**** BEGIN LOGGING AT Thu Nov 02 02:59:57 2006 Nov 02 06:51:46 ~seen marceln Nov 02 06:52:00 marceln was last seen on IRC in channel #nslu2-linux, 16d 14h 49m 14s ago, saying: 'I have compiled it and after that installed it on my wlhdd'. Nov 02 08:22:20 The Wiki links towards the OpenEmbedded HOWTO fails on me: http://www.nslu2-linux.org/wiki/OpenSlug/HomePage Nov 02 08:22:33 (and then click on OpenEmbedded) Nov 02 08:25:43 heh, nice one **** BEGIN LOGGING AT Thu Nov 02 09:20:15 2006 Nov 02 09:25:30 wiki fixed Nov 02 12:28:09 Building openslug-image for openslug/ixp4xx: Nov 02 12:28:11 In file included from umount.c:30: Nov 02 12:28:12 | nfsmount.h:6:1: error: unterminated #ifndef Nov 02 13:10:43 morning! Nov 02 13:12:30 morning Nov 02 13:13:21 o/ Nov 02 13:13:51 A cold one in the UK today Nov 02 13:16:10 how cold? Nov 02 13:16:56 -3 C in Chicago... for what it's worth :) Nov 02 13:17:13 It's starting to freeze here overnight in Ottawa, Ontario, Canada Nov 02 13:17:19 I can't wait to go ski... Nov 02 13:17:57 But you need snow for that -- we're on the wrong side of the Great Lakes for much snow. Just cold. Nov 02 13:18:06 it's 0C @ 8:38 Nov 02 13:18:17 yeah. plus the hills are pretty small around here. Nov 02 13:18:27 Head north, no? Nov 02 13:18:35 But Tremblant is not too far. Nov 02 13:18:57 Ah, but the best ski place is a long way from both of us: Whistler in B.C. Nov 02 13:19:11 yep. Nov 02 13:19:35 Never been but my friend goes every year. **** BEGIN LOGGING AT Thu Nov 02 13:33:32 2006 Nov 02 15:03:49 03bzhou * r4364 10optware/trunk/make/py-genshi.mk: py-genshi: upstream upgrade to 0.3.4 Nov 02 15:07:00 03bzhou * r4365 10optware/trunk/make/git-core.mk: git-core: upstream upgrade to 1.4.3.3 Nov 02 15:54:13 <[g2]> GPSFan ping Nov 02 19:04:45 http://lists.arm.linux.org.uk/pipermail/linux-arm-kernel/2006-November/036802.html Nov 02 19:10:44 what is the disadvantage of initrd? Nov 02 19:19:27 Space is one, sure. But we're saving a fair bit by using the open NPE driver (and if possible, putting the firmware in other partitions) Nov 02 19:23:48 An advantage is that it's much easier to make changes to how the startup process works without having to care about getting it upstream. Nov 02 19:25:28 morning, ByronT Nov 02 19:26:40 morning Nov 02 19:27:56 another advantage of initrd would be that it's easier to add support for new devices. Just change the initrd and you're ready to go. Nov 02 19:29:32 ah Nov 02 19:29:40 * NAiL reads the answer on L-A-K Nov 02 22:03:15 eno: did something happen recently with perl? I cant make perl-stage: Nov 02 22:03:35 make[1]: *** Keine Regel, um »perl_Configure« zu erstellen. Schluss. Nov 02 22:04:18 "no rule to make perl_Configure" Nov 02 22:15:47 looks like the cross makefile doesn't get copied in place anymore Nov 02 22:22:35 03gda * r4366 10optware/trunk/make/perl.mk: perl: problems building on ubuntu Nov 02 22:22:59 ubuntu is strange Nov 02 22:25:48 03gda * r4367 10optware/trunk/make/spamassassin.mk: spamassassin: 3.1.7 Nov 02 22:30:23 [g2]: ping Nov 02 22:30:57 <[g2]> GPSFan pong Nov 02 22:32:30 [g2]: did you & blaster8 ever figure out what was wrong with my config? Nov 02 22:32:44 <[g2]> GPSFan I never looked at it Nov 02 22:33:11 <[g2]> I just provided the file as blaster8 seem interested in the diff Nov 02 22:33:19 [g2]: oh, no matter, now that there is a working one it's somewhat academic. Nov 02 22:33:31 <[g2]> and with the working config he could easily diff Nov 02 22:34:10 quetion 2 did you & rwhitby ever figure out what to do about multiple machines in the same kernel & the machine id shim? Nov 02 22:34:11 <[g2]> GPSFan btw, right now the repo is missing the EEPROM notifier in the avila/loft config with the MAC patches, and the ixmac_open patch Nov 02 22:34:47 <[g2]> GPSFan I'm unclear what you mean by question 2. Nov 02 22:35:08 <[g2]> the machine id shim was missing for BE Nov 02 22:35:10 [g2]: k, I haven't looked at the repo since yesterday, been doing "real life" related work Nov 02 22:35:36 how do you add the shim> I just used hexedit on the zImage file. Nov 02 22:35:50 GPSFan: machine id shim works - Loft BE support was added to the Makefile in the svn kernel repo Nov 02 22:36:55 rwhitby: yes, I know it works, I was asking what the mechanism was for adding it to the zImage file (something in Makefile perhaps?) Nov 02 22:37:10 <[g2]> GPSFan timeout Nov 02 22:37:30 <[g2]> Redboot should pass the proper machine id Nov 02 22:37:46 <[g2]> we can fix this for your Redboot Nov 02 22:38:10 GPSFan: for other targets, yes, the Makefile adds the correct shim. For the Loft, [g2] has fixed RedBoot to send the correct machine id. Nov 02 22:38:49 <[g2]> rwhitby, I'm saying GPSFan should ping GW about passing the right machine ID Nov 02 22:39:08 <[g2]> it's not an IXDP425 platform Nov 02 22:39:29 GPSFan: do you have a Loft, or a different GW board? Nov 02 22:39:34 [g2]: ah that's a thought. Nov 02 22:41:08 <[g2]> rwhitby the Loft is a custom GW2348 board that has USB 2.0 Nov 02 22:41:36 [g2]: I know that. I'm asking if GPSFan has a Loft, or a different GW board. Nov 02 22:41:43 <[g2]> GW2348 Nov 02 22:41:50 [g2]: I already asked Gateworks tech support how I could get a hold of the source for the distributed software. Still no reply after a week. Nov 02 22:42:00 rwhitby: GW2348-4 Nov 02 22:42:32 <[g2]> GPSFan they ship the RedBoot source as part of their BSP package Nov 02 22:45:36 [g2]: they also distribute the linux kernel, busybox and other GPL software on their boards, with no source supplied and no offer of source either. Nov 02 22:46:01 <[g2]> GPSFan that's a snapgear distro Nov 02 22:46:10 <[g2]> they distribute the source with the BSP Nov 02 22:46:26 [g2]: how does one get a bsp? Nov 02 22:46:54 <[g2]> same way you get a board, call and order it Nov 02 22:48:00 [g2]: they could have said that in a reply to my e-mail, is all I'm saying. Nov 02 22:48:19 <[g2]> well maybe they are busy Nov 02 22:48:31 <[g2]> GPSFan you are absolutely right Nov 02 22:48:58 <[g2]> but sometimes it takes awhile to get an e-mail back Nov 02 22:49:10 <[g2]> re: dsaxena Nov 02 22:50:05 [g2]: that's ok I haven't gotten upset of bothered about it, since I was able to get a working kernel and userland built with minimal help. Then I disconered you here onnslu2-linux, and things got much easier and there was a community to boot. I'm delighted at the performance so far. Nov 02 22:50:50 <[g2]> yeah the hw rocks. Nov 02 22:51:10 <[g2]> GPSFan did you see the PMs ? Nov 02 22:51:26 several pings Nov 02 22:51:36 <[g2]> nod Nov 02 22:51:41 * [g2] misses the pongs Nov 02 22:51:50 <[g2]> I wanted to play ping-pong Nov 02 22:52:50 ah, well I was busy again today, and probably tomorrow too, it's getting on toward winter, and there's stuff outside that has to happen befor too long. Nov 02 23:37:46 <[g2]> dwery still up ? Nov 02 23:38:02 yes Nov 02 23:38:39 <[g2]> are you interested in JTAG or bootloaders at all ? Nov 02 23:39:15 yes Nov 02 23:39:58 <[g2]> the openexz guys are swearing by u-boot. specifically because of USB support Nov 02 23:40:09 oh Nov 02 23:40:20 <[g2]> I think the denx guy is adding the open NPE drivers to the ixdp425 platform Nov 02 23:40:45 <[g2]> we've got jtag'able platforms I've reloaded RedBoot many times Nov 02 23:41:06 <[g2]> we've got the p-e-r-f-e-c-t development platform Nov 02 23:41:36 <[g2]> the only thing missing is openocd support for hw debugging of the bootloader, the that may not be too far off Nov 02 23:42:12 <[g2]> although we probably don't need that at all in this case Nov 02 23:42:41 yep.. we are starting to get results! Nov 02 23:42:59 <[g2]> get results ? Nov 02 23:49:46 yep.. I mean.. all pieces ar getting together Nov 02 23:50:20 we need a second stage bootloader with IXP NPE support. Nov 02 23:51:35 <[g2]> rwhitby maybe you could port it for the second stage Nov 02 23:52:22 <[g2]> I think being able to have an image on USB and just have the bootloader automatically flash it would be handy Nov 02 23:52:42 yep - I've asked Marc about usb host support Nov 02 23:52:48 <[g2]> not to mention usb networking and possibly usb serial console Nov 02 23:53:08 <[g2]> rwhitby usb support is supposed to already be in u-boot Nov 02 23:53:53 ok, so we would then need skip region support in u-boot :-) Nov 02 23:54:31 * [g2] dunno about region support Nov 02 23:56:18 [g2]: it's the bit we added to Apex to allow it to skip the 16 byte sercomm header which ends up in the middle of a > 1MB kernel on the NSLU2, but is required for the first stage bootloader (Linksys RedBoot) to boot. Nov 02 23:56:45 <[g2]> rwhitby ah.. yeah Nov 02 23:57:00 (so only a problem for the NSLU2) Nov 03 00:02:48 <[g2]> I guess the USB support looks like mostly tty support Nov 03 00:03:02 <[g2]> it's is probably OK **** ENDING LOGGING AT Fri Nov 03 02:59:57 2006