intel/compiler: Remove cs_prog_data->threads

At this point all drivers are doing this math on their own -- since
most of them need to cover the variable group size case, in which at
compile time the group size (and number of threads) is not defined.

Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4504>

diff --git a/src/intel/compiler/brw_compiler.h b/src/intel/compiler/brw_compiler.h
index 2e34b16dd44e2e2ad12facce2ddabb176c89bc9b..03512471d80c79b5e6faec859ceb069615559b12 100644 (file)
--- a/src/intel/compiler/brw_compiler.h
+++ b/src/intel/compiler/brw_compiler.h
@@ -906,7 +906,6 @@ struct brw_cs_prog_data {
    unsigned local_size[3];
    unsigned max_variable_local_size;
    unsigned simd_size;
-   unsigned threads;
    unsigned slm_size;
    bool uses_barrier;
    bool uses_num_work_groups;

diff --git a/src/intel/compiler/brw_fs.cpp b/src/intel/compiler/brw_fs.cpp
index 323fdb56ff5513abe5c4b329dfedb4a3ea038000..4e13dcca54adcd0268f2b083d2ca4d9b814df295 100644 (file)
--- a/src/intel/compiler/brw_fs.cpp
+++ b/src/intel/compiler/brw_fs.cpp
@@ -8864,22 +8864,6 @@ cs_fill_push_const_info(const struct gen_device_info *devinfo,
              prog_data->nr_params);
 }
 
-static void
-cs_set_simd_size(struct brw_cs_prog_data *cs_prog_data, unsigned size)
-{
-   cs_prog_data->simd_size = size;
-
-   unsigned group_size;
-   if (cs_prog_data->uses_variable_group_size) {
-      group_size = cs_prog_data->max_variable_local_size;
-   } else {
-      group_size = cs_prog_data->local_size[0] *
-                   cs_prog_data->local_size[1] *
-                   cs_prog_data->local_size[2];
-   }
-   cs_prog_data->threads = DIV_ROUND_UP(group_size, size);
-}
-
 static nir_shader *
 compile_cs_to_nir(const struct brw_compiler *compiler,
                   void *mem_ctx,
@@ -8972,7 +8956,7 @@ brw_compile_cs(const struct brw_compiler *compiler, void *log_data,
          assert(v8->max_dispatch_width >= 32);
 
          v = v8;
-         cs_set_simd_size(prog_data, 8);
+         prog_data->simd_size = 8;
          cs_fill_push_const_info(compiler->devinfo, prog_data);
       }
    }
@@ -9002,7 +8986,7 @@ brw_compile_cs(const struct brw_compiler *compiler, void *log_data,
          assert(v16->max_dispatch_width >= 32);
 
          v = v16;
-         cs_set_simd_size(prog_data, 16);
+         prog_data->simd_size = 16;
          cs_fill_push_const_info(compiler->devinfo, prog_data);
       }
    }
@@ -9034,7 +9018,7 @@ brw_compile_cs(const struct brw_compiler *compiler, void *log_data,
          }
       } else {
          v = v32;
-         cs_set_simd_size(prog_data, 32);
+         prog_data->simd_size = 32;
          cs_fill_push_const_info(compiler->devinfo, prog_data);
       }
    }

diff --git a/src/mesa/drivers/dri/i965/genX_state_upload.c b/src/mesa/drivers/dri/i965/genX_state_upload.c
index 53c570742793ce5c75650c99bb43e02e520aca62..ac83337b6910cd5d10336a1bb1fcb90b5fe7abb2 100644 (file)
--- a/src/mesa/drivers/dri/i965/genX_state_upload.c
+++ b/src/mesa/drivers/dri/i965/genX_state_upload.c
@@ -4267,9 +4267,6 @@ genX(upload_cs_state)(struct brw_context *brw)
    const unsigned threads =
       DIV_ROUND_UP(brw_cs_group_size(brw), cs_prog_data->simd_size);
 
-   if (!cs_prog_data->uses_variable_group_size)
-      assert(cs_prog_data->threads == threads);
-
    if (INTEL_DEBUG & DEBUG_SHADER_TIME) {
       brw_emit_buffer_surface_state(
          brw, &stage_state->surf_offset[