I guess this means it’s installed on my machine, but for some reason isn’t finding/loading that hybrid_drv_video.so file?
~> sudo zypper se intel-hybrid-driver
S | Name | Summary | Type
---+---------------------+----------------------------------------------+--------
| intel-hybrid-driver | VA driver for Intel G45 & HD Graphics family | package
@pattim The next step is to install it and see if that suffices. sudo zypper in intel-hybrid-driver
I still think that you should take steps to get your hybrid graphics working as it was intended. You could start a new topic for that (with suitable descriptive title) when ready.
This seems less informative. I double clicked on MPlayer to go to fullscreen at 11:02:15
Jan 06 11:02:16 linux-lhkc polkit-kde-authentication-agent-1[4016]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc kglobalaccel5[3981]: The X11 connection broke: I/O error (code 1)
Jan 06 11:02:16 linux-lhkc akonadi_followupreminder_agent[4387]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc akonadi_ical_resource[4388]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc akonadi_birthdays_resource[4385]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc kded5[3959]: The X11 connection broke: I/O error (code 1)
Jan 06 11:02:16 linux-lhkc kaccess[4168]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc akonadi_control[4307]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc xembedsniproxy[4019]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc gmenudbusmenuproxy[4015]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc kwin_x11[3960]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc kscreen_backend_launcher[4163]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc pulseaudio[4195]: X11 I/O error exit handler called, preparing to tear down X11 modules
Jan 06 11:02:16 linux-lhkc pulseaudio[4195]: X11 I/O error handler called
Jan 06 11:02:16 linux-lhkc kwalletd5[3846]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc kglobalaccel5[3981]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc kalendarac[4170]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc akonadi_maildir_resource[4390]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc akonadi_akonotes_resource[4383]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc python3[4139]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc kded5[4098]: X connection to :0 broken (explicit kill or server shutdown).
Jan 06 11:02:16 linux-lhkc akonadi_contacts_resource[4386]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc kded5[3959]: after 106 requests (106 known processed) with 0 events remaining.
Jan 06 11:02:16 linux-lhkc kded5[3959]: XIO: fatal IO error 2 (No such file or directory) on X server ":0"
Jan 06 11:02:16 linux-lhkc kglobalaccel5[3981]: after 8 requests (8 known processed) with 0 events remaining.
Jan 06 11:02:16 linux-lhkc kglobalaccel5[3981]: XIO: fatal IO error 25 (Inappropriate ioctl for device) on X server ":0"
Jan 06 11:02:16 linux-lhkc ksmserver[4332]: gkrellm: Fatal IO error 104 (Connection reset by peer) on X server :0.
Jan 06 11:02:16 linux-lhkc dolphin[4779]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc akonadi_indexing_agent[4389]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc akonadiserver[4344]: org.kde.pim.akonadiserver: Subscriber "SpecialCollectionsMonitor - 94750043937808" disconnected
Jan 06 11:02:16 linux-lhkc akonadiserver[4344]: org.kde.pim.akonadiserver: Subscriber "Archive Mail Kernel ETM - 94520373393200" disconnected
Jan 06 11:02:16 linux-lhkc akonadi_migration_agent[4406]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc akonadi_maildispatcher_agent[4402]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc akonadiserver[4344]: org.kde.pim.akonadiserver: Subscriber "SpecialCollectionsMonitor - 94407153947584" disconnected
Jan 06 11:02:16 linux-lhkc akonadi_mailmerge_agent[4404]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc akonadi_notes_agent[4408]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc akonadi_newmailnotifier_agent[4407]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc akonadi_mailfilter_agent[4403]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc akonadi_unifiedmailbox_agent[4411]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc akonadi_archivemail_agent[4384]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:16 linux-lhkc akonadi_sendlater_agent[4410]: The X11 connection broke (error 1). Did the X11 server die?
Jan 06 11:02:06 linux-lhkc kioslave5[4968]: libpng warning: iCCP: known incorrect sRGB profile
Jan 06 11:02:03 linux-lhkc systemd[2302]: Started SMPlayer - Media Player.
S | Name | Summary | Type
---+---------------------+----------------------------------------------+--------
i+ | intel-hybrid-driver | VA driver for Intel G45 & HD Graphics family | package
Hi Deano! No, MPlayer and VLC (and youtube, etc.) play fine on laptop screen and external monitor. The problem happens when I use the external monitor (which is normal since laptop screen is small). When I double click to switch one of these vids players to full screen (or on youtube hit the expand-to-fs arrows) then X breaks. Note: this problem is new to a fresh install upgrade to 15.6.
I wind up logged-off, and many the apps are not restarted when I log back in.
I finally succeeded in capturing the error display which flashed briefly moments after clicking to full-screen, by recording a video with my cellphone and pulling out a screen capture. So apparently there’s a memory issue with the way LEAP set itself up during boot (stolen memory) that’s new to 15.6? This kills Nouveau (which is the video driver I’ve used for the last several versions of LEAP)?
Also, I just now tried running a fullscreen video game (Descent II, from the games repo) and it did the same thing. The switch from KDE to full screen triggers that error and forcibly logs me off.
I’m really not sure what to advise here. I assume that htis has nothing to do with the desktop environment, and more to do with using the nouveau driver, and the hybrid graphics architecture. Is hybrid or discrete mode (or similar) set in the BIOS? I’m out of my comfort zone with hybrid graphics, so hopefully someone with similar hardware can advise further.
BTW, is secure boot enabled or disabled currently? (This may be good to know prior to installing/using the nvidia driver.)
Not sure… I think secure boot is set (or unset) by the installer during install? Or should I be searching BIOS for that? If it’s a “stolen memory” (whatever that is in linux) conflict, then it “seems” as fundamental as a boot issue…?
The only thing I really did differently at the most recent install of 15.6 is I switched to XFS from BtrFS… but I don’t see how that would affect video.
Set in BIOS (but unrelated to the error message you have reported. It will impact if/when you switch to nvidia driver. You posted this issue in the ‘Multimedia’ forum category. I recommend starting a new topic in the ‘Hardware’ forum category with a suitable descriptive title. Share the graphics card results from inxi -Ga, and laptop model etc. Even though this manifests when playing video content (full screen) or when connecting to an external monitor via HDMI output, I think it needs to be treated as a ‘hybrid graphics issue’, so you need support for getting that configured appropriately.
OK. Will do. I have never used hybrid graphics on this box before 15.6, just the default install, which I assume goes under the name “Nouveau” - by that I mean I never tried the bumblebee (I think that’s the right word).
But I will move this to the recommended forum. Thank you!!
I assume that you meant that you’ve never undertaken the steps required to take advantage of the hybrid graphics implementation that your laptop comes with. You have an Intel graphics chipset (part of the laptop’s CPU architecture), and a more powerful NVIDIA graphics dedicated GPU. I’m not sure how old your laptop is, so the hybrid technology may vary slightly with newer hybrid implementations. Maybe you were only using the Intel graphics, and the NVIDIA GPU remained effectively dormant. In any case the xrandr output you shared showed the laptop outputs connected like this
Yes, that’s what I meant. I suppose hardware and code change with time. I guess the default install now sets things up video-wise in a way that’s no longer fully compatible with my old system (purchased ~2017). I know the hybrid character of my hardware is the most obvious thing to notice, but I’m guessing, unless the newest code (15.6) now enforces switching on such hardware, that it’s not related to switching between intel and nvidia, mostly because everything works fine on the laptop screen. Thanks for the suggestion - I’ll try to decode those xrandr outputs and how they might relate to the link you gave me.
EDIT: Well I found a fix - from the link deano gave me (THANK YOU!!) I saw that it was an option to disable one of the two vid systems via BIOS - so I checked there and disabled “video switching.” After reboot, things seemed to be fixed except for significant flicker on the external monitor - changing refresh rate didn’t affect this. Installing the Nvidia G05 video driver from the NVidia repo mostly fixed the flicker. This is the first LEAP version for which I’ve had to change any of this; but fullscreen video and games work now on 15.6.
I also tried logging-in using Weyland but that eats up almost 100% of CPU time, so I switched back to X11.
I’m tempted to try installing the Nvidia G06 driver to see if that fixes the small amount of flicker on the external monitor, but I’m worried that, if incompatible, I could brick my system. Can anyone tell what driver I should be using from the info I supplied via xrandr? Quadro K610m
Compatible NVIDIA drivers for particular hardware generations are described here https://en.opensuse.org/SDB:NVIDIA_drivers#Determination_of_driver_version
I’m don’t have NVIDIA hardware, but checking the NVIDIA driver search engine (linked there), yields the following results for your hardware…
Driver Version:390.157 Release Date:Tue Nov 22, 2022
I understand that that represents the G04 driver, but open to correction about that. I’m sure that others familiar with such hardware could advise further here.
I assume you could install it the hard way (as described in the wiki), but as you mentioned only the later drivers are built and provided for Leap 15.6
Thanks again, Deano - I’m always looking for page dates to be sure the information isn’t anachronistic, but I’ll assume this is right. Thank you for that link!!!
I used to do things the “hard” way (using the nvidia***.run file) when I was on a tower but haven’t attempted it on a laptop. Now I know that I have to do the extra steps of disabling intel and doing the runfiles. (That means I’m not using Nouveau, any more, right? Just making sure I understand…)
Still, something changed 15.5 → 15.6 (which we won’t find out here) that caused me to need to disable the intel video. 16 is supposed to be a radical change? I saw that SLE 15 was supposed to last into the 2030’s? Is tumbleweed a bad idea for “older” hardware like mine?
**Warning:** Legacy drivers G03 and G04 won't work with the actual kernel 6.6.x series. Use free nouveau drivers instead!
…that suggests to me it’s a bad idea to use NVidia drivers as linux continues to change radically. That’s why I wanted to trouble-shoot the Nouveau error I mentioned earlier… I think Nouveau is being updated with the kernels.