Update time op to more accurately get device time

[pfultz2]

Attaching events to the launch kernel has a lot of overhead. Instead run the op in the loop, and then add a hip event at the end of the loop.

[codecov]

Codecov Report

Merging #2104 (1d24562) into develop (c58e7d8) will decrease coverage by 0.01%.
Report is 4 commits behind head on develop.
The diff coverage is 100.00%.

❗ Current head 1d24562 differs from pull request most recent head 1704518. Consider uploading reports for the commit 1704518 to get more accurate results

@@             Coverage Diff             @@
##           develop    #2104      +/-   ##
===========================================
- Coverage    91.45%   91.45%   -0.01%     
===========================================
  Files          433      433              
  Lines        16175    16174       -1     
===========================================
- Hits         14793    14792       -1     
  Misses        1382     1382              

Files	Coverage Δ
src/api/api.cpp	72.36% <ø> (ø)
src/api/include/migraphx/migraphx.hpp	98.63% <ø> (ø)
src/include/migraphx/auto_register.hpp	100.00% <ø> (ø)
src/include/migraphx/float_equal.hpp	100.00% <ø> (ø)
src/include/migraphx/generate.hpp	76.92% <ø> (ø)
src/include/migraphx/instruction_ref.hpp	100.00% <ø> (ø)
src/include/migraphx/matcher.hpp	97.13% <ø> (-0.02%)	⬇️
src/include/migraphx/op/nonmaxsuppression.hpp	98.47% <ø> (ø)
src/include/migraphx/op/roialign.hpp	99.16% <ø> (ø)
src/include/migraphx/optional.hpp	100.00% <ø> (ø)
... and 3 more

[migraphx-bot]

Test	Batch	Rate new 1d4c58	Rate old e07b7c	Diff	Compare
torchvision-resnet50	64	2,325.51	2,320.06	0.23%	✅
torchvision-resnet50_fp16	64	5,362.56	5,353.68	0.17%	✅
torchvision-densenet121	32	1,848.72	1,842.18	0.35%	✅
torchvision-densenet121_fp16	32	3,403.46	3,412.09	-0.25%	✅
torchvision-inceptionv3	32	1,297.32	1,293.34	0.31%	✅
torchvision-inceptionv3_fp16	32	2,525.45	2,539.42	-0.55%	✅
cadene-inceptionv4	16	620.44	619.92	0.08%	✅
cadene-resnext64x4	16	588.83	589.07	-0.04%	✅
slim-mobilenet	64	7,204.93	7,219.97	-0.21%	✅
slim-nasnetalarge	64	236.53	236.23	0.13%	✅
slim-resnet50v2	64	2,555.84	2,556.43	-0.02%	✅
bert-mrpc-onnx	8	824.99	825.47	-0.06%	✅
bert-mrpc-tf	1	389.17	388.10	0.28%	✅
pytorch-examples-wlang-gru	1	299.44	296.27	1.07%	✅
pytorch-examples-wlang-lstm	1	315.47	316.06	-0.19%	✅
torchvision-resnet50_1	1	548.49	547.74	0.14%	✅
torchvision-inceptionv3_1	1	301.50	305.30	-1.25%	✅
cadene-dpn92_1	1	356.75	351.99	1.35%	✅
cadene-resnext101_1	1	219.66	218.06	0.73%	✅
slim-vgg16_1	1	224.45	223.86	0.26%	✅
slim-mobilenet_1	1	1,520.15	1,498.13	1.47%	✅
slim-inceptionv4_1	1	214.91	213.83	0.51%	✅
onnx-taau-downsample	1	305.05	306.25	-0.39%	✅
dlrm-criteoterabyte	1	21.69	21.68	0.03%	✅
dlrm-criteoterabyte_fp16	1	40.75	40.62	0.32%	✅
agentmodel	1	5,871.54	5,817.78	0.92%	✅
unet_fp16	2	55.17	55.18	-0.02%	✅
resnet50v1_fp16	1	758.35	759.76	-0.19%	✅
bert_base_cased_fp16	64	971.02	970.69	0.03%	✅
bert_large_uncased_fp16	32	305.04	305.01	0.01%	✅
bert_large_fp16	1	166.77	167.09	-0.20%	✅
distilgpt2_fp16	16	1,351.59	1,277.85	5.77%	🔆

Check results before merge 🔆

[migraphx-bot]

    :white_check_mark:bert-mrpc-onnx: PASSED: MIGraphX meets tolerance

    :white_check_mark:bert-mrpc-tf: PASSED: MIGraphX meets tolerance

    :white_check_mark:pytorch-examples-wlang-gru: PASSED: MIGraphX meets tolerance

    :white_check_mark:pytorch-examples-wlang-lstm: PASSED: MIGraphX meets tolerance

    :white_check_mark:torchvision-resnet50_1: PASSED: MIGraphX meets tolerance

    :white_check_mark:torchvision-inceptionv3_1: PASSED: MIGraphX meets tolerance

    :white_check_mark:cadene-dpn92_1: PASSED: MIGraphX meets tolerance

    :white_check_mark:cadene-resnext101_1: PASSED: MIGraphX meets tolerance

    :white_check_mark:slim-vgg16_1: PASSED: MIGraphX meets tolerance

    :white_check_mark:slim-mobilenet_1: PASSED: MIGraphX meets tolerance

    :white_check_mark:slim-inceptionv4_1: PASSED: MIGraphX meets tolerance

    :white_check_mark:dlrm-criteoterabyte: PASSED: MIGraphX meets tolerance

    :white_check_mark:agentmodel: PASSED: MIGraphX meets tolerance

    :white_check_mark:unet: PASSED: MIGraphX meets tolerance

    :white_check_mark:resnet50v1: PASSED: MIGraphX meets tolerance

🔴bert_base_cased_fp16: FAILED: MIGraphX is not within tolerance - check verbose output

🔴bert_large_uncased_fp16: FAILED: MIGraphX is not within tolerance - check verbose output

    :white_check_mark:bert_large: PASSED: MIGraphX meets tolerance

🔴distilgpt2_fp16: FAILED: MIGraphX is not within tolerance - check verbose output

[causten]

this fails when I tested it with mlir

root@ce81174f8f3c:/workspace/AMDMIGraphX# MIGRAPHX_ENABLE_MLIR=1 migraphx-driver perf /models/mlperf/resnet50_v1.onnx
Compiling ...
Reading: /models/mlperf/resnet50_v1.onnx
Benchmarking gpu::mlir_op: 21 configs
terminate called after throwing an instance of 'migraphx::version_2_7_0::exception'
what(): /workspace/AMDMIGraphX/src/targets/gpu/include/migraphx/gpu/context.hpp:333: get_elapsed_ms: Failed hipEventElapsedTime: invalid resource handle
Aborted (core dumped)

[pfultz2]

this fails when I tested it with mlir

Fixed.