[llvm-commits] [llvm] r172534 - /llvm/trunk/test/Transforms/LoopStrengthReduce/post-inc-icmpzero.ll

Renato Golin rengolin at systemcall.org
Tue Jan 15 08:52:47 PST 2013


Hi Dmitri,

The output is not identical because the command doesn't specify the
triple, but the relationship between the constructs is still the same.

Regardless if this was intended to run on ARM, I think that having a
pattern-match approach is preferably to a specific name, especially
when concerning SSA register names.

cheers,
--renato

On 15 January 2013 16:41, Dmitri Gribenko <gribozavr at gmail.com> wrote:
> On Tue, Jan 15, 2013 at 5:22 PM, Renato Golin <rengolin at systemcall.org> wrote:
>> Author: rengolin
>> Date: Tue Jan 15 09:22:45 2013
>> New Revision: 172534
>>
>> URL: http://llvm.org/viewvc/llvm-project?rev=172534&view=rev
>> Log:
>> Pattern-matched variables in post-inc-icmpzero.ll
>>
>> Test was failing for clang-native-arm-cortex-a9 build-bot configuration.
>> The reason for the failure was the test was using hardcoded names.
>> The attached patch fixes this failure by replacing the hard-coded variables
>> names with pattern-matched variable names.
>
> Is the generated IR otherwise identical? (i.e., there is no extra
> instruction named %3)  It is not possible to see from the buildbot
> output.
>
> Dmitri
>
> --
> main(i,j){for(i=2;;i++){for(j=2;j<i;j++){if(!(i%j)){j=0;break;}}if
> (j){printf("%d\n",i);}}} /*Dmitri Gribenko <gribozavr at gmail.com>*/



More information about the llvm-commits mailing list