[llvm] r288884 - Fix a warning introduced in r288874

[Philip Reames via llvm-commits]

One of the self-host bots which broke with the same message has now 
cycled clean.  I'm assuming this is fixed.

Again, I would appreciate help from a clang developer to extract a LLVM 
IR test case which illustrates the problem so that I can reduce it and 
add it to the LLVM test suite.

Philip


On 12/06/2016 08:59 PM, Philip Reames via llvm-commits wrote:
>
> Hopefully fixed in 288886.  Help extracting a test case would still be 
> appreciated.
>
> On 12/06/2016 08:35 PM, Philip Reames via llvm-commits wrote:
>>
>> Yes.  I'm pretty sure I know which change and will reintroduce the 
>> check I removed.  However, if someone could extract a IR test case 
>> which triggers this, I'd appreciate it.  I don't keep a prebuilt copy 
>> of clang around and it looks like only the clang tests catch this.  
>> I'd like to understand how exactly we got here.
>>
>> Philip
>>
>>
>> On 12/06/2016 08:29 PM, Peter Collingbourne wrote:
>>> Hi Philip,
>>>
>>> One of our chromium bots has started failing with an assertion 
>>> failure in LVI.
>>>
>>> https://build.chromium.org/p/chromium.fyi/builders/ThinLTO%20Linux%20ToT/builds/635/steps/compile/logs/stdio
>>>
>>> Could it be possible that one of your recent changes has caused this?
>>>
>>> I'll see if I can get you a reproducer if you can't figure it out.
>>>
>>> Peter
>>>
>>>
>>> On Dec 6, 2016 20:21, "Philip Reames via llvm-commits" 
>>> <llvm-commits at lists.llvm.org <mailto:llvm-commits at lists.llvm.org>> 
>>> wrote:
>>>
>>>     Author: reames
>>>     Date: Tue Dec  6 22:11:22 2016
>>>     New Revision: 288884
>>>
>>>     URL: http://llvm.org/viewvc/llvm-project?rev=288884&view=rev
>>>     <http://llvm.org/viewvc/llvm-project?rev=288884&view=rev>
>>>     Log:
>>>     Fix a warning introduced in r288874
>>>
>>>
>>>     Modified:
>>>         llvm/trunk/lib/Analysis/LazyValueInfo.cpp
>>>
>>>     Modified: llvm/trunk/lib/Analysis/LazyValueInfo.cpp
>>>     URL:
>>>     http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Analysis/LazyValueInfo.cpp?rev=288884&r1=288883&r2=288884&view=diff
>>>     <http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Analysis/LazyValueInfo.cpp?rev=288884&r1=288883&r2=288884&view=diff>
>>>     ==============================================================================
>>>     --- llvm/trunk/lib/Analysis/LazyValueInfo.cpp (original)
>>>     +++ llvm/trunk/lib/Analysis/LazyValueInfo.cpp Tue Dec  6
>>>     22:11:22 2016
>>>     @@ -186,7 +186,6 @@ private:
>>>            if (NewR.isEmptySet())
>>>              markOverdefined();
>>>            else {
>>>     -        bool changed = Range != NewR;
>>>              Range = std::move(NewR);
>>>            }
>>>            return;
>>>
>>>
>>>     _______________________________________________
>>>     llvm-commits mailing list
>>>     llvm-commits at lists.llvm.org <mailto:llvm-commits at lists.llvm.org>
>>>     http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits
>>>     <http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits>
>>>
>>
>>
>>
>> _______________________________________________
>> llvm-commits mailing list
>> llvm-commits at lists.llvm.org
>> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits
>
>
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at lists.llvm.org
> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20161206/bc29fea0/attachment.html>