<div dir="ltr">On Wed, Apr 24, 2013 at 1:10 PM, Adam Sampson <span dir="ltr"><<a href="mailto:ats@offog.org" target="_blank">ats@offog.org</a>></span> wrote:<br><div class="gmail_extra"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div><span style="color:rgb(34,34,34)">for KRoC from Linux worked pretty well when I last tried it[1], so that</span><br></div>
might be an easier option than preserving the native Windows build if<br>
anyone still needs Win32 binaries.<br></blockquote><div><br></div><div>I did some reading on that recently, but haven't had time to go further. (I found it while reading up on MSYS/MinGW.) Getting a Linux machine doing a cross-build is fine, and something I think we could maintain longer term. The current build scene is rough.</div>
<div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
It's in the "mingw" branch -- which didn't get merged at the time<br>
because it involved a truly horrible hack to make CCSP work without<br>
signals or pipes, but if it's the TVM toolchain you're after then that<br>
bit can be quietly ignored...<br></blockquote><div><br></div><div style>That would probably be the way we'd go; the only reason for doing this would be to support native o-pi work that lets students do assignments like (for example) "Life on Mars." We could do it all "in the cloud," but that's another rewrite/massive project, so a Linux->Windows cross-build is a saner approach at this time, in my mind.</div>
<div style><br></div><div style>Cheers,</div><div style>M</div></div></div></div>