__DEPRECATED__ 2.6.15-nitro2 NO_MORE_POSTING_PLEASE

[kopp]

It compiled just fine and I've got no problem with it so far. I still have to test it under high load

Great job Tiger683

Thanks!

[esc*]

installed new nitro, everything just works but i get all the time i boot says filesystem check error on my /home ext3.
something cannot fix automatically press ctrl+d to continue or type root password to fix problem manually. i tryed e2fsck on /home the output looks like clear and then booted but no effect. startup stops that ext3 partition checks. i see theres some ext3-get-blocks patches on -mm patches. is there simple way to fix this? or is it the kernel problem? 2.6.14-nitro-r2 dont get these errors.

[Tiger683]

installed new nitro, everything just works but i get all the time i boot says filesystem check error on my /home ext3.
something cannot fix automatically press ctrl+d to continue or type root password to fix problem manually. i tryed e2fsck on /home the output looks like clear and then booted but no effect. startup stops that ext3 partition checks. i see theres some ext3-get-blocks patches on -mm patches. is there simple way to fix this? or is it the kernel problem? 2.6.14-nitro-r2 dont get these errors.

try reverting the ext3 patches from mm, but if i were you , i'd remount your /home readonly and fsck it with all checking options enabled, as it's not normal

[satanskin]

Tiger,
Any words or thoughts on a netfilter patch for command owner match support?

[Pete M]

Thank you Tiger683

nitro-sources-2.6.15-r2 "Freshness" running just fine here

Samsung X20 Laptop
nitro-sources-2.6.15-r2
x11-base/xorg-x11-7.0
media-sound/alsa-driver-1.0.11_rc2
KDE 3.5

Code: Select all

# dmesg | grep agp
Linux agpgart interface v0.101 (c) Dave Jones
agpgart: Detected an Intel 915GM Chipset.
agpgart: Detected 7932K stolen memory.
agpgart: AGP aperture is 256M @ 0xc0000000

Code: Select all

# dmesg | grep drm
[drm] Initialized drm 1.0.0 20040925
[drm] Initialized i915 1.1.0 20040405 on minor 0:

Code: Select all

# glxinfo | grep direct
direct rendering: Yes

Plus this kernel is super fast

Once again many thanks

Pete

[Tiger683]

Tiger,
Any words or thoughts on a netfilter patch for command owner match support?

I'll let you know, but it doesn't look goot for now, not even a word as to in which git it got removed, but it must have been somewhere between 2.6.13 and 2.6.14

[tuam]

try reverting the ext3 patches from mm, but if i were you , i'd remount your /home readonly and fsck it with all checking options enabled, as it's not normal

Yes, fsck on ext3 only checks the journal. Use "fsck -f" to check the whole filesystem.

FF,

Dan - suffered from corruption this morning - iel

[Cinder6]

Under Processor type and features, there is now an option for AMD MCE features. If we use AMD(64, in this case), can we go ahead and disable the Intel MCE features and just use the AMD ones?

[Tiger683]

Cinder6: only one way to find out

[tuam]

Compiled, booted, inserted usb stick works as expected, no kioslaves, hal, etc.. update needed. I don't think I have to worry about the rest.

Thanks a lot, Tiger683!

FF,

Daniel

[Pse]

Under Processor type and features, there is now an option for AMD MCE features. If we use AMD(64, in this case), can we go ahead and disable the Intel MCE features and just use the AMD ones?

I think they provide different features. I'm not sure whether those provided by Intel are provided by AMD processors as well... It'd be nice to find out how to use them, though :S

[troebel]

Installed, had some problems with nvidia, alsa and vmware, solved the first 2 by re-emerging nvidia-kernel and f*cking around with the kernel alsa configuration, but vmware is a bit harder on me
I tried running vmware-config.pl, which solved the problems during boot. The only problem that persists is when I start a virtual PC (better: TRY to start one) it tells me this:

Code: Select all

Could not open /dev/vmmon: No such file or directory.
Please make sure that the kernel module `vmmon' is loaded.

I've seen this problem before on the forum, so it will be just a matter of time to get rid of this problem too.
This is the first nitro kernel I've ever tried, and so far I like it! Props to the nitro devs!

oh, one minor (eyecandy) "problem" is that my bootsplash appears later in the boot process. The gentoo kernel didn't show any text during boot, whereas the nitro one does show the first few lines before the splash comes up.

[Tiger683]

you must reemerge the vmware module.

[troebel]

you mean

Code: Select all

emerge vmware-workstation

or just the vmmon module? I thought vmware-config.pl did that (the module)?
BTW, vmmon is up and running

[Tiger683]

satanskin:
Surprise!!!!!
Ok, i found the patch that removed the support from mainline.
It was a bit work to port it to the new api of filedescriptor handling, but in the end i also
added a Kconfig option, and it compiles fine.
I need your feedback badly though, if it works, i'll clean it up a little
and off it goes to lkml for reinclusion as experimental

readd-ipt_owner-pid-sid-comm-matching-with-Kconfig-opt.patch

The reason it was removed is that it's broken on smp, so making the Kconfig opt depend on !SMP will probably
make it able to get back in...

[troebel]

Well, I tried both re-emerging vmware-workstation and running vmware-config.pl, with no luck.
So after reading this topic I edited the configscript and it works fine now

[Tiger683]

Well, I tried both re-emerging vmware-workstation and running vmware-config.pl, with no luck.
So after reading this topic I edited the configscript and it works fine now

Happy to hear it

K guys, until somewhere in about 8-10 hours, even the bunny has to sleep

[satanskin]

satanskin:
Surprise!!!!!
Ok, i found the patch that removed the support from mainline.
It was a bit work to port it to the new api of filedescriptor handling, but in the end i also
added a Kconfig option, and it compiles fine.
I need your feedback badly though, if it works, i'll clean it up a little
and off it goes to lkml for reinclusion as experimental

readd-ipt_owner-pid-sid-comm-matching-with-Kconfig-opt.patch

The reason it was removed is that it's broken on smp, so making the Kconfig opt depend on !SMP will probably
make it able to get back in...

Patched cleanly, compiled cleanly, boots clean, and seems to be running just fine so far. Thank you so much! Could you try to keep this patch in further releases as well please? At least until they decide to re-add it to vanilla. Thanks again!

[troebel]

The kernel runs great, but I can't get it to sleep. I forgot to compile Suspend2 into it, so when I tried to suspend, I was told to enable it in the kernel. So I added Suspend2 and the swap writer, tried to compile and got this:

Code: Select all

  CHK     include/linux/version.h
  SPLIT   include/linux/autoconf.h -> include/config/*
  CHK     include/linux/compile.h
dnsdomainname: Unknown host
  CHK     usr/initramfs_list
  GZIP    kernel/config_data.gz
  IKCFG   kernel/config_data.h
  CC      kernel/configs.o
  LD      kernel/built-in.o
  LD      drivers/acpi/built-in.o
  LD      drivers/built-in.o
  GEN     .version
  CHK     include/linux/compile.h
dnsdomainname: Unknown host
  UPD     include/linux/compile.h
  CC      init/version.o
  LD      init/built-in.o
  LD      .tmp_vmlinux1
arch/x86_64/kernel/built-in.o: In function `set_up_temporary_mappings':
suspend.c:(.text+0x12135): undefined reference to `get_safe_page'
suspend.c:(.text+0x1218b): undefined reference to `get_safe_page'
suspend.c:(.text+0x12251): undefined reference to `get_safe_page'
arch/x86_64/kernel/built-in.o: In function `swsusp_arch_resume':
: undefined reference to `restore_image'
make: *** [.tmp_vmlinux1] Error 1

Anyone has an idea on how to fix this??

[Thetargos]

I've ran into a problem when using the broken out patches... I can't bulid sched.o, here's the output:

Code: Select all

  CC      kernel/sched.o
kernel/sched.c: In function ‘effective_prio’:
kernel/sched.c:718: error: ‘PF_FREEZE’ not declared here (first use in this function)
kernel/sched.c:718: error: (Each undeclared identifier is only reported once
kernel/sched.c:718: error: for each function it appears in.)
make[1]: *** [kernel/sched.o] Error 1
make: *** [kernel] Error 2

I'm not sure what is the problem here... I will try to apply the condensed patch and see how it goes (though I really need the broken-out), any thoughts on what may be causing this problem?

Thanks.

[nesl247]

Can anyone explain the difference between Broken out and the regular version?

[Thetargos]

Can anyone explain the difference between Broken out and the regular version?

Broken out is the whole set of patches (individually) that integrate the whole patchset. The regular is a condensation of those.

[alamuru420123]

Alright ... the mouse wasn't detected initially but after upgrading udev to 079, and reemerging nvidia, it's working great. But one question ... I really don't feel any difference in the performance. How do I benchmark it? nbench?

[Kuroki Kaze]

CC mm/shmem.o
CC kernel/seccomp.o
CC kernel/dyn-tick.o
kernel/dyn-tick.c: In function `do_dyn_reprogram':
kernel/dyn-tick.c:109: error: `per_cpu__irq_stat' undeclared (first use in this function)
kernel/dyn-tick.c:109: error: (Each undeclared identifier is reported only once
kernel/dyn-tick.c:109: error: for each function it appears in.)

ps: prebuild:
cp /usr/src/linux/include/asm-i386/dyn-tick.h /usr/src/linux/include/asm/dyn-tick.h
cp /usr/src/linux/include/asm-i386/timer.h /usr/src/linux/include/asm/timer.h
-- make didnt find it...
______________________________
amd64, gcc-3.4.5, glibc-2.3.6-r2

[Thetargos]

CC mm/shmem.o
CC kernel/seccomp.o
CC kernel/dyn-tick.o
kernel/dyn-tick.c: In function `do_dyn_reprogram':
kernel/dyn-tick.c:109: error: `per_cpu__irq_stat' undeclared (first use in this function)
kernel/dyn-tick.c:109: error: (Each undeclared identifier is reported only once
kernel/dyn-tick.c:109: error: for each function it appears in.)

ps: prebuild:
cp /usr/src/linux/include/asm-i386/dyn-tick.h /usr/src/linux/include/asm/dyn-tick.h
cp /usr/src/linux/include/asm-i386/timer.h /usr/src/linux/include/asm/timer.h
-- make didnt find it...
______________________________
amd64, gcc-3.4.5, glibc-2.3.6-r2

Are you building for 32-bits? If I correctly recall, the dynticks code is not supported oon x86_64 architectures running in 64-bit mode.