mirror of
https://github.com/tiagovignatti/intel-gpu-tools.git
synced 2025-06-10 09:26:10 +00:00
If we autotune the workload to only take 0.1s and then repeat the measurements over 2s, we can bound the benchmark runtime. (Roughly of course! Sometimes the dispartity between main memory CPU bandwidth, and GPU execution bandwidth throws off the runtime, but that's the purpose of the benchmark!) Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
These benchmarks are designed to be used from within a benchmark framework like http://cgit.freedesktop.org/~mperes/ezbench/. A typical way to run them would be: $ ~/ezbench/ezbench.sh -p ./linux,git -m '' -b gem: -r 15 which executes the set of gem benchmarks, 15 times each, using HEAD of ./linux.git as the reference commit.