Bug 92172 - Intel Graphic Drivers don't detect Chrontel CH7511B DP to LVDS bridge
Summary: Intel Graphic Drivers don't detect Chrontel CH7511B DP to LVDS bridge
Status: CLOSED WONTFIX
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: unspecified
Hardware: x86-64 (AMD64) Linux (All)
: medium blocker
Assignee: Intel GFX Bugs mailing list
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2015-09-29 01:00 UTC by Adolfo
Modified: 2017-07-24 22:45 UTC (History)
2 users (show)

See Also:
i915 platform: BYT
i915 features: display/DP


Attachments
dmesg file (204.39 KB, text/plain)
2015-09-29 01:00 UTC, Adolfo
no flags Details
VBIOS dump (64.00 KB, text/plain)
2015-09-29 01:00 UTC, Adolfo
no flags Details

Description Adolfo 2015-09-29 01:00:13 UTC
Created attachment 118507 [details]
dmesg file

O.S: Ubuntu 14.10
Kernel 4.2
Intel Bay Trail I

Using DP to LVDS converter CH7511B
1) There is signal on the monitor before entering to Operating System
2) There is signal on the monitor if using Windows 7
3) There is no signal on the monitor if using Ubuntu 14.10

-xrandr shows output disconnected 

-attached dmesg with drm.debug=0xe
Comment 1 Adolfo 2015-09-29 01:00:43 UTC
Created attachment 118508 [details]
VBIOS dump
Comment 2 Jani Nikula 2015-09-29 07:59:44 UTC
Do you mean eDP to LVDS?

Please try drm-intel-nightly branch of http://cgit.freedesktop.org/drm-intel
Comment 3 Adolfo 2015-10-02 00:18:39 UTC
Hello Jani

Thanks for your comments, the CH7511B has been configured as DP port, I will reply back as soon as possible with the results of the nightly branch.

Best Regards,
Adolfo.
Comment 4 Adolfo 2015-10-09 21:35:03 UTC
Hello Jani

We have tested with the latest build but the problem persist
Comment 5 Jani Nikula 2015-10-15 09:29:44 UTC
Adolfo, from upstream perspective this is an out of the ordinary use case. If you want priority for this issue, please raise this with your usual support channels, referencing the bug. Thank you.
Comment 6 Adolfo 2015-10-15 17:25:11 UTC
Hello Jani

I'm not familiar with Bugzilla yet, I though I have set the Importance incorrectly the first time I opened the Bug report. 
I apologize for any inconvenience that I have caused for you.
Comment 7 Jim Bride 2015-10-19 23:04:20 UTC
Do you have a spec for the panel that the customer is using?  Also, has the customer engaged with Chrontel about this issue?  I notice that there is a EEPROM on the that needs to be programmed correctly with the EDID and some power-up timing / sequencing information that may need some different settings to work with Linux.  Chrontel would be the best source of information / help on the EEPROM contents.  Until we are sure that the EEPROM is being loaded with the correct information for Linux and for the specific panel the customer is using, any driver debug isn't likely to be very fruitful.
Comment 8 Adolfo 2015-10-20 12:18:02 UTC
The customer sent me the ROM with the image of the CH7511B is that useful for you.
I have requested more details about the programming, customer just changed the resolution using the Chrontel Programming Tool, so the rest of the configurations should be the default. In any case I will advice the customer to contact Chrontel for more information, is there any parameter that you are interested in?

Thanks.
Comment 9 Adolfo 2015-10-26 14:40:06 UTC
Hello

Customer used the kernel parameter nomodeset, and used xserver instead of KMS now he is getting video output on the LVDS panel that is attached to CH7511B.

Best Regards,
Adolfo.
Comment 10 Jani Nikula 2015-11-03 11:04:31 UTC
Fixed? Great news... but what was the fix?
Comment 11 Adolfo 2015-11-09 20:22:20 UTC
Hello Jani

On this case the customer didn't provide more feedback, the issue is fixed for the customer because he opted for using nomodeset and let the xserver handle the initialization. Is there any other status that would be more appropiate?
Comment 12 Jani Nikula 2015-11-10 08:31:41 UTC
(In reply to Adolfo from comment #11)
> Hello Jani
> 
> On this case the customer didn't provide more feedback, the issue is fixed
> for the customer because he opted for using nomodeset and let the xserver
> handle the initialization. Is there any other status that would be more
> appropiate?

Thanks for the follow-up. I feel "fixed" is wrong because we didn't really fix anything. I'll choose wontfix. We can revisit that and reopen if needed, but I don't think we'll work on this anymore now.


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.