i965: Update compute workgroup size limit calculation for SIMD32.

This should have the side effect of enabling the ARB_compute_shader
extension on Gen8+ hardware and all Gen7 platforms that didn't
previously expose it (VLV and IVB GT1) due to the number of hardware
threads per subslice being insufficient in SIMD16 mode.

v2: Bump workgroup size limit for GLES too (Jordan).

Reviewed-by: Jason Ekstrand <[email protected]>
Reviewed-by: Jordan Justen <[email protected]>

1 files changed, 3 insertions, 11 deletions

diff --git a/src/mesa/drivers/dri/i965/brw_context.c b/src/mesa/drivers/dri/i965/brw_context.c
index 552e5ec8925..2504dceb6a6 100644
--- a/src/mesa/drivers/dri/i965/brw_context.c
+++ b/src/mesa/drivers/dri/i965/brw_context.c
@@ -709,18 +709,10 @@ static void
 brw_initialize_cs_context_constants(struct brw_context *brw, unsigned max_threads)
 {
    struct gl_context *ctx = &brw->ctx;
-
-   /* For ES, we set these constants based on SIMD8.
-    *
-    * TODO: Once we can always generate SIMD16, we should update this.
-    *
-    * For GL, we assume we can generate a SIMD16 program, but this currently
-    * is not always true. This allows us to run more test cases, and will be
-    * required based on desktop GL compute shader requirements.
+   /* Maximum number of scalar compute shader invocations that can be run in
+    * parallel in the same subslice assuming SIMD32 dispatch.
     */
-   const int simd_size = ctx->API == API_OPENGL_CORE ? 16 : 8;
-
-   const uint32_t max_invocations = simd_size * max_threads;
+   const uint32_t max_invocations = 32 * max_threads;
    ctx->Const.MaxComputeWorkGroupSize[0] = max_invocations;
    ctx->Const.MaxComputeWorkGroupSize[1] = max_invocations;
    ctx->Const.MaxComputeWorkGroupSize[2] = max_invocations;