Patchwork [PATCHv2,2/6] cpus: flush all requests on each vm stop

login
register
mail settings
Submitter Michael S. Tsirkin
Date Nov. 30, 2010, 2:05 p.m.
Message ID <20101130140531.GA20536@redhat.com>
Download mbox | patch
Permalink /patch/73776/
State New
Headers show

Comments

Michael S. Tsirkin - Nov. 30, 2010, 2:05 p.m.
On Tue, Nov 30, 2010 at 11:46:48AM -0200, Marcelo Tosatti wrote:
> On Tue, Nov 30, 2010 at 03:34:29PM +0200, Michael S. Tsirkin wrote:
> > On Tue, Nov 30, 2010 at 10:45:40AM -0200, Marcelo Tosatti wrote:
> > > On Wed, Nov 24, 2010 at 05:52:58PM +0200, Michael S. Tsirkin wrote:
> > > > Make sure disk is in consistent state.
> > > > 
> > > > Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
> > > > Tested-by: Jason Wang <jasowang@redhat.com>
> > > > ---
> > > >  cpus.c |    2 ++
> > > >  1 files changed, 2 insertions(+), 0 deletions(-)
> > > > 
> > > > diff --git a/cpus.c b/cpus.c
> > > > index 91a0fb1..d421a96 100644
> > > > --- a/cpus.c
> > > > +++ b/cpus.c
> > > > @@ -110,6 +110,8 @@ static void do_vm_stop(int reason)
> > > >          cpu_disable_ticks();
> > > >          vm_running = 0;
> > > >          pause_all_vcpus();
> > > > +        qemu_aio_flush();
> > > > +        bdrv_flush_all();
> > > 
> > > Can you move these after vm_state_notify? qemu-kvm stops vcpus there.
> > > 
> > > >          vm_state_notify(0, reason);
> > > >          monitor_protocol_event(QEVENT_STOP, NULL);
> > > >      }
> > 
> > Like this:
> > 
> > cpus: flush all requests on each vm stop
> > 
> > Make sure disk is in consistent state.
> > 
> > Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
> > Tested-by: Jason Wang <jasowang@redhat.com>
> > 
> > ---
> > 
> > diff --git a/cpus.c b/cpus.c
> > index 91a0fb1..d421a96 100644
> > --- a/cpus.c
> > +++ b/cpus.c
> > @@ -110,6 +110,8 @@ static void do_vm_stop(int reason)
> >          cpu_disable_ticks();
> >          vm_running = 0;
> >          pause_all_vcpus();
> > +        qemu_aio_flush();
> > +        bdrv_flush_all();
> >          vm_state_notify(0, reason);
> >          monitor_protocol_event(QEVENT_STOP, NULL);
> >      }
> 
> No, after vm_state_notify.


Like this then:


    cpus: flush all requests on each vm stop
    
    Flush all requests once we have stopped all
    cpus and devices.
    Make sure disk is in consistent state.
    
    Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
    Tested-by: Jason Wang <jasowang@redhat.com>
Marcelo Tosatti - Dec. 3, 2010, 4:30 p.m.
On Tue, Nov 30, 2010 at 04:05:31PM +0200, Michael S. Tsirkin wrote:
> On Tue, Nov 30, 2010 at 11:46:48AM -0200, Marcelo Tosatti wrote:
> > On Tue, Nov 30, 2010 at 03:34:29PM +0200, Michael S. Tsirkin wrote:
> > > On Tue, Nov 30, 2010 at 10:45:40AM -0200, Marcelo Tosatti wrote:
> > > > On Wed, Nov 24, 2010 at 05:52:58PM +0200, Michael S. Tsirkin wrote:
> > > > > Make sure disk is in consistent state.
> > > > > 
> > > > > Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
> > > > > Tested-by: Jason Wang <jasowang@redhat.com>
> > > > > ---
> > > > >  cpus.c |    2 ++
> > > > >  1 files changed, 2 insertions(+), 0 deletions(-)
> > > > > 
> > > > > diff --git a/cpus.c b/cpus.c
> > > > > index 91a0fb1..d421a96 100644
> > > > > --- a/cpus.c
> > > > > +++ b/cpus.c
> > > > > @@ -110,6 +110,8 @@ static void do_vm_stop(int reason)
> > > > >          cpu_disable_ticks();
> > > > >          vm_running = 0;
> > > > >          pause_all_vcpus();
> > > > > +        qemu_aio_flush();
> > > > > +        bdrv_flush_all();
> > > > 
> > > > Can you move these after vm_state_notify? qemu-kvm stops vcpus there.
> > > > 
> > > > >          vm_state_notify(0, reason);
> > > > >          monitor_protocol_event(QEVENT_STOP, NULL);
> > > > >      }
> > > 
> > > Like this:
> > > 
> > > cpus: flush all requests on each vm stop
> > > 
> > > Make sure disk is in consistent state.
> > > 
> > > Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
> > > Tested-by: Jason Wang <jasowang@redhat.com>
> > > 
> > > ---
> > > 
> > > diff --git a/cpus.c b/cpus.c
> > > index 91a0fb1..d421a96 100644
> > > --- a/cpus.c
> > > +++ b/cpus.c
> > > @@ -110,6 +110,8 @@ static void do_vm_stop(int reason)
> > >          cpu_disable_ticks();
> > >          vm_running = 0;
> > >          pause_all_vcpus();
> > > +        qemu_aio_flush();
> > > +        bdrv_flush_all();
> > >          vm_state_notify(0, reason);
> > >          monitor_protocol_event(QEVENT_STOP, NULL);
> > >      }
> > 
> > No, after vm_state_notify.
> 
> 
> Like this then:
> 
> 
>     cpus: flush all requests on each vm stop
>     
>     Flush all requests once we have stopped all
>     cpus and devices.
>     Make sure disk is in consistent state.
>     
>     Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
>     Tested-by: Jason Wang <jasowang@redhat.com>
> 
> diff --git a/cpus.c b/cpus.c
> index 91a0fb1..0309189 100644
> --- a/cpus.c
> +++ b/cpus.c
> @@ -111,6 +111,8 @@ static void do_vm_stop(int reason)
>          vm_running = 0;
>          pause_all_vcpus();
>          vm_state_notify(0, reason);
> +        qemu_aio_flush();
> +        bdrv_flush_all();
>          monitor_protocol_event(QEVENT_STOP, NULL);
>      }
>  }

ACK.

Patch

diff --git a/cpus.c b/cpus.c
index 91a0fb1..0309189 100644
--- a/cpus.c
+++ b/cpus.c
@@ -111,6 +111,8 @@  static void do_vm_stop(int reason)
         vm_running = 0;
         pause_all_vcpus();
         vm_state_notify(0, reason);
+        qemu_aio_flush();
+        bdrv_flush_all();
         monitor_protocol_event(QEVENT_STOP, NULL);
     }
 }