=== ubuntulog [i=ubuntulo@trider-g7.fabbione.net] has joined #ubuntu-kernel === Topic for #ubuntu-kernel: Ubuntu kernel development discussion ONLY | Kernel Wiki: https://wiki.ubuntu.com/KernelTeam | Latest kernel upload: 2.6.22-9.25 | Latest news: -rt and -xen kernels re-added | New Kernel Team machine: http://kernel.ubuntu.com/ === Topic (#ubuntu-kernel): set by lamont at Fri Aug 3 16:12:28 2007 === #ubuntu-kernel [freenode-info] please register your nickname...don't forget to auto-identify! http://freenode.net/faq.shtml#nicksetup === fabbione [i=fabbione@gordian.fabbione.net] has joined #ubuntu-kernel === Starting logfile irclogs/ubuntu-kernel.log === ubuntulog [i=ubuntulo@ubuntu/bot/ubuntulog] has joined #ubuntu-kernel === Topic for #ubuntu-kernel: Ubuntu kernel development discussion ONLY | Kernel Wiki: https://wiki.ubuntu.com/KernelTeam | Latest kernel upload: 2.6.22-9.25 | Latest news: -rt and -xen kernels re-added | New Kernel Team machine: http://kernel.ubuntu.com/ === Topic (#ubuntu-kernel): set by lamont at Fri Aug 3 16:12:28 2007 === johanbr [n=j@blk-224-159-249.eastlink.ca] has joined #ubuntu-kernel === bullgard4 [n=detlef@p54BF09DB.dip0.t-ipconnect.de] has joined #ubuntu-kernel === joris_ [i=joris@082-146-097-013.dyn.adsl.xs4all.be] has joined #ubuntu-kernel [08:09] I have a rather odd feisty problem. Machine with a core2quad cpu with a 2.6.20-16-generic kernel works very fast, altough with only 3.2 out of 8GB of ram. Installing the 2.6.20-16-server kernel halves disk IO speed (measurably) and subjectively decimates CPU performance... I'm at a complete loss [08:09] I filed a bugreport containing some more info https://launchpad.net/ubuntu/+bugs?field.searchtext=2.6.20-16-server&orderby=-importance&search=Search&field.status%3Alist=New&field.status%3Alist=Incomplete&field.status%3Alist=Confirmed&field.status%3Alist=Triaged&field.status%3Alist=In+Progress&field.status%3Alist=Fix+Committed&field.assignee=&field.bug_reporter=&field.omit_dupes=on&field.has_patch=&field.has_no_package= [08:10] wrong url, sorry [08:12] joris_: are you trying 32-bit or 64-bit kernel? [08:13] 32bit [08:13] Try 64-bit [08:13] hey BenC [08:14] hey fabbione [08:14] (hmmz, I can't seem to find my own bugreport on launchpad..) [08:15] bug id #132577 [08:16] BenC: do you mean to install an x86_64 optimized kernel, or switching the entire distro to 64bit? [08:16] well, you have to do the entire distro [08:16] ouch [08:16] but you could just do a kernel, but it wont be easy [08:17] it's supposed to become a fat server (for thin clients) for a school... they'll need some 32bit apps like flash [08:18] BenC: apt-pinning feisty/amd64 for linux-image*? [08:18] no you can't use pinning for this [08:18] this is taking a .deb by hand and dpkg-deb extract/control -> build [08:19] and having to continue doing it for security updates [08:19] hmm... not fun [08:19] you can't use a chroot for the 32-bit thin clients? [08:19] is it a known problem? is it worth the effort to try to bugfix this? [08:19] joris_: are you using LTSP ? [08:20] or some manually crafted thin client installation? [08:20] it's not a bug...just 32-bit can only handle but so much (the slowdown on -server is a bug though) [08:20] BenC: possibly yes, but I'm having enough trouble getting LTSP up and running without doing an arch transposition [08:20] fabbione: no, edubuntu ltsp :) [08:20] joris_: ok. i suggest you wait one hour or so and talk to ogra [08:21] joris_: i think he did put some work into running an amd64 server with 32bit thin client for flash and stuff [08:21] aha [08:21] he'll be on here? [08:21] he is no kernel expert but he is the LTSP guru [08:21] nice :) [08:21] in #edubuntu or #ubuntu-devel [08:21] I'll see if I can remotely reinstall the machine to amd64 [08:21] but he is not awake yet for what i can see [08:22] i'd recommend to talk to ogra first [08:22] and see what's the best way to handle it [08:22] well, I don't know yet if going to amd64 will fix the lack of performance [08:34] joris_: you could try booting an amd64 livecd and see what it would do [08:35] pretty sure it would handle the memory a lot better [08:37] joris_: Your BIOS doesn't set MTRRs for the upper memory [08:37] So it's all flagged as uncacheable, and performance sucks [08:37] It's unlikely that using a 64-bit kernel will help her [08:37] e [08:37] ah, there's a better answer [08:38] We really need PAT support in Linux, but nobody's written an implementation the maintainer is happy with [08:45] mjg59: hmmm [08:46] mjg59: that's not good [08:46] and this is why I recommended using a server board, but they wanted to save 100 bucks [08:47] mjg59: the effect is really surprisingly strong btw [08:47] Yes, it will be [08:47] The machine will be basically unusable [08:47] it's not like a 10% slowdown, it's more like a 95% slowdown [08:47] it is indeed === cobman [n=justinas@194.176.50.52] has joined #ubuntu-kernel === martink [n=martin@84-75-131-205.dclient.hispeed.ch] has joined #ubuntu-kernel [08:56] mjg59: anything I can try to objectively measure this? [08:57] joris_: Not really - it'll depend on workloads and where in memory stuff gets allocated [08:58] The bottom ~4GB is probably flagged cacheable, but the kernel will be loaded higher [08:58] So if your code is in the cacheable region and doesn't make any syscalls, it'll probably run fairly fast [08:59] hmmm [09:00] without programming C there is no way to put a benchmark util in upper or lower memory, I presume [09:01] take out half the memory and see if it improves vastly? [09:01] There's no way to do it at all [09:01] You're at the mercy of the kernel in terms of where your app ends up [09:02] Mithrandir: that's what happens with the desktop kernel, presumably [09:02] mjg59: ok :( [09:02] It's a genuine failing in Linux to implement PAT, but it's also a BIOS issue to fail to set the MTRRs sanely [09:02] joris_: presumably, yes. [09:02] You can boot with different values of mem= to find the maximum you can get away with [09:03] joris_: Adding the output of cat /proc/mtrr to the bug would be helpful [09:03] mjg59: is there any logic in the values to try? [09:03] mjg59: will do so [09:03] joris_: Look at /proc/mtrr. Figure out the highest address that's flagged write-back or write-through. [09:03] That's the maximum amount of RAM you can use [09:03] okay [09:04] as soon as the machine finishes installing sysinfo (half an hour and counting) [09:09] mjg59: it's attached [09:09] joris_: Thanks - what was the bug number, again? [09:09] https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.20/+bug/132577 [09:10] what does write-combining mean? [09:12] joris_: Looks like you're not going to get more than 3584MB [09:12] joris_: http://en.wikipedia.org/wiki/Write-combining :) [09:13] ouch :( [09:13] The mtrrs simply don't cover more than that [09:16] moin [09:18] mjg59: do you know off hat what unit/syntax does the kernel mem= parameter accept? [09:19] 'mem=xyzM' ? [09:19] Yeah [09:20] mem=3584M it is then [09:21] Might end up needing to be slightly lower than that [09:21] But give that a go first [09:22] okay, here goes nothing [09:24] that worked :) [09:24] Excellent [09:25] it's most likely an intel 965-based motherboard [09:26] Yeah [09:26] All you can really do at this point is check for a bios update, I'm afraid [09:27] yep... makes sense :/ [09:27] Or wait for PAT support [09:27] I'll break the bad news === EtienneG [i=etienne@ubuntu/member/EtienneG] has joined #ubuntu-kernel === blenderhead001 [n=blenderh@adsl-068-209-133-121.sip.jax.bellsouth.net] has joined #ubuntu-kernel === Nafallo [n=nafallo@ubuntu/member/nafallo] has joined #ubuntu-kernel === Lure_ [n=lure@89-212-18-142.dynamic.dsl.t-2.net] has joined #ubuntu-kernel === asac_ [n=asac@debian/developer/asac] has joined #ubuntu-kernel === EtienneG [i=etienne@ubuntu/member/EtienneG] has joined #ubuntu-kernel === EtienneG [i=etienne@ubuntu/member/EtienneG] has joined #ubuntu-kernel === Keybuk [n=scott@quest.netsplit.com] has joined #ubuntu-kernel === zul [n=chuck@ubuntu/member/zul] has joined #ubuntu-kernel [02:31] morning [03:05] morning zul === IntuitiveNipple [n=TJ@alexandros.tjworld.net] has joined #ubuntu-kernel [03:22] hi lamont how goes it? [03:22] generally pretty well [03:23] good === Lure_ [n=lure@89-212-18-142.dynamic.dsl.t-2.net] has joined #ubuntu-kernel === tuxmaniac [n=aanjhan@unaffiliated/tuxmaniac] has joined #ubuntu-kernel === ThrobbingBrain66 [n=bradley@CPE-24-208-44-208.new.res.rr.com] has joined #ubuntu-kernel === calc [n=ccheney@conr-adsl-209-169-124-200.consolidated.net] has joined #ubuntu-kernel === instabin|work [n=instabin@c-68-32-128-115.hsd1.pa.comcast.net] has joined #ubuntu-kernel [04:27] Is any one working on getting the nvidia 100.14.11 drivers in to gutsy [04:31] Yes. === johanbr [n=j@JBrannlund.MathStat.Dal.Ca] has joined #ubuntu-kernel === chrisle [n=chris@e178122004.adsl.alicedsl.de] has joined #ubuntu-kernel === chrisle [n=chris@e178122004.adsl.alicedsl.de] has joined #ubuntu-kernel === BFTD [n=thomas@67-150-254-75.oak.mdsg-pacwest.com] has joined #ubuntu-kernel === dhon__ [n=dhon@60-240-97-109.static.tpgi.com.au] has joined #ubuntu-kernel === pmjdebruijn [n=pmjdebru@ds9.pcode.nl] has joined #ubuntu-kernel === tuxmaniac [n=aanjhan@unaffiliated/tuxmaniac] has joined #ubuntu-kernel === bronson [n=bronson@adsl-76-202-198-213.dsl.pltn13.sbcglobal.net] has joined #ubuntu-kernel === Starting logfile irclogs/ubuntu-kernel.log === ubuntulog [i=ubuntulo@ubuntu/bot/ubuntulog] has joined #ubuntu-kernel === Topic for #ubuntu-kernel: Ubuntu kernel development discussion ONLY | Kernel Wiki: https://wiki.ubuntu.com/KernelTeam | Latest kernel upload: 2.6.22-9.25 | Latest news: -rt and -xen kernels re-added | New Kernel Team machine: http://kernel.ubuntu.com/ === Topic (#ubuntu-kernel): set by lamont at Fri Aug 3 16:12:28 2007 === m0rg0th [n=m0rg0th@219.64.120.181] has joined #ubuntu-kernel === dorileo [n=dorileo@201.40.162.47] has joined #ubuntu-kernel === chrisle [n=chris@e178122004.adsl.alicedsl.de] has joined #ubuntu-kernel === m0rg0th [n=m0rg0th@219.64.120.181] has joined #ubuntu-kernel === IntuitiveNipple [n=TJ@alexandros.tjworld.net] has joined #ubuntu-kernel === zul [n=chuck@ubuntu/member/zul] has joined #ubuntu-kernel === NativeAngels [n=Native@host86-146-52-246.range86-146.btcentralplus.com] has joined #ubuntu-kernel [10:42] hello [10:42] is there anyone in here who knows about unrealirc and neostats === verwilst [n=verwilst@d51A4A585.access.telenet.be] has joined #ubuntu-kernel [10:42] NativeAngels: This sounds like the wrong place to ask [10:43] i know [10:43] sorry [10:44] hellow! [10:44] hello [10:44] i'm building the -xen kernel for amd64 [10:44] i did an apt-get source linux-source-2.6.22 [10:44] and then debuild === jweyrich [n=phoenix@189.5.194.90] has joined #ubuntu-kernel [10:44] but this will take aaaages :) [10:44] is there an easy way to only build the -xen kernel? [10:44] ): [10:45] it's done "build-generic build-server custom-build-rt custom-source-rt" so far :) [10:45] btw any reason why -xen is i386-only? [10:45] and not amd64? === Instabin [n=instabin@oh-76-5-123-160.dhcp.embarqhsd.net] has joined #ubuntu-kernel [11:14] verwilst: because it doesn't compile on amd64 AFAIK :) [11:14] oh? [11:14] we'll soon know =- [11:14] :) [11:14] i'm compiling it now [11:15] I might be wrong, but I'm sure there must be a good reason [11:16] i could really use an amd64 version of the -xen though :) [11:16] the 2.6.18 from xensource has ext3 corruption [11:16] as i experienced :p [11:16] Xen patches seem to be constantly outdated compared to current kernels anyway :) [11:17] yeah [11:17] maybe i should just use a 32bit system instead of a 64bit.. [11:17] but 64bit is a lot faster imo [11:17] does the -generic have pae? [11:23] hm, what's the kernel config option for PAE again? [11:24] anyway, there are -server kernels too [11:25] i need to use -xen, stupid me :) [11:25] 32bit, with -xen [11:25] can i go over 4GB RAM with it? [11:27] I have no -xen kernels installed === phoenixx [n=phoenix@189.5.194.90] has joined #ubuntu-kernel [11:30] I guess they support > 4 GiB of RAM if the patch allows for it === tormod [n=tormod@80-219-112-71.dclient.hispeed.ch] has joined #ubuntu-kernel === tormod [n=tormod@80-219-112-71.dclient.hispeed.ch] has left #ubuntu-kernel [] === cjwatson_ [n=cjwatson@82-69-40-219.dsl.in-addr.zen.co.uk] has joined #ubuntu-kernel === cobman [n=justinas@194.176.50.62] has joined #ubuntu-kernel