[clang] [X86][AMX-AVX512] Add required/min vector width 512 constraint (PR #173532)

[via cfe-commits]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-backend-x86

Author: Phoebe Wang (phoebewang)

<details>
<summary>Changes</summary>

We use ZMM registers in AMX-AVX512 instructions.

Fixes: https://godbolt.org/z/GsY167Yzn

---
Full diff: https://github.com/llvm/llvm-project/pull/173532.diff


2 Files Affected:

- (modified) clang/include/clang/Basic/BuiltinsX86_64.td (+2-2) 
- (modified) clang/lib/Headers/amxavx512intrin.h (+1-1) 


``````````diff
diff --git a/clang/include/clang/Basic/BuiltinsX86_64.td b/clang/include/clang/Basic/BuiltinsX86_64.td
index 062060e6afbbe..2e68a5a016ef0 100644
--- a/clang/include/clang/Basic/BuiltinsX86_64.td
+++ b/clang/include/clang/Basic/BuiltinsX86_64.td
@@ -239,7 +239,7 @@ let Features = "amx-complex", Attributes = [NoThrow] in {
   def tcmmrlfp16ps_internal : X86Builtin<"_Vector<256, int>(unsigned short, unsigned short, unsigned short, _Vector<256, int>, _Vector<256, int>, _Vector<256, int>)">;
 }
 
-let Features = "amx-avx512,avx10.2", Attributes = [NoThrow] in {
+let Features = "amx-avx512,avx10.2", Attributes = [NoThrow, RequiredVectorWidth<512>] in {
   def tcvtrowd2ps_internal : X86Builtin<"_Vector<16, float>(unsigned short, unsigned short, _Vector<256, int>, unsigned int)">;
   def tcvtrowps2bf16h_internal : X86Builtin<"_Vector<32, __bf16>(unsigned short, unsigned short, _Vector<256, int>, unsigned int)">;
   def tcvtrowps2bf16l_internal : X86Builtin<"_Vector<32, __bf16>(unsigned short, unsigned short, _Vector<256, int>, unsigned int)">;
@@ -297,7 +297,7 @@ let Features = "amx-complex", Attributes = [NoThrow] in {
   def tcmmrlfp16ps : X86Builtin<"void(_Constant unsigned char, _Constant unsigned char, _Constant unsigned char)">;
 }
 
-let Features = "amx-avx512,avx10.2", Attributes = [NoThrow] in {
+let Features = "amx-avx512,avx10.2", Attributes = [NoThrow, RequiredVectorWidth<512>] in {
   def tcvtrowd2ps : X86Builtin<"_Vector<16, float>(_Constant unsigned char, unsigned int)">;
   def tcvtrowps2bf16h : X86Builtin<"_Vector<32, __bf16>(_Constant unsigned char, unsigned int)">;
   def tcvtrowps2bf16l : X86Builtin<"_Vector<32, __bf16>(_Constant unsigned char, unsigned int)">;
diff --git a/clang/lib/Headers/amxavx512intrin.h b/clang/lib/Headers/amxavx512intrin.h
index 373a54f46290f..5f21c25b05ad1 100644
--- a/clang/lib/Headers/amxavx512intrin.h
+++ b/clang/lib/Headers/amxavx512intrin.h
@@ -16,7 +16,7 @@
 
 #define __DEFAULT_FN_ATTRS_AVX512                                              \
   __attribute__((__always_inline__, __nodebug__,                               \
-                 __target__("amx-avx512,avx10.2")))
+                 __target__("amx-avx512,avx10.2"), __min_vector_width__(512)))
 
 /// Moves a row from a tile register to a zmm destination register, converting
 ///    the int32 source elements to fp32. The row of the tile is selected by a

``````````

</details>


https://github.com/llvm/llvm-project/pull/173532