[llvm-bugs] [Bug 34617] New: [X86] [ASM INTEL SYNTAX] clang generates incorrect assembler code when x86-asm-syntax=intel

[via llvm-bugs]

https://bugs.llvm.org/show_bug.cgi?id=34617

            Bug ID: 34617
           Summary: [X86] [ASM INTEL SYNTAX] clang generates incorrect
                    assembler code when x86-asm-syntax=intel
           Product: clang
           Version: trunk
          Hardware: PC
                OS: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: LLVM Codegen
          Assignee: unassignedclangbugs at nondot.org
          Reporter: konstantin.belochapka at sony.com
                CC: llvm-bugs at lists.llvm.org

Consider the following source code:
-----------------------------------------------------------------
char X[ 4 ];
extern char testCall084(long    n, char* y);
volatile char* px;
char test084(long x0) {
    char rc = testCall084(x0, X);
    px =  X;
    return rc;
}

Running:

$clang -S -O1 -mllvm --x86-asm-syntax=intel source.cpp

Produces the following assembler source:
--------------------------------------------------------------
_Z7test084l:
        push    rax
        mov     esi, X
        call    _Z11testCall084lPc
        mov     qword ptr [rip + px], X
        pop     rcx
        ret

GNU assembler interprets: 
1       mov     esi, X
2       mov     qword ptr [rip + px], X
as:
1       mov     esi, [X]
2       mov     qword ptr [rip + px], [X]

which produces incorrect binary code for first instruction
and error message:
"too many memory references for `mov'" for second instruction.

-- 
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/20170915/7023d62f/attachment.html>