On Wed, 22 Aug 2018 17:56:12 +0200 Christian Borntraeger <borntrae...@de.ibm.com> wrote:
> This breaks qemu-io test for me. > > #0 0x000003ff98f3e2d4 in raise () at /lib64/libc.so.6 > #1 0x000003ff98f239a8 in abort () at /lib64/libc.so.6 > #2 0x000003ff98f3632e in __assert_fail_base () at /lib64/libc.so.6 > #3 0x000003ff98f363ac in () at /lib64/libc.so.6 > #4 0x000000000108301a in qemu_balloon_inhibit (state=state@entry=false) at > /home/cborntra/REPOS/qemu/balloon.c:56 > #5 0x00000000012026f2 in postcopy_ram_incoming_cleanup > (mis=mis@entry=0x163bc5c0) at > /home/cborntra/REPOS/qemu/migration/postcopy-ram.c:542 > #6 0x00000000011eed5e in process_incoming_migration_co (opaque=<optimized > out>) at /home/cborntra/REPOS/qemu/migration/migration.c:407 > #7 0x0000000001374000 in coroutine_trampoline (i0=<optimized out>, > i1=<optimized out>) at /home/cborntra/REPOS/qemu/util/coroutine-ucontext.c:116 > #8 0x000003ff98f53c02 in __makecontext_ret () at /lib64/libc.so.6 > > Cant see right now whats wrong. I'd have to guess it's the assert added to verify increments and decrements are balanced, which suggests postcopy is de-inhibiting more than it's inhibiting :-\ Thanks, Alex
