Re: Bug#886294: transition: nodejs

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|

Re: Bug#886294: transition: nodejs

Emilio Pozuelo Monfort-4
On 23/01/18 19:21, Jérémy Lal wrote:

> cc-ing s390x team to ask (re)building nodejs-8.9.3~dfsg-11 on zemlinsky.
>
>             but on s390x you're getting
>             illegal instructions on zemlinsky, which is a Z10 mainframe. Looks
>             like newer
>             node possibly bumped the baseline, or just accidentally introduced
>             instructions
>             not supported by our baseline.
>
>
>         Starting investigations about that. Hopefully it's a change that could
>         have been
>         backward-compatible.
>
>
> nodejs/v8 somewhat officially support s390x down to z196.
> I removed the added march=z196 flag and uploaded it into nodejs-8.9.3~dfsg-11.
> It would be wonderful to build it on zemlinsky to see what happens.

Still fails: https://buildd.debian.org/status/package.php?p=node-srs&suite=sid

Our baseline is z10, and z196 is newer. So if upstream now requires z196, we
have three options:

- Revert / fix that so upstream works with z10 again
- Remove nodejs from s390x
- Bump our baseline

See go and rustc for similar problems.

Cheers,
Emilio

Reply | Threaded
Open this post in threaded view
|

Re: Bug#886294: transition: nodejs

Bastian Blank
On Wed, Jan 24, 2018 at 01:24:06PM +0100, Emilio Pozuelo Monfort wrote:
> Our baseline is z10, and z196 is newer. So if upstream now requires z196, we
> have three options:
>
> - Revert / fix that so upstream works with z10 again
> - Remove nodejs from s390x
> - Bump our baseline
>
> See go and rustc for similar problems.

All the other buildds are already z13 (either 2964-NE1 or 2964-N63).

Bastian

--
The idea of male and female are universal constants.
                -- Kirk, "Metamorphosis", stardate 3219.8

Reply | Threaded
Open this post in threaded view
|

Re: Bug#886294: transition: nodejs

Philipp Kern-6
On 01/24/2018 06:51 PM, Bastian Blank wrote:

> On Wed, Jan 24, 2018 at 01:24:06PM +0100, Emilio Pozuelo Monfort wrote:
>> Our baseline is z10, and z196 is newer. So if upstream now requires z196, we
>> have three options:
>>
>> - Revert / fix that so upstream works with z10 again
>> - Remove nodejs from s390x
>> - Bump our baseline
>>
>> See go and rustc for similar problems.
>
> All the other buildds are already z13 (either 2964-NE1 or 2964-N63).
Yup. But the more crucial part is what to do about the user-sided
baseline. And I think we're mostly at the point where there isn't an
advantage to keeping the baseline low except a lot of churn on our side
in which no-one else is interested in. People on older platforms could
still use stable for a while.

Does anyone happen to know what the story with Hercules is[1]? I don't
worry about qemu.

Kind regards and thanks
Philipp Kern

[1] Yes, I feel like I should know.


signature.asc (523 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Bug#886294: transition: nodejs

Aurelien Jarno
In reply to this post by Emilio Pozuelo Monfort-4
On 2018-01-24 13:24, Emilio Pozuelo Monfort wrote:

> On 23/01/18 19:21, Jérémy Lal wrote:
> > cc-ing s390x team to ask (re)building nodejs-8.9.3~dfsg-11 on zemlinsky.
> >
> >             but on s390x you're getting
> >             illegal instructions on zemlinsky, which is a Z10 mainframe. Looks
> >             like newer
> >             node possibly bumped the baseline, or just accidentally introduced
> >             instructions
> >             not supported by our baseline.
> >
> >
> >         Starting investigations about that. Hopefully it's a change that could
> >         have been
> >         backward-compatible.
> >
> >
> > nodejs/v8 somewhat officially support s390x down to z196.
> > I removed the added march=z196 flag and uploaded it into nodejs-8.9.3~dfsg-11.
> > It would be wonderful to build it on zemlinsky to see what happens.
>
> Still fails: https://buildd.debian.org/status/package.php?p=node-srs&suite=sid

I requeued nodejs on zemlinsky and it also failed. After a quick check,
it appears that It uses he load/store on condition 1 facility directly,
ie not in GCC generated code. The change is therefore intentional. This
facility has been introduced with the z196.

> Our baseline is z10, and z196 is newer. So if upstream now requires z196, we
> have three options:

Officially our baseline is still z900.

> - Revert / fix that so upstream works with z10 again
> - Remove nodejs from s390x
> - Bump our baseline
>
> See go and rustc for similar problems.

Bumping the baseline to z196 looks like the easiest way and as you said,
it would also fix go, rustc and maybe more software. However we discussed
raising the ISA to z10 about one year and a half ago, and the conclusion
was that we still have users with older machines. I'll try to restart
the discussion again.

Aurelien

--
Aurelien Jarno                          GPG: 4096R/1DDD8C9B
[hidden email]                 http://www.aurel32.net

Reply | Threaded
Open this post in threaded view
|

Re: Bug#886294: transition: nodejs

Philipp Kern-6
On 2018-01-25 11:36, Aurelien Jarno wrote:
> Bumping the baseline to z196 looks like the easiest way and as you
> said,
> it would also fix go, rustc and maybe more software. However we
> discussed
> raising the ISA to z10 about one year and a half ago, and the
> conclusion
> was that we still have users with older machines. I'll try to restart
> the discussion again.

What's the venue to have this discussion in? :)

Kind regards and thanks
Philipp Kern

Reply | Threaded
Open this post in threaded view
|

Re: Bug#886294: transition: nodejs

Emilio Pozuelo Monfort-4
On 29/01/18 20:15, Philipp Kern wrote:
> On 2018-01-25 11:36, Aurelien Jarno wrote:
>> Bumping the baseline to z196 looks like the easiest way and as you said,
>> it would also fix go, rustc and maybe more software. However we discussed
>> raising the ISA to z10 about one year and a half ago, and the conclusion
>> was that we still have users with older machines. I'll try to restart
>> the discussion again.
>
> What's the venue to have this discussion in? :)

[hidden email] ?

FWIW you two (Philipp and Aurelien) are the two current s390x porters, so I
think it's mostly your call.

Cheers,
Emilio