staging evaluation in progress

[Marius Bakke]

Hello!

I just started a 'staging' evaluation:

https://hydra.gnu.org/jobset/gnu/staging

Fairly minor changes this round, highlights include Wayland 1.15 and
GStreamer 1.14. We narrowly missed Mesa 17.3.9 which was scheduled for
today but delayed, hopefully 17.3.8 doesn't introduce any new bugs.

Results should start ticking in tomorrow.

Efraim Flashner (2):
build-system/meson: Use 'target-arm32?' for armhf-linux case.
Revert "build-system/meson: Use 'target-arm32?' for armhf-linux case."

Marius Bakke (15):
gnu: gtk+: Update to 3.22.29.
gnu: libxkbcommon: Update to 0.8.0.
gnu: mesa: Update to 17.3.7.
gnu: meson: Update to 0.45.1.
gnu: gstreamer: Update to 1.14.0.
gnu: nss, nss-certs: Update to 3.36.1.
gnu: mesa: Update to 17.3.8.
gnu: alsa-lib: Update to 1.1.6.
gnu: alsa-utils: Update to 1.1.6.
gnu: alsa-plugins: Update to 1.1.6.
gnu: atk: Update to 2.28.1.
build-system/meson: Don't override LDFLAGS if already set.
gnu: gdk-pixbuf: Update to 2.36.12.
gnu: libinput: Update to 1.10.3.
gnu: eudev: Update to 3.2.5.

Rutger Helling (2):
gnu: wayland: Update to 1.15.0.
gnu: wayland-protocols: Update to 1.13.

[Ludovic Courtès]

Hello Marius,

Post by Marius Bakke
https://hydra.gnu.org/jobset/gnu/staging
Fairly minor changes this round, highlights include Wayland 1.15 and
GStreamer 1.14. We narrowly missed Mesa 17.3.9 which was scheduled for
today but delayed, hopefully 17.3.8 doesn't introduce any new bugs.

Awesome! Hydra has apparently built stuff. I’ve enabled it just now on
berlin.guixsd.org as well, let’s see how it goes.

Thank you!

Ludo’.

[Mark H Weaver]

Hi Marius,

Post by Marius Bakke
https://hydra.gnu.org/jobset/gnu/staging
Fairly minor changes this round, highlights include Wayland 1.15 and
GStreamer 1.14. We narrowly missed Mesa 17.3.9 which was scheduled for
today but delayed, hopefully 17.3.8 doesn't introduce any new bugs.
Results should start ticking in tomorrow.

The main issue I see so far is that 'gst-plugins-base' seems to
consistently fail the "elements/opus" test on i686-linux. It failed
twice in a row, anyway:

https://hydra.gnu.org/build/2635798

Thanks for working on it.

Mark

[Marius Bakke]

Post by Mark H Weaver
Hi Marius,

Post by Marius Bakke
https://hydra.gnu.org/jobset/gnu/staging
Fairly minor changes this round, highlights include Wayland 1.15 and
GStreamer 1.14. We narrowly missed Mesa 17.3.9 which was scheduled for
today but delayed, hopefully 17.3.8 doesn't introduce any new bugs.
Results should start ticking in tomorrow.

The main issue I see so far is that 'gst-plugins-base' seems to
consistently fail the "elements/opus" test on i686-linux. It failed
https://hydra.gnu.org/build/2635798

I can reproduce this failure locally, and could not find related bug
reports or git commits upstream. For now I downgraded to 1.12.5 so we
can proceed, and will report the i686 and armhf issues upstream.

[Mark H Weaver]

Post by Marius Bakke

Post by Mark H Weaver
Hi Marius,

Post by Marius Bakke
https://hydra.gnu.org/jobset/gnu/staging
Fairly minor changes this round, highlights include Wayland 1.15 and
GStreamer 1.14. We narrowly missed Mesa 17.3.9 which was scheduled for
today but delayed, hopefully 17.3.8 doesn't introduce any new bugs.
Results should start ticking in tomorrow.

The main issue I see so far is that 'gst-plugins-base' seems to
consistently fail the "elements/opus" test on i686-linux. It failed
https://hydra.gnu.org/build/2635798

I can reproduce this failure locally, and could not find related bug
reports or git commits upstream. For now I downgraded to 1.12.5 so we
can proceed, and will report the i686 and armhf issues upstream.

Instead of downgrading gstreamer, I think it would be better to simply
disable that test on i686 for now. Most likely, it is due to tests that
are intolerant of the double rounding that occurs on i686 without SSE2,
where the old x87 FP instructions are used instead. The double rounding
happens because x87 operations are performed on 80-bit double-extended
precision floating-point numbers, which must then be rounded a second
time when they are converted to 64-bit doubles as used in C.

What do you think?

Mark

[Efraim Flashner]

Post by Mark H Weaver

Post by Marius Bakke

Post by Mark H Weaver
Hi Marius,

Post by Marius Bakke
https://hydra.gnu.org/jobset/gnu/staging
Fairly minor changes this round, highlights include Wayland 1.15 and
GStreamer 1.14. We narrowly missed Mesa 17.3.9 which was scheduled for
today but delayed, hopefully 17.3.8 doesn't introduce any new bugs.
Results should start ticking in tomorrow.

The main issue I see so far is that 'gst-plugins-base' seems to
consistently fail the "elements/opus" test on i686-linux. It failed
https://hydra.gnu.org/build/2635798

I can reproduce this failure locally, and could not find related bug
reports or git commits upstream. For now I downgraded to 1.12.5 so we
can proceed, and will report the i686 and armhf issues upstream.

Instead of downgrading gstreamer, I think it would be better to simply
disable that test on i686 for now. Most likely, it is due to tests that
are intolerant of the double rounding that occurs on i686 without SSE2,
where the old x87 FP instructions are used instead. The double rounding
happens because x87 operations are performed on 80-bit double-extended
precision floating-point numbers, which must then be rounded a second
time when they are converted to 64-bit doubles as used in C.
What do you think?
Mark

somewhat random tests fail on aarch64 also, different ones from the
other architectures. I'd suggest we wait it out until 1.14.1 and watch
upstream for bug fixes.

[Mark H Weaver]

Post by Efraim Flashner

Post by Mark H Weaver

Post by Marius Bakke

Post by Mark H Weaver
Hi Marius,

Post by Marius Bakke
https://hydra.gnu.org/jobset/gnu/staging
Fairly minor changes this round, highlights include Wayland 1.15 and
GStreamer 1.14. We narrowly missed Mesa 17.3.9 which was scheduled for
today but delayed, hopefully 17.3.8 doesn't introduce any new bugs.
Results should start ticking in tomorrow.

The main issue I see so far is that 'gst-plugins-base' seems to
consistently fail the "elements/opus" test on i686-linux. It failed
https://hydra.gnu.org/build/2635798

I can reproduce this failure locally, and could not find related bug
reports or git commits upstream. For now I downgraded to 1.12.5 so we
can proceed, and will report the i686 and armhf issues upstream.

Instead of downgrading gstreamer, I think it would be better to simply
disable that test on i686 for now. Most likely, it is due to tests that
are intolerant of the double rounding that occurs on i686 without SSE2,
where the old x87 FP instructions are used instead. The double rounding
happens because x87 operations are performed on 80-bit double-extended
precision floating-point numbers, which must then be rounded a second
time when they are converted to 64-bit doubles as used in C.
What do you think?
Mark

somewhat random tests fail on aarch64 also, different ones from the
other architectures. I'd suggest we wait it out until 1.14.1 and watch
upstream for bug fixes.

Do we know if there are any security fixes in 1.14.0 that are not in
1.12.5? Is the 1.12.x branch still receiving security updates?

Mark

[Marius Bakke]

Post by Mark H Weaver

Post by Efraim Flashner

Post by Mark H Weaver

Post by Marius Bakke

Post by Mark H Weaver
Hi Marius,

Post by Marius Bakke
https://hydra.gnu.org/jobset/gnu/staging
Fairly minor changes this round, highlights include Wayland 1.15 and
GStreamer 1.14. We narrowly missed Mesa 17.3.9 which was scheduled for
today but delayed, hopefully 17.3.8 doesn't introduce any new bugs.
Results should start ticking in tomorrow.

The main issue I see so far is that 'gst-plugins-base' seems to
consistently fail the "elements/opus" test on i686-linux. It failed
https://hydra.gnu.org/build/2635798

I can reproduce this failure locally, and could not find related bug
reports or git commits upstream. For now I downgraded to 1.12.5 so we
can proceed, and will report the i686 and armhf issues upstream.

Instead of downgrading gstreamer, I think it would be better to simply
disable that test on i686 for now. Most likely, it is due to tests that
are intolerant of the double rounding that occurs on i686 without SSE2,
where the old x87 FP instructions are used instead. The double rounding
happens because x87 operations are performed on 80-bit double-extended
precision floating-point numbers, which must then be rounded a second
time when they are converted to 64-bit doubles as used in C.
What do you think?
Mark

somewhat random tests fail on aarch64 also, different ones from the
other architectures. I'd suggest we wait it out until 1.14.1 and watch
upstream for bug fixes.

Do we know if there are any security fixes in 1.14.0 that are not in
1.12.5? Is the 1.12.x branch still receiving security updates?

As far as I could tell, the 1.12.5 release was mostly security fixes
backported from 1.14.

See the log in
<https://cgit.freedesktop.org/gstreamer/gst-plugins-base/log/?h=1.12>.

[Marius Bakke]

Hello Mark,

Post by Mark H Weaver

Post by Marius Bakke

Post by Mark H Weaver
Hi Marius,

Post by Marius Bakke
https://hydra.gnu.org/jobset/gnu/staging
Fairly minor changes this round, highlights include Wayland 1.15 and
GStreamer 1.14. We narrowly missed Mesa 17.3.9 which was scheduled for
today but delayed, hopefully 17.3.8 doesn't introduce any new bugs.
Results should start ticking in tomorrow.

The main issue I see so far is that 'gst-plugins-base' seems to
consistently fail the "elements/opus" test on i686-linux. It failed
https://hydra.gnu.org/build/2635798

I can reproduce this failure locally, and could not find related bug
reports or git commits upstream. For now I downgraded to 1.12.5 so we
can proceed, and will report the i686 and armhf issues upstream.

Instead of downgrading gstreamer, I think it would be better to simply
disable that test on i686 for now. Most likely, it is due to tests that
are intolerant of the double rounding that occurs on i686 without SSE2,
where the old x87 FP instructions are used instead. The double rounding
happens because x87 operations are performed on 80-bit double-extended
precision floating-point numbers, which must then be rounded a second
time when they are converted to 64-bit doubles as used in C.
What do you think?

Thanks for the insight here. My rationale for downgrading was that
there are a handful other tests that fail on armhf; but I just realized
that we haven't had "gst-plugins-base" working on armhf for a long time.

1.12.5 is also "newer" than 1.14.0. Since there are problems on aarch64
too I went ahead and started the branch with 1.12.5.

Let's revisit this issue on 'core-updates'.

[Marius Bakke]

Hi,
I had pushed the following two texlive-bin patches to staging on Wed Apr
11.
gnu: texlive-bin: Patch texlua shebangs.
gnu: texlive-bin: Use ghostscript executable "gs" in ps2eps.
These patches are missing from the list of patches you mentioned. Did
something go wrong?

The initial email was sent April 16th, while the texlive patches were
added on April 21st. So everything is in order :-)

[Arun Isaac]

Post by Marius Bakke
The initial email was sent April 16th, while the texlive patches were
added on April 21st. So everything is in order :-)

Ok. Thank you for the clarification! :-)