Description of problem: I have an Intel graphic chipset and an Nvidia GTX 940 MX. Currently, Mageia uses only the Intel chipset. After trying to get the Nvidia card to work by installing bumblebee-nvidia, reboot is impossible and just freezes at some point. I'm forced to reboot in failsafe mode to remove bumblebee to get Mageia to work again. I tried to get some help on IRC, in particular from Akien, but no luck... Version-Release number of selected component (if applicable): bumblebee-nvidia 3.2.1 (15.20150120.2)
1. What is the output of lspcidrake -v | grep Card 2. Please attach the journal logs from when you tried to boot with bumblebee. If that was 4 boots ago, then run journalctl -ab -4 > journal.txt and attach journal.txt to this report. (Adjust "-4" to the needed number, "-0" is current boot, "-1" is one boot ago.)
Keywords: (none) => NEEDINFOCC: (none) => marja11Assignee: bugsquad => rverschelde
lspcidrake -v | grep Card Gives me: Card:Intel 810 and later: Intel Corporation|HD Graphics 530 [DISPLAY_VGA] (vendor:8086 device:191b subv:1558 subd:5500) (rev: 06) `lspci` also gives me: 01:00.0 3D controller: NVIDIA Corporation GM108M [GeForce 940MX] (rev a2) Also, I'm not sure if this is related, but I appear to have graphic glitches except when an external display is plugged using HDMI.
Created attachment 8732 [details] SysD journal of the failing boot
CC'ing tmb as I need some help to debug this, we tried various things on IRC but that didn't work :/
CC: (none) => tmb
Ping.
Still no news about this?
Created attachment 8818 [details] SysD journal kernel 4.4.36
Still having the freeze with kernel 4.4.36.
I just tested it also does this with package bumbblebee-nouveau
CC'ing kernel team as I have no clue how to debug this further, I can't reproduce it myself. Also reported upstream by Augier, but no answer so far: https://github.com/Bumblebee-Project/Bumblebee/issues/826
CC: (none) => kernelSee Also: (none) => https://github.com/Bumblebee-Project/Bumblebee/issues/826
Thanks to upstream comment, I managed to workaround this bug by passing `apci_osi=! acpi=Windows 2009` option to kernel.
(In reply to Augier from comment #11) > Thanks to upstream comment, I managed to workaround this bug by passing > `apci_osi=! acpi=Windows 2009` option to kernel. What's weird is that "apci_osi" doesn't seem to exist, it is not listed here: https://www.kernel.org/doc/html/latest/admin-guide/kernel-parameters.html Shouldn't that be "acpi_osi=!" ?
I have absolutely no idea. I just works. The trick is given in this comment: https://github.com/Bumblebee-Project/Bumblebee/issues/764#issuecomment-234494238
See Also: (none) => https://github.com/Bumblebee-Project/Bumblebee/issues/764#issuecomment-234494238
(In reply to Augier from comment #13) > I have absolutely no idea. I just works. The trick is given in this comment: > https://github.com/Bumblebee-Project/Bumblebee/issues/764#issuecomment- > 234494238 Marja's comment was about the command being `a*c*pi_osi`, not `a*p*ci_osi` as you wrote initially. Could you document the workaround in https://wiki.mageia.org/en/Bumblebee ?
> Marja's comment was about the command being `a*c*pi_osi`, not `a*p*ci_osi` as you wrote initially. Sure, my mistake. It's `acpi_osi=! acpi_osi="Windows 2009"`. > Could you document the workaround in https://wiki.mageia.org/en/Bumblebee ? Sure.
(In reply to Augier from comment #15) > > Marja's comment was about the command being `a*c*pi_osi`, not `a*p*ci_osi` as you wrote initially. > > Sure, my mistake. It's `acpi_osi=! acpi_osi="Windows 2009"`. > > > Could you document the workaround in https://wiki.mageia.org/en/Bumblebee ? > > Sure. That was done, thanks :-) Does this (Mageia 5) bug report need to stay open for Mageia 6 and/or later? Please close it as OLD if not (or change "Version:")
No reply, so closing as OLD
Status: NEW => RESOLVEDResolution: (none) => OLD