<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 12pt;
font-family:Calibri
}
--></style></head>
<body class='hmmessage'><div dir='ltr'><div>Hi Renato,</div><div><br></div><div> Ok so far I have tested with RC1 and RC2 they both show the regression.</div><div><br></div><div> Ok I will try to create a repo and get back in touch.</div><div><br></div><div>Thanks</div><div><br></div><div>Dan</div><div><br><br></div><div>> From: renato.golin@linaro.org<br>> Date: Wed, 24 Aug 2016 13:39:56 +0100<br>> Subject: Re: [cfe-dev] Regression in Release Candidate 2<br>> To: dan@walms.co.uk<br>> CC: hans@chromium.org; cfe-dev@lists.llvm.org<br>> <br>> On 24 August 2016 at 11:56, Dan Walmsley via cfe-dev<br>> <cfe-dev@lists.llvm.org> wrote:<br>> ><br>> > Hi,<br>> ><br>> > I noticed a regression in this build...<br>> ><br>> > libclang, code completeAt...<br>> > the results returned inbuilt libraries, i.e. memset<br>> ><br>> > all come back as char* functionname( void)<br>> ><br>> > if I go back to older build (3.7.1) then this information returns.<br>> <br>> Hi Dan,<br>> <br>> Can you narrow down the problem a bit more?<br>> <br>> Is this only with the RC2? Or also with RC1? Does it work with 3.8.0<br>> or 3.8.1? Some version of trunk?<br>> <br>> Also, that is a *very* high level description of the problem and it's<br>> really hard to reproduce it. This may be a problem in backwards<br>> compatibility of libclang (which shouldn't happen), or it could be<br>> that the completeAt usage was assuming something that wasn't true.<br>> <br>> If you could create a bug explaining the process, stating the most<br>> recent version that you can see it working, and how to reproduce<br>> without any additional tools (ie. by using libclang directly,<br>> preferably with a piece of code that does it), then Clang folks could<br>> have a better look at that.<br>> <br>> cheers,<br>> --renato<br></div> </div></body>
</html>