<div dir="ltr"><div>Both Clang/LLVM 3.4 -> Clang/LLVM 3.5<br></div><div>And i will also try using MCJIT.</div></div><div class="gmail_extra"><br><div class="gmail_quote">2014-09-17 18:56 GMT+06:00 Anton Smirnov <span dir="ltr"><<a href="mailto:dev@antonsmirnov.name" target="_blank">dev@antonsmirnov.name</a>></span>:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Hi, Tim.<div><br></div><div>I've used Clang 3.4 final release and now i'm going to test it with 3.5 release (since i've read about arm64 improvements).</div><div>I will report my results.</div><div><br></div><div>BTW, is it possible to get smth like "hello world" output even with apple restrictions?</div><div><br></div><div>Regards, Anton.</div></div><div class="HOEnZb"><div class="h5"><div class="gmail_extra"><br><div class="gmail_quote">2014-09-17 18:42 GMT+06:00 Tim Northover <span dir="ltr"><<a href="mailto:t.p.northover@gmail.com" target="_blank">t.p.northover@gmail.com</a>></span>:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi Anton,<br>
<br>
I've added the llvmdev list, since the issues you're seeing are coming<br>
from the backend, which is more their side.<br>
<span><br>
On 17 September 2014 08:43, Anton Smirnov <<a href="mailto:dev@antonsmirnov.name" target="_blank">dev@antonsmirnov.name</a>> wrote:<br>
> i've changed lli arguments to the next (instead of default):<br>
><br>
> return llvm_interpret(<br>
> InputFile,<br>
> std::vector<std::string>(),<br>
> false, // ForceInterpreter<br>
> false, // UseMCJIT<br>
</span>> [...]<br>
<span>> Now i'm having:<br>
><br>
> Unhandled instruction encoding format!<br>
> UNREACHABLE executed at<br>
> /Users/asmirnov/Documents/dev/src/llvm_34_ios/lib/Target/ARM/ARMCodeEmitter.cpp:547!<br>
<br>
</span>This one at least is understandable. Your options imply (I couldn't<br>
find any "llvm_interpret" function, so there's some guesswork) that<br>
you're using the old JIT. That's been discouraged for a while, and<br>
it's been removed completely now in trunk.<br>
<br>
It's entirely possible it could randomly fall over (not all<br>
instructions are supported), and probably not even worth worrying<br>
about why. I'd just flip that "UseMCJIT" option.<br>
<br>
The interpreter failure you were seeing earlier is harder to explain<br>
(there are various options), but if we're lucky it won't happen in<br>
MCJIT mode. Then we don't have to worry about that one either.<br>
<br>
Cheers.<br>
<span><font color="#888888"><br>
Tim.<br>
</font></span></blockquote></div><br></div>
</div></div></blockquote></div><br></div>