View previous topic :: View next topic |
Author |
Message |
Eskarel Apprentice
Joined: 07 May 2004 Posts: 257 Location: Perth Australia
|
Posted: Sat Nov 06, 2004 4:54 am Post subject: |
|
|
According to the release notes this kernel is supposed to have the patch which fixes the nmap issue. I've been testing some various things to see if that same issue is related to the problems I've been having with, well azureus in particularl. They both heavily use the same kernel methods I would assume.
Well I've tested the nmap crash with this kernel and it still dies, at least with a preemptable kernel. 2.6.10-mm2 doesn't(which is what I'm currently testing) so the problem has been fixed, but 2.6.10 has the no binary drivers BS in it.
Could someone who knows more about deep kernel magic check this out for me. The patch doesn't seem to be working or else for some reason it's not installing. |
|
Back to top |
|
|
Clent n00b
Joined: 03 Nov 2004 Posts: 3
|
Posted: Sat Nov 06, 2004 1:59 pm Post subject: |
|
|
sibov wrote: |
My solution was to connect via ssh from an other system and to
reemerge me nvidia-kernel and a restart.
|
I've got an ati card, with the kernel built-in driver (no ati-drivers, otherwise i won't have 3d accel under xorg 6. :/
Quote: |
please make sure your symlink in /usr/src/linux --points-to->> linux-2.6.9-nitro2
|
This one is already ok.
Thx however, i might try to use ati-drivers just to see if xorg launches or not. |
|
Back to top |
|
|
stalcair n00b
Joined: 07 Aug 2003 Posts: 57
|
Posted: Sat Nov 06, 2004 2:16 pm Post subject: Re: to all those with reiser4 lockups |
|
|
eean wrote: |
Yes, something like this, except I don't use ReiserFS 4 (I do have a couple of ReiserFS 3 partitions). And while SSH does not work, the num lock key and LED do work, which is kind of odd. Generally if there's a full on kernel panic that breaks SSH, the num lock light does not work either.
It seems to happen while the system is multitasking under heavy load. |
Well, after switching my Reiser4 partitions to ext3 and trying for a few days with no Reiser4 compiled in or as module, I am noticing improvement. My box appears to actually run faster overall, something I didn't anticipate. OpenOffice (always has been on a ext3 partition) loads considerably faster and I haven't yet had a runaway process forcing a hard reboot. (btw, my advice is to enable the magic key debugging, I intend on using that to more gently reboot my system in the future, that is something that could use a place in the docs IMHO)
You have a very odd situation... however, are you sure that SSH is actually not working or is timing out? With 2 of my boxes (one has never had reiser4) that display this problem, I really can't do anything once those processes hit 99.9. All remote and local login attempts fail, however if I am still in the box remote or local then I can do whatever I want. Granted if it is anything but type echo "this is not cool" then it does not respond. If I run a ps aux (or just about anything else) it freezes attempting I assume to run that process through the gauntlet of 99.9 % CPU utilization procs. Only because I have tested runing a grep -iR at /usr/src/linux but with a very high nice level and I could later run ps if the grep I ran was the offending process.
And any attempt to kill it (<CTRL><D> or <CTRL><Z>) does nothing. If I am in another session and I kill or kill -9 that process it does nothing. When the opportunity has been available I noticed that the process where always R. I have yet to see a S, Z or D. |
|
Back to top |
|
|
Goya13 n00b
Joined: 12 Aug 2004 Posts: 59
|
Posted: Sat Nov 06, 2004 8:44 pm Post subject: |
|
|
First off, love the patchset!
I just have a quick question about the alsa snapshot - is it stable, anything new in there and where can I get more info about it?
I just wanna know if the kernel upgrade I just made to nitro2 is the reason why my sound went kaput.
Thanks a bunch. |
|
Back to top |
|
|
mcoulman n00b
Joined: 20 Sep 2002 Posts: 34 Location: Maui
|
Posted: Sat Nov 06, 2004 9:01 pm Post subject: Azureus et al. BT clients |
|
|
Eskarel wrote: | According to the release notes this kernel is supposed to have the patch which fixes the nmap issue. I've been testing some various things to see if that same issue is related to the problems I've been having with, well azureus in particularl. |
I'll confim this: Azureus works great with -nitro1, Azureus, or any other BT client for that matter, crashes within a few moments of starting a download on -nitro2.
I don't have any of the reported issues with nvidia drivers, xorg start up, the alsa snapshot, vesafb-tng, or splash on -nitro2. All work like in -nitro1.
Anyone got a clue on the download thing? |
|
Back to top |
|
|
hinken Tux's lil' helper
Joined: 16 Jan 2004 Posts: 109
|
Posted: Sat Nov 06, 2004 9:40 pm Post subject: |
|
|
Well, I use sun - jdk 1.5 (5.0) and I have no problem using azureus with this kernel.
1.5 breaks a lot of other things(rhino,log4j) but azureus works dandy! |
|
Back to top |
|
|
mcoulman n00b
Joined: 20 Sep 2002 Posts: 34 Location: Maui
|
Posted: Sat Nov 06, 2004 10:02 pm Post subject: Azureus et al. BT clients |
|
|
hinken wrote: | Well, I use sun - jdk 1.5 (5.0) and I have no problem using azureus with this kernel.
1.5 breaks a lot of other things(rhino,log4j) but azureus works dandy! |
Azureus "works," but it can't download. Other BT clients aren't all java either. Try reading for content.
This seems to be a hw issue. I'm building -nitro2 on another machine, with a different nic, to test.
Azureus, at least, seems fine on the RTL-8139 nic on my notebook with -nitro2. It crashes after a few seconds of downloading on a nVidia Corporation nForce2 Ethernet Controller (rev a1) [forcedeth kernel module] on -nitro2 on the desktop machine (Asus AMD NForce2 MB).
Others who have reported this problem please chime in with your hw to help isolate the bug. :)
Last edited by mcoulman on Sat Nov 06, 2004 11:44 pm; edited 1 time in total |
|
Back to top |
|
|
hermes_jr l33t
Joined: 26 Feb 2004 Posts: 701 Location: Moscow, Russia
|
Posted: Sat Nov 06, 2004 10:22 pm Post subject: |
|
|
Guys, please take a look at my nitro prob . I can't find anything similar to this in forums... |
|
Back to top |
|
|
Apexman Tux's lil' helper
Joined: 07 Apr 2004 Posts: 141
|
|
Back to top |
|
|
Eskarel Apprentice
Joined: 07 May 2004 Posts: 257 Location: Perth Australia
|
Posted: Sun Nov 07, 2004 4:48 am Post subject: |
|
|
You can temporary fix Azureus by cranking max connects down to 100 and max connect per torrent down to 30 and only running a few torrents at a time, it's reasonably stable then.
Still crashed in mm-sources though, so whatever it is is either my network card or some still as yet unpatched kernel problem. I try to get sysrq outputs, but the only the print doesn't seem to work, I can unmount and sync(I think) and boot(I know) but that seems to be about it. |
|
Back to top |
|
|
TecHunter Tux's lil' helper
Joined: 15 Feb 2003 Posts: 124
|
Posted: Sun Nov 07, 2004 10:34 am Post subject: |
|
|
hey, I still can't use vesafb-tng
the error msg of dmesg about vesafb are:
Code: |
Nov 7 18:20:54 wayne vesafb: NVidia Corporation, NV11 Board, Chip Rev B2 (OEM: NVidia)
Nov 7 18:20:54 wayne vesafb: VBE version: 3.0
Nov 7 18:20:54 wayne vesafb: protected mode interface info at c000:c650
Nov 7 18:20:54 wayne vesafb: pmi: set display start = c00cc695, set palette = c00cc71a
Nov 7 18:20:54 wayne vesafb: pmi: ports = b4c3 b503 ba03 c003 c103 c403 c503 c603 c703 c803 c903 cc03 ce03 cf03 d003 d103 d203 d303 d403 d503 da03 ff03
Nov 7 18:20:54 wayne vesafb: hardware supports DCC2 transfers
Nov 7 18:20:54 wayne vesafb: monitor limits: vf = 160 Hz, hf = 92 kHz, clk = 190 MHz
Nov 7 18:20:54 wayne vesafb: scrolling: redraw
Nov 7 18:20:54 wayne vesafb: cannot reserve video memory at 0x0
Nov 7 18:20:54 wayne vesafb: abort, cannot ioremap video memory 0x0 @ 0x0
Nov 7 18:20:54 wayne vesafb: probe of vesafb0 failed with error -5
|
_________________ Gentoo is GREAT!!! |
|
Back to top |
|
|
jdkbx n00b
Joined: 03 May 2004 Posts: 40
|
Posted: Sun Nov 07, 2004 8:55 pm Post subject: |
|
|
nice dude, no xmms skipping anymore, everything seems stable so far. |
|
Back to top |
|
|
l0tus_ n00b
Joined: 03 Nov 2004 Posts: 22
|
Posted: Sun Nov 07, 2004 10:29 pm Post subject: |
|
|
not sure what to make of this error
Code: | CHK include/linux/version.h
make[1]: `arch/i386/kernel/asm-offsets.s' is up to date.
CHK include/asm-i386/asm_offsets.h
CHK include/linux/compile.h
dnsdomainname: Unknown host
CC arch/i386/kernel/process.o
include/linux/dcache.h: In function `dget':
include/linux/dcache.h:276: warning: asm operand 1 probably doesn't match constraints
include/linux/dcache.h: In function `dget_parent':
include/linux/dcache.h:276: warning: asm operand 1 probably doesn't match constraints
include/linux/fs.h: In function `simple_transaction_set':
include/linux/fs.h:1704: warning: asm operand 1 probably doesn't match constraints
include/asm/fixmap.h: In function `virt_to_fix':
include/asm/fixmap.h:151: warning: asm operand 1 probably doesn't match constraints
include/linux/mm.h: In function `put_page':
include/linux/mm.h:318: warning: asm operand 1 probably doesn't match constraints
arch/i386/kernel/process.c: In function `release_thread':
arch/i386/kernel/process.c:343: warning: asm operand 1 probably doesn't match constraints
arch/i386/kernel/process.c:343: error: impossible constraint in `asm'
make[1]: *** [arch/i386/kernel/process.o] Error 1
make: *** [arch/i386/kernel] Error 2 |
tryring to make nitro2, can anyone help me out with what these errors might be reffering to? |
|
Back to top |
|
|
l0tus_ n00b
Joined: 03 Nov 2004 Posts: 22
|
Posted: Sun Nov 07, 2004 11:30 pm Post subject: |
|
|
fixed. turns out it was the -fforce-addr CFLAG i was using that was causing the problem. i use a pentium3 |
|
Back to top |
|
|
John5788 Advocate
Joined: 06 Apr 2004 Posts: 2140 Location: 127.0.0.1
|
Posted: Mon Nov 08, 2004 1:25 am Post subject: |
|
|
im curious, why do 4kb stacks have to be disabled in order to use reiser4? and whats the difference between 4kb and 8kb stacks? _________________ John5788 |
|
Back to top |
|
|
devsk Advocate
Joined: 24 Oct 2003 Posts: 3003 Location: Bay Area, CA
|
|
Back to top |
|
|
grzewho l33t
Joined: 31 Dec 2002 Posts: 626 Location: /home/g
|
Posted: Mon Nov 08, 2004 1:27 pm Post subject: |
|
|
the compilation went fine for me, the kernel boots perfectly, xorg starts, but dies when running fullscreen glx-based xscreensaver modules. the windowed glx apps work perfectly. i`m using latest nvidia, which built fine
any hints ? _________________
Code: | USE="freedom -software_patents" emerge --deep --update world |
|
|
Back to top |
|
|
Crazor Tux's lil' helper
Joined: 23 Apr 2003 Posts: 131
|
Posted: Mon Nov 08, 2004 3:11 pm Post subject: |
|
|
http://www.gc-linux.org/down/isobel/kexec/2.6.9/
there you get a kexec patch for 2.6.9 which applies to nitro sources. only thing that failed was some insert prior to swsusp2 code because the patch looked for swsusp1, but with common sense and no coding experience in C i applied it by hand.
should be integrated in the next nitro, kexec is really nice (start new kernel, i.e. reboot without going through bios post). search bugzilla for init script etc. |
|
Back to top |
|
|
randomguy n00b
Joined: 17 Oct 2004 Posts: 14
|
Posted: Mon Nov 08, 2004 6:13 pm Post subject: |
|
|
Can anyone help me with my problem?
When I try to boot it gets stuck / stops at "Calculating module dependences", no matter how long I wait, it won't go past... |
|
Back to top |
|
|
fimbulvetr n00b
Joined: 03 Feb 2004 Posts: 51 Location: South Dakota
|
Posted: Mon Nov 08, 2004 7:04 pm Post subject: |
|
|
Hey guys, been using nitro for quite some time but this is the first time I've tried the cisco vpn client. Anyone sucessfully using this?
Here's what I get on modprobe:
Code: |
divert: not allocating divert_blk for non-ethernet device cipsec0
Cisco Systems VPN Client Version 4.6.00 (0045) kernel module loaded |
Here is what I get when I do "vpnclient connect foo":
Code: | Badness in local_bh_enable at kernel/softirq.c:141
[<b0121f29>] local_bh_enable+0x89/0x90
[<d0bcff0d>] handle_vpnup+0x9d/0x1e0 [cisco_ipsec]
[<d0bcfcac>] interceptor_ioctl+0x15c/0x160 [cisco_ipsec]
[<b0542e04>] dev_ifsioc+0x374/0x3e0
[<b0543035>] dev_ioctl+0x1c5/0x320
[<b0538497>] sock_ioctl+0x277/0x290
[<b017fa0a>] sys_ioctl+0xca/0x230
[<b0114e80>] do_page_fault+0x0/0x5bf
[<b01052a1>] syscall_call+0x7/0xb
Badness in local_bh_enable at kernel/softirq.c:141
[<b0121f29>] local_bh_enable+0x89/0x90
[<b05407df>] dev_remove_pack+0xf/0x20
[<d0bcff64>] handle_vpnup+0xf4/0x1e0 [cisco_ipsec]
[<d0bcfcac>] interceptor_ioctl+0x15c/0x160 [cisco_ipsec]
[<b0542e04>] dev_ifsioc+0x374/0x3e0
[<b0543035>] dev_ioctl+0x1c5/0x320
[<b0538497>] sock_ioctl+0x277/0x290
[<b017fa0a>] sys_ioctl+0xca/0x230
[<b0114e80>] do_page_fault+0x0/0x5bf
[<b01052a1>] syscall_call+0x7/0xb
bad: scheduling while atomic!
[<b05c67dc>] schedule+0x42c/0x440
[<b05c68d8>] wait_for_completion+0x78/0xd0
[<b01199c0>] default_wake_function+0x0/0x20
[<b012e95e>] __kernel_text_address+0x2e/0x40
[<b01199c0>] default_wake_function+0x0/0x20
[<b01057b8>] show_trace+0x58/0xa0
[<b012e452>] synchronize_kernel+0x32/0x3a
[<b01058bc>] dump_stack+0x1c/0x20
[<b012e410>] wakeme_after_rcu+0x0/0x10
[<b05407df>] dev_remove_pack+0xf/0x20
[<d0bcff64>] handle_vpnup+0xf4/0x1e0 [cisco_ipsec]
[<d0bcfcac>] interceptor_ioctl+0x15c/0x160 [cisco_ipsec]
[<b0542e04>] dev_ifsioc+0x374/0x3e0
[<b0543035>] dev_ioctl+0x1c5/0x320
[<b0538497>] sock_ioctl+0x277/0x290
[<b017fa0a>] sys_ioctl+0xca/0x230
[<b0114e80>] do_page_fault+0x0/0x5bf
[<b01052a1>] syscall_call+0x7/0xb
|
Any suggestions?
Thanks
fim _________________ :wq |
|
Back to top |
|
|
fimbulvetr n00b
Joined: 03 Feb 2004 Posts: 51 Location: South Dakota
|
Posted: Mon Nov 08, 2004 7:13 pm Post subject: |
|
|
randomguy wrote: | Can anyone help me with my problem?
When I try to boot it gets stuck / stops at "Calculating module dependences", no matter how long I wait, it won't go past... |
Rebuild your kernel, do make, make modules_install and make install.
Might be a corrupt/missing module or something. _________________ :wq |
|
Back to top |
|
|
grzewho l33t
Joined: 31 Dec 2002 Posts: 626 Location: /home/g
|
Posted: Mon Nov 08, 2004 8:01 pm Post subject: |
|
|
or run modules-update manually to see it`s effects _________________
Code: | USE="freedom -software_patents" emerge --deep --update world |
|
|
Back to top |
|
|
mojo n00b
Joined: 21 Nov 2002 Posts: 47 Location: Heidelberg, Germany
|
Posted: Mon Nov 08, 2004 8:09 pm Post subject: |
|
|
@fimbulvetr
I am using the latest nitro-sources and have no problems running the cisco vpn client 4.0.4 B. I am sorry but I can't help you. I never had this problem with cisco... _________________ Hat der alte Hexenmeister sich doch einmal wegbegeben
Und nun sollen seine Geister auch nach meinem willen leben
Seine Wort und Werke merkt ich und den Brauch
Und mit Geistestaerke tu ich wunder auch
-Goethe |
|
Back to top |
|
|
Pepek Guru
Joined: 25 Apr 2004 Posts: 313 Location: Tarnowskie Góry - Poland - Europe
|
Posted: Tue Nov 09, 2004 1:59 am Post subject: |
|
|
Hi all !
As I read in this topic is problem with win4lin and software-suspend 2 with nitro2. I read that win4lin + nitro2 works fine, swsusp2 + nitro2 works fine too, but win4lin + swsusp2 + nitro2 doesn't compile. So I've made some patch for that problems. This patch include software_suspend-2.1.2 and win4lin. It's now work fine for me with fbsplash and vesafb-tng, but I don't use win4lin and swsusp, so I don't test that and this post is request for tests. You can download patch here : http://www.pepek.neostrada.pl/software_suspend-2.1.2-and-win4lin3-for-2.6.9-nitro2.tar.bz2 and this is archive with software_suspend directory and patch is in that directory. For patch your 2.6.9-nitro2 kernel try sth like this : Code: | cd /usr/src/linux
patch -p1 <path_to_patch |
Cheers. _________________ For Polish users :
$ man mas
Nie ma strony manuala dla mas
For Other users :
Sorry for my english, I know it's bad, but I think it's at least understandable |
|
Back to top |
|
|
Chaotix n00b
Joined: 06 Aug 2004 Posts: 13
|
Posted: Tue Nov 09, 2004 2:35 am Post subject: |
|
|
Mmmmm.... Nitro-2 _________________ Don't you love the smell of warm hard drives in the morning? |
|
Back to top |
|
|
|