Mesa (master): radv: copy indirect lowering settings from radeonsi

Timothy Arceri tarceri at kemper.freedesktop.org
Thu Oct 19 21:03:31 UTC 2017


Module: Mesa
Branch: master
Commit: 087e010b2b3dd83a539f97203909d6c43b5da87c
URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=087e010b2b3dd83a539f97203909d6c43b5da87c

Author: Timothy Arceri <tarceri at itsqueeze.com>
Date:   Thu Oct 19 09:27:04 2017 +1100

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 <asmith at feralinteractive.com>
Reviewed-by: Bas Nieuwenhuizen <bas at basnieuwenhuizen.nl>

---

 src/amd/vulkan/radv_shader.c | 27 ++++++++++++++++++++++++++-
 1 file changed, 26 insertions(+), 1 deletion(-)

diff --git a/src/amd/vulkan/radv_shader.c b/src/amd/vulkan/radv_shader.c
index a86ba2a01c..3688680107 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);




More information about the mesa-commit mailing list