On 10/10/2017 05:19 PM, Paolo Bonzini wrote: > On 10/10/2017 10:58, Marek Vasut wrote: >>> #0 0x00007ffff7931945 in pthread_cond_wait@@GLIBC_2.3.2 () at >>> /lib64/libpthread.so.0 >>> #1 0x00005555557cf9c5 in qemu_cond_wait (cond=cond@entry=0x555555b5f9c0 >>> <qemu_pause_cond>, mutex=mutex@entry=0x555555b5fa40 <qemu_global_mutex>) at >>> util/qemu-thread-posix.c:161 >>> #2 0x0000555555606ee8 in pause_all_vcpus () at cpus.c:1620 >>> #3 0x00005555555cc43a in main (argc=<optimized out>, argv=<optimized out>, >>> envp=<optimized out>) >>> at vl.c:4906 >>> >>> Looks like the CPU does not get stopped right? Unfortunately, I've got no >>> good clue how to debug that further right now... >> I'd look at pause_all_vcpus() and check why it is stuck. >> I'm a bit overloaded at the moment, so I don't have time to dig in it. >> > > It's actually the CPU thread that is stuck because the > start-of-translation-block test is wrong and never succeeds: > > 0x7fd715fff0c0: mov -0x14(%rsp),%ebp > 0x7fd715fff0c4: test %ebp,%ebp > 0x7fd715fff0c6: jl 0x7fd715fff0e4 > > It should be using %r14, not %rsp. I've sent a patch.
Oh nice, thank you :) -- Best regards, Marek Vasut
