The start of CPUState reset.hold happens before any subclasses, which can result in attempting to dump uninitialized data. In the case of Sparc, this will SIGSEGV.
r~ Richard Henderson (2): hw/core: Dump cpu_reset in the reset.exit phase hw/core: Use qemu_log_trylock/unlock in cpu_common_reset_exit hw/core/cpu-common.c | 21 ++++++++++++++++----- 1 file changed, 16 insertions(+), 5 deletions(-) -- 2.43.0
