From e63404cfda97050b9d516232ceb4f918f48edcfc Mon Sep 17 00:00:00 2001 From: Joshua Bakita Date: Fri, 9 Jun 2023 18:55:07 -0400 Subject: Add copy_contender benchmark, Makefile, and copy_testbench header Benchmark supports: => Generating a chain of copies in a single context - No timing pause between copies => Copies to, from, or between device(s) => Variable size and number of copies => Logging how many milliseconds each copy takes *** Warning *** * Relies on implicit synchronization of unidirectional copies across * streams. Verify that this is still the case with CUDA before * trusting results. One can check by verifying that the runtime for * one iteration roughly matches the time for each iteration in a * multi-iteration run. *** --- copy_experiments/Makefile | 3 +++ 1 file changed, 3 insertions(+) create mode 100644 copy_experiments/Makefile (limited to 'copy_experiments/Makefile') diff --git a/copy_experiments/Makefile b/copy_experiments/Makefile new file mode 100644 index 0000000..2358c60 --- /dev/null +++ b/copy_experiments/Makefile @@ -0,0 +1,3 @@ + +copy_contender: copy_contender.cu copy_testbench.h + /usr/local/cuda/bin/nvcc copy_contender.cu -o $@ -lcuda -g -- cgit v1.2.2