[Mlir-commits] [mlir] [mlir] Add guard for using posixmemalign <= Android API 28 (PR #194263)

[llvmlistbot at llvm.org]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->
@llvm/pr-subscribers-mlir

@llvm/pr-subscribers-mlir-execution-engine

Author: cqwrteur (trcrsired)

<details>
<summary>Changes</summary>

aligned_alloc does not exist for Android before 28 too just like apple

---
Full diff: https://github.com/llvm/llvm-project/pull/194263.diff


1 Files Affected:

- (modified) mlir/lib/ExecutionEngine/CRunnerUtils.cpp (+1-1) 


``````````diff
diff --git a/mlir/lib/ExecutionEngine/CRunnerUtils.cpp b/mlir/lib/ExecutionEngine/CRunnerUtils.cpp
index 41c619566b55d..272137d4b1d7c 100644
--- a/mlir/lib/ExecutionEngine/CRunnerUtils.cpp
+++ b/mlir/lib/ExecutionEngine/CRunnerUtils.cpp
@@ -152,7 +152,7 @@ extern "C" void *mlirAlloc(uint64_t size) { return malloc(size); }
 extern "C" void *mlirAlignedAlloc(uint64_t alignment, uint64_t size) {
 #ifdef _WIN32
   return _aligned_malloc(size, alignment);
-#elif defined(__APPLE__)
+#elif defined(__APPLE__) || (defined(__ANDROID_API__) && __ANDROID_API__ <= 28)
   // aligned_alloc was added in MacOS 10.15. Fall back to posix_memalign to also
   // support older versions.
   void *result = nullptr;

``````````

</details>


https://github.com/llvm/llvm-project/pull/194263