| juliusb | woo! got u-boot running on my mor1kx pronto espresso pipeline in the cycle accurate model | 01:33 |
|---|---|---|
| juliusb | also proves u-boot can be made to run on the no delay version of the tool chain with a few hacks to the start.S | 01:34 |
| juliusb | eCos kinda boots too | 03:17 |
| juliusb | it relies on UART for printing :( | 03:17 |
| juliusb | ah and it's showing up bugs in the new core too by the looks of things :) | 03:35 |
| stekern | juliusb: doesn't u-boot rely on uart for printing? | 10:20 |
| juliusb | it does, and it worked I guess because I saw output | 11:26 |
| stekern | I'm playing with getting 'fbi' working between dayjob-tasks | 11:58 |
| stekern | It runs now, but since my display is at home I can't tell if it works ;) | 11:59 |
| stekern | cross compiling with or1k-linux is a breeze nowdays | 12:01 |
| stekern | ./configure --host=or1k --prefix=/path/to/sysroot/usr | 12:02 |
| stekern | and of you go | 12:02 |
| stekern | s/of/off | 12:02 |
| stekern | ./configure --host=or1k-linux --prefix=/path/to/sysroot/usr | 12:03 |
| stekern | only annoyance is the or32 in all config.guess an config.sub... | 12:03 |
| juliusb | oh cool!! | 12:04 |
| juliusb | haha | 12:04 |
| stekern | find ./ -type f -name '*' -print | xargs sed -i 's/or32/or1k/' | 12:05 |
| stekern | fixes it though ;) | 12:05 |
| stekern | that old stuff has some -coff flag which probably should be -elf too | 12:06 |
| stekern | don't know what impact that has though | 12:06 |
| stekern | if I get this rolling and rte has a projector with hdmi input, I could use the atlys board to show my pdf-sides | 12:09 |
| stekern | =) | 12:09 |
| stekern | https://www.kraxel.org/blog/linux/fbida/ | 12:14 |
| juliusb | awesome man | 12:19 |
| stekern | so are the problems you see in eCos pronto-specific or mor1kx-specific? | 13:14 |
| stekern | I have never tested eCos on mor1kx | 13:15 |
| juliusb | umm to do with the pipeline I think | 13:23 |
| juliusb | not the code | 13:23 |
| juliusb | eCos looks good, actually, another source of test code for the build | 13:23 |
| juliusb | a bit easier to build than the GCC stuff | 13:24 |
| juliusb | stekern: OK we have the go ahead to use the RTE offices all day Sunday too | 13:59 |
| juliusb | so I'm going to re-arrange some sessions to run on the Sunday | 13:59 |
| juliusb | although, the restaurant booking is at 8 | 13:59 |
| juliusb | so we'll be in the RTE offices until about 7:30 I would assume | 13:59 |
| juliusb | which is about an hour longer than I assumed | 13:59 |
| juliusb | but maybe we can space out teh stuff on Saturday a little more | 13:59 |
| juliusb | and then we'll have more tmie for the LLVM presentation on Saturday, too | 14:00 |
| juliusb | so, what do you say we do the dynamic linking stuff on Sunday? | 14:00 |
| juliusb | stekern: are you and simoncook really going to fit your LLVM talk into 30 minutes? what do you say I allocate 50 minutes, but you can have up to an hour | 15:23 |
| juliusb | Ok new schedule up | 15:28 |
| juliusb | moer breathing space on Saturday | 15:28 |
| _franck_ | juliusb: you can count on me for the dinner on saturday | 16:01 |
| stekern | juliusb: I'm not sure how much time simoncook needs, but in the original schedule it didn't matter so much since if the LLVM talk would run long it would only cut into the dynamic linking, which I could dynamically adjust then. | 16:25 |
| stekern | but what you have now is probably ok | 16:26 |
| juliusb | cool, yeah there'll be plenty of time hopefully | 16:29 |
| juliusb | the break time isn't strict, either, it just allows time for questions and follow on but people won't feel obliged to pay strict attention I guess | 16:29 |
| jeremybennett | juliusb: Looks good | 18:07 |
| olofk | juliusb: You can decrease the time allocated for my talk if you need more time. I think I will have a hard time filling up 40 minutes | 18:44 |
| juliusb | olofk: OK, well we can surely fill it up with discussion I would expect | 18:49 |
| juliusb | or start the following session a bit early | 18:49 |
| olofk | Yeah, I don't think it will be a problem to fill up the time. Just wanted to tell you if you needed to move things around | 18:50 |
| juliusb | nah im happy to leave it like that | 18:53 |
| juliusb | lots of extra time around the place | 18:53 |
| juliusb | jeremybennett: are you around? | 19:33 |
| juliusb | jeremybennett: I just wanted to ask you a quick question about something in Verilator. Was wondering if I could give you a call | 19:36 |
| jeremybennett | juliusb: Just on phone - free shortly. | 19:50 |
| jeremybennett | Actually - free now | 19:50 |
| jeremybennett | Skype or real phone is fine | 19:50 |
| juliusb | Ah have to run, speak to you later or tomorrow | 20:06 |
| juliusb | has anyone here played with the eCos port for OpenRISC? | 23:39 |
| olofk | I was just going to ask if someone knew about a good place to find clipart, but your question seems way more interesting | 23:45 |
| olofk | Aren't some of the guys who ported ecos coming to the conference? | 23:45 |
Generated by irclog2html.py 2.15.2 by Marius Gedminas - find it at mg.pov.lt!