[llvm-commits] [test-suite] r164834 - in /test-suite/trunk/MultiSource/Benchmarks: ./ TSVC/ TSVC/ControlFlow-dbl/ TSVC/ControlFlow-flt/ TSVC/ControlLoops-dbl/ TSVC/ControlLoops-flt/ TSVC/CrossingThresholds-dbl/ TSVC/CrossingThresholds-flt/ TSVC/Equivalencing-dbl/ TSVC/Equivalencing-flt/ TSVC/Expansion-dbl/ TSVC/Expansion-flt/ TSVC/GlobalDataFlow-dbl/ TSVC/GlobalDataFlow-flt/ TSVC/IndirectAddressing-dbl/ TSVC/IndirectAddressing-flt/ TSVC/InductionVariable-dbl/ TSVC/InductionVariable-flt/ TSVC/LinearDependence-dbl/ TSVC/...

[Hal Finkel]

On Fri, 28 Sep 2012 14:08:11 -0500
John Criswell <criswell at illinois.edu> wrote:

> On 9/28/12 1:33 PM, Hal Finkel wrote:
> > Author: hfinkel
> > Date: Fri Sep 28 13:33:27 2012
> > New Revision: 164834
> >
> > URL: http://llvm.org/viewvc/llvm-project?rev=164834&view=rev
> > Log:
> > Add the TSVC loop benchmarks to the test suite.
> 
> Please be sure to add a LICENSE.TXT file to the TSVC directory and 
> update the top-level LICENSE.TXT file in test-suite.

Done, thanks! r164838.

 -Hal

> 
> -- John T.
> 
> >
> > Compared to the original benchmark, I've made the following
> > enhancements:
> >
> >   - I made the number of test iterations a command-line parameter
> > (this makes it easy to adjust the run time), and also the output
> > precision
> >   - I made it possible to run the tests using both float or double
> >   - I split out the different classes of loops into separate tests
> > (I did this primarily to make performance improvements and
> > regressions noticeable)
> >   - I removed the timing information from the output
> >
> >
> >
> 



-- 
Hal Finkel
Postdoctoral Appointee
Leadership Computing Facility
Argonne National Laboratory