<html><head><meta http-equiv="Content-Type" content="text/html charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class="">Hi James,<div class=""><br class=""></div><div class=""><br class=""><div><blockquote type="cite" class=""><div class="">On Aug 31, 2016, at 1:33 PM, James Molloy <<a href="mailto:james@jamesmolloy.co.uk" class="">james@jamesmolloy.co.uk</a>> wrote:</div><br class="Apple-interchange-newline"><div class="">Hi,<br class=""><br class="">Printf could call main. If it did so, demotion to a local would be illegal. <br class=""><br class="">You can either compile in c++ mode (which mandates that main is not recursive) or add '-force-attribute=main:norecurse' to get this behaviour. <br class=""></div></blockquote><div><br class=""></div><div>Thanks for the hint! You’re right, building in c++ mode vs c mode is the key point...</div><div><br class=""></div><div>— </div><div>Mehdi</div><div><br class=""></div><br class=""><blockquote type="cite" class=""><div class=""><br class="">James<br class=""><div class="gmail_quote"><div dir="ltr" class="">On Wed, 31 Aug 2016 at 20:52, Mehdi Amini via llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" class="">llvm-dev@lists.llvm.org</a>> wrote:<br class=""></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word" class=""><div class=""><blockquote type="cite" class=""><div class="">On Aug 30, 2016, at 2:57 PM, Ryan Taylor <<a href="mailto:ryta1203@gmail.com" target="_blank" class="">ryta1203@gmail.com</a>> wrote:</div><br class=""></blockquote></div></div><div style="word-wrap:break-word" class=""><div class=""><blockquote type="cite" class=""><div class=""><div dir="ltr" class=""><div class="">Yes, the full test case is:</div><div class=""><br class=""></div><div class="">static int x = 100;</div><div class="">int y = whatever;</div><div class=""><br class=""></div><div class="">int main() {</div><div class=""> x = -101;</div><div class=""> y = whatever that's not whatever above;</div><div class=""> printf("%d\n", y);</div><div class=""> printf("%d\n", x);</div><div class=""> return 0;</div><div class="">}</div><div class=""><br class=""></div></div></div></blockquote></div></div><div style="word-wrap:break-word" class=""><div class=""><blockquote type="cite" class=""><div class=""><div dir="ltr" class=""><div class="">You are correct, in the above test case the globalopt does not make the transformation…. </div></div></div></blockquote><div class=""><br class=""></div><div class="">I’m more annoyed that it does not turn x into an alloca. With this code:</div><div class=""><br class=""></div><div class="">$ cat /tmp/test.c</div><div class=""></div></div></div><div style="word-wrap:break-word" class=""><div class=""><div class=""><div class="">static int x = 100;</div></div></div></div><div style="word-wrap:break-word" class=""><div class=""><div class=""><div class="">int y = 2;</div></div></div></div><div style="word-wrap:break-word" class=""><div class=""><div class=""><div class="">int main() {</div><div class=""> x = -101;</div></div></div></div><div style="word-wrap:break-word" class=""><div class=""><div class=""><div class=""> y = 1; </div></div></div></div><div style="word-wrap:break-word" class=""><div class=""><div class=""><div class=""> printf("%d\n", y);</div><div class=""> printf("%d\n", x);</div><div class=""> return 0;</div><div class="">}</div></div></div></div><div style="word-wrap:break-word" class=""><div class=""><div class=""><div class="">$ clang -O3 -S -emit-llvm -o - /tmp/test.c </div><div class="">; ModuleID = '/tmp/test.c'<br class="">source_filename = "/tmp/test.c"<br class="">target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"<br class="">target triple = "x86_64-apple-macosx10.12.0"<br class=""><br class="">@y = global i32 2, align 4<br class="">@x = internal unnamed_addr global i1 false, align 4<br class="">@.str = private unnamed_addr constant [4 x i8] c"%d\0A\00", align 1<br class=""><br class="">; Function Attrs: nounwind ssp uwtable<br class="">define i32 @main() #0 {<br class=""> store i1 true, i1* @x, align 4<br class=""> store i32 1, i32* @y, align 4, !tbaa !2<br class=""> %1 = tail call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str, i64 0, i64 0), i32 1)<br class=""> %2 = load i1, i1* @x, align 4<br class=""> %3 = select i1 %2, i32 -101, i32 100<br class=""> %4 = tail call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str, i64 0, i64 0), i32 %3)<br class=""> ret i32 0<br class="">}<br class=""><br class=""></div></div></div></div><div style="word-wrap:break-word" class=""><div class=""><br class=""><blockquote type="cite" class=""><div class=""><div dir="ltr" class=""><div class="">however, I think the original issue still stands, it really shouldn't be doing it here either.</div></div></div></blockquote><div class=""><br class=""></div></div></div><div style="word-wrap:break-word" class=""><div class=""><div class="">Yes.</div><div class=""><br class=""></div><div class="">— </div></div></div><div style="word-wrap:break-word" class=""><div class=""><div class="">Mehdi</div></div></div><div style="word-wrap:break-word" class=""><div class=""><div class=""><br class=""></div><div class=""><br class=""></div><br class=""><blockquote type="cite" class=""><div class=""><div class="gmail_extra"><br class=""><div class="gmail_quote">On Tue, Aug 30, 2016 at 5:52 PM, Mehdi Amini <span dir="ltr" class=""><<a href="mailto:mehdi.amini@apple.com" target="_blank" class="">mehdi.amini@apple.com</a>></span> wrote:<br class=""><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class=""><br class="">
> On Aug 30, 2016, at 1:01 PM, Ryan Taylor via llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" target="_blank" class="">llvm-dev@lists.llvm.org</a>> wrote:<br class="">
><br class="">
</span><span class="">> Given some code:<br class="">
><br class="">
> static int x = 100;<br class="">
><br class="">
> int main() {<br class="">
> x = 101;<br class="">
> printf("%d\n", x);<br class="">
> }<br class="">
><br class="">
> results in:<br class="">
><br class="">
> %0 = load i1<br class="">
> %1 = select %0, 101, 100<br class="">
<br class="">
</span>If x is only touched in one function, I’d expect globlaopt to turn it into an alloca instead of leaving it global and shrinking it.<br class="">
Did you oversimplify the real case where you see this for the sake of the example here?<br class="">
<br class="">
—<br class="">
Mehdi<br class="">
<span class=""><br class="">
<br class="">
> ...<br class="">
> ...<br class="">
><br class="">
> 1) What architecture(s) does this benefit?<br class="">
> 2) What's the best way to circumvent this in the backend (currently I am just not allowing this opt function to run)? I have tried a few setOperationAction methods to no result. Rather not just comment out core code because it makes porting more difficult.<br class="">
><br class="">
> -Ryan<br class="">
</span>> _______________________________________________<br class="">
> LLVM Developers mailing list<br class="">
> <a href="mailto:llvm-dev@lists.llvm.org" target="_blank" class="">llvm-dev@lists.llvm.org</a><br class="">
> <a href="http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev" rel="noreferrer" target="_blank" class="">http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a><br class="">
<br class="">
</blockquote></div><br class=""></div>
</div></blockquote></div></div>_______________________________________________<br class="">
LLVM Developers mailing list<br class="">
<a href="mailto:llvm-dev@lists.llvm.org" target="_blank" class="">llvm-dev@lists.llvm.org</a><br class="">
<a href="http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev" rel="noreferrer" target="_blank" class="">http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a><br class="">
</blockquote></div>
</div></blockquote></div><br class=""></div></body></html>