latest GOOD state of series i386 cleanup, arm cleanup, s390 cleanup

latest GOOD state of series i386 cleanup, arm cleanup, s390 cleanup

[Claudio Fontana]

Hello all,

due to my inactivity for a few weeks coupled likely with the upstream processes around qemu-6.0 now the series:

1) i386 cleanup
2) arm cleanup and experimental kvm-only build
3) s390 cleanup

have become stale and hard to rebase on latest master.
This effect is compounded by the fact that lots of broken tests in master have been added.

In the interest of not losing work,
I provide here the latest known good state of these series:

For the i386 cleanup:
https://gitlab.com/hw-claudio/qemu/-/pipelines/293603386

Tests started breaking horribly since about 1/2 weeks.
The latest version of the cleanup is reachable here:

https://github.com/qemu/qemu.git branch "i386_cleanup_9"

In my understanding, Paolo has now picked up this one.

For the ARM cleanup and experimental kvm-only build:

https://gitlab.com/hw-claudio/qemu/-/pipelines/293603376

https://github.com/qemu/qemu.git branch "arm_cleanup_v15"

Again here tests started misbehaving in the same timeframe.

The state of ARM cleanup is still experimental, maybe Liang or Philippe you can adopt this one?

For the s390x cleanup:

https://gitlab.com/hw-claudio/qemu/-/pipelines/293603379

https://github.com/qemu/qemu.git branch "s390_cleanup_v1"

Al, can you take this one to the finish line?

Thanks,

Claudio
-- 
Claudio Fontana
Engineering Manager Virtualization, SUSE Labs Core

SUSE Software Solutions Italy Srl

Re: latest GOOD state of series i386 cleanup, arm cleanup, s390 cleanup

[Claudio Fontana]

On 5/17/21 11:53 AM, Claudio Fontana wrote:
> Hello all,
> 
> due to my inactivity for a few weeks coupled likely with the upstream processes around qemu-6.0 now the series:
> 
> 1) i386 cleanup
> 2) arm cleanup and experimental kvm-only build
> 3) s390 cleanup
> 
> have become stale and hard to rebase on latest master.
> This effect is compounded by the fact that lots of broken tests in master have been added.
> 
> In the interest of not losing work,
> I provide here the latest known good state of these series:
> 
> For the i386 cleanup:
> https://gitlab.com/hw-claudio/qemu/-/pipelines/293603386
> 
> Tests started breaking horribly since about 1/2 weeks.
> The latest version of the cleanup is reachable here:
> 
> https://github.com/qemu/qemu.git branch "i386_cleanup_9"
> 
> In my understanding, Paolo has now picked up this one.
> 
> For the ARM cleanup and experimental kvm-only build:
> 
> https://gitlab.com/hw-claudio/qemu/-/pipelines/293603376
> 
> https://github.com/qemu/qemu.git branch "arm_cleanup_v15"
> 
> Again here tests started misbehaving in the same timeframe.
> 
> The state of ARM cleanup is still experimental, maybe Liang or Philippe you can adopt this one?
> 
> For the s390x cleanup:
> 
> https://gitlab.com/hw-claudio/qemu/-/pipelines/293603379
> 
> https://github.com/qemu/qemu.git branch "s390_cleanup_v1"
> 
> Al, can you take this one to the finish line?
> 
> Thanks,
> 
> Claudio
> 

Adding also Alex Bennee in Cc:

Thanks,

Claudio

Re: latest GOOD state of series i386 cleanup, arm cleanup, s390 cleanup

[Al Cho via]

[-- Attachment #1: Type: text/plain, Size: 2251 bytes --]

Hi Claudio,

> https://gitlab.com/hw-claudio/qemu/-/pipelines/293603379
>
> https://github.com/qemu/qemu.git branch "s390_cleanup_v1"
>
> Al, can you take this one to the finish line?

Yes, I will take it.

Cheers,
             AL
________________________________
From: Claudio Fontana <cfontana@suse.de>
Sent: Monday, May 17, 2021 5:54 PM
To: Richard Henderson <richard.henderson@linaro.org>; Philippe Mathieu-Daudé <f4bug@amsat.org>; Paolo Bonzini <pbonzini@redhat.com>; Peter Maydell <peter.maydell@linaro.org>
Cc: qemu-devel <qemu-devel@nongnu.org>; Al Cho <ACho@suse.com>; Liang Yan <LYan@suse.com>; Alex Bennee <alex.bennee@linaro.org>
Subject: Re: latest GOOD state of series i386 cleanup, arm cleanup, s390 cleanup

On 5/17/21 11:53 AM, Claudio Fontana wrote:
> Hello all,
>
> due to my inactivity for a few weeks coupled likely with the upstream processes around qemu-6.0 now the series:
>
> 1) i386 cleanup
> 2) arm cleanup and experimental kvm-only build
> 3) s390 cleanup
>
> have become stale and hard to rebase on latest master.
> This effect is compounded by the fact that lots of broken tests in master have been added.
>
> In the interest of not losing work,
> I provide here the latest known good state of these series:
>
> For the i386 cleanup:
> https://gitlab.com/hw-claudio/qemu/-/pipelines/293603386
>
> Tests started breaking horribly since about 1/2 weeks.
> The latest version of the cleanup is reachable here:
>
> https://github.com/qemu/qemu.git branch "i386_cleanup_9"
>
> In my understanding, Paolo has now picked up this one.
>
> For the ARM cleanup and experimental kvm-only build:
>
> https://gitlab.com/hw-claudio/qemu/-/pipelines/293603376
>
> https://github.com/qemu/qemu.git branch "arm_cleanup_v15"
>
> Again here tests started misbehaving in the same timeframe.
>
> The state of ARM cleanup is still experimental, maybe Liang or Philippe you can adopt this one?
>
> For the s390x cleanup:
>
> https://gitlab.com/hw-claudio/qemu/-/pipelines/293603379
>
> https://github.com/qemu/qemu.git branch "s390_cleanup_v1"
>
> Al, can you take this one to the finish line?
>
> Thanks,
>
> Claudio
>

Adding also Alex Bennee in Cc:

Thanks,

Claudio


[-- Attachment #2: Type: text/html, Size: 4852 bytes --]

Re: latest GOOD state of series i386 cleanup, arm cleanup, s390 cleanup

[Alex Bennée]

Claudio Fontana <cfontana@suse.de> writes:

> On 5/17/21 11:53 AM, Claudio Fontana wrote:
>> Hello all,
>> 
>> due to my inactivity for a few weeks coupled likely with the upstream processes around qemu-6.0 now the series:
>> 
>> 1) i386 cleanup
>> 2) arm cleanup and experimental kvm-only build
>> 3) s390 cleanup
>> 
>> have become stale and hard to rebase on latest master.
>> This effect is compounded by the fact that lots of broken tests in
>> master have been added.

Which tests are these? I know master suffers a bit from occasional
falling red but to my knowledge everything should be green (at least
from my last PR anyway ;-).

>> 
>> In the interest of not losing work,
>> I provide here the latest known good state of these series:
>> 
>> For the i386 cleanup:
>> https://gitlab.com/hw-claudio/qemu/-/pipelines/293603386
>> 
>> Tests started breaking horribly since about 1/2 weeks.

The pipeline only shows one failed test (checkpatch) which is an
allowfail I believe. /me is confused.

>> The latest version of the cleanup is reachable here:
>> 
>> https://github.com/qemu/qemu.git branch "i386_cleanup_9"
>> 
>> In my understanding, Paolo has now picked up this one.
>> 
>> For the ARM cleanup and experimental kvm-only build:
>> 
>> https://gitlab.com/hw-claudio/qemu/-/pipelines/293603376
>> 
>> https://github.com/qemu/qemu.git branch "arm_cleanup_v15"
>> 
>> Again here tests started misbehaving in the same timeframe.
>> 
>> The state of ARM cleanup is still experimental, maybe Liang or Philippe you can adopt this one?
>> 
>> For the s390x cleanup:
>> 
>> https://gitlab.com/hw-claudio/qemu/-/pipelines/293603379
>> 
>> https://github.com/qemu/qemu.git branch "s390_cleanup_v1"
>> 
>> Al, can you take this one to the finish line?
>> 
>> Thanks,
>> 
>> Claudio
>> 
>
> Adding also Alex Bennee in Cc:
>
> Thanks,
>
> Claudio


-- 
Alex Bennée

Re: latest GOOD state of series i386 cleanup, arm cleanup, s390 cleanup

[Claudio Fontana]

On 5/18/21 4:02 PM, Alex Bennée wrote:
> 
> Claudio Fontana <cfontana@suse.de> writes:
> 
>> On 5/17/21 11:53 AM, Claudio Fontana wrote:
>>> Hello all,
>>>
>>> due to my inactivity for a few weeks coupled likely with the upstream processes around qemu-6.0 now the series:
>>>
>>> 1) i386 cleanup
>>> 2) arm cleanup and experimental kvm-only build
>>> 3) s390 cleanup
>>>
>>> have become stale and hard to rebase on latest master.
>>> This effect is compounded by the fact that lots of broken tests in
>>> master have been added.
> 
> Which tests are these? I know master suffers a bit from occasional
> falling red but to my knowledge everything should be green (at least
> from my last PR anyway ;-).
> 
>>>
>>> In the interest of not losing work,
>>> I provide here the latest known good state of these series:
>>>
>>> For the i386 cleanup:
>>> https://gitlab.com/hw-claudio/qemu/-/pipelines/293603386
>>>
>>> Tests started breaking horribly since about 1/2 weeks.
> 
> The pipeline only shows one failed test (checkpatch) which is an
> allowfail I believe. /me is confused.

Hi Alex, yes, I pointed to the last pipeline that works :-)

> 
>>> The latest version of the cleanup is reachable here:
>>>
>>> https://github.com/qemu/qemu.git branch "i386_cleanup_9"
>>>
>>> In my understanding, Paolo has now picked up this one.
>>>
>>> For the ARM cleanup and experimental kvm-only build:
>>>
>>> https://gitlab.com/hw-claudio/qemu/-/pipelines/293603376
>>>
>>> https://github.com/qemu/qemu.git branch "arm_cleanup_v15"
>>>
>>> Again here tests started misbehaving in the same timeframe.
>>>
>>> The state of ARM cleanup is still experimental, maybe Liang or Philippe you can adopt this one?
>>>
>>> For the s390x cleanup:
>>>
>>> https://gitlab.com/hw-claudio/qemu/-/pipelines/293603379
>>>
>>> https://github.com/qemu/qemu.git branch "s390_cleanup_v1"
>>>
>>> Al, can you take this one to the finish line?
>>>
>>> Thanks,
>>>
>>> Claudio
>>>
>>
>> Adding also Alex Bennee in Cc:
>>
>> Thanks,
>>
>> Claudio
> 
> 

Re: latest GOOD state of series i386 cleanup, arm cleanup, s390 cleanup

[Alex Bennée]

Claudio Fontana <cfontana@suse.de> writes:

> On 5/18/21 4:02 PM, Alex Bennée wrote:
>> 
>> Claudio Fontana <cfontana@suse.de> writes:
>> 
>>> On 5/17/21 11:53 AM, Claudio Fontana wrote:
>>>> Hello all,
>>>>
>>>> due to my inactivity for a few weeks coupled likely with the upstream processes around qemu-6.0 now the series:
>>>>
>>>> 1) i386 cleanup
>>>> 2) arm cleanup and experimental kvm-only build
>>>> 3) s390 cleanup
>>>>
>>>> have become stale and hard to rebase on latest master.
>>>> This effect is compounded by the fact that lots of broken tests in
>>>> master have been added.
>> 
>> Which tests are these? I know master suffers a bit from occasional
>> falling red but to my knowledge everything should be green (at least
>> from my last PR anyway ;-).
>> 
>>>>
>>>> In the interest of not losing work,
>>>> I provide here the latest known good state of these series:
>>>>
>>>> For the i386 cleanup:
>>>> https://gitlab.com/hw-claudio/qemu/-/pipelines/293603386
>>>>
>>>> Tests started breaking horribly since about 1/2 weeks.
>> 
>> The pipeline only shows one failed test (checkpatch) which is an
>> allowfail I believe. /me is confused.
>
> Hi Alex, yes, I pointed to the last pipeline that works :-)
>
>> 
>>>> The latest version of the cleanup is reachable here:
>>>>
>>>> https://github.com/qemu/qemu.git branch "i386_cleanup_9"
>>>>
>>>> In my understanding, Paolo has now picked up this one.
>>>>
>>>> For the ARM cleanup and experimental kvm-only build:
>>>>
>>>> https://gitlab.com/hw-claudio/qemu/-/pipelines/293603376
>>>>
>>>> https://github.com/qemu/qemu.git branch "arm_cleanup_v15"
>>>>
>>>> Again here tests started misbehaving in the same timeframe.
>>>>
>>>> The state of ARM cleanup is still experimental, maybe Liang or
>>>> Philippe you can adopt this one?

I've done a re-base onto the current master (and my testing/next):

  https://github.com/stsquad/qemu/tree/review/arm_cleanup_v15

which is currently working it's way through my CI:

  https://gitlab.com/stsquad/qemu/-/pipelines/306727076

As I've got patches waiting for this re-factor I'm happy to take the
series on if you've run out of time/patience ;-)

-- 
Alex Bennée

Re: latest GOOD state of series i386 cleanup, arm cleanup, s390 cleanup

[Cho Yu-Chen]

I also done a branch for current master from [RFC v3 00/13] s390x cleanup

https://gitlab.com/alcho.tw/qemu/-/tree/s390_cleanup_v4

and it's also built and test success passed through my CI:

https://gitlab.com/alcho.tw/qemu/-/pipelines/307149915

Cheers,
           AL


Alex Bennée <alex.bennee@linaro.org> 於 2021年5月21日 週五 上午12:07寫道：
>
>
> Claudio Fontana <cfontana@suse.de> writes:
>
> > On 5/18/21 4:02 PM, Alex Bennée wrote:
> >>
> >> Claudio Fontana <cfontana@suse.de> writes:
> >>
> >>> On 5/17/21 11:53 AM, Claudio Fontana wrote:
> >>>> Hello all,
> >>>>
> >>>> due to my inactivity for a few weeks coupled likely with the upstream processes around qemu-6.0 now the series:
> >>>>
> >>>> 1) i386 cleanup
> >>>> 2) arm cleanup and experimental kvm-only build
> >>>> 3) s390 cleanup
> >>>>
> >>>> have become stale and hard to rebase on latest master.
> >>>> This effect is compounded by the fact that lots of broken tests in
> >>>> master have been added.
> >>
> >> Which tests are these? I know master suffers a bit from occasional
> >> falling red but to my knowledge everything should be green (at least
> >> from my last PR anyway ;-).
> >>
> >>>>
> >>>> In the interest of not losing work,
> >>>> I provide here the latest known good state of these series:
> >>>>
> >>>> For the i386 cleanup:
> >>>> https://gitlab.com/hw-claudio/qemu/-/pipelines/293603386
> >>>>
> >>>> Tests started breaking horribly since about 1/2 weeks.
> >>
> >> The pipeline only shows one failed test (checkpatch) which is an
> >> allowfail I believe. /me is confused.
> >
> > Hi Alex, yes, I pointed to the last pipeline that works :-)
> >
> >>
> >>>> The latest version of the cleanup is reachable here:
> >>>>
> >>>> https://github.com/qemu/qemu.git branch "i386_cleanup_9"
> >>>>
> >>>> In my understanding, Paolo has now picked up this one.
> >>>>
> >>>> For the ARM cleanup and experimental kvm-only build:
> >>>>
> >>>> https://gitlab.com/hw-claudio/qemu/-/pipelines/293603376
> >>>>
> >>>> https://github.com/qemu/qemu.git branch "arm_cleanup_v15"
> >>>>
> >>>> Again here tests started misbehaving in the same timeframe.
> >>>>
> >>>> The state of ARM cleanup is still experimental, maybe Liang or
> >>>> Philippe you can adopt this one?
>
> I've done a re-base onto the current master (and my testing/next):
>
>   https://github.com/stsquad/qemu/tree/review/arm_cleanup_v15
>
> which is currently working it's way through my CI:
>
>   https://gitlab.com/stsquad/qemu/-/pipelines/306727076
>
> As I've got patches waiting for this re-factor I'm happy to take the
> series on if you've run out of time/patience ;-)
>
> --
> Alex Bennée
>

Re: latest GOOD state of series i386 cleanup, arm cleanup, s390 cleanup

[Claudio Fontana]

On 5/20/21 6:02 PM, Alex Bennée wrote:
> 
> Claudio Fontana <cfontana@suse.de> writes:
> 
>> On 5/18/21 4:02 PM, Alex Bennée wrote:
>>>
>>> Claudio Fontana <cfontana@suse.de> writes:
>>>
>>>> On 5/17/21 11:53 AM, Claudio Fontana wrote:
>>>>> Hello all,
>>>>>
>>>>> due to my inactivity for a few weeks coupled likely with the upstream processes around qemu-6.0 now the series:
>>>>>
>>>>> 1) i386 cleanup
>>>>> 2) arm cleanup and experimental kvm-only build
>>>>> 3) s390 cleanup
>>>>>
>>>>> have become stale and hard to rebase on latest master.
>>>>> This effect is compounded by the fact that lots of broken tests in
>>>>> master have been added.
>>>
>>> Which tests are these? I know master suffers a bit from occasional
>>> falling red but to my knowledge everything should be green (at least
>>> from my last PR anyway ;-).
>>>
>>>>>
>>>>> In the interest of not losing work,
>>>>> I provide here the latest known good state of these series:
>>>>>
>>>>> For the i386 cleanup:
>>>>> https://gitlab.com/hw-claudio/qemu/-/pipelines/293603386
>>>>>
>>>>> Tests started breaking horribly since about 1/2 weeks.
>>>
>>> The pipeline only shows one failed test (checkpatch) which is an
>>> allowfail I believe. /me is confused.
>>
>> Hi Alex, yes, I pointed to the last pipeline that works :-)
>>
>>>
>>>>> The latest version of the cleanup is reachable here:
>>>>>
>>>>> https://github.com/qemu/qemu.git branch "i386_cleanup_9"
>>>>>
>>>>> In my understanding, Paolo has now picked up this one.
>>>>>
>>>>> For the ARM cleanup and experimental kvm-only build:
>>>>>
>>>>> https://gitlab.com/hw-claudio/qemu/-/pipelines/293603376
>>>>>
>>>>> https://github.com/qemu/qemu.git branch "arm_cleanup_v15"
>>>>>
>>>>> Again here tests started misbehaving in the same timeframe.
>>>>>
>>>>> The state of ARM cleanup is still experimental, maybe Liang or
>>>>> Philippe you can adopt this one?
> 
> I've done a re-base onto the current master (and my testing/next):
> 
>   https://github.com/stsquad/qemu/tree/review/arm_cleanup_v15
> 
> which is currently working it's way through my CI:
> 
>   https://gitlab.com/stsquad/qemu/-/pipelines/306727076
> 
> As I've got patches waiting for this re-factor I'm happy to take the
> series on if you've run out of time/patience ;-)
> 

I don't see any other takers so it makes sense, thanks a lot!

Maybe Liang you could help Alex along the way in promoting this work from experimental state to ready for actual integration?

Thanks,

Claudio