View previous topic :: View next topic |
Author |
Message |
mirko_3 l33t
Joined: 02 Nov 2003 Posts: 605 Location: Birreria
|
Posted: Mon Mar 07, 2005 5:12 pm Post subject: |
|
|
I don't know if this has anything to do with the patchset, but I sure know that this never happened before:
Code: |
[ 7204.711856] oom-killer: gfp_mask=0x80d2
[ 7204.711860] DMA per-cpu:
[ 7204.711863] cpu 0 hot: low 2, high 6, batch 1
[ 7204.711865] cpu 0 cold: low 0, high 2, batch 1
[ 7204.711866] Normal per-cpu:
[ 7204.711868] cpu 0 hot: low 32, high 96, batch 16
[ 7204.711870] cpu 0 cold: low 0, high 32, batch 16
[ 7204.711871] HighMem per-cpu: empty
[ 7204.711874]
[ 7204.711875] Free pages: 6108kB (0kB HighMem)
[ 7204.711878] Active:2094 inactive:111583 dirty:0 writeback:3390 unstable:0 free:1527 slab:6219 mapped:106729 pagetables:703
[ 7204.711882] DMA free:2068kB min:88kB low:108kB high:132kB active:8372kB inactive:0kB present:16384kB pages_scanned:10409 all_unreclaimable? yes
[ 7204.711885] lowmem_reserve[]: 0 495 495
[ 7204.711889] Normal free:4040kB min:2804kB low:3504kB high:4204kB active:4kB inactive:446332kB present:507888kB pages_scanned:396 all_unreclaimable? no
[ 7204.711892] lowmem_reserve[]: 0 0 0
[ 7204.711895] HighMem free:0kB min:128kB low:160kB high:192kB active:0kB inactive:0kB present:0kB pages_scanned:0 all_unreclaimable? no
[ 7204.711897] lowmem_reserve[]: 0 0 0
[ 7204.711899] DMA: 1*4kB 0*8kB 1*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 1*2048kB 0*4096kB = 2068kB
[ 7204.711906] Normal: 214*4kB 38*8kB 6*16kB 1*32kB 1*64kB 1*128kB 0*256kB 1*512kB 0*1024kB 1*2048kB 0*4096kB = 4040kB
[ 7204.711913] HighMem: empty
[ 7204.711915] Swap cache: add 4649, delete 1255, find 160/175, race 0+0
[ 7204.711917] Free swap = 1034148kB
[ 7204.711919] Total swap = 1052216kB
[ 7204.711921] Out of Memory: Killed process 32663 (wine-preloader).
|
(I couln't understand where the relevant part started...)
Anyway, I was playing starcraft, and suddenly wine got killed. Anyone can explain this "out of memory"?
Thanks... _________________ Non fa male! Non fa male! |
|
Back to top |
|
|
fallow Bodhisattva
Joined: 08 Jan 2004 Posts: 2208 Location: Poland
|
Posted: Mon Mar 07, 2005 5:28 pm Post subject: |
|
|
Tiger683 wrote: |
Code: |
#include <asm-i386/march-default/mach_pci_blacklist.h>
|
be rather
Code: |
#include <asm-i386/mach-default/mach_pci_blacklist.h>
|
cheers,
T |
Yeah - thanks , You have right , the "r" letter removed . I must go to the some rest finally or more sleep
cheers. _________________ "Time is a companion that goes with us on a journey. It reminds us to cherish each moment, because it will never come again. What we leave behind is not as important as how we have lived" J-L. Picard |
|
Back to top |
|
|
mhennings n00b
Joined: 22 Feb 2005 Posts: 15 Location: Germany
|
Posted: Mon Mar 07, 2005 5:47 pm Post subject: |
|
|
fallow wrote: |
mhennings wrote: | Did anyone else notice skipping sound with this kernel?
I am currently using genetic nicksched with cfq.
I've been using ck-sources and/or nitro for a while now, but remeber that this did not happen with earlier love-sources for me. | nope. Try to renice You audio player for -10 for ex.
|
I dislike renicing. I think responsiveness should not lead to a less fair scheduling. But maybe its not a problem with the scheduling -- when switching to the oss wrapper the skipping went away.
fallow wrote: |
mhennings wrote: | BTW: Is it possible to look into the results of the genetic |
yeap Code: |
Enterprise ~ # cd /proc/genetic
Enterprise genetic # ls
as-ioscheduler nick-sched
|
|
Thanks Nice to watch.
I'd like to play with the parameters like generation size and mutation rate, since it seems not to adapt very fast or consequently. I would expect a higher generation size with a longer life time to perform better.
Where do i modify it? |
|
Back to top |
|
|
fallow Bodhisattva
Joined: 08 Jan 2004 Posts: 2208 Location: Poland
|
Posted: Mon Mar 07, 2005 6:18 pm Post subject: |
|
|
mhennings wrote: |
I'd like to play with the parameters like generation size and mutation rate, since it seems not to adapt very fast or consequently. I would expect a higher generation size with a longer life time to perform better.
Where do i modify it? |
genetic library is in /lib/genetic.c
implementation /include/linux/sched.h & /kernel/sched.c
intarface to proc /kernel/sysctl.c
lifetime of the child :
Code: |
#define NICK_SCHED_CHILD_LIFESPAN (4 * HZ)
|
You can change mutation rate with *mutation_rate functions from genetic lib 0.4, not yet implemented - I`m going to do it with next version.
generation size If You mean number of children for each - via Code: | #define NICK_SCHED_NUM_CHILDREN 8 |
If You are going to experiment with it , please come back to this thread with the results
cheers. _________________ "Time is a companion that goes with us on a journey. It reminds us to cherish each moment, because it will never come again. What we leave behind is not as important as how we have lived" J-L. Picard |
|
Back to top |
|
|
jonnevers Veteran
Joined: 02 Jan 2003 Posts: 1594 Location: Gentoo64 land
|
Posted: Mon Mar 07, 2005 6:23 pm Post subject: |
|
|
Zinic wrote: | Sad attempt at a love-sources wall paper (1280x1024) -> http://jpserver.net/love_wallpaper_1280x1024.jpg
Just like to say how much I enjoy this patchest. Love the love. Thank you so much for all of your hard work and spent hours. Keep it up!
-Zin |
this looks like a powerpoint presentation slide... |
|
Back to top |
|
|
Hirion n00b
Joined: 05 Mar 2005 Posts: 4
|
|
Back to top |
|
|
feld Guru
Joined: 29 Aug 2004 Posts: 593 Location: USA
|
Posted: Mon Mar 07, 2005 6:36 pm Post subject: |
|
|
jonnevers wrote: | Zinic wrote: | Sad attempt at a love-sources wall paper (1280x1024) -> http://jpserver.net/love_wallpaper_1280x1024.jpg
Just like to say how much I enjoy this patchest. Love the love. Thank you so much for all of your hard work and spent hours. Keep it up!
-Zin |
this looks like a powerpoint presentation slide... |
rofl.... be nice!
-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 |
|
|
fallow Bodhisattva
Joined: 08 Jan 2004 Posts: 2208 Location: Poland
|
Posted: Mon Mar 07, 2005 6:42 pm Post subject: |
|
|
gentoo_till_die wrote: | Hi fallow,
You should keep the patches from 2.6.11-rc4-love1 about hostap thingy :
And madwifi...too..that would be nice ..thanks , I love 2.6.11-rc4-love1, if you can keep that patches for wireless needs on the 2.6.11 the stable one, that'll be fantastic.
danke |
Ok , yes this relase haven`t net-dev patch , but orinnoco and madwifi-cvs.
I added hostap driver if You like it to this update. ( few lines down )
Robin79 wrote: | does sk98lin work on this one? |
I don`t know - haven`t - I added it now by some request - compiled ok - but I not tested it.
ok first unofficial update :
-> I tested it on my machine , so don`t know about ohters <-
Code: |
--2.6.11-love1-unofficial-upd----
---requests-------------------------
usbhid-readd-kconfig.patch //Some people want to choose poll from menu//
hostap-linux-2.6.2.patch //by gentoo_till_die request//
sk98lin_v7.09_2.6.11_patch thx Michael L. ;)
--add-ons--[mm]--------------------
mach-blacklist (x86_64) try_fix#1
bk-usb.patch
copy_pte_range-latency-fix.patch
irqpoll.patch
properly-share-process-and-session-keyrings-with-clone_thread.patch
put-newly-registered-shrinkers-at-the-tail-of-the-list.patch
run-softirqs-on-proper-processor-on-offline.patch
serialize-access-to-ide-devices.patch
setup_per_zone_lowmem_reserve-oops-fix.patch
simpler-topdown-mmap-layout-allocator.patch
smpbootc-cleanups.patch
swapspace-layout-improvements.patch
sys_setpriority-euid-semantics-fix.patch
tty-output-lossage-fix.patch
vm-pageout-throttling.patch
|
patch : http://vivid.dat.pl/fallow/pub/love-sources/2.6.11-love1/upd1/2.6.11-love1-upd1.bz2
this info : http://vivid.dat.pl/fallow/pub/love-sources/2.6.11-love1/upd1/2.6.11-love1-upd1.txt
new ebuild ( the same name ) for use of upd1 patch : http://vivid.dat.pl/fallow/pub/love-sources/2.6.11-love1/upd1/love-sources-2.6.11-r1.ebuild
going to some rest
cheers
[EDIT]
ohhh You overtaked me hehe
[/EDIT] _________________ "Time is a companion that goes with us on a journey. It reminds us to cherish each moment, because it will never come again. What we leave behind is not as important as how we have lived" J-L. Picard |
|
Back to top |
|
|
VoVaN l33t
Joined: 02 Jul 2003 Posts: 701 Location: The Netherlands
|
Posted: Mon Mar 07, 2005 6:57 pm Post subject: reiser4 |
|
|
Could somebody point me to the only reiser4 patch from the latest love sources? Is that possible to get it separately? The one from love sources seems to be only the one which could be compiled _cleanly_ with vanilla 2.6.11. Thanx |
|
Back to top |
|
|
feld Guru
Joined: 29 Aug 2004 Posts: 593 Location: USA
|
Posted: Mon Mar 07, 2005 6:59 pm Post subject: |
|
|
bk-usb supposed to fix any usb problems? perhaps it will make my usbmouse able to have its polling changed again like it could in 2.6.10?
-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 |
|
|
krejler Tux's lil' helper
Joined: 10 Nov 2003 Posts: 142 Location: Denmark
|
Posted: Mon Mar 07, 2005 7:20 pm Post subject: |
|
|
feld wrote: | bk-usb supposed to fix any usb problems? perhaps it will make my usbmouse able to have its polling changed again like it could in 2.6.10?
-Feld |
Could you please PM me the exact problems you are facing?
Thanks. |
|
Back to top |
|
|
Harry_Hakbijl n00b
Joined: 07 Mar 2005 Posts: 55 Location: Leuven, Vlaanderen
|
Posted: Mon Mar 07, 2005 7:45 pm Post subject: |
|
|
fallow wrote: |
Yeah - thanks , You have right , the "r" letter removed . I must go to the some rest finally or more sleep
cheers. |
Thank you, I'll try it tomorrow!!
Thank's for the support! |
|
Back to top |
|
|
sobers_2002 Veteran
Joined: 16 Mar 2004 Posts: 1128
|
Posted: Mon Mar 07, 2005 7:51 pm Post subject: |
|
|
Master One wrote: | rmh3093 wrote: | has anyone else noticed their idle temps are lower with nick sched vs. staircase and with genetic over deadline |
So you use genetic nicksched + genetic as on Reiser4?
Did you try dealine as i/o sched instead as well?
From what I've read so far, genetic nicksched + genetic as should be quite unstable, resulting in random hardlocks on heavy i/o use.
That's why I decided to go for genetic nicksched + deadline. |
how do u use 2 I/O schedulers at the same time??? _________________ Pdict - dockable dictionary client for linux
FREE97WIN: Use this code on Dreamhost and you get $97 off !! |
|
Back to top |
|
|
feld Guru
Joined: 29 Aug 2004 Posts: 593 Location: USA
|
Posted: Mon Mar 07, 2005 8:16 pm Post subject: |
|
|
nicksched is a CPU scheduler. it is genetic enabled if wanted in this kernel.
deadline, as / genetic as, and cfq are the i/o schedulers.
-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 |
|
|
Arainach l33t
Joined: 08 Jul 2004 Posts: 609
|
Posted: Mon Mar 07, 2005 9:05 pm Post subject: |
|
|
Well, I'm a convert. Love sources run every bit as fast as Nitro sources, possibly a bit more stable, and my ATI drivers compile on them without modification. I'm hooked. Great release! _________________ Gentoo: Stage3 w/ NPTL & udev, gcc 3.4.4 full rebuild
Kernel: 2.6.15-gentoo-r1 w/ 1G-Lowmem Patch
System: Athlon XP 2.2Ghz/1GB Corsair Value/160GB, 250GB WD IDE/128MB GeForce 6800/Sony 17" Trinitron G200 @ 1280x1024x75Hz |
|
Back to top |
|
|
kaktyc Tux's lil' helper
Joined: 14 Aug 2004 Posts: 100
|
Posted: Mon Mar 07, 2005 11:14 pm Post subject: |
|
|
Very nice! love-kernel rocks!
I found that it's faster than gentoo-dev and more stable than nitro Congratulations!
Reiser4 works fine only with deadline. I didn't tried CFQ, but with genetic as I have lockups.
Nice work! |
|
Back to top |
|
|
gimpel Advocate
Joined: 15 Oct 2004 Posts: 2720 Location: Munich, Bavaria
|
|
Back to top |
|
|
kontol n00b
Joined: 19 Dec 2004 Posts: 31
|
Posted: Tue Mar 08, 2005 1:06 am Post subject: |
|
|
Mmm..I've got error trying to enable hostap on kernel :
Code: |
ld: drivers/net/wireless/hostap/hostap.o: No such file: No such file or directory
make[3]: ***[drivers/net/wireless/built-in-o] Error1
make[2]: ***[drivers/net/wireless] Error 2
make[1]: ***[drivers/net] Error 2
make: *** [drivers] Error 2
|
|
|
Back to top |
|
|
dhaemon n00b
Joined: 06 Feb 2005 Posts: 18
|
Posted: Tue Mar 08, 2005 3:56 am Post subject: |
|
|
Love Rocks!!! Finally real speed improvements... I can finally run all my server apps(apache etc) while playing demanding games, listening to music, and watching a movie... YES ^_^
But somthing is bothering me. While everything seems to work fine, my dmesg shows a bunch of:
Code: |
scheduling while atomic: swapper/0x00000001/0
[ 1964.216628]
[ 1964.216629] Call Trace:<ffffffff803f05b4>{schedule+116} <ffffffff8010c93a>{cpu_idle+74}
[ 1964.216638] <ffffffff8058a8f0>{start_kernel+624} <ffffffff8058a26a>{x86_64_start_kernel+362}
|
Code: |
Linux laika.g.la 2.6.11-love1 #2 Mon Mar 7 22:05:57 Local time zone must be set--see zic manua x86_64 AMD Athlon(tm) 64 Processor 3500+ AuthenticAMD GNU/Linux
|
Any comments on this? |
|
Back to top |
|
|
DrWoland l33t
Joined: 13 Nov 2004 Posts: 603
|
Posted: Tue Mar 08, 2005 5:17 am Post subject: |
|
|
Just installed Love as part of my re-install, so far so good. Booted up successfully with all the Genetic stuff enabled and on RR4, I'm very pleased so far. Nvidia seems to be working fine as well, at least as far as startx goes with the nvidia driver.
I have a question about AGPGART. I used to have NvAGP set to 1 and when I did cat /proc/driver/nvidia/agp/status, it would say that it was enabled and using NVIDIA driver. Now, I notice that in recent kernels, there's an NVidia option under AGPGART, which I enabled. Is that the same thing as using the NVIDIA driver? Because now, proc returns AGPGART as the driver, and if I try to use NvAGP 1, it returns disabled, saying that another instance of AGP is running or whatever. Is compiling it into the kernel as part of AGPGART the same thing as what I used to do? I didn't have AGPGART compiled into the kernel at all on 2.6.8 I believe, just used NvAGP 1.
Oh yeah, the Nvidia drivers is also what pushed me from Nitro to Love - the last version of Nitro is REAAAAL iffy with Nvidia, even though they compiled it into the kernel. Module refuses to start/work all together. _________________ I'm not a Guru, I just ask a lot of questions. |
|
Back to top |
|
|
Archangel1 Veteran
Joined: 21 Apr 2004 Posts: 1212 Location: Work
|
Posted: Tue Mar 08, 2005 6:45 am Post subject: |
|
|
Nice release fallow - running sweetly here.
It's good to see love-sources are still about, I have a sneaking suspicion that they've been a tad slicker for me than others I've tried.
Going to drop it onto my laptop now to play with some of the other toys in there (swsusp2 + orinoco). Should be good
DrWoland wrote: | Oh yeah, the Nvidia drivers is also what pushed me from Nitro to Love - the last version of Nitro is REAAAAL iffy with Nvidia, even though they compiled it into the kernel. Module refuses to start/work all together. |
I agree... I never had any problems previously, but with this new feature of theirs it doesn't want to go at all on here. A bit unfortunate really since AFAIK they were trying to make it easier... _________________ What are you, stupid? |
|
Back to top |
|
|
Archangel1 Veteran
Joined: 21 Apr 2004 Posts: 1212 Location: Work
|
Posted: Tue Mar 08, 2005 7:13 am Post subject: |
|
|
Oooeerrr... It's refusing to compile with swsusp2 enabled;
Code: | kernel/power/suspend_swap.c: In function `swapwriter_allocate_header_space':
kernel/power/suspend_swap.c:594: error: too few arguments to function `get_swap_page'
kernel/power/suspend_swap.c: In function `swapwriter_allocate_storage':
kernel/power/suspend_swap.c:681: error: too few arguments to function `get_swap_page'
make[2]: *** [kernel/power/suspend_swap.o] Error 1
|
Edit:
Further investigation has indicated that the get_swap_page function in mm/swapfile.c has grown a couple of arguments now, where it had none in the 2.6.10 kernels I have handy.
Not quite sure how other people have managed to make this work? It doesn't seem like it's ever going to work, but someone earlier on in the thread said they had swsusp going? _________________ What are you, stupid? |
|
Back to top |
|
|
Master One l33t
Joined: 25 Aug 2003 Posts: 754 Location: Austria
|
Posted: Tue Mar 08, 2005 8:39 am Post subject: |
|
|
I'm also getting this error now with upd1: Code: | CC kernel/power/suspend_swap.o
kernel/power/suspend_swap.c: In function `swapwriter_allocate_header_space':
kernel/power/suspend_swap.c:594: Fehler: zu wenig Argumente für Funktion »get_swap_page«
kernel/power/suspend_swap.c: In function `swapwriter_allocate_storage':
kernel/power/suspend_swap.c:681: Fehler: zu wenig Argumente für Funktion »get_swap_page«
make[2]: *** [kernel/power/suspend_swap.o] Fehler 1
make[1]: *** [kernel/power] Fehler 2
make: *** [kernel] Fehler 2 |
It wasn't there with the first 2.6.11-love release, so it has something to do with the swapspace optimizations. I simply disabled all the swsusp2 features, but let swsusp2 still be built into the kernel, since I need it to have ipw2200 get compiled correctly (I don't use swsusp2, because it does not work with with ATI's binary drivers, and on the other hand, I like to use swap-encryption, which would get real fuzzy with having to use an initrd to make swsusp2 work with an encrypted swap partition). _________________ Las torturas mentales de la CIA |
|
Back to top |
|
|
VoVaN l33t
Joined: 02 Jul 2003 Posts: 701 Location: The Netherlands
|
Posted: Tue Mar 08, 2005 8:52 am Post subject: Re: reiser4 |
|
|
gimpel wrote: | VoVaN wrote: | Could somebody point me to the only reiser4 patch from the latest love sources? Is that possible to get it separately? The one from love sources seems to be only the one which could be compiled _cleanly_ with vanilla 2.6.11. Thanx |
why not just use love
|
A lot of stuff I really don't need... I'm running now 2.6.10 with reiser4-for-2.6.10-2 from namesys and the latest ck patch, but namesys doesn't have (yet) pach for 2.6.11 kernel.
Thank you for the advise, but of course I tried both of them already, but for both of them I'm getting about the same warning during kernel compilation
Code: |
fs/reiser4/page_cache.c: In function `set_page_dirty_internal':
fs/reiser4/page_cache.c:471: warning: passing arg 1 of `_write_lock_irq' from incompatible pointer type
fs/reiser4/page_cache.c:485: warning: passing arg 1 of `_write_unlock_irq' from incompatible pointer type
....
fs/reiser4/as_ops.c: In function `reiser4_clear_page_dirty':
fs/reiser4/as_ops.c:76: warning: passing arg 1 of `_read_lock_irqsave' from incompatible pointer type
fs/reiser4/as_ops.c:78: warning: passing arg 1 of `_read_unlock_irqrestore' from incompatible pointer type
fs/reiser4/as_ops.c:83: warning: passing arg 1 of `_read_unlock_irqrestore' from incompatible pointer type
fs/reiser4/as_ops.c: In function `reiser4_set_page_dirty':
fs/reiser4/as_ops.c:109: warning: passing arg 1 of `_read_lock_irq' from incompatible pointer type
fs/reiser4/as_ops.c:118: warning: passing arg 1 of `_read_unlock_irq' from incompatible pointer type
fs/reiser4/as_ops.c: In function `reiser4_releasepage':
fs/reiser4/as_ops.c:544: warning: passing arg 1 of `_write_lock_irq' from incompatible pointer type
fs/reiser4/as_ops.c:550: warning: passing arg 1 of `_write_unlock_irq' from incompatible pointer type
....
fs/reiser4/plugin/file/file.c: In function `sync_page_list':
fs/reiser4/plugin/file/file.c:1181: warning: passing arg 1 of `_read_lock_irq' from incompatible pointer type
fs/reiser4/plugin/file/file.c:1193: warning: passing arg 1 of `_read_unlock_irq' from incompatible pointer type
fs/reiser4/plugin/file/file.c:1200: warning: passing arg 1 of `_read_lock_irq' from incompatible pointer type
fs/reiser4/plugin/file/file.c:1203: warning: passing arg 1 of `_read_unlock_irq' from incompatible pointer type
|
I don't think that what we can simply ignore... and I don't have these warnings when I'm compiling love sources . |
|
Back to top |
|
|
fallow Bodhisattva
Joined: 08 Jan 2004 Posts: 2208 Location: Poland
|
Posted: Tue Mar 08, 2005 10:32 am Post subject: |
|
|
OK - working on fixes of swsup2 in update1 and HOSTAP fix and these xfs mencuconfig error
will be availabe soon ad upd2
@VoVaN : please don`t post here error about OHTER sources . I look at this and I was been thinked that is error from love and don`t know what is going on - cause I don`t remember reiser4 compilation problem.
Please just make new thread - and If You want reiser4code from love - I can make a patch for U
@kaktyc : thanx
@DrWoland : Code: |
Option "NvAGP" "integer"
Configure AGP support. Integer argument can be one of:
0 : disable agp
1 : use NVIDIA's internal AGP support, if possible
2 : use AGPGART, if possible
3 : use any agp support (try AGPGART, then NVIDIA's AGP)
Please note that NVIDIA's internal AGP support cannot
work if AGPGART is either statically compiled into your
kernel or is built as a module, but loaded into your
kernel (some distributions load AGPGART into the kernel
at boot up). Default: 3 (the default was 1 until after
1.0-1251). |
@dhaemon : thanks , and : this issue is on genetic_nicksched ?
cheers. _________________ "Time is a companion that goes with us on a journey. It reminds us to cherish each moment, because it will never come again. What we leave behind is not as important as how we have lived" J-L. Picard |
|
Back to top |
|
|
|