1. Patchew
  2. QEMU
  3. View series

[Qemu-devel] [PATCH v2 0/6] s390x/pci: remaining hot/un)plug patches

David Hildenbrand posted 6 patches 5 years, 2 months ago
Test docker-mingw@fedora passed
Test asan passed
Test checkpatch passed
Test docker-clang@ubuntu passed
Patches applied successfully (tree, apply log)
git fetch https://github.com/patchew-project/qemu tags/patchew/20190130155733.32742-1-david@redhat.com
Maintainers: Christian Borntraeger <borntraeger@de.ibm.com>, Richard Henderson <rth@twiddle.net>, Halil Pasic <pasic@linux.ibm.com>, Cornelia Huck <cohuck@redhat.com>, Collin Walling <walling@linux.ibm.com>, David Hildenbrand <david@redhat.com>
hw/s390x/s390-pci-bus.c | 233 ++++++++++++++++++++++++++--------------
hw/s390x/s390-pci-bus.h |   4 +-
2 files changed, 152 insertions(+), 85 deletions(-)
[Qemu-devel] [PATCH v2 0/6] s390x/pci: remaining hot/un)plug patches
Posted by David Hildenbrand 5 years, 2 months ago 
These are all the patches that are not yet upstream (@Conny you might
already picked some, including them for the full picture) and after
a good discussion yesterday, including a patch t get rid of the release
timer. I ran a couple of sanity tests on this series.

#1 and #2 fix hotplugging of PCI bridges.
#3 warns when "zpci=off"
#4 refactors unplugging
#5 get's rid of the release timer
#6 processes all unplug requests on reboot

@Colin, can you review/ack? Especially Patch #4 is needed for qdev
patches already on the list. ("[PATCH RFCv2 0/9] qdev: Hotplug handler
chaining + virtio-pmem")

David Hildenbrand (6):
  s390x/pci: Fix primary bus number for PCI bridges
  s390x/pci: Fix hotplugging of PCI bridges
  s390x/pci: Warn when adding PCI devices without the 'zpci' feature
  s390x/pci: Introduce unplug requests and split unplug handler
  s390x/pci: Drop release timer and replace it with a flag
  s390x/pci: Unplug remaining requested devices on pcihost reset

 hw/s390x/s390-pci-bus.c | 233 ++++++++++++++++++++++++++--------------
 hw/s390x/s390-pci-bus.h |   4 +-
 2 files changed, 152 insertions(+), 85 deletions(-)

-- 
2.17.2
Re: [Qemu-devel] [PATCH v2 0/6] s390x/pci: remaining hot/un)plug patches
Posted by Cornelia Huck 5 years, 2 months ago 
On Wed, 30 Jan 2019 16:57:27 +0100
David Hildenbrand <david@redhat.com> wrote:

> These are all the patches that are not yet upstream (@Conny you might
> already picked some, including them for the full picture) and after

I usually keep s390-next up-to-date, so if the patches are not there, I
have not yet queued them.

> a good discussion yesterday, including a patch t get rid of the release
> timer. I ran a couple of sanity tests on this series.
> 
> #1 and #2 fix hotplugging of PCI bridges.
> #3 warns when "zpci=off"
> #4 refactors unplugging
> #5 get's rid of the release timer
> #6 processes all unplug requests on reboot
> 
> @Colin, can you review/ack? Especially Patch #4 is needed for qdev
> patches already on the list. ("[PATCH RFCv2 0/9] qdev: Hotplug handler
> chaining + virtio-pmem")
> 
> David Hildenbrand (6):
>   s390x/pci: Fix primary bus number for PCI bridges
>   s390x/pci: Fix hotplugging of PCI bridges
>   s390x/pci: Warn when adding PCI devices without the 'zpci' feature
>   s390x/pci: Introduce unplug requests and split unplug handler
>   s390x/pci: Drop release timer and replace it with a flag
>   s390x/pci: Unplug remaining requested devices on pcihost reset
> 
>  hw/s390x/s390-pci-bus.c | 233 ++++++++++++++++++++++++++--------------
>  hw/s390x/s390-pci-bus.h |   4 +-
>  2 files changed, 152 insertions(+), 85 deletions(-)
>
Re: [Qemu-devel] [PATCH v2 0/6] s390x/pci: remaining hot/un)plug patches
Posted by Cornelia Huck 5 years, 2 months ago 
On Wed, 30 Jan 2019 16:57:27 +0100
David Hildenbrand <david@redhat.com> wrote:

> These are all the patches that are not yet upstream (@Conny you might
> already picked some, including them for the full picture) and after
> a good discussion yesterday, including a patch t get rid of the release
> timer. I ran a couple of sanity tests on this series.
> 
> #1 and #2 fix hotplugging of PCI bridges.
> #3 warns when "zpci=off"
> #4 refactors unplugging
> #5 get's rid of the release timer
> #6 processes all unplug requests on reboot
> 
> @Colin, can you review/ack? Especially Patch #4 is needed for qdev
> patches already on the list. ("[PATCH RFCv2 0/9] qdev: Hotplug handler
> chaining + virtio-pmem")
> 
> David Hildenbrand (6):
>   s390x/pci: Fix primary bus number for PCI bridges
>   s390x/pci: Fix hotplugging of PCI bridges
>   s390x/pci: Warn when adding PCI devices without the 'zpci' feature
>   s390x/pci: Introduce unplug requests and split unplug handler
>   s390x/pci: Drop release timer and replace it with a flag
>   s390x/pci: Unplug remaining requested devices on pcihost reset
> 
>  hw/s390x/s390-pci-bus.c | 233 ++++++++++++++++++++++++++--------------
>  hw/s390x/s390-pci-bus.h |   4 +-
>  2 files changed, 152 insertions(+), 85 deletions(-)
> 

Thanks, applied the remaining patches.
Re: [Qemu-devel] [PATCH v2 0/6] s390x/pci: remaining hot/un)plug patches
Posted by David Hildenbrand 5 years, 2 months ago 
On 05.02.19 10:55, Cornelia Huck wrote:
> On Wed, 30 Jan 2019 16:57:27 +0100
> David Hildenbrand <david@redhat.com> wrote:
> 
>> These are all the patches that are not yet upstream (@Conny you might
>> already picked some, including them for the full picture) and after
>> a good discussion yesterday, including a patch t get rid of the release
>> timer. I ran a couple of sanity tests on this series.
>>
>> #1 and #2 fix hotplugging of PCI bridges.
>> #3 warns when "zpci=off"
>> #4 refactors unplugging
>> #5 get's rid of the release timer
>> #6 processes all unplug requests on reboot
>>
>> @Colin, can you review/ack? Especially Patch #4 is needed for qdev
>> patches already on the list. ("[PATCH RFCv2 0/9] qdev: Hotplug handler
>> chaining + virtio-pmem")
>>
>> David Hildenbrand (6):
>>   s390x/pci: Fix primary bus number for PCI bridges
>>   s390x/pci: Fix hotplugging of PCI bridges
>>   s390x/pci: Warn when adding PCI devices without the 'zpci' feature
>>   s390x/pci: Introduce unplug requests and split unplug handler
>>   s390x/pci: Drop release timer and replace it with a flag
>>   s390x/pci: Unplug remaining requested devices on pcihost reset
>>
>>  hw/s390x/s390-pci-bus.c | 233 ++++++++++++++++++++++++++--------------
>>  hw/s390x/s390-pci-bus.h |   4 +-
>>  2 files changed, 152 insertions(+), 85 deletions(-)
>>
> 
> Thanks, applied the remaining patches.
> 

Thanks a lot Conny!

-- 

Thanks,

David / dhildenb
Re: [Qemu-devel] [qemu-s390x] [PATCH v2 0/6] s390x/pci: remaining hot/un)plug patches
Posted by Collin Walling 5 years, 2 months ago 
On 1/30/19 10:57 AM, David Hildenbrand wrote:
> These are all the patches that are not yet upstream (@Conny you might
> already picked some, including them for the full picture) and after
> a good discussion yesterday, including a patch t get rid of the release
> timer. I ran a couple of sanity tests on this series.
> 
> #1 and #2 fix hotplugging of PCI bridges.
> #3 warns when "zpci=off"
> #4 refactors unplugging
> #5 get's rid of the release timer
> #6 processes all unplug requests on reboot
> 
> @Colin, can you review/ack? Especially Patch #4 is needed for qdev
> patches already on the list. ("[PATCH RFCv2 0/9] qdev: Hotplug handler
> chaining + virtio-pmem")

I've given my r-b for 4, 5, and 6. IIRC, the other patches in this
series looked fine from previous versions, but I'd like to look at them
more carefully before ack'ing. Please allow me some time to devote to
them in the near future. Trying my best to balance things :)

> 
> David Hildenbrand (6):
>    s390x/pci: Fix primary bus number for PCI bridges
>    s390x/pci: Fix hotplugging of PCI bridges
>    s390x/pci: Warn when adding PCI devices without the 'zpci' feature
>    s390x/pci: Introduce unplug requests and split unplug handler
>    s390x/pci: Drop release timer and replace it with a flag
>    s390x/pci: Unplug remaining requested devices on pcihost reset
> 
>   hw/s390x/s390-pci-bus.c | 233 ++++++++++++++++++++++++++--------------
>   hw/s390x/s390-pci-bus.h |   4 +-
>   2 files changed, 152 insertions(+), 85 deletions(-)
>
Re: [Qemu-devel] [qemu-s390x] [PATCH v2 0/6] s390x/pci: remaining hot/un)plug patches
Posted by David Hildenbrand 5 years, 2 months ago 
On 31.01.19 21:43, Collin Walling wrote:
> On 1/30/19 10:57 AM, David Hildenbrand wrote:
>> These are all the patches that are not yet upstream (@Conny you might
>> already picked some, including them for the full picture) and after
>> a good discussion yesterday, including a patch t get rid of the release
>> timer. I ran a couple of sanity tests on this series.
>>
>> #1 and #2 fix hotplugging of PCI bridges.
>> #3 warns when "zpci=off"
>> #4 refactors unplugging
>> #5 get's rid of the release timer
>> #6 processes all unplug requests on reboot
>>
>> @Colin, can you review/ack? Especially Patch #4 is needed for qdev
>> patches already on the list. ("[PATCH RFCv2 0/9] qdev: Hotplug handler
>> chaining + virtio-pmem")
> 
> I've given my r-b for 4, 5, and 6. IIRC, the other patches in this
> series looked fine from previous versions, but I'd like to look at them
> more carefully before ack'ing. Please allow me some time to devote to
> them in the near future. Trying my best to balance things :)

Thanks, patch 4 was the most important one for me (needed for the qdev
reworks). All the other stuff was fallout from testing and code review.

> 
>>
>> David Hildenbrand (6):
>>    s390x/pci: Fix primary bus number for PCI bridges
>>    s390x/pci: Fix hotplugging of PCI bridges
>>    s390x/pci: Warn when adding PCI devices without the 'zpci' feature
>>    s390x/pci: Introduce unplug requests and split unplug handler
>>    s390x/pci: Drop release timer and replace it with a flag
>>    s390x/pci: Unplug remaining requested devices on pcihost reset
>>
>>   hw/s390x/s390-pci-bus.c | 233 ++++++++++++++++++++++++++--------------
>>   hw/s390x/s390-pci-bus.h |   4 +-
>>   2 files changed, 152 insertions(+), 85 deletions(-)
>>
> 


-- 

Thanks,

David / dhildenb
Re: [Qemu-devel] [qemu-s390x] [PATCH v2 0/6] s390x/pci: remaining hot/un)plug patches
Posted by Cornelia Huck 5 years, 2 months ago 
On Thu, 31 Jan 2019 15:43:16 -0500
Collin Walling <walling@linux.ibm.com> wrote:

> On 1/30/19 10:57 AM, David Hildenbrand wrote:
> > These are all the patches that are not yet upstream (@Conny you might
> > already picked some, including them for the full picture) and after
> > a good discussion yesterday, including a patch t get rid of the release
> > timer. I ran a couple of sanity tests on this series.
> > 
> > #1 and #2 fix hotplugging of PCI bridges.
> > #3 warns when "zpci=off"
> > #4 refactors unplugging
> > #5 get's rid of the release timer
> > #6 processes all unplug requests on reboot
> > 
> > @Colin, can you review/ack? Especially Patch #4 is needed for qdev
> > patches already on the list. ("[PATCH RFCv2 0/9] qdev: Hotplug handler
> > chaining + virtio-pmem")  
> 
> I've given my r-b for 4, 5, and 6. IIRC, the other patches in this
> series looked fine from previous versions, but I'd like to look at them
> more carefully before ack'ing. Please allow me some time to devote to
> them in the near future. Trying my best to balance things :)

Great, thanks for looking. If patches 4-6 can be applied without
patches 1-3 (I think they can?), I'll go ahead and queue them.

(I'll probably send a pull request next week, will be happy to include
the other patches as well.)
Re: [Qemu-devel] [qemu-s390x] [PATCH v2 0/6] s390x/pci: remaining hot/un)plug patches
Posted by David Hildenbrand 5 years, 2 months ago 
On 01.02.19 09:35, Cornelia Huck wrote:
> On Thu, 31 Jan 2019 15:43:16 -0500
> Collin Walling <walling@linux.ibm.com> wrote:
> 
>> On 1/30/19 10:57 AM, David Hildenbrand wrote:
>>> These are all the patches that are not yet upstream (@Conny you might
>>> already picked some, including them for the full picture) and after
>>> a good discussion yesterday, including a patch t get rid of the release
>>> timer. I ran a couple of sanity tests on this series.
>>>
>>> #1 and #2 fix hotplugging of PCI bridges.
>>> #3 warns when "zpci=off"
>>> #4 refactors unplugging
>>> #5 get's rid of the release timer
>>> #6 processes all unplug requests on reboot
>>>
>>> @Colin, can you review/ack? Especially Patch #4 is needed for qdev
>>> patches already on the list. ("[PATCH RFCv2 0/9] qdev: Hotplug handler
>>> chaining + virtio-pmem")  
>>
>> I've given my r-b for 4, 5, and 6. IIRC, the other patches in this
>> series looked fine from previous versions, but I'd like to look at them
>> more carefully before ack'ing. Please allow me some time to devote to
>> them in the near future. Trying my best to balance things :)
> 
> Great, thanks for looking. If patches 4-6 can be applied without
> patches 1-3 (I think they can?), I'll go ahead and queue them.

Yes, I think they can, patch 1-3 mainly deals with plugging, these ones
with unplugging. Thanks Conny!

> 
> (I'll probably send a pull request next week, will be happy to include
> the other patches as well.)
> 


-- 

Thanks,

David / dhildenb
Re: [Qemu-devel] [qemu-s390x] [PATCH v2 0/6] s390x/pci: remaining hot/un)plug patches
Posted by Collin Walling 5 years, 2 months ago 
On 2/1/19 4:18 AM, David Hildenbrand wrote:
> On 01.02.19 09:35, Cornelia Huck wrote:
>> On Thu, 31 Jan 2019 15:43:16 -0500
>> Collin Walling <walling@linux.ibm.com> wrote:
>>
>>> On 1/30/19 10:57 AM, David Hildenbrand wrote:
>>>> These are all the patches that are not yet upstream (@Conny you might
>>>> already picked some, including them for the full picture) and after
>>>> a good discussion yesterday, including a patch t get rid of the release
>>>> timer. I ran a couple of sanity tests on this series.
>>>>
>>>> #1 and #2 fix hotplugging of PCI bridges.
>>>> #3 warns when "zpci=off"
>>>> #4 refactors unplugging
>>>> #5 get's rid of the release timer
>>>> #6 processes all unplug requests on reboot
>>>>
>>>> @Colin, can you review/ack? Especially Patch #4 is needed for qdev
>>>> patches already on the list. ("[PATCH RFCv2 0/9] qdev: Hotplug handler
>>>> chaining + virtio-pmem")
>>>
>>> I've given my r-b for 4, 5, and 6. IIRC, the other patches in this
>>> series looked fine from previous versions, but I'd like to look at them
>>> more carefully before ack'ing. Please allow me some time to devote to
>>> them in the near future. Trying my best to balance things :)
>>
>> Great, thanks for looking. If patches 4-6 can be applied without
>> patches 1-3 (I think they can?), I'll go ahead and queue them.
> 
> Yes, I think they can, patch 1-3 mainly deals with plugging, these ones
> with unplugging. Thanks Conny!
> 
>>
>> (I'll probably send a pull request next week, will be happy to include
>> the other patches as well.)
>>
> 
> 

I noted the conflict you ran into when applying. I'll try to get to
reviewing the other patches either by today or Monday. If for some
reason I don't get to it by then, please ping me.

Patchew 2.1-devel-0870f84

© 2016 - 2024 Red Hat, Inc.