[Mesa-dev] [PATCH 5/5] radeonsi/nir: add missing pieces for bindless image/sampler support

[Timothy Arceri]

---
 src/gallium/drivers/radeonsi/si_shader_nir.c | 33 +++++++++++++++++++-
 1 file changed, 32 insertions(+), 1 deletion(-)

diff --git a/src/gallium/drivers/radeonsi/si_shader_nir.c b/src/gallium/drivers/radeonsi/si_shader_nir.c
index 2c95c62d99..256ef28bb1 100644
--- a/src/gallium/drivers/radeonsi/si_shader_nir.c
+++ b/src/gallium/drivers/radeonsi/si_shader_nir.c
@@ -912,6 +912,28 @@ si_nir_lookup_interp_param(struct ac_shader_abi *abi,
 		LLVMGetParam(ctx->main_fn, interp_param_idx) : NULL;
 }
 
+static LLVMValueRef
+get_bindless_index(struct ac_shader_abi *abi,
+		   struct si_shader_context *ctx, unsigned base_index,
+		   unsigned constant_index, LLVMValueRef dynamic_index)
+{
+	LLVMValueRef offset = LLVMConstInt(ctx->i32, base_index * 4, 0);
+	LLVMValueRef index = LLVMBuildAdd(ctx->ac.builder, dynamic_index,
+					  LLVMConstInt(ctx->ac.i32, constant_index, 0), "");
+
+	/* Bindless uniforms are 64bit so multiple index by 8 */
+	index = LLVMBuildMul(ctx->ac.builder, index, LLVMConstInt(ctx->i32, 8, 0), "");
+	offset = LLVMBuildAdd(ctx->ac.builder, offset, index, "");
+
+	LLVMValueRef ubo_index = abi->load_ubo(abi, ctx->ac.i32_0);
+
+	LLVMValueRef ret = ac_build_buffer_load(&ctx->ac, ubo_index, 1, NULL, offset,
+						NULL, 0, false, false, true, true);
+
+	return LLVMBuildBitCast(ctx->ac.builder, ret, ctx->i32, "");
+}
+
+
 static LLVMValueRef
 si_nir_load_sampler_desc(struct ac_shader_abi *abi,
 		         unsigned descriptor_set, unsigned base_index,
@@ -937,8 +959,17 @@ si_nir_load_sampler_desc(struct ac_shader_abi *abi,
 		LLVMValueRef list =
 			LLVMGetParam(ctx->main_fn, ctx->param_bindless_samplers_and_images);
 
-		/* dynamic_index is the bindless handle */
+		dynamic_index = dynamic_index ? dynamic_index : ctx->ac.i32_0;
+		dynamic_index = get_bindless_index(abi, ctx, base_index,
+						   constant_index, dynamic_index);
+
 		if (image) {
+			/* For simplicity, bindless image descriptors use fixed
+			 * 16-dword slots for now.
+			 */
+			dynamic_index = LLVMBuildMul(ctx->ac.builder, dynamic_index,
+					     LLVMConstInt(ctx->i32, 2, 0), "");
+
 			return si_load_image_desc(ctx, list, dynamic_index, desc_type,
 						  dcc_off, true);
 		}
-- 
2.20.1