Re: [libvirt] [PATCH] BlockJob: Support sync/async virDomainBlockJobAbort

2012-02-27 Thread Stefan Hajnoczi
On Fri, Jan 13, 2012 at 8:51 PM, Adam Litke a...@us.ibm.com wrote: Qemu has changed the semantics of the block_job_cancel API.  Originally, the operation was synchronous (ie. upon command completion, the operation was guaranteed to be completely stopped).  With the new semantics, a

Re: [libvirt] [PATCH] BlockJob: Support sync/async virDomainBlockJobAbort

2012-01-20 Thread Adam Litke
On Thu, Jan 19, 2012 at 05:39:43PM -0700, Eric Blake wrote: On 01/13/2012 01:51 PM, Adam Litke wrote: Qemu has changed the semantics of the block_job_cancel API. Originally, the operation was synchronous (ie. upon command completion, the operation was guaranteed to be completely

Re: [libvirt] [PATCH] BlockJob: Support sync/async virDomainBlockJobAbort

2012-01-19 Thread Adam Litke
On Wed, Jan 18, 2012 at 12:59:29PM -0500, Federico Simoncelli wrote: - Original Message - From: Adam Litke a...@us.ibm.com To: Dave Allan dal...@redhat.com Cc: libvir-list@redhat.com Sent: Friday, January 13, 2012 9:51:23 PM Subject: [libvirt] [PATCH] BlockJob: Support sync

Re: [libvirt] [PATCH] BlockJob: Support sync/async virDomainBlockJobAbort

2012-01-19 Thread Eric Blake
On 01/13/2012 01:51 PM, Adam Litke wrote: Qemu has changed the semantics of the block_job_cancel API. Originally, the operation was synchronous (ie. upon command completion, the operation was guaranteed to be completely stopped). With the new semantics, a block_job_cancel merely requests

Re: [libvirt] [PATCH] BlockJob: Support sync/async virDomainBlockJobAbort

2012-01-19 Thread Eric Blake
On 01/19/2012 02:00 PM, Adam Litke wrote: Without the VIR_DOMAIN_BLOCK_JOB_ABORT_ASYNC flag, libvirt will internally poll using qemu's query-block-jobs API and will not return until the operation has been completed. Why do you raise the event VIR_DOMAIN_BLOCK_JOB_CANCELLED also in the case

Re: [libvirt] [PATCH] BlockJob: Support sync/async virDomainBlockJobAbort

2012-01-18 Thread Federico Simoncelli
- Original Message - From: Adam Litke a...@us.ibm.com To: Dave Allan dal...@redhat.com Cc: libvir-list@redhat.com Sent: Friday, January 13, 2012 9:51:23 PM Subject: [libvirt] [PATCH] BlockJob: Support sync/async virDomainBlockJobAbort Qemu has changed the semantics

Re: [libvirt] [PATCH] BlockJob: Support sync/async virDomainBlockJobAbort

2012-01-16 Thread Dave Allan
On Fri, Jan 13, 2012 at 04:30:56PM -0600, Adam Litke wrote: On Fri, Jan 13, 2012 at 04:48:30PM -0500, Dave Allan wrote: On Fri, Jan 13, 2012 at 02:51:23PM -0600, Adam Litke wrote: Qemu has changed the semantics of the block_job_cancel API. Originally, the operation was synchronous

[libvirt] [PATCH] BlockJob: Support sync/async virDomainBlockJobAbort

2012-01-13 Thread Adam Litke
Qemu has changed the semantics of the block_job_cancel API. Originally, the operation was synchronous (ie. upon command completion, the operation was guaranteed to be completely stopped). With the new semantics, a block_job_cancel merely requests that the operation be cancelled and an event is

Re: [libvirt] [PATCH] BlockJob: Support sync/async virDomainBlockJobAbort

2012-01-13 Thread Dave Allan
On Fri, Jan 13, 2012 at 02:51:23PM -0600, Adam Litke wrote: Qemu has changed the semantics of the block_job_cancel API. Originally, the operation was synchronous (ie. upon command completion, the operation was guaranteed to be completely stopped). With the new semantics, a block_job_cancel

Re: [libvirt] [PATCH] BlockJob: Support sync/async virDomainBlockJobAbort

2012-01-13 Thread Adam Litke
On Fri, Jan 13, 2012 at 04:48:30PM -0500, Dave Allan wrote: On Fri, Jan 13, 2012 at 02:51:23PM -0600, Adam Litke wrote: Qemu has changed the semantics of the block_job_cancel API. Originally, the operation was synchronous (ie. upon command completion, the operation was guaranteed to be