KonstantinSeurer
/
mesa
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

radv: take LDS into account for compute shader occupancy stats 

Ported from d205faeb6c.

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
This commit is contained in:
Timothy Arceri 2019-02-01 22:04:39 +11:00
parent a53d68d318
commit 9b9ccee4d6
3 changed files with 14 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
src/amd/vulkan/radv_nir_to_llvm.c
View File

@ -3372,9 +3372,9 @@ ac_setup_rings(struct radv_shader_context *ctx)
}
}
static unsigned
ac_nir_get_max_workgroup_size(enum chip_class chip_class,
const struct nir_shader *nir)
unsigned
radv_nir_get_max_workgroup_size(enum chip_class chip_class,
const struct nir_shader *nir)
{
switch (nir->info.stage) {
case MESA_SHADER_TESS_CTRL:

3
src/amd/vulkan/radv_private.h
View File

@ -1934,6 +1934,9 @@ void radv_compile_nir_shader(struct ac_llvm_compiler *ac_llvm,
int nir_count,
const struct radv_nir_compiler_options *options);
unsigned radv_nir_get_max_workgroup_size(enum chip_class chip_class,
const struct nir_shader *nir);
/* radv_shader_info.h */
struct radv_shader_info;

10
src/amd/vulkan/radv_shader.c
View File

@ -733,7 +733,8 @@ generate_shader_stats(struct radv_device *device,
gl_shader_stage stage,
struct _mesa_string_buffer *buf)
{
unsigned lds_increment = device->physical_device->rad_info.chip_class >= CIK ? 512 : 256;
enum chip_class chip_class = device->physical_device->rad_info.chip_class;
unsigned lds_increment = chip_class >= CIK ? 512 : 256;
struct ac_shader_config *conf;
unsigned max_simd_waves;
unsigned lds_per_wave = 0;
@ -746,12 +747,17 @@ generate_shader_stats(struct radv_device *device,
lds_per_wave = conf->lds_size * lds_increment +
align(variant->info.fs.num_interp * 48,
lds_increment);
} else if (stage == MESA_SHADER_COMPUTE) {
unsigned max_workgroup_size =
ac_nir_get_max_workgroup_size(chip_class, variant->nir);
lds_per_wave = (conf->lds_size * lds_increment) /
DIV_ROUND_UP(max_workgroup_size, 64);
}
if (conf->num_sgprs)
max_simd_waves =
MIN2(max_simd_waves,
ac_get_num_physical_sgprs(device->physical_device->rad_info.chip_class) / conf->num_sgprs);
ac_get_num_physical_sgprs(chip_class) / conf->num_sgprs);
if (conf->num_vgprs)
max_simd_waves =