| View previous topic :: View next topic |
| Author |
Message |
priestjim
Tux's lil' helper
Joined: 09 Jan 2005
Posts: 100
Location: Never never land
|
 Posted: Tue May 10, 2005 11:55 am Post subject: |
 |
|
Get the reiserfs patches too! reiser3 people here... 
_________________
My soul is my antenna, I am the instrument + the guitar is my amplifier.
-- Ottmar Liebert |
|
| Back to top |
|
 |
rmh3093
Advocate
Joined: 06 Aug 2003
Posts: 2135
Location: Rochester/Buffalo, NY
|
| Posted: Tue May 10, 2005 1:41 pm Post subject: |
|
|
I dont really know how CFQ vs. Deadline vs. AS benchmarks have been done in the past, but the whole idea of a benchmark dosent sound right. How can you accurately mimic a normal work load. If my computer just booted at its sitting at the desktop waiting for me to start using it and I launch applications, deadline starts programs the fastest. If you start glxgears you can notice that the very first fps is usually much lower than your max but the rest of the fps are at the max of your card. BUT if I can an 'emerge world' going in the background and xmms paying some mp3 and gaim is open and i start using firfox or something else, genetic sched accomodates those new processes smoother, with out skipping xmms, I too like genetic as.
Are all of these reiserfs and ext3 patches you want from -mm?
_________________
Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
| Back to top |
|
|
rmh3093
Advocate
Joined: 06 Aug 2003
Posts: 2135
Location: Rochester/Buffalo, NY
|
| Posted: Tue May 10, 2005 11:40 pm Post subject: |
|
|
bump new version
_________________
Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
| Back to top |
|
|
DaSmurf
n00b
Joined: 15 Apr 2005
Posts: 17
|
| Posted: Wed May 11, 2005 1:24 am Post subject: |
|
|
First of all, thanks for these sources, they rock!
I'm currently running 2.6.12-rc4-skunk1 (+vraxel1 patches for v4l2)
I got the ebuild for skunk2, ran into the problem with asm_offsets.h, but solved that by copying the file like you said. Quickly after that, I got this:
| Code: |
arch/i386/kernel/setup.c: In function `setup_bootmem_allocator':
arch/i386/kernel/setup.c:1150: error: `__PHYSICAL_START' undeclared (first use in this function)
arch/i386/kernel/setup.c:1150: error: (Each undeclared identifier is reported only once
arch/i386/kernel/setup.c:1150: error: for each function it appears in.)
make[1]: *** [arch/i386/kernel/setup.o] Error 1
make: *** [arch/i386/kernel] Error 2
|
I dont know what I need to disable/enable in my config, or if this is just a code problem |
|
| Back to top |
|
|
mpalencia
n00b
Joined: 16 Feb 2005
Posts: 51
Location: Colombia
|
| Posted: Wed May 11, 2005 3:00 am Post subject: |
|
|
| rmh3093 wrote: | | bump new version |
I have one question rmh3093 (or any other guru) why don´t I see the cddvd-cmdfilter-drop.patch anymore?
It used to be in 2.6.12-rc3, but cant find it in con´s site, and It´s not in nitro, and neither in skunk
Is it not necesary anymore?
BTW cool patchset, glad to see you´re back and supportiing genetic.
_________________
Lucky Livecd 32 and 64 bit versions (reiser4 suppport)
http://forums.gentoo.org/viewtopic-p-2986139.html#2986139
There are 2 things infinite: the universe and human stupidity, but I am not sure about the first one. (Einstein) |
|
| Back to top |
|
|
rmh3093
Advocate
Joined: 06 Aug 2003
Posts: 2135
Location: Rochester/Buffalo, NY
|
| Posted: Wed May 11, 2005 3:46 am Post subject: |
|
|
| DaSmurf wrote: | First of all, thanks for these sources, they rock!
I'm currently running 2.6.12-rc4-skunk1 (+vraxel1 patches for v4l2)
I got the ebuild for skunk2, ran into the problem with asm_offsets.h, but solved that by copying the file like you said. Quickly after that, I got this:
| Code: |
arch/i386/kernel/setup.c: In function `setup_bootmem_allocator':
arch/i386/kernel/setup.c:1150: error: `__PHYSICAL_START' undeclared (first use in this function)
arch/i386/kernel/setup.c:1150: error: (Each undeclared identifier is reported only once
arch/i386/kernel/setup.c:1150: error: for each function it appears in.)
make[1]: *** [arch/i386/kernel/setup.o] Error 1
make: *** [arch/i386/kernel] Error 2
|
I dont know what I need to disable/enable in my config, or if this is just a code problem |
that is from the 1g_lowmem1_i386.diff patch, disable it for now and it should compile i will work on a fix
_________________
Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
| Back to top |
|
|
rmh3093
Advocate
Joined: 06 Aug 2003
Posts: 2135
Location: Rochester/Buffalo, NY
|
| Posted: Wed May 11, 2005 3:47 am Post subject: |
|
|
| mpalencia wrote: | | rmh3093 wrote: | | bump new version |
I have one question rmh3093 (or any other guru) why don´t I see the cddvd-cmdfilter-drop.patch anymore?
It used to be in 2.6.12-rc3, but cant find it in con´s site, and It´s not in nitro, and neither in skunk
Is it not necesary anymore?
BTW cool patchset, glad to see you´re back and supportiing genetic.
|
i had noticed that too, i assumed it was included in -rc4 (is it even needed anymore though?)
_________________
Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
| Back to top |
|
|
Cintra
Advocate
Joined: 03 Apr 2004
Posts: 2111
Location: Norway
|
| Posted: Wed May 11, 2005 4:56 am Post subject: |
|
|
| desertstalker wrote: | Something has happend to the reiser 3 in this release. I get kernel oops when attempting to mount root fs (resier3)
Worked fine int rc3-r2 (r3 wouldnt compile)
Thanks |
http://forums.gentoo.org/viewtopic-t-334521.html
_________________
"I am not bound to please thee with my answers" W.S. |
|
| Back to top |
|
|
HecHacker1
Apprentice
Joined: 26 Jun 2003
Posts: 213
Location: UCSD
|
| Posted: Wed May 11, 2005 6:52 am Post subject: |
|
|
yeah! for skunk sources. I dig the logo on startup. Anyways, the lowmem gave me the same error as above. And I agree with others here, Genetic rocks when it comes to usability. A good test is opening beep-media-player or xmms and play an internet stream. Then go to Hardocp.com in firefox and resize the window from the left, making the windows smaller and bigger.
The flash and layout of that page is harsh on cpu when resizing, especially because I have my composite enabled in xorg too.
When I do that with any other combination of scheduler my music pauses and skips/crackles/pops. When I do it with Genetic, I get no problems with beep-media-player. |
|
| Back to top |
|
|
DaSmurf
n00b
Joined: 15 Apr 2005
Posts: 17
|
| Posted: Wed May 11, 2005 7:27 am Post subject: |
|
|
Good call on the 1G lowmem error. I got another one for ya :-/
| Code: |
CC drivers/char/tty_io.o
drivers/char/tty_io.c: In function `tty_ioctl':
drivers/char/tty_io.c:2379: error: `TIOCGDEV' undeclared (first use in this function)
drivers/char/tty_io.c:2379: error: (Each undeclared identifier is reported only once
drivers/char/tty_io.c:2379: error: for each function it appears in.)
make[2]: *** [drivers/char/tty_io.o] Error 1
make[1]: *** [drivers/char] Error 2
make: *** [drivers] Error 2
|
The only thing in the config that I can find even mentioning tty is the "number of tty devices" in skunk options.
I tried with and without legacy pty support, no change.
I also tried with CFQ and Genetic anticipatory I/O controllers, they arent causing the problem either. |
|
| Back to top |
|
|
kiel.wells
Tux's lil' helper
Joined: 10 Sep 2004
Posts: 97
|
| Posted: Wed May 11, 2005 7:46 am Post subject: |
|
|
I get this upon compiling the newest set (rc4-skunk2): | Code: | glock23c linux-2.6.12-rc4-skunk2: sudo make && sudo make modules && sudo make modules_install && sudo make install
CHK include/linux/version.h
SPLIT include/linux/autoconf.h -> include/config/*
CC arch/x86_64/kernel/asm-offsets.s
In file included from include/linux/irq.h:21,
from include/asm/hardirq.h:6,
from include/linux/hardirq.h:6,
from arch/x86_64/kernel/asm-offsets.c:10:
include/asm/irq.h:16:25: irq_vectors.h: No such file or directory
In file included from include/asm/hardirq.h:6,
from include/linux/hardirq.h:6,
from arch/x86_64/kernel/asm-offsets.c:10:
include/linux/irq.h:75: error: `NR_IRQS' undeclared here (not in a function)
In file included from include/linux/irq.h:84,
from include/asm/hardirq.h:6,
from include/linux/hardirq.h:6,
from arch/x86_64/kernel/asm-offsets.c:10:
include/asm/hw_irq.h:28: error: `NR_IRQ_VECTORS' undeclared here (not in a function)
In file included from include/asm/suspend.h:7,
from include/linux/suspend.h:5,
from arch/x86_64/kernel/asm-offsets.c:11:
include/asm/desc.h: In function `__set_tss_desc':
include/asm/desc.h:53: warning: cast from pointer to integer of different size
include/asm/desc.h: In function `set_ldt_desc':
include/asm/desc.h:61: warning: cast from pointer to integer of different size
arch/x86_64/kernel/asm-offsets.c:12:21: asm/pda.h: No such file or directory
arch/x86_64/kernel/asm-offsets.c:16:22: asm/ia32.h: No such file or directory
arch/x86_64/kernel/asm-offsets.c: In function `main':
arch/x86_64/kernel/asm-offsets.c:39: error: dereferencing pointer to incomplete type
arch/x86_64/kernel/asm-offsets.c:40: error: dereferencing pointer to incomplete type
arch/x86_64/kernel/asm-offsets.c:41: error: dereferencing pointer to incomplete type
arch/x86_64/kernel/asm-offsets.c:42: error: dereferencing pointer to incomplete type
arch/x86_64/kernel/asm-offsets.c:43: error: dereferencing pointer to incomplete type
arch/x86_64/kernel/asm-offsets.c:44: error: dereferencing pointer to incomplete type
arch/x86_64/kernel/asm-offsets.c:45: error: dereferencing pointer to incomplete type
make[1]: *** [arch/x86_64/kernel/asm-offsets.s] Error 1
make: *** [arch/x86_64/kernel/asm-offsets.s] Error 2
|
I'm not very knowledgeable about kernel stuff, so any help is much appreciated.
_________________
AMD Athlon64 3000+ Winchester Core @ 1.813 GHz
Asus A8N5X
2GB RAM (4x512) PC3200 400MHz
1 200 GB HDD (SATA), 1 250GB HDD (SATA)
1 NEC DVD-RW, 1 DVD-R/CD-RW
420W Enermax ATX 2.0 |
|
| Back to top |
|
|
priestjim
Tux's lil' helper
Joined: 09 Jan 2005
Posts: 100
Location: Never never land
|
| Posted: Wed May 11, 2005 10:14 am Post subject: |
|
|
For your problem read the first page of this topic. I says how to correct the error. BTW make modules is not needed in the 2.6 tree. They are built automatically.
_________________
My soul is my antenna, I am the instrument + the guitar is my amplifier.
-- Ottmar Liebert |
|
| Back to top |
|
|
kiel.wells
Tux's lil' helper
Joined: 10 Sep 2004
Posts: 97
|
| Posted: Wed May 11, 2005 11:39 am Post subject: |
|
|
unfortunately, /include/asm-x86_64/asm-offsets.h doesn't exist for me. i also looked in /include/asm-ia64/, but no dice.
/*edit
and i also looked in the sources for rc4-skunk1, figuring they can't be so much different that i can't use them....
i'll check rc3-skunk2
edit*/
_________________
AMD Athlon64 3000+ Winchester Core @ 1.813 GHz
Asus A8N5X
2GB RAM (4x512) PC3200 400MHz
1 200 GB HDD (SATA), 1 250GB HDD (SATA)
1 NEC DVD-RW, 1 DVD-R/CD-RW
420W Enermax ATX 2.0 |
|
| Back to top |
|
|
priestjim
Tux's lil' helper
Joined: 09 Jan 2005
Posts: 100
Location: Never never land
|
| Posted: Wed May 11, 2005 12:41 pm Post subject: |
|
|
Some comments :
1. Your patched directory is not clean. There are the .config files and the include/asm directory. When diffing the dirs please make an "make clean distclean mrproper" before. Plus include/asm should be a symlink created automatically in compile time, not a directory. Being a directory makes "make distclean" fail.
2. IPMI doesn't work. Linking vmlinux, I get this :
| Code: |
drivers/built-in.o(.text+0x2e455): In function `ipmi_new_smi':
: undefined reference to `class_device_create'
drivers/built-in.o(.text+0x2e47f): In function `ipmi_smi_gone':
: undefined reference to `class_device_destroy'
drivers/built-in.o(.init.text+0x1d12): In function `init_ipmi_devintf':
: undefined reference to `class_create'
drivers/built-in.o(.init.text+0x1d92): In function `init_ipmi_devintf':
: undefined reference to `class_destroy'
drivers/built-in.o(.init.text+0x1dc7): In function `init_ipmi_devintf':
: undefined reference to `class_destroy'
drivers/built-in.o(.exit.text+0x1e0): In function `cleanup_ipmi':
: undefined reference to `class_destroy'
|
3. Suggestion : provide the patch against 2.6.12-rc4 and not against 2.6.11. This will make a smaller download and it will be easier for us to try different patchsets with less reverse-patching.
_________________
My soul is my antenna, I am the instrument + the guitar is my amplifier.
-- Ottmar Liebert |
|
| Back to top |
|
|
rmh3093
Advocate
Joined: 06 Aug 2003
Posts: 2135
Location: Rochester/Buffalo, NY
|
| Posted: Wed May 11, 2005 12:42 pm Post subject: |
|
|
after organic chem and my bio tests this morning i will see if I can fix this asm-offsets error and the 1gigmem bug
_________________
Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
| Back to top |
|
|
rmh3093
Advocate
Joined: 06 Aug 2003
Posts: 2135
Location: Rochester/Buffalo, NY
|
| Posted: Wed May 11, 2005 12:51 pm Post subject: |
|
|
| priestjim wrote: | Some comments :
1. Your patched directory is not clean. There are the .config files and the include/asm directory. When diffing the dirs please make an "make clean distclean mrproper" before. Plus include/asm should be a symlink created automatically in compile time, not a directory. Being a directory makes "make distclean" fail.
|
i always run make clean mrproper then a script that cleans up .orig and .reg files but i will add disclean, about the patch over 2.6.12-rc4... I used plugsched for 2.6.11 and applied that first to 2.6.11 because its the largest patch, doing it in any other order causes like 57 of 78 regjected patches, i would rater manually apply single -mm patches or -rc patches over the sched patch because the only patches that fail then are sched related patches that dont get included anyway... i think I can still make my patch over 2.6.12-rc4 though, i just need to tinker with the ebuild a little
_________________
Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
| Back to top |
|
|
priestjim
Tux's lil' helper
Joined: 09 Jan 2005
Posts: 100
Location: Never never land
|
| Posted: Wed May 11, 2005 1:28 pm Post subject: |
|
|
Check the ipmi bug too. Just so you know I am a slack user not a gentoo one, so I don't use the ebuilds, just the patch But you can apply the patches in any order you want and then create a nice 2.6.12-rc4 tree and diff -Naur the trees (the patched and the vanilla rc4) and make the patch nice and small . BTW the mm sched patches do not apply to plugsched. But they do to ingosched which is a part of plugsched so...check that out too
Cheers,
priestjim
_________________
My soul is my antenna, I am the instrument + the guitar is my amplifier.
-- Ottmar Liebert |
|
| Back to top |
|
|
_sys/sid
Guru
Joined: 27 Aug 2004
Posts: 346
Location: Asola (Mantova)
|
| Posted: Wed May 11, 2005 2:36 pm Post subject: |
|
|
| Code: |
nathan@nebula [nathan] $ uname -a
Linux nebula 2.6.12-rc4-skunk2 #42 SMP Wed May 11 16:13:23 CEST 2005 i686 Intel(R) Pentium(R) 4 CPU 2.60GHz GenuineIntel GNU/Linux
nathan@nebula [nathan] $ cat /proc/version
Linux version 2.6.12-rc4-skunk2 (root@nebula) (gcc version 3.3.5-20050130 (Gentoo Linux 3.3.5.20050130-r1, ssp-3.3.5.20050130-1, pie-8.7.7.1)) #42 SMP Wed May 11 16:13:23 CEST 2005
nathan@nebula [nathan] $ cat /proc/scheduler
staircase
nathan@nebula [nathan] $ cat /sys/block/hdd/queue/scheduler
noop [anticipatory] deadline cfq
nathan@nebula [nathan] $ mount
/dev/hda8 on / type reiserfs (rw,noatime)
none on /proc type proc (rw)
none on /sys type sysfs (rw)
none on /dev type ramfs (rw)
none on /dev/pts type devpts (rw)
cachedir on /lib/splash/cache type tmpfs (rw)
/dev/hda6 on /var type reiser4 (rw,noatime)
/dev/hda7 on /home type reiserfs (rw,noatime)
/dev/hda2 on /usr/portage type reiser4 (rw,noatime)
none on /dev/shm type tmpfs (rw)
none on /proc/bus/usb type usbfs (rw)
/dev/hda3 on /boot type ext2 (rw,noatime)
nathan@nebula [nathan] $ qpkg -I -v nvidia
media-video/nvidia-glx-1.0.7174-r4 *
media-video/nvidia-kernel-1.0.7174 *
media-video/nvidia-settings-1.0.7167 *
|
_________________
http://th30z.netsons.org/ |
|
| Back to top |
|
|
priestjim
Tux's lil' helper
Joined: 09 Jan 2005
Posts: 100
Location: Never never land
|
| Posted: Wed May 11, 2005 4:16 pm Post subject: |
|
|
I made up a list with all the skunk requests so far :
-parts of rc4-mm when it gets out (I've got a feeling that it is going to be long but I don't know why).
-parts of the ck patchset
-kj patches (kernel janitor's)
-swsusp2
-plugsched
-inotify
-badram
-genetic lib (io + cpu)
-vesafb + fbsplash
-reiser4, fuse (maybe lufs and squashfs?)
-orinoco drivers
-maybe lirc
-kexec
-various patches (cpad, mouse polling patch, tty, acpi dsdt etc. + some nice patches I saw in nitro -> irqbal_fast, slabtune, fasync)
I think that is complete. What do you say?
Edit : Forgot varstacks
_________________
My soul is my antenna, I am the instrument + the guitar is my amplifier.
-- Ottmar Liebert
Last edited by priestjim on Wed May 11, 2005 7:12 pm; edited 1 time in total |
|
| Back to top |
|
|
rmh3093
Advocate
Joined: 06 Aug 2003
Posts: 2135
Location: Rochester/Buffalo, NY
|
| Posted: Wed May 11, 2005 4:48 pm Post subject: |
|
|
new version, just fixes to -skunk2
_________________
Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
| Back to top |
|
|
feld
Guru
Joined: 29 Aug 2004
Posts: 593
Location: WI, USA
|
| Posted: Wed May 11, 2005 6:32 pm Post subject: |
|
|
thank you very much for this update; its looks very nice, except something is broken.
| Code: |
arch/i386/mm/pgtable.c: In function `show_mem':
arch/i386/mm/pgtable.c:27: warning: unused variable `badram'
CC arch/i386/mm/fault.o
CC arch/i386/mm/ioremap.o
CC arch/i386/mm/extable.o
CC arch/i386/mm/pageattr.o
CC arch/i386/mm/mmap.o
LD arch/i386/mm/built-in.o
CC arch/i386/mach-default/setup.o
CC arch/i386/mach-default/topology.o
LD arch/i386/mach-default/built-in.o
CC kernel/sched.o
kernel/sched.c: In function `show_schedstat':
kernel/sched.c:103: error: structure has no member named `ttwu_local'
make[1]: *** [kernel/sched.o] Error 1
make: *** [kernel] Error 2 |
keep on truckin and keep on including plugsched and genetics!
If I had the skills and nobody wanted to do it i'd start one of my own but i'm
a) dumb
b) short on time
c) dumb
thx again!
-Feld
_________________
< bmg505> I think the first line in reiserfsck is
if (random(65535)< 65500) { hose(partition); for (i=0;i<100000000;i++) print_crap(); } |
|
| Back to top |
|
|
priestjim
Tux's lil' helper
Joined: 09 Jan 2005
Posts: 100
Location: Never never land
|
| Posted: Wed May 11, 2005 7:24 pm Post subject: |
|
|
I see you have made the patchset apply on top of rc4. Good job
_________________
My soul is my antenna, I am the instrument + the guitar is my amplifier.
-- Ottmar Liebert |
|
| Back to top |
|
|
rmh3093
Advocate
Joined: 06 Aug 2003
Posts: 2135
Location: Rochester/Buffalo, NY
|
| Posted: Wed May 11, 2005 8:32 pm Post subject: |
|
|
| feld wrote: | thank you very much for this update; its looks very nice, except something is broken.
| Code: |
arch/i386/mm/pgtable.c: In function `show_mem':
arch/i386/mm/pgtable.c:27: warning: unused variable `badram'
CC arch/i386/mm/fault.o
CC arch/i386/mm/ioremap.o
CC arch/i386/mm/extable.o
CC arch/i386/mm/pageattr.o
CC arch/i386/mm/mmap.o
LD arch/i386/mm/built-in.o
CC arch/i386/mach-default/setup.o
CC arch/i386/mach-default/topology.o
LD arch/i386/mach-default/built-in.o
CC kernel/sched.o
kernel/sched.c: In function `show_schedstat':
kernel/sched.c:103: error: structure has no member named `ttwu_local'
make[1]: *** [kernel/sched.o] Error 1
make: *** [kernel] Error 2 |
keep on truckin and keep on including plugsched and genetics!
If I had the skills and nobody wanted to do it i'd start one of my own but i'm
a) dumb
b) short on time
c) dumb
thx again!
-Feld |
I cant figure out what is triggering that error, i cant recreate it. However, try inserting the line below @ line 67 of "kernel/sched.c" and let me know how that goes
| Code: |
unsigned long ttwu_local;
|
_________________
Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
| Back to top |
|
|
DaSmurf
n00b
Joined: 15 Apr 2005
Posts: 17
|
| Posted: Wed May 11, 2005 10:18 pm Post subject: |
|
|
| Code: | | Linux stacker 2.6.12-rc4-skunk3-kraxel1 #2 Wed May 11 18:06:57 EDT 2005 i686 AMD Athlon(tm) AuthenticAMD GNU/Linux |
Thank you so much for fixing those errors, I've never had a system so responsive.
Ingosched + Genetic as rules! |
|
| Back to top |
|
|
kiel.wells
Tux's lil' helper
Joined: 10 Sep 2004
Posts: 97
|
| Posted: Thu May 12, 2005 12:23 am Post subject: |
|
|
She done compiled, now I just gotta reboot.
Thanks
/*edit
Well, I got a kernel panic when I tried to boot into it, but not with my previous skunk sources kernel.
I'm not quite sure why it panic'ed, but It'd be nice to find out why, as I want to try genetic scheduler
*/
_________________
AMD Athlon64 3000+ Winchester Core @ 1.813 GHz
Asus A8N5X
2GB RAM (4x512) PC3200 400MHz
1 200 GB HDD (SATA), 1 250GB HDD (SATA)
1 NEC DVD-RW, 1 DVD-R/CD-RW
420W Enermax ATX 2.0 |
|
| Back to top |
|
|
|
Unsupported Software
