Here is a new nvidia-current backport. This is now classified long lived stable branch beginning with 410.66 SRPMS: ldetect-lst-0.3.7.9-3.mga6.src.rpm nvidia-current-410.66-1.mga6.nonfree.src.rpm x86_64: ldetect-lst-0.3.7.9-3.mga6.x86_64.rpm ldetect-lst-devel-0.3.7.9-3.mga6.x86_64.rpm dkms-nvidia-current-410.66-1.mga6.nonfree.x86_64.rpm nvidia-current-cuda-opencl-410.66-1.mga6.nonfree.x86_64.rpm nvidia-current-devel-410.66-1.mga6.nonfree.x86_64.rpm nvidia-current-doc-html-410.66-1.mga6.nonfree.x86_64.rpm x11-driver-video-nvidia-current-410.66-1.mga6.nonfree.x86_64.rpm it adds additional support for RTX2070 cards and some Quadro cards based on Turing For all supported cards by this driver: http://us.download.nvidia.com/XFree86/Linux-x86_64/410.66/README/supportedchips.html I've already confirmed it works with a RTX2080
Mageia 6, x86_64. Works with GTX 970.
CC: (none) => tarazed25
Working OK with GeForce GTX 1080Ti.
Working fine with GTX 965M.
Also working OK on a GT 650M.
GTX 770 works with 410.66.
Hit a problem with a GTX 555. Could not reboot. Could not boot to run level 3 either. Stuck at a black screen with no interaction possible except Ctrl-Alt-Del. Gave up in the end and started a re-installation of Mageia 6. Shall try again later.
After reinstalling the system on this hardware (Alienware X51 - GTX 555) it was impossible to configure X to use nvidia - module not properly installed. To be expected as it is one of the Fermi cards. So how do we get these to switch to a driver which does support it? We are supposed to be safe until 2022. How do we get hold of the alternative driver that ldetect-lst is supposed to switch to, or maybe it is not available yet?
Would a downgrade work?
No. Downgrade did not work because there were two branches of the kernel tree. Got rid of the 4.18 kernel files and then removed dkms-nvidia-current. Reinstalled dkms-nvidia-current and evrything seemed to be working - the nvidia-410.57 driver was installed. Ran drakx11 and again thing ran fine, installation without errors apparently but then a conflicting message "The proprietary driver was not properly installed, defaulting to free software driver." Have run out of ideas. Help!
Re comment #9. # dkms status nvidia-current, 410.57-1.mga6.nonfree, 4.14.78-desktop-1.mga6, x86_64: installed vboxadditions, 5.2.18-1.mga6, 4.14.78-desktop-1.mga6, x86_64: installed-binary from 4.14.78-desktop-1.mga6 virtualbox, 5.2.18-1.mga6, 4.14.78-desktop-1.mga6, x86_64: installed-binary from 4.14.78-desktop-1.mga6 xtables-addons, 2.13-3.mga6, 4.14.78-desktop-1.mga6, x86_64: installed-binary from 4.14.78-desktop-1.mga6
(In reply to Len Lawrence from comment #7) > After reinstalling the system on this hardware (Alienware X51 - GTX 555) it > was impossible to configure X to use nvidia - module not properly installed. > To be expected as it is one of the Fermi cards. So how do we get these to > switch to a driver which does support it? We are supposed to be safe until > 2022. How do we get hold of the alternative driver that ldetect-lst is > supposed to switch to, or maybe it is not available yet? For Fermi cards, stay with the driver in nonfree/updates That is the R390 branch that will be supported until 2022 The R390 will be named nvidia390 in mga7, but I wont introduce driver name changes in a stable release. That's why I provide this one in backports mostly for RTX (Turing) support...
Does not work with GF108 [GeForce GT 730] backports update ldetect and nvidia 4.10 and stop booting on that loop (I can only reboot in level 3 to downgrade) : oct. 22 15:55:43 gaia.terre sddm[1888]: QProcess: Destroyed while process ("/usr/libexec/sddm-helper") is still running. oct. 22 15:55:43 gaia.terre audit[13112]: USER_AUTH pid=13112 uid=0 auid=4294967295 ses=4294967295 msg='op=PAM:authentication grantors=pam_permit acct="sddm" exe="/usr/libexec/sddm-helper" hostname=? addr=? terminal=? res=success' oct. 22 15:55:43 gaia.terre audit[13112]: USER_ACCT pid=13112 uid=0 auid=4294967295 ses=4294967295 msg='op=PAM:accounting grantors=pam_permit acct="sddm" exe="/usr/libexec/sddm-helper" hostname=? addr=? terminal=? res=success' oct. 22 15:55:43 gaia.terre sddm[1888]: QProcess: Destroyed while process ("/usr/libexec/sddm-helper") is still running. oct. 22 15:55:43 gaia.terre audit[13127]: USER_AUTH pid=13127 uid=0 auid=4294967295 ses=4294967295 msg='op=PAM:authentication grantors=pam_permit acct="sddm" exe="/usr/libexec/sddm-helper" hostname=? addr=? terminal=? res=success' oct. 22 15:55:43 gaia.terre audit[13127]: USER_ACCT pid=13127 uid=0 auid=4294967295 ses=4294967295 msg='op=PAM:accounting grantors=pam_permit acct="sddm" exe="/usr/libexec/sddm-helper" hostname=? addr=? terminal=? res=success' oct. 22 15:55:43 gaia.terre sddm[1888]: QProcess: Destroyed while process ("/usr/libexec/sddm-helper") is still running. oct. 22 15:55:43 gaia.terre audit[13142]: USER_AUTH pid=13142 uid=0 auid=4294967295 ses=4294967295 msg='op=PAM:authentication grantors=pam_permit acct="sddm" exe="/usr/libexec/sddm-helper" hostname=? addr=? terminal=? res=success' oct. 22 15:55:43 gaia.terre audit[13142]: USER_ACCT pid=13142 uid=0 auid=4294967295 ses=4294967295 msg='op=PAM:accounting grantors=pam_permit acct="sddm" exe="/usr/libexec/sddm-helper" hostname=? addr=? terminal=? res=success' oct. 22 15:55:43 gaia.terre sddm[1888]: QProcess: Destroyed while process ("/usr/libexec/sddm-helper") is still running. oct. 22 15:55:43 gaia.terre audit[13157]: USER_AUTH pid=13157 uid=0 auid=4294967295 ses=4294967295 msg='op=PAM:authentication grantors=pam_permit acct="sddm" exe="/usr/libexec/sddm-helper" hostname=? addr=? terminal=? res=success' oct. 22 15:55:43 gaia.terre audit[13157]: USER_ACCT pid=13157 uid=0 auid=4294967295 ses=4294967295 msg='op=PAM:accounting grantors=pam_permit acct="sddm" exe="/usr/libexec/sddm-helper" hostname=? addr=? terminal=? res=success
CC: (none) => andresalaun
(In reply to andre salaun from comment #12) > Does not work with > GF108 [GeForce GT 730] > Thats expected. GF... meand Geforce Fermi
@tmb, comment #11. Wghich driver ts that. I tried 410.57 and found that it installed but could not be used. See this extract from the journal. It looks weird to me. Oct 22 16:31:07 belexeuli urpme[12697]: called with: dkms-nvidia-current Oct 22 16:31:23 belexeuli [RPM][12697]: erase nvidia-current-doc-html-410.57-1.mga6.nonfree.x86_64: success Oct 22 16:31:23 belexeuli [RPM][12697]: erase x11-driver-video-nvidia-current-410.57-1.mga6.nonfree.x86_64: success Oct 22 16:31:23 belexeuli [RPM][12697]: erase dkms-nvidia-current-410.57-1.mga6.nonfree.x86_64: success Oct 22 16:31:24 belexeuli [RPM][12697]: erase nvidia-current-doc-html-410.57-1.mga6.nonfree.x86_64: success Oct 22 16:31:25 belexeuli [RPM][12697]: erase x11-driver-video-nvidia-current-410.57-1.mga6.nonfree.x86_64: success Oct 22 16:31:35 belexeuli [RPM][12697]: erase dkms-nvidia-current-410.57-1.mga6.nonfree.x86_64: success Oct 22 16:32:04 belexeuli urpmi[14286]: called with: --X dkms-nvidia-current Oct 22 16:33:10 belexeuli [RPM][14286]: install dkms-nvidia-current-410.57-1.mga6.nonfree.x86_64: success Oct 22 16:33:10 belexeuli [RPM][14286]: install dkms-nvidia-current-410.57-1.mga6.nonfree.x86_64: success x11-driver-video-nvidia-current Oct 22 16:41:10 belexeuli drakx11[32138]: those kernel module packages can be installed: dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current x11-driver-video-nouveau x11-driver-video-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current dkms-nvidia-current x11-driver-video-nvidia-current Oct 22 16:41:28 belexeuli [RPM][32347]: install x11-driver-video-nvidia-current-410.57-1.mga6.nonfree.x86_64: success Oct 22 16:41:28 belexeuli [RPM][32347]: install nvidia-current-doc-html-410.57-1.mga6.nonfree.x86_64: success Oct 22 16:41:28 belexeuli [RPM][32347]: install x11-driver-video-nvidia-current-410.57-1.mga6.nonfree.x86_64: success Oct 22 16:41:29 belexeuli [RPM][32347]: install x11-driver-video-nvidia-current-410.57-1.mga6.nonfree.x86_64: success Oct 22 16:41:29 belexeuli [RPM][32347]: install nvidia-current-doc-html-410.57-1.mga6.nonfree.x86_64: success Oct 22 16:41:33 belexeuli drakx11[32138]: removed files/directories /etc/ld.so.conf.d/nvidia.conf Oct 22 16:41:33 belexeuli drakx11[32138]: removed files/directories /etc/ld.so.conf.d/nvidia_legacy.conf
(In reply to Thomas Backlund from comment #13) > (In reply to andre salaun from comment #12) > > Does not work with > > GF108 [GeForce GT 730] > > > > Thats expected. GF... meand Geforce Fermi Thank you added to skip.list
I must have tried this about a dozen times this afternoon but 410.57 refuses to install properly. Everything looks normal until XFdrake is involved. That always complains that the driver module is improperly installed.
Got it. Finally figured out what must be happening - 410.57 must be coming from somewhere. I always have Backports enabled. Disabled it and tried again. This time 390.87 turned up, and that works. It does not seem desirable to have Backports disabled - so can 410.57 be removed? It is no use here.
Re comment #17. Ah. That would be a candidate for the skip.list. Unsure of the syntax though. Using /410.57/
I have tested with version 410.57 and kernel 4.18.12 and the system remains frozen. Mageia-Prime does not work and I had to return to kernel 4.14.70 and driver 390.87 that work perfectly. I'm waiting for updates on this, because I can not play games with vulkan on Steam as I've already reported on another bug. All tested with Nvidia Gt 820 M.
CC: (none) => joselp
Enough tests, validating and moved to backports
Resolution: (none) => FIXEDWhiteboard: (none) => MGA6-64-OKStatus: NEW => RESOLVEDKeywords: (none) => validated_backport