| View previous topic :: View next topic |
| Author |
Message |
Phlogiston
Veteran
Joined: 27 Jan 2004
Posts: 1925
Location: Europe, Swizerland
|
 Posted: Wed Aug 16, 2006 1:47 pm Post subject: |
 |
|
| vipernicus wrote: |
I understand and appreciate your efforts. |
But you don't know a solution? =) I will post the kernel error soon.
_________________
Workstation: 5.1 SurroundSound, LIRC remote control; Laptop [IBM-T43]: patched sources, s2disk/ram, fingerprint sensor |
|
| Back to top |
|
 |
vipernicus
Veteran
Joined: 17 Jan 2005
Posts: 1462
Location: Your College IT Dept.
|
|
| Back to top |
|
|
Phlogiston
Veteran
Joined: 27 Jan 2004
Posts: 1925
Location: Europe, Swizerland
|
| Posted: Wed Aug 16, 2006 2:09 pm Post subject: |
|
|
| vipernicus wrote: | | Phlogiston wrote: | | But you don't know a solution? =) I will post the kernel error soon. |
Well, at the moment, I am at work, and unable to work on your issue. Also, like alot of users, I don't use Suspend2.
|
I fixed it  [squashfs]
lol 
_________________
Workstation: 5.1 SurroundSound, LIRC remote control; Laptop [IBM-T43]: patched sources, s2disk/ram, fingerprint sensor |
|
| Back to top |
|
|
wrc1944
Advocate
Joined: 15 Aug 2002
Posts: 2637
Location: Gainesville, Florida
|
| Posted: Wed Aug 16, 2006 2:14 pm Post subject: |
|
|
On an ide ATA 100 box, with this kernel I'm getting hangups on shutdown. something about:
EIP: 00e0:[<c011ade07] Tainted VLI
bunch of lines (about 50 or so), then it hangs at:
<0>Kernel panic - not syncing: Fatal exception
Otherwise, kernel runs fine on this box, but only this shutdown problem at the end of the shutdown routine.
Googling on this suggests possibly :
It bugs out in spinlock.h,
spinlock debugging is enabled.
Another reference mentions possible motherboard rtc clock problems???
or possible shutdown script??
_________________
Main box- ASRock 880GM-LE AM3
Phenom II x6 1090T, 3.2 GHz, 8GB GSkill DDR3 1333mhz
Samsung SATA 500GB, Radeon HD 4670 1GB DDR3
Gentoo ~x86, ~amd64, glibc-2.15-r1, gcc-4.7.0, kernels 3.3.7, 3.4.0 w/auto-cgroups |
|
| Back to top |
|
|
wrc1944
Advocate
Joined: 15 Aug 2002
Posts: 2637
Location: Gainesville, Florida
|
| Posted: Thu Aug 17, 2006 8:01 pm Post subject: |
|
|
Just compiled 2.6.18-rc4-no1, and it does the same thing- hangs at the final stage of shutdown.
2.6.17-no1 does not hang. This is on one of my ide boxes.
However, the same 2.6.18-rc4 kernels that hang on the ide box don't hang on my sata box. Kernel config, Gentoo installation, and hardware is exactly the same on both boxes, except of course for the sata drive support added into my config file.
This has really got me stumped. 
Guess I'll try a plain rc4, and a plain rc4-mm1 in the next day ot two, if I have time.
_________________
Main box- ASRock 880GM-LE AM3
Phenom II x6 1090T, 3.2 GHz, 8GB GSkill DDR3 1333mhz
Samsung SATA 500GB, Radeon HD 4670 1GB DDR3
Gentoo ~x86, ~amd64, glibc-2.15-r1, gcc-4.7.0, kernels 3.3.7, 3.4.0 w/auto-cgroups |
|
| Back to top |
|
|
Phlogiston
Veteran
Joined: 27 Jan 2004
Posts: 1925
Location: Europe, Swizerland
|
| Posted: Fri Aug 18, 2006 8:40 am Post subject: |
|
|
DAMN after hard work and hours I found out that it's actually not your patchset that broke suspend2. It was the tp_smapi patch all the time. (v0.27). The bug occured while doing atomic copy and said: BUG: Scheduling while atomic.
After removing that patch it works =) Although think there are too much patches in this patchset, so I created my own one. Thanks for the broken out directory!
_________________
Workstation: 5.1 SurroundSound, LIRC remote control; Laptop [IBM-T43]: patched sources, s2disk/ram, fingerprint sensor |
|
| Back to top |
|
|
vipernicus
Veteran
Joined: 17 Jan 2005
Posts: 1462
Location: Your College IT Dept.
|
| Posted: Fri Aug 18, 2006 11:38 am Post subject: |
|
|
| Phlogiston wrote: | DAMN after hard work and hours I found out that it's actually not your patchset that broke suspend2. It was the tp_smapi patch all the time. (v0.27). The bug occured while doing atomic copy and said: BUG: Scheduling while atomic.
After removing that patch it works =) Although think there are too much patches in this patchset, so I created my own one. Thanks for the broken out directory! |
I don't see a tp_smapi patch in my patchset! So, would it look better for you, if I had 15 rollup patches? 
_________________
Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog |
|
| Back to top |
|
|
Phlogiston
Veteran
Joined: 27 Jan 2004
Posts: 1925
Location: Europe, Swizerland
|
| Posted: Fri Aug 18, 2006 12:01 pm Post subject: |
|
|
| vipernicus wrote: | | Phlogiston wrote: | DAMN after hard work and hours I found out that it's actually not your patchset that broke suspend2. It was the tp_smapi patch all the time. (v0.27). The bug occured while doing atomic copy and said: BUG: Scheduling while atomic.
After removing that patch it works =) Although think there are too much patches in this patchset, so I created my own one. Thanks for the broken out directory! |
I don't see a tp_smapi patch in my patchset! So, would it look better for you, if I had 15 rollup patches? |
Hehe that's what I said... But I never thought that it could be the tp_smapi patch.
Furthermore if I had used latest version it would have been no problem because with that one it works like a charm now. 
_________________
Workstation: 5.1 SurroundSound, LIRC remote control; Laptop [IBM-T43]: patched sources, s2disk/ram, fingerprint sensor |
|
| Back to top |
|
|
waspers
n00b
Joined: 18 Aug 2006
Posts: 3
|
| Posted: Fri Aug 18, 2006 12:27 pm Post subject: |
|
|
i may have missed this in a thread many days ago which is hard to find but...
i have no /sys/cpusched
how do i enable such an interface?
thanks |
|
| Back to top |
|
|
cheater1034
Veteran
Joined: 09 Sep 2004
Posts: 1551
|
| Posted: Fri Aug 18, 2006 12:38 pm Post subject: |
|
|
| waspers wrote: | i may have missed this in a thread many days ago which is hard to find but...
i have no /sys/cpusched
how do i enable such an interface?
thanks |
If you're running 2.6.18-rc4-viper, and have all the schedulers compiled in like default, then the interface will/should be there.
_________________
JOIN IRC!: #zen-sources on irc.rizon.net
zen-kernel.org
Zen bugs/requests?: http://zen-kernel.org/issues
--
Join: ARCFPWDLVEFOTTSBNA - Anti-Rice club for people who don't like volcanoes except for on the TVs sometimes but not always |
|
| Back to top |
|
|
martin.k
Guru
Joined: 28 Nov 2004
Posts: 492
Location: Wylatowo, Polska
|
| Posted: Fri Aug 18, 2006 12:52 pm Post subject: |
|
|
| wrc1944 wrote: | On an ide ATA 100 box, with this kernel I'm getting hangups on shutdown. something about:
EIP: 00e0:[<c011ade07] Tainted VLI
bunch of lines (about 50 or so), then it hangs at:
<0>Kernel panic - not syncing: Fatal exception
Otherwise, kernel runs fine on this box, but only this shutdown problem at the end of the shutdown routine.
Googling on this suggests possibly :
It bugs out in spinlock.h,
spinlock debugging is enabled.
Another reference mentions possible motherboard rtc clock problems???
or possible shutdown script?? |
I had the very same shutdown errors on my box 
_________________
linux-2.6.17 +ck +R4 +lockless +genetic-as +... więcej nie pamiętam 
LRU #299256 |
|
| Back to top |
|
|
wrc1944
Advocate
Joined: 15 Aug 2002
Posts: 2637
Location: Gainesville, Florida
|
| Posted: Fri Aug 18, 2006 5:35 pm Post subject: |
|
|
UPDATE:
On the same box, I just compiled 2.6.18-rc4 vanilla, and 2.6.18-rc4-mm1. BOTH shutdown normally.
I just shutdown rc4-viper1, and got the same thing and hang I posted about above.
I then booted and shutdown rc4-no1, and it still hangs, but now I get some more types of errors. Maybe this will give a clue for somebody more knowledgable than I am
Lots of lines (20-30) like:
| Code: | Bug: Unable to handle kernel paging request at virtual address 1874b3db
Bug: Unable to handle kernel NULL pointer dereference at virtual address 1874b3db
Oops: 0000 [# (10 digits here)]
4K_STACKS PREEMPT
Code: Bad EIP value. |
Then it hangs with the kernel panic I posted above.
Apparently, this is something unique to both rc4-viper1 and rc4-no1, but only on my ide ATA100 drive box. On my SATA drive box, these kernels don't hang at shutdown.
_________________
Main box- ASRock 880GM-LE AM3
Phenom II x6 1090T, 3.2 GHz, 8GB GSkill DDR3 1333mhz
Samsung SATA 500GB, Radeon HD 4670 1GB DDR3
Gentoo ~x86, ~amd64, glibc-2.15-r1, gcc-4.7.0, kernels 3.3.7, 3.4.0 w/auto-cgroups |
|
| Back to top |
|
|
vipernicus
Veteran
Joined: 17 Jan 2005
Posts: 1462
Location: Your College IT Dept.
|
| Posted: Fri Aug 18, 2006 8:07 pm Post subject: |
|
|
| wrc1944 wrote: | UPDATE:
On the same box, I just compiled 2.6.18-rc4 vanilla, and 2.6.18-rc4-mm1. BOTH shutdown normally.
I just shutdown rc4-viper1, and got the same thing and hang I posted about above.
I then booted and shutdown rc4-no1, and it still hangs, but now I get some more types of errors. Maybe this will give a clue for somebody more knowledgable than I am
Lots of lines (20-30) like:
| Code: | Bug: Unable to handle kernel paging request at virtual address 1874b3db
Bug: Unable to handle kernel NULL pointer dereference at virtual address 1874b3db
Oops: 0000 [# (10 digits here)]
4K_STACKS PREEMPT
Code: Bad EIP value. |
Then it hangs with the kernel panic I posted above.
Apparently, this is something unique to both rc4-viper1 and rc4-no1, but only on my ide ATA100 drive box. On my SATA drive box, these kernels don't hang at shutdown. |
Are you running with 4K stacks?
_________________
Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog |
|
| Back to top |
|
|
wrc1944
Advocate
Joined: 15 Aug 2002
Posts: 2637
Location: Gainesville, Florida
|
| Posted: Fri Aug 18, 2006 8:46 pm Post subject: |
|
|
I haven't seen that option in make xconfig or menuconfig under kernel hacking in a long long time. Just to be sure, I just re-checked all my config files, and it's not there. I guess I just sort of assumed it was no longer relevant, since the option disappeared.
Is there another way to check or change this, like a boot option on the grub.conf kernel line?
BTW, I just noticed that 4K_Stacks PREEMPT Oops today for the first time. I'm almost certain it wasn't there since this started happening.
EDIT: Another thought, after briefly googling around trying to update my knowledge on 4k stacks.
The ide box with the problems only has 512mb ram, while the sata box has 1GB- could that make a difference? Also some web site discussions seem to imply mm kernels now default to 4k, and this causes problems on some systems/apps.
Or, is there something about sata drives and it's serial data transfer method that avoids this problem.
Does anyone running Gentoo and rc4-viper1 on an ide ATA100 or less, with 512mb ram or less have this problem?
_________________
Main box- ASRock 880GM-LE AM3
Phenom II x6 1090T, 3.2 GHz, 8GB GSkill DDR3 1333mhz
Samsung SATA 500GB, Radeon HD 4670 1GB DDR3
Gentoo ~x86, ~amd64, glibc-2.15-r1, gcc-4.7.0, kernels 3.3.7, 3.4.0 w/auto-cgroups |
|
| Back to top |
|
|
zxy
Veteran
Joined: 06 Jan 2006
Posts: 1160
Location: in bed in front of the computer
|
| Posted: Sat Aug 19, 2006 9:29 am Post subject: |
|
|
sky2 drver for marvel network cards is still rboken.
First it works ok, but after some time it starts to work bad.
--- EDIT ---
More explanation:
Under heavy load the network breaks. It starts to work very slow (almost impossible to surf the net).
emerge --sync can't finish. Goes through downloading, but then it maybe outputs 20-30 lines, then stops and after some time retries again.
_________________
Nature does not hurry, yet everything is accomplished.
Lao Tzu
Last edited by zxy on Sun Aug 20, 2006 8:14 pm; edited 1 time in total |
|
| Back to top |
|
|
Dinini
Apprentice
Joined: 11 Jun 2003
Posts: 236
|
| Posted: Sat Aug 19, 2006 3:46 pm Post subject: |
|
|
| Phlogiston wrote: | | As I wrote: I got vanilla 2.6.18-rc4 and applied ___only___ suspend patch to show and prove that patches in your viper patchset break suspend/standby functionality here. Then I applied so more patches to get my own patchset with working suspend and figuring out wich of your included patches breaks the functionality. Are you with me? Can you follow? And then I could not get squashfs to work on vanilla.... |
I had been trying similar myself. Unfortunately I had to start work again so I gave up and dropped back to 2.6.18-rc4 (no patches) which does suspend/hibernate properly for me. Thanks for the effort. |
|
| Back to top |
|
|
Nick C
Guru
Joined: 18 Mar 2005
Posts: 526
Location: Portsmouth, England
|
| Posted: Mon Aug 21, 2006 12:49 am Post subject: |
|
|
| Just gave it a try but still getting problems playing dvds, off to try the new no-sources to see if that works otherwise back to 2.6.17 |
|
| Back to top |
|
|
a7thson
Apprentice
Joined: 08 Apr 2006
Posts: 163
Location: your pineal gland
|
| Posted: Mon Aug 21, 2006 2:51 am Post subject: |
|
|
Just a big FYI here, in defense of 2.6.18-rc4-viper1:
I'm not writing this kernel off despite some unfavorable kernel comparison results (then again I don't really like generic benchmarks per se anyway)... I ran for 3d 18h 56m on viper+lockless with preempt, and it survived a complete "emerge -e system" under gcc-4.1.1/glibc-2.4-r4 (glibc upgrade and I wanted to recompile with hardened support again), running at high cpu load for several hours compiling away without issue. It succeeded in copying several gig of files to XFS, archiving a 1.5GB /home directory with tar (to XFS). mplayer gives seamless DVD playback (Donnie Darko if you're curious ), VCD and DivX playback worked fine while doing emerge --sync... this kernel in general handled everything I threw at it and detects and is compatible with basically everything in my SATA-based system. It slices, it dices, it even cuts. I'm keeping this around for stability and look forward to future tweaks on viper-sources, keep up the good work, and don't be discouraged by the test results, I predict that viper shall rise again 
that said, I'm going to try out emission-sources for sure
_________________
Dell Inspiron 9300 | Pentium-M 1.6GHz 1.5G DDR2 | Gentoo 2.6.21-r3 | gcc-4.1.2/glibc-2.5-r3 | Xorg 7.2
Conroe E6600 | MSI P6N Platinum | 2GB DDR2-800 | XFX 7900GS | x86_64 | Xorg 7.2 XFCE4+beryl+nvidia-drivers |
|
| Back to top |
|
|
iphitus
Apprentice
Joined: 03 Aug 2005
Posts: 226
|
| Posted: Mon Aug 21, 2006 7:19 am Post subject: |
|
|
| a7thson wrote: | Just a big FYI here, in defense of 2.6.18-rc4-viper1:
I'm not writing this kernel off despite some unfavorable kernel comparison results (then again I don't really like generic benchmarks per se anyway)...
|
they weren't very scientific or accurate anyway.
| Quote: | I ran for 3d 18h 56m on viper+lockless with preempt, and it survived a complete "emerge -e system" under gcc-4.1.1/glibc-2.4-r4 (glibc upgrade and I wanted to recompile with hardened support again), running at high cpu load for several hours compiling away without issue. It succeeded in copying several gig of files to XFS, archiving a 1.5GB /home directory with tar (to XFS). mplayer gives seamless DVD playback (Donnie Darko if you're curious ), VCD and DivX playback worked fine while doing emerge --sync... this kernel in general handled everything I threw at it and detects and is compatible with basically everything in my SATA-based system. It slices, it dices, it even cuts. I'm keeping this around for stability and look forward to future tweaks on viper-sources, keep up the good work, and don't be discouraged by the test results, I predict that viper shall rise again
|
any kernel should be able to handle that, if it doesnt, then there's a problem. 3d uptime isnt groundbreaking either -- again, something any kernel should acheive.
James |
|
| Back to top |
|
|
immudium
Guru
Joined: 12 Oct 2004
Posts: 330
Location: Utah
|
| Posted: Mon Aug 21, 2006 8:34 am Post subject: |
|
|
| a7thson wrote: | Just a big FYI here, in defense of 2.6.18-rc4-viper1:
I'm not writing this kernel off despite some unfavorable kernel comparison results (then again I don't really like generic benchmarks per se anyway)... |
Yeah, I think viper got docked several points because he had problems with some of the hardware on his only testing machine which I believe is a laptop. Granted that is a legitimate problem, but the way he's presented the numbers makes it look like Viper sources is the slowest and there is no way that viper sources is the slowest. Maybe it's the least favorable for his particular hardware and setup, but for the things that I use it with, it's very stable and fast and I've run a sufficient number of my own benchmarks and done enough of my own stability tests to be satisfied with viper's performance. |
|
| Back to top |
|
|
a7thson
Apprentice
Joined: 08 Apr 2006
Posts: 163
Location: your pineal gland
|
| Posted: Mon Aug 21, 2006 1:26 pm Post subject: |
|
|
| iphitus wrote: | | a7thson wrote: | Just a big FYI here, in defense of 2.6.18-rc4-viper1:
I'm not writing this kernel off despite some unfavorable kernel comparison results (then again I don't really like generic benchmarks per se anyway)...
|
they weren't very scientific or accurate anyway.
|
(nods) and my accounts are anecdotal ("it feels faster", it works for me blahblahblah), which is even worse and less objective! I just didn't think viper got a fair hearing, and regardless of results will continue to keep viper in rotation for awhile. the brute fact is, it's DAMN hard to benchmark something as nebulous as 'a kernel' when its numbers depend so much on how the system is used. benchmark utilities have their uses, but most modern testing is pretty linear (highly parallel spawning of identical worker threads doesn't really count) compared to "real-world" multitasking even by the average hobby hacker, consisting of heterogenous applications and workloads etc. this is all, of course, IMO.
| Quote: |
any kernel should be able to handle that, if it doesnt, then there's a problem. 3d uptime isnt groundbreaking either -- again, something any kernel should acheive.
|
agreed. then again, it hasn't even been out for more than 10 days in fact, so a "real" test is pretty much impossible. it doesn't match my time with 2.6.16-beyond-git12 however, which due to accidents more than anything and a severe lack of time on my part, ran for over a month solid without issues (and is still a personal favorite due to inclusion of grsecurity among other things). speaking of which... hopefully we see a beyond-dev-sources for 2.6.18-rc4 kernel coming down the "tubes" sometime soon as well
_________________
Dell Inspiron 9300 | Pentium-M 1.6GHz 1.5G DDR2 | Gentoo 2.6.21-r3 | gcc-4.1.2/glibc-2.5-r3 | Xorg 7.2
Conroe E6600 | MSI P6N Platinum | 2GB DDR2-800 | XFX 7900GS | x86_64 | Xorg 7.2 XFCE4+beryl+nvidia-drivers |
|
| Back to top |
|
|
vipernicus
Veteran
Joined: 17 Jan 2005
Posts: 1462
Location: Your College IT Dept.
|
|
| Back to top |
|
|
a7thson
Apprentice
Joined: 08 Apr 2006
Posts: 163
Location: your pineal gland
|
| Posted: Fri Aug 25, 2006 3:56 pm Post subject: |
|
|
| vipernicus wrote: | | Planning on next release, should we keep Plugsched, or drop it for a Tunable Staircase? |
My vote, given that this is (I think) intended as an experimental kernel and testbed for e-mission anyway, is for tunable staircase. Another possibility (though it involves more work) would be to select one or the other with USE flags in the ebuild... USE="lockless +plugsched -staircase" or something along these lines. Then one could "tune" the broad kernel options at emerge and fine-tune the parameters when configuring it, since based on whatever kernel config patch was/was not applied, the appropriate options would be available to tweak.
That's the long response - short version, "tunable staircase"
_________________
Dell Inspiron 9300 | Pentium-M 1.6GHz 1.5G DDR2 | Gentoo 2.6.21-r3 | gcc-4.1.2/glibc-2.5-r3 | Xorg 7.2
Conroe E6600 | MSI P6N Platinum | 2GB DDR2-800 | XFX 7900GS | x86_64 | Xorg 7.2 XFCE4+beryl+nvidia-drivers |
|
| Back to top |
|
|
Tanisete
Guru
Joined: 12 Mar 2004
Posts: 312
|
| Posted: Sun Aug 27, 2006 8:52 am Post subject: |
|
|
Plugshed is running great in rc4-viper1 (at least for me) so my vote is to maintain it.
Last edited by Tanisete on Mon Aug 28, 2006 8:40 pm; edited 1 time in total |
|
| Back to top |
|
|
buthus
Guru
Joined: 17 May 2005
Posts: 429
Location: Bell, Germany
|
| Posted: Mon Aug 28, 2006 4:13 pm Post subject: |
|
|
| Code: | | Portage 2.1.1_rc1-r1 (default-linux/x86/2006.0, gcc-4.1.1/vanilla, glibc-2.4-r3, 2.6.18-rc4-viper i686) |
Works great, thanks vipernicus! Nice work! |
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
Unsupported Software
