[PULL 0/6] ppc-for-9.2-2 queue

Nicholas Piggin posted 6 patches 1 week, 3 days ago
hw/pci-host/mv64361.c       |  1 +
hw/ppc/pegasos2.c           | 30 +++++++++++++++++++++++++++++-
hw/ppc/pnv_core.c           | 11 +++++++++--
hw/ppc/pnv_nest_pervasive.c |  2 +-
hw/ppc/spapr.c              |  2 +-
hw/ppc/spapr_cpu_core.c     |  1 +
target/ppc/cpu.h            |  8 ++++++--
target/ppc/excp_helper.c    |  7 +++++++
8 files changed, 55 insertions(+), 7 deletions(-)
[PULL 0/6] ppc-for-9.2-2 queue
Posted by Nicholas Piggin 1 week, 3 days ago
The following changes since commit ba54a7e6b86884e43bed2d2f5a79c719059652a8:

  Merge tag 'net-pull-request' of https://github.com/jasowang/qemu into staging (2024-11-26 14:06:40 +0000)

are available in the Git repository at:

  https://gitlab.com/npiggin/qemu.git tags/pull-ppc-for-9.2-2-20241127

for you to fetch changes up to 0805136a44d39adc2467f23ac3c65e680e45d0a2:

  hw/ppc/pegasos2: Fix IRQ routing from pci.0 (2024-11-27 02:49:36 +1000)

----------------------------------------------------------------
* Assorted small ppc fixes

----------------------------------------------------------------
BALATON Zoltan (1):
      hw/ppc/pegasos2: Fix IRQ routing from pci.0

Glenn Miles (1):
      target/ppc: Fix THREAD_SIBLING_FOREACH for multi-socket

Harsh Prateek Bora (1):
      ppc/spapr: fix drc index mismatch for partially enabled vcpus

Nicholas Piggin (3):
      target/ppc: Fix non-maskable interrupt while halted
      ppc/pnv: Fix direct controls quiesce
      ppc/pnv: Add xscom- prefix to pervasive-control region name

 hw/pci-host/mv64361.c       |  1 +
 hw/ppc/pegasos2.c           | 30 +++++++++++++++++++++++++++++-
 hw/ppc/pnv_core.c           | 11 +++++++++--
 hw/ppc/pnv_nest_pervasive.c |  2 +-
 hw/ppc/spapr.c              |  2 +-
 hw/ppc/spapr_cpu_core.c     |  1 +
 target/ppc/cpu.h            |  8 ++++++--
 target/ppc/excp_helper.c    |  7 +++++++
 8 files changed, 55 insertions(+), 7 deletions(-)
Re: [PULL 0/6] ppc-for-9.2-2 queue
Posted by Michael Tokarev 1 week, 2 days ago
26.11.2024 20:12, Nicholas Piggin:

> * Assorted small ppc fixes
> 
> ----------------------------------------------------------------
> BALATON Zoltan (1):
>        hw/ppc/pegasos2: Fix IRQ routing from pci.0
> 
> Glenn Miles (1):
>        target/ppc: Fix THREAD_SIBLING_FOREACH for multi-socket
> 
> Harsh Prateek Bora (1):
>        ppc/spapr: fix drc index mismatch for partially enabled vcpus
> 
> Nicholas Piggin (3):
>        target/ppc: Fix non-maskable interrupt while halted
>        ppc/pnv: Fix direct controls quiesce
>        ppc/pnv: Add xscom- prefix to pervasive-control region name
anything here worth to pick for qemu-stable, or is
all this "small enough" for this?

Thanks,

/mjt
Re: [PULL 0/6] ppc-for-9.2-2 queue
Posted by Harsh Prateek Bora 1 week, 2 days ago
Hi Michael,

On Wed, 27 Nov, 2024, 3:14 pm Michael Tokarev, <mjt@tls.msk.ru> wrote:

> 26.11.2024 20:12, Nicholas Piggin:
>
> > * Assorted small ppc fixes
> >
> > ----------------------------------------------------------------
> > BALATON Zoltan (1):
> >        hw/ppc/pegasos2: Fix IRQ routing from pci.0
> >
> > Glenn Miles (1):
> >        target/ppc: Fix THREAD_SIBLING_FOREACH for multi-socket
> >
> > Harsh Prateek Bora (1):
> >        ppc/spapr: fix drc index mismatch for partially enabled vcpus
> >


I think at least this ^^^ can be picked for stable.
Nick may confirm about other patches.

Thanks
Harsh


> > Nicholas Piggin (3):
> >        target/ppc: Fix non-maskable interrupt while halted
> >        ppc/pnv: Fix direct controls quiesce
> >        ppc/pnv: Add xscom- prefix to pervasive-control region name
> anything here worth to pick for qemu-stable, or is
> all this "small enough" for this?
>
> Thanks,
>
> /mjt
>
>
Re: [PULL 0/6] ppc-for-9.2-2 queue
Posted by Nicholas Piggin 4 days, 21 hours ago
On Wed Nov 27, 2024 at 11:22 PM AEST, Harsh Prateek Bora wrote:
> Hi Michael,
>
> On Wed, 27 Nov, 2024, 3:14 pm Michael Tokarev, <mjt@tls.msk.ru> wrote:
>
> > 26.11.2024 20:12, Nicholas Piggin:
> >
> > > * Assorted small ppc fixes
> > >
> > > ----------------------------------------------------------------
> > > BALATON Zoltan (1):
> > >        hw/ppc/pegasos2: Fix IRQ routing from pci.0
> > >
> > > Glenn Miles (1):
> > >        target/ppc: Fix THREAD_SIBLING_FOREACH for multi-socket
> > >
> > > Harsh Prateek Bora (1):
> > >        ppc/spapr: fix drc index mismatch for partially enabled vcpus
> > >
>
>
> I think at least this ^^^ can be picked for stable.
> Nick may confirm about other patches.

I would say these two as well.

target/ppc: Fix non-maskable interrupt while halted
target/ppc: Fix THREAD_SIBLING_FOREACH for multi-socket

Thanks,
Nick
Re: [PULL 0/6] ppc-for-9.2-2 queue
Posted by Michael Tokarev 4 days, 20 hours ago
02.12.2024 08:44, Nicholas Piggin wrote:
> On Wed Nov 27, 2024 at 11:22 PM AEST, Harsh Prateek Bora wrote:
>> Hi Michael,
>>
>> On Wed, 27 Nov, 2024, 3:14 pm Michael Tokarev, <mjt@tls.msk.ru> wrote:
>>
>>> 26.11.2024 20:12, Nicholas Piggin:
>>>
>>>> * Assorted small ppc fixes
>>>>
>>>> ----------------------------------------------------------------
>>>> BALATON Zoltan (1):
>>>>         hw/ppc/pegasos2: Fix IRQ routing from pci.0
>>>>
>>>> Glenn Miles (1):
>>>>         target/ppc: Fix THREAD_SIBLING_FOREACH for multi-socket
>>>>
>>>> Harsh Prateek Bora (1):
>>>>         ppc/spapr: fix drc index mismatch for partially enabled vcpus
>>>>
>>
>>
>> I think at least this ^^^ can be picked for stable.
>> Nick may confirm about other patches.
> 
> I would say these two as well.

Thank you both for help!

> target/ppc: Fix non-maskable interrupt while halted
> target/ppc: Fix THREAD_SIBLING_FOREACH for multi-socket

I picked the 3 mentioned changes to stable series.

With the last one there's a small issue though:

  2fc0a78a57 target/ppc: Fix THREAD_SIBLING_FOREACH for multi-socket

It fixes an issue introduced in v8.0.0-2045-gb769d4c8f4, so
should be picked up for 8.2.x.  But this one lacks:
  v9.0.0-2635-gfeb37fdc82 ppc: Add a core_index to CPUPPCState for SMT vCPUs
  v9.0.0-2638-gc26504afd5 ppc/pnv: Add a big-core mode that joins two regular cores

What do you suggest to do with 8.2.x in this case?

Thanks,

/mjt

Re: [PULL 0/6] ppc-for-9.2-2 queue
Posted by BALATON Zoltan 1 week, 2 days ago
On Wed, 27 Nov 2024, Michael Tokarev wrote:
> 26.11.2024 20:12, Nicholas Piggin:
>
>> * Assorted small ppc fixes
>> 
>> ----------------------------------------------------------------
>> BALATON Zoltan (1):
>>        hw/ppc/pegasos2: Fix IRQ routing from pci.0

This depends on a change (exporting IRQState) that was merged during this 
cycle so probably would not apply for earlier versions. The issue fixed is 
also rarely needed (mainly for GPU passthrough) so I think this one is not 
needed for stable. I can't judge the others.

Regards,
BALATON Zoltan

>> 
>> Glenn Miles (1):
>>        target/ppc: Fix THREAD_SIBLING_FOREACH for multi-socket
>> 
>> Harsh Prateek Bora (1):
>>        ppc/spapr: fix drc index mismatch for partially enabled vcpus
>> 
>> Nicholas Piggin (3):
>>        target/ppc: Fix non-maskable interrupt while halted
>>        ppc/pnv: Fix direct controls quiesce
>>        ppc/pnv: Add xscom- prefix to pervasive-control region name
> anything here worth to pick for qemu-stable, or is
> all this "small enough" for this?
>
> Thanks,
>
> /mjt
>
>
Re: [PULL 0/6] ppc-for-9.2-2 queue
Posted by Peter Maydell 1 week, 3 days ago
On Tue, 26 Nov 2024 at 17:13, Nicholas Piggin <npiggin@gmail.com> wrote:
>
> The following changes since commit ba54a7e6b86884e43bed2d2f5a79c719059652a8:
>
>   Merge tag 'net-pull-request' of https://github.com/jasowang/qemu into staging (2024-11-26 14:06:40 +0000)
>
> are available in the Git repository at:
>
>   https://gitlab.com/npiggin/qemu.git tags/pull-ppc-for-9.2-2-20241127
>
> for you to fetch changes up to 0805136a44d39adc2467f23ac3c65e680e45d0a2:
>
>   hw/ppc/pegasos2: Fix IRQ routing from pci.0 (2024-11-27 02:49:36 +1000)
>
> ----------------------------------------------------------------
> * Assorted small ppc fixes


Applied, thanks.

Please update the changelog at https://wiki.qemu.org/ChangeLog/9.2
for any user-visible changes.

PS: it's helpful if you can send pullreqs to arrive by lunchtime-ish
UK time on release-candidate day, given how long it takes to get
them through CI. Otherwise they're likely to miss getting tagged
before end of day my time.

-- PMM