<div dir="ltr"><br><div class="gmail_extra"><br><div class="gmail_quote">On Tue, Feb 9, 2016 at 7:25 AM, Sanjay Patel <span dir="ltr"><<a href="mailto:spatel@rotateright.com" target="_blank">spatel@rotateright.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">spatel added a subscriber: spatel.<br>
spatel added a comment.<br>
<span class=""><br>
In <a href="http://reviews.llvm.org/D15393#347116" rel="noreferrer" target="_blank">http://reviews.llvm.org/D15393#347116</a>, @silvas wrote:<br>
<br>
> One interesting thing to also note: the 11, 9, and 12 rows have such bloated instruction sizes primarily because of needing to store an immediate "0" as an imm32. (we have some open PR's about this I think).<br>
<br>
<br>
</span>Excellent data visualization - thanks! If you have a chance, it would be interesting to recompile the app that you're looking at with this patch applied to see if that app benefits.<br></blockquote><div><br></div><div><div>My findings are in this PDF of my Mathematica notebook: <a href="https://drive.google.com/file/d/0B8v10qJ6EXRxVVc1NVpMNk1UV0k/view?usp=sharing">https://drive.google.com/file/d/0B8v10qJ6EXRxVVc1NVpMNk1UV0k/view?usp=sharing</a></div></div><div><br></div><div>The summary w.r.t. text size is that there is about a 1% text size improvement.</div><div><br></div><div>Also, there is a measurable performance improvement on the level that I was testing on (sorry, I didn't do a more in-depth test across multiple levels).</div><div>The 99'th percentile frame time is improved by a whopping 1.6%! The median and 90'th percentile only improve about half a percent, but that is still a very respectable improvement for general-purpose backend tuning!</div><div><br></div><div>Note: this was for a regular O3 build, not a peak performance build (LTO+PGO). However, this is likely to be more important during LTO since there is generally greater inlining and so greater pressure on stack frames. Sorry, didn't have enough time to measure all the {configurations} x {levels} like when I'm doing an in-depth analysis!</div><div><br></div><div>-- Sean Silva</div><div><br></div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
<br>
The related instruction-size bugzillas that I filed are:<br>
<a href="https://llvm.org/bugs/show_bug.cgi?id=24447" rel="noreferrer" target="_blank">https://llvm.org/bugs/show_bug.cgi?id=24447</a><br>
<a href="https://llvm.org/bugs/show_bug.cgi?id=24448" rel="noreferrer" target="_blank">https://llvm.org/bugs/show_bug.cgi?id=24448</a><br>
<a href="https://llvm.org/bugs/show_bug.cgi?id=24449" rel="noreferrer" target="_blank">https://llvm.org/bugs/show_bug.cgi?id=24449</a><br>
<br>
We made some progress on the example in the last report, but that may not apply to the cases that you were looking at. It's possible that this helps too:<br>
<a href="http://reviews.llvm.org/rL256969" rel="noreferrer" target="_blank">http://reviews.llvm.org/rL256969</a><br>
<br>
<br>
<a href="http://reviews.llvm.org/D15393" rel="noreferrer" target="_blank">http://reviews.llvm.org/D15393</a><br>
<br>
<br>
<br>
</blockquote></div><br></div></div>