<div dir="ltr">Hi David,<div><br></div><div>I've added a testcase in r207697. I realised why I coudn't reproduce it - it only showed up when emitting ELF, not assembly.</div><div><br></div><div>Thanks for the prod!</div>
<div><br></div><div>James</div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On 30 April 2014 19:41, David Blaikie <span dir="ltr"><<a href="mailto:dblaikie@gmail.com" target="_blank">dblaikie@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="">On Wed, Apr 30, 2014 at 11:25 AM, James Molloy <<a href="mailto:James.Molloy@arm.com">James.Molloy@arm.com</a>> wrote:<br>
> Hi David,<br>
><br>
> It looks like it should be trivial to test, however I failed :(<br>
><br>
> This was originally noticed because of a disparity between the layout of a constant pool and a MOVK'd immediate.<br>
<br>
</div>Sounds like you've got a reproduction, then.<br>
<div class=""><br>
> When writing a testcase (albeit after the fact - should probably have done it at the same time) I couldn't get it to trigger.<br>
><br>
> I can try harder to get a testcase tomorrow when I'm not as bleary-eyed as I was when I tried before though if you like?<br>
<br>
</div>That'd be swell, thanks.<br>
<div class=""><br>
> I definitely agree with you that everything should be tested if possible. It's just that stuff in the backend gets difficult to test.<br>
<br>
</div>Yeah - there are sometimes good reasons not to commit tests for some<br>
of this stuff (mostly around the inability to run small pieces of the<br>
backend in isolation so tests tend to be rather fragile due to other<br>
optimizations in the backends causing the test to fail or to no longer<br>
test their intent) - but this particular one seems like a pretty<br>
significant/important distinction. (which endianness we use... )<br>
<div class="HOEnZb"><div class="h5"><br>
><br>
> Cheers,<br>
><br>
> James<br>
><br>
>> -----Original Message-----<br>
>> From: David Blaikie [mailto:<a href="mailto:dblaikie@gmail.com">dblaikie@gmail.com</a>]<br>
>> Sent: 30 April 2014 18:55<br>
>> To: James Molloy<br>
>> Cc: <a href="mailto:llvm-commits@cs.uiuc.edu">llvm-commits@cs.uiuc.edu</a><br>
>> Subject: Re: [llvm] r207627 - [ARM64] Fix stupid copy-pasto in<br>
>> ARM64MCAsmInfo.cpp - aarch64_be -> arm64_be<br>
>><br>
>> On Wed, Apr 30, 2014 at 3:15 AM, James Molloy <<a href="mailto:james.molloy@arm.com">james.molloy@arm.com</a>><br>
>> wrote:<br>
>> > Author: jamesm<br>
>> > Date: Wed Apr 30 05:15:46 2014<br>
>> > New Revision: 207627<br>
>> ><br>
>> > URL: <a href="http://llvm.org/viewvc/llvm-project?rev=207627&view=rev" target="_blank">http://llvm.org/viewvc/llvm-project?rev=207627&view=rev</a><br>
>> > Log:<br>
>> > [ARM64] Fix stupid copy-pasto in ARM64MCAsmInfo.cpp - aarch64_be -><br>
>> arm64_be<br>
>><br>
>> I'm assuming this is testable?<br>
>><br>
>> > Modified:<br>
>> > llvm/trunk/lib/Target/ARM64/MCTargetDesc/ARM64MCAsmInfo.cpp<br>
>> ><br>
>> > Modified:<br>
>> llvm/trunk/lib/Target/ARM64/MCTargetDesc/ARM64MCAsmInfo.cpp<br>
>> > URL: <a href="http://llvm.org/viewvc/llvm-" target="_blank">http://llvm.org/viewvc/llvm-</a><br>
>> project/llvm/trunk/lib/Target/ARM64/MCTargetDesc/ARM64MCAsmInfo.cp<br>
>> p?rev=207627&r1=207626&r2=207627&view=diff<br>
>> ><br>
>> ==========================================================<br>
>> ====================<br>
>> > --- llvm/trunk/lib/Target/ARM64/MCTargetDesc/ARM64MCAsmInfo.cpp<br>
>> (original)<br>
>> > +++ llvm/trunk/lib/Target/ARM64/MCTargetDesc/ARM64MCAsmInfo.cpp<br>
>> Wed Apr 30 05:15:46 2014<br>
>> > @@ -66,7 +66,7 @@ const MCExpr *ARM64MCAsmInfoDarwin::getE<br>
>> ><br>
>> > ARM64MCAsmInfoELF::ARM64MCAsmInfoELF(StringRef TT) {<br>
>> > Triple T(TT);<br>
>> > - if (T.getArch() == Triple::aarch64_be)<br>
>> > + if (T.getArch() == Triple::arm64_be)<br>
>> > IsLittleEndian = false;<br>
>> ><br>
>> > // We prefer NEON instructions to be printed in the short form.<br>
>> ><br>
>> ><br>
>> > _______________________________________________<br>
>> > llvm-commits mailing list<br>
>> > <a href="mailto:llvm-commits@cs.uiuc.edu">llvm-commits@cs.uiuc.edu</a><br>
>> > <a href="http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits" target="_blank">http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits</a><br>
><br>
><br>
> -- IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.<br>
><br>
> ARM Limited, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2557590<br>
> ARM Holdings plc, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2548782<br>
<br>
_______________________________________________<br>
llvm-commits mailing list<br>
<a href="mailto:llvm-commits@cs.uiuc.edu">llvm-commits@cs.uiuc.edu</a><br>
<a href="http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits" target="_blank">http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits</a><br>
</div></div></blockquote></div><br></div>