[Bug ld/19264] New: [2.26/2.27 Regression] ld hangs (indefinite loop) when building util-linux test case on powerpc64le-linux-gnu

[doko at debian dot org]

https://sourceware.org/bugzilla/show_bug.cgi?id=19264

            Bug ID: 19264
           Summary: [2.26/2.27 Regression] ld hangs (indefinite loop) when
                    building util-linux test case on powerpc64le-linux-gnu
           Product: binutils
           Version: 2.26
            Status: NEW
          Severity: normal
          Priority: P2
         Component: ld
          Assignee: unassigned at sourceware dot org
          Reporter: doko at debian dot org
  Target Milestone: ---

Created attachment 8794
  --> https://sourceware.org/bugzilla/attachment.cgi?id=8794&action=edit
test case

seen when trying to run the test suite for util-linux on powerpc64le-linux-gnu
using binutils from the 2.26 branch.

from the downstream bug report:

"""
strace shows it repeatedly calling brk(), so presumably trying to allocate more
memory in very small increments.

It doesn't appear to be lto-related. The symptom is reproducible when calling
gcc with -fno-lto.

I do not yet have a useful backtrace from ld for this, because for some reason
recent versions of binutils are not getting dbgsym packages in the archive.

Rebuilding with nostrip, I get a backtrace that looks like this, with at least
partial symbols.
(gdb) bt
#0 0x00003fffb3b8885c in vfprintf () from /lib/powerpc64le-linux-gnu/libc.so.6
#1 0x00003fffb3c714b8 in __vsprintf_chk ()
   from /lib/powerpc64le-linux-gnu/libc.so.6
#2 0x00003fffb3c713ac in __sprintf_chk ()
   from /lib/powerpc64le-linux-gnu/libc.so.6
#3 0x00003fffb3daa338 in ?? ()
#4 0x00003fffb3db963c in ?? ()
#5 0x0000000010032168 in gldelf64lppc_find_exp_assignment (
    exp=<optimized out>) at eelf64lppc.c:1845
#6 0x000000001002824c in ldemul_hll (name=<optimized out>)
    at ../../ld/ldemul.c:39
Backtrace stopped: frame did not save the PC
(gdb)
"""

-- 
You are receiving this mail because:
You are on the CC list for the bug.