devtools/eigen
1
0
Fork 0
mirror of https://gitlab.com/libeigen/eigen.git synced 2026-04-10 11:34:33 +08:00
Code Issues Packages Projects Releases Wiki Activity

[SYCL] Rebasing the SYCL support branch on top of the Einge upstream master branch.

Browse Source 
* Unifying all loadLocalTile from lhs and rhs to an extract_block function.
* Adding get_tensor operation which was missing in TensorContractionMapper.
* Adding the -D method missing from cmake for Disable_Skinny Contraction operation.
* Wrapping all the indices in TensorScanSycl into Scan parameter struct.
* Fixing typo in Device SYCL
* Unifying load to private register for tall/skinny no shared
* Unifying load to vector tile for tensor-vector/vector-tensor operation
* Removing all the LHS/RHS class for extracting data from global
* Removing Outputfunction from TensorContractionSkinnyNoshared.
* Combining the local memory version of tall/skinny and normal tensor contraction into one kernel.
* Combining the no-local memory version of tall/skinny and normal tensor contraction into one kernel.
* Combining General Tensor-Vector and VectorTensor contraction into one kernel.
* Making double buffering optional for Tensor contraction when local memory is version is used.
* Modifying benchmark to accept custom Reduction Sizes
* Disabling AVX optimization for SYCL backend on the host to allow SSE optimization to the host
* Adding Test for SYCL
* Modifying SYCL CMake
This commit is contained in:
Mehdi Goli
2019-11-28 10:08:54 +00:00
parent ea51a9eace
commit 00f32752f7
56 changed files with 7336 additions and 4826 deletions
Download Patch File Download Diff File Expand all files Collapse all files

30
Eigen/src/Core/util/ConfigureVectorization.h
View File

@@ -240,15 +240,19 @@
#define EIGEN_VECTORIZE_SSE4_2
#endif
#ifdef __AVX__
#define EIGEN_VECTORIZE_AVX
#ifndef EIGEN_USE_SYCL
#define EIGEN_VECTORIZE_AVX
#endif
#define EIGEN_VECTORIZE_SSE3
#define EIGEN_VECTORIZE_SSSE3
#define EIGEN_VECTORIZE_SSE4_1
#define EIGEN_VECTORIZE_SSE4_2
#endif
#ifdef __AVX2__
#define EIGEN_VECTORIZE_AVX2
#define EIGEN_VECTORIZE_AVX
#ifndef EIGEN_USE_SYCL
#define EIGEN_VECTORIZE_AVX2
#define EIGEN_VECTORIZE_AVX
#endif
#define EIGEN_VECTORIZE_SSE3
#define EIGEN_VECTORIZE_SSSE3
#define EIGEN_VECTORIZE_SSE4_1
@@ -267,19 +271,23 @@
#error Please enable FMA in your compiler flags (e.g. -mfma): compiling with AVX512 alone without SSE/AVX FMA is not supported (bug 1638).
#endif
#endif
#define EIGEN_VECTORIZE_AVX512
#define EIGEN_VECTORIZE_AVX2
#define EIGEN_VECTORIZE_AVX
#ifndef EIGEN_USE_SYCL
#define EIGEN_VECTORIZE_AVX512
#define EIGEN_VECTORIZE_AVX2
#define EIGEN_VECTORIZE_AVX
#endif
#define EIGEN_VECTORIZE_FMA
#define EIGEN_VECTORIZE_SSE3
#define EIGEN_VECTORIZE_SSSE3
#define EIGEN_VECTORIZE_SSE4_1
#define EIGEN_VECTORIZE_SSE4_2
#ifdef __AVX512DQ__
#define EIGEN_VECTORIZE_AVX512DQ
#endif
#ifdef __AVX512ER__
#define EIGEN_VECTORIZE_AVX512ER
#ifndef EIGEN_USE_SYCL
#ifdef __AVX512DQ__
#define EIGEN_VECTORIZE_AVX512DQ
#endif
#ifdef __AVX512ER__
#define EIGEN_VECTORIZE_AVX512ER
#endif
#endif
#endif