Re: armv7-on-aarch64 stuck at urdlck: I got a replication of the "ampere2" bulk build hangup problem on a Windows DevKit 2023

On Jul 20, 2024, at 01:57, Konstantin Belousov <[email protected]> wrote:

> [Everything and everybody in Cc: are stripped for good].
> 
> On Fri, Jul 19, 2024 at 10:38:36PM -0700, Mark Millard wrote:
>> 0x201375c0 - 0x2014092c is .bss in /lib/libthr.so.3
>> 
>> (gdb) bt
>> #0  0x201aeec0 in __pthread_map_stacks_exec () from /lib/libc.so.7
>> #1  0x2005d1e4 in ?? () from /libexec/ld-elf.so.1
>> Backtrace stopped: previous frame identical to this frame (corrupt stack?)
>> (gdb) disass
>> Dump of assembler code for function __pthread_map_stacks_exec:
>> => 0x201aeec0 <+0>: ldr r0, [pc, #8] @ 0x201aeed0 <__pthread_map_stacks_exec+16>
>>   0x201aeec4 <+4>: add r0, pc, r0
>>   0x201aeec8 <+8>: ldr r0, [r0, #156] @ 0x9c
>>   0x201aeecc <+12>: bx r0
>>   0x201aeed0 <+16>: andseq r6, r7, r4, lsr #12
>> End of assembler dump.
>> 
> 
> Do the following:
> 1. Rebuild rtld/libc/libthr with the debugging info and no optimization,
>   i.e. ensure that flags are "-O0 -g" or "-Og -g" and not -O2.  See
>   the first comment in libexec/rtld-elf/Makefile for the hint how to
>   do it.

I did a full buildworld with "-Og -g" via temporary
use of:

diff --git a/share/mk/sys.mk b/share/mk/sys.mk
index 44db9266784f..9c6c7ce575a4 100644
--- a/share/mk/sys.mk
+++ b/share/mk/sys.mk
@@ -145,7 +145,8 @@ CC          ?=      c89
CFLAGS         ?=      -O
.else
CC             ?=      cc
-CFLAGS         ?=      -O2 -pipe
+#CFLAGS                ?=      -O2 -pipe
+CFLAGS         ?=      -Og -g -pipe
.if defined(NO_STRICT_ALIASING)
CFLAGS         +=      -fno-strict-aliasing
.endif

I installed the result armv7 world into a
directory tree and installed pkg and cairo.

> 2. Reproduce the issue

The dlopen_test.c based case does not fail under the world
built with "-Og -g":

# cc -g -std=c11 -pedantic -Wall -pthread dlopen_test.c ; ./a.out
# 

> under gdb

(gdb) run
Starting program: /root/a.out  [Inferior 1 (process 36680) exited normally]
(gdb) 

So it does not reproduce in gdb when buildworld was based
on "-Og -g".

> , and backtrace all threads from userspace.
>   I only need userspace backtrace, not either kernel-side stacks nor
>   the syscall history.
> 
> Are you sure that the issue is specific to armv7, might be it takes more
> efforts to reproduce on host native?


===
Mark Millard
marklmi at yahoo.com