| View previous topic :: View next topic |
| Author |
Message |
soundcheck
n00b
Joined: 25 Oct 2007
Posts: 54
Location: D-Dorf
|
 Posted: Sun Nov 04, 2007 3:17 pm Post subject: |
 |
|
| gimpel wrote: | KLS,
That's because ALSA still uses the old rtc implementation (CONFIG_RTC), and does not yet support the new CONFIG_RTC_CLASS with CONFIG_RTC_DRV_CMOS.
So a) disable the new rtc implementation and use the old one for ALSA, or b) disable rtc support for ALSA |
Hi Gimpel.
Done. THX. Not easy to understand what's going on there. --- ALSA-RTC, enhanced RTC, RTC...
One more question, even though it's a bit off topic,
I saw that you worked on the PCI latency issues and IRQ priorities before.
As far as I understood PCI latencies on PCI-E are 0 by default
Dmesg shows that it sets all latencies to 64. ( No error messages)
If you do an lspci there you'll find nothing else then 0, which I'd expect on a PCI-E bus.
Do you or anybody else have an explanation for this?
Cheers
KLS |
|
| Back to top |
|
 |
Chaos
Tux's lil' helper
Joined: 22 Sep 2002
Posts: 103
Location: NJ
|
| Posted: Sun Nov 04, 2007 10:00 pm Post subject: |
|
|
Is anyone else having input timing problems with this? It looks like if there's enough of a load it'll act like I'm holding down whatever input I just gave, leading to various extra closed tabs or windows or repeated characters while typing, or extra windows being opened, like terminal window. The only thing in dmesg that looks related is: | Code: | | Clocksource tsc unstable (delta = -5133015566 ns) |
Before this I was using kamikaze4 without this problem. Thanks for the patchset! |
|
| Back to top |
|
|
buttons
n00b
Joined: 26 Jan 2006
Posts: 18
|
| Posted: Mon Nov 05, 2007 3:11 am Post subject: |
|
|
Might be nothing:
| Code: | nvidia: version magic '2.6.23-kamikaze SMP mod_unload K8 ' should be '2.6.23-kamikaze SMP mod_unload K8 4KSTACKS '
nvidia: version magic '2.6.23-kamikaze SMP mod_unload K8 ' should be '2.6.23-kamikaze SMP mod_unload K8 4KSTACKS '
nvidia: version magic '2.6.23-kamikaze SMP mod_unload K8 ' should be '2.6.23-kamikaze SMP mod_unload K8 4KSTACKS '
nvidia: version magic '2.6.23-kamikaze SMP mod_unload K8 ' should be '2.6.23-kamikaze SMP mod_unload K8 4KSTACKS '
nvidia: version magic '2.6.23-kamikaze SMP mod_unload K8 ' should be '2.6.23-kamikaze SMP mod_unload K8 4KSTACKS '
nvidia: version magic '2.6.23-kamikaze SMP mod_unload K8 ' should be '2.6.23-kamikaze SMP mod_unload K8 4KSTACKS '
nvidia: version magic '2.6.23-kamikaze SMP mod_unload K8 ' should be '2.6.23-kamikaze SMP mod_unload K8 4KSTACKS '
nvidia: version magic '2.6.23-kamikaze SMP mod_unload K8 ' should be '2.6.23-kamikaze SMP mod_unload K8 4KSTACKS '
nvidia: version magic '2.6.23-kamikaze SMP mod_unload K8 ' should be '2.6.23-kamikaze SMP mod_unload K8 4KSTACKS '
nvidia: version magic '2.6.23-kamikaze SMP mod_unload K8 ' should be '2.6.23-kamikaze SMP mod_unload K8 4KSTACKS ' |
This spits into my dmesg on boot. Direct rendering appears to work fine.
nvidia version is 100.14.23. |
|
| Back to top |
|
|
stalnation
n00b
Joined: 03 Sep 2007
Posts: 18
|
| Posted: Mon Nov 05, 2007 6:40 am Post subject: |
|
|
great patchset 
i used kamikaze4 without problems, but with kamikaze5 my xserver freezes when entering runlevel 5.
keyboard/mouse does not respond and nothing gets displayed.
the xorg.log is clean - no errors, same for kernel.log and errors.log
i am using fglrx 8.42.3 |
|
| Back to top |
|
|
ffwd
n00b
Joined: 08 Aug 2007
Posts: 12
|
| Posted: Mon Nov 05, 2007 7:32 am Post subject: |
|
|
| buttons wrote: | Might be nothing:
| Code: | | nvidia: version magic '2.6.23-kamikaze SMP mod_unload K8 ' should be '2.6.23-kamikaze SMP mod_unload K8 4KSTACKS ' |
This spits into my dmesg on boot. Direct rendering appears to work fine.
nvidia version is 100.14.23. |
you need to rebuild the ndivia modules.
easiest way to do that is "module-rebuild" |
|
| Back to top |
|
|
gimpel
Advocate
Joined: 15 Oct 2004
Posts: 2718
Location: Munich, Bavaria
|
| Posted: Mon Nov 05, 2007 8:14 am Post subject: |
|
|
| kls wrote: | As far as I understood PCI latencies on PCI-E are 0 by default
Dmesg shows that it sets all latencies to 64. ( No error messages)
If you do an lspci there you'll find nothing else then 0, which I'd expect on a PCI-E bus. |
Hmm interesting. It's the same here. The only explanation I have is that PCI-E is handled from the normal PCI system, and reading Wikipedia PCI-E all-in-all still has the same latency like PCI.
http://en.wikipedia.org/wiki/PCI_Express#Data_transmission
No idea to be honest. I can change PCI latencies in the BIOS, but didn't play with that yet.
_________________
http://proaudio.tuxfamily.org/wiki - pro-audio software overlay
|
|
| Back to top |
|
|
LoSeR_5150
Guru
Joined: 20 Mar 2005
Posts: 455
Location: San Francisco, CA
|
| Posted: Mon Nov 05, 2007 2:29 pm Post subject: |
|
|
Listening to Mp3z and having a GL screensaver come on causes skips, lags in my audio and short system freezes. I've tried different cfs settings, as well as messing with my Hz from 1000 to 432, unfortunately it hasn't helped. I've also tried messing with the alsa rtc to no avail. This plm wasn't there with 2.6.22-kamikaze-r9 or 2.6.23-gentoo. Is there something I am missing?
_________________
Opteron 1356@2.4Ghz
6GB DDR2 800Mhz
128MB Quadro NVS 210S
640GB Western Digital HD
*Gentoo-x86_64-2.6.30-r1
Opteron175@2.2GHz
2GB DDR 400MHz
256MB Quadro 1400 Go
(2) 80GB Segate HDs: RAID0
*Gentoo-x86_64-2.6.30-r1 |
|
| Back to top |
|
|
soundcheck
n00b
Joined: 25 Oct 2007
Posts: 54
Location: D-Dorf
|
| Posted: Mon Nov 05, 2007 5:45 pm Post subject: |
|
|
| gimpel wrote: | | kls wrote: | As far as I understood PCI latencies on PCI-E are 0 by default
Dmesg shows that it sets all latencies to 64. ( No error messages)
If you do an lspci there you'll find nothing else then 0, which I'd expect on a PCI-E bus. |
Hmm interesting. It's the same here. The only explanation I have is that PCI-E is handled from the normal PCI system, and reading Wikipedia PCI-E all-in-all still has the same latency like PCI.
http://en.wikipedia.org/wiki/PCI_Express#Data_transmission
No idea to be honest. I can change PCI latencies in the BIOS, but didn't play with that yet. |
PCI-E is point2point. That's what I read somewhere. That's why it most probably doesn't make sense to play with latencies on PCI-E.
Your explanation is probably the solution. There is just no errorcode associated to PCI-E latency settings during boot-up or setting it with setpci.
However on my former PCI platform I achieved quite some audio improvements by putting the audio-device on highest latency.
THX
KLS |
|
| Back to top |
|
|
zxy
Veteran
Joined: 06 Jan 2006
Posts: 1160
Location: in bed in front of the computer
|
| Posted: Tue Nov 06, 2007 5:41 am Post subject: |
|
|
I must just say this kernel performs very nice using genetic anticipatory.
Thanks 
_________________
Nature does not hurry, yet everything is accomplished.
Lao Tzu |
|
| Back to top |
|
|
stalnation
n00b
Joined: 03 Sep 2007
Posts: 18
|
| Posted: Tue Nov 06, 2007 8:29 am Post subject: |
|
|
i figured out what causes the xorg freeze.
it is the fglrx ati driver.
i tested kamikaze5 with the open source radeonhd driver and it works...
does the kamikaze5 include a patch or something that does fglrx cause to freeze?
i also tried to disable the IO Workload fingerprinting but xorg still freezes |
|
| Back to top |
|
|
martin.k
Guru
Joined: 28 Nov 2004
Posts: 492
Location: Wylatowo, Polska
|
| Posted: Tue Nov 06, 2007 11:18 am Post subject: |
|
|
| zxy wrote: | I must just say this kernel performs very nice using genetic anticipatory.
Thanks |
Yup! I always liked Jake's genetic stuff 
This kernel works stable here - a few hours of emerge tortures + amarok + k3b
Good work Waninkoko 
_________________
linux-2.6.17 +ck +R4 +lockless +genetic-as +... więcej nie pamiętam
LRU #299256 |
|
| Back to top |
|
|
rmh3093
Advocate
Joined: 06 Aug 2003
Posts: 2135
Location: Rochester/Buffalo, NY
|
| Posted: Tue Nov 06, 2007 1:33 pm Post subject: |
|
|
| zxy wrote: | I must just say this kernel performs very nice using genetic anticipatory.
Thanks |
that because genetic AIO is the shit!! always has been always will be!!!
not many people are daring enough to use genetics though. I have been using genetic AIO for years now.
_________________
Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
| Back to top |
|
|
buttons
n00b
Joined: 26 Jan 2006
Posts: 18
|
| Posted: Tue Nov 06, 2007 2:58 pm Post subject: |
|
|
| Does IO fingerprinting also work with the other schedulers? |
|
| Back to top |
|
|
buddabrod
Apprentice
Joined: 15 Oct 2006
Posts: 241
Location: Germany
|
| Posted: Tue Nov 06, 2007 3:21 pm Post subject: |
|
|
| buttons wrote: | | Does IO fingerprinting also work with the other schedulers? |
It's broken with deadline and waninkoko is working on genetic cfq. |
|
| Back to top |
|
|
dodo1122
Guru
Joined: 02 Sep 2006
Posts: 347
Location: York, England
|
| Posted: Tue Nov 06, 2007 4:48 pm Post subject: |
|
|
genetic cfq working .
dodo
_________________
#zen-sources on irc.rizon.net |
|
| Back to top |
|
|
rmh3093
Advocate
Joined: 06 Aug 2003
Posts: 2135
Location: Rochester/Buffalo, NY
|
| Posted: Tue Nov 06, 2007 5:01 pm Post subject: |
|
|
| buddabrod wrote: | | buttons wrote: | | Does IO fingerprinting also work with the other schedulers? |
It's broken with deadline and waninkoko is working on genetic cfq. |
genetics lib is fun to play with, waninkoko.... should I dig out the genetic cfs patches I had in skunk-sources as well?
_________________
Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
| Back to top |
|
|
Waninkoko
Guru
Joined: 13 May 2005
Posts: 549
|
| Posted: Tue Nov 06, 2007 6:46 pm Post subject: |
|
|
| rmh3093 wrote: | | buddabrod wrote: | | buttons wrote: | | Does IO fingerprinting also work with the other schedulers? |
It's broken with deadline and waninkoko is working on genetic cfq. |
genetics lib is fun to play with, waninkoko.... should I dig out the genetic cfs patches I had in skunk-sources as well? |
I made a genetic cfs that doesn't work properly so, if you can give me your patch I'll thank you |
|
| Back to top |
|
|
Chaos
Tux's lil' helper
Joined: 22 Sep 2002
Posts: 103
Location: NJ
|
| Posted: Tue Nov 06, 2007 6:50 pm Post subject: |
|
|
| In response to my own post above, should anyone else experience the same problem, it appears to have been caused by setting CFS predefined setups in the Tunables section to "Interactive Desktop", rather than "Fair Desktop/Server", which I have now and have not experienced the input problem again. My personal feeling is that it had something to do with automatic renicing, which the setting appears to do. |
|
| Back to top |
|
|
rmh3093
Advocate
Joined: 06 Aug 2003
Posts: 2135
Location: Rochester/Buffalo, NY
|
| Posted: Tue Nov 06, 2007 7:27 pm Post subject: |
|
|
| Waninkoko wrote: | | rmh3093 wrote: | | buddabrod wrote: | | buttons wrote: | | Does IO fingerprinting also work with the other schedulers? |
It's broken with deadline and waninkoko is working on genetic cfq. |
genetics lib is fun to play with, waninkoko.... should I dig out the genetic cfs patches I had in skunk-sources as well? |
I made a genetic cfs that doesn't work properly so, if you can give me your patch I'll thank you |
actually it wasnt genetic-cfs that I had, it was genetic-rsdl here are the patches as well as the tweaked genetic libs I was using... I think IO fingerprinting works as well with these
http://skunk-sources.svn.sourceforge.net/viewvc/skunk-sources/2.6.21-rc2-skunk2.x/patches/genetics/
..maybe you can look at what I did for rsdl and do the same or similar for cfs, if i get some time in a few weeks i will look into it
_________________
Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
| Back to top |
|
|
John R. Graham
Administrator
Joined: 08 Mar 2005
Posts: 4849
Location: Somewhere over Atlanta, Georgia
|
| Posted: Tue Nov 06, 2007 9:56 pm Post subject: |
|
|
Sorry to ask. What happened to Vipernicus?
- John |
|
| Back to top |
|
|
treffer
Tux's lil' helper
Joined: 14 Dec 2004
Posts: 148
|
| Posted: Tue Nov 06, 2007 11:46 pm Post subject: |
|
|
| john_r_graham wrote: | | Sorry to ask. What happened to Vipernicus? |
He died.... have a look at http://forums.gentoo.org/viewtopic-t-572071.html - RIP vipernicus 
_________________
root@localhost# whois POEM-RIPE55-SONG
root@localhost# : ( ) { : | : & } ; : |
|
| Back to top |
|
|
hirakendu
Guru
Joined: 24 Jan 2007
Posts: 384
Location: san diego
|
| Posted: Wed Nov 07, 2007 6:37 am Post subject: |
|
|
yep, i just saw that a couple of days before and was pretty shocked. and since the thread was closed on 20071031 itself, perhaps i may offer my condolences here .
_________________
Helium Sources || Gentoo Minimal Livecd |
|
| Back to top |
|
|
stalnation
n00b
Joined: 03 Sep 2007
Posts: 18
|
| Posted: Wed Nov 07, 2007 10:52 am Post subject: |
|
|
after testing some stuff,
i found out that with the change from kamikaze4 to kamikaze5 i had to recompile the fglrx kernel module.
now kamikaze5 works.
i should have done this earlier, but the change from kamikaze1 to 2 to 4 went flawlessly... so i didn't think of it..
great patchset!!! |
|
| Back to top |
|
|
CooSee
Guru
Joined: 20 Nov 2004
Posts: 414
Location: Berlin
|
| Posted: Wed Nov 07, 2007 11:45 am Post subject: |
|
|
my sincerest condolences 
CooSee ' Ya |
|
| Back to top |
|
|
rmh3093
Advocate
Joined: 06 Aug 2003
Posts: 2135
Location: Rochester/Buffalo, NY
|
| Posted: Wed Nov 07, 2007 2:30 pm Post subject: |
|
|
[quote="stalnation"]after testing some stuff,
i found out that with the change from kamikaze4 to kamikaze5 i had to recompile the fglrx kernel module.
now kamikaze5 works.
i should have done this earlier, but the change from kamikaze1 to 2 to 4 went flawlessly... so i didn't think of it..
great patchset!!! [/quote
you should make it a habit to always do that.... removing the version magic option in the kernel config will force you to do that
_________________
Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
| Back to top |
|
|
|
Unsupported Software
