[llvm-testresults] Build using GCC 4.2.0 RC3 on Linux x86 32 Bit

Aaron Gray angray at beeb.net
Wed May 9 11:03:21 PDT 2007


>>>  Have you thought about running a nightly tester? It will do this for 
>>> you automatically and also provide more information about the build and 
>>> what errors you are getting. Just the stats are ok, but its better to 
>>> know which tests are failing.
>>
>> No. Thats sounds good.
>
> Actually, I just realized you are using Cygwin. I don't know if anyone 
> else is running a nightly tester on cygwin. It would be cool to see that 
> (and if there are any problems).

Yes, I am doing Cygwin as well as Linux 32 bit x86.

Cygwin does not look too good though. There are alot of errors and some 
possible critical faults.

>>>  Also, how are you building the llvm-gcc frontend? Are you rebuilding it 
>>> each time you run this?
>>
>> I am building llvm-gcc4 rather than llvm-gcc3. I am only rebuilding it 
>> each time theres SVN updates.
>
> I meant llvm-gcc4 by default :) llvm-gcc3 will be dead as of 2.0. So its 
> good you are building llvm-gcc4.
>
>>>  What configure and build options?
>>
>> The proceedure I am using is near enought the same as the Cygwin 
>> proceedure :-
>>
>>       http://www.aarongray.org/LLVM/BuildingLLVMonCygwin.html
>>
>>>  I'm seeing build errors on linux right now.
>>
>> There are two errors one with 'bit intrinsics' 
>> (llvm/test/CodeGen/Generic/bit-intrinsics.ll) which I think should be a 
>> expected failure for now. The other error 
>> (llvm/test/Transforms/LICM/scalar_promote.ll) I do not know about.
>
> Ah ok. I meant compile errors which you don't seem to get. Interesting.

I was building from mainstream CVS and SVN.

> Thanks for the test results!

Do you want me to test the candidate release CVS on Linux ? I was going to 
wait for tar balls but I can do a CVS test if that would help.

> -Tanya
>
>
> -- 
> No virus found in this incoming message.
> Checked by AVG Free Edition. Version: 7.5.467 / Virus Database: 
> 269.6.6/794 - Release Date: 08/05/2007 14:23
>
> 




More information about the llvm-testresults mailing list