Created attachment 41041 [details]
heres full dmesg when im inside X
Hello! Sorry if i made a mistake this is my first bug report.
im using 82865g chipset
i have installed kernel 18.104.22.168
lastest intel 2.13.902
slackware distro slackware-current
heres what happens when system boots up in runlevel 3 everything is okay no kms error
when inside X windows i do get
[drm:i915_report_and_clear_eir] *ERROR* EIR stuck: 0x00000010, masking [ 289.466006] render error detected, EIR: 0x00000010
after a while then when i do dmesg|grep drm nothing is returned
i dont have xorg.conf
if theres anything else that you need please let me know.
thank you very much
in attachment im sending you full dmesg log
Created attachment 41042 [details]
heres full Xorg.log
Created attachment 41044 [details]
heres full dmesg when im system boots in runlevel 3 all ok
this attachment show full dmesg when system boots in runlevel 3 theres no error untill i log in inside X
Thankyou for wasting my time.
(In reply to comment #3)
> Thankyou for wasting my time.
Could you please tell me why the error msg appears when i enter inside x?
thank you very much
Not until I've had time to analyse the information. The error indicates a page table error, but as I have neither the actual PGTBL_ER register to tell me which exact error it was, nor the contents of the various display registers along with what buffers have been allocated within the GTT, there is a lot of guesswork involved. In the meantime, grab a /sys/kernel/debug/dri/0/i915_error_state from drm-intel-next which contains additional information to diagnose this condition and *STOP* faffing around with the summary. (The summary is used by our scripts to classify bugs and for me to be able to immediately recognise bugs.)
Created attachment 41087 [details]
heres i915_error_state log
Thank you for letting me know i attached the file as requested.If theres anything else i can do to help you please let me know. Thank you very much.
I believe this is fixed by:
Author: Chris Wilson <email@example.com>
Date: Tue Dec 14 16:09:31 2010 +0000
drm/i915: Wait for vblank before unpinning old fb
Be paranoid and ensure that the vblank has passed and the scanout has
switched to the new fb, before unpinning the old one and possibly
tearing down its PTEs.
Signed-off-by: Chris Wilson <firstname.lastname@example.org>