Hi Ben,<br><br>On Tuesday, June 14, 2016, Ben Coman <<a href="mailto:btc@openinworld.com">btc@openinworld.com</a>> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><br>
Some other possible hints of transitions between structures is that<br>
only some methods have pragmas?? (Like <var: #cogMethod type:.....)</blockquote><div><br></div><div>Unrelated. These pragmas merely confer C types in variables when Smalltalk/Slang is translated to C.</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
Only some methods are marked for <inline>.</blockquote><div><br></div><div>Right. Some methods must be inlined, some are inlined automatically. It's a bit chaotic, but us unrelated to the transitions.</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
I notice a function... convertToMachineCodeFrame: cogMethod bcpc:</blockquote><div><br></div><div>Now you're talking :-). The interpreter jits an interpreted method that contains a loop by counting backward branches and jitting by default on the 20th iteration.</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
cheers- ben<br>
</blockquote><br><br>-- <br><div dir="ltr"><div><span style="font-size:small;border-collapse:separate"><div>_,,,^..^,,,_<br></div><div>best, Eliot</div></span></div></div><br>