pytorch

7335f079 - [pt][quant] qmul and qadd should preserve input memory format (#34834)

[pt][quant] qmul and qadd should preserve input memory format (#34834) Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/34834 They should keep the activations in channelLast format, i.e., the same as input tensors to these operations. ### Before ``` ------------------------- --------------- --------------- --------------- --------------- --------------- --------------- Name Self CPU total % Self CPU total CPU total % CPU total CPU time avg Number of Calls ------------------------- --------------- --------------- --------------- --------------- --------------- --------------- quantize_per_tensor 0.06% 129.181us 0.06% 129.181us 129.181us 1 quantized::conv2d 21.74% 47.744ms 21.74% 47.744ms 408.067us 117 quantized::add_scalar 16.36% 35.930ms 16.36% 35.930ms 520.726us 69 quantized::relu6 0.69% 1.515ms 0.69% 1.515ms 21.959us 69 quantized::mul_scalar 6.08% 13.364ms 6.08% 13.364ms 193.676us 69 quantized::mul 53.17% 116.781ms 53.17% 116.781ms 1.269ms 92 adaptive_avg_pool2d 0.02% 42.700us 1.61% 3.527ms 146.948us 24 _adaptive_avg_pool2d 1.59% 3.484ms 1.59% 3.484ms 145.169us 24 sigmoid 0.08% 173.702us 0.08% 173.702us 7.552us 23 quantized::add 0.20% 445.648us 0.20% 445.648us 27.853us 16 dropout 0.00% 2.598us 0.00% 2.598us 2.598us 1 view 0.00% 10.311us 0.00% 10.311us 10.311us 1 dequantize 0.00% 4.645us 0.00% 4.645us 4.645us 1 ------------------------- --------------- --------------- --------------- --------------- --------------- --------------- Self CPU time total: 219.627ms ``` ### After ``` ------------------------- --------------- --------------- --------------- --------------- --------------- --------------- Name Self CPU total % Self CPU total CPU total % CPU total CPU time avg Number of Calls ------------------------- --------------- --------------- --------------- --------------- --------------- --------------- quantize_per_tensor 0.12% 155.807us 0.12% 155.807us 155.807us 1 quantized::conv2d 25.50% 31.981ms 25.50% 31.981ms 273.343us 117 quantized::add_scalar 44.53% 55.840ms 44.53% 55.840ms 809.281us 69 quantized::relu6 1.25% 1.570ms 1.25% 1.570ms 22.749us 69 quantized::mul_scalar 10.73% 13.449ms 10.73% 13.449ms 194.914us 69 quantized::mul 16.67% 20.904ms 16.67% 20.904ms 227.220us 92 adaptive_avg_pool2d 0.03% 41.713us 0.69% 862.922us 35.955us 24 _adaptive_avg_pool2d 0.65% 821.209us 0.65% 821.209us 34.217us 24 sigmoid 0.15% 182.344us 0.15% 182.344us 7.928us 23 quantized::add 0.34% 431.939us 0.34% 431.939us 26.996us 16 dropout 0.00% 1.936us 0.00% 1.936us 1.936us 1 view 0.01% 10.281us 0.01% 10.281us 10.281us 1 dequantize 0.00% 4.562us 0.00% 4.562us 4.562us 1 ------------------------- --------------- --------------- --------------- --------------- --------------- --------------- Self CPU time total: 125.394ms ``` ghstack-source-id: 100305788 Test Plan: buck test //caffe2/test:quantized Differential Revision: D20473713 fbshipit-source-id: c878fbb8f5a1a33f0cdac2657cc61e97ceb6c183