<div class="gmail_quote">
<div class="gmail_quote">
<div class="im">2011/4/7 Siarhei Siamashka <span dir="ltr"><<a href="mailto:siarhei.siamashka@gmail.com" target="_blank">siarhei.siamashka@gmail.com</a>></span><br>
<blockquote style="BORDER-LEFT: #ccc 1px solid; MARGIN: 0px 0px 0px 0.8ex; PADDING-LEFT: 1ex" class="gmail_quote">Hmm, I just realized that you probably suggested using simple function<br>calls to the shared parts of code, right? It could be beneficial in<br>
the cases like this, but requires to always have lr register free for<br>such purpose, and we are short on registers in some fast paths.<br>Hopefully i-cache misses are not a big problem in pixman NEON<br>optimized fast paths. Or do you have some statistics proving that they<br>
are?<br></blockquote></div>
<div> </div>
<div>I just wanted to point that macro expension of basic building blocks can be a potential problem causing larger code size.</div>
<div>As you mentioned, we can think of using lr register or some dynamic runtime code fetching.</div>
<div>In practice, i-cache miss (including i-TLB miss) is not that a big problem especially for these kind of code blocks in loop.</div>
<div>I'm a bit gone too much in that point.</div>
<div> </div>
<div>What can be the best approach to maximizing code utilization in binary level (not source level)??</div>
<div>It does not seem that we have better choice than dynamic code fetching.</div>
<div>And is it really worth doing that?</div>
<div><br>-- <br>Best Regards, </div>
<div>Taekyun Kim</div></div></div>