[llvm-bugs] [Bug 51324] New: Macros in Intel intrinsic headers missing enclosing parentheses

[via llvm-bugs]

https://bugs.llvm.org/show_bug.cgi?id=51324

            Bug ID: 51324
           Summary: Macros in Intel intrinsic headers missing enclosing
                    parentheses
           Product: clang
           Version: trunk
          Hardware: PC
                OS: All
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: Headers
          Assignee: unassignedclangbugs at nondot.org
          Reporter: ehandal at gmail.com
                CC: craig.topper at gmail.com, llvm-bugs at lists.llvm.org,
                    richard-llvm at metafoo.co.uk

I've seen macros like this in smmintrin.h and xmmintrin.h:

#define _mm_round_ps(X, M) \
  (__m128)__builtin_ia32_roundps((__v4sf)(__m128)(X), (M))

Specifically, macros where the result is casted are missing parentheses
enclosing the cast, leading to unexpected behavior and compile errors if any
operators with higher precedence than the cast are used directly on the return
value.

It's possible more of the Intel intrinsic headers have this issue, but I only
checked the two headers above.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20210803/4371cd3f/attachment.html>