2.6.11-rc1-love1 aka The end of the world

[OneOfOne]

Yep its back!

based on 2.6.11-rc1-mm1
2.6.11-rc1-mm1_to_staircase10.4.diff | staircase 10.4 upped to 2.6.11-rc1-mm1
schedrange.diff | Infrastructure for more policies *
schedbatch2.7.diff | Batch scheduling *
schediso2.10.diff | Isochronous scheduling *
mwII.diff | mapped watermarkII, More gentle ram eviction sooner and less file cache (unlikely to swap).
1g_lowmem1_i386.diff | 1gb lowmem support
defaultcfq.diff | set CFQ as the default IO scheduler
cddvd-cmdfilter-drop.patch | fix burning as non-root
chmp-r5-FULL.patch | Configurable USB HID Mouse Polling
EXPORT_SYMBOL(ipt_find_target); | fix for net/sched/ipt.ko http://lkml.org/lkml/2005/1/14/127
patch-2.6.10-qnet1.bz2 | major patch-o-matic/iptables/iproute patches (http://kem.p.lodz.pl/~peter/qnet/)
2.6.10-ipp2p_0.7.1.patch.bz2 | ipp2p updates (iptable patch http://kem.p.lodz.pl/~peter/qnet/2.6.10 ... .patch.bz2)
vesafb-tng-0.9-rc5-2.6.10-rc2.patch | vesafb-tng ( http://dev.gentoo.org/~spock/projects/vesafb-tng/ )
fbsplash-0.9.1-2.6.11-rc1-mm1.patch | fbsplash ( http://dev.gentoo.org/~spock/projects/gensplash/ )
config-nr-tty-devices.patch | how many /dev/tty's
squashfs2.1-r2-patch | latest squashfs v2.1-r2
2.6.11-rc1-mm1-cfq-ts-20c-quickndirty.patch | cfq-ts (thanks krejler)
isobatch_ionice.diff / rt_ionice.diff | iso nice support
menuconfig-NAME-v2.1-dev5.patch | name in menuconfig
reiser4fix | fix lock up while compiling OpenOffice.Org
nvidia_6111-6629_compat2.diff | compile fix for none-gentoo users
2.6.11-rc1-love1-remove-new-agpgart-features*.patch | fixes nvidia (it works with agpgart driver only for now) and removes devfs
# notes :
# i know i know, I was always against staircase, but I've got to admit, it's lately a very nice scheduler and i like that
# it actually obeys the nice values.
# note, sometimes X is automagiclly niced to -1, make sure to nice to 0 or you will notice some lag.
# for schedbatch/schediso emerge schedtool and read the notes on http://members.optusnet.com.au/ckolivas/kernel/
# **WARNING**WARNING** devfs is gone, so triple check that you either have a static /dev or use udev.

More random notes :
**WARNING** devfs is gone **/WARNING**
lufs is dropped in the favor of fuse (sshfs-fuse), http://love-sources.q2h.com/ebuilds/sys-fs/
Nvidia only works with agpgart driver for now (actually not 100% sure, i might be wrong).
If you use CONFIG_IOPRIO_WRITE (Support for full write io fairness) which is a good idea and want to set ionice, use : http://omfg.linux.dk/pub/misc/ionice.c
Tons of props goes to krejler for helping me and TheCoop/TPC for testing and for Con for making staircase
And one final random note, staircase is very strict about niceness, so make sure X is niced to 0.

patch : http://love-sources.q2h.com/2.6.11-rc1-love1.bz2 (applies on top of 2.6.11-rc1-mm1)
ebuild : http://love-sources.q2h.com/love-source ... -r1.ebuild

(yes i know i was against staircase, but i have to admit it got much better)

---edit----
fix for compiling on amd64
http://omfg.linux.dk/pub/patches/2.6.11 ... sues.patch (thx krejler one more time )

---- one more time
nvidia patch (for nvidia-kernel) : http://omfg.linux.dk/pub/patches/2.6.11 ... dfix.patch
ati drivers patch :
http://omfg.linux.dk/pub/patches/2.6.11 ... -fix.patch
---^^^^ UPDATED
peace

[TheCoop]

ditto, this is a cracking release. Even got nvidia sorted

[OneOfOne]

fixed the ebuild..

peace

[Tuti]

works great here! good job!

[vrln]

Glad to see love-sources making a comeback with style :) Good job, everything is working here. Actually this is the first mm based patchset that has worked for me... Finally love with staircase, which is something I've used for a long time.

as usual with -mm, some odd messages in dmesg for me:
(just copypasted from dmesg)

FDC 0 is a post-1991 82077
ram0: unknown partition table
ram1: unknown partition table
ram2: unknown partition table
ram3: unknown partition table
ram4: unknown partition table
ram5: unknown partition table
ram6: unknown partition table
ram7: unknown partition table
ram8: unknown partition table
ram9: unknown partition table
ram10: unknown partition table
ram11: unknown partition table
ram12: unknown partition table
ram13: unknown partition
RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize table
ram14: unknown partition table
ram15: unknown partition table
Probing IDE interface ide3...
ide3: Wait for ready failed before probe !
Probing IDE interface ide4...
ide4: Wait for ready failed before probe !
Probing IDE interface ide5...
ide5: Wait for ready failed before probe !
Uniform CD-ROM driver Revision: 3.20
cdrom: open failed.
hdd: ATAPI 48X DVD-ROM drive, 256kB Cache, UDMA(33)
cdrom: open failed.

But as already said, everything still works so far, including fbsplash which I couldn't get to work myself, even with the latest patch :)

[Zyne]

love-sources have always been my favourite!!!!
thanks for making a new release m8!
greatly appreciated!

time to build a new kernel now

[Pink]

Haven't tried it - I'm running a stock 2.6.11-rc1 with win4lin but i wanted to say Welcome Back OneofOne - it's been too long

[MrStaticVoid]

Working well here. Thanks!

[discomfitor]

Love-sources is back and the world is right again. Hallelujah!

[Evil Dark Archon]

I've got a big compile error when compiling this on amd64, unfortunately it's too long to post the entire thing here but suffice it to say, in the 32bit emulation code, there some lingering traces of devfs, more specifically fs/compat_ioctl.c is looking for the devfs_fs header and there are a lot of undeclared functions and non-constant constants also arch/x86_64/ia32/ia32_ioctl.c is affected.

[krejler]

I've got a big compile error when compiling this on amd64, unfortunately it's too long to post the entire thing here but suffice it to say....

Haha, having second thoughts?
The lenghy post was a bit better for finding a possible sollution, though.

Well, I can't really say that I've compile-tested this, but I think it might be the answer:

http://omfg.linux.dk/pub/patches/2.6.11 ... sues.patch

Please report back.

Thanks.

[Evil Dark Archon]

thank you, that fixed it, the reason i couldn't keep the long error up was because the forum software cut it off, but your fix was right on.

[krejler]

thank you, that fixed it, the reason i couldn't keep the long error up was because the forum software cut it off, but your fix was right on.

Great!

[Zyne]

I seem to have a problem getting my ATI-drivers to run...
for some reason the module doesn't show up in /lib/modules/2.6.11-rc1-mm1/video/

does anyone else have this problem?
I gotta admit that I added the vesafb-tng and the 1GB low memory patch manually afterwards, but I didn't get any rejected hunks, so I don't think that that could be the problem...

I tried ati-drivers-3.14.6 and 3.14.1-r2 (someone from the forum made that for the pci patch)


any idea's?

[krejler]

I seem to have a problem getting my ATI-drivers to run...
for some reason the module doesn't show up in /lib/modules/2.6.11-rc1-mm1/video/

does anyone else have this problem?
I gotta admit that I added the vesafb-tng and the 1GB low memory patch manually afterwards, but I didn't get any rejected hunks, so I don't think that that could be the problem...

I tried ati-drivers-3.14.6 and 3.14.1-r2 (someone from the forum made that for the pci patch)


any idea's?

First of all, vesafb-tng is already in... Why are you trying to add it once more?

Second of all, the module oughten't show up in /lib/modules/2.6.11-rc1-mm1/video/. I really don't know where ATi's installer (or portage, for that matter) puts the module, but if I'm to follow your suggestion, I'd say somewhere around /lib/modules/2.6.11-rc1-love1/video/.

I really have no experience with the ATi-drivers, as I implied earlier on. Did it fail to compile? Did it only fail to install?

I can't really help you that much, without a better description of the problem.

Thanks.

[charlieg]

Yep its back!

based on 2.6.11-rc1-mm1
2.6.11-rc1-mm1_to_staircase10.4.diff | staircase 10.4 upped to 2.6.11-rc1-mm1

WTF? Staircase scheduler? With -nitro including -mm these days, is there actually any difference between -love and -nitro? The two patchsets seem to have ended up on the same path! They'll soon be damn near identical!

[Zyne]

First of all, vesafb-tng is already in... Why are you trying to add it once more?

That's weird, cause it wasn't in there when I first compiled it...

Second of all, the module oughten't show up in /lib/modules/2.6.11-rc1-mm1/video/.

I agree with you on this, BUT appearantly they do, unless if you manually edit the Makefile.

maybe I just have to start over again...

edit: I think I figured it out...
I manually applied all the patches, and now all seems to work great... I must've done something wrong with the ebuild before...

edit2: well, now I noticed vesafg-tng is indeed built into the kernel, but my ATI-drivers still aren't working...

[krejler]

Yep its back!

based on 2.6.11-rc1-mm1
2.6.11-rc1-mm1_to_staircase10.4.diff | staircase 10.4 upped to 2.6.11-rc1-mm1

WTF? Staircase scheduler? With -nitro including -mm these days, is there actually any difference between -love and -nitro? The two patchsets seem to have ended up on the same path! They'll soon be damn near identical!

[sarcasm]
Excuse me?

You must have overseen this:

Love-sources is back and the world is right again. Hallelujah!

As far as I know, Darckness is currently maintaining -nitro. It may just be me, but I clearly am able to notice how Darckness points out that nitro has been defeated...
[/sarcasm]

[krejler]

First of all, vesafb-tng is already in... Why are you trying to add it once more?

That's weird, cause it wasn't in there when I first compiled it...

Second of all, the module oughten't show up in /lib/modules/2.6.11-rc1-mm1/video/.

I agree with you on this, BUT appearantly they do, unless if you manually edit the Makefile.

maybe I just have to start over again...

edit: I think I figured it out...
I manually applied all the patches, and now all seems to work great... I must've done something wrong with the ebuild before...

edit2: well, now I noticed vesafg-tng is indeed built into the kernel, but my ATI-drivers still aren't working...

Phew, I fixed it.

It _SHOULD_ work. I have successfully built the module, but it fails to load, because it is unable to detect the ATi card that I do not have.

http://omfg.linux.dk/pub/patches/2.6.11 ... ifix.patch

EDIT:

edit: I think I figured it out...
I manually applied all the patches, and now all seems to work great... I must've done something wrong with the ebuild before...

Aww, why didn't I notice that earlier?

Oh well, this patch should work fine for other ATi lusers.

EDIT2:

edit2: well, now I noticed vesafg-tng is indeed built into the kernel, but my ATI-drivers still aren't working...

Aww, why didn't I notice THAT EARLIER TOO?
Haha! I really should learn to read better.

Anyway, It should work for you. Please report back!

[Zyne]

Oh well, this patch should work fine for other ATi lusers.

wow... Humor really seems to rise with time...

and for those of you that are wondering.. You have to apply this patch on the ATI-drivers, and not on the kernel sources... might not be clear for everyone...


I'll report back in a few seconds, but at least the module is there now...


edit: it's not workign AT ALL...
I can load the module fine, but from the moment I start xorg, the system just hardlocks... I can't even ssh in my box anymore

[krejler]

Oh well, this patch should work fine for other ATi lusers.

wow... Humor really seems to rise with time...

and for those of you that are wondering.. You have to apply this patch on the ATI-drivers, and not on the kernel sources... might not be clear for everyone...


I'll report back in a few seconds, but at least the module is there now...


edit: it's not workign AT ALL...
I can load the module fine, but from the moment I start xorg, the system just hardlocks... I can't even ssh in my box anymore

That's weird.

As I pointed out earlier, I don't have an ATi card, so I am unable to test it myself. I also don't know a whole lot about the drivers either.

Do the drivers provide their own AGP-interface, like Nvidia's NvAGP, or do they work through the kernel's AGPGART?

If that's the case, you might want to use the kernel's AGPGART, since the native implementation could be broken like NvAGP is.

Are other ATi-people having the same issues? If so, please respond.

[ahorn]

i can not build nvidia-kernel on amd64 platform.

Code: Select all

NVIDIA: calling KBUILD...
make CC=cc  KBUILD_VERBOSE=1 -C /usr/src/linux M=/var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv modules
make[2]: Warnung: Kein Jobserver verfügbar: setzen -j1. Fügen »+« zur Ursprungsregel hinzu.
mkdir -p /var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv/.tmp_versions
make -f scripts/Makefile.build obj=/var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv
echo \#define NV_COMPILER \"`cc -v 2>&1 | tail -n 1`\" > /var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv/nv_compiler.h
  cc -Wp,-MD,/var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv/.nv.o.d -nostdinc -isystem /usr/lib/gcc/x86_64-pc-linux-gnu/3.4.3/include -D__KERNEL__ -Iinclude  -Wall -Wstrict-prototypes -Wno-trigraphs -fno-strict-aliasing -fno-common -ffreestanding -O2     -fomit-frame-pointer -march=k8 -mno-red-zone -mcmodel=kernel -pipe -fno-reorder-blocks        -Wno-sign-compare -fno-asynchronous-unwind-tables -funit-at-a-time -mno-sse -mno-mmx -mno-sse2 -mno-3dnow -Wdeclaration-after-statement -I/var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv -Wall -Wimplicit -Wreturn-type -Wswitch -Wformat -Wchar-subscripts -Wparentheses -Wno-multichar  -Werror -O -fno-common -mno-red-zone -MD   -Wno-cast-qual -Wno-error -D_LOOSE_KERNEL_NAMES -D__KERNEL__ -DMODULE  -mcmodel=kernel -DNTRM -D_GNU_SOURCE -D_LOOSE_KERNEL_NAMES -D__KERNEL__ -DMODULE  -DNV_MAJOR_VERSION=1 -DNV_MINOR_VERSION=0 -DNV_PATCHLEVEL=6629  -DNV_UNIX  -DNV_LINUX  -DNV_INT64_OK  -DNVCPU_X86_64  -DNV_64_BITS    -UDEBUG -U_DEBUG -DNDEBUG -DNV_REMAP_PFN_RANGE_PRESENT -DNV_CHANGE_PAGE_ATTR_PRESENT -DNV_PCI_DISABLE_DEVICE_PRESENT -DNV_CLASS_SIMPLE_CREATE_PRESENT -DNV_PCI_GET_CLASS_PRESENT  -DMODULE -DKBUILD_BASENAME=nv -DKBUILD_MODNAME=nvidia -c -o /var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv/.tmp_nv.o /var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv/nv.c
/var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv/nv.c: In function `_get_phys_address':
/var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv/nv.c:2529: Fehler: structure hat kein Element namens »pud«
/var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv/nv.c: In function `nv_agp_init':
/var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv/nv.c:3010: Warnung: »inter_module_put« ist veraltet (deklariert bei include/linux/module.h:578)
make[3]: *** [/var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv/nv.o] Fehler 1
make[2]: *** [_module_/var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv] Fehler 2
NVIDIA: left KBUILD.
nvidia.ko failed to build!
make[1]: *** [module] Fehler 1
make: *** [module] Fehler 2

!!! ERROR: media-video/nvidia-kernel-1.0.6629-r1 failed.
!!! Function linux-mod_src_compile, Line 330, Exitcode 2
!!! Unable to make                                  IGNORE_CC_MISMATCH=yes V=1 SYSSRC=/usr/src/linux SYSOUT=/usr/src/linux clean module.
!!! If you need support, post the topmost build error, NOT this status message

with latest nitro or morph it also doesnt work. so maybe its the mm-patchset.

[krejler]

i can not build nvidia-kernel on amd64 platform.

Code: Select all

NVIDIA: calling KBUILD...
make CC=cc  KBUILD_VERBOSE=1 -C /usr/src/linux M=/var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv modules
make[2]: Warnung: Kein Jobserver verfügbar: setzen -j1. Fügen »+« zur Ursprungsregel hinzu.
mkdir -p /var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv/.tmp_versions
make -f scripts/Makefile.build obj=/var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv
echo \#define NV_COMPILER "`cc -v 2>&1 | tail -n 1`" > /var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv/nv_compiler.h
  cc -Wp,-MD,/var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv/.nv.o.d -nostdinc -isystem /usr/lib/gcc/x86_64-pc-linux-gnu/3.4.3/include -D__KERNEL__ -Iinclude  -Wall -Wstrict-prototypes -Wno-trigraphs -fno-strict-aliasing -fno-common -ffreestanding -O2     -fomit-frame-pointer -march=k8 -mno-red-zone -mcmodel=kernel -pipe -fno-reorder-blocks        -Wno-sign-compare -fno-asynchronous-unwind-tables -funit-at-a-time -mno-sse -mno-mmx -mno-sse2 -mno-3dnow -Wdeclaration-after-statement -I/var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv -Wall -Wimplicit -Wreturn-type -Wswitch -Wformat -Wchar-subscripts -Wparentheses -Wno-multichar  -Werror -O -fno-common -mno-red-zone -MD   -Wno-cast-qual -Wno-error -D_LOOSE_KERNEL_NAMES -D__KERNEL__ -DMODULE  -mcmodel=kernel -DNTRM -D_GNU_SOURCE -D_LOOSE_KERNEL_NAMES -D__KERNEL__ -DMODULE  -DNV_MAJOR_VERSION=1 -DNV_MINOR_VERSION=0 -DNV_PATCHLEVEL=6629  -DNV_UNIX  -DNV_LINUX  -DNV_INT64_OK  -DNVCPU_X86_64  -DNV_64_BITS    -UDEBUG -U_DEBUG -DNDEBUG -DNV_REMAP_PFN_RANGE_PRESENT -DNV_CHANGE_PAGE_ATTR_PRESENT -DNV_PCI_DISABLE_DEVICE_PRESENT -DNV_CLASS_SIMPLE_CREATE_PRESENT -DNV_PCI_GET_CLASS_PRESENT  -DMODULE -DKBUILD_BASENAME=nv -DKBUILD_MODNAME=nvidia -c -o /var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv/.tmp_nv.o /var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv/nv.c
/var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv/nv.c: In function `_get_phys_address':
/var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv/nv.c:2529: Fehler: structure hat kein Element namens »pud«
/var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv/nv.c: In function `nv_agp_init':
/var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv/nv.c:3010: Warnung: »inter_module_put« ist veraltet (deklariert bei include/linux/module.h:578)
make[3]: *** [/var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv/nv.o] Fehler 1
make[2]: *** [_module_/var/tmp/portage/nvidia-kernel-1.0.6629-r1/work/NVIDIA-Linux-x86_64-1.0-6629-pkg2/usr/src/nv] Fehler 2
NVIDIA: left KBUILD.
nvidia.ko failed to build!
make[1]: *** [module] Fehler 1
make: *** [module] Fehler 2

!!! ERROR: media-video/nvidia-kernel-1.0.6629-r1 failed.
!!! Function linux-mod_src_compile, Line 330, Exitcode 2
!!! Unable to make                                  IGNORE_CC_MISMATCH=yes V=1 SYSSRC=/usr/src/linux SYSOUT=/usr/src/linux clean module.
!!! If you need support, post the topmost build error, NOT this status message

with latest nitro or morph it also doesnt work. so maybe its the mm-patchset.

Hello!

This patch might fix your issue:
http://omfg.linux.dk/pub/patches/nvidia-pudpatch.diff

It is supposed to be applied onto the nvidia driver, not the kernel.

Please report back.

EDIT:

It doesn't apply on the heavily patched nvidia-kernel. Give me a moment.

[discomfitor]

As far as I know, Darckness is currently maintaining -nitro. It may just be me, but I clearly am able to notice how Darckness points out that nitro has been defeated...
[/sarcasm]

I wasn't aware that this was a competition...

[neenee]

works fine for me nice one oneofone and krejler