[___DEPRECATED____] 2.6.14-nitro1

Message

Tiger683 · Post by **Tiger683** » Wed Nov 09, 2005 9:11 pm

nitro-sources-2.6.14-r1 "Grizzly"

Disclaimer
This patchset is in no way affiliated with the Gentoo Foundation and so it's
use does not imply any more or less official support from the gentoo developers.
Do NOT, EVER, post bugs regarding a kernel patched to nitro-sources to
_bugs.gentoo.org_. Post your problems in this thread instead.

What is it?
Custom kernel patchset, based upon ck-patchset from Con Kolivas and
many additional features usually requested by users.

THIS RELEASE HAS VESAFB-TNG DISABLED, AS IT CAUSED PANIC ON BOOT
USE VESAFB INSTEAD, FOR 1280x1024 RESOLUTION APPEND "vga=0x31A" TO KERNEL BOOTPARAMS
INSTEAD OF "video=vesafb: etc...."

What's in it?

Code: Select all

#
# ck patchset
#
->2.6.14-ck4.patch
...........1g_lowmem1_i386.diff
...........2.6.14_to_staircase12.1.diff
...........hz-default_values.patch
...........hz-update_values.patch
...........isobatch_ionice2.diff
...........mm-batch_prio.patch
...........mm-kswapd_inherit_prio.patch
...........mm-prio_dependant_scan.patch
...........mm-swap_prefetch-18.patch
...........pdflush-tweaks.patch
...........rt_ionice.diff
...........sched-iso3.2.patch
...........sched-staircase12.1_12.2.patch
...........sched-staircase12.2_13.patch
...........schedbatch2.9.diff
...........schedrange.diff
...........scsi-low_hz_fix.diff
...........smp-nice-support7.diff
...........vm-background_scan-1.diff
...........vm-lots_watermark.diff
...........vm-mapped.diff
...........net-fix_zero-size_datagram_reception.patch
...........patch-2.6.14.1

#
#Some stuff from mm-tree
#
->01-mm_patches.patch
...........activate-sata300-tx4-in-sata_promise.patch
...........add-a-file_permission-helper.patch
...........add-a-vfs_permission-helper.patch
...........add-vfs_-helpers-for-xattr-operations.patch
...........add-via-vt6410-support.patch
...........agp-performance-fixes.patch
...........as-cooperating-processes.patch
...........eagle-and-adi-930-usb-adsl-modem-driver-tidies.patch
...........eagle-and-adi-930-usb-adsl-modem-driver.patch
...........git-acpi-build-fix-3.patch
...........git-acpi-pciehprm_acpi-fix.patch
...........git-acpi-update-8250_acpi.patch
...........git-acpi.patch
...........git-agpgart.patch
...........git-alsa.patch
...........git-blktrace-fixup.patch
...........git-blktrace.patch
...........git-input.patch
...........git-libata-all.patch
...........ide-promise-flushing-hang-fix.patch
...........increase-maximum-kmalloc-size-to-256k-fix.patch
...........increase-maximum-kmalloc-size-to-256k.patch
...........ingo-nfs-stuff.patch
...........ktimers-kt2-export-mktime.patch
...........ktimers-kt2.patch
...........linus.patch
...........memory-leak-in-dentry_open.patch
...........pci-gart-fix.patch
...........vfs-pass-file-pointer-to-filesystem-from-ftruncate.patch
...........vfs-split-dentry-locking-documentation.patch
...........vfs-update-overview-document.patch
...........vxfs-use-generic_ro_fops.patch
...........workaround-for-pnp-device-interrupt.patch
...........write_inode_now-write-inode-if-not-bdi_cap_no_writeback.patch

#
# The rest
#
->03-reiser4-07.11.2005.patch

->04-daconfig-2.3.0.patch

->05-linux-2.6.13-e100-badeeprom.diff

->06-mppe_mppc-iptables_layer7-2.6.14.diff

->07-squashfs_2.2-r2.diff

->08-unionfs-20051014-1448.diff

->09-shfs-0.35_for_2.6.14-rc4.diff

->10-svgalib_helper-1.9.21.diff

->11-usbhid-readd-kconfig.diff

->12-shutup-unneeded-warnings.diff

->13-config-nr-tty-devices.diff

->14-Nitro-FB-logo.diff

->15-fixes

->16-suspend2-2.2-rc9-for-2.6.14.patch

->18-fbsplash-0.9.2-r5-2.6.14.patch

IMPORTANT UPDATES AND EDITS:
*** experimental/testing release, vesafb-tng disabled
*** Typo, came from my copy/paste-happiness
** To be continued...

Howto install

GENTOO:
1)create directory <OVERLAY_DIR>/sys-kernel/nitro-sources (substitute <OVERLAY_DIR> according to overlay path)
2) PROMPT>cd <OVERLAY_DIR>/sys-kernel/nitro-sources
3) PROMPT>wget http://www.stud.uni-karlsruhe.de/~uyavl ... -r1.ebuild
4) PROMPT>ebuild nitro-sources-2.6.14-r1.ebuild digest
5) PROMPT>USE="symlink" emerge nitro-sources
6) PROMPT>cd /usr/src/linux
7) configure, compile and install your kernel
if in doubt about how to do it, consult the documentation in the kernel source directory

ALL OTHER DISTROS:
1) PROMPT>cd /usr/src
2) PROMPT>wget http://www.kernel.org/pub/linux/kernel/ ... 14.tar.bz2
3) PROMPT>wget http://www.stud.uni-karlsruhe.de/~uyavl ... nitro1.bz2
4) PROMPT>tar -xvjf linux-2.6.14.tar.bz2 && rm linux-2.6.14.tar.bz2
5) PROMPT>mv linux-2.6.14 linux-2.6.14-nitro1 && rm linux && ln -s linux-2.6.14-nitro1 linux
6) PROMPT>bunzip2 patch-2.6.14-nitro1.bz2 && cd linux
7) PROMPT>patch -p1 -i ../patch-2.6.14-nitro1
configure, compile and install your kernel
if in doubt about how to do it, consult the documentation in the kernel source directory

NOTE:
for rpm based distros, try doing "make rpm" after menuconfig, although this is untested

And now, even more important part:

LINKS
MANIFEST FILE
DIGEST FILE
EBUILD
PATCH
BROKEN OUT TARBALL
SERIES (DETAILED PATCH LISTING)

Let the compilefest begin !

cheers

T

SchAmane · Post by **SchAmane** » Wed Nov 09, 2005 9:15 pm

Nice, but without vesa-tng it make this kernel little bit bad usable for desktops

Tiger683 · Post by **Tiger683** » Wed Nov 09, 2005 9:15 pm

SchAmane wrote:Nice, but without vesa-tng it make this kernel little bit bad usable for desktops

i wouldnt say so,
for example in my case with normal vesafb with kernel params

Code: Select all

vga=0x31A

looks EXACTLY the same as vesafb-tng with

Code: Select all

video=vesafb:mtrr,ywrap,pmipal,1280x1024-24@75

AND fbsplash works with normal vesafb as well.

dont be too lazy and change your grub.conf

Just in case anyone wants to find out how broken vesafb-tng is in this one:

1) go to drivers/video/Kconfig

2) change

Code: Select all

config FB_VESA_TNG
	bool "vesafb-tng"
	depends on !X86_64 && 0
	select FB_MODE_HELPERS

to

Code: Select all

config FB_VESA_TNG
	bool "vesafb-tng"
	depends on !X86_64
	select FB_MODE_HELPERS

b3cks · Post by **b3cks** » Wed Nov 09, 2005 10:11 pm

<grrml> Just came from the pub... Will test this release maybe tomorrow... night or so.

But thx already...

cheater1034 · Post by **cheater1034** » Wed Nov 09, 2005 11:31 pm

Compiling Now

Thankyou for this release

Thanks, I love nitro, I hope to get it back into the conrad guide

BeteNoire · Post by **BeteNoire** » Wed Nov 09, 2005 11:48 pm

I get this message when I make menuconfig:

Code: Select all

scripts/kconfig/mconf arch/i386/Kconfig
nitro/Kconfig:298: unexpected 'endmenu' statement

Is it a result that I've copied nitro config from previous version?
I've just removed one endmenu statement from the end of the file, now its OK.
And nitro works fine! Thanks!

Which scheduler do you, guys, use by default? Which is better for you and why?

John5788 · Post by **John5788** » Thu Nov 10, 2005 1:20 am

compiled, everything works. in the begining right before my progress bar loads, all the colors turn tutty fruity then go back to normal.

Cinder6 · Post by **Cinder6** » Thu Nov 10, 2005 5:01 am

Sweet, looking forward to using it. Been running with acid2 since that was released, but it still isn't nitro

satanskin · Post by **satanskin** » Thu Nov 10, 2005 6:28 am

Anyone that used acid2 tried this one yet? How you think it compares?

BeteNoire · Post by **BeteNoire** » Thu Nov 10, 2005 8:25 am

My tvcard doesn't work with nitro

I have Pixelvie Ultra Pro with Conexant chipset.
I use kdetv and tvtime and get messages: bad parameter or sth...
I'll paste peace of log when I reboot into nitro again, because now checking newest gentoo-sources - tv works.

Tiger683 · Post by **Tiger683** » Thu Nov 10, 2005 8:36 am

Just to clear things up,
this release is VERY experimental, i just
pushed it out the door, because i wanted
to hear some bugreports about
stuff i cant test myself.

The next release will have all the mm patches cut off.
linus.patch might be the main reason for the broken things like
vesafb-tng and (probably) the reported TVcard problem.
then, the git patches must go too, but as long as noone
screams badly for the libata patch, i think it should be ok....

cheers

T

BeteNoire · Post by **BeteNoire** » Thu Nov 10, 2005 9:52 am

Ok, here's the output of KdeTV when I run it from Xterminal:

Code: Select all

kdetv: WARNING: lirc: Failed to initialize
kdetv: WARNING: V4L2Dev: VIDIOC_QBUF failed: Wrong address
kdetv: WARNING: V4L2Dev: VIDIOC_QBUF failed: Wrong address
kdetv: WARNING: V4L2Dev: VIDIOC_QBUF failed: Wrong address
kdetv: WARNING: V4L2Dev: VIDIOC_DQBUF failed: Bad argument
kdetv: WARNING: V4L2Dev::enqueueBuffer(): broken id: 3066813048
(...)
kdetv: WARNING: V4L2Dev: VIDIOC_DQBUF failed: Bad argument
kdetv: WARNING: V4L2Dev: VIDIOC_QBUF failed: Wrong address
Too many errors. Ending V4L2 grabbing.
V4L2Grabber::~V4L2Grabber(): wait().
V4L2Grabber::~V4L2Grabber(): deleted.

Tiger683 · Post by **Tiger683** » Thu Nov 10, 2005 10:10 am

Didn't you forget to reemerge lirc?
->first line of the error

/EDIT: Also, the dmesg output concerning your particular peace of hardware
would be more interesting, userspace apps don't include info about
kernel drivers (mostly, and even if, then not enough)

BeteNoire · Post by **BeteNoire** » Thu Nov 10, 2005 12:30 pm

I have lirc just... for fun

When I bought tvcard I've recompiled my software with USE=lirc but haven't use it yet

Anyway... KdeTV works fine with gentoo-sources, line about lirc is just a warning.
Here's what I get when I run Tvtime:

Code: Select all

videoinput: Can't read frame. Error was: Bad argument (0).
videoinput: Can't free frame 0: Wrong adress
videoinput: Can't free frame 1: Wrong adress
videoinput: Can't free frame 2: Wrong adress
videoinput: Can't free frame 3: Wrong adress

and... hundreds such lines...
Dmesg is:

Code: Select all

cx88[0]: ioctl 0xc0445624 (v4l2, rw, VIDIOC_QUERYCTRL)
CORE IOCTL: 0xc0445624
cx88[0]: ioctl 0xc0445624 (v4l2, rw, VIDIOC_QUERYCTRL)
CORE IOCTL: 0x80045626
cx88[0]: ioctl 0x80045626 (v4l2, r-, VIDIOC_G_INPUT)
CORE IOCTL: 0xc0045627
cx88[0]: ioctl 0xc0045627 (v4l2, rw, VIDIOC_S_INPUT)
CORE IOCTL: 0xc054561d
cx88[0]: ioctl 0xc054561d (v4l2, rw, VIDIOC_G_TUNER)
CORE IOCTL: 0xc0045627
cx88[0]: ioctl 0xc0045627 (v4l2, rw, VIDIOC_S_INPUT)
CORE IOCTL: 0x80045626
cx88[0]: ioctl 0x80045626 (v4l2, r-, VIDIOC_G_INPUT)
CORE IOCTL: 0x40085618
cx88[0]: ioctl 0xc0445624 (v4l2, rw, VIDIOC_QUERYCTRL)
CORE IOCTL: 0xc008561b
cx88[0]: ioctl 0xc008561b (v4l2, rw, VIDIOC_G_CTRL)
CORE IOCTL: 0xc0445624
cx88[0]: ioctl 0xc0445624 (v4l2, rw, VIDIOC_QUERYCTRL)
CORE IOCTL: 0xc008561b
cx88[0]: ioctl 0xc008561b (v4l2, rw, VIDIOC_G_CTRL)
CORE IOCTL: 0xc02c5638
cx88[0]: ioctl 0xc02c5638 (v4l2, rw, VIDIOC_???)
CORE IOCTL: 0xc054561d
cx88[0]: ioctl 0xc054561d (v4l2, rw, VIDIOC_G_TUNER)
CORE IOCTL: 0xc008561c
cx88[0]: ioctl 0xc008561c (v4l2, rw, VIDIOC_S_CTRL)
CORE IOCTL: 0xc0445624
cx88[0]: ioctl 0xc0445624 (v4l2, rw, VIDIOC_QUERYCTRL)
CORE IOCTL: 0xc008561c

Again hundreds of them...

NiLuJe · Post by **NiLuJe** » Thu Nov 10, 2005 12:54 pm

Panic @ boot, related to the ir module from the bt878 DVB modules I think.

oRDeX · Post by **oRDeX** » Thu Nov 10, 2005 2:31 pm

i'll try it this evening..

predatorfreak · Post by **predatorfreak** » Thu Nov 10, 2005 3:39 pm

Hey, wait a second, if vesa-tng is so busted, why did you even include it in such a state? I mean, if it's so busted, you should just remove it.

mayday147 · Post by **mayday147** » Thu Nov 10, 2005 4:14 pm

This nitro sources has increased my boot time by ~20 seconds. This is fantastic. Great job.

*edit*
My tvcard isn't working neither. I mean, I have just the sound, but no image:

Code: Select all

$tvtime
videoinput: Can't read frame. Error was: Invalid argument (0).
videoinput: Can't free frame 0: Bad address
videoinput: Can't free frame 1: Bad address
videoinput: Can't free frame 2: Bad address
videoinput: Can't free frame 3: Bad address
videoinput: Can't read frame. Error was: Invalid argument (0).
videoinput: Can't free frame 0: Bad address
videoinput: Can't free frame 1: Bad address
videoinput: Can't free frame 2: Bad address
videoinput: Can't free frame 3: Bad address
videoinput: Can't read frame. Error was: Invalid argument (0).
videoinput: Can't free frame 0: Bad address
videoinput: Can't free frame 1: Bad address
videoinput: Can't free frame 2: Bad address
videoinput: Can't free frame 3: Bad address
videoinput: Can't free frame 0: Bad address
videoinput: Can't free frame 1: Bad address
videoinput: Can't free frame 2: Bad address
videoinput: Can't free frame 3: Bad address
(..........)

Tiger683 · Post by **Tiger683** » Thu Nov 10, 2005 4:45 pm

predatorfreak wrote:Hey, wait a second, if vesa-tng is so busted, why did you even include it in such a state? I mean, if it's so busted, you should just remove it.

Because as soon as i fixed it, the incremental patch will be LOTS smaller......

Oh, BTW, it doesn't appear in Kconfig, so noone will accidentally select it, no worries

sobers_2002 · Post by **sobers_2002** » Thu Nov 10, 2005 8:32 pm

am testng and will be reporting back shortly!!

btw @tiger : i heard reiser4 now supports 4k stack, so shouldn't the patch that blocks 4k stack from appearing with reiser4 selected go now?

SchAmane · Post by **SchAmane** » Thu Nov 10, 2005 8:49 pm

sobers_2002 wrote:btw @tiger : i heard reiser4 now supports 4k stack, so shouldn't the patch that blocks 4k stack from appearing with reiser4 selected go now?

Yes, i would ask same. Any opinion about this Tiger ?

This nitro is very stable and smooth to me. Its defenetively stables nitro of all latest nitros. I can run very high load on system and it dont produce any system locks. Seems like Threadsheduling is somehow tweeked.

predatorfreak · Post by **predatorfreak** » Thu Nov 10, 2005 9:03 pm

Tiger683 wrote:
predatorfreak wrote:Hey, wait a second, if vesa-tng is so busted, why did you even include it in such a state? I mean, if it's so busted, you should just remove it.
Because as soon as i fixed it, the incremental patch will be LOTS smaller......
Oh, BTW, it doesn't appear in Kconfig, so noone will accidentally select it, no worries

Yes, but for the time being, you're introducing broken code into you're kernel tree, I'd just keep the patch on the side and hack at it, creating either incramental updates or just updating the entire patch. Right now, you've introduced broken code which could possibly create bugs in other, unforseen places. Bad idea.

priestjim · Post by **priestjim** » Thu Nov 10, 2005 9:11 pm

Where's genetic?

predatorfreak · Post by **predatorfreak** » Thu Nov 10, 2005 9:16 pm

priestjim wrote:Where's genetic?

Last I checked, genetic was considered to be useless outside of finding good tunable values or for finding values for an extremely odd workload, which means, most patchsets should NOT include genetic for desktop usage, unless the creator has an extremely odd workload.

Tiger683 · Post by **Tiger683** » Thu Nov 10, 2005 9:51 pm

predatorfreak wrote:
priestjim wrote:Where's genetic?
Last I checked, genetic was considered to be useless outside of finding good tunable values or for finding values for an extremely odd workload, which means, most patchsets should NOT include genetic for desktop usage, unless the creator has an extremely odd workload.

correct, genetic is useless piece of code causing more overhead than any (placebo-) gain.

regarding vesafb-tng: i would've done it, but i wanted to push it out soon enough to be the first patchset on ck4

[_DEPRECATED__] 2.6.14-nitro1

[_DEPRECATED__] 2.6.14-nitro1