2.6.7_rc1-love1 aka 'S0 1337'

[OneOfOne]

Here comes the love!
guess you guys will be stuck with me for a while since steel300 will be busy. :-/
here are the notes :

linux-2.6.6.tar.bz2 | 2.6.6
patch-2.6.7-rc1.bz2 | 2.6.7-rc1 patch
2.6.7-rc1-mm1.bz2 | 2.6.7-rc1-mm1
v31.bz2 | new nicksched
elv-select.bz2 | elevator selector, send your hugs and kisses to steel300!
squashfs2.0-linux-2.6.6.patch.bz2 | SquashFS v2
acerhk.patch.bz2 | Acer Hotkeys
acpi-dsdt-initrd-patch-v0.4-2.6.4-eric.patch.bz2 | Custom acpi dsdt patch
subfs-genstef.bz2 | SubFS (death to supermount!)
BadRAM-2.6.5.2.patch.bz2 | a workaround bad ram
bd-claim.patch.bz2 | EVMS fix
fbdev-mode-switching-fix.patch.bz2 | reversed for bootsplash
bootsplash{2,3}.diff | fixes for bootsplash (thanks LSD`)
bootsplash-3.1.4-2.6.6.diff.bz2 | bootsplash
config-nr-tty-devices.patch.bz2 | for a cleaner /dev
ipw2100-0.44 | ipw2100
layer7-kernelpatch-nf-0.4.0.bz2 | layer7 netfilter target
lirc-2.6.5-mm1-20040406.bz2 | new lirc patch
4105_lirc_infrared-2.6.5-rc2.patch | new lirc patch, ty g-d-s
4905_speakup_accessibility.patch | speakup patch
4720_gcloop-2.6-20040330.patch | gcloop
love_ppm.patch.bz2 | love-sources' logo
lufs-0.9.7-2.6.0-test9.patch.bz2 | lufs
menuconfig-NAME-v1.0.patch.bz2 | name in menuconfig patch
2.6.x-bring-up-compiler-option.patch.bz2 | extra CFLAGS hack
sii-3512-update.patch.bz2 | siimage patch
patch-2.4.x-vesafb-rrc.bz2 | vesafb refresh hack
patch-2.6.4-am9.bz2 | auto swappiness
patch-2.6.4-amd76x_pm.bz2 | AMD Power NOW!
viafb_02.diff.bz2 | via framebuffer
via-v4l-1.4a-drm.patch.bz2 | via v4l stuff
acx100-0.2.0pre7-linux2.6.2.diff.bz2 | ACX100 Driver
iteraid_1.44.patch.bz2 | Giga Raid
linux-2.6.4-Walken.diff.bz2 | guess
orinoco-0.13e-patch.diff.bz2 | couldnt find the newer driver :-/
mki-adapter26_1_3_5.patch.bz2 | stuff needed by win4lin
Kernel-Win4Lin3-2.6.6.patch.bz2 | Win4Lin *sigh*, why can't people use vmware?

elv-select is back!
and brand new v31 nicksched.
known problems :
reiser4 is gone, i'll try to pull a bk snapshot and see how it goes.
win4lin *might be* broken.
bootsplash doesn't work until the bootsplash init script starts
known features :
supermount is gone for GOOD, i included submount which is less evil but that's a temp fix until ivman ( http://ivman.sf.net ) gets more popular.

ivman is generally a daemon version of gnome-volume-manager which uses udev/dbus/hal to auto mount/umount and its the right way to do it without all the gnome deps.
(i know the name is lame )

edit: oops forgot the urls
http://oneofone.limitlessfx.com/love-so ... -love1.bz2
http://oneofone.limitlessfx.com/love-so ... -r1.ebuild
http://oneofone.limitlessfx.com/love-so ... 1-r1.notes

edit:
fixed release is up to fix regparam problems, please redigest and download.
enjoy
peace

[discomfitor]

Checking it out now...I have high hopes.
Link here if anyone wants it:
http://oneofone.limitlessfx.com/love-sources/

UPDATE: Tried it, boots fine and seems to work alright. DID NOT work with the 5341 nvidia stuff, had massive errors so I went back to 5336. The bug with wine accessing the cdrom is still there, as it has been since 2.6.6-love2, and I still don't know what is causing it.

[ejohnson]

OneofOne, I still get the Reiser4 small keys warning on bootup. I'll try the patch posted in the last thread.

Mirrored with ebuild here

EDIT: Oh shi. I should really read the changlogs

[Sgaduuw]

/me is waiting for the first persons to ask where supermount is

[Jake]

Is there a new version of elv-select? steel300 dropped the old one because of USB drive issues.

[Evil Dark Archon]

win4lin is broken

Code: Select all

arch/i386/mki/mki.c:92: error: conflicting types for 'do_exit'
include/linux/kernel.h:61: error: previous declaration of 'do_exit' was here
arch/i386/mki/mki.c:92: error: conflicting types for 'do_exit'
include/linux/kernel.h:61: error: previous declaration of 'do_exit' was here
make[1]: *** [arch/i386/mki/mki.o] Error 1
make: *** [arch/i386/mki] Error 2

[HydroSan]

/me is waiting for the first persons to ask where supermount is

I've used Submount for the last while and have never looked back.

[RainmaN]

Almost finishes compiling... but

Code: Select all

fs/built-in.o(.init.text+0xcaa): In function `proc_misc_init':
: undefined reference to `proc_schedstat_operations'
make: *** [.tmp_vmlinux1] Error 1

[silverpig]

I got it compiled and installed but the boot hangs on pci hotplugging for some reason.

[UncleOwen]

DID NOT work with the 5341 nvidia stuff,

What 5341 stuff do you mean? AFAICS the latest version of the nvidia-driver is 5336, both on nvidia.com and in portage.

[silverpig]

Compiled, installed, running (just had to remove hotplug from the startup)... Just can't get ati opengl to work right... Argh.

Re-emerged the drivers, opengl-update ati works fine, restarted X, but I still get mesa running my 3d.

[silverpig]

Is this a kernel problem or a me problem?

from dmesg:

Code: Select all

fglrx: module license 'Proprietary. (C) 2002 - ATI Technologies, Starnberg, GERMANY' taints kernel.
[fglrx] Maximum main memory to use for locked dma buffers: 930 MBytes.
Unable to handle kernel NULL pointer dereference at virtual address 00000000
 printing eip:
f8cda280
*pde = 00000000
Oops: 0002 [#1]
Modules linked in: fglrx 3c59x sg scsi_mod
CPU:    0
EIP:    0060:[<f8cda280>]    Tainted: P   VLI
EFLAGS: 00010286   (2.6.7-rc1-love1)
EIP is at __ke_memset+0x10/0x20 [fglrx]
eax: 00000000   ebx: 00000000   ecx: f8d02768   edx: 00000000
esi: f8d02240   edi: 00000000   ebp: c02d30c0   esp: f7bd9f24
ds: 007b   es: 007b   ss: 0068
Process modprobe (pid: 4777, threadinfo=f7bd8000 task=f7bdb790)
Stack: f8d02760 f8d02630 f8ce4742 f8d02768 00000000 00000010 f8d025e0 f8d02240
       f8d02630 f8d025e0 f8cdbfa7 f8d02760 00000000 00000d28 00000000 00000282
       f8d025e0 f8d025e0 f8c52257 f8d025e0 00000000 0000001b 0000001a 0000001c
Call Trace:
 [<f8ce4742>] _r6x_IsDeviceSupported+0x22/0x1790 [fglrx]
 [<f8cdbfa7>] firegl_init+0xe7/0x110 [fglrx]
 [<f8c52257>] firegl_init_module+0xe7/0x17d [fglrx]
 [<c0127083>] sys_init_module+0xf3/0x1c0
 [<c010556b>] syscall_call+0x7/0xb
                                                                               
Code: ff ff 89 06 83 c6 04 39 ef 72 e5 8b 04 24 83 c4 08 5b 5e 5f 5d c3 90 8d 74 26 00 83 ec 08 89 1c 24 89 7c 24 04 89 c3 89 df 89 d0 <f3> aa 89 d8 8b 7c 24
04 8b 1c 24 83 c4 08 c3 90 83 ec 08 81 f9

[col]

yes win4lin stuff is broken....

[gun26]

I get the same error as Rainman:

Code: Select all

GEN     .version
  CHK     include/linux/compile.h
  UPD     include/linux/compile.h
  CC      init/version.o
  LD      init/built-in.o
  LD      .tmp_vmlinux1
fs/built-in.o(.init.text+0xe6e): In function `proc_misc_init':
: undefined reference to `proc_schedstat_operations'
make: *** [.tmp_vmlinux1] Error 1

Also, for whatever it's worth (not much to me) compressloop support gives the same old "undefined reference to blk_run_queues" error as the last bunch of Love versions,, and Win4lin is broken again.

[Legoguy]

Ditto to the "undefined reference to `proc_schedstat_operations'"...

[hjlane3]

I'd really like the config_hz patch back in love-sources. I need it for my laptop to not make weird noises when running off battery. :-\

[Evil Dark Archon]

fastest kernel for me ever, but i can't get any out of tree modules working, btw, i reversed the move-__this-module-to-modpost patch (this is all referring to my test build of evil-sources, which is based on this patchset

[eikketk]

Lame? LAME? Tell me how I *should* call it then!!! "The Black Dog and Sheep's Volume Manager" doesn't sound really professional isn't it?

[DarwinianCoeus]

how come walken is still being patched in? when he was first added, i thought maybe it was his birthday. does someone (steel?) have a secret crush?

[Admiral LSD]

bootsplash{2,3}.diff | fixes for bootsplash (thanks LSD`)

heh, cool but I must say the patches aren't my work. Dinini created them for an earlier mm/love.

Is this a kernel problem or a me problem?

from dmesg:

Code: Select all

fglrx: module license 'Proprietary. (C) 2002 - ATI Technologies, Starnberg, GERMANY' taints kernel.
[fglrx] Maximum main memory to use for locked dma buffers: 930 MBytes.
Unable to handle kernel NULL pointer dereference at virtual address 00000000
 printing eip:
f8cda280
*pde = 00000000
Oops: 0002 [#1]
Modules linked in: fglrx 3c59x sg scsi_mod
CPU:    0
EIP:    0060:[<f8cda280>]    Tainted: P   VLI
EFLAGS: 00010286   (2.6.7-rc1-love1)
EIP is at __ke_memset+0x10/0x20 [fglrx]
eax: 00000000   ebx: 00000000   ecx: f8d02768   edx: 00000000
esi: f8d02240   edi: 00000000   ebp: c02d30c0   esp: f7bd9f24
ds: 007b   es: 007b   ss: 0068
Process modprobe (pid: 4777, threadinfo=f7bd8000 task=f7bdb790)
Stack: f8d02760 f8d02630 f8ce4742 f8d02768 00000000 00000010 f8d025e0 f8d02240
       f8d02630 f8d025e0 f8cdbfa7 f8d02760 00000000 00000d28 00000000 00000282
       f8d025e0 f8d025e0 f8c52257 f8d025e0 00000000 0000001b 0000001a 0000001c
Call Trace:
 [<f8ce4742>] _r6x_IsDeviceSupported+0x22/0x1790 [fglrx]
 [<f8cdbfa7>] firegl_init+0xe7/0x110 [fglrx]
 [<f8c52257>] firegl_init_module+0xe7/0x17d [fglrx]
 [<c0127083>] sys_init_module+0xf3/0x1c0
 [<c010556b>] syscall_call+0x7/0xb
                                                                               
Code: ff ff 89 06 83 c6 04 39 ef 72 e5 8b 04 24 83 c4 08 5b 5e 5f 5d c3 90 8d 74 26 00 83 ec 08 89 1c 24 89 7c 24 04 89 c3 89 df 89 d0 <f3> aa 89 d8 8b 7c 24
04 8b 1c 24 83 c4 08 c3 90 83 ec 08 81 f9

That might be because you have CONFIG_REGPARM enabled. Love reverses the patch that makes this the permanent kernel default so go in to "Processor Type and Features" and check to see if the "Use Register Arguments" option is diabled.

[Chaosite]

Hey,

Can you update the ACX100 driver? You seem to be including pre7, which is kinda old by now.

Latest version can be found here:

http://lisas.de/~andi/acx100/acx100-0.2 ... _9.tar.bz2

(Thats where new versions get posted)

Thanks!

[Pink]

The win4lin problem, at least on my machine, is the same as last time:

Code: Select all

  CC [M]  arch/i386/mki-adapter26/mki-main.o
  CC [M]  arch/i386/mki-adapter26/mki26.o
arch/i386/mki-adapter26/mki26.c: In function `mkia_remove_page_ref':
arch/i386/mki-adapter26/mki26.c:1537: error: structure has no member named `count'
make[1]: *** [arch/i386/mki-adapter26/mki26.o] Error 1
make: *** [arch/i386/mki-adapter26] Error 2

So apply the same fix I posted last time and it compiles just fine:

Code: Select all

(Line 1537 arch/i386/mki-adapter26/mki26.c)


Change

	MKIA_ASSERT(atomic_read(&pageinfop->counter) >= 2);

to

	MKIA_ASSERT(page_count(pageinfop) >= 2);

However, I cannot test it as I am on a reiser4 system, but as it compiles, it should work ok.

The other win4lin problem posted above by Evil Dark Archon, I don't know, I can't reproduce it...

[PrakashP]

Yup, OneOfOne forgot to take out permanenet regparm...

[sxz]

Well, the reason i prefered supermount instead of submount, was because gnome didn`t like it
First of all, when i opened a cd with nautilus, the window closed every few seconds, i don`t know why Secondly, when i opened gnome`s system monitor, it would display all my removable media, even if it wasn`t there. And if, say a floppy, was there, it would get accessed every few seconds, probably to refresh the information or smth. Supermount, on the other hand, got rid of these problems.
Sorry for the offtopic

[OneOfOne]

Hey,

Can you update the ACX100 driver? You seem to be including pre7, which is kinda old by now.

Latest version can be found here:

http://lisas.de/~andi/acx100/acx100-0.2 ... _9.tar.bz2

(Thats where new versions get posted)

Thanks!

thanks for the patch, i'll include it next love.

damn, sorry about the regparm patch, i'll release a fix in a few.

@win4lin problems, i honest have no idea about that error, but i'm open for fixes or we'll have to wait until they release a new patch.
@proc_misc_init problems, can you please post your config? try to turn on Kernel hacking -> "Collect scheduler statistics" or if its already on try to turn it off.

peace