INVOCATIONS)
+      return MAX_PIPELINE_STATISTICS - 1;
+   else
+      return target - GL_VERTICES_SUBMITTED_ARB;
+}
+
+static void
+emit_pipeline_stat(struct brw_context *brw, drm_intel_bo *bo,
+                   int stream, int target, int idx)
+{
+   /* One source of confusion is the tessellation shader statistics. The
+    * hardware has no statistics specific to the TE unit. Ideally we could have
+    * the HS primitives for TESS_CONTROL_SHADER_PATCHES_ARB, and the DS
+    * invocations as the register for TESS_CONTROL_SHADER_PATCHES_ARB.
+    * Unfortunately we don't have HS primitives, we only have HS invocations.
+    */
+
+   /* Everything except GEOMETRY_SHADER_INVOCATIONS can be kept in a simple
+    * lookup table
+    */
+   static const uint32_t target_to_register[] = {
+      IA_VERTICES_COUNT,   /* VERTICES_SUBMITTED */
+      IA_PRIMITIVES_COUNT, /* PRIMITIVES_SUBMITTED */
+      VS_INVOCATION_COUNT, /* VERTEX_SHADER_INVOCATIONS */
+      0, /* HS_INVOCATION_COUNT,*/  /* TESS_CONTROL_SHADER_PATCHES */
+      0, /* DS_INVOCATION_COUNT,*/  /* TESS_EVALUATION_SHADER_INVOCATIONS */
+      GS_PRIMITIVES_COUNT, /* GEOMETRY_SHADER_PRIMITIVES_EMITTED */
+      PS_INVOCATION_COUNT, /* FRAGMENT_SHADER_INVOCATIONS */
+      CS_INVOCATION_COUNT, /* COMPUTE_SHADER_INVOCATIONS */
+      CL_INVOCATION_COUNT, /* CLIPPING_INPUT_PRIMITIVES */
+      CL_PRIMITIVES_COUNT, /* CLIPPING_OUTPUT_PRIMITIVES */
+      GS_INVOCATION_COUNT /* This one is special... */
+   };
+   STATIC_ASSERT(ARRAY_SIZE(target_to_register) == MAX_PIPELINE_STATISTICS);
+   uint32_t reg = target_to_register[pipeline_target_to_index(target)];
+   assert(reg != 0);
+
+   /* Emit a flush to make sure various parts of the pipeline are complete and
+    * we get an accurate value
+    */
+   intel_batchbuffer_emit_mi_flush(brw);
+
+   brw_store_register_mem64(brw, bo, reg, idx);
+}
+
+
 /**
  * Wait on the query object's BO and calculate the final result.
  */
@@ -169,9 +218,35 @@ gen6_queryobj_get_results(struct gl_context *ctx,
 
    case GL_PRIMITIVES_GENERATED:
    case GL_TRANSFORM_FEEDBACK_PRIMITIVES_WRITTEN:
+   case GL_VERTICES_SUBMITTED_ARB:
+   case GL_PRIMITIVES_SUBMITTED_ARB:
+   case GL_VERTEX_SHADER_INVOCATIONS_ARB:
+   case GL_GEOMETRY_SHADER_INVOCATIONS:
+   case GL_GEOMETRY_SHADER_PRIMITIVES_EMITTED_ARB:
+   case GL_CLIPPING_INPUT_PRIMITIVES_ARB:
+   case GL_CLIPPING_OUTPUT_PRIMITIVES_ARB:
+   case GL_COMPUTE_SHADER_INVOCATIONS_ARB:
       query->Base.Result = results[1] - results[0];
       break;
 
+   case GL_FRAGMENT_SHADER_INVOCATIONS_ARB:
+      query->Base.Result = (results[1] - results[0]);
+      /* Implement the "WaDividePSInvocationCountBy4:HSW,BDW" workaround:
+       * "Invocation counter is 4 times actual.  WA: SW to divide HW reported
+       *  PS Invocations value by 4."
+       *
+       * Prior to Haswell, invocation count was counted by the WM, and it
+       * buggily counted invocations in units of subspans (2x2 unit). To get the
+       * correct value, the CS multiplied this by 4. With HSW the logic moved,
+       * and correctly emitted the number of pixel shader invocations, but,
+       * whomever forgot to undo the multiply by 4.
+       */
+      if (brw->gen >= 8 || brw->is_haswell)
+         query->Base.Result /= 4;
+      break;
+
+   case GL_TESS_CONTROL_SHADER_PATCHES_ARB:
+   case GL_TESS_EVALUATION_SHADER_INVOCATIONS_ARB:
    default:
       unreachable("Unrecognized query target in brw_queryobj_get_results()");
    }
@@ -240,6 +315,20 @@ gen6_begin_query(struct gl_context *ctx, struct gl_query_object *q)
       write_xfb_primitives_written(brw, query->bo, query->Base.Stream, 0);
       break;
 
+   case GL_VERTICES_SUBMITTED_ARB:
+   case GL_PRIMITIVES_SUBMITTED_ARB:
+   case GL_VERTEX_SHADER_INVOCATIONS_ARB:
+   case GL_GEOMETRY_SHADER_INVOCATIONS:
+   case GL_GEOMETRY_SHADER_PRIMITIVES_EMITTED_ARB:
+   case GL_FRAGMENT_SHADER_INVOCATIONS_ARB:
+   case GL_CLIPPING_INPUT_PRIMITIVES_ARB:
+   case GL_CLIPPING_OUTPUT_PRIMITIVES_ARB:
+   case GL_COMPUTE_SHADER_INVOCATIONS_ARB:
+      emit_pipeline_stat(brw, query->bo, query->Base.Stream, query->Base.Target, 0);
+      break;
+
+   case GL_TESS_CONTROL_SHADER_PATCHES_ARB:
+   case GL_TESS_EVALUATION_SHADER_INVOCATIONS_ARB:
    default:
       unreachable("Unrecognized query target in brw_begin_query()");
    }
@@ -278,6 +367,21 @@ gen6_end_query(struct gl_context *ctx, struct gl_query_object *q)
       write_xfb_primitives_written(brw, query->bo, query->Base.Stream, 1);
       break;
 
+   case GL_VERTICES_SUBMITTED_ARB:
+   case GL_PRIMITIVES_SUBMITTED_ARB:
+   case GL_VERTEX_SHADER_INVOCATIONS_ARB:
+   case GL_GEOMETRY_SHADER_PRIMITIVES_EMITTED_ARB:
+   case GL_FRAGMENT_SHADER_INVOCATIONS_ARB:
+   case GL_COMPUTE_SHADER_INVOCATIONS_ARB:
+   case GL_CLIPPING_INPUT_PRIMITIVES_ARB:
+   case GL_CLIPPING_OUTPUT_PRIMITIVES_ARB:
+   case GL_GEOMETRY_SHADER_INVOCATIONS:
+      emit_pipeline_stat(brw, query->bo,
+                         query->Base.Stream, query->Base.Target, 1);
+      break;
+
+   case GL_TESS_CONTROL_SHADER_PATCHES_ARB:
+   case GL_TESS_EVALUATION_SHADER_INVOCATIONS_ARB:
    default:
       unreachable("Unrecognized query target in brw_end_query()");
    }