Re: [PATCH V13 2/9] meson.build: Re-enable KVM support for MIPS

Re: [PATCH V13 2/9] meson.build: Re-enable KVM support for MIPS

[Philippe Mathieu-Daudé]

On 11/20/20 5:28 AM, Huacai Chen wrote:
> On Wed, Nov 18, 2020 at 1:17 AM Philippe Mathieu-Daudé <f4bug@amsat.org> wrote:
>> On 10/7/20 10:39 AM, Huacai Chen wrote:
>>> After converting from configure to meson, KVM support is lost for MIPS,
>>> so re-enable it in meson.build.
>>>
>>> Fixes: fdb75aeff7c212e1afaaa3a43 ("configure: remove target configuration")
>>> Fixes: 8a19980e3fc42239aae054bc9 ("configure: move accelerator logic to meson")
>>> Cc: aolo Bonzini <pbonzini@redhat.com>
>>> Signed-off-by: Huacai Chen <chenhc@lemote.com>
>>> ---
>>>  meson.build | 2 ++
>>>  1 file changed, 2 insertions(+)
>>>
>>> diff --git a/meson.build b/meson.build
>>> index 17c89c8..b407ff4 100644
>>> --- a/meson.build
>>> +++ b/meson.build
>>> @@ -59,6 +59,8 @@ elif cpu == 's390x'
>>>    kvm_targets = ['s390x-softmmu']
>>>  elif cpu in ['ppc', 'ppc64']
>>>    kvm_targets = ['ppc-softmmu', 'ppc64-softmmu']
>>> +elif cpu in ['mips', 'mips64']
>>> +  kvm_targets = ['mips-softmmu', 'mipsel-softmmu', 'mips64-softmmu', 'mips64el-softmmu']
>>
>> Are you sure both 32-bit hosts and targets are supported?
>>
>> I don't have hardware to test. If you are not working with
>> 32-bit hardware I'd remove them.
> When I add MIPS64 KVM support (Loongson-3 is MIPS64), MIPS32 KVM is
> already there. On the kernel side, MIPS32 KVM is supported, but I
> don't know whether it can work well.

Well, from the history, you inherited from it:

commit 1fa639e5618029e944ac68d27e32a99dcb85a349
Author: James Hogan <jhogan@kernel.org>
Date:   Sat Dec 21 15:53:06 2019 +0000

    MAINTAINERS: Orphan MIPS KVM CPUs

    I haven't been active for 18 months, and don't have the hardware
    set up to test KVM for MIPS, so mark it as orphaned and remove
    myself as maintainer. Hopefully somebody from MIPS can pick this up.


commit 134f7f7da12aad99daafbeb2a7ba9dbc6bd40abc
Author: Aleksandar Markovic <aleksandar.m.mail@gmail.com>
Date:   Mon Feb 24 12:50:58 2020 +0100

    MAINTAINERS: Reactivate MIPS KVM CPUs

    Reactivate MIPS KVM maintainership with a modest goal of keeping
    the support alive, checking common KVM code changes against MIPS
    functionality, etc. (hence the status "Odd Fixes"), with hope that
    this component will be fully maintained at some further, but not
    distant point in future.


commit 15d983dee95edff1dc4c0bed71ce02fff877e766
Author: Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>
Date:   Wed Jul 1 20:25:58 2020 +0200

    MAINTAINERS: Adjust MIPS maintainership (Huacai Chen & Jiaxun Yang)

    Huacai Chen and Jiaxun Yang step in as new energy [1].


commit ca263c0fb9f33cc746e6e3d968b7db80072ecf86
Author: Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>
Date:   Wed Oct 7 22:37:21 2020 +0200

    MAINTAINERS: Remove myself

    I have been working on project other than QEMU for some time,
    and would like to devote myself to that project. It is impossible
    for me to find enough time to perform maintainer's duties with
    needed meticulousness and patience.


QEMU deprecation process is quite slow, if we release mips-softmmu
and mipsel-softmmu binaries with KVM support in 5.2, and you can not
test them, you will still have to maintain them during 2021...

If you don't have neither the hardware nor the time, I suggest you
to only release it on 64-bit hosts. Personally I'd even only
announce KVM supported on the little-endian binary only, because
AFAIK you don't test big-endian KVM neither.

Your call as a maintainer, but remember last RC tag is next
Tuesday (Nov 24) in *4* days, then we release 5.2:
https://wiki.qemu.org/Planning/5.2#Release_Schedule

Regards,

Phil.

Re: [PATCH V13 2/9] meson.build: Re-enable KVM support for MIPS

[Huacai Chen]

+CC Jiaxun

Hi, Jiaxun,

What do you think about?

Huacai

On Fri, Nov 20, 2020 at 6:55 PM Philippe Mathieu-Daudé <f4bug@amsat.org> wrote:
>
> On 11/20/20 5:28 AM, Huacai Chen wrote:
> > On Wed, Nov 18, 2020 at 1:17 AM Philippe Mathieu-Daudé <f4bug@amsat.org> wrote:
> >> On 10/7/20 10:39 AM, Huacai Chen wrote:
> >>> After converting from configure to meson, KVM support is lost for MIPS,
> >>> so re-enable it in meson.build.
> >>>
> >>> Fixes: fdb75aeff7c212e1afaaa3a43 ("configure: remove target configuration")
> >>> Fixes: 8a19980e3fc42239aae054bc9 ("configure: move accelerator logic to meson")
> >>> Cc: aolo Bonzini <pbonzini@redhat.com>
> >>> Signed-off-by: Huacai Chen <chenhc@lemote.com>
> >>> ---
> >>>  meson.build | 2 ++
> >>>  1 file changed, 2 insertions(+)
> >>>
> >>> diff --git a/meson.build b/meson.build
> >>> index 17c89c8..b407ff4 100644
> >>> --- a/meson.build
> >>> +++ b/meson.build
> >>> @@ -59,6 +59,8 @@ elif cpu == 's390x'
> >>>    kvm_targets = ['s390x-softmmu']
> >>>  elif cpu in ['ppc', 'ppc64']
> >>>    kvm_targets = ['ppc-softmmu', 'ppc64-softmmu']
> >>> +elif cpu in ['mips', 'mips64']
> >>> +  kvm_targets = ['mips-softmmu', 'mipsel-softmmu', 'mips64-softmmu', 'mips64el-softmmu']
> >>
> >> Are you sure both 32-bit hosts and targets are supported?
> >>
> >> I don't have hardware to test. If you are not working with
> >> 32-bit hardware I'd remove them.
> > When I add MIPS64 KVM support (Loongson-3 is MIPS64), MIPS32 KVM is
> > already there. On the kernel side, MIPS32 KVM is supported, but I
> > don't know whether it can work well.
>
> Well, from the history, you inherited from it:
>
> commit 1fa639e5618029e944ac68d27e32a99dcb85a349
> Author: James Hogan <jhogan@kernel.org>
> Date:   Sat Dec 21 15:53:06 2019 +0000
>
>     MAINTAINERS: Orphan MIPS KVM CPUs
>
>     I haven't been active for 18 months, and don't have the hardware
>     set up to test KVM for MIPS, so mark it as orphaned and remove
>     myself as maintainer. Hopefully somebody from MIPS can pick this up.
>
>
> commit 134f7f7da12aad99daafbeb2a7ba9dbc6bd40abc
> Author: Aleksandar Markovic <aleksandar.m.mail@gmail.com>
> Date:   Mon Feb 24 12:50:58 2020 +0100
>
>     MAINTAINERS: Reactivate MIPS KVM CPUs
>
>     Reactivate MIPS KVM maintainership with a modest goal of keeping
>     the support alive, checking common KVM code changes against MIPS
>     functionality, etc. (hence the status "Odd Fixes"), with hope that
>     this component will be fully maintained at some further, but not
>     distant point in future.
>
>
> commit 15d983dee95edff1dc4c0bed71ce02fff877e766
> Author: Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>
> Date:   Wed Jul 1 20:25:58 2020 +0200
>
>     MAINTAINERS: Adjust MIPS maintainership (Huacai Chen & Jiaxun Yang)
>
>     Huacai Chen and Jiaxun Yang step in as new energy [1].
>
>
> commit ca263c0fb9f33cc746e6e3d968b7db80072ecf86
> Author: Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>
> Date:   Wed Oct 7 22:37:21 2020 +0200
>
>     MAINTAINERS: Remove myself
>
>     I have been working on project other than QEMU for some time,
>     and would like to devote myself to that project. It is impossible
>     for me to find enough time to perform maintainer's duties with
>     needed meticulousness and patience.
>
>
> QEMU deprecation process is quite slow, if we release mips-softmmu
> and mipsel-softmmu binaries with KVM support in 5.2, and you can not
> test them, you will still have to maintain them during 2021...
>
> If you don't have neither the hardware nor the time, I suggest you
> to only release it on 64-bit hosts. Personally I'd even only
> announce KVM supported on the little-endian binary only, because
> AFAIK you don't test big-endian KVM neither.
>
> Your call as a maintainer, but remember last RC tag is next
> Tuesday (Nov 24) in *4* days, then we release 5.2:
> https://wiki.qemu.org/Planning/5.2#Release_Schedule
>
> Regards,
>
> Phil.

Re: [PATCH V13 2/9] meson.build: Re-enable KVM support for MIPS

[Philippe Mathieu-Daudé]

Hi Huacai,

We are going to tag QEMU v6.0-rc0 today.

I only have access to a 64-bit MIPS in little-endian to
test KVM.

Can you test the other configurations please?
- 32-bit BE
- 32-bit LE
- 64-bit BE

Thanks!

Phil.

On 11/22/20 4:31 AM, Huacai Chen wrote:
> +CC Jiaxun
> 
> Hi, Jiaxun,
> 
> What do you think about?
> 
> Huacai
> 
> On Fri, Nov 20, 2020 at 6:55 PM Philippe Mathieu-Daudé <f4bug@amsat.org> wrote:
>>
>> On 11/20/20 5:28 AM, Huacai Chen wrote:
>>> On Wed, Nov 18, 2020 at 1:17 AM Philippe Mathieu-Daudé <f4bug@amsat.org> wrote:
>>>> On 10/7/20 10:39 AM, Huacai Chen wrote:
>>>>> After converting from configure to meson, KVM support is lost for MIPS,
>>>>> so re-enable it in meson.build.
>>>>>
>>>>> Fixes: fdb75aeff7c212e1afaaa3a43 ("configure: remove target configuration")
>>>>> Fixes: 8a19980e3fc42239aae054bc9 ("configure: move accelerator logic to meson")
>>>>> Cc: aolo Bonzini <pbonzini@redhat.com>
>>>>> Signed-off-by: Huacai Chen <chenhc@lemote.com>
>>>>> ---
>>>>>  meson.build | 2 ++
>>>>>  1 file changed, 2 insertions(+)
>>>>>
>>>>> diff --git a/meson.build b/meson.build
>>>>> index 17c89c8..b407ff4 100644
>>>>> --- a/meson.build
>>>>> +++ b/meson.build
>>>>> @@ -59,6 +59,8 @@ elif cpu == 's390x'
>>>>>    kvm_targets = ['s390x-softmmu']
>>>>>  elif cpu in ['ppc', 'ppc64']
>>>>>    kvm_targets = ['ppc-softmmu', 'ppc64-softmmu']
>>>>> +elif cpu in ['mips', 'mips64']
>>>>> +  kvm_targets = ['mips-softmmu', 'mipsel-softmmu', 'mips64-softmmu', 'mips64el-softmmu']
>>>>
>>>> Are you sure both 32-bit hosts and targets are supported?
>>>>
>>>> I don't have hardware to test. If you are not working with
>>>> 32-bit hardware I'd remove them.
>>> When I add MIPS64 KVM support (Loongson-3 is MIPS64), MIPS32 KVM is
>>> already there. On the kernel side, MIPS32 KVM is supported, but I
>>> don't know whether it can work well.
>>
>> Well, from the history, you inherited from it:
>>
>> commit 1fa639e5618029e944ac68d27e32a99dcb85a349
>> Author: James Hogan <jhogan@kernel.org>
>> Date:   Sat Dec 21 15:53:06 2019 +0000
>>
>>     MAINTAINERS: Orphan MIPS KVM CPUs
>>
>>     I haven't been active for 18 months, and don't have the hardware
>>     set up to test KVM for MIPS, so mark it as orphaned and remove
>>     myself as maintainer. Hopefully somebody from MIPS can pick this up.
>>
>>
>> commit 134f7f7da12aad99daafbeb2a7ba9dbc6bd40abc
>> Author: Aleksandar Markovic <aleksandar.m.mail@gmail.com>
>> Date:   Mon Feb 24 12:50:58 2020 +0100
>>
>>     MAINTAINERS: Reactivate MIPS KVM CPUs
>>
>>     Reactivate MIPS KVM maintainership with a modest goal of keeping
>>     the support alive, checking common KVM code changes against MIPS
>>     functionality, etc. (hence the status "Odd Fixes"), with hope that
>>     this component will be fully maintained at some further, but not
>>     distant point in future.
>>
>>
>> commit 15d983dee95edff1dc4c0bed71ce02fff877e766
>> Author: Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>
>> Date:   Wed Jul 1 20:25:58 2020 +0200
>>
>>     MAINTAINERS: Adjust MIPS maintainership (Huacai Chen & Jiaxun Yang)
>>
>>     Huacai Chen and Jiaxun Yang step in as new energy [1].
>>
>>
>> commit ca263c0fb9f33cc746e6e3d968b7db80072ecf86
>> Author: Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>
>> Date:   Wed Oct 7 22:37:21 2020 +0200
>>
>>     MAINTAINERS: Remove myself
>>
>>     I have been working on project other than QEMU for some time,
>>     and would like to devote myself to that project. It is impossible
>>     for me to find enough time to perform maintainer's duties with
>>     needed meticulousness and patience.
>>
>>
>> QEMU deprecation process is quite slow, if we release mips-softmmu
>> and mipsel-softmmu binaries with KVM support in 5.2, and you can not
>> test them, you will still have to maintain them during 2021...
>>
>> If you don't have neither the hardware nor the time, I suggest you
>> to only release it on 64-bit hosts. Personally I'd even only
>> announce KVM supported on the little-endian binary only, because
>> AFAIK you don't test big-endian KVM neither.
>>
>> Your call as a maintainer, but remember last RC tag is next
>> Tuesday (Nov 24) in *4* days, then we release 5.2:
>> https://wiki.qemu.org/Planning/5.2#Release_Schedule
>>
>> Regards,
>>
>> Phil.
> 

Re: [PATCH V13 2/9] meson.build: Re-enable KVM support for MIPS

[Jiaxun Yang]

On Tue, Mar 23, 2021, at 9:56 PM, Philippe Mathieu-Daudé wrote:
> Hi Huacai,
> 
> We are going to tag QEMU v6.0-rc0 today.
> 
> I only have access to a 64-bit MIPS in little-endian to
> test KVM.
> 
> Can you test the other configurations please?
> - 32-bit BE
> - 32-bit LE
> - 64-bit BE

+syq
As Loongson doesn't have Big-Endian processor and Loongson 3A won't run 32bit kernel.

Probably wecan test on boston or malta board? 

Thanks. 


> 
> Thanks!
> 
> Phil.
> 
> 
[...]

-- 
- Jiaxun

Re: [PATCH V13 2/9] meson.build: Re-enable KVM support for MIPS

[YunQiang Su]

Jiaxun Yang <jiaxun.yang@flygoat.com> 于2021年3月24日周三 上午9:29写道：
>
>
>
> On Tue, Mar 23, 2021, at 9:56 PM, Philippe Mathieu-Daudé wrote:
> > Hi Huacai,
> >
> > We are going to tag QEMU v6.0-rc0 today.
> >
> > I only have access to a 64-bit MIPS in little-endian to
> > test KVM.
> >
> > Can you test the other configurations please?
> > - 32-bit BE
> > - 32-bit LE
> > - 64-bit BE
>

How to run the test? just run a VM with KVM support on these kernel?

> +syq
> As Loongson doesn't have Big-Endian processor and Loongson 3A won't run 32bit kernel.
>
> Probably wecan test on boston or malta board?
>
> Thanks.
>
>
> >
> > Thanks!
> >
> > Phil.
> >
> >
> [...]
>
> --
> - Jiaxun