/srv/irclogs.ubuntu.com/2006/03/05/#ubuntu-kernel.txt

=== psusi [n=phreak@54.161.205.68.cfl.res.rr.com] has joined #ubuntu-kernel
=== mgalvin [n=mgalvin@ubuntu/member/mgalvin] has joined #ubuntu-kernel
=== mgalvin [n=mgalvin@ubuntu/member/mgalvin] has joined #ubuntu-kernel
bluefoxicydoes anyone think that the Linux kernel mainline should have built-in memory compression for memory stress?01:27
bluefoxicyI did some test with Rodrigo's patch, tweaked it to be able to compress in larger blocks01:28
bluefoxicyusing the WK4x4 and WKdm compression algorithms (lz0 I didn't like, it corrupted memory on occasion) I got 20% or so at 8k blocks01:29
bluefoxicybut with 16k blocks I averaged 35%, and with 32 I got 40%.  Beyond there, there wasn't much gain.01:29
=== mgalvin [n=mgalvin@ubuntu/member/mgalvin] has joined #ubuntu-kernel
airliedjust buy more RAM.. it's cheap...01:34
airliedor fix apps to not chew it all up..01:34
airliedI don't think the gain of in-memory compression is that majorly useful for most people..01:34
=== airlied remembers win95 tools to do it..
bluefoxicyairlied:  in-memory compression can get up to 40% memory gains on the compressed area.  If you allow typically 25% of memory to be in-memory compressed, this may be significant.01:38
airliedbut what about the process using that memory?01:40
bluefoxicyFor example, at 256M this allows 299M memory to be used; at 512M this allows 597M to be used; and at 1 gig this allows 1194M to be used.  This is an average of 170M per gigabyte.01:40
airliedit doesn't really...01:40
bluefoxicythe process using that memory would hit an invalid pagefault trying to access it01:40
bluefoxicywhich would trigger decompression.01:40
airliedfi you are under memory pressure you are going to run out, unless you fix the problem..01:40
bluefoxicyThis happens when the linux kernel swaps memory to disk as well; execpt that the the memory is retrieved from disk instead of decompressed.01:41
bluefoxicyYou won't necessarily run out under memory pressure01:41
bluefoxicyLinux is ALWAYS under memory pressure; disk cache typically uses whatever programs don't01:41
airliedI'm just not seeing the advantages outweighing the complexity of it..01:41
bluefoxicyIf you have ample memory already, then the system just becomes faster, as it can simply compress disk cache until memory pressure increases enough to dump it.01:42
bluefoxicyI'm currently thinking of liveCDs, although database applications may benefit largely as well.01:42
bluefoxicyLiveCDs may run on systems without swap.  While you may say it is "nice" to try to use less than 25% for memory compression, without swap memory pressure suddenly results in one of two things:  OOM kills or eating more memory for compressed cache01:43
airliedbut if I'm running a firefox and an openoffice and I've got no RAM left I don't think I'll be able to do much compression ...01:43
bluefoxicyA system with 256M of memory runs somewhat slow on a LiveCD.  There is a LOT of reading from the CD, because disk cache is constantly invalidated.01:43
bluefoxicyairlied:  remember I'm talking about compressing 32K blocks.  The WK4x4 algorithm in Rodrigo CAstro's patch used about 128K of memory for this (it errantly over-allocated 300k, which it didn't use)01:44
bluefoxicyat any rate, given severe pressure (growing disk cache, growing program memory) and no swap, a livecd system may compensate by compressing memory and using up to 90% of memory for a compressed store.  This may slow programs down by perhaps 10% between pagefaults and decompression routines (remember we're spending a fraction of a millisecond decompressing a compressed 32k block or vice versa)01:47
airliedwell I'll remain sceptical that the good outweights the bad... and I doubt it'll turn up in mainline any time soon..01:47
bluefoxicyhowever the apparent memory capacity of such a system becomes 410MB01:47
bluefoxicydisk access is much, much longer than the compression, decompression, and spaztic pagefaults until you use a lot of memory for a compressed store.  I don't like anything over 50%, heh.01:48
airliedbut if an apps working set is larger than 10% you'll end up compress trashing it..01:48
bluefoxicyyes, you would01:48
bluefoxicybut if an app's working set is larger than 10%, you're risking OOM kills.01:48
bluefoxicywell, maybe not01:49
bluefoxicythat would be 25 megs on a 256 meg system01:49
airliedtrue... I can see it being rejected though as RAM is cheap..01:49
bluefoxicyHowever, if an application is working with a lot of libraries (openOffice, gnome applets, firefox.. . . these feel this a lot), then every time you do something your CD drive spins01:49
bluefoxicywhich means you wait about 5-10 seconds, instead of a few miliseconds01:50
bluefoxicyon the subject of database systems, many of these would like to mmap() large areas of the database into memory01:50
bluefoxicyas your server may have 8GB of memory and a 256GB database (there are multi-TB databases on 64GB of RAM, btw), you could feasibly map in 256GB of the database (64-bit system will do this, you have 256TiB of VM space and the kernel gives access to 87TiB)01:51
bluefoxicybut only 8GB could be resident at once, minus other stuff in memory.01:51
bluefoxicydatabase servers such as this could feasibly use half of memory max as a compressed store, since most of that would become disk cache for the database anyway; they'd have about 10GiB (2GiB extra) of effective ram01:52
bluefoxicyRAM may be cheap, but CPU clocks are even cheaper01:52
bluefoxicyairlied:  ram is cheap yeah, it's nice to get something for free though ;)  It's been nice talking though, just felt like running through it to see if anyone actually cared.01:54
airliedbluefoxicy: yes it's an intersting idea.. just not sure it'll get mainlined :-)02:02
=== dolson [n=dana@d235-185-252.home1.cgocable.net] has joined #ubuntu-kernel
dolsonif I am compiling a customized dapper kernel that I want to be the default over any of the ubuntu standard ones, what should I put in the EXTRAVERSION field or kpkg parameters to do this?02:04
cjbI don't think what you put in EXTRAVERSION will affect where it goes in the grub menu.  If you want it to be the grub default, make it that in /boot/grub/menu.lst.02:06
dolsonwell then when a new kernel comes out, it takes over as the default02:07
=== BenC [n=bcollins@debian/developer/bcollins] has joined #ubuntu-kernel
dolsonBenC: hey man, would you answer a question for me? you said you would in your emails to me :)02:22
BenCsure02:22
dolsoncool. ok, I've got a version of Ingo's patch that applies to Dapper's kernel, and I want to compile my kernel and have it as the default kernel at bootup, even if a new Ubuntu kernel comes out. how do I do this? is it in EXTRAVERSION, make-kpkg, or in menuconfig?02:23
BenCwell, building requires make-kpkg02:24
BenCbut whether or not it is the default depends on your boot loader02:24
BenCif it's grub, then edit /boot/grub/menu.lst appropriately02:25
dolsonok, but then if there's a kernel update, it takes priority and becomes the default02:25
BenCnot if you setup menu.lst correctly02:27
BenClook at the default config option, and read what it says about saveddefault02:29
dolsonalright. I was hoping to not have to add a section about configuring grub to my wiki, but I guess that's the only way02:30
infinitydolson: No one's stopping you from removing the Ubuntu kernels completely once yours is installed.02:35
dolsonI know, that but I'm not going to advise that in a tutorial02:35
BenCthe only other way, maybe, is to set the extra version to something like -99-foo02:43
BenC--append_version for make-kpkg02:44
BenCbut I can't gurantee that we wont hit -104-* before dapper is released :)02:44
dolsonheh.02:44
dolsonis there a need to change the localversion (in menuconfig) or EXTRAVERSION in the Makefile, or is just --append_version enough02:45
BenC--append_version works02:52
dolsonalright thanks Ben02:55
=== dolson [n=dana@d235-185-252.home1.cgocable.net] has left #ubuntu-kernel ["leaving..]
mgalvinBenC: has the fix for #28660 been uploaded yet b/c it does not appear to be fixed on the current (2.6.15-16-686) kernel I have installed?04:18
=== mxpxpod [n=bryan@unaffiliated/mxpxpod] has joined #ubuntu-kernel
=== karlheg [n=karlheg@131.252.226.60] has joined #ubuntu-kernel
=== karlheg [n=karlheg@131.252.226.60] has left #ubuntu-kernel ["Leaving"]
=== BenC [n=bcollins@debian/developer/bcollins] has joined #ubuntu-kernel
=== HAL [n=HAL@CPE-144-131-214-239.qld.bigpond.net.au] has joined #ubuntu-kernel
HALanyone here06:13
HALneed a little assistance with kernel stuff06:14
HALQ1: what is the command to report kernel version06:14
fabbioneuname -a06:15
HALcool06:15
fabbionebut you should really ask stuff like this in #ubuntu :)06:15
HALtried in there they all seam to be having there own little discussions06:16
HALthanks for that06:16
fabbionewelcome06:17
bluefoxicyholy shit that scared me06:23
bluefoxicyfor a second i thought I had an e-mail from Richard Stallman and I was like "Oh no not this moron"06:23
bluefoxicysleep06:24
=== HAL [n=HAL@CPE-144-131-214-239.qld.bigpond.net.au] has left #ubuntu-kernel ["Leaving"]
=== ispiked_ [n=ispiked@Lee-11-133.rh.ncsu.edu] has joined #ubuntu-kernel
=== JaneW [n=JaneW@196.36.161.235] has joined #ubuntu-kernel
=== chmj [n=chmj@196.44.1.98] has joined #ubuntu-kernel
=== Keybuk [n=scott@c51008D91.inet.catch.no] has joined #ubuntu-kernel
=== CataEnry [n=Enrico@host83-39.pool8248.interbusiness.it] has joined #ubuntu-kernel
=== johnm [n=johnm@gentoo/developer/johnm] has joined #ubuntu-kernel
=== raphink [n=raphink@ubuntu/member/raphink] has joined #ubuntu-kernel
raphinkhi guys12:02
raphinkI'm currently working on a minor bug on update-grub12:02
raphinkand would like some comments on how to fix it12:02
fabbioneraphink: i think zul is working on grub, but he is not around right now. he is on .ca time12:03
raphinkok12:03
raphinkfabbione: do you know of update-grub ?12:03
raphinkI think you might be able to help me the issue is pretty simple12:04
raphinkupdate-grub detects existing ^splashimage= lines12:04
fabbioneraphink: sorry i am busy and i don't have time for it :/12:04
fabbioneyou will have to wait for zul12:04
raphinkand add them to the buffer12:04
raphinkah ok12:04
fabbioneotherwise file a bug12:04
raphink:(12:04
raphinkthe bug exists fabbione, I'm fixing it 12:04
raphink;)12:04
raphinkbut I need some guidelines from the kernel guys12:04
raphinkcause I don't want to break update-grub12:05
Keybukraphink: fabbione says he's busy, and suggested you speak to zul when he's around12:06
raphinkKeybuk: yes I can understand what fabbione said ;)12:06
Keybukraphink: ah, then why did you carry on bugging him?12:06
raphinkI don't need english->english translation :)12:06
Keybukif somebody says they're busy and don't have time to help you, it's generally quite rude to carry on talking to them12:07
=== chmj [n=chmj@196.44.1.98] has joined #ubuntu-kernel
raphinkI've uploaded a fix for the grub package. Who shall I assign the bug to so the fix gets uploaded ?01:01
Keybuk... if you've already uploaded it ?01:02
Keybukleave it unassigned, it'll get assigned to the appropriate person01:02
raphinkI've uploaded to malone01:02
Keybukah, leave it unassigned01:03
raphinkI doubt it'll get assigned soon, otherwise it would already be01:03
Keybukwhat makes you think that?01:03
raphinkcause this bug is not very very new ;)01:04
Keybukthere's nobody in particular who knows/cares about grub01:04
raphinkKeybuk: I mean is there an equivalent to motu-reviewers for main?01:04
Keybukyes, the default assigness01:05
Keybuk(ie. "nobody")01:05
raphinkthis is a quite small patch, could you have a look at it ?01:05
Keybukno, sorry01:05
KeybukI don't know grub01:05
raphinkit's just a piece of bash in update-grub01:06
raphinkto better detect and replace the splashimage line01:06
Keybukhaving attached the patch to the bug, it should get noticed on the ubuntu-bugs mailing list01:06
raphinkit has nothing to do with grub itself01:06
Keybukhave you talked to mvo about it?  he was dealing with splashimage stuff01:06
Keybuk(according to the changelog)01:06
raphinkreally?01:06
raphinkhmm01:07
raphinkI'll go talk to him01:08
=== mjg59 [n=mjg59@cavan.codon.org.uk] has joined #ubuntu-kernel
fabbioneBenC: new OCFS2 rocks :)01:50
fabbioneit doesn't crash anymore on super load01:50
BenCsweet01:50
fabbionei am checking if i have something for you01:50
fabbionei don't hounestly rememeber if i did some push/pull :)01:51
fabbioneoh crap01:53
fabbionei do have some stuff01:53
fabbionei did a pull from david but it did download all the 1500 objects again01:54
fabbioneand now it's pushing them to roockery01:54
fabbionecommit 0351e92ae406408854d0a4a6d7251f8e04308a7901:56
fabbioneAuthor: Ben Collins <bcollins@ubuntu.com>01:56
fabbioneDate:   Mon Feb 27 18:39:56 2006 -050001:56
fabbione    Blah01:56
fabbioneAHHAHA01:56
fabbionelovely01:56
fabbioneBenC: please pull from me01:56
fabbionei have a couple of changes/fixed01:56
fabbionefixes01:56
fabbionemore sun4v and one tg3 fix01:56
BenCfabbione: that stupid file wouldn't go back to "no change"...I've no idea why it had a delta01:56
BenCok01:56
fabbione(for sun blade 2500(01:56
mjg59BenC: Did those diffs I sent look ok?01:56
BenCmjg59: yeah01:57
mjg59Rock01:57
fabbionereceiving file list ... rsync: link_stat "/scm/linux/kernel/git/bcollins/ubuntu-2.6/objects/." (in pub) failed: No such file or directory (2)02:14
fabbionesomething wrong on master.k.o ben?02:14
fabbionenever mind02:16
BenCI'm doing a git-prune02:23
=== zul [n=chuck@ubuntu/member/zul] has joined #ubuntu-kernel
fabbioneok02:24
fabbioneit was only for 30 secs or so02:24
fabbioneit disappeared immediatly after02:24
zulheylo02:26
fabbionehi zul02:29
zulapparently my ears were burning in the log file02:30
=== CataEnry [n=cataenry@host83-39.pool8248.interbusiness.it] has joined #ubuntu-kernel
=== doko [n=doko@dslb-084-059-090-097.pools.arcor-ip.net] has joined #ubuntu-kernel
=== mxpxpod [n=BryanFor@unaffiliated/mxpxpod] has joined #ubuntu-kernel
zulBenC: ping03:36
BenCzul: pong03:36
zulim getting the following error when im doing a pull03:36
zulinux/kernel/git/bcollins/ubuntu-2.603:36
zulrsync: link_stat "/scm/linux/kernel/git/bcollins/ubuntu-2.6.git/refs/heads/ubuntu-fixes" (in pub) failed: No such file or directory (2)03:36
zulrsync error: some files could not be transferred (code 23) at main.c(1173)03:36
BenCcurrently?03:37
zulyep...ill try again thought if you want03:37
BenCoh, yeah, rm -f .git/refs/heads/ubuntu-fixes03:37
BenCactually cd .git/refs/heads/03:38
BenCrm everything except for master and origin03:38
zuldone03:38
=== ubuntulog [i=ubuntulo@trider-g7.fabbione.net] has joined #ubuntu-kernel
=== Topic for #ubuntu-kernel: Ubuntu kernel development discussion ONLY | New git tree for dapper: https://wiki.ubuntu.com/KernelGitGuide | 2.6.15-16.23 uploaded (PPC FTBFS fixed) | Daily Diet of Destruction: http://people.ubuntu.com/~bcollins/kernels-daily/
=== Topic (#ubuntu-kernel): set by BenC at Mon Feb 20 16:39:14 2006
=== fabbione [i=fabbione@gordian.fabbione.net] has joined #ubuntu-kernel
=== BenC [n=bcollins@debian/developer/bcollins] has joined #ubuntu-kernel
=== BenC [n=bcollins@debian/developer/bcollins] has joined #ubuntu-kernel
=== CataEnry [n=Enrico@host83-39.pool8248.interbusiness.it] has joined #ubuntu-kernel
=== zul [n=chuck@ubuntu/member/zul] has joined #ubuntu-kernel
=== CataEnry [n=cataenry@host83-39.pool8248.interbusiness.it] has joined #ubuntu-kernel
=== mgalvin [n=mgalvin@ubuntu/member/mgalvin] has joined #ubuntu-kernel
fabbioneBenC: 09:41
fabbionersync: link_stat "/scm/linux/kernel/git/bcollins/ubuntu-2.6.git/refs/heads/ubuntu-fixes" (in pub) failed: No such file or directory (2)09:41
fabbioneagain?09:42
=== ispiked_ [n=ispiked@Lee-11-133.rh.ncsu.edu] has joined #ubuntu-kernel
=== netzmeister [n=netzmeis@p549FAC7E.dip0.t-ipconnect.de] has joined #ubuntu-kernel
fabbioneBenC: can you please pull from me again?10:14
fabbionei have got the SPARC64 IIIi SMP bug fix10:15
fabbioneso we will make elmo and the porting machine more happy :)10:15
lamontBenC: it would be wonderful if the next upload included the hppa patches - it'll help the DC buildd machines out some, as well as making it possible for me to use my new test machine...10:26
BenCfabbione: yeah10:59
BenCfabbione: also, cd .git/refs/heads/ and rm everything but master and origin10:59
BenClamont: it will be10:59
fabbioneBenC: i did try that too... it didn't work.. i had to clone again11:15
=== netzmeister [n=netzmeis@p549FAC7E.dip0.t-ipconnect.de] has left #ubuntu-kernel ["...]

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!