[opensuse-factory] *Updated* NVIDIA gfx driver RPMs available for Tumbleweed ...
Hi Since today there are updated NVIDIA gfx driver RPMs available for Tumbleweed. Feel free to give them a/another try by adding the repository manually via zypper ar https://download.nvidia.com/opensuse/tumbleweed nvidia-tumbleweed (if not created yet) and install them via zypper inr (or "zypper up", if already installed) The kernel module is being built during installation (as it's done with the Leap packages). But the kernel module is also being rebuilt and reinstalled after a kernel update has been done, since we don't necessarily keep the kABI stable for TW. This has been implemented by making use of RPM's trigger scripts. Issues fixed in this update: - gdm could not be started Known issues/limitations: https://download.nvidia.com/opensuse/tumbleweed nvidia-tumbleweed is a non-browsable directory, so please don't tell me it doesn't exist after pasting it into your favorite web-browser. ;-) Packages are only available for the latest long lived branch (currently 384.59). Reason is, that with TW we switched to libglvnd, which is not supported by the legacy driver series 340.xx and 304.xx. If you had NVIDIA driver installed manually when updating TW from Mesa to libglvnd+Mesa, you'll see an orphaned libGL.so.1.2* in /usr/lib64 and /usr/lib. This may have happened unnoticed. So please check this after running nvidia-installer --uninstall and remove these files, if available via rm -f /usr/lib*/libGL.so.1.2* before installing the NVIDIA RPMs. This is already documented in https://en.opensuse.org/SDB:NVIDIA_the_hard_way You can provide feedback here in this mail thread or via bugzilla. If you use Bugzilla, please use Product: openSUSE Tumbleweed Component: X11 3rd party Driver Thanks, Stefan Public Key available ------------------------------------------------------ Stefan Dirsch (Res. & Dev.) SUSE LINUX GmbH Tel: 0911-740 53 0 Maxfeldstraße 5 FAX: 0911-740 53 479 D-90409 Nürnberg http://www.suse.de Germany --------------------------------------------------------------- SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg) --------------------------------------------------------------- -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Hi, On mardi 15 août 2017 14:18:04 CEST Stefan Dirsch wrote:
Hi
Since today there are updated NVIDIA gfx driver RPMs available for Tumbleweed. Feel free to give them a/another try by adding the repository manually via
zypper ar https://download.nvidia.com/opensuse/tumbleweed nvidia-tumbleweed
(if not created yet) and install them via
zypper inr
(or "zypper up", if already installed)
The kernel module is being built during installation (as it's done with the Leap packages). But the kernel module is also being rebuilt and reinstalled after a kernel update has been done, since we don't necessarily keep the kABI stable for TW. This has been implemented by making use of RPM's trigger scripts.
Issues fixed in this update:
- gdm could not be started
Known issues/limitations:
There was a package update to 384.69. Don't install it if you're using KDE. It causes segfaults in kscreenlocker_greet (the process that spawns and asks for your password when your session is locked). Also read https://bugs.kde.org/384005 Christophe -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On samedi, 26 août 2017 16.01:31 h CEST Christophe Giboudeaux wrote:
Hi,
On mardi 15 août 2017 14:18:04 CEST Stefan Dirsch wrote:
Hi
Since today there are updated NVIDIA gfx driver RPMs available for Tumbleweed. Feel free to give them a/another try by adding the repository manually via
zypper ar https://download.nvidia.com/opensuse/tumbleweed
nvidia-tumbleweed
(if not created yet) and install them via
zypper inr
(or "zypper up", if already installed)
The kernel module is being built during installation (as it's done with the Leap packages). But the kernel module is also being rebuilt and reinstalled after a kernel update has been done, since we don't necessarily keep the kABI stable for TW. This has been implemented by making use of RPM's trigger scripts.
Issues fixed in this update:
- gdm could not be started
Known issues/limitations: There was a package update to 384.69. Don't install it if you're using KDE. It causes segfaults in kscreenlocker_greet (the process that spawns and asks for your password when your session is locked).
Also read https://bugs.kde.org/384005
Christophe
Unfortunately, once installed there's no way to go back. Perhaps something that could be improved (if nvidia agree), is to have at least two version of the driver. So in such case, impacted users can downgrade easily. btw: screen locker crash here too. -- Bruno Friedmann Ioda-Net Sàrl www.ioda-net.ch Bareos Partner, openSUSE Member, fsfe fellowship GPG KEY : D5C9B751C4653227 irc: tigerfoot -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 27/08/17 03:51 AM, Bruno Friedmann wrote:
On samedi, 26 août 2017 16.01:31 h CEST Christophe Giboudeaux wrote:
Hi,
On mardi 15 août 2017 14:18:04 CEST Stefan Dirsch wrote:
Hi
Since today there are updated NVIDIA gfx driver RPMs available for Tumbleweed. Feel free to give them a/another try by adding the repository manually via
zypper ar https://download.nvidia.com/opensuse/tumbleweed
nvidia-tumbleweed
(if not created yet) and install them via
zypper inr
(or "zypper up", if already installed)
The kernel module is being built during installation (as it's done with the Leap packages). But the kernel module is also being rebuilt and reinstalled after a kernel update has been done, since we don't necessarily keep the kABI stable for TW. This has been implemented by making use of RPM's trigger scripts.
Issues fixed in this update:
- gdm could not be started
Known issues/limitations: There was a package update to 384.69. Don't install it if you're using KDE. It causes segfaults in kscreenlocker_greet (the process that spawns and asks for your password when your session is locked).
Also read https://bugs.kde.org/384005
Christophe
Unfortunately, once installed there's no way to go back. Perhaps something that could be improved (if nvidia agree), is to have at least two version of the driver. So in such case, impacted users can downgrade easily.
btw: screen locker crash here too.
For the time being. You could always remove the rpms and use the .run drivers downloaded from the Nvidia site. ;-) -- Cheers! Roman -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Bruno Friedmann wrote:
Unfortunately, once installed there's no way to go back. Perhaps something that could be improved (if nvidia agree), is to have at least two version of the driver. So in such case, impacted users can downgrade easily.
btw: screen locker crash here too. The option keeppackages=1 in the Nvidia repository can be used too. Unfortunately I also forgot to set this option. If keeppackages=1 is set, you could manually downgrade the packages from the nvidia repository cache in /var/cache/zypp/packages/nvidia* and later lock the installed packages with YaST or Zypper.
nvidia-tumbleweed.repo: [nvidia-tumbleweed] enabled=1 autorefresh=0 baseurl=https://download.nvidia.com/opensuse/tumbleweed type=rpm-md keeppackages=1 Greetings, Björn -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Sat, Aug 26, 2017 at 04:01:31PM +0200, Christophe Giboudeaux wrote:
Since today there are updated NVIDIA gfx driver RPMs available for Tumbleweed. Feel free to give them a/another try by adding the repository manually via
There was a package update to 384.69.
Ouch!
Don't install it if you're using KDE. It causes segfaults in kscreenlocker_greet (the process that spawns and asks for your password when your session is locked).
Can you do me a favor and test, whether this also happens with the manual installation of the driver via NVIDIA's .run file? Uninstall the NVIDIA RPMs before. If the result is the same also try the previous driver version 384.59. http://download.nvidia.com/XFree86/Linux-x86/384.59/NVIDIA-Linux-x86-384.59.... http://download.nvidia.com/XFree86/Linux-x86_64/384.59/NVIDIA-Linux-x86_64-3... So we know, whether we see a driver regression unrelated to package changes here.
Also read https://bugs.kde.org/384005
Thanks! Stefan Public Key available ------------------------------------------------------ Stefan Dirsch (Res. & Dev.) SUSE LINUX GmbH Tel: 0911-740 53 0 Maxfeldstraße 5 FAX: 0911-740 53 479 D-90409 Nürnberg http://www.suse.de Germany --------------------------------------------------------------- SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg) --------------------------------------------------------------- -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On lundi, 28 août 2017 10.42:51 h CEST Stefan Dirsch wrote:
On Sat, Aug 26, 2017 at 04:01:31PM +0200, Christophe Giboudeaux wrote:
Since today there are updated NVIDIA gfx driver RPMs available for Tumbleweed. Feel free to give them a/another try by adding the repository manually via
There was a package update to 384.69.
Ouch!
Don't install it if you're using KDE. It causes segfaults in kscreenlocker_greet (the process that spawns and asks for your password when your session is locked).
Can you do me a favor and test, whether this also happens with the manual installation of the driver via NVIDIA's .run file? Uninstall the NVIDIA RPMs before. If the result is the same also try the previous driver version 384.59.
http://download.nvidia.com/XFree86/Linux-x86/384.59/NVIDIA-Linux-x86-384.59. run http://download.nvidia.com/XFree86/Linux-x86_64/384.59/NVIDIA-Linux-x86_64-> 384.59.run
So we know, whether we see a driver regression unrelated to package changes here.
Also read https://bugs.kde.org/384005
Thanks!
Stefan
Also there's request at nvidia for informations too https://devtalk.nvidia.com/default/topic/1023193/linux/384-69-broke-kde-scre... -- Bruno Friedmann Ioda-Net Sàrl www.ioda-net.ch Bareos Partner, openSUSE Member, fsfe fellowship GPG KEY : D5C9B751C4653227 irc: tigerfoot -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On lundi, 28 août 2017 11.04:07 h CEST Bruno Friedmann wrote:
On lundi, 28 août 2017 10.42:51 h CEST Stefan Dirsch wrote:
On Sat, Aug 26, 2017 at 04:01:31PM +0200, Christophe Giboudeaux wrote:
Since today there are updated NVIDIA gfx driver RPMs available for Tumbleweed. Feel free to give them a/another try by adding the repository manually via
There was a package update to 384.69.
Ouch!
Don't install it if you're using KDE. It causes segfaults in kscreenlocker_greet (the process that spawns and asks for your password when your session is locked).
Can you do me a favor and test, whether this also happens with the manual installation of the driver via NVIDIA's .run file? Uninstall the NVIDIA RPMs before. If the result is the same also try the previous driver version 384.59.
http://download.nvidia.com/XFree86/Linux-x86/384.59/NVIDIA-Linux-x86-384.5 9. run http://download.nvidia.com/XFree86/Linux-x86_64/384.59/NVIDIA-Linux-x86_64 -> 384.59.run
So we know, whether we see a driver regression unrelated to package changes here.
Also read https://bugs.kde.org/384005
Thanks!
Stefan
Also there's request at nvidia for informations too
https://devtalk.nvidia.com/default/topic/1023193/linux/384-69-broke-kde-scre en-locker-possibly-other-qt-based-software-on-linux/
Now strangely, I dup to today snapshot, and kscreen-locker has work two times? -- Bruno Friedmann Ioda-Net Sàrl www.ioda-net.ch Bareos Partner, openSUSE Member, fsfe fellowship GPG KEY : D5C9B751C4653227 irc: tigerfoot -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Mon, Aug 28, 2017 at 05:07:32PM +0200, Bruno Friedmann wrote:
On lundi, 28 août 2017 11.04:07 h CEST Bruno Friedmann wrote:
On lundi, 28 août 2017 10.42:51 h CEST Stefan Dirsch wrote:
On Sat, Aug 26, 2017 at 04:01:31PM +0200, Christophe Giboudeaux wrote:
Since today there are updated NVIDIA gfx driver RPMs available for Tumbleweed. Feel free to give them a/another try by adding the repository manually via
There was a package update to 384.69.
Ouch!
Don't install it if you're using KDE. It causes segfaults in kscreenlocker_greet (the process that spawns and asks for your password when your session is locked).
Can you do me a favor and test, whether this also happens with the manual installation of the driver via NVIDIA's .run file? Uninstall the NVIDIA RPMs before. If the result is the same also try the previous driver version 384.59.
http://download.nvidia.com/XFree86/Linux-x86/384.59/NVIDIA-Linux-x86-384.5 9. run http://download.nvidia.com/XFree86/Linux-x86_64/384.59/NVIDIA-Linux-x86_64 -> 384.59.run
So we know, whether we see a driver regression unrelated to package changes here.
Also read https://bugs.kde.org/384005
Thanks!
Stefan
Also there's request at nvidia for informations too
https://devtalk.nvidia.com/default/topic/1023193/linux/384-69-broke-kde-scre en-locker-possibly-other-qt-based-software-on-linux/
Now strangely, I dup to today snapshot, and kscreen-locker has work two times?
And I can't reproduce this problem with 384.69 RPMs either. :-( Stefan Public Key available ------------------------------------------------------ Stefan Dirsch (Res. & Dev.) SUSE LINUX GmbH Tel: 0911-740 53 0 Maxfeldstraße 5 FAX: 0911-740 53 479 D-90409 Nürnberg http://www.suse.de Germany --------------------------------------------------------------- SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg) --------------------------------------------------------------- -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Il giorno Mon, 28 Aug 2017 17:53:12 +0200 Stefan Dirsch <sndirsch@suse.de> ha scritto:
And I can't reproduce this problem with 384.69 RPMs either. :-(
Apparently, the issue was in the way the screenlocker set a seccomp sandbox (rather, that such a sandbox was too restrictive)[1]. Upstream has made a patch, and will make a new tarball for it. Once the code is merged upstream, we'll get the patch in our packages. [1] https://phabricator.kde.org/D7616 -- Luca Beltrame - KDE Forums team GPG key ID: A29D259B
Am Samstag, den 26.08.2017, 16:01 +0200 schrieb Christophe Giboudeaux:
Hi,
On mardi 15 août 2017 14:18:04 CEST Stefan Dirsch wrote:
There was a package update to 384.69. Don't install it if you're using KDE. It causes segfaults in kscreenlocker_greet (the process that spawns and asks for your password when your session is locked).
does anybody know if this workaround will be available for the tumbleweed KDE packages: https://bugs.kde.org/show_bug.cgi?id=384005#c15 -- Best Regards DI Rainer Klier Research & Development, DevOps Namirial GmbH
Am Freitag, den 01.09.2017, 12:03 +0000 schrieb Rainer Klier:
[This sender failed our fraud detection checks and may not be who they appear to be. Learn about spoofing at http://ak a.ms/LearnAboutSpoofing]
Am Samstag, den 26.08.2017, 16:01 +0200 schrieb Christophe Giboudeaux:
There was a package update to 384.69. Don't install it if you're using KDE. It causes segfaults in kscreenlocker_greet (the process that spawns and asks for your password when your session is locked).
does anybody know if this workaround will be available for the tumbleweed KDE packages: https://bugs.kde.org/show_bug.cgi?id=384005#c15
i just saw new package of kscreenlocker version 5.10.5-89.1 in http://download.opensuse.org/repositories/KDE:/Frameworks 5/openSUSE_Factory with this changelog: Thu Aug 31 14:00:00 2017 fabian@ritter-vogt.de - Add patch to allow the NVIDIA driver to poke through seccomp and thus not crashing (kde#384005): * 0001-Don-t-dissallow-open-with-write-flag-syscall-on-NVID.patch THANKS, fabian! -- Best Regards DI Rainer Klier Research & Development, DevOps Namirial GmbH
On Tue 15 Aug 2017 02:18:04 PM CDT, Stefan Dirsch wrote: <snip>
Known issues/limitations:
https://download.nvidia.com/opensuse/tumbleweed nvidia-tumbleweed is a non-browsable directory, so please don't tell me it doesn't exist after pasting it into your favorite web-browser. ;-)
Packages are only available for the latest long lived branch (currently 384.59). Reason is, that with TW we switched to libglvnd, which is not supported by the legacy driver series 340.xx and 304.xx.
<snip> Hi Just a couple of observations, the directory can be browsed via; https://http.download.nvidia.com/opensuse/ ;) Secondly, I have patched the 340.102 driver since I have a legacy card (8800 GT) working fine (again as far as I can tell) with Tumbleweed and the 4.12.8 kernel. Is there a potential impact using this unsupported legacy card and libglvnd? -- Cheers Malcolm °¿° SUSE Knowledge Partner (Linux Counter #276890) openSUSE Leap 42.2 | GNOME 3.20.2 | 4.4.79-18.26-default HP 255 G4 Notebook | A6-6310 X4 @ 1.80 GHz | AMD Radeon R4 up 10 days 12:17, 1 user, load average: 0.51, 0.54, 0.52 -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Mon, Aug 28, 2017 at 11:19:29AM -0500, Malcolm wrote:
Known issues/limitations:
https://download.nvidia.com/opensuse/tumbleweed nvidia-tumbleweed is a non-browsable directory, so please don't tell me it doesn't exist after pasting it into your favorite web-browser. ;-)
Packages are only available for the latest long lived branch (currently 384.59). Reason is, that with TW we switched to libglvnd, which is not supported by the legacy driver series 340.xx and 304.xx.
<snip>
Hi Just a couple of observations, the directory can be browsed via; https://http.download.nvidia.com/opensuse/ ;)
This is news to me. Maybe they changed their webserver setup recently.
Secondly, I have patched the 340.102 driver since I have a legacy card (8800 GT) working fine (again as far as I can tell) with Tumbleweed and the 4.12.8 kernel. Is there a potential impact using this unsupported legacy card and libglvnd?
Well, I would be interested having a look at your patch. Thanks, Stefan Public Key available ------------------------------------------------------ Stefan Dirsch (Res. & Dev.) SUSE LINUX GmbH Tel: 0911-740 53 0 Maxfeldstraße 5 FAX: 0911-740 53 479 D-90409 Nürnberg http://www.suse.de Germany --------------------------------------------------------------- SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg) --------------------------------------------------------------- -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Mon 28 Aug 2017 10:03:36 PM CDT, Stefan Dirsch wrote:
Secondly, I have patched the 340.102 driver since I have a legacy card (8800 GT) working fine (again as far as I can tell) with Tumbleweed and the 4.12.8 kernel. Is there a potential impact using this unsupported legacy card and libglvnd?
Well, I would be interested having a look at your patch.
Hi The patch is here; http://paste.opensuse.org/view/raw/27535948 Then; ./NVIDIA-Linux-x86_64-340.102.run --apply-patch nv_patch_340.102_linux_kernel_4.12.patch ./NVIDIA-Linux-x86_64-340.102-custom.run -aq Rebased from <https://devtalk.nvidia.com/default/topic/1019362/fully-working-patch-for-nvidia-driver-340-102-compiler-installer-file-and-linux-kernel-4-12/> -- Cheers Malcolm °¿° SUSE Knowledge Partner (Linux Counter #276890) openSUSE Leap 42.2 | GNOME 3.20.2 | 4.4.79-18.26-default HP 255 G4 Notebook | A6-6310 X4 @ 1.80 GHz | AMD Radeon R4 up 10 days 17:13, 1 user, load average: 0.47, 0.42, 0.35 -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Mon, Aug 28, 2017 at 03:55:45PM -0500, Malcolm wrote:
On Mon 28 Aug 2017 10:03:36 PM CDT, Stefan Dirsch wrote:
Secondly, I have patched the 340.102 driver since I have a legacy card (8800 GT) working fine (again as far as I can tell) with Tumbleweed and the 4.12.8 kernel. Is there a potential impact using this unsupported legacy card and libglvnd?
Well, I would be interested having a look at your patch.
Hi The patch is here; http://paste.opensuse.org/view/raw/27535948
Oh. You only have a patch for the kernel. I thought you would also have a patch against our package sources, so the driver works somehow although we now switched to libglvnd with TW. I never tried 340.xx installer on TW with libglvnd. I would expect issues. But maybe I'm wrong. Thanks, Stefan
Then; ./NVIDIA-Linux-x86_64-340.102.run --apply-patch nv_patch_340.102_linux_kernel_4.12.patch ./NVIDIA-Linux-x86_64-340.102-custom.run -aq
Rebased from <https://devtalk.nvidia.com/default/topic/1019362/fully-working-patch-for-nvi...
Public Key available ------------------------------------------------------ Stefan Dirsch (Res. & Dev.) SUSE LINUX GmbH Tel: 0911-740 53 0 Maxfeldstraße 5 FAX: 0911-740 53 479 D-90409 Nürnberg http://www.suse.de Germany --------------------------------------------------------------- SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg) --------------------------------------------------------------- -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Tue, Aug 29, 2017 at 10:11:37AM +0200, Stefan Dirsch wrote:
On Mon, Aug 28, 2017 at 03:55:45PM -0500, Malcolm wrote:
On Mon 28 Aug 2017 10:03:36 PM CDT, Stefan Dirsch wrote:
Secondly, I have patched the 340.102 driver since I have a legacy card (8800 GT) working fine (again as far as I can tell) with Tumbleweed and the 4.12.8 kernel. Is there a potential impact using this unsupported legacy card and libglvnd?
Well, I would be interested having a look at your patch.
Hi The patch is here; http://paste.opensuse.org/view/raw/27535948
Oh. You only have a patch for the kernel. I thought you would also have a patch against our package sources, so the driver works somehow although we now switched to libglvnd with TW.
I never tried 340.xx installer on TW with libglvnd. I would expect issues. But maybe I'm wrong.
Then; ./NVIDIA-Linux-x86_64-340.102.run --apply-patch nv_patch_340.102_linux_kernel_4.12.patch ./NVIDIA-Linux-x86_64-340.102-custom.run -aq
Rebased from <https://devtalk.nvidia.com/default/topic/1019362/fully-working-patch-for-nvi...
Ok. Just had a try with such a manual installation. Seems to work for me as well. NVIDIA's installer moves away /usr/lib64/libEGL.so.1.0.0 /usr/lib64/libGL.so.1.0.0 /usr/lib64/libGLESv1_CM.so.1.0.0 /usr/lib64/libGLESv2.so.2.0.0 /usr/lib64/xorg/modules/libglamoregl.so and relinks /usr/lib64/libEGL.so.1 /usr/lib64/libGL.so.1 /usr/lib64/libGLESv1_CM.so.1 /usr/lib64/libGLESv2.so.2 /usr/lib64/xorg/modules/extensions/libglx.so as if it wouldn't be a glvnd installation. Of course once you update Mesa this results in a messed-up system again. So, in theory it would be possible to prepare G03/340.xx packages also for TW. But it would be a lot of work I'm afraid. Thanks, Stefan Public Key available ------------------------------------------------------ Stefan Dirsch (Res. & Dev.) SUSE LINUX GmbH Tel: 0911-740 53 0 Maxfeldstraße 5 FAX: 0911-740 53 479 D-90409 Nürnberg http://www.suse.de Germany --------------------------------------------------------------- SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg) --------------------------------------------------------------- -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
вівторок, 15 серпня 2017 р. 15:18:04 EEST Stefan Dirsch написано:
Since today there are updated NVIDIA gfx driver RPMs available for Tumbleweed. Feel free to give them a/another try by adding the repository manually via
zypper ar https://download.nvidia.com/opensuse/tumbleweed nvidia-tumbleweed
Hi, is there any hints to install it on "optimus" laptop? I'm aware that switching between GPU's would not work, I want to try pure nvidia on parallel installation (for tests and Vulkan). I've removed bumblebee, bbswitch, bumblebee-nvidia packages. Restarted and installed packages from nvidia repository. There was still Intel driver used. I've run nvidia-xconfig (which looks kinda buggy), and got strange /etc/X11/ xorg.conf [1]. Tried to boot with it anyway but X hadn't started and got fast switching from one to other tty. It's almost unable to login while that. Restarted without X, renamed xorg.conf and added manually to xorg.conf.d only essential parts [2]. But after restart got the very same behaviour — fast switching between tty's. Is that something of bumblebee left or something other is wrong? There is no any /usr/lib*/libGL.so.1.2* files. Intel i7 6700HQ NVidia GTX970M openSUSE Tumbleweed x86_64 boot options: acpi_osi=! "acpi_osi=Windows 2009" SDDM KDE Plasma 5 [1] http://susepaste.org/38919395 [2] http://susepaste.org/42224432 -- Kind regards, Mykola Krachkovsky -- Найкращі побажання, Микола Крачковський
Sorry, forgot to add Xorg.0.log [1]. Bad part is at the end. [ 84.459] (EE) No devices detected. [ 84.459] (EE) Fatal server error: [ 84.459] (EE) no screens found(EE) [1] http://susepaste.org/73527391 -- Kind regards, Mykola Krachkovsky -- Найкращі побажання, Микола Крачковський
02.09.2017 15:58, Mykola Krachkovsky пишет:
вівторок, 15 серпня 2017 р. 15:18:04 EEST Stefan Dirsch написано:
Since today there are updated NVIDIA gfx driver RPMs available for Tumbleweed. Feel free to give them a/another try by adding the repository manually via
zypper ar https://download.nvidia.com/opensuse/tumbleweed nvidia-tumbleweed
Hi, is there any hints to install it on "optimus" laptop? I'm aware that switching between GPU's would not work, I want to try pure nvidia on parallel installation (for tests and Vulkan).
This really depends on your precise optimus (hardware) implementation and what you want to do. nVidia proprietary driver does not support render offload, so you can only switch complete X output running on dGPU to screen connected to iGPU device. You can neither use both iGPU and dGPU simultaneously (e.g. start *one* program on dGPU and display output on screen connected to iGPU) nor start program (or complete X) on iGPU and display output on screen connected to dGPU. bumblebee bypasses this problem by using "sideband" channel to display content rendered on dGPU on screen connected to iGPU. So in general you cannot replace bumblebee with proprietary drivers + glvnd. You can use nouveau, although you still need to manually manage which GPU will be doing rendering, same as with bumblebee (by explicitly starting program on dGPU). If dGPU has own screen, I presume it should be possible to use both of them, although I am unsure about to which extent it will be possible to combine them in single "desktop".
I've removed bumblebee, bbswitch, bumblebee-nvidia packages. Restarted and installed packages from nvidia repository. There was still Intel driver used. I've run nvidia-xconfig (which looks kinda buggy), and got strange /etc/X11/ xorg.conf [1]. Tried to boot with it anyway but X hadn't started and got fast switching from one to other tty. It's almost unable to login while that. Restarted without X, renamed xorg.conf and added manually to xorg.conf.d only essential parts [2]. But after restart got the very same behaviour — fast switching between tty's.
Is that something of bumblebee left or something other is wrong? There is no any /usr/lib*/libGL.so.1.2* files.
Intel i7 6700HQ NVidia GTX970M
openSUSE Tumbleweed x86_64 boot options: acpi_osi=! "acpi_osi=Windows 2009" SDDM KDE Plasma 5
[1] http://susepaste.org/38919395 [2] http://susepaste.org/42224432
субота, 2 вересня 2017 р. 17:21:24 EEST Andrei Borzenkov написано:
02.09.2017 15:58, Mykola Krachkovsky пишет:
Hi, is there any hints to install it on "optimus" laptop? I'm aware that switching between GPU's would not work, I want to try pure nvidia on parallel installation (for tests and Vulkan).
This really depends on your precise optimus (hardware) implementation and what you want to do.
Maybe I was not clear enough, I want to run X on NVidia-only. What else about implementation I could provide? Screen connected to Intel GPU I suppose, at least there is no problem when NVidia GPU is completely switched off by bumblebee.
nVidia proprietary driver does not support render offload, so you can only switch complete X output running on dGPU to screen connected to iGPU device. You can neither use both iGPU and dGPU simultaneously (e.g. start *one* program on dGPU and display output on screen connected to iGPU) nor start program (or complete X) on iGPU and display output on screen connected to dGPU.
bumblebee bypasses this problem by using "sideband" channel to display content rendered on dGPU on screen connected to iGPU.
So in general you cannot replace bumblebee with proprietary drivers + glvnd. You can use nouveau, although you still need to manually manage which GPU will be doing rendering, same as with bumblebee (by explicitly starting program on dGPU).
If dGPU has own screen, I presume it should be possible to use both of them, although I am unsure about to which extent it will be possible to combine them in single "desktop".
I'm not interested in running both Intel and NVidia, that's what bumblebee does (although not very well). NVidia-only on optimus laptop. So I didn't really get why I need that, sorry. -- Kind regards, Mykola Krachkovsky -- Найкращі побажання, Микола Крачковський
02.09.2017 23:43, Mykola Krachkovsky пишет:
субота, 2 вересня 2017 р. 17:21:24 EEST Andrei Borzenkov написано:
02.09.2017 15:58, Mykola Krachkovsky пишет:
Hi, is there any hints to install it on "optimus" laptop? I'm aware that switching between GPU's would not work, I want to try pure nvidia on parallel installation (for tests and Vulkan).
This really depends on your precise optimus (hardware) implementation and what you want to do.
Maybe I was not clear enough, I want to run X on NVidia-only. What else about implementation I could provide?
You may have hardware switch to select between iGPU and dGPU. In this case you simply need to flip it, presumably via BIOS settings.
Screen connected to Intel GPU I suppose, at least there is no problem when NVidia GPU is completely switched off by bumblebee.
Assuming no hardware multiplexor, your xorg.conf is obviously wrong. You must have both iGPU and dGPU active in X configuration. Did you read https://us.download.nvidia.com/XFree86/Linux-x86_64/384.69/README/randr14.ht...
неділя, 3 вересня 2017 р. 09:34:25 EEST Andrei Borzenkov написано:
You may have hardware switch to select between iGPU and dGPU. In this case you simply need to flip it, presumably via BIOS settings.
I will check, but I don't want to switch to dGPU, as my main openSUSE Tumbleweed uses bumblebee and parallel (I'm trying to get worked) openSUSE with NVidia only is just for bug testing and Vulkan.
Assuming no hardware multiplexor, your xorg.conf is obviously wrong. You must have both iGPU and dGPU active in X configuration. Did you read https://us.download.nvidia.com/XFree86/Linux-x86_64/384.69/README/randr14.ht ml?
As I've managed to eliminate errors from Xorg.0.log it looks like this is the issue of blank black screen. Thanks, I will try to use that. Strange Arch wiki has nothing about that or I missed something. -- Kind regards, Mykola Krachkovsky -- Найкращі побажання, Микола Крачковський
03.09.2017 11:26, Mykola Krachkovsky пишет:
Strange Arch wiki has nothing about that
https://wiki.archlinux.org/index.php/NVIDIA_Optimus#Using_nvidia describes exactly that.
неділя, 3 вересня 2017 р. 11:35:23 EEST Andrei Borzenkov написано:
03.09.2017 11:26, Mykola Krachkovsky пишет:
Strange Arch wiki has nothing about that
https://wiki.archlinux.org/index.php/NVIDIA_Optimus#Using_nvidia describes exactly that.
I've done that (except "Alternative configuration") and added /usr/share/sddm/ scripts/Xsetup. I haven't found any GPU switches in the UEFI. So I've tried to use configuration from [1]. And there is no any difference, still blank black screen. I can even login by blindly entering the password, KDE starts, but still screen is off. Xorg.0.log has no noticeable errors [2]. I've run from tty: `DISPLAY=:0 xranrd --listproviders` Providers: number : 2 Provider 0: id: 0x205; cap: 0x1 (Source Output); crtcs: 0; outputs: 0; associated providers: 0; name: NVIDIA-0 Provider 1: id: 0x46; cap: 0xf (Source Output, Sink Output, Source Offload, Sink Offload); crtcs: 3; outputs: 4; associated providers: 0; name: modesetting output 0x42 output 0x43 output 0x44 output 0x45 and output looks ok (needed capabilities are present) but output just numbered, there are no names eDP-1, DP-1, HDMI-1, HDMI-2. I can run `DISPLAY=:0 scrot` but got 640x480 black file, `DISPLAY=:0 glxspheres` runs and says that NVidia driver is used but got only 5 fps. Full content of xorg.conf is [3]. What I've done wrong? [1] https://us.download.nvidia.com/XFree86/Linux-x86_64/384.69/README/ randr14.html [2] http://susepaste.org/60627857 [3] http://susepaste.org/83934734 -- Kind regards, Mykola Krachkovsky -- Найкращі побажання, Микола Крачковський
03.09.2017 12:53, Mykola Krachkovsky пишет:
неділя, 3 вересня 2017 р. 11:35:23 EEST Andrei Borzenkov написано:
03.09.2017 11:26, Mykola Krachkovsky пишет:
Strange Arch wiki has nothing about that
https://wiki.archlinux.org/index.php/NVIDIA_Optimus#Using_nvidia describes exactly that.
I've done that (except "Alternative configuration") and added /usr/share/sddm/ scripts/Xsetup.
I haven't found any GPU switches in the UEFI.
So I've tried to use configuration from [1]. And there is no any difference, still blank black screen. I can even login by blindly entering the password, KDE starts, but still screen is off. Xorg.0.log has no noticeable errors [2]. I've run from tty: `DISPLAY=:0 xranrd --listproviders`
Providers: number : 2 Provider 0: id: 0x205; cap: 0x1 (Source Output); crtcs: 0; outputs: 0; associated providers: 0; name: NVIDIA-0 Provider 1: id: 0x46; cap: 0xf (Source Output, Sink Output, Source Offload, Sink Offload); crtcs: 3; outputs: 4; associated providers: 0; name: modesetting output 0x42 output 0x43 output 0x44 output 0x45
It appears that these two providers are not linked (associated providers 0). Are you sure "xrandr --setprovideroutputsource" was executed? Did you test running it manually? Usual recommendation is to follow it with "xrandr --auto" to reconfigure. I currently have modesetting/mouveau which does not offer "source output" capability so I cannot test (and it is using DRI3 anyway so does not require manual xrandr invocation). I may try to reconfigure for nvidia driver but it takes time.
and output looks ok (needed capabilities are present) but output just numbered, there are no names eDP-1, DP-1, HDMI-1, HDMI-2.
I can run `DISPLAY=:0 scrot` but got 640x480 black file, `DISPLAY=:0 glxspheres` runs and says that NVidia driver is used but got only 5 fps.
Full content of xorg.conf is [3]. What I've done wrong?
[1] https://us.download.nvidia.com/XFree86/Linux-x86_64/384.69/README/ randr14.html [2] http://susepaste.org/60627857 [3] http://susepaste.org/83934734
понеділок, 4 вересня 2017 р. 06:35:09 EEST Andrei Borzenkov написано:
It appears that these two providers are not linked (associated providers 0). Are you sure "xrandr --setprovideroutputsource" was executed? Did you test running it manually? Usual recommendation is to follow it with "xrandr --auto" to reconfigure.
I've tried to run them manually, associated providers are set to 1 for both providers but got error after `xrandr --auto`: X Error of failed request: BadMatch (invalid parameter attributes) Major opcode of failed request: 140 (RANDR) Minor opcode of failed request: 7 (RRSetScreenSize) Serial number of failed request: 31 Current serial number in output stream: 32 Will examine that problem lately. Setting Virtual size in subsection doesn't change anything (except note in Xorg.0.log). -- Kind regards, Mykola Krachkovsky -- Найкращі побажання, Микола Крачковський
On Tue, Sep 5, 2017 at 11:43 AM, Mykola Krachkovsky <w01dnick@gmail.com> wrote:
понеділок, 4 вересня 2017 р. 06:35:09 EEST Andrei Borzenkov написано:
It appears that these two providers are not linked (associated providers 0). Are you sure "xrandr --setprovideroutputsource" was executed? Did you test running it manually? Usual recommendation is to follow it with "xrandr --auto" to reconfigure.
I've tried to run them manually, associated providers are set to 1 for both providers but got error after `xrandr --auto`:
X Error of failed request: BadMatch (invalid parameter attributes) Major opcode of failed request: 140 (RANDR) Minor opcode of failed request: 7 (RRSetScreenSize) Serial number of failed request: 31 Current serial number in output stream: 32
Will examine that problem lately. Setting Virtual size in subsection doesn't change anything (except note in Xorg.0.log).
As a wild guess - modesetting driver defaults to DRI3; I presume nVidia supports DRI2 (I may be wrong here). I could not find how to force modesetting to use DRI2; but you could try Intel driver whether it works, it can also be forced to use DRI2 with Option "DRI" "2". -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
вівторок, 5 вересня 2017 р. 12:50:57 EEST Andrei Borzenkov написано:
As a wild guess - modesetting driver defaults to DRI3; I presume nVidia supports DRI2 (I may be wrong here). I could not find how to force modesetting to use DRI2; but you could try Intel driver whether it works, it can also be forced to use DRI2 with Option "DRI" "2".
Adding DRI 2 doesn't change much. For some reason `xrandr --listmonitors` says: Monitors: 0 Adding manually by `xrandr --setmonitor Monitor0 1920/344x1080/194+0+0 eDP-1-1` Works ok, but `xrandr --auto` still fails with same error and new error at beginning: Configure crtc 0 failed Changing --crtc 0/1/2 — doesn't work. xrandr --output eDP-1-1 --crtc X --primary /* other outputs --off */ Says "Configure crtc X failed", where X = 0, 1, 2 (all `xrandr --verbose` says). Adding Monitor Section and Monitor to nvidia screen changes nothing (except new Modeline in `xrandr --verbose`). Any hints? Maybe it's better to write somewhere on NVidia forum? -- Kind regards, Mykola Krachkovsky -- Найкращі побажання, Микола Крачковський
Am Sonntag, 3. September 2017, 12:53:54 schrieb Mykola Krachkovsky:
неділя, 3 вересня 2017 р. 11:35:23 EEST Andrei Borzenkov написано:
03.09.2017 11:26, Mykola Krachkovsky пишет:
Strange Arch wiki has nothing about that
https://wiki.archlinux.org/index.php/NVIDIA_Optimus#Using_nvidia describes exactly that.
I've done that (except "Alternative configuration") and added /usr/share/sddm/ scripts/Xsetup.
Please note that we use /etc/X11/xdm/Xsetup by default in openSUSE. So changing /usr/share/sddm/scripts/Xsetup will normally have no effect in openSUSE... Add you custom command(s) to /etc/X11/xdm/Xsetup instead, or configure sddm to use /usr/share/sddm/scripts/Xsetup in /etc/sddm.conf, option "DisplayCommand". (although that may have other side-effects, and that file will be overwritten by updates I think) Kind Regards, Wolfgang -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
понеділок, 4 вересня 2017 р. 11:50:22 EEST Wolfgang Bauer написано:
Please note that we use /etc/X11/xdm/Xsetup by default in openSUSE. So changing /usr/share/sddm/scripts/Xsetup will normally have no effect in openSUSE...
Thanks! -- Kind regards, Mykola Krachkovsky -- Найкращі побажання, Микола Крачковський
субота, 2 вересня 2017 р. 15:58:04 EEST Mykola Krachkovsky написано:
Hi, is there any hints to install it on "optimus" laptop? I'm aware that switching between GPU's would not work, I want to try pure nvidia on parallel installation (for tests and Vulkan).
So I've tried to follow Arch wiki recommendation for using NVidia (only) on optimus laptops [1]. After that there is no tty switching but tty7 (where X is supposed to be) is completely black, no backlight. Switching to tty1-6 works fine. Everything works except X. Logs says "(EE) NVIDIA(0): Failed to initialize the GLX module" [2]. Google didn't help much, maybe something left after bumblebee? [1] https://wiki.archlinux.org/index.php/NVIDIA_Optimus [2] http://susepaste.org/5512040 -- Kind regards, Mykola Krachkovsky -- Найкращі побажання, Микола Крачковський
Op zaterdag 2 september 2017 22:55:13 CEST schreef Mykola Krachkovsky:
субота, 2 вересня 2017 р. 15:58:04 EEST Mykola Krachkovsky написано:
Hi, is there any hints to install it on "optimus" laptop? I'm aware that switching between GPU's would not work, I want to try pure nvidia on parallel installation (for tests and Vulkan).
So I've tried to follow Arch wiki recommendation for using NVidia (only) on optimus laptops [1]. After that there is no tty switching but tty7 (where X is supposed to be) is completely black, no backlight. Switching to tty1-6 works fine. Everything works except X. Logs says "(EE) NVIDIA(0): Failed to initialize the GLX module" [2]. Google didn't help much, maybe something left after bumblebee?
[1] https://wiki.archlinux.org/index.php/NVIDIA_Optimus [2] http://susepaste.org/5512040
Reinstall the Mesa packages. -- Gertjan Lettink, a.k.a. Knurpht openSUSE Board Member openSUSE Forums Team -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
субота, 2 вересня 2017 р. 23:57:49 EEST Knurpht - Gertjan Lettink написано:
Reinstall the Mesa packages.
I've reinstalled Mesa packages, nothing changed. Reinstalled nvidia packages, glvnd — also nothing. But while reinstalling nvidia I've noticed message about alternatives. So I've checked them and found libglx.so alternative was set into manual mode to libglx_xorg, reset it to auto mode (libglx_nvidia by priority) and there is no errors in Xorg.0.log, but screen is still blank and black. -- Kind regards, Mykola Krachkovsky -- Найкращі побажання, Микола Крачковський
03.09.2017 11:18, Mykola Krachkovsky пишет:
субота, 2 вересня 2017 р. 23:57:49 EEST Knurpht - Gertjan Lettink написано:
Reinstall the Mesa packages.
I've reinstalled Mesa packages, nothing changed. Reinstalled nvidia packages, glvnd — also nothing. But while reinstalling nvidia I've noticed message about alternatives. So I've checked them and found libglx.so alternative was set into manual mode to libglx_xorg,
Apparently this is done intentionally on optimus systems: # On Optimus systems disable NVIDIA driver/libs completely by default (bnc#902667) if lspci -n | grep -e '^..:..\.. 0300: ' | cut -d " " -f3 | cut -d ":" -f1 | grep -q 8086; then /usr/sbin/update-alternatives \ --set libglx.so /usr/lib64/xorg/modules/extensions/xorg/xorg-libglx.so sed -i 's/\(^\/.*\)/#\1/g' /etc/ld.so.conf.d/nvidia-gfxG04.conf fi Ubuntu provides nvidia-prime package with tools to switch configuration between iGPU and dGPU that effectively does just change library links and adds xrandr call to X server setup.
reset it to auto mode (libglx_nvidia by priority) and there is no errors in Xorg.0.log, but screen is still blank and black.
On 03/09/17 04:30 AM, Andrei Borzenkov wrote:
03.09.2017 11:18, Mykola Krachkovsky пишет:
субота, 2 вересня 2017 р. 23:57:49 EEST Knurpht - Gertjan Lettink написано:
Reinstall the Mesa packages.
I've reinstalled Mesa packages, nothing changed. Reinstalled nvidia packages, glvnd — also nothing. But while reinstalling nvidia I've noticed message about alternatives. So I've checked them and found libglx.so alternative was set into manual mode to libglx_xorg,
Apparently this is done intentionally on optimus systems:
# On Optimus systems disable NVIDIA driver/libs completely by default (bnc#902667) if lspci -n | grep -e '^..:..\.. 0300: ' | cut -d " " -f3 | cut -d ":" -f1 | grep -q 8086; then /usr/sbin/update-alternatives \ --set libglx.so /usr/lib64/xorg/modules/extensions/xorg/xorg-libglx.so sed -i 's/\(^\/.*\)/#\1/g' /etc/ld.so.conf.d/nvidia-gfxG04.conf fi
Ubuntu provides nvidia-prime package with tools to switch configuration between iGPU and dGPU that effectively does just change library links and adds xrandr call to X server setup.
reset it to auto mode (libglx_nvidia by priority) and there is no errors in Xorg.0.log, but screen is still blank and black.
We have opensuse-prime which was ported from Ubuntu. You just have to search for the package. -- Cheers! Roman -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
04.09.2017 20:01, Roman Bysh пишет:
We have opensuse-prime which was ported from Ubuntu. You just have to search for the package.
it is not found by s.o.o. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Mon, Sep 4, 2017 at 8:40 PM, Andrei Borzenkov <arvidjaar@gmail.com> wrote:
04.09.2017 20:01, Roman Bysh пишет:
We have opensuse-prime which was ported from Ubuntu. You just have to search for the package.
it is not found by s.o.o.
It's actually suse-prime https://software.opensuse.org/package/suse-prime Robert -- http://robert.muntea.nu/ -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Mon, Sep 04, 2017 at 09:54:35PM +0300, Robert Munteanu wrote:
On Mon, Sep 4, 2017 at 8:40 PM, Andrei Borzenkov <arvidjaar@gmail.com> wrote:
04.09.2017 20:01, Roman Bysh пишет:
We have opensuse-prime which was ported from Ubuntu. You just have to search for the package.
it is not found by s.o.o.
It's actually suse-prime
Thanks. Sounds interesting! I had a quick look at home:Knurpht:Extras / suse-prime Not sure, whether the other ones found in obs package home:adra / suse-prime package home:bosim:suse-prime / suse-prime package home:bosim:suse-prime / suse-prime-alt project home:bosim:suse-prime are more current or better maintained. I'm afraid scripts need to be adjusted for libglvnd usage in TW. glx Xserver module still needs to be the appropriate one, but selecting the appropriate user libs (libGL, etc.) is now obviously handled differently, i.e. no longer by a ld.so.conf snippet /etc/ld.so.conf.d/nvidia-gfxG0*. Also we switched from "intel" to "modesetting" driver for all GPUs > GEN3. I haven't tried the package/scripts yet, since I currently have no Optimus hardware available for testing. Thanks, Stefan Public Key available ------------------------------------------------------ Stefan Dirsch (Res. & Dev.) SUSE LINUX GmbH Tel: 0911-740 53 0 Maxfeldstraße 5 FAX: 0911-740 53 479 D-90409 Nürnberg http://www.suse.de Germany --------------------------------------------------------------- SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg) --------------------------------------------------------------- -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Op maandag 4 september 2017 19:40:39 CEST schreef Andrei Borzenkov:
04.09.2017 20:01, Roman Bysh пишет:
We have opensuse-prime which was ported from Ubuntu. You just have to search for the package.
it is not found by s.o.o.
https://software.opensuse.org/package/suse-prime -- Gertjan Lettink, a.k.a. Knurpht openSUSE Board Member openSUSE Forums Team -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Knurpht - Gertjan Lettink wrote:
Op maandag 4 september 2017 19:40:39 CEST schreef Andrei Borzenkov:
04.09.2017 20:01, Roman Bysh пишет:
We have opensuse-prime which was ported from Ubuntu. You just have to search for the package. it is not found by s.o.o. https://software.opensuse.org/package/suse-prime Currently this package is only available for Leap 42.2 and Tumbleweed. It would be nice to have this also for Leap 42.3.
Greetings, Björn -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Il 05/09/2017 20:59, Bjoern Voigt ha scritto:
Knurpht - Gertjan Lettink wrote:
Op maandag 4 september 2017 19:40:39 CEST schreef Andrei Borzenkov:
04.09.2017 20:01, Roman Bysh пишет:
We have opensuse-prime which was ported from Ubuntu. You just have to search for the package. it is not found by s.o.o. https://software.opensuse.org/package/suse-prime Currently this package is only available for Leap 42.2 and Tumbleweed. It would be nice to have this also for Leap 42.3.
Greetings, Björn
At the end, I think that there are only few script, so you can extract them and move to the right position. Start here: https://build.opensuse.org/package/show/home%3AKnurpht%3AExtras/suse-prime HTH, Daniele. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Op dinsdag 5 september 2017 21:18:56 CEST schreef Daniele:
Il 05/09/2017 20:59, Bjoern Voigt ha scritto:
Knurpht - Gertjan Lettink wrote:
Op maandag 4 september 2017 19:40:39 CEST schreef Andrei Borzenkov:
04.09.2017 20:01, Roman Bysh пишет:
We have opensuse-prime which was ported from Ubuntu. You just have to search for the package.
it is not found by s.o.o.
Currently this package is only available for Leap 42.2 and Tumbleweed. It would be nice to have this also for Leap 42.3.
Greetings, Björn
At the end, I think that there are only few script, so you can extract them and move to the right position. Start here: https://build.opensuse.org/package/show/home%3AKnurpht%3AExtras/suse-prime
HTH,
Daniele. Or, branch the package to your own account and have OBS build it for 42.3 -- Gertjan Lettink, a.k.a. Knurpht
openSUSE Board Member openSUSE Forums Team -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 04/09/17 01:40 PM, Andrei Borzenkov wrote:
04.09.2017 20:01, Roman Bysh пишет:
We have opensuse-prime which was ported from Ubuntu. You just have to search for the package.
it is not found by s.o.o.
Good one :-) -- Cheers! Roman -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Op dinsdag 5 september 2017 22:38:33 CEST schreef Roman Bysh:
On 04/09/17 01:40 PM, Andrei Borzenkov wrote:
04.09.2017 20:01, Roman Bysh пишет:
We have opensuse-prime which was ported from Ubuntu. You just have to search for the package.
it is not found by s.o.o.
Good one :-) it's called suse-prime
-- Gertjan Lettink, a.k.a. Knurpht openSUSE Board Member openSUSE Forums Team -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 05/09/17 04:55 PM, Knurpht - Gertjan Lettink wrote:
Op dinsdag 5 september 2017 22:38:33 CEST schreef Roman Bysh:
On 04/09/17 01:40 PM, Andrei Borzenkov wrote:
04.09.2017 20:01, Roman Bysh пишет:
We have opensuse-prime which was ported from Ubuntu. You just have to search for the package.
it is not found by s.o.o.
Good one :-) it's called suse-prime
Your are correct. I introduced this package several months ago in a thread.. I was typing this from memory. Looks like my memory ain't what it used to be. Cheers! Roman -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 15.08.2017 14:18, Stefan Dirsch wrote:
Since today there are updated NVIDIA gfx driver RPMs available for Tumbleweed.
Great project! But I miss packages for G02 (and G03) adapters. Background: Leap does not support 32-bit systems. So you have to use Tumbleweed (if you like to stay with openSUSE). Usually the these 32-bit systems have graphics cards of former generations (e.g. I still use a GeForce 6200) and it makes no sense to invest in newer cards here. Any chance to get drivers for these cards? Regards, Steffen BTW: The repos for Leap still contain G02 and G03 packages. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Tue, Oct 24, 2017 at 02:22:02PM +0200, Steffen Sledz wrote:
On 15.08.2017 14:18, Stefan Dirsch wrote:
Since today there are updated NVIDIA gfx driver RPMs available for Tumbleweed.
Great project! But I miss packages for G02 (and G03) adapters.
Driver versions used for G02 and G03 packages (legacy drivers!) lack the support for libglvnd, which we're now using for Tumbleweed. It would be *a lot* of work trying to provide packages for them, and I'm even not sure whether it's doable at all. Currently I don't plan to invest my time into looking at this.
BTW: The repos for Leap still contain G02 and G03 packages.
We don't use libglvnd yet on Leap. That's the reason. Thanks, Stefan Public Key available ------------------------------------------------------ Stefan Dirsch (Res. & Dev.) SUSE LINUX GmbH Tel: 0911-740 53 0 Maxfeldstraße 5 FAX: 0911-740 53 479 D-90409 Nürnberg http://www.suse.de Germany --------------------------------------------------------------- SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg) --------------------------------------------------------------- -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
participants (15)
-
Andrei Borzenkov -
Bjoern Voigt -
Bruno Friedmann -
Christophe Giboudeaux -
Daniele -
Knurpht - Gertjan Lettink -
Luca Beltrame -
Malcolm -
Mykola Krachkovsky -
Rainer Klier -
Robert Munteanu -
Roman Bysh -
Stefan Dirsch -
Steffen Sledz -
Wolfgang Bauer