Bug 603386 - [Crestline] Login screen no longer cross-fades from final plymouth screen on Intel graphics
Summary: [Crestline] Login screen no longer cross-fades from final plymouth screen on ...
Alias: None
Product: Fedora
Classification: Fedora
Component: xorg-x11-drv-intel
Version: 15
Hardware: All
OS: Linux
Target Milestone: ---
Assignee: Adam Jackson
QA Contact: Fedora Extras Quality Assurance
Whiteboard: RejectedBlocker RejectedNTH
Depends On:
Blocks: F15Target
TreeView+ depends on / blocked
Reported: 2010-06-12 19:44 UTC by James
Modified: 2018-04-11 07:16 UTC (History)
8 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Last Closed: 2011-04-18 18:22:33 UTC
Type: ---

Attachments (Terms of Use)
/var/log/messages (68.43 KB, text/plain)
2010-06-15 10:33 UTC, James
no flags Details
/var/log/Xorg.0.log (58.89 KB, text/plain)
2010-06-15 10:33 UTC, James
no flags Details
dmesg output with drm.debug=0x04 (59.74 KB, text/plain)
2010-06-15 10:34 UTC, James
no flags Details
gdm :0.log (44.57 KB, text/plain)
2010-06-15 10:34 UTC, James
no flags Details
gdm :0-greeter.log (4.65 KB, text/plain)
2010-06-15 10:35 UTC, James
no flags Details
gdm :0-slave.log (98 bytes, text/plain)
2010-06-15 10:35 UTC, James
no flags Details

Description James 2010-06-12 19:44:26 UTC
Description of problem:
OK, this is only a cosmetic thing, but under F12, the login screen used to cross-fade in nicely from whatever plymouth had left at the end of the boot process. Under F13, this is also the case on my ATI graphics machine, but with Intel graphics the screen turns black and then presents the login screen. (Intel X3100 graphics.)

[Filing under Intel driver since it seems to be working on the non-Intel hardware I can lay my hands on.]

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

How reproducible:

Steps to Reproduce:
1. Start the machine.

Comment 2 Matěj Cepl 2010-06-14 09:33:11 UTC
Thanks for the bug report.  We have reviewed the information you have provided above, and there is some additional information we require that will be helpful in our diagnosis of this issue.

Please add drm.debug=0x04 to the kernel command line, restart computer, and attach

* your X server config file (/etc/X11/xorg.conf, if available),
* GDM log files (/var/log/gdm/*.log),
* X server log file (/var/log/Xorg.*.log)
* output of the dmesg command, and
* system log (/var/log/messages)

to the bug report as individual uncompressed file attachments using the bugzilla file attachment link above.

We will review this issue again once you've had a chance to attach this information.

Thanks in advance.

Comment 3 James 2010-06-15 10:33:38 UTC
Created attachment 424100 [details]

Comment 4 James 2010-06-15 10:33:58 UTC
Created attachment 424101 [details]

Comment 5 James 2010-06-15 10:34:25 UTC
Created attachment 424102 [details]
dmesg output with drm.debug=0x04

Comment 6 James 2010-06-15 10:34:46 UTC
Created attachment 424103 [details]
gdm :0.log

Comment 7 James 2010-06-15 10:35:05 UTC
Created attachment 424104 [details]
gdm :0-greeter.log

Comment 8 James 2010-06-15 10:35:31 UTC
Created attachment 424105 [details]
gdm :0-slave.log

Comment 9 James 2010-06-15 10:36:01 UTC
No xorg.conf present.

Comment 10 Adam Williamson 2010-10-01 20:33:29 UTC
This was discussed at the 2010-10-01 blocker review meeting. This is a minor cosmetic issue which does not meet any of the release criteria; you still get a functioning desktop.

Comment 11 Andrew Duggan 2010-10-03 02:41:32 UTC
The release criteria should really be amended to include any known regression  automatically as a release blocker.

See http://fedoraproject.org/wiki/Features/IntelKMS (From F11) 

Under the test plan for that feature in F11 it says this:

"... that Fedora transitions into X from plymouth without flashes or blackouts,..."

This means with this bug Fedora 14 doesn't meet the criteria under which Fedora 11 was released even though the process then was not at all formal. 

Although personally I would rather this bug be closed as wontfix as that feature (KMS) from Fedora 11 broke Hibernate on Intel for two full releases, and now that is fixed, I can't say I have a lot of confidence that a fix for this bug won't break hibernate again for Intel.

It is rather ironic that in the name of a feature (KMS in F11) that Fedora broke important functionality for laptops with Intel video (hibernate), can't pass one of the test items important enough to make it into its original test plan and now it is dismissed with a hand-wave.  -- Any you all wonder why you lose users?

Comment 12 Mads Kiilerich 2011-03-04 19:15:01 UTC
I see the same on F15 on 915GM/GMS/910GML

Shouldn't this be a F15 target?



Comment 13 Mads Kiilerich 2011-03-07 15:23:41 UTC
A couple of not intel-specific comments:

/etc/X11/prefdm still contains "plymouth quit". AFAIK that should be handled by gdm in F15.

No matter what, Xorg is started with
/usr/bin/Xorg :0 -br -verbose -auth /var/run/gdm/auth-for-gdm-x/database -nolisten tcp
AFAIK it should be started with -nr - but that has perhaps been changed to -br?

Comment 14 Adam Williamson 2011-03-07 17:02:13 UTC
"/etc/X11/prefdm still contains "plymouth quit". AFAIK that should be handled by
gdm in F15."

and if you're not using gdm?

Comment 15 Mads Kiilerich 2011-03-07 17:13:35 UTC
gdm has been patched to support smooth fading. The content of prefdm indicates that kdm also has been patched. Users using other login managers do currently not get the same smooth experience. AFAIK all this hasn't changed since F12/13.

But I guess you are right, plymouth quit without --retain-splash should still be run for "other" login managers. I didn't want to pretend that I was an expert who knows all the details of the plan.

Comment 16 Adam Jackson 2011-04-18 18:22:33 UTC
Fixed in -intel 2.14.0-5.fc15.

Note You need to log in before you can comment on or make changes to this bug.