[clang] [clang][CodeGen] fix UB in aarch64 bfloat16 scalar conversion (PR #89062)
via cfe-commits
cfe-commits at lists.llvm.org
Wed Apr 17 05:12:15 PDT 2024
llvmbot wrote:
<!--LLVM PR SUMMARY COMMENT-->
@llvm/pr-subscribers-clang
Author: None (nihui)
<details>
<summary>Changes</summary>
do not bitcast 16bit `bfloat16` to 32bit `int32_t` directly
bitcast to `int16_t`, and then upcast to `int32_t`
Fix ASAN runtime error when calling vcvtah_f32_bf16
`==21842==ERROR: AddressSanitizer: stack-buffer-overflow on address 0x007fda1dd063 at pc 0x005c0361c234 bp 0x007fda1dd030 sp 0x007fda1dd028
`
without patch
```c
__ai __attribute__((target("bf16"))) float32_t vcvtah_f32_bf16(bfloat16_t __p0) {
float32_t __ret;
bfloat16_t __reint = __p0;
int32_t __reint1 = *(int32_t *) &__reint << 16;
__ret = *(float32_t *) &__reint1;
return __ret;
}
```
with this patch
```c
__ai __attribute__((target("bf16"))) float32_t vcvtah_f32_bf16(bfloat16_t __p0) {
float32_t __ret;
bfloat16_t __reint = __p0;
int32_t __reint1 = (int32_t)(*(int16_t *) &__reint) << 16;
__ret = *(float32_t *) &__reint1;
return __ret;
}
```
fix issue https://github.com/llvm/llvm-project/issues/61983
---
Full diff: https://github.com/llvm/llvm-project/pull/89062.diff
1 Files Affected:
- (modified) clang/include/clang/Basic/arm_neon.td (+1-1)
``````````diff
diff --git a/clang/include/clang/Basic/arm_neon.td b/clang/include/clang/Basic/arm_neon.td
index 6d655c39360d3b..6390ba3f9fe5e5 100644
--- a/clang/include/clang/Basic/arm_neon.td
+++ b/clang/include/clang/Basic/arm_neon.td
@@ -275,7 +275,7 @@ def OP_VCVT_BF16_F32_HI_A32
(call "vget_low", $p0))>;
def OP_CVT_F32_BF16
- : Op<(bitcast "R", (op "<<", (bitcast "int32_t", $p0),
+ : Op<(bitcast "R", (op "<<", (cast "int32_t", (bitcast "int16_t", $p0)),
(literal "int32_t", "16")))>;
//===----------------------------------------------------------------------===//
``````````
</details>
https://github.com/llvm/llvm-project/pull/89062
More information about the cfe-commits
mailing list