experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Automated rollback of commit bb73a513b8

Browse Source 
PiperOrigin-RevId: 293243640
Change-Id: I6385c19eef7387ce1af6eb285522006cc18d937e
This commit is contained in:
Robert David 2020-02-04 15:32:01 -08:00 committed by TensorFlower Gardener
parent cd5281bd47
commit 1e1d9c93ab
1 changed files with 4 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
tensorflow/lite/kernels/internal/optimized/sse_check.h
View File

@ -15,20 +15,14 @@ limitations under the License.
#ifndef TENSORFLOW_LITE_KERNELS_INTERNAL_OPTIMIZED_SSE_CHECK_H_
#define TENSORFLOW_LITE_KERNELS_INTERNAL_OPTIMIZED_SSE_CHECK_H_
#if defined(__SSE4_1__)
// SSE 4.1 available: Use the SSE code.
#if defined(__SSSE3__)
// SSSE 3 available: Use the SSE code.
#define SSE_OR_PORTABLE(funcname, ...) Sse##funcname(__VA_ARGS__)
#else
#include "tensorflow/lite/kernels/internal/optimized/neon_check.h"
// No SSE 4.1 available: Fall back to NEON_OR_PORTABLE, potentially used with
// NEON_2_SSE translator library. As the library requires SSSE3, the fallback is
// generally using Portable code, only a narrow subset of processors supporting
// SSSE3 but no SSE4.1 is affected - but that includes the android_x86 ABI (not
// android_x86_64).
#define SSE_OR_PORTABLE(...) NEON_OR_PORTABLE(__VA_ARGS__)
// No SSSE 3 available: Use Portable code
#define SSE_OR_PORTABLE(...) Portable##funcname(__VA_ARGS__)
#endif
#endif // TENSORFLOW_LITE_KERNELS_INTERNAL_OPTIMIZED_SSE_CHECK_H_