**** BEGIN LOGGING AT Sat Feb 04 10:59:57 2006 Feb 04 17:58:50 koen, who would be the right person to contact for a bug in ipkg (or where should I report it). Feb 04 17:59:00 to make it sweeter, I'll include the fix as well... Feb 04 17:59:13 Post it on the OE bugtracker Feb 04 17:59:14 eFfeM: pigi@frumar.it Feb 04 17:59:23 or post it in the hh.org bugtracker Feb 04 17:59:24 tnx Feb 04 17:59:29 oh Feb 04 17:59:33 ok Feb 04 18:03:18 found why ipkg sometimes prints garbage chars when installing/upgrading Feb 04 18:03:39 oh, why? :)= Feb 04 18:04:14 char message_out[15] ; Feb 04 18:04:14 if ( message ) Feb 04 18:04:14 strncpy( message_out,"Upgrading ",strlen("Upgrading ") ); Feb 04 18:04:26 message_out is on the stack so can containa anything Feb 04 18:04:47 and strnpy will add a 0 byte as strlen will return the exact # of chars Feb 04 18:05:00 better use strcpy, also saves a few cycles for strlen Feb 04 18:06:30 which version is that? Feb 04 18:06:58 0.99.155, the most current one in oe Feb 04 18:07:08 actually the number does not even appear in bugzilla Feb 04 18:07:35 eFfeM: http://handhelds.org/cgi-bin/cvsweb.cgi/familiar/dist/ipkg/C/ChangeLog?rev=1.218&content-type=text/x-cvsweb-markup Feb 04 18:08:16 maybe .156 fixes the issues, but I'm not sure Feb 04 18:09:36 i think 156 will fix it: Fixed a length problem for strncpy when "Installing" option added. This fix bug #1456. Thanks to hrw for signaling Feb 04 18:09:44 i'll check Feb 04 18:12:38 if it doesn't fix it assign the bug to pigi and he'll look into it Feb 04 18:20:38 koen, found out that the latest version is 158, I'm now rebuilding that one, will add/commit the 158 bb file if it works Feb 04 18:21:28 ok Feb 04 18:21:39 I'm waiting for .159 Feb 04 18:22:17 http://handhelds.org/~bugzilla/show_bug.cgi?id=1515 Feb 04 18:23:30 hm, made the bb file but it does not build. Feb 04 18:23:33 [slug@Woonkamer ipkg]$ cat ipkg_0.99.158.bb Feb 04 18:23:33 include ipkg.inc Feb 04 18:23:33 PR = "r0" Feb 04 18:23:53 exactly as in the 155.bb file (which uses PV etc Feb 04 18:24:20 btw: does not build means: bb ipkg does not start the build process Feb 04 18:26:16 the odd thing is that bb -b ipkg_0.99.158.bb does start the build Feb 04 18:27:50 koen, checked the src, it fixes the problem but not in a very elegant way. Feb 04 18:27:54 now the code reads: Feb 04 18:27:57 char message_out[15] ; Feb 04 18:27:57 memset(message_out,'\x0',15); Feb 04 18:27:57 if ( message ) Feb 04 18:27:57 strncpy( message_out,"Upgrading ",strlen("Upgrading ") ); Feb 04 18:28:18 still feel that leaving out the memset and replacing strncpy by strcpy is much much much better Feb 04 18:31:45 hm cleaned 158, retried bb ipkg, still no action Feb 04 18:34:28 i think i'm lost with bb. did a bb -cclean ipkg followed by a bb ipkg, and it still builds 155 even though there is a 158.bb file Feb 04 18:34:32 why does this happen? Feb 04 21:25:12 koen, something wrong with packages not propagating to OE? Or am I just impatient? Feb 04 21:25:33 can't find Revision: 8fb0e135d3da67bc1fa37bd6ad67902d8cf02bcb in CiA; checked it in at 1407 Feb 04 21:25:41 or did i do something wrong? Feb 04 21:25:53 * koen checks Feb 04 21:26:17 * koen spots the error Feb 04 21:26:25 gr8 Feb 04 21:26:32 the nslu2 people stopped serving the dreambox branch Feb 04 21:26:43 without telling me to update the sync scripts Feb 04 21:27:13 ah, ok, so the polling script stopped working Feb 04 21:37:02 anybody an idea why madplay is not in the unstable feed while it is in the conf file Feb 04 21:38:33 hmm Feb 04 21:38:38 has it been added recently? Feb 04 21:38:46 (the autobuilds are currently failing) Feb 04 21:40:54 it is in slugos-packages, i think it was there since the creation of slugos-packages Feb 04 21:42:47 cdstatus is also not in the feed, but it is in slugos-packages Feb 04 21:43:30 (due to a disk problem I'm rebuilding my system so I am bumping into those issues) Feb 04 21:43:31 hmm Feb 04 21:43:34 odd Feb 04 21:43:54 I'd check it right now if I didn't forget my laptop PSU when I went home :-P Feb 04 21:44:11 As soon as I get coLinux up and running, I'll check it out Feb 04 21:46:42 it doesn't bother me very much, just build the packages myself, but it is not the way it should be Feb 04 21:47:46 btw i did glance over the groff issues in slugbug. I seem to recall I had one of them a while ago, but last week when I recompiled it, it worked like a charm for me Feb 04 21:48:34 maybe it's time to retry putting it in the feed Feb 04 21:48:36 It'd be nice Feb 04 22:00:09 i definitely like it that man works on the slug Feb 04 22:02:20 I'll check and see if it compiles on nudi now Feb 04 22:02:46 good Feb 04 22:03:32 Of course, even if it compiles doesn't mean that it actually works :-P Feb 04 22:03:43 but if it works, I'll try backporting to 2.7-beta Feb 04 22:04:34 NOTE: package groff-1.19.2: completed Feb 04 22:04:34 NOTE: build 200602041401: completed Feb 04 22:04:40 Looks OK. Feb 04 22:12:34 NAiL, if there is a runtime error it is probably because man and/or groff is picky on where the macro files (tmac.n or so) is stored **** ENDING LOGGING AT Sun Feb 05 10:59:59 2006