2.6.23-kamikaze1: "Die computer, die die die!"

Waninkoko · Guru Joined: 13 May 2005 Posts: 549

· KAMIKAZE-SOURCES ::.
· 2.6.23-kamikaze1: "Die computer, die die die!"

New version of "kamikaze-sources" !!

NOTE: reiser4 patch doesn't work properly!!

Links:

ilikenwf · Tux's lil' helper Joined: 04 Oct 2007 Posts: 119

Sweet! I am excited!

Now to just wait for 2.6.23-1 to fix the dev's screwups.

mbar · Veteran Joined: 19 Jan 2005 Posts: 1944 Location: Poland

what's wrong with reiser4 for .23?

tallica · Posted: Wed Oct 10, 2007 4:40 pm Post subject:

Thanks

Gonna try this one.

Waninkoko · Guru Joined: 13 May 2005 Posts: 549

Janax · Apprentice Joined: 17 Aug 2004 Posts: 162 Location: Iowa

Just a question - do you have plans for uvesafb?

Thanks for the quick update on the new kernel - trying now!
_________________
Americans for Fair Taxation because the current tax system is not only burdensome but corrupt as well!

Waninkoko · Guru Joined: 13 May 2005 Posts: 549

dodo1122 · Posted: Wed Oct 10, 2007 5:23 pm Post subject:

heh, can't wait for reiser4 patch for 2.6.23

will have to mask the .23 for now...

dodo

tallica · Posted: Wed Oct 10, 2007 5:42 pm Post subject:

I'm getting these errors:

wrc1944 · Posted: Wed Oct 10, 2007 5:47 pm Post subject:

Compiled and booted fine, but on reboot it kind of hung for about 90 seconds here:

Oct 10 12:36:17 gentoo VP_IDE: not 100% native mode: will probe irqs later
Oct 10 12:36:17 gentoo VP_IDE: VIA vt8237 (rev 00) IDE UDMA133 controller on pci0000:00:0f.1
Oct 10 12:36:17 gentoo ide0: BM-DMA at 0xc800-0xc807, BIOS settings: hda:pio, hdb:pio
Oct 10 12:36:17 gentoo ide1: BM-DMA at 0xc808-0xc80f, BIOS settings: hdc:DMA, hdd:pio

Oct 10 12:36:17 gentoo Probing IDE interface ide0...
Oct 10 12:36:17 gentoo Probing IDE interface ide1...

and later at udev setup the same thing happened- also in 2-3 other places I had shorter hangs. This has never happened before. I'm using my 2622-kamikaze1 config file, but did disable a few things i didn;t need- unrelated to any ytpe of booting hangup AFAIK.

Anyway, kernel booted OK, and seems to be running fine so far.

Any ideas as to why this boot hanging would occur? It seemed like it must have been going through drastically longer probing routines.

I have a 26.23-rc4 with (ingo RT patch) kernel on another box that this didn't happen on, and no other kernels on this box do it either.

It must be either something in 2.6.23 final, something in kamikaze1, or one of the few little things I changed in my config file (disabled CPU Frequency scaling, some ethernet modules, and some ACPI stuff). Really don't see how any of my config weaks could cause a 90 second hang on IDE and BM-DMA stuff.

Googling hasn't clued me in yet- will keep trying to figure this out, with recompile without the little changes I made in config, and compiling a clean and simple 2.6.23 vanilla.
_________________
Main box- ASRock 880GM-LE AM3
Phenom II x6 1090T, 3.2 GHz, 8GB GSkill DDR3 1333mhz
Samsung SATA 500GB, Radeon HD 6570 2GB DDR3
Gentoo ~x86, ~amd64, glibc-2.17, gcc-4.7.2, kernel 3.8.11, 3.9.0

wrc1944 · Posted: Wed Oct 10, 2007 6:20 pm Post subject:

Rebooted, and same thing happens. (I was wrong on the udev stuff- that's OK). It seems to be related to the hard drives, and my cd/dvd burner. That's where it hangs up on probing them, for a total of about 3 minutes.

Could this have something to do with the ATA/SATA patches? Once it gets done with htis section, it proceeds normally. I have noticed windows/apps don't open as snappily now- there is a .5 or so second delay sometimes. Here's a little more context from /var/log/messages:

wrc1944 · Posted: Wed Oct 10, 2007 7:23 pm Post subject:

OK- compiled a vanilla 2.6.23 kernel with only these patches applied from your kamikazi1 repository:

sched-cfs-devel.patch
sched-cfs-boost.patch
sched-cfs-boost-tunables.patch
ckpatches-2.6.23.patch
genpatches-2.6.23.patch
hrtimers-2.6.23.patch
cpu-support-rollup-2.6.23.patch
enable-4k-stacks-default-2.6.23.patch

Seems to have solved the boot hang problems I had. It paused briefly at the same place, but only for about 15-20 seconds (seems normal time to probe), as compared to 2-3 minutes before with the full kamikaze1. No other problems during boot.

Also, none of the delays in window/apps opening are now ocurring.

My conclusion is it's not in 2.6.23 vanilla, or my config (used same config I just used in 2.6.23-kamikaze1), but must be in one of the other patches from kamikaze1 I didn't use this time.

Anyone else running into this- or is it just on this box?

EDIT: I just noticed this in the log- This time, at the exact point where it did hang, it switched to high resolution mode on CPU, whereas last time it didn't!

Oct 10 15:09:38 gentoo VP_IDE: chipset revision 6
Oct 10 15:09:38 gentoo VP_IDE: not 100% native mode: will probe irqs later
Oct 10 15:09:38 gentoo VP_IDE: VIA vt8237 (rev 00) IDE UDMA133 controller on pci0000:00:0f.1
Oct 10 15:09:38 gentoo ide0: BM-DMA at 0xc800-0xc807, BIOS settings: hda:pio, hdb:pio
Oct 10 15:09:38 gentoo ide1: BM-DMA at 0xc808-0xc80f, BIOS settings: hdc:DMA, hdd:pio
Oct 10 15:09:38 gentoo Probing IDE interface ide0...
[Oct 10 15:09:38 gentoo Switched to high resolution mode on CPU 0
Oct 10 15:09:38 gentoo Probing IDE interface ide1...
Oct 10 15:09:38 gentoo hdc: LITE-ON DVDRW SHW-1635S, ATAPI CD/DVD-ROM drive
Oct 10 15:09:38 gentoo hdc: selected mode 0x44
_________________
Main box- ASRock 880GM-LE AM3
Phenom II x6 1090T, 3.2 GHz, 8GB GSkill DDR3 1333mhz
Samsung SATA 500GB, Radeon HD 6570 2GB DDR3
Gentoo ~x86, ~amd64, glibc-2.17, gcc-4.7.2, kernel 3.8.11, 3.9.0

prestige · Tux's lil' helper Joined: 24 Sep 2005 Posts: 95

Dottout · l33t Joined: 07 Mar 2006 Posts: 882

Waninkoko · Guru Joined: 13 May 2005 Posts: 549

wrc1944, I don't know which patch is doing that. I'm suspecting in TuxOnIce patch.

prestige, post your .config. TuxOnIce compiles fine in my laptop.

Dottout, I'll fix that for kamikaze2.

Dottout · l33t Joined: 07 Mar 2006 Posts: 882

great, did you find already somewhere a solution? if so, pls give me the link, Ill patch kamikaze1 for now

kernelOfTruth · Posted: Wed Oct 10, 2007 7:48 pm Post subject:

yay !

gonna try it soon

@waninkoko:

you tried 2.6.23-rc1-mm1' reiser4 + namesys' patches ?

ftp://ftp.namesys.com/pub/reiser4-for-2.6/2.6.23-rc1-mm1
_________________
Unofficial minimal livecd x86/amd64 w/reiser4+truecrypt (by Neo2)
2.6.37.2_plus_v1: BFS, CFS,THP,compaction, zcache or TOI
Hardcore Linux user since 2004

prestige · Tux's lil' helper Joined: 24 Sep 2005 Posts: 95

Dottout · l33t Joined: 07 Mar 2006 Posts: 882

according to kernel changelog:

ocin · Guru Joined: 01 Jan 2006 Posts: 500

kernelOfTruth · Posted: Wed Oct 10, 2007 8:23 pm Post subject:

you're my hero

finally energy-saving on the intel HD

_________________
Unofficial minimal livecd x86/amd64 w/reiser4+truecrypt (by Neo2)
2.6.37.2_plus_v1: BFS, CFS,THP,compaction, zcache or TOI
Hardcore Linux user since 2004

brot · Guru Joined: 06 Apr 2004 Posts: 318

verynice

Works great here, except the "sandbox issue" - but i dont know if it is a bug of portage or the ebuild. Anyways.. thank you for your work Waninkoko
_________________
Use your free CPU-Time on the World Community Grid and make this a better world

wrc1944 · Posted: Wed Oct 10, 2007 8:36 pm Post subject:

Waninkoko,
Since my log reffered to the cpu mode where apparently something was prevented the high-resolution mode (whatever that means, or if it's even desirable??), I was suspecting:

# -mm patches
cpufreq-allow-ondemand-and-conservative-cpufreq-governors-to-be-used-as-default.patch
allow-ondemand-and-conservative-cpufreq-governors-to-be-used-as-default-kconfig-fix.patch

Or maybe the hrtimers was conflicting with one of the other patches??

I thought the TuxOnIce patch had something to do with suspend/hibernation, at least that's what my brief googling led me to believe. In any case, I never enable the suspend or power management stuff in my configs, as I never have had much need for it.

This is pretty much beyond my level of kernel patch knowledge or insight.
_________________
Main box- ASRock 880GM-LE AM3
Phenom II x6 1090T, 3.2 GHz, 8GB GSkill DDR3 1333mhz
Samsung SATA 500GB, Radeon HD 6570 2GB DDR3
Gentoo ~x86, ~amd64, glibc-2.17, gcc-4.7.2, kernel 3.8.11, 3.9.0

enderandrew · l33t Joined: 25 Oct 2005 Posts: 731

There don't appear to be hardly any changes from 2.6.23-rc8 and rc9 or the final 2.6.23 judging from the changelogs.

Reiser4 works in 2.6.23-rc8-mm1 so why not steal that patch?
_________________
Nihilism makes me smile.

tallica · Posted: Wed Oct 10, 2007 9:41 pm Post subject: