View previous topic :: View next topic |
Author |
Message |
hououin n00b
Joined: 24 Jun 2018 Posts: 7
|
Posted: Sun Jun 24, 2018 7:42 pm Post subject: |
|
|
dantrell wrote: | hououin wrote: | Really? In my environment build file it was:
CMAKE_BUILD_TYPE="Gentoo" |
This led me to check the cmake-utils.eclass and I found that it's indeed being set to Gentoo there.
Further down, I found this so I guess there are changes on my end to make.
Thanks for the hint. |
Happy to help!!
Now I'm running into a new problem though
dev-libs/gjs keeps failing. Still trying to figure out exactly why.
Here is the log: https://pastebin.com/0QrWY46i |
|
Back to top |
|
|
Shibotto Apprentice
Joined: 19 Jun 2015 Posts: 155 Location: CET/CEST
|
Posted: Sun Jun 24, 2018 8:49 pm Post subject: |
|
|
dantrell wrote: | Flisk wrote: | Yeah, "Suspend when laptop lid is closed" is set to on. |
I'm out of guesses then since I don't have a laptop to test this with.
What I can say though is that I haven't patched GNOME Settings Daemon in relation to this so the problem might be elogind (or maybe just the elogind version used). If you have time, you can probably rule this out by booting up GNOME with systemd using a Fedora live image. |
Sorry to always catch up late, but I can reproduce this too. I know it showed up with a certain version of elogind, which I blacklisted, but I can't check which one right now because I don't have my laptop with me. Strange enough, but maybe somewhat useful, this problem didn't show up on Void with a recent elogind.
I'll update you tomorrow (according to my timezone in about 11 hours). |
|
Back to top |
|
|
Juippisi Developer
Joined: 30 Sep 2005 Posts: 722 Location: /home
|
Posted: Mon Jun 25, 2018 6:37 am Post subject: |
|
|
It was silent for a long time ;)
now I ran into problem emerging dev-libs/appstream-glib-0.7.9:
Code: |
>>> Emerging (1 of 1) dev-libs/appstream-glib-0.7.9::dantrell-gnome
>>> Failed to emerge dev-libs/appstream-glib-0.7.9, Log file:
>>> '/tmp/portage/dev-libs/appstream-glib-0.7.9/temp/build.log'
>>> Jobs: 0 of 1 complete, 1 failed Load avg: 0.22, 0.25, 0.10
* Package: dev-libs/appstream-glib-0.7.9
* Repository: dantrell-gnome
* Maintainer: gnome@gentoo.org
* USE: abi_x86_64 amd64 elibc_glibc introspection kernel_linux userland_GNU
* FEATURES: network-sandbox preserve-libs sandbox userpriv usersandbox
>>> Unpacking source...
>>> Unpacking appstream-glib-0.7.9.tar.xz to /tmp/portage/dev-libs/appstream-glib-0.7.9/work
>>> Source unpacked in /tmp/portage/dev-libs/appstream-glib-0.7.9/work
>>> Preparing source in /tmp/portage/dev-libs/appstream-glib-0.7.9/work/appstream-glib-0.7.9 ...
* Disabling deprecation warnings ...
[ ok ]
>>> Source prepared.
>>> Configuring source in /tmp/portage/dev-libs/appstream-glib-0.7.9/work/appstream-glib-0.7.9 ...
grep: ./configure: No such file or directory
grep: ./configure.*: No such file or directory
grep: ./configure: No such file or directory
grep: ./configure: No such file or directory
grep: ./configure: No such file or directory
grep: ./configure: No such file or directory
grep: ./configure: No such file or directory
* ERROR: dev-libs/appstream-glib-0.7.9::dantrell-gnome failed (configure phase):
* no configure script found
*
* Call stack:
* ebuild.sh, line 119: Called src_configure
* environment, line 2713: Called gnome2_src_configure '--enable-builder' '--enable-firmware' '--enable-fonts' '--disable-rpm' '--disable-static' '--enable-dep11' '--enable-man' '--disable-nls' '--enable-introspection' '--disable-stemmer' '--with-bashcompletiondir=/usr/share/bash-completion/completions'
* environment, line 1883: Called econf '--enable-builder' '--enable-firmware' '--enable-fonts' '--disable-rpm' '--disable-static' '--enable-dep11' '--enable-man' '--disable-nls' '--enable-introspection' '--disable-stemmer' '--with-bashcompletiondir=/usr/share/bash-completion/completions'
* phase-helpers.sh, line 724: Called die
* The specific snippet of code:
* die "no configure script found"
*
|
In the ebuild it calls for gnome2_configure but when I extract it, the source looks to require meson to build it. |
|
Back to top |
|
|
Spargeltarzan Guru
Joined: 23 Jul 2017 Posts: 317
|
Posted: Mon Jun 25, 2018 9:28 am Post subject: |
|
|
CapSel,
I am facing the same issue with gnome-keyring after I login to my session in GDM. Every time I need to enter my password a second time for gnome-keyring - did you find already something in the meantime how to fix? (also Gnome 3.28 - with all former release I didn't have any issues) _________________ ___________________
Regards
Spargeltarzan
Notebook: Lenovo YOGA 900-13ISK: Gentoo stable amd64, GNOME systemd, KVM/QEMU
Desktop-PC: Intel Core i7-4770K, 8GB Ram, AMD Radeon R9 280X, ZFS Storage, GNOME openrc, Dantrell, Xen |
|
Back to top |
|
|
hououin n00b
Joined: 24 Jun 2018 Posts: 7
|
Posted: Mon Jun 25, 2018 2:29 pm Post subject: |
|
|
hououin wrote: | dantrell wrote: | hououin wrote: | Really? In my environment build file it was:
CMAKE_BUILD_TYPE="Gentoo" |
This led me to check the cmake-utils.eclass and I found that it's indeed being set to Gentoo there.
Further down, I found this so I guess there are changes on my end to make.
Thanks for the hint. |
Happy to help!!
Now I'm running into a new problem though
dev-libs/gjs keeps failing. Still trying to figure out exactly why.
Here is the log: https://pastebin.com/0QrWY46i |
No idea why this worked but when I emerged gjs this way I didn't run into compile time errors.
CFLAGS="-O2 -march=native" && CXXFLAGS=$CFLAGS emerge -v1 gjs
Going from here I got gnome 3.28 working!
I have a question about sound now. Does pulseaudio require consolekit to work? From what I understand consolekit is not available in 3.28. I get this in syslog.
Code: |
[pulseaudio] module-alsa-card.c: Failed to find a working profile.
[pulseaudio] module.c: Failed to load module "module-alsa-card" (argument: "device_id="1" name="pci-0000_01_00.1" card_name="alsa_card.pci-0000_01_00.1" namereg_fail=false tsched=yes fixed_latency_range=no ignore_dB=no rred_volume=yes use_ucm=yes card_properties="module-udev-detect.discovered=1""): initialization failed. |
|
|
Back to top |
|
|
Juippisi Developer
Joined: 30 Sep 2005 Posts: 722 Location: /home
|
Posted: Tue Jun 26, 2018 6:21 am Post subject: |
|
|
hououin wrote: | Does pulseaudio require consolekit to work? |
No it doesnt. Maybe something wrong with your pulseaudio config, or kernel settings? |
|
Back to top |
|
|
dantrell l33t
Joined: 01 Jun 2007 Posts: 915 Location: Earth
|
Posted: Tue Jun 26, 2018 7:10 am Post subject: |
|
|
Shibotto wrote: | Strange enough, but maybe somewhat useful, this problem didn't show up on Void with a recent elogind. |
I guess I'll be checking Void Linux to see if they are doing anything special.
Juippisi wrote: | It was silent for a long time |
We can always count on Meson to mess things up.
Juippisi wrote: | now I ran into problem emerging dev-libs/appstream-glib-0.7.9 |
Fixed.
hououin wrote: | No idea why this worked but when I emerged gjs this way I didn't run into compile time errors.
CFLAGS="-O2 -march=native" && CXXFLAGS=$CFLAGS emerge -v1 gjs |
Does your make.conf have sane CFLAGS?
hououin wrote: | Going from here I got gnome 3.28 working! |
Nice.
Juippisi wrote: | hououin wrote: | Does pulseaudio require consolekit to work? |
No it doesnt. Maybe something wrong with your pulseaudio config, or kernel settings? |
Yeah, I'm almost 100% sure PulseAudio can run out of the box without ConsoleKit or systemd. _________________ Dantrell B. |
|
Back to top |
|
|
Shibotto Apprentice
Joined: 19 Jun 2015 Posts: 155 Location: CET/CEST
|
Posted: Tue Jun 26, 2018 7:27 am Post subject: |
|
|
dantrell wrote: | Shibotto wrote: | Strange enough, but maybe somewhat useful, this problem didn't show up on Void with a recent elogind. |
I guess I'll be checking Void Linux to see if they are doing anything special. |
sys-auth/elogind-233.7-r1 is the last one behaving correctly on my Gentoo system. |
|
Back to top |
|
|
simonvanderveldt Apprentice
Joined: 26 Jan 2016 Posts: 151
|
Posted: Tue Jun 26, 2018 7:43 am Post subject: |
|
|
Small question: Is there anything that needs to be done in the profiles to get the new default of Python 3.6?
It has already been applied to the gentoo tree (https://github.com/gentoo/gentoo/commit/c533252f86bbbf8277892e504a133cdd50b687bd), not sure if there's something needed to make it work with the gnome without systemd profiles, but my emerge --info says I'm still using a default of 3.5. |
|
Back to top |
|
|
Flisk n00b
Joined: 10 Jun 2018 Posts: 6
|
Posted: Tue Jun 26, 2018 9:37 am Post subject: |
|
|
Shibotto wrote: | sys-auth/elogind-233.7-r1 is the last one behaving correctly on my Gentoo system. |
Same here, closing the lid with elogind 233.7-r1 installed suspends my laptop as expected. |
|
Back to top |
|
|
dantrell l33t
Joined: 01 Jun 2007 Posts: 915 Location: Earth
|
Posted: Tue Jun 26, 2018 3:29 pm Post subject: |
|
|
simonvanderveldt wrote: | Small question: Is there anything that needs to be done in the profiles to get the new default of Python 3.6?
It has already been applied to the gentoo tree (https://github.com/gentoo/gentoo/commit/c533252f86bbbf8277892e504a133cdd50b687bd), not sure if there's something needed to make it work with the gnome without systemd profiles, but my emerge --info says I'm still using a default of 3.5. |
The move to Python 3.6 was reverted in base which we inherit from.
For details see Gentoo bug #656406.
Flisk wrote: | Shibotto wrote: | sys-auth/elogind-233.7-r1 is the last one behaving correctly on my Gentoo system. |
Same here, closing the lid with elogind 233.7-r1 installed suspends my laptop as expected. |
This looks like elogind issue #51 and also explains why Void Linux doesn't have this problem now. _________________ Dantrell B. |
|
Back to top |
|
|
Spargeltarzan Guru
Joined: 23 Jul 2017 Posts: 317
|
Posted: Tue Jun 26, 2018 4:58 pm Post subject: |
|
|
Do you think the newer version of elogind might also be responsible for the non-working auto-unlocking of gnome-keyring after logging in in gdm? _________________ ___________________
Regards
Spargeltarzan
Notebook: Lenovo YOGA 900-13ISK: Gentoo stable amd64, GNOME systemd, KVM/QEMU
Desktop-PC: Intel Core i7-4770K, 8GB Ram, AMD Radeon R9 280X, ZFS Storage, GNOME openrc, Dantrell, Xen |
|
Back to top |
|
|
hououin n00b
Joined: 24 Jun 2018 Posts: 7
|
Posted: Tue Jun 26, 2018 5:23 pm Post subject: |
|
|
dantrell wrote: |
Yeah, I'm almost 100% sure PulseAudio can run out of the box without ConsoleKit or systemd. |
Yes I found out that I had forgotten to mount /boot after building kernel with necessary changes. Thanks! |
|
Back to top |
|
|
simonvanderveldt Apprentice
Joined: 26 Jan 2016 Posts: 151
|
Posted: Wed Jun 27, 2018 7:31 am Post subject: |
|
|
dantrell wrote: | The move to Python 3.6 was reverted in base which we inherit from.
For details see Gentoo bug #656406. |
Ohw, I didn't know that, thanks for the pointer and sorry for the noise. We'll see how quickly it gets sorted. |
|
Back to top |
|
|
hououin n00b
Joined: 24 Jun 2018 Posts: 7
|
Posted: Sat Jun 30, 2018 11:53 pm Post subject: |
|
|
Noticed that when I log into my Google account using Gnome Online Accounts I get this after I enter my username and it freezes up:
Code: | Jun 30 18:51:10 okabe kernel: [ 6361.607682] WebKitWebProces[12881]: segfault at bbadbeef ip 00007fc97a5144cb sp 00007ffed44ae9c0 error 6 in libjavascriptcoregtk-4.0.so.18.7.11[7fc977a72000+3a2f000] |
|
|
Back to top |
|
|
cono n00b
Joined: 15 Sep 2016 Posts: 8
|
Posted: Tue Jul 03, 2018 3:31 pm Post subject: |
|
|
hououin wrote: | Noticed that when I log into my Google account using Gnome Online Accounts I get this after I enter my username and it freezes up:
Code: | Jun 30 18:51:10 okabe kernel: [ 6361.607682] WebKitWebProces[12881]: segfault at bbadbeef ip 00007fc97a5144cb sp 00007ffed44ae9c0 error 6 in libjavascriptcoregtk-4.0.so.18.7.11[7fc977a72000+3a2f000] |
|
With recent update also getting tonz of these errors when launching my evolution-owa and trying to navigate via emails.
Using this profile:
dantrell-gnome-3-20:default/linux/amd64/17.0/desktop/gnome/3.20/extended |
|
Back to top |
|
|
dantrell l33t
Joined: 01 Jun 2007 Posts: 915 Location: Earth
|
Posted: Wed Jul 04, 2018 9:11 am Post subject: |
|
|
Flisk wrote: | dantrell wrote: | Another solution is a patch like this one which allows suspending despite that block. If you can test it and confirm that it works, I'll add it as a Quality of Life patch on my end. |
I built gnome-settings-daemon with this patch and enabled /org/gnome/settings-daemon/plugins/power/lid-close-suspend-with-external-monitor both in my regular session and in the gdm session. |
Since this was tested, I added it across the board on my end.
If anyone who applied that patch experiences a patch failure on a gnome-base/gnome-settings-daemon update, this is why.
dantrell wrote: | Flisk wrote: | Shibotto wrote: | sys-auth/elogind-233.7-r1 is the last one behaving correctly on my Gentoo system. |
Same here, closing the lid with elogind 233.7-r1 installed suspends my laptop as expected. |
This looks like elogind issue #51 and also explains why Void Linux doesn't have this problem now. |
I have keyword masked the affected versions until elogind issue #51 is resolved (it was reopened ~2 days ago).
Spargeltarzan wrote: | Do you think the newer version of elogind might also be responsible for the non-working auto-unlocking of gnome-keyring after logging in in gdm? |
If it worked before then it might be.
It might also just be a change in GNOME 3.28 itself.
hououin wrote: | Noticed that when I log into my Google account using Gnome Online Accounts I get this after I enter my username and it freezes up:
Code: | Jun 30 18:51:10 okabe kernel: [ 6361.607682] WebKitWebProces[12881]: segfault at bbadbeef ip 00007fc97a5144cb sp 00007ffed44ae9c0 error 6 in libjavascriptcoregtk-4.0.so.18.7.11[7fc977a72000+3a2f000] |
|
cono wrote: | With recent update also getting tonz of these errors when launching my evolution-owa and trying to navigate via emails. |
Either something broke between WebKitGTK+ 2.20.1 and 2.20.3 (as there have definitely been issues with the 2.20.z series) or a dependency didn't get rebuilt.
I'm going to bet that things probably still work with 2.18.6 which is safe enough to use as long as you are mindful of the security advisories (or if you know that the security advisories don't apply to you).
I added 2.20.1 and 2.21.4 so that there are 4 versions (2.18.6, 2.20.1, 2.20.3 and 2.21.4) to choose from. _________________ Dantrell B. |
|
Back to top |
|
|
Ultimator n00b
Joined: 23 Jun 2018 Posts: 26
|
Posted: Wed Jul 04, 2018 8:17 pm Post subject: |
|
|
dantrell wrote: | I have keyword masked the affected versions until elogind issue #51 is resolved (it was reopened ~2 days ago). |
This change downgrades elogind to 233.7-r1. That version didn't build for me at first. I got the following error message.
Code: | In file included from src/basic/missing.h:1197:0,
from src/basic/util.h:46,
from src/basic/hashmap.h:29,
from src/basic/locale-util.c:35:
src/basic/missing_syscall.h:62:19: error: static declaration of ‘memfd_create’ follows non-static declaration
static inline int memfd_create(const char *name, unsigned int flags) {
^~~~~~~~~~~~
In file included from /usr/include/bits/mman-linux.h:115:0,
from /usr/include/bits/mman.h:45,
from /usr/include/sys/mman.h:41,
from src/basic/locale-util.c:30:
/usr/include/bits/mman-shared.h:46:5: note: previous declaration of ‘memfd_create’ was here
int memfd_create (const char *__name, unsigned int __flags) __THROW; |
However I found a patch for a similiar issue (https://github.com/systemd/systemd/commit/5187dd2c403caf92d09f3491e41f1ceb3f10491f) and adapted it.
The package builds now and suspend also works when closing the lid. I uploaded it to github just in case sb encounters the same issue: https://gist.github.com/Ultimator14/1f0e52f40bee049e778a334cc3850017 |
|
Back to top |
|
|
cono n00b
Joined: 15 Sep 2016 Posts: 8
|
Posted: Wed Jul 04, 2018 11:47 pm Post subject: |
|
|
dantrell wrote: |
I added 2.20.1 and 2.21.4 so that there are 4 versions (2.18.6, 2.20.1, 2.20.3 and 2.21.4) to choose from. |
2.18.6 fails with linker errors
Code: |
lib/libjavascriptcoregtk-4.0.so.18.6.15: error: undefined reference to 'JSC::HeapCell::vm() const'
lib/libjavascriptcoregtk-4.0.so.18.6.15: error: undefined reference to 'JSC::B3::BasicBlock::fallThrough() const'
collect2: error: ld returned 1 exit status
|
also checked 2.20.1 and 2.20.3, both gives:
Code: |
[135202.941189] WebKitWebProces[1466]: segfault at bbadbeef ip 00007fb9f1ee6c9c sp 00007ffd167895c0 error 6 in libjavascriptcoregtk-4.0.so.18.7.11[7fb9f0f54000+1816000]
[137877.190704] WebKitWebProces[3835]: segfault at bbadbeef ip 00007f5fdf28092c sp 00007ffc7d6c02d0 error 6 in libjavascriptcoregtk-4.0.so.18.7.9[7f5fde2ef000+1813000]
|
Haven't tried 2.21.4 yet. Too slow compilation |
|
Back to top |
|
|
dantrell l33t
Joined: 01 Jun 2007 Posts: 915 Location: Earth
|
Posted: Thu Jul 05, 2018 1:20 am Post subject: |
|
|
Thanks for the feedback.
I have applied this patch on my end.
cono wrote: | 2.18.6 fails with linker errors |
I recognize one of these issues so I think I can probably correct this.
Edit: Actually, I think I'll need your build log for this one.
cono wrote: | Too slow compilation |
This is the biggest reason why it's a bit much for me to diagnose WebkitGTK+ issues. _________________ Dantrell B. |
|
Back to top |
|
|
Spargeltarzan Guru
Joined: 23 Jul 2017 Posts: 317
|
Posted: Thu Jul 05, 2018 8:38 am Post subject: |
|
|
I can't emerge webkit-gtk 2.18.6 too
Upgraded to 2.21.4 successfully. _________________ ___________________
Regards
Spargeltarzan
Notebook: Lenovo YOGA 900-13ISK: Gentoo stable amd64, GNOME systemd, KVM/QEMU
Desktop-PC: Intel Core i7-4770K, 8GB Ram, AMD Radeon R9 280X, ZFS Storage, GNOME openrc, Dantrell, Xen
Last edited by Spargeltarzan on Thu Jul 05, 2018 9:37 am; edited 1 time in total |
|
Back to top |
|
|
cono n00b
Joined: 15 Sep 2016 Posts: 8
|
Posted: Thu Jul 05, 2018 9:33 am Post subject: |
|
|
dantrell wrote: |
Edit: Actually, I think I'll need your build log for this one.
|
Here is mine: https://www.cono.org.ua/build.log let me know if you need more details. |
|
Back to top |
|
|
Spargeltarzan Guru
Joined: 23 Jul 2017 Posts: 317
|
Posted: Thu Jul 05, 2018 2:25 pm Post subject: |
|
|
Hi,
I wonder if someone experiences similar udev errors:
Code: |
[ 1.824932] udevd[2052]: invalid key/value pair in file /usr/lib64/udev/rules.d/61-gdm.rules on line 2, starting at character 175 ('[')
[ 1.830082] udevd[2052]: RUN{builtin}: 'uaccess' unknown /lib64/udev/rules.d/73-seat-late.rules:15
[ 3.037387] udevd[2053]: invalid key/value pair in file /usr/lib64/udev/rules.d/61-gdm.rules on line 2, starting at character 175 ('[')
[ 3.037836] udevd[2053]: RUN{builtin}: 'uaccess' unknown /lib64/udev/rules.d/73-seat-late.rules:15
[ 3.276212] usb 3-4: string descriptor 0 read error: -22
|
They are present quite long, they do not disturb, but I am curious what might be the cause and how to fix.
I guess we use all openrc-0.29? Did someone try already a newer openrc version? _________________ ___________________
Regards
Spargeltarzan
Notebook: Lenovo YOGA 900-13ISK: Gentoo stable amd64, GNOME systemd, KVM/QEMU
Desktop-PC: Intel Core i7-4770K, 8GB Ram, AMD Radeon R9 280X, ZFS Storage, GNOME openrc, Dantrell, Xen |
|
Back to top |
|
|
alfredthetomato n00b
Joined: 06 Jul 2018 Posts: 2
|
Posted: Fri Jul 06, 2018 12:07 pm Post subject: |
|
|
Spargeltarzan wrote: | Hi,
I wonder if someone experiences similar udev errors:
Code: |
[ 1.824932] udevd[2052]: invalid key/value pair in file /usr/lib64/udev/rules.d/61-gdm.rules on line 2, starting at character 175 ('[')
[ 1.830082] udevd[2052]: RUN{builtin}: 'uaccess' unknown /lib64/udev/rules.d/73-seat-late.rules:15
[ 3.037387] udevd[2053]: invalid key/value pair in file /usr/lib64/udev/rules.d/61-gdm.rules on line 2, starting at character 175 ('[')
[ 3.037836] udevd[2053]: RUN{builtin}: 'uaccess' unknown /lib64/udev/rules.d/73-seat-late.rules:15
[ 3.276212] usb 3-4: string descriptor 0 read error: -22
|
They are present quite long, they do not disturb, but I am curious what might be the cause and how to fix.
I guess we use all openrc-0.29? Did someone try already a newer openrc version? |
Try adding /lib64/elogind/elogind-uaccess-command to your initramfs. |
|
Back to top |
|
|
dantrell l33t
Joined: 01 Jun 2007 Posts: 915 Location: Earth
|
Posted: Sun Jul 08, 2018 12:14 am Post subject: |
|
|
cono wrote: | 2.18.6 fails with linker errors |
Spargeltarzan wrote: | I can't emerge webkit-gtk 2.18.6 too |
This issue has been corrected.
P.S. I previously mentioned that there have definitely been issues with the 2.20.z series. Said series also builds ~2x faster than the 2.18.z series. Make of that what you will.
Spargeltarzan wrote: | I wonder if someone experiences similar udev errors |
alfredthetomato wrote: | Try adding /lib64/elogind/elogind-uaccess-command to your initramfs. |
This is now reminding me of elogind issue #10.
It was fixed in 235.3 so I guess I have some backporting to do.
Edit: And... backported. I'll leave the testing to you. _________________ Dantrell B. |
|
Back to top |
|
|
|