Hi,

Here is a new wiki page for benchmarking with cbuild. I separated it into one part describing how to use the autobuilders, and one part for setting stuff up manually, for instance for running on the ursas.

https://wiki.linaro.org/WorkingGroups/ToolChain/Benchmarks/RunningBenchmarksWithCbuild

Regards
Åsa


On 27 February 2012 22:27, Michael Hope <michael.hope@linaro.org> wrote:
Here's brief notes on running different benchmark variants across the
auto builders.  Asa, could you pull these plus your notes into a wiki
page?

Spawn a job:
 http://ex.seabright.co.nz/helpers/scheduler/spawn

Merge requests are automatically built.  Otherwise, drop arbitrary
tarballs into cbuild@orion:~/snapshots and spawn <tarball name minus
extension>.  For example, scp gcc-4.6.3.tar.gz
cbuild@orion:~/snapshots; spawn gcc-4.6.3 into a9-builder.

Jobs:
 * gcc-version - build and test GCC
 * benchmarks-gcc-version - run coremark, denbench, eembc against the
already built version
 * benchmarks-spec2000-gcc-version - run spec2000

Queues:
 * a9-builders: anything that can naive build a A9 compiler
 * a9-ref: reference A9 boards
 * a8-ref: reference A8 boards

Variables:
 * BENCHMARKS = list, such as coremark spec2000 pybench - run these
benchmarks instead of the defaults

Variants:
 * By default we build o3-neon
 * See http://bazaar.launchpad.net/~linaro-toolchain-dev/cbuild/trunk/view/head:/lib/all-variants.mk
for all names
 * Spawn a job with VARIANT_SRC = all and VARIANT_LIST = glob-pattern

Examples:
 * VARIANT_LIST = o3-neon o3-vfpv3 (compare NEON with VFPv3D32)
 * VARIANT_LIST = o3-neon-cortexa8 o3-neon-cortexa9 (compare
-mtune=cortexa8 vs -mtune=cortexa9(
 * VARIANT_LIST = o3-neon-novect o3-neon (compare with/without the vectoriser)

-- Michael

_______________________________________________
linaro-toolchain mailing list
linaro-toolchain@lists.linaro.org
http://lists.linaro.org/mailman/listinfo/linaro-toolchain