https://github.com/tcgu-amd updated
https://github.com/llvm/llvm-project/pull/158145
>From 2790a7a05ab9453176a29ce7f5c3def7e4c9a1ae Mon Sep 17 00:00:00 2001
From: Tim Gu
Date: Thu, 18 Sep 2025 13:03:11 -0400
Subject: [PATCH] Removed custom type checking & changed type string from
__fp16 to _Fl
https://github.com/tcgu-amd created
https://github.com/llvm/llvm-project/pull/158145
Addresses https://github.com/ROCm/ROCm/issues/5253.
The builtin expects a vector _Float16 of length 2, but clang does not emit
errors when the wrong argument types is supplied (e.g. half2). This causes the
c
@@ -436,4 +440,32 @@ void SemaAMDGPU::handleAMDGPUMaxNumWorkGroupsAttr(Decl *D,
addAMDGPUMaxNumWorkGroupsAttr(D, AL, AL.getArgAsExpr(0), YExpr, ZExpr);
}
+
+bool SemaAMDGPU::checkAMDGCNAtomicFaddV2F16Type(CallExpr *TheCall) {
+ // Check that the pointer argument is a pointe
@@ -436,4 +440,32 @@ void SemaAMDGPU::handleAMDGPUMaxNumWorkGroupsAttr(Decl *D,
addAMDGPUMaxNumWorkGroupsAttr(D, AL, AL.getArgAsExpr(0), YExpr, ZExpr);
}
+
+bool SemaAMDGPU::checkAMDGCNAtomicFaddV2F16Type(CallExpr *TheCall) {
+ // Check that the pointer argument is a pointe
