Re: [Qemu-devel] [PULL v2 00/34] Misc patches for 2019-08-20

[Peter Maydell]

On Tue, 20 Aug 2019 at 16:28, Paolo Bonzini <address@hidden> wrote:
>
> The following changes since commit 17dc57990320edaad52ac9ea808be9719c91cea6:
>
>   Merge remote-tracking branch 
> 'remotes/huth-gitlab/tags/pull-request-2019-08-20' into staging (2019-08-20 
> 14:14:20 +0100)
>
> are available in the git repository at:
>
>
>   git://github.com/bonzini/qemu.git tags/for-upstream
>
> for you to fetch changes up to f496f8aca311b0b1fab979657694069871426395:
>
>   x86: Intel AVX512_BF16 feature enabling (2019-08-20 17:26:30 +0200)
>
> ----------------------------------------------------------------
> * New KVM PV features (Marcelo, Wanpeng)
> * valgrind fixes (Andrey)
> * Remove clock reset notifiers (David)
> * KConfig and Makefile cleanups (Paolo)
> * Replay and icount improvements (Pavel)
> * x86 FP fixes (Peter M.)
> * TCG locking assertions (Roman)
> * x86 support for mmap-ed -kernel/-initrd (Stefano)
> * Other cleanups (Wei Yang, Yan Zhao, Tony)
> * LSI fix for infinite loop (Prasad)
> * ARM migration fix (Catherine)
> * AVX512_BF16 feature (Jing)
>
> ----------------------------------------------------------------

The linux-user build seems to fail 'make check-tcg':
e104462:bionic:qemu-for-merges$ make -C build/all-linux-static/ check-tcg
make: Entering directory
'/home/petmay01/linaro/qemu-for-merges/build/all-linux-static'
make[1]: Entering directory '/home/petmay01/linaro/qemu-for-merges/slirp'
make[1]: Nothing to be done for 'all'.
make[1]: Leaving directory '/home/petmay01/linaro/qemu-for-merges/slirp'
  BUILD   debian10
  BUILD   debian-buster-arm64-cross
  BUILD   TCG tests for aarch64-linux-user
  BUILD   aarch64 guest-tests with aarch64-linux-gnu-gcc
  RUN     TCG tests for aarch64-linux-user
  BUILD   aarch64 guest-tests with aarch64-linux-gnu-gcc
  RUN     tests for aarch64
  TEST    test-mmap (default) on aarch64
qemu-aarch64: /home/petmay01/linaro/qemu-for-merges/cpus-common.c:219:
cpu_exec_start: Assertion `!qemu_mutex_iothread_locked()' failed.
/home/petmay01/linaro/qemu-for-merges/tests/tcg/multiarch/Makefile.target:30:
recipe for target 'run-test-mmap' failed
make[2]: *** [run-test-mmap] Error 127
/home/petmay01/linaro/qemu-for-merges/tests/tcg/Makefile.include:71:
recipe for target 'run-guest-tests' failed
make[1]: *** [run-guest-tests] Error 2
/home/petmay01/linaro/qemu-for-merges/tests/Makefile.include:1075:
recipe for target 'run-tcg-tests-aarch64-linux-user' failed
make: *** [run-tcg-tests-aarch64-linux-user] Error 2
make: Leaving directory
'/home/petmay01/linaro/qemu-for-merges/build/all-linux-static'

Looks like we hit the assertion immediately on startup before
we try to do anything:

Thread 2 (Thread 0x7ffff7ff9700 (LWP 9252)):
#0  0x0000000060340b49 in syscall ()
#1  0x00000000601492af in qemu_futex_wait (f=0x628c9838
<rcu_call_ready_event>, val=4294967295)
    at /home/petmay01/linaro/qemu-for-merges/include/qemu/futex.h:29
#2  0x0000000060149476 in qemu_event_wait (ev=0x628c9838 <rcu_call_ready_event>)
    at /home/petmay01/linaro/qemu-for-merges/util/qemu-thread-posix.c:442
#3  0x0000000060152794 in call_rcu_thread (opaque=0x0) at
/home/petmay01/linaro/qemu-for-merges/util/rcu.c:260
#4  0x0000000060149629 in qemu_thread_start (args=0x628d3f20) at
/home/petmay01/linaro/qemu-for-merges/util/qemu-thread-posix.c:502
#5  0x00000000602b19fb in start_thread (arg=0x7ffff7ff9700) at
pthread_create.c:463
#6  0x000000006034292f in clone ()

Thread 1 (Thread 0x628d2940 (LWP 9248)):
#0  0x00000000602d52b7 in raise ()
#1  0x00000000602d5951 in abort ()
#2  0x00000000602c884c in __assert_fail_base ()
#3  0x00000000602c88c2 in __assert_fail ()
#4  0x00000000600f2438 in cpu_exec_start (cpu=0x628f5cd0) at
/home/petmay01/linaro/qemu-for-merges/cpus-common.c:219
#5  0x0000000060094fc7 in cpu_loop (env=0x628fdfa0) at
/home/petmay01/linaro/qemu-for-merges/linux-user/i386/cpu_loop.c:93
#6  0x0000000060064f2c in main (argc=6, argv=0x7fffffffe3b8,
envp=0x7fffffffe3f0)
    at /home/petmay01/linaro/qemu-for-merges/linux-user/main.c:838

This is because for linux-user we're using the stubs/ version
of qemu_mutex_iothread_locked(), which always returns 'true',
so the assertion can never pass...

thanks
-- PMM