| [09:27] <lool> Excellent http://fl0rian.wordpress.com/2009/03/29/a-smart-beaglebrick/ | |
| === Meiz__n810 is now known as Meiz_n810 | |
| [10:48] <ogra> lool, cute ! | |
| [10:49] <lool> ogra: Hey you know that exec -s issue? | |
| [10:49] <lool> ogra: It was really what I suspected | |
| [10:49] <ogra> ?? | |
| [10:49] <lool> That config for ramdisk size was too small, even after bumping it | |
| [10:50] <lool> By 72192 bytes! | |
| [10:50] <ogra> so you have solved the need for -s ? | |
| [10:50] <lool> ogra: Not 100% confirmed, but I found the bug which caused the error message during RAMDISK unpacking | |
| [10:50] <lool> (the ENOSPC) | |
| [10:50] <ogra> ah | |
| [10:51] <ogra> i still think that redboot is in our way | |
| [10:51] <lool> In fact I had solved it when I asked for bumping that config, but we didn't bump it enough for the cheer size of our initrds | |
| [10:51] <lool> I don't think redboot matters | |
| [10:51] <lool> You can load an initrd without -r/-s | |
| [10:51] <ogra> with the patched redboot ? | |
| [10:51] <lool> No; with kernel cmdline args | |
| [10:51] <ogra> i thought with our version it was mandatory | |
| [10:51] <lool> But I prefer using ATAG personally, more elegant IMO | |
| [10:51] <ogra> indeed | |
| [10:52] <lool> It's mandatory for the ATAG ramdisk thingy to be passed; but you can also tell the kernel "BTW there's an initrd at this random addres" | |
| [10:52] <ogra> hmm | |
| [10:54] <lool> Anyway, I prefer encoding this in the redboot bootscript's commands rather than the cmdline; the former doesn't require encoding the size of RAM or the addresses | |
| [10:54] <lool> (physicaladdresses) | |
| [10:54] <ogra> yup | |
| === mcasadevall is now known as NCommander | |
| [21:13] <strk> does anyone have any idea why htons should return 'unsigned int' in ubuntu-arm ? | |
| [21:14] <strk> ref: https://savannah.gnu.org/bugs/index.php?26051 | |
| [21:31] <suihkulokki> strk: #include <arpa/inet.h> | |
| [21:31] <suihkulokki> perhaps you'r missing that? | |
| [21:32] <strk> I tried that, didn't help | |
| [21:36] <suihkulokki> http://paste.debian.net/31967/ | |
| [21:39] <strk> g++ mentioned "unsigned int", btw | |
| [21:40] <strk> which isn't what the compiler thinks an undefined function returns, right ? | |
| [21:40] <strk> that'd be 'int' | |
| [21:51] <strk> maybe is a jaunty-specific bug | |
| [21:55] <strk> g++ complains on reinterpret_cast: error: invalid cast from type 'unsigned int' to type 'uint16_t' | |
| [21:56] <strk> oh-ha | |
| [21:58] <strk> http://rafb.net/p/JeNuNd73.html | |
| [21:58] <strk> test.cpp:7: error: invalid cast from type 'uint16_t' to type 'uint16_t' (!!!!) | |
| [22:00] <suihkulokki> haha | |
| [22:00] <suihkulokki> I'm happy I don't deal with c++ usually | |
| [22:29] <strk> found it ! | |
| [22:29] <strk> it's -O2 !!! | |
| [22:29] <strk> g++ (Ubuntu 4.3.2-2ubuntu11) 4.3.3 20090111 (prerelease) | |
| [22:31] <strk> -O2 makes that short into an int | |
| [22:33] <strk> remains a short with g++ (Ubuntu 4.3.2-1ubuntu12) 4.3.2 | |
| [22:39] <Martyn> re | |