Conversation
pkgs/applications/networking/instant-messengers/telegram/tdesktop/tg_owt.nix
Outdated
Show resolved
Hide resolved
This comment was marked as spam.
This comment was marked as spam.
|
qtwebengine build error: thrown from random locations ... Detailsor qtwebengine-6.3.1.error.log.gz similar to QTBUG-103573: GCC internal compiler error: Segmentation fault when trying to compile Qt 6.3 simlar to gentoo bug 663770: dev-qt/qtwebengine-5.11.1 fails to build with insufficient memory caused by setting NINJAFLAGS, even with high values like → reverting "fix ninja flags" |
|
review: qt6 compiles + telegram works |
|
Would the ninja fix solve the problems we are having building qtwebengine on hydra? |
There was a problem hiding this comment.
webengine tends to hog even really good build servers. We shouldn't merge this like this.
There was a problem hiding this comment.
im working on this in my branch qt6-631-debug-qtwebengine
the main build process is ninja
so currently im exploring ninja-tokenpool (implements jobclient + jobserver)
to limit the job count across multiple ninja processes
todo: add jobclients to python and node builders in qtwebengine
- src/3rdparty/chromium/third_party/blink/renderer/bindings/scripts/bind_gen/task_queue.py
- src/3rdparty/chromium/third_party/devtools-frontend/src/node_modules/rollup-plugin-terser/rollup-plugin-terser.js
by default, these builders use all cpu cores
There was a problem hiding this comment.
I think building is better than being well behaved. This build break blocks quite a few packages. qtwebengine is expensive to build but it isn't that bad.
There was a problem hiding this comment.
status update:
i have added jobclient to
src/3rdparty/chromium/third_party/devtools-frontend/src/node_modules/rollup-plugin-terser/rollup-plugin-terser.js
src/3rdparty/chromium/third_party/blink/renderer/bindings/scripts/bind_gen/task_queue.py
src/3rdparty/chromium/mojo/public/tools/mojom/mojom_parser.py
src/3rdparty/gn/src/util/worker_pool.cc
this is mostly working ...
in rare cases, the workers dont release their tokens back to the jobserver
which results in underloading, for example load is 20 when it should be 32
i hope to fix/workaround this in the next ~2 days
but feel free to merge this PR as is
There was a problem hiding this comment.
in rare cases, the workers dont release their tokens back to the jobserver
What is the chance that the number of tokens drops to zero or a very low number? It would be annoying if that could hang a build.
There was a problem hiding this comment.
here, the problem are bugs in my jobclients
the load slowly drops to 1 = no parallelism (but the build does not hang)
but its also a well-known weakness/bug in the jobserver protocol
milahu/gnumake-tokenpool#1 (comment)
There was a problem hiding this comment.
I guess that almost any time a job fails the whole build is going to fail anyways so it shouldn't be too big of a problem. In that I case I think it would be fine with a warning. But if you think you can fix it that is even better.
There was a problem hiding this comment.
fixed in milahu/gnumake-tokenpool@247c42b
in javascript, fs.readSync can block when the pipe is empty
now the max load is 32 == NIX_BUILD_CORES
some outliers around T=500, probably cmake, but meh
There was a problem hiding this comment.
I tried building e206f87 and it has stalled out as well
same here : (
with qt6-631-debug-qtwebengine-2 i see that the javascript workers are blocking ...
load Load rss time spr cpr proc
20.0 18.1 9869 471 123 2 bash 1: bash -e /nix/store/9krlzvny65gdc8s7kpb6lkx8cd02c25b-default-builder.sh # {'cwd': '/build/qtwebengine-everywhere-src-6.3.1/build', 'exe': '/nix/store/j9rc786ylq8cid4zgcn6idknilkbd4ax-bash-5.1-p16/bin/bash'}
20.0 18.1 9863 425 121 1 ninja 13856: ninja -j32 -l32 --tokenpool-master # {'exe': '/nix/store/k38101zidsncnv0qvcia82n1nmvkb57a-ninja-tokenpool-unstable-2022-02-13/bin/ninja'}
20.0 18.1 9858 378 120 40 ninja 17319: ninja -C /build/qtwebengine-everywhere-src-6.3.1/build/src/core/Release/x86_64 QtWebEngineCore # {'cwd': '/build/qtwebengine-everywhere-src-6.3.1/build/src/core/Release/x86_64', 'exe': '/nix/store/k38101zidsncnv0qvcia82n1nmvkb57a-ninja-tokenpool-unstable-2022-02-13/bin/ninja'}
1.0 1.1 956 229 2 1 python3 31547: python3 ../../../../../src/3rdparty/chromium/third_party/blink/renderer/bindings/scripts/generate_bindings.py --web_idl_database gen/third_party/blink/renderer/bindings/web_idl_database.pickle --root_src_dir ../../../../../src/3rdparty/chromium/ --root_gen_dir gen --output_reldir core=third_party/blink/renderer/bindings/core/v8/ --output_reldir modules=third_party/blink/renderer/bindings/modules/v8/ --format_generated_files enumeration callback_function callback_interface dictionary interface namespace typedef union # {'exe': '/nix/store/i6svmmq24wl4i10dx0m8blrpl3m3fq4b-python3-3.10.5/bin/python3.10'}
1.0 1.0 790 228 1 0 python3 31861: python3 ../../../../../src/3rdparty/chromium/third_party/blink/renderer/bindings/scripts/generate_bindings.py --web_idl_database gen/third_party/blink/renderer/bindings/web_idl_database.pickle --root_src_dir ../../../../../src/3rdparty/chromium/ --root_gen_dir gen --output_reldir core=third_party/blink/renderer/bindings/core/v8/ --output_reldir modules=third_party/blink/renderer/bindings/modules/v8/ --format_generated_files enumeration callback_function callback_interface dictionary interface namespace typedef union # {'exe': '/nix/store/i6svmmq24wl4i10dx0m8blrpl3m3fq4b-python3-3.10.5/bin/python3.10'}
0.0 0.0 132 225 3 1 python3 32929: python3 ../../../../../src/3rdparty/chromium/third_party/node/node.py ../../../../../src/3rdparty/chromium/third_party/devtools-frontend/src/node_modules/rollup/dist/bin/rollup --silent --config ../../../../../src/3rdparty/chromium/third_party/devtools-frontend/src/scripts/build/rollup.config.js --input gen/third_party/devtools-frontend/src/front_end/entrypoints/formatter_worker/FormatterActions.prebundle.js --file gen/third_party/devtools-frontend/src/front_end/entrypoints/formatter_worker/FormatterActions.js --configDCHECK # {'exe': '/nix/store/i6svmmq24wl4i10dx0m8blrpl3m3fq4b-python3-3.10.5/bin/python3.10'}
0.0 0.0 122 224 2 1 node 32930: node ../../../../../src/3rdparty/chromium/third_party/devtools-frontend/src/node_modules/rollup/dist/bin/rollup --config ../../../../../src/3rdparty/chromium/third_party/devtools-frontend/src/scripts/build/rollup.config.js --input gen/third_party/devtools-frontend/src/front_end/entrypoints/formatter_worker/FormatterActions.prebundle.js --file gen/third_party/devtools-frontend/src/front_end/entrypoints/formatter_worker/FormatterActions.js --configDCHECK # {'exe': '/nix/store/2ph85rmc3j8n5k679y4zp0qvqzrmwwkn-nodejs-16.15.0/bin/node'}
0.0 0.0 25 224 1 0 node 32995: node /build/qtwebengine-everywhere-src-6.3.1/src/3rdparty/chromium/third_party/devtools-frontend/src/node_modules/jest-worker/build/workers/processChild.js # {'exe': '/nix/store/2ph85rmc3j8n5k679y4zp0qvqzrmwwkn-nodejs-16.15.0/bin/node'}
0.0 0.0 149 223 3 1 python3 33114: python3 ../../../../../src/3rdparty/chromium/third_party/node/node.py ../../../../../src/3rdparty/chromium/third_party/devtools-frontend/src/node_modules/rollup/dist/bin/rollup --silent --config ../../../../../src/3rdparty/chromium/third_party/devtools-frontend/src/scripts/build/rollup.config.js --input gen/third_party/devtools-frontend/src/front_end/core/platform/platform.prebundle.js --file gen/third_party/devtools-frontend/src/front_end/core/platform/platform.js --configDCHECK # {'exe': '/nix/store/i6svmmq24wl4i10dx0m8blrpl3m3fq4b-python3-3.10.5/bin/python3.10'}
0.0 0.0 139 223 2 1 node 33136: node ../../../../../src/3rdparty/chromium/third_party/devtools-frontend/src/node_modules/rollup/dist/bin/rollup --config ../../../../../src/3rdparty/chromium/third_party/devtools-frontend/src/scripts/build/rollup.config.js --input gen/third_party/devtools-frontend/src/front_end/core/platform/platform.prebundle.js --file gen/third_party/devtools-frontend/src/front_end/core/platform/platform.js --configDCHECK # {'exe': '/nix/store/2ph85rmc3j8n5k679y4zp0qvqzrmwwkn-nodejs-16.15.0/bin/node'}
0.0 0.0 25 223 1 0 node 33512: node /build/qtwebengine-everywhere-src-6.3.1/src/3rdparty/chromium/third_party/devtools-frontend/src/node_modules/jest-worker/build/workers/processChild.js # {'exe': '/nix/store/2ph85rmc3j8n5k679y4zp0qvqzrmwwkn-nodejs-16.15.0/bin/node'}
0.0 0.0 136 223 3 1 python3 33246: python3 ../../../../../src/3rdparty/chromium/third_party/node/node.py ../../../../../src/3rdparty/chromium/third_party/devtools-frontend/src/node_modules/rollup/dist/bin/rollup --silent --config ../../../../../src/3rdparty/chromium/third_party/devtools-frontend/src/scripts/build/rollup.config.js --input gen/third_party/devtools-frontend/src/front_end/models/heap_snapshot_model/heap_snapshot_model.prebundle.js --file gen/third_party/devtools-frontend/src/front_end/models/heap_snapshot_model/heap_snapshot_model.js --configDCHECK # {'exe': '/nix/store/i6svmmq24wl4i10dx0m8blrpl3m3fq4b-python3-3.10.5/bin/python3.10'}
0.0 0.0 126 223 2 1 node 33267: node ../../../../../src/3rdparty/chromium/third_party/devtools-frontend/src/node_modules/rollup/dist/bin/rollup --config ../../../../../src/3rdparty/chromium/third_party/devtools-frontend/src/scripts/build/rollup.config.js --input gen/third_party/devtools-frontend/src/front_end/models/heap_snapshot_model/heap_snapshot_model.prebundle.js --file gen/third_party/devtools-frontend/src/front_end/models/heap_snapshot_model/heap_snapshot_model.js --configDCHECK # {'exe': '/nix/store/2ph85rmc3j8n5k679y4zp0qvqzrmwwkn-nodejs-16.15.0/bin/node'}
0.0 0.0 25 222 1 0 node 33645: node /build/qtwebengine-everywhere-src-6.3.1/src/3rdparty/chromium/third_party/devtools-frontend/src/node_modules/jest-worker/build/workers/processChild.js # {'exe': '/nix/store/2ph85rmc3j8n5k679y4zp0qvqzrmwwkn-nodejs-16.15.0/bin/node'}
note the time column. the workers hang since 200 seconds with average load zero (2nd column)
blocked by ... ninja-build/ninja#2164
|
|
|
This pull request has been mentioned on NixOS Discourse. There might be relevant details there: https://discourse.nixos.org/t/nix-build-profiling-memory-and-cpu-usage/19934/1 |
|
Result of 2 packages marked as broken and skipped:
41 packages built:
|
|
This appears to fix the broken qtwebengine build so I think we should merge it. We can make other improvements as we go. Maybe file tickets for the failure to respect the cores config? |
There was a problem hiding this comment.
Are there plans to upstream these patches? Do we want to maintain these fairly significant changes to the build process?
To be honest I'm not a huge fan of the significant changes and complicated patching.
There was a problem hiding this comment.
i linked upstream issues in my first post
what is worse?
- wait for upstream
- overload nix build servers
to review my patches, you would have to review ...
- 300 lines in tokenpool.py
- 260 lines in tokenpool.js
- tokenpool-gnu-make.cc etc ...
these are 99% copy-paste from the tokenpool-master fork via ninja-build/ninja#1139
650 lines of patches
$ ( cd pkgs/development/libraries/qt-6/modules/patches/qtwebengine && wc -l * )
129 0001-task_queue.py-add-jobclient.patch
24 0002-build_inspector_overlay.py-fix-inherit-fds.patch
282 0003-gn-add-jobclient.patch
95 0004-node.py-add-debug.patch
121 0005-mojom_parser.py-add-jobclient.patch
651 total
There was a problem hiding this comment.
Yeah, I agree. It isn't pretty either way.
I think my preferences would be:
- Accept PR as-is, with patches going upstream so that we can update to whatever upstream does once it happens. This way we just have temporary maintenance burden.
- Suck it up and live with the poor resource usage.
- Maintain these patches out-of-tree forever.
But I'm not really sure how much of a problem the poor scheduling is. Also if you are happy to maintain these patches for a long time then worst case is that we rip them out if we don't want to maintain them in the future.
I would love to have more opinions here.
The existing package itself is not broken. It's a ulimit issue with Nix itself. See here: Further, these build patches seem to cause the builder to stall out completely at random places during the build. Each time I try to build it, it stalls in a different place, so I cannot complete the build because it just stalls permanently. Considering that, I am not in favor of the build patches, as without them the build completes just fine, although it does seem to progress more slowly. I would take a reliable build over a fast one any day. |
|
I tried building e206f87 and it has stalled out as well. Definitely can't be merged with the stalling. |
yes, e45d046 works but ignores NIX_BUILD_CORES |
|
Ok, what do you think about resetting this PR to e45d046? Then we can deal with the other improvements in a different PR to be considered in isolation? |
|
yes. just give me a few more minutes, ideally i just messed up the rebase |
|
Result of 2 packages marked as broken and skipped:
41 packages built:
|
|
please try to build qt6-631-debug-qtwebengine-3 |
|
I get an error on 8f48223 log% nix-build -A qt6.qtwebengine these 12 derivations will be built: /nix/store/40wc4c6y9gldy8vsx7swkgclih03p4hn-source.drv /nix/store/83sj2lb3lc73g1hpgjj89qh53dbsfps8-qtbase-6.3.1.drv /nix/store/5nk779ghm8dygif8fkyi6ibg88ba2krd-qtserialport-6.3.1.drv /nix/store/x4sawdx3xzwxrf52nx9lwsghfs6dhm10-qtshadertools-6.3.1.drv /nix/store/cidc9vdjgyq2zi5jsjpf29fjp9cgm8ng-qtdeclarative-6.3.1.drv /nix/store/6arridv0q268jrh87ddj5f1nkgn3k0fq-qtpositioning-6.3.1.drv /nix/store/b1yfbdw04nfih0z668jm696glaim78kj-qtwebsockets-6.3.1.drv /nix/store/lv9g4pr6pydbrp83ylxl3qj0z7clnxk0-python3.9-gnumake-tokenpool-0.0.1.drv /nix/store/lzi42xh5y2r9vi49b3fb1q5fcsij7gbs-python3.9-nix-build-profiler-0.0.1.drv /nix/store/rx5h1zrdqlhk73j66q3d1y81rds1jn7c-qtwebchannel-6.3.1.drv /nix/store/x3hz687a372718w9qwp8qxxi8wkfvl14-ninja-tokenpool-unstable-2022-02-13.drv /nix/store/kwcwnrhd9pw1cy46j3y9lzazk9d5sjrk-qtwebengine-6.3.1.drv these 15 paths will be fetched (24.15 MiB download, 98.48 MiB unpacked): /nix/store/0p53nw75f5ixvwzwv5xynbmkd56f4xx1-stoken-0.92 /nix/store/46acy2byia8xb99z4addfrlrz4fq7kbs-pipewire-0.3.51-lib /nix/store/4xp79p3zg45ffiw6xmrpn33rm328i569-ffmpeg-4.4.1-dev /nix/store/5179ywqpvqlbfa2hasr1as74rq0ld4n2-networkmanager-1.38.0 /nix/store/53sb4zq7201bs5wg102yy03af26d67vd-ffmpeg-4.4.1 /nix/store/b633fhyfxcw95hybskdq15zdy3jsyhi0-pipewire-0.3.51-dev /nix/store/cw1nbvyxva8gqiazmwdif91vkh0q9p0p-libproxy-0.4.17 /nix/store/l34rlzmpy1criasx6l0vjb63dfqvvj63-ffmpeg-4.4.1-bin /nix/store/mkzxyh06njyq485gavmvjcg67lzdqaqv-ffmpeg-4.4.1 /nix/store/npqzzphxiwfhz3sshhx3vh2zp1x13hjm-SDL2-2.0.22 /nix/store/q1xladjz5x2rzs3icr1y43ymn384g29b-pipewire-0.3.51 /nix/store/vlclik0ciwfhplj06y9iv9py6h5n9594-gtk+3-3.24.34-dev /nix/store/wyl82a2pqhr2awwavbklh1cz16rhab2q-openconnect-8.20 /nix/store/x38hs5vn93l68p2mw0i5xb535myz2p3k-libproxy-0.4.17-dev /nix/store/z61p740r7wld3rljzqknlj4jc2wp25hs-cups-2.4.2-dev copying path '/nix/store/z61p740r7wld3rljzqknlj4jc2wp25hs-cups-2.4.2-dev' from 'https://cache.nixos.org'... copying path '/nix/store/mkzxyh06njyq485gavmvjcg67lzdqaqv-ffmpeg-4.4.1' from 'https://cache.nixos.org'... building '/nix/store/x3hz687a372718w9qwp8qxxi8wkfvl14-ninja-tokenpool-unstable-2022-02-13.drv'... building '/nix/store/40wc4c6y9gldy8vsx7swkgclih03p4hn-source.drv'... copying path '/nix/store/0p53nw75f5ixvwzwv5xynbmkd56f4xx1-stoken-0.92' from 'https://cache.nixos.org'... unpacking sources unpacking source archive /nix/store/wgyxh790z2sqgk4mwk85m5x10cgjnbjm-source source root is source patching sources configuring no configure script, doing nothing building bootstrapping ninja... copying path '/nix/store/vlclik0ciwfhplj06y9iv9py6h5n9594-gtk+3-3.24.34-dev' from 'https://cache.nixos.org'... copying path '/nix/store/wyl82a2pqhr2awwavbklh1cz16rhab2q-openconnect-8.20' from 'https://cache.nixos.org'... trying https://github.com/milahu/gnumake-tokenpool/archive/7eba965adfafa429333dd40d7d920085578185ba.tar.gz % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0 copying path '/nix/store/5179ywqpvqlbfa2hasr1as74rq0ld4n2-networkmanager-1.38.0' from 'https://cache.nixos.org'... 100 105k 100 105k 0 0 321k 0 --:--:-- --:--:-- --:--:-- 321k unpacking source archive /build/7eba965adfafa429333dd40d7d920085578185ba.tar.gz building '/nix/store/lv9g4pr6pydbrp83ylxl3qj0z7clnxk0-python3.9-gnumake-tokenpool-0.0.1.drv'... Sourcing python-remove-tests-dir-hook Sourcing python-catch-conflicts-hook.sh Sourcing python-remove-bin-bytecode-hook.sh Sourcing setuptools-build-hook Using setuptoolsBuildPhase Using setuptoolsShellHook Sourcing pip-install-hook Using pipInstallPhase Sourcing python-imports-check-hook.sh Using pythonImportsCheckPhase Sourcing python-namespaces-hook Sourcing setuptools-check-hook Using setuptoolsCheckPhase unpacking sources unpacking source archive /nix/store/759126rs3vqw3abx68lsq3i9s5v1gm4s-source do not know how to unpack source archive /nix/store/759126rs3vqw3abx68lsq3i9s5v1gm4s-source error: builder for '/nix/store/lv9g4pr6pydbrp83ylxl3qj0z7clnxk0-python3.9-gnumake-tokenpool-0.0.1.drv' failed with exit code 1 error: 1 dependencies of derivation '/nix/store/lzi42xh5y2r9vi49b3fb1q5fcsij7gbs-python3.9-nix-build-profiler-0.0.1.drv' failed to build error: 1 dependencies of derivation '/nix/store/kwcwnrhd9pw1cy46j3y9lzazk9d5sjrk-qtwebengine-6.3.1.drv' failed to build |
|
youre hitting NixOS/nix#6572 workaround |
|
Same issue: log8:48:15% nix-build -E ' let pkgs = import <nixpkgs> {}; in pkgs.fetchFromGitHub { owner = "milahu"; repo = "gnumake-tokenpool"; rev = "4eb559ae323bef153cbe1d0a5e3496b377fb7856"; sha256 = "v3UqPi4fCnN86xmp4eOSz8IXNH70E8Kvvq6HVOiCiwk="; } ' /nix/store/nqw44wdq34xd0klarc6li51lv8iyihiv-source 0 kevinryzen:~/p/nixpkgs#8f482236935 8:48:16% nix-build -A qt6.qtwebengine these 12 derivations will be built: /nix/store/40wc4c6y9gldy8vsx7swkgclih03p4hn-source.drv /nix/store/83sj2lb3lc73g1hpgjj89qh53dbsfps8-qtbase-6.3.1.drv /nix/store/5nk779ghm8dygif8fkyi6ibg88ba2krd-qtserialport-6.3.1.drv /nix/store/x4sawdx3xzwxrf52nx9lwsghfs6dhm10-qtshadertools-6.3.1.drv /nix/store/cidc9vdjgyq2zi5jsjpf29fjp9cgm8ng-qtdeclarative-6.3.1.drv /nix/store/6arridv0q268jrh87ddj5f1nkgn3k0fq-qtpositioning-6.3.1.drv /nix/store/b1yfbdw04nfih0z668jm696glaim78kj-qtwebsockets-6.3.1.drv /nix/store/lv9g4pr6pydbrp83ylxl3qj0z7clnxk0-python3.9-gnumake-tokenpool-0.0.1.drv /nix/store/lzi42xh5y2r9vi49b3fb1q5fcsij7gbs-python3.9-nix-build-profiler-0.0.1.drv /nix/store/rx5h1zrdqlhk73j66q3d1y81rds1jn7c-qtwebchannel-6.3.1.drv /nix/store/x3hz687a372718w9qwp8qxxi8wkfvl14-ninja-tokenpool-unstable-2022-02-13.drv /nix/store/kwcwnrhd9pw1cy46j3y9lzazk9d5sjrk-qtwebengine-6.3.1.drv these 3 paths will be fetched (8.32 MiB download, 26.03 MiB unpacked): /nix/store/4xp79p3zg45ffiw6xmrpn33rm328i569-ffmpeg-4.4.1-dev /nix/store/53sb4zq7201bs5wg102yy03af26d67vd-ffmpeg-4.4.1 /nix/store/l34rlzmpy1criasx6l0vjb63dfqvvj63-ffmpeg-4.4.1-bin copying path '/nix/store/53sb4zq7201bs5wg102yy03af26d67vd-ffmpeg-4.4.1' from 'https://cache.nixos.org'... building '/nix/store/x3hz687a372718w9qwp8qxxi8wkfvl14-ninja-tokenpool-unstable-2022-02-13.drv'... building '/nix/store/40wc4c6y9gldy8vsx7swkgclih03p4hn-source.drv'... building '/nix/store/83sj2lb3lc73g1hpgjj89qh53dbsfps8-qtbase-6.3.1.drv'... unpacking sources unpacking source archive /nix/store/wgyxh790z2sqgk4mwk85m5x10cgjnbjm-source source root is source patching sources configuring no configure script, doing nothing building bootstrapping ninja... |
|
sorry, i had the wrong version Details |
|
That built for me. |
|
Any objections to merging this PR as-is? We can then evaluate other improvements as independent followups? |
|
fine with me. the fix for qtwebengine still needs some polish = time |
|
This is failing on nixos-unstable with:
|
|
This appears to come from https://sourcegraph.com/github.com/qt/qtwebengine@126584f52e883a8caaacb79c34e9e0779260d61f/-/blob/cmake/Functions.cmake?L1226-1229&subtree=true which uses |
|
It's not clear to me why this worked before merging. Doesn't seem like it would be a merge conflict. |
see the condition edit: similar upstream issue https://bugreports.qt.io/browse/QTBUG-97568 in this case, i guess upstream would accept a patch |
|
I'm still not sure why this was different before and after the merge though. I built on the same nix-daemon for both. Unless nixpkgs-review is setting funny options. |
|
Yes, but unless I am missing something it shouldn't be the (only) problem here.
|
|
shrug ... edit: possibly also a symptom of cpu overloading |
But from the log on my machine. The hydra log does have "Open files limit 1024" though. |
|
Hmm, It seems that the code attempts to extract the integer output from ulimitOutput into function(check_for_ulimit)
message("-- Checking 'ulimit -n'")
execute_process(COMMAND bash -c "ulimit -n"
OUTPUT_VARIABLE ulimitOutput
)
string(REGEX MATCHALL "[0-9]+" limit "${ulimitOutput}")
message(" -- Open files limit ${limit}")
if(NOT (QT_FEATURE_use_gold_linker OR QT_FEATURE_use_lld_linker) AND ulimitOutput LESS 4096)
# ... snip ...
endif()
endfunction()But the output of |
|
Ah, I found the problem. |

Description of changes
use latest patch release of qt
https://www.qt.io/blog/qt-6.3.1-released
...
upstream issues
patches for jobclient support were sent to ...
Child packages
tdesktop 3.7.3 throws build error → update to pre-release 3.7.6
tdesktop 3.7.6 throws build error → update tdesktop.tg_owt (upstream issue)
Review
Details
Things done
sandbox = trueset innix.conf? (See Nix manual)nix-shell -p nixpkgs-review --run "nixpkgs-review rev HEAD". Note: all changes have to be committed, also see nixpkgs-review usage./result/bin/)nixos/doc/manual/md-to-db.shto update generated release notes