[Mesa-dev] [PATCH 2/2] ac/nir: avoid loading unused VS input components

[Samuel Pitoiset]

Totals from affected shaders:
SGPRS: 142032 -> 137651 (-3.08 %)
VGPRS: 93280 -> 92992 (-0.31 %)
Spilled SGPRs: 104 -> 129 (24.04 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Private memory VGPRs: 0 -> 0 (0.00 %)
Scratch size: 0 -> 0 (0.00 %) dwords per thread
Code Size: 3309852 -> 3301444 (-0.25 %) bytes
LDS: 0 -> 0 (0.00 %) blocks
Max Waves: 22178 -> 22206 (0.13 %)
Wait states: 0 -> 0 (0.00 %)

This increases SGPRs spilling a bit with Talos, but I have
some other ideas that might reduce it.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset at gmail.com>
---

This needs to be applied after "radeonsi: load the right number of components
for VS inputs and TBOs"

 src/amd/common/ac_nir_to_llvm.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/amd/common/ac_nir_to_llvm.c b/src/amd/common/ac_nir_to_llvm.c
index 54c5f84886..e21023de0c 100644
--- a/src/amd/common/ac_nir_to_llvm.c
+++ b/src/amd/common/ac_nir_to_llvm.c
@@ -5348,6 +5348,9 @@ handle_vs_input_decl(struct nir_to_llvm_context *ctx,
 	int index = variable->data.location - VERT_ATTRIB_GENERIC0;
 	int idx = variable->data.location;
 	unsigned attrib_count = glsl_count_attribute_slots(variable->type, true);
+	uint8_t input_usage_mask =
+		ctx->shader_info->info.vs.input_usage_mask[variable->data.location];
+	unsigned num_channels = util_last_bit(input_usage_mask);
 
 	variable->data.driver_location = idx * 4;
 
@@ -5372,7 +5375,9 @@ handle_vs_input_decl(struct nir_to_llvm_context *ctx,
 		input = ac_build_buffer_load_format(&ctx->ac, t_list,
 						    buffer_index,
 						    ctx->ac.i32_0,
-						    4, true);
+						    num_channels, true);
+
+		input = ac_build_expand_to_vec4(&ctx->ac, input, num_channels);
 
 		for (unsigned chan = 0; chan < 4; chan++) {
 			LLVMValueRef llvm_chan = LLVMConstInt(ctx->ac.i32, chan, false);
-- 
2.16.1