2.6.18-rc4-no2 "To be..." | (Super Fix/Stability Attempt?)

Message

cheater1034 · Post by **cheater1034** » Sun Aug 20, 2006 5:01 pm

2.6.18-rc4-no2 "To be..." | ...My attempt at more stability
this is a testing release
If you have a problem please report it, otherwise I can not help you with it

Release
- A very solid release coming up into the mix, Based off of 2.6.18-rc4-mm1 but including
several important and not so important fixes, some being from -mm2. Nothing real experimental
was pushed into this release, just a ton of fixes and the new version of -mm. I'm really
opt about the stability of this release in comparison to 2.6.18-rc4-no1, I went out of my
way to assure it as stable as possible this release. (marked unstable due to -rc release only)

Patchset
- No-sources is the "successor" of love-sources, This kernel is for "power users", those who like living on the bleeding edge. It contains many experimental patches that can be slightly volatile at times, make your [soda||pop] real bland, etc.

- No-sources is a kernel patchset based on Andrew Morton's -mm patchset. It contains
numerous user-requested features, other things gathered from various sources, and
some of our own items/patches to make it no-certified.

"No-sources was founded during the end of love-source's reign by JasonF and myself.
Our intention was to create a patchset that would be able to continue on regardless
of maintainer and still live in the spirit of -love's original -mm base and featureset.
Given there have been numerous changes throughout the releases, including different
maintainers, but I believe through this philosophy a useful patchset can continue to be made."
-joecool

Changelog > 2.6.18-rc4-no1
* Based off of a stable -mm release (2.6.18-rc4-mm1)
* Fixes intellimouse broken support found in all .18-rc kernels.
* Includes about 46 misc. fixes/cleanups for the -mm patchset, some including major fixes
* Dropped lockless pagecache from this release because it still has bugs. (We'll wait for
some fixes to the direct data port from Nick Piggin first)
* NoSched series re-done into 11 patches
* Added the max_delay schedstat from nicksched to ingosched and staircase
* Added a second patch to fix ati drivers.
* Re-did the squashfs 3 port, moved a few things for statfs. It still gives a warning, but
the warning should not be anything fatal.
* Deadline is default again
* Made a new improved version of the config-nr-tty-devices patch
* ACPI DSDT seems to have been dropped, probally by accident. Apply manually if needed from
the 2.6.18-rc2-no1 broken-out.

Patches

Code: Select all

# No-Sched Series:
01_NoSched-prep.patch
02_NoSched-Makefile.patch
03_NoSched-Kconfig.patch
04_NoSched-implement-ingosched.patch
05_NoSched-implement-staircase.patch
06_NoSched-implement-staircase-tunables.patch
07_NoSched-implement-staircase-sched-iso.patch
08_NoSched-implement-staircase-idleprio.patch
09_NoSched-implement-ifdefs.patch
10_NoSched-implement-max_delay-stat.patch
11_NoSched-implement-max_delay-stat-2.patch

# Hot-Fixes/Simple Cleanups for -mm
trivial-kzalloc-opportunity.patch
lock-ticogwinsz.patch
stop-the-tty-vanishing-under-procfs-access.patch
fix-crash-case.patch
lock-tty-directly-in-acct-layer.patch
multiprobe-sanitizer.patch
kprobes-i386-add-KPROBE_END_for_popsection.patch
kprobes-x86_64-add-KPROBE_END_for_popsection.patch
ppc_fix-firewire-and-suspend.patch
char-better-error-codes.patch
fix-docs-for-suid_non_dumpable.patch
net_restrict-device-names-from-having-whitespace.patch
revert-intellimouse-4-support_broken.patch
revert-simplify-update_times-avoid-jiffies-jiffies_64-aliasing-problem.patch
do-not-check-unpopulated-zones-for-draining-and-counter.patch
ecryptfs-crypto-functions-mutex-fixes.patch
fix-wrong-error-code-on-interrupted-close-syscalls-fix.patch
fs-xfs-xfs_dmapih-removal-of-old-code.patch
gregkh-usb-usb-cypress-bugfix.patch
hdaps-add-explicit-hardware-configuration-functions-fix.patch
swsusp-fix-swap_type_of.patch
swsusp-read-speedup-fix-fix-2.patch
swsusp-use-memory-bitmaps-during-resume-fix.patch
swsusp-use-suspend_console.patch
pm-make-it-possible-to-disable-console-suspending.patch
pm-make-it-possible-to-disable-console-suspending-fix.patch
i386-detect-clock-skew-during-suspend.patch
optimize-free_one_page.patch
reduce-max_nr_zones-remove-display-of-counters-for-unconfigured-zones-s390-fix-fix.patch
rtmutex-clean-up-and-remove-some-extra-spinlocks-more.patch
zvc-overstep-counters.patch
fix-____call_usermodehelper-errors-being-silently-ignored.patch
fix-ext2-mounts-at-16t.patch
fix-ext3-mounts-at-16t.patch
fix-memory-leak-in-vc_resize-vc_allocate.patch
fix-possible-udf-deadlock-and-memory-corruption.patch
i-o-error-attempting-to-read-last-partial-block-of-a-file-in-an-iso9660-file-system.patch
mxser-upgrade-to-191.patch
__dequeue_signal-cleanup.patch
reparent_to_init-use-has_rt_policy.patch
kill-extraneous-printk-in-kernel_restart.patch
tty-make-termios_sem-a-mutex.patch
tty-make-termios_sem-a-mutex-fix.patch
for-2618-revert-drop-tasklist-lock-in-do_sched_setscheduler.patch
do_sched_setscheduler-dont-take-tasklist_lock.patch

# from Spock
fbsplash-0.9.2-r5.patch | gensplash
vesafb-tng-1.0-rc2.patch | vesa-tng

# from -ck
kconfig-expose_vmsplit-option.patch | improve explanations and make vmsplit visible
mm-convert_swappiness_to_mapped.patch | convert swappiness to mapped watermark
mm-lots_watermark.patch | Add the pages_lots watermark

# from -tiny
cpu-support.patch
movsl-mask.patch
mtrr-cpusupport.patch

# Thinkpad Patches
ibm-fancontrol.patch
ibm_acpi-thermal.patch
linux-phc-0.2.6-2.6.18-rc4-no2.patch

# Convert joydev for digital presses (optional/Kconfig Option)
linux-joydev-analogdigital.patch
linux-joydev-analogdigital-fix.patch

# Fix support for ATI-Drivers
fix-ati-build-1.patch
fix-ati-build-2.patch

# Random Stuff
HZ-864-Kconfig-Option.patch
add-clockinfo-to-timerc.patch
prevent-backwards-time.patch
no-preferred-desktop-opts.patch
config-nr-tty-devices-r3.patch
realtime-lsm.patch


# File Systems
squashfs-3.0.patch
unionfs-svn.patch

# USB
usb-mouse-polling.patch
usbxchange5.patch

# from DaMouse
daconfig-2.3.1.patch

# tag
no-tag.patch

# Build Fixes.
#
# To fix all build errors from the other patches.
final-build-fixes.patch

Downloads:
EBUILD
PATCH
DETAILED PATCHLIST
BROKEN-OUT

HOWTO
GENTOO and ANY DISTRIBUTION
A how-to can be found at:
http://gentoo-wiki.com/HOWTO_no-sources

Good Luck!
Lets hear feedback, results, problems, etc.

mb.Tab · Post by **mb.Tab** » Sun Aug 20, 2006 5:06 pm

**installing** Will be back for a report !

PS.: Thanks for doing all this work

/E kernel performces really well and the problem with my freeezing mouse is gone, too

So thanks again cheater

mbar · Post by **mbar** » Sun Aug 20, 2006 7:07 pm

yeah

*subscribes*

Nick C · Post by **Nick C** » Sun Aug 20, 2006 7:16 pm

will update and give it a try when i get home, got a couple of things to test but will report back if the dvd reading problem i had with the last release is fixed or not.

wrc1944 · Post by **wrc1944** » Sun Aug 20, 2006 7:51 pm

Compiles, boots, and runs fine here on my wireless ide ATA100 box. No longer hangs at shutdown like rc4-no1 and rc4-viper1 did, as I described on the viper thread. No sense in repeating all that here.

Don't know what fixed the problem.

azdruid · Post by **azdruid** » Sun Aug 20, 2006 9:13 pm

Compiles and boots perfectly. Tested with NVidia binary drivers and madwifi-ng.

Seems to be quite stable as well. Thanks for the excellent release, cheater.

freigeist · Post by **freigeist** » Sun Aug 20, 2006 10:05 pm

madwifi-ng now working here too...bootimes are still 15 seconds longer than with other patchset...

cheater1034 · Post by **cheater1034** » Sun Aug 20, 2006 11:07 pm

freigeist wrote:madwifi-ng now working here too...bootimes are still 15 seconds longer than with other patchset...

I don't know about 15 seconds, that's sort of like nothing. It's probally something extra enabled in -mm, that doesn't come with vanilla causing it, or something else hanging, boot time also will vary a lot depending on dhcp fetching the IP and stuff.

I honestly have no suggestions because 15 seconds is nothing, it's probally just initializing extra things from -no/mm (you need to take into consideration this is like 1500 patches over vanilla)

a7thson · Post by **a7thson** » Mon Aug 21, 2006 12:40 am

I must have bad luck... 2.6.18-rc4-no2 failed in a number of ways on this system... after successfully drawing most of a window, Xserver locked hard and the system had to be manually shut off. firewire support and sbp2 failed and the SATA-based cdrw/dvd was inaccessible. a quick check of Xorg.log revealed that ati-drivers apparently loaded fine, so that is probably not the issue.

in general then, here are the problems so far:
1) IEEE1394 subsystem produces strange error messages and is nonfunctional
2) SATA CDRW/DVD is detected and recognized by udev but creation of special devices fails and the dvd player is inaccessible
3) ati-drivers (8.27.10-r1) appear to load fine but system hard-locks X server after drawing a partial screen (mouse initially works, and within a few minutes this locks as well. no other input seems to have any effect and it is not possible to kill the x server)

what did seem to work:
1) framebuffer support under vesafb-tng widescreen resolution (1440x900) whoohoo!

2) external USB hard drives were detected and mounted properly
3) XFS modules loaded, external drive mounted and I was able to access filesystem

I copied the exact config from 2.6.18-rc4-viper1 and noticed a large number of config options were no longer present when I was configuring no-sources, including this entire subsystem:

Code: Select all

.config:822:warning: trying to assign nonexistent symbol SCSI_SATA
.config:823:warning: trying to assign nonexistent symbol SCSI_SATA_AHCI
.config:824:warning: trying to assign nonexistent symbol SCSI_SATA_SVW
.config:825:warning: trying to assign nonexistent symbol SCSI_ATA_PIIX
.config:826:warning: trying to assign nonexistent symbol SCSI_SATA_MV
.config:827:warning: trying to assign nonexistent symbol SCSI_SATA_NV
.config:828:warning: trying to assign nonexistent symbol SCSI_PDC_ADMA
.config:830:warning: trying to assign nonexistent symbol SCSI_SATA_QSTOR
.config:831:warning: trying to assign nonexistent symbol SCSI_SATA_PROMISE
.config:832:warning: trying to assign nonexistent symbol SCSI_SATA_SX4
.config:833:warning: trying to assign nonexistent symbol SCSI_SATA_SIL
.config:834:warning: trying to assign nonexistent symbol SCSI_SATA_SIL24
.config:835:warning: trying to assign nonexistent symbol SCSI_SATA_SIS
.config:836:warning: trying to assign nonexistent symbol SCSI_SATA_ULI
.config:837:warning: trying to assign nonexistent symbol SCSI_SATA_VIA
.config:838:warning: trying to assign nonexistent symbol SCSI_SATA_VITESSE
.config:839:warning: trying to assign nonexistent symbol SCSI_SATA_INTEL_COMBINED

apparently this subsystem was removed? moved? I took the config from another 2.6.18-rc4 kernel so this seems a little weird to me, unless perhaps mm- has done something with these.

and here is the relevant part of the firewire subsystem in the kernel log. note that no other kernel has produced these error messages for me before, and the configuration of the IEEE1394 subsystem is totally unchanged from a config that works under another kernel.

Code: Select all

2006-08-20 18:23:45 -05:00; neuros; kern.err; kernel: ohci1394: fw-host0: Running dma failed because Node ID is not valid
2006-08-20 18:23:45 -05:00; neuros; kern.info; kernel: ohci1394: fw-host0: AT dma reset ctx=0, aborting transmission
2006-08-20 18:23:45 -05:00; neuros; kern.debug; kernel: ieee1394: Current remote IRM is not 1394a-2000 compliant, resetting...
2006-08-20 18:23:45 -05:00; neuros; kern.debug; kernel: ieee1394: Host added: ID:BUS[0-00:1023]  GUID[334fc0000edc04e1]

Note that the "Host added" line is identical to what's produced on a working kernel (hopefully that's a good thing!)

I hope that you can offer some suggestions as to why my system and this no-sources don't play nicely together. I am a little hesitant to boot into the kernel again for the time being, I really don't like lockups

I won't have much time to work with this until tomorrow night so apologies in advance if I'm slow to respond to suggestions until then.

cheater1034 · Post by **cheater1034** » Mon Aug 21, 2006 12:42 am

-rc4-mm1 notes wrote:- Warning: all the Serial ATA Kconfig options have been renamed. If you
blindly run `make oldconfig' you won't have any disks.

Also,
1) IEEE1394 subsystem produces strange error messages and is nonfunctional

ieee1394 should work fine as long as you're using the one in the kernel?

p4r0l3 · Post by **p4r0l3** » Mon Aug 21, 2006 12:45 am

wrc1944 wrote:Compiles, boots, and runs fine here on my wireless ide ATA100 box. No longer hangs at shutdown like rc4-no1 and rc4-viper1 did, as I described on the viper thread. No sense in repeating all that here.

Don't know what fixed the problem.

Same thing had been happening to me. Compiled your new kernel with my old config, everything booted up perfectly the first time. Don't think that has ever happened to me before!

The shutdown hang was a bit annoying, that was fixed and i'm no longer getting random system hangs like I would from time to time in no1.

Props go out to you on this one cheater!

EDIT: to the above poster, i can confirm that the latest ati-drivers (in portage, not the new new ones, havent tried those) work fine for me

bigbob73 · Post by **bigbob73** » Mon Aug 21, 2006 12:47 am

working here

a7thson · Post by **a7thson** » Mon Aug 21, 2006 1:11 am

cheater1034 wrote:
-rc4-mm1 notes wrote:- Warning: all the Serial ATA Kconfig options have been renamed. If you
blindly run `make oldconfig' you won't have any disks.
Also,
1) IEEE1394 subsystem produces strange error messages and is nonfunctional

ieee1394 should work fine as long as you're using the one in the kernel?

I had ATA working, luckily, after seeing this pile of missing config options and wondering "wtf?" then reconfiguring it (and it worked).
yes, this is straight kernel module ieee1394, not an external from portage or custom ebuild

broch · Post by **broch** » Mon Aug 21, 2006 1:50 am

is xfs fixed in -mm already? It was badly broken (and reported as such). So i could not test your previous patch.

cheater1034 · Post by **cheater1034** » Mon Aug 21, 2006 2:03 am

broch wrote:is xfs fixed in -mm already? It was badly broken (and reported as such). So i could not test your previous patch.

I honestly don't know what caused the couple xfs problems, but the last release wasn't even off of a -mm release, and this -mm has the latest xfs tree and any additional fixes and it should work fine

cheater1034 · Post by **cheater1034** » Mon Aug 21, 2006 2:10 am

a7thson wrote:
cheater1034 wrote:
-rc4-mm1 notes wrote:- Warning: all the Serial ATA Kconfig options have been renamed. If you
blindly run `make oldconfig' you won't have any disks.
Also,
1) IEEE1394 subsystem produces strange error messages and is nonfunctional

ieee1394 should work fine as long as you're using the one in the kernel?
I had ATA working, luckily, after seeing this pile of missing config options and wondering "wtf?" then reconfiguring it (and it worked).
yes, this is straight kernel module ieee1394, not an external from portage or custom ebuild

I can't help you on ieee1394 without any dmesg or system info

a7thson · Post by **a7thson** » Mon Aug 21, 2006 2:27 am

cheater1034 wrote: I can't help you on ieee1394 without any dmesg or system info

yeah, I realize that. unfortunately I was stupid and didn't copy dmesg before I hit 'startx' and ended up with a frozen system. I have to have a usable machine for a couple hours, but will give it another go tomorrow after work, and will post appropriate information from the dmesg log and whatever else I can gather. here is emerge --info (I removed some fluff, overlays and my mirror list

otherwise this is everything) .. running under a different kernel but it should give some basic information to get started.

Code: Select all

#emerge --info
Portage 2.1.1_pre5-r2 (default-linux/x86/2006.0, gcc-4.1.1, glibc-2.4-r4, 2.6.18-rc4-viper i686)
=================================================================
System uname: 2.6.18-rc4-viper i686 Intel(R) Pentium(R) M processor 1.60GHz
Gentoo Base System version 1.12.4
Last Sync: Sun, 20 Aug 2006 16:20:01 +0000
distcc 2.18.3 i686-pc-linux-gnu (protocols 1 and 2) (default port 3632) [disabled]
ccache version 2.3 [enabled]
app-admin/eselect-compiler: [Not Present]
dev-lang/python:     2.4.3-r1
dev-python/pycrypto: 2.0.1-r5
dev-util/ccache:     2.3
dev-util/confcache:  0.4.2-r1
sys-apps/sandbox:    1.2.17
sys-devel/autoconf:  2.13, 2.59-r7
sys-devel/automake:  1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6-r2
sys-devel/binutils:  2.16.1-r3, 2.17
sys-devel/gcc-config: 1.3.13-r3
sys-devel/libtool:   1.5.22
virtual/os-headers:  2.6.17
ACCEPT_KEYWORDS="x86"
AUTOCLEAN="yes"
CBUILD="i686-pc-linux-gnu"
CFLAGS="-O2 -march=pentium-m -mtune=pentium-m -mmmx -pipe -ftracer -ffast-math -mno-tls-direct-seg-refs -funroll-loops -fomit-frame-pointer"
CHOST="i686-pc-linux-gnu"
CONFIG_PROTECT="/env /etc /share/config /shutdown /usr/lib/X11/xkb /usr/lib/mozilla/defaults/pref /usr/share/config /usr/share/texmf/dvipdfm/config/ /usr/share/texmf/dvips/config/ /usr/share/texmf/tex/generic/config/ /usr/share/texmf/tex/platex/config/ /usr/share/texmf/xdvi/"
CONFIG_PROTECT_MASK="/etc/env.d /etc/gconf /etc/revdep-rebuild /etc/splash /etc/terminfo /usr/X11R6/bin/startx"
CXXFLAGS="-O2 -march=pentium-m -mtune=pentium-m -mmmx -pipe -ftracer -ffast-math -mno-tls-direct-seg-refs -funroll-loops -fomit-frame-pointer -fvisibility-inlines-hidden"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoconfig buildpkg ccache confcache distlocks metadata-transfer parallel-fetch sandbox sfperms strict userpriv usersandbox"
LDFLAGS="-Wl,-O1"
MAKEOPTS="-j3"
PKGDIR="/usr/portage/packages"
PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --compress --force --whole-file --delete --delete-after --stats --timeout=180 --exclude='/distfiles' --exclude='/local' --exclude='/packages'"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
PORTDIR_OVERLAY="/usr/local/portage"
SYNC="rsync://rsync.us.gentoo.org/gentoo-portage"
USE="3dnow 3dnowext 7zip X aac aalib acpi aim alsa apache2 apm avi bash-completion berkdb bitmap-fonts bzip2 bzlib cddb cdparanoia cdr cgi cli crypt cups curl curlwrappers dga dhcp divx4linux dlloader dri dvd dvdread eds elibc_glibc emacs emboss encode esd faac fbcon firefox foomaticdb fortran ftp fuse gd gdbm gif gimp glibc-omitfp glut gmp gnome gnuplot gpm gstreamer gtk gtk2 guile hardened hardenedphp iconv icq image imlib ipv6 isdnlog jabber java javadoc jpeg kernel_linux ldap libcaca libg++ libgd libwww logrotate lua lzw mad mikmod mime mixer mmx mmxext modplug motif mozilla mp3 mpeg mpeg4 mplayer msn multislot mysql ncurses net network neural nfs nls nntp nptl nptlonly odbc ogg oggvorbis opengl oss pam pcap pcmcia pcre pdf pdflib perl php pic pie png pnp posix pppd pygtk python qt qt3 qt4 quicktime readline real reflection samba screen sdl session shared sharedmem sid snmp soap sockets socks5 spell spl sqlite sse sse2 ssl svga sysvipc tcl tcpd theora truetype truetype-fonts type1-fonts udev unicode usb userland_GNU userlocales vcd vorbis win32codecs x86 xemacs xine xinetd xml xmms xorg xpm xv xvid yahoo zip zlib"
Unset:  CTARGET, EMERGE_DEFAULT_OPTS, INSTALL_MASK, LANG, LC_ALL, PORTAGE_RSYNC_EXTRA_OPTS

Code: Select all

neuros linux # lspci
00:00.0 Host bridge: Intel Corporation Mobile 915GM/PM/GMS/910GML Express Processor to DRAM Controller (rev 03)
00:01.0 PCI bridge: Intel Corporation Mobile 915GM/PM Express PCI Express Root Port (rev 03)
00:1d.0 USB Controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) USB UHCI #1 (rev 03)
00:1d.1 USB Controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) USB UHCI #2 (rev 03)
00:1d.2 USB Controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) USB UHCI #3 (rev 03)
00:1d.3 USB Controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) USB UHCI #4 (rev 03)
00:1d.7 USB Controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) USB2 EHCI Controller (rev 03)
00:1e.0 PCI bridge: Intel Corporation 82801 Mobile PCI Bridge (rev d3)
00:1e.2 Multimedia audio controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) AC'97 Audio Controller (rev 03)
00:1f.0 ISA bridge: Intel Corporation 82801FBM (ICH6M) LPC Interface Bridge (rev 03)
00:1f.2 IDE interface: Intel Corporation 82801FBM (ICH6M) SATA Controller (rev 03)
00:1f.3 SMBus: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) SMBus Controller (rev 03)
01:00.0 VGA compatible controller: ATI Technologies Inc M22 [Radeon Mobility M300]
03:00.0 Ethernet controller: Broadcom Corporation BCM4401-B0 100Base-TX (rev 02)
03:01.0 CardBus bridge: Ricoh Co Ltd RL5c476 II (rev b3)
03:01.1 FireWire (IEEE 1394): Ricoh Co Ltd R5C552 IEEE 1394 Controller (rev 08)
03:01.2 Class 0805: Ricoh Co Ltd R5C822 SD/SDIO/MMC/MS/MSPro Host Adapter (rev 17)
03:03.0 Network controller: Intel Corporation PRO/Wireless 2200BG (rev 05)

irondog · Post by **irondog** » Mon Aug 21, 2006 9:48 am

Why do all these newer kernels make the boot process more verbose? Every time the system loads the nvidia kernel module the kernel begins spitting out kernel messages in the console. Why?

vipernicus · Post by **vipernicus** » Mon Aug 21, 2006 11:36 am

irondog wrote:Why do all these newer kernels make the boot process more verbose? Every time the system loads the nvidia kernel module the kernel begins spitting out kernel messages in the console. Why?

I think it has something to do with the newer Gentoo baselayouts.

irondog · Post by **irondog** » Mon Aug 21, 2006 11:55 am

vipernicus wrote:I think it has something to do with the newer Gentoo baselayouts.

Nope. It happens after modprobe nvidia. Something in the nvidia kernel module lowers the loglevel of the kernel. This only happens with newer kernels. Try removing nvidia.ko from /lib/modules and the problem disappears.

Or try insmodding nvidia.ko on a console, right after that the kernel loglevel is lowered as with dmesg -n 8.

Phlogiston · Post by **Phlogiston** » Mon Aug 21, 2006 12:34 pm

irondog wrote:
vipernicus wrote:I think it has something to do with the newer Gentoo baselayouts.
Nope. It happens after modprobe nvidia. Something in the nvidia kernel module lowers the loglevel of the kernel. This only happens with newer kernels. Try removing nvidia.ko from /lib/modules and the problem disappears.

Or try insmodding nvidia.ko on a console, right after that the kernel loglevel is lowered as with dmesg -n 8.

Yes thats true. I noticed the same here and not only with nvidia. Furthermore this is a kernel issue.

irondog · Post by **irondog** » Mon Aug 21, 2006 12:55 pm

I found it in the source of the kernel. The kernel devs have decided to turn on all debug messages when the kernel is tainted somehow. This will be the default behaviour in kernel 2.6.18-torvalds if we don't act now.

I solved it by commenting it out.

Code: Select all

void add_taint(unsigned flag)
{
        //debug_locks_off(); /* can't trust the integrity of the kernel anymore */
        tainted |= flag;
}

I'll create a patch to configure the kernel to behave like this or not.
I think this isn't the preferred behavior for everybody so it should be possible to configure this kind of stuff.

irondog · Post by **irondog** » Mon Aug 21, 2006 1:50 pm

Here it is for 2.6.18-rc4-vanilla:
http://tienstra4.flatnet.tudelft.nl/~ge ... .patch.txt

zxy · Post by **zxy** » Mon Aug 21, 2006 2:32 pm

My keboard goes nuts with this kernel. I can hardly type. I press once and get 2,3,4,5, ore more same keystrokes.
Looks like its something with time going wrong. I hardly login. 60s pass by in 10 or 15 s. Like time is running faster for the computer.

It is the same in console and in X.

Any ideas?

cheater1034 · Post by **cheater1034** » Mon Aug 21, 2006 2:51 pm

zxy wrote:My keboard goes nuts with this kernel. I can hardly type. I press once and get 2,3,4,5, ore more same keystrokes.
Looks like its something with time going wrong. I hardly login. 60s pass by in 10 or 15 s. Like time is running faster for the computer.

It is the same in console and in X.

Any ideas?

Try this patch:

Code: Select all

Signed-off-by: Roman Zippel <zippel@linux-m68k.org>

---
 kernel/time/ntp.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
Index: linux-2.6-mm/kernel/time/ntp.c
===================================================================
--- linux-2.6-mm.orig/kernel/time/ntp.c
+++ linux-2.6-mm/kernel/time/ntp.c
@@ -41,7 +41,7 @@ static long time_reftime;		/* time at la
 long time_adjust;
 
 #define CLOCK_TICK_OVERFLOW	(LATCH * HZ - CLOCK_TICK_RATE)
-#define CLOCK_TICK_ADJUST	(((s64)CLOCK_TICK_OVERFLOW * NSEC_PER_SEC) / CLOCK_TICK_RATE)
+#define CLOCK_TICK_ADJUST	(((s64)CLOCK_TICK_OVERFLOW * NSEC_PER_SEC) / (s64)CLOCK_TICK_RATE)
 
 static void ntp_update_frequency(void)
 {