| --- Log opened Sat May 18 00:00:03 2013 | ||
| mor1kx | [mor1kx] skristiansson pushed 3 new commits to master: https://github.com/openrisc/mor1kx/compare/01f2a14ded13...ba1cd267baf4 | 06:07 |
|---|---|---|
| mor1kx | mor1kx/master 4946f6b Stefan Kristiansson: dcache: remove autotemplate from tag and way ram instantiations... | 06:07 |
| mor1kx | mor1kx/master 7091a0d Stefan Kristiansson: dcache: tidy up WRITE state logic | 06:07 |
| mor1kx | mor1kx/master ba1cd26 Stefan Kristiansson: cappuccino/wb_mux: remove redundant check for lsu_valid_i | 06:07 |
| stekern | hmm, is it expected that dropbear is *alot* slower than openssh? | 08:52 |
| juliusb | hmm, not sure | 08:58 |
| stekern | anyways, I put dropbear in there to see if it fails in the same way as openssh, and it does | 08:59 |
| stekern | seems both tries to use the legacy pty system | 09:00 |
| stekern | but openssh doesn't make noise about it, dropbear does | 09:00 |
| stekern | i.e., it tries to open /dev/ptyXX instead of /dev/pty/XX | 09:00 |
| stekern | cappuccino is surprisingly stable, generating ssh keys works ;) | 09:01 |
| juliusb | oh yeah, some config wrong or something? | 09:01 |
| juliusb | nice :) you've worked hard on it, so I'm not surprised | 09:01 |
| stekern | yeah, something like that, could be that it doesn't get to pick the right option when crosscompiling | 09:01 |
| stekern | I want to get irssi and apache up running on the board | 09:03 |
| juliusb | haha awesome | 09:03 |
| stekern | irssi itself isn't a big deal i guess, but it requires perl | 09:03 |
| stekern | native gcc crashes, but so it did for the jor1kx guy as well, so i guess that's a compiler bug | 09:05 |
| stekern | jor1k | 09:06 |
| stekern | got roo used to put the x in the end ;) | 09:06 |
| stekern | ah, I figured it out, uClibc needs to be built with UCLIBC_HAS_LIBUTIL set | 11:29 |
| stekern | but after that, dropbear segfaults :( | 11:29 |
| juliusb | oh wow, that's all? | 11:30 |
| stekern | openssh works perfectly though | 11:30 |
| stekern | and since that proved to be faster too, I'll just use that :) | 11:30 |
| stekern | space isn't really a problem, I've got plenty of it on my ws | 11:31 |
| stekern | rootfs is around 350mb now | 11:31 |
| juliusb | awesome | 11:31 |
| stekern | ok, next; cross compile perl | 11:34 |
| stekern | it needs ssh to access the target | 11:34 |
| stekern | perl's config scripts are really a pain... | 14:14 |
| stekern | this made it alot more sane http://perlcross.berlios.de/ | 14:37 |
| stekern | a simple "hello world" perl script works at least | 14:41 |
| stekern | oh, now I remember... irssi needs glib, which needs libffi, which needs to be ported... | 15:51 |
| stekern | a fun task, but not for today | 15:52 |
| stekern | I really would like to figure out why native gcc segfaults too | 15:55 |
| stekern | what was the status of gdb server? | 15:56 |
| hno | Is there strace support in uClibc? linux gdbserver relies on it. | 17:16 |
| stekern | think so | 17:21 |
| stekern | gdb isn't really my table though | 17:21 |
| hno | saw some patches earlier. Don't remember them merged, but memory isn't entirely reliable (was many months ago I looked into this) | 17:24 |
| stekern | yeah, _franck_ was asking about it iirc | 17:55 |
| stekern | but I think we had support for it in the old toolchain, but I've never used it | 17:56 |
| stekern | jeremybennett knows for sure =P | 17:57 |
| stekern | otoh, I'm ostly just interested at what address gcc segfaults, there are other ways to get that information | 17:58 |
| stekern | +m | 17:58 |
| _franck_ | stekern: https://github.com/blueCommand/or1k-src/commits/gdb-franck | 19:14 |
| _franck_ | the gdbserver should work | 19:15 |
| stekern | _franck_: oh, nice | 19:36 |
| stekern | thanks | 19:36 |
| stekern | will take a look tomorrow, time for bed here now | 19:36 |
| --- Log closed Sun May 19 00:00:05 2013 | ||
Generated by irclog2html.py 2.15.2 by Marius Gedminas - find it at mg.pov.lt!