[Bug libc/24040] New: riscv64: unterminated call chain in __thread_start

[wilson at gcc dot gnu.org]

https://sourceware.org/bugzilla/show_bug.cgi?id=24040

            Bug ID: 24040
           Summary: riscv64: unterminated call chain in __thread_start
           Product: glibc
           Version: 2.30
            Status: NEW
          Severity: normal
          Priority: P2
         Component: libc
          Assignee: unassigned at sourceware dot org
          Reporter: wilson at gcc dot gnu.org
                CC: drepper.fsp at gmail dot com
  Target Milestone: ---

This looks like the same bug as 23125, except in __thread_start instead of
_start.  Presumably the same fix for 23125 will work, except applied to the
file sysdeps/unix/sysv/linux/riscv/clone.S instead of sysdeps/riscv/start.S.

I found this while working on elfutils.  With the patches I submitted today,
running the elfutils testsuite, I get

FAIL: run-backtrace-native-core.sh
==================================

/usr/bin/coredumpctl
0x200001d000    0x200001e000    linux-vdso.so.1
0x2000000000    0x200001c168    ld-linux-riscv64-lp64d.so.1
0x200004d000    0x2000175208    libc.so.6
0x2000031000    0x200004c470    libpthread.so.0
0x2aaaaaa000    0x2aaaaad0c0    backtrace-child
TID 1581:
# 0 0x2000040394        raise
# 1 0x2aaaaaabbe - 1    sigusr2
# 2 0x2aaaaaaca2 - 1    stdarg
# 3 0x2aaaaaacb0 - 1    backtracegen
# 4 0x2aaaaaacb8 - 1    start
# 5 0x2000037e9a - 1    start_thread
# 6 0x20000f472e - 1    __thread_start
# 7 0x20000f472e - 1    __thread_start
# 8 0x20000f472e - 1    __thread_start
# 9 0x20000f472e - 1    __thread_start
#10 0x20000f472e - 1    __thread_start
#11 0x20000f472e - 1    __thread_start
#12 0x20000f472e - 1    __thread_start
#13 0x20000f472e - 1    __thread_start
#14 0x20000f472e - 1    __thread_start
#15 0x20000f472e - 1    __thread_start
#16 0x20000f472e - 1    __thread_start
/home/jimw/FOSS/elfutils/X-hifiveu017/tests/backtrace: Too many frames: 17

Incidentally, the run-backtrace-native.sh test fails on a system without the
fix for 23125, failing the same way, except with a repeating _start at the end.

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