Bug 103033 - [NVA8] nouveau can't start my discrete nvidia card (BIOS / ACPI error)
Summary: [NVA8] nouveau can't start my discrete nvidia card (BIOS / ACPI error)
Status: NEW
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/nouveau (show other bugs)
Version: unspecified
Hardware: Other All
: medium normal
Assignee: Nouveau Project
QA Contact: Xorg Project Team
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2017-09-28 22:07 UTC by Alexandre Dutriaux
Modified: 2017-09-28 22:39 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments
dmesg with nouveau.debug=trace + acpidump + xorg.conf (100.92 KB, application/zip)
2017-09-28 22:07 UTC, Alexandre Dutriaux
no flags Details

Note You need to log in before you can comment on or make changes to this bug.
Description Alexandre Dutriaux 2017-09-28 22:07:27 UTC
Created attachment 134558 [details]
dmesg with nouveau.debug=trace + acpidump + xorg.conf

I've got a portable computer with an Intel graphic card and a nvidia discrete card (G 105M).
I've installed Fedora 26. 
Everything should run fine with nouveau and Prime.

However, xrandr --listproviders only send me 1 provider, which is the Intel card. 

I've seen a problem with the BIOS and the nvidia card : 

[    2.777871] nouveau 0000:01:00.0: Invalid PCI ROM header signature: expecting 0xaa55, got 0x0000
[    2.777887] nouveau 0000:01:00.0: bios: unable to locate usable image
[    2.777895] nouveau 0000:01:00.0: bios ctor failed, -22

-----------
$ lspci |grep -E "3D|VGA"
00:02.0 VGA compatible controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07)
01:00.0 VGA compatible controller: NVIDIA Corporation GT218M [GeForce G 105M] (rev a2)

$ xrandr --listproviders
Providers: number : 1
Provider 0: id: 0x46 cap: 0xb, Source Output, Sink Output, Sink Offload crtcs: 3 outputs: 4 associated providers: 0 name:Intel
Comment 1 Ilia Mirkin 2017-09-28 22:39:48 UTC
The issue is that shadow_fetch() fails for ACPI. It tries to read a 0xf800-sized image and apparently can't. Both the fast and slow methods end up the same. Also it's NPDE is set to "HYB$", although looking at the code that shouldn't really matter.

However something's there, since it gets far enough to e.g. check the NPDE. Either the _ROM method is buggy, or we're driving it slightly incorrectly.


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.