The build system for qtwebengine is complex, but in summary it runs the usual 'make' command which itself runs 'ninja' in a subprocess to actually build the source. The problem is that the 'make' is run with the standard '-j jobs' option which causes it to buffer the output of any subprocess until it completes, and 'ninja' is run with the same parallel option. Since 'ninja' builds the entire source, which can take several hours, no progress output is shown until it completes, at which point all of the output is dumped. It would be far more useful to see the progress output as it is generated. Partial process tree while building: 1067453 /usr/bin/python3.11 /usr/lib/python-exec/python3.11/emerge -a -1 --autounm 1068696 \_ /usr/bin/python3.11 /usr/lib/portage/python3.11/pid-ns-init 1068697 1068697 \_ /usr/bin/python3.11 /usr/lib/portage/python3.11/pid-ns-init 250 25 1068700 \_ [dev-qt/qtwebengine-5.15.9_p20230505] sandbox /usr/lib/portage 1068701 \_ bash /usr/lib/portage/python3.11/ebuild.sh compile 1068717 \_ bash /usr/lib/portage/python3.11/ebuild.sh compile 1068724 \_ bash /usr/lib/portage/python3.11/ebuild-helpers/emake 1068726 \_ make -j4 1068728 \_ make -f Makefile 1069350 \_ make -f Makefile 1069360 \_ make -f Makefile.gn_run 1069435 \_ ninja -j4 -l0 -v -C /var/tmp/p 1069461 \_ /usr/bin/x86_64-pc-linux-gcc \_ [more gcc processes] Tracing shows that process 1068726 (make -j4) is waiting for 1069360 (make -f Makefile.gn_run) to finish before it displays its output, but process 1069435 (ninja) will not finish until the complete source is built. Reproducible: Always Steps to Reproduce: 1. emerge qtwebengine 2. Wait until the configuration phase is finished and the message "Done. Made 14717 targets from 2324 files in 11065ms" is displayed. 3. Wait several hours with no progress output. 4. Observer that the progress is dumped in one black just before the compile phase finishes. Actual Results: Long delay with no progress output at step 3. Expected Results: Progress output displayed while the build is in progress. Workaround: set the qtwebengine build environment to NINJAFLAGS="-j6 -l8 -v" MAKEOPTS="-j1" at the cost of not performing the first part of the build (creating 'gn') in parallel. ebuild version: dev-qt/qtwebengine-5.15.9_p20230505 emerge --info: Portage 3.0.46 (python 3.11.3-final-0, default/linux/amd64/17.1/systemd/merged-usr, gcc-12, glibc-2.36-r8, 5.15.69-gentoo-x86_64 x86_64) ================================================================= System uname: Linux-5.15.69-gentoo-x86_64-x86_64-AMD_A8-9600_RADEON_R7,_10_COMPUTE_CORES_4C+6G-with-glibc2.36 KiB Mem: 7586900 total, 2312712 free KiB Swap: 16777212 total, 16751100 free Timestamp of repository gentoo: Tue, 20 Jun 2023 10:30:01 +0000 Head commit of repository gentoo: 3cac361206f51b91780d3bcc50a41f98f0e191f1 sh bash 5.1_p16-r4 ld GNU ld (Gentoo 2.39 p6) 2.39.0 app-misc/pax-utils: 1.3.5::gentoo app-shells/bash: 5.1_p16-r4::gentoo dev-lang/perl: 5.36.0-r2::gentoo dev-lang/python: 3.10.11::gentoo, 3.11.3::gentoo dev-lang/rust-bin: 1.69.0::gentoo dev-util/cmake: 3.26.4-r1::gentoo dev-util/meson: 1.1.1::gentoo sys-apps/baselayout: 2.13-r1::gentoo sys-apps/sandbox: 2.30-r1::gentoo sys-apps/systemd: 253.3-r1::gentoo sys-devel/autoconf: 2.71-r6::gentoo sys-devel/automake: 1.16.5::gentoo sys-devel/binutils: 2.39-r5::gentoo sys-devel/binutils-config: 5.5::gentoo sys-devel/gcc: 12.2.1_p20230428-r1::gentoo sys-devel/gcc-config: 2.10::gentoo sys-devel/libtool: 2.4.7-r1::gentoo sys-devel/llvm: 16.0.5::gentoo sys-devel/make: 4.4.1-r1::gentoo sys-kernel/linux-headers: 6.1::gentoo (virtual/os-headers) sys-libs/glibc: 2.36-r8::gentoo Repositories: gentoo location: /mnt/auto/portage sync-type: rsync sync-uri: rsync://rsync.gentoo.org/gentoo-portage priority: -1000 volatile: True sync-rsync-verify-max-age: 24 sync-rsync-verify-metamanifest: yes sync-rsync-extra-opts: sync-rsync-verify-jobs: 1 ACCEPT_KEYWORDS="amd64" ACCEPT_LICENSE="@FREE fairuse freedist NVIDIA-r2 Kermit bh-luxi google-chrome AdobeFlash-11.x all-rights-reserved unRAR free-noncomm linux-fw-redistributable no-source-code" CBUILD="x86_64-pc-linux-gnu" CFLAGS="-O2 -march=amdfam10 -mtune=native -pipe -fno-diagnostics-show-option -fno-diagnostics-show-caret" CHOST="x86_64-pc-linux-gnu" CONFIG_PROTECT="/etc /usr/share/gnupg/qualified.txt" CONFIG_PROTECT_MASK="/etc/ca-certificates.conf /etc/dconf /etc/env.d /etc/fonts/fonts.conf /etc/gconf /etc/gentoo-release /etc/revdep-rebuild /etc/sandbox.d" CXXFLAGS="-O2 -march=amdfam10 -mtune=native -pipe -fno-diagnostics-show-option -fno-diagnostics-show-caret" DISTDIR="/var/tmp/distfiles" ENV_UNSET="CARGO_HOME DBUS_SESSION_BUS_ADDRESS DISPLAY GDK_PIXBUF_MODULE_FILE GOBIN GOPATH PERL5LIB PERL5OPT PERLPREFIX PERL_CORE PERL_MB_OPT PERL_MM_OPT XAUTHORITY XDG_CACHE_HOME XDG_CONFIG_HOME XDG_DATA_HOME XDG_RUNTIME_DIR XDG_STATE_HOME" FCFLAGS="-O2 -pipe" FEATURES="assume-digests binpkg-docompress binpkg-dostrip binpkg-logs buildpkg-live config-protect-if-modified distlocks ebuild-locks fixlafiles ipc-sandbox merge-sync multilib-strict network-sandbox news parallel-fetch pid-sandbox preserve-libs protect-owned qa-unresolved-soname-deps sandbox sfperms splitdebug strict unknown-features-warn unmerge-logs unmerge-orphans userfetch userpriv usersandbox usersync xattr" FFLAGS="-O2 -pipe" GENTOO_MIRRORS="http://www.mirrorservice.org/sites/distfiles.gentoo.org/ http://mirror.bytemark.co.uk/gentoo/" LANG="en_GB.UTF-8" LDFLAGS="-Wl,-O1 -Wl,--as-needed" LEX="flex" LINGUAS="en en_GB de fr" MAKEFLAGS="-j6" PKGDIR="/var/cache/binpkgs" PORTAGE_CONFIGROOT="/" PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --omit-dir-times --compress --force --whole-file --delete --stats --human-readable --timeout=180 --exclude=/distfiles --exclude=/local --exclude=/packages --exclude=/.git" PORTAGE_TMPDIR="/var/tmp" SHELL="/bin/bash" USE="X aac acl alsa amd64 bash-completion bzip2 cdr cli crypt cups dbus dri exif ffmpeg fortran ftp gdbm gif iconv jpeg jpeg2k libtirpc mmx mmxext mp3 multilib ncurses nls nptl ogg opengl openmp pam pcre pdf png readline samba seccomp sse sse2 sse3 ssl svg systemd test-rust tiff truetype udev unicode usb vim-syntax vorbis wmf xattr xface xml xpm zlib" ABI_X86="64" ADA_TARGET="gnat_2021" APACHE2_MODULES="authn_core authz_core socache_shmcb unixd actions alias auth_basic authn_alias authn_anon authn_dbm authn_default authn_file authz_dbm authz_default authz_groupfile authz_host authz_owner authz_user autoindex cache cgi cgid dav dav_fs dav_lock deflate dir disk_cache env expires ext_filter file_cache filter headers include info log_config logio mem_cache mime mime_magic negotiation rewrite setenvif speling status unique_id userdir usertrack vhost_alias" CALLIGRA_FEATURES="karbon sheets words" COLLECTD_PLUGINS="df interface irq load memory rrdtool swap syslog" CPU_FLAGS_X86="mmx mmxext sse sse2 sse3" ELIBC="glibc" GPSD_PROTOCOLS="ashtech aivdm earthmate evermore fv18 garmin garmintxt gpsclock greis isync itrax mtk3301 nmea ntrip navcom oceanserver oldstyle oncore rtcm104v2 rtcm104v3 sirf skytraq superstar2 timing tsip tripmate tnt ublox ubx" GRUB_PLATFORMS="pc" INPUT_DEVICES="evdev" KERNEL="linux" L10N="en en-GB de fr" LCD_DEVICES="bayrad cfontz cfontz633 glk hd44780 lb216 lcdm001 mtxorb ncurses text" LIBREOFFICE_EXTENSIONS="presenter-console presenter-minimizer" LUA_SINGLE_TARGET="lua5-1" LUA_TARGETS="lua5-1" OFFICE_IMPLEMENTATION="libreoffice" PHP_TARGETS="php7-4 php8-0" POSTGRES_TARGETS="postgres12 postgres13" PYTHON_SINGLE_TARGET="python3_11" PYTHON_TARGETS="python3_11" RUBY_TARGETS="ruby26" VIDEO_CARDS="dummy fbdev vesa vga radeon amdgpu radeonsi" XTABLES_ADDONS="quota2 psd pknock lscan length2 ipv4options ipset ipp2p iface geoip fuzzy condition tee tarpit sysrq proto steal rawnat logmark ipmark dhcpmac delude chaos account" Unset: ADDR2LINE, AR, ARFLAGS, AS, ASFLAGS, CC, CCLD, CONFIG_SHELL, CPP, CPPFLAGS, CTARGET, CXX, CXXFILT, ELFEDIT, EMERGE_DEFAULT_OPTS, EXTRA_ECONF, F77FLAGS, FC, GCOV, GPROF, INSTALL_MASK, LC_ALL, LD, LFLAGS, LIBTOOL, MAKE, MAKEOPTS, NM, OBJCOPY, OBJDUMP, PORTAGE_BINHOST, PORTAGE_BUNZIP2_COMMAND, PORTAGE_COMPRESS, PORTAGE_COMPRESS_FLAGS, PORTAGE_RSYNC_EXTRA_OPTS, RANLIB, READELF, RUSTFLAGS, SIZE, STRINGS, STRIP, YACC, YFLAGS
Hm, is this because of the default in newer Portage? But you have MAKEOPTS="-j6"...
From the --info output: MAKEFLAGS="-j6" Unset: ... MAKEOPTS, ... Portage doesn't take into account MAKEFLAGS when deciding on whether to use default values or not[1]. The question is, should it do that? 1: https://github.com/gentoo/portage/blob/f537f520422af0900049c760b435ab3e0938240d/lib/portage/package/ebuild/doebuild.py#L588
Ignore the MAKEFLAGS="-j6" in the emerge --info output, this is set in my make.conf but does not seem to be used anywhere - it may be a historical artifact. MAKEOPTS is not set in make.conf which means that it is calculated automatically by doebuild.py (for 4 CPUs). The option for Ninja is manually set in the package environment file, any value will work here.
No, make will respect MAKEFLAGS. It's relevant to whether the default for output synchronisation we added in Portage should be used or not.
Confirmed that the same happens (make -j4, ninja -j4, no progress output) if neither MAKEFLAGS nor MAKEOPTS is set in make.conf or environment.
Yes, that is expected behavior. Portage is "blind" to MAKEFLAGS in make.conf, and makes it's decisions purely on whether MAKEOPTS is set or not. In the case out output synchronization, it adds an extra check if GNUMAKEFLAGS is set to avoid clobbering it on the off chance someone has GNU Make-specific options set (so setting either one will prevent the default output synchronization from being set as well).
Instead of setting MAKEOPTS="-j1" in the qtwebengine env file, you should be able to set MAKEOPTS="-j$(nproc)" and see the same progress output because now MAKEOPTS is set which prevents the output synchronization default from taking effect.
Confirmed that with no MAKE* settings in make.conf but setting NINJAFLAGS="-j7 -l8 -v" MAKEOPTS="-j5" in package env file (random numbers just to make it clear where they come from) results in parallel build of both gn and engine but with logging as expected: 9739 \_ make -j5 10450 \_ ninja -j7 -l8 -v -C /var/tmp/portage/dev-qt/qtwebengine- with the environment of process 9739 containing MAKEOPTS=-j5 Removing MAKEOPTS from the package anv file and allowing it to be automatically calculated gives (no logging in this case): 12244 \_ make -j4 12954 \_ ninja -j7 -l8 -v -C /var/tmp/portage/dev-qt/qtwebengine- with the environment of process 12244 containing MAKEOPTS=-j4 GNUMAKEFLAGS=--output-sync=line
Same problem (no logging until build complete) also seen with net-libs/nodejs, same workaround.
(In reply to Oskari Pirhonen from comment #6) > Yes, that is expected behavior. Portage is "blind" to MAKEFLAGS in > make.conf, and makes it's decisions purely on whether MAKEOPTS is set or > not. In the case out output synchronization, it adds an extra check if > GNUMAKEFLAGS is set to avoid clobbering it on the off chance someone has GNU > Make-specific options set (so setting either one will prevent the default > output synchronization from being set as well). Let's go off MAKEFLAGS as well, even though it's odd for people to set it.
The bug has been referenced in the following commit(s): https://gitweb.gentoo.org/proj/portage.git/commit/?id=87d0a7447f1464157f18a8ba0991719777003325 commit 87d0a7447f1464157f18a8ba0991719777003325 Author: Oskari Pirhonen <xxc3ncoredxx@gmail.com> AuthorDate: 2023-08-13 03:11:15 +0000 Commit: Sam James <sam@gentoo.org> CommitDate: 2023-08-19 15:05:19 +0000 doebuild: gate -Oline behind MAKEFLAGS check Add a check to see if MAKEFLAGS is set (alongside the existing MAKEOPTS and GNUMAKEFLAGS checks) before enabling output synchronization. This is only tangentially related to the bug below via being a (likely rare) case of MAKEFLAGS being set in the wild. Bug: https://bugs.gentoo.org/909009 Signed-off-by: Oskari Pirhonen <xxc3ncoredxx@gmail.com> Signed-off-by: Sam James <sam@gentoo.org> lib/portage/package/ebuild/doebuild.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
Note that this bug shouldn't be automatically closed with the portage-3.0.51 release IMO since it alone doesn't really address the underlying issue. An idea I have for actually fixing it is to remove any output synchronization in the relevant ebuilds themselves. Maybe by either checking for the option in the vars or just unconditionally appending `--output-sync=none`.
The bug has been closed via the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=a6a68c0255b6d3132dd555b1519a035204bffde2 commit a6a68c0255b6d3132dd555b1519a035204bffde2 Author: Sam James <sam@gentoo.org> AuthorDate: 2023-08-20 07:57:47 +0000 Commit: Sam James <sam@gentoo.org> CommitDate: 2023-08-20 07:58:22 +0000 sys-apps/portage: add 3.0.51 Bug: https://bugs.gentoo.org/587484 Closes: https://bugs.gentoo.org/673624 Closes: https://bugs.gentoo.org/691434 Closes: https://bugs.gentoo.org/835927 Closes: https://bugs.gentoo.org/909009 Closes: https://bugs.gentoo.org/911335 Closes: https://bugs.gentoo.org/911629 Closes: https://bugs.gentoo.org/911938 Signed-off-by: Sam James <sam@gentoo.org> sys-apps/portage/Manifest | 1 + sys-apps/portage/portage-3.0.51.ebuild | 234 +++++++++++++++++++++++++++++++++ 2 files changed, 235 insertions(+)
We should pass -Onone like we do in nodejs: commit 223940a28118fc34c3cc2c7b630b83ddc6d9de99 Author: Sam James <sam@gentoo.org> Date: Wed Nov 29 05:12:21 2023 +0000 net-libs/nodejs: adapt to newer Portage's make output synchronisation Newer Portage defaults to GNU Make's output synchronisation if no MAKEOPTS are set by the user. nodejs's Makefile fires up a ninja process which then does all the work, so everything gets buffered and shown at once (once all of nodejs is built). Pass -Onone to override the Portage default as it's not helpful here. Closes: https://bugs.gentoo.org/917245 Signed-off-by: Sam James <sam@gentoo.org>
I assume qt6 doesn't need to do anything here? I don't think it still use make for anything (unless I'm missing something that gn does) and output looks normal to me.
