On 11/02/12 18:59, Eric Blake wrote:
On 11/01/2012 10:22 AM, Peter Krempa wrote:
When pausing the guest while migration is running (to speed up convergence) the virDomainSuspend API checks if the migration job is active before entering the job. This could cause a possible race if the virDomainSuspend is called while the job is active but ends before the Suspend API enters the job (this would require that the migration is aborted). This would cause a incorrect event to be emitted.
Not just aborted, but also completed normally. At any rate, I agree that we shouldn't be relying on the state of job.asyncJob until we have obtained the sync job.
--- src/qemu/qemu_driver.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-)
[...]
In that case, this code is fixing a pre-existing bug, in preparation for the work in patch 10. ACK as-is.
Thanks. I pushed this one. I should have ordered it at the beginning of the series to avoid confusion. Peter