summaryrefslogtreecommitdiffstats
path: root/src/amd
diff options
context:
space:
mode:
authorTimothy Arceri <[email protected]>2017-10-19 09:27:04 +1100
committerTimothy Arceri <[email protected]>2017-10-20 08:01:26 +1100
commit087e010b2b3dd83a539f97203909d6c43b5da87c (patch)
tree6eba6aa693c41c498107f2d01dcc9895eaf994eb /src/amd
parent5549b47d7b373ab1ded2ffd31ebbc77b876fa901 (diff)
radv: copy indirect lowering settings from radeonsi
It looks the original indirect mask was probably copied from ANV. Sascha Willems demo results: tessellation ~4000 -> ~4200 fps V2: continue lowering local indirects due to llvm deficiencies. Tested-by: Alex Smith <[email protected]> Reviewed-by: Bas Nieuwenhuizen <[email protected]>
Diffstat (limited to 'src/amd')
-rw-r--r--src/amd/vulkan/radv_shader.c27
1 files changed, 26 insertions, 1 deletions
diff --git a/src/amd/vulkan/radv_shader.c b/src/amd/vulkan/radv_shader.c
index a86ba2a01c0..3688680107f 100644
--- a/src/amd/vulkan/radv_shader.c
+++ b/src/amd/vulkan/radv_shader.c
@@ -245,8 +245,33 @@ radv_shader_compile_to_nir(struct radv_device *device,
nir_shader_gather_info(nir, entry_point->impl);
+ /* While it would be nice not to have this flag, we are constrained
+ * by the reality that LLVM 5.0 doesn't have working VGPR indexing
+ * on GFX9.
+ */
+ bool llvm_has_working_vgpr_indexing =
+ device->physical_device->rad_info.chip_class <= VI;
+
+ /* TODO: Indirect indexing of GS inputs is unimplemented.
+ *
+ * TCS and TES load inputs directly from LDS or offchip memory, so
+ * indirect indexing is trivial.
+ */
nir_variable_mode indirect_mask = 0;
- indirect_mask |= nir_var_shader_in;
+ if (nir->stage == MESA_SHADER_GEOMETRY ||
+ (nir->stage != MESA_SHADER_TESS_CTRL &&
+ nir->stage != MESA_SHADER_TESS_EVAL &&
+ !llvm_has_working_vgpr_indexing)) {
+ indirect_mask |= nir_var_shader_in;
+ }
+
+ /* TODO: We shouldn't need to do this, however LLVM isn't currently
+ * smart enough to handle indirects without causing excess spilling
+ * causing the gpu to hang.
+ *
+ * See the following thread for more details of the problem:
+ * https://lists.freedesktop.org/archives/mesa-dev/2017-July/162106.html
+ */
indirect_mask |= nir_var_local;
nir_lower_indirect_derefs(nir, indirect_mask);