[CP] Upstream LLVM commit upgrades and gfx950 target addition

ROCm · Dec 13, 2024 · 2de5803 · 2de5803
1 parent dbc771e
commit 2de5803
Show file tree

Hide file tree

Showing 6 changed files with 11 additions and 9 deletions.
diff --git a/cmake/llvm-hash.txt b/cmake/llvm-hash.txt
@@ -1 +1 @@
-b5cc222d7429fe6f18c787f633d5262fac2e676f
+86b69c31642e98f8357df62c09d118ad1da4e16a
diff --git a/lib/Conversion/TritonToTritonGPU/TritonGPUConversion.cpp b/lib/Conversion/TritonToTritonGPU/TritonGPUConversion.cpp
@@ -56,20 +56,19 @@ TritonGPUTypeConverter::TritonGPUTypeConverter(MLIRContext *context,
   // This will create newArg, and map(origArg, newArg)
   addArgumentMaterialization([&](OpBuilder &builder,
                                  RankedTensorType tensorType, ValueRange inputs,
-                                 Location loc) -> std::optional<Value> {
+                                 Location loc) -> Value {
     llvm_unreachable("Argument rematerialization should not happen in Triton "
                      "-> TritonGPU conversion");
-    return std::nullopt;
+    return {};
   });
 
   // If the origValue still has live user(s), use this to
   // convert origValue to newValue
   addSourceMaterialization([&](OpBuilder &builder, RankedTensorType tensorType,
-                               ValueRange inputs,
-                               Location loc) -> std::optional<Value> {
+                               ValueRange inputs, Location loc) -> Value {
     llvm_unreachable("Source rematerialization should not happen in Triton -> "
                      "TritonGPU Conversion");
-    return std::nullopt;
+    return {};
   });
 
   // This will be called when (desiredType != newOperandType)
@@ -79,7 +78,7 @@ TritonGPUTypeConverter::TritonGPUTypeConverter(MLIRContext *context,
                                ValueRange inputs, Location loc) {
     auto cast =
         builder.create<triton::gpu::ConvertLayoutOp>(loc, tensorType, inputs);
-    return std::optional<Value>(cast.getResult());
+    return cast.getResult();
   });
 }
 

diff --git a/test/TritonGPU/amd/amd-convert-buffer-ops.mlir b/test/TritonGPU/amd/amd-convert-buffer-ops.mlir
@@ -42,7 +42,7 @@ module attributes {"triton_gpu.num-ctas" = 1 : i32, "triton_gpu.num-warps" = 4 :
     %1 = arith.muli %0, %c1024_i32 : i32
     %sub = arith.subi %1, %c128_i32 : i32
     %cmp = arith.cmpi sgt, %sub, %c0_i32 : i32
-    "llvm.intr.assume"(%cmp) : (i1) -> ()
+    llvm.intr.assume %cmp : i1
     %2 = tt.splat %sub : i32 -> tensor<1024xi32, #blocked>
     %3 = tt.make_range {end = 1024 : i32, start = 0 : i32} : tensor<1024xi32, #blocked>
     // CHECK: %[[offset:.*]] = arith.addi

diff --git a/test/lib/Instrumentation/GPUHello.cpp b/test/lib/Instrumentation/GPUHello.cpp
@@ -61,7 +61,7 @@ bool GpuHello::runOnModule(Module &module) {
 
 PassPluginLibraryInfo getPassPluginInfo() {
   const auto callback = [](PassBuilder &pb) {
-    pb.registerOptimizerLastEPCallback([&](ModulePassManager &mpm, auto) {
+    pb.registerOptimizerLastEPCallback([&](ModulePassManager &mpm, auto, auto) {
       mpm.addPass(GpuHello());
       return true;
     });

diff --git a/third_party/amd/backend/include/hsa/amd_hsa_elf.h b/third_party/amd/backend/include/hsa/amd_hsa_elf.h
@@ -130,6 +130,8 @@ enum : unsigned {
   EF_AMDGPU_MACH_AMDGCN_GFX1151       = 0x04a,
   EF_AMDGPU_MACH_AMDGCN_GFX941        = 0x04b,
   EF_AMDGPU_MACH_AMDGCN_GFX942        = 0x04c,
+  EF_AMDGPU_MACH_AMDGCN_GFX950        = 0x04f,
+  // clang-format on
 
   // First/last AMDGCN-based processors.
   EF_AMDGPU_MACH_AMDGCN_FIRST = EF_AMDGPU_MACH_AMDGCN_GFX600,

diff --git a/third_party/amd/lib/TritonAMDGPUToLLVM/TargetUtils.cpp b/third_party/amd/lib/TritonAMDGPUToLLVM/TargetUtils.cpp
@@ -11,6 +11,7 @@ ISAFamily deduceISAFamily(llvm::StringRef arch) {
 
   // CDNA ISA cases
   switch (kind) {
+  case llvm::AMDGPU::GK_GFX950:
   case llvm::AMDGPU::GK_GFX942:
   case llvm::AMDGPU::GK_GFX941:
   case llvm::AMDGPU::GK_GFX940:
Original file line number	Diff line number	Diff line change
		@@ -1 +1 @@
		b5cc222d7429fe6f18c787f633d5262fac2e676f
		86b69c31642e98f8357df62c09d118ad1da4e16a