| # 00:08:36 |
imMute |
quits : Ping timeout: 260 seconds |
| # 00:12:38 |
ccole |
quits : Quit: leaving |
| # 00:14:51 |
imMute |
joins #crosstool-ng |
| # 00:14:52 |
imMute |
quits : Changing host |
| # 00:14:52 |
imMute |
joins #crosstool-ng |
| # 01:27:21 |
Doug_ |
joins #crosstool-ng |
| # 01:31:33 |
Doug_ |
quits : Client Quit |
| # 02:34:52 |
mnt_real |
quits : Quit: Ex-Chat |
| # 04:49:27 |
sfan5|OFF |
is now known as: sfan5 |
| # 06:36:09 |
bhundven |
joins #crosstool-ng |
| # 06:38:04 |
bhundven |
should ./configure fail if zlib1g-dev is not installed on the build host? |
| # 06:39:12 |
bhundven |
I'm working on a fix for elf2flt, because it does depend on host being installed |
| # 06:39:42 |
bhundven |
but I'd rather not be specific if that should just be installed anyways. |
| # 07:43:12 |
smartin_ |
joins #crosstool-ng |
| # 08:29:00 |
bhundven |
quits : Quit: Leaving. |
| # 09:26:42 |
dalias |
quits : Ping timeout: 258 seconds |
| # 09:42:59 |
dalias |
joins #crosstool-ng |
| # 09:43:22 |
y_morin |
joins #crosstool-ng |
| # 09:47:38 |
dalias |
quits : Ping timeout: 245 seconds |
| # 09:50:48 |
dalias |
joins #crosstool-ng |
| # 10:15:00 |
Weasel[DK] |
quits : Ping timeout: 258 seconds |
| # 10:21:31 |
muep |
joins #crosstool-ng |
| # 10:34:45 |
smartin_ |
y_morin: hi |
| # 10:38:29 |
smartin_ |
does the companion libs checks work? i mean ppl seems not really helpful with this since it requires -fpermissive, but the checks fail because of the build done with -fpermissive... :( |
| # 10:46:35 |
y_morin |
smartin_: Hmmm. Patch? ;-) |
| # 10:46:54 |
y_morin |
smartin_: No, it's een a long time since I last ran th complibs checks. |
| # 10:47:25 |
y_morin |
smartin_: ppl took like 1h30' here, when the toolchain would otherwise be built in ~20' |
| # 10:48:11 |
y_morin |
smartin_: I'm in my hack-ct-ng-like-crazy WE, so I'll get to it in a moment... |
| # 13:15:57 |
smartin_ |
y_morin: nice, let me know if you want my build log and/or defconfig |
| # 13:58:52 |
alan_o |
joins #crosstool-ng |
| # 14:37:56 |
kos_tom |
y_morin: "hack-ct-ng-like-crazy WE". Sounds fun :-) |
| # 14:40:36 |
y_morin |
kos_tom: No much. I'm so late, with so much back-buffer... :-( |
| # 15:34:30 |
smartin_ |
y_morin: i'm testing some fix... unfortunately my machine seems slower than yours ;) |
| # 16:26:23 |
smartin_ |
y_morin: i think i have a patch, ppl check is done and the build is still running :) |
| # 16:33:39 |
doc2 |
joins #crosstool-ng |
| # 17:09:55 |
y_morin |
smartin_: Good! :-) |
| # 17:27:13 |
plfiorini |
quits : Ping timeout: 245 seconds |
| # 17:47:13 |
smartin_ |
always wonders how to do a pull --rebase with hg... |
| # 17:50:49 |
bhundven |
joins #crosstool-ng |
| # 17:54:42 |
y_morin |
uses MQs for that. |
| # 17:55:45 |
y_morin |
smartin_: ^^^, although I have to agree that now I know about 'git rebase --onto' and 'git rebase -i', it realy something I'm miossing in Mercurial. |
| # 17:56:26 |
y_morin |
Hg does have a rebase, but I have never been really tempted to use it so far. |
| # 18:08:45 |
smartin_ |
y_morin: well... i find my way out ;) |
| # 18:09:14 |
smartin_ |
y_morin: you've just got a mail :) |
| # 18:10:29 |
smartin_ |
btw, my build successfully ended few minutes ago |
| # 18:11:11 |
kos_tom |
y_morin: I'm sure you'll end up using git for ct-ng :-) |
| # 18:12:34 |
smartin_ |
just sees a flying-troll... |
| # 18:14:42 |
kos_tom |
smartin_: where ? :-) |
| # 18:15:35 |
smartin_ |
in Paris sky, don't you see at Toulouse? and at Rennes? |
| # 18:16:42 |
smartin_ |
s/see at/see it at/ |
| # 18:18:36 |
smartin_ |
kos_tom: is your brother maintaining code.bulix? I sent him a patch for paste.py |
| # 18:43:48 |
kos_tom |
smartin_: yes, my brother is maitining code.bulix.org |
| # 19:32:58 |
y_morin |
smartin_: troll's still flying around! ;-) |
| # 20:31:41 |
y_morin |
smartin_: what's your host distro? |
| # 20:39:06 |
smartin_ |
y_morin: for today's builds, it a debian squeeze chroot |
| # 20:40:01 |
y_morin |
smartin_: I'm running on Squeeze, too, and never had your -fpermissive issue. |
| # 20:42:20 |
smartin_ |
well, it's more like a warning turned into an error during the check phase. |
| # 20:44:42 |
y_morin |
smartin_: Ah, right, during the checks, right, forgot it, right, thanks! :-) |
| # 21:14:32 |
y_morin |
smartin_: OK, check performed OK here, will commit soonish. Thanks! |
| # 21:17:08 |
smartin_ |
y_morin: you're welcome |
| # 21:50:52 |
sfan5 |
is now known as: sfan5|OFF |
| # 21:51:43 |
smartin_ |
quits : Quit: leaving |
| # 22:19:55 |
bhundven |
hey yann |
| # 22:25:55 |
y_morin |
bhundven: Howdy! LTNS! :-) |
| # 22:38:45 |
plfiorini |
joins #crosstool-ng |
| # 22:38:52 |
bhundven |
:) |
| # 22:40:15 |
bhundven |
y_morin: I found that elf2flt requires zlib1g-dev on debian. I was adding a check to configure.ac to flip a kconfig option, but then I asked last night here if I should just fore-go the kconfig part, and just bail if is missing? |
| # 22:43:24 |
y_morin |
bhundven: not in configure.ac: elf2flt runs on host machine, while ./configure can only check on the build machine. |
| # 22:43:41 |
y_morin |
bhundven: so, if you want to add a check, it can only be at runtime. |
| # 22:44:07 |
bhundven |
ah |
| # 22:44:08 |
bhundven |
ok |
| # 22:44:16 |
y_morin |
bhundven: I mean, at ct-ng run-time |
| # 22:44:19 |
bhundven |
right |
| # 22:44:36 |
bhundven |
I mean |
| # 22:45:07 |
bhundven |
zlib1g-dev (or zlib-devel on rpm systems), might be needed by other things. |
| # 22:45:14 |
bhundven |
not just elf2flt |
| # 22:45:26 |
bhundven |
on build or host |
| # 22:45:44 |
bhundven |
some things package their own zlib source |
| # 22:46:09 |
bhundven |
so this could be a path to telling those packages to use the host zlib or the packaged zlib |
| # 22:46:23 |
y_morin |
bhundven: Yes, like gcc. |
| # 22:46:29 |
bhundven |
right |
| # 22:47:43 |
y_morin |
bhundven: There's already a toggle in gcc to use the system zlib, BTW. |
| # 22:47:52 |
bhundven |
ok |
| # 22:48:59 |
y_morin |
bhundven: Back to the zlib1g-dev package: if you need to check for its presence, then you need to check at runtime with something like: ${host}gcc -E -xc - <<<'#include ' |
| # 22:49:23 |
bhundven |
k |
| # 22:49:29 |
y_morin |
bhundven: if it returns 0, then host gcc can find zlib.h in its sysroot. Otherwise, bummer. |
| # 22:49:33 |
bhundven |
yup |
| # 22:49:37 |
bhundven |
makes sense |
| # 22:50:05 |
y_morin |
bhundven: ./configure can not check, because ./configure does not know the host-gcc, as it can be set in the menuconfig. |
| # 22:50:14 |
bhundven |
yup |
| # 22:50:18 |
bhundven |
cross-canadian |
| # 22:50:28 |
y_morin |
bhundven: Indeed. |
| # 22:50:37 |
bhundven |
ok, let me redo stuff |
| # 22:51:06 |
y_morin |
bhundven: sure. Thanks! |
| # 22:53:26 |
bhundven |
anyone here using a bare-metal arm cortex toolchain with cmsis? |
| # 22:54:17 |
y_morin |
bhundven: cmsis==? |
| # 22:54:32 |
bhundven |
http://www.arm.com/cmsis/ |
| # 22:57:43 |
y_morin |
bhundven: It looks like a (small) C-library to me (like newlib), right? |
| # 22:57:54 |
bhundven |
suplimental |
| # 22:58:00 |
bhundven |
kinda like the avr32 headers |
| # 22:58:13 |
bhundven |
*supplemental |
| # 22:58:35 |
y_morin |
bhundven: OK, I see. No I have not used it. |
| # 22:59:00 |
bhundven |
this guy explains it better: http://www.youtube.com/watch?v=ttXW58-drYw |
| # 22:59:38 |
bhundven |
the funky thing is it seems that each hardware vendor has their own version of cmsis |
| # 22:59:48 |
bhundven |
*seems*, I might be wrong. |
| # 23:01:24 |
y_morin |
bhundven: sounds to be expected: each chipco will provide a cmsys to access their peripherals, but all will provide the same upper API/ABI, as I understand it. |
| # 23:01:39 |
y_morin |
s/cmsys/cmsis/ |
| # 23:01:48 |
bhundven |
:) |
| # 23:02:07 |
bhundven |
right, but I'm hoping I'm wrong |
| # 23:02:35 |
bhundven |
so I'm trying to find others using cmsis so I can learn more |
| # 23:12:59 |
bhundven |
ok, gotta go. bbl |
| # 23:13:01 |
bhundven |
quits : Quit: Leaving. |
| # 23:15:04 |
y_morin |
quits : Quit: Nighty Night! |