[mlir][nvvm] Expose MLIR_NVPTXCOMPILER_ENABLED in mlir-config.h.

[ingomueller-net]

This is another follow-up of #83004, which made the same change for MLIR_CUDA_CONVERSIONS_ENABLED. As the previous PR, this PR commit exposes mentioned CMake variable through mlir-config.h and uses the macro that is introduced with the same name. This replaces the macro MLIR_NVPTXCOMPILER_ENABLED, which the CMake files previously defined manually.

[llvmbot]

@llvm/pr-subscribers-mlir

@llvm/pr-subscribers-mlir-llvm

Author: Ingo Müller (ingomueller-net)

Changes

This is another follow-up of #83004, which made the same change for MLIR_CUDA_CONVERSIONS_ENABLED. As the previous PR, this PR commit exposes mentioned CMake variable through mlir-config.h and uses the macro that is introduced with the same name. This replaces the macro MLIR_NVPTXCOMPILER_ENABLED, which the CMake files previously defined manually.

---

Full diff: https://github.com/llvm/llvm-project/pull/84007.diff

4 Files Affected:

• (modified) mlir/include/mlir/Config/mlir-config.h.cmake (+3)
• (modified) mlir/lib/Target/LLVM/CMakeLists.txt (-1)
• (modified) mlir/lib/Target/LLVM/NVVM/Target.cpp (+6-5)
• (modified) utils/bazel/llvm-project-overlay/mlir/BUILD.bazel (+9-2)

diff --git a/mlir/include/mlir/Config/mlir-config.h.cmake b/mlir/include/mlir/Config/mlir-config.h.cmake
index 4a7d75e2266823..03fda997313ba9 100644
--- a/mlir/include/mlir/Config/mlir-config.h.cmake
+++ b/mlir/include/mlir/Config/mlir-config.h.cmake
@@ -33,4 +33,7 @@
    and targets. */
 #cmakedefine01 MLIR_ENABLE_CUDA_CONVERSIONS
 
+/* If set, enables features that depend on the NVIDIA's PTX compiler. */
+#cmakedefine01 MLIR_ENABLE_NVPTXCOMPILER
+
 #endif
diff --git a/mlir/lib/Target/LLVM/CMakeLists.txt b/mlir/lib/Target/LLVM/CMakeLists.txt
index cc2c3a00a02eaf..e0657c895e8a38 100644
--- a/mlir/lib/Target/LLVM/CMakeLists.txt
+++ b/mlir/lib/Target/LLVM/CMakeLists.txt
@@ -93,7 +93,6 @@ if(MLIR_ENABLE_CUDA_CONVERSIONS)
   # Define the `CUDAToolkit` path.
   target_compile_definitions(obj.MLIRNVVMTarget
     PRIVATE
-    MLIR_NVPTXCOMPILER_ENABLED=${MLIR_ENABLE_NVPTXCOMPILER}
     __DEFAULT_CUDATOOLKIT_PATH__="${MLIR_CUDAToolkit_ROOT}"
   )
 endif()
diff --git a/mlir/lib/Target/LLVM/NVVM/Target.cpp b/mlir/lib/Target/LLVM/NVVM/Target.cpp
index d5b6645631edd6..e438ce84af1b5f 100644
--- a/mlir/lib/Target/LLVM/NVVM/Target.cpp
+++ b/mlir/lib/Target/LLVM/NVVM/Target.cpp
@@ -22,6 +22,7 @@
 #include "mlir/Target/LLVMIR/Dialect/NVVM/NVVMToLLVMIRTranslation.h"
 #include "mlir/Target/LLVMIR/Export.h"
 
+#include "llvm/Config/llvm-config.h"
 #include "llvm/Support/FileSystem.h"
 #include "llvm/Support/FileUtilities.h"
 #include "llvm/Support/FormatVariadic.h"
@@ -431,7 +432,7 @@ NVPTXSerializer::compileToBinary(const std::string &ptxCode) {
   return SmallVector<char, 0>(fatbin.begin(), fatbin.end());
 }
 
-#if MLIR_NVPTXCOMPILER_ENABLED == 1
+#if MLIR_ENABLE_NVPTXCOMPILER
 #include "nvPTXCompiler.h"
 
 #define RETURN_ON_NVPTXCOMPILER_ERROR(expr)                                    \
@@ -510,7 +511,7 @@ NVPTXSerializer::compileToBinaryNVPTX(const std::string &ptxCode) {
   RETURN_ON_NVPTXCOMPILER_ERROR(nvPTXCompilerDestroy(&compiler));
   return binary;
 }
-#endif // MLIR_NVPTXCOMPILER_ENABLED == 1
+#endif // MLIR_ENABLE_NVPTXCOMPILER
 
 std::optional<SmallVector<char, 0>>
 NVPTXSerializer::moduleToObject(llvm::Module &llvmModule) {
@@ -556,12 +557,12 @@ NVPTXSerializer::moduleToObject(llvm::Module &llvmModule) {
     return SmallVector<char, 0>(bin.begin(), bin.end());
   }
 
-    // Compile to binary.
-#if MLIR_NVPTXCOMPILER_ENABLED == 1
+  // Compile to binary.
+#if MLIR_ENABLE_NVPTXCOMPILER
   return compileToBinaryNVPTX(*serializedISA);
 #else
   return compileToBinary(*serializedISA);
-#endif // MLIR_NVPTXCOMPILER_ENABLED == 1
+#endif // MLIR_ENABLE_NVPTXCOMPILER
 }
 #endif // MLIR_ENABLE_CUDA_CONVERSIONS
 
diff --git a/utils/bazel/llvm-project-overlay/mlir/BUILD.bazel b/utils/bazel/llvm-project-overlay/mlir/BUILD.bazel
index 8a8dd6e10c48aa..6875963905c49c 100644
--- a/utils/bazel/llvm-project-overlay/mlir/BUILD.bazel
+++ b/utils/bazel/llvm-project-overlay/mlir/BUILD.bazel
@@ -36,8 +36,14 @@ expand_template(
         "#cmakedefine MLIR_GREEDY_REWRITE_RANDOMIZER_SEED ${MLIR_GREEDY_REWRITE_RANDOMIZER_SEED}": "/* #undef MLIR_GREEDY_REWRITE_RANDOMIZER_SEED */",
         "#cmakedefine01 MLIR_ENABLE_PDL_IN_PATTERNMATCH": "#define MLIR_ENABLE_PDL_IN_PATTERNMATCH 1",
     } | if_cuda_available(
-        {"#cmakedefine01 MLIR_ENABLE_CUDA_CONVERSIONS": "#define MLIR_ENABLE_CUDA_CONVERSIONS 1"},
-        {"#cmakedefine01 MLIR_ENABLE_CUDA_CONVERSIONS": "#define MLIR_ENABLE_CUDA_CONVERSIONS 0"},
+        {
+            "#cmakedefine01 MLIR_ENABLE_CUDA_CONVERSIONS": "#define MLIR_ENABLE_CUDA_CONVERSIONS 1",
+            "#cmakedefine01 MLIR_ENABLE_NVPTXCOMPILER": "#define MLIR_ENABLE_NVPTXCOMPILER 1",
+        },
+        {
+            "#cmakedefine01 MLIR_ENABLE_CUDA_CONVERSIONS": "#define MLIR_ENABLE_CUDA_CONVERSIONS 0",
+            "#cmakedefine01 MLIR_ENABLE_NVPTXCOMPILER": "#define MLIR_ENABLE_NVPTXCOMPILER 0",
+        },
     ),
     template = "include/mlir/Config/mlir-config.h.cmake",
 )
@@ -6182,6 +6188,7 @@ cc_library(
         ":config",
         "//llvm:NVPTXCodeGen",
         "//llvm:Support",
+        "//llvm:config",
     ],
 )
 

[ingomueller-net]

On first thought, that doesn't sound plausible. How are the two macros related? If you're using cmake, then it is most likely rather caused by #84011.

I am not sure about this specific case. Very generally, we either need to define the macro, either in a file or through the compiler/cmake/bazel, or turn the #if into #ifdef.

[PeimingLiu]

On first thought, that doesn't sound plausible. How are the two macros related? If you're using cmake, then it is most likely rather caused by #84011.

I am not sure about this specific case. Very generally, we either need to define the macro, either in a file or through the compiler/cmake/bazel, or turn the #if into #ifdef.

Sorry, I commented on the wrong PR, it should be caused by #84011