[RFC,04/52] i386/WHPX: Fix error message when fail to set ProcessorCount

From: Zhao Liu <zhao1.liu@intel.com>

003f230 (machine: Tweak the order of topology members in struct
CpuTopology) changes the meaning of MachineState.smp.cores from "the
number of cores in one package" to "the number of cores in one die"
and doesn't fix other uses of MachineState.smp.cores. And because of
the introduction of cluster, now smp.cores just means "the number of
cores in one cluster". This clearly does not fit the semantics here.

And before this error message, WHvSetPartitionProperty() is called to
set prop.ProcessorCount.

So the error message should show the prop.ProcessorCount other than
"cores per cluster" or "cores per package".

Cc: Sunil Muthuswamy <sunilmut@microsoft.com>
Signed-off-by: Zhao Liu <zhao1.liu@intel.com>
---
 target/i386/whpx/whpx-all.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

On Mon, Feb 13, 2023 at 05:49:47PM +0800, Zhao Liu wrote:
> From: Zhao Liu <zhao1.liu@intel.com>
> 
> 003f230 (machine: Tweak the order of topology members in struct
> CpuTopology) changes the meaning of MachineState.smp.cores from "the
> number of cores in one package" to "the number of cores in one die"
> and doesn't fix other uses of MachineState.smp.cores. And because of
> the introduction of cluster, now smp.cores just means "the number of
> cores in one cluster". This clearly does not fit the semantics here.
> 
> And before this error message, WHvSetPartitionProperty() is called to
> set prop.ProcessorCount.
> 
> So the error message should show the prop.ProcessorCount other than
> "cores per cluster" or "cores per package".
> 
> Cc: Sunil Muthuswamy <sunilmut@microsoft.com>
> Signed-off-by: Zhao Liu <zhao1.liu@intel.com>
> ---
>  target/i386/whpx/whpx-all.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)

This patch and the 3 patches before it all look like basic bug
fixes to current code, not really part of the new hybrid topology
feature work.

As such I'd suggest sending these first four patches as a separate
series, so the bug fixes can be merged fairly quickly. I expect the
rest of the hybrid topology series is going to take a long time to
get agreement on, so no need to delay the easy bug fixes.

> 
> diff --git a/target/i386/whpx/whpx-all.c b/target/i386/whpx/whpx-all.c
> index e738d83e8191..fc349f887e47 100644
> --- a/target/i386/whpx/whpx-all.c
> +++ b/target/i386/whpx/whpx-all.c
> @@ -2613,8 +2613,8 @@ static int whpx_accel_init(MachineState *ms)
>          sizeof(WHV_PARTITION_PROPERTY));
>  
>      if (FAILED(hr)) {
> -        error_report("WHPX: Failed to set partition core count to %d,"
> -                     " hr=%08lx", ms->smp.cores, hr);
> +        error_report("WHPX: Failed to set partition processor count to %d,"
> +                     " hr=%08lx", prop.ProcessorCount, hr);
>          ret = -EINVAL;
>          goto error;
>      }
> -- 
> 2.34.1
> 
> 

With regards,
Daniel

On Mon, Feb 13, 2023 at 01:41:19PM +0000, Daniel P. Berrangé wrote:
> Date: Mon, 13 Feb 2023 13:41:19 +0000
> From: "Daniel P. Berrangé" <berrange@redhat.com>
> Subject: Re: [RFC 04/52] i386/WHPX: Fix error message when fail to set
>  ProcessorCount
> 
> On Mon, Feb 13, 2023 at 05:49:47PM +0800, Zhao Liu wrote:
> > From: Zhao Liu <zhao1.liu@intel.com>
> > 
> > 003f230 (machine: Tweak the order of topology members in struct
> > CpuTopology) changes the meaning of MachineState.smp.cores from "the
> > number of cores in one package" to "the number of cores in one die"
> > and doesn't fix other uses of MachineState.smp.cores. And because of
> > the introduction of cluster, now smp.cores just means "the number of
> > cores in one cluster". This clearly does not fit the semantics here.
> > 
> > And before this error message, WHvSetPartitionProperty() is called to
> > set prop.ProcessorCount.
> > 
> > So the error message should show the prop.ProcessorCount other than
> > "cores per cluster" or "cores per package".
> > 
> > Cc: Sunil Muthuswamy <sunilmut@microsoft.com>
> > Signed-off-by: Zhao Liu <zhao1.liu@intel.com>
> > ---
> >  target/i386/whpx/whpx-all.c | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> This patch and the 3 patches before it all look like basic bug
> fixes to current code, not really part of the new hybrid topology
> feature work.
> 
> As such I'd suggest sending these first four patches as a separate
> series, so the bug fixes can be merged fairly quickly. I expect the
> rest of the hybrid topology series is going to take a long time to
> get agreement on, so no need to delay the easy bug fixes.

Yes, I'll split out the first four. Thanks!

Zhao

> 
> > 
> > diff --git a/target/i386/whpx/whpx-all.c b/target/i386/whpx/whpx-all.c
> > index e738d83e8191..fc349f887e47 100644
> > --- a/target/i386/whpx/whpx-all.c
> > +++ b/target/i386/whpx/whpx-all.c
> > @@ -2613,8 +2613,8 @@ static int whpx_accel_init(MachineState *ms)
> >          sizeof(WHV_PARTITION_PROPERTY));
> >  
> >      if (FAILED(hr)) {
> > -        error_report("WHPX: Failed to set partition core count to %d,"
> > -                     " hr=%08lx", ms->smp.cores, hr);
> > +        error_report("WHPX: Failed to set partition processor count to %d,"
> > +                     " hr=%08lx", prop.ProcessorCount, hr);
> >          ret = -EINVAL;
> >          goto error;
> >      }
> > -- 
> > 2.34.1
> > 
> > 
> 
> With regards,
> Daniel
> -- 
> |: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
> |: https://libvirt.org         -o-            https://fstop138.berrange.com :|
> |: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|
>