| # 00:19:12 | fireandfuel | joins #crosstool-ng |
| # 00:19:31 | fireandfuel | parts #crosstool-ng |
| # 00:23:22 | dprice | quits : Quit: Leaving. |
| # 00:53:42 | dprice | joins #crosstool-ng |
| # 01:01:06 | dprice | quits : Ping timeout: 240 seconds |
| # 01:12:44 | dprice | joins #crosstool-ng |
| # 01:39:42 | dprice | quits : Quit: Leaving. |
| # 01:45:32 | dprice | joins #crosstool-ng |
| # 03:16:44 | dprice | quits : Quit: Leaving. |
| # 03:30:43 | devcoder | quits : Quit: devcoder |
| # 03:33:48 | alan_o | quits : Quit: Leaving |
| # 06:49:42 | dprice | joins #crosstool-ng |
| # 06:55:07 | dprice1 | joins #crosstool-ng |
| # 06:57:06 | dprice | quits : Ping timeout: 240 seconds |
| # 07:01:59 | diorcety | quits : Quit: Leaving. |
| # 07:52:14 | diorcety | joins #crosstool-ng |
| # 08:11:06 | dprice1 | quits : Ping timeout: 240 seconds |
| # 08:20:28 | dprice1 | joins #crosstool-ng |
| # 08:56:06 | linuxjacques | quits : Remote host closed the connection |
| # 09:08:35 | dprice1 | quits : Quit: Leaving. |
| # 09:09:59 | dprice | joins #crosstool-ng |
| # 09:12:51 | sh4rm4 | quits : Ping timeout: 276 seconds |
| # 09:13:11 | dprice | quits : Client Quit |
| # 09:22:02 | sh4rm4 | joins #crosstool-ng |
| # 09:26:36 | Net147 | joins #crosstool-ng |
| # 09:43:23 | dprice | joins #crosstool-ng |
| # 10:09:59 | sim | joins #crosstool-ng |
| # 10:10:05 | sim | hi |
| # 10:11:58 | sim | I am trying to generate a multilib toolchain for x86 target |
| # 10:12:00 | ChanServ | quits : *.net *.split |
| # 10:12:38 | sim | but while compiling libc for 32-bit I get the following error related with tls: |
| # 10:12:50 | sim | http://pastebin.com/XMtNzXQF |
| # 10:13:42 | sim | it appear that gcc generated during the first pass fails to assemblate tls instruction when called with the -m32 option |
| # 10:14:08 | sim | any hints are welcome |
| # 10:15:53 | ChanServ | joins #crosstool-ng |
| # 10:16:15 | dprice | quits : Quit: Leaving. |
| # 10:18:13 | kos_tom | sim: I am not sure ct-ng has official multilib support to build a x86/x86-64 toolchain. |
| # 10:18:31 | kos_tom | sim: it could probably be made to work, but some effort might be needed. |
| # 10:20:38 | Net147 | quits : *.net *.split |
| # 10:20:38 | sh4rm4 | quits : *.net *.split |
| # 10:21:58 | Net147 | joins #crosstool-ng |
| # 10:21:58 | sh4rm4 | joins #crosstool-ng |
| # 10:26:42 | Net147 | quits : *.net *.split |
| # 10:26:42 | sh4rm4 | quits : *.net *.split |
| # 10:27:15 | Net147 | joins #crosstool-ng |
| # 10:27:15 | sh4rm4 | joins #crosstool-ng |
| # 10:27:52 | Net147 | quits : Quit: HydraIRC -> http://www.hydrairc.com <- Wibbly Wobbly IRC |
| # 10:28:17 | codyps | quits : *.net *.split |
| # 10:28:17 | burek | quits : *.net *.split |
| # 10:28:42 | codyps | joins #crosstool-ng |
| # 10:28:42 | burek | joins #crosstool-ng |
| # 11:21:33 | mingwandroid | joins #crosstool-ng |
| # 11:59:25 | Net147 | joins #crosstool-ng |
| # 12:40:26 | sfan5|OFF | is now known as: sfan5 |
| # 12:48:18 | Net147 | quits : Quit: HydraIRC -> http://www.hydrairc.com <- Now with extra fish! |
| # 12:58:24 | sh4rm4 | quits : Ping timeout: 276 seconds |
| # 13:03:32 | sh4rm4 | joins #crosstool-ng |
| # 13:22:28 | alan_o | joins #crosstool-ng |
| # 13:47:49 | burek | parts #crosstool-ng |
| # 16:21:53 | diorcety | quits : Quit: Leaving. |
| # 16:39:29 | linuxjacques | joins #crosstool-ng |
| # 16:39:54 | linuxjacques | is now known as: Guest2417 |
| # 16:40:10 | Guest2417 | quits : Changing host |
| # 16:40:10 | Guest2417 | joins #crosstool-ng |
| # 16:40:28 | Guest2417 | is now known as: linuxjacques |
| # 17:45:58 | y_morin | joins #crosstool-ng |
| # 17:49:23 | diorcety | joins #crosstool-ng |
| # 19:01:57 | mingwandroid | parts #crosstool-ng |
| # 19:12:27 | dprice | joins #crosstool-ng |
| # 19:26:15 | diorcety | y_morin: i will try to do the modifications as soon as possible |
| # 19:30:28 | codyps | quits : Ping timeout: 246 seconds |
| # 19:36:49 | smartin | joins #crosstool-ng |
| # 19:52:07 | y_morin | diorcety: hello! OK, thank you! :-) |
| # 19:52:26 | y_morin | diorcety: of course, if there are good reasons for doing the way you already did, just say so! ;-) |
| # 19:52:52 | y_morin | goes back to his familly life for a while... Back in ~1h... |
| # 19:56:03 | mingwandroid | joins #crosstool-ng |
| # 20:04:01 | codyps | joins #crosstool-ng |
| # 20:38:37 | dprice | quits : Quit: Leaving. |
| # 20:40:45 | dprice | joins #crosstool-ng |
| # 21:02:29 | dprice | quits : Quit: Leaving. |
| # 21:04:23 | y_morin | diorcety: I think you missed the most important comment in the patch review: do we really need to do anything in the 'kernel' part? Can't we just do it in the 'libc' part? |
| # 21:04:51 | y_morin | diorcety: thus, there is no need to: |
| # 21:04:55 | y_morin | 1) share the tarball between kernel and libc, |
| # 21:05:25 | y_morin | 2) split the build between 'kernel' and 'libc' |
| # 21:05:33 | y_morin | 3) Hmm.. There's no 3 for now... |
| # 21:05:53 | y_morin | diorcety: and IMHO it will make the code much cleaner, by having everything in a single script. |
| # 21:11:52 | diorcety | we can |
| # 21:12:02 | diorcety | i keep 2 part as previous mingw |
| # 21:12:21 | diorcety | in kernel or libc? |
| # 21:13:22 | y_morin | diorcety: I think the best is to move all to the libc config and script. |
| # 21:14:08 | diorcety | added to this when LIBC_none is selected it is usefull to do cc1 and cc2 (i'm asking ...) |
| # 21:14:29 | y_morin | diorcety: basically, do_libc_start_files would be responsible for what is currently in do_kernel_headers |
| # 21:14:53 | y_morin | diorcety: no, we still need to say 'kernel is mingw' (or something like that). |
| # 21:15:17 | y_morin | diorcety: maybe, instead of 'kernel is mingw' we can call it 'Windows' |
| # 21:15:57 | y_morin | diorcety: the main title is 'Operating System' so saying 'Windows' there is correct, I would think. |
| # 21:20:30 | y_morin | diorcety: All config options should go in the 'C library' sub-menu, and all build instruction (except the tuple thingy) should go in the C library build script. |
| # 21:21:01 | y_morin | diorcety: Looking at the code how it is right now makes it look like that would be possible. |
| # 21:21:14 | y_morin | Of course, I may have miss something... ;-) |
| # 21:32:52 | dprice | joins #crosstool-ng |
| # 21:40:06 | dprice | quits : Ping timeout: 240 seconds |
| # 21:47:30 | sh4rm4 | quits : Ping timeout: 276 seconds |
| # 21:58:18 | dprice | joins #crosstool-ng |
| # 22:01:18 | sh4rm4 | joins #crosstool-ng |
| # 22:10:58 | diorcety | y_morin: ok |
| # 22:11:07 | diorcety | y_morin: and what about LIBC_none stuff? |
| # 22:11:45 | y_morin | diorcety: You mean, targetting 'Windows' without actually building the minngw runtime? |
| # 22:12:03 | diorcety | no it's totally another question |
| # 22:12:44 | y_morin | diorcety: sorry, I don't understand the question. Can you re-state, please? |
| # 22:13:20 | diorcety | if we don't select a libc (none) |
| # 22:13:28 | y_morin | diorcety: Oh, I think I understand, now! |
| # 22:13:49 | diorcety | build cc1 and cc2 is it usefull? |
| # 22:13:54 | y_morin | diorcety: you mean, in this case, there's no point in building core-1 and core-2, and final should be enough? |
| # 22:14:01 | diorcety | yes |
| # 22:14:08 | y_morin | diorcety: good question! ;-) |
| # 22:14:19 | y_morin | diorcety: not sure about that. We'd have to test. |
| # 22:14:26 | diorcety | i'am asking that because in darwin case we don't need cc1 and cc2.. i'm wondering in general case of LIBC_none |
| # 22:14:27 | y_morin | diorcety: probably, it should be OK. |
| # 22:15:00 | y_morin | diorcety: for Darwin, there's no C library? |
| # 22:15:12 | diorcety | the libc is in the sdk |
| # 22:15:24 | diorcety | you only build the gcc |
| # 22:15:59 | y_morin | diorcety: OK. So, there's still a 'C library', but it comes pre-built only, and the do_libc step only extracts that? |
| # 22:16:23 | diorcety | it's one package |
| # 22:16:48 | diorcety | containing all frameworks includes, library even a gcc (for darwin host) |
| # 22:17:21 | diorcety | we only build a gcc for build machine in this case |
| # 22:17:57 | y_morin | diorcety: OK, I get it. And binutils (or the equivalent). |
| # 22:18:41 | diorcety | as cc1 and cc2 are static ... and we have only shared libs (no static libs on darwin) ... it didn't work ... so i disabled cc1 cc2 ... it was not mandatory |
| # 22:18:44 | diorcety | cctools |
| # 22:18:57 | diorcety | i'm working on a cleaner cctools with mingwandroid |
| # 22:19:09 | y_morin | OK. Looks promising! :-) |
| # 22:19:11 | diorcety | not a big 2Mo patch as currently |
| # 22:19:11 | mingwandroid | yeah. I've not had much of a chance to do much yet though. |
| # 22:19:38 | mingwandroid | diorcety: it won't be 2Mo, but the patch will be quite big as it adds e.g. autotools support.. |
| # 22:19:57 | diorcety | yes put it will be clean |
| # 22:20:04 | smartin | quits : Quit: leaving |
| # 22:20:09 | diorcety | but |
| # 22:20:15 | mingwandroid | yes, clean for sure (I hope you'll agree)... |
| # 22:20:27 | mingwandroid | actually, I'll generate a preview of it right now... |
| # 22:21:30 | y_morin | diorcety, mingwandroid: could you guys post a prelimiary patch to the list, so I can see what you're doing, and offer suggestions? |
| # 22:21:47 | y_morin | diorcety, mingwandroid: or put your changes to a repo I can clone from. |
| # 22:22:10 | mingwandroid | we've got one on github, but I've got a few projects on the go at once. |
| # 22:22:39 | y_morin | github.. Sigh... We're using Mercurial for ct-ng! ;-) |
| # 22:22:48 | y_morin | was just kidding! |
| # 22:22:51 | diorcety | :D |
| # 22:23:16 | y_morin | Gah... Cygwin under Win7-64 under qemu/kvm is sloooowww as helll... :-( |
| # 22:23:47 | mingwandroid | y_morin: cygwin is slow as hell... fullstop... |
| # 22:23:55 | y_morin | 15 minutes to build a package that builds in under 25 seconds on native Linux... |
| # 22:24:03 | diorcety | y_morin: you can see the progression on https://github.com/diorcety/crosstool-ng |
| # 22:24:26 | mingwandroid | https://github.com/mingwandroid/toolchain4 |
| # 22:24:32 | y_morin | mingwandroid: yes, Cygwin is slow, but it was not that slow on Win7-32 |
| # 22:24:33 | diorcety | multiple branches.. we split the differents stuff |
| # 22:24:45 | mingwandroid | that's my project that we're basing it on. |
| # 22:24:46 | diorcety | -s |
| # 22:25:04 | y_morin | diorcety: OK, thanks, I'll give it a look. |
| # 22:25:10 | mingwandroid | and here's some builds: |
| # 22:25:11 | mingwandroid | http://mingw-and-ndk.googlecode.com/files/multiarch-darwin11-cctools127.2-gcc42-5666.3-llvmgcc42-2336.1-Linux-120724.tar.xz |
| # 22:25:20 | mingwandroid | http://mingw-and-ndk.googlecode.com/files/multiarch-darwin11-cctools127.2-gcc42-5666.3-llvmgcc42-2336.1-Darwin-120615.7z |
| # 22:25:27 | mingwandroid | http://mingw-and-ndk.googlecode.com/files/multiarch-darwin11-cctools127.2-gcc42-5666.3-llvmgcc42-2336.1-Windows-120614.7z |
| # 22:25:46 | mingwandroid | (supports both iOS and OSX) |
| # 22:26:23 | diorcety | we try to stay less intrusive than possible in the current ct-ng |
| # 22:27:50 | y_morin | diorcety: you're working on master? |
| # 22:29:13 | diorcety | yes final work will be on it. But for more general work (llvm for example). We have created another branch |
| # 22:30:48 | diorcety | y_morin: if you have time to look at current work on master and give us a feedback. it would be nice. |
| # 22:31:13 | diorcety | (about the changes of ct-ng ... not additional stufff |
| # 22:32:19 | y_morin | diorcety: I'm currently browsing master... |
| # 22:34:30 | mingwandroid | guys, my patch is bigger than 2MO... |
| # 22:35:47 | diorcety | :( |
| # 22:35:58 | mingwandroid | but an awful lot of it is just the addition of autotools... and the removal of old build files. |
| # 22:36:24 | diorcety | y_morin: the main changes are binutils stuff and companion libs |
| # 22:37:12 | diorcety | y_morin: i have also add additional build/host flags (cflags/ldflags) |
| # 22:37:55 | diorcety | i have certainly done crappy modifications |
| # 22:39:40 | y_morin | diorcety: I'm currently looking at master. Don;t worry, I already have a bit to say about it! ;-) I'm preparing a mail. |
| # 22:40:07 | y_morin | diorcety, mingwandroid: could you post a short mail to the list with the pointer to your repository? |
| # 22:40:34 | y_morin | diorcety, mingwandroid: thus I can reply to it (so others don;t think I replying out of the blue!) |
| # 22:40:43 | y_morin | *I am replying... |
| # 22:42:17 | mingwandroid | cool. |
| # 22:43:15 | y_morin | diorcety, mingwandroid: basically, I am only commenting on the patchset /workflow/, not the content per-say. Nothing really bad, you'll see. ;-) |
| # 22:43:22 | mingwandroid | looking through my changes, the patch size can be hugely reduced. e.g. I've removed *all* the old Makefiles and there's 477 of them. |
| # 22:43:26 | diorcety | notice that i have completly eject elf2flt and sstrip from binutils |
| # 22:43:50 | diorcety | :D |
| # 22:44:38 | diorcety | y_morin: thanks ... it is better to start on solid base |
| # 22:45:14 | y_morin | diorcety: :-) |
| # 22:47:59 | diorcety | ok i send a email |
| # 22:48:35 | y_morin | diorcety: Thanks! Wait a bit for the reply, though, I'm not ready. Takes a bit of time reviewing your stuff! ;-) |
| # 22:48:36 | devcoder | joins #crosstool-ng |
| # 22:55:26 | y_morin | diorcety: I meant, do not expect a reply right now, Im still reviwing ;-) |
| # 22:55:43 | diorcety | no pb |
| # 22:55:47 | y_morin | ;-) |
| # 22:56:16 | diorcety | i worked on clang stuff .. but i quit |
| # 23:00:11 | y_morin | diorcety: gcc-5666.3: what is *this*? |
| # 23:00:55 | diorcety | special gcc version from apple |
| # 23:00:58 | diorcety | :D |
| # 23:02:12 | y_morin | Sigh... :-( |
| # 23:02:25 | tkil | quits : Quit: Leaving |
| # 23:02:52 | y_morin | is going for a nicotine-break. BAD! :-( |
| # 23:03:14 | diorcety | y_morin: ADDICT ! |
| # 23:07:37 | y_morin | diorcety: yes... :-( |
| # 23:08:17 | y_morin | diorcety: OK, thanks for the mail! |
| # 23:08:31 | mingwandroid | y_morin: Apple's stuff is a brave new world I'm afraid :-( |
| # 23:08:42 | y_morin | mingwandroid: indeed. :/ |
| # 23:08:55 | mingwandroid | y_morin: I'll subscribe, is the ML moderated? |
| # 23:09:17 | y_morin | mingwandroid: no it's not moderated. But HTML is forbiden. |
| # 23:09:36 | y_morin | mingwandroid: http://sourceware.org/ml/crossgcc/ |
| # 23:09:40 | mingwandroid | ok. I'll reply with the my existing binaries in case anyone wants to find bugs in them! |
| # 23:10:32 | mingwandroid | actually, I can't really reply as I didn't get the original mail. diorcety: can you reply to your own mail to include links to my binaries? |
| # 23:12:56 | mingwandroid | I got to go guys. Will be back on tomorrow. I hope to get my python cross compilation thing finished (for a while) tomorrow, then I can devote a lot more of my time to this. |
| # 23:13:23 | y_morin | mingwandroid: See ya! |
| # 23:13:24 | mingwandroid | I also hope to bring full support to Windows as a build machine FWIW. |
| # 23:15:21 | diorcety | me too |
| # 23:15:27 | diorcety | too tired |
| # 23:15:28 | y_morin | diorcety: Bye! |
| # 23:15:31 | diorcety | see you |
| # 23:15:33 | diorcety | and thanks |
| # 23:16:26 | y_morin | Cheers@! |
| # 23:16:30 | y_morin | s/@// |
| # 23:18:32 | devcoder | quits : Remote host closed the connection |
| # 23:18:44 | devcoder | joins #crosstool-ng |
| # 23:44:52 | dprice | quits : Quit: Leaving. |
| # 23:55:19 | dprice | joins #crosstool-ng |