**** BEGIN LOGGING AT Sun Jun 05 23:59:56 2005 Jun 05 23:59:56 I'll try the latest image as soon as it finishes building Jun 06 00:00:05 I saw it in a bulk changeset. Jun 06 00:00:06 and let you know Jun 06 00:00:25 we just need Tiersten to add the usb devices patch, and we've got a real R29-based unslung without the genesys rubbish they did Jun 06 00:00:52 we should then benchmark that with and without the interrupt-driven ixp driver Jun 06 00:01:14 next thing I'm going to do is put dropbear into the image Jun 06 00:02:31 and make the uNSLUng password stick everywhere Jun 06 00:03:02 If you have dropbear it's possible to auto-flash just by 'scp zImage /boot/zImage' and have a watcher process pick it up and flash (devio) it into the correct places. Jun 06 00:03:03 bbl Jun 06 00:03:26 jbowler-away: not on a running jffs2 though Jun 06 00:03:44 but yeah, with an external rootfs we could do that Jun 06 00:03:44 right... that's the precondition. Jun 06 00:04:03 that's why I want to get rid of the linksys web upgrade interface, so that we can eventually put a proper one in there. Jun 06 00:04:27 Hum. No it works on a running jffs2 rootfs (if there's space for the image) - just mount -o ro / Jun 06 00:05:01 jbowler-away: do we have devio in the unslung firmware yet? Jun 06 00:05:03 ... probably not reliable though Jun 06 00:05:22 you should make it a separate .bb and we will add it to unslung too Jun 06 00:05:26 devio: I'm moving it into a separate package Jun 06 00:05:42 cool - when you do, please add it to unslung Jun 06 00:06:05 In unslung-image? Jun 06 00:06:09 then you'll have to teach me how to use it one day Jun 06 00:06:14 yeah, in unslung-image Jun 06 00:06:26 I was going to write a man page too ;-) Jun 06 00:06:26 I assume it's pretty small Jun 06 00:06:49 ok, I'll be back later on .. Jun 06 00:07:08 It's 25540 bytes with the built-in help. The help text is about 10kbytes Jun 06 00:25:15 Well, it is hanging at S40networking, I may have damaged the udhcpc script (since I boot with DHCP). Jun 06 00:38:07 my build is still going Jun 06 00:40:37 The udhcpc script isn't even running on my system - so there is a hang before ifup tries to start it. Jun 06 00:44:14 hmm, I have static IPs on my slugs Jun 06 00:46:00 Yes, it's hanging in ifup, so I'll swap back to static and see it that helps... Jun 06 00:48:21 hmm, so what does openslug use the linksys firmware for ? Jun 06 00:48:31 my openslug build is downloading R29 Jun 06 00:49:05 that sounds like a bug Jun 06 00:49:09 RedBoot Jun 06 00:49:32 Oh yeah, we put it back in there dont we\/. Jun 06 00:50:04 Yes, along with SysConf. Some brave person with JTAG could try just putting /dev/zero in there - it should work ;-) Jun 06 00:51:40 well the way I flash it would work - fis write -f 0x50060000 -b 0x01060000 -l 0x7a0000 Jun 06 00:51:57 flashing Jun 06 00:53:07 booting... Jun 06 00:54:05 CRAP Jun 06 00:54:21 Hung? Jun 06 00:54:30 crap crap crap Jun 06 00:55:03 Alignment trap: ifup (799) PC=0x40140784 Instr=0xe5903004 Address=0xbeffff2a FSR 0x013 Jun 06 00:55:11 it's scrolling that so fast I can barely copy it Jun 06 00:55:42 Well, that's easy to turn off - alignment.sh Jun 06 00:55:42 (if I quit minicom it clears the screen) Jun 06 00:56:10 oh lol Jun 06 00:56:18 I hadn't thought to try ctrl-C Jun 06 00:56:29 that snapped it out of it Jun 06 00:57:08 But why would a compiler change create a mis-aligned access... Jun 06 00:57:46 network modules aren't loaded Jun 06 00:59:41 They wouldn't be if ifup crapped out before it touched eth0 Jun 06 00:59:51 even lo isn't up Jun 06 01:00:33 Same problem then - ifup is failing. Try echo 0 >/proc/cpu/alignment then re-run ifup -a Jun 06 01:00:46 ok, trying Jun 06 01:01:11 root@LKG0FB08F:/lib/modules/2.6.11.2/kernel/drivers/net# ifup -a Jun 06 01:01:11 Segmentation fault Jun 06 01:01:26 something is fscked Jun 06 01:02:37 ifup is in busybox (well, ifup is busybox) Jun 06 01:02:55 I manually configured eth0 and it works Jun 06 01:03:41 Ok, I remember seeing the add of "patches/ifupdown-spurious-environ.patch" Jun 06 01:04:20 "you make it, we break it guaranteed!" Jun 06 01:05:25 why are fat and vfat modules autoloaded ? Jun 06 01:05:36 I don't want/need those autoloaded Jun 06 01:05:43 this is new I think Jun 06 01:06:13 Because you didn't specify -t to turnup? Jun 06 01:06:55 It should only happen if you do 'mount /dev/sda2' (etc) and rely on the kernel to work out what type of FS it is. Jun 06 01:07:17 I haven't tried to mount anything, and don't have a disk attached Jun 06 01:07:18 I did fix turnup though so it doesn't do it (it always passes -t to mount). Jun 06 01:07:57 The patch removes 'static char **environ = NULL' from ifupdown.c (!) Jun 06 01:08:43 lol Jun 06 01:08:55 nice Jun 06 01:10:57 I don't see why the environ it picks up in exchange should be non-aligned (though obviously that would explain everything.) Jun 06 01:11:31 wish I had strace installed already Jun 06 01:12:42 Ah... "add gcc4 patches", and it is one of them. It clearly isn't spurious, since it hides the extern environ ;-) Jun 06 01:15:14 * jacques builds strace ipkg. Jun 06 01:16:23 ok this is bad Jun 06 01:16:26 strace used to work Jun 06 01:16:30 now I get: Jun 06 01:16:40 root@LKG0FB08F:/var/tmp# strace ls Jun 06 01:16:40 execve("/bin/ls", ["ls"], [/* 9 vars */]) = 0 Jun 06 01:16:40 syscall: unknown syscall trap 0xe5bcf620 Jun 06 01:17:00 WTF Jun 06 01:17:59 NOTE: Applying patch 'arm-syscallent.patch' Jun 06 01:18:04 apparently that did little good Jun 06 01:18:19 seems like all the oe patches are screwing us Jun 06 01:19:23 why didn't it apply the other two patches in the dir: arm-compilation.patch arm-syscall.patch Jun 06 01:20:23 It was changed on April 1 ;-) Jun 06 01:21:18 The change seems not to have been any change though. strace doesn't seem to have changed for a long time. Jun 06 01:21:39 My openslug is now past rcS (having un-spurioned the static char **environ) Jun 06 01:21:41 Booted. Jun 06 01:21:57 I thought this was a new version of strace from what I used last Jun 06 01:22:36 damn there are so many oe issues I am forgetting them Jun 06 01:22:39 Ah. The old one might have been bk rm'ed Jun 06 01:23:25 so how do we get that patch removed for our build? Jun 06 01:23:28 Weird. Somehow during the boot ext3, ext2, and all the fats do get loaded. Jun 06 01:23:39 patch: I'll commit my change. Jun 06 01:23:58 which file did you change? the busybox .bb ? Jun 06 01:24:07 put in a special case for NSLU2 ? Jun 06 01:27:00 Yes - changeset 1.3435 (just pushed) Jun 06 01:27:16 'k thanks Jun 06 01:28:15 got it Jun 06 01:29:04 There's going to be a merge conflict on that file when OE fixes it, but it should be fairly obvious how to resolve it. Jun 06 01:29:16 i agree Jun 06 01:42:34 flashing new image Jun 06 01:45:07 looking much better Jun 06 01:46:05 it's up with network interfaces Jun 06 01:46:18 still has those ugly fat modules tho :-P Jun 06 01:46:49 And ext2/ext3 Jun 06 01:47:37 Ah, it's that 1 block initrd which isn't anything. Jun 06 01:48:41 No. It isn't that - if I boot off (reiserfs) disk nothing is mounted. Jun 06 01:48:57 s/mounted/loaded/ Jun 06 01:50:32 yeah I have the ext[23] modules loaded too Jun 06 01:56:26 I bet it's the 'mount -o remount' in /etc/init.d/checkroot.sh Jun 06 01:58:10 you think it goes through vfat and ext2 before getting to jffs2 ? Jun 06 01:58:16 could be Jun 06 01:59:07 Yes - change 'auto' in /etc/fstab (on the flash partition) to 'jffs2' and the modules aren't loaded. Jun 06 02:00:17 That's an easy fix. Jun 06 02:00:36 Time for bed... Jun 06 02:00:58 nice work jbowler-zzz sleep well Jun 06 02:35:00 Attempting a build now. Jun 06 02:35:52 * rwhitby attempts a build - first time since the gcc debacle Jun 06 02:36:01 * dyoung me toos Jun 06 02:37:14 finally got my wl530g today Jun 06 02:37:42 cool Jun 06 02:37:42 just deciding whether to put it in as the internet router unchanged, allowing me to hack more on the wiley, or whether to open it up Jun 06 02:38:49 or alternately recompile the firmware for it with dropbear included .... Jun 06 02:45:44 it's got the nice asus restore mode, so bricking is not a concern Jun 06 02:46:03 and I do really need to fix the timezones on it anyway ... Jun 06 02:54:21 ooh - I just telnetted to it and got a command line Jun 06 02:54:33 Sash command shell (version 1.1.1) Jun 06 02:54:42 jacques, any new leads for some LWSF Action? Jun 06 02:55:22 Linux version 2.4.22-uc0 (root@localhost.localdomain) (gcc version 2.95.3 20010315 (release)(ColdFire patches - 20010318 from http://fiddes.net/coldfire/)(uClinux XIP and shared lib patches from http://www.snapgear.com/)) #1369 Wed Aug 18 21:32:58 CDT 2004 Jun 06 02:56:10 coldfire - I thought that was a motorola thing Jun 06 02:56:19 dyoung not yet :-\ Jun 06 02:56:34 coldfire is a Motorola/Freescale thing Jun 06 02:57:49 isn't that a MIPS device? Jun 06 02:57:58 the wl530g Jun 06 03:08:10 http://www.chupa.nl/phpwiki/index.php/WL530g Jun 06 03:08:21 will update that as I learn more tonight Jun 06 03:09:40 Did you void the warranty to take some internal pictures yet? Jun 06 03:10:34 the FCC entry has internal pictures Jun 06 03:10:36 IIRC, ColdFire is BE also. Jun 06 03:11:04 ColdFire is a 68K-style instruction set, not MIPS, nor ARM Jun 06 03:11:06 do you have the FCC Link handy? Jun 06 03:11:21 nope, but I remember it was easy to find on the fcc site Jun 06 03:12:09 MSQ is the Asus id Jun 06 03:12:22 so it's probably MSQWL530G Jun 06 03:12:43 yeah I am just wondering why that kernel has the coldfire patches Jun 06 03:13:09 https://gullfoss2.fcc.gov/prod/oet/cf/eas/reports/ViewExhibitReport.cfm?mode=Exhibits&RequestTimeout=500&calledFromFrame=N&application_id=983996&fcc_id='MSQWL530G' Jun 06 03:13:23 photos at https://gullfoss2.fcc.gov/prod/oet/forms/blobs/retrieve.cgi?attachment_id=483265&native_or_pdf=pdf Jun 06 03:14:21 bbiab Jun 06 03:14:57 weird, that one looks like the pictures for the 500. Jun 06 03:15:58 hmm - that's not riht Jun 06 03:16:00 right Jun 06 03:18:50 https://gullfoss2.fcc.gov/prod/oet/cf/tcb/reports/Tcb731GrantForm.cfm?mode=COPY&RequestTimeout=500&application_id=903874&fcc_id=MSQWL530G Jun 06 03:18:56 https://gullfoss2.fcc.gov/prod/oet/forms/blobs/retrieve.cgi?attachment_id=495191&native_or_pdf=pdf Jun 06 03:20:07 double decker PCB Jun 06 03:20:37 bbiab Jun 06 03:24:21 Maybe it is a coldfire in there, no pics with the sheild off of the CPU board. Jun 06 03:24:48 that kind of makes it interesting as another BE toy. Jun 06 03:33:52 dyoung: all the pictures are there - you can even read the chips Jun 06 03:34:43 marvell 88w8510 Jun 06 03:34:58 http://www.marvell.com/products/wireless/libertas/Libertas_88W8000G_88W8510.pdf Jun 06 03:35:09 jacques: I too wonder why it has coldfire patches Jun 06 03:35:27 I expect they are unrelated Jun 06 03:38:47 I added cat /proc/kmsg to the wiki page at http://www.chupa.nl/phpwiki/index.php/WL530g Jun 06 03:40:46 Those coldfire patches must have been added as part of the standard uclinux build. Jun 06 03:41:39 how'd your builds go? Jun 06 03:42:00 unslung finished Jun 06 03:42:17 cool Jun 06 03:42:18 modutils-cross failed for unslung for me Jun 06 03:42:27 Its still working on openslug Jun 06 03:42:38 I didnt actually flash these images yet though. Jun 06 03:45:33 I'll try an unslung build Jun 06 03:50:49 directory contents for wl530g are up (if anyone is still interested - let me know if you still are) Jun 06 03:54:45 sounds like a fun toy. Jun 06 04:10:02 it's just the thrill of the chase at the moment :-) Jun 06 04:13:03 Yeah, its true. Jun 06 04:13:40 I'm still tempted to buy a 2nd Actiontec for "testing". Jun 06 04:13:59 openslug built. Jun 06 04:26:46 modutils failed for me too in unslung Jun 06 04:27:41 openembedded/packages/modutils/files/gcc4.patch is not there Jun 06 04:28:42 gcc4.patch is in modutils-2.4.27/ I dunno why the build isn't finding it Jun 06 04:34:49 wonder why it worked for me.... Jun 06 04:35:56 how new is your bitbake? Jun 06 04:36:26 $ bitbake --version Jun 06 04:36:26 BitBake Build Tool Core version 1.2.2, bitbake version 1.2.9 Jun 06 04:36:54 I pilfered it from nudi around 1.5 weeks ago Jun 06 04:37:30 (it was one of the times that berlios svn was down) Jun 06 04:38:14 ok so that's probably not it then Jun 06 04:38:39 bitbake --version Jun 06 04:38:39 BitBake Build Tool Core version 1.3.0, bitbake version 1.3.1 Jun 06 04:39:15 interesting. Jun 06 04:52:07 * jacques goes to bed - later guys. Jun 06 04:56:44 night jacques Jun 06 05:26:32 any unslung packge experts around? Jun 06 05:27:40 03jeanfabrice * 10unslung/Makefile: Promote Neon, clips and sed for wiley Jun 06 05:36:21 * dyoung-zzzz heh hehs Jun 06 05:36:43 jf-work, youre a unslung package expert right? Jun 06 05:37:02 dyoung-zzzz.. humm not as good as jp30 I think Jun 06 05:39:10 Drat. hes on travel, so i cant bug him now. Jun 06 05:39:58 okay. He's the wizard and I'm the pupil so maybe I have a chance to help you Jun 06 05:40:31 I'm just trying to clean up some of the mess I made some months ago. Jun 06 05:40:58 ctrlproxy. I revisited it. Jun 06 05:41:34 its busted for reasons that completely escape me; becuase it builds as a openslug package okay. Jun 06 05:43:01 dyoung-zzzz: the problem is on your build env ? Jun 06 05:43:26 and it doesn't build as an unslung optware ? Jun 06 05:44:32 It gives me a whole slew of error messages. Jun 06 05:45:34 How can I help you ? do you want me to try building it on my own env. ? Jun 06 05:50:34 I am a bit curisous if it will build for wiley.... Jun 06 05:51:05 I can check Jun 06 05:52:45 cvs does not have sources/ctrlproxy Jun 06 05:59:24 crtlproxy.mk reference configure.in.patch. do we need or is it a typo ? Jun 06 06:00:34 oh well. Jun 06 06:00:37 I guess I'll give up for now Jun 06 06:05:23 03jeanfabrice * 10unslung/Makefile: Promote postgresql, rrdtool and libart for wiley Jun 06 06:10:16 03jeanfabrice * 10unslung/Makefile: correct typo in an optware name for wiley Jun 06 06:25:35 what could be the reason of segfaulting all apps after unsling? Jun 06 06:27:37 what version of unslung firmware? Jun 06 06:27:59 4.20b Jun 06 06:28:10 from the binary download? Jun 06 06:28:24 yes Jun 06 06:28:36 with a newly formatted disk? Jun 06 06:28:40 but the apps are not segfaulting while running Jun 06 06:28:57 the apps are segfaulting if i want to telnet or ssh to my box Jun 06 06:29:12 the the disk contains data, there is no way to format the disk Jun 06 06:30:04 and you ran "/sbin/unsling disk1" ? Jun 06 06:30:14 yes Jun 06 06:30:52 did you have any directories in the data partition which had the same name as the directories in the root filesystem? Jun 06 06:31:01 but i installed the package ldconfig, perhaps that's the problem Jun 06 06:31:46 no definitly not, but it would be better to copy the rootfs into /share/hdd/data/unslung instead of /share/hdd/data Jun 06 06:31:49 well, that should be an easy theory to test Jun 06 06:31:59 but there is no directory with the same name on my disk Jun 06 06:32:25 what happens if you ipkg remove ldconfig? Jun 06 06:33:58 thats not possible at the moment because i can only boot without a disk attached Jun 06 06:34:19 i connect the disk after booting but then it's not mounted at the right place Jun 06 06:34:50 look in the .ipk file to see what it installs, and then remove those files manually Jun 06 06:35:07 (assuming it didn't overwrite anything outside of /opt) Jun 06 06:39:01 i could do a mount -o bind /share/hdd/data/opt /opt Jun 06 06:39:44 hmm, but the ipkg database is wrong then Jun 06 06:41:01 the dirty way could be a chroot /share/hdd/data Jun 06 06:41:11 if i do that then the package list is ok Jun 06 06:41:32 should i do a chroot then remove the package and then exit? Jun 06 06:42:34 dunno - I know nothing about the ldconfig package Jun 06 06:44:22 Aiee, segfault! You should probably report this as a bug to the developer Jun 06 06:44:22 Connection to 10.20.10.100 closed. Jun 06 06:44:29 it's the same Jun 06 06:44:34 must be another reason Jun 06 06:45:01 start with a clean slate (reinstall firmware), and step forward until it reoccurs Jun 06 06:46:27 :/ Jun 06 06:47:36 that's why we mark it as beta .... Jun 06 06:58:26 03unslung * 10unslung/package-strip.pl: Added Jun 06 07:07:32 there is no problem with a beta, but i installed it twice and got two several problems, i think it's the wrong way to reinstall the firmware, is there a better way to debug the problem? Jun 06 07:09:19 unpack the ldconfig .ipk file and see what it installed. Look at the postinst script to see what it did. Undo those things. Jun 06 07:15:45 ok, i mounted the disk and the conf part and run dropbear manually and dont fork it into background Jun 06 07:15:59 [720] Jun 06 15:21:54 Not forking Jun 06 07:15:59 [721] Jun 06 15:21:56 Child connection from 10.20.2.100:51934 Jun 06 07:15:59 [721] Jun 06 15:22:00 password auth succeeded for 'root' Jun 06 07:15:59 [721] Jun 06 15:22:01 syslogin_perform_logout: logout(ttyp1) returned an error: No such file or directory Jun 06 07:15:59 [721] Jun 06 15:22:01 exit after auth (root): Exited normally Jun 06 07:16:21 that is the screen of the box Jun 06 07:16:31 Aiee, segfault! You should probably report this as a bug to the developer Jun 06 07:16:32 Connection to 10.20.10.100 closed. Jun 06 07:16:36 thats the message at the client Jun 06 08:50:01 which files should be in /share/hdd/conf? Jun 06 08:50:34 or better question, what is "format disk" of the webinterface doing with the conf partition? Jun 06 09:13:00 please can anybody take a look into /share/hdd/conf and tell me the few files which are in there? Jun 06 09:27:48 mr_claus: unslung4.2? Jun 06 09:27:58 yes Jun 06 09:28:18 but i only need to know whats copied to the partition to get the disk working again Jun 06 09:28:28 i deleted my conf stuff and dont find the backup files Jun 06 09:28:53 i dont need the config, i can configure it new but i want to use the disk again without formatting Jun 06 09:29:58 mr_claus: http://www.jrbnet.de/texte/ll.txt Jun 06 09:30:51 mr_claus: without .sda2root .unslung initrd Jun 06 09:31:23 thx Jun 06 09:33:15 whats inside of the subdirs? Jun 06 09:35:40 mr_claus: same link Jun 06 09:39:14 whats in share.info? is there an information about the disk? Jun 06 09:40:36 we should change the chanel to #nslu2-german, may be easier, and more on topic Jun 06 11:50:33 ~seen bob_tm Jun 06 11:50:49 bob_tm <~chatzilla@181.80-203-27.nextgentel.com> was last seen on IRC in channel #nslu2-linux, 12d 7h 19m 41s ago, saying: 'Thanks!!'. Jun 06 12:38:53 03ingeba * 10unslung/make/inetutils.mk: Added zlib to dependency list, since it would be used if staged before Inetutils configure is run Jun 06 15:11:05 mr_claus should try removing /etc/ld.so.cache Jun 06 15:41:41 Applying 1 revisions to packages/modutils/files/gcc4.patch Jun 06 15:41:50 i guess modutils should build again - I will try Jun 06 15:45:04 starting fresh unslung-image build Jun 06 16:26:51 hey rwhitby Jun 06 16:39:46 my fresh unslung-image build completed Jun 06 16:40:20 yeah, mine just finished too Jun 06 16:40:43 mine woulda been faster but it had to download R29 :-) Jun 06 16:40:46 <[g2]> was that a wipe of tmp or just a pull and build ? Jun 06 16:41:10 wipe of tmp and ccache Jun 06 16:41:16 as usual for me Jun 06 16:41:21 <[g2]> cool Jun 06 16:52:56 mine was just a pull Jun 06 16:53:23 starting fresh clean openslug-image build Jun 06 16:57:04 heh - I have the patience of a saint :-) Jun 06 16:57:18 <[g2]> you do Jun 06 16:57:29 <[g2]> I read many of your responses on the ml :) Jun 06 18:14:16 openslug build completed a bit ago Jun 06 19:32:32 rwhitby-away: Running 5.1. Sent you a code snippet that uses an autoadjusting sleep for drive ready. Jun 06 19:32:49 It works for me. Let me know what you think. Jun 06 19:38:40 We need an unslung 5.x database. Jun 06 19:40:49 ka6sox-away: rwhitby-away: We need a 5.x database. Jun 06 19:43:21 good point Jun 06 20:23:09 Thanks dyoung. Jun 06 20:24:13 I sent a code snippet to Rod, but maybe I should have sent it to you. Jun 06 20:51:34 hmm something's broke with ssh in openslug Jun 06 20:51:53 root@LKG0FB08F:~# ssh 67.171.198.193 Jun 06 20:51:53 ssh: connection to root@67.171.198.193:22 exited: No auth methods could be used. Jun 06 20:51:58 this used to work Jun 06 20:52:12 as did strace, etc etc :-( Jun 06 20:54:18 I don't even know how to get error logging from ssh on the slug Jun 06 21:16:24 OK, at some point gentoo changed my sshd config Jun 06 21:16:39 to have this: PasswordAuthentication no Jun 06 21:16:47 not too useful in this case Jun 06 21:45:11 * jacques decides to read the turnup source. Jun 06 22:13:06 hmmm, is there supposed to be a loopback route in openslug? Jun 06 22:13:24 because there aint one Jun 06 22:36:43 wow, turnup is cool Jun 06 22:40:14 Yeah. Makes me wonder if jbowler is human, superhuman, or robot. Jun 06 22:40:38 while youre in there, look at devio.c too Jun 06 22:40:43 thats pretty cool too. Jun 06 22:41:04 says jbowler, "I got carried away..." Jun 06 22:42:30 on the other hand, the openslug ipkg repos seems rather broken Jun 06 22:42:56 I can see the strace ipk on the site, but it's not in the list Jun 06 22:43:10 http://ipkg.nslu2-linux.org/feeds/openslug/oe/strace_4.5.9-r1_armeb.ipk Jun 06 22:43:57 a-fscking-ha! Jun 06 22:44:08 and, as I thought, this version of strace DOES work Jun 06 22:44:17 unlike what's broken in oe right now Jun 06 22:45:40 (I installed the ipk manually) Jun 06 22:52:55 thankfully noone's blown away that repo. Jun 06 22:53:12 yeah Jun 06 23:20:39 * jacques wonders why procps isn't in openslug-packages Jun 06 23:24:02 * jacques decides to build his own. Jun 06 23:26:08 * jacques gets lonely since it's so quiet in here. Jun 06 23:27:09 ah Jun 06 23:27:20 procps isn't in openslug-packages because it doesn't build :- Jun 06 23:27:34 I thought it used to... Jun 06 23:28:09 fails the package step Jun 06 23:28:28 install -D --mode a=rx --strip proc/libproc.so.3.2.1 /home/packages/oe/nslu2/bb_openslug_default/tmp/work/procps-3.2.1-r1/image/lib//libproc.so.3.2.1 Jun 06 23:28:28 strip: Unable to recognise the format of the input file /home/packages/oe/nslu2/bb_openslug_default/tmp/work/procps-3.2.1-r1/image/lib//libproc.so.3.2.1 Jun 06 23:28:28 install: strip failed Jun 06 23:28:44 MF POS bastage dumbass.... Jun 06 23:29:03 S-fscking-igh!!! Jun 06 23:54:25 is it not big-endian-ified or something? Jun 06 23:54:49 install --strip will never work when cross-building Jun 06 23:54:52 I think it used to work. Jun 06 23:54:58 because it uses the native strip command Jun 06 23:55:10 oh yeah. Jun 06 23:55:14 it's something we have to patch in many many apps in unslung cvs Jun 06 23:55:26 right. Jun 06 23:55:37 and something frankly the oe devels should now have allowed to get through Jun 06 23:55:40 not Jun 06 23:56:06 you can cheat though right? And do make install-strip or something like that, whihc inheriets the $STRIP from someplace? Jun 06 23:56:12 or am I makingi that up.... Jun 06 23:56:13 I'm in a bad mood right now becauwse I seem to have lost my openslug devel env configs Jun 06 23:56:42 you just do the strip as a separate command and use the cross-strip in your toolchain Jun 06 23:59:14 Oh yeah, I'm remembering a differnt thing for cheating about. **** ENDING LOGGING AT Mon Jun 06 23:59:56 2005