--- Log opened Sun Aug 19 00:00:03 2007
00:45 -!- sepp [n=sepp@Z45d4.z.pppool.de] has quit ["Lost terminal"]
01:34 -!- wbx [n=wbx@static-87-79-68-60.netcologne.de] has joined #t2
02:22 -!- wbx [n=wbx@static-87-79-68-60.netcologne.de] has quit ["leaving"]
02:25 -!- tri__ [n=tri@p57ADC33D.dip0.t-ipconnect.de] has joined #t2
02:39 -!- yokoy [n=yokoy@p57BC3E2E.dip0.t-ipconnect.de] has joined #t2
02:44 -!- tri_ [n=tri@p57AD9890.dip0.t-ipconnect.de] has quit [Read error: 110 (Connection timed out)]
02:51 -!- yokoy_ [n=yokoy@p57BC1A38.dip0.t-ipconnect.de] has quit [Read error: 110 (Connection timed out)]
05:33 -!- sepp [n=sepp@Z64d2.z.pppool.de] has joined #t2
07:34 -!- emte [n=emte@66.183.8.167] has joined #t2
07:42 -!- emte [n=emte@66.183.8.167] has quit [Remote closed the connection]
07:43 -!- emte [n=emte@66.183.8.167] has joined #t2
08:39 -!- emte [n=emte@66.183.8.167] has quit [Remote closed the connection]
08:44 -!- emte [n=emte@66.183.8.167] has joined #t2
08:56 -!- emte [n=emte@66.183.8.167] has quit [Remote closed the connection]
09:07 -!- DeeKey [n=denis@84-50-19-235-dsl.kjj.estpak.ee] has left #t2 []
09:07 -!- DeeKey [n=denis@84-50-19-235-dsl.kjj.estpak.ee] has joined #t2
09:11 < DeeKey> hi
10:26 < rxr> hi
10:36 < CIA-24> rene * r25676 /trunk/package/editors/zile/zile.desc: * updated zile (2.2.40 -> 2.2.41)
10:37 < tri__> moin
10:37 < CIA-24> rene * r25677 /trunk/package/graphic/libpng/libpng.desc: * updated libpng (1.2.18 -> 1.2.19)
10:37 < rxr> hi tri__
10:53 < CIA-24> rene * r25678 /trunk/package/ (12 files in 12 dirs):
10:53 < CIA-24> * bbkeys, ccrtp, gnome2-user-docs, ipsec-tools, libofx, libtomcrypt,
10:53 < CIA-24> libtomfastmath, libzrtpcpp, openssh, pearpc, pmount and vice to cross
10:53 < CIA-24> build
11:00 < CIA-24> susan * r25679 /trunk/package/doc/howtos/howtos.desc: * updated howtos (20070811 -> 20070818)
11:05 < CIA-24> susan * r25680 /trunk/package/develop/hte/hte.desc: * updated hte (2.0.8 -> 2.0.9)
11:05 < CIA-24> susan * r25681 /trunk/package/security/axtls/axtls.desc: * updated axtls (1.1.4 -> 1.1.6)
11:06 < DeeKey> Strange thing happens, when I use var_append GCC_WRAPPER_INSERT " " "-lm"
11:06 < DeeKey> argument "-lm" is added to the end of the string! Like, I was using APPEND.
11:06 < DeeKey> http://pastebin.com/m3e29f135
11:11 < rxr> when you add it with the T2 wrapper you will not see it in the log anyway,
11:11 < rxr> the -lm in your log comes fom the makefile or so
11:11 < rxr> you see what the T2 wrapper does in the bebinning
11:11 < rxr> beginning of the log
11:12 < rxr> as there we call gcc once with GCC_WRAPPER_DEBUG=1 yet
11:12 < rxr> so you should have a trace there
11:52 < CIA-24> susan * r25682 /trunk/package/emulators/vice/vice.desc: * updated vice (1.21 -> 1.22)
12:25 < sepp> moin :)
12:28 < rxr> moin moin sepp
12:28 [Users #t2]
12:28 [@ChanServ] [ Dallur ] [ Fatal ] [ mtr ] [ rxr ] [ tri__ ]
12:28 [ CIA-24 ] [ DeeKey ] [ LMJ ] [ pdinoto_ ] [ sepp ] [ valentin]
12:28 [ cyda ] [ Enqlave] [ mqueiros] [ Ragnar|away] [ TobiX] [ yokoy ]
12:28 -!- Irssi: #t2: Total of 18 nicks [1 ops, 0 halfops, 0 voices, 17 normal]
12:29 < sepp> moin rxr
12:36 < DeeKey> hello all!
12:37 < sepp> hi DeeKey :)
13:44 [Users #t2]
13:44 [@ChanServ] [ Dallur ] [ Fatal ] [ mtr ] [ rxr ] [ tri__ ]
13:44 [ CIA-24 ] [ DeeKey ] [ LMJ ] [ pdinoto_ ] [ sepp ] [ valentin]
13:44 [ cyda ] [ Enqlave] [ mqueiros] [ Ragnar|away] [ TobiX] [ yokoy ]
13:44 -!- Irssi: #t2: Total of 18 nicks [1 ops, 0 halfops, 0 voices, 17 normal]
14:03 < DeeKey> strange, recently comedilib was updated to version 0.8.0, but i am not able to compile it.
14:03 < DeeKey> maybe it is due to I am using uclibc?!
14:04 < DeeKey> I have the same compiler error as described here: https://mail.comedi.org/pipermail/comedi/2007-May/008477.html
14:04 < DeeKey> but I cannot solve it..
14:06 < sepp> just -lm missing?
14:07 < sepp> var_append GCC_WRAPPER_APPEND ' ' '-lm'
14:07 < DeeKey> yeh, but I have already tryed to add -lm as a wrapper!
14:07 < sepp> that does not work?
14:07 < DeeKey> nope
14:07 < sepp> hmmm
14:07 < DeeKey> and INSERT also dows not work
14:08 < sepp> you can test it with debug.sh
14:08 < DeeKey> same result
14:09 < sepp> then i don't know
14:09 < DeeKey> could that be uclibc problem?
14:09 < sepp> i never used uclibc
14:55 < DeeKey> Why comedilib (in t2 just comedi) is being built in the first stage?! It is configured to be built in the 5-th stage!
15:06 < sepp> do you run a cross build?
15:07 < sepp> otherwise it should not
15:22 < rxr> re
15:23 < rxr> could be, yes
15:24 < rxr> hm - uClibc appears to have nearbyint, at least when I grep thru it's sources
15:29 < DeeKey> no, i do not run cross build
15:41 < rxr> my could be was for the line when I read "maybe not in uclibc" -)
15:41 < rxr> :-)
15:41 < rxr> the trunk uclibc appears to have this function
15:41 < rxr> you can check with objdump -T or so on the uclibc libm binary
15:44 < DeeKey> you propose to checkout UCLIBC from trunk? Actually I have checked the version of uclibc
15:44 < DeeKey> it is the latst
15:44 < DeeKey> latest
15:47 < rxr> no, I meant the release version we have in trunk appears to have the symbol
15:47 < rxr> I propose to check if your binary of libm.so or .a has the symbol objdump -t or -T will tell you
15:47 < rxr> you said you checked insert as well - but maybe the -lm must be in the beginning or so as the order is sometimes important
15:52 < CIA-24> rene * r25683 /trunk/package/ (5 files in 5 dirs):
15:52 < CIA-24> * marked libspectrum, pamtester, smbldap-tools, xf86-input-evtouch,
15:52 < CIA-24> xf86-input-fujitouch to cross build
15:53 < DeeKey> Insert puts -lm at the beginning of the string, isnt it?
15:53 < rxr> depending what insert
15:53 < rxr> GCC_WRAPPER_INSERT, yes :-)
15:56 < DeeKey> the only string in comedi.conf is:
15:56 < DeeKey> var_append GCC_WRAPPER_APPEND " " "-lm"
15:56 < rxr> you could try var_append GCC_WRAPPER_INSERT " " "-lm"
15:56 < rxr> to see if it makes a difference
15:58 < DeeKey> BTW: why comedilib is build in first stage?
15:58 < DeeKey> == 16:57:17 =[1]=> Building scientific/comedi [0.8.0 7.0-trunk].
15:58 < DeeKey> with insert same result as with append
16:00 < rxr> and in the top of the log file you have a log trace that -lm is inserted by the wrapper, correct ?
16:01 < DeeKey> I will post log
16:06 < DeeKey> http://pastebin.com/me17a6ac
16:07 < rxr> Insert: -lm
16:07 < DeeKey> I dont like the string INSERT = '--help?:-pipe -Os -march=pentium2 -lm'
16:07 < DeeKey> where -lm is at the end
16:07 < DeeKey> I used insert lm!
16:08 < rxr> you used var_append GCC_WRAPPER_INSERT
16:08 < rxr> but that is no problem it ends up after -Os and -march=pentium2 in the arguments
16:08 < rxr> that is ok
16:08 < DeeKey> yes
16:09 < rxr> maybe there is a problem with the symbol in the libm, maybe uclibc does not support it ?
16:09 < rxr> objdump -T build/.../lib/libm.so* | grep nearby
16:12 < DeeKey> strange, I dont have libm.so in my build, but I have it in my distro:
16:12 < DeeKey> locate libm.so
16:12 < DeeKey> /lib/tls/i686/cmov/libm.so.6
16:12 < DeeKey> /lib/libm.so.6
16:12 < DeeKey> /usr/lib/libm.so
16:12 < DeeKey> /usr/i386-uclibc-linux/lib/libm.so.0
16:12 < DeeKey> /usr/i386-uclibc-linux/lib/libm.so
16:12 < rxr> hm
16:12 < rxr> in that in build/... ?
16:13 < rxr> haha:
16:13 < rxr> http://www.efytimes.com/efytimes/21160/news.htm
16:13 < rxr> interview with Linus Torvalds:
16:13 < DeeKey> no locate searches in the whole distro!
16:13 < rxr> Microsoft OLE DB Provider for ODBC Drivers error '80004005'
16:13 < rxr> [Microsoft][ODBC SQL Server Driver][SQL Server]Transaction (Process ID 72) was deadlocked on lock resources with another process and has been chosen as the deadlock victim. Rerun the transaction.
16:13 < rxr> /efytimes/lefthome.asp, line 284
16:13 < rxr> :-(((
16:13 < rxr> dont you locate
16:13 < rxr> look at build/.../lib
16:14 < DeeKey> ok
16:37 -!- emte [n=emte@66.183.8.167] has joined #t2
16:44 < rxr> hi emte
16:44 [Users #t2]
16:44 [@ChanServ] [ DeeKey ] [ LMJ ] [ Ragnar|away] [ tri__ ]
16:44 [ CIA-24 ] [ emte ] [ mqueiros] [ rxr ] [ valentin]
16:44 [ cyda ] [ Enqlave] [ mtr ] [ sepp ] [ yokoy ]
16:44 [ Dallur ] [ Fatal ] [ pdinoto_] [ TobiX ]
16:44 -!- Irssi: #t2: Total of 19 nicks [1 ops, 0 halfops, 0 voices, 18 normal]
16:44 < rxr> puh, so I cleaned my laptop t2/trunk again
16:45 < rxr> had 400++ noisy local changes due to ecessive CROSS testing ...
16:45 < rxr> back to sane levels :-)
16:46 < emte> hey
16:46 < emte> brb
16:47 -!- emte [n=emte@66.183.8.167] has quit [Remote closed the connection]
16:49 < DeeKey> objdump -T build/.../lib/libm.so* | grep nearby
16:49 < DeeKey> does not give nay output
16:49 < DeeKey> but in ../lib there is libm.so.0 link to
16:50 < DeeKey> libm-0.9.29.so
16:52 < rxr> is the symbol in it ?
16:52 < rxr> maybe uclibc has some "WITH FULL FEATURED LIBM" option or so ..
16:54 < DeeKey> no
17:05 < rxr> my ARM uclibc has it neither
17:06 < rxr> DO_C99_MATH
17:06 < rxr> apparently you need to config uclibc with DO_C99_MATH in order to get those functions
17:06 < rxr> maybe we should make that the default, hm?
17:07 < rxr> and people can deselect it to further strip uclibc, down ?
17:07 < rxr> you can create a target/../uclibc.config
17:07 < rxr> with
17:07 < rxr> X DO_C99_MATH
17:07 < rxr> and rebuild uclibc, that should give you said function
17:18 < CIA-24> rene * r25684 /trunk/package/xorg/beforelight/destdir.patch: * fixed beforelight DESTDIR handling for cross builds
17:53 < CIA-24> rene * r25685 /trunk/package/ (20 files in 10 dirs):
17:53 < CIA-24> * fixed enblend, libnjb, typespeed, beforelight, pam_pkcs11, rssh,
17:53 < CIA-24> opendbx, libtextcat, dfu-programmer, gnuchess, and sipsak to cross
17:53 < CIA-24> build
17:57 < rxr> 1364 packaged marked CROSS
18:10 < CIA-24> rene * r25686 /trunk/package/base/suspend/suspend.desc: * marked suspend x86, x86-64 only
18:30 < CIA-24> rene * r25687 /trunk/package/kde/amarok/amarok.desc: * updated amarok (1.4.6 -> 1.4.7)
18:48 < CIA-24> rene * r25688 /trunk/package/graphic/lcms/lcms.desc: * updated lcms (1.16 -> 1.17)
18:49 < CIA-24> rene * r25689 /trunk/package/network/pmacct/ (. pmacct.cache pmacct.desc): * added pmacct (0.11.2) - A libpcap, NetFlow, and sFlow IPv4/IPv6 accounting package
18:52 < CIA-24> rene * r25690 /trunk/package/editors/ed/ed.desc: * updated ed (0.7 -> 0.8)
19:30 < DeeKey> Hi rxr. Just returned from a bicycle ride.
19:31 < DeeKey> do I understand you correctly
19:32 < DeeKey> I need to create you can create a file in target dir (for me target/rtai/uclibc.config)
19:32 < DeeKey> and insert the string "X DO_C99_MATH" in it?
19:33 < rxr> yes, for now
19:34 < rxr> if it fixes your problem I think we probably should change the default uclibc to build with that enabled by default
19:35 < DeeKey> ok I will try to build it. Hope it helps :)
19:35 < CIA-24> rene * r25691 /trunk/package/cad/brlcad/brlcad.desc: * updated brlcad (7.10.0 -> 7.10.2)
20:49 -!- happycube [n=cpage@netblock-68-183-40-90.dslextreme.com] has joined #t2
20:59 < DeeKey> rxr: X DO_C99_MATH did not help
20:59 < DeeKey> http://pastebin.com/m6d4943ad
21:00 -!- tri__ [n=tri@p57ADC33D.dip0.t-ipconnect.de] has quit [Read error: 104 (Connection reset by peer)]
21:00 < DeeKey> Should it be written in logs, that uclibc.conf is being used?
21:06 < sepp> the things the wrapper appends are not listed in the gcc -c bla ... bla log
21:06 < sepp> just at the beginning
21:15 < DeeKey> no,i was talking about the method which rxr pointed to me:
21:15 < DeeKey> (06:06:46 PM) rxr: you can create a target/../uclibc.config
21:15 < DeeKey> (06:06:49 PM) rxr: with
21:15 < DeeKey> (06:06:53 PM) rxr: X DO_C99_MATH
21:29 < sepp> ok
21:39 < LMJ> moin moin
21:39 < DeeKey> hi
23:25 -!- yokoy [n=yokoy@p57BC3E2E.dip0.t-ipconnect.de] has quit [Read error: 110 (Connection timed out)]
23:27 < happycube> weird... uclibc isn't cross-compiling (for arm) on my amd64
23:27 < happycube> i'm trying to build a neo1973 image
23:27 < happycube> You should not call uname during the build!
23:27 < happycube> You should not call uname during the build!
23:27 < happycube> oops
23:28 < happycube> /bin/sh: line 1: 2516 Segmentation fault arm-t2-linux-gnueabi-ld --warn-unresolved-symbols -o /dev/null -b binary /dev/null >/dev/null 2>&1
23:28 < happycube> seems to actually build ok other than that
23:35 < DeeKey> You should not call uname during the build!
23:35 < DeeKey> this is ok...
23:36 < DeeKey> it is not a T2 project... These are common wornings.
--- Log closed Mon Aug 20 00:00:04 2007