[llvm-bugs] [Bug 37760] New: regcall calling convention mismatch between Clang and Intel C++ compiler (AVX512)

[via llvm-bugs]

https://bugs.llvm.org/show_bug.cgi?id=37760

            Bug ID: 37760
           Summary: regcall calling convention mismatch between Clang and
                    Intel C++ compiler (AVX512)
           Product: libraries
           Version: trunk
          Hardware: PC
                OS: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: Backend: X86
          Assignee: unassignedbugs at nondot.org
          Reporter: wenzel.jakob at epfl.ch
                CC: llvm-bugs at lists.llvm.org

Dear LLVM/Clang developers,

the 'regcall' calling convention currently does not match the behavior of the
Intel compiler when working with values that use AVX512-style ZMM registers.
This is problematic for two reasons:

1. 'regcall' functions compiled with Clang and ICPC are not ABI-compatible.

2. The 'regcall' calling convention is used to improve performance in
vectorized code involving function calls. It currently seems impossible to
obtain this benefit when using Clang & AVX512.

How to reproduce:
-----------------

Consider the following simple snippet which has 256 and 512 bit versions of a
function that just forwards its argumetns.

/// ---------------------

#include <immintrin.h>

struct Vector2_256 { __m256 x[2]; };
struct Vector2_512 { __m512 x[2]; };

__attribute__((regcall)) void f_256(Vector2_256 x);
__attribute__((regcall)) void f_512(Vector2_512 x);

__attribute__((regcall)) void call_f_256(Vector2_256 x) { return f_256(x); }
__attribute__((regcall)) void call_f_512(Vector2_512 x) { return f_512(x); }

/// ---------------------

With Clang trunk, this compiles to

$ clang++ test.cpp -march=skx -S -o - -O3 -fomit-frame-pointer
# (with minor cleanups)
__Z22__regcall3__call_f_25611Vector2_256:
        jmp     __Z17__regcall3__f_25611Vector2_256

__Z22__regcall3__call_f_51211Vector2_512:
        pushq   %rbp
        movq    %rsp, %rbp
        pushq   %rsp
        andq    $-64, %rsp
        subq    $192, %rsp
        vmovaps 16(%rbp), %zmm0
        vmovaps 80(%rbp), %zmm1
        vmovaps %zmm1, 64(%rsp)
        vmovaps %zmm0, (%rsp)
        vzeroupper
        callq   __Z17__regcall3__f_51211Vector2_512
        leaq    -8(%rbp), %rsp
        popq    %rsp
        popq    %rbp
        retq


In other words, the 256 bit version is correct, while the 512-bit version
fetches the arguments and passes them on the stack (which it shouldn't do.)

On ICPC, I get

_Z22__regcall2__call_f_25611Vector2_256:
        jmp       _Z17__regcall2__f_25611Vector2_256

_Z22__regcall2__call_f_51211Vector2_512:
        jmp       _Z17__regcall2__f_51211Vector2_512

-- 
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/20180610/c545b843/attachment-0001.html>