Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
No console framebuffer with linux-3.12.x and i915
View unanswered posts
View posts from last 24 hours

Goto page 1, 2  Next  
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware
View previous topic :: View next topic  
Author Message
pappy_mcfae
Watchman
Watchman


Joined: 27 Dec 2007
Posts: 5894
Location: Pomona, California.

PostPosted: Mon Dec 02, 2013 10:22 am    Post subject: No console framebuffer with linux-3.12.x and i915 Reply with quote

I have been looking, but I seem to be not looking in the right place.

I upgraded to 3.12.2 today, and thereafter, I lost my console framebuffer. The machine boots. If I allow it time to finish booting up to where I log in (about two minutes), I can log in, then start X. X comes up normally. When I exit X, my framebuffer is there. While X is running, I can use <ctl><alt><Fn>, and switch to other console sessions, and the console is there.

I have been searching and looking, but I'm not seeing anything. If anyone has a clue what the deal is, please let me know. I know that it is kernel related, as 3.11.10 boots just fine, and after installing 3.12.0, I found it has the same issue. I don't think it's a bug as much as someone changed up a few things with the intel driver code, and now, I can only get a frame buffer if I can hold on until I can log in, then hit the up arrow key once (startx).

Thanks in advance.

Cheers,
Pappy
_________________
SITE LIST:
Main: http://www.kernel-seeds.org
Mirror: http://kernel-seeds.grytpype-thynne.org/
Mirror 2: http://kernel-seeds.gentoostudio.org/
Mirror 3: http://www.elilabs.com/~pappy/
Mirror 4: http://62.3.120.142/~seeds/
Back to top
View user's profile Send private message
chithanh
Developer
Developer


Joined: 05 Aug 2006
Posts: 1561
Location: Berlin, Germany

PostPosted: Wed Dec 04, 2013 3:12 pm    Post subject: Reply with quote

Maybe CONFIG_FRAMEBUFFER_CONSOLE is set to m and the module is not loaded until X starts? If you have enabled printk timestamps, dmesg might tell at which time the framebuffer console initialized.
Back to top
View user's profile Send private message
pappy_mcfae
Watchman
Watchman


Joined: 27 Dec 2007
Posts: 5894
Location: Pomona, California.

PostPosted: Sun Dec 08, 2013 8:41 am    Post subject: Reply with quote

Nope. That's easy stuff. I think I may have an idea, though.

Cheers,
Pappy
_________________
SITE LIST:
Main: http://www.kernel-seeds.org
Mirror: http://kernel-seeds.grytpype-thynne.org/
Mirror 2: http://kernel-seeds.gentoostudio.org/
Mirror 3: http://www.elilabs.com/~pappy/
Mirror 4: http://62.3.120.142/~seeds/
Back to top
View user's profile Send private message
DepreTux
n00b
n00b


Joined: 15 Aug 2011
Posts: 8

PostPosted: Sun Dec 08, 2013 7:56 pm    Post subject: Reply with quote

Ooops, I started another thread about the same issue:

https://forums.gentoo.org/viewtopic-t-977764.html

The last kernel update (3.12.4) doesn't seem to bring any fixes to i915, so I'm not compiling it.
Back to top
View user's profile Send private message
wuzzerd
Guru
Guru


Joined: 05 Jan 2005
Posts: 430
Location: New Mexico

PostPosted: Sun Dec 08, 2013 9:09 pm    Post subject: Reply with quote

These are my settings under 3.12.3
Code:
CONFIG_X86_SYSFB=y
CONFIG_FB=y
CONFIG_FB_BOOT_VESA_SUPPORT=y
CONFIG_FB_CFB_FILLRECT=y
CONFIG_FB_CFB_COPYAREA=y
CONFIG_FB_CFB_IMAGEBLIT=y
CONFIG_FB_MODE_HELPERS=y
CONFIG_FB_TILEBLITTING=y
CONFIG_FB_VESA=y

CONFIG_DRM=y
CONFIG_DRM_KMS_HELPER=y
CONFIG_DRM_I915=y
CONFIG_DRM_I915_KMS=y
CONFIG_SND_HDA_I915=y
Code:
 x11-drivers/xf86-video-intel-2.99.903

Several times I had to block the later x11 drivers for various reasons.

Screen bleeding through.
X-server crashing on keyboard events/switching to virtual console.
And (forgot this one) xorg using my boot video size instead on size in xorg.conf. 648x480 Ha!

Hope this helps.
Back to top
View user's profile Send private message
pappy_mcfae
Watchman
Watchman


Joined: 27 Dec 2007
Posts: 5894
Location: Pomona, California.

PostPosted: Tue Dec 10, 2013 1:23 am    Post subject: Reply with quote

nomodeset also works for me, unless I want to run X after booting. One of these days, I may understand why they have to break things so badly in the kernel, then take forever to give the actual working fix. I am still playing with it, but due to that reality, I can't recommend anyone use 3.12.x kernels until they fix this issue.

Cheers,
Pappy
_________________
SITE LIST:
Main: http://www.kernel-seeds.org
Mirror: http://kernel-seeds.grytpype-thynne.org/
Mirror 2: http://kernel-seeds.gentoostudio.org/
Mirror 3: http://www.elilabs.com/~pappy/
Mirror 4: http://62.3.120.142/~seeds/
Back to top
View user's profile Send private message
pappy_mcfae
Watchman
Watchman


Joined: 27 Dec 2007
Posts: 5894
Location: Pomona, California.

PostPosted: Tue Dec 10, 2013 9:12 am    Post subject: Reply with quote

After checking the documentation that comes with the kernel, I can say that they obviously never tried it.

Adding:
Code:
append="video=intelfb:mode=1280x800-32@60.0,accel,hwcursor,vram=8"
, as recommended in the documentation, did nothing to mitigate the situation.

If I use nomodeset, I can't start X. Not only that, I get default 640x480 (stretched to hell). If I use the above, or nothing appended, as long as I allow two minutes for full boot, I can log in and up-key my way to an X session. If I make a mistake, then it takes forever to get things back to right, and only a three finger salute gets me back to where I can boot to a kernel that, um, works.

I sometimes have to wonder about the minds of those who do kernel development. Do they harbor some strange sadistic streak, or what?

Cheers,
Pappy
_________________
SITE LIST:
Main: http://www.kernel-seeds.org
Mirror: http://kernel-seeds.grytpype-thynne.org/
Mirror 2: http://kernel-seeds.gentoostudio.org/
Mirror 3: http://www.elilabs.com/~pappy/
Mirror 4: http://62.3.120.142/~seeds/
Back to top
View user's profile Send private message
DepreTux
n00b
n00b


Joined: 15 Aug 2011
Posts: 8

PostPosted: Thu Dec 12, 2013 8:16 pm    Post subject: Reply with quote

I just upgraded to 3.12.5 and I'm still experiencing the same regression.

But my problem differs from yours in that if I do a startx on the blank screen, I only get a the pointer to show, but the rest of my wm doesn't display.

Maybe we should try the mailing list?
Back to top
View user's profile Send private message
pappy_mcfae
Watchman
Watchman


Joined: 27 Dec 2007
Posts: 5894
Location: Pomona, California.

PostPosted: Tue Dec 17, 2013 8:35 am    Post subject: Reply with quote

I've started a bug report on this at kernel.org. I added a link to this thread.

Cheers,
Pappy
_________________
SITE LIST:
Main: http://www.kernel-seeds.org
Mirror: http://kernel-seeds.grytpype-thynne.org/
Mirror 2: http://kernel-seeds.gentoostudio.org/
Mirror 3: http://www.elilabs.com/~pappy/
Mirror 4: http://62.3.120.142/~seeds/
Back to top
View user's profile Send private message
TheNiceGuy
n00b
n00b


Joined: 10 Mar 2014
Posts: 3

PostPosted: Mon Mar 10, 2014 8:39 pm    Post subject: Reply with quote

Hi, first post here.

I have the same problem. I search everywhere and I couldn't find anything. Like OP, my computer boots up normally but there is no framebuffer. The strange thing is that it will happen sometimes, but most of the time. I can log in, and startx blindly. If I close the xserver the framebuffer is there. I tested a couple of boot kernel parameters, it didn't work. I tried playing with the brightness, it didn't do anything.

I noticed that you made a bug repport and the dev were asking for a kernel bisect. It seems that nobody did it yet. So I decided to give it a try, my first bisect.
Gentoo's wiki had a great page on how to bisect the kernel: Kernel git-bisect I did the bisect from 3.11 to 3.12.

After a couple of kernel compilling and testing, git bisect gave me the following bad revision:27c053aa8d18d1fa7b83041e36bad20bcdf55514. Which didn't make any sens since it doesn't seems to touch the i915 driver, which is what I have. I tested the kernel, the problem was still there, as expected.

I checked the commit under it, a09e9a7a4b907f2dfa9bdb2b98a1828ab4b340b2, bingo, there were some pretty big changes on the i915 driver. I tested it, the problem was still there.

If that commit made the problem, the one under it should work, right? I tested the commit 9ab073bc45b8b523cc39658925bb44bef35ca657 and the problem wasn't there. There is most likely a problem in the commit above.

That was the kernel's config I use: .config. If you try both of these commit and tell us if it work that might help to fix this problem.

I hope it's been usefull.

EDIT: typos
Back to top
View user's profile Send private message
DepreTux
n00b
n00b


Joined: 15 Aug 2011
Posts: 8

PostPosted: Mon Mar 10, 2014 9:33 pm    Post subject: Reply with quote

Did you try with the more recent 3.13 kernels yet?

I actually went back to 3.2, since I was trying to upgrade from that version to the latest lts...

I will do the bisect and report back ASAP.
Back to top
View user's profile Send private message
TheNiceGuy
n00b
n00b


Joined: 10 Mar 2014
Posts: 3

PostPosted: Mon Mar 10, 2014 9:51 pm    Post subject: Reply with quote

Yes, I've been running on 3.13.5 even if the problem is there. I even tried the lastest git kernel.
Back to top
View user's profile Send private message
ff11
n00b
n00b


Joined: 10 Mar 2014
Posts: 12

PostPosted: Mon Mar 10, 2014 10:42 pm    Post subject: Reply with quote

One bug that crush my hope! I tried the 3.13.6, and it still there.
The 3.11.10 is in EOL, then it has removed from sys-kernel/gentoo-sources.
Maybe we have to go forever with the longterm 3.10.X
I have never felt so sad with linux before.

I think this is related with the "Splitting DRM and KMS device nodes" from the 3.12 (ref: https://dvdhrm.wordpress.com/2013/09/01/splitting-drm-and-kms-device-nodes/ )
Back to top
View user's profile Send private message
pappy_mcfae
Watchman
Watchman


Joined: 27 Dec 2007
Posts: 5894
Location: Pomona, California.

PostPosted: Mon Mar 10, 2014 11:01 pm    Post subject: Reply with quote

Well, I'm glad someone is doing something with this bug. The bug at kernel-dot-org is going nowhere. I have not tried the latest 3.13.x, but I know that 3.13.0 is still a no go. I suppose I could mess around and see if the latest version, and see what happens.

As for a sad state of affairs, yes. Because XP is about to die, the windows side of this machine is practically dead. I could set it back up for Vista, but I think I'd rather chew glass. Yeah, it has more memory and a bigger hard drive, but it's Vista.

Now, even Linux is turning this machine obsolete. That totally sucks. I guess it's time to bite the bullet, and upgrade.

Bummer!

Cheers,
Pappy
_________________
SITE LIST:
Main: http://www.kernel-seeds.org
Mirror: http://kernel-seeds.grytpype-thynne.org/
Mirror 2: http://kernel-seeds.gentoostudio.org/
Mirror 3: http://www.elilabs.com/~pappy/
Mirror 4: http://62.3.120.142/~seeds/
Back to top
View user's profile Send private message
Anon-E-moose
Advocate
Advocate


Joined: 23 May 2008
Posts: 2101
Location: Dallas area

PostPosted: Mon Mar 10, 2014 11:12 pm    Post subject: Reply with quote

Is this what you're seeing? https://bbs.archlinux.org/viewtopic.php?id=173914

If so the last post is interesting. Not sure whether it will fix anything as I don't use intel graphics.
_________________
Asus m5a99fx, FX 8320 - amd64-multilib, 3.9.1-zen, eudev
glibc-2.17, gcc-4.7.3-r1, xorg-server-1.15, lxde, nouveau, oss4
Back to top
View user's profile Send private message
ff11
n00b
n00b


Joined: 10 Mar 2014
Posts: 12

PostPosted: Mon Mar 10, 2014 11:38 pm    Post subject: Reply with quote

Anon-E-moose wrote:
Is this what you're seeing? https://bbs.archlinux.org/viewtopic.php?id=173914

If so the last post is interesting. Not sure whether it will fix anything as I don't use intel graphics.

Yes and No! The result of black screen is kind the same and work on 3.11.X like this bug, but my hardware have just 1 GPU (intel gen3 onboard card only), then I don't have issue with radeon and I don't have "video.backlight=vendor" kernel parameter.

I'm searching for the bug, doing some tests. Thanks God, i have knowledge for do it. But I will have to spend a lot of time for what I have seen of this bug (Oh! My headache!).
Back to top
View user's profile Send private message
Anon-E-moose
Advocate
Advocate


Joined: 23 May 2008
Posts: 2101
Location: Dallas area

PostPosted: Tue Mar 11, 2014 12:06 am    Post subject: Reply with quote

If it has to do with the backlight it may not have anything to do with the kernel parameters per se.

One way would be to take it out of config, recompile and see what happens.

I have these options on (don't know why, but they're on)
Code:
CONFIG_DRM_NOUVEAU_BACKLIGHT=y
CONFIG_FB_BACKLIGHT=y
CONFIG_BACKLIGHT_LCD_SUPPORT=y
CONFIG_BACKLIGHT_CLASS_DEVICE=y


Anyway, hope y'all solve it.
_________________
Asus m5a99fx, FX 8320 - amd64-multilib, 3.9.1-zen, eudev
glibc-2.17, gcc-4.7.3-r1, xorg-server-1.15, lxde, nouveau, oss4
Back to top
View user's profile Send private message
ff11
n00b
n00b


Joined: 10 Mar 2014
Posts: 12

PostPosted: Tue Mar 11, 2014 12:17 am    Post subject: Reply with quote

Anon-E-moose wrote:
If it has to do with the backlight it may not have anything to do with the kernel parameters per se.

One way would be to take it out of config, recompile and see what happens.

I have these options on (don't know why, but they're on)
Code:
CONFIG_DRM_NOUVEAU_BACKLIGHT=y
CONFIG_FB_BACKLIGHT=y
CONFIG_BACKLIGHT_LCD_SUPPORT=y
CONFIG_BACKLIGHT_CLASS_DEVICE=y


Anyway, hope y'all solve it.


Well, it's not this kind of bug. I have spend my whole day rebuilding kernel, search logs, kernel codes, making tests...

After the system boot, if i type the login, pass and start the xorg (even without see the screen), I can see the pointer of the mouse very well and clear.
Back to top
View user's profile Send private message
till
n00b
n00b


Joined: 19 Sep 2007
Posts: 14

PostPosted: Tue Mar 11, 2014 12:28 am    Post subject: Reply with quote

disabling CONFIG_X86_SYSFB did the trick for me.
_________________
Gruß Till
Back to top
View user's profile Send private message
ff11
n00b
n00b


Joined: 10 Mar 2014
Posts: 12

PostPosted: Tue Mar 11, 2014 12:33 am    Post subject: Reply with quote

till wrote:
disabling CONFIG_X86_SYSFB did the trick for me.


"CONFIG_X86_SYSFB is not set" for me -_-"
By the way, if you are using the KMS you should disable all frame buffer devices drivers, including VGA, Intel, nVidia, and ATI. It's the normal.
Back to top
View user's profile Send private message
ff11
n00b
n00b


Joined: 10 Mar 2014
Posts: 12

PostPosted: Tue Mar 11, 2014 1:28 am    Post subject: Reply with quote

Before anyone ask, I also tried force the brightness to the max using setpci direct on the hardware, but he result is just one high brightness black screen -_-"
Back to top
View user's profile Send private message
DepreTux
n00b
n00b


Joined: 15 Aug 2011
Posts: 8

PostPosted: Tue Mar 11, 2014 6:24 am    Post subject: Reply with quote

The problem is gone for me with 3.12.14

I did turn on the PCI_QUIRKS option, so it might have been just that.
Back to top
View user's profile Send private message
ff11
n00b
n00b


Joined: 10 Mar 2014
Posts: 12

PostPosted: Tue Mar 11, 2014 2:23 pm    Post subject: Reply with quote

DepreTux wrote:
The problem is gone for me with 3.12.14

I did turn on the PCI_QUIRKS option, so it might have been just that.


This bug is not about the PCI_QUIRKS option. PCI_QUIRKS is always on at my kernel's configs (even the older config I have, case the doc help have: "Disable this only if your target machine is unaffected by PCI quirks", then I never disable it)

Or the 3.12.14 fix, or is not the same bug.
Well, because be stuck on the 3.12.(14-X) isn't the end of the problem, I will test new codes only on the last stable releases (now is 3.13.6, and don't work). And I will be waiting for some more releases before came back to try solve this. In this time, I'm using the 3.10.33.
Back to top
View user's profile Send private message
ff11
n00b
n00b


Joined: 10 Mar 2014
Posts: 12

PostPosted: Sat Mar 15, 2014 6:06 pm    Post subject: Reply with quote

Because we don't have kernel > 3.13.6 yet, I have tested the 3.12.14:
1* After install and reboot, it's work! ^^
2* I reboot 1 more time to confirm, and it don't work anymore. I rebooted more times and keep don't working -_-"
3* I rebooted without boot splash, then worked! ^^
4* I rebooted without boot splash for confirm, and it don't work anymore. I rebooted more times without boot splash and keep don't working -_-" (sometimes, 10% of the tests, it's work in this case)
5* I removed all boot splash things from initramfs, and it keep don't working.
6* I don't know what to do using just the free little time, then I'm back to 3.10.33

I'm not sure, but I feel that this bug is happening just on gen3 intel GPU only (915G/GM, 945G/GM, G/Q33). The other generations seems just have other minor bugs reported (more easy to solve).
Back to top
View user's profile Send private message
TheNiceGuy
n00b
n00b


Joined: 10 Mar 2014
Posts: 3

PostPosted: Sat Mar 15, 2014 9:01 pm    Post subject: Reply with quote

This is pretty much the same here, it will work sometimes, but it mostly doesn't work.
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware All times are GMT
Goto page 1, 2  Next
Page 1 of 2

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum