[llvm-bugs] [Bug 40229] New: lld-link fails to detect an attributes mismatch which link.exe does detect

via llvm-bugs llvm-bugs at lists.llvm.org
Fri Jan 4 12:05:09 PST 2019


https://bugs.llvm.org/show_bug.cgi?id=40229

            Bug ID: 40229
           Summary: lld-link fails to detect an attributes mismatch which
                    link.exe does detect
           Product: lld
           Version: unspecified
          Hardware: PC
                OS: Windows NT
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: COFF
          Assignee: unassignedbugs at nondot.org
          Reporter: brucedawson at chromium.org
                CC: llvm-bugs at lists.llvm.org

Building Chrome on Windows (x86 or x64) with link.exe (use_lld=false) the
following warning is hit:

libvpx_yasm.lib(vpx_subpixel_8t_ssse3.o) : warning LNK4078: multiple '.rodata'
sections found with different attributes (40500040)
LINK : error LNK1218: warning treated as error; no output file generated

This suggests two issues:
1) We are creating .rodata sections with different attributes in
vpx_subpixel_8t_ssse3.o
2) lld-link.exe is not noticing this discrepancy

Because this is only a warning I can move past it by adding /ignore:4078 to
common_linker_setup but I don't want to commit to such a "fix" because I don't
understand how we are ending up with different attributes and what they mean.

The gn args I used are listed here:

use_lld = false
is_component_build = false
is_debug = false
target_cpu = "x86"
enable_nacl = false

I filed crbug.com/919180 to track fixing libvpx to avoid the different
attributes, but fixing lld-link.exe to detect the different attributes would
also be valuable since it would ensure that we find these glitches in a timely
manner. The time frame for the regression is unknown.

-- 
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/20190104/09b506be/attachment.html>


More information about the llvm-bugs mailing list