<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=""><br class=""><div><blockquote type="cite" class=""><div class="">On Jan 20, 2015, at 11:10 AM, David Blaikie <<a href="mailto:dblaikie@gmail.com" class="">dblaikie@gmail.com</a>> wrote:</div><br class="Apple-interchange-newline"><div class=""><div dir="ltr" class=""><br class=""><div class="gmail_extra"><br class=""><div class="gmail_quote">On Tue, Jan 20, 2015 at 10:58 AM, Adrian Prantl <span dir="ltr" class=""><<a href="mailto:aprantl@apple.com" target="_blank" class="">aprantl@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"><br class="">
<br class="">
> Begin forwarded message:<br class="">
><br class="">
> Subject: Re: [llvm] r226588 - Add an assertion and prefer a crash over an infinite loop.<br class="">
> From: "Duncan P. N. Exon Smith" <<a href="mailto:dexonsmith@apple.com" class="">dexonsmith@apple.com</a>><br class="">
> Date: January 20, 2015 at 10:53:38 AM PST<br class="">
> Cc: <a href="mailto:llvm-commits@cs.uiuc.edu" class="">llvm-commits@cs.uiuc.edu</a><br class="">
> To: Adrian Prantl <<a href="mailto:aprantl@apple.com" class="">aprantl@apple.com</a>><br class="">
<div class=""><div class="h5">><br class="">
><br class="">
>> On 2015 Jan 20, at 10:03, Adrian Prantl <<a href="mailto:aprantl@apple.com" class="">aprantl@apple.com</a>> wrote:<br class="">
>><br class="">
>> Author: adrian<br class="">
>> Date: Tue Jan 20 12:03:37 2015<br class="">
>> New Revision: 226588<br class="">
>><br class="">
>> URL: <a href="http://llvm.org/viewvc/llvm-project?rev=226588&view=rev" target="_blank" class="">http://llvm.org/viewvc/llvm-project?rev=226588&view=rev</a><br class="">
>> Log:<br class="">
>> Add an assertion and prefer a crash over an infinite loop.<br class="">
>><br class="">
>> Modified:<br class="">
>> llvm/trunk/lib/IR/DebugInfo.cpp<br class="">
>><br class="">
>> Modified: llvm/trunk/lib/IR/DebugInfo.cpp<br class="">
>> URL: <a href="http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/IR/DebugInfo.cpp?rev=226588&r1=226587&r2=226588&view=diff" target="_blank" class="">http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/IR/DebugInfo.cpp?rev=226588&r1=226587&r2=226588&view=diff</a><br class="">
>> ==============================================================================<br class="">
>> --- llvm/trunk/lib/IR/DebugInfo.cpp (original)<br class="">
>> +++ llvm/trunk/lib/IR/DebugInfo.cpp Tue Jan 20 12:03:37 2015<br class="">
>> @@ -527,12 +527,15 @@ bool DISubprogram::Verify() const {<br class="">
>> while ((IA = DL.getInlinedAt()))<br class="">
>> DL = DebugLoc::getFromDILocation(IA);<br class="">
>> DL.getScopeAndInlinedAt(Scope, IA);<br class="">
>> + assert(Scope && "debug location has no scope");<br class="">
>> assert(!IA);<br class="">
>> while (!DIDescriptor(Scope).isSubprogram()) {<br class="">
>> DILexicalBlockFile D(Scope);<br class="">
>> Scope = D.isLexicalBlockFile()<br class="">
>> ? D.getScope()<br class="">
>> : DebugLoc::getFromDILexicalBlock(Scope).getScope();<br class="">
>> + if (!Scope)<br class="">
>> + llvm_unreachable("lexical block file has no scope");<br class="">
><br class="">
> It's strange that this isn't simply an assertion:<br class="">
><br class="">
> assert(Scope && "lexical block file has no scope");<br class="">
><br class="">
> But I guess that's so that it crashes even when NDEBUG? (Is that<br class="">
> reliable?)<br class="">
<br class="">
</div></div>Yes, that’s the intention (and llvm_unreachable is meant to be used that way).<br class=""></blockquote><div class=""><br class="">I don't believe that's the intention of the LLVM style guide - generally we try to remove conditionals like this and replace the conditional with the assertion so the dead code isn't exercised in release builds.<br class=""></div></div></div></div></div></blockquote><div><br class=""></div><div>Isn’t the point of “unreachable” to let the compiler do whatever it wants in this case?</div><div>I thought LLVM was free to just drop the test entirely in release mode?</div><div><br class=""></div><div>Mehdi</div><div><br class=""></div><br class=""><blockquote type="cite" class=""><div class=""><div dir="ltr" class=""><div class="gmail_extra"><div class="gmail_quote"><div class=""> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<span class="">><br class="">
> If that's the reason, you should probably add a comment to that<br class="">
> effect since otherwise I'm likely to switch it to an assert next<br class="">
> time I look at it ;).<br class="">
><br class="">
> (Also, do you know when this situation arises?)<br class="">
<br class="">
</span>Happened to me when I compiled an old .ll file without the MDLocation upgrade.<br class="">
<span class="HOEnZb"><font color="#888888" class=""><br class="">
-- adrian<br class="">
</font></span><div class="HOEnZb"><div class="h5"><br class="">
><br class="">
>> }<br class="">
>> if (!DISubprogram(Scope).describes(F))<br class="">
>> return false;<br class="">
>><br class="">
>><br class="">
>> _______________________________________________<br class="">
>> llvm-commits mailing list<br class="">
>> <a href="mailto:llvm-commits@cs.uiuc.edu" class="">llvm-commits@cs.uiuc.edu</a><br class="">
>> <a href="http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits" target="_blank" class="">http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits</a><br class="">
><br class="">
<br class="">
<br class="">
_______________________________________________<br class="">
llvm-commits mailing list<br class="">
<a href="mailto:llvm-commits@cs.uiuc.edu" class="">llvm-commits@cs.uiuc.edu</a><br class="">
<a href="http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits" target="_blank" class="">http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits</a><br class="">
</div></div></blockquote></div><br class=""></div></div>
_______________________________________________<br class="">llvm-commits mailing list<br class=""><a href="mailto:llvm-commits@cs.uiuc.edu" class="">llvm-commits@cs.uiuc.edu</a><br class="">http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits<br class=""></div></blockquote></div><br class=""></body></html>