diff mbox series

pc:piix4: Update smbus I/O space after a migration

Message ID 20181126182844.23109-1-minyard@acm.org (mailing list archive)
State New, archived
Headers show
Series pc:piix4: Update smbus I/O space after a migration | expand

Commit Message

Corey Minyard Nov. 26, 2018, 6:28 p.m. UTC
From: Corey Minyard <cminyard@mvista.com>

Otherwise it won't be set up correctly and won't work after
miigration.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Igor Mammedov <imammedo@redhat.com>
---
 hw/acpi/piix4.c | 1 +
 1 file changed, 1 insertion(+)

A rather obvious fix, in hindsight :).

Comments

Corey Minyard Nov. 29, 2018, 3:45 p.m. UTC | #1
On 11/26/18 12:28 PM, minyard@acm.org wrote:
> From: Corey Minyard <cminyard@mvista.com>
>
> Otherwise it won't be set up correctly and won't work after
> miigration.

Any opinions on this?  I think this would be good for 3.1. Without
this when a migration occurs on piix4 the SMBus address space
appears in the wrong place.  I'm not sure how bad that is, but
SMBus certainly ceases to work.

-corey


> Signed-off-by: Corey Minyard <cminyard@mvista.com>
> Cc: Michael S. Tsirkin <mst@redhat.com>
> Cc: Igor Mammedov <imammedo@redhat.com>
> ---
>   hw/acpi/piix4.c | 1 +
>   1 file changed, 1 insertion(+)
>
> A rather obvious fix, in hindsight :).
>
> diff --git a/hw/acpi/piix4.c b/hw/acpi/piix4.c
> index e330f24c71..2f4dd03b83 100644
> --- a/hw/acpi/piix4.c
> +++ b/hw/acpi/piix4.c
> @@ -173,6 +173,7 @@ static int vmstate_acpi_post_load(void *opaque, int version_id)
>       PIIX4PMState *s = opaque;
>   
>       pm_io_space_update(s);
> +    smbus_io_space_update(s);
>       return 0;
>   }
>
Michael S. Tsirkin Nov. 29, 2018, 3:49 p.m. UTC | #2
On Thu, Nov 29, 2018 at 09:45:17AM -0600, Corey Minyard wrote:
> On 11/26/18 12:28 PM, minyard@acm.org wrote:
> > From: Corey Minyard <cminyard@mvista.com>
> > 
> > Otherwise it won't be set up correctly and won't work after
> > miigration.
> 
> Any opinions on this?  I think this would be good for 3.1. Without
> this when a migration occurs on piix4 the SMBus address space
> appears in the wrong place.  I'm not sure how bad that is, but
> SMBus certainly ceases to work.
> 
> -corey


I have this queued. Thanks!

> 
> > Signed-off-by: Corey Minyard <cminyard@mvista.com>
> > Cc: Michael S. Tsirkin <mst@redhat.com>
> > Cc: Igor Mammedov <imammedo@redhat.com>
> > ---
> >   hw/acpi/piix4.c | 1 +
> >   1 file changed, 1 insertion(+)
> > 
> > A rather obvious fix, in hindsight :).
> > 
> > diff --git a/hw/acpi/piix4.c b/hw/acpi/piix4.c
> > index e330f24c71..2f4dd03b83 100644
> > --- a/hw/acpi/piix4.c
> > +++ b/hw/acpi/piix4.c
> > @@ -173,6 +173,7 @@ static int vmstate_acpi_post_load(void *opaque, int version_id)
> >       PIIX4PMState *s = opaque;
> >       pm_io_space_update(s);
> > +    smbus_io_space_update(s);
> >       return 0;
> >   }
>
Eric Blake Nov. 29, 2018, 4:43 p.m. UTC | #3
On 11/29/18 9:49 AM, Michael S. Tsirkin wrote:
> On Thu, Nov 29, 2018 at 09:45:17AM -0600, Corey Minyard wrote:
>> On 11/26/18 12:28 PM, minyard@acm.org wrote:
>>> From: Corey Minyard <cminyard@mvista.com>
>>>
>>> Otherwise it won't be set up correctly and won't work after
>>> miigration.
>>
>> Any opinions on this?  I think this would be good for 3.1. Without
>> this when a migration occurs on piix4 the SMBus address space
>> appears in the wrong place.  I'm not sure how bad that is, but
>> SMBus certainly ceases to work.
>>
>> -corey
> 
> 
> I have this queued. Thanks!

We've missed -rc3. Is the problem this patch addresses severe enough to 
warrant -rc4? Is it a regression since 3.0?
Peter Maydell Dec. 3, 2018, 3:14 p.m. UTC | #4
On Thu, 29 Nov 2018 at 16:44, Eric Blake <eblake@redhat.com> wrote:
>
> On 11/29/18 9:49 AM, Michael S. Tsirkin wrote:
> > On Thu, Nov 29, 2018 at 09:45:17AM -0600, Corey Minyard wrote:
> >> On 11/26/18 12:28 PM, minyard@acm.org wrote:
> >>> From: Corey Minyard <cminyard@mvista.com>
> >>>
> >>> Otherwise it won't be set up correctly and won't work after
> >>> miigration.
> >>
> >> Any opinions on this?  I think this would be good for 3.1. Without
> >> this when a migration occurs on piix4 the SMBus address space
> >> appears in the wrong place.  I'm not sure how bad that is, but
> >> SMBus certainly ceases to work.
> >>
> >> -corey
> >
> >
> > I have this queued. Thanks!
>
> We've missed -rc3. Is the problem this patch addresses severe enough to
> warrant -rc4? Is it a regression since 3.0?

The git history of the file suggests it's been like this for
at least five years, so I would suggest this is not worth
putting into -rc4 (we will have an rc4). Happy to hear
contrary views, but I don't want to bloat rc4 out more
than we can avoid...

thanks
-- PMM
diff mbox series

Patch

diff --git a/hw/acpi/piix4.c b/hw/acpi/piix4.c
index e330f24c71..2f4dd03b83 100644
--- a/hw/acpi/piix4.c
+++ b/hw/acpi/piix4.c
@@ -173,6 +173,7 @@  static int vmstate_acpi_post_load(void *opaque, int version_id)
     PIIX4PMState *s = opaque;
 
     pm_io_space_update(s);
+    smbus_io_space_update(s);
     return 0;
 }