| commit | 4e2f73ce3d70083dacb99815c9633af48aae32b4 | |
| author | Glauber Costa <glommer@redhat.com> | |
| Tue, 25 Aug 2009 22:13:36 +0000 (25 18:13 -0400) | ||
| committer | Anthony Liguori <aliguori@us.ibm.com> | |
| Fri, 4 Sep 2009 14:37:27 +0000 (4 09:37 -0500) | ||
| tree | 3c12b36ab4212f3c160c2f0270bae6491c57e1c9 | treesnapshot (tar.gz zip) |
| parent | 67b3b71d51844b9413278add178a49723a97aad9 | commitdiff |
do not issue ioctl from within the io thread
According to Documentation/kvm/api.txt, (and well, to common sense),
we should not be calling vcpu ioctls from within the iothread.
Since vcpu initialization issues a vcpu ioctl, move it a little bit
further in time to prevent it.
Signed-off-by: Glauber Costa <glommer@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
According to Documentation/kvm/api.txt, (and well, to common sense),
we should not be calling vcpu ioctls from within the iothread.
Since vcpu initialization issues a vcpu ioctl, move it a little bit
further in time to prevent it.
Signed-off-by: Glauber Costa <glommer@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
| vl.c | diffblobblamehistory |