[Mesa-dev] [PATCH] ac/nir: Add workaround for GFX9 buffer views.
Bas Nieuwenhuizen
bas at basnieuwenhuizen.nl
Wed Apr 4 20:19:16 UTC 2018
On GFX9 whether the buffer size is interpreted as elements or bytes
depends on whether IDXEN is enabled in the instruction. If the index
is a constant zero, LLVM optimizes IDXEN to 0.
Now the size in elements is interpreted in bytes which of course
results in out of bounds accesses.
The correct fix is most likely to disable the LLVM optimization,
but we need something to work with LLVM <= 6.0.
radeonsi does the max between stride and element count on the CPU
but that results in the size intrinsics returning the wrong size
for the buffer. This would cause CTS errors for radv.
v2: Also include the store changes.
Fixes: e38685cc62e 'Revert "radv: disable support for VEGA for now."'
(backported from 4503ff760c794c3bb15b978a47c530037d56498e for 17.3)
---
src/amd/common/ac_llvm_build.c | 20 ++++++++++++++++++++
src/amd/common/ac_llvm_build.h | 8 ++++++++
src/amd/common/ac_nir_to_llvm.c | 36 ++++++++++++++++++++++++++++++------
src/amd/common/ac_shader_abi.h | 4 ++++
4 files changed, 62 insertions(+), 6 deletions(-)
diff --git a/src/amd/common/ac_llvm_build.c b/src/amd/common/ac_llvm_build.c
index e5cd23e025..f193f71c3e 100644
--- a/src/amd/common/ac_llvm_build.c
+++ b/src/amd/common/ac_llvm_build.c
@@ -960,6 +960,26 @@ LLVMValueRef ac_build_buffer_load_format(struct ac_llvm_context *ctx,
AC_FUNC_ATTR_READONLY);
}
+LLVMValueRef ac_build_buffer_load_format_gfx9_safe(struct ac_llvm_context *ctx,
+ LLVMValueRef rsrc,
+ LLVMValueRef vindex,
+ LLVMValueRef voffset,
+ bool can_speculate)
+{
+ LLVMValueRef elem_count = LLVMBuildExtractElement(ctx->builder, rsrc, LLVMConstInt(ctx->i32, 2, 0), "");
+ LLVMValueRef stride = LLVMBuildExtractElement(ctx->builder, rsrc, LLVMConstInt(ctx->i32, 1, 0), "");
+ stride = LLVMBuildLShr(ctx->builder, stride, LLVMConstInt(ctx->i32, 16, 0), "");
+
+ LLVMValueRef new_elem_count = LLVMBuildSelect(ctx->builder,
+ LLVMBuildICmp(ctx->builder, LLVMIntUGT, elem_count, stride, ""),
+ elem_count, stride, "");
+
+ LLVMValueRef new_rsrc = LLVMBuildInsertElement(ctx->builder, rsrc, new_elem_count,
+ LLVMConstInt(ctx->i32, 2, 0), "");
+
+ return ac_build_buffer_load_format(ctx, new_rsrc, vindex, voffset, can_speculate);
+}
+
/**
* Set range metadata on an instruction. This can only be used on load and
* call instructions. If you know an instruction can only produce the values
diff --git a/src/amd/common/ac_llvm_build.h b/src/amd/common/ac_llvm_build.h
index aa2a2899ab..d4264f2879 100644
--- a/src/amd/common/ac_llvm_build.h
+++ b/src/amd/common/ac_llvm_build.h
@@ -188,6 +188,14 @@ LLVMValueRef ac_build_buffer_load_format(struct ac_llvm_context *ctx,
LLVMValueRef voffset,
bool can_speculate);
+/* load_format that handles the stride & element count better if idxen is
+ * disabled by LLVM. */
+LLVMValueRef ac_build_buffer_load_format_gfx9_safe(struct ac_llvm_context *ctx,
+ LLVMValueRef rsrc,
+ LLVMValueRef vindex,
+ LLVMValueRef voffset,
+ bool can_speculate);
+
LLVMValueRef
ac_get_thread_id(struct ac_llvm_context *ctx);
diff --git a/src/amd/common/ac_nir_to_llvm.c b/src/amd/common/ac_nir_to_llvm.c
index b696cb8a45..8ee80b1a67 100644
--- a/src/amd/common/ac_nir_to_llvm.c
+++ b/src/amd/common/ac_nir_to_llvm.c
@@ -2257,11 +2257,19 @@ static LLVMValueRef build_tex_intrinsic(struct ac_nir_context *ctx,
struct ac_image_args *args)
{
if (instr->sampler_dim == GLSL_SAMPLER_DIM_BUF) {
- return ac_build_buffer_load_format(&ctx->ac,
- args->resource,
- args->addr,
- LLVMConstInt(ctx->ac.i32, 0, false),
- true);
+ if (ctx->abi->gfx9_stride_size_workaround) {
+ return ac_build_buffer_load_format_gfx9_safe(&ctx->ac,
+ args->resource,
+ args->addr,
+ ctx->ac.i32_0,
+ true);
+ } else {
+ return ac_build_buffer_load_format(&ctx->ac,
+ args->resource,
+ args->addr,
+ ctx->ac.i32_0,
+ true);
+ }
}
args->opcode = ac_image_sample;
@@ -3613,8 +3621,23 @@ static void visit_image_store(struct ac_nir_context *ctx,
glc = i1true;
if (glsl_get_sampler_dim(type) == GLSL_SAMPLER_DIM_BUF) {
+ LLVMValueRef rsrc = get_sampler_desc(ctx, instr->variables[0], AC_DESC_BUFFER, true, true);
+
+ if (ctx->abi->gfx9_stride_size_workaround) {
+ LLVMValueRef elem_count = LLVMBuildExtractElement(ctx->ac.builder, rsrc, LLVMConstInt(ctx->ac.i32, 2, 0), "");
+ LLVMValueRef stride = LLVMBuildExtractElement(ctx->ac.builder, rsrc, LLVMConstInt(ctx->ac.i32, 1, 0), "");
+ stride = LLVMBuildLShr(ctx->ac.builder, stride, LLVMConstInt(ctx->ac.i32, 16, 0), "");
+
+ LLVMValueRef new_elem_count = LLVMBuildSelect(ctx->ac.builder,
+ LLVMBuildICmp(ctx->ac.builder, LLVMIntUGT, elem_count, stride, ""),
+ elem_count, stride, "");
+
+ rsrc = LLVMBuildInsertElement(ctx->ac.builder, rsrc, new_elem_count,
+ LLVMConstInt(ctx->ac.i32, 2, 0), "");
+ }
+
params[0] = ac_to_float(&ctx->ac, get_src(ctx, instr->src[2])); /* data */
- params[1] = get_sampler_desc(ctx, instr->variables[0], AC_DESC_BUFFER, true, true);
+ params[1] = rsrc;
params[2] = LLVMBuildExtractElement(ctx->ac.builder, get_src(ctx, instr->src[0]),
ctx->ac.i32_0, ""); /* vindex */
params[3] = ctx->ac.i32_0; /* voffset */
@@ -6645,6 +6668,7 @@ LLVMModuleRef ac_translate_nir_to_llvm(LLVMTargetMachineRef tm,
ctx.abi.load_ssbo = radv_load_ssbo;
ctx.abi.load_sampler_desc = radv_get_sampler_desc;
ctx.abi.clamp_shadow_reference = false;
+ ctx.abi.gfx9_stride_size_workaround = ctx.ac.chip_class == GFX9;
if (shader_count >= 2)
ac_init_exec_full_mask(&ctx.ac);
diff --git a/src/amd/common/ac_shader_abi.h b/src/amd/common/ac_shader_abi.h
index 14517d5570..b04dc076de 100644
--- a/src/amd/common/ac_shader_abi.h
+++ b/src/amd/common/ac_shader_abi.h
@@ -92,6 +92,10 @@ struct ac_shader_abi {
/* Whether to clamp the shadow reference value to [0,1]on VI. Radeonsi currently
* uses it due to promoting D16 to D32, but radv needs it off. */
bool clamp_shadow_reference;
+
+ /* Whether to workaround GFX9 ignoring the stride for the buffer size if IDXEN=0
+ * and LLVM optimizes an indexed load with constant index to IDXEN=0. */
+ bool gfx9_stride_size_workaround;
};
#endif /* AC_SHADER_ABI_H */
--
2.16.2
More information about the mesa-dev
mailing list