Re: Stable backport request for new OVMF version

Re: Stable backport request for new OVMF version

[Ian Campbell]

Forgot to CC the list, now done...

On Thu, 2015-07-23 at 14:56 +0100, Ian Campbell wrote:
> The version of OVMF in 4.5 (and presumably earlier) doesn't build with
> the gcc in modern Linux distros (I noticed it with 4.9 from
> Debian/Jessie).
> 
> I think we should update Config.mk:OVMF_UPSTREAM_REVISION to
> cb9a7ebabcd6b8a49dc0854b2f9592d732b5afbd in all stable branches. That
> version is the one currently in master.
> 
> One aspect of the specific failure I've tripped over (Debian Jessie) is
> that it will be a blocker for updating osstest to using the new version
> of Debian on the test hosts, which in turn is a prerequisite for arm64
> support.
> 
> Ian.

Re: Stable backport request for new OVMF version

[Jan Beulich]

>>> On 23.07.15 at 15:56, <ian.campbell@citrix.com> wrote:
> The version of OVMF in 4.5 (and presumably earlier) doesn't build with
> the gcc in modern Linux distros (I noticed it with 4.9 from
> Debian/Jessie).
> 
> I think we should update Config.mk:OVMF_UPSTREAM_REVISION to
> cb9a7ebabcd6b8a49dc0854b2f9592d732b5afbd in all stable branches. That
> version is the one currently in master.

To be honest I didn't even know we support use of OVMF in any
released version (nor in unstable). When did this change?

Jan

> One aspect of the specific failure I've tripped over (Debian Jessie) is
> that it will be a blocker for updating osstest to using the new version
> of Debian on the test hosts, which in turn is a prerequisite for arm64
> support.
> 
> Ian.

Re: Stable backport request for new OVMF version

[Ian Campbell]

On Thu, 2015-07-23 at 08:26 -0600, Jan Beulich wrote:
> > 
> > > > On 23.07.15 at 15:56, <ian.campbell@citrix.com> wrote:
> > The version of OVMF in 4.5 (and presumably earlier) doesn't build 
> > with
> > the gcc in modern Linux distros (I noticed it with 4.9 from
> > Debian/Jessie).
> > 
> > I think we should update Config.mk:OVMF_UPSTREAM_REVISION to
> > cb9a7ebabcd6b8a49dc0854b2f9592d732b5afbd in all stable branches. 
> > That
> > version is the one currently in master.
> 
> To be honest I didn't even know we support use of OVMF in any
> released version (nor in unstable). When did this change?

It was present in 4.3. I don't think we build it by default since its
toolchain is a bit picky, but we turn it on and test it in osstest from
4.4 onwards.

You can ask for it with bios="ovmf" in your hvm guest's config.

Ian.

Re: Stable backport request for new OVMF version

[Jan Beulich]

>>> On 23.07.15 at 16:43, <ian.campbell@citrix.com> wrote:
> On Thu, 2015-07-23 at 08:26 -0600, Jan Beulich wrote:
>> > 
>> > > > On 23.07.15 at 15:56, <ian.campbell@citrix.com> wrote:
>> > The version of OVMF in 4.5 (and presumably earlier) doesn't build 
>> > with
>> > the gcc in modern Linux distros (I noticed it with 4.9 from
>> > Debian/Jessie).
>> > 
>> > I think we should update Config.mk:OVMF_UPSTREAM_REVISION to
>> > cb9a7ebabcd6b8a49dc0854b2f9592d732b5afbd in all stable branches. 
>> > That
>> > version is the one currently in master.
>> 
>> To be honest I didn't even know we support use of OVMF in any
>> released version (nor in unstable). When did this change?
> 
> It was present in 4.3. I don't think we build it by default since its
> toolchain is a bit picky, but we turn it on and test it in osstest from
> 4.4 onwards.
> 
> You can ask for it with bios="ovmf" in your hvm guest's config.

Well, that means it can be used by the adventurous, but it doesn't
mean we consider it supported.

Jan

Re: Stable backport request for new OVMF version

[Ian Campbell]

On Thu, 2015-07-23 at 08:51 -0600, Jan Beulich wrote:
> > 
> > > > On 23.07.15 at 16:43, <ian.campbell@citrix.com> wrote:
> > On Thu, 2015-07-23 at 08:26 -0600, Jan Beulich wrote:
> > > > 
> > > > > > On 23.07.15 at 15:56, <ian.campbell@citrix.com> wrote:
> > > > The version of OVMF in 4.5 (and presumably earlier) doesn't 
> > > > build 
> > > > with
> > > > the gcc in modern Linux distros (I noticed it with 4.9 from
> > > > Debian/Jessie).
> > > > 
> > > > I think we should update Config.mk:OVMF_UPSTREAM_REVISION to
> > > > cb9a7ebabcd6b8a49dc0854b2f9592d732b5afbd in all stable 
> > > > branches. 
> > > > That
> > > > version is the one currently in master.
> > > 
> > > To be honest I didn't even know we support use of OVMF in any
> > > released version (nor in unstable). When did this change?
> > 
> > It was present in 4.3. I don't think we build it by default since 
> > its
> > toolchain is a bit picky, but we turn it on and test it in osstest 
> > from
> > 4.4 onwards.
> > 
> > You can ask for it with bios="ovmf" in your hvm guest's config.
> 
> Well, that means it can be used by the adventurous, but it doesn't
> mean we consider it supported.

TBH, I'm more concerned that it is a blocker for upgrade osstest to
Debian Jessie, which in turn is a prerequisite for testing arm64.

Ian.

Re: Stable backport request for new OVMF version

[Fabio Fantoni]

Il 23/07/2015 16:51, Jan Beulich ha scritto:
>>>> On 23.07.15 at 16:43, <ian.campbell@citrix.com> wrote:
>> On Thu, 2015-07-23 at 08:26 -0600, Jan Beulich wrote:
>>>>>> On 23.07.15 at 15:56, <ian.campbell@citrix.com> wrote:
>>>> The version of OVMF in 4.5 (and presumably earlier) doesn't build
>>>> with
>>>> the gcc in modern Linux distros (I noticed it with 4.9 from
>>>> Debian/Jessie).
>>>>
>>>> I think we should update Config.mk:OVMF_UPSTREAM_REVISION to
>>>> cb9a7ebabcd6b8a49dc0854b2f9592d732b5afbd in all stable branches.
>>>> That
>>>> version is the one currently in master.
>>> To be honest I didn't even know we support use of OVMF in any
>>> released version (nor in unstable). When did this change?
>> It was present in 4.3. I don't think we build it by default since its
>> toolchain is a bit picky, but we turn it on and test it in osstest from
>> 4.4 onwards.
>>
>> You can ask for it with bios="ovmf" in your hvm guest's config.
> Well, that means it can be used by the adventurous, but it doesn't
> mean we consider it supported.
>
> Jan
>
>

Even if used only by the adventurous left it unable to build with newer 
gcc don't seems very good, even if the adventurous should be able to 
change Config.mk to use update ovmf git that don't fail the build.

Re: Stable backport request for new OVMF version

[Jan Beulich]

>>> On 23.07.15 at 16:58, <ian.campbell@citrix.com> wrote:
> On Thu, 2015-07-23 at 08:51 -0600, Jan Beulich wrote:
>> > 
>> > > > On 23.07.15 at 16:43, <ian.campbell@citrix.com> wrote:
>> > On Thu, 2015-07-23 at 08:26 -0600, Jan Beulich wrote:
>> > > > 
>> > > > > > On 23.07.15 at 15:56, <ian.campbell@citrix.com> wrote:
>> > > > The version of OVMF in 4.5 (and presumably earlier) doesn't 
>> > > > build 
>> > > > with
>> > > > the gcc in modern Linux distros (I noticed it with 4.9 from
>> > > > Debian/Jessie).
>> > > > 
>> > > > I think we should update Config.mk:OVMF_UPSTREAM_REVISION to
>> > > > cb9a7ebabcd6b8a49dc0854b2f9592d732b5afbd in all stable 
>> > > > branches. 
>> > > > That
>> > > > version is the one currently in master.
>> > > 
>> > > To be honest I didn't even know we support use of OVMF in any
>> > > released version (nor in unstable). When did this change?
>> > 
>> > It was present in 4.3. I don't think we build it by default since 
>> > its
>> > toolchain is a bit picky, but we turn it on and test it in osstest 
>> > from
>> > 4.4 onwards.
>> > 
>> > You can ask for it with bios="ovmf" in your hvm guest's config.
>> 
>> Well, that means it can be used by the adventurous, but it doesn't
>> mean we consider it supported.
> 
> TBH, I'm more concerned that it is a blocker for upgrade osstest to
> Debian Jessie, which in turn is a prerequisite for testing arm64.

If it was explicitly unsupported, I could say "do whatever you need
to do to it", but with not being sure what status it has I'm kind of
wondering what other changes (i.e. apart from fixing the build) the
tree(s) would get which then may break someone. But in the end
it's probably okay to do the change.

Jan

Re: Stable backport request for new OVMF version

[Wei Liu]

On Thu, Jul 23, 2015 at 09:28:15AM -0600, Jan Beulich wrote:
> >>> On 23.07.15 at 16:58, <ian.campbell@citrix.com> wrote:
> > On Thu, 2015-07-23 at 08:51 -0600, Jan Beulich wrote:
> >> > 
> >> > > > On 23.07.15 at 16:43, <ian.campbell@citrix.com> wrote:
> >> > On Thu, 2015-07-23 at 08:26 -0600, Jan Beulich wrote:
> >> > > > 
> >> > > > > > On 23.07.15 at 15:56, <ian.campbell@citrix.com> wrote:
> >> > > > The version of OVMF in 4.5 (and presumably earlier) doesn't 
> >> > > > build 
> >> > > > with
> >> > > > the gcc in modern Linux distros (I noticed it with 4.9 from
> >> > > > Debian/Jessie).
> >> > > > 
> >> > > > I think we should update Config.mk:OVMF_UPSTREAM_REVISION to
> >> > > > cb9a7ebabcd6b8a49dc0854b2f9592d732b5afbd in all stable 
> >> > > > branches. 
> >> > > > That
> >> > > > version is the one currently in master.
> >> > > 
> >> > > To be honest I didn't even know we support use of OVMF in any
> >> > > released version (nor in unstable). When did this change?
> >> > 
> >> > It was present in 4.3. I don't think we build it by default since 
> >> > its
> >> > toolchain is a bit picky, but we turn it on and test it in osstest 
> >> > from
> >> > 4.4 onwards.
> >> > 
> >> > You can ask for it with bios="ovmf" in your hvm guest's config.
> >> 
> >> Well, that means it can be used by the adventurous, but it doesn't
> >> mean we consider it supported.
> > 
> > TBH, I'm more concerned that it is a blocker for upgrade osstest to
> > Debian Jessie, which in turn is a prerequisite for testing arm64.
> 
> If it was explicitly unsupported, I could say "do whatever you need
> to do to it", but with not being sure what status it has I'm kind of
> wondering what other changes (i.e. apart from fixing the build) the
> tree(s) would get which then may break someone. But in the end
> it's probably okay to do the change.
> 

It's marked as experimental, not supported and not built by default --
if that makes you feel more comfortable.

Wei.

> Jan

Re: Stable backport request for new OVMF version

[Jan Beulich]

>>> On 23.07.15 at 17:31, <wei.liu2@citrix.com> wrote:
> On Thu, Jul 23, 2015 at 09:28:15AM -0600, Jan Beulich wrote:
>> >>> On 23.07.15 at 16:58, <ian.campbell@citrix.com> wrote:
>> > On Thu, 2015-07-23 at 08:51 -0600, Jan Beulich wrote:
>> >> > 
>> >> > > > On 23.07.15 at 16:43, <ian.campbell@citrix.com> wrote:
>> >> > On Thu, 2015-07-23 at 08:26 -0600, Jan Beulich wrote:
>> >> > > > 
>> >> > > > > > On 23.07.15 at 15:56, <ian.campbell@citrix.com> wrote:
>> >> > > > The version of OVMF in 4.5 (and presumably earlier) doesn't 
>> >> > > > build 
>> >> > > > with
>> >> > > > the gcc in modern Linux distros (I noticed it with 4.9 from
>> >> > > > Debian/Jessie).
>> >> > > > 
>> >> > > > I think we should update Config.mk:OVMF_UPSTREAM_REVISION to
>> >> > > > cb9a7ebabcd6b8a49dc0854b2f9592d732b5afbd in all stable 
>> >> > > > branches. 
>> >> > > > That
>> >> > > > version is the one currently in master.
>> >> > > 
>> >> > > To be honest I didn't even know we support use of OVMF in any
>> >> > > released version (nor in unstable). When did this change?
>> >> > 
>> >> > It was present in 4.3. I don't think we build it by default since 
>> >> > its
>> >> > toolchain is a bit picky, but we turn it on and test it in osstest 
>> >> > from
>> >> > 4.4 onwards.
>> >> > 
>> >> > You can ask for it with bios="ovmf" in your hvm guest's config.
>> >> 
>> >> Well, that means it can be used by the adventurous, but it doesn't
>> >> mean we consider it supported.
>> > 
>> > TBH, I'm more concerned that it is a blocker for upgrade osstest to
>> > Debian Jessie, which in turn is a prerequisite for testing arm64.
>> 
>> If it was explicitly unsupported, I could say "do whatever you need
>> to do to it", but with not being sure what status it has I'm kind of
>> wondering what other changes (i.e. apart from fixing the build) the
>> tree(s) would get which then may break someone. But in the end
>> it's probably okay to do the change.
>> 
> 
> It's marked as experimental, not supported and not built by default --
> if that makes you feel more comfortable.

It does. So Ian, feel free to go ahead.

Jan

Re: Stable backport request for new OVMF version

[Ian Campbell]

On Thu, 2015-07-23 at 09:37 -0600, Jan Beulich wrote:
> > 
> > It's marked as experimental, not supported and not built by default 
> > --
> > if that makes you feel more comfortable.
> 
> It does. So Ian, feel free to go ahead.

Thanks. I suppose you were addressing Ian J here since I don't
typically do stable release work.

I'm happy to do the grunt work on this occasion though...

Ian.

Re: Stable backport request for new OVMF version

[Jan Beulich]

>>> On 23.07.15 at 17:59, <ian.campbell@citrix.com> wrote:
> On Thu, 2015-07-23 at 09:37 -0600, Jan Beulich wrote:
>> > 
>> > It's marked as experimental, not supported and not built by default 
>> > --
>> > if that makes you feel more comfortable.
>> 
>> It does. So Ian, feel free to go ahead.
> 
> Thanks. I suppose you were addressing Ian J here since I don't
> typically do stable release work.

No, since this was mainly for ARM (for which I recall you doing
backports), I did indeed mean you.

> I'm happy to do the grunt work on this occasion though...

Thanks.

Jan

Re: Stable backport request for new OVMF version

[Ian Campbell]

On Thu, 2015-07-23 at 10:11 -0600, Jan Beulich wrote:
> > 
> > > > On 23.07.15 at 17:59, <ian.campbell@citrix.com> wrote:
> > On Thu, 2015-07-23 at 09:37 -0600, Jan Beulich wrote:
> > > > 
> > > > It's marked as experimental, not supported and not built by 
> > > > default 
> > > > --
> > > > if that makes you feel more comfortable.
> > > 
> > > It does. So Ian, feel free to go ahead.
> > 
> > Thanks. I suppose you were addressing Ian J here since I don't
> > typically do stable release work.
> 
> No, since this was mainly for ARM (for which I recall you doing
> backports), I did indeed mean you.

OVMF isn't for ARM, it's x86.

I don't think we build OVMF for ARM under any circumstances. (I was
hoping to add that to raisin at some point, but that's another story)

> 
> > I'm happy to do the grunt work on this occasion though...
> 
> Thanks.
> 
> Jan
> 

Re: Stable backport request for new OVMF version

[Jan Beulich]

>>> On 23.07.15 at 18:23, <ian.campbell@citrix.com> wrote:
> On Thu, 2015-07-23 at 10:11 -0600, Jan Beulich wrote:
>> > 
>> > > > On 23.07.15 at 17:59, <ian.campbell@citrix.com> wrote:
>> > On Thu, 2015-07-23 at 09:37 -0600, Jan Beulich wrote:
>> > > > 
>> > > > It's marked as experimental, not supported and not built by 
>> > > > default 
>> > > > --
>> > > > if that makes you feel more comfortable.
>> > > 
>> > > It does. So Ian, feel free to go ahead.
>> > 
>> > Thanks. I suppose you were addressing Ian J here since I don't
>> > typically do stable release work.
>> 
>> No, since this was mainly for ARM (for which I recall you doing
>> backports), I did indeed mean you.
> 
> OVMF isn't for ARM, it's x86.
> 
> I don't think we build OVMF for ARM under any circumstances. (I was
> hoping to add that to raisin at some point, but that's another story)

But didn't you say it's a prereq for ARM64? Or was that only
implicitly through the intended Debian update?

Jan

Re: Stable backport request for new OVMF version

[Wei Liu]

On Fri, Jul 24, 2015 at 01:47:41AM -0600, Jan Beulich wrote:
> >>> On 23.07.15 at 18:23, <ian.campbell@citrix.com> wrote:
> > On Thu, 2015-07-23 at 10:11 -0600, Jan Beulich wrote:
> >> > 
> >> > > > On 23.07.15 at 17:59, <ian.campbell@citrix.com> wrote:
> >> > On Thu, 2015-07-23 at 09:37 -0600, Jan Beulich wrote:
> >> > > > 
> >> > > > It's marked as experimental, not supported and not built by 
> >> > > > default 
> >> > > > --
> >> > > > if that makes you feel more comfortable.
> >> > > 
> >> > > It does. So Ian, feel free to go ahead.
> >> > 
> >> > Thanks. I suppose you were addressing Ian J here since I don't
> >> > typically do stable release work.
> >> 
> >> No, since this was mainly for ARM (for which I recall you doing
> >> backports), I did indeed mean you.
> > 
> > OVMF isn't for ARM, it's x86.
> > 
> > I don't think we build OVMF for ARM under any circumstances. (I was
> > hoping to add that to raisin at some point, but that's another story)
> 
> But didn't you say it's a prereq for ARM64? Or was that only
> implicitly through the intended Debian update?
> 

The prerequisite for ARM64 (in fact, running ARM64 test case in OSSTest)
is OSSTest needs to upgrade to Debian Jessie. Ian wasn't clear about
that in his reply. :-)

Wei.

> Jan

Re: Stable backport request for new OVMF version

[Jan Beulich]

>>> On 24.07.15 at 09:56, <wei.liu2@citrix.com> wrote:
> On Fri, Jul 24, 2015 at 01:47:41AM -0600, Jan Beulich wrote:
>> >>> On 23.07.15 at 18:23, <ian.campbell@citrix.com> wrote:
>> > On Thu, 2015-07-23 at 10:11 -0600, Jan Beulich wrote:
>> >> > 
>> >> > > > On 23.07.15 at 17:59, <ian.campbell@citrix.com> wrote:
>> >> > On Thu, 2015-07-23 at 09:37 -0600, Jan Beulich wrote:
>> >> > > > 
>> >> > > > It's marked as experimental, not supported and not built by 
>> >> > > > default 
>> >> > > > --
>> >> > > > if that makes you feel more comfortable.
>> >> > > 
>> >> > > It does. So Ian, feel free to go ahead.
>> >> > 
>> >> > Thanks. I suppose you were addressing Ian J here since I don't
>> >> > typically do stable release work.
>> >> 
>> >> No, since this was mainly for ARM (for which I recall you doing
>> >> backports), I did indeed mean you.
>> > 
>> > OVMF isn't for ARM, it's x86.
>> > 
>> > I don't think we build OVMF for ARM under any circumstances. (I was
>> > hoping to add that to raisin at some point, but that's another story)
>> 
>> But didn't you say it's a prereq for ARM64? Or was that only
>> implicitly through the intended Debian update?
>> 
> 
> The prerequisite for ARM64 (in fact, running ARM64 test case in OSSTest)
> is OSSTest needs to upgrade to Debian Jessie. Ian wasn't clear about
> that in his reply. :-)

So what we really want is a backport of b28da5fc1c + 28e5d9a9ad.

Ian - is this needed on 4.4 and 4.5, or just the latter?

Jan

Re: Stable backport request for new OVMF version

[Ian Campbell]

On Fri, 2015-07-24 at 02:09 -0600, Jan Beulich wrote:
> > 
> > > > On 24.07.15 at 09:56, <wei.liu2@citrix.com> wrote:
> > On Fri, Jul 24, 2015 at 01:47:41AM -0600, Jan Beulich wrote:
> > > > > > On 23.07.15 at 18:23, <ian.campbell@citrix.com> wrote:
> > > > On Thu, 2015-07-23 at 10:11 -0600, Jan Beulich wrote:
> > > > > > 
> > > > > > > > On 23.07.15 at 17:59, <ian.campbell@citrix.com> wrote:
> > > > > > On Thu, 2015-07-23 at 09:37 -0600, Jan Beulich wrote:
> > > > > > > > 
> > > > > > > > It's marked as experimental, not supported and not 
> > > > > > > > built by 
> > > > > > > > default 
> > > > > > > > --
> > > > > > > > if that makes you feel more comfortable.
> > > > > > > 
> > > > > > > It does. So Ian, feel free to go ahead.
> > > > > > 
> > > > > > Thanks. I suppose you were addressing Ian J here since I 
> > > > > > don't
> > > > > > typically do stable release work.
> > > > > 
> > > > > No, since this was mainly for ARM (for which I recall you 
> > > > > doing
> > > > > backports), I did indeed mean you.
> > > > 
> > > > OVMF isn't for ARM, it's x86.
> > > > 
> > > > I don't think we build OVMF for ARM under any circumstances. (I 
> > > > was
> > > > hoping to add that to raisin at some point, but that's another 
> > > > story)
> > > 
> > > But didn't you say it's a prereq for ARM64? Or was that only
> > > implicitly through the intended Debian update?
> > > 
> > 
> > The prerequisite for ARM64 (in fact, running ARM64 test case in 
> > OSSTest)
> > is OSSTest needs to upgrade to Debian Jessie. Ian wasn't clear 
> > about
> > that in his reply. :-)

That's right, sorry for not being clear.

> So what we really want is a backport of b28da5fc1c + 28e5d9a9ad.

Yes, that looks like the right pair.

> Ian - is this needed on 4.4 and 4.5, or just the latter?

Osstest tests OVMF from 4.4 onwards, so I think both.

Ian.