LinuxQuestions.org
Page 1 of 2 1 2
Show 50 post(s) from this thread on one page

LinuxQuestions.org (/questions/)
-   Slackware (https://www.linuxquestions.org/questions/slackware-14/)
-   -   Problem with 2.6.36 and 2.6.37 kernels on EeePC 1005HA, Slackware 13.1 (https://www.linuxquestions.org/questions/slackware-14/problem-with-2-6-36-and-2-6-37-kernels-on-eeepc-1005ha-slackware-13-1-a-854970/)

Ilgar 01-08-2011 04:35 AM
Problem with 2.6.36 and 2.6.37 kernels on EeePC 1005HA, Slackware 13.1
 
Hi all,

I have an Asus EeePC 1005HA (with i915 graphics) running on Slackware 13.1. I wonder if anyone is able to use this combination with a 2.6.36 or 2.6.37 kernel.

I am now using my custom-compiled 2.6.35.6 kernel, which is the last problem-free version among those that I have tried. The computer was freezing (at seemingly random times) with 2.6.36 and 2.6.36.2, so I decided to wait for the next major version. Yesterday I tried 2.6.37, and this time my screen goes black after leaving the computer idle for a while. It's probably the monitor powersave kicking in (I've got XFCE power manager running), but the screen doesn't wake up. It is not a hard freeze however, the system shuts down normally if I press the power button.

So, is there anyone who can use these kernels on a comparable system?

markush 01-08-2011 05:06 AM
Hello Ilgar,

some thougths: freezing often is caused by Ram-problems. You may check the Ram with an adequate tool (I use Slax).

Did you experience this problems before you installed Slackware, or is this the first installation on a new machine?

Did you change somthing with the hardware?

Markus

Ilgar 01-08-2011 06:11 AM
Hi markush,

No, the hardware is the same, as are the kernel settings (I do make oldconfig to carry the old settings over to the new kernel). I believe the RAM is fine, I get the lockups with 2.6.36.x only, not with .35.x or .37. I don't have any of these issues with 2.6.35.6 or Windows. I have been running Slack on this machine all the time, all was good so far.

The only other oddity I can think of is a screen corruption I get when I visit a certain web page with flash, using Firefox. I noticed this just recently, but the bug has probably been there for much longer (it is very likely related to Flash and not to a problem with the i915 chip. Konqueror seems to work fine on the same page).

guanx 01-08-2011 06:56 AM
What about upgrading the intel video driver? It's available at http://xorg.freedesktop.org/archive/individual/driver/ under the name xf86-video-intel-*.

Ilgar 01-08-2011 07:16 AM
Thank you guanx, I will try doing that as a last resort. It will require a new version of libdrm at least, and I want to stick to the stock packages as much as possible. By the way, I realized that I wasn't very precise about my graphics chipset, it is an

Code:
Intel Corporation Mobile 945GM/GMS/GME, 943/940GML Express Integrated Graphics Controller (rev 03).

brixtoncalling 01-08-2011 08:34 AM
Ilgar, after you reboot to the command line check /var/log/Xorg.0.log. I've noticed a similar freeze when Xscreensaver runs or I put flash in full screen.

Here is an excerpt from mine after this happens:
Code:
[ 74536.527] 0: /usr/bin/X (xorg_backtrace+0x3b) [0x80e6d0b]
[ 74536.527] 1: /usr/bin/X (0x8048000+0x5daf5) [0x80a5af5]
[ 74536.527] 2: (vdso) (__kernel_rt_sigreturn+0x0) [0xffffe40c]
[ 74536.527] 3: /usr/lib/libdrm_intel.so.1 (0xb73fb000+0x53eb) [0xb74003eb]
[ 74536.527] 4: /usr/lib/libdrm_intel.so.1 (0xb73fb000+0x5645) [0xb7400645]
[ 74536.527] 5: /usr/lib/libdrm_intel.so.1 (drm_intel_bo_exec+0x2e) [0xb73fc77e]
[ 74536.527] 6: /usr/lib/xorg/modules/drivers/intel_drv.so (0xb7393000+0x764b) [0xb739a64b]
[ 74536.527] 7: /usr/lib/xorg/modules/drivers/intel_drv.so (0xb7393000+0xbbaf) [0xb739ebaf]
[ 74536.527] 8: /usr/bin/X (CallCallbacks+0x56) [0x8074ae6]
[ 74536.527] 9: /usr/bin/X (FlushAllOutput+0x38) [0x80a46a8]
[ 74536.527] 10: /usr/bin/X (0x8048000+0x27d80) [0x806fd80]
[ 74536.527] 11: /usr/bin/X (0x8048000+0x1aec5) [0x8062ec5]
[ 74536.527] 12: /lib/libc.so.6 (__libc_start_main+0xe6) [0xb749bc66]
[ 74536.527] 13: /usr/bin/X (0x8048000+0x1aaa1) [0x8062aa1]
[ 74536.527]
Fatal server error:
[ 74536.527] Caught signal 3 (Quit). Server aborting

guanx 01-08-2011 08:53 AM
Quote:
Originally Posted by Ilgar (Post 4217328)
Thank you guanx, I will try doing that as a last resort. It will require a new version of libdrm at least, and I want to stick to the stock packages as much as possible. By the way, I realized that I wasn't very precise about my graphics chipset, it is an

Code:
Intel Corporation Mobile 945GM/GMS/GME, 943/940GML Express Integrated Graphics Controller (rev 03).
I am using an AOpen MoDT with exactly the same chipset.
Code:
00:02.0 VGA compatible controller: Intel Corporation Mobile 945GM/GMS, 943/940GML Express Integrated Graphics Controller (rev 03)
Many video bugs disappeared after compiling and installing the new libdrm and intel video driver.
.

Ilgar 01-08-2011 10:47 AM
@brixtoncalling: With 2.6.37, when the screen goes blank, I do not see anything wrong in the logs. I'm now sure that it is the screensaver turning it off, but it doesn't wake up. It is not a freeze either, as I can "blindly" switch to a console and run commands. I was getting a real freeze with 2.6.36 but I removed that kernel a while ago and cannot test is now.

@guanx: While the bugs referenced in the driver changelog do not seem to apply to my situation, I am going to update libdrm and the driver. Let's hope that it will improve things over here, as it did for you. I will report back here.

Thank you all...

Ilgar 01-08-2011 11:18 AM
Unfortunately upgrading libdrm and the Intel driver didn't work: The screen still doesn't wake up after screensaver turns it off, and I can not see anything suspicious in the X log.

brixtoncalling 01-08-2011 11:22 AM
Ilgar, it seems our problems are slightly different. Mine also is not a hard freeze -- sometimes I can switch to console and even if I can't, the power button causes X to close and the normal shutdown to commence. I'm also trying a different kernel, but the freeze on my computer only occurs occasionally so I have to wait to see if a new kernel will help.

guanx 01-08-2011 11:26 AM
Quote:
Originally Posted by Ilgar (Post 4217500)
Unfortunately upgrading libdrm and the Intel driver didn't work: The screen still doesn't wake up after screensaver turns it off, and I can not see anything suspicious in the X log.
Sorry to hear that. But as a bug it's well behaved -- can be reproduced easily.

Ilgar 01-08-2011 12:25 PM
@brixtoncalling: Did you try guanx's suggestion to upgrade libdrm and drivers? In the changelog I've seen some references to bugs fitting your description.

I decided to keep the new libdrm and intel driver installed. Since everything works well with 2.6.35.6, I believe there must be a kernel element in all this. I hope they fix it soon enough.

ponce 01-08-2011 12:38 PM
actually, with 2.6.37, it is.
also alsa fails here (realtek chip, intel-hda, ALC268).

Ilgar 01-08-2011 01:09 PM
Thanks ponce, so at least the bug is actively being worked on... As for the audio, it works here.

brixtoncalling 01-08-2011 01:37 PM
Quote:
Originally Posted by Ilgar (Post 4217560)
@brixtoncalling: Did you try guanx's suggestion to upgrade libdrm and drivers? In the changelog I've seen some references to bugs fitting your description.

I decided to keep the new libdrm and intel driver installed. Since everything works well with 2.6.35.6, I believe there must be a kernel element in all this. I hope they fix it soon enough.
No, I haven't gotten around to it yet, Ilgar. My problem is infrequent, so I haven't worried too much. I'm trying a 2.6.36 kernel right now since 2.6.34 was problematic. I wonder if it has anything to do with the X.org updates in current?


All times are GMT -5. The time now is 12:07 AM.
Page 1 of 2 1 2
Show 50 post(s) from this thread on one page