[Mesa-dev] [PATCH 3/3] radeonsi: use #pragma pack to pack si_shader_key

[Marek Olšák]

From: Marek Olšák <marek.olsak at amd.com>

sizeof(struct si_shader_key):
  Before reverting the 2 commits: 120 bytes
  After reverting the 2 commits: 128 bytes
  With #pragma pack: 107 bytes

I'm not sure if memcmp with a byte-aligned size is a good idea.
---
 src/gallium/drivers/radeonsi/si_shader.h | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/src/gallium/drivers/radeonsi/si_shader.h b/src/gallium/drivers/radeonsi/si_shader.h
index 185bb8d..6432126 100644
--- a/src/gallium/drivers/radeonsi/si_shader.h
+++ b/src/gallium/drivers/radeonsi/si_shader.h
@@ -371,20 +371,25 @@ struct si_shader_selector {
  * GFX6 - with GS:   ES |     |     | GS | VS | PS
  *      - with tess: LS | HS  | VS  |    |    | PS
  *      - with both: LS | HS  | ES  | GS | VS | PS
  * GFX9 - with GS:   -> |     |     | GS | VS | PS
  *      - with tess: -> | HS  | VS  |    |    | PS
  *      - with both: -> | HS  | ->  | GS | VS | PS
  *
  * -> = merged with the next stage
  */
 
+/* Use the byte alignment for all following structure members for optimal
+ * shader key memory footprint.
+ */
+#pragma pack(push, 1)
+
 /* Common VS bits between the shader key and the prolog key. */
 struct si_vs_prolog_bits {
 	unsigned	instance_divisors[SI_MAX_ATTRIBS];
 };
 
 /* Common TCS bits between the shader key and the epilog key. */
 struct si_tcs_epilog_bits {
 	unsigned	prim_mode:3;
 	unsigned	tes_reads_tess_factors:1;
 };
@@ -507,20 +512,23 @@ struct si_shader_key {
 		/* For shaders where monolithic variants have better code.
 		 *
 		 * This is a flag that has no effect on code generation,
 		 * but forces monolithic shaders to be used as soon as
 		 * possible, because it's in the "opt" group.
 		 */
 		unsigned	prefer_mono:1;
 	} opt;
 };
 
+/* Restore the pack alignment to default. */
+#pragma pack(pop)
+
 struct si_shader_config {
 	unsigned			num_sgprs;
 	unsigned			num_vgprs;
 	unsigned			spilled_sgprs;
 	unsigned			spilled_vgprs;
 	unsigned			private_mem_vgprs;
 	unsigned			lds_size;
 	unsigned			spi_ps_input_ena;
 	unsigned			spi_ps_input_addr;
 	unsigned			float_mode;
-- 
2.7.4