Switch to using hipBLASLt as the default instead of rocBLAS

[ahsan-ca]

No description provided.

[causten]

Can you update the env flag documentation to indicate that it is to now disable hipblaslt

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 92.29%. Comparing base (879f306) to head (b15fde8).

Additional details and impacted files

@@           Coverage Diff            @@
##           develop    #3783   +/-   ##
========================================
  Coverage    92.29%   92.29%           
========================================
  Files          519      519           
  Lines        22233    22233           
========================================
  Hits         20520    20520           
  Misses        1713     1713           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[migraphx-bot]

Test	Batch	Rate new b15fde	Rate old 879f30	Diff	Compare
torchvision-resnet50	64	3,232.32	3,237.49	-0.16%	✅
torchvision-resnet50_fp16	64	6,866.95	6,885.53	-0.27%	✅
torchvision-densenet121	32	2,437.39	2,438.04	-0.03%	✅
torchvision-densenet121_fp16	32	4,210.23	4,201.39	0.21%	✅
torchvision-inceptionv3	32	1,612.90	1,614.18	-0.08%	✅
torchvision-inceptionv3_fp16	32	2,684.56	2,690.16	-0.21%	✅
cadene-inceptionv4	16	749.77	750.10	-0.04%	✅
cadene-resnext64x4	16	808.49	809.38	-0.11%	✅
slim-mobilenet	64	6,664.40	6,664.81	-0.01%	✅
slim-nasnetalarge	64	198.99	199.11	-0.06%	✅
slim-resnet50v2	64	3,424.46	3,426.01	-0.05%	✅
bert-mrpc-onnx	8	1,134.75	1,145.27	-0.92%	✅
bert-mrpc-tf	1	500.00	479.30	4.32%	🔆
pytorch-examples-wlang-gru	1	482.12	470.08	2.56%	✅
pytorch-examples-wlang-lstm	1	448.46	438.03	2.38%	✅
torchvision-resnet50_1	1	793.65	803.96	-1.28%	✅
cadene-dpn92_1	1	427.83	433.48	-1.30%	✅
cadene-resnext101_1	1	387.60	389.71	-0.54%	✅
onnx-taau-downsample	1	372.86	373.58	-0.19%	✅
dlrm-criteoterabyte	1	31.79	31.80	-0.04%	✅
dlrm-criteoterabyte_fp16	1	51.04	51.06	-0.03%	✅
agentmodel	1	8,746.53	8,764.93	-0.21%	✅
unet_fp16	2	58.33	58.01	0.55%	✅
resnet50v1_fp16	1	1,041.67	1,021.79	1.95%	✅
resnet50v1_int8	1	787.40	780.83	0.84%	✅
bert_base_cased_fp16	64	1,167.46	1,172.44	-0.43%	✅
bert_large_uncased_fp16	32	360.77	362.45	-0.46%	✅
bert_large_fp16	1	215.98	201.10	7.40%	🔆
distilgpt2_fp16	16	2,191.78	2,219.84	-1.26%	✅
yolov5s	1	535.18	521.90	2.55%	✅
tinyllama	1	49.14	43.62	12.67%	🔆
vicuna-fastchat	1	43.67	43.83	-0.35%	✅
whisper-tiny-encoder	1	436.68	412.03	5.98%	🔆
whisper-tiny-decoder	1	448.43	410.73	9.18%	🔆

Check results before merge 🔆

[migraphx-bot]

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

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

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

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

     ✅ torchvision-resnet50_1: PASSED: MIGraphX meets tolerance

     ✅ cadene-dpn92_1: PASSED: MIGraphX meets tolerance

     ✅ cadene-resnext101_1: PASSED: MIGraphX meets tolerance

     ✅ dlrm-criteoterabyte: PASSED: MIGraphX meets tolerance

     ✅ agentmodel: PASSED: MIGraphX meets tolerance

     ✅ unet: PASSED: MIGraphX meets tolerance

     ✅ resnet50v1: PASSED: MIGraphX meets tolerance

     ✅ bert_base_cased_fp16: PASSED: MIGraphX meets tolerance

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

     ✅ bert_large: PASSED: MIGraphX meets tolerance

     ✅ yolov5s: PASSED: MIGraphX meets tolerance

     ✅ tinyllama: PASSED: MIGraphX meets tolerance

     ✅ vicuna-fastchat: PASSED: MIGraphX meets tolerance

     ✅ whisper-tiny-encoder: PASSED: MIGraphX meets tolerance

     ✅ whisper-tiny-decoder: PASSED: MIGraphX meets tolerance

     ✅ distilgpt2_fp16: PASSED: MIGraphX meets tolerance

[CharlieL7]

The change looks fine but why not rename the env variable?