pytorch

OSSForks/pytorch

Fork 0

mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-06 12:20:52 +01:00

Commit Graph

Author	SHA1	Message	Date
Tongliang Liao	55511004d1	Resolve errors in perfkernel for Windows (#16031 ) Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/16031 1. MSVC only has _mm_prefetch(const char*, int). Fixed in both python codegen and C++ files. 2. uint32_t in "cvtsh_ss_bugfix.h" requires "#include <cstdint>". 3. Some files use gflags headers. Add dependency via c10. 4. Isolate arch flags with interface library and private compile options. Pull Request resolved: https://github.com/pytorch/pytorch/pull/15753 Reviewed By: dskhudia Differential Revision: D13636233 Pulled By: jspark1105 fbshipit-source-id: cdcbd4240e07b749554a2a5676c11af88f23c31d	2019-01-16 21:51:00 -08:00
Jongsoo Park	1e0eab5df8	minimize header file includes from _avx2.cc (#14950 ) Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/14950 Minimize the number of headers included from _avx2.cc files to avoid accidental compilation of functions defined the header files reused by other translation units that can lead to illegal instruction errors. Reviewed By: dskhudia Differential Revision: D13394483 fbshipit-source-id: 67149a6fb51f7f047e745bfe395cb6dd4ae7c1ae	2018-12-13 00:18:11 -08:00
Wei Wen	0e30fa6f3c	Faster random number generation in fused_rowwise_random_quantization_ops (#10634 ) Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/10634 ``` Trying example: test_speed_of_rand_quantization(self=<caffe2.caffe2.python.operator_test.rand_quantization_op_speed_test.TestSpeedFloatToFusedRandRowwiseQuantized testMethod=test_speed_of_rand_quantization>, bitwidth_=2, random_=True, data_shape_=array([1024, 1224]), gc=, dc=[, device_type: 1]) Sub+Scale+Sum time: 1.9944190979003908 ms Quantizing time: 2.080512046813965 ms (1.0431669296609765X) De-quantizing time: 0.7375001907348633 ms (0.36978195380863577X) ``` ``` Trying example: test_speed_of_rand_quantization(self=<caffe2.caffe2.python.operator_test.rand_quantization_op_speed_test.TestSpeedFloatToFusedRandRowwiseQuantized testMethod=test_speed_of_rand_quantization>, bitwidth_=1, random_=True, data_shape_=array([1024, 1224]), gc=device_type: 1, dc=[, device_type: 1]) Sub+Scale+Sum time: 1.6691923141479492 ms Quantizing time: 7.500243186950684 ms (4.493336761366071X) De-quantizing time: 1.1209726333618164 ms (0.6715658967876477X) ``` Reviewed By: jspark1105 Differential Revision: D8849770 fbshipit-source-id: 2bb2bac7e633f647f38e419ce980b8958f3bcae2	2018-08-22 13:15:46 -07:00

Author

SHA1

Message

Date

Tongliang Liao

55511004d1

Resolve errors in perfkernel for Windows (#16031 )

Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/16031

1. MSVC only has _mm_prefetch(const char*, int). Fixed in both python codegen and C++ files.
2. uint32_t in "cvtsh_ss_bugfix.h" requires "#include <cstdint>".
3. Some files use gflags headers. Add dependency via c10.
4. Isolate arch flags with interface library and private compile options.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/15753

Reviewed By: dskhudia

Differential Revision: D13636233

Pulled By: jspark1105

fbshipit-source-id: cdcbd4240e07b749554a2a5676c11af88f23c31d

2019-01-16 21:51:00 -08:00

Jongsoo Park

1e0eab5df8

minimize header file includes from _avx2.cc (#14950 )

Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/14950

Minimize the number of headers included from _avx2.cc files to avoid accidental compilation of functions defined the header files reused by other translation units that can lead to illegal instruction errors.

Reviewed By: dskhudia

Differential Revision: D13394483

fbshipit-source-id: 67149a6fb51f7f047e745bfe395cb6dd4ae7c1ae

2018-12-13 00:18:11 -08:00

Wei Wen

0e30fa6f3c

Faster random number generation in fused_rowwise_random_quantization_ops (#10634 )

Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/10634

```
Trying example: test_speed_of_rand_quantization(self=<caffe2.caffe2.python.operator_test.rand_quantization_op_speed_test.TestSpeedFloatToFusedRandRowwiseQuantized testMethod=test_speed_of_rand_quantization>, bitwidth_=2, random_=True, data_shape_=array([1024, 1224]), gc=, dc=[, device_type: 1])
Sub+Scale+Sum time: 1.9944190979003908 ms
Quantizing time: 2.080512046813965 ms (1.0431669296609765X)
De-quantizing time: 0.7375001907348633 ms (0.36978195380863577X)
```

```
Trying example: test_speed_of_rand_quantization(self=<caffe2.caffe2.python.operator_test.rand_quantization_op_speed_test.TestSpeedFloatToFusedRandRowwiseQuantized testMethod=test_speed_of_rand_quantization>, bitwidth_=1, random_=True, data_shape_=array([1024, 1224]), gc=device_type: 1, dc=[, device_type: 1])
Sub+Scale+Sum time: 1.6691923141479492 ms
Quantizing time: 7.500243186950684 ms (4.493336761366071X)
De-quantizing time: 1.1209726333618164 ms (0.6715658967876477X)
```

Reviewed By: jspark1105

Differential Revision: D8849770

fbshipit-source-id: 2bb2bac7e633f647f38e419ce980b8958f3bcae2

2018-08-22 13:15:46 -07:00

3 Commits