[00:42] pbuckley: export $(dpkg-architecture -aarmel); export CROSS_COMPILE=arm-linux-gnueabi-; fakeroot debian/rules clean binary-omap4 [00:42] pbuckley: you'll need gcc-arm-linux-gnueabihf installed [00:42] err, gcc-arm-linux-gnueabi [00:43] for armhf, s/armel/armhf/ and it's gnueabihf [00:44] thank you [00:45] im stil waiting for my first compile i kicked off about 4 hours ago [00:45] so cross compiling is really the only way [00:45] i can do this in any sort of real time frame [01:29] i wonder if distcc is still around [01:31] sweet it is.. anyone have success with it on arm? [01:31] it works [01:32] ty [01:32] it's not very useful though, if you have 40 arm boxes to hook up with it you're probably also in posession of a rather powerful PC too :) [01:32] well [01:32] true.. but even in ec2 the kernel is taking forever.. ive gotten spoiled with near instant builds [01:32] going to spin up a couple more instances [01:33] and distcc this shiznit [01:33] Neko: some stuff can't be cross-compiled, though [01:34] Although the same stuff probably also has trouble with distcc... [01:34] and that;s where you break out qemu to emulate arm enough to get the bits that can't be cross-compiled done [01:34] I guess. [01:34] Not all arches get so much love in qemu :-) [01:34] suse manage it in their build service just fine, ubuntu are getting this stuff up and running. compiling natively on arm boxes is just looking for a long wait no matter how many you have. [01:35] when you hit quad or eight core 1.8GHz monsters with 4GB RAM and real hard disks, yeah, it'll make sense again, but they don't exist yet. Not quite yet :) [01:35] AFAIK Debian still relies heavily on native builds [01:36] and it'd still be faster on a PC. Nobody does a "native" Android build [01:36] * pbuckley drools about the thought of an eightcore 4gb ram monster [01:36] yeah we own the armhf build farm.. [01:36] Well, android are crack-smoking java monkeys, what do you expect [01:36] and it works.... but there's a better way [01:36] whats the better way? [01:36] cross-compile :] [01:37] heh [01:37] use your 8-core hyperthreaded xeon with 32gb ram and a 4TB raid array and build all of android in 25 minutes. [01:37] I expect it'll become trendier in Debian once multilib stuff is done [01:38] Neko: that still takes 25 minutes? You can build openwrt on an *atom* in that time [01:38] openwrt is a little smaller though [01:39] building a 6MB ROM for a router is a bit different to something that comes out about 300MB worth of apps and java crap [01:39] spitting a kernel out takes a couple minutes for a full distclean build.. you'd never do that on ARM even if you had 50 boxes doing it [01:40] im at about 5 hours for building 3.2.0 on my pandaboard [01:40] it takes 80 minutes on an Efika MX :) [01:40] Those extra 294MB are just bloat ;-) [01:41] twb: thats the java ;) [01:41] have 50 of them and it might take half an hour, maybe 10 minutes if you're lucky [01:41] twb, I did say java crap :D [01:41] efika mx you say [01:41] * pbuckley googles [01:41] Now now, just because dalvik walks like a duck and quacks like a duck, doesn't mean Oracle can sue Google ;-) [01:42] pbuckley: The Efika's much slower than your Panda, I wouldn't bother with the googling. [01:42] pbuckley: If it has taken 5 hours on your panda, you're doing something wrong. https://launchpad.net/ubuntu/+source/linux-ti-omap4/3.2.0-1405.7/+build/3121438 [01:42] And that system isn't even optimized. I think it is running natty. [01:43] GrueMaster: apt-get --only-source source linux-ti-omap4 (something); cd /usr/src/linux; make oldconfig; make [01:43] is basically all i did [01:43] Are you running on SD or USB Sata? [01:43] SD [01:44] GrueMaster: he is probably building with debug on [01:44] Ah. That explains it. You'll see almost 10x performance increase on USB. [01:44] That ALWAYS gets me [01:45] I'm regretting buying a pandaboard, not because it's slow (usb disk makes it fun to use again) but because everything just seems not to be as good as you'd think [01:45] awesome CPU, awful TI peripherals [01:45] except the wireless, that works really well [01:45] Looking at the buildlog from the link I posted, the buildd is running Natty (Kernel version: 2.6.38-1209-omap4 #20-Ubuntu). [01:46] The panda is essentially a cellphone dev platform. When thought of along those terms, it is actually quite good. [01:46] it's a shame nobody's making a real omap4 desktoppy kind of system that's been optimized for usage and not just developing phones [01:46] the other TI dev board is much better than the panda for that though too [01:47] like.. if you had the choice you'd buy the MX53 EVK and not bother with the Quickstart [01:47] if you were developing phones that is [01:47] Actually, there are a lot of non-phone systems out there (kindle fire, nook tablet). And I heard of one system coming soon that will just plug into a spare hdmi port on your tv. [01:47] I thought omap4 was a better platform than the dragonball [01:47] not a big enough market for it [01:47] tablets are just big phones at the end of the day, same development model [01:48] omap4 needs a trimslice, efika mx kind of thing [01:48] True. Can't wait for omap5. Could be interesting. [01:48] well I'm duty bound to wait for MX6, but I can't help thinking most companies will run to nVidia before TI at the moment [01:48] all the new tablets in the pipeline seem to be tegra3, if that is at all realistic [01:49] I remember when they were all going to be MX51, hahahaha.. [01:49] what happens at CES, stays at CES [01:49] Yea, well.... [01:50] a tegra3 pandaboard like board would be fun [01:50] *I* remember in like 2008 when asus demoed an 9" eeepc running arm [01:50] And then I had to wait four years for the tf101 [01:50] so whats the trick to turn debug mode off? [01:51] pbuckley: make menuconfig, find it in the kernel hacking section [01:51] It basically means whether it builds with -g [01:53] twb: That is on by default in all of our kernels, from what I can tell. [01:53] GrueMaster: it is on standard ubuntu and debian kernels, yes [01:54] but it makes compile take an order of magnitude longer, ESPECIALLY if you have slow I/O [01:54] ah k [01:54] thanks [01:54] CONFIG_DEBUG_KERNEL=y I guess, I don't have menuconfig in front of me right now [01:54] But IME you can't just edit .config you need to use the UI [01:55] E: Package 'ncurses' has no installation candidate [01:55] oh wait [01:55] that changed to [01:55] I think you can if you run make oldconfig afterwards. [01:55] ncurses-base [01:55] wow that package name changed again [01:56] libncurses5-dev or so [01:56] though im a bit surprised it doesnt get installed with apt-get build-dep [01:56] ncurses-base is just terminfo files [01:56] build-dep of a kernel *source* package? [01:57] well know.. but of the linux-kernel package [01:57] s/know/no [01:57] linux-kernel is a metapackage, no build deps [01:57] ah [01:58] libncurses5-dev isn't a build-dep of the kernel because the kernel build doesn't call menuconfig. [01:58] fair enough [02:00] infinity: right, thanks [02:00] flag changed.. thank you for the tip [02:00] :) [02:01] and on that note its time to go home [02:12] If building natively, you can also do "make localmodconfig" and "make localyesconfig" to compile all used modules in, and disable all unused modules [02:31] GrueMaster: the rebuilt kernel works [02:31] GrueMaster: the device is detected [02:31] Excellent! [04:28] hi there :) i bump into a problem building linux kernel for arm xilinx,can anyone help me? [05:17] hi there guy, i needs some help please.sorry but i am new in this area [05:17] ur help would be greatly appreciated ,thank you [05:19] Im building a linux kernel [05:19] this the error i have,please show my mistake [05:19] nidsub-VirtualBox:/opt/codesourcery/linux-2.6-xlnx$ sudo make ARCH=arm [05:19] scripts/kconfig/conf --silentoldconfig Kconfig [05:19] CHK include/linux/version.h [05:19] UPD include/linux/version.h [05:19] CHK include/generated/utsrelease.h [05:19] UPD include/generated/utsrelease.h [05:19] Generating include/generated/mach-types.h [05:19] CC kernel/bounds.s [05:19] cc1: error: unrecognized command line option ‘-mlittle-endian’ [05:19] cc1: error: unrecognized command line option ‘-mno-thumb-interwork’ [05:19] kernel/bounds.c:1:0: error: unknown ABI (aapcs-linux) for -mabi= switch [05:19] kernel/bounds.c:1:0: error: bad value (armv5t) for -march= switch [05:19] make[1]: *** [kernel/bounds.s] Error 1 [05:19] make: *** [prepare0] Error 2 [05:23] here the detail on the step i took ..from this webpage [05:24] http://wiki.xilinx.com/zynq-linux [05:24] nidsub: you need to learn about a pastebin [05:24] !pastebin | nidsub [05:24] nidsub: For posting multi-line texts into the channel, please use http://paste.ubuntu.com | To post !screenshots use http://imagebin.org/?page=add | !pastebinit to paste directly from command line | Make sure you give us the URL for your paste - see also the channel topic. [05:25] ohh, im really sorry [05:25] :( sorry guys [05:32] http://paste.ubuntu.com/832267/ === Quintasan_ is now known as Quintasan === dduffey_afk is now known as dduffey [15:44] XavB: thanks for your help the other day, my board has been running flawlessly since installing the arm-extras :-) [15:44] XavB: yeah !!! [15:45] Person987: so you are running the 3.1 kernel, right? [15:45] how can I check? :-) [15:45] uname -a returns Linux ubuntu-desktop 3.1.0-1282-omap4 #10 ... [15:47] XavB: yes, thats it. It was locking up every 30min and since upgrading it has not had a single problem. I even left my OpenGL app running all day when I went to work and it was still running that night. [15:48] Person987: just perfect... :D [15:48] XavB: just ordered another board for my dad's birthday :-) [15:48] ;) [15:53] PePe === shadeslayer is now known as shadeslayer_ === shadeslayer_ is now known as shadeslayer [17:14] GrueMaster: around? [17:15] Very much so, sadly. Need to lose weight. [17:15] Oh, you meant something else? [17:15] GrueMaster: I got a second usb serial cable [17:16] GrueMaster: I have the same issue as well, not being able to hook on to the serial port [17:16] GrueMaster: stty hangs [17:16] Very odd. [17:17] I have not seen that issue and I have 7-8 usb-serial cables (I lost track). [17:17] Do you have any other serial devices you can test that cable with? [17:17] GrueMaster: baud_rate = baud_base / divisor? [17:17] Not that I know of. [17:17] I just pulled an ancient desktop I have [17:18] I am seeing if it will boot of a USB flash drive [17:18] You will need a crossover to connect to a regular desktop. [17:18] (or are you trying it as a host?) [17:18] trying it as a host [17:19] let it export serial console [17:19] see if I can connect to it [17:19] I am wondering if setserial for some reason isnt detecting pl2303 [17:20] I tried my second laptop booting lucid [17:20] I have the same issue with it as well [17:20] I don't see why it would have issues. Works fine here on all my systems from Lucid forward. [17:20] That is disturbing. [17:21] Bus 005 Device 008: ID 067b:2303 Prolific Technology, Inc. PL2303 Serial Port [17:21] from lsusb [17:21] Most of the usb serial cables I have are from Trendnet. I also have 1 4-port usb-serial cable. All of them use the pl2303 chip. [17:22] the keyspan which I borrowed works right out of the box [17:22] I think I will order another cable [21:18] plop [21:19] janimo, I've been pointed to your package of the linux 3.0.19 kernel for arm, and wanted to give it a try ; it's unfortunately an armel package -- could you also provide an armhf variant? [21:19] (I'm now subscribed to the mailing-list, so will follow there...) [21:20] mailing list? [21:22] pbuckley, https://launchpad.net/~ac100 [21:23] ah cool thanks [21:25] pbuckley, no problem ; I wasn't on that team 15min ago ;-) [21:27] thihttp://www.engadget.com/2010/06/21/toshibas-ac100-8-hour-smartbook-runs-android-2-1-on-a-1ghz-tegr/ [21:27] this the device? [21:28] pbuckley, yes [21:28] the android was a joke, and killed the device commercially [21:28] i would love to find a tegra3 board [21:29] (besides the tf201) [21:35] GrueMaster: in hiding? === Jack87|Away is now known as Jack87 === Jack87 is now known as Jack87|Away [21:43] dioxin: Nope, just back from lunch. [21:44] GrueMaster: would you be able to guide me in the ways of local repositories? I've got ubuntu 10.04 on the spare box, just doing first round of updates now [21:46] dioxin: Here is my ubumirror.conf: http://paste.ubuntu.com/833211/ [21:47] With that, you just add a crontab job for the specific ubumirror app you want to run (ubuports for ports.ubuntu.com). [21:50] can I just sudo apt-get install ubuports ? [21:51] dioxin: sudo apt-get install ubumirror. ubuports is one script in the package. The package isn't very big. [21:52] My cron job is: [21:52] cat /etc/cron.d/ubumirror [21:52] # Update archive mirror every two hours. [21:52] 0 */2 * * * root /usr/bin/ubuports >> /var/log/ubumirror/cron.log [21:52] postfix local only? [21:52] (from install of ubumirror) [21:53] The /var/log/ubumirror/cron.log file only traps output from the cron job. ubuports (and all of the other ubu scripts) create their own logs in the same directory. [21:53] postfix local (unless you want it to run a larger mail server). [21:54] ubumirror scripts use that to send failure notices. [21:57] where does the ubumirror.conf go? /usr/bin/ubuports ? [21:57] I recommend using my ubumirror.conf and change the destination paths appropriate to your environment. Mine is set to ignore everything except armel/armhf on ports.ubuntu.com. It also pulls the daily arm images from cdimage.ubuntu.com (using ubucdimage script), but I don't recommend it as it can be slow. Better to use zsync and pull as needed. [21:57] ubumirror.conf goes in /etc [21:58] See dpkg-query -L ubumirror for all the file locations. [21:58] done it in the /etc/ directory [21:59] ... only need to wait 1 minute... [21:59] changed the 2 to a 1 [21:59] My conf file does not mirror the source files, but it does get everything needed for netboot install. First run will take several hours (like overnight). [22:00] how can I tell its running? ps -U root ? [22:00] I found that every 2 hours is enough. [22:00] ps ax|fgrep rsync [22:00] Or tail -f /var/log/ubumirror/ubuports.log [22:02] GrueMaster: my kernel compile finally finished ;) [22:02] only took 7 hours [22:03] pbuckley: Cool. how long did it take? [22:03] went alot faster after i shutdown firefox [22:03] about to boot it.. hopefully those patches help [22:08] next time ill be sure to turn off that kernel hacking debug stuff [22:08] GrueMaster: I'm getting errors that /server/mirrors/linux-distro blahblah no such file or directory [22:09] Yea, those paths are on my mirror. You need to edit /etc/ubumirror.conf and change all /server/mirror paths to something sane on your server. [22:10] My server has 4 partitions: /server/data /server/mirror /server/media /server/roots. I doubt your system will be the same. [22:11] Once you have the mirror, it will look like /ports.ubuntu.com/ubuntu-ports/* [22:11] I've just added /server/mirror/linux-distros/Ubuntu/ports.ubuntu-com/ubuntu-ports :D [22:12] Well, copy exact is one way to go. :P [22:13] I've just built the box specially for this task, it dont really matter :D [22:13] On my system, in my www directory (created by the web server), I have a symlink to /server/mirror/linux-distros/Ubuntu/ports.ubuntu-com/ubuntu-ports. That way, the mirror resolves to http:///ubuntu-ports, which makes changing the sources.list easier. [22:14] I guess I have to sort out the symlink for the apache later..... [22:17] ok I think I just added the symlink :D [22:40] GrueMaster: do you guys have a x86 >> armv7 cross compiler up and running yet? [22:41] dioxin: on x86 host just apt-get install g++-arm-linux-gnueabi [22:41] What he said. :P [22:42] kk I might have to try that once I get my hexa server back up and running :D [22:43] it really needs the other CPU tho :( [22:43] soon my precious! SOOOOON [22:43] heh [22:44] is distributed compilation supported out of the box for x86 to arm? [22:44] (me nub, be gentle) [22:49] I have no idea. Usually, cross compiling is only recommended for small projects or projects that don't do any post-compile processing (like running unit tests, etc). [22:49] And those projects normally fail on a distcc type environment anyways. [22:50] boo :( [22:51] I've got soooo much x86 horse power available :( [22:53] dioxin: unless you are doing a really really really big compile, simply have a dual core box and passing a -j4 to make will be plenty [22:53] dioxin: i can do a complete android build in under 20 minutes on a dual core 2ghz box [22:54] I've a 2.4 Ghz Hex core with hyperthreading ... so I might be able to do that in 3 minutes :) [22:55] I was expecting a full build to take much longer === dduffey is now known as dduffey_afk [23:31] GrueMaster: how do I tell the repo is completely downloaded.... the sync stopped running.... [23:32] sudo /usr/bin/ubuports & stopped almost immediately... [23:32] On your arm system, edit /etc/sources.list to point to it instead of ports.ubuntu.com and run apt-get update for starters. [23:33] Also, check the ubumirror logs (/var/log/ubumirror/ubuports.log.* [23:33] ) [23:36] kk its running again, I'll check properly in the morning [23:36] night peeps [23:39] night