Zen Kernel Discussion/Support Thread - Part 3

Message

cheater1034 · Post by **cheater1034** » Sat Feb 06, 2010 11:31 pm

continued from part 2

Zen Kernel Gentoo Discussion and Support Thread Part 3
-----------------------------------
WEBSITE
http://zen-kernel.org! Register! Read documentation for more information!
-----------------------------------
Zen Ebuilds!
Zen ebuilds can now be found in the portage tree. It's simple: add the entry for sys-kernel/zen-sources to package.keywords, sync, and install.
-----------------------------------
Zen Git Trees!
Zen (unstable): git://zen-kernel.org/kernel/zen.git
Zen (stable): git://zen-kernel.org/kernel/zen-stable.git
------------------------------------

Please post results here, IRC, and all around about problems, no problems, tests, voting, supporting, etc.

JOIN US
ON IRC: #zen-sources, irc.rizon.net
WEBSITE: http://zen-kernel.org
Mailing List: http://groups.google.com/group/zen_kernel or zen_kernel+subscribe@googlegroups.com
Issue Tracker: http://zen-kernel.org/issues

pappy_mcfae · Post by **pappy_mcfae** » Sun Feb 07, 2010 12:04 am

Got to make sure I don't miss out on this thread.

Blessed be!
Pappy

kernelOfTruth · Post by **kernelOfTruth** » Tue Feb 09, 2010 2:51 pm

@dusanc:

as I already wrote - it just popped up in my mind

I looked through the mailing list and he only wrote that it's "work[ing] incorrectly" - maybe also that it's broken in zen

guys - don't take things too literally

@dodo:

thanks for bumping to 2.6.33-rc7-zen1 !

Saundersx · Post by **Saundersx** » Thu Feb 11, 2010 6:43 pm

I think I might have found a kernel regression or something. I don't have any kernel dumps unfortunately but here's my symptoms.

I was using 2.6.32-zen6 but it would freeze intermittently, no rhyme or reason, under load or just sitting here. HD IO would also make things drag, something that has not happened in quite a while. I recompiled the kernel with CFS/BFS, CFQ/BFQ, Preempt/Voluntary, slub/slqb, tickless (with cfs), no tickless (Bfs). All variations suffered from the same thing.

Now when it froze it would always exhibit the same behavior, screen would freeze solid and mouse would freeze. Alt-Sys-REISU would do nothing but Alt-Sys-B would reboot the system. Never anything in the logs, even running watch -n0.5 "dmesg|tail -n 30" wouldn't catch anything if I left it on the screen. Just an instant freeze and it could happen within 10 min but always within about 6 hours.

So I rolled back to 2.6.31-zen12, same exact issues.

Then I rolled back to my last working one, 2.6.31-zen9, this has worked perfectly. I have had an uptime of about 3 days and completed an "emerge -auveDN system" on here and another box using distcc and upgraded this system last night including the kde 4.4 packages. No problem, and heavy IO had little effect on system responsiveness.

Here's my current config - http://pastebin.com/m40ea3016 - just happen to be using cfs right now, this has been roughly the same for all kernels
portage info - http://pastebin.com/m569699e5
and lilo append string - append="noresume splash=silent acpi_enforce_resources=lax" - lax part needed to control my fans

pappy_mcfae · Post by **pappy_mcfae** » Thu Feb 11, 2010 9:02 pm

Do you really need all those devices? Somehow, I think not.

Take a look at one of my kernel seeds, set it up with your devices, then start adding all that extra stuff until you break the kernel again...then backtrack.

Also, how can you tell if you boot if you don't have a frame buffer or frame buffer console support? While not totally needed if your machine is going to boot directly to X, if you plan on doing ANYTHING in the CLI world, you have to have a frame buffer and a console.

Click on the link to my page in my sig, and check out how I set up a .config. Also check out my how-to.

Blessed be!
Pappy

Saundersx · Post by **Saundersx** » Fri Feb 12, 2010 1:24 am

pappy_mcfae wrote:Do you really need all those devices? Somehow, I think not.

Take a look at one of my kernel seeds, set it up with your devices, then start adding all that extra stuff until you break the kernel again...then backtrack.

Also, how can you tell if you boot if you don't have a frame buffer or frame buffer console support? While not totally needed if your machine is going to boot directly to X, if you plan on doing ANYTHING in the CLI world, you have to have a frame buffer and a console.

Click on the link to my page in my sig, and check out how I set up a .config. Also check out my how-to.

Blessed be!
Pappy

I just went through your howto and excluding the networking part they are pretty damn close. And I can see the console, I just dont use a framebuffer, 99% of the time I'm in X.

And what devices are you talking about? For the most part they are geared towards this machine but if your talking about the networking stuff yeah I just turned most of it on. I have a bridge on here as well as vpn, fun fun!

ebrsc · Post by **ebrsc** » Fri Feb 12, 2010 2:08 am

Saundersx wrote:I think I might have found a kernel regression or something. I don't have any kernel dumps unfortunately but here's my symptoms.

I was using 2.6.32-zen6 but it would freeze intermittently, no rhyme or reason, under load or just sitting here. HD IO would also make things drag, something that has not happened in quite a while. I recompiled the kernel with CFS/BFS, CFQ/BFQ, Preempt/Voluntary, slub/slqb, tickless (with cfs), no tickless (Bfs). All variations suffered from the same thing.

Now when it froze it would always exhibit the same behavior, screen would freeze solid and mouse would freeze. Alt-Sys-REISU would do nothing but Alt-Sys-B would reboot the system. Never anything in the logs, even running watch -n0.5 "dmesg|tail -n 30" wouldn't catch anything if I left it on the screen. Just an instant freeze and it could happen within 10 min but always within about 6 hours.

So I rolled back to 2.6.31-zen12, same exact issues.

Then I rolled back to my last working one, 2.6.31-zen9, this has worked perfectly. I have had an uptime of about 3 days and completed an "emerge -auveDN system" on here and another box using distcc and upgraded this system last night including the kde 4.4 packages. No problem, and heavy IO had little effect on system responsiveness.

Here's my current config - http://pastebin.com/m40ea3016 - just happen to be using cfs right now, this has been roughly the same for all kernels
portage info - http://pastebin.com/m569699e5
and lilo append string - append="noresume splash=silent acpi_enforce_resources=lax" - lax part needed to control my fans

This is what you need http://kernel.osuosl.org/pub/software/s ... isect.html to find out the guilty commit. it will be way easier to solve your issue. and please report it upstream if someone didn't do it yet.

pappy_mcfae · Post by **pappy_mcfae** » Fri Feb 12, 2010 4:56 am

I counted four different network drivers. Not saying that you can't stuff four net cards into your machine, but I can only do so on one. Not having frame buffers, well, if that's your style, who am I to slam it?

Did the seed work out better than your setup?

BB!
P

leprosys · Post by **leprosys** » Fri Feb 12, 2010 6:16 am

I can't boot the kernel, when I upgrade to p6 version and try with the 9999 with the same problem, with the p5 works.

The message:
http://imagebin.org/84480

My config file:
http://leprosys.net/files/config

Some suggestion?

EDIT:
I solved my problem, reconfiguring the kernel.

Saundersx · Post by **Saundersx** » Fri Feb 12, 2010 7:36 am

pappy_mcfae wrote:I counted four different network drivers. Not saying that you can't stuff four net cards into your machine, but I can only do so on one. Not having frame buffers, well, if that's your style, who am I to slam it?

Did the seed work out better than your setup?

BB!
P

There was an extra one that came out of nowhere, disabled that. Otherwise its the onboard nforce one and a usb based one I play with. Dunno about the forth. And no I never tried the seed but I did make a few changes from the howto, see how those work out. I geared everything back towards bfs again too.

As long as this stays stable I'm going to wait for 2.6.33-zen1, if the problem continues then I'll dig around for the guilty commit. If I wasn't shooting completely blind (no crashlogs) I could probably figure this out

pappy_mcfae · Post by **pappy_mcfae** » Fri Feb 12, 2010 7:44 am

Yes, provide the patches for 2.6.32-zen7, as I don't see them on the zen site, and I don't see it in either of my zen git repositories. Then post the results of lspci -n, cat /proc/cpuinfo, as well as your /etc/fstab file, and I'll set you up with a kernel that works. I'm currently running 2.6.32-zen6, and obviously, I'm not having no-boot issues.

Blessed be!
Pappy

Saundersx · Post by **Saundersx** » Sat Feb 13, 2010 9:13 am

pappy_mcfae wrote:Yes, provide the patches for 2.6.32-zen7, as I don't see them on the zen site, and I don't see it in either of my zen git repositories. Then post the results of lspci -n, cat /proc/cpuinfo, as well as your /etc/fstab file, and I'll set you up with a kernel that works. I'm currently running 2.6.32-zen6, and obviously, I'm not having no-boot issues.

Blessed be!
Pappy

Unless you have the same hardware as me the fact that you have no-boot? (freezing?) issues doesn't prove anything. And no I don't need a kernel built lol I posted mine and if there is something glaringly wrong just point it out.

pappy_mcfae · Post by **pappy_mcfae** » Sat Feb 13, 2010 9:43 am

If you send me the patches, I can take a look. I use make xconfig, and without your source, I can do nothing. So, you can either send the patches and I can tell you what I see, or you can continue to experiment until you find the fixes yourself. It makes no difference to me. My computers are running just fine.

BB!
POh, and without lspci -n, how can you say this is different hardware? You have no idea.

timbo · Post by **timbo** » Sat Feb 20, 2010 8:49 pm

Hi all

Been using zen-sources on and off for a while... now gentoo-sources works perfect but zen is just that bit faster BUT one nagging problem I've had over multiple versions of zen-sources is that when I shut down the PC it doesn't actually power off, it'll sit there after shutting down all the services with "power off" at the top of the screen. Of course gentoo-sources actually power's off my machine.

Any ideas, I've checked and rechecked the .config but either a bug or something I can't see for the trees...

Tim

angry_vincent · Post by **angry_vincent** » Sun Feb 21, 2010 11:26 am

timbo wrote:Hi all

Been using zen-sources on and off for a while... now gentoo-sources works perfect but zen is just that bit faster BUT one nagging problem I've had over multiple versions of zen-sources is that when I shut down the PC it doesn't actually power off, it'll sit there after shutting down all the services with "power off" at the top of the screen. Of course gentoo-sources actually power's off my machine.

Any ideas, I've checked and rechecked the .config but either a bug or something I can't see for the trees...

Tim

I can confirm this with exactly the same problem. After turning off comp (poweroff) all services is down, i have Poweroff message on screen but looks like comp is working. I spent 2 days trying to find out the solution and only one thing solved: with CFS scheduler everything fine, if i back to BFS, problem returns, so problem there.

platojones · Post by **platojones** » Sun Feb 21, 2010 11:29 am

Hah...same problem here...power off doesn't shut down my PC either with Zen sources.

chris256 · Post by **chris256** » Sun Feb 21, 2010 12:21 pm

Hello ,

I have the same problem too, but not all the time.

By the way , bfs has been updated ( 314 ).

niick · Post by **niick** » Sun Feb 21, 2010 3:19 pm

Just for the record I also have the poweroff problem. It is intermittent but it fails about 80-90% of the time.

Waninkoko · Post by **Waninkoko** » Sun Feb 21, 2010 5:59 pm

BFS updated in zen.git

kernelOfTruth · Post by **kernelOfTruth** » Sun Feb 21, 2010 6:08 pm

Waninkoko wrote:BFS updated in zen.git

thanks !

timbo · Post by **timbo** » Mon Feb 22, 2010 4:57 am

Well at least I'm not sitting in the playground with this odd problem all on my own... thanks all

cheater1034 · Post by **cheater1034** » Mon Feb 22, 2010 5:04 am

timbo wrote:Well at least I'm not sitting in the playground with this odd problem all on my own... thanks all

This could be caused by the bfs cpu scheduler, maybe try a few reboots with cfs.

timbo · Post by **timbo** » Mon Feb 22, 2010 5:12 am

Yes most likely is gentoo-sources = fine, zen-sources = bad. major difference between the two for me is BFS... angry_vincent even commented he tried cfs was ok...

whiteghost · Post by **whiteghost** » Mon Feb 22, 2010 9:18 am

2.6.32-zen6 works fine for me.
i have tried 2.6.33-rc5-zen1 and 2.6.33-rc7-zen1 but i cannot compile
nvidia-drivers on these versions. anyone else tried?

Code: Select all

 >>> Emerging (123 of 124) x11-drivers/nvidia-drivers-190.53-r1
 * NVIDIA-Linux-x86_64-190.53-pkg2.run RMD160 SHA1 SHA256 size ;-) ...   [ ok ]
 * checking ebuild checksums ;-) ...                                     [ ok ]
 * checking auxfile checksums ;-) ...                                    [ ok ]
 * checking miscfile checksums ;-) ...                                   [ ok ]
 * CPV:  x11-drivers/nvidia-drivers-190.53-r1
 * REPO: gentoo
 * USE:  acpi amd64 elibc_glibc gtk kernel_linux multilib userland_GNU
 * Determining the location of the kernel source code
 * Found kernel source directory:
 *     /usr/src/linux
 * Found kernel object directory:
 *     /lib/modules/2.6.33-rc7-zen1/build
 * Found sources for kernel version:
 *     2.6.33-rc7-zen1
 * Checking for MTRR support ...                                          [ ok ]
>>> Unpacking source...
>>> Unpacking NVIDIA-Linux-x86_64-190.53-pkg2.run to /var/tmp/portage/x11-drivers/nvidia-drivers-190.53-r1/work/NVIDIA-Linux-x86_64-190.53-pkg2
>>> Source unpacked in /var/tmp/portage/x11-drivers/nvidia-drivers-190.53-r1/work
>>> Preparing source in /var/tmp/portage/x11-drivers/nvidia-drivers-190.53-r1/work/NVIDIA-Linux-x86_64-190.53-pkg2 ...
 * Applying NVIDIA_glx-defines.patch ...                                  [ ok ]
 * Applying NVIDIA_glx-glheader.patch ...                                 [ ok ]
 * Converting NVIDIA-Linux-x86_64-190.53-pkg2/usr/src/nv/Makefile.kbuild to use M= instead of SUBDIRS= ...                                                [ ok ]
>>> Source prepared.
>>> Configuring source in /var/tmp/portage/x11-drivers/nvidia-drivers-190.53-r1/work/NVIDIA-Linux-x86_64-190.53-pkg2 ...
>>> Source configured.
>>> Compiling source in /var/tmp/portage/x11-drivers/nvidia-drivers-190.53-r1/work/NVIDIA-Linux-x86_64-190.53-pkg2 ...
 * Preparing nvidia module
make -j5 HOSTCC=x86_64-pc-linux-gnu-gcc CROSS_COMPILE=x86_64-pc-linux-gnu- LDFLAGS= IGNORE_CC_MISMATCH=yes V=1 SYSSRC=/usr/src/linux SYSOUT=/lib/modules/2.6.33-rc7-zen1/build HOST_CC=x86_64-pc-linux-gnu-gcc clean module 
If you are using a Linux 2.4 kernel, please make sure
you either have configured kernel sources matching your
kernel or the correct set of kernel headers installed
on your system.

If you are using a Linux 2.6 kernel, please make sure
you have configured kernel sources matching your kernel
installed on your system. If you specified a separate
output directory using either the "KBUILD_OUTPUT" or
the "O" KBUILD parameter, make sure to specify this
directory with the SYSOUT environment variable or with
the equivalent nvidia-installer command line option.

Depending on where and how the kernel sources (or the
kernel headers) were installed, you may need to specify
their location with the SYSSRC environment variable or
the equivalent nvidia-installer command line option.

*** Unable to determine the target kernel version. ***

make: *** [select_makefile] Error 1
 * ERROR: x11-drivers/nvidia-drivers-190.53-r1 failed:
 *   Unable to emake HOSTCC=x86_64-pc-linux-gnu-gcc CROSS_COMPILE=x86_64-pc-linux-gnu- LDFLAGS=  IGNORE_CC_MISMATCH=yes V=1 SYSSRC=/usr/src/linux 		SYSOUT=/lib/modules/2.6.33-rc7-zen1/build HOST_CC=x86_64-pc-linux-gnu-gcc clean module
 * 
 * Call stack:
 *     ebuild.sh, line   54:  Called src_compile
 *   environment, line 4146:  Called linux-mod_src_compile
 *   environment, line 3080:  Called die
 * The specific snippet of code:
 *               eval "emake HOSTCC=\"$(tc-getBUILD_CC)\" 			CROSS_COMPILE=${CHOST}- 						LDFLAGS=\"$(get_abi_LDFLAGS)\" 						${BUILD_FIXES} 		${BUILD_PARAMS} 						${BUILD_TARGETS} " || die "Unable to emake HOSTCC="$(tc-getBUILD_CC)" CROSS_COMPILE=${CHOST}- LDFLAGS="$(get_abi_LDFLAGS)" ${BUILD_FIXES} ${BUILD_PARAMS} ${BUILD_TARGETS}";
 * 
 * If you need support, post the output of 'emerge --info =x11-drivers/nvidia-drivers-190.53-r1',
 * the complete build log and the output of 'emerge -pqv =x11-drivers/nvidia-drivers-190.53-r1'.
 * The complete build log is located at '/var/tmp/portage/x11-drivers/nvidia-drivers-190.53-r1/temp/build.log'.
 * The ebuild environment file is located at '/var/tmp/portage/x11-drivers/nvidia-drivers-190.53-r1/temp/environment'.
 * S: '/var/tmp/portage/x11-drivers/nvidia-drivers-190.53-r1/work/NVIDIA-Linux-x86_64-190.53-pkg2'

>>> Failed to emerge x11-drivers/nvidia-drivers-190.53-r1, Log file:

>>>  '/var/tmp/portage/x11-drivers/nvidia-drivers-190.53-r1/temp/build.log'

duryodhana · Post by **duryodhana** » Mon Feb 22, 2010 9:31 am

2.6.32-zen6 works fine for me.
i have tried 2.6.33-rc5-zen1 and 2.6.33-rc7-zen1 but i cannot compile
nvidia-drivers on these versions. anyone else tried?

Here, I unmasked nvidia-drivers-195.36.03 and it compiled fine with 2.6.33-rc7-zen1.

Zen Kernel Discussion/Support Thread - Part 3

Zen Kernel Discussion/Support Thread - Part 3

Kernel panic