Bug 1373031

Summary: 4.7.2 kernel upgrade fails output to display with Intel GM45
Product: [Fedora] Fedora Reporter: ben <know1>
Component: xorg-x11-drv-intelAssignee: Adam Jackson <ajax>
Status: CLOSED ERRATA QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: high Docs Contact:
Priority: unspecified    
Version: 24CC: ajax, gansalmon, ichavero, itamar, james, jonathan, kernel-maint, madhu.chinakonda, mchehab, xgl-maint
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2016-10-01 15:30:20 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Attachments:
Description Flags
this is what I get after boot
none
output of journalctl -b-1
none
output of; su -c 'lspci -nn'
none
intel_reg_dumper run on kernel 4.6.7-300 none

Description ben 2016-09-04 23:33:27 UTC
Created attachment 1197728 [details]
this is what I get after boot

Description of problem:
Display has no output since upgrading to 4.7.2 kernel from 4.6.*

Version-Release number of selected component (if applicable):


How reproducible:
Every kernel boot with no edit to boot parameters 

Steps to Reproduce:
1.Boot & select 4.7.2 kernel
2.n/a
3.n/a

Actual results:
no output on display

Expected results:
RGB interface on LVDS1
(just being techie there!)


Additional info:
4.7.2 works if nomodeset is inserted to boot command (with restricted graphical result!)

Intel Corporation 82801 Mobile PCI Bridge

id:	
display:0
description:	VGA compatible controller
product:	Mobile 4 Series Chipset Integrated Graphics Controller [8086:2A42]
vendor:	Intel Corporation [8086]
physical id:	
2
bus info:	
pci@0000:00:02.0
version:	07
width:	64 bits
clock:	33MHz
capabilities:	msi pm vga_controller bus_master cap_list

journal logs available!

Comment 2 ben 2016-09-08 22:03:05 UTC
I initially put this as the driver then changed it to kernel because this is when the bug occured after receiving the update. I was (made) aware of that link (not long after) changing the Component.

Comment 3 James 2016-09-14 21:24:27 UTC
(In reply to ben from comment #2)
> I initially put this as the driver then changed it to kernel because this is
> when the bug occured after receiving the update. I was (made) aware of that
> link (not long after) changing the Component.

Have you tried the patch mention upstream at FDo? (Got things going for me.)

https://bugs.freedesktop.org/show_bug.cgi?id=97642

Comment 4 ben 2016-09-15 16:36:34 UTC
(In reply to James from comment #3)
>
> 
> Have you tried the patch mention upstream at FDo? (Got things going for me.)
> 
> https://bugs.freedesktop.org/show_bug.cgi?id=97642

What machine have you used with that patch please, before I (do the work needed to) try it out.

(haven't patched kernels succesfully before!)

Comment 5 James 2016-09-15 17:37:02 UTC
(In reply to ben from comment #4)
> (In reply to James from comment #3)
> >
> > 
> > Have you tried the patch mention upstream at FDo? (Got things going for me.)
> > 
> > https://bugs.freedesktop.org/show_bug.cgi?id=97642
> 
> What machine have you used with that patch please, before I (do the work
> needed to) try it out.
> 
> (haven't patched kernels succesfully before!)

Lenovo C50-30, Intel 5500 graphics.

Comment 6 ben 2016-09-15 22:38:32 UTC
Oh sorry for not reading! thanks for the info.
Looks like that's the only hope so far..I'll check it out as I'm still only able to use it with 4.6.7-300.. and still no luck with the Psychotic Stoned Sheep.

Comment 7 ben 2016-09-19 16:13:40 UTC
Created attachment 1202546 [details]
output of journalctl -b-1

Comment 8 ben 2016-09-19 16:16:08 UTC
Created attachment 1202548 [details]
output of; su -c 'lspci -nn'

Comment 9 ben 2016-09-19 16:23:37 UTC
Created attachment 1202549 [details]
intel_reg_dumper run on kernel 4.6.7-300

unable to run intel_reg_dumper and see output with kernel 4.7* !!

Comment 10 ben 2016-09-29 22:04:52 UTC
Am able to use with external Samsung display via the gameport/vga and have had to turn off the internal Intel i915 GM45 mobile 4, currently running Kernel-4.7.4-200 The internal display (still) only running on kernel-4.6

Comment 11 ben 2016-09-29 23:30:26 UTC
As a workaround, this works.. kinda! resolution is slightly off and destop eyecandy is limited or doesn't work?
http://forums.fedoraforum.org/showpost.php?p=1768450&postcount=2

Comment 12 ben 2016-09-30 14:13:21 UTC
Using VGA port with ext monitor & 4.7 kernel

& unplugging the plugging back the ext monitor I get this in the journal;

Sep 30 14:58:30 localhost.localdomain kdeinit5[1185]: Unable to get EDID for output "LVDS1"
Sep 30 14:58:30 localhost.localdomain kdeinit5[1185]: Failed to register device: "device id 'xrandr-LVDS1' already exists"
Sep 30 14:58:30 localhost.localdomain kdeinit5[1185]: Unable to get EDID for output "LVDS1"
Sep 30 14:58:30 localhost.localdomain kdeinit5[1185]: Failed to register device: "device id 'xrandr-LVDS1' already exists"
Sep 30 14:58:30 localhost.localdomain kdeinit5[1185]: Unable to get EDID for output "LVDS1"
Sep 30 14:58:30 localhost.localdomain kdeinit5[1185]: Failed to register device: "device id 'xrandr-LVDS1' already exists"
Sep 30 14:58:30 localhost.localdomain kdeinit5[1185]: Unable to get EDID for output "LVDS1"
Sep 30 14:58:30 localhost.localdomain kdeinit5[1185]: Failed to register device: "device id 'xrandr-LVDS1' already exists"
Sep 30 14:58:30 localhost.localdomain kdeinit5[1185]: Failed to create ICC profile on cmsCreateRGBProfile
Sep 30 14:58:30 localhost.localdomain kdeinit5[1185]: Failed to create ICC profile on cmsCreateRGBProfile
Sep 30 14:58:30 localhost.localdomain kdeinit5[1185]: Failed to create ICC profile on cmsCreateRGBProfile
Sep 30 14:58:30 localhost.localdomain kdeinit5[1185]: Failed to create ICC profile on cmsCreateRGBProfile

Comment 13 James 2016-10-01 09:41:40 UTC
Fixed for me in 4.7.5-200.fc24.x86_64...

Comment 14 ben 2016-10-01 14:49:05 UTC
Confirmed fixed with kernel-4.7.5-200
(so it was a kernel issue!!)

Comment 15 Josh Boyer 2016-10-01 15:30:20 UTC
(In reply to ben from comment #14)
> Confirmed fixed with kernel-4.7.5-200
> (so it was a kernel issue!!)

It usually is.  The bugzilla component is still correct per the link already shared.  Thank you for confirming it's fixed.