Re: [gentoo-user] Re: Highlight certain packages being upgraded
On Monday, 10 July 2023 02:25:43 BST Nikos Chantziaras wrote:
> On 08/07/2023 11:33, Dale wrote:
> > that excessively long qt package
>
> Off-topic, but just in case you mean qtwebengine, I was able to get rid
> of it by putting "-webengine" in my USE flags. After a world update, a
> depclean should then remove it from the system.
>
> You might have to juggle a few other USE flags in specific packages to
> make it happen though, I forgot. It's been a while.
Nice idea, but it's not for me I'm afraid:
$ emerge -cvp qtwebengine
Calculating dependencies ... done!
dev-qt/qtwebengine-5.15.10_p20230623 pulled in by:
app-office/kalendar-23.04.3 requires >=dev-qt/qtwebengine-5.15.9:5
kde-apps/akregator-23.04.3 requires >=dev-qt/qtwebengine-5.15.9:5
kde-apps/grantlee-editor-23.04.3 requires >=dev-qt/
qtwebengine-5.15.9:5[widgets]
kde-apps/kaccounts-providers-23.04.3 requires >=dev-qt/
qtwebengine-5.15.9:5
kde-apps/kdepim-runtime-23.04.3 requires >=dev-qt/
qtwebengine-5.15.9:5[widgets]
kde-apps/kmail-23.04.3 requires >=dev-qt/qtwebengine-5.15.9:5[widgets]
kde-apps/kontact-23.04.3 requires >=dev-qt/qtwebengine-5.15.9:5[widgets]
kde-apps/libksieve-23.04.3 requires >=dev-qt/qtwebengine-5.15.9:5[widgets]
kde-apps/messagelib-23.04.3 requires >=dev-qt/
qtwebengine-5.15.9:5[widgets]
net-libs/signon-ui-0.15_p20171022-r1 requires dev-qt/qtwebengine:5
www-client/falkon-23.04.3 requires >=dev-qt/
qtwebengine-5.15.9:5=[widgets], >=dev-qt/qtwebengine-5.15.9:5/5.15=[widgets]
--
Regards,
Peter.
Re: [gentoo-user] Re: Highlight certain packages being upgraded
On Monday, 10 July 2023 04:45:52 BST Dale wrote:
> Nikos Chantziaras wrote:
> > On 08/07/2023 11:33, Dale wrote:
> >> that excessively long qt package
> >
> > Off-topic, but just in case you mean qtwebengine, I was able to get
> > rid of it by putting "-webengine" in my USE flags. After a world
> > update, a depclean should then remove it from the system.
> >
> > You might have to juggle a few other USE flags in specific packages to
> > make it happen though, I forgot. It's been a while.
>
> That's a idea but since everything works well enough, I don't mind the
> compile times. After all, I nap while it does it anyway. Someone else
> may see that info and find it interesting tho. Maybe someone with a
> laptop or a really under-powered system. I never thought about trying
> to get rid of it. I didn't know it was possible even.
>
> Thanks.
>
> Dale
>
> :-) :-)
Thanks for the tip, Nikos. I wouldn't mind trying this, because qtwebengine
is a pain to emerge on resource constrained hardware:
>>> Running pre-merge checks for dev-qt/qtwebengine-5.15.10_p20230623
* Checking for at least 20 GiB RAM ...
* Amount of main memory is insufficient, but amount
* of main memory combined with swap is sufficient.
* Build process may make computer very slow! [ ok ]
* Checking for at least 7 GiB disk space at "/var/tmp/portage/dev-qt/
qtwebengine-5.15.10_p20230623/temp" ... [ ok ]
* Checking for at least 150 MiB disk space at "/usr" ... [ ok ]
However, I wonder what will break if I were to do this:
$ euse -i webengine
global use flags (searching: webengine)
no matching entries found
local use flags (searching: webengine)
[- ] webengine
app-misc/recoll: Use dev-qt/qtwebengine for fancy result list display
[- ] 1.34.6-r1 [gentoo]
[- ] 1.35.0 [gentoo]
[- ] webengine
app-office/libalkimia: Enable online quotes using dev-qt/qtwebengine
[- ] (0/8) 8.1.1-r1 [gentoo]
[- ] webengine
app-text/kbibtex: Use dev-qt/qtwebengine for HTML previews
[- ] (5) 0.10.0 [gentoo]
[- ] webengine
dev-python/QtPy: Pull in QtWebEngine and QtWebEngineWidgets modules
[- ] 2.3.1 [gentoo]
[- ] webengine
dev-python/pyside2: Build QtWebEngine and QtWebEngineWidgets modules
[- ] 5.15.9 [gentoo]
[- ] 5.15.10 [gentoo]
[- ] webengine
dev-python/pyside6: Build QtWebEngine and QtWebEngineWidgets modules
[- ] 6.5.0 [gentoo]
[- ] 6.5.1.1 [gentoo]
[- ] webengine
dev-qt/qt-creator: Use dev-qt/qtwebengine to view documentation
[- ] 8.0.2 [gentoo]
[- ] webengine
dev-qt/qt-docs: Install documentation for dev-qt/qtwebengine
[- ] (5) 5.15.2_p202011130614 [gentoo]
[- ] webengine
kde-apps/kaccounts-providers: Enable Nextcloud KAccounts plugin using dev-
qt/qtwebengine
[+ B] (5) 22.12.3 [gentoo]
[+ B] (5) 23.04.2 [gentoo]
[+ B] (5) 23.04.3 [gentoo]
[- ] webengine
kde-apps/kdecore-meta: Enable www-client/falkon which depends on dev-qt/
qtwebengine
[+ B] (5) 22.12.3 [gentoo]
[+ B] (5) 23.04.2 [gentoo]
[+ B] (5) 23.04.3 [gentoo]
[- ] webengine
kde-apps/kdeedu-meta: Enable packages requiring dev-qt/qtwebengine
[+ B] (5) 22.12.3 [gentoo]
[+ B] (5) 23.04.2 [gentoo]
[+ B] (5) 23.04.3 [gentoo]
[- ] webengine
kde-apps/kdenetwork-meta: Enable support for Google Drive integration via
kde-misc/kio-gdrive
[+ B] (5) 22.12.3 [gentoo]
[+ B] (5) 23.04.2 [gentoo]
[+ B] (5) 23.04.3 [gentoo]
[- ] webengine
kde-apps/kdesdk-meta: Enable dev-util/kdevelop, requiring dev-qt/
qtwebengine
[- ] (5) 22.12.3 [gentoo]
[- ] (5) 23.04.2 [gentoo]
[- ] (5) 23.04.3 [gentoo]
[- ] webengine
kde-apps/kdeutils-meta: Enable kde-apps/kimagemapeditor, requiring dev-qt/
qtwebengine
[+ B] (5) 22.12.3 [gentoo]
[+ B] (5) 23.04.2 [gentoo]
[+ B] (5) 23.04.3 [gentoo]
[- ] webengine
kde-apps/marble: Use dev-qt/qtwebengine for embedded web browser
[+ B] (5/22.12) 22.12.3 [gentoo]
[+ B] (5/23.04) 23.04.2 [gentoo]
[+ B] (5/23.04) 23.04.3 [gentoo]
[- ] webengine
kde-plasma/discover: Enable webflow support using dev-qt/qtwebview and
dev-qt/qtwebengine instead of default URL handler
[- ] (5) 5.27.5-r1 [gentoo]
[- ] (5) 5.27.6 [gentoo]
[- ] webengine
kde-plasma/kdeplasma-addons: Enable dictionary and web browser applets
using
dev-qt/qtwebengine
[- ] (5) 5.27.5 [gentoo]
[- ] (5) 5.
Re: [gentoo-user] Load average revisited
Peter, On Tuesday, 2023-05-02 15:04:54 +0100, you wrote: > ... > Meanwhile, is it possible to set things up so that, say, qtwebengine is never > compiled at the same time as anything else? I don't want to rely on my > noticing and intervening, and besides, it isn't always possible just to > --exclude it. What about # emerge -1u qtwebengine && emerge -u @world This will first update any dependencies of "qtwebengine" and only update "qtwebengine" itself when all its dependencies are dealt with, before it will deal with the rest. Sincerely, Rainer
Re: [gentoo-user] Error emerging qtwebengine-5.12.5
Raphael MD writes: > I'm trying to update KDE, but the emerge is stopping at qtwebengine-5.12.5 > compile process. > > Here is the compile log. > > https://pastebin.com/wKbJuAAS try to emerge a newer dev-libs/mpfr before qtwebengine. -- https://www.rohleder.de/ - irc:mroh@freenet - pgp:755E2DE5D0D585C52E7828307C7AFFBEFEF2CB25 please quote in https://en.wikipedia.org/wiki/Posting_style#Interleaved_style Q: How does a Unix guru have sex? A: unzip;strip;touch;finger;mount;fsck;more;yes;umount;sleep
Re: [gentoo-user] Load average revisited
On Tue, 2023-05-02 at 19:06 +0200, Dr Rainer Woitok wrote: > What about > > # emerge -1u qtwebengine && emerge -u @world > > This will first update any dependencies of "qtwebengine" and only update > "qtwebengine" itself when all its dependencies are dealt with, before it > will deal with the rest. What if qtwebengine's (many) dependencies also need to be updated first? I think trying to outsmart the dependency graphing in portage is inviting frustration.
Re: [gentoo-user] Not enough RAM for dev-qt/qtwebengine build
On Thu, Jun 13, 2019 at 05:15:41PM +0300, Alexey Eschenko wrote > Thank you. Didn't think about that. Don't know why though. My > MAKEOPTS was "-j32". Looks like that was too many for package like > qtwebengine. Solved the problem with creating specific environment > for qtwebengine and setting it up in /etc/portage/package.env/ It was > vry long build process but this time it finished successfully. I > think I'll try to find more appropriate value for qtwebengine which > could be used with 32GB of RAM. Please use plain text, not HTML. According to https://blogs.gentoo.org/ago/2013/01/14/makeopts-jcore-1-is-not-the-best-optimization/ the fastest compiles come with setting MAKEOPTS to the number of cores in your machine. E.g. for a dual core cpu, use "-j2", for a 4 core cpu use "-j4", etc. To check the number of cpus in your machine, execute... grep -c ^flags /proc/cpuinfo -- Walter Dnes I don't run "desktop environments"; I run useful applications
Re: [gentoo-user] gui-libs/egl-wayland and x11-drivers/nvidia-drivers dependency conflict.
While at it. Unrelated really but why not. Why does genlop -c show two running for this? (chroot) root@fireball / # genlop -c Currently merging 200 out of 257 * dev-qt/qtwebengine-5.15.12_p20240122 current merge time: 1 hour, 22 minutes and 33 seconds. ETA: 4 hours, 58 minutes and 39 seconds. Currently merging 200 out of 257 * dev-qt/qtwebengine-5.15.12_p20240122 current merge time: 1 hour, 22 minutes and 34 seconds. ETA: 4 hours, 58 minutes and 38 seconds. (chroot) root@fireball / # I notice there is a slot for version 6 and when it was compiling, it showed only one in the list like below. However, slot 5 always shows there is two running like above. I've never seen it do this with any other package, large or small. It only does it with slot 5 version of qtwebengine. This is slot 6 from earlier. (chroot) root@fireball / # genlop -c Currently merging 195 out of 257 * dev-qt/qtwebengine-6.6.2 current merge time: 2 hours, 59 minutes and 44 seconds. ETA: 3 hours, 21 minutes and 28 seconds. (chroot) root@fireball / # By the way, that time isn't exactly right. It takes a while but not that long. Anyone else see this? It's done this for a good while. Just kinda weird. o_O Dale :-) :-)
Re: [gentoo-user] Not enough RAM for dev-qt/qtwebengine build
Thank you. Didn't think about that. Don't know why though.My MAKEOPTS was "-j32". Looks like that was too many for package like qtwebengine.Solved the problem with creating specific environment for qtwebengine and setting it up in /etc/portage/package.env/It was vry long build process but this time it finished successfully. I think I'll try to find more appropriate value for qtwebengine which could be used with 32GB of RAM.Thanks again! 13.06.2019, 13:40, "tedheadster" :Alexey, you can check /etc/portage/make.conf and see if there is aMAKEOPTS="-j8" or similar variable. If not, add it and make the numberof jobs small, like "-j2". I have a similar problem when I buildsys-devel/binutils; it is a huge memory hog.- Matthew
Re: [gentoo-user] Not enough RAM for dev-qt/qtwebengine build
On Thu, 13 Jun 2019 13:19:24 +0300, Alexey Eschenko wrote: > For some time I have problems with dev-qt/qtwebengine (at least 5.12.3) > build. As far as I can see it fails to build due to memory exhaustion. > Although I have 32 GB of RAM (at least 20 of them is almost always > free) looks like it's not enough. It's strange because I have no > problems with Firefox/Chromium/Libreoffice builds. Do you have $PORTAGE_TMPDIR on a tmpfs filesystem? I had a similar problem, on a slightly less well-endowed system, and had to use portage.env to instruct that ebuild, as well as chromium and libreoffice, to use a directory on my SSD for PORTAGE_TMPDIR. % cat /etc/portage/package.env/qt dev-qt/qtwebengine disk-tmpdir.conf % cat /etc/portage/env/disk-tmpdir.conf PORTAGE_TMPDIR="/mnt/scratch" -- Neil Bothwick Assassins do it from behind. pgpYY1ppJLU_2.pgp Description: OpenPGP digital signature
Re: [gentoo-user] Not enough RAM for dev-qt/qtwebengine build
On Sat, Jun 15, 2019, at 14:19, Walter Dnes wrote:
> On Thu, Jun 13, 2019 at 05:15:41PM +0300, Alexey Eschenko wrote
> > Thank you. Didn't think about that. Don't know why though. My
> > MAKEOPTS was "-j32". Looks like that was too many for package like
> > qtwebengine. Solved the problem with creating specific environment
> > for qtwebengine and setting it up in /etc/portage/package.env/ It was
> > vry long build process but this time it finished successfully. I
> > think I'll try to find more appropriate value for qtwebengine which
> > could be used with 32GB of RAM.
>
> Please use plain text, not HTML.
>
> According to
> https://blogs.gentoo.org/ago/2013/01/14/makeopts-jcore-1-is-not-the-best-optimization/
> the fastest compiles come with setting MAKEOPTS to the number of cores
> in your machine. E.g. for a dual core cpu, use "-j2", for a 4 core cpu
> use "-j4", etc. To check the number of cpus in your machine, execute...
>
> grep -c ^flags /proc/cpuinfo
That's a good rule but not necessarily always true. My old machine was an
i7-3930K (6 cores, 12 threads) w/ 32G RAM. I had /var/tmp on tmpfs. I
benchmarked firefox, chromium, and some other big projects once and -j13 was
consistently the fastest on that box.
As that blog post says:
> I’m just saying, ${core} + 1 is not the best optimization for me
> and the test confirms the part:“but this guideline isn’t always perfect”
Depends on available RAM, how fast your disk is, etc.
Alec
[gentoo-user] Error emerging qtwebengine-5.12.5
Hello, I'm trying to update KDE, but the emerge is stopping at qtwebengine-5.12.5 compile process. Here is the compile log. https://pastebin.com/wKbJuAAS dmesg http://dpaste.com/27D5XHK emerge --info http://dpaste.com/2VKVS57 Many Thanks M.S. Raphael Mejias Dias Nuclear Engineer | Reactors Secure e-mail: raphael.mejias.d...@protonmail.com PGP Key for raph...@gmail.com: https://pgp.mit.edu/pks/lookup?op=get&search=0x87BC5A746072F951
Re: [gentoo-user] switch to profile 17.0 complete, completely painless
On 06/12/2017 17:38, Andrés Becerra Sandoval wrote: > > > 2017-12-06 2:18 GMT-05:00 Dale <mailto:rdalek1...@gmail.com>>: > > Raffaele Belardi wrote: > > One (~x86) LXDE system completed the switch with no problem, the > other (~amd64) built all > > except two packaged (sdlmame and torcs) which did not build with > gcc-7.2 even before the > > switch to 17.0. > > > > Gentoo devs and arch testers did a good job as usual. > > > > I'll do the switch on the Gnome system in the next days but up to > now I can say that the > > switch to 17.0 is a _lot_ less painful than switching major > compiler version. > > > > raffaele > > > > > > > I'm having trouble with these: > > net-libs/webkit-gtk-2.4.11-r200 > dev-qt/qtwebengine-5.9.3 > net-libs/webkit-gtk > > Those three, I've had to adjust the USE flags and it may or may not be > profile switch related. If I had to guess, it just happened to pop up > and isn't related to the switch. They are back in the rear compiling as > I type. > > > Dale, > > How did you merge qtwebengine? I had trouble with qtwebengine. It seemed to be an issue with icu-60.* but I wasn't prepared to downgrade to icu-59.* just as a test (icu changes here tend to trigger a rebuild of half of world), and a recent bug on b.g.o. backed up with I was thinking. It built fine with this in package.use: =dev-qt/qtwebengine-5.9.3 -system-ffmpeg -system-icu Yes, I did do it, favoured bundled libs instead of system ones. But I was also having similar issues with bundled vs system ffmpeg for kodi, and this was the easiest way to get past it and finish a 17.0 migration -- Alan McKinnon alan.mckin...@gmail.com
Re: [gentoo-user] Re: Netflix Error Code O7355 with Opera
On 6/11/20 7:45 AM, Michael wrote: > I figured since qtwebengine uses the same rendering engine and I spend enough > time compiling that package anyway, because KDE won't do without it, I might > as well ditch Chromium. I haven't looked back. ;-) Since you already have qtwebengine built, you could always try falkon as well.
[gentoo-user] 100% CPU load in qtwebengine
Hello list, On this box I have this: # grep '\-j' /etc/portage/make.conf EMERGE_DEFAULT_OPTS="--jobs --load-average=4 [...] " MAKEOPTS="-j4 -l4" That seems to work well, except for a 20s period at the beginning of emerging qtwebengine, during which CPU load goes to 100%, according to gkrellm. It seems that the ebuild runs a process other than make, ignoring make.conf. Does anyone here know what that might be, and why it disregards my preferences? -- Regards, Peter. signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] Not enough RAM for dev-qt/qtwebengine build
That was my first move. Unfortunately it didn't help.But in the first answer in this thread I saw a good idea about decreasing number of parallel build threads.I have MAKEOPTS="-j32" and it's probably the cause of my problem with this package.Right now I'm trying separate build environment with decreased job count value and it looks promissing.I'll leave the message with results in the reply to that message. 13.06.2019, 14:13, "Neil Bothwick" :On Thu, 13 Jun 2019 13:19:24 +0300, Alexey Eschenko wrote: For some time I have problems with dev-qt/qtwebengine (at least 5.12.3) build. As far as I can see it fails to build due to memory exhaustion. Although I have 32 GB of RAM (at least 20 of them is almost always free) looks like it's not enough. It's strange because I have no problems with Firefox/Chromium/Libreoffice builds.Do you have $PORTAGE_TMPDIR on a tmpfs filesystem? I had a similarproblem, on a slightly less well-endowed system, and had to useportage.env to instruct that ebuild, as well as chromium and libreoffice,to use a directory on my SSD for PORTAGE_TMPDIR.% cat /etc/portage/package.env/qtdev-qt/qtwebengine disk-tmpdir.conf% cat /etc/portage/env/disk-tmpdir.confPORTAGE_TMPDIR="/mnt/scratch" --Neil BothwickAssassins do it from behind.
Re: [gentoo-user] Error emerging qtwebengine-5.12.5
On Thursday, 21 November 2019 17:35:30 GMT Raphael MD wrote: > Hello, > > I'm trying to update KDE, but the emerge is stopping at qtwebengine-5.12.5 > compile process. I don't know if your error is related to the gcc version you're using. Better switch to 9.2 and try again. PS. MAKEOPTS="-j1" appears to be quite restrictive for a Ryzen CPU. It wouldn't cause your problem, but will slow down every emerge as it will compile one job at a time. -- Regards, Mick signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] Re: Highlight certain packages being upgraded
On Mon, 2023-07-10 at 04:25 +0300, Nikos Chantziaras wrote: > > that excessively long qt package > Off-topic, but just in case you mean qtwebengine, I was able to get > rid of it by putting "-webengine" in my USE flags. I got rid of it by switching to a flatpak version of the singular desktop application that required it on my system. Normally I'll always choose a native package but qtwebengine builds for a program that I run a couple of times a week skews the reward/effort ratio.
Re: [gentoo-user] Why falkon?
On Monday, 19 November 2018 23:40:04 GMT Neil Bothwick wrote: > On Mon, 19 Nov 2018 09:27:30 +, Peter Humphrey wrote: > > Would someone kindly enloghten me as to why the presence of qtwebengine > > now requires installation of www-client/falkon? --->8 > It doesn't, it's the kdecore-meta package that depends on it, controlled > by the webengine USE flag. qtwebengine doesn't require falkon at all, > it's the other way round. Hmm. I see what you mean. Maybe this is the straw that makes me strip out all the stuff I don't need, by removing meta-packages and just installing what I want. After all, there's a limit to the number of web browsers a body might need. -- Regards, Peter.
[gentoo-user] Compile large packages as last package
Hello list, I would like to tag or be able to prioritise (not via "nice" or "renice") large packages. Currently, one system is compiling package 245 of 279 and nothing else. "qlop --running --verbose --time" shows me, that it is "qtwebengine", which will compile for a few hours, blocking all other packages. I set "MAKEOPTS" AND "EMERGE_DEFAULT_OPTS" like so: MAKEOPTS="--jobs 8 --load-average 7.2" EMERGE_DEFAULT_OPTS="--jobs 8 --load-average 7.2" Is there any way to tell "portage", that packages like "qtwebengine", "(ungoogled-)chromium", "firefox" and so on are always compiled as last package? Regards -Ramon -- GPG public key: 5983 98DA 5F4D A464 38FD CF87 155B E264 13E6 99BF OpenPGP_signature Description: OpenPGP digital signature
Re: [gentoo-user] Reproducible Installation Lists?
Michael writes:
> On Saturday, 26 February 2022 22:47:52 GMT Ramces Tampo-og Red wrote:
>
>> Yeah, I was just thinking about that since building a powerful, new
>> computer around my area is prohibitively expensive. But getting old,
>> prebuilt computers is ludicrously cheap. I figured that I can get a few
>> of them for $50-75 and just plug them to an ethernet switch to do the
>> compiling for me.
>
> Distributed compiling may not be as useful as you think. Not all phases of a
> build can be distributed, pre-processing and linking will still take a lot of
> time, some packages will not compile over distcc and will fail, any gains in
> compiling time could be eaten away by network losses, etc. On the other hand
> a 'better' PC with more RAM and a faster CPU with more cores could prove
> transformative in its performance impact, when used as a building server for
> binary packages to be installed thereafter on slower systems.
>
> I have found older PCs with limited resources eventually reach an EOL as far
> as their capability to emerge large packages. I have a very old Core 2 Duo
> Pentium laptop with 4G RAM, which even with MAKEOPTS="-j1" takes forever to
> build qtwebengine:
>
> genlop -t dev-qt/qtwebengine
>
> Fri Feb 4 20:06:46 2022 >>> dev-qt/qtwebengine-5.15.2_p20211216
>merge time: 1 day, 3 hours, 12 minutes and 7 seconds.
>
> Chromium got so slow over the years, I stopped emerging it long ago.
>
> Give distcc a go if you have a spare PC to experiment and want to test %
> improvements with your setup, but personally I wouldn't invest money on it.
> Instead I'd save up for a faster machine. YMMV.
That sucks. I'll try it out but I guess there's no way out on building a
proper rig that can handle the compile jobs.
Though to be honest, I've been compiling all of my stuff in both my X200
and T400 laptops and both are Core 2 Duo machines. The only real issue
that I've had with those is with compiling Qtwebengine which took around
12-14 hours. Other than that the compile times are negligible enough
that I don't really notice it that much.
Either way, thanks for the heads up.
Cheers!
--
|Give them an inch and they will take a mile.|
||
|-{gopher,gemini}://kalayaan.xyz-|
|-C4AE 5D53 46A0 01DF 6E92 CB46 92D7 9FBB AB9F 3E37-|
signature.asc
Description: PGP signature
Re: [gentoo-user] Not enough RAM for dev-qt/qtwebengine build
On Thu, 13 Jun 2019 14:48:04 +0300, Alexey Eschenko wrote: > That was my first move. Unfortunately it didn't help. > But in the first answer in this thread I saw a good idea about > decreasing number of parallel build threads. I have MAKEOPTS="-j32" and > it's probably the cause of my problem with this package. Right now I'm > trying separate build environment with decreased job count value and it > looks promissing. I'll leave the message with results in the reply to > that message. 13.06.2019, 14:13, "Neil Bothwick" : You can set that in package.env too, I do this for Chromium: % cat /etc/portage/package.env/chromium www-client/chromium alert-done.conf disk-tmpdir.conf j2.conf % cat /etc/portage/env/j2.conf MAKEOPTS="-j2" > > On Thu, 13 Jun 2019 13:19:24 +0300, Alexey Eschenko wrote: > > For some time I have problems with dev-qt/qtwebengine (at least > > 5.12.3) > >> build. As far as I can see it fails to build due to memory > >> exhaustion. Although I have 32 GB of RAM (at least 20 of them is > >> almost always free) looks like it's not enough. It's strange because > >> I have no problems with Firefox/Chromium/Libreoffice builds. > > > > > > Do you have $PORTAGE_TMPDIR on a tmpfs filesystem? I had a similar > > problem, on a slightly less well-endowed system, and had to use > > portage.env to instruct that ebuild, as well as chromium and > > libreoffice, to use a directory on my SSD for PORTAGE_TMPDIR. > > > > % cat /etc/portage/package.env/qt > > dev-qt/qtwebengine disk-tmpdir.conf > > > > % cat /etc/portage/env/disk-tmpdir.conf > > PORTAGE_TMPDIR="/mnt/scratch" > > > > -- > > Neil Bothwick > > > > > > Assassins do it from behind. -- Neil Bothwick Always proofread carefully to see if you any words out. pgpaUi1VKaQay.pgp Description: OpenPGP digital signature
Re: [gentoo-user] Compile large packages as last package
Am Sat, Aug 14, 2021 at 09:20:23PM +0200 schrieb Ramon Fischer: > Hello list, > > I would like to tag or be able to prioritise (not via "nice" or "renice") > large packages. > > Currently, one system is compiling package 245 of 279 and nothing else. > "qlop --running --verbose --time" shows me, that it is "qtwebengine", which > will compile for a few hours, blocking all other packages. > […] > Is there any way to tell "portage", that packages like "qtwebengine", > "(ungoogled-)chromium", "firefox" and so on are always compiled as last > package? In the case of qtwebengine, this might be impossible, because it is a dependency for other packages. If they require the newer version (although not very likely in this case), then it has to be built first. Perhaps the logic changes if you put it into @world. Then it can be compiled with other big packages (big applications) in one go, and only after that you build the rest by doing the deep update. One possibility is a wrapper script in which you store those packages you want to handle that way. It will check whether there is an update for them and—if there is—mask all versions that are newer than the currently installed one in a dedicated package.mask.d file. Then it runs a world update, after which it removes the mask file and finally runs another update pass. -- Grüße | Greetings | Qapla’ Please do not share anything from, with or about me on any social network. These valuable information, put in a Braille line, would give a flat, polished surface. Thank you. (SelfHTML forum) signature.asc Description: PGP signature
Re: [gentoo-user] Reproducible Installation Lists?
On Saturday, 26 February 2022 22:47:52 GMT Ramces Tampo-og Red wrote: > Yeah, I was just thinking about that since building a powerful, new > computer around my area is prohibitively expensive. But getting old, > prebuilt computers is ludicrously cheap. I figured that I can get a few > of them for $50-75 and just plug them to an ethernet switch to do the > compiling for me. Distributed compiling may not be as useful as you think. Not all phases of a build can be distributed, pre-processing and linking will still take a lot of time, some packages will not compile over distcc and will fail, any gains in compiling time could be eaten away by network losses, etc. On the other hand a 'better' PC with more RAM and a faster CPU with more cores could prove transformative in its performance impact, when used as a building server for binary packages to be installed thereafter on slower systems. I have found older PCs with limited resources eventually reach an EOL as far as their capability to emerge large packages. I have a very old Core 2 Duo Pentium laptop with 4G RAM, which even with MAKEOPTS="-j1" takes forever to build qtwebengine: genlop -t dev-qt/qtwebengine Fri Feb 4 20:06:46 2022 >>> dev-qt/qtwebengine-5.15.2_p20211216 merge time: 1 day, 3 hours, 12 minutes and 7 seconds. Chromium got so slow over the years, I stopped emerging it long ago. Give distcc a go if you have a spare PC to experiment and want to test % improvements with your setup, but personally I wouldn't invest money on it. Instead I'd save up for a faster machine. YMMV. signature.asc Description: This is a digitally signed message part.
[gentoo-user] Re: Compile large packages as last package
On 14/08/2021 22:20, Ramon Fischer wrote: Is there any way to tell "portage", that packages like "qtwebengine", "(ungoogled-)chromium", "firefox" and so on are always compiled as last package? The simplest way is to exclude those packages in the first update, and then allow them in the second: emerge -uDU @world --exclude "qtwebengine firefox chromium" && emerge -uDU @world The dependency tracker of portage will of course also exclude packages that depend on the excluded packages, unless they themselves have updates pending. In that case, they *might* get built twice; once against the current version of the excluded packages, and then perhaps again on the second run, if there's rebuild triggers involved. Most of the time though, you won't run into cases of redundant rebuilds. Rebuild triggers are not very common.
Re: [gentoo-user] 100% CPU load in qtwebengine
On Thursday, 23 May 2024 14:07:16 BST Peter Humphrey wrote: > Hello list, > > On this box I have this: > > # grep '\-j' /etc/portage/make.conf > EMERGE_DEFAULT_OPTS="--jobs --load-average=4 [...] " > MAKEOPTS="-j4 -l4" > > That seems to work well, except for a 20s period at the beginning of > emerging qtwebengine, during which CPU load goes to 100%, according to > gkrellm. > > It seems that the ebuild runs a process other than make, ignoring make.conf. > Does anyone here know what that might be, and why it disregards my > preferences? Does this happen while the source archive is being decompressed? You can check in top (press 'c' then 'Shift+v') to see what command/child process is eating up CPU time. Use Page Up/Down to navigate through the long list. signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] switch to profile 17.0 complete, completely painless
On Wed, 6 Dec 2017 18:20:54 +0200, Alan McKinnon wrote: > I had trouble with qtwebengine. It seemed to be an issue with icu-60.* > but I wasn't prepared to downgrade to icu-59.* just as a test (icu > changes here tend to trigger a rebuild of half of world), and a recent > bug on b.g.o. backed up with I was thinking. > > It built fine with this in package.use: > > =dev-qt/qtwebengine-5.9.3 -system-ffmpeg -system-icu > > Yes, I did do it, favoured bundled libs instead of system ones. But I > was also having similar issues with bundled vs system ffmpeg for kodi, > and this was the easiest way to get past it and finish a 17.0 migration Or you could have searched bgo... https://bugs.gentoo.org/639220 -- Neil Bothwick Synonym: a word you use when you can't spell the other one. pgp4rxj2sLiHX.pgp Description: OpenPGP digital signature
Re: [gentoo-user] Why falkon?
On Mon, 19 Nov 2018 09:27:30 +, Peter Humphrey wrote: > Would someone kindly enloghten me as to why the presence of qtwebengine > now requires installation of www-client/falkon? > > $ cat $(equery w kdecore-meta) > # Copyright 1999-2018 Gentoo Authors > # Distributed under the terms of the GNU General Public License v2 > > EAPI=6 > > inherit kde5-meta-pkg > > DESCRIPTION="kdecore - merge this to pull in the most basic > applications" KEYWORDS="~amd64 ~x86" > IUSE="+handbook +webengine webkit" > > RDEPEND=" > $(add_kdeapps_dep dolphin) > $(add_kdeapps_dep kdialog) > $(add_kdeapps_dep keditbookmarks) > $(add_kdeapps_dep kfind) > $(add_kdeapps_dep konsole) > $(add_kdeapps_dep kwrite) > handbook? ( $(add_kdeapps_dep khelpcenter) ) > webengine? ( www-client/falkon ) It doesn't, it's the kdecore-meta package that depends on it, controlled by the webengine USE flag. qtwebengine doesn't require falkon at all, it's the other way round. -- Neil Bothwick "I need your clothes, your boots, and your tagline!" pgpmvhixCpSbP.pgp Description: OpenPGP digital signature
Re: [gentoo-user] Not enough RAM for dev-qt/qtwebengine build
Alec Ten Harmsel wrote:
> On Sat, Jun 15, 2019, at 14:19, Walter Dnes wrote:
>> On Thu, Jun 13, 2019 at 05:15:41PM +0300, Alexey Eschenko wrote
>>> Thank you. Didn't think about that. Don't know why though. My
>>> MAKEOPTS was "-j32". Looks like that was too many for package like
>>> qtwebengine. Solved the problem with creating specific environment
>>> for qtwebengine and setting it up in /etc/portage/package.env/ It was
>>> vry long build process but this time it finished successfully. I
>>> think I'll try to find more appropriate value for qtwebengine which
>>> could be used with 32GB of RAM.
>> Please use plain text, not HTML.
>>
>> According to
>> https://blogs.gentoo.org/ago/2013/01/14/makeopts-jcore-1-is-not-the-best-optimization/
>> the fastest compiles come with setting MAKEOPTS to the number of cores
>> in your machine. E.g. for a dual core cpu, use "-j2", for a 4 core cpu
>> use "-j4", etc. To check the number of cpus in your machine, execute...
>>
>> grep -c ^flags /proc/cpuinfo
> That's a good rule but not necessarily always true. My old machine was an
> i7-3930K (6 cores, 12 threads) w/ 32G RAM. I had /var/tmp on tmpfs. I
> benchmarked firefox, chromium, and some other big projects once and -j13 was
> consistently the fastest on that box.
>
> As that blog post says:
>
>> I’m just saying, ${core} + 1 is not the best optimization for me
>> and the test confirms the part:“but this guideline isn’t always perfect”
> Depends on available RAM, how fast your disk is, etc.
>
> Alec
>
>
As a AMD CPU user, I always set mine to number of cores plus one. That
seems to always be the most efficient for me. I have portages work
directory on tmpfs for all but a couple large packages. I used to make
exceptions for Firefox, Seamonkey, Libreoffice and qtweb something or
other. Since I upgraded my memory to 32GBs I removed all but
Libreoffice. My biggest problem was when more than one of those wanted
to compile at the same time.
The best way to know what to set it to, test it. Set it to cores plus
one and test. Then set it to half the number of cores, twice the number
of cores etc until you find that sweet spot. One could even do that
during normal updates although it may not be as accurate. I suspect if
ten people with ten different systems tested this, we'd get half a dozen
different results, maybe more.
The best thing in my opinion, start emerge, go to bed and then hope it
is done when you wake up. lol
Dale
:-) :-)
Re: [gentoo-user] Not enough RAM for dev-qt/qtwebengine build
On June 13, 2019 11:48:04 AM UTC, Alexey Eschenko wrote: >That was my first move. Unfortunately it didn't help. > >But in the first answer in this thread I saw a good idea about >decreasing number of parallel build threads. > >I have MAKEOPTS="-j32" and it's probably the cause of my problem with >this package. > >Right now I'm trying separate build environment with decreased job >count value and it looks promissing. > >I'll leave the message with results in the reply to that message. > When using --jobs (I had issues with -j once because a different build tool used it for something else) it is a good idea to also add --load-average. This will limit the amount of jobs when the load average goes up. (And that goes up fast when swap is touched) -- Joost > >13.06.2019, 14:13, "Neil Bothwick" : > >On Thu, 13 Jun 2019 13:19:24 +0300, Alexey Eschenko wrote: > > >For some time I have problems with dev-qt/qtwebengine (at least 5.12.3) > build. As far as I can see it fails to build due to memory exhaustion. > Although I have 32 GB of RAM (at least 20 of them is almost always > free) looks like it's not enough. It's strange because I have no > problems with Firefox/Chromium/Libreoffice builds. > > >Do you have $PORTAGE_TMPDIR on a tmpfs filesystem? I had a similar >problem, on a slightly less well-endowed system, and had to use >portage.env to instruct that ebuild, as well as chromium and >libreoffice, >to use a directory on my SSD for PORTAGE_TMPDIR. > >% cat /etc/portage/package.env/qt >dev-qt/qtwebengine disk-tmpdir.conf > >% cat /etc/portage/env/disk-tmpdir.conf >PORTAGE_TMPDIR="/mnt/scratch" > > > >-- >Neil Bothwick > > >Assassins do it from behind. -- Sent from my Android device with K-9 Mail. Please excuse my brevity.
Re: [gentoo-user] Cannot build qtwebengine: ‘WEBP_EXTERN’ does not name a type
On Mon, Apr 29, 2019 at 06:54:08PM +0200, Frank Steinmetzger wrote: > I’ll try a libwebp downgrade now. Since that didn’t work either, I tried to recreate the error by building a minimal c program, which only includes libwebp headers and opened/closed a webp object. I was so rusty with handling gcc that I didn’t get it to build, but when running `locate libwebp.a`, I found out that I had an old version of the lib installed in /usr/local. I can’t remember what I needed it for and so I uninstalled it. And whaddaya know, qtwebengine now compiles happily. I’m already at file 8981/15727, while the error popped up at 6592/15727, which is way earlier. Yay. -- Gruß | Greetings | Qapla’ Please do not share anything from, with or about me on any social network. NOTHING is the complete absence of SOMETHING. Without presence of absence there is no NOTHING. signature.asc Description: PGP signature
Re: [gentoo-user] 100% CPU load in qtwebengine
On Thursday, 23 May 2024 20:13:27 BST Michael wrote: > On Thursday, 23 May 2024 14:07:16 BST Peter Humphrey wrote: > > Hello list, > > > > On this box I have this: > > > > # grep '\-j' /etc/portage/make.conf > > EMERGE_DEFAULT_OPTS="--jobs --load-average=4 [...] " > > MAKEOPTS="-j4 -l4" > > > > That seems to work well, except for a 20s period at the beginning of > > emerging qtwebengine, during which CPU load goes to 100%, according to > > gkrellm. > > > > It seems that the ebuild runs a process other than make, ignoring > > make.conf. Does anyone here know what that might be, and why it > > disregards my preferences? > > Does this happen while the source archive is being decompressed? It could be; the .tar.xz file is 288MB - but I didn't think bzip2 was multithreaded*. I tried to check by rerunning the emerge, but it found a Gentoo binary and went to fetch that. * And app-alternatives/bzip2 has installed bzip2. -- Regards, Peter. signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] switch to profile 17.0 complete, completely painless
2017-12-06 2:18 GMT-05:00 Dale : > Raffaele Belardi wrote: > > One (~x86) LXDE system completed the switch with no problem, the other > (~amd64) built all > > except two packaged (sdlmame and torcs) which did not build with gcc-7.2 > even before the > > switch to 17.0. > > > > Gentoo devs and arch testers did a good job as usual. > > > > I'll do the switch on the Gnome system in the next days but up to now I > can say that the > > switch to 17.0 is a _lot_ less painful than switching major compiler > version. > > > > raffaele > > > > > > > I'm having trouble with these: > > net-libs/webkit-gtk-2.4.11-r200 > dev-qt/qtwebengine-5.9.3 > net-libs/webkit-gtk > > Those three, I've had to adjust the USE flags and it may or may not be > profile switch related. If I had to guess, it just happened to pop up > and isn't related to the switch. They are back in the rear compiling as > I type. > > Dale, How did you merge qtwebengine? -- Andrés Becerra Sandoval
[gentoo-user] Cannot build qtwebengine: ‘WEBP_EXTERN’ does not name a type
Hi folks, it’s been a while since I had a problem for which I needed a nudge. :) Due to some other lib upgrade, I need to rebuild qtwebengine. But it always fails with: | In file included from ../../3rdparty/chromium/third_party/skia/src/images/SkWebpEncoder.cpp:42: | /usreincnude/webp/mux.h:101:1: Since libpcap has been upgradederror: ‘WEBP_EXTERN’ does not name a type | WEBP_EXTERN int WebPGetMuxVersion(void); | ^~~ | | /usr/include/webp/mux.h: In function ‘WebPMux* WebPMuxCreate(const WebPData*, int)’: | /usr/include/webp/mux.h:138:10: error: ‘WebPMuxCreateInternal’ was not declared in this scope |return WebPMuxCreateInternal(bitstream, copy_data, WEBP_MUX_ABI_VERSION); | ^ And a number more of similar problems, hinting at a missing or broken include file in libpcap. I tried upgrading libpcap from 1.8 to 1.9 as a remedy, but no luck there. There is also no relevant entry on BGO and I asked twice on IRC with this, but noone answered. So you are my last hope, Gentoo-list Kenobi. What could I be missing? Cheers. -- Gruß | Greetings | Qapla’ Please do not share anything from, with or about me on any social network. This is a lousy party! I’ll leave as soon as I find my trousers. signature.asc Description: PGP signature
Re: [gentoo-user] Error emerging qtwebengine-5.12.5
Hello, Am Donnerstag, 21. November 2019, 19:40:43 CET schrieb Mick: > On Thursday, 21 November 2019 17:35:30 GMT Raphael MD wrote: > > Hello, > > > > I'm trying to update KDE, but the emerge is stopping at qtwebengine-5.12.5 > > compile process. > > I don't know if your error is related to the gcc version you're using. > Better switch to 9.2 and try again. > > PS. MAKEOPTS="-j1" appears to be quite restrictive for a Ryzen CPU. It > wouldn't cause your problem, but will slow down every emerge as it will > compile one job at a time. I've got the same error. Reason was (as far as i can interpret it) that a new gcc (v9) was compiled. The mpfr-Part (seen by the assertion abort on an mpfr- test in the log) was updated at the same time. But gcc v8 - still active - was not recompiled afterwards. Solution - in my case - recompile gcc v8: emerge -1 =gcc-8* Other solution (untested) - switch to gcc v9. The MAKEOPTS depends on how much main memory you have on your host - besides on how much cores your CPU has. kind regards Petric
Re: [gentoo-user] long compiles
Chromium and qtwebengine have the longest build times that I have encountered On Tue, 12 Sep 2023 at 1:01 AM, Dale wrote: > Alan McKinnon wrote: > > After my long time away from Gentoo, I thought perhaps some packages that > always took ages to compile would have improved. I needed to change to > ~amd64 anyway (dumb n00b mistake leaving it at amd64). So that's what I did > and let emerge do it's thing. > > chromium has been building since 10:14, it's now 21:16 and still going so > 9 hours at least on this machine to build a browser - almost as bad as > openoffice at it's worst (regularly took 12 hours). Nodejs also took a > while, but I didn't record time. > > > What other packages have huge build times? > > -- > Alan McKinnon > alan dot mckinnon at gmail dot com > > > > I have some software you don't likely use that takes a while but one that > is common is qtwebengine or something. If it's not that one, it's qtweb > something. It takes about 4 hours, sometimes 5 or so. > > I think the software takes longer to compile so that we will build new > rigs. ROFL > > Dale > > :-) :-) >
Re: [gentoo-user] switch to profile 17.0 complete, completely painless
Andrés Becerra Sandoval wrote: > > > 2017-12-06 2:18 GMT-05:00 Dale <mailto:rdalek1...@gmail.com>>: > > Raffaele Belardi wrote: > > One (~x86) LXDE system completed the switch with no problem, the > other (~amd64) built all > > except two packaged (sdlmame and torcs) which did not build with > gcc-7.2 even before the > > switch to 17.0. > > > > Gentoo devs and arch testers did a good job as usual. > > > > I'll do the switch on the Gnome system in the next days but up > to now I can say that the > > switch to 17.0 is a _lot_ less painful than switching major > compiler version. > > > > raffaele > > > > > > > I'm having trouble with these: > > net-libs/webkit-gtk-2.4.11-r200 > dev-qt/qtwebengine-5.9.3 > net-libs/webkit-gtk > > Those three, I've had to adjust the USE flags and it may or may not be > profile switch related. If I had to guess, it just happened to pop up > and isn't related to the switch. They are back in the rear > compiling as > I type. > > > Dale, > > How did you merge qtwebengine? > > > -- > Andrés Becerra Sandoval > You may have already fixed this but the way I got it to work was this: -system-icu I added that to my make.conf USE line and it seems to compile fine, but with that disabled for now. Dale :-) :-)
Re: [gentoo-user] Why falkon?
On Tuesday, 20 November 2018 15:26:14 GMT Peter Humphrey wrote: > On Monday, 19 November 2018 23:40:04 GMT Neil Bothwick wrote: > > On Mon, 19 Nov 2018 09:27:30 +, Peter Humphrey wrote: > > > Would someone kindly enloghten me as to why the presence of qtwebengine > > > now requires installation of www-client/falkon? > > --->8 > > > It doesn't, it's the kdecore-meta package that depends on it, controlled > > by the webengine USE flag. qtwebengine doesn't require falkon at all, > > it's the other way round. > > Hmm. I see what you mean. Maybe this is the straw that makes me strip out > all the stuff I don't need, by removing meta-packages and just installing > what I want. After all, there's a limit to the number of web browsers a > body might need. I did that, and got rid of a large number of packages I don't need. A word of advice though, if I may, to anyone else who tries slimming plasma down: you do need kdedesktop, not just kdeworkspace. When I tried just the former of those two, I got an "all shell packages missing" error (or something like that) at plasma startup via sddm, together with a black screen. Windows opened and closed all right, but the system clearly wasn't happy. -- Regards, Peter.
Re: [gentoo-user] Ungoogled-chromium, anyone?
On Monday, 13 April 2020 11:46:45 BST Peter Humphrey wrote: > Hello list, > > I discovered this package today and wondered whether anyone here had any > experience of it. Interesting to see this project exists. I thought Chromium was essentially un-Googled, but obviously there's more there to take place to strip Google's tentacles from the browser. > Which overlay to get it from? According to: https://github.com/Eloston/ungoogled-chromium/blob/master/README.md The overlay they mention is pf4public: https://github.com/PF4Public/gentoo-overlay > How stable is it? Does it > really "privatise" chromium? Does it allow extensions like ublock-origin? > > I'm happy with firefox, but it never hurts to have a choice. I'd be also interested to know if Falkon/QtWebEngine is more or less /un- Googled/ than the ungoogled-chromium browser. I've gradually grown reluctant to spend many-many hours emerging Chromium and have removed it from a number of my older boxen. Especially since the 'jumbo- build' USE flag was dropped, Chromium's compilation takes twice as long as the many hours it used to take. I'd rather not waste that much electricity when QtWebEngine is built as a default with Plasma/KDE anyway. signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] long compiles
qtwebengine! yes that one took forever also. It also said my 16G of RAM was smaller than the 16G it needed. Weird. Anyways I enabled a swapfile and left it to run overnight Alan On Mon, Sep 11, 2023 at 9:31 PM Dale wrote: > Alan McKinnon wrote: > > After my long time away from Gentoo, I thought perhaps some packages that > always took ages to compile would have improved. I needed to change to > ~amd64 anyway (dumb n00b mistake leaving it at amd64). So that's what I did > and let emerge do it's thing. > > chromium has been building since 10:14, it's now 21:16 and still going so > 9 hours at least on this machine to build a browser - almost as bad as > openoffice at it's worst (regularly took 12 hours). Nodejs also took a > while, but I didn't record time. > > > What other packages have huge build times? > > -- > Alan McKinnon > alan dot mckinnon at gmail dot com > > > > I have some software you don't likely use that takes a while but one that > is common is qtwebengine or something. If it's not that one, it's qtweb > something. It takes about 4 hours, sometimes 5 or so. > > I think the software takes longer to compile so that we will build new > rigs. ROFL > > Dale > > :-) :-) > -- Alan McKinnon alan dot mckinnon at gmail dot com
Re: [gentoo-user] 100% CPU load in qtwebengine
On Friday, 24 May 2024 11:52:55 BST Peter Humphrey wrote: > On Thursday, 23 May 2024 20:13:27 BST Michael wrote: > > On Thursday, 23 May 2024 14:07:16 BST Peter Humphrey wrote: > > > Hello list, > > > > > > On this box I have this: > > > > > > # grep '\-j' /etc/portage/make.conf > > > EMERGE_DEFAULT_OPTS="--jobs --load-average=4 [...] " > > > MAKEOPTS="-j4 -l4" > > > > > > That seems to work well, except for a 20s period at the beginning of > > > emerging qtwebengine, during which CPU load goes to 100%, according to > > > gkrellm. > > > > > > It seems that the ebuild runs a process other than make, ignoring > > > make.conf. Does anyone here know what that might be, and why it > > > disregards my preferences? > > > > Does this happen while the source archive is being decompressed? > > It could be; the .tar.xz file is 288MB - but I didn't think bzip2 was > multithreaded*. I tried to check by rerunning the emerge, but it found a > Gentoo binary and went to fetch that. > > * And app-alternatives/bzip2 has installed bzip2. The archive is compressed with xz, which in later versions can run in a multithreaded fashion. I don't know if emerge calls upon it to operate with multiple threads (e.g. xz --threads 0 foo.xz). PS. The bzip2 is single threaded and a slow compressor to boot, but pbzip2 is multithreaded. signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] Error emerging qtwebengine-5.12.5
Mick 3:41 PM (...) >I don't know if your error is related to the gcc version you're using. Better >switch to 9.2 and try again. I've updated GCC to 9.2 and this has solved my problem. Thanks!
Re: [gentoo-user] switch to profile 17.0 complete, completely painless
Andrés Becerra Sandoval wrote: > > > 2017-12-06 2:18 GMT-05:00 Dale <mailto:rdalek1...@gmail.com>>: > > Raffaele Belardi wrote: > > One (~x86) LXDE system completed the switch with no problem, the > other (~amd64) built all > > except two packaged (sdlmame and torcs) which did not build with > gcc-7.2 even before the > > switch to 17.0. > > > > Gentoo devs and arch testers did a good job as usual. > > > > I'll do the switch on the Gnome system in the next days but up > to now I can say that the > > switch to 17.0 is a _lot_ less painful than switching major > compiler version. > > > > raffaele > > > > > > > I'm having trouble with these: > > net-libs/webkit-gtk-2.4.11-r200 > dev-qt/qtwebengine-5.9.3 > net-libs/webkit-gtk > > Those three, I've had to adjust the USE flags and it may or may not be > profile switch related. If I had to guess, it just happened to pop up > and isn't related to the switch. They are back in the rear > compiling as > I type. > > > Dale, > > How did you merge qtwebengine? > > > -- > Andrés Becerra Sandoval > LOL. So far, I haven't been able to get it too. I'm starting to disable USE flags to see if I can either get it to compile OR get rid of the package. I don't think the later is possible since it has a wide range of things depending on it. If I get a fix, I'll post it and let you try it. I got my hammer out. Dale :-) :-)
Re: [gentoo-user] Why falkon?
On Thursday, 22 November 2018 17:55:22 GMT Peter Humphrey wrote: > On Tuesday, 20 November 2018 15:26:14 GMT Peter Humphrey wrote: > > On Monday, 19 November 2018 23:40:04 GMT Neil Bothwick wrote: > > > On Mon, 19 Nov 2018 09:27:30 +, Peter Humphrey wrote: > > > > Would someone kindly enloghten me as to why the presence of > > > > qtwebengine > > > > now requires installation of www-client/falkon? > > > > --->8 > > > > > It doesn't, it's the kdecore-meta package that depends on it, controlled > > > by the webengine USE flag. qtwebengine doesn't require falkon at all, > > > it's the other way round. > > > > Hmm. I see what you mean. Maybe this is the straw that makes me strip out > > all the stuff I don't need, by removing meta-packages and just installing > > what I want. After all, there's a limit to the number of web browsers a > > body might need. > > I did that, and got rid of a large number of packages I don't need. > > A word of advice though, if I may, to anyone else who tries slimming plasma > down: you do need kdedesktop, not just kdeworkspace. When I tried just the > former of those two, ... That should have been the latter, of course. > ... I got an "all shell packages missing" error (or > something like that) at plasma startup via sddm, together with a black > screen. Windows opened and closed all right, but the system clearly wasn't > happy. -- Regards, Peter.
Re: [gentoo-user] Ungoogled-chromium, anyone?
On Monday, 13 April 2020 11:59:24 BST Michael wrote: > On Monday, 13 April 2020 11:46:45 BST Peter Humphrey wrote: > > Hello list, > > > > I discovered this package today and wondered whether anyone here had any > > experience of it. > > Interesting to see this project exists. I thought Chromium was essentially > un-Googled, but obviously there's more there to take place to strip Google's > tentacles from the browser. > > > Which overlay to get it from? > > According to: > > https://github.com/Eloston/ungoogled-chromium/blob/master/README.md > > The overlay they mention is pf4public: > > https://github.com/PF4Public/gentoo-overlay But zugaina.org also mention chaoslab and alinefr. That's why I asked "which" overlay, not "what" overlay. :) > > How stable is it? Does it > > really "privatise" chromium? Does it allow extensions like ublock-origin? > > > > I'm happy with firefox, but it never hurts to have a choice. > > I'd be also interested to know if Falkon/QtWebEngine is more or less /un- > Googled/ than the ungoogled-chromium browser. > > I've gradually grown reluctant to spend many-many hours emerging Chromium > and have removed it from a number of my older boxen. Especially since the > 'jumbo- build' USE flag was dropped, Chromium's compilation takes twice as > long as the many hours it used to take. I'd rather not waste that much > electricity when QtWebEngine is built as a default with Plasma/KDE anyway. Me too, but I do have a copy of google-chrome around here somewhere... -- Regards, Peter.
[gentoo-user] almost free launch: an idea to lower build time, and rice, at the same time
DISCLAIMER: I am not claiming that this idea is new. It is probably not new. --- Even though some of its details might be new for a Linux distribution, it's all based on boring well-established bits of known science. But regardless of its newness, I think it's worth sharing with the hope that it may re-kindle the fire in a nerd's heart (or a group of nerds) so that they develop this for me (or us). GOAL: - Reduce compile time, rice (e.g. fancy USE, make.conf, etc), and yet not increase dev overhead. CURRENT SITUATION: -- If you use *-bin packages, you cannot rice, and must compile on your own. THE APPROACH: - 1. Some nerd (or a group of nerds) makes (or make) a package, maybe call it `almostfreelunch.ebuild`. 2. Say you want to compile qtwebengine. You do: `almostfreelunch -aqvDuNt --backbrack=1000 qtwebengine`. 3. The app, `almostfreelunch`, will lookup your build setup (e.g. USE flags, make.conf settings, etc) for all packages that you are about to build on your system as you are about to install that qtwebengine. 4. The app will upload that info to a central server, which looks up the popularity of certain configurations. E.g. see the distribution of compile-time configurations for a given package. The central server will then figure out things like, qtwebengine is commonly compiled for x86-64 with certain USE flags and other settings in make.conf. 5. If the server figures out that the package that `almostfreelunch` is about to compile is popular enough with the specific build settings that is about to happen, the server will reply to the app and tell it "hi, upload to me your bins when cooked, plz". But if the build setting is not popular enough, it will reply "nothx". This way, the central server will not end up with too much undesired binaries with uncommon build-time settings. 6. The central server will also collect multiple binary packages from multiple people who use `almostfreelunch` for the same packages and the same build-time options. I.e. multiple qtwebengine with identical build-time settings (e.g. same USE flags, make.conf, etc). 7. The central server will perform statistical analysis against all of the uploaded binaries, of the same packages and the same claimed build-time settings, to cross-check those binaries to obtain a statistical confidence in identifying which of the binaries is the good one, and which ones are outliers outlier. Outliers might exist because of users with buggy compilers, or malicious users that intentionally try to inject malware/bugs into their binaries. 8. Thanks to information theory, we will be able to figure out how much redundancy is needed in order to numerically calculate confidence value that shows how trusty a given binary is. E.g. if a package, with specific build-time options, as a very large number of binary submissions that are also extremely similar (i.e. only differ in trivial aspects due to certain randomness in how compilers work), then the central server can calculate a high confidence value for it. Else, the confidence value drops. 9. If a user invokes `almostfreelunch -aqvDuNt --backbrack=1000 qtwebengine` and the central server tells the user that there is an already compiled package with the same settings, then the server simply tells the user, and shows him the confidence associated with the fitness of the binary (based on calculations in stepss (6) to (8)). By default, bins with too-low confidence values will be masked and proper colours will be used to adequately scare the users from low-confidence packages. 10. If at step (9) the user likes the confidence of the pre-compiled binary package, the user can simply download the binary package, blazing fast, with all the nice UES and make.conf flags that he has. Else, the user is free to compile his own version, and upload his own binary, to help the server enhance its confidence as calculated in steps (6) to (8). NOTES: -- * The statistical analysis in step (5) can also consider the compile time of packages. So the minimum popularity required for a specific package build is weighted while considering the total build time. This way, too slow-to-build packages will end up getting a lower minimum popularity than those small packages. Choosing the sweet-spot trade-off is a matter of optimizing resources of the central server. * The statistical analysis in steps (6) to (8) could also be further enhanced by ranking individual users who upload the binaries. Users, who upload bins, could optionally also sign their packages, and henceforth be identified by the central server. Eventually, statistics can be used to also calculate a confidence measure on how trusty a user is. This ca
[gentoo-user] Not enough RAM for dev-qt/qtwebengine build
Hi.For some time I have problems with dev-qt/qtwebengine (at least 5.12.3) build. As far as I can see it fails to build due to memory exhaustion. Although I have 32 GB of RAM (at least 20 of them is almost always free) looks like it's not enough. It's strange because I have no problems with Firefox/Chromium/Libreoffice builds.The build ends with full (32GB/32GB) memory usage and significant system lag (mouse/keyboard input, output, video and sound is freezing for ~10+ seconds) every time and then Portage is marking it as failed with something like this in the end of the logs: https://cp.skobk.in/2096/ae027ec300f014f6Also I have something like this in the kernel log:Jun 13 12:42:24 skobkin-pc kernel: oom-kill:constraint=CONSTRAINT_NONE,nodemask=(null),cpuset=/,mems_allowed=0,global_oom,task_memcg=/user.slice/user-1000.slice/session-1.scope,task=cc1plus,pid=39585,uid=250Jun 13 12:42:24 skobkin-pc kernel: Out of memory: Killed process 39585 (cc1plus) total-vm:1715764kB, anon-rss:1615872kB, file-rss:0kB, shmem-rss:0kBJun 13 12:42:24 skobkin-pc kernel: oom_reaper: reaped process 39585 (cc1plus), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kBIt's the first time when I encountered a problem like this on my system and now I'm thinking how to deal with it. Has anyone dealt with this? Is there any solutions other than buying more RAM (I don't need more RAM for my work/entertainment right now)? Thanks in advance.
Re: [gentoo-user] Not enough RAM for dev-qt/qtwebengine build
Alexey, you can check /etc/portage/make.conf and see if there is a MAKEOPTS="-j8" or similar variable. If not, add it and make the number of jobs small, like "-j2". I have a similar problem when I build sys-devel/binutils; it is a huge memory hog. - Matthew
Re: [gentoo-user] Not enough RAM for dev-qt/qtwebengine build
Alexey Eschenko writes: > Hi. > > For some time I have problems with dev-qt/qtwebengine (at least 5.12.3) > build. As far as I can see it fails to build due to memory exhaustion. > Although I have 32 GB of RAM (at least 20 of them is almost always free) > looks like it's not enough. It's strange because I have > no problems with Firefox/Chromium/Libreoffice builds. > > The build ends with full (32GB/32GB) memory usage and significant system lag > (mouse/keyboard input, output, video and sound is freezing for ~10+ seconds) > every time and then Portage is marking it as failed with something like this > in the end of the logs: > https://cp.skobk.in/2096/ae027ec300f014f6 > > Also I have something like this in the kernel log: > > Jun 13 12:42:24 skobkin-pc kernel: > oom-kill:constraint=CONSTRAINT_NONE,nodemask=(null),cpuset=/,mems_allowed=0,global_oom,task_memcg=/user.slice/user-1000.slice/session-1.scope,task=cc1plus,pid=39585,uid=250 > Jun 13 12:42:24 skobkin-pc kernel: Out of memory: Killed process 39585 > (cc1plus) total-vm:1715764kB, anon-rss:1615872kB, file-rss:0kB, shmem-rss:0kB > Jun 13 12:42:24 skobkin-pc kernel: oom_reaper: reaped process 39585 > (cc1plus), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB > > It's the first time when I encountered a problem like this on my system and > now I'm thinking how to deal with it. Has anyone dealt with this? Is there > any solutions other than buying more RAM (I don't need more RAM for my > work/entertainment right now)? > > Thanks in advance. I can't really help with the problem, but I've built the same package recently with just 4GB of RAM. (It takes a long time.) So most likely it's something in your portage settings that's causing this. - Lasse
Re: [gentoo-user] Not enough RAM for dev-qt/qtwebengine build
On Thu, 13 Jun 2019 12:59:18 -0500, Dale wrote: > Neil Bothwick wrote: > > You can set that in package.env too, I do this for Chromium: > > > > % cat /etc/portage/package.env/chromium > > www-client/chromium alert-done.conf disk-tmpdir.conf j2.conf > > > > % cat /etc/portage/env/j2.conf > > MAKEOPTS="-j2" > > Has that changed? I have a package.env FILE that contains a list of > packages and the path to a file in /etc/portage/env/ that tells what to > change that for package/ Looks like this: > > > root@fireball / # cat /etc/portage/package.env/package.env > #www-client/seamonkey ../env/single.conf > #www-client/firefox ../env/single.conf > #www-client/firefox ../env/notmpfs.conf > #www-client/seamonkey ../env/notmpfs.conf > app-office/libreoffice ../env/notmpfs.conf > #sys-devel/gcc ../env/notmpfs.conf > dev-qt/qtwebengine ../env/notmpfs.conf > #dev-qt/qtwebkit ../env/notmpfs.conf > #sci-electronics/kicad ../env/notmpfs.conf > > root@fireball / # ls -al /etc/portage/env/ > total 16 > drwxr-xr-x 2 root root 4096 Sep 3 2017 . > drwxr-xr-x 14 root root 4096 May 31 00:40 .. > -rw-r--r-- 1 root root 35 Sep 3 2017 notmpfs.conf > -rw-r--r-- 1 root root 31 Feb 12 2017 single.conf > > root@fireball / # cat /etc/portage/env/notmpfs.conf > PORTAGE_TMPDIR="/var/tmp/notmpfs" > > root@fireball / # cat /etc/portage/env/single.conf > #EMERGE_DEFAULT_OPTS="-j1" > > root@fireball / # > > > I think I'm doing this different than what you posted. It works but is > it being changed and I need to update before it stops working? The only difference is the way you are specifying the paths. Portage looks in /etc/portage/env for these anyway, you are telling it to load /etc/portage/env../env/whatever.conf, which is the same thing. -- Neil Bothwick Pound for pound, the amoeba is the most vicious animal on the earth. pgpQ21gLQx3Hr.pgp Description: OpenPGP digital signature
Re: [gentoo-user] switch to profile 17.0 complete, completely painless
El dic 6, 2017 11:20 AM, "Helmut Jarausch" escribió: On 12/06/2017 04:38:02 PM, Andrés Becerra Sandoval wrote: > > I'm having trouble with these: > > net-libs/webkit-gtk-2.4.11-r200 > this needs an addtional patch webkit-gtk-2.4.11-icu59.patch which I've attached dev-qt/qtwebengine-5.9.3 > This needs a tiny change, I've attached my ebuild net-libs/webkit-gtk > which version? Version 2.18.3 installed just fine here. Helmut Thank you Helmut!
Re: [gentoo-user] Not enough RAM for dev-qt/qtwebengine build
It's the first time when I encountered a problem like this on my system and now I'm thinking how to deal with it. Has anyone dealt with this? Is there any solutions other than buying more RAM (I don't need more RAM for my work/entertainment right now)? Thanks in advance. You can use more swap (files) before buying more RAM. -- Sent with eQmail-1.10.2 - a fork of djb's famous qmail
[gentoo-user] Re: Highlight certain packages being upgraded
On 08/07/2023 11:33, Dale wrote: that excessively long qt package Off-topic, but just in case you mean qtwebengine, I was able to get rid of it by putting "-webengine" in my USE flags. After a world update, a depclean should then remove it from the system. You might have to juggle a few other USE flags in specific packages to make it happen though, I forgot. It's been a while.
Re: [gentoo-user] Cannot build qtwebengine: ‘WEBP_EXTERN’ does not name a type
On Mon, Apr 29, 2019 at 08:53:26AM +0100, Mick wrote: > On Sunday, 28 April 2019 21:25:58 BST Frank Steinmetzger wrote: > > Hi folks, > > > > it’s been a while since I had a problem for which I needed a nudge. :) > > Due to some other lib upgrade, I need to rebuild qtwebengine. > > > > But it always fails with: > > | In file included from > > | ../../3rdparty/chromium/third_party/skia/src/images/SkWebpEncoder.cpp:42: > > | /usreincnude/webp/mux.h:101:1: Since libpcap has been upgradederror: > > | ‘WEBP_EXTERN’ does not name a type| > > | WEBP_EXTERN int WebPGetMuxVersion(void); > > | ^~~ > > | > > | /usr/include/webp/mux.h: In function ‘WebPMux* WebPMuxCreate(const > > | WebPData*, int)’: /usr/include/webp/mux.h:138:10: error: > > | ‘WebPMuxCreateInternal’ was not declared in this scope| > > |return WebPMuxCreateInternal(bitstream, copy_data, > > |WEBP_MUX_ABI_VERSION); > > | > > | ^ > > > > And a number more of similar problems, hinting at a missing or broken > > include file in libpcap. I tried upgrading libpcap from 1.8 to 1.9 as a > > remedy, but no luck there. > > These are the libcap versions listed in my system and there is no 1.8 or 1.9 > versions here: We both made a mistake here. Notice I mentioned libpcap (with two “p”s), you only read one. And my mistake was to look at libpcap at all, when I should have been looking at libwebp (oh those damn Ps). After all, that is the lib which contains the culprit mux.h and which was upgraded from 0.5 to 1.0, causing a rebuild in the first place. (In hindsight, why should a network library have a file about muxing anyway). I’ll try a libwebp downgrade now. -- Gruß | Greetings | Qapla’ Please do not share anything from, with or about me on any social network. How many people can read hex if only you and dead people can read hex? deae people. signature.asc Description: PGP signature
Re: [gentoo-user] Not enough RAM for dev-qt/qtwebengine build
Neil Bothwick wrote: > On Thu, 13 Jun 2019 14:48:04 +0300, Alexey Eschenko wrote: > >> That was my first move. Unfortunately it didn't help. >> But in the first answer in this thread I saw a good idea about >> decreasing number of parallel build threads. I have MAKEOPTS="-j32" and >> it's probably the cause of my problem with this package. Right now I'm >> trying separate build environment with decreased job count value and it >> looks promissing. I'll leave the message with results in the reply to >> that message. 13.06.2019, 14:13, "Neil Bothwick" : > You can set that in package.env too, I do this for Chromium: > > % cat /etc/portage/package.env/chromium > www-client/chromium alert-done.conf disk-tmpdir.conf j2.conf > > % cat /etc/portage/env/j2.conf > MAKEOPTS="-j2" > Has that changed? I have a package.env FILE that contains a list of packages and the path to a file in /etc/portage/env/ that tells what to change that for package/ Looks like this: root@fireball / # cat /etc/portage/package.env/package.env #www-client/seamonkey ../env/single.conf #www-client/firefox ../env/single.conf #www-client/firefox ../env/notmpfs.conf #www-client/seamonkey ../env/notmpfs.conf app-office/libreoffice ../env/notmpfs.conf #sys-devel/gcc ../env/notmpfs.conf dev-qt/qtwebengine ../env/notmpfs.conf #dev-qt/qtwebkit ../env/notmpfs.conf #sci-electronics/kicad ../env/notmpfs.conf root@fireball / # ls -al /etc/portage/env/ total 16 drwxr-xr-x 2 root root 4096 Sep 3 2017 . drwxr-xr-x 14 root root 4096 May 31 00:40 .. -rw-r--r-- 1 root root 35 Sep 3 2017 notmpfs.conf -rw-r--r-- 1 root root 31 Feb 12 2017 single.conf root@fireball / # cat /etc/portage/env/notmpfs.conf PORTAGE_TMPDIR="/var/tmp/notmpfs" root@fireball / # cat /etc/portage/env/single.conf #EMERGE_DEFAULT_OPTS="-j1" root@fireball / # I think I'm doing this different than what you posted. It works but is it being changed and I need to update before it stops working? Dale :-) :-)
Re: [gentoo-user] Not enough RAM for dev-qt/qtwebengine build
Neil Bothwick wrote: > On Thu, 13 Jun 2019 12:59:18 -0500, Dale wrote: > >> Neil Bothwick wrote: >>> You can set that in package.env too, I do this for Chromium: >>> >>> % cat /etc/portage/package.env/chromium >>> www-client/chromium alert-done.conf disk-tmpdir.conf j2.conf >>> >>> % cat /etc/portage/env/j2.conf >>> MAKEOPTS="-j2" >> Has that changed? I have a package.env FILE that contains a list of >> packages and the path to a file in /etc/portage/env/ that tells what to >> change that for package/ Looks like this: >> >> >> root@fireball / # cat /etc/portage/package.env/package.env >> #www-client/seamonkey ../env/single.conf >> #www-client/firefox ../env/single.conf >> #www-client/firefox ../env/notmpfs.conf >> #www-client/seamonkey ../env/notmpfs.conf >> app-office/libreoffice ../env/notmpfs.conf >> #sys-devel/gcc ../env/notmpfs.conf >> dev-qt/qtwebengine ../env/notmpfs.conf >> #dev-qt/qtwebkit ../env/notmpfs.conf >> #sci-electronics/kicad ../env/notmpfs.conf >> >> root@fireball / # ls -al /etc/portage/env/ >> total 16 >> drwxr-xr-x 2 root root 4096 Sep 3 2017 . >> drwxr-xr-x 14 root root 4096 May 31 00:40 .. >> -rw-r--r-- 1 root root 35 Sep 3 2017 notmpfs.conf >> -rw-r--r-- 1 root root 31 Feb 12 2017 single.conf >> >> root@fireball / # cat /etc/portage/env/notmpfs.conf >> PORTAGE_TMPDIR="/var/tmp/notmpfs" >> >> root@fireball / # cat /etc/portage/env/single.conf >> #EMERGE_DEFAULT_OPTS="-j1" >> >> root@fireball / # >> >> >> I think I'm doing this different than what you posted. It works but is >> it being changed and I need to update before it stops working? > The only difference is the way you are specifying the paths. Portage > looks in /etc/portage/env for these anyway, you are telling it to > load /etc/portage/env../env/whatever.conf, which is the same thing. > > OK. If something was changing, I wanted to update if needed. I sort of think your way is simpler tho. It took me a while to figure out how to make mine work. At the time, I couldn't find a lot of examples. Thanks for the info. Dale :-) :-)
Re: [gentoo-user] Re: Compile large packages as last package
Thank you for your ideas!
I was actually hoping for a neat hack with "/etc/portage/env/" and
"/etc/portage/package.env/", where you can set environment variables.
I will try out the following solution:
$ < "/etc/portage/package.env/no_tmpfs.conf
# custom - 20181121 - rfischer: list packages, which are too big for
tmpfs
#app-emulation/qemu-kv no_tmpfs.conf
#app-office/libreoffice no_tmpfs.conf
#dev-java/icedtea no_tmpfs.conf
#dev-lang/ghc no_tmpfs.conf
#dev-lang/rust no_tmpfs.conf
#mail-client/thunderbird no_tmpfs.conf
#sci-libs/tensorflow no_tmpfs.conf
#sys-devel/gcc no_tmpfs.conf
#www-client/firefox no_tmpfs.conf
#www-client/ungoogled-chromium no_tmpfs.conf #throttle_make_emerge.conf
$ < "${HOME}/bin/update.sh"
[...]
large_package_list=$(/bin/grep --extended-regexp --only-matching
"[a-z]+-[a-z]+\/[-0-9a-zA-Z]+" "/etc/portage/package.env/no_tmpfs.conf")
[...]
/usr/bin/emerge --ask --update --deep --newuse --tree --verbose
--exclude="${large_package_list//$'\n'/ }" @world
/usr/bin/emerge --ask --update --deep --newuse --tree --verbose @world
-Ramon
On 15/08/2021 17:48, Nikos Chantziaras wrote:
On 14/08/2021 22:20, Ramon Fischer wrote:
Is there any way to tell "portage", that packages like "qtwebengine",
"(ungoogled-)chromium", "firefox" and so on are always compiled as
last package?
The simplest way is to exclude those packages in the first update, and
then allow them in the second:
emerge -uDU @world --exclude "qtwebengine firefox chromium" && emerge
-uDU @world
The dependency tracker of portage will of course also exclude packages
that depend on the excluded packages, unless they themselves have
updates pending. In that case, they *might* get built twice; once
against the current version of the excluded packages, and then perhaps
again on the second run, if there's rebuild triggers involved.
Most of the time though, you won't run into cases of redundant
rebuilds. Rebuild triggers are not very common.
--
GPG public key: 5983 98DA 5F4D A464 38FD CF87 155B E264 13E6 99BF
OpenPGP_signature
Description: OpenPGP digital signature
Re: [gentoo-user] Re: Compile large packages as last package
>Thank you for your ideas!
>
>I was actually hoping for a neat hack with "/etc/portage/env/" and
>"/etc/portage/package.env/", where you can set environment variables.
Did you look at example 2 in
https://wiki.gentoo.org/wiki//etc/portage/package.env ?
It seems to address your problem.
DaveF
>
>I will try out the following solution:
>
>$ < "/etc/portage/package.env/no_tmpfs.conf
># custom - 20181121 - rfischer: list packages, which are too big for
>tmpfs
>#app-emulation/qemu-kv no_tmpfs.conf
>#app-office/libreoffice no_tmpfs.conf
>#dev-java/icedtea no_tmpfs.conf
>#dev-lang/ghc no_tmpfs.conf
>#dev-lang/rust no_tmpfs.conf
>#mail-client/thunderbird no_tmpfs.conf
>#sci-libs/tensorflow no_tmpfs.conf
>#sys-devel/gcc no_tmpfs.conf
>#www-client/firefox no_tmpfs.conf
>#www-client/ungoogled-chromium no_tmpfs.conf #throttle_make_emerge.conf
>
>$ < "${HOME}/bin/update.sh"
>[...]
>large_package_list=$(/bin/grep --extended-regexp --only-matching
>"[a-z]+-[a-z]+\/[-0-9a-zA-Z]+" "/etc/portage/package.env/no_tmpfs.conf")
>[...]
>/usr/bin/emerge --ask --update --deep --newuse --tree --verbose
>--exclude="${large_package_list//$'\n'/ }" @world
>/usr/bin/emerge --ask --update --deep --newuse --tree --verbose @world
>
>-Ramon
>
>On 15/08/2021 17:48, Nikos Chantziaras wrote:
>> On 14/08/2021 22:20, Ramon Fischer wrote:
>>> Is there any way to tell "portage", that packages like "qtwebengine",
>>> "(ungoogled-)chromium", "firefox" and so on are always compiled as
>>> last package?
>>
>> The simplest way is to exclude those packages in the first update, and
>> then allow them in the second:
>>
>> emerge -uDU @world --exclude "qtwebengine firefox chromium" && emerge
>> -uDU @world
>>
>> The dependency tracker of portage will of course also exclude packages
>> that depend on the excluded packages, unless they themselves have
>> updates pending. In that case, they *might* get built twice; once
>> against the current version of the excluded packages, and then perhaps
>> again on the second run, if there's rebuild triggers involved.
>>
>> Most of the time though, you won't run into cases of redundant
>> rebuilds. Rebuild triggers are not very common.
>>
>>
>
>--
>GPG public key: 5983 98DA 5F4D A464 38FD CF87 155B E264 13E6 99BF
>
>
>
>>> application/pgp-signature attachment, name=OpenPGP_signature
Re: Re: [gentoo-user] gst-plugins-bad-1.11.90 is blocking gst-plugins-base-1.12.3
Now I've an error: tux ~ # emerge --verbose-conflicts @preserved-rebuild * IMPORTANT: 8 news items need reading for repository 'gentoo'. * Use eselect news read to view new items. Calculating dependencies... done! [ebuild R ] dev-libs/botan-1.10.17 [ebuild U ] media-gfx/imagemagick-7.0.7.6 [6.9.9.0] [ebuild R ] dev-python/pillow-4.2.1-r1 [ebuild U ] dev-qt/qtgui-5.9.2 [5.7.1-r1] USE="libinput* -vnc%" [ebuild R ] media-video/ffmpeg-3.3.4 [ebuild U ] dev-qt/qtwidgets-5.9.2 [5.7.1] USE="gtk%*" [ebuild U ] media-libs/gst-plugins-ugly-1.12.3 [1.10.5] [ebuild U ] dev-qt/qtdeclarative-5.9.2 [5.7.1] [ebuild U ] dev-qt/qtprintsupport-5.9.2 [5.7.1] [ebuild U ] media-plugins/gst-plugins-x264-1.12.3 [1.10.5] [ebuild U ] media-video/vlc-2.2.6-r2 [2.2.6] USE="qt5*" [ebuild U ] dev-qt/qtwebchannel-5.9.2 [5.7.1] [ebuild R ] net-analyzer/wireshark-2.4.2 [ebuild U ] xfce-base/xfwm4-4.13.0-r1 [4.12.3-r1] USE="opengl%* -xpresent%" [ebuild U ] app-editors/vim-8.0.1188 [8.0.0386] USE="terminal%*" PYTHON_SINGLE_TARGET="python3_4%* -python2_7% -python3_5% -python3_6%" [ebuild U ] dev-qt/qtwebengine-5.9.2 [5.7.1-r2] [ebuild R ] dev-db/postgresql-9.6.5-r1 [ebuild U ] media-video/obs-studio-20.0.1-r1 [20.0.1] [ebuild U ] net-print/cups-filters-1.17.9 [1.16.4] USE="-pclm%" [ebuild U ] media-gfx/graphviz-2.40.1 [2.38.0-r1] [ebuild R ] www-servers/nginx-1.12.1 [ebuild R ] dev-lang/php-7.0.24 [ebuild R ] dev-db/mariadb-10.2.9 !!! Multiple package instances within a single package slot have been pulled !!! into the dependency graph, resulting in a slot conflict: dev-qt/qtgui:5 (dev-qt/qtgui-5.7.1-r1:5/5.7::gentoo, installed) pulled in by ~dev-qt/qtgui-5.7.1 required by (dev-qt/qtwebkit-5.7.1:5/5.7::gentoo, installed) ^ ^ ~dev-qt/qtgui-5.7.1[gles2=] required by (dev-qt/qtopengl-5.7.1:5/5.7::gentoo, installed) ^ ^ ~dev-qt/qtgui-5.7.1 required by (dev-qt/qthelp-5.7.1:5/5.7::gentoo, installed) ^ ^ ~dev-qt/qtgui-5.7.1[gles2=] required by (dev-qt/qtprintsupport-5.7.1:5/5.7::gentoo, installed) ^ ^ ~dev-qt/qtgui-5.7.1 required by (dev-qt/qtquickcontrols-5.7.1:5/5.7::gentoo, installed) ^ ^ ~dev-qt/qtgui-5.7.1[gles2=] required by (dev-qt/qtmultimedia-5.7.1:5/5.7::gentoo, installed) ^ ^ ~dev-qt/qtgui-5.7.1[gles2=,png=,xcb?] required by (dev-qt/qtwidgets-5.7.1:5/5.7::gentoo, installed) ^ ^ ~dev-qt/qtgui-5.7.1[gles2=] required by (dev-qt/qtdeclarative-5.7.1:5/5.7::gentoo, installed) ^ ^ ~dev-qt/qtgui-5.7.1 required by (dev-qt/qtsvg-5.7.1:5/5.7::gentoo, installed) ^ ^ ~dev-qt/qtgui-5.7.1[xcb] required by (dev-qt/qtx11extras-5.7.1:5/5.7::gentoo, installed) ^ ^ (dev-qt/qtgui-5.9.2:5/5.9::gentoo, ebuild scheduled for merge) pulled in by ~dev-qt/qtgui-5.9.2[gles2=,png=,xcb?] required by (dev-qt/qtwidgets-5.9.2:5/5.9::gentoo, ebuild scheduled for merge) ^ ^ ~dev-qt/qtgui-5.9.2[gles2=] required by (dev-qt/qtdeclarative-5.9.2:5/5.9::gentoo, ebuild scheduled for merge) ^ ^ ~dev-qt/qtgui-
Re: [gentoo-user] Not enough RAM for dev-qt/qtwebengine build
> > I can't really help with the problem, but I've built the same package > recently with just 4GB of RAM. (It takes a long time.) So most likely > it's something in your portage settings that's causing this. > > - Lasse Agreed. I’ve built it recently on 16GB of RAM. My MAKEOPTS is -j13 -l5 to allow good distribution of effort across 2 other machines with distcc. Only about 7-8 GB appears to be actively used on the machine running the emerge. John
Re: [gentoo-user] Not enough RAM for dev-qt/qtwebengine build
> > You can use more swap (files) before buying more RAM. > > I have been doing this too. It only get used during the big builds. To create a 32G swap file and enable it (OP can do this now as the build runs, to keep OOM away) # fallocate -l 32G && chmod 600 && mkswap && swapon use 'watch free -m' to see if its being used. Then after the build is over, delete it/add it to fstab/whatever
[gentoo-user] Why falkon?
Hello list, Would someone kindly enloghten me as to why the presence of qtwebengine now requires installation of www-client/falkon? $ cat $(equery w kdecore-meta) # Copyright 1999-2018 Gentoo Authors # Distributed under the terms of the GNU General Public License v2 EAPI=6 inherit kde5-meta-pkg DESCRIPTION="kdecore - merge this to pull in the most basic applications" KEYWORDS="~amd64 ~x86" IUSE="+handbook +webengine webkit" RDEPEND=" $(add_kdeapps_dep dolphin) $(add_kdeapps_dep kdialog) $(add_kdeapps_dep keditbookmarks) $(add_kdeapps_dep kfind) $(add_kdeapps_dep konsole) $(add_kdeapps_dep kwrite) handbook? ( $(add_kdeapps_dep khelpcenter) ) webengine? ( www-client/falkon ) webkit? ( $(add_kdeapps_dep konqueror 'webengine?') kde-misc/kwebkitpart:5 ) " -- Regards, Peter.
Re: [gentoo-user] Not enough RAM for dev-qt/qtwebengine build
> > More swap will help, but the basic problem here is the HUGE number of jobs > specified. If each job eats say up to 2G when it gets going, then -j32 > will > require >64G RAM to keep running without thrashing the swap device, or > running > OOM. > Yes agree. If the swapfile is on an SSD then there's no thrashing. For me the swap file only gets used on big builds and /proc/sys/vm/swappiness is tweaked via /etc/sysctl.conf so SSD wear is not a big concern.
Re: [gentoo-user] Compile large packages as last package
On Sat, 14 Aug 2021 21:20:23 +0200, Ramon Fischer wrote: > Is there any way to tell "portage", that packages like "qtwebengine", > "(ungoogled-)chromium", "firefox" and so on are always compiled as last > package? You can use the --exclude option for emerge. Maybe a script that runs emerge @world with --exclude then again without? While I haven't tried it with a script, I do use --exclude when chromium is in the list because my laptop only has 8GB RAM, which means chromium takes ages. -- Neil Bothwick Isn't 'Criminal Lawyer' rather redundant? pgp7adfDN1yWU.pgp Description: OpenPGP digital signature
Re: [gentoo-user] switch to profile 17.0 complete, completely painless
On Wed, 6 Dec 2017 01:18:02 -0600, Dale wrote: > I'm having trouble with these: > > net-libs/webkit-gtk-2.4.11-r200 > dev-qt/qtwebengine-5.9.3 > net-libs/webkit-gtk > > Those three, I've had to adjust the USE flags and it may or may not be > profile switch related. If I had to guess, it just happened to pop up > and isn't related to the switch. I did too, but webkit-gtk was failing a rebuild before the switch, so I think you're right. Solutions for them can be found on bgo. -- Neil Bothwick SITCOM: Single Income, Two Children, Oppressive Mortgage pgpuAvE89oNJ6.pgp Description: OpenPGP digital signature
Re: [gentoo-user] Going through these one by one.
On 2020.11.19 17:37, Adam Carter wrote: On Fri, Nov 20, 2020 at 8:36 AM Steven Lembark wrote: > > Based on: > > 2020-04-22 Python 3.7 to become the default target > > I'd have thought that using: > > PYTHON_TARGETS="python2_7 python3_7" > PYTHON_SINGLE_TARGET="python3_7" > > I would remove those entries and try again. As a Gnome ~amd64 user, python 2.7 was depcleaned on the 11th as the last dependency was removed. I don't think that's quite fully true - as it's still a build time dependency for a small number of packages, chromium and qtwebengine being somewhat notable examples. However, removing all explicity python target flags is still a reasonable thing to try.
Re: [gentoo-user] almost free launch: an idea to lower build time, and rice, at the same time
Le mar. 5 nov. 2019 à 01:02, Caveman Al Toraboran a écrit : > > > DISCLAIMER: I am not claiming that this idea is new. It is probably not new. > --- Even though some of its details might be new for a Linux > distribution, it's all based on boring well-established bits of > known science. But regardless of its newness, I think it's worth > sharing with the hope that it may re-kindle the fire in a nerd's > heart (or a group of nerds) so that they develop this for me (or > us). > > > > GOAL: > - > Reduce compile time, rice (e.g. fancy USE, make.conf, etc), and yet not > increase dev overhead. > > > CURRENT SITUATION: > -- > If you use *-bin packages, you cannot rice, and must compile on your own. > > > THE APPROACH: > - > 1. Some nerd (or a group of nerds) makes (or make) a package, maybe call it >`almostfreelunch.ebuild`. > > 2. Say you want to compile qtwebengine. You do: `almostfreelunch -aqvDuNt >--backbrack=1000 qtwebengine`. > > 3. The app, `almostfreelunch`, will lookup your build setup (e.g. USE flags, >make.conf settings, etc) for all packages that you are about to build on >your system as you are about to install that qtwebengine. > > 4. The app will upload that info to a central server, which looks up the >popularity of certain configurations. E.g. see the distribution of >compile-time configurations for a given package. The central server will >then figure out things like, qtwebengine is commonly compiled for x86-64 >with certain USE flags and other settings in make.conf. > > 5. If the server figures out that the package that `almostfreelunch` is about >to compile is popular enough with the specific build settings that is about >to happen, the server will reply to the app and tell it "hi, upload to me >your bins when cooked, plz". But if the build setting is not popular >enough, it will reply "nothx". This way, the central server will not end > up >with too much undesired binaries with uncommon build-time settings. > > 6. The central server will also collect multiple binary packages from multiple >people who use `almostfreelunch` for the same packages and the same >build-time options. I.e. multiple qtwebengine with identical build-time >settings (e.g. same USE flags, make.conf, etc). > > 7. The central server will perform statistical analysis against all of the >uploaded binaries, of the same packages and the same claimed build-time >settings, to cross-check those binaries to obtain a statistical confidence >in identifying which of the binaries is the good one, and which ones are >outliers outlier. Outliers might exist because of users with buggy >compilers, or malicious users that intentionally try to inject malware/bugs >into their binaries. > > 8. Thanks to information theory, we will be able to figure out how much >redundancy is needed in order to numerically calculate confidence value > that >shows how trusty a given binary is. E.g. if a package, with specific >build-time options, as a very large number of binary submissions that are >also extremely similar (i.e. only differ in trivial aspects due to certain >randomness in how compilers work), then the central server can calculate a >high confidence value for it. Else, the confidence value drops. > > 9. If a user invokes `almostfreelunch -aqvDuNt --backbrack=1000 qtwebengine` >and the central server tells the user that there is an already compiled >package with the same settings, then the server simply tells the user, and >shows him the confidence associated with the fitness of the binary (based > on >calculations in stepss (6) to (8)). By default, bins with too-low >confidence values will be masked and proper colours will be used to >adequately scare the users from low-confidence packages. > > 10. If at step (9) the user likes the confidence of the pre-compiled binary >package, the user can simply download the binary package, blazing fast, > with >all the nice UES and make.conf flags that he has. Else, the user is free > to >compile his own version, and upload his own binary, to help the server >enhance its confidence as calculated in steps (6) to (8). > > > NOTES: > -- > * The statistical analysis in step (5) can also consider the compile time of > packages. So the minimum popularity required for a specific package build > is > weighted while considering the total build time. This way, too > slow-to-build > packages will en
Re: [gentoo-user] Cannot build qtwebengine: ‘WEBP_EXTERN’ does not name a type
On Sunday, 28 April 2019 21:25:58 BST Frank Steinmetzger wrote: > Hi folks, > > it’s been a while since I had a problem for which I needed a nudge. :) > Due to some other lib upgrade, I need to rebuild qtwebengine. > > But it always fails with: > | In file included from > | ../../3rdparty/chromium/third_party/skia/src/images/SkWebpEncoder.cpp:42: > | /usreincnude/webp/mux.h:101:1: Since libpcap has been upgradederror: > | ‘WEBP_EXTERN’ does not name a type| > | WEBP_EXTERN int WebPGetMuxVersion(void); > | ^~~ > | > | /usr/include/webp/mux.h: In function ‘WebPMux* WebPMuxCreate(const > | WebPData*, int)’: /usr/include/webp/mux.h:138:10: error: > | ‘WebPMuxCreateInternal’ was not declared in this scope| > |return WebPMuxCreateInternal(bitstream, copy_data, > |WEBP_MUX_ABI_VERSION); > | > | ^ > > And a number more of similar problems, hinting at a missing or broken > include file in libpcap. I tried upgrading libpcap from 1.8 to 1.9 as a > remedy, but no luck there. These are the libcap versions listed in my system and there is no 1.8 or 1.9 versions here: [I] sys-libs/libcap Available versions: 2.25[pam static-libs ABI_MIPS="n32 n64 o32" ABI_PPC="32 64" ABI_S390="32 64" ABI_X86="32 64 x32"] 2.26-r2 ^t[pam static-libs ABI_MIPS="n32 n64 o32" ABI_PPC="32 64" ABI_S390="32 64" ABI_X86="32 64 x32"] ~2.27 ^t[pam static-libs ABI_MIPS="n32 n64 o32" ABI_PPC="32 64" ABI_S390="32 64" ABI_X86="32 64 x32"] Installed versions: 2.26-r2^t(13:18:30 04/19/19)(pam -static-libs ABI_MIPS="-n32 -n64 -o32" ABI_PPC="-32 -64" ABI_S390="-32 -64" ABI_X86="64 -32 -x32") Homepage:http://www.friedhoff.org/posixfilecaps.html Description: POSIX 1003.1e capabilities [I] sys-libs/libcap-ng Available versions: 0.7.8 ^t[python static-libs PYTHON_TARGETS="python2_7 python3_5 python3_6"] ["python? ( || ( python_targets_python2_7 python_targets_python3_5 python_targets_python3_6 ) )"] ~0.7.9 [python static-libs PYTHON_TARGETS="python2_7 python3_5 python3_6 python3_7"] ["python? ( || ( python_targets_python2_7 python_targets_python3_5 python_targets_python3_6 python_targets_python3_7 ) )"] Installed versions: 0.7.8^t(13:27:50 03/23/19)(-python -static-libs PYTHON_TARGETS="python2_7 python3_6 -python3_5") Homepage:https://people.redhat.com/sgrubb/libcap-ng/ Description: POSIX 1003.1e capabilities Instead of manually emerging/upgrading libraries it may be better to run 'emerge -a -v @preserved-rebuild' and let portage decide which version is required, through its thread of dependencies. > There is also no relevant entry on BGO and I > asked twice on IRC with this, but noone answered. Search the emerge output for "Error" to find out all errors printed out by emerge. With multicore/multithread CPUs it is easy to miss "Error 1" further up the page. HTH. -- Regards, Mick signature.asc Description: This is a digitally signed message part.
[gentoo-user] Re: /var/tmp on tmpfs
On 08/02/18 21:17, Dale wrote: I have 16GBs of memory here and have /var/tmp/portage/ on tmpfs, no ccache. With the growing size of packages, I've had to put several on regular spinning rust to make sure enough space is available. This is my list, so far. www-client/firefox www-client/seamonkey app-office/libreoffice sys-devel/gcc dev-qt/qtwebengine dev-qt/qtwebkit I'm on 16GB as well, with a 9GB /var/tmp/portage tmpfs, and all of the above build fine. No need to build them on disk. Note that tmpfs does not consume any memory if there's no files in it! So you can make your tmpfs larger. Here, 9GB is enough to build the above. However, that's of course for when not using --jobs for emering. Just regular -jN in MAKEOPTS.
[gentoo-user] Removing or stopping binary emerges from being in emerge.log
Howdy, I have a chroot environment that I do updates in. Once the updates are done, I copy the binaries and distfiles over to my running system and use the -k option to update everything in my real system. It comes in real handy when libreoffice, Firefox, qtwebengine and other large time consuming packages are being updated. The bad thing is, I have the full length of build time in the chroot but the binary install on my running system. Is there a way to either stop it from logging binary updates or removing them after it is done? I'd rather it not keep those times in either place really. I can't find a emerge option. It seems to record everything regardless. My reason for this, the binary install times throws off genlop -c and its estimates. Anybody have ideas? Thanks. Dale :-) :-)
Re: [gentoo-user] Jobs and load-average
On Wednesday, 15 February 2023 13:18:24 GMT Rich Freeman wrote: > First, keep in mind that --jobs=16 + -j16 can result in up to 256 > (16*16) tasks running at once. Of course, that is worst case and most > of the time you'll have way less than that. Yes, I was aware of that, but why didn't --load-average=32 take precedence? --->8 > Some packages that I build with either a greatly reduced -j setting or > a non-tmpfs build directory are: > sys-cluster/ceph > dev-python/scipy > dev-python/pandas > app-office/calligra > net-libs/nodejs > dev-qt/qtwebengine > dev-qt/qtwebkit > dev-lang/spidermonkey > www-client/chromium > app-office/libreoffice > sys-devel/llvm > dev-lang/rust (I use the rust binary these days as this has gotten > really out of hand) > x11-libs/gtk+ Thanks for the list, Rich. -- Regards, Peter.
Re: [gentoo-user] Re: Highlight certain packages being upgraded
Nikos Chantziaras wrote: > On 08/07/2023 11:33, Dale wrote: >> that excessively long qt package > Off-topic, but just in case you mean qtwebengine, I was able to get > rid of it by putting "-webengine" in my USE flags. After a world > update, a depclean should then remove it from the system. > > You might have to juggle a few other USE flags in specific packages to > make it happen though, I forgot. It's been a while. > > > That's a idea but since everything works well enough, I don't mind the compile times. After all, I nap while it does it anyway. Someone else may see that info and find it interesting tho. Maybe someone with a laptop or a really under-powered system. I never thought about trying to get rid of it. I didn't know it was possible even. Thanks. Dale :-) :-)
Re: [gentoo-user] zoom?
On 25/3/20 8:42 pm, Michael wrote:
On Wednesday, 25 March 2020 12:38:35 GMT Jorge Almeida wrote:
On Wed, Mar 25, 2020 at 12:16 PM Alarig Le Lay wrote:
On mer. 25 mars 11:51:33 2020, Jorge Almeida wrote:
Did someone try to install zoom? (relevant to many people during the
current crisis)
https://support.zoom.us/hc/en-us/articles/204206269-Installing-Zoom-on-L
inux
I downloaded an archive (cannot find the URL again; the site is that
bad) and the directory doesn't even contain a REDME...)
Jorge Almeida
Why didn’t you try net-im/zoom?
I don't have that in /usr/portage. emerge--sync'ing now...
Have you looked at using WebRTC with Zoom, rather than installing their code
on your PC?
Anyone able to install this? I can't get mesa and the libglvnd
dispatcher to coexist.
It does work in chromium, but I think some of the options are not available
BillK.
rattus ~ # emerge net-im/zoom -vp
These are the packages that would be merged, in order:
Calculating dependencies... done!
[ebuild N ] media-libs/libglvnd-1.3.1::gentoo USE="X -test"
ABI_X86="(64) -32 (-x32)" 0 KiB
[ebuild N ] dev-qt/qtwebchannel-5.14.1:5/5.14::gentoo USE="qml
-debug -test" 198 KiB
[ebuild N ] dev-qt/qtpositioning-5.14.1:5/5.14::gentoo USE="qml
-debug -geoclue -test" 5,976 KiB
[ebuild N ] dev-qt/qtscript-5.14.1:5/5.14::gentoo USE="jit -debug
-scripttools -test" 2,593 KiB
[ebuild N ] dev-qt/qtwebengine-5.14.1:5/5.14::gentoo USE="alsa
pulseaudio system-ffmpeg system-icu widgets -bindist -debug -designer
-geolocation -jumbo-build -test" 236,757 KiB
[ebuild N ~] net-im/zoom-3.5.374815.0324-r2::gentoo USE="pulseaudio"
70,908 KiB
[blocks B ] media-libs/mesa[-libglvnd(-)]
("media-libs/mesa[-libglvnd(-)]" is blocking media-libs/libglvnd-1.3.1)
Total: 6 packages (6 new), Size of downloads: 316,430 KiB
Conflict: 1 block (1 unsatisfied)
* Error: The above package list contains packages which cannot be
* installed at the same time on the same system.
(media-libs/mesa-19.3.5:0/0::gentoo, installed) pulled in by
media-libs/mesa[egl,X(+)] required by
(dev-qt/qtwebengine-5.14.1:5/5.14::gentoo, ebuild scheduled for merge)
USE="alsa pulseaudio system-ffmpeg system-icu widgets -bindist -debug
-designer -geolocation -jumbo-build -test" ABI_X86="(64)"
(media-libs/libglvnd-1.3.1:0/0::gentoo, ebuild scheduled for merge)
pulled in by
media-libs/libglvnd required by
(net-im/zoom-3.5.374815.0324-r2:0/0::gentoo, ebuild scheduled for merge)
USE="pulseaudio" ABI_X86="(64)"
For more information about Blocked Packages, please refer to the following
section of the Gentoo Linux x86 Handbook (architecture is irrelevant):
https://wiki.gentoo.org/wiki/Handbook:X86/Working/Portage#Blocked_packages
Re: [gentoo-user] zoom?
On 1/4/20 4:55 pm, Neil Bothwick wrote:
> On Wed, 1 Apr 2020 09:12:46 +0800, William Kenworthy wrote:
>
>> [blocks B ] media-libs/mesa[-libglvnd(-)]
>> ("media-libs/mesa[-libglvnd(-)]" is blocking media-libs/libglvnd-1.3.1)
> It looks like you need to emerge mesa with USE="libglvnd".
>
>
Hi Neil,
no, there is something else at play - I have tried adding it to
package.accept_keywords, package.provided etc. but they fail. If I use
--nodeps it wants to overwrite a number of mesa files so I am thinking
its not actually needed.
Trying to install just zoom and its deps without the --nodeps flag for
each package - working so far but I cant tell until qtwebengine finishes
compiling ... hopefully before Easter :)
I am hoping that mesa will supply the files and libglvnd is not needed
and the ebuild just needs fixing.
I thought someone else must have run across this already?
BillK
signature.asc
Description: OpenPGP digital signature
Re: [gentoo-user] zoom?
> > no, there is something else at play - I have tried adding it to > package.accept_keywords, package.provided etc. but they fail. If I use > --nodeps it wants to overwrite a number of mesa files so I am thinking > its not actually needed. > > Trying to install just zoom and its deps without the --nodeps flag for > each package - working so far but I cant tell until qtwebengine finishes > compiling ... hopefully before Easter :) > > I am hoping that mesa will supply the files and libglvnd is not needed > and the ebuild just needs fixing. > > I thought someone else must have run across this already? > I had some weird blocks during a world update (was it eselect-opencl?) that i got past with media-libs/mesa -libglvnd in package.use. I then tried removing the package.use entry and rebuilding mesa - no blockers this time and the rebuild worked. Everything seems happy now, but I was thinking ebuild bug at the time.
Re: [gentoo-user] long compiles
Alan McKinnon wrote: > After my long time away from Gentoo, I thought perhaps some packages > that always took ages to compile would have improved. I needed to > change to ~amd64 anyway (dumb n00b mistake leaving it at amd64). So > that's what I did and let emerge do it's thing. > > chromium has been building since 10:14, it's now 21:16 and still going > so 9 hours at least on this machine to build a browser - almost as bad > as openoffice at it's worst (regularly took 12 hours). Nodejs also > took a while, but I didn't record time. > > > What other packages have huge build times? > > -- > Alan McKinnon > alan dot mckinnon at gmail dot com I have some software you don't likely use that takes a while but one that is common is qtwebengine or something. If it's not that one, it's qtweb something. It takes about 4 hours, sometimes 5 or so. I think the software takes longer to compile so that we will build new rigs. ROFL Dale :-) :-)
Re: [gentoo-user] long compiles
On Mon, 11 Sep 2023 21:19:27 +0200, Alan McKinnon wrote: > chromium has been building since 10:14, it's now 21:16 and still going > so 9 hours at least on this machine to build a browser - almost as bad > as openoffice at it's worst (regularly took 12 hours). Nodejs also took > a while, but I didn't record time. Chromium is definitely the worst, and strangely variable. The last few compiles have taken between 6 and 14 hours. Since it takes longer than everything else to build, it is usually compiling on its own, so parallel emerges aren't a factor. Qtwebengine is also bad, not surprising as it is a cut down Chromium. Emerging world with --exclude then timing build to coincide with sleep helps, although I haven't quite reached the age where I need 14 hours of sleep a day. -- Neil Bothwick If it isn't broken, I can fix it. pgpb9yNuiaxzs.pgp Description: OpenPGP digital signature
[gentoo-user] Compiling/loading problem
Hello list, I'm having problems with libreoffice. I thought at first it was a space problem, after the recent thread on running out of it, but I now have /tmp as part of /, not separate, and /var/tmp/portage on a 100GB partition of a SATA SSD. The root partition has 35GB spare. Before I try again with /tmp on its own partition, does the following extract from the emerge log file ring any bells? --->8 [CAT] CustomTarget/sysui/share/libreofficedev/create_tree.sh mkdir -p /mnt/scratch/tmp/portage/app-office/libreoffice-6.4.7.2/work/ libreoffice-6.4.7.2/ workdir/CustomTarget/sysui/share/libreofficedev/ /usr/lib/gcc/x86_64-pc-linux-gnu/10.3.0/../../../../x86_64-pc-linux-gnu/bin/ ld: /usr/lib/g cc/x86_64-pc-linux-gnu/10.3.0/../../../../lib64/libpoppler.so: undefined reference to `std ::__throw_bad_array_new_length()@GLIBCXX_3.4.29' echo "#!/bin/sh" > /mnt/scratch/tmp/portage/app-office/libreoffice-6.4.7.2/work/ libreoffic e-6.4.7.2/workdir/CustomTarget/sysui/share/libreoffice/create_tree.sh collect2: error: ld returned 1 exit status --->8 There's also qtwebengine to get through when I have this one fixed. -- Regards, Peter.
[gentoo-user] Re: long compiles
On 11/09/2023 22:19, Alan McKinnon wrote: chromium has been building since 10:14, it's now 21:16 and still going so 9 hours at least on this machine to build a browser - almost as bad as openoffice at it's worst (regularly took 12 hours). Nodejs also took a while, but I didn't record time. What's your CPU and how much RAM? Even on my older system I had (an 4-core i5 2500K) libreoffice took like 2 hours or so to build. What other packages have huge build times? IIRC, dev-qt/qtwebengine is one of the heaviest when it comes to build times. Anyway, a nice way to cut down on build times is to build on tmpfs. To do that however with heavy packages like that, I had to upgrade to 32GB RAM. There was a large price drop in the memory market a couple months ago, so I snatched a 32GB DDR4 3600 kit (2x16GB) for like 80€. So now with plenty of RAM, I configured a 14GB tmpfs in /var/tmp/portage. I never hit swap when emerging.
Re: [gentoo-user] Controlling emerges
On Tue, Sep 19, 2023 at 5:48 AM Peter Humphrey wrote: > > On Tuesday, 19 September 2023 10:14:42 BST William Kenworthy wrote: > > That is where you set per package compiler parameters by overriding > > make.conf settings. > > And which make.conf setting might achieve what I want? Careful reading of the > make.conf man page hasn't revealed anything relevant. > There isn't one. At best there is -l which regulates jobs by system load, but there is nothing that takes into account RAM use. I just use package.env to limit jobs on packages that I know are RAM-hungry. Right now my list includes: calligra qtwebengine qtwebkit ceph nodejs passwdqc scipy pandas spidermonkey (It has been ages since I've pruned the list, and of course what is "too much RAM" will vary.) The other thing I will tweak is avoiding building in a tmpfs. Obviously anything that is RAM constrained is a good contender for not using a tmpfs, but there are also packages that just have really large build directories that otherwise don't need to much RAM when building. -- Rich
Re: [gentoo-user] zoom?
On Wednesday, April 1, 2020 3:12:46 AM CEST William Kenworthy wrote:
> On 25/3/20 8:42 pm, Michael wrote:
> > On Wednesday, 25 March 2020 12:38:35 GMT Jorge Almeida wrote:
> >> On Wed, Mar 25, 2020 at 12:16 PM Alarig Le Lay
wrote:
> >>> On mer. 25 mars 11:51:33 2020, Jorge Almeida wrote:
> >>>> Did someone try to install zoom? (relevant to many people during the
> >>>> current crisis)
> >>>>
> >>>> https://support.zoom.us/hc/en-us/articles/204206269-Installing-Zoom-on->
> >>>> >>>> L
> >>>> inux
> >>>>
> >>>> I downloaded an archive (cannot find the URL again; the site is that
> >>>> bad) and the directory doesn't even contain a REDME...)
> >>>>
> >>>> Jorge Almeida
> >>>
> >>> Why didn’t you try net-im/zoom?
> >>
> >> I don't have that in /usr/portage. emerge--sync'ing now...
> >
> > Have you looked at using WebRTC with Zoom, rather than installing their
> > code on your PC?
>
> Anyone able to install this? I can't get mesa and the libglvnd
> dispatcher to coexist.
>
>
> It does work in chromium, but I think some of the options are not available
>
>
> BillK.
>
>
>
> rattus ~ # emerge net-im/zoom -vp
>
> These are the packages that would be merged, in order:
>
> Calculating dependencies... done!
> [ebuild N ] media-libs/libglvnd-1.3.1::gentoo USE="X -test"
> ABI_X86="(64) -32 (-x32)" 0 KiB
> [ebuild N ] dev-qt/qtwebchannel-5.14.1:5/5.14::gentoo USE="qml
> -debug -test" 198 KiB
> [ebuild N ] dev-qt/qtpositioning-5.14.1:5/5.14::gentoo USE="qml
> -debug -geoclue -test" 5,976 KiB
> [ebuild N ] dev-qt/qtscript-5.14.1:5/5.14::gentoo USE="jit -debug
> -scripttools -test" 2,593 KiB
> [ebuild N ] dev-qt/qtwebengine-5.14.1:5/5.14::gentoo USE="alsa
> pulseaudio system-ffmpeg system-icu widgets -bindist -debug -designer
> -geolocation -jumbo-build -test" 236,757 KiB
> [ebuild N~] net-im/zoom-3.5.374815.0324-r2::gentoo USE="pulseaudio"
> 70,908 KiB
> [blocks B ] media-libs/mesa[-libglvnd(-)]
> ("media-libs/mesa[-libglvnd(-)]" is blocking media-libs/libglvnd-1.3.1)
>
> Total: 6 packages (6 new), Size of downloads: 316,430 KiB
> Conflict: 1 block (1 unsatisfied)
>
> * Error: The above package list contains packages which cannot be
> * installed at the same time on the same system.
>
>(media-libs/mesa-19.3.5:0/0::gentoo, installed) pulled in by
> media-libs/mesa[egl,X(+)] required by
> (dev-qt/qtwebengine-5.14.1:5/5.14::gentoo, ebuild scheduled for merge)
> USE="alsa pulseaudio system-ffmpeg system-icu widgets -bindist -debug
> -designer -geolocation -jumbo-build -test" ABI_X86="(64)"
>
>(media-libs/libglvnd-1.3.1:0/0::gentoo, ebuild scheduled for merge)
> pulled in by
> media-libs/libglvnd required by
> (net-im/zoom-3.5.374815.0324-r2:0/0::gentoo, ebuild scheduled for merge)
> USE="pulseaudio" ABI_X86="(64)"
>
>
> For more information about Blocked Packages, please refer to the following
> section of the Gentoo Linux x86 Handbook (architecture is irrelevant):
>
> https://wiki.gentoo.org/wiki/Handbook:X86/Working/Portage#Blocked_packages
This is because of the "libglvnd" activities for Xorg.
I had a few blockers on my desktop and old laptop as well, but calmly doing a
full update and helping it through is possible.
Masking this flag will not help in the long-term as the flag will disappear
and this will be the norm because it gets rid of the "eselect opengl" hack.
--
Joost
Re: [gentoo-user] Re: Compile large packages as last package [SOLVED]
Thank you for the hint, but I am very aware of this.
As the subject says, I want to compile large packages as last package,
since it sometimes happens, that a large package may get compiled as
package 245 of 300 for example; blocking other small packages.
I just re-used "/etc/portage/package.env/no_tmpfs.conf", since it
already contains a list of large packages, which most likely need a long
time to compile.
I should have been more elaborate in my last e-mail:
I was actually hoping for a neat hack, where you can tag or
prioritise (not nice or renice) packages in "/etc/portage/env/" or
"/etc/portage/package.env/" to compile them as last packages.
The solution with "--exclude" is working well so far; I had no redundant
rebuilds. :)
Maybe I need to tweak it later on, if rebuilds occur frequently, but
this for another time.
-Ramon
On 08/09/2021 17:24, David M. Fellows wrote:
Thank you for your ideas!
I was actually hoping for a neat hack with "/etc/portage/env/" and
"/etc/portage/package.env/", where you can set environment variables.
Did you look at example 2 in
https://wiki.gentoo.org/wiki//etc/portage/package.env ?
It seems to address your problem.
DaveF
I will try out the following solution:
$ < "/etc/portage/package.env/no_tmpfs.conf
# custom - 20181121 - rfischer: list packages, which are too big for
tmpfs
#app-emulation/qemu-kv no_tmpfs.conf
#app-office/libreoffice no_tmpfs.conf
#dev-java/icedtea no_tmpfs.conf
#dev-lang/ghc no_tmpfs.conf
#dev-lang/rust no_tmpfs.conf
#mail-client/thunderbird no_tmpfs.conf
#sci-libs/tensorflow no_tmpfs.conf
#sys-devel/gcc no_tmpfs.conf
#www-client/firefox no_tmpfs.conf
#www-client/ungoogled-chromium no_tmpfs.conf #throttle_make_emerge.conf
$ < "${HOME}/bin/update.sh"
[...]
large_package_list=$(/bin/grep --extended-regexp --only-matching
"[a-z]+-[a-z]+\/[-0-9a-zA-Z]+" "/etc/portage/package.env/no_tmpfs.conf")
[...]
/usr/bin/emerge --ask --update --deep --newuse --tree --verbose
--exclude="${large_package_list//$'\n'/ }" @world
/usr/bin/emerge --ask --update --deep --newuse --tree --verbose @world
-Ramon
On 15/08/2021 17:48, Nikos Chantziaras wrote:
On 14/08/2021 22:20, Ramon Fischer wrote:
Is there any way to tell "portage", that packages like "qtwebengine",
"(ungoogled-)chromium", "firefox" and so on are always compiled as
last package?
The simplest way is to exclude those packages in the first update, and
then allow them in the second:
emerge -uDU @world --exclude "qtwebengine firefox chromium" && emerge
-uDU @world
The dependency tracker of portage will of course also exclude packages
that depend on the excluded packages, unless they themselves have
updates pending. In that case, they *might* get built twice; once
against the current version of the excluded packages, and then perhaps
again on the second run, if there's rebuild triggers involved.
Most of the time though, you won't run into cases of redundant
rebuilds. Rebuild triggers are not very common.
--
GPG public key: 5983 98DA 5F4D A464 38FD CF87 155B E264 13E6 99BF
application/pgp-signature attachment, name=OpenPGP_signature
--
GPG public key: 5983 98DA 5F4D A464 38FD CF87 155B E264 13E6 99BF
OpenPGP_signature
Description: OpenPGP digital signature
Re: [gentoo-user] long compiles
Hi. Nothing compares to Chromium (browser) in terms of compilation times. On my system with 12 core threads it takes about 8 hours to compile - which is 4 times longer than 10 years ago with 2 core threads ;) Libreoffice takes a few hours, but less than half of chromium. Nothing gets close to Chromium. But otherwise webkitgtk and qtwebengine are to big ones - but still about a quarter of Chromium. Kristian Poul Herkild Den 11.09.2023 kl. 21.19 skrev Alan McKinnon: After my long time away from Gentoo, I thought perhaps some packages that always took ages to compile would have improved. I needed to change to ~amd64 anyway (dumb n00b mistake leaving it at amd64). So that's what I did and let emerge do it's thing. chromium has been building since 10:14, it's now 21:16 and still going so 9 hours at least on this machine to build a browser - almost as bad as openoffice at it's worst (regularly took 12 hours). Nodejs also took a while, but I didn't record time. What other packages have huge build times? -- Alan McKinnon alan dot mckinnon at gmail dot com
Re: [gentoo-user] printing pdfs
On Sat, May 30, 2020, at 14:54, james wrote: > # qlist -ICv dev-qt/ > dev-qt/designer-5.14.2 > dev-qt/linguist-5.14.2 > dev-qt/linguist-tools-5.14.2 > dev-qt/qt-creator-4.8.2 > dev-qt/qt-docs-5.14.2_p202003291239 > dev-qt/qt3d-5.14.2 > dev-qt/qtchooser-66 > dev-qt/qtconcurrent-5.14.2 > dev-qt/qtcore-5.14.2 > dev-qt/qtdatavis3d-5.14.2 > dev-qt/qtdbus-5.14.2 > dev-qt/qtdeclarative-5.14.2 > dev-qt/qtdiag-5.14.2 > dev-qt/qtgraphicaleffects-5.14.2 > dev-qt/qtgui-5.14.2 > dev-qt/qthelp-5.14.2 > dev-qt/qtimageformats-5.14.2 > dev-qt/qtlockedfile-2.4.1_p20171024 > dev-qt/qtmultimedia-5.14.2 > dev-qt/qtnetwork-5.14.2 > dev-qt/qtopengl-5.14.2 > dev-qt/qtpositioning-5.14.2 > dev-qt/qtprintsupport-5.14.2 > dev-qt/qtquickcontrols-5.14.1 > dev-qt/qtscript-5.14.2 > dev-qt/qtsingleapplication-2.6.1_p20171024 > dev-qt/qtsql-5.14.2 > dev-qt/qtsvg-5.14.1 > dev-qt/qttest-5.14.2 > dev-qt/qtwayland-5.14.2 > dev-qt/qtwebchannel-5.14.2 > dev-qt/qtwebengine-5.14.2 > dev-qt/qtwebkit-5.212.0_pre20200309-r1 > dev-qt/qtwidgets-5.14.2 > dev-qt/qtx11extras-5.14.2 Most of your Qt packages have the 5.14.2 version, except dev-qt/qtquickcontrols and dev-qt/qtsvg which are still at 5.14.1. Please try upgrading them: # emerge -av1 =dev-qt/qtquickcontrols-5.14.2 =dev-qt/qtsvg-5.14.2 -- https://fturco.net/
Re: [gentoo-user] zoom?
On 2/4/20 12:51 pm, J. Roeleveld wrote:
On Wednesday, April 1, 2020 3:12:46 AM CEST William Kenworthy wrote:
On 25/3/20 8:42 pm, Michael wrote:
On Wednesday, 25 March 2020 12:38:35 GMT Jorge Almeida wrote:
On Wed, Mar 25, 2020 at 12:16 PM Alarig Le Lay
wrote:
On mer. 25 mars 11:51:33 2020, Jorge Almeida wrote:
Did someone try to install zoom? (relevant to many people during the
current crisis)
https://support.zoom.us/hc/en-us/articles/204206269-Installing-Zoom-on-> >>>> L
inux
I downloaded an archive (cannot find the URL again; the site is that
bad) and the directory doesn't even contain a REDME...)
Jorge Almeida
Why didn’t you try net-im/zoom?
I don't have that in /usr/portage. emerge--sync'ing now...
Have you looked at using WebRTC with Zoom, rather than installing their
code on your PC?
Anyone able to install this? I can't get mesa and the libglvnd
dispatcher to coexist.
It does work in chromium, but I think some of the options are not available
BillK.
rattus ~ # emerge net-im/zoom -vp
These are the packages that would be merged, in order:
Calculating dependencies... done!
[ebuild N ] media-libs/libglvnd-1.3.1::gentoo USE="X -test"
ABI_X86="(64) -32 (-x32)" 0 KiB
[ebuild N ] dev-qt/qtwebchannel-5.14.1:5/5.14::gentoo USE="qml
-debug -test" 198 KiB
[ebuild N ] dev-qt/qtpositioning-5.14.1:5/5.14::gentoo USE="qml
-debug -geoclue -test" 5,976 KiB
[ebuild N ] dev-qt/qtscript-5.14.1:5/5.14::gentoo USE="jit -debug
-scripttools -test" 2,593 KiB
[ebuild N ] dev-qt/qtwebengine-5.14.1:5/5.14::gentoo USE="alsa
pulseaudio system-ffmpeg system-icu widgets -bindist -debug -designer
-geolocation -jumbo-build -test" 236,757 KiB
[ebuild N~] net-im/zoom-3.5.374815.0324-r2::gentoo USE="pulseaudio"
70,908 KiB
[blocks B ] media-libs/mesa[-libglvnd(-)]
("media-libs/mesa[-libglvnd(-)]" is blocking media-libs/libglvnd-1.3.1)
Total: 6 packages (6 new), Size of downloads: 316,430 KiB
Conflict: 1 block (1 unsatisfied)
* Error: The above package list contains packages which cannot be
* installed at the same time on the same system.
(media-libs/mesa-19.3.5:0/0::gentoo, installed) pulled in by
media-libs/mesa[egl,X(+)] required by
(dev-qt/qtwebengine-5.14.1:5/5.14::gentoo, ebuild scheduled for merge)
USE="alsa pulseaudio system-ffmpeg system-icu widgets -bindist -debug
-designer -geolocation -jumbo-build -test" ABI_X86="(64)"
(media-libs/libglvnd-1.3.1:0/0::gentoo, ebuild scheduled for merge)
pulled in by
media-libs/libglvnd required by
(net-im/zoom-3.5.374815.0324-r2:0/0::gentoo, ebuild scheduled for merge)
USE="pulseaudio" ABI_X86="(64)"
For more information about Blocked Packages, please refer to the following
section of the Gentoo Linux x86 Handbook (architecture is irrelevant):
https://wiki.gentoo.org/wiki/Handbook:X86/Working/Portage#Blocked_packages
This is because of the "libglvnd" activities for Xorg.
I had a few blockers on my desktop and old laptop as well, but calmly doing a
full update and helping it through is possible.
Masking this flag will not help in the long-term as the flag will disappear
and this will be the norm because it gets rid of the "eselect opengl" hack.
--
Joost
In my case libglvnd-1.3.1 wanted to install ~20 files that would have
replaced ones already installed by Mesa-19.3.5, hence the merge
failure. I already had that Mesa version installed - did you see this?
BillK
Re: [gentoo-user] Re: btop fails to compile
On Monday, 5 December 2022 06:30:22 GMT Walter Dnes wrote: > On Wed, Nov 30, 2022 at 01:50:02PM +0100, Jochen Kirchner wrote > > > this is my make.conf: (its a web - and mail server) > > > > MAKEOPTS="-j17 -l17" > > Ouch!!! How much ram do you have on that machine? The Gentoo install > handbook has a dire warning at... > > https://wiki.gentoo.org/wiki/Handbook:AMD64/Full/Installation#MAKEOPTS > > ...that you need approx 2 gigabytes free ram for each increment in > "MAKEOPTS". > > > * Warning > > Using a large number of jobs can significantly impact memory > > consumption. A good recommendation is to have at least 2 GiB of RAM > > for every job specified (so, e.g. -j6 requires at least 12 GiB). To > > avoid running out of memory, lower the number of jobs to fit the > > available memory. > > If you have a fancy-schmancy "desktop environment" allow another 3 or > 4 Gigs, especially if you're simultaneously running Chrome or > calculating large spreadsheets or processing large documents. For > "MAKEOPTS=-j17" you'll need at least 36-to-40 gigabytes. Well, I've seen that warning, and I run a plasma desktop, but I set -j48 -- jobs=48 --load-average=72 without noticeable swapping in this 64GB, 24-core Ryzen 9 5900X machine. The one exception is qtwebengine, which wants 48GB for itself. -- Regards, Peter.
Re: [gentoo-user] Jobs and load-average
On Thu, 16 Feb 2023 09:53:30 + Peter Humphrey wrote: > On Wednesday, 15 February 2023 13:18:24 GMT Rich Freeman wrote: > > > First, keep in mind that --jobs=16 + -j16 can result in up to 256 > > (16*16) tasks running at once. Of course, that is worst case and most > > of the time you'll have way less than that. > > Yes, I was aware of that, but why didn't --load-average=32 take precedence? This only means that emerge would not schedule additional package job (where a package job means something like `emerge gcc`) when load average > 32, howwever if a job is scheduled it's running, independently of the current load. While having it in MAKEOPTS, it would be handled by the make system, which schedules single build jobs, and would stop scheduling additional jobs, when the load is too high. Extreme case: emerge chromium firefox qtwebengine --> your load when you do this is pretty much close to 0, i.e. all 3 packages are being merged simultaneously and each will be built with -j16. I.e. for a long time you will have about 3*16=48 single build jobs running in parallel, i.e. you should see a load going towards 48, when you do not have anything in your MAKEOPTS. Cheers Andreas
[gentoo-user] Re: google-chrome can render pages after update
On 2023-06-12, Michael wrote: >> It seems to be a variation on this bug which affects only AMD GPUs: >> >> https://bugs.gentoo.org/907431 >> >> Clearing the GPU driver cache or using the >> -disable-gpu-driver-bug-workarounds option avoids the problem. >> >> In my case, It wasn't a mesa update that triggered the problem. I >> think it was the llvm update (I haven't confirmed that). > > Did you (re)compile anything graphics related using llvm, which > might be used by the Chrome binary? No -- but as I understand it, mesa uses llvm (at runtime) to generate GPU object code. Based on the work-around, it looks like compiled GPU object code is cached by Chrome/Chromium, and updates to mesa and/or llvm can result attempts to use old, incompatible GPU object code. As pages are rendered, there was a constant stream of "link failure" messages on the console window where Chrome is running. > I don't have chrome/chromium installed here to directly compare > notes and so far qtwebengine appears to work fine after updating > llvm this morning, as does www-client/microsoft-edge. Firefix-bin still worked fine also. It only seemed to affect Chrome/Chromium or it's derivitives. -- Grant
[gentoo-user] Re: gst-plugins-bad-1.11.90 is blocking gst-plugins-base-1.12.3
Am Tue, 17 Oct 2017 16:35:27 +0200
schrieb Hubert Hauser :
> I've got error:
>
> tux ~ # emerge @preserved-rebuild
>
> * IMPORTANT: 8 news items need reading for repository 'gentoo'.
> * Use eselect news read to view new items.
>
> Calculating dependencies... done!
> [ebuild R ] dev-libs/botan-1.10.17
> [ebuild U ] media-gfx/imagemagick-7.0.7.6 [6.9.9.0]
> [ebuild R ] dev-python/pillow-4.2.1-r1
> [ebuild U ] media-libs/gst-plugins-base-1.12.3 [1.10.5]
> [ebuild U ] dev-qt/qtgui-5.9.2 [5.7.1-r1] USE="libinput* -vnc%"
> [ebuild R ] media-video/ffmpeg-3.3.4
> [ebuild U ] dev-qt/qtwidgets-5.9.2 [5.7.1] USE="gtk%*"
> [ebuild U ] media-libs/gst-plugins-ugly-1.12.3 [1.10.5]
> [ebuild U ] dev-qt/qtdeclarative-5.9.2 [5.7.1]
> [ebuild U ] dev-qt/qtprintsupport-5.9.2 [5.7.1]
> [ebuild U ] media-video/vlc-2.2.6-r2 [2.2.6] USE="qt5*"
> [ebuild U ] media-plugins/gst-plugins-x264-1.12.3 [1.10.5]
> [ebuild U ] dev-qt/qtwebchannel-5.9.2 [5.7.1]
> [ebuild R ] net-analyzer/wireshark-2.4.2
> [ebuild U ] xfce-base/xfwm4-4.13.0-r1 [4.12.3-r1] USE="opengl%*
> -xpresent%"
> [ebuild U ] app-editors/vim-8.0.1188 [8.0.0386] USE="terminal%*"
> PYTHON_SINGLE_TARGET="python3_4%* -python2_7% -python3_5% -python3_6%"
> [ebuild U ] dev-qt/qtwebengine-5.9.2 [5.7.1-r2]
> [ebuild R ] dev-db/postgresql-9.6.5-r1
> [ebuild U ] media-video/obs-studio-20.0.1-r1 [20.0.1]
> [ebuild U ] net-print/cups-filters-1.17.9 [1.16.4] USE="-pclm%"
> [ebuild U ] media-gfx/graphviz-2.40.1 [2.38.0-r1]
> [ebuild R ] www-servers/nginx-1.12.1
> [ebuild R ] dev-lang/php-7.0.24
> [ebuild R ] dev-db/mariadb-10.2.9
> [blocks B ] (" media-libs/gst-plugins-base-1.12.3)
>
> !!! Multiple package instances within a single package slot have been
> pulled !!! into the dependency graph, resulting in a slot conflict:
>
> dev-qt/qtgui:5
>
> (dev-qt/qtgui-5.7.1-r1:5/5.7::gentoo, installed) pulled in by
> ~dev-qt/qtgui-5.7.1 required by
> (dev-qt/qtquickcontrols-5.7.1:5/5.7::gentoo, installed)
> ^
> ^
>
>
> (and 9 more with the same problem)
>
> (dev-qt/qtgui-5.9.2:5/5.9::gentoo, ebuild scheduled for merge)
> pulled in by
> ~dev-qt/qtgui-5.9.2 required by
> (dev-qt/qtwebengine-5.9.2:5/5.9::gentoo, ebuild scheduled for merge)
> ^
> ^
>
>
> (and 3 more with the same problem)
>
> dev-qt/qtwidgets:5
>
> (dev-qt/qtwidgets-5.7.1:5/5.7::gentoo, installed) pulled in by
> ~dev-qt/qtwidgets-5.7.1 required by
> (dev-qt/qtwebkit-5.7.1:5/5.7::gentoo, installed)
> ^
> ^
>
>
> (and 6 more with the same problem)
>
> (dev-qt/qtwidgets-5.9.2:5/5.9::gentoo, ebuild scheduled for merge)
> pulled in by
> ~dev-qt/qtwidgets-5.9.2 required by
> (dev-qt/qtwebengine-5.9.2:5/5.9::gentoo, ebuild scheduled for merge)
> ^
> ^
>
>
> (and 2 more with the same problem)
>
> dev-qt/qtprintsupport:5
>
> (dev-qt/qtprintsupport-5.7.1:5/5.7::gentoo, installed) pulled in by
> ~dev-qt/qtprintsupport-5.7.1 required by
> (dev-qt/qtwebkit-5.7.1:5/5.7::gentoo, installed)
> ^
> ^
>
>
>
> (dev-qt/qtprintsupport-5.9.2:5/5.9::gentoo, ebuild scheduled for
> merge) pulled in by
> ~dev-qt/qtprintsupport-5.9.2 required by
> (dev-qt/qtwebengine-5.9.2:5/5.9::gentoo, ebuild scheduled for merge)
> ^
> ^
>
>
>
> dev-qt/qtdeclarative:5
>
> (dev-qt/qtdeclarative-5.7.1:5/5.7::gentoo, installed) pulled in by
> ~dev-qt/qtdeclarative-5.7.1 required by
> (dev-qt/qtquickcontrols-
Re: [gentoo-user] Not enough RAM for dev-qt/qtwebengine build
On Thursday, 13 June 2019 12:59:41 BST Adam Carter wrote: > > You can use more swap (files) before buying more RAM. > > I have been doing this too. It only get used during the big builds. > > To create a 32G swap file and enable it (OP can do this now as the build > runs, to keep OOM away) > # fallocate -l 32G && chmod 600 && mkswap > && swapon > > use 'watch free -m' to see if its being used. Then after the build is over, > delete it/add it to fstab/whatever More swap will help, but the basic problem here is the HUGE number of jobs specified. If each job eats say up to 2G when it gets going, then -j32 will require >64G RAM to keep running without thrashing the swap device, or running OOM. The reasonable thing for big builds is to add an environment variable as already explained, which restricts the number of jobs. Or for an one off run something like: MAKEOPTS="-j8 -l7.8" emerge -1aNDv On a i7 (4 cores x 2 threads) I use -j2 for huge compiles like chromium, because the PC only has 8G RAM. Smaller packages will emerge with -j12 without breaking into a sweat, although the optimum setting for this PC is "- j8 -l7.8". -- Regards, Mick signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] per package parallel build
Raffaele BELARDI wrote: > > 8-core CPU: > > EMERGE_DEFAULT_OPTS="--quiet-build --keep-going --jobs 9 --load-average 9" > > MAKEOPTS="-j9 -l9" > > > > Works fine except when both Firefox and Thunderbird need update, in > that case emerge typically tries to build them in parallel and one > gets OOM killed due to insufficient swap space (1G swap, 16G RAM). I > will increase the swap but I’d like to know: > > Is there a way to tell emerge to normally run 9 parallel jobs but > limit to 1 when it is building one of the two monsters? > > > > Thanks, > > > > raffaele > This may help, along with Adams info. https://wiki.gentoo.org/wiki//etc/portage/package.env This is a bit of mine that may also help. I hope you can follow what that means. The one for single may interest you the most. root@fireball / # cat /etc/portage/package.env/package.env #www-client/seamonkey ../env/single.conf #www-client/firefox ../env/single.conf #www-client/firefox ../env/notmpfs.conf #www-client/seamonkey ../env/notmpfs.conf app-office/libreoffice ../env/notmpfs.conf #sys-devel/gcc ../env/notmpfs.conf dev-qt/qtwebengine ../env/notmpfs.conf #dev-qt/qtwebkit ../env/notmpfs.conf #sci-electronics/kicad ../env/notmpfs.conf root@fireball / # cat /etc/portage/env/notmpfs.conf PORTAGE_TMPDIR="/var/tmp/notmpfs" root@fireball / # cat /etc/portage/env/single.conf #EMERGE_DEFAULT_OPTS="-j1" root@fireball / # Note, some of mine are commented out so don't duplicate that. I upgraded from 16GBs to 32GBs and no longer need some exceptions. Dale :-) :-)
Re: [gentoo-user] Removing or stopping binary emerges from being in emerge.log
Dale schrieb am 06.03.22 um 06:53: I have a chroot environment that I do updates in. Once the updates are done, I copy the binaries and distfiles over to my running system and use the -k option to update everything in my real system. It comes in real handy when libreoffice, Firefox, qtwebengine and other large time consuming packages are being updated. The bad thing is, I have the full length of build time in the chroot but the binary install on my running system. Is there a way to either stop it from logging binary updates or removing them after it is done? I'd rather it not keep those times in either place really. I can't find a emerge option. It seems to record everything regardless. My reason for this, the binary install times throws off genlop -c and its estimates. Anybody have ideas? There is a long-standing bug [1] regrading this issue but given genlop currently is not actively developed I don't think there will be a solution soon. It should be possible to exclude binary merges as they can be identified in emerge.log which is read by genlop to generate the output. Also I don't think there is an option in portage to not log binary merges. [1] https://bugs.gentoo.org/120899 -- Best Daniel
Re: [gentoo-user] Would a Thinkpad X200 be too much trouble too run gentoo on?
On Thursday, 21 April 2022 17:00:04 BST Michael Orlitzky wrote: > On Thu, 2022-04-21 at 08:24 -0700, cal wrote: > > Do you have any other (more powerful) machines at home that you could > > set up as a distcc cluster? > > My desktop is only slightly more powerful. I don't really mind the > webkit-gtk build time since it's shared between epiphany and evolution. > I just run that (and/or gcc) overnight when I need to. I have an old Acer laptop with a Core 2 Duo P7550 @2.26GHz CPU and only 4G RAM. It will compile everything, even rust and qtwebengine, although it may take more than a day to achieve this. Hence I use a more modern PC to build binaries which I then transfer and emerge on the laptop in minutes. > > In addition to the usual problem packages others have called out, the > > main problem I ran into was heat dissipation > > I leave it on top of a giant fan or near an open window, weather > permitting. Or, use some blocks/books to suspend it off the desk. A couple of inches may be enough. Alternatively, you can buy a cooling pad. Some come with USB powered fans too, which help drop the CPU temperature by another couple of degrees. signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] Highlight certain packages being upgraded
On Sat, Jul 8, 2023 at 4:33 AM Dale wrote: > Howdy, > > I was wondering. Is there a way to highlight certain packages that are > about to be upgraded? Example, I like to know when some larger packages > like Firefox, LOo, that excessively long qt package and a couple others > are going to be upgraded. Some that are listed in the world file show > up in a darker green and are easier to see however, some are not. They > are dependencies of another package but I'd like to have them stick out > in the list of packages to be upgraded. I don't recall ever seeing > anyone mention this as a feature of emerge or heard of a way to > configure such a thing either. That said, it could be possible and just > not well known. > > I'd like to be able to have those packages show up as red or something > like that. If it is possible. > > Thoughts? Ever heard of such a thing? > > Dale > > :-) :-) How about first making a list of all files needing to be rebuilt with something like emerge -pDuNv @world > builds Then make a file “long” containing names of demanding builds like firefox and qtwebengine, one to a line. Finally, execute grep -f long builds which will print the names of those long builds if they’re due for rebuilding or upgrading. HTH John Blinka >
Re: [gentoo-user] long compiles
On Mon, Sep 11, 2023 at 10:05 PM Neil Bothwick wrote: > On Mon, 11 Sep 2023 21:19:27 +0200, Alan McKinnon wrote: > > > chromium has been building since 10:14, it's now 21:16 and still going > > so 9 hours at least on this machine to build a browser - almost as bad > > as openoffice at it's worst (regularly took 12 hours). Nodejs also took > > a while, but I didn't record time. > > Chromium is definitely the worst, and strangely variable. The last few > compiles have taken between 6 and 14 hours. Since it takes longer than > everything else to build, it is usually compiling on its own, so parallel > emerges aren't a factor. > > Qtwebengine is also bad, not surprising as it is a cut down Chromium. > Emerging world with --exclude then timing build to coincide with sleep > helps, although I haven't quite reached the age where I need 14 hours of > sleep a day. > > > -- > Neil Bothwick > > If it isn't broken, I can fix it. > Yup, that jibes with what I see. Oh well, just means that the need for overnight compiles did not go away haha Thanks to every one else that replied too - everyone said much the same thing so I figured one replay to rule them all was the best way Alan -- Alan McKinnon alan dot mckinnon at gmail dot com
Re: [gentoo-user] Memory manager
On Sun, 20 Oct 2019 18:01:01 +0100, Mick wrote: > Now, in a gentoo scenario, say a mammoth compile like Chromium, with a > large count of jobs specified for it, you could end up swapping part or > all of one or more jobs into memory, only to swap it out again in order > to process it. The compile keeps swapping in and out a job at a time in > order to carry on compiling. The disk thrashing is now continuous and > indeed interacting with your desktop will be painful - potentially > waiting for minutes at a time before an application responds. The way > out of this bottleneck is to either increase your RAM, or minimise the > use of memory by reducing the job count in MAKEOPTS. Shutting down > desktop applications and login out of any desktop sessions to release > RAM will also help. > > On a laptop with 4G RAM compiling Chromium is quite challenging when > even a single gcc job could grow to 3G or more. Swapping and a disk > I/O bottleneck becomes unavoidable and moving the compile of binaries > to a bigger PC becomes a rather wise solution. That's why I have Chromium, as well and LO and qtwebengine, set to use my SSD for PORTAGE_TMPDIR on this laptop, which is limited to 8GB. MAKEOPTS is also constricted for Chromium. As a result, the packages build more quickly with minimal impact on using the system. -- Neil Bothwick Top Oxymorons Number 3: Working vacation pgpSZT2Se7d0N.pgp Description: OpenPGP digital signature
Re: [gentoo-user] Google Mail still not recognising Gentoo KMail
On Thursday, 7 November 2019 10:47:19 GMT Антон Кулешов wrote: > Hello, > > Create 'Application-specific password required' in your google account and > use it instead of your google password. Try this link > myaccount.google.com/apppasswords A couple of months ago Kmail stopped working with OAuth2. I assumed Gmail changed their OAuth2 implementation and Kmail hadn't caught up with the changes. The way I managed this change so far, has been to set my Google Account security settings to allow "less secure apps": https://support.google.com/accounts/answer/6010255?hl=en Then change the Kmail/Settings/Receiving/Advanced/Authentication from "Gmail" to "Plain". The other relevant settings are "SSL/TLS" and port "993". When selecting "Gmail" authentication, it used to be the case qtwebengine would launch a GUI in which you would login into the Gmail security web GUI to set up an application specific password. I don't know if this works now, it didn't last time I tried it, a few months ago. I suspect it won't work if you have "less secure app access" enabled. Anyway, I better send this message before I start tweaking my Google security settings and end up locking myself out of my account! LOL! PS. TBH I'm finding all this Google omniscience troublesome and their security settings tiresome. Perhaps it's time I took my email data elsewhere. -- Regards, Mick signature.asc Description: This is a digitally signed message part.
[gentoo-user] Load average revisited
Afternoon all, It seems that portage's self-protection from overload is incomplete - or so I've been led to believe by some pretty odd goings-on. Injudicious fiddling with -j, --jobs and --load-average can easily cause miscompilation of packages without causing an OOM or any other sign of problems. For some time, I had those values set to take advantage of the 24 threads and 64GB of RAM in this machine (not to mention 8 + 50GB swap, little of which appears ever to be used), resulting in the load average rising into three digits at times - over 200, even. But no sign of difficulty was shown and all appeared to have gone to plan. Until I ran the system, when odd errors would show. One example I remember is Firefox having a bizarre colour scheme (the window frame, not the page display) and missing the three upper-right buttons. That wasn't corrected by recompiling Firefox, so I assume the problem was lower down. I think I have some safe values now (time will tell), but it's worrying that compilation errors can go undetected. Of course I don't know where to start looking for the problem; I just hope someone else does. Meanwhile, is it possible to set things up so that, say, qtwebengine is never compiled at the same time as anything else? I don't want to rely on my noticing and intervening, and besides, it isn't always possible just to --exclude it. -- Regards, Peter.
Re: [gentoo-user] long compiles
On 11/09/2023 20:46, Alan McKinnon wrote: qtwebengine! yes that one took forever also. It also said my 16G of RAM was smaller than the 16G it needed. Weird. Anyways I enabled a swapfile and left it to run overnight 16GB physical ram <> 16GB usable ram for the compile ... I concur with others that tmpfs is the way to go - I don't think my system is set up that way just now, but I always have googols of swap (twice max physical ram per disk) so I just declare a huge ramdisk for compiling on. My current system has four ram slots, two maxed out with 16GB chips each, so that makes 128GB swap partitions per disk (4 of them) equals 512GB swap ... (Yes the people at openSUSE said I was being stupid with that much swap) But declare a 128GB ramdisk, and it'll spill over as required, but anything that fits in ram will compile very quick. And yes, I also used to have my systems configured so they had one shared portage area, matching make.conf, and were set up to "use binary if it exists, else build one". I had the opposite problem though, my nice fast system had a habit of crashing, so I used the old slow one to build most things, because it was more reliable. Hey ho. There's all sorts of tricks, some work for some people, others work for others. Cheers, Wol
Re: [gentoo-user] Re: Netflix Error Code O7355 with Opera
On Thursday, 11 June 2020 08:57:45 BST Neil Bothwick wrote: > On Thu, 11 Jun 2020 03:19:43 -0400, Walter Dnes wrote: > > > I don't watch Netflix on my laptop, but I've just tried it in Chromium > > > and it seems to be working fine. I suspect it's either the widevone or > > > proprietary-codecs USE flag. > > > > > I have Chrome installed. I looked at Chromium. It wants even more > > > > stuff on top of what Chrome has pulled in! No way. > > Unless you're using it as your main browser, it's not worth even > considering the extended build times. I do use it as my main browser, so > I use chromium and firefox-bin. > > Incidentally, I found an a reliable way of killing Firefox on this > laptop, run a chromium build in the background :) TBH the way Chromium has been bloating it would kill pretty much anything alive on a PC. With 16G RAM I had to add swap, reduce number of jobs, move / var/tmp/portage to a disk and various other resource gymnastics to stop users cursing at me for "doing things on their workstation and slowing it down"! I figured since qtwebengine uses the same rendering engine and I spend enough time compiling that package anyway, because KDE won't do without it, I might as well ditch Chromium. I haven't looked back. ;-) Mind you I don't play/watch/pay Netflix anyway. signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] Re: Netflix Error Code O7355 with Opera
On Thu, 11 Jun 2020 13:45:53 +0100, Michael wrote: > > Incidentally, I found an a reliable way of killing Firefox on this > > laptop, run a chromium build in the background :) > > TBH the way Chromium has been bloating it would kill pretty much > anything alive on a PC. With 16G RAM I had to add swap, reduce number > of jobs, move / var/tmp/portage to a disk and various other resource > gymnastics to stop users cursing at me for "doing things on their > workstation and slowing it down"! It's worse with 8GB, but manageable with package.env. It does cause slowdowns and pauses from time to time, but they were short-lived... except with Firefox. Once Firefox got hit by a lack of resources, it wouldn't recover and stayed unresponsive until I shut t down and restarted it. I recently started using distcc, specifically for Chromium but I now use it for a few other packages too. This laptop is now happily building Chromium and it was only this discussion that reminded me that it was doing so :) > I figured since qtwebengine uses the same rendering engine and I spend > enough time compiling that package anyway, because KDE won't do without > it, I might as well ditch Chromium. I haven't looked back. ;-) Life would be easier if the rendering engine was separated from the browser so it only had to be built once. -- Neil Bothwick A TRUE Klingon warrior does not comment his code! pgp4GNx5wP6Eq.pgp Description: OpenPGP digital signature
