<div dir="ltr"><br><div class="gmail_extra"><br><br><div class="gmail_quote">On Thu, Sep 4, 2014 at 5:36 AM, "C. Bergström" <span dir="ltr"><<a href="mailto:cbergstrom@pathscale.com" target="_blank">cbergstrom@pathscale.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><br>
Hi<br>
<br>
Would patches for cleaning up unnecessary #include for libc++ be accepted? In general it seems like it has followed the rather poorly designed GNU STL rather than being more strict.<br>
<br>
A simple preprocessed hello world which only #include <iostream><br>
<br>
CC -E hello.cxx -std=c++11 > hello-libcxx.e.cxx<br>
<br>
du -sh *.e.cxx<br>
532K hello-stdcxx.e.cxx<br>
1.3M hello-libcxx.e.cxx<br>
<br>
The effects of this cascade into being a measurable performance difference later on in the compiler. (My numbers may not be exactly llvm upstream though - ymmv) Basically the Module size and other things are larger as a result.<br>
<br>
Thoughts/feedback?<br></blockquote><div><br></div><div>The code for the two libraries is different though. How did your measurements correct for that, so that you can so surely attribute it to unnecessary includes, and not simply differences in implementation?</div>
<div><br></div><div>-- Sean Silva</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
______________________________<u></u>_________________<br>
cfe-dev mailing list<br>
<a href="mailto:cfe-dev@cs.uiuc.edu" target="_blank">cfe-dev@cs.uiuc.edu</a><br>
<a href="http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev" target="_blank">http://lists.cs.uiuc.edu/<u></u>mailman/listinfo/cfe-dev</a><br>
</blockquote></div><br></div></div>