Add mon_cross_ctx_copies benchmark and make target

Benchmark supports:
=> Generating parallel copies from CPU to GPU in seperate process
=> Granular timestamping of when each portion of the copy completes
=> Monitoring of copy process from either CPU or GPU
=> Tested on platforms including the GV100 and GV11B

Formerly known as diff_ctx_diff_stream.

*** Warning ***
* The current implementation requires some hardcoded parameters
* (such as the number of monitoring threads on the GPU) to be
* updated and recompiled when switching test platform.
***

1 files changed, 8 insertions, 0 deletions

diff --git a/copy_experiments/Makefile b/copy_experiments/Makefile
index b6b5a38..6ccec54 100644
--- a/copy_experiments/Makefile
+++ b/copy_experiments/Makefile
@@ -1,2 +1,10 @@
+.PHONY: all
+
+all: copy_contender mon_cross_ctx_copies
+
 copy_contender: copy_contender.cu copy_testbench.h ../testbench.h
         /usr/local/cuda/bin/nvcc copy_contender.cu -o $@ -lcuda -g
+
+mon_cross_ctx_copies: mon_cross_ctx_copies.cu copy_testbench.h ../testbench.h
+        /usr/local/cuda/bin/nvcc mon_cross_ctx_copies.cu -o $@ -lcuda -lpthread -g
+