View previous topic :: View next topic |
Author |
Message |
yngwin Retired Dev
Joined: 19 Dec 2002 Posts: 4572 Location: Suzhou, China
|
|
Back to top |
|
|
vipernicus Veteran
Joined: 17 Jan 2005 Posts: 1462 Location: Your College IT Dept.
|
|
Back to top |
|
|
santaclaws Apprentice
Joined: 05 Jan 2007 Posts: 161 Location: Deeper Underground
|
Posted: Mon Feb 26, 2007 7:11 am Post subject: |
|
|
vipernicus wrote: | santaclaws wrote: | I found another issue but I do not know if it only concerns my laptop:
If I run "dmesg" the output at the begining is cut off. That was not the case before.
Example:
Code: |
localhost santaclaws # dmesg
is hidden behind transparent bridge #01 (-#01) (try 'pci=assign-busses')
Please report the result to linux-kernel to fix this permanently
PCI: Bus #06 (-#09) is hidden behind transparent bridge #01 (-#01) (try 'pci=assign-busses')
Please report the result to linux-kernel to fix this permanently
ACPI: PCI Interrupt Routing Table [\_SB_.PCI0._PRT]
(..)
|
But the performance, especially the interactivity during compilation, is amazing! |
Very odd. What terminal emulator are you using? |
I am using xterm.. _________________ Software is like sex. It is better when ist is free. |
|
Back to top |
|
|
vipernicus Veteran
Joined: 17 Jan 2005 Posts: 1462 Location: Your College IT Dept.
|
Posted: Mon Feb 26, 2007 7:45 am Post subject: |
|
|
santaclaws wrote: | vipernicus wrote: | santaclaws wrote: | I found another issue but I do not know if it only concerns my laptop:
If I run "dmesg" the output at the begining is cut off. That was not the case before.
Example:
Code: |
localhost santaclaws # dmesg
is hidden behind transparent bridge #01 (-#01) (try 'pci=assign-busses')
Please report the result to linux-kernel to fix this permanently
PCI: Bus #06 (-#09) is hidden behind transparent bridge #01 (-#01) (try 'pci=assign-busses')
Please report the result to linux-kernel to fix this permanently
ACPI: PCI Interrupt Routing Table [\_SB_.PCI0._PRT]
(..)
|
But the performance, especially the interactivity during compilation, is amazing! |
Very odd. What terminal emulator are you using? |
I am using xterm.. |
You don't think it has anything to do with your terminal emulator? _________________ Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog |
|
Back to top |
|
|
santaclaws Apprentice
Joined: 05 Jan 2007 Posts: 161 Location: Deeper Underground
|
Posted: Mon Feb 26, 2007 8:44 am Post subject: |
|
|
vipernicus wrote: | santaclaws wrote: | vipernicus wrote: | santaclaws wrote: | I found another issue but I do not know if it only concerns my laptop:
If I run "dmesg" the output at the begining is cut off. That was not the case before.
Example:
Code: |
localhost santaclaws # dmesg
is hidden behind transparent bridge #01 (-#01) (try 'pci=assign-busses')
Please report the result to linux-kernel to fix this permanently
PCI: Bus #06 (-#09) is hidden behind transparent bridge #01 (-#01) (try 'pci=assign-busses')
Please report the result to linux-kernel to fix this permanently
ACPI: PCI Interrupt Routing Table [\_SB_.PCI0._PRT]
(..)
|
But the performance, especially the interactivity during compilation, is amazing! |
Very odd. What terminal emulator are you using? |
I am using xterm.. |
You don't think it has anything to do with your terminal emulator? |
I do not know - I have never had that strange issue before.
I realized it for the first time when I switched to viper4 and I did not do any terminal upgrades or things like that.. _________________ Software is like sex. It is better when ist is free. |
|
Back to top |
|
|
santaclaws Apprentice
Joined: 05 Jan 2007 Posts: 161 Location: Deeper Underground
|
Posted: Mon Feb 26, 2007 8:51 am Post subject: |
|
|
I just rebooted my computer into nicus2 kernel and (guess) - dmesg works as it should.
Code: |
localhost santaclaws # dmesg
Linux version 2.6.20-nicus2 (root@localhost) (gcc-Version 4.1.2 (Gentoo 4.1.2)) #1 PREEMPT Thu Feb 22 11:43:35 CET 2007
BIOS-provided physical RAM map:
sanitize start
(..)
|
Maybe it is an issue of viper4? _________________ Software is like sex. It is better when ist is free. |
|
Back to top |
|
|
Phlogiston Veteran
Joined: 27 Jan 2004 Posts: 1925 Location: Europe, Swizerland
|
Posted: Mon Feb 26, 2007 9:03 am Post subject: |
|
|
System seems to work but my dmesg reports some strange things:
Code: |
20%...40%...60%...80%...<4>printk: 33321 messages suppressed.
hibernate: page allocation failure. order:0, mode:0x8020
[<c0146d18>] __alloc_pages+0x279/0x28a
[<c011c810>] profile_tick+0x39/0x55
[<c0146d76>] get_zeroed_page+0x1a/0x35
[<c013cb1c>] suspend_prepare_readahead+0xd/0x34
[<c013d4d0>] suspend_rw_buffer+0x19b/0x2e8
[<c013d804>] suspend_bio_read_chunk+0x6d/0xa4
[<c0254e87>] crypto_decompress+0x1d/0x22
[<c013c194>] suspend_compress_read_chunk+0x4d/0x101
[<c013791d>] worker_rw_loop+0x1fd/0x4f7
[<c026dd72>] get_next_bit_on+0x96/0x132
[<c0137d51>] do_rw_loop+0x139/0x1cb
[<c013766e>] rw_init_modules+0x4b/0xfd
[<c0137eaa>] read_pageset+0xc7/0x11d
[<c0137f1b>] read_pageset2+0x1b/0x4f
[<c013b376>] copyback_post+0x8a/0xa6
[<c0136e20>] do_suspend2_step+0x2e5/0x510
[<c0137257>] __suspend2_try_suspend+0x90/0xc5
[<c013636d>] suspend2_attr_store+0x185/0x1c4
[<c018f08d>] sysfs_write_file+0x9a/0xc3
[<c018eff3>] sysfs_write_file+0x0/0xc3
[<c015d59f>] vfs_write+0x8a/0x136
[<c015dada>] sys_write+0x41/0x67
[<c0102c24>] sysenter_past_esp+0x5d/0x81
[<c0400033>] sctp_assoc_lookup_laddr+0x1/0x5a
=======================
Mem-info:
DMA per-cpu:
CPU 0: Hot: hi: 0, btch: 1 usd: 0 Cold: hi: 0, btch: 1 usd: 0
Normal per-cpu:
CPU 0: Hot: hi: 186, btch: 31 usd: 127 Cold: hi: 62, btch: 15 usd: 49
Active:0 inactive:0 dirty:7 writeback:0 unstable:0 free:676 slab:4325 mapped:17425 pagetables:482
DMA free:1744kB min:96kB low:120kB high:144kB lots:192kB active:0kB inactive:0kB present:16256kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 427
Normal free:960kB min:2592kB low:3240kB high:3888kB lots:5184kB active:0kB inactive:0kB present:437772kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0
DMA: 0*4kB 0*8kB 1*16kB 0*32kB 1*64kB 1*128kB 0*256kB 1*512kB 1*1024kB 0*2048kB 0*4096kB = 1744kB
Normal: 0*4kB 0*8kB 0*16kB 0*32kB 1*64kB 1*128kB 1*256kB 1*512kB 0*1024kB 0*2048kB 0*4096kB = 960kB
Swap cache: add 2008, delete 1972, find 10410/10424, race 0+0
Free swap = 1137456kB
Total swap = 1663152kB
Free swap: 1137456kB
114400 pages of RAM
0 pages of HIGHMEM
18680 reserved pages
83231 pages shared
36 pages swap cached
7 pages dirty
0 pages writeback
17425 pages mapped
4325 pages slab
482 pages pagetables
100%...done.
Cleaning up...
Restarting tasks ... done.
Suspend2 debugging info:
- Suspend core : 2.2.9.7
- Kernel Version : 2.6.20-viper4-IBM-T43
- Compiler vers. : 4.1
- Attempt number : 2
- Parameters : 0 81936 0 1 0 0
- Overall expected compression percentage: 40.
- Compressor is 'lzf'.
Compressed 357699584 bytes into 201194738 (43 percent compression).
- SwapAllocator active.
Swap available for image: 415038 pages.
- I/O speed: Write 61 MB/s, Read 61 MB/s.
- Extra pages : -51 used/18800.
[drm] Loading R300 Microcode
ata1.00: configured for UDMA/100
ata1: EH complete
SCSI device sda: 117210240 512-byte hdwr sectors (60012 MB)
sda: Write Protect is off
sda: Mode Sense: 00 3a 00 00
SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[drm] Loading R300 Microcode
|
_________________ Workstation: 5.1 SurroundSound, LIRC remote control; Laptop [IBM-T43]: patched sources, s2disk/ram, fingerprint sensor |
|
Back to top |
|
|
IvanMajhen Guru
Joined: 10 Jun 2006 Posts: 392 Location: Croatia
|
Posted: Mon Feb 26, 2007 10:34 am Post subject: |
|
|
This is only when you use suspend2?
Are you using RSDL?
Suspend2 works perfectly for me and RSDL 0.19. Didn't try later version because I don't wanna break this stable and fast kernel. |
|
Back to top |
|
|
Phlogiston Veteran
Joined: 27 Jan 2004 Posts: 1925 Location: Europe, Swizerland
|
Posted: Mon Feb 26, 2007 9:09 pm Post subject: |
|
|
IvanMajhen wrote: | This is only when you use suspend2?
Are you using RSDL?
Suspend2 works perfectly for me and RSDL 0.19. Didn't try later version because I don't wanna break this stable and fast kernel. |
I'm not using rsdl. This happened with suspend2 sometimes. But now it didn't for about 4 suspends... _________________ 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: Mon Feb 26, 2007 9:21 pm Post subject: |
|
|
Phlogiston wrote: | IvanMajhen wrote: | This is only when you use suspend2?
Are you using RSDL?
Suspend2 works perfectly for me and RSDL 0.19. Didn't try later version because I don't wanna break this stable and fast kernel. |
I'm not using rsdl. This happened with suspend2 sometimes. But now it didn't for about 4 suspends... |
Try IngoSched and Staircase, may be an issue with PlugSched's schedulers. _________________ Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog |
|
Back to top |
|
|
Phlogiston Veteran
Joined: 27 Jan 2004 Posts: 1925 Location: Europe, Swizerland
|
Posted: Mon Feb 26, 2007 9:34 pm Post subject: |
|
|
vipernicus wrote: | Phlogiston wrote: | IvanMajhen wrote: | This is only when you use suspend2?
Are you using RSDL?
Suspend2 works perfectly for me and RSDL 0.19. Didn't try later version because I don't wanna break this stable and fast kernel. |
I'm not using rsdl. This happened with suspend2 sometimes. But now it didn't for about 4 suspends... |
Try IngoSched and Staircase, may be an issue with PlugSched's schedulers. |
Nope, I'm using Staircase
BTW: I'd like to have a stable system I don't like all those "strange" new patches _________________ Workstation: 5.1 SurroundSound, LIRC remote control; Laptop [IBM-T43]: patched sources, s2disk/ram, fingerprint sensor |
|
Back to top |
|
|
RobertDavid Apprentice
Joined: 08 Aug 2005 Posts: 271 Location: Czech Republic
|
Posted: Mon Feb 26, 2007 10:48 pm Post subject: |
|
|
RSDL .23 looks like more stable than .19 (had some modules issue with .19), now I didnt have any problem for few hours... Also I found out that glxgears performance improved from 1020 to 1060 fps with new kernel (with RSDL .19 and modules issue that was 600fps) _________________ Arch & Fluxbox & 2.6.24-rc-zen!!!!
robertek.brevnov.net |
|
Back to top |
|
|
vipernicus Veteran
Joined: 17 Jan 2005 Posts: 1462 Location: Your College IT Dept.
|
Posted: Mon Feb 26, 2007 11:28 pm Post subject: |
|
|
RobertDavid wrote: | RSDL .23 looks like more stable than .19 (had some modules issue with .19), now I didnt have any problem for few hours... Also I found out that glxgears performance improved from 1020 to 1060 fps with new kernel (with RSDL .19 and modules issue that was 600fps) |
Yeah, it's running very very well for me. _________________ Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog |
|
Back to top |
|
|
na641 Apprentice
Joined: 27 Jun 2002 Posts: 171 Location: Eugene, OR
|
Posted: Tue Feb 27, 2007 12:01 am Post subject: |
|
|
There are so many cpusched's to choose from... what is recommend for desktop usage? I take it the RSDL is nice seeing as how viper is using it and most people seem to use it. Selecting "Show plugsched schedulers" completely changes the whole menu... which is very confusing. Whats the difference? I understand the idea behind the pluggable schedulers... but why does selecting the plugsched completely remove the original scheduler options (staircase, rsdl, etc.).
basically i would like to know which is best/preferred for desktop usage. I apologize in advance, i understand the best way to do this would be to compile a kernel for each option and test each one... but i just don't have the time to do all that :\
Any help would be greatly appreciated. |
|
Back to top |
|
|
vipernicus Veteran
Joined: 17 Jan 2005 Posts: 1462 Location: Your College IT Dept.
|
Posted: Tue Feb 27, 2007 1:35 am Post subject: |
|
|
na641 wrote: | There are so many cpusched's to choose from... what is recommend for desktop usage? I take it the RSDL is nice seeing as how viper is using it and most people seem to use it. Selecting "Show plugsched schedulers" completely changes the whole menu... which is very confusing. Whats the difference? I understand the idea behind the pluggable schedulers... but why does selecting the plugsched completely remove the original scheduler options (staircase, rsdl, etc.).
basically i would like to know which is best/preferred for desktop usage. I apologize in advance, i understand the best way to do this would be to compile a kernel for each option and test each one... but i just don't have the time to do all that :\
Any help would be greatly appreciated. |
I recommend Staircase for generic desktops. RSDL for very low latencies.
PlugSched is a different animal than Ingosched, Staircase, and RSDL. Even though it has multiple schedulers, it is one thing. It has various schedulers created by Peter Williams, and some other popular schedulers like Nicksched (also good for desktops).
There are really 4 scheduler interfaces: Ingo, Staircase, RSDL, and PlugSched. _________________ Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog |
|
Back to top |
|
|
vipernicus Veteran
Joined: 17 Jan 2005 Posts: 1462 Location: Your College IT Dept.
|
|
Back to top |
|
|
tranquilcool Veteran
Joined: 25 Mar 2005 Posts: 1245
|
Posted: Wed Feb 28, 2007 2:21 pm Post subject: |
|
|
thanks viper. i needed some e-learning on these stuff. _________________ this is a strange strange world. |
|
Back to top |
|
|
vipernicus Veteran
Joined: 17 Jan 2005 Posts: 1462 Location: Your College IT Dept.
|
|
Back to top |
|
|
bollucks l33t
Joined: 27 Oct 2004 Posts: 606
|
Posted: Thu Mar 01, 2007 4:47 am Post subject: |
|
|
Since lots are using RSDL, don't you think the author of the patch would want some feedback? There's been surprisingly little on the -ck mailing list considering how many patchsets are taking it for a spin. |
|
Back to top |
|
|
lxxb n00b
Joined: 12 Jul 2006 Posts: 19
|
Posted: Thu Mar 01, 2007 2:44 pm Post subject: |
|
|
Great kernel, I used viper3 for a month with no problems (haven't tested suspend2 yet...). I am trying to setup 2.6.20-viper4 on a Thinkpad T60. {EDIT: I am using the same .config file for viper3 and viper4}. I am getting the following compilation problem:
Code: | ...snip...
LD [M] /var/tmp/portage/app-laptop/thinkpad-5.9-r1/work/thinkpad-5.9/2.6/drivers/smapi.o
Building modules, stage 2.
MODPOST 4 modules
CC /var/tmp/portage/app-laptop/thinkpad-5.9-r1/work/thinkpad-5.9/2.6/drivers/rtcmosram.mod.o
CC /var/tmp/portage/app-laptop/thinkpad-5.9-r1/work/thinkpad-5.9/2.6/drivers/superio.mod.o
CC /var/tmp/portage/app-laptop/thinkpad-5.9-r1/work/thinkpad-5.9/2.6/drivers/smapi.mod.o
CC /var/tmp/portage/app-laptop/thinkpad-5.9-r1/work/thinkpad-5.9/2.6/drivers/thinkpad.mod.o
LD [M] /var/tmp/portage/app-laptop/thinkpad-5.9-r1/work/thinkpad-5.9/2.6/drivers/smapi.ko
LD [M] /var/tmp/portage/app-laptop/thinkpad-5.9-r1/work/thinkpad-5.9/2.6/drivers/rtcmosram.ko
LD [M] /var/tmp/portage/app-laptop/thinkpad-5.9-r1/work/thinkpad-5.9/2.6/drivers/superio.ko
LD [M] /var/tmp/portage/app-laptop/thinkpad-5.9-r1/work/thinkpad-5.9/2.6/drivers/thinkpad.ko
make[1]: Leaving directory `/usr/src/linux-2.6.20-viper4'
>>> Source compiled.
>>> Test phase [not enabled]: app-laptop/thinkpad-5.9-r1
>>> Install thinkpad-5.9-r1 into /var/tmp/portage/app-laptop/thinkpad-5.9-r1/image/ category app-laptop
* Installing thinkpad module
install: cannot stat `thinkpad.': No such file or directory
!!! ERROR: app-laptop/thinkpad-5.9-r1 failed.
Call stack:
ebuild.sh, line 1614: Called dyn_install ...snip...
|
Any ideas? |
|
Back to top |
|
|
vipernicus Veteran
Joined: 17 Jan 2005 Posts: 1462 Location: Your College IT Dept.
|
Posted: Thu Mar 01, 2007 3:21 pm Post subject: |
|
|
bollucks wrote: | Since lots are using RSDL, don't you think the author of the patch would want some feedback? There's been surprisingly little on the -ck mailing list considering how many patchsets are taking it for a spin. |
He's gotten feedback from me in PM, we've been able to point out at least one bug to him. (the hotplug_cpu one)
bollucks wrote: | Great kernel, I used viper3 for a month with no problems (haven't tested suspend2 yet...). I am trying to setup 2.6.20-viper4 on a Thinkpad T60. {EDIT: I am using the same .config file for viper3 and viper4}. I am getting the following compilation problem:
Code:
...snip...
LD [M] /var/tmp/portage/app-laptop/thinkpad-5.9-r1/work/thinkpad-5.9/2.6/drivers/smapi.o |
These patches should already be in -viper4, especially tp-smapi. _________________ Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog |
|
Back to top |
|
|
kernelOfTruth Watchman
Joined: 20 Dec 2005 Posts: 6111 Location: Vienna, Austria; Germany; hello world :)
|
Posted: Thu Mar 01, 2007 7:43 pm Post subject: |
|
|
vipernicus, since I'm going to use your kernel on my rig for some time & my laptop [if it works, haven't tested it yet] (I call it kot, "kerneloftruth", since it has all features I need ), is there a possibility to get a dyntick patch for it?
Quote: | "The tickless kernel feature (CONFIG_NO_HZ) enables 'on-demand' timer interrupts: if there is no timer to be expired for say 1.5 seconds when the system goes idle, then the system will stay totally idle for 1.5 seconds. This should bring cooler CPUs and power savings: on our (x86) testboxes we have measured the effective IRQ rate to go from HZ to 1-2 timer interrupts per second.
"This feature is implemented by driving 'low res timer wheel' processing via special per-CPU high-res timers, which timers are reprogrammed to the next-low-res-timer-expires interval. This tickless-kernel design is SMP-safe in a natural way and has been developed on SMP systems from the beginning."
|
I think this would gain at least 20-30 minutes of battery life, (if not already included via hpet)
thanks in advance _________________ https://github.com/kernelOfTruth/ZFS-for-SystemRescueCD/tree/ZFS-for-SysRescCD-4.9.0
https://github.com/kernelOfTruth/pulseaudio-equalizer-ladspa
Hardcore Gentoo Linux user since 2004 |
|
Back to top |
|
|
vipernicus Veteran
Joined: 17 Jan 2005 Posts: 1462 Location: Your College IT Dept.
|
Posted: Thu Mar 01, 2007 7:52 pm Post subject: |
|
|
kernelOfTruth wrote: | vipernicus, since I'm going to use your kernel on my rig for some time & my laptop [if it works, haven't tested it yet] (I call it kot, "kerneloftruth", since it has all features I need ), is there a possibility to get a dyntick patch for it?
Quote: | "The tickless kernel feature (CONFIG_NO_HZ) enables 'on-demand' timer interrupts: if there is no timer to be expired for say 1.5 seconds when the system goes idle, then the system will stay totally idle for 1.5 seconds. This should bring cooler CPUs and power savings: on our (x86) testboxes we have measured the effective IRQ rate to go from HZ to 1-2 timer interrupts per second.
"This feature is implemented by driving 'low res timer wheel' processing via special per-CPU high-res timers, which timers are reprogrammed to the next-low-res-timer-expires interval. This tickless-kernel design is SMP-safe in a natural way and has been developed on SMP systems from the beginning."
|
I think this would gain at least 20-30 minutes of battery life, (if not already included via hpet)
thanks in advance |
I've been looking into this and will probably attempt to port it over for viper5. In the meantime you can try using 100-300hz options. _________________ Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog |
|
Back to top |
|
|
kernelOfTruth Watchman
Joined: 20 Dec 2005 Posts: 6111 Location: Vienna, Austria; Germany; hello world :)
|
|
Back to top |
|
|
vipernicus Veteran
Joined: 17 Jan 2005 Posts: 1462 Location: Your College IT Dept.
|
Posted: Thu Mar 01, 2007 7:58 pm Post subject: |
|
|
kernelOfTruth wrote: | thanks
viper4, btw., is also doing fine with exec-shield: "bulletproof pinguin" |
Great, though, I'll probably hold out for grsecurity and PAX. I'll most likely have PAX in next release. _________________ Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog |
|
Back to top |
|
|
|