FS#63207 - kernel 5.2.1 graphics i915 detection failed - unless followed by suspending or hibernating

Attached to Project: Arch Linux
Opened by carlos (osly) - Tuesday, 16 July 2019, 11:26 GMT
Last edited by Antonio Rojas (arojas) - Tuesday, 15 October 2019, 17:23 GMT
Task Type Bug Report
Category Kernel
Status Closed
Assigned To No-one
Architecture x86_64
Severity Critical
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 1
Private No

Details

Description:


Additional info:
* kernel 5.2.1
* kernel driver in use: i915
* kernel modules: i915

Steps to reproduce:



5.2.1.arch1-1


Having a modeset on the kernel params did not resolve anything, as this just added a considerable lack of performance. This was not an option.

Detection of graphics seemed to work after suspending (hibernating) the system.

The only workaround that seemed to work in my case, was to enable log messages during initialization of the kernel parameters.

lspci

VGA compatible controller: Intel Corporation Core Processor
kernel driver in use: i915
Kernel modules: i915
This task depends upon

Closed by  Antonio Rojas (arojas)
Tuesday, 15 October 2019, 17:23 GMT
Reason for closing:  Fixed
Comment by loqs (loqs) - Tuesday, 16 July 2019, 16:19 GMT Comment by carlos (osly) - Wednesday, 17 July 2019, 00:05 GMT
it seems odd that this issue started after linux (5.1.16.arch1-1 -> 5.2.arch2-1)
and rolled over to the current kernel
Comment by loqs (loqs) - Wednesday, 17 July 2019, 00:11 GMT
Why does that seem odd to you? 5.2 was the update to a new release which introduced an issue that has not been resolved in the first stable release of the 5.2.Y series.
Comment by carlos (osly) - Wednesday, 17 July 2019, 00:29 GMT
hello. but mine, unlike whatever freeze issue was introduced on 5.2.Y series, is not non-deterministic
Comment by carlos (osly) - Thursday, 18 July 2019, 11:11 GMT
May I know the reason behind the closure request? Can I safely revert the kernel params as I originally had it so my graphics are detected as before? If not, the closure request is just compulsive.

If graphics are detected only after wakeups, then my issue can't be closed. If log messages are disabled, and my graphics are not detected, my issue can't be closed either.
Comment by carlos (osly) - Monday, 05 August 2019, 12:36 GMT Comment by carlos (osly) - Thursday, 10 October 2019, 22:13 GMT
After I submitted this closure request, I realized that with xf86-video-vesa, this is still an ongoing issue. With xf86-video-intel though, a quiet parameter works, but I've had some other glitches and performance issues. Because of this, I'd rather omit the quiet parameter with xf86-video-vesa. The issue can be closed.

Loading...