=== dyfet` is now known as dyfet === Meiz_n810 is now known as Meiz_n900 === Meiz_n900 is now known as Meiz_n810 [10:36] gcc 4.4.1 seems to be generating some screwy code for me. [10:37] Is there a newer release I could try? [11:10] erikcorry: Is this reported? [11:10] erikcorry: 4.4.1 is the latest we have, but we add patches [11:12] I'm trying to reduce it. [11:12] Fails to compile Google V8 [11:12] In -O2 mode [11:13] I already added -fno-strict-aliasing [14:02] Gotcha. [14:02] Who should I mail the reduced test case to? [14:02] erikcorry: Upstream? :-) [14:02] Or should I just file a bug on gcc's site? [14:02] erikcorry: I'd file a bug against gcc-4.4 upstream [14:03] Yeah [14:03] It only happens with -O3 [14:04] erikcorry: You could find out which subopt breaks it [14:04] erikcorry: gcc -c -Q -Os --help=optimizers will give you a list [14:04] Thx [14:04] or -O3 [14:05] Woah! [14:05] 163 optimizations! === cbrake_away is now known as cbrake [14:05] erikcorry: Have fun :) [14:05] thats what these gentoo guys say too all the time ;) [14:05] and they they switch them all on ;) [14:25] -fno-tree-sink makes it go away. [14:26] That's actually part of -O1 === playya_ is now known as playya [16:54] hey guys, the latest rootstock is hanging for me on startup complaining about rsyslogd [16:54] pastebin at : http://pastebin.com/m21cd01a3 [16:55] looks like a dbus issue [16:55] is that during build or the installed rootfs on a machine ? [16:56] it's on an sd card, so during startup of the sheevaplug that I am running it off of [16:56] build was successful [16:56] what release did you build ? [16:56] jaunty or karmic [16:57] karmic [16:57] karmic wont run on sheeva ... its ARMv6 now [16:58] oh ok [16:58] we use -march=armv6 -mfloat-abi=softfp and -mfpu=vfp now [16:58] I'm guessing it's the same for debian? [16:58] so thats likely where your illegal instruction comes from ... [16:58] no [16:58] debian uses v5 or even something smaller afaik [16:59] so the img files that are build for ubuntu are all armv6 as well then [17:00] jalto: for karmic [17:00] but yes [17:01] alright that's for the help [17:01] jaunty will work on your sheeva though [17:01] jalto: jaunty is fine though [17:02] I was having trouble with rootstock for jaunty as well, but I need to revisit it and report back [17:02] yes, tell me if you run into any issues [18:11] ogra, I tried to get jaunty up and running and it hangs on starting kernel log daemon [18:12] the kernel I'm using comes from http://sheeva.with-linux.com/sheeva/ [18:12] jalto, did you enable a serial console ? [18:12] 2.6.30.5 [18:13] in grub? [18:13] in rootstock :) [18:13] it has a --serial option, else the system wont set up serial in init [18:14] i.e the only login prompt you would see after starting klogd would be on a VGA output ... [18:14] which is hard to get in a sheeva i imagine [18:15] ogra thanks a tty might be useful [18:15] the other option is to have openssh-server installed and know the IP [18:15] so you can log in via ssh [18:16] back to building [18:16] do you have your tgz around ? [18:17] just untar it in a dir ... [18:17] https://wiki.ubuntu.com/ARM/BuildEABIChroot ... [18:17] chroot into it and install openssh-server [18:17] (and set up /etc/network/interfaces if needed) [18:18] there are jaunty packages in my ppa [18:18] thanks it's on an sd card so I can just pop it into my desktop and play with it [18:19] yeah [18:19] note that qemu-arm-static doesnt work with mono packages yet ... in case oyu plan to do any mono stuff [18:43] ogra, it pinged fine, but my build doesn't have an /etc/ssh folder with configs, so it wouldn't accept ssh [18:44] not sure if ssh was installed [18:45] its not by default [18:46] if you didnt add it to the --seed option [18:46] if you install it through qemu-arm-static you need to make sure /proc is mounted inside your chroot [18:46] else it wont generate keys [18:46] (and make sure to unmount /proc again afterwards) [18:47] ok [18:48] well the fact that it pinged was a good sign [18:48] indeed [18:49] anyway, its getting late here, i'll go afk now ... [18:50] good night [18:50] and thanks for the help [18:50] see you around (i hope) ;) [22:22] ogra: just noticed something about my console-kit-dae problem [22:22] on dmesg: console-kit-dae invoked oom-killer: gfp_mask=0x1201d2, order=0, oomkilladj=0 [22:23] it seems like the oomkiller killed it, but it still lives on and eats all cpu... [22:26] Hmm [22:27] ali12341: Would be best to start afresh and monitor what's happening on a fresh system; could be a process eating all the RAM, being killed and then another one using all CPU waiting for it [22:27] that happened during boot [22:27] this machine only has 64mb [22:28] Ah that might be why [22:28] * lool sleep & [22:28] it seems like actually gdm got killed [22:33] the funny thing is i can actually log in to the desktop and everything works more or less [22:33] it's really slow though, due to the runaway process