View previous topic :: View next topic |
Author |
Message |
Lovechild Advocate
Joined: 17 May 2002 Posts: 2858 Location: Århus, Denmark
|
Posted: Tue May 27, 2003 7:54 pm Post subject: |
|
|
AlterEgo wrote: | First looks:
>Tux-framebuffer bootlogo is still gone
>IPtables and Iproute still fail to build running this kernel.
>Furthermore, without making any change (that I'm aware of) I no longer can compile glibc. It's failing with a lot of references to /lib/modules/2.5.70.
It ends like this:
Quote: |
gcc ../sysdeps/generic/libc-start.c -c -std=gnu99 -O2 -Wall -Winline -Wstrict-prototypes -Wwrite-strings -freorder-blocks -march=i686 -I../nptl -I../include -I. -I/var/tmp/portage/glibc-2.3.2-r1/work/glibc-2.3.2/buildhere/csu -I.. -I../libio -I/var/tmp/portage/glibc-2.3.2-r1/work/glibc-2.3.2/buildhere -I../sysdeps/i386/elf -I../nptl/sysdeps/unix/sysv/linux/i386/i686 -I../nptl/sysdeps/unix/sysv/linux/i386 -I../nptl/sysdeps/unix/sysv/linux -I../nptl/sysdeps/pthread -I../sysdeps/pthread -I../nptl/sysdeps/unix/sysv -I../nptl/sysdeps/unix -I../nptl/sysdeps/i386/i686 -I../nptl/sysdeps/i386 -I../sysdeps/unix/sysv/linux/i386 -I../sysdeps/unix/sysv/linux -I../sysdeps/gnu -I../sysdeps/unix/common -I../sysdeps/unix/mman -I../sysdeps/unix/inet -I../sysdeps/unix/sysv/i386 -I../sysdeps/unix/sysv -I../sysdeps/unix/i386 -I../sysdeps/unix -I../sysdeps/posix -I../sysdeps/i386/i686/fpu -I../sysdeps/i386/i686 -I../sysdeps/i386/i486 -I../nptl/sysdeps/i386/i486 -I../sysdeps/i386/fpu -I../sysdeps/i386 -I../sysdeps/wordsize-32 -I../sysdeps/ieee754/ldbl-96 -I../sysdeps/ieee754/dbl-64 -I../sysdeps/ieee754/flt-32 -I../sysdeps/ieee754 -I../sysdeps/generic/elf -I../sysdeps/generic -nostdinc -isystem /usr/lib/gcc-lib/i686-pc-linux-gnu/3.2.3/include -isystem /lib/modules/2.5.70/build/include -D_LIBC_REENTRANT -D_LIBC_REENTRANT -include ../include/libc-symbols.h -DHAVE_INITFINI -o /var/tmp/portage/glibc-2.3.2-r1/work/glibc-2.3.2/buildhere/csu/libc-start.o
In file included from ../sysdeps/unix/sysv/linux/sys/sysctl.h:29,
from ../include/sys/sysctl.h:2,
from ../sysdeps/unix/sysv/linux/dl-osinfo.h:22,
from ../sysdeps/generic/libc-start.c:33:
/lib/modules/2.5.70/build/include/linux/sysctl.h:39: parse error before '*' token
/lib/modules/2.5.70/build/include/linux/sysctl.h:41: parse error before '*' token
/lib/modules/2.5.70/build/include/linux/sysctl.h:42: parse error before '*' token
/lib/modules/2.5.70/build/include/linux/sysctl.h:43: parse error before '*' token
/lib/modules/2.5.70/build/include/linux/sysctl.h:46: parse error before '}' token
../sysdeps/generic/libc-start.c: In function `__libc_start_main':
../sysdeps/generic/libc-start.c:166: warning: assignment of read-only variable `ptr'
make[2]: *** [/var/tmp/portage/glibc-2.3.2-r1/work/glibc-2.3.2/buildhere/csu/libc-start.o] Error 1
make[2]: Leaving directory `/var/tmp/portage/glibc-2.3.2-r1/work/glibc-2.3.2/csu'
make[1]: *** [csu/subdir_lib] Error 2
make[1]: Leaving directory `/var/tmp/portage/glibc-2.3.2-r1/work/glibc-2.3.2'
make: *** [all] Error 2
|
When I boot the same box with kernel 2.5.69, the error is almost the same, but also strikingly different:
Quote: |
gcc version.c -c -std=gnu99 -O2 -Wall -Winline -Wstrict-prototypes -Wwrite-strings -freorder-blocks -march=i686 -I../include -I. -I/var/tmp/portage/glibc-2.3.2-r1/work/glibc-2.3.2/buildhere/csu -I.. -I../libio -I/var/tmp/portage/glibc-2.3.2-r1/work/glibc-2.3.2/buildhere -I../sysdeps/i386/elf -I../nptl/sysdeps/unix/sysv/linux/i386/i686 -I../nptl/sysdeps/unix/sysv/linux/i386 -I../nptl/sysdeps/unix/sysv/linux -I../nptl/sysdeps/pthread -I../sysdeps/pthread -I../nptl/sysdeps/unix/sysv -I../nptl/sysdeps/unix -I../nptl/sysdeps/i386/i686 -I../nptl/sysdeps/i386 -I../sysdeps/unix/sysv/linux/i386 -I../sysdeps/unix/sysv/linux -I../sysdeps/gnu -I../sysdeps/unix/common -I../sysdeps/unix/mman -I../sysdeps/unix/inet -I../sysdeps/unix/sysv/i386 -I../sysdeps/unix/sysv -I../sysdeps/unix/i386 -I../sysdeps/unix -I../sysdeps/posix -I../sysdeps/i386/i686/fpu -I../sysdeps/i386/i686 -I../sysdeps/i386/i486 -I../nptl/sysdeps/i386/i486 -I../sysdeps/i386/fpu -I../sysdeps/i386 -I../sysdeps/wordsize-32 -I../sysdeps/ieee754/ldbl-96 -I../sysdeps/ieee754/dbl-64 -I../sysdeps/ieee754/flt-32 -I../sysdeps/ieee754 -I../sysdeps/generic/elf -I../sysdeps/generic -nostdinc -isystem /usr/lib/gcc-lib/i686-pc-linux-gnu/3.2.3/include -isystem /usr/src/linux/include -D_LIBC_REENTRANT -D_LIBC_REENTRANT -include ../include/libc-symbols.h -DHAVE_INITFINI -o /var/tmp/portage/glibc-2.3.2-r1/work/glibc-2.3.2/buildhere/csu/version.o
In file included from ../sysdeps/unix/sysv/linux/sys/sysctl.h:29,
from ../include/sys/sysctl.h:2,
from ../sysdeps/unix/sysv/linux/dl-osinfo.h:22,
from ../sysdeps/generic/libc-start.c:33:
/usr/src/linux/include/linux/sysctl.h:39: parse error before '*' token
/usr/src/linux/include/linux/sysctl.h:41: parse error before '*' token
/usr/src/linux/include/linux/sysctl.h:42: parse error before '*' token
/usr/src/linux/include/linux/sysctl.h:43: parse error before '*' token
/usr/src/linux/include/linux/sysctl.h:46: parse error before '}' token
../sysdeps/generic/libc-start.c: In function `__libc_start_main':
../sysdeps/generic/libc-start.c:166: warning: assignment of read-only variable `ptr'
make[2]: *** [/var/tmp/portage/glibc-2.3.2-r1/work/glibc-2.3.2/buildhere/csu/libc-start.o] Error 1
make[2]: *** Waiting for unfinished jobs....
make[2]: Leaving directory `/var/tmp/portage/glibc-2.3.2-r1/work/glibc-2.3.2/csu'
make[1]: *** [csu/subdir_lib] Error 2
make[1]: Leaving directory `/var/tmp/portage/glibc-2.3.2-r1/work/glibc-2.3.2'
make: *** [all] Error 2
|
The statistics:
Code: |
Portage 2.0.48 (default-x86-1.4, gcc-3.2.3, glibc-2.3.2-r1)
=================================================================
System uname: 2.5.70 i686 AMD Athlon(tm) XP 2000+
USE="x86 oss 3dnow apm avi crypt cups encode gif gpm jpeg kde gnome libg++ mikmod mmx mpeg ncurses nls pdflib png quicktime spell truetype xml2 xmms xv zlib gdbm berkdb slang readline arts tetex svga tcltk java guile X sdl tcpd pam libwww ssl perl python esd imlib oggvorbis gtk qt motif opengl cdr acpi alsa dga doc freetype gtk2 libvorbis mozilla nocardbus nptl samba scanner smooth sse usb wmf xml"
COMPILER="gcc3"
CHOST="i686-pc-linux-gnu"
CFLAGS="-march=i686 -O2"
CXXFLAGS=""
ACCEPT_KEYWORDS="x86 ~x86"
MAKEOPTS="-j2"
AUTOCLEAN="yes"
FEATURES="sandbox ccache"
|
Who knows what' s happened? |
known bug..
https://bugs.gentoo.org/show_bug.cgi?id=21326
I must warn you that my system with nptl caused every app to segfault. |
|
Back to top |
|
|
vt n00b
Joined: 07 Oct 2002 Posts: 39 Location: Vilnius, Lithuania
|
Posted: Tue May 27, 2003 8:28 pm Post subject: |
|
|
yes, it is a good kernel.
but if you experience oopses, or modules do no load, add
Code: | EXPORT_SYMBOL(mmu_cr4_features); |
in kernel/ksyms.c
P.S. i use mm-1
Last edited by vt on Tue May 27, 2003 8:32 pm; edited 1 time in total |
|
Back to top |
|
|
AlterEgo Veteran
Joined: 25 Apr 2002 Posts: 1619
|
Posted: Tue May 27, 2003 8:31 pm Post subject: |
|
|
Thanks Lovechild,
I should have checked bugs.gentoo myself
I'm rebuilding glibc from a chrooted environment now -> fixes the problem as well. |
|
Back to top |
|
|
Lovechild Advocate
Joined: 17 May 2002 Posts: 2858 Location: Århus, Denmark
|
Posted: Tue May 27, 2003 8:37 pm Post subject: |
|
|
damn, nvidia-kernel dislikes mm1
Code: |
gcc -c -Wall -Wimplicit -Wreturn-type -Wswitch -Wformat -Wchar-subscripts -Wparentheses -Wcast-qual -Wno-multichar -O -MD -D__KERNEL__ -DMODULE -D_LOOSE_KERNEL_NAMES -DKBUILD_MODNAME="nvidia" -DNTRM -D_GNU_SOURCE -D_LOOSE_KERNEL_NAMES -D__KERNEL__ -DMODULE -DNV_MAJOR_VERSION=1 -DNV_MINOR_VERSION=0 -DNV_PATCHLEVEL=4348 -DNV_UNIX -DNV_LINUX -DNV_INT64_OK -DNVCPU_X86 -DREMAP_PAGE_RANGE_5 -I. -I/usr/src/linux/include -I/usr/src/linux/include/asm/mach-default -Wno-cast-qual os-agp.c
os-agp.c:31: parse error before "agp_memory"
os-agp.c:31: warning: no semicolon at end of struct or union
os-agp.c:38: parse error before '}' token
os-agp.c:38: warning: type defaults to `int' in declaration of `agp_priv_data'
os-agp.c:38: warning: data definition has no type or storage class
os-agp.c:48: parse error before "agpinfo"
os-agp.c:48: warning: type defaults to `int' in declaration of `agpinfo'
os-agp.c:48: warning: data definition has no type or storage class
os-agp.c: In function `KernInitAGP':
os-agp.c:124: warning: passing arg 1 of pointer to function from incompatible pointer type
os-agp.c:132: request for member `aper_base' in something not a structure or union
os-agp.c:133: request for member `aper_size' in something not a structure or union
os-agp.c:155: request for member `aper_base' in something not a structure or union
os-agp.c:165: request for member `aper_size' in something not a structure or union
os-agp.c:188: request for member `mode' in something not a structure or union
os-agp.c:189: request for member `mode' in something not a structure or union
os-agp.c:191: request for member `mode' in something not a structure or union
os-agp.c:192: request for member `mode' in something not a structure or union
os-agp.c:194: request for member `mode' in something not a structure or union
os-agp.c:196: request for member `aper_base' in something not a structure or union
os-agp.c:198: request for member `aper_size' in something not a structure or union
os-agp.c: In function `KernAllocAGPPages':
os-agp.c:263: `agp_memory' undeclared (first use in this function)
os-agp.c:263: (Each undeclared identifier is reported only once
os-agp.c:263: for each function it appears in.)
os-agp.c:263: `ptr' undeclared (first use in this function)
os-agp.c:265: `data' undeclared (first use in this function)
os-agp.c:301: request for member `aper_base' in something not a structure or union
os-agp.c: In function `KernMapAGPPages':
os-agp.c:340: `agp_data' undeclared (first use in this function)
os-agp.c:340: parse error before ')' token
os-agp.c:347: request for member `aper_base' in something not a structure or union
os-agp.c: In function `KernMapAGPNopage':
os-agp.c:390: `agp_data' undeclared (first use in this function)
os-agp.c:390: parse error before ')' token
os-agp.c:391: `agp_memory' undeclared (first use in this function)
os-agp.c:391: `agp_memory_ptr' undeclared (first use in this function)
os-agp.c: In function `KernFreeAGPPages':
os-agp.c:434: `agp_memory' undeclared (first use in this function)
os-agp.c:434: `ptr' undeclared (first use in this function)
os-agp.c:435: `data' undeclared (first use in this function)
make: *** [os-agp.o] Error 1
!!! ERROR: media-video/nvidia-kernel-1.0.4363-r1 failed.
!!! Function src_compile, Line 104, Exitcode 2
!!! (no error message)
|
anyone have a fix for this ? |
|
Back to top |
|
|
vt n00b
Joined: 07 Oct 2002 Posts: 39 Location: Vilnius, Lithuania
|
Posted: Tue May 27, 2003 9:09 pm Post subject: |
|
|
best fix: radeon
secondbest: www.minion.de |
|
Back to top |
|
|
darktux Veteran
Joined: 16 Nov 2002 Posts: 1086 Location: Coimbra, Portugal
|
Posted: Tue May 27, 2003 9:13 pm Post subject: |
|
|
It works fine with me Lovechild.
which version of the nvidia-kernel package are you using? I had one new today, along with 2.5.70-mm1, which probably fixed things. _________________ Lego my ego, and I'll lego your knowledge
www.tuxslare.org - My reborn website |
|
Back to top |
|
|
Lovechild Advocate
Joined: 17 May 2002 Posts: 2858 Location: Århus, Denmark
|
Posted: Tue May 27, 2003 9:22 pm Post subject: |
|
|
darktux wrote: | It works fine with me Lovechild.
which version of the nvidia-kernel package are you using? I had one new today, along with 2.5.70-mm1, which probably fixed things. |
doing a big fat emerge sync to see if one has migrated to the european servers yet - I'm having a few problems with the rsync servers - I get partial rsyncs at times - so I tend to keep my own tree aside the official one, for a few days untill I'm sure I've gotten the entire thing.
-edit-
got the updated version now and it compiles. |
|
Back to top |
|
|
darktux Veteran
Joined: 16 Nov 2002 Posts: 1086 Location: Coimbra, Portugal
|
Posted: Tue May 27, 2003 10:03 pm Post subject: |
|
|
You're welcome _________________ Lego my ego, and I'll lego your knowledge
www.tuxslare.org - My reborn website |
|
Back to top |
|
|
sindre Guru
Joined: 01 Nov 2002 Posts: 315 Location: Norway
|
Posted: Tue May 27, 2003 10:16 pm Post subject: |
|
|
Whoa! I just changed from cfq to as, now the kernel panics are gone. Seeing that as might be better anyway I'll just stick with that. Anyone else care to test if elevator=cfq gives them kernel panics in 2.5.70-mm1? |
|
Back to top |
|
|
Lovechild Advocate
Joined: 17 May 2002 Posts: 2858 Location: Århus, Denmark
|
Posted: Tue May 27, 2003 10:25 pm Post subject: |
|
|
sindre wrote: | Whoa! I just changed from cfq to as, now the kernel panics are gone. Seeing that as might be better anyway I'll just stick with that. Anyone else care to test if elevator=cfq gives them kernel panics in 2.5.70-mm1? |
Jens Axboe is planning a complete rewrite of CFQ, afterall the first patch was simply a proof of concept implementation, back when aa, akpm and he were debating SFQ scheduling on lkml. So I would expect CFQ to go into 2.7, in a majorly reworked version, of course there will be some sort of framework that would allow you to change the scheduler. |
|
Back to top |
|
|
lurid Guru
Joined: 12 Mar 2003 Posts: 595 Location: Florida
|
Posted: Tue May 27, 2003 10:55 pm Post subject: |
|
|
The first time I tried to re-emerge nvidia-kernel after booting 2.5.70-mm1 the kernel died while emerge was running ldconfig (ie. it built the nividia-module, but portage died afterwards) which caused some heafty damage. Nothing booting the Gentoo cd couldn't fix though. Trying it a second time the kernel died while in the middle of compiling it.
This doesn't seem to be an nvidia problem to me, it seems to be a kernel problem. Running ldconfig or compiling shouldn't cause it to die. For the record I was using the CFQ elevator and used the same .config file from 2.5.69-mm3. I'll try switching to AS and see if I can get through the emerge.
EDIT:
Ok, turning off the CFQ elevator seems to do the trick. Everything is working well.. not to sure about anything being "snappier" because everything was pretty fast to begin with (was using 2.5.69-mm3 and got used to it). I see the annoying "sound skips while browsing the web" thing is still around. Are we ever going to get this fixed? I know we're still in dev-mode but Linus is talking like hes ready to release this kernel really soon. This bug can't go unfixed. I'd guess that 90% of computer users spend 90% of their time on the computer bowsing the web and listening to music. |
|
Back to top |
|
|
IntergalacticWalrus Guru
Joined: 07 Jan 2003 Posts: 513 Location: Montreal QC (Canada)
|
Posted: Wed May 28, 2003 1:08 am Post subject: |
|
|
Damn, I just read through this topic and now I feel like trying 2.4.70-mm1 too...
However I'd like to know: you guys talk a lot about some elevator setting, but where exactly do you set it? Where do I put this "elevator=XXX"? |
|
Back to top |
|
|
idl Retired Dev
Joined: 24 Dec 2002 Posts: 1728 Location: Nottingham, UK
|
Posted: Wed May 28, 2003 1:12 am Post subject: |
|
|
IntergalacticWalrus wrote: | Damn, I just read through this topic and now I feel like trying 2.4.70-mm1 too...
However I'd like to know: you guys talk a lot about some elevator setting, but where exactly do you set it? Where do I put this "elevator=XXX"? |
grub.conf _________________ a.k.a port001
Found a bug? Please report it: Gentoo Bugzilla |
|
Back to top |
|
|
dolbz Tux's lil' helper
Joined: 29 Mar 2003 Posts: 138 Location: Nuneaton,UK
|
Posted: Wed May 28, 2003 1:20 am Post subject: |
|
|
so what exactly is this elevator? anyone link me to somewhere or a quick explanation?
Thanks
Dolbz |
|
Back to top |
|
|
cyfred Retired Dev
Joined: 23 Aug 2002 Posts: 596
|
Posted: Wed May 28, 2003 1:30 am Post subject: |
|
|
Well ive been using mm1 for 12hours quite happily except for two little things regarding my mice. (its a laptop machine with a ps2 touch pad and a usb external mouse)
1) On boot up if I have mousedev compiled statically the system tries to load it as a module. This fails, so I compiled mousedev as a module. But now I can only use the ps2 mouse if I have the usb mouse plugged in. mousedev is in modules.autoload.d/kernel-2.5 and does get loaded... It seems there is some "mice" controller that just doesnt activate the ps2 mouse until something is loaded during inserting the usb mouse (wether during boot or when running).
2) Touch pad clicking doesnt work. As in i cant just tap the touch pad and have that recognised as a cllick. Thats no big deal but it is annoying... as im used to it now...
Other than that im quite happy
Well actually there was one other thing. On boot up I recieved
Code: | xargs: environment is too large for exec... |
I just patched the findutils/xargs/xargs.c source file to stop arg_max being limited to 20 * 1024 and now the error is gone. But is that the best way to do this? Or should I be looking to fix something else? |
|
Back to top |
|
|
idl Retired Dev
Joined: 24 Dec 2002 Posts: 1728 Location: Nottingham, UK
|
Posted: Wed May 28, 2003 1:41 am Post subject: |
|
|
dolbz wrote: | so what exactly is this elevator? anyone link me to somewhere or a quick explanation?
Thanks
Dolbz |
Elevators (also knows as schedulers) in a nutshell handle the priority levels of threaded processes and deal with them respectivley.
Couldn't find any sites with clear descriptions for you though, sorry _________________ a.k.a port001
Found a bug? Please report it: Gentoo Bugzilla |
|
Back to top |
|
|
ebrostig Bodhisattva
Joined: 20 Jul 2002 Posts: 3152 Location: Orlando, Fl
|
Posted: Wed May 28, 2003 2:02 am Post subject: |
|
|
sindre wrote: | Whoa! I just changed from cfq to as, now the kernel panics are gone. Seeing that as might be better anyway I'll just stick with that. Anyone else care to test if elevator=cfq gives them kernel panics in 2.5.70-mm1? |
I ran with the CFQ elevator and it locked up my PC solid twice. I had the same problem with 2.5.69-mm3 at least, mm8 was fine.
Currently back on 2.5.69-mm8.
So you are saying that the as elevator works? Might have to try that
Erik _________________ 'Yes, Firefox is indeed greater than women. Can women block pops up for you? No. Can Firefox show you naked women? Yes.' |
|
Back to top |
|
|
idl Retired Dev
Joined: 24 Dec 2002 Posts: 1728 Location: Nottingham, UK
|
Posted: Wed May 28, 2003 2:06 am Post subject: |
|
|
ebrostig wrote: | So you are saying that the as elevator works? Might have to try that |
Works fine here _________________ a.k.a port001
Found a bug? Please report it: Gentoo Bugzilla |
|
Back to top |
|
|
lurid Guru
Joined: 12 Mar 2003 Posts: 595 Location: Florida
|
Posted: Wed May 28, 2003 2:09 am Post subject: |
|
|
Yeah, it locked me up hard twice too. I had the CFQ elevator enabled with 2.5.69-mm3 and I just changed the grub.conf entry to reflect the new kernel, not thinking twice about the elevator setting. I discovered this thread, switched it to AS and its running quite well now. |
|
Back to top |
|
|
jimlynch11 Guru
Joined: 21 Feb 2003 Posts: 590 Location: massachusetts
|
Posted: Wed May 28, 2003 2:18 am Post subject: |
|
|
whats all this "as" talk? |
|
Back to top |
|
|
rvalles Tux's lil' helper
Joined: 19 Feb 2003 Posts: 121
|
Posted: Wed May 28, 2003 2:18 am Post subject: |
|
|
Quote: |
May 27 16:51:27 nyu Linux agpgart interface v0.100 (c) Dave Jones
May 27 16:51:41 nyu login(pam_unix)[3734]: session closed for user tux
May 27 16:51:43 nyu pam_console[3734]: getgrnam failed for cdwriter
May 27 16:52:26 nyu mtrr: no MTRR for c8000000,4000000 found
May 27 16:52:53 nyu nvidia: no version for "struct_module" found: kernel tainted.
May 27 16:52:53 nyu nvidia: no version magic, tainting kernel.
May 27 16:52:53 nyu nvidia: module license 'NVIDIA' taints kernel.
May 27 16:52:53 nyu 0: nvidia: loading NVIDIA Linux x86 nvidia.o Kernel Module 1.0-4363 Sat Apr 19 17:46:46 PDT 2003
May 27 16:53:09 nyu login(pam_unix)[4104]: session opened for user tux by (uid=0)
May 27 16:53:12 nyu login(pam_unix)[4104]: session closed for user tux
May 27 16:53:14 nyu pam_console[4104]: getgrnam failed for cdwriter
May 27 16:53:14 nyu modprobe: FATAL: Module agpgart already in kernel.
May 27 16:53:14 nyu 0: NVRM: AGPGART: unable to retrieve symbol table
|
And X starts without AGP
On the other hand...
Quote: |
root@nyu sh # cdrecord dev=ATAPI: -scanbus
Cdrecord 2.01a14 (i686-pc-linux-gnu) Copyright (C) 1995-2003 Jörg Schilling
scsidev: 'ATAPI:'
devname: 'ATAPI'
scsibus: -1 target: -1 lun: -1
Warning: Using ATA Packet interface.
Warning: The related libscg interface code is in pre alpha.
Warning: There may be fatal problems.
Using libscg version 'schily-0.7'
scsibus0:
0,0,0 0) 'ADAPTEC ' 'ACB-5500 ' 'FAKE' NON CCS Disk
0,1,0 1) *
0,2,0 2) *
0,3,0 3) *
0,4,0 4) *
0,5,0 5) *
0,6,0 6) *
0,7,0 7) *
|
Yes, it detects this instead of my plextor cd writer. CD burning is not working.
And, there's still...
Quote: |
root@nyu sh # sensors
Can't access /proc file
/proc/sys/dev/sensors/chips or /proc/bus/i2c unreadable;
Make sure you have done 'modprobe i2c-proc'!
root@nyu sh # modprobe i2c-proc
FATAL: Module i2c_proc not found.
|
And I cannot see a option for i2c_proc in menuconfig
If you know how to solve any of those three problems, let me know.
By the way, I'm running vanilla 2.5.70. |
|
Back to top |
|
|
ebrostig Bodhisattva
Joined: 20 Jul 2002 Posts: 3152 Location: Orlando, Fl
|
Posted: Wed May 28, 2003 2:33 am Post subject: |
|
|
jimlynch11 wrote: | whats all this "as" talk? |
the 2.5.x series includes 3 different scheduling algorithms, the old default one called Deadline, and 2 new ones, Anticipatory (or as) and CFQ. The default is the Anticipatory scheduler.
They can be switched at boot time with the following option to your kernel line in grub.conf:
elevator=cfq (or as or deadline)
If you want to learn more about these, try to search the Linux kernel mailinglist.
Ok, so I'm back up with 2.5.70-mm1 and the as elevator and so far it seems fine. I even did a few load runs with Oracle 9.2.0.3 and it seems smoother than before. I guess there has been some work on the Anticipatory scheduler too. Going to leave it as my default for now.
Thanks people for testing the as too
Erik _________________ 'Yes, Firefox is indeed greater than women. Can women block pops up for you? No. Can Firefox show you naked women? Yes.' |
|
Back to top |
|
|
sindre Guru
Joined: 01 Nov 2002 Posts: 315 Location: Norway
|
Posted: Wed May 28, 2003 2:49 am Post subject: |
|
|
Something that's been bugging me for a while: What is Huge TLB Page support? It's an option in the 2.5.x Processor type and features section. Enable or disable? Also there's this new /sys directory. What's it's purpose? Seems to me as an ordered mix between /dev and /proc. |
|
Back to top |
|
|
ebrostig Bodhisattva
Joined: 20 Jul 2002 Posts: 3152 Location: Orlando, Fl
|
Posted: Wed May 28, 2003 3:02 am Post subject: |
|
|
sindre wrote: | Something that's been bugging me for a while: What is Huge TLB Page support? It's an option in the 2.5.x Processor type and features section. Enable or disable? Also there's this new /sys directory. What's it's purpose? Seems to me as an ordered mix between /dev and /proc. |
No, huge TLB pages is method of reducing the number of memory pages on systems with huge amounts of memory.
Look at it this way:
Your system has 64GB of RAM, the default page size is 4096 bytes. The number of pages that the kernel has to consider (it has all the pages put on a doubly linked list. one for free and one for used pages) each time it is looking for continues pages to allocate to a process would be enormous on a 64Gb system. Thus the kernel would have to spend more time looping through the linked list than serving other tasks and the system would become very sluggish)
The solution is to make the memory pages larger (1 MB) thus the number of entries on the linked lists becomes fewer and the time it takes to loop through them shorter, hence a performance gain on systems with lareg amounts of memory.
Bottom line, unless you have memory beyond the 4GB range, you will not benefit from Huge TLB pages
I hope I made myself understandable, else please flame me to /dev/null
Erik _________________ 'Yes, Firefox is indeed greater than women. Can women block pops up for you? No. Can Firefox show you naked women? Yes.' |
|
Back to top |
|
|
IntergalacticWalrus Guru
Joined: 07 Jan 2003 Posts: 513 Location: Montreal QC (Canada)
|
Posted: Wed May 28, 2003 4:14 am Post subject: |
|
|
Just tried 2.5.70-mm1. A few problems occured:
- I get a kernel panic when I load my NIC drivers (e100).
- My ALSA drivers won't get loaded by the alsasound boot script. They seem fine if I load them manually with modprobe, though.
Basically problems with modules only (yes, I do have module-init-tools installed). And why did they change so much about modules anyway? The .ko extension, the new tools, god knows what else... ugh. When 2.6 will be out, migrating from 2.4 won't be too smooth cause a lot of things will be broken, that's for sure. |
|
Back to top |
|
|
|