mirror/eigen
2
0
Fork 0
mirror of https://gitlab.com/libeigen/eigen.git synced 2024-12-27 07:29:52 +08:00
Code Issues Packages Projects Releases Wiki Activity
11,617 Commits 11 Branches 89 Tags 168 MiB
675b72e44b
Commit Graph

2387 Commits

Author SHA1 Message Date
Rasmus Munk Larsen
ab773c7e91 Extend support for Packet16b: 
* Add ptranspose<*,4> to support matmul and add unit test for Matrix<bool> * Matrix<bool>
* work around a bug in slicing of Tensor<bool>.
* Add tensor tests

This speeds up matmul for boolean matrices by about 10x

name                            old time/op             new time/op             delta
BM_MatMul<bool>/8                267ns ± 0%              479ns ± 0%  +79.25%          (p=0.008 n=5+5)
BM_MatMul<bool>/32              6.42µs ± 0%             0.87µs ± 0%  -86.50%          (p=0.008 n=5+5)
BM_MatMul<bool>/64              43.3µs ± 0%              5.9µs ± 0%  -86.42%          (p=0.008 n=5+5)
BM_MatMul<bool>/128              315µs ± 0%               44µs ± 0%  -85.98%          (p=0.008 n=5+5)
BM_MatMul<bool>/256             2.41ms ± 0%             0.34ms ± 0%  -85.68%          (p=0.008 n=5+5)
BM_MatMul<bool>/512             18.8ms ± 0%              2.7ms ± 0%  -85.53%          (p=0.008 n=5+5)
BM_MatMul<bool>/1k               149ms ± 0%               22ms ± 0%  -85.40%          (p=0.008 n=5+5)
 2020-04-28 16:12:47 +00:00
Eugene Zhulenev
3c02fefec5 Add async evaluation support to TensorSlicingOp. 
Device::memcpy is not async-safe and might lead to deadlocks. Always evaluate slice expression in async mode.
 2020-04-22 19:55:01 +00:00
Changming Sun
b1aa07a8d3 Fix a bug in TensorIndexList.h 2020-04-13 18:22:03 +00:00
jangsoopark
39142904cc Resolve C4346 when building eigen on windows 2020-04-08 14:55:39 +09:00
Deven Desai
7158ed4e0e Fixing HIP breakage caused by the recent commit that introduces Packet4h2 as the Eigen::Half packet type 2020-03-12 01:06:24 +00:00
Sami Kama
b733b8b680 remove duplicate pset1 for half and add some comments about why we need expose pmul/add/div/min/max on host 2020-03-10 20:28:43 +00:00
Cédric Hubert
98bfc5aaa8 Update MarketIO.h 2020-02-28 12:41:51 +00:00
Ilya Tokar
eb6cc29583 Avoid a division in NonBlockingThreadPool::Steal. 
Looking at profiles we spend ~10-20% of Steal on simply computing
random % size. We can reduce random 32-bit int into [0, size) range with
a single multiplication and shift. This transformation is described in
https://lemire.me/blog/2016/06/27/a-fast-alternative-to-the-modulo-reduction/
 2020-02-14 16:02:57 -05:00
Eugene Zhulenev
f584bd9b30 Fail at compile time if default executor tries to use non-default device 2020-02-06 22:43:24 +00:00
Eugene Zhulenev
3fda850c46 Remove dead code from TensorReduction.h 2020-01-29 18:45:31 +00:00
Jeff Daily
b5df8cabd7 fix hip-clang compilation due to new HIP scalar accessor 2020-01-20 21:08:52 +00:00
Deven Desai
6d284bb1b7 Fix for HIP breakage - 200115. Adding a missing EIGEN_DEVICE_FUNC attr 2020-01-16 00:51:43 +00:00
Srinivas Vasudevan
f6c6de5d63 Ensure Igamma does not NaN or Inf for large values. 2020-01-14 21:32:48 +00:00
Eugene Zhulenev
b9362fb8f7 Convert StridedLinearBufferCopy::Kind to enum class 2020-01-13 11:43:24 -08:00
Matthew Powelson
2ea5a715cf Properly initialize b vector in SplineFitting 
InterpolateWithDerivative does not initialize the be vector correctly. This issue is discussed In stackoverflow question 48382939.
 2020-01-09 21:29:04 +00:00
Ilya Tokar
19876ced76 Bug #1785: Introduce numext::rint. 
This provides a new op that matches std::rint and previous behavior of
pround. Also adds corresponding unsupported/../Tensor op.
Performance is the same as e. g. floor (tested SSE/AVX).
 2020-01-07 21:22:44 +00:00
mehdi-goli
d0ae052da4 [SYCL Backend] 
* Adding Missing operations for vector comparison in SYCL. This caused compiler error for vector comparison when compiling SYCL
 * Fixing the compiler error for placement new in TensorForcedEval.h This caused compiler error when compiling SYCL backend
 * Reducing the SYCL warning by  removing the abort function inside the kernel
 * Adding Strong inline to functions inside SYCL interop.
 2020-01-07 15:13:37 +00:00
Deven Desai
636e2bb3fa Fix for HIP breakage - 191220 
The breakage was introduced by the following commit :

ae07801dd8

After the commit, HIPCC errors out on some tests with the following error

```
Building HIPCC object unsupported/test/CMakeFiles/cxx11_tensor_device_1.dir/cxx11_tensor_device_1_generated_cxx11_tensor_device.cu.o
In file included from /home/rocm-user/eigen/unsupported/test/cxx11_tensor_device.cu:17:
In file included from /home/rocm-user/eigen/unsupported/Eigen/CXX11/Tensor💯
/home/rocm-user/eigen/unsupported/Eigen/CXX11/src/Tensor/TensorBlock.h:129:12: error: no matching constructor for initialization of 'Eigen::internal::TensorBlockResourceRequirements'
    return {merge(lhs.shape_type, rhs.shape_type),           // shape_type
           ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/home/rocm-user/eigen/unsupported/Eigen/CXX11/src/Tensor/TensorBlock.h:75:8: note: candidate constructor (the implicit copy constructor) not viable: requires 1 argument, but 3 were provided
struct TensorBlockResourceRequirements {
       ^
/home/rocm-user/eigen/unsupported/Eigen/CXX11/src/Tensor/TensorBlock.h:75:8: note: candidate constructor (the implicit move constructor) not viable: requires 1 argument, but 3 were provided
/home/rocm-user/eigen/unsupported/Eigen/CXX11/src/Tensor/TensorBlock.h:75:8: note: candidate constructor (the implicit copy constructor) not viable: requires 5 arguments, but 3 were provided
/home/rocm-user/eigen/unsupported/Eigen/CXX11/src/Tensor/TensorBlock.h:75:8: note: candidate constructor (the implicit default constructor) not viable: requires 0 arguments, but 3 were provided
...
...
```

The fix is to explicitly decalre the (implicitly called) constructor as a device func
 2019-12-20 21:28:00 +00:00
Christoph Hertzberg
1e9664b147 Bug #1796: Make matrix squareroot usable for Map and Ref types 2019-12-20 18:10:22 +01:00
Christoph Hertzberg
d86544d654 Reduce code duplication and avoid confusing Doxygen 2019-12-19 19:48:39 +01:00
Christoph Hertzberg
dde279f57d Hide recursive meta templates from Doxygen 2019-12-19 19:47:23 +01:00
Christoph Hertzberg
a3273aeff8 Fix trivial shadow warning 2019-12-19 19:13:11 +01:00
Eugene Zhulenev
7a65219a2e Fix TensorPadding bug in squeezed reads from inner dimension 2019-12-19 05:43:57 +00:00
Eugene Zhulenev
73e55525e5 Return const data pointer from TensorRef evaluator.data() 2019-12-18 23:19:36 +00:00
Eugene Zhulenev
ae07801dd8 Tensor block evaluation cost model 2019-12-18 20:07:00 +00:00
Jeff Daily
de07c4d1c2 fix compilation due to new HIP scalar accessor 2019-12-17 20:27:30 +00:00
Eugene Zhulenev
788bef6ab5 Reduce block evaluation overhead for small tensor expressions 2019-12-17 19:06:14 +00:00
Eugene Zhulenev
381f8f3139 Initialize non-trivially constructible types when allocating a temp buffer. 2019-12-12 01:31:30 +00:00
Eugene Zhulenev
64272c7f40 Squeeze reads from two inner dimensions in TensorPadding 2019-12-11 16:54:51 -08:00
Eugene Zhulenev
963ba1015b Add back accidentally deleted default constructor to TensorExecutorTilingContext. 2019-12-11 18:47:55 +00:00
Eugene Zhulenev
c9220c035f Remove block memory allocation required by removed block evaluation API 2019-12-10 17:15:55 -08:00
Eugene Zhulenev
1c879eb010 Remove V2 suffix from TensorBlock 2019-12-10 15:40:23 -08:00
Eugene Zhulenev
dbca11e880 Remove TensorBlock.h and old TensorBlock/BlockMapper 2019-12-10 14:31:44 -08:00
Deven Desai
c49f0d851a Fix for HIP breakage detected on 191210 
The following commit introduces compile errors when running eigen with hipcc

2918f85ba9

hipcc errors out because it requies the device attribute on the methods within the TensorBlockV2ResourceRequirements struct instroduced by the commit above. The fix is to add the device attribute to those methods
 2019-12-10 22:14:05 +00:00
Eugene Zhulenev
2918f85ba9 Do not use std::vector in getResourceRequirements 2019-12-09 16:19:55 -08:00
Artem Belevich
8056a05b54 Undo the block size change. 
.z *is* used by the EigenContractionKernelInternal().
 2019-12-09 11:10:29 -08:00
Eugene Zhulenev
dbb703d44e Add async evaluation support to TensorSelectOp 2019-12-09 18:36:13 +00:00
Janek Kozicki
11d6465326 fix AlignedVector3 inconsisent interface with other Vector classes, default constructor and operator- were missing. 2019-12-06 21:07:39 +01:00
Eugene Zhulenev
bb7ccac3af Add recursive work splitting to EvalShardedByInnerDimContext 2019-12-05 14:51:49 -08:00
Artem Belevich
25230d1862 Improve performance of contraction kernels 
* Force-inline implementations. They pass around pointers to shared memory
  blocks. Without inlining compiler must operate via generic pointers.
  Inlining allows compiler to detect that we're operating on shared memory
  which allows generation of substantially faster code.

* Fixed a long-standing typo which resulted in launching 8x more kernels
  than we needed (.z dimension of the block is unused by the kernel).
 2019-12-05 12:48:34 -08:00
Eugene Zhulenev
8f4536e852 Capture TensorMap by value inside tensor expression AST 2019-12-03 16:39:05 -08:00
Rasmus Munk Larsen
4e696901f8 Remove __host__ annotation for device-only function. 2019-12-03 14:33:19 -08:00
Rasmus Munk Larsen
ead81559c8 Use EIGEN_DEVICE_FUNC macro instead of __device__. 2019-12-03 12:08:22 -08:00
Mehdi Goli
00f32752f7 [SYCL] Rebasing the SYCL support branch on top of the Einge upstream master branch. 
* 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
 2019-11-28 10:08:54 +00:00
Eugene Zhulenev
5496d0da0b Add async evaluation support to TensorReverse 2019-11-26 15:02:24 -08:00
Eugene Zhulenev
bc66c88255 Add async evaluation support to TensorPadding/TensorImagePatch/TensorShuffling 2019-11-26 11:41:57 -08:00
Hans Johnson
8c8cab1afd STYLE: Convert CMake-language commands to lower case 
Ancient CMake versions required upper-case commands.  Later command names
became case-insensitive.  Now the preferred style is lower-case.
 2019-10-31 11:36:37 -05:00
Gael Guennebaud
c3f6fcf2c0 bug #1747: one more fix for MSVC regarding the Bessel implementation. 2019-11-15 11:12:35 +01:00
Gael Guennebaud
b9837ca9ae bug #1281: fix AutoDiffScalar's make_coherent for nested expression of constant ADs. 2019-11-14 14:58:08 +01:00
Eugene Zhulenev
13c3327f5c Remove legacy block evaluation support 2019-11-12 10:12:28 -08:00
Rasmus Munk Larsen
0ed0338593 Fix a race in async tensor evaluation: Don't run on_done() until after device.deallocate() / evaluator.cleanup() complete, since the device might be destroyed after on_done() runs. 2019-11-11 12:26:41 -08:00
Eugene Zhulenev
c952b8dfda Break loop dependence in TensorGenerator block access 2019-11-11 10:32:57 -08:00
Rasmus Munk Larsen
cc3d0e6a40 Add EIGEN_HAS_INTRINSIC_INT128 macro 
Add a new EIGEN_HAS_INTRINSIC_INT128 macro, and use this instead of __SIZEOF_INT128__. This fixes related issues with TensorIntDiv.h when building with Clang for Windows, where support for 128-bit integer arithmetic is advertised but broken in practice.
 2019-11-06 14:24:33 -08:00
Rasmus Munk Larsen
ee404667e2 Rollback or PR-746 and partial rollback of 668ab3fc47 
.

std::array is still not supported in CUDA device code on Windows.
 2019-11-05 17:17:58 -08:00
Rasmus Larsen
0c9745903a Merged in ezhulenev/eigen-01 (pull request PR-746) 
Remove internal::smart_copy and replace with std::copy
 2019-11-04 20:18:38 +00:00
Eugene Zhulenev
73ecb2c57d Cleanup includes in Tensor module after switch to C++11 and above 2019-10-29 15:49:54 -07:00
Eugene Zhulenev
e7ed4bd388 Remove internal::smart_copy and replace with std::copy 2019-10-29 11:25:24 -07:00
Eugene Zhulenev
fbc0a9a3ec Fix CXX11Meta compilation with MSVC 2019-10-28 18:30:10 -07:00
Eugene Zhulenev
bd864ab42b Prevent potential ODR in TensorExecutor 2019-10-28 15:45:09 -07:00
Mehdi Goli
6332aff0b2 This PR fixes: 
* The specialization of array class in the different namespace for GCC<=6.4
* The implicit call to `std::array` constructor using the initializer list for GCC <=6.1
 2019-10-23 15:56:56 +01:00
Rasmus Larsen
8e4e29ae99 Merged in deven-amd/eigen-hip-fix-191018 (pull request PR-738) 
Fix for the HIP build+test errors.
 2019-10-22 22:18:38 +00:00
Rasmus Munk Larsen
97c0c5d485 Add block evaluation V2 to TensorAsyncExecutor. 
Add async evaluation to a number of ops.
 2019-10-22 12:42:44 -07:00
Deven Desai
102cf2a72d Fix for the HIP build+test errors. 
The errors were introduced by this commit :

After the above mentioned commit, some of the tests started failing with the following error


```
Built target cxx11_tensor_reduction
Building HIPCC object unsupported/test/CMakeFiles/cxx11_tensor_reduction_gpu_5.dir/cxx11_tensor_reduction_gpu_5_generated_cxx11_tensor_reduction_gpu.cu.o
In file included from /home/rocm-user/eigen/unsupported/test/cxx11_tensor_reduction_gpu.cu:16:
In file included from /home/rocm-user/eigen/unsupported/Eigen/CXX11/Tensor:117:
/home/rocm-user/eigen/unsupported/Eigen/CXX11/src/Tensor/TensorBlockV2.h:155:5: error: the field type is not amp-compatible
    DestinationBufferKind m_kind;
    ^
/home/rocm-user/eigen/unsupported/Eigen/CXX11/src/Tensor/TensorBlockV2.h:211:3: error: the field type is not amp-compatible
  DestinationBuffer m_destination;
  ^
```


For some reason HIPCC does not like device code to contain enum types which do not have the base-type explicitly declared. The fix is trivial, explicitly state "int" as the basetype
 2019-10-22 19:21:27 +00:00
Rasmus Munk Larsen
668ab3fc47 Drop support for c++03 in Eigen tensor. Get rid of some code used to emulate c++11 functionality with older compilers. 2019-10-18 16:42:00 -07:00
Eugene Zhulenev
df0e8b8137 Propagate block evaluation preference through rvalue tensor expressions 2019-10-17 11:17:33 -07:00
Eugene Zhulenev
0d2a14ce11 Cleanup Tensor block destination and materialized block storage allocation 2019-10-16 17:14:37 -07:00
Eugene Zhulenev
02431cbe71 TensorBroadcasting support for random/uniform blocks 2019-10-16 13:26:28 -07:00
Eugene Zhulenev
d380c23b2c Block evaluation for TensorGenerator/TensorReverse/TensorShuffling 2019-10-14 14:31:59 -07:00
Gael Guennebaud
39fb9eeccf bug #1747: fix compilation with MSVC 2019-10-14 22:50:23 +02:00
Eugene Zhulenev
a411e9f344 Block evaluation for TensorGenerator + TensorReverse + fixed bug in tensor reverse op 2019-10-10 10:56:58 -07:00
Eugene Zhulenev
33e1746139 Block evaluation for TensorChipping + fixed bugs in TensorPadding and TensorSlicing 2019-10-09 12:45:31 -07:00
Gael Guennebaud
f0a4642bab Implement c++03 compatible fix for changeset 7a43af1a33 2019-10-09 16:00:57 +02:00
Eugene Zhulenev
f74ab8cb8d Add block evaluation to TensorEvalTo and fix few small bugs 2019-10-07 15:34:26 -07:00
Brian Zhao
3afb640b56 Fixing incorrect size in Tensor documentation. 2019-10-04 21:30:35 -07:00
Rasmus Munk Larsen
20c4a9118f Use "pdiv" rather than operator/ to support packet types. 2019-10-04 16:54:03 -07:00
Eugene Zhulenev
98bdd7252e Fix compilation warnings and errors with clang in TensorBlockV2 code and tests 2019-10-04 10:15:33 -07:00
Eugene Zhulenev
60ae24ee1a Add block evaluation to TensorReshaping/TensorCasting/TensorPadding/TensorSelect 2019-10-02 12:44:06 -07:00
Eugene Zhulenev
6e40454a6e Add beta to TensorContractionKernel and make memset optional 2019-10-02 11:06:02 -07:00
Rasmus Munk Larsen
13ef08e5ac Move implementation of vectorized error function erf() to SpecialFunctionsImpl.h. 2019-09-27 13:56:04 -07:00
Eugene Zhulenev
71d5bedf72 Fix compilation warnings and errors with clang in TensorBlockV2 2019-09-25 11:25:22 -07:00
Deven Desai
5e186b1987 Fix for the HIP build+test errors. 
The errors were introduced by this commit : d38e6fbc27


After the above mentioned commit, some of the tests started failing with the following error


```
Building HIPCC object unsupported/test/CMakeFiles/cxx11_tensor_reduction_gpu_5.dir/cxx11_tensor_reduction_gpu_5_generated_cxx11_tensor_reduction_gpu.cu.o
In file included from /home/rocm-user/eigen/unsupported/test/cxx11_tensor_reduction_gpu.cu:16:
In file included from /home/rocm-user/eigen/unsupported/Eigen/CXX11/Tensor:29:
In file included from /home/rocm-user/eigen/unsupported/Eigen/CXX11/../SpecialFunctions:70:
/home/rocm-user/eigen/unsupported/Eigen/CXX11/../src/SpecialFunctions/SpecialFunctionsHalf.h:28:22: error: call to 'erf' is ambiguous
  return Eigen::half(Eigen::numext::erf(static_cast<float>(a)));
                     ^~~~~~~~~~~~~~~~~~
/home/rocm-user/eigen/unsupported/test/../../Eigen/src/Core/MathFunctions.h:1600:7: note: candidate function [with T = float]
float erf(const float &x) { return ::erff(x); }
      ^
/home/rocm-user/eigen/unsupported/Eigen/CXX11/../src/SpecialFunctions/SpecialFunctionsImpl.h:1897:5: note: candidate function [with Scalar = float]
    erf(const Scalar& x) {
    ^
In file included from /home/rocm-user/eigen/unsupported/test/cxx11_tensor_reduction_gpu.cu:16:
In file included from /home/rocm-user/eigen/unsupported/Eigen/CXX11/Tensor:29:
In file included from /home/rocm-user/eigen/unsupported/Eigen/CXX11/../SpecialFunctions:75:
/home/rocm-user/eigen/unsupported/Eigen/CXX11/../src/SpecialFunctions/arch/GPU/GpuSpecialFunctions.h:87:23: error: call to 'erf' is ambiguous
  return make_double2(erf(a.x), erf(a.y));
                      ^~~
/home/rocm-user/eigen/unsupported/test/../../Eigen/src/Core/MathFunctions.h:1603:8: note: candidate function [with T = double]
double erf(const double &x) { return ::erf(x); }
       ^
/home/rocm-user/eigen/unsupported/Eigen/CXX11/../src/SpecialFunctions/SpecialFunctionsImpl.h:1897:5: note: candidate function [with Scalar = double]
    erf(const Scalar& x) {
    ^
In file included from /home/rocm-user/eigen/unsupported/test/cxx11_tensor_reduction_gpu.cu:16:
In file included from /home/rocm-user/eigen/unsupported/Eigen/CXX11/Tensor:29:
In file included from /home/rocm-user/eigen/unsupported/Eigen/CXX11/../SpecialFunctions:75:
/home/rocm-user/eigen/unsupported/Eigen/CXX11/../src/SpecialFunctions/arch/GPU/GpuSpecialFunctions.h:87:33: error: call to 'erf' is ambiguous
  return make_double2(erf(a.x), erf(a.y));
                                ^~~
/home/rocm-user/eigen/unsupported/test/../../Eigen/src/Core/MathFunctions.h:1603:8: note: candidate function [with T = double]
double erf(const double &x) { return ::erf(x); }
       ^
/home/rocm-user/eigen/unsupported/Eigen/CXX11/../src/SpecialFunctions/SpecialFunctionsImpl.h:1897:5: note: candidate function [with Scalar = double]
    erf(const Scalar& x) {
    ^
3 errors generated.
```


This PR fixes the compile error by removing the "old" implementation for "erf" (assuming that the "new" implementation is what we want going forward. from a GPU point-of-view both implementations are the same).

This PR also fixes what seems like a cut-n-paste error in the aforementioned commit
 2019-09-25 15:39:13 +00:00
Eugene Zhulenev
f35b9ab510 Fix a bug in a packed block type in TensorContractionThreadPool 2019-09-24 16:54:36 -07:00
Rasmus Larsen
d38e6fbc27 Merged in rmlarsen/eigen (pull request PR-704) 
Add generic PacketMath implementation of the Error Function (erf).
 2019-09-24 23:40:29 +00:00
Rasmus Munk Larsen
591a554c68 Add TODO to cleanup FMA cost modelling. 2019-09-24 16:39:25 -07:00
Eugene Zhulenev
c64396b4c6 Choose TensorBlock StridedLinearCopy type statically 2019-09-24 16:04:29 -07:00
Eugene Zhulenev
c97b208468 Add new TensorBlock api implementation + tests 2019-09-24 15:17:35 -07:00
Eugene Zhulenev
ef9dfee7bd Tensor block evaluation V2 support for unary/binary/broadcsting 2019-09-24 12:52:45 -07:00
Christoph Hertzberg
e4c1b3c1d2 Fix implicit conversion warnings and use pnegate to negate packets 2019-09-23 16:07:43 +02:00
Christoph Hertzberg
ba0736fa8e Fix (or mask away) conversion warnings introduced in 553caeb6a3 
.
 2019-09-23 15:58:05 +02:00
Rasmus Munk Larsen
1d5af0693c Add support for asynchronous evaluation of tensor casting expressions. 2019-09-19 13:54:49 -07:00
Rasmus Munk Larsen
6de5ed08d8 Add generic PacketMath implementation of the Error Function (erf). 2019-09-19 12:48:30 -07:00
Srinivas Vasudevan
df0816b71f Merging eigen/eigen. 2019-09-16 19:33:29 -04:00
Srinivas Vasudevan
6e215cf109 Add Bessel functions to SpecialFunctions. 
- Split SpecialFunctions files in to a separate BesselFunctions file.

In particular add:
    - Modified bessel functions of the second kind k0, k1, k0e, k1e
    - Bessel functions of the first kind j0, j1
    - Bessel functions of the second kind y0, y1
 2019-09-14 12:16:47 -04:00
Eugene Zhulenev
bf8866b466 Fix maybe-unitialized warnings in TensorContractionThreadPool 2019-09-13 14:29:55 -07:00
Eugene Zhulenev
553caeb6a3 Use ThreadLocal container in TensorContractionThreadPool 2019-09-13 12:14:44 -07:00
Srinivas Vasudevan
facdec5aa7 Add packetized versions of i0e and i1e special functions. 
- In particular refactor the i0e and i1e code so scalar and vectorized path share code.
  - Move chebevl to GenericPacketMathFunctions.


A brief benchmark with building Eigen with FMA, AVX and AVX2 flags

Before:

CPU: Intel Haswell with HyperThreading (6 cores)
Benchmark                  Time(ns)        CPU(ns)     Iterations
-----------------------------------------------------------------
BM_eigen_i0e_double/1            57.3           57.3     10000000
BM_eigen_i0e_double/8           398            398        1748554
BM_eigen_i0e_double/64         3184           3184         218961
BM_eigen_i0e_double/512       25579          25579          27330
BM_eigen_i0e_double/4k       205043         205042           3418
BM_eigen_i0e_double/32k     1646038        1646176            422
BM_eigen_i0e_double/256k   13180959       13182613             53
BM_eigen_i0e_double/1M     52684617       52706132             10
BM_eigen_i0e_float/1             28.4           28.4     24636711
BM_eigen_i0e_float/8             75.7           75.7      9207634
BM_eigen_i0e_float/64           512            512        1000000
BM_eigen_i0e_float/512         4194           4194         166359
BM_eigen_i0e_float/4k         32756          32761          21373
BM_eigen_i0e_float/32k       261133         261153           2678
BM_eigen_i0e_float/256k     2087938        2088231            333
BM_eigen_i0e_float/1M       8380409        8381234             84
BM_eigen_i1e_double/1            56.3           56.3     10000000
BM_eigen_i1e_double/8           397            397        1772376
BM_eigen_i1e_double/64         3114           3115         223881
BM_eigen_i1e_double/512       25358          25361          27761
BM_eigen_i1e_double/4k       203543         203593           3462
BM_eigen_i1e_double/32k     1613649        1613803            428
BM_eigen_i1e_double/256k   12910625       12910374             54
BM_eigen_i1e_double/1M     51723824       51723991             10
BM_eigen_i1e_float/1             28.3           28.3     24683049
BM_eigen_i1e_float/8             74.8           74.9      9366216
BM_eigen_i1e_float/64           505            505        1000000
BM_eigen_i1e_float/512         4068           4068         171690
BM_eigen_i1e_float/4k         31803          31806          21948
BM_eigen_i1e_float/32k       253637         253692           2763
BM_eigen_i1e_float/256k     2019711        2019918            346
BM_eigen_i1e_float/1M       8238681        8238713             86


After:

CPU: Intel Haswell with HyperThreading (6 cores)
Benchmark                  Time(ns)        CPU(ns)     Iterations
-----------------------------------------------------------------
BM_eigen_i0e_double/1            15.8           15.8     44097476
BM_eigen_i0e_double/8            99.3           99.3      7014884
BM_eigen_i0e_double/64          777            777         886612
BM_eigen_i0e_double/512        6180           6181         100000
BM_eigen_i0e_double/4k        48136          48140          14678
BM_eigen_i0e_double/32k      385936         385943           1801
BM_eigen_i0e_double/256k    3293324        3293551            228
BM_eigen_i0e_double/1M     12423600       12424458             57
BM_eigen_i0e_float/1             16.3           16.3     43038042
BM_eigen_i0e_float/8             30.1           30.1     23456931
BM_eigen_i0e_float/64           169            169        4132875
BM_eigen_i0e_float/512         1338           1339         516860
BM_eigen_i0e_float/4k         10191          10191          68513
BM_eigen_i0e_float/32k        81338          81337           8531
BM_eigen_i0e_float/256k      651807         651984           1000
BM_eigen_i0e_float/1M       2633821        2634187            268
BM_eigen_i1e_double/1            16.2           16.2     42352499
BM_eigen_i1e_double/8           110            110        6316524
BM_eigen_i1e_double/64          822            822         851065
BM_eigen_i1e_double/512        6480           6481         100000
BM_eigen_i1e_double/4k        51843          51843          10000
BM_eigen_i1e_double/32k      414854         414852           1680
BM_eigen_i1e_double/256k    3320001        3320568            212
BM_eigen_i1e_double/1M     13442795       13442391             53
BM_eigen_i1e_float/1             17.6           17.6     41025735
BM_eigen_i1e_float/8             35.5           35.5     19597891
BM_eigen_i1e_float/64           240            240        2924237
BM_eigen_i1e_float/512         1424           1424         485953
BM_eigen_i1e_float/4k         10722          10723          65162
BM_eigen_i1e_float/32k        86286          86297           8048
BM_eigen_i1e_float/256k      691821         691868           1000
BM_eigen_i1e_float/1M       2777336        2777747            256


This shows anywhere from a 50% to 75% improvement on these operations.

I've also benchmarked without any of these flags turned on, and got similar
performance to before (if not better).

Also tested packetmath.cpp + special_functions to ensure no regressions.
 2019-09-11 18:34:02 -07:00
Deven Desai
cdb377d0cb Fix for the HIP build+test errors introduced by the ndtri support. 
The fixes needed are
 * adding EIGEN_DEVICE_FUNC attribute to a couple of funcs (else HIPCC will error out when non-device funcs are called from global/device funcs)
 * switching to using ::<math_func> instead std::<math_func> (only for HIPCC) in cases where the std::<math_func> is not recognized as a device func by HIPCC
 * removing an errant "j" from a testcase (don't know how that made it in to begin with!)
 2019-09-06 16:03:49 +00:00
Eugene Zhulenev
d918bd9a8b Update ThreadLocal to use separate Initialize/Release callables 2019-09-10 16:13:32 -07:00
Eugene Zhulenev
e3dec4dcc1 ThreadLocal container that does not rely on thread local storage 2019-09-09 15:18:14 -07:00
Srinivas Vasudevan
e38dd48a27 PR 681: Add ndtri function, the inverse of the normal distribution function. 2019-08-12 19:26:29 -04:00
Eugene Zhulenev
47fefa235f Allow move-only done callback in TensorAsyncDevice 2019-09-03 17:20:56 -07:00
Eugene Zhulenev
f68f2bba09 TensorMap constness should not change underlying storage constness 2019-09-03 11:08:09 -07:00
Alberto Luaces
c694be1214 Fixed Tensor documentation formatting. 2019-07-23 09:24:06 +00:00
Eugene Zhulenev
79c402e40e Fix shadow warnings in TensorContractionThreadPool 2019-08-30 15:38:31 -07:00
Eugene Zhulenev
edf2ec28d8 Fix block mapper type name in TensorExecutor 2019-08-30 15:29:25 -07:00
Eugene Zhulenev
f0b36fb9a4 evalSubExprsIfNeededAsync + async TensorContractionThreadPool 2019-08-30 15:13:38 -07:00
Eugene Zhulenev
619cea9491 Revert accidentally removed <memory> header from ThreadPool 2019-08-30 14:51:17 -07:00
Eugene Zhulenev
66665e7e76 Asynchronous expression evaluation with TensorAsyncDevice 2019-08-30 14:49:40 -07:00
Eugene Zhulenev
bc40d4522c Const correctness in TensorMap<const Tensor<T, ...>> expressions 2019-08-28 17:46:05 -07:00
Eugene Zhulenev
6e77f9bef3 Remove shadow warnings in TensorDeviceThreadPool 2019-08-28 10:32:19 -07:00
Rasmus Larsen
84fefdf321 Merged in ezhulenev/eigen-01 (pull request PR-683) 
Asynchronous parallelFor in Eigen ThreadPoolDevice
 2019-08-26 21:49:17 +00:00
maratek
8b5ab0e4dd Fix get_random_seed on Native Client 
Newlib in Native Client SDK does not provide ::random function.
Implement get_random_seed for NaCl using ::rand, similarly to Windows version.
 2019-08-23 15:25:56 -07:00
Eugene Zhulenev
6901788013 Asynchronous parallelFor in Eigen ThreadPoolDevice 2019-08-22 10:50:51 -07:00
Eugene Zhulenev
071311821e Remove XSMM support from Tensor module 2019-08-19 11:44:25 -07:00
Rasmus Munk Larsen
eab7e52db2 [Eigen] Vectorize evaluation of coefficient-wise functions over tensor blocks if the strides are known to be 1. Provides up to 20-25% speedup of the TF cross entropy op with AVX. 
A few benchmark numbers:

name                              old time/op             new time/op             delta
BM_Xent_16_10000_cpu              448µs ± 3%              389µs ± 2%  -13.21%
(p=0.008 n=5+5)
BM_Xent_32_10000_cpu              575µs ± 6%              454µs ± 3%  -21.00%          (p=0.008 n=5+5)
BM_Xent_64_10000_cpu              933µs ± 4%              712µs ± 1%  -23.71%          (p=0.008 n=5+5)
 2019-08-07 12:57:42 -07:00
Rasmus Munk Larsen
0987126165 Clean up unnecessary namespace specifiers in TensorBlock.h. 2019-08-07 12:12:52 -07:00
Rasmus Munk Larsen
e2999d4c38 Fix performance regressions due to https://bitbucket.org/eigen/eigen/pull-requests/662. 
The change caused the device struct to be copied for each expression evaluation, and caused, e.g., a 10% regression in the TensorFlow multinomial op on GPU:


Benchmark                       Time(ns)        CPU(ns)     Iterations
----------------------------------------------------------------------
BM_Multinomial_gpu_1_100000_4     128173         231326           2922  1.610G items/s

VS

Benchmark                       Time(ns)        CPU(ns)     Iterations
----------------------------------------------------------------------
BM_Multinomial_gpu_1_100000_4     146683         246914           2719  1.509G items/s
 2019-08-02 11:18:13 -07:00
Eugene Zhulenev
3cd148f983 Fix expression evaluation heuristic for TensorSliceOp 2019-07-09 12:10:26 -07:00
Eugene Zhulenev
6083014594 Add outer/inner chipping optimization for chipping dimension specified at runtime 2019-07-03 11:35:25 -07:00
Deven Desai
7eb2e0a95b adding the EIGEN_DEVICE_FUNC attribute to the constCast routine. 
Not having this attribute results in the following failures in the `--config=rocm` TF build.

```
In file included from tensorflow/core/kernels/cross_op_gpu.cu.cc:20:
In file included from ./tensorflow/core/framework/register_types.h:20:
In file included from ./tensorflow/core/framework/numeric_types.h:20:
In file included from ./third_party/eigen3/unsupported/Eigen/CXX11/Tensor:1:
In file included from external/eigen_archive/unsupported/Eigen/CXX11/Tensor:140:
external/eigen_archive/unsupported/Eigen/CXX11/src/Tensor/TensorChipping.h:356:37: error:  'Eigen::constCast':  no overloaded function has restriction specifiers that are compatible with the ambient context 'data'
    typename Storage::Type result = constCast(m_impl.data());
                                    ^
external/eigen_archive/unsupported/Eigen/CXX11/src/Tensor/TensorChipping.h:356:37: error:  'Eigen::constCast':  no overloaded function has restriction specifiers that are compatible with the ambient context 'data'
external/eigen_archive/unsupported/Eigen/CXX11/src/Tensor/TensorAssign.h:148:56: note: in instantiation of member function 'Eigen::TensorEvaluator<const Eigen::TensorChippingOp<1, Eigen::TensorMap<Eigen::Tensor<int, 2, 1, long>, 16, MakePointer> >, Eigen::Gpu\
Device>::data' requested here
    return m_rightImpl.evalSubExprsIfNeeded(m_leftImpl.data());

```

Adding the EIGEN_DEVICE_FUNC attribute resolves those errors
 2019-07-02 20:02:46 +00:00
Gael Guennebaud
ef8aca6a89 Merged in codeplaysoftware/eigen (pull request PR-667) 
[SYCL] :

Approved-by: Gael Guennebaud <g.gael@free.fr>
Approved-by: Rasmus Larsen <rmlarsen@google.com>
 2019-07-02 12:45:23 +00:00
Eugene Zhulenev
4ac93f8edc Allocate non-const scalar buffer for block evaluation with DefaultDevice 2019-07-01 10:55:19 -07:00
Mehdi Goli
9ea490c82c [SYCL] : 
* Modifying TensorDeviceSYCL to use `EIGEN_THROW_X`.
  * Modifying TensorMacro to use `EIGEN_TRY/CATCH(X)` macro.
  * Modifying TensorReverse.h to use `EIGEN_DEVICE_REF` instead of `&`.
  * Fixing the SYCL device macro in SpecialFunctionsImpl.h.
 2019-07-01 16:27:28 +01:00
Eugene Zhulenev
81a03bec75 Fix TensorReverse on GPU with m_stride[i]==0 2019-06-28 15:50:39 -07:00
Rasmus Munk Larsen
74a9dd1102 Fix preprocessor condition to only generate a warning when calling eigen::GpuDevice::synchronize() from device code, but not when calling from a non-GPU compilation unit. 2019-06-28 11:56:21 -07:00
Rasmus Munk Larsen
70d4020ad9 Remove comma causing warning in c++03 mode. 2019-06-28 11:39:45 -07:00
Eugene Zhulenev
6e7c76481a Merge with Eigen head 2019-06-28 11:22:46 -07:00
Eugene Zhulenev
878845cb25 Add block access to TensorReverseOp and make sure that TensorForcedEval uses block access when preferred 2019-06-28 11:13:44 -07:00
Mehdi Goli
7d08fa805a [SYCL] This PR adds the minimum modifications to the Eigen unsupported module required to run it on devices supporting SYCL. 
* Abstracting the pointer type so that both SYCL memory and pointer can be captured.
* Converting SYCL virtual pointer to SYCL device memory in Eigen evaluator class.
* Binding SYCL placeholder accessor to command group handler by using bind method in Eigen evaluator node.
* Adding SYCL macro for controlling loop unrolling.
* Modifying the TensorDeviceSycl.h and SYCL executor method to adopt the above changes.
 2019-06-28 10:08:23 +01:00
Christoph Hertzberg
adec097c61 Remove extra comma (causes warnings in C++03) 2019-06-26 16:14:28 +02:00
Eugene Zhulenev
229db81572 Optimize evaluation strategy for TensorSlicingOp and TensorChippingOp 2019-06-25 15:41:37 -07:00
Rasmus Munk Larsen
b08527b0c1 Clean up CUDA/NVCC version macros and their use in Eigen, and a few other CUDA build failures. 2019-05-31 15:26:06 -07:00
Michael Tesch
c5019f722b Use pade for matrix exponential also for complex values. 2019-05-08 17:04:55 +02:00
Rasmus Larsen
e92486b8c3 Merged in rmlarsen/eigen (pull request PR-643) 
Make Eigen build with cuda 10 and clang.

Approved-by: Justin Lebar <justin.lebar@gmail.com>
 2019-05-20 17:02:39 +00:00
Eugene Zhulenev
01654d97fa Prevent potential division by zero in TensorExecutor 2019-05-17 14:02:25 -07:00
Eugene Zhulenev
96a276803c Always evaluate Tensor expressions with broadcasting via tiled evaluation code path 2019-05-16 16:15:45 -07:00
Rasmus Munk Larsen
ab0a30e429 Make Eigen build with cuda 10 and clang. 2019-05-15 13:32:15 -07:00
Rasmus Munk Larsen
e5ac8cbd7a A) fix deadlocks in thread pool caused by EventCount 
This fixed 2 deadlocks caused by sloppiness in the EventCount logic.
Both most likely were introduced by cl/236729920 which includes the new EventCount algorithm:
01da8caf00

bug #1 (Prewait):
Prewait must not consume existing signals.
Consider the following scenario.
There are 2 thread pool threads (1 and 2) and 1 external thread (3). RunQueue is empty.
Thread 1 checks the queue, calls Prewait, checks RunQueue again and now is going to call CommitWait.
Thread 2 checks the queue and now is going to call Prewait.
Thread 3 submits 2 tasks, EventCount signals is set to 1 because only 1 waiter is registered the second signal is discarded).
Now thread 2 resumes and calls Prewait and takes away the signal.
Thread 1 resumes and calls CommitWait, there are no pending signals anymore, so it blocks.
As the result we have 2 tasks, but only 1 thread is running.

bug #2 (CancelWait):
CancelWait must not take away a signal if it's not sure that the signal was meant for this thread.
When one thread blocks and another submits a new task concurrently, the EventCount protocol guarantees only the following properties (similar to the Dekker's algorithm):
(a) the registered waiter notices presence of the new task and does not block
(b) the signaler notices presence of the waiters and wakes it
(c) both the waiter notices presence of the new task and signaler notices presence of the waiter
[it's only that both of them do not notice each other must not be possible, because it would lead to a deadlock]
CancelWait is called for cases (a) and (c). For case (c) it is OK to take the notification signal away, but it's not OK for (a) because nobody queued a signals for us and we take away a signal meant for somebody else.
Consider:
Thread 1 calls Prewait, checks RunQueue, it's empty, now it's going to call CommitWait.
Thread 3 submits 2 tasks, EventCount signals is set to 1 because only 1 waiter is registered the second signal is discarded).
Thread 2 calls Prewait, checks RunQueue, discovers the tasks, calls CancelWait and consumes the pending signal (meant for thread 1).
Now Thread 1 resumes and calls CommitWait, since there are no signals it blocks.
As the result we have 2 tasks, but only 1 thread is running.

Both deadlocks are only a problem if the tasks require parallelism. Most computational tasks do not require parallelism, i.e. a single thread will run task 1, finish it and then dequeue and run task 2.

This fix undoes some of the sloppiness in the EventCount that was meant to reduce CPU consumption by idle threads, because we now have more threads running in these corner cases. But we still don't have pthread_yield's and maybe the strictness introduced by this change will actually help to reduce tail latency because we will have threads running when we actually need them running.



B) fix deadlock in thread pool caused by RunQueue

This fixed a deadlock caused by sloppiness in the RunQueue logic.
Most likely this was introduced with the non-blocking thread pool.
The deadlock only affects workloads that require parallelism.
Most computational tasks don't require parallelism.

PopBack must not fail spuriously. If it does, it can effectively lead to single thread consuming several wake up signals.
Consider 2 worker threads are blocked.
External thread submits a task. One of the threads is woken.
It tries to steal the task, but fails due to a spurious failure in PopBack (external thread submits another task and holds the lock).
The thread executes blocking protocol again (it won't block because NonEmptyQueueIndex is precise and the thread will discover pending work, but it has called PrepareWait).
Now external thread submits another task and signals EventCount again.
The signal is consumed by the first thread again. But now we have 2 tasks pending but only 1 worker thread running.

It may be possible to fix this in a different way: make EventCount::CancelWait forward wakeup signal to a blocked thread rather then consuming it. But this looks more complex and I am not 100% that it will fix the bug.
It's also possible to have 2 versions of PopBack: one will do try_to_lock and another won't. Then worker threads could first opportunistically check all queues with try_to_lock, and only use the blocking version before blocking. But let's first fix the bug with the simpler change.
 2019-05-08 10:16:46 -07:00
Christoph Hertzberg
e6667a7060 Fix stupid shadow-warnings (with old clang versions) 2019-05-07 18:32:19 +02:00
Christoph Hertzberg
e54dc24d62 Restore C++03 compatibility 2019-05-07 18:30:44 +02:00
Rasmus Larsen
ac50afaffa Merged in ezhulenev/eigen-01 (pull request PR-633) 
Check if gpu_assert was overridden in TensorGpuHipCudaDefines
 2019-04-29 16:29:35 +00:00
Eugene Zhulenev
01d7e6ee9b Check if gpu_assert was overridden in TensorGpuHipCudaDefines 2019-04-25 11:19:17 -07:00
Eugene Zhulenev
8ead5bb3d8 Fix doxygen warnings to enable statis code analysis 2019-04-24 12:42:28 -07:00
Rasmus Munk Larsen
144ca33321 Remove deprecation annotation from typedef Eigen::Index Index, as it would generate too many build warnings. 2019-04-24 08:50:07 -07:00
Eugene Zhulenev
a7b7f3ca8a Add missing EIGEN_DEPRECATED annotations to deprecated functions and fix few other doxygen warnings 2019-04-23 17:23:19 -07:00
Anuj Rawat
8c7a6feb8e Adding lowlevel APIs for optimized RHS packet load in TensorFlow 
SpatialConvolution

Low-level APIs are added in order to optimized packet load in gemm_pack_rhs
in TensorFlow SpatialConvolution. The optimization is for scenario when a
packet is split across 2 adjacent columns. In this case we read it as two
'partial' packets and then merge these into 1. Currently this only works for
Packet16f (AVX512) and Packet8f (AVX2). We plan to add this for other
packet types (such as Packet8d) also.

This optimization shows significant speedup in SpatialConvolution with
certain parameters. Some examples are below.

Benchmark parameters are specified as:
Batch size, Input dim, Depth, Num of filters, Filter dim

Speedup numbers are specified for number of threads 1, 2, 4, 8, 16.

AVX512:

Parameters                  | Speedup (Num of threads: 1, 2, 4, 8, 16)
----------------------------|------------------------------------------
128,   24x24,  3, 64,   5x5 |2.18X, 2.13X, 1.73X, 1.64X, 1.66X
128,   24x24,  1, 64,   8x8 |2.00X, 1.98X, 1.93X, 1.91X, 1.91X
 32,   24x24,  3, 64,   5x5 |2.26X, 2.14X, 2.17X, 2.22X, 2.33X
128,   24x24,  3, 64,   3x3 |1.51X, 1.45X, 1.45X, 1.67X, 1.57X
 32,   14x14, 24, 64,   5x5 |1.21X, 1.19X, 1.16X, 1.70X, 1.17X
128, 128x128,  3, 96, 11x11 |2.17X, 2.18X, 2.19X, 2.20X, 2.18X

AVX2:

Parameters                  | Speedup (Num of threads: 1, 2, 4, 8, 16)
----------------------------|------------------------------------------
128,   24x24,  3, 64,   5x5 | 1.66X, 1.65X, 1.61X, 1.56X, 1.49X
 32,   24x24,  3, 64,   5x5 | 1.71X, 1.63X, 1.77X, 1.58X, 1.68X
128,   24x24,  1, 64,   5x5 | 1.44X, 1.40X, 1.38X, 1.37X, 1.33X
128,   24x24,  3, 64,   3x3 | 1.68X, 1.63X, 1.58X, 1.56X, 1.62X
128, 128x128,  3, 96, 11x11 | 1.36X, 1.36X, 1.37X, 1.37X, 1.37X

In the higher level benchmark cifar10, we observe a runtime improvement
of around 6% for AVX512 on Intel Skylake server (8 cores).

On lower level PackRhs micro-benchmarks specified in TensorFlow
tensorflow/core/kernels/eigen_spatial_convolutions_test.cc, we observe
the following runtime numbers:

AVX512:

Parameters                                                     | Runtime without patch (ns) | Runtime with patch (ns) | Speedup
---------------------------------------------------------------|----------------------------|-------------------------|---------
BM_RHS_NAME(PackRhs, 128, 24, 24, 3, 64, 5, 5, 1, 1, 256, 56)  |  41350                     | 15073                   | 2.74X
BM_RHS_NAME(PackRhs, 32, 64, 64, 32, 64, 5, 5, 1, 1, 256, 56)  |   7277                     |  7341                   | 0.99X
BM_RHS_NAME(PackRhs, 32, 64, 64, 32, 64, 5, 5, 2, 2, 256, 56)  |   8675                     |  8681                   | 1.00X
BM_RHS_NAME(PackRhs, 32, 64, 64, 30, 64, 5, 5, 1, 1, 256, 56)  |  24155                     | 16079                   | 1.50X
BM_RHS_NAME(PackRhs, 32, 64, 64, 30, 64, 5, 5, 2, 2, 256, 56)  |  25052                     | 17152                   | 1.46X
BM_RHS_NAME(PackRhs, 32, 256, 256, 4, 16, 8, 8, 1, 1, 256, 56) |  18269                     | 18345                   | 1.00X
BM_RHS_NAME(PackRhs, 32, 256, 256, 4, 16, 8, 8, 2, 4, 256, 56) |  19468                     | 19872                   | 0.98X
BM_RHS_NAME(PackRhs, 32, 64, 64, 4, 16, 3, 3, 1, 1, 36, 432)   | 156060                     | 42432                   | 3.68X
BM_RHS_NAME(PackRhs, 32, 64, 64, 4, 16, 3, 3, 2, 2, 36, 432)   | 132701                     | 36944                   | 3.59X

AVX2:

Parameters                                                     | Runtime without patch (ns) | Runtime with patch (ns) | Speedup
---------------------------------------------------------------|----------------------------|-------------------------|---------
BM_RHS_NAME(PackRhs, 128, 24, 24, 3, 64, 5, 5, 1, 1, 256, 56)  | 26233                      | 12393                   | 2.12X
BM_RHS_NAME(PackRhs, 32, 64, 64, 32, 64, 5, 5, 1, 1, 256, 56)  |  6091                      |  6062                   | 1.00X
BM_RHS_NAME(PackRhs, 32, 64, 64, 32, 64, 5, 5, 2, 2, 256, 56)  |  7427                      |  7408                   | 1.00X
BM_RHS_NAME(PackRhs, 32, 64, 64, 30, 64, 5, 5, 1, 1, 256, 56)  | 23453                      | 20826                   | 1.13X
BM_RHS_NAME(PackRhs, 32, 64, 64, 30, 64, 5, 5, 2, 2, 256, 56)  | 23167                      | 22091                   | 1.09X
BM_RHS_NAME(PackRhs, 32, 256, 256, 4, 16, 8, 8, 1, 1, 256, 56) | 23422                      | 23682                   | 0.99X
BM_RHS_NAME(PackRhs, 32, 256, 256, 4, 16, 8, 8, 2, 4, 256, 56) | 23165                      | 23663                   | 0.98X
BM_RHS_NAME(PackRhs, 32, 64, 64, 4, 16, 3, 3, 1, 1, 36, 432)   | 72689                      | 44969                   | 1.62X
BM_RHS_NAME(PackRhs, 32, 64, 64, 4, 16, 3, 3, 2, 2, 36, 432)   | 61732                      | 39779                   | 1.55X

All benchmarks on Intel Skylake server with 8 cores.
 2019-04-20 06:46:43 +00:00
Rasmus Munk Larsen
039ee52125 Tweak cost model for tensor contraction when parallelizing over the inner dimension. 
https://bitbucket.org/snippets/rmlarsen/MexxLo
 2019-04-12 13:35:10 -07:00
Jonathon Koyle
9a3f06d836 Update TheadPoolDevice example to include ThreadPool creation and passing pointer into constructor. 2019-04-10 10:02:33 -06:00
Deven Desai
66a885b61e adding EIGEN_DEVICE_FUNC to the recently added TensorContractionKernel constructor. Not having the EIGEN_DEVICE_FUNC attribute on it was leading to compiler errors when compiling Eigen in the ROCm/HIP path 2019-04-08 13:45:08 +00:00
Eugene Zhulenev
629ddebd15 Add missing semicolon 2019-04-02 15:04:26 -07:00