EESSI / software-layer

Software layer of the EESSI project
https://eessi.github.io/docs/software_layer
GNU General Public License v2.0
20 stars 43 forks source link

{2023.06}[2023a] Paraver 4.11.4 #570

Open boegel opened 1 month ago

eessi-bot-aws[bot] commented 1 month ago

Instance eessi-bot-mc-aws is configured to build:

eessi-bot-aws[bot] commented 1 month ago

Instance eessi-bot-mc-azure is configured to build:

boegel commented 1 month ago

bot: build repo:eessi.io-2023.06-software arch:x86_64/intel/skylake_avx512

eessi-bot-aws[bot] commented 1 month ago
Updates by the bot instance eessi-bot-mc-aws (click for details) - received bot command `build repo:eessi.io-2023.06-software arch:x86_64/intel/skylake_avx512` from `boegel` - expanded format: `build repository:eessi.io-2023.06-software architecture:x86_64/intel/skylake_avx512` - handling command `build repository:eessi.io-2023.06-software architecture:x86_64/intel/skylake_avx512` resulted in: - submitted job `10464`, for details & status see https://github.com/EESSI/software-layer/pull/570#issuecomment-2106248866
eessi-bot-aws[bot] commented 1 month ago
Updates by the bot instance eessi-bot-mc-azure (click for details) - received bot command `build repo:eessi.io-2023.06-software arch:x86_64/intel/skylake_avx512` from `boegel` - expanded format: `build repository:eessi.io-2023.06-software architecture:x86_64/intel/skylake_avx512` - handling command `build repository:eessi.io-2023.06-software architecture:x86_64/intel/skylake_avx512` resulted in: - no jobs were submitted
eessi-bot-aws[bot] commented 1 month ago
New job on instance eessi-bot-mc-aws for architecture x86_64-intel-skylake_avx512 for repository eessi.io-2023.06-software in job dir /project/def-users/SHARED/jobs/2024.05/pr_570/10464 date job status comment
May 12 13:33:27 UTC 2024 submitted job id 10464 awaits release by job manager
May 12 13:34:05 UTC 2024 released job awaits launch by Slurm scheduler
May 12 13:39:12 UTC 2024 running job 10464 is running
May 12 13:58:58 UTC 2024 finished
:cry: FAILURE (click triangle for details)
Reason
EasyConfig not found during missing installation check. Are you sure all PRs referenced have been merged in EasyBuild?
Details
:white_check_mark: job output file slurm-10464.out
:x: found message matching ERROR:
:x: found message matching FAILED:
:x: found message matching required modules missing:
:x: no message matching No missing installations
:white_check_mark: found message matching .tar.gz created!
Artefacts
eessi-2023.06-software-linux-x86_64-intel-skylake_avx512-1715521365.tar.gzsize: 8 MiB (8568967 bytes)
entries: 514
modules under _2023.06/software/linux/x86_64/intel/skylakeavx512/modules/all
elfutils/0.189-GCCcore-12.3.0.lua
Graphene/1.10.8-GCCcore-12.3.0.lua
GStreamer/1.22.5-GCC-12.3.0.lua
software under _2023.06/software/linux/x86_64/intel/skylakeavx512/software
elfutils/0.189-GCCcore-12.3.0
Graphene/1.10.8-GCCcore-12.3.0
GStreamer/1.22.5-GCC-12.3.0
other under _2023.06/software/linux/x86_64/intel/skylakeavx512
no other files in tarball
May 12 13:58:58 UTC 2024 test result
:grin: SUCCESS (click triangle for details)
ReFrame Summary
[ PASSED ] Ran 10/10 test case(s) from 10 check(s) (0 failure(s), 0 skipped, 0 aborted)
Details
:white_check_mark: job output file slurm-10464.out
:x: found message matching ERROR:
:white_check_mark: no message matching [\s*FAILED\s*].*Ran .* test case
boegel commented 1 month ago

bot: build repo:eessi.io-2023.06-software arch:aarch64/neoverse_v1

eessi-bot-aws[bot] commented 1 month ago
Updates by the bot instance eessi-bot-mc-aws (click for details) - received bot command `build repo:eessi.io-2023.06-software arch:aarch64/neoverse_v1` from `boegel` - expanded format: `build repository:eessi.io-2023.06-software architecture:aarch64/neoverse_v1` - handling command `build repository:eessi.io-2023.06-software architecture:aarch64/neoverse_v1` resulted in: - submitted job `10466`, for details & status see https://github.com/EESSI/software-layer/pull/570#issuecomment-2106252623
eessi-bot-aws[bot] commented 1 month ago
Updates by the bot instance eessi-bot-mc-azure (click for details) - received bot command `build repo:eessi.io-2023.06-software arch:aarch64/neoverse_v1` from `boegel` - expanded format: `build repository:eessi.io-2023.06-software architecture:aarch64/neoverse_v1` - handling command `build repository:eessi.io-2023.06-software architecture:aarch64/neoverse_v1` resulted in: - no jobs were submitted
eessi-bot-aws[bot] commented 1 month ago
New job on instance eessi-bot-mc-aws for architecture aarch64-neoverse_v1 for repository eessi.io-2023.06-software in job dir /project/def-users/SHARED/jobs/2024.05/pr_570/10466 date job status comment
May 12 13:46:22 UTC 2024 submitted job id 10466 awaits release by job manager
May 12 13:46:25 UTC 2024 released job awaits launch by Slurm scheduler
May 12 13:51:37 UTC 2024 running job 10466 is running
May 12 14:03:11 UTC 2024 finished
:cry: FAILURE (click triangle for details)
Reason
EasyConfig not found during missing installation check. Are you sure all PRs referenced have been merged in EasyBuild?
Details
:white_check_mark: job output file slurm-10466.out
:x: found message matching ERROR:
:x: found message matching FAILED:
:x: found message matching required modules missing:
:x: no message matching No missing installations
:white_check_mark: found message matching .tar.gz created!
Artefacts
eessi-2023.06-software-linux-aarch64-neoverse_v1-1715522079.tar.gzsize: 8 MiB (8577407 bytes)
entries: 513
modules under _2023.06/software/linux/aarch64/neoversev1/modules/all
elfutils/0.189-GCCcore-12.3.0.lua
Graphene/1.10.8-GCCcore-12.3.0.lua
GStreamer/1.22.5-GCC-12.3.0.lua
software under _2023.06/software/linux/aarch64/neoversev1/software
elfutils/0.189-GCCcore-12.3.0
Graphene/1.10.8-GCCcore-12.3.0
GStreamer/1.22.5-GCC-12.3.0
other under _2023.06/software/linux/aarch64/neoversev1
no other files in tarball
May 12 14:03:11 UTC 2024 test result
:grin: SUCCESS (click triangle for details)
ReFrame Summary
[ PASSED ] Ran 10/10 test case(s) from 10 check(s) (0 failure(s), 0 skipped, 0 aborted)
Details
:white_check_mark: job output file slurm-10466.out
:x: found message matching ERROR:
:white_check_mark: no message matching [\s*FAILED\s*].*Ran .* test case
boegel commented 1 month ago

Building of GST-plugins-base-1.22.5-GCC-12.3.0.eb dependency fails with:

[146/754] Generating gst-libs/gst/tag/GstTag-1.0.gir with a custom command (wrapped by meson to set env)
FAILED: gst-libs/gst/tag/GstTag-1.0.gir
...
/bin/bash: /cvmfs/software.eessi.io/versions/2023.06/compat/linux/aarch64/usr/lib64/libtinfo.so.6: no version information available (required by /bin/bash)
/bin/bash: /lib/aarch64-linux-gnu/libc.so.6: version `GLIBC_2.33' not found (required by /cvmfs/software.eessi.io/versions/2023.06/compat/linux/aarch64/usr/lib64/libtinfo.so.6)
...
ninja: build stopped: subcommand failed.

This is a problem that looks very similar to what we ran into in with at-spi2-core #428, which was fixed in #454...

ocaisa commented 1 month ago

This now requires a sync with the default branch for CI or builds to pass

bedroge commented 3 weeks ago

Building of GST-plugins-base-1.22.5-GCC-12.3.0.eb dependency fails with:

[146/754] Generating gst-libs/gst/tag/GstTag-1.0.gir with a custom command (wrapped by meson to set env)
FAILED: gst-libs/gst/tag/GstTag-1.0.gir
...
/bin/bash: /cvmfs/software.eessi.io/versions/2023.06/compat/linux/aarch64/usr/lib64/libtinfo.so.6: no version information available (required by /bin/bash)
/bin/bash: /lib/aarch64-linux-gnu/libc.so.6: version `GLIBC_2.33' not found (required by /cvmfs/software.eessi.io/versions/2023.06/compat/linux/aarch64/usr/lib64/libtinfo.so.6)
...
ninja: build stopped: subcommand failed.

This is a problem that looks very similar to what we ran into in with at-spi2-core #428, which was fixed in #454...

Yes, it indeed looks like the same problem. For at-spi2-core I've fixed it by adding a --lib-dirs-envvar=FILTER_LD_LIBRARY_PATH (which prevents that $LD_LIBRARY_PATH is being set) to gir_extra_args near the affected call to gnome.generate_gir. I see several meson.build files in the sources of gst-plugins-base that call gnome.generate_gir, though, but all of them set gir_extra_args = gir_init_section + something_else. The gir_init_section is only defined in the meson.build of the top-level dir, so you can probably add the same thing there.

However, since we're now running into the same thing again, we could perhaps also look into patching GObject-Retrospection itself, to prevent that we have to do this same fix over and over again in several other packages...

bedroge commented 3 weeks ago

By setting --lib-dirs-envvar you're basically preventing that runtime_path_envvar contains LD_LIBRARY_PATH (https://gitlab.gnome.org/GNOME/gobject-introspection/-/blob/1.76.1/giscanner/ccompiler.py?ref_type=tags#L224), and in turn that means that LD_LIBRARY_PATH won't be set as an environment variable a bit further down: https://gitlab.gnome.org/GNOME/gobject-introspection/-/blob/1.76.1/giscanner/ccompiler.py?ref_type=tags#L267

We could try patching this code somehow, but I'm not sure how we can detect here that we're filtering LD_LIBRARY_PATH? Maybe we can check in that if statement if LD_LIBRARY_PATH is already empty (i.e. if it's not in os.environ), and if so, never set it anyway? Because if we're not filtering it, the GObject-Introspection module itself should already have set LD_LIBRARY_PATH.

@boegel thoughts?

edit: gave it a quicky try with at-spi2-core by disabling the hook and changing the else on line https://gitlab.gnome.org/GNOME/gobject-introspection/-/blob/1.76.1/giscanner/ccompiler.py?ref_type=tags#L271 to elif envvar != 'LD_LIBRARY_PATH':, and that indeed worked fine.