[llvm-bugs] [Bug 37598] New: clang emitting RIP-relative relocation for weak external on Windows x86-64

via llvm-bugs llvm-bugs at lists.llvm.org
Fri May 25 17:10:50 PDT 2018


https://bugs.llvm.org/show_bug.cgi?id=37598

            Bug ID: 37598
           Summary: clang emitting RIP-relative relocation for weak
                    external on Windows x86-64
           Product: libraries
           Version: trunk
          Hardware: PC
                OS: All
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: Backend: X86
          Assignee: unassignedbugs at nondot.org
          Reporter: smeenai at fb.com
                CC: compnerd at compnerd.org, llvm-bugs at lists.llvm.org,
                    rnk at google.com

% cat reduced.ll
@Weak = extern_weak global i32
define i32* @getWeak() {
  ret i32* @Weak
}

% clang -target x86_64-windows-msvc -S -o - reduced.ll
        .text
        .def     getWeak;
        .scl    2;
        .type   32;
        .endef
        .globl  getWeak                 # -- Begin function getWeak
        .p2align        4, 0x90
getWeak:                                # @getWeak
# %bb.0:
        leaq    Weak(%rip), %rax
        retq
                                        # -- End function
        .weak   Weak

The issue is that LLVM implements weak symbols for COFF through the weak
external mechanism, and it emits an absolute symbol with the value of 0 to
serve as the fallback symbol. We'll therefore end up with a RIP-relative
relocation against an absolute symbol in the case where Weak isn't defined.

link errors out for this:
error LNK2016: absolute symbol 'Weak' used as target of REL32 relocation in
section 1

LLD accepts it silently; I'll file a separate bug for that.

-- 
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/20180526/5e93defb/attachment-0001.html>


More information about the llvm-bugs mailing list