Compute-benchmarks use already compiled UR #2146
Conversation
github-actions
bot
added
ci/cd
lslusarczyk
force-pushed
the
bench_no_recompile_ur
branch
2 times, most recently
from
e9582f3 to
91cd46a
Compare
lslusarczyk
force-pushed
the
bench_no_recompile_ur
branch
from
91cd46a to
c03d7d0
Compare
| def __init__(self, directory): | ||
| self.directory = directory | ||
| self.adapter_path = os.path.join(options.ur_dir, 'build', 'lib', f"libur_adapter_{options.ur_adapter_name}.so") | ||
| for libs_dir_name in ['lib', 'lib64']: |
There was a problem hiding this comment.
I think the make install fix is fine, but, in this case, not doing the install makes this deterministic - the file will be located in the lib directory.
Why the change to do a make install?
There was a problem hiding this comment.
can you make 'get lib path' into a generic function?
There was a problem hiding this comment.
I think the make install fix is fine, but, in this case, not doing the install makes this deterministic - the file will be located in the lib directory. Why the change to do a
make install?
Not doing install will fail benchmark in an assert in the next line with message like could not find adapter file /home/lslusarczyk/src2/ur_install/lib64/libur_adapter_level_zero.so (and in similar lib paths) which looks what I would like to se, that is sample full path which was searched and still being a short message
"make install" used as cleanest way of using binaries and includes of already built UR
can you make 'get lib path' into a generic function?
yes, separate function added
lslusarczyk
force-pushed
the
bench_no_recompile_ur
branch
from
7488aa7 to
053d092
Compare
|
Compute Benchmarks level_zero run (with params: ): |
|
Compute Benchmarks level_zero run (): SummaryNo diffs to calculate performance change (result is better) Performance change in benchmark groupsRelative perf in group api (6): cannot calculate
Relative perf in group memory (4): cannot calculate
Relative perf in group miscellaneous (1): cannot calculate
Relative perf in group Velocity-Bench (5): cannot calculate
Relative perf in group Runtime (24): cannot calculate
Relative perf in group MicroBench (15): cannot calculate
Relative perf in group Pattern (14): cannot calculate
Relative perf in group ScalarProduct (6): cannot calculate
Relative perf in group SYCL2020 (2): cannot calculate
Relative perf in group USM (17): cannot calculate
Relative perf in group VectorAddition (3): cannot calculate
Relative perf in group Polybench (12): cannot calculate
Relative perf in group ReductionAtomic (4): cannot calculate
Relative perf in group Kmeans (1): cannot calculate
Relative perf in group LinearRegressionCoeff (1): cannot calculate
Relative perf in group LinearRegression (1): cannot calculate
Relative perf in group MatmulChain (1): cannot calculate
Relative perf in group MolecularDynamics (1): cannot calculate
DetailsBenchmark details - environment, command, output...api_overhead_benchmark_sycl SubmitKernel out of orderEnvironment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/api_overhead_benchmark_sycl --test=SubmitKernel --csv --noHeaders --Ioq=0 --DiscardEvents=0 --MeasureCompletion=0 --iterations=100000 --Profiling=0 --NumKernels=10 --KernelExecTime=1 Output:TestCase,Mean,Median,StdDev,Min,Max,Type api_overhead_benchmark_sycl SubmitKernel in orderEnvironment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/api_overhead_benchmark_sycl --test=SubmitKernel --csv --noHeaders --Ioq=1 --DiscardEvents=0 --MeasureCompletion=0 --iterations=100000 --Profiling=0 --NumKernels=10 --KernelExecTime=1 Output:TestCase,Mean,Median,StdDev,Min,Max,Type api_overhead_benchmark_ur SubmitKernel out of orderEnvironment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/api_overhead_benchmark_ur --test=SubmitKernel --csv --noHeaders --Ioq=0 --DiscardEvents=0 --MeasureCompletion=0 --iterations=100000 --Profiling=0 --NumKernels=10 --KernelExecTime=1 Output:TestCase,Mean,Median,StdDev,Min,Max,Type api_overhead_benchmark_ur SubmitKernel in orderEnvironment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/api_overhead_benchmark_ur --test=SubmitKernel --csv --noHeaders --Ioq=1 --DiscardEvents=0 --MeasureCompletion=0 --iterations=100000 --Profiling=0 --NumKernels=10 --KernelExecTime=1 Output:TestCase,Mean,Median,StdDev,Min,Max,Type memory_benchmark_sycl QueueInOrderMemcpy from Device to Device, size 1024Environment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/memory_benchmark_sycl --test=QueueInOrderMemcpy --csv --noHeaders --iterations=10000 --IsCopyOnly=0 --sourcePlacement=Device --destinationPlacement=Device --size=1024 --count=100 Output:TestCase,Mean,Median,StdDev,Min,Max,Type memory_benchmark_sycl QueueInOrderMemcpy from Host to Device, size 1024Environment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/memory_benchmark_sycl --test=QueueInOrderMemcpy --csv --noHeaders --iterations=10000 --IsCopyOnly=0 --sourcePlacement=Host --destinationPlacement=Device --size=1024 --count=100 Output:TestCase,Mean,Median,StdDev,Min,Max,Type memory_benchmark_sycl QueueMemcpy from Device to Device, size 1024Environment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/memory_benchmark_sycl --test=QueueMemcpy --csv --noHeaders --iterations=10000 --sourcePlacement=Device --destinationPlacement=Device --size=1024 Output:TestCase,Mean,Median,StdDev,Min,Max,Type memory_benchmark_sycl StreamMemory, placement Device, type Triad, size 10240Environment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/memory_benchmark_sycl --test=StreamMemory --csv --noHeaders --iterations=10000 --type=Triad --size=10240 --memoryPlacement=Device --useEvents=0 --contents=Zeros Output:TestCase,Mean,Median,StdDev,Min,Max,Type api_overhead_benchmark_sycl ExecImmediateCopyQueue out of order from Device to Device, size 1024Environment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/api_overhead_benchmark_sycl --test=ExecImmediateCopyQueue --csv --noHeaders --iterations=100000 --ioq=0 --IsCopyOnly=1 --MeasureCompletionTime=0 --src=Device --dst=Device --size=1024 Output:TestCase,Mean,Median,StdDev,Min,Max,Type api_overhead_benchmark_sycl ExecImmediateCopyQueue in order from Device to Host, size 1024Environment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/api_overhead_benchmark_sycl --test=ExecImmediateCopyQueue --csv --noHeaders --iterations=100000 --ioq=1 --IsCopyOnly=1 --MeasureCompletionTime=0 --src=Host --dst=Host --size=1024 Output:TestCase,Mean,Median,StdDev,Min,Max,Type miscellaneous_benchmark_sycl VectorSumEnvironment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/miscellaneous_benchmark_sycl --test=VectorSum --csv --noHeaders --iterations=1000 --numberOfElementsX=512 --numberOfElementsY=256 --numberOfElementsZ=256 Output:TestCase,Mean,Median,StdDev,Min,Max,Type Velocity-Bench HashtableEnvironment Variables:Command:/home/test-user/bench_workdir/hashtable/hashtable_sycl --no-verify Output:hashtable - total time for whole calculation: 0.371992 s Velocity-Bench BitcrackerEnvironment Variables:Command:/home/test-user/bench_workdir/bitcracker/bitcracker -f /home/test-user/bench_workdir/velocity-bench-repo/bitcracker/hash_pass/img_win8_user_hash.txt -d /home/test-user/bench_workdir/velocity-bench-repo/bitcracker/hash_pass/user_passwords_60000.txt -b 60000 Output:---------> BitCracker: BitLocker password cracking tool <--------- ==================================
|
|
Compute Benchmarks level_zero_v2 run (with params: ): |
|
Compute Benchmarks level_zero_v2 run (): SummaryNo diffs to calculate performance change (result is better) Performance change in benchmark groupsRelative perf in group api (6): cannot calculate
Relative perf in group memory (4): cannot calculate
Relative perf in group Velocity-Bench (5): cannot calculate
Relative perf in group Runtime (20): cannot calculate
Relative perf in group MicroBench (16): cannot calculate
Relative perf in group Pattern (14): cannot calculate
Relative perf in group ScalarProduct (6): cannot calculate
Relative perf in group USM (17): cannot calculate
Relative perf in group SYCL2020 (2): cannot calculate
Relative perf in group VectorAddition (3): cannot calculate
Relative perf in group Polybench (12): cannot calculate
Relative perf in group ReductionAtomic (4): cannot calculate
Relative perf in group Kmeans (1): cannot calculate
Relative perf in group LinearRegressionCoeff (1): cannot calculate
Relative perf in group LinearRegression (1): cannot calculate
Relative perf in group MatmulChain (1): cannot calculate
Relative perf in group MolecularDynamics (1): cannot calculate
DetailsBenchmark details - environment, command, output...api_overhead_benchmark_sycl SubmitKernel out of orderEnvironment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/api_overhead_benchmark_sycl --test=SubmitKernel --csv --noHeaders --Ioq=0 --DiscardEvents=0 --MeasureCompletion=0 --iterations=100000 --Profiling=0 --NumKernels=10 --KernelExecTime=1 Output:TestCase,Mean,Median,StdDev,Min,Max,Type api_overhead_benchmark_sycl SubmitKernel in orderEnvironment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/api_overhead_benchmark_sycl --test=SubmitKernel --csv --noHeaders --Ioq=1 --DiscardEvents=0 --MeasureCompletion=0 --iterations=100000 --Profiling=0 --NumKernels=10 --KernelExecTime=1 Output:TestCase,Mean,Median,StdDev,Min,Max,Type api_overhead_benchmark_ur SubmitKernel out of orderEnvironment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/api_overhead_benchmark_ur --test=SubmitKernel --csv --noHeaders --Ioq=0 --DiscardEvents=0 --MeasureCompletion=0 --iterations=100000 --Profiling=0 --NumKernels=10 --KernelExecTime=1 Output:TestCase,Mean,Median,StdDev,Min,Max,Type api_overhead_benchmark_ur SubmitKernel in orderEnvironment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/api_overhead_benchmark_ur --test=SubmitKernel --csv --noHeaders --Ioq=1 --DiscardEvents=0 --MeasureCompletion=0 --iterations=100000 --Profiling=0 --NumKernels=10 --KernelExecTime=1 Output:TestCase,Mean,Median,StdDev,Min,Max,Type memory_benchmark_sycl QueueInOrderMemcpy from Device to Device, size 1024Environment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/memory_benchmark_sycl --test=QueueInOrderMemcpy --csv --noHeaders --iterations=10000 --IsCopyOnly=0 --sourcePlacement=Device --destinationPlacement=Device --size=1024 --count=100 Output:TestCase,Mean,Median,StdDev,Min,Max,Type memory_benchmark_sycl QueueInOrderMemcpy from Host to Device, size 1024Environment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/memory_benchmark_sycl --test=QueueInOrderMemcpy --csv --noHeaders --iterations=10000 --IsCopyOnly=0 --sourcePlacement=Host --destinationPlacement=Device --size=1024 --count=100 Output:TestCase,Mean,Median,StdDev,Min,Max,Type memory_benchmark_sycl QueueMemcpy from Device to Device, size 1024Environment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/memory_benchmark_sycl --test=QueueMemcpy --csv --noHeaders --iterations=10000 --sourcePlacement=Device --destinationPlacement=Device --size=1024 Output:TestCase,Mean,Median,StdDev,Min,Max,Type memory_benchmark_sycl StreamMemory, placement Device, type Triad, size 10240Environment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/memory_benchmark_sycl --test=StreamMemory --csv --noHeaders --iterations=10000 --type=Triad --size=10240 --memoryPlacement=Device --useEvents=0 --contents=Zeros Output:TestCase,Mean,Median,StdDev,Min,Max,Type api_overhead_benchmark_sycl ExecImmediateCopyQueue out of order from Device to Device, size 1024Environment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/api_overhead_benchmark_sycl --test=ExecImmediateCopyQueue --csv --noHeaders --iterations=100000 --ioq=0 --IsCopyOnly=1 --MeasureCompletionTime=0 --src=Device --dst=Device --size=1024 Output:TestCase,Mean,Median,StdDev,Min,Max,Type api_overhead_benchmark_sycl ExecImmediateCopyQueue in order from Device to Host, size 1024Environment Variables:Command:/home/test-user/bench_workdir/compute-benchmarks-build/bin/api_overhead_benchmark_sycl --test=ExecImmediateCopyQueue --csv --noHeaders --iterations=100000 --ioq=1 --IsCopyOnly=1 --MeasureCompletionTime=0 --src=Host --dst=Host --size=1024 Output:TestCase,Mean,Median,StdDev,Min,Max,Type Velocity-Bench HashtableEnvironment Variables:Command:/home/test-user/bench_workdir/hashtable/hashtable_sycl --no-verify Output:hashtable - total time for whole calculation: 0.369433 s Velocity-Bench BitcrackerEnvironment Variables:Command:/home/test-user/bench_workdir/bitcracker/bitcracker -f /home/test-user/bench_workdir/velocity-bench-repo/bitcracker/hash_pass/img_win8_user_hash.txt -d /home/test-user/bench_workdir/velocity-bench-repo/bitcracker/hash_pass/user_passwords_60000.txt -b 60000 Output:---------> BitCracker: BitLocker password cracking tool <--------- ==================================
|
lslusarczyk
force-pushed
the
bench_no_recompile_ur
branch
from
053d092 to
1360515
Compare
|
I've compared level_zero_v2 benchmarks on main and on my PR.
Results are also comparable. E.g. "api_overhead_benchmark_sycl/ur SubmitKernel out of order/in order.
So it seems I've not broken anything I think. |
separate PR: #2169 |
lslusarczyk
force-pushed
the
bench_no_recompile_ur
branch
from
1360515 to
bf46d7f
Compare
|
the cmake change got merged, please rebase. |
already rebased |
3 participants
Utilize support for using UR binary through find_package added in intel/compute-benchmarks#21
Before this change compute-benchmarks always builds UR from sources which is unnecessary since we already build UR as part of SYCL on UR CI.
Add installing adapters by "cmake --install" which is needed by new benchmarks code.