| --- Log opened Tue Jan 20 00:00:04 2015 | ||
| -!- E11fangirl is now known as Nala_ | 04:30 | |
| -!- Nala_ is now known as E11fangirl | 04:30 | |
| stekern | olofk: yeah... that is a bug in cgen, and I've proposed a patch that fixed it | 05:13 |
|---|---|---|
| stekern | (a patch to cgen that does the same as what now is committed to the generated file as a workaround) | 05:15 |
| stekern | but the cgen maintainer never seemed to be pleased with it | 05:16 |
| olofk | aha, so the binutils version is changed manually | 05:26 |
| olofk | Found your mail on the cgen ml | 05:28 |
| olofk | hmm... failed to install the or1k-elf tc again | 08:47 |
| olofk | What on earth am I doing wrong? | 08:47 |
| poke53282 | You decided to compile software, which are a historic mess with tons of obsolete workarounds and bad design decisions? | 09:00 |
| ams | haha | 09:01 |
| olofk | poke53282: You might be right. I will file a bug report for that | 09:01 |
| olofk | [PATCH gcc] Fix bad design decisions | 09:02 |
| olofk | [PATCH gcc] Remove obsolete workarounds | 09:02 |
| poke53282 | :) | 09:03 |
| ams | [BUG ams] Not long enough beard | 09:03 |
| olofk | [BUG olofk] Too much food ends up in beard | 09:04 |
| ams | mm.. food | 09:04 |
| ams | beard food | 09:04 |
| ams | mm.. | 09:04 |
| olofk | mm.. beard beer | 09:04 |
| olofk | Who put the beer in beard | 09:04 |
| ams | who put the beard in the beer sounds worse .. | 09:05 |
| olofk | wallento: Still have problems with upstream newlib. Using your git repo works fine | 09:05 |
| wallento | hey, okay, I will check | 09:06 |
| wallento | at some point the several for-upstream and master branches got confusing :) | 09:06 |
| olofk | I just did a diff against newlib-2.2.0-1, but I can't find anything that could explain this | 09:11 |
| olofk | Only a few things regarding (cache?) flushing and some sync_cas stuff that differed | 09:11 |
| olofk | wallento: Pastie here if it tells you anything http://pastie.org/9842931 | 09:12 |
| wallento | wohaa, thats weird | 09:12 |
| wallento | I will check | 09:13 |
| wallento | seems like a missing -D__SYSCALL__BLABLA | 09:13 |
| olofk | awesome | 09:13 |
| wallento | I will also do the continuous integration for the upstream then | 09:14 |
| wallento | and prepare official releases | 09:14 |
| wallento | but first fix ;) | 09:14 |
| olofk | Yeah, it would be great if you could test and use the upstream versions as much as possible | 09:14 |
| wallento | yeah, got it | 09:15 |
| wallento | or1k-*-elf | or1knd-*-elf) | 09:15 |
| wallento | sys_dir=or1k | 09:15 |
| wallento | newlib_cflags="${newlib_cflags} -DREENTRANT_SYSCALLS_PROVIDED " | 09:15 |
| wallento | have_crt0="no" | 09:15 |
| wallento | ;; | 09:15 |
| wallento | somehow didn't make it to configure-host | 09:15 |
| wallento | thats weird | 09:16 |
| wallento | it was in the submitted patch | 09:16 |
| wallento | i will get back to them | 09:16 |
| olofk | Cool! | 09:17 |
| olofk | I can change it locally and rebuild to make sure | 09:18 |
| olofk | _franck__: Do you want to apply the fixes to wb_altera_ddr_wrapper? | 09:22 |
| olofk | Bonus points if you put it in a separate repo and put a version tag on it | 09:23 |
| wallento | yeah, I just sent the patch to the list | 09:25 |
| poke53282 | lol, got this error during the compilation of openjdk: java.lang.RuntimeException: time is more than 10 years from present: 1104530400000 | 09:27 |
| olofk | haha | 09:27 |
| poke53282 | the problem started on 01-01-2015: https://bugs.gentoo.org/show_bug.cgi?id=534118 | 09:27 |
| poke53282 | but it looks like, that I use the newest version. | 09:28 |
| poke53282 | What does this tell you about Java? | 09:28 |
| olofk | That was a fu bug | 09:30 |
| olofk | fun | 09:30 |
| olofk | Seems like they got a fix | 09:30 |
| poke53282 | Yes | 09:30 |
| olofk | I should add that to FuseSoC too. Just add a static file with a date, and check against that when compiling | 09:31 |
| poke53282 | :) | 09:31 |
| olofk | If someone uses FuseSoC in 2025 something must be seriously wrong :) | 09:31 |
| olofk | hmm.. now gcc fails to install instead | 09:33 |
| poke53282 | Well, we use Fortran77 code here, which would still run on punch cards. Yes, there is something seriously wrong, but no need to disallow compiling. | 09:33 |
| ams | nothing wrong with punch cards .. you need to plan for the zombie invasion | 09:34 |
| olofk | make[2]: Entering directory '/home/olof/code/or1k/tcnew/bld/gcc/or1k-elf/libstdc++-v3' | 09:34 |
| olofk | make[2]: *** No rule to make target 'install'. Stop. | 09:34 |
| olofk | How is this even possible? Did someone steal the isntall target while I was compiling? | 09:36 |
| ams | building in source tree? | 09:37 |
| ams | didn't run configure? | 09:37 |
| ams | make -jN where N is big? | 09:37 |
| olofk | Hm.. could be -j5 | 09:37 |
| ams | do a make configure-host | 09:38 |
| ams | or some such .. | 09:38 |
| ams | make configure-target | 09:38 |
| poke53282 | ams: the code has some nice statements like "rewind 3", which means that the unit 3 with punch cards (which contains data) should be reset. | 09:38 |
| olofk | I noticed that my script didn't clean out the gcc build directory between the stage 1 and stage 2 builds. That could be a reason | 09:38 |
| ams | poke53282: nod, i've done punch card programming .. even debugged them and taped them .. | 09:39 |
| poke53282 | In C code, this is a fseek command. | 09:39 |
| ams | (i like punch card programming ..) | 09:42 |
| poke53282 | ams: you must be old. | 09:42 |
| ams | many people think that | 09:43 |
| olofk | Well ams also likes wind-up grammophones :) | 09:43 |
| ams | olofk: how did you know? | 09:43 |
| ams | seriously ... gf loves 'em, so do i .. i fix them for her when she finds one | 09:43 |
| _franck__ | olofk: "separate repo" ? | 09:48 |
| olofk | _franck__: Yes. I've started to realize that it's starting to become pretty messy to have a lot of RTL code in orpsoc-cores | 09:49 |
| olofk | But it's fine to just apply the patches to orpsoc-cores if you want | 09:50 |
| _franck__ | at least it is centralized | 09:50 |
| _franck__ | because if github/fjullien desapear, wb_altera_ddr_wrapper would desapear too | 09:51 |
| olofk | Don't scare us like that :) | 09:51 |
| _franck__ | s/desapear/disappear | 09:52 |
| _franck__ | :) | 09:52 |
| olofk | Yes, that's a benefit of having it centralized, but if we are worried about code disappearing we could also clone the repo | 09:53 |
| _franck__ | I pushed the fix now. I'll move the repo later | 09:53 |
| olofk | cool. No hurry with moving . Just thought I should mention it | 09:54 |
| stekern | olofk: libstdc++ always fail for me if I try to resume a halted build or try to compile it with anything > -j1 | 09:59 |
| olofk | stekern: Cool. Is that a bug in our code or in upstream? | 10:01 |
| stekern | no idea | 10:01 |
| stekern | I'm not in a hurry, I can wait for -j1 ;) | 10:01 |
| olofk | I guess it's quicker than continiously rebuilding with -j5 :) | 10:02 |
| wallento | olofk: its upstream now | 10:02 |
| wallento | should build from cvs | 10:02 |
| wallento | I think the git is delayed.. | 10:02 |
| olofk | Yep Saw the mail | 10:02 |
| olofk | Good thing that they decided on monthly snapshots. Hopefully we will only have to wait a month for a tar ball release | 10:03 |
| wallento | yeah, but nevertheless people will depend on our build | 10:06 |
| wallento | I will create a 2.2.0 + or1k release | 10:06 |
| olofk | stekern: Still fails without -j5. | 10:46 |
| olofk | Fuck this shit. I'm disabling c++ now | 10:47 |
| stekern | but did you start the build over from beginning? | 10:48 |
| olofk | Yep. And cleaned out the target directory | 10:49 |
| olofk | And no other toolchains in the path | 10:49 |
| olofk | ok, so it installs now, but fails to find stdio.h instead | 11:16 |
| wallento | newlib git is also updated | 11:49 |
| wallento | everything upstream now | 11:49 |
| olofk | wallento: Where can I find newlib git? | 11:49 |
| wallento | git://sourceware.org/git/newlib.git | 11:50 |
| olofk | I tried adding the patch to configure.host to newlib-2.2.0-1, but it still fails | 11:50 |
| olofk | cool. I'll try that | 11:50 |
| olofk | wallento: Nope. Still fails :( | 12:02 |
| wallento | mmh, whats the error? | 12:02 |
| olofk | wallento: I think this might be the relevant part http://pastie.org/9843322 | 12:04 |
| wallento | okay, seems some autotools are missing | 12:04 |
| wallento | i will check | 12:04 |
| wallento | I was thinking upstream executes them | 12:04 |
| wallento | ah, got it | 12:09 |
| wallento | next round | 12:13 |
| olofk | :) | 12:14 |
| wallento | its a rather simple line, see my new patch | 12:14 |
| olofk | Do you want me to test it first? | 12:14 |
| wallento | naaa | 12:14 |
| wallento | it will work :) | 12:14 |
| wallento | :-p | 12:14 |
| olofk | Famouse last words :) | 12:14 |
| wallento | I also created the continuous integration for this | 12:14 |
| olofk | With the upstream git repo? | 12:16 |
| wallento | yes | 12:17 |
| wallento | binutils and newlib | 12:17 |
| olofk | good | 12:18 |
| olofk | The git repo isn't updated yet AFAICS | 13:04 |
| olofk | There! | 13:16 |
| wallento | mmh, on my computer or1k-gcc doesn't build anymore | 13:17 |
| wallento | pod2man fails | 13:17 |
| wallento | strange | 13:17 |
| olofk | ehhmm.... wallento... ehmm.. it... kind of... still doesn't work | 13:20 |
| wallento | cmon | 13:21 |
| wallento | :-D | 13:21 |
| olofk | or1k-support.h is missing | 13:21 |
| wallento | are you sure? | 13:23 |
| wallento | ahhhh, shit | 13:23 |
| wallento | that was a patch that I never submitted | 13:23 |
| wallento | I think | 13:23 |
| wallento | let me check | 13:23 |
| wallento | it is there but cannot be found, correct? | 13:23 |
| olofk | ../../../../../../../newlib-git/newlib/libc/sys/or1k/mlock.c:19:26: fatal error: or1k-support.h: No such file or directory #include <or1k-support.h> | 13:23 |
| wallento | okay, let me fix this, but add some new features also | 13:25 |
| olofk | There are a few fixes in mlock.c | 13:25 |
| wallento | namely mmu management | 13:25 |
| olofk | Do you want me to try it before you send it upstream? :) | 13:25 |
| _franck__ | wallento: new_features == new_problems | 13:28 |
| _franck__ | :) | 13:28 |
| wallento | it is only (i|d)mmu_flush | 13:28 |
| wallento | in assembly | 13:28 |
| wallento | that is hopefully no new errors | 13:29 |
| wallento | problem is I wanted to change something in or1k-headers for this | 13:29 |
| wallento | by the way there are GPL headers in or1k-headers, for the nops | 13:29 |
| wallento | we need a blind monkey to rewrite them | 13:29 |
| wallento | ;) | 13:30 |
| olofk | What? I thought the idea of or1k-headers was to remove the GPL stuff | 13:30 |
| wallento | yes, but or1k-nops.h was moved there | 13:37 |
| olofk | ah | 13:37 |
| wallento | I did NOP_PUTC by heart | 13:37 |
| wallento | the others I didn't remember and was not supposed to look at the preivous one | 13:37 |
| wallento | :-D | 13:38 |
| wallento | olofk: git@github.com:openrisc/newlib.git | 13:38 |
| wallento | branch for-upstream | 13:38 |
| olofk | wallento: olofk the build bot is on it! | 13:38 |
| olofk | wallento: Fail! | 13:45 |
| olofk | Looks like it can't find sys/or1k | 13:56 |
| wallento | you're kidding :) | 13:58 |
| olofk | I wish I did | 13:58 |
| wallento | I still can't build gcc from the repo | 13:59 |
| wallento | did something change in ubuntu's pod2man recently? | 13:59 |
| wallento | gfdl.pod around line 53: Expected text after =item, not a number | 13:59 |
| olofk | Don't know. Gentoo user here | 13:59 |
| wallento | strange | 13:59 |
| olofk | That hasn't been a problem in any of the ~30 gcc compilations today :) | 14:00 |
| poke53282 | olofk: Since my mainboard died with my Atom N270, I haven't touched gentoo anymore. | 14:02 |
| poke53282 | which gcc version you try to compile? | 14:02 |
| poke53282 | wallent: Can you compüile with --disable-docs ? | 14:03 |
| wallento | mmh, its gone now | 14:04 |
| wallento | even better :) | 14:04 |
| wallento | olofk: do it | 14:34 |
| olofk | Do what? | 14:34 |
| wallento | build | 14:36 |
| wallento | was just a git hickup | 14:37 |
| wallento | upstream was not in, i.e., the auto-generated configure | 14:37 |
| olofk | aha | 14:46 |
| olofk | Got to go now, but I should just pull your for-upstream branch and build, right? | 14:46 |
| wallento | yes | 14:50 |
| wallento | for me it build now | 14:50 |
| wallento | for a clean tree | 14:50 |
| wallento | olofk: i just build successfully from for-upstream, upstream binutils and or1k-gcc | 15:39 |
| wallento | olofk: http://bugzilla.opencores.org/show_bug.cgi?id=31 works now (I assume I meant objcopy ages ago) | 15:45 |
| olofk | wallento: Good job with the last upstreaming hickups, and thanks for looking at the bug report | 18:26 |
| --- Log closed Wed Jan 21 00:00:06 2015 | ||
Generated by irclog2html.py 2.15.2 by Marius Gedminas - find it at mg.pov.lt!