[llvm] [AMDGPU] Introduce "amdgpu-sw-lower-lds" pass to lower LDS accesses to use device global memory. (PR #87265)

Matt Arsenault via llvm-commits llvm-commits at lists.llvm.org
Wed May 8 09:51:48 PDT 2024


================
@@ -679,6 +680,11 @@ void AMDGPUTargetMachine::registerPassBuilderCallbacks(
 
         if (EarlyInlineAll && !EnableFunctionCalls)
           PM.addPass(AMDGPUAlwaysInlinePass());
+
+#if __has_feature(address_sanitizer)
----------------
arsenm wrote:

There is no conditional compilation. The pass must be unconditionally run. You can skip functions inside the pass itself based on the sanitize_address function attribute on individual functions 

https://github.com/llvm/llvm-project/pull/87265


More information about the llvm-commits mailing list