Re: [Qemu-devel] intermittent make check failure: "tcg_handle_interrupt: assertion failed: (qemu_mutex_iothread_locked())"

[Thomas Huth <thuth@redhat.com>]

On 28.02.2017 20:30, Thomas Huth wrote:
> On 28.02.2017 20:10, Peter Maydell wrote:
>> I got a make check failure on aarch64 host running a sparc64 test:
>>
>> TEST: tests/prom-env-test... (pid=13573)
>>   /sparc64/prom-env/sun4u:                                             **
>> ERROR:/home/pm215/qemu/translate-common.c:34:tcg_handle_interrupt:
>> assertion failed: (qemu_mutex_iothread_locked())
>> Broken pipe
>> FAIL
>> GTester: last random seed: R02Sa5fa983185fe5e65cfb5b7fcb39ed3d1
>> (pid=13579)
>> FAIL: tests/prom-env-test
>>
>> This is with commit 9514f2648ca05b3.
>>
>> It didn't reproduce on a rerun of 'make check' and it didn't
>> look like the merge in question was particularly relevant to
>> the failure, so I went ahead and pushed it to master.
>>
>> I'm (perhaps unfairly) assuming that this is fallout from MTTCG
>> landing. (prom-env-test is one of the handful of tests in
>> 'make check' that actually runs TCG code.)
> 
> Maybe worth mentioning that the prom-env test for sparc64 has just been
> enabled a couple of commits ago (6b591ad613010f13). There was an issue
> with that test for the sparc64 target last year, but IIRC it only
> affected 32-bit hosts and has been fixed, so I think/hope this is a
> different issue now...

OK, I think this is definitively not related to the prom-env test, but
to MTTCG (or something else in TCG): I can reproduce this problem
reliably when I run QEMU in s390 mode with the Moon Buggy disk image
from the QEMU advent calendar
(http://www.qemu-advent-calendar.org/2014/download/s390-moon-buggy.tar.xz):

qemu-system-s390x -nographic -kernel s390-bb.kernel \
                  -initrd s390-moon-buggy.initrd

... starts the kernel, but aborts very soon with:

ERROR:qemu/translate-common.c:34:tcg_handle_interrupt: assertion failed:
(qemu_mutex_iothread_locked())
Aborted (core dumped)

 Thomas