| View previous topic :: View next topic |
| Author |
Message |
NathanZachary
Moderator
Joined: 30 Jan 2007
Posts: 2609
|
 Posted: Sat Mar 19, 2011 1:20 pm Post subject: |
 |
|
Strange. I'm performing a new installation, and can't get past udev in my first system update. However, it looks like the bug has been fixed with 164-r2:
https://bugs.gentoo.org/show_bug.cgi?id=359407
I'll try it now. 
_________________
“Truth, like infinity, is to be forever approached but never reached.” --Jean Ayres (1972)
---avatar cropped from =AimanStudio--- |
|
| Back to top |
|
 |
NathanZachary
Moderator
Joined: 30 Jan 2007
Posts: 2609
|
| Posted: Sat Mar 19, 2011 1:28 pm Post subject: |
|
|
In case anyone is interested in the actual portion that solved the udev problem, I mention it in the other thread about it.
_________________
“Truth, like infinity, is to be forever approached but never reached.” --Jean Ayres (1972)
---avatar cropped from =AimanStudio--- |
|
| Back to top |
|
|
VinzC
Watchman
Joined: 17 Apr 2004
Posts: 5098
Location: Dark side of the mood
|
| Posted: Sat Mar 19, 2011 1:34 pm Post subject: |
|
|
| NathanZachary wrote: | Strange. I'm performing a new installation, and can't get past udev in my first system update. However, it looks like the bug has been fixed with 164-r2:
https://bugs.gentoo.org/show_bug.cgi?id=359407
I'll try it now. |
Aaah, that explains why. On my system nothing exists, which depends on V4L1 so I guess that's why udev compiles fine.
_________________
Gentoo addict: tomorrow I quit, I promise!... Just one more emerge...
1739! |
|
| Back to top |
|
|
Aquous
l33t
Joined: 08 Jan 2011
Posts: 700
|
| Posted: Sat Mar 19, 2011 3:27 pm Post subject: |
|
|
Is anyone else owning a Radeon GPU experiencing a strange flicker when switching from a TTY to X?
Apart from that, this kernel rocks. I've been running it for less than ten minutes and already my desktop feels smoother somehow.  |
|
| Back to top |
|
|
gorkypl
Guru
Joined: 04 Oct 2010
Posts: 444
Location: Kraków, PL
|
| Posted: Sat Mar 19, 2011 3:38 pm Post subject: |
|
|
| Aquous wrote: | | Is anyone else owning a Radeon GPU experiencing a strange flicker when switching from a TTY to X? |
no - everything OK here with 6.14.1 open drivers and 2.6.38 kernel |
|
| Back to top |
|
|
Aquous
l33t
Joined: 08 Jan 2011
Posts: 700
|
| Posted: Sat Mar 19, 2011 4:00 pm Post subject: |
|
|
Weird. I see my TTY flash through my X screen for like a fifth of a second on my HD 5450 (mesa & libdrm from ~amd64, DDX from amd64).
Ah well. It's no problem at all, really, just a bit odd. This kernel (still) rocks. |
|
| Back to top |
|
|
Gusar
Advocate
Joined: 09 Apr 2005
Posts: 2665
Location: Slovenia
|
| Posted: Sat Mar 19, 2011 5:07 pm Post subject: |
|
|
| VinzC wrote: | | Finally, even if this patch might happen to prove useless for a majority of users, it doubtlessly comes handy with Gentoo just because compiling is a common task. No need to renice, just compile as much as you want and watch. That's the point. |
But you yourself said this could already be done before, by just setting a sane -j. So where is the miracleness of this patch? Set a sane -j and possibly use BFS and there's no need for any "miracles". |
|
| Back to top |
|
|
VinzC
Watchman
Joined: 17 Apr 2004
Posts: 5098
Location: Dark side of the mood
|
| Posted: Sat Mar 19, 2011 7:13 pm Post subject: |
|
|
| VinzC wrote: | | Finally, even if this patch might happen to prove useless for a majority of users, it doubtlessly comes handy with Gentoo just because compiling is a common task. No need to renice, just compile as much as you want and watch. That's the point. |
| Gusar wrote: | | But you yourself said this could already be done before, by just setting a sane -j. So where is the miracleness of this patch? Set a sane -j and possibly use BFS and there's no need for any "miracles". |
One difference is at which value of -j my system stops being responsive. The purpose of the patch is to demonstrate how a system can still be responsive under an uncommonly heavy load and not only compiling. Before that my system would have been unresponsive with a low -j value (but still higher than 3), even without compiling hugin. Now with even -j16 it remains usable... except while compiling hugin. Another difference is now you don't have to *do* anything.
_________________
Gentoo addict: tomorrow I quit, I promise!... Just one more emerge...
1739! |
|
| Back to top |
|
|
Gusar
Advocate
Joined: 09 Apr 2005
Posts: 2665
Location: Slovenia
|
| Posted: Sat Mar 19, 2011 8:31 pm Post subject: |
|
|
| Quote: | | he purpose of the patch is to demonstrate how a system can still be responsive under an uncommonly heavy load and not only compiling. |
High -j is not "unusually high load", it's a ridiculous artificial scenario.
| Quote: | | Another difference is now you don't have to *do* anything. |
Before you followed documentation that told you how to set -j (number of cores plus one). I don't see how things are different now. In fact, you now had to specifically create an artificial scenario to show that the patch does something. |
|
| Back to top |
|
|
PaulBredbury
Watchman
Joined: 14 Jul 2005
Posts: 7310
|
| Posted: Sat Mar 19, 2011 9:13 pm Post subject: |
|
|
| ulatencyd looks like it could be interesting, talking about "automatic" scheduling. |
|
| Back to top |
|
|
agent_jdh
Veteran
Joined: 08 Aug 2002
Posts: 1783
Location: Scotland
|
| Posted: Sat Mar 19, 2011 9:19 pm Post subject: |
|
|
Just to add my 2c here ... on VinzC's side - noticed a considerable responsiveness improvement with 2.6.38 over pf-sources-2.6.37 (using bfs etc) while backing up files from my fileserver onto an external eSata (NTFS) hard drive connected to my desktop. I have typically 3 Firefox windows open with say a dozen tabs in each, and with pf-sources, Firefox was _much_ less responsive while the copy was happening. Using kernel 2.6.38, with essentially the same config (no bfs obviously) but with CONFIG_SCHED_AUTOGROUP enabled, Firefox was much more responsive during the same copy procedure, in fact it behaved pretty much like Firefox was the only thing running.
This is on a Core i5 760 so it is not a slow machine. The culprit of the high cpu usage during the actual copy would appear to be the ntfs3g driver.
I'll be interested to see a 2.6.38-based pf-sources kernel.
_________________
Jingle Jangle Jewellery |
|
| Back to top |
|
|
PaulBredbury
Watchman
Joined: 14 Jul 2005
Posts: 7310
|
| Posted: Sat Mar 19, 2011 9:23 pm Post subject: |
|
|
| agent_jdh wrote: | | Firefox was _much_ less responsive while the copy was happening |
Use the ionice command, as in my examples above. |
|
| Back to top |
|
|
agent_jdh
Veteran
Joined: 08 Aug 2002
Posts: 1783
Location: Scotland
|
| Posted: Sat Mar 19, 2011 9:55 pm Post subject: |
|
|
| PaulBredbury wrote: | | agent_jdh wrote: | | Firefox was _much_ less responsive while the copy was happening |
Use the ionice command, as in my examples above. |
Yeah, but now I don't have to. That has to be progress, right?
_________________
Jingle Jangle Jewellery |
|
| Back to top |
|
|
Gusar
Advocate
Joined: 09 Apr 2005
Posts: 2665
Location: Slovenia
|
| Posted: Sat Mar 19, 2011 10:02 pm Post subject: |
|
|
| Just one thing I'm curious about... Have you tried a 2.6.38 kernel but without autogroup? To make sure it's really the autogroup thing that brings the improvement and not some other goodies the 38 kernel has. |
|
| Back to top |
|
|
Anon-E-moose
Watchman
Joined: 23 May 2008
Posts: 6222
Location: Dallas area
|
| Posted: Sat Mar 19, 2011 10:03 pm Post subject: |
|
|
Use whatever works and makes you happy. 
_________________
UM780, 6.12 zen kernel, gcc 13, openrc, wayland |
|
| Back to top |
|
|
VinzC
Watchman
Joined: 17 Apr 2004
Posts: 5098
Location: Dark side of the mood
|
| Posted: Sun Mar 20, 2011 12:20 am Post subject: |
|
|
| Gusar wrote: | | High -j is not "unusually high load", it's a ridiculous artificial scenario. |
Just remove «unusual» then  . Now you don't seem convinced (kay, I got it) but even Linus Torvalds saw a big improvement. I just confirmed his findings... if they still have to be confirmed!
_________________
Gentoo addict: tomorrow I quit, I promise!... Just one more emerge...
1739! |
|
| Back to top |
|
|
jormartr
Apprentice
Joined: 02 Jan 2008
Posts: 174
|
| Posted: Sun Mar 20, 2011 9:56 pm Post subject: |
|
|
| jcTux wrote: | | NathanZachary wrote: | It seems that there is a problem with 2.6.38 though. udev won't build with it:
It seems that the problem is with videodev.h being removed from 2.6.38:
http://comments.gmane.org/gmane.linux.hotplug.devel/16670
The error I get on my new installation is:
| Code: |
extras/v4l_id/v4l_id.c:31:28: error: linux/videodev.h: No such file or directory
|
There are other errors, but it fails to build (with or without the 'extra' USE flag). 
Anyone have suggestions? |
Udev compiles fine here | Code: | | USE="extras -devfs-compat -old-hd-rules (-selinux) -test" |
| Code: | uname -r
2.6.38-gentoo |
|
I have just had the same problem, do you have sys-kernel/kernel-headers unmasked to its last version? It worked for me, disabling the unmask, and using the standard version (also, after, i downgraded kernel to 2.6.36 stable). |
|
| Back to top |
|
|
VoidMage
Watchman
Joined: 14 Oct 2006
Posts: 6196
|
| Posted: Sun Mar 20, 2011 10:05 pm Post subject: |
|
|
| There's already a revision of udev with a fix in the tree. |
|
| Back to top |
|
|
PaulBredbury
Watchman
Joined: 14 Jul 2005
Posts: 7310
|
| Posted: Wed Mar 23, 2011 1:24 pm Post subject: |
|
|
Further evidence that 2.6.38 is fast - "transparent huge pages".
Of course, 2.6.38+BFS will be even better  |
|
| Back to top |
|
|
VinzC
Watchman
Joined: 17 Apr 2004
Posts: 5098
Location: Dark side of the mood
|
| Posted: Wed Mar 23, 2011 8:47 pm Post subject: |
|
|
| PaulBredbury wrote: | Further evidence that 2.6.38 is fast - "transparent huge pages".
Of course, 2.6.38+BFS will be even better |
I guess you mean Zen Sources? I can't wait to check it. I was using it until 2.6.35 or 2.6.36 after which I could no longer see it in portage .
OTOH I've had to revert to 2.6.37 temporarily for I've had a couple of kernel Oopses while the screensaver was active. The issue always occured while I was cancelling the screen saver or stopped it by moving my mouse or pressing a key. I don't know if that's because of the video driver, the video BIOS or because I'm using GL screensavers or anything else. Maybe I should enable crashdumps...
_________________
Gentoo addict: tomorrow I quit, I promise!... Just one more emerge...
1739!
Last edited by VinzC on Thu Mar 24, 2011 7:18 am; edited 1 time in total |
|
| Back to top |
|
|
cach0rr0
Bodhisattva
Joined: 13 Nov 2008
Posts: 4123
Location: Houston, Republic of Texas
|
| Posted: Wed Mar 23, 2011 9:01 pm Post subject: |
|
|
| VinzC wrote: | | PaulBredbury wrote: | Further evidence that 2.6.38 is fast - "transparent huge pages".
Of course, 2.6.38+BFS will be even better |
I guess you mean Zen Sources? I can't wait to check it. |
zen or ck-sources, either one once it hits .38
kernelOfTruth actually has a patchset he seems to be maintaining him, possible he includes BFS as well (I haven't looked)
_________________
Lost configuring your system?
dump lspci -n here | see Pappy's guide | Link Stash |
|
| Back to top |
|
|
Anon-E-moose
Watchman
Joined: 23 May 2008
Posts: 6222
Location: Dallas area
|
| Posted: Wed Mar 23, 2011 9:14 pm Post subject: |
|
|
Con has stated that he hasn't started working on BFS for 2.6.38 yet, (it came out quicker than he expected) and he was busy on some other project, but it sounds like he will be looking at porting it soon.
_________________
UM780, 6.12 zen kernel, gcc 13, openrc, wayland |
|
| Back to top |
|
|
PaulBredbury
Watchman
Joined: 14 Jul 2005
Posts: 7310
|
| Posted: Wed Mar 23, 2011 9:26 pm Post subject: |
|
|
| I use pf kernel patchset, for its convenience. |
|
| Back to top |
|
|
VinzC
Watchman
Joined: 17 Apr 2004
Posts: 5098
Location: Dark side of the mood
|
| Posted: Thu Mar 24, 2011 7:20 am Post subject: |
|
|
Thanks. Will try it.
_________________
Gentoo addict: tomorrow I quit, I promise!... Just one more emerge...
1739! |
|
| Back to top |
|
|
Yamakuzure
Advocate
Joined: 21 Jun 2006
Posts: 2305
Location: Adendorf, Germany
|
| Posted: Thu Mar 24, 2011 12:13 pm Post subject: |
|
|
| Gusar wrote: | | Quote: | | he purpose of the patch is to demonstrate how a system can still be responsive under an uncommonly heavy load and not only compiling. |
High -j is not "unusually high load", it's a ridiculous artificial scenario. |
No and no. Sorry, but on a quad-core cpu the recommended value is -j9 with which my laptop became completely unusable before the "cgroups-console-hack" was invented whenever I emerged packages with --jobs in the EMERGE_DEFAULT_OPTS. With the console-hack (now superseded by 2.6.38 with auto scheduler) I can keep on working like there was no load. In the meantime portage did a world update, vmware is running with windows xp (needed for a cisco vpn tunnel to a customer), several openoffice documents are open, kmail, knode, amarok is playing music and typing this text goes without problem. Load of my system: 18.5 to 25.0 -- unthinkable before those cgroups methods came up.
And this is not a "ridiculous artificial scenario", it is something I have twice a week. You know, I _do_ want to get the updates finished as soon as possible without having to go for a walk for hours because I can't use my machine...
But I do have a problem with gentoo-sources-2.6.38 : Although the load distribution works, all programs I start need 5 to 10 times longer than before the kernel upgrade. How can this be?
_________________
Edited 220,176 times by Yamakuzure |
|
| Back to top |
|
|
|
Kernel & Hardware
