diff options
author | Thomas Huth <thuth@redhat.com> | 2016-02-18 22:01:38 +0100 |
---|---|---|
committer | David Gibson <david@gibson.dropbear.id.au> | 2016-02-25 13:58:44 +1100 |
commit | 072ed5f260865c8aac8c7cc0f23353b8d5753580 (patch) | |
tree | 952d7c905b636f4e5e6d25c61ee16ef14190bfa8 /.exrc | |
parent | 9897e462645fb8d91a4cef2a1ea3a74cc9149c13 (diff) | |
download | qemu-072ed5f260865c8aac8c7cc0f23353b8d5753580.zip qemu-072ed5f260865c8aac8c7cc0f23353b8d5753580.tar.gz qemu-072ed5f260865c8aac8c7cc0f23353b8d5753580.tar.bz2 |
ppc/kvm: Use error_report() instead of cpu_abort() for user-triggerable errors
Setting the KVM_CAP_PPC_PAPR capability can fail if either the KVM
kernel module does not support it, or if the specified vCPU type
is not a 64-bit Book3-S CPU type. For example, the user can trigger
it easily with "-M pseries -cpu G2leLS" when using the kvm-pr kernel
module. So the error should not be reported with cpu_abort() since
this function is rather meant for reporting programming errors than
reporting user-triggerable errors (it prints out all CPU registers
and then calls abort() to kills the program - two things that the
normal user does not expect here) . So let's use error_report() with
exit(1) here instead.
A similar problem exists in the code that sets the KVM_CAP_PPC_EPR
capability, so while we're at it, fix that, too.
Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Diffstat (limited to '.exrc')
0 files changed, 0 insertions, 0 deletions