| View previous topic :: View next topic |
| Author |
Message |
GentooBox
Veteran
Joined: 22 Jun 2003
Posts: 1168
Location: Denmark
|
 Posted: Sat Jan 03, 2004 6:46 pm Post subject: |
 |
|
To Love-sources maintainer:
When you release a new love-sources, then set all options in .config to =y and compile the kernel before you release it.
if any compile problems comes up, then see whats wrong, and fix it.
that will minimize compile problems for us.
you should also check rejected chunks of code then you patch the kernel.
_________________
Encrypt, lock up everything and duct tape the rest |
|
| Back to top |
|
 |
bssteph
l33t
Joined: 26 Feb 2003
Posts: 652
Location: Wisconsin
|
| Posted: Sat Jan 03, 2004 6:52 pm Post subject: |
|
|
| StringCheesian wrote: | I'm getting some freaky results here. It boots up fast, but as KDE CVS is starting up, the mouse cursor seems to do about 5 to 10 frames per second and everything is in slow, jerky motion. But as soon as I start Konqueror everything goes back to fast and responsive again. I have KDE set to keep 1 instance of Konqueror loaded into memory.
This didn't happen with 2.6.0-mm2 or 2.6.0-love1. Could this have something to do with that swappiness patch?
edit: it happens when I play a wav file too! That stupid swappiness patch is evil! It's a terrible idea, or at least a terrible implementation of an ok idea. I can't stand this, I'm going back to 2.6.0-love1 or love2. |
Let's not jump to conclusions. cat /proc/sys/vm/swappiness to see what the level currently is. The closer to 60, the less the patch is really mattering. IIRC without the autoregulator, the setting defaults to 60. And if you want to, you can try to unapply the patch. I find it still cleanly applies, it should still unapply from the love patch.
I haven't been paying the most attention so I don't know if it's been in for 2.6.0-final, but VM swappiness has been around for a while in the test series. I worked it in during my reign as maintainer, so if you were around for those days, you used it. |
|
| Back to top |
|
|
Manco
Tux's lil' helper
Joined: 30 Apr 2003
Posts: 115
Location: Rotterdam the Netherlands
|
| Posted: Sat Jan 03, 2004 7:04 pm Post subject: |
|
|
| GentooBox wrote: | To Love-sources maintainer:
When you release a new love-sources, then set all options in .config to =y and compile the kernel before you release it.
if any compile problems comes up, then see whats wrong, and fix it.
that will minimize compile problems for us.
you should also check rejected chunks of code then you patch the kernel. |
This is without any doubt a supertip for the maintainer, if the maintainer does this before a release of new love-sources, that could be perfect for us 
_________________
Mr. White: The choice between doing ten years and taking out some stupid motherfucker, ain't no choice at all. But I ain't no madman. |
|
| Back to top |
|
|
GentooBox
Veteran
Joined: 22 Jun 2003
Posts: 1168
Location: Denmark
|
| Posted: Sat Jan 03, 2004 7:09 pm Post subject: |
|
|
| Manco wrote: | | GentooBox wrote: | To Love-sources maintainer:
When you release a new love-sources, then set all options in .config to =y and compile the kernel before you release it.
if any compile problems comes up, then see whats wrong, and fix it.
that will minimize compile problems for us.
you should also check rejected chunks of code then you patch the kernel. |
This is without any doubt a supertip for the maintainer, if the maintainer does this before a release of new love-sources, that could be perfect for us |
Well, he should also compile all options as modules. not only in kernel.
I really hate when i get an error message when trying to compile a new kernel. - i have the reiserfs4 debug.o compile failture. and i have nearly no hair left 
_________________
Encrypt, lock up everything and duct tape the rest |
|
| Back to top |
|
|
Wedge_
Advocate
Joined: 08 Aug 2002
Posts: 3614
Location: Scotland
|
| Posted: Sat Jan 03, 2004 7:56 pm Post subject: |
|
|
Aren't you being a little harsh here? steel300 is under no obligation to perfect every release before he makes it available. Remember that he's doing most of this on his own, and that love-sources is supposed to be for people who want to be at the sharp end of kernel development - if you can't put up with the occasional problem stick with gentoo-dev-sources/mm-sources/development-sources instead.
I'd like to try Reiser4 again, but from the quick look I had it doesn't seem to be a particulary simple fix.
_________________
Per Ardua Ad Astra
The Earth is the cradle of the mind, but we cannot live forever in a cradle - Konstantin E. Tsiolkovsky
Gentoo Radeon FAQ |
|
| Back to top |
|
|
nepenthe
Guru
Joined: 14 Jul 2002
Posts: 317
Location: Sanford, ME
|
| Posted: Sat Jan 03, 2004 8:23 pm Post subject: |
|
|
| Yea give the man a break.. I say if he wants and has time perhaps he could try that route.. but instaed of him spending whatever length of time trying to make sure everything compiles... he could have released it, and let others help as well. It's a community thing... |
|
| Back to top |
|
|
StringCheesian
l33t
Joined: 21 Oct 2003
Posts: 887
|
| Posted: Sat Jan 03, 2004 8:34 pm Post subject: |
|
|
| bssteph wrote: | Let's not jump to conclusions. cat /proc/sys/vm/swappiness to see what the level currently is. The closer to 60, the less the patch is really mattering. IIRC without the autoregulator, the setting defaults to 60. And if you want to, you can try to unapply the patch. I find it still cleanly applies, it should still unapply from the love patch.
I haven't been paying the most attention so I don't know if it's been in for 2.6.0-final, but VM swappiness has been around for a while in the test series. I worked it in during my reign as maintainer, so if you were around for those days, you used it. |
You're right. It seems to only happen when something tries to play a sound. It could be ALSA or emu10k1, in that case. I've got an SB Live! . |
|
| Back to top |
|
|
Halanegri
Guru
Joined: 03 Mar 2003
Posts: 351
Location: Reykjavík, Iceland
|
| Posted: Sat Jan 03, 2004 8:43 pm Post subject: |
|
|
Still no fix for the reiser4 compilation error?  |
|
| Back to top |
|
|
merkaba
n00b
Joined: 10 Jul 2003
Posts: 56
|
| Posted: Sat Jan 03, 2004 8:51 pm Post subject: |
|
|
seems to be running fine for me after those first couple bumps.
unfortunately right now *we're* QC for a lot of this stuff.
i think if you use genkernel then it will copy /usr/src/linux/.config into /etc/kernels/config-kernel-version automagically for you.
since i've been doing make menuconfig by hand lately, i just copy the .config file myself. so when it's time for the next version just go down to the bottom of the list and choose the load alternate configuration option when you make menuconfig.
of course if you're going from 2.4 to 2.6 for the first time it'd be better to start with a fresh config. and it's always a good idea to look over the existing settings before you accept them. |
|
| Back to top |
|
|
GentooBox
Veteran
Joined: 22 Jun 2003
Posts: 1168
Location: Denmark
|
| Posted: Sat Jan 03, 2004 9:18 pm Post subject: |
|
|
| Wedge_ wrote: | Aren't you being a little harsh here? steel300 is under no obligation to perfect every release before he makes it available. Remember that he's doing most of this on his own, and that love-sources is supposed to be for people who want to be at the sharp end of kernel development - if you can't put up with the occasional problem stick with gentoo-dev-sources/mm-sources/development-sources instead.
I'd like to try Reiser4 again, but from the quick look I had it doesn't seem to be a particulary simple fix. |
I dont hold anything against Steel - i think he does a wonderfull job.
but why release a Source that is incomplete?
not only Steel, but also any other kernel programmer should test the result of their work before they release it.
_________________
Encrypt, lock up everything and duct tape the rest |
|
| Back to top |
|
|
Wedge_
Advocate
Joined: 08 Aug 2002
Posts: 3614
Location: Scotland
|
| Posted: Sat Jan 03, 2004 9:30 pm Post subject: |
|
|
| Halanegri wrote: | | Still no fix for the reiser4 compilation error? |
I'm trying to fix it just now, I'll post back if I get anywhere.
Edit: It looks like there are several problems with it, not just one. For now, if you really want to try Reiser4, then grab all.diff.gz from here and patch 2.6.0, that should work fine.
_________________
Per Ardua Ad Astra
The Earth is the cradle of the mind, but we cannot live forever in a cradle - Konstantin E. Tsiolkovsky
Gentoo Radeon FAQ |
|
| Back to top |
|
|
merkaba
n00b
Joined: 10 Jul 2003
Posts: 56
|
| Posted: Sat Jan 03, 2004 9:42 pm Post subject: |
|
|
| GentooBox wrote: |
I dont hold anything against Steel - i think he does a wonderfull job.
but why release a Source that is incomplete?
not only Steel, but also any other kernel programmer should test the result of their work before they release it. |
i think steel is doing a great job so far, but there's always room for improvement. that's probably why this other thread was started. unfortunately, it's started to degrade into another "what should we call it" topic. i'd hoped to see more ideas about improving the quality of these releases.
personally, i'd like to see something like a website with its own phpbb where the initial cowboy release and testing could be done to round out these rough edges before a (possibly time-based) release would be anounced here. |
|
| Back to top |
|
|
nepenthe
Guru
Joined: 14 Jul 2002
Posts: 317
Location: Sanford, ME
|
| Posted: Sat Jan 03, 2004 10:09 pm Post subject: |
|
|
| GentooBox wrote: | | Wedge_ wrote: | Aren't you being a little harsh here? steel300 is under no obligation to perfect every release before he makes it available. Remember that he's doing most of this on his own, and that love-sources is supposed to be for people who want to be at the sharp end of kernel development - if you can't put up with the occasional problem stick with gentoo-dev-sources/mm-sources/development-sources instead.
I'd like to try Reiser4 again, but from the quick look I had it doesn't seem to be a particulary simple fix. |
I dont hold anything against Steel - i think he does a wonderfull job.
but why release a Source that is incomplete
not only Steel, but also any other kernel programmer should test the result of their work before they release it. |
so you would rather he not release anything? steel is the maintainer...
I bet alot of people would rather he releases as is, as opposed to waiting... |
|
| Back to top |
|
|
nepenthe
Guru
Joined: 14 Jul 2002
Posts: 317
Location: Sanford, ME
|
| Posted: Sat Jan 03, 2004 10:10 pm Post subject: |
|
|
| merkaba wrote: | | GentooBox wrote: |
I dont hold anything against Steel - i think he does a wonderfull job.
but why release a Source that is incomplete?
not only Steel, but also any other kernel programmer should test the result of their work before they release it. |
i think steel is doing a great job so far, but there's always room for improvement. that's probably why this other thread was started. unfortunately, it's started to degrade into another "what should we call it" topic. i'd hoped to see more ideas about improving the quality of these releases.
personally, i'd like to see something like a website with its own phpbb where the initial cowboy release and testing could be done to round out these rough edges before a (possibly time-based) release would be anounced here. |
I know a fellow love-sources users has registerd love-sources.org |
|
| Back to top |
|
|
MrF
n00b
Joined: 28 Aug 2002
Posts: 73
Location: Auckland, NZ
|
| Posted: Sat Jan 03, 2004 10:28 pm Post subject: |
|
|
| This new version fails to boot for me. It locks hard at "bringing up eth0", I have a Intel pro 100/1000 (e1000?). Previously I was running 2.6.0-love1. hmmm..... |
|
| Back to top |
|
|
StringCheesian
l33t
Joined: 21 Oct 2003
Posts: 887
|
| Posted: Sun Jan 04, 2004 12:01 am Post subject: |
|
|
| GentooBox wrote: | When you release a new love-sources, then set all options in .config to =y and compile the kernel before you release it.
...
I dont hold anything against Steel - i think he does a wonderfull job.
but why release a Source that is incomplete? |
Well, volunteering and colaboration is what free/open source is all about, right? **hint, hint **. It was your idea, ya know.
And I disagree that steel should have to do any extensive testing. I downloaded love-sources to try out and test a kernel that I knew full well to be experimental and untested. I mean, testing/trying it out it is what all of us came here for, right?
I would be happy to volunteer to compile love-sources after each release with everything selected and report my results, but I doubt I'm knowledgable enough to be a valuable tester (especially if my terrible diagnosis in my first post above is any indication  ). |
|
| Back to top |
|
|
bssteph
l33t
Joined: 26 Feb 2003
Posts: 652
Location: Wisconsin
|
| Posted: Sun Jan 04, 2004 1:21 am Post subject: |
|
|
| StringCheesian wrote: | | bssteph wrote: | Let's not jump to conclusions. cat /proc/sys/vm/swappiness to see what the level currently is. The closer to 60, the less the patch is really mattering. IIRC without the autoregulator, the setting defaults to 60. And if you want to, you can try to unapply the patch. I find it still cleanly applies, it should still unapply from the love patch.
I haven't been paying the most attention so I don't know if it's been in for 2.6.0-final, but VM swappiness has been around for a while in the test series. I worked it in during my reign as maintainer, so if you were around for those days, you used it. |
You're right. It seems to only happen when something tries to play a sound. It could be ALSA or emu10k1, in that case. I've got an SB Live! . |
I have a emu10k1-based chip (Audigy) and with my 2.6.0-mm2-based patchset* I haven't had any problems. I've been meaning to try out the latest love, but right now all my machines are busy with world updates. It's possible that some badness unrelated to love was added since 2.6.0.
* Yeah, I've been rolling my own patches... just a little thing to get what I want in. |
|
| Back to top |
|
|
supernovus
Apprentice
Joined: 13 Jul 2003
Posts: 150
Location: inside my head
|
| Posted: Sun Jan 04, 2004 5:16 am Post subject: |
|
|
Well I applied the patch provided to fix the bootsplash error, and it got a lot farther through, and then near the end died with this error:
| Code: |
GEN .version
CHK include/linux/compile.h
UPD include/linux/compile.h
CC init/version.o
LD init/built-in.o
LD .tmp_vmlinux1
drivers/built-in.o(.text+0x950b9): In function `splash_getraw':
: undefined reference to `con2fb_map'
drivers/built-in.o(.text+0x958d2): In function `splash_verbose':
: undefined reference to `con2fb_map'
drivers/built-in.o(.text+0x95de0): In function `splash_status':
: undefined reference to `con2fb_map'
drivers/built-in.o(.text+0x95f10): In function `splash_read_proc':
: undefined reference to `con2fb_map'
drivers/built-in.o(.text+0x960f7): In function `splash_write_proc':
: undefined reference to `con2fb_map'
make: *** [.tmp_vmlinux1] Error 1
|
Is anyone else having this problem, or is it just me?
EDIT: Never mind, I fiddled with the configuration a bit and it compiled cleanly. Except for one strange non-fatal is seems error, anyone have any clue what this means:
| Code: |
AS arch/i386/boot/setup.o
arch/i386/boot/setup.S: Assembler messages:
arch/i386/boot/setup.S:165: Warning: value 0x37ffffff truncated to 0x37ffffff
LD arch/i386/boot/setup
|
It doesn't appear to stop anything from working, so I'm guessing it's nothing too serious.
_________________
Remove OTW
Last edited by supernovus on Sun Jan 04, 2004 5:27 am; edited 1 time in total |
|
| Back to top |
|
|
steel300
Veteran
Joined: 10 Jul 2003
Posts: 1155
|
| Posted: Sun Jan 04, 2004 5:20 am Post subject: |
|
|
I believe that it's time for an update. After almost getting arrested yesterday trying to get the latest love-sources all fixed up, I'm spending a sleepless night making everything work. Here are my notes so far:
| Quote: | Patched 2.6.0 to 2.6.1-rc1
Testing: Patch applied cleanly and compiles cleanly
Description: Latest prepatch update to the 2.6.0 kernel
Patched 2.6.1-rc1 to 2.6.1-rc1-mm1
Testing: Patch applied cleanly and compiles cleanly
Description: Newest mm patches, many bug fixes
Applied patch-2.6.0-am-5 (Con's VM Swappiness)
Testing: Patch applied cleanly and compiles cleanly
Description: Con Koliva's VM swappiness patch - system loads more
programs into memory to increase performance under load
Applied elv-select
Testing: Patch applied cleanly and compiles cleanly
Description: Nick Piggin's runtime selectable I/O scheduler
Applied radeon-igp-patch
Testing: Patch applied with a few errors, but compiles cleanly, staying
in for now, unless someone else determine's that it's faulty
Description: igp fix for ATI's radeon drm drivers
Applied latest alsa-patches (as of 2003-12-30)
Testing: Patch applied cleanly and compiles cleanly
Description: Latest alsa snapshot, fixes many minor inconveniences,
namely the emu10k1 patches
Applied bootsplash patches
Testing: Patch failed in fb.h, tried to compile anyway, compiling
failed, patched fb.h by hand and trying to compile again, compiling
again, worked
Description: If you don't know what bootsplash is, use the search
feature of phpBB
Applied nvidia-agp patch
Testing: Patch applied cleanly and compiles cleanly
Description: Fixes some problems with EPoX 8rda(+) boards, ati
drivers, and agp
Applied orinoco monitor mode patch
Testing: Patch applied cleanly and compiles cleanly
Description: Enable monitor mode on orinoco-hermes wireless card
chipsets
|
I still have to get reiser4 working, udf packet writing, and finish appliying the patches that apply and compile, but I just haven't gotten around to them yet (vesa_fb hack, etc..).
Also, does anyon use a seagate sata drive? If so, I'll throw the siimage patch back in to speed up disk access.
_________________
Rationality is the recognition of the fact that nothing can alter the truth and nothing can take precedence over that act of perceiving it. |
|
| Back to top |
|
|
taskara
Advocate
Joined: 10 Apr 2002
Posts: 3763
Location: Australia
|
| Posted: Sun Jan 04, 2004 5:47 am Post subject: |
|
|
| steel300 wrote: | | Also, does anyon use a seagate sata drive? If so, I'll throw the siimage patch back in to speed up disk access. |
oh yes please!
also it will also have the latest siliconimage-patch-fix won't it? because I think this is now included in 2.6.0-mm1 mm-sources
_________________
Kororaa install method - have Gentoo up and running quickly and easily, fully automated with an installer!
Last edited by taskara on Sun Jan 04, 2004 5:48 am; edited 1 time in total |
|
| Back to top |
|
|
steel300
Veteran
Joined: 10 Jul 2003
Posts: 1155
|
| Posted: Sun Jan 04, 2004 5:47 am Post subject: |
|
|
[quote=supernovus] | Code: |
AS arch/i386/boot/setup.o
arch/i386/boot/setup.S: Assembler messages:
arch/i386/boot/setup.S:165: Warning: value 0x37ffffff truncated to 0x37ffffff
LD arch/i386/boot/setup |
[/quote]
The value that was originally there (0x37ffffffff) was too large to fit in the memory space allocated to it, so the compiler cut off the trailing pieces to make it fit. It's not fatal. In this situation it won't make a difference.
_________________
Rationality is the recognition of the fact that nothing can alter the truth and nothing can take precedence over that act of perceiving it. |
|
| Back to top |
|
|
steel300
Veteran
Joined: 10 Jul 2003
Posts: 1155
|
| Posted: Sun Jan 04, 2004 5:49 am Post subject: |
|
|
| taskara wrote: | | steel300 wrote: | | Also, does anyon use a seagate sata drive? If so, I'll throw the siimage patch back in to speed up disk access. |
oh yes please!
also it will also have the latest siliconimage-patch-fix won't it? because I think this is now included in 2.6.0-mm1 mm-sources |
That's the patch I'm talking about. Give me bit to get everything else working, then that's going in.
_________________
Rationality is the recognition of the fact that nothing can alter the truth and nothing can take precedence over that act of perceiving it. |
|
| Back to top |
|
|
taskara
Advocate
Joined: 10 Apr 2002
Posts: 3763
Location: Australia
|
| Posted: Sun Jan 04, 2004 5:58 am Post subject: |
|
|
| steel300 wrote: | | taskara wrote: | | steel300 wrote: | | Also, does anyon use a seagate sata drive? If so, I'll throw the siimage patch back in to speed up disk access. |
oh yes please!
also it will also have the latest siliconimage-patch-fix won't it? because I think this is now included in 2.6.0-mm1 mm-sources |
That's the patch I'm talking about. Give me bit to get everything else working, then that's going in. |
ahh ok.. I think there are a number of patches..
that one I linked fixes the slow and freezing issues
then there are these:
ide-siimage-seagate.patch
ide-siimage-stack-fix.patch
ide-siimage-sil3114.patch
all of which I think are already included in mm-sources
- not sure ov the siimage driver version tho.. the latest is 1.09 (20th Dec 2003)
I guess part of using mm-sources as a base, we need to know exactly what patches they are already using
good luck with reiser4
cheers
_________________
Kororaa install method - have Gentoo up and running quickly and easily, fully automated with an installer! |
|
| Back to top |
|
|
supernovus
Apprentice
Joined: 13 Jul 2003
Posts: 150
Location: inside my head
|
| Posted: Sun Jan 04, 2004 6:22 am Post subject: |
|
|
Well I just want to say, while I haven't gotten the silent splash image working yet, I have everything else working just great, and must say this is the fastest performing love-sources yet, keep up the good work!
_________________
Remove OTW |
|
| Back to top |
|
|
dabooty
Guru
Joined: 15 May 2003
Posts: 482
Location: Belgium
|
| Posted: Sun Jan 04, 2004 11:18 am Post subject: |
|
|
crap, compilation fails on packet writing.
Might try this out again in a few days
_________________
registered user #284425
get yourself counted
http://counter.li.org
------
#emerge -pv solves a lot of questions beforehand |
|
| Back to top |
|
|
|
Kernel & Hardware
