On Ubuntu 22.04.1 x86_64 (with glibc 2.35), I run into:
...
(gdb) PASS: gdb.base/corefile.exp: $_exitcode is void
bt^M
#0 __pthread_kill_implementation (...) at ./nptl/pthread_kill.c:44^M
#1 __pthread_kill_internal (...) at ./nptl/pthread_kill.c:78^M
#2 __GI___pthread_kill (...) at ./nptl/pthread_kill.c:89^M
#3 0x00007f4985e1a476 in __GI_raise (...) at ../sysdeps/posix/raise.c:26^M
#4 0x00007f4985e007f3 in __GI_abort () at ./stdlib/abort.c:79^M
#5 0x0000556b4ea4b504 in func2 () at gdb.base/coremaker.c:153^M
#6 0x0000556b4ea4b516 in func1 () at gdb.base/coremaker.c:159^M
#7 0x0000556b4ea4b578 in main (...) at gdb.base/coremaker.c:171^M
(gdb) PASS: gdb.base/corefile.exp: backtrace
up^M
#1 __pthread_kill_internal (...) at ./nptl/pthread_kill.c:78^M
78 in ./nptl/pthread_kill.c^M
(gdb) FAIL: gdb.base/corefile.exp: up
...
The problem is that the regexp used here:
...
gdb_test "up" "#\[0-9\]* *\[0-9xa-fH'\]* in .* \\(.*\\).*" "up"
...
does not fit the __pthread_kill_internal line which lacks the instruction
address due to inlining.
Fix this by making the regexp less strict.
Tested on x86_64-linux.
# file correctly. I don't think the other tests do this.
gdb_test "bt" "abort.*func2.*func1.*main.*" "backtrace"
-gdb_test "up" "#\[0-9\]* *\[0-9xa-fH'\]* in .* \\(.*\\).*" "up"
+gdb_test "up" "#\[0-9\]* *(\[0-9xa-fH'\]* in)? .* \\(.*\\).*" "up"
# Test ability to read mmap'd data
# test reinit_frame_cache
gdb_load ${binfile}
-gdb_test "up" "#\[0-9\]* *\[0-9xa-fH'\]* in .* \\(.*\\).*" "up (reinit)"
+gdb_test "up" "#\[0-9\]* *(\[0-9xa-fH'\]* in)? .* \\(.*\\).*" "up (reinit)"
gdb_test "core" "No core file now."