# 00:19:01 |
y_morin |
quits : Quit: Night All! |
# 07:16:06 |
esbenh |
joins #crosstool-ng |
# 07:39:19 |
Buglouse |
quits : Ping timeout: 246 seconds |
# 08:00:09 |
kos_tom |
joins #crosstool-ng |
# 10:15:54 |
kos_tom |
quits : Quit: Leaving |
# 11:08:03 |
kos_tom |
joins #crosstool-ng |
# 12:14:51 |
daggs1 |
joins #crosstool-ng |
# 12:15:24 |
daggs1 |
hello guys, |
# 12:15:58 |
daggs1 |
does anyone knows about a successfully compilation of gcc-4.6.0 with crosstool? |
# 12:26:57 |
kos_tom |
I've built a ARM toolchain with Linaro gcc 4.6 successfully. |
# 12:27:04 |
kos_tom |
it's able to build a kernel that boots |
# 12:27:20 |
kos_tom |
(but userspace programs don't work, but I suspect it's due to uClibc 0.9.32-rc with NPTL rather than gcc 4.6) |
# 12:27:40 |
kos_tom |
i'm assuming you're talking about crosstool-ng and not crosstool, of course |
# 12:29:31 |
daggs1 |
yes |
# 12:29:47 |
daggs1 |
any idea about non Linaro? |
# 12:33:30 |
kos_tom |
no, sorry, I haven't tried the stock gcc 4.6 |
# 12:38:40 |
daggs1 |
do you know how can I disable java building in ppl? |
# 14:09:14 |
Y_MORIN |
joins #crosstool-ng |
# 14:10:35 |
Y_MORIN |
quits : Client Quit |
# 14:11:15 |
y_morin |
joins #crosstool-ng |
# 14:51:02 |
kos_tom_ |
joins #crosstool-ng |
# 14:54:00 |
kos_tom |
quits : Ping timeout: 240 seconds |
# 15:03:35 |
daggs1 |
y_morin: are you here? |
# 15:03:48 |
y_morin |
Is it possible to mute the join-bell, please?daggs1: in a meeting... |
# 15:03:58 |
y_morin |
Woops... |
# 15:04:03 |
y_morin |
daggs1: in a meeting... |
# 15:04:09 |
daggs1 |
ok |
# 15:04:21 |
y_morin |
daggs1: I know about java PPL. |
# 15:04:35 |
daggs1 |
y_morin: got passed it |
# 15:04:44 |
y_morin |
There's a patch floating around... |
# 15:04:48 |
daggs1 |
me issue is now with the kernel headers |
# 15:04:59 |
daggs1 |
y_morin: I don't need it to I've changed the ppl script |
# 15:05:08 |
daggs1 |
s/me/my/g |
# 15:06:18 |
daggs1 |
I'm going home now, I'll reconnect in ~ 15 min or so |
# 15:09:06 |
daggs1 |
brb soon |
# 15:09:09 |
daggs1 |
quits : Remote host closed the connection |
# 15:56:32 |
kos_tom_ |
quits : Quit: Leaving |
# 15:56:41 |
daggs1 |
joins #crosstool-ng |
# 15:59:04 |
daggs1 |
y_morin, back |
# 15:59:33 |
y_morin |
daggs: still in meeting... |
# 16:01:53 |
daggs1 |
ok, I'm going for a jogg |
# 16:06:27 |
Buglouse |
joins #crosstool-ng |
# 16:14:18 |
y_morin |
quits : Quit: Back on-line in a few hours |
# 16:47:02 |
linuxjacques |
joins #crosstool-ng |
# 19:45:22 |
y_morin |
joins #crosstool-ng |
# 20:14:36 |
daggs1 |
y_morin, hello |
# 20:14:44 |
y_morin |
daggs1: hey! |
# 20:14:53 |
daggs1 |
how are you? |
# 20:15:37 |
y_morin |
Hey, tired, bad day... |
# 20:15:46 |
daggs1 |
sorry to hear |
# 20:15:50 |
y_morin |
You? |
# 20:16:05 |
y_morin |
Hmm. nothing too bad, don;t worry! |
# 20:16:38 |
daggs1 |
actually, good day, I was able to find a dubious bug |
# 20:16:53 |
y_morin |
daggs1: what was the issue earlier? Disabling Java interface in PPL? |
# 20:17:08 |
daggs1 |
y_morin, I've got passed that |
# 20:17:25 |
y_morin |
Oh, good. :-) |
# 20:17:28 |
daggs1 |
question is, in the last month or so, did you make any changes to the linux-headers? |
# 20:18:45 |
y_morin |
daggs1: I added some latest kernel versions, yes. But did not touch the build script. Last touched in Jan 14, 2011. |
# 20:19:21 |
y_morin |
Which reminds me I'll have to add support for 3.x shortly... :-) |
# 20:19:22 |
daggs1 |
mmm strange, I get this now: |
# 20:19:25 |
daggs1 |
==> Executing: 'make' '-C' '/home/dagg/workspace/crosstool-ng/build/src/linux-custom' 'O=/home/dagg/workspace/crosstool-ng/build/x86_64-vSMP-linux-gnu/build/build-kernel-headers' 'ARCH=x86' 'INSTALL_HDR_PATH=/home/dagg/workspace/crosstool-ng/opt/x86_64-vSMP-linux-gnu/gcc-4.6.0-glibc-2.11.1/x86_64-vSMP-linux-gnu/sysroot/usr' 'V=0' 'headers_install' |
# 20:19:25 |
daggs1 |
[ALL ] make[1]: Entering directory `/home/dagg/workspace/crosstool-ng/build/src/linux-custom' |
# 20:19:25 |
daggs1 |
[ALL ] /home/dagg/workspace/crosstool-ng/build/src/linux-custom/Makefile:1443: *** mixed implicit and normal rules. Stop. |
# 20:19:26 |
daggs1 |
[ERROR] make[1]: *** [headers_install] Error 2 |
# 20:19:30 |
daggs1 |
[ALL ] make[1]: Leaving directory `/home/dagg/workspace/crosstool-ng/build/src/linux-custom' |
# 20:19:32 |
daggs1 |
great :) |
# 20:19:43 |
y_morin |
daggs1: expected! |
# 20:19:53 |
y_morin |
daggs1: you have make-3.82, right? |
# 20:19:56 |
daggs1 |
ok, any idea how to get passed it? |
# 20:20:05 |
y_morin |
daggs1: and you are using an 'old' kernel? |
# 20:20:08 |
daggs1 |
yup |
# 20:20:11 |
daggs1 |
on both |
# 20:20:21 |
y_morin |
Ahah! This is magic! :-) |
# 20:20:28 |
y_morin |
Lemme dig it back for you... |
# 20:20:46 |
linuxjacques |
is there a way around it without reverting to make 3.81 ? that's what I did |
# 20:20:58 |
y_morin |
daggs1, linuxjacques: http://www.gossamer-threads.com/lists/linux/kernel/1263151 |
# 20:21:41 |
y_morin |
We should probably add that patch to the patches/ directory for all impacted kernels... |
# 20:22:33 |
daggs1 |
y_morin, good idea |
# 20:22:34 |
linuxjacques |
ah OK - my problem was trying to build 2.6.27.xx for powerpc - the powerpc-specific makefiles have that problem and I guess even though .27 is "long term support" nobody cared to update them |
# 20:22:45 |
daggs1 |
this should be applied on the kernel right? |
# 20:22:53 |
y_morin |
daggs1: yes, indeed. |
# 20:23:12 |
daggs1 |
I need to update both files? |
# 20:23:17 |
y_morin |
linuxjacques: no idea about the PPC part and 2.6.27. |
# 20:23:32 |
y_morin |
daggs1: I'm digging the patch that was really applied upstream.... |
# 20:24:05 |
daggs1 |
y_morin, I use a pretty old kernel for my cross tool |
# 20:24:25 |
y_morin |
daggs1, linuxjacques: here it is: https://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=3c955b407a084810f57260d61548cc92c14bc627;hp=3643f849d7da5c12da589beb03c12303fe79b841 |
# 20:24:55 |
linuxjacques |
y_morin: BTW, did anything ever come of our toolchain tuple issue (as with e500v2) ? |
# 20:25:40 |
y_morin |
linuxjacques: not that I know of. F5, please... |
# 20:27:03 |
y_morin |
linuxjacques: search for make in there: https://git.kernel.org/?p=linux%2Fkernel%2Fgit%2Ftorvalds%2Flinux-2.6.git&a=search&h=HEAD&st=commit&s=3.82 |
# 20:27:07 |
linuxjacques |
y_morin: sorry I don't understand the F5 thing - you want me to refresh your memory about the issue? |
# 20:27:16 |
y_morin |
linuxjacques: Yes... :-/ |
# 20:27:58 |
y_morin |
linuxjacques: Above you'll see the fixes for make-3.82, and one of the first is for PPC. |
# 20:28:30 |
y_morin |
daggs1, linuxjacques: alternatively, you can ask crostool-NG to build 3.81 for you. It's in the comanion tools. |
# 20:28:44 |
daggs1 |
ok |
# 20:28:52 |
daggs1 |
I'll see if I can apply it first |
# 20:29:16 |
linuxjacques |
y_morin: it's from May 2nd - http://thread.gmane.org/gmane.comp.lib.uclibc.buildroot/30512 |
# 20:29:26 |
y_morin |
daggs1: patch is the way to go, IMHO. |
# 20:29:56 |
linuxjacques |
oops, Apr 30th, not May 2nd |
# 20:30:27 |
y_morin |
linuxjacques: OK, I remember, now, that was about buildroot not using the proper tuple. |
# 20:31:49 |
linuxjacques |
yep |
# 20:32:36 |
linuxjacques |
there was talk about changing the way BR handles tuples, but I didn't see any follow-up |
# 20:32:38 |
y_morin |
linuxjacques: there have been no response to this, either from Jacmet or kos_tom. |
# 20:32:44 |
y_morin |
indeed. |
# 20:32:45 |
linuxjacques |
yeah :-( |
# 20:33:04 |
daggs1 |
crap |
# 20:33:15 |
y_morin |
linuxjacques: I'll try to poke them... |
# 20:33:17 |
daggs1 |
my kernel is too old :-S |
# 20:33:29 |
linuxjacques |
y_morin: thanks! |
# 20:33:45 |
y_morin |
daggs1: Then update... |
# 20:33:47 |
y_morin |
hides... |
# 20:33:56 |
y_morin |
daggs1: the kernel on your build machine? |
# 20:33:59 |
daggs1 |
the kernel I need to use I mean |
# 20:34:05 |
y_morin |
Or the kernel to get headers from? |
# 20:34:15 |
daggs1 |
headers |
# 20:34:24 |
y_morin |
What version did you select? |
# 20:34:29 |
daggs1 |
2.6.22 |
# 20:34:32 |
daggs1 |
custom |
# 20:34:50 |
y_morin |
Arch? libc? NPTL? |
# 20:34:52 |
daggs1 |
Companion tools->make will do the trick for me? |
# 20:35:15 |
daggs1 |
fedora, libc, NPTL |
# 20:35:40 |
y_morin |
daggs1: oh, you mean the patch to fix make-3.82 does not apply to your kernel? |
# 20:35:46 |
daggs1 |
yes |
# 20:36:07 |
y_morin |
Oh, sorry. I thought glibc/gcc/whatever did not build because of too old the kernel. |
# 20:36:38 |
y_morin |
daggs1: yes, comptools -> make will build 3.81. |
# 20:36:54 |
daggs1 |
y_morin, ok, how do I instruct ct-ng to use it? |
# 20:37:40 |
y_morin |
daggs1: that's automatic. If you select it, ct-ng builds and uses it in place of your host make. |
# 20:38:12 |
daggs1 |
okie, if it works, my boss will really like me and I'll send you a beer :) |
# 20:39:01 |
y_morin |
daggs1: beer! Hmmm! Where are you? :-] |
# 20:39:09 |
daggs1 |
isreal |
# 20:39:43 |
daggs1 |
deadly silent... |
# 20:39:54 |
y_morin |
daggs1: That's a bit far for me... |
# 20:39:56 |
daggs1 |
did I said something wrong? |
# 20:40:09 |
daggs1 |
you're from france right? |
# 20:40:20 |
y_morin |
has other conversations in other windows, mails to read... |
# 20:40:25 |
y_morin |
Yes! |
# 20:40:56 |
daggs1 |
not that far, only one or two latitude lines in the map to the right |
# 20:41:33 |
y_morin |
daggs1: It's even earer on a smaller map! :-) |
# 20:41:40 |
y_morin |
*nearer |
# 20:41:49 |
daggs1 |
cheaper too :) |
# 20:41:53 |
y_morin |
:-) |
# 20:43:36 |
y_morin |
Israel is not a place I would probably visit soon... It too hot and sunny for me. I prefer going north. Iceland would be my destination of choice... |
# 20:44:01 |
daggs1 |
yeah, pretty hot :-S |
# 20:44:29 |
daggs1 |
good thing I work of a sw company, free AC :) |
# 20:48:19 |
y_morin |
daggs1: Yeah! :-) |
# 20:48:44 |
daggs1 |
so far so good, binutils is compiling now |
# 20:50:52 |
daggs1 |
going to bed, thanks for the help :) hope it works |
# 20:53:09 |
y_morin |
daggs1: Good! See ya! |
# 20:54:44 |
daggs1 |
fails :( |
# 20:54:46 |
daggs1 |
[ALL ] *** Error: Headers not exportable for this architecture (x86) |
# 20:55:31 |
daggs1 |
I had to link x86_64 to x86 as the tar doesn't have it |
# 20:56:04 |
daggs1 |
but it is the same tar which worked before for gcc-4.5.2 |
# 20:57:47 |
y_morin |
daggs1: Hmmm. Weird. 2.6.22 you said? |
# 21:48:34 |
y_morin |
quits : Quit: Nite all! :-) |
# 22:48:28 |
linuxjacques |
quits : Ping timeout: 252 seconds |