Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
2.6.18-rc4-viper1: Ah, Now That Hit The Spot | Plugsched!
View unanswered posts
View posts from last 24 hours

Goto page 1, 2, 3, 4, 5  Next  
Reply to topic    Gentoo Forums Forum Index Unsupported Software
View previous topic :: View next topic  
Author Message
vipernicus
Veteran
Veteran


Joined: 17 Jan 2005
Posts: 1462
Location: Your College IT Dept.

PostPosted: Fri Aug 11, 2006 2:43 am    Post subject: 2.6.18-rc4-viper1: Ah, Now That Hit The Spot | Plugsched! Reply with quote

Viper Sources is pretty much my playground for new patches.

The main goal of this patchset is pretty much the same as E-Mission Sources: to
offer latest features, decrease latencies, decrease overhead, and improve interactivity. The only difference is experimental versions of everything. I boot test each release and run for an extended amount of time.


/* Notes */
Alot of changes in this version, I've been in contact with Peter Williams lately, the creater of Plugsched,
he claims that all schedulers are running at or around 100% performance, so I thought I'd give it a shot.

Added in latest genpatches, plugsched, iosched fixes rollup, newer lockless, ck desktop patches,
and build fixes. No extra fixes are needed to 'emerge ati-drivers'!


2.6.18-rc4-viper1: Ah, Now That Hit The Spot
Code:
/* series */
# Latest Genpatches for 2.6.18
1705_sparc-U1-hme-lockup.patch
4000_deprecate-sk98lin.patch
4105_dm-bbr.patch
4110_promise-pdc2037x.patch
4200_fbsplash-0.9.2-r5.patch
4205_vesafb-tng-1.0-rc2.patch
4300_squashfs-3.0.patch
4405_alpha-sysctl-uac.patch
#
# Latest Plugsched by Peter Williams
plugsched-6.4-for-2.6.18-rc2.patch
#
# Latest IOSched Improvements by Jens Axboe
iosched-rollup-2.6.18-rc3-1
#
# Nick Piggin Patches from -mm
radix-tree-rcu-lockless-readside.patch
redo-radix-tree-fixes.patch
adix-tree-rcu-lockless-readside-update.patch
radix-tree-rcu-lockless-readside-semicolon.patch
adix-tree-rcu-lockless-readside-update-tidy.patch
adix-tree-rcu-lockless-readside-fix-2.patch
mm-non-syncing-lock_page.patch
mm-speculative-get_page.patch
mm-speculative-get_page-uninlining.patch
mm-speculative-get_page-fix.patch
mm-lockless-pagecache.patch
#
# More patches from -mm
mm-vm_bug_on.patch
sleazy-fpu-feature-i386-support.patch
sleazy-fpu-feature-x86_64-support.patch
vectorize-aio_read-aio_write-fileop-methods.patch
vectorize-aio_read-aio_write-fileop-methods-xfs-fix.patch
vectorize-aio_read-aio_write-fileop-methods-hypfs-fix.patch
remove-readv-writev-methods-and-use-aio_read-aio_write.patch
streamline-generic_file_-interfaces-and-filemap.patch
add-vector-aio-support.patch
add-vector-aio-support-fix.patch
add-address_space_operationsbatch_write.patch
add-address_space_operationsbatch_write-fix.patch
pass-io-size-to-batch_write-address-space-operation.patch
#
# Fixes
mmc-queue-fix.patch
dm-mmc-fix.patch
dm-bbr-temp-fix.patch
ati-drivers-fix.patch
#
# -ck patches
sched-add-above-background-load-function.patch
mm-swap_prefetch-33.patch
kconfig-expose_vmsplit_option.patch
track_mutexes-1.patch
mm-convert_swappiness_to_mapped.patch
mm-lots_watermark.diff
mm-kswapd_inherit_prio-1.patch
mm-prio_dependant_scan-1.patch
mm-background_scan-2.patch
mm-decrease_minimum_dirty_ratio.patch
mm-set_zero_dirty_ratio.patch
mm-filesize_dependant_lru_cache_add.patch
#
# Latest Reiser4 Patch from -mm
reiser4-export-handle_ra_miss.patch
reiser4-sb_sync_inodes.patch
reiser4-export-remove_from_page_cache.patch
reiser4-export-radix_tree_preload.patch
reiser4-export-find_get_pages.patch
make-copy_from_user_inatomic-not-zero-the-tail-on-i386-vs-reiser4.patch
reiser4.patch
reiser4-hardirq-include-fix.patch
reiser4-fix-trivial-tyops-which-were-hard-to-hit.patch
reiser4-run-truncate_inode_pages-in-reiser4_delete_inode.patch
reiser4-bug-fixes.patch
reiser4-write-via-do_sync_write.patch
reiser4-fix-gcc-ws-compains.patch
fs-reiser4-possible-cleanups.patch
reiser4-get_sb_dev-fix.patch
reiser4-vs-zoned-allocator.patch
reiser4-vs-streamline-generic_file_-interfaces-and-filemap.patch
reiser4-vs-streamline-generic_file_-interfaces-and-filemap-fix.patch
#
# Unionfs
unionfs-latest.patch
#
# Inode-Diet from -mm
inode_diet-replace-inodeugeneric_ip-with-inodei_private.patch
inode-diet-move-i_pipe-into-a-union.patch
inode-diet-move-i_bdev-into-a-union.patch
inode-diet-move-i_cdev-into-a-union.patch
inode-diet-eliminate-i_blksize-and-use-a-per-superblock-default.patch
inode-diet-eliminate-i_blksize-and-use-a-per-superblock-default-fix.patch
inode_diet-replace-inodeugeneric_ip-with-inodei_private-reiser4.patch
inode-diet-eliminate-i_blksize-and-use-a-per-superblock-default-reiser4.patch
inode-diet-squashfs-3.0.patch
inode-diet-unionfs-latest.patch
#
# Suspend2
2010-suspend2-freezer-upgrade.patch
2020-kmod-freeze.patch
3000-crypto-api-lzf-support.patch
3020-cryptoapi-deflate.patch
3030-reboot-handler-hook.patch
3050-init-hooks.patch
3060-reset-kswapd-max-order-after-resume.patch
3080-dynamic-pageflags.patch
9900-Suspend2.patch
#
# Realtime Linux Security Model
realtime-lsm.patch
#
# patches from -tiny
cpu-support-rollup.patch
#
# Centrino Undervolting Patch
linux-phc-0.2.6-2.6.18.patch
#
# Ooo, pretty
lxdialog-refractor-color-support.patch
lxdialog-color-themes-blackbg.patch
lxdialog-color-themes-classic.patch
daconfig-2.3.1-2.6.17-rc4
name-tag.patch


Enjoy!

Patch:
http://vipernicus.evolution-mission.org/patches/2.6.18-rc4/viper1/linux-2.6.18-rc4-viper1.bz2

Ebuild:
http://vipernicus.evolution-mission.org/patches/2.6.18-rc4/viper1/viper-sources-2.6.18_rc4.ebuild

Broken-Out:
http://vipernicus.evolution-mission.org/patches/2.6.18-rc4/viper1/linux-2.6.18-rc4-viper1-broken-out.tar.bz2

Broken-Out Directory:
http://vipernicus.evolution-mission.org/patches/2.6.18-rc4/viper1/broken-out

Making it work:

Ebuild Method:
Code:
mkdir -p /usr/local/portage/sys-kernel/viper-sources
cd /usr/local/portage/sys-kernel/viper-sources
wget http://vipernicus.evolution-mission.org/patches/2.6.18-rc4/viper1/viper-sources-2.6.18_rc4.ebuild
ebuild viper-sources-2.6.18_rc4.ebuild digest
USE="symlink" emerge viper-sources
cd /usr/src/linux
make MENUCONFIG_COLOR=blackbg menuconfig


Manual Method:
Code:
cd /usr/src
rm linux
wget http://www.kernel.org/pub/linux/kernel/v2.6/linux-2.6.17.tar.bz2
tar -xvjf linux-2.6.17.tar.bz2
mv linux-2.6.17 linux-2.6.18-rc4-viper1
ln -s linux-2.6.18-rc4-viper1 linux
cd linux
wget http://www.kernel.org/pub/linux/kernel/v2.6/testing/patch-2.6.18-rc4.bz2
wget http://vipernicus.evolution-mission.org/patches/2.6.18-rc4/viper1/linux-2.6.18-rc4-viper1.bz2
bzcat patch-2.6.18-rc4.bz2 | patch -p1
bzcat linux-2.6.18-rc4-viper1.bz2 | patch -p1
rm -rf *.bz2
make MENUCONFIG_COLOR=blackbg menuconfig

_________________
Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog


Last edited by vipernicus on Sat Aug 12, 2006 4:24 pm; edited 6 times in total
Back to top
View user's profile Send private message
cheater1034
Veteran
Veteran


Joined: 09 Sep 2004
Posts: 1558

PostPosted: Fri Aug 11, 2006 2:45 pm    Post subject: Reply with quote

Fails to build on:
Code:
.tmp.Kallsyms
include/asm/types.h:4 something about  typedef _Bool bool


Just thought I'd report it.
_________________
IRC!: #zen-sources on irc.rizon.net
zen-kernel.org
--
Lost in android development land.
Back to top
View user's profile Send private message
Dinini
Apprentice
Apprentice


Joined: 11 Jun 2003
Posts: 236

PostPosted: Fri Aug 11, 2006 4:01 pm    Post subject: Reply with quote

I was hoping it was because I am using Ubuntu rather than gentoo on this notebook and the compiler versions are different.

affected line is: include/asm-i386/types.h:4:typedef _Bool bool;
problem is from the generic-boolean.patch
Back to top
View user's profile Send private message
vipernicus
Veteran
Veteran


Joined: 17 Jan 2005
Posts: 1462
Location: Your College IT Dept.

PostPosted: Fri Aug 11, 2006 5:11 pm    Post subject: Reply with quote

The problem is now fixed, redownload patch.
_________________
Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog
Back to top
View user's profile Send private message
vipernicus
Veteran
Veteran


Joined: 17 Jan 2005
Posts: 1462
Location: Your College IT Dept.

PostPosted: Fri Aug 11, 2006 8:22 pm    Post subject: Reply with quote

Ran some tests with interbench. So far, Staircase CPUSched is running as well as vanilla. But Nicksched's latencies shine! I recommend you guys to give Nicksched a try with this release. It is running smooth as silk here ;)
_________________
Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog


Last edited by vipernicus on Sat Aug 12, 2006 2:10 am; edited 1 time in total
Back to top
View user's profile Send private message
jnengland77
n00b
n00b


Joined: 27 Jan 2005
Posts: 8

PostPosted: Sat Aug 12, 2006 1:50 am    Post subject: Reply with quote

Nice, this kernel works great. I've been dispointed with the 2.6.17 kernels but this one is nice and Nickshed makes it even better!

thanks,
jnengland77
Back to top
View user's profile Send private message
a7thson
Apprentice
Apprentice


Joined: 08 Apr 2006
Posts: 176
Location: your pineal gland

PostPosted: Sat Aug 12, 2006 2:12 am    Post subject: Reply with quote

vipernicus wrote:
Ran some tests on interactivity. So far, staircase is running as good as it does on vanilla. But Nicksched's latencies shine! I recommend you guys to give nicksched a try with this release. It is running smooth as silk here ;)

Downloaded with a hacked-together ebuild and it's compiling as I write this. I went ahead with nicksched (which I've never played with previously) and will attempt to reboot into the new kernel shortly, with a quick update to this telling whether or not it succeeded (and/or what blew up).

The SD/MMC modules buit without issue, thanks for including the relevant patches. ati-drivers requires the same patch I mentioned earlier, and then builds fine. up-to-date ipw2200 drivers, qemu kernel module (a binary), arpstar all emerged without issue. net-fs/shfs still broken. more information forthcoming ;-) Now to try it all before it's time to sleep...
_________________
i7-3610QM | E5-2670 | FX-8300
Back to top
View user's profile Send private message
vipernicus
Veteran
Veteran


Joined: 17 Jan 2005
Posts: 1462
Location: Your College IT Dept.

PostPosted: Sat Aug 12, 2006 2:30 am    Post subject: Reply with quote

a7thson wrote:
vipernicus wrote:
Ran some tests on interactivity. So far, staircase is running as good as it does on vanilla. But Nicksched's latencies shine! I recommend you guys to give nicksched a try with this release. It is running smooth as silk here ;)

Downloaded with a hacked-together ebuild and it's compiling as I write this. I went ahead with nicksched (which I've never played with previously) and will attempt to reboot into the new kernel shortly, with a quick update to this telling whether or not it succeeded (and/or what blew up).

The SD/MMC modules buit without issue, thanks for including the relevant patches. ati-drivers requires the same patch I mentioned earlier, and then builds fine. up-to-date ipw2200 drivers, qemu kernel module (a binary), arpstar all emerged without issue. net-fs/shfs still broken. more information forthcoming ;-) Now to try it all before it's time to sleep...


I've already included an ati-fix. I did some Makefile trickery to get it to work:

This is what my included in rc4-viper1 looks like:

ati-drivers-fix.patch
Code:
diff -purN linux-2.6.17/Makefile linux-2.6.17-test/Makefile
--- linux-2.6.17/Makefile   2006-08-10 21:08:17.000000000 -0500
+++ linux-2.6.17-test/Makefile   2006-08-10 21:12:00.000000000 -0500
@@ -860,7 +860,8 @@ endef
 define filechk_version.h
    (echo \#define LINUX_VERSION_CODE $(shell                             \
    expr $(VERSION) \* 65536 + $(PATCHLEVEL) \* 256 + $(SUBLEVEL));     \
-   echo '#define KERNEL_VERSION(a,b,c) (((a) << 16) + ((b) << 8) + (c))';)
+   echo '#define KERNEL_VERSION(a,b,c) (((a) << 16) + ((b) << 8) + (c))'; \
+   echo \#define UTS_RELEASE \"$(KERNELRELEASE)\";)
 endef
 
 include/linux/version.h: $(srctree)/Makefile FORCE
diff -purN linux-2.6.17/include/linux/mm.h linux-2.6.17-test/include/linux/mm.h
--- oldtree/include/linux/mm.h   2006-08-10 21:08:17.000000000 -0500
+++ newtree/include/linux/mm.h   2006-08-10 21:08:17.000000000 -0500
@@ -165,6 +165,7 @@
 #define VM_NONLINEAR   0x00800000   /* Is non-linear (remap_file_pages) */
 #define VM_MAPPED_COPY   0x01000000   /* T if mapped copy of data (nommu mmap) */
 #define VM_INSERTPAGE   0x02000000   /* The vma has had "vm_insert_page()" done on it */
+#define VM_SHM          0x00000000      /* ATI Drivers won't work unless you supply this def. */
 
 #ifndef VM_STACK_DEFAULT_FLAGS      /* arch can override this */
 #define VM_STACK_DEFAULT_FLAGS VM_DATA_DEFAULT_FLAGS


This way when you build the kernel it automatically adds the UTS_RELEASE to your version.h.
_________________
Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog
Back to top
View user's profile Send private message
a7thson
Apprentice
Apprentice


Joined: 08 Apr 2006
Posts: 176
Location: your pineal gland

PostPosted: Sat Aug 12, 2006 3:15 am    Post subject: Reply with quote

vipernicus wrote:

I've already included an ati-fix. I did some Makefile trickery to get it to work:

cool. that is definitely a cleaner approach than the patch that I found and used in the previous version to get ati-drivers to build. good deal! this builds fine, I didn't realize that I was actually building the version from my local overlay with the older patch, somehow nothing broke but I'll recompile it again just to be sure.
_________________
i7-3610QM | E5-2670 | FX-8300
Back to top
View user's profile Send private message
a7thson
Apprentice
Apprentice


Joined: 08 Apr 2006
Posts: 176
Location: your pineal gland

PostPosted: Sat Aug 12, 2006 3:31 pm    Post subject: Reply with quote

#uname -a
Linux neuros 2.6.18-rc4-viper #1 PREEMPT Fri Aug 11 20:58:23 CDT 2006 i686 Intel(R) Pentium(R) M processor 1.60GHz GNU/Linux

et voila - nicksched is in place for the time being and so far very nice. MMC support, vmware kernel module, ati-drivers build perfectly.

system config: SATA, firewire, usb-storage, ipw2200, ext3+reiser+XFS+fat32, SD/MMC reader, ATI Radeon X300
external modules working: ieee80211, ipw2200, ati-drivers, arpstar (built, not tested), kqemu (qemu kernel module, tested), vmmon (vmware kernel module, tested).
still broken: shfs (does anyone know the status of this project, it seems to have been last updated Feb 2006)

so far my overall impression is... :-D
_________________
i7-3610QM | E5-2670 | FX-8300
Back to top
View user's profile Send private message
vipernicus
Veteran
Veteran


Joined: 17 Jan 2005
Posts: 1462
Location: Your College IT Dept.

PostPosted: Sat Aug 12, 2006 4:28 pm    Post subject: Reply with quote

Ebuild is now online, thanks to a7thson
_________________
Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog
Back to top
View user's profile Send private message
vipernicus
Veteran
Veteran


Joined: 17 Jan 2005
Posts: 1462
Location: Your College IT Dept.

PostPosted: Sun Aug 13, 2006 5:23 am    Post subject: Reply with quote

zxy wrote:
Freezes during compile
Code:
  CC      drivers/base/bus.o

on amd64. (gcc 4.1.1)


Interesting, it just stops there while compiling the kernel?
_________________
Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog
Back to top
View user's profile Send private message
zxy
Veteran
Veteran


Joined: 06 Jan 2006
Posts: 1160
Location: in bed in front of the computer

PostPosted: Sun Aug 13, 2006 6:10 am    Post subject: Reply with quote

@ Vipernicus: No, (well, yes), i had a harddrive full. (to many kernel sources...) :roll:
_________________
Nature does not hurry, yet everything is accomplished.
Lao Tzu
Back to top
View user's profile Send private message
vipernicus
Veteran
Veteran


Joined: 17 Jan 2005
Posts: 1462
Location: Your College IT Dept.

PostPosted: Sun Aug 13, 2006 6:13 am    Post subject: Reply with quote

zxy wrote:
@ Vipernicus: No, (well, yes), i had a harddrive full. (to many kernel sources...) :roll:


Sounds like a personal problem :lol:
_________________
Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog
Back to top
View user's profile Send private message
zxy
Veteran
Veteran


Joined: 06 Jan 2006
Posts: 1160
Location: in bed in front of the computer

PostPosted: Sun Aug 13, 2006 6:23 am    Post subject: Reply with quote

Works, and it works GOOOD. (for now at least)

One question. No-op sheduler has no description. Could somebody say something abot it.

Thanks
_________________
Nature does not hurry, yet everything is accomplished.
Lao Tzu
Back to top
View user's profile Send private message
rituko_a
n00b
n00b


Joined: 04 Dec 2005
Posts: 16
Location: Moscow, Russia

PostPosted: Sun Aug 13, 2006 1:54 pm    Post subject: Reply with quote

Code:
  CC [M]  /var/tmp/portage/alsa-driver-1.0.12_rc1-r1/work/alsa-driver-1.0.12rc1/acore/pcm_native.o
  CC [M]  /var/tmp/portage/alsa-driver-1.0.12_rc1-r1/work/alsa-driver-1.0.12rc1/pci/ac97/ac97_pcm.o
  CC [M]  /var/tmp/portage/alsa-driver-1.0.12_rc1-r1/work/alsa-driver-1.0.12rc1/pci/ac97/ac97_patch.o
/var/tmp/portage/alsa-driver-1.0.12_rc1-r1/work/alsa-driver-1.0.12rc1/acore/pcm_native.c:3598: error: unknown field ‘writev’ specified in initializer
/var/tmp/portage/alsa-driver-1.0.12_rc1-r1/work/alsa-driver-1.0.12rc1/acore/pcm_native.c:3598: warning: initialization from incompatible pointer type
/var/tmp/portage/alsa-driver-1.0.12_rc1-r1/work/alsa-driver-1.0.12rc1/acore/pcm_native.c:3618: error: unknown field ‘readv’ specified in initializer
/var/tmp/portage/alsa-driver-1.0.12_rc1-r1/work/alsa-driver-1.0.12rc1/acore/pcm_native.c:3618: warning: initialization from incompatible pointer type
make[4]: *** [/var/tmp/portage/alsa-driver-1.0.12_rc1-r1/work/alsa-driver-1.0.12rc1/acore/pcm_native.o] Error 1
make[3]: *** [/var/tmp/portage/alsa-driver-1.0.12_rc1-r1/work/alsa-driver-1.0.12rc1/acore] Error 2
make[3]: *** Waiting for unfinished jobs....


I have such error when compiling alsa-driver-1.0.12_rc1-r1 against this kernel. How can I fix it? :\
gcc is 4.1.1,
CFLAGS="-O2 -march=athlon-xp -fomit-frame-pointer -pipe"
ALSA_CARDS="emu10k1 via82xx"
Back to top
View user's profile Send private message
vipernicus
Veteran
Veteran


Joined: 17 Jan 2005
Posts: 1462
Location: Your College IT Dept.

PostPosted: Sun Aug 13, 2006 1:56 pm    Post subject: Reply with quote

rituko_a wrote:
Code:
  CC [M]  /var/tmp/portage/alsa-driver-1.0.12_rc1-r1/work/alsa-driver-1.0.12rc1/acore/pcm_native.o
  CC [M]  /var/tmp/portage/alsa-driver-1.0.12_rc1-r1/work/alsa-driver-1.0.12rc1/pci/ac97/ac97_pcm.o
  CC [M]  /var/tmp/portage/alsa-driver-1.0.12_rc1-r1/work/alsa-driver-1.0.12rc1/pci/ac97/ac97_patch.o
/var/tmp/portage/alsa-driver-1.0.12_rc1-r1/work/alsa-driver-1.0.12rc1/acore/pcm_native.c:3598: error: unknown field ‘writev’ specified in initializer
/var/tmp/portage/alsa-driver-1.0.12_rc1-r1/work/alsa-driver-1.0.12rc1/acore/pcm_native.c:3598: warning: initialization from incompatible pointer type
/var/tmp/portage/alsa-driver-1.0.12_rc1-r1/work/alsa-driver-1.0.12rc1/acore/pcm_native.c:3618: error: unknown field ‘readv’ specified in initializer
/var/tmp/portage/alsa-driver-1.0.12_rc1-r1/work/alsa-driver-1.0.12rc1/acore/pcm_native.c:3618: warning: initialization from incompatible pointer type
make[4]: *** [/var/tmp/portage/alsa-driver-1.0.12_rc1-r1/work/alsa-driver-1.0.12rc1/acore/pcm_native.o] Error 1
make[3]: *** [/var/tmp/portage/alsa-driver-1.0.12_rc1-r1/work/alsa-driver-1.0.12rc1/acore] Error 2
make[3]: *** Waiting for unfinished jobs....


I have such error when compiling alsa-driver-1.0.12_rc1-r1 against this kernel. How can I fix it? :\
gcc is 4.1.1,
CFLAGS="-O2 -march=athlon-xp -fomit-frame-pointer -pipe"
ALSA_CARDS="emu10k1 via82xx"


Follow the Gentoo Linux Alsa Guide:
http://www.gentoo.org/doc/en/alsa-guide.xml

Don't use 'emerge alsa-driver', use the built in kernel drivers, it makes life easier.

This might help you along:
Code:
<M> Advanced Linux Sound Architecture
<M>   Sequencer support             
<M>     Sequencer dummy client       
<M>   OSS Mixer API                 
<M>   OSS PCM (digital audio) API   
[*]     OSS PCM (digital audio) API - Include plugin system
[*]   OSS Sequencer API             
<M>   RTC Timer support             
[*]     Use RTC as default sequencer timer
[ ]   Dynamic device file minor numbers
[*]   Support old ALSA API           
[*]   Verbose procfs contents       
[ ]   Verbose printk                 
[ ]   Debug                         
      Generic devices  --->         
      PCI devices  --->             
      USB devices  --->             


Under PCI devices:

Code:
<M> Emu10k1 (SB Live!, Audigy, E-mu APS)

_________________
Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog
Back to top
View user's profile Send private message
rituko_a
n00b
n00b


Joined: 04 Dec 2005
Posts: 16
Location: Moscow, Russia

PostPosted: Sun Aug 13, 2006 2:47 pm    Post subject: Reply with quote

Thanks, that helped =)
Back to top
View user's profile Send private message
markus42
n00b
n00b


Joined: 11 Jul 2006
Posts: 53

PostPosted: Mon Aug 14, 2006 6:55 am    Post subject: Reply with quote

I thought what the hey, I'll give it a go in my Gentoo VM running on a Windows XP host.

Linux gentoovm 2.6.18-rc4-viper #1 Mon Aug 14 16:23:41 EST 2006 i686 Intel(R) Pentium(R) 4 CPU 3.00GHz GNU/Linux using Nicksched is running sort of OK with one glitch. The VMWare tools software must be recompiled for the running kernel. This has worked flawlessly so far for gentoo-sources and emission-sources.

I get the following error with viper-sources when trying to compile VMwareTools-1.0.0-28343...
/tmp/vmware-config5/vmhgfs-only/driver.c: In function `HgfsChangeFileAttributes':
/tmp/vmware-config5/vmhgfs-only/driver.c:763: error: structure has no member named `i_blksize'
/tmp/vmware-config5/vmhgfs-only/driver.c: In function `HgfsInitializeInode':
/tmp/vmware-config5/vmhgfs-only/driver.c:835: error: structure has no member named `u'
etc...

Any clues?
Back to top
View user's profile Send private message
darklegion
Guru
Guru


Joined: 14 Nov 2004
Posts: 468

PostPosted: Mon Aug 14, 2006 7:17 am    Post subject: Reply with quote

Great to see a kernel with nicksched back again. :)
A useful tweak for nicksched is to set the base_timeslice to a value lower than 256 such as 128 or 64 for better interactivity.64 can cause problems with some applications though, so its safer to set it 128.
As root :
Code:

echo 128 > /sys/cpusched/nicksched/base_timeslice

You can also put this in /etc/conf.d/local.start to set it everytime at boot.
Back to top
View user's profile Send private message
vipernicus
Veteran
Veteran


Joined: 17 Jan 2005
Posts: 1462
Location: Your College IT Dept.

PostPosted: Mon Aug 14, 2006 12:02 pm    Post subject: Reply with quote

markus42 wrote:
I thought what the hey, I'll give it a go in my Gentoo VM running on a Windows XP host.

Linux gentoovm 2.6.18-rc4-viper #1 Mon Aug 14 16:23:41 EST 2006 i686 Intel(R) Pentium(R) 4 CPU 3.00GHz GNU/Linux using Nicksched is running sort of OK with one glitch. The VMWare tools software must be recompiled for the running kernel. This has worked flawlessly so far for gentoo-sources and emission-sources.

I get the following error with viper-sources when trying to compile VMwareTools-1.0.0-28343...
/tmp/vmware-config5/vmhgfs-only/driver.c: In function `HgfsChangeFileAttributes':
/tmp/vmware-config5/vmhgfs-only/driver.c:763: error: structure has no member named `i_blksize'
/tmp/vmware-config5/vmhgfs-only/driver.c: In function `HgfsInitializeInode':
/tmp/vmware-config5/vmhgfs-only/driver.c:835: error: structure has no member named `u'
etc...

Any clues?


Looks like it doesn't like inode-diet patches. Do vmware-modules compile? I've had reports that they work well on rc4-viper1. Have any of you ran into problems with nvidia-drivers compiling?
_________________
Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog
Back to top
View user's profile Send private message
enderandrew
l33t
l33t


Joined: 25 Oct 2005
Posts: 731

PostPosted: Mon Aug 14, 2006 1:38 pm    Post subject: Reply with quote

With those patches it removes the i_blksize from the block structure, but I've found that plenty of file systems still make calls to it. Just remove them and you should be good.
_________________
Nihilism makes me smile.
Back to top
View user's profile Send private message
Buddman
Tux's lil' helper
Tux's lil' helper


Joined: 04 Dec 2005
Posts: 115

PostPosted: Mon Aug 14, 2006 5:19 pm    Post subject: Reply with quote

I keep getting odd issues with Alsa. Sometimes when I boot the kernel, I get errors to where my sound plays really low/slow after loading into desktop, and if I reboot sometimes the problem fixes itself, sometimes it remains. This happens whether I load my alsa driver in the kernel or as a module. Anyone else see this issue?

I'm using the ice1724 driver BTW for an M-Audio Revolution. Doesn't happen with ck source kernel or gentoo's.
Back to top
View user's profile Send private message
immudium
Guru
Guru


Joined: 12 Oct 2004
Posts: 332
Location: Utah

PostPosted: Mon Aug 14, 2006 5:30 pm    Post subject: Reply with quote

darklegion wrote:
Great to see a kernel with nicksched back again. :)
A useful tweak for nicksched is to set the base_timeslice to a value lower than 256 such as 128 or 64 for better interactivity.64 can cause problems with some applications though, so its safer to set it 128.
As root :
Code:

echo 128 > /sys/cpusched/nicksched/base_timeslice

You can also put this in /etc/conf.d/local.start to set it everytime at boot.


Thanks for the tip. I wasn't sure what to expect with this scheduler never having used it before and noticed there were small delays when switching contexts between apps that I didn't notice with other schedulers. It appears to me to be much more desktop friendly with that setting, but I will have to test some more.

Anyway, other than that, this release has been running great for me the last few days. Reiser4 + lockless is running without issue despite my efforts to break it by yanking the power in the middle of writing data, copying big data blocks from ntfs, and lots of I/O. Stuff that has easily broken reiser4 in the past hasn't caused any inconsistencies at least that can be detected by multiple fsck tests. Also, I haven't had any problems at all compiling the 1.0.8762 nvidia drivers with this release.
Back to top
View user's profile Send private message
vipernicus
Veteran
Veteran


Joined: 17 Jan 2005
Posts: 1462
Location: Your College IT Dept.

PostPosted: Mon Aug 14, 2006 5:49 pm    Post subject: Reply with quote

immudium wrote:
Anyway, other than that, this release has been running great for me the last few days. Reiser4 + lockless is running without issue despite my efforts to break it by yanking the power in the middle of writing data, copying big data blocks from ntfs, and lots of I/O. Stuff that has easily broken reiser4 in the past hasn't caused any inconsistencies at least that can be detected by multiple fsck tests. Also, I haven't had any problems at all compiling the 1.0.8762 nvidia drivers with this release.


Great, it's good to know you've had good results.
_________________
Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Unsupported Software All times are GMT
Goto page 1, 2, 3, 4, 5  Next
Page 1 of 5

 
Jump to:  
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