mirror/eigen
2
0
Fork 0
mirror of https://gitlab.com/libeigen/eigen.git synced 2025-01-24 14:45:14 +08:00
Code Issues Packages Projects Releases Wiki Activity
10,620 Commits 11 Branches 89 Tags 171 MiB
e0f5a2a456
Commit Graph

6019 Commits

Author SHA1 Message Date
Gael Guennebaud
f159cf3d75 Artificially increase l1-blocking size for AVX512. +10% speedup with current kernels. 
With a 6pX4 kernel (not committed yet), this provides a +20% speedup.
 2018-12-11 15:36:27 +01:00
Gael Guennebaud
0a7e7af6fd Properly set the number of registers for AVX512 2018-12-11 15:33:17 +01:00
Gael Guennebaud
7166496f70 bug #1643: fix compilation issue with gcc and no optimizaion 2018-12-11 13:24:42 +01:00
Gael Guennebaud
0d90637838 enable spilling workaround on architectures with SSE/AVX 2018-12-10 23:22:44 +01:00
Gael Guennebaud
bff90bf270 workaround "may be used uninitialized" warning 2018-12-08 18:58:28 +01:00
Gael Guennebaud
81c27325ae bug #1641: fix testing of pandnot and fix pandnot for complex on SSE/AVX/AVX512 2018-12-08 14:27:48 +01:00
Gael Guennebaud
426bce7529 fix EIGEN_GEBP_2PX4_SPILLING_WORKAROUND for non vectorized type, and non x86/64 target 2018-12-08 09:44:21 +01:00
Gael Guennebaud
956678a4ef bug #1515: disable gebp's 3pX4 micro kernel for MSVC<=19.14 because of register spilling. 2018-12-07 18:03:36 +01:00
Gael Guennebaud
7b6d0ff1f6 Enable FMA with MSVC (through /arch:AVX2). To make this possible, I also has to turn the #warning regarding AVX512-FMA to a #error. 2018-12-07 15:14:50 +01:00
Gael Guennebaud
f233c6194d bug #1637: workaround register spilling in gebp with clang>=6.0+AVX+FMA 2018-12-07 10:01:09 +01:00
Gael Guennebaud
ae59a7652b bug #1638: add a warning if avx512 is enabled without SSE/AVX FMA 2018-12-07 09:23:28 +01:00
Gael Guennebaud
4e7746fe22 bug #1636: fix gemm performance issue with gcc>=6 and no FMA 2018-12-07 09:15:46 +01:00
Gael Guennebaud
cbf2f4b7a0 AVX512f includes FMA but GCC does not define __FMA__ with -mavx512f only 2018-12-06 18:21:56 +01:00
Gael Guennebaud
1d683ae2f5 Fix compilation with avx512f only, i.e., no AVX512DQ 2018-12-06 18:11:07 +01:00
Gael Guennebaud
c53eececb0 Implement AVX512 vectorization of std::complex<float/double> 2018-12-06 15:58:06 +01:00
Gael Guennebaud
3fba59ea59 temporarily re-disable SSE/AVX vectorization of complex<> on AVX512 -> this needs to be fixed though! 2018-12-06 00:13:26 +01:00
Gael Guennebaud
1ac2695ef7 bug #1636: fix compilation with some ABI versions. 2018-12-06 00:05:10 +01:00
Rasmus Munk Larsen
47d8b741b2 #elif -> #else to fix GPU build. 2018-12-05 13:19:31 -08:00
Christoph Hertzberg
c1d356e8b4 bug #1635: Use infinity from Numtraits instead of creating it manually. 2018-12-05 15:01:04 +01:00
Rasmus Munk Larsen
b57b31cce9 Merged in ezhulenev/eigen-01 (pull request PR-553) 
Do not disable alignment with EIGEN_GPUCC

Approved-by: Rasmus Munk Larsen <rmlarsen@google.com>
 2018-12-04 23:47:19 +00:00
Eugene Zhulenev
0bb15bb6d6 Update checks in ConfigureVectorization.h 2018-12-03 17:10:40 -08:00
Eugene Zhulenev
fd0fbfa9b5 Do not disable alignment with EIGEN_GPUCC 2018-12-03 15:54:10 -08:00
Christoph Hertzberg
919414b9fe bug #785: Make Cholesky decomposition work for empty matrices 2018-12-03 16:18:15 +01:00
Gael Guennebaud
0ea7ae7213 Add missing padd for Packet8i (it was implicitly generated by clang and gcc) 2018-11-30 21:52:25 +01:00
Gael Guennebaud
ab4df3e6ff bug #1634: remove double copy in move-ctor of non movable Matrix/Array 2018-11-30 21:25:51 +01:00
Gael Guennebaud
c785464430 Add packet sin and cos to Altivec/VSX and NEON 2018-11-30 16:21:33 +01:00
Gael Guennebaud
69ace742be Several improvements regarding packet-bitwise operations: 
- add unit tests
- optimize their AVX512f implementation
- add missing implementations (half, Packet4f, ...)
 2018-11-30 15:56:08 +01:00
Gael Guennebaud
fa87f9d876 Add psin/pcos on AVX512 -> almost for free, at last! 2018-11-30 14:33:13 +01:00
Gael Guennebaud
c68bd2fa7a Cleanup 2018-11-30 14:32:31 +01:00
Gael Guennebaud
f91500d303 Fix pandnot order in AVX512 2018-11-30 14:32:06 +01:00
Gael Guennebaud
b477d60bc6 Extend the generic psin_float code to handle cosine and make SSE and AVX use it (-> this adds pcos for AVX) 2018-11-30 11:26:30 +01:00
Gael Guennebaud
e19ece822d Disable fma gcc's workaround for gcc >= 8 (based on GEMM benchmarks) 2018-11-28 17:56:24 +01:00
Gael Guennebaud
41052f63b7 same for pmax 2018-11-28 17:17:28 +01:00
Gael Guennebaud
3e95e398b6 pmin/pmax o SSE: make sure to use AVX instruction with AVX enabled, and disable gcc workaround for fixed gcc versions 2018-11-28 17:14:20 +01:00
Gael Guennebaud
aa6097395b Add missing SSE/AVX type-casting in AVX512 mode 2018-11-28 16:09:08 +01:00
Gael Guennebaud
48fe78c375 bug #1630: fix linspaced when requesting smaller packet size than default one. 2018-11-28 13:15:06 +01:00
Eugene Zhulenev
80f1651f35 Use explicit packet type in SSE/PacketMath pldexp 2018-11-27 17:25:49 -08:00
Benoit Jacob
a4159dba08 do not read buffers out of bounds -- load only the 4 bytes we know exist here. Could also have done a vld1_lane_f32 but doing so here, without the overhead of initializing the unused lane, would have triggered used-of-uninitialized-value errors in tools such as ASan. Note that this code is sub-optimal before or after this change: we should be reading either 2 or 4 float32 values per load-instruction (2 for ARM in-order cores with an affinity for 8-byte loads; 4 for ARM out-of-order cores able to dual-issue 16-byte load instructions with arithmetic instructions). Before or after this patch, we are only loading 4 bytes of useful data here (even if before this patch, we were technically loading 8, only to use only the 4 first). 2018-11-27 16:53:14 -05:00
Gael Guennebaud
b131a4db24 bug #1631: fix compilation with ARM NEON and clang, and cleanup the weird pshiftright_and_cast and pcast_and_shiftleft functions. 2018-11-27 23:45:00 +01:00
Gael Guennebaud
a1a5fbbd21 Update pshiftleft to pass the shift as a true compile-time integer. 2018-11-27 22:57:30 +01:00
Gael Guennebaud
fa7fd61eda Unify SSE/AVX psin functions. 
It is based on the SSE version which is much more accurate, though very slightly slower.
This changeset also includes the following required changes:
 - add packet-float to packet-int type traits
 - add packet float<->int reinterpret casts
 - add faster pselect for AVX based on blendv
 2018-11-27 22:41:51 +01:00
Benoit Jacob
7b1cb8a440 fix the build on 64-bit ARM when NEON is disabled 2018-11-27 11:11:02 -05:00
Gael Guennebaud
b5695a6008 Unify Altivec/VSX pexp(double) with default implementation 2018-11-27 13:53:05 +01:00
Gael Guennebaud
7655a8af6e cleanup 2018-11-26 23:21:29 +01:00
Gael Guennebaud
502f92fa10 Unify SSE and AVX pexp for double. 2018-11-26 23:12:44 +01:00
Gael Guennebaud
4a347a0054 Unify NEON's pexp with generic implementation 2018-11-26 22:15:44 +01:00
Gael Guennebaud
5c8406babc Unify Altivec/VSX's pexp with generic implementation 2018-11-26 16:47:13 +01:00
Gael Guennebaud
cf8b85d5c5 Unify SSE and AVX implementation of pexp 2018-11-26 16:36:19 +01:00
Gael Guennebaud
c2f35b1b47 Unify Altivec/VSX's plog with generic implementation, and enable it! 2018-11-26 15:58:11 +01:00
Gael Guennebaud
c24e98e6a8 Unify NEON's plog with generic implementation 2018-11-26 15:02:16 +01:00
Gael Guennebaud
2c44c40114 First step toward a unification of packet log implementation, currently only SSE and AVX are unified. 
To this end, I added the following functions: pzero, pcmp_*, pfrexp, pset1frombits functions.
 2018-11-26 14:21:24 +01:00
Gael Guennebaud
5f6045077c Make SSE/AVX pandnot(A,B) consistent with generic version, i.e., "A and not B" 2018-11-26 14:14:07 +01:00
Gael Guennebaud
0836a715d6 bug #1611: fix plog(0) on NEON 2018-11-26 09:08:38 +01:00
Patrik Huber
95566eeed4 Fix typos 2018-11-23 22:22:14 +00:00
Gael Guennebaud
ccabdd88c9 Fix reserved usage of double __ in macro names 2018-11-23 16:01:47 +01:00
Gael Guennebaud
a7842daef2 Fix several uninitialized member from ctor 2018-11-23 15:10:28 +01:00
Gael Guennebaud
a476054879 bug #1624: improve matrix-matrix product on ARM 64, 20% speedup 2018-11-23 10:25:19 +01:00
Gael Guennebaud
4b2cebade8 Workaround weird MSVC bug 2018-11-21 15:53:37 +01:00
Deven Desai
e7e6809e6b ROCm/HIP specfic fixes + updates 
1. Eigen/src/Core/arch/GPU/Half.h

   Updating the HIPCC implementation half so that it can declared as a __shared__ variable


2. Eigen/src/Core/util/Macros.h, Eigen/src/Core/util/Memory.h

   introducing a EIGEN_USE_STD(func) macro that calls
   - std::func be default
   - ::func when eigen is being compiled with HIPCC

   This change was requested in the previous HIP PR
   (https://bitbucket.org/eigen/eigen/pull-requests/518/pr-with-hip-specific-fixes-for-the-eigen/diff)


3. unsupported/Eigen/CXX11/src/Tensor/TensorDeviceThreadPool.h

   Removing EIGEN_DEVICE_FUNC attribute from pure virtual methods as it is not supported by HIPCC


4. unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h

   Disabling the template specializations of InnerMostDimReducer as they run into HIPCC link errors
 2018-11-19 18:13:59 +00:00
Gael Guennebaud
6a510fe69c Make MaxPacketSize a true upper bound, even for fixed-size inputs 2018-11-16 11:25:32 +01:00
Mark D Ryan
670d56441c PR 544: Set requestedAlignment correctly for SliceVectorizedTraversals 
Commit aa110e681b
 optimised the multiplication of small dyanmically
sized matrices by restricting the packet size to a maximum of 4, increasing
the chances that SIMD instructions are used in the computation.  However, it
introduced a mismatch between the packet size and the requestedAlignment.  This
mismatch can lead to crashes when the destination is not aligned.  This patch
fixes the issue by ensuring that the AssignmentTraits are correctly computed
when using a restricted packet size.
* * *
Bind LinearPacketType to MaxPacketSize

This commit applies any packet size limit specified when instantiating
copy_using_evaluator_traits to the LinearPacketType, providing that the
size of the destination is not known at compile time.
* * *
Add unit test for restricted packet assignment

A new unit test is added to check that multiplication of small dynamically
sized matrices works correctly when the packet size is restricted to 4 and
the destination is unaligned.
 2018-11-13 16:15:08 +01:00
Nikolaus Demmel
3dc0845046 Fix typo in comment on EIGEN_MAX_STATIC_ALIGN_BYTES 2018-11-14 18:11:30 +01:00
Gael Guennebaud
7fddc6a51f typo 2018-11-14 14:43:18 +01:00
Gael Guennebaud
449f948b2a help doxygen linking to DenseBase::NulllaryExpr 2018-11-14 14:42:59 +01:00
luz.paz"
f67b19a884 [PATCH 1/2] Misc. typos 
From 68d431b4c14ad60a778ee93c1f59ecc4b931950e Mon Sep 17 00:00:00 2001
Found via `codespell -q 3 -I ../eigen-word-whitelist.txt` where the whitelists consists of:
```
als
ans
cas
dum
lastr
lowd
nd
overfl
pres
preverse
substraction
te
uint
whch
```
---
 CMakeLists.txt                                | 26 +++++++++----------
 Eigen/src/Core/GenericPacketMath.h            |  2 +-
 Eigen/src/SparseLU/SparseLU.h                 |  2 +-
 bench/bench_norm.cpp                          |  2 +-
 doc/HiPerformance.dox                         |  2 +-
 doc/QuickStartGuide.dox                       |  2 +-
 .../Eigen/CXX11/src/Tensor/TensorChipping.h   |  6 ++---
 .../Eigen/CXX11/src/Tensor/TensorDeviceGpu.h  |  2 +-
 .../src/Tensor/TensorForwardDeclarations.h    |  4 +--
 .../src/Tensor/TensorGpuHipCudaDefines.h      |  2 +-
 .../Eigen/CXX11/src/Tensor/TensorReduction.h  |  2 +-
 .../CXX11/src/Tensor/TensorReductionGpu.h     |  2 +-
 .../test/cxx11_tensor_concatenation.cpp       |  2 +-
 unsupported/test/cxx11_tensor_executor.cpp    |  2 +-
 14 files changed, 29 insertions(+), 29 deletions(-)
 2018-09-18 04:15:01 -04:00
Rasmus Munk Larsen
77b447c24e Add optimized version of logistic function for float. As an example, this is about 50% faster than the existing version on Haswell using AVX. 2018-11-12 13:42:24 -08:00
Gael Guennebaud
0105146915 Fix warning in c++03 2018-11-10 09:11:38 +01:00
Gael Guennebaud
784a3f13cf bug #1619: fix mixing of const and non-const generic iterators 2018-11-09 21:45:10 +01:00
Gael Guennebaud
db9a9a12ba bug #1619: make const and non-const iterators compatible 2018-11-09 16:49:19 +01:00
Gael Guennebaud
bd9a00718f Let doxygen sees lastN 2018-11-09 11:35:48 +01:00
Gael Guennebaud
a368848473 Recent xcode versions does support EIGEN_HAS_STATIC_ARRAY_TEMPLATE 2018-11-09 10:33:17 +01:00
Gael Guennebaud
f62a0f69c6 Fix max-size in indexed-view 2018-11-08 18:40:22 +01:00
Gael Guennebaud
bf495859ff Merged in glchaves/eigen (pull request PR-539) 
Vectorize row-by-row gebp loop iterations on 16 packets as well
 2018-11-07 07:21:15 +00:00
Gustavo Lima Chaves
4ad359237a Vectorize row-by-row gebp loop iterations on 16 packets as well 
Signed-off-by: Gustavo Lima Chaves <gustavo.lima.chaves@intel.com>
Signed-off-by: Mark D. Ryan <mark.d.ryan@intel.com>
 2018-11-06 10:48:42 -08:00
Matthieu Vigne
8d7a73e48e bug #1617: Fix SolveTriangular.solveInPlace crashing for empty matrix. 
This made FullPivLU.kernel() crash when used on the zero matrix.
Add unit test for FullPivLU.kernel() on the zero matrix.
 2018-10-31 20:28:18 +01:00
Christoph Hertzberg
66b28e290d bug #1618: Use different power-of-2 check to avoid MSVC warning 2018-11-01 13:23:19 +01:00
Christian von Schultz
4a40b3785d Collapsed revision (based on pull request PR-325) 
* Support compiling without IO streams

Add the preprocessor definition EIGEN_NO_IO which, if defined,
disables all use of the IO streams part of the standard library.
 2018-10-22 21:14:40 +02:00
Rasmus Munk Larsen
14054e217f Do not rely on the compiler generating __device__ functions for constexpr in Cuda (via EIGEN_CONSTEXPR_ARE_DEVICE_FUNC. This breaks several target in the TensorFlow Cuda build, e.g., 
INFO: From Compiling tensorflow/core/kernels/maxpooling_op_gpu.cu.cc:
/b/f/w/run/external/eigen_archive/Eigen/src/Core/arch/GPU/Half.h(197): error: calling a __host__ function("std::equal_to<float> ::operator () const") from a __global__ function("tensorflow::_NV_ANON_NAMESPACE::MaxPoolGradBackwardNoMaskNHWC< ::Eigen::half> ") is not allowed

/b/f/w/run/external/eigen_archive/Eigen/src/Core/arch/GPU/Half.h(197): error: identifier "std::equal_to<float> ::operator () const" is undefined in device code"

/b/f/w/run/external/eigen_archive/Eigen/src/Core/arch/GPU/Half.h(197): error: calling a __host__ function("std::equal_to<float> ::operator () const") from a __global__ function("tensorflow::_NV_ANON_NAMESPACE::MaxPoolGradBackwardNoMaskNCHW< ::Eigen::half> ") is not allowed

/b/f/w/run/external/eigen_archive/Eigen/src/Core/arch/GPU/Half.h(197): error: identifier "std::equal_to<float> ::operator () const" is undefined in device code

4 errors detected in the compilation of "/tmp/tmpxft_00000011_00000000-6_maxpooling_op_gpu.cu.cpp1.ii".
ERROR: /tmpfs/tensor_flow/tensorflow/core/kernels/BUILD:3753:1: output 'tensorflow/core/kernels/_objs/pooling_ops_gpu/maxpooling_op_gpu.cu.pic.o' was not created
ERROR: /tmpfs/tensor_flow/tensorflow/core/kernels/BUILD:3753:1: Couldn't build file tensorflow/core/kernels/_objs/pooling_ops_gpu/maxpooling_op_gpu.cu.pic.o: not all outputs were created or valid
 2018-10-22 16:18:24 -07:00
Rasmus Munk Larsen
9caafca550 Merged in rmlarsen/eigen (pull request PR-532) 
Only set EIGEN_CONSTEXPR_ARE_DEVICE_FUNC for clang++ if cxx_relaxed_constexpr is available.
 2018-10-19 21:37:14 +00:00
Christoph Hertzberg
449ff74672 Fix most Doxygen warnings. Also add links to stable documentation from unsupported modules (by using the corresponding Doxytags file). 
Manually grafted from d107a371c6
 2018-10-19 21:10:28 +02:00
Rasmus Munk Larsen
d8f285852b Only set EIGEN_CONSTEXPR_ARE_DEVICE_FUNC for clang++ if cxx_relaxed_constexpr is available. 2018-10-18 16:55:02 -07:00
Gael Guennebaud
0f780bb0b4 Fix float-to-double warning 2018-10-16 09:19:45 +02:00
Gael Guennebaud
a39e0f7438 bug #1612: fix regression in "outer-vectorization" of partial reductions for PacketSize==1 (aka complex<double>) 2018-10-16 01:04:25 +02:00
Gael Guennebaud
d2d570c116 Remove useless (and broken) resize 2018-10-16 00:42:48 +02:00
Gael Guennebaud
f0fb95135d Iterative solvers: unify and fix handling of multiple rhs. 
m_info was not properly computed and the logic was repeated in several places.
 2018-10-15 23:47:46 +02:00
Gael Guennebaud
3a33db4de5 merge 2018-10-15 09:22:27 +02:00
Mark D Ryan
aa110e681b PR 526: Speed up multiplication of small, dynamically sized matrices 
The Packet16f, Packet8f and Packet8d types are too large to use with dynamically
sized matrices typically processed by the SliceVectorizedTraversal specialization of
the dense_assignment_loop.  Using these types is likely to lead to little or no
vectorization.  Significant slowdown in the multiplication of these small matrices can
be observed when building with AVX and AVX512 enabled.

This patch introduces a new dense_assignment_kernel that is used when
computing small products whose operands have dynamic dimensions.  It ensures that the
PacketSize used is no larger than 4, thereby increasing the chance that vectorized
instructions will be used when computing the product.

I tested all 969 possible combinations of M, K, and N that are handled by the
dense_assignment_loop on x86 builds.  Although a few combinations are slowed down
by this patch they are far outnumbered by the cases that are sped up, as the
following results demonstrate.


Disabling Packed8d on AVX512 builds:

Total Cases:             969
Better:                  511
Worse:                   85
Same:                    373
Max Improvement:         169.00% (4 8 6)
Max Degradation:         36.50% (8 5 3)
Median Improvement:      35.46%
Median Degradation:      17.41%
Total FLOPs Improvement: 19.42%


Disabling Packet16f and Packed8f on AVX512 builds:

Total Cases:             969
Better:                  658
Worse:                   5
Same:                    306
Max Improvement:         214.05% (8 6 5)
Max Degradation:         22.26% (16 2 1)
Median Improvement:      60.05%
Median Degradation:      13.32%
Total FLOPs Improvement: 59.58%


Disabling Packed8f on AVX builds:

Total Cases:             969
Better:                  663
Worse:                   96
Same:                    210
Max Improvement:         155.29% (4 10 5)
Max Degradation:         35.12% (8 3 2)
Median Improvement:      34.28%
Median Degradation:      15.05%
Total FLOPs Improvement: 26.02%
 2018-10-12 15:20:21 +02:00
Eugene Zhulenev
d9392f9e55 Fix code format 2018-11-02 14:51:35 -07:00
Eugene Zhulenev
118520f04a Workaround nbcc+msvc compiler bug 2018-11-02 14:48:28 -07:00
Christoph Hertzberg
24dc076519 Explicitly convert 0 to Scalar for custom types 2018-10-12 10:22:19 +02:00
Gael Guennebaud
43633fbaba Fix warning with AVX512f 2018-10-11 10:13:48 +02:00
Gael Guennebaud
97e2c808e9 Fix avx512 plog(NaN) to return NaN instead of +inf 2018-10-11 10:13:13 +02:00
Gael Guennebaud
b3f66d29a5 Enable avx512 plog with clang 2018-10-11 10:12:21 +02:00
Gael Guennebaud
f0aa7e40fc Fix regression in changeset 5335659c47 2018-10-10 23:47:30 +02:00
Gael Guennebaud
ce243ee45b bug #520: add diagmat +/- diagmat operators. 2018-10-10 23:38:22 +02:00
Gael Guennebaud
5335659c47 Merged in ezhulenev/eigen-02 (pull request PR-525) 
Fix bug in partial reduction of expressions requiring evaluation
 2018-10-10 20:59:00 +00:00
Gael Guennebaud
eec0dfd688 bug #632: add specializations for res ?= dense +/- sparse and res ?= sparse +/- dense. 
They are rewritten as two compound assignment to by-pass hybrid dense-sparse iterator.
 2018-10-10 22:50:15 +02:00
Eugene Zhulenev
8e6dc2c81d Fix bug in partial reduction of expressions requiring evaluation 2018-10-10 13:23:52 -07:00
Eugene Zhulenev
2bf1a31d81 Use void type if stl-style iterators are not supported 2018-10-10 10:31:40 -07:00
Rasmus Munk Larsen
e8918743c1 Merged in ezhulenev/eigen-01 (pull request PR-523) 
Compile time detection for unimplemented stl-style iterators
 2018-10-09 23:42:01 +00:00
Eugene Zhulenev
c0ca8a9fa3 Compile time detection for unimplemented stl-style iterators 2018-10-09 15:28:23 -07:00
Gael Guennebaud
1dd1f8e454 bug #65: add vectorization of partial reductions along the outer-dimension, for instance: colmajor_mat.rowwise().mean() 2018-10-09 23:36:50 +02:00
Gael Guennebaud
bfa2a81a50 Make redux_vec_unroller more flexible regarding packet-type 2018-10-09 23:30:41 +02:00
Christoph Hertzberg
f6359ad795 Small Doxygen fixes 2018-10-09 19:33:35 +02:00
Gael Guennebaud
7a882c05ab Fix compilation on CUDA 2018-10-09 17:02:16 +02:00
Gael Guennebaud
e00487f7d2 bug #1603: add parenthesis around ternary operator in function body as well as a harmless attempt to make MSVC happy. 2018-10-08 22:27:04 +02:00
Gael Guennebaud
649d4758a6 merge 2018-10-08 17:35:18 +02:00
Gael Guennebaud
774bb9d6f7 fix a doxygen issue 2018-10-08 09:30:15 +02:00
Gael Guennebaud
bcb7c66b53 Workaround gcc's alloc-size-larger-than= warning 2018-10-07 21:55:59 +02:00
Gael Guennebaud
6512c5e136 Implement a better workaround for GCC's bug #87544 2018-10-07 15:00:05 +02:00
Gael Guennebaud
409132bb81 Workaround gcc bug making it trigger an invalid warning 2018-10-07 09:23:15 +02:00
Gael Guennebaud
c6a1ab4036 Workaround MSVC compilation issue 2018-10-06 13:49:17 +02:00
Gael Guennebaud
e21766c6f5 Clarify doc of rowwise/colwise/vectorwise. 2018-10-05 23:12:09 +02:00
Gael Guennebaud
d92f004ab7 Simplify API by removing allCols/allRows and reusing rowwise/colwise to define iterators over rows/columns 2018-10-05 23:11:21 +02:00
Gael Guennebaud
3e64b1fc86 Move iterators to internal, improve doc, make unit test c++03 friendly 2018-10-03 15:13:15 +02:00
Gael Guennebaud
2b2b4d0580 fix unused warning 2018-10-03 14:16:21 +02:00
Gael Guennebaud
5f26f57598 Change the logic of A.reshaped<Order>() to be a simple alias to A.reshaped<Order>(AutoSize,fix<1>). 
This means that now AutoOrder is allowed, and it always return a column-vector.
 2018-10-03 11:41:47 +02:00
Gael Guennebaud
0481900e25 Add pointer-based iterator for direct-access expressions 2018-10-02 23:44:36 +02:00
Gael Guennebaud
8c38528168 Factorize RowsProxy/ColsProxy and related iterators using subVector<>(Index) 2018-10-02 14:03:26 +02:00
Gael Guennebaud
12487531ce Add templated subVector<Vertical/Horizonal>(Index) aliases to col/row(Index) methods (plus subVectors<>() to retrieve the number of rows/columns) 2018-10-02 14:02:34 +02:00
Gael Guennebaud
37e29fc893 Use Index instead of ptrdiff_t or int, fix random-accessors. 2018-10-02 13:29:32 +02:00
Gael Guennebaud
de2efbc43c bug #1605: workaround ABI issue with vector types (aka __m128) versus scalar types (aka float) 2018-10-01 23:45:55 +02:00
Gael Guennebaud
b0c66adfb1 bug #231: initial implementation of STL iterators for dense expressions 2018-10-01 23:21:37 +02:00
Christoph Hertzberg
564ca71e39 Merged in deven-amd/eigen/HIP_fixes (pull request PR-518) 
PR with HIP specific fixes (for the eigen nightly regression failures in HIP mode)
 2018-10-01 16:51:04 +00:00
Deven Desai
94898488a6 This commit contains the following (HIP specific) updates: 
- unsupported/Eigen/CXX11/src/Tensor/TensorReductionGpu.h
  Changing "pass-by-reference" argument to be "pass-by-value" instead
  (in a  __global__ function decl).
  "pass-by-reference" arguments to __global__ functions are unwise,
  and will be explicitly flagged as errors by the newer versions of HIP.

- Eigen/src/Core/util/Memory.h
- unsupported/Eigen/CXX11/src/Tensor/TensorContraction.h
  Changes introduced in recent commits breaks the HIP compile.
  Adding EIGEN_DEVICE_FUNC attribute to some functions and
  calling ::malloc/free instead of the corresponding std:: versions
  to get the HIP compile working again

- unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h
  Change introduced a recent commit breaks the HIP compile
  (link stage errors out due to failure to inline a function).
  Disabling the recently introduced code (only for HIP compile), to get
  the eigen nightly testing going again.
  Will submit another PR once we have te proper fix.

- Eigen/src/Core/util/ConfigureVectorization.h
  Enabling GPU VECTOR support when HIP compiler is in use
  (for both the host and device compile phases)
 2018-10-01 14:28:37 +00:00
Gael Guennebaud
af3ad4b513 oops, I've been too fast in previous copy/paste 2018-09-27 09:28:57 +02:00
Gael Guennebaud
24b163a877 #pragma GCC diagnostic push/pop is not supported prioro to gcc 4.6 2018-09-27 09:23:54 +02:00
Gael Guennebaud
41c3a2ffc1 Fix documentation of reshape to vectors. 2018-09-25 16:35:44 +02:00
Christoph Hertzberg
2c083ace3e Provide EIGEN_OVERRIDE and EIGEN_FINAL macros to mark virtual function overrides 2018-09-24 18:01:17 +02:00
Gael Guennebaud
626942d9dd fix alignment issue in ploaddup for AVX512 2018-09-28 16:57:32 +02:00
Gael Guennebaud
84a1101b36 Merge with default. 2018-09-23 21:52:58 +02:00
Gael Guennebaud
795e12393b Fix logic in diagonal*dense product in a corner case. 
The problem was for: diag(1x1) * mat(1,n)
 2018-09-22 16:44:33 +02:00
Gael Guennebaud
bac36d0996 Demangle Travseral and Unrolling in Redux 2018-09-21 23:03:45 +02:00
Gael Guennebaud
1bf12880ae Add reshaped<>() shortcuts when returning vectors and remove the reshaping version of operator()(all) 2018-09-21 16:50:04 +02:00
Gael Guennebaud
371068992a Add more debug output 2018-09-21 14:32:39 +02:00
Gael Guennebaud
b00e48a867 Improve slice-vectorization logic for redux (significant speed-up for reduxion of blocks) 2018-09-21 13:45:56 +02:00
Gael Guennebaud
a488d59787 merge with default Eigen 2018-09-21 11:51:49 +02:00
Gael Guennebaud
47720e7970 Doc fixes 2018-09-21 11:48:22 +02:00
Gael Guennebaud
3ec2985914 Merged indexing cleanup (pull request PR-506) 2018-09-21 09:36:05 +00:00
Gael Guennebaud
651e5d4866 Fix EIGEN_MAKE_ALIGNED_OPERATOR_NEW_IF_VECTORIZABLE_FIXED_SIZE for AVX512 or AVX with malloc aligned on 8 bytes only. 
This change also make it future proof for AVX1024
 2018-09-21 10:33:22 +02:00
Gael Guennebaud
f0ef3467de Fix doc 2018-09-20 22:57:28 +02:00
Gael Guennebaud
617f75f117 Add indexing namespace 2018-09-20 22:57:10 +02:00
Gael Guennebaud
0c56d22e2e Fix shadowing 2018-09-20 22:56:21 +02:00
Gael Guennebaud
9419f506d0 Fix regression introduced by the previous fix for AVX512. 
It brokes the complex-complex case on SSE.
 2018-09-20 17:32:34 +02:00
Gael Guennebaud
e38d1ab4d1 Workaround increases required alignment warning 2018-09-20 17:07:33 +02:00
Gael Guennebaud
71496b0e25 Fix gebp kernel for real+complex in case only reals are vectorized (e.g., AVX512). 
This commit also removes "half-packet" from data-mappers: it was not used and conceptually broken anyways.
 2018-09-20 17:01:24 +02:00
Gael Guennebaud
5a30eed17e Fix warnings in AVX512 2018-09-20 16:58:51 +02:00
Gael Guennebaud
c3a19527a2 Fix doc wrt previous change 2018-09-19 11:49:26 +02:00
Gael Guennebaud
dfa8439e4d Update reshaped API to use RowMajor/ColMajor directly as integral values instead of introducing RowOrder/ColOrder types. 
The API changed from A.respahed(rows,cols,RowOrder) to A.template reshaped<RowOrder>(rows,cols).
 2018-09-19 11:49:26 +02:00
Gael Guennebaud
297ca62319 ease transition by adding placeholders::all/last/and as deprecated 2018-09-17 16:24:52 +02:00
Gael Guennebaud
2014c7ae28 Move all, last, end from Eigen::placeholders namespace to Eigen::, and rename end to lastp1 to avoid conflicts with std::end. 2018-09-15 14:35:10 +02:00
Gael Guennebaud
82772e8d9d Rename Symbolic namespace to symbolic to be consistent with numext namespace 2018-09-15 14:16:20 +02:00
Gael Guennebaud
3e8188fc77 bug #1600: initialize m_info to InvalidInput by default, even though m_info is not accessible until it has been initialized (assert) 2018-09-18 21:24:48 +02:00
Christoph Hertzberg
d7378aae8e Provide EIGEN_ALIGNOF macro, and give handmade_aligned_malloc the possibility for alignments larger than the standard alignment. 2018-09-14 20:17:47 +02:00
Gael Guennebaud
1141bcf794 Fix conjugate-gradient for very small rhs 2018-09-13 23:53:28 +02:00
Deven Desai
c64fe9ea1f Updates to fix HIP-clang specific compile errors. 
Compiling the eigen unittests with hip-clang (HIP with clang as the underlying compiler instead of hcc or nvcc), results in compile errors. The changes in this commit fix those compile errors. The main change is to convert a few instances of "__device__" to "EIGEN_DEVICE_FUNC"
 2018-08-30 20:22:16 +00:00
Gael Guennebaud
5927eef612 Enable std::result_of for msvc 2015 and later 2018-09-13 09:44:46 +02:00
Christoph Hertzberg
3adece4827 Fix misleading indentation of errorCode and make it loop-local 2018-09-12 14:41:38 +02:00
Christoph Hertzberg
7e9c9fbb2d Disable type-limits warnings for g++ < 4.8 2018-09-12 14:40:39 +02:00
Justin Carpentier
4827bec776 LLT: correct doc and add missing reference for the return type of rankUpdate 
---
 Eigen/src/Cholesky/LLT.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
 2018-09-11 09:33:21 +02:00
luz.paz"
43fd42a33b Fix doxy and misc. typos 
Found via `codespell -q 3 -I ../eigen-word-whitelist.txt`
---
 Eigen/src/Core/ProductEvaluators.h |  4 ++--
 Eigen/src/Core/arch/GPU/Half.h     |  2 +-
 Eigen/src/Core/util/Memory.h       |  2 +-
 Eigen/src/Geometry/Hyperplane.h    |  2 +-
 Eigen/src/Geometry/Transform.h     |  2 +-
 Eigen/src/Geometry/Translation.h   | 12 ++++++------
 doc/PreprocessorDirectives.dox     |  2 +-
 doc/TutorialGeometry.dox           |  2 +-
 test/boostmultiprec.cpp            |  2 +-
 test/triangular.cpp                |  2 +-
 10 files changed, 16 insertions(+), 16 deletions(-)
 2018-08-01 21:34:47 -04:00
Jiandong Ruan
6dcd2642aa bug #1526 - CUDA compilation fails on CUDA 9.x SDK when arch is set to compute_60 and/or above 2018-09-08 12:05:33 -07:00
Alexey Frunze
ec38f07b79 bug #1595: Don't use C++11's std::isnan() in MIPS/MSA packet math. 
This removes reliance on C++11 and improves generated code.
 2018-09-06 15:40:09 -07:00
cgs1019
c6066ac411 Make param name and docs constistent for JacobiRotation::makeGivens 
Previously the rendered math in the doc string called the optional return value
'r', while the actual parameter and the doc string text referred to the
parameter as 'z'. This changeset renames all the z's to r's to match the math.
 2018-09-06 11:04:17 -04:00
Christoph Hertzberg
ddbc564386 Fixed a few more shadowing warnings when compiling with g++ (and c++03) 2018-08-30 16:33:03 +02:00
Mehdi Goli
7ec8b40ad9 Collapsed revision 
* Separating SYCL math function.
* Converting function overload to function specialisation.
* Applying the suggested design.
 2018-08-28 14:20:48 +01:00
Christoph Hertzberg
73ca600bca Fix numerous shadow-warnings for GCC<=4.8 2018-08-28 18:32:39 +02:00
Christoph Hertzberg
ef4d79fed8 Disable/ReenableStupidWarnings did not work properly, when included recursively 2018-08-28 18:26:22 +02:00
Gael Guennebaud
befaf83f5f bug #1590: fix collision with some system headers defining the macro FP32 2018-08-28 13:21:28 +02:00
Christoph Hertzberg
42f3ee4fb8 Old gcc versions have problems with recursive #pragma GCC diagnostic push/pop 
Workaround: Don't include "DisableStupidWarnings.h" before including other main-headers
 2018-08-28 11:44:15 +02:00
Alexey Frunze
050bcf6126 bug #1584: Improve random (avoid undefined behavior). 2018-08-08 20:19:32 -07:00
Christoph Hertzberg
ad4a08fb68 Use Intel cast intrinsics, since MSVC does not allow direct casting. 
Reported by David Winkler.
 2018-08-24 19:04:33 +02:00
Christoph Hertzberg
41f1cc67b8 Assertion depended on a not yet initialized value 2018-08-17 16:42:53 +02:00
Christoph Hertzberg
595cae9b09 Silence logical-op-parentheses warning 2018-08-17 16:30:32 +02:00
Justin Carpentier
eabc7a4031 PR 465: Fix issue in RowMajor assignment in plain_matrix_type_row_major::type 
The type should be RowMajor
 2018-08-10 14:30:06 +02:00
Rasmus Munk Larsen
c49e93440f SuiteSparse defines the macro SuiteSparse_long to control what type is used for 64bit integers. The default value of this macro on non-MSVC platforms is long and __int64 on MSVC. CholmodSupport defaults to using long for the long variants of CHOLMOD functions. This creates problems when SuiteSparse_long is different than long. So the correct thing to do here is 
to use SuiteSparse_long as the type instead of long.
 2018-08-13 15:53:31 -07:00
Gael Guennebaud
3ec60215df Merged in rmlarsen/eigen2 (pull request PR-466) 
Move sigmoid functor to core and rename it to 'logistic'.
 2018-08-13 21:28:20 +00:00
Rasmus Munk Larsen
d6e283ba96 sigmoid -> logistic 2018-08-13 11:14:50 -07:00
Rasmus Munk Larsen
bfc5091dd5 Cast to diagonalSize to RealScalar instead Scalar. 2018-08-09 14:46:17 -07:00
Rasmus Munk Larsen
8603d80029 Cast diagonalSize() to Scalar before multiplication. Without this, automatic differentiation in Ceres breaks because Scalar is a custom type that does not support multiplication by Index. 2018-08-09 11:09:10 -07:00
Mehdi Goli
67711eaa31 Fixing typo. 2018-08-08 11:38:10 +01:00
Mehdi Goli
22031ab59a Adding EIGEN_UNROLL_LOOP macro. 2018-08-08 11:07:27 +01:00
Rasmus Munk Larsen
fa68342ef8 Move sigmoid functor to core. 2018-08-03 17:31:23 -07:00
Rasmus Munk Larsen
7f8b53fd0e bug #1580: Fix cuda clang build. STL is not supported, so std::equal_to and std::not_equal breaks compilation. 
Update the definition of EIGEN_CONSTEXPR_ARE_DEVICE_FUNC to exclude clang.
See also PR 450.
 2018-08-01 12:36:24 -07:00
Mehdi Goli
01358300d5 Creating separate SYCL required PR for uncontroversial files. 2018-08-03 16:59:15 +01:00
Gael Guennebaud
62169419ab Fix two regressions introduced in previous merges: bad usage of EIGEN_HAS_VARIADIC_TEMPLATES and linking issue. 2018-08-01 23:35:34 +02:00
Benoit Steiner
17221115c9 Merged in codeplaysoftware/eigen-upstream-pure/eigen_variadic_assert (pull request PR-447) 
Adding variadic version of assert which can take a parameter pack as its input.
 2018-08-01 16:41:54 +00:00
Mehdi Goli
af96018b49 Using the suggested modification. 2018-08-01 16:04:44 +01:00
Mehdi Goli
c84509d7cc Adding new arch/SYCL headers, used for SYCL vectorization. 2018-08-01 12:40:54 +01:00
Mehdi Goli
3a197a60e6 variadic version of assert which can take a parameter pack as its input. 2018-08-01 12:19:14 +01:00
Alexey Frunze
7b91c11207 bug #1578: Improve prefetching in matrix multiplication on MIPS. 2018-07-24 18:36:44 -07:00
Mark D Ryan
bc615e4585 Re-enable FMA for fast sqrt functions 2018-07-30 13:21:00 +02:00
Mark D Ryan
e79c5149bf Fix AVX512 implementations of psqrt 
This commit fixes the AVX512 implementations of psqrt in the same
way that 3ed67cb0bb
 fixed the AVX2 version of this function.  The
AVX512 versions of psqrt incorrectly return -0.0 for negative
values, instead of NaN.  Fixing the issues requires adding
some additional instructions that slow down the algorithms.  A
similar test to the one used in 3ed67cb0bb
 shows that the
corrected Packet16f code runs at 73% of the speed of the existing code,
while the corrected Packed8d function runs at 68% of the original.
 2018-06-25 05:05:02 -07:00
Rasmus Munk Larsen
2ebcb911b2 Add pcast packet op for NEON. 2018-07-26 14:28:48 -07:00
Christoph Hertzberg
fd4fe7cbc5 Fixed issue which made documentation not getting built anymore 2018-07-24 22:56:15 +02:00
Gael Guennebaud
4ca3e48f42 fix typo 2018-07-23 16:51:57 +02:00
Gael Guennebaud
c747cde69a Add lastN shorcuts to seq/seqN. 2018-07-23 16:20:25 +02:00
Eugene Zhulenev
2bf864f1eb Disable type traits for stdlibc++ <= 4.9.3 2018-07-20 10:11:44 -07:00
Gael Guennebaud
509a5fa77f Fix IsRelocatable without C++11 2018-07-19 18:47:38 +02:00
Gael Guennebaud
2ca2592009 Fix determination of EIGEN_HAS_TYPE_TRAITS 2018-07-19 18:47:18 +02:00
Gael Guennebaud
5e5987996f Fix stupid error in Quaternion move ctor 2018-07-19 18:33:53 +02:00
Alexey Frunze
1f523e7304 Add MIPS changes missing from previous merge. 2018-07-18 12:27:50 -07:00
Eugene Zhulenev
086ded5c85 Disable type traits for GCC < 5.1.0 2018-07-18 16:32:55 -07:00
Gael Guennebaud
863580fe88 bug #1432: fix conservativeResize for non-relocatable scalar types. For those we need to by-pass realloc routines and fall-back to allocate as new - copy - delete. The remaining problem is that we don't have any mechanism to accurately determine whether a type is relocatable or not, so currently let's be super conservative using either RequireInitialization or std::is_trivially_copyable 2018-07-18 23:33:07 +02:00
Gael Guennebaud
a503fc8725 bug #1575: fix regression introduced in bug #1573 patch. Move ctor/assignment should not be defaulted. 2018-07-18 23:26:13 +02:00
Gael Guennebaud
308725c3c9 More clearly disable the inclusion of src/Core/arch/CUDA/Complex.h without CUDA 2018-07-18 13:51:36 +02:00
Deven Desai
f124f07965 applying EIGEN_DECLARE_TEST to *gpu* tests 
Also, a few minor fixes for GPU tests running in HIP mode.

1. Adding an include for hip/hip_runtime.h in the Macros.h file
   For HIP __host__ and __device__ are macros which are defined in hip headers.
   Their definitions need to be included before their use in the file.

2. Fixing the compile failure in TensorContractionGpu introduced by the commit to
   "Fuse computations into the Tensor contractions using output kernel"

3. Fixing a HIP/clang specific compile error by making the struct-member assignment explicit
 2018-07-17 14:16:48 -04:00
Gael Guennebaud
2b2cd85694 bug #1573: add noexcept move constructor and move assignment operator to Quaternion 2018-07-17 11:11:33 +02:00
Gael Guennebaud
5539587b1f Some warning fixes 2018-07-17 10:29:12 +02:00
Gael Guennebaud
40797dbea3 bug #1572: use c++11 atomic instead of volatile if c++11 is available, and disable multi-threaded GEMM on non-x86 without c++11. 2018-07-17 00:11:20 +02:00
Gael Guennebaud
a87cff20df Fix GeneralizedEigenSolver when requesting for eigenvalues only. 2018-07-14 09:38:49 +02:00
Rasmus Munk Larsen
4a3952fd55 Relax the condition to not only work on Android. 2018-07-13 11:24:07 -07:00
Rasmus Munk Larsen
02a9443db9 Clang produces incorrect Thumb2 assembler when using alloca. 
Don't define EIGEN_ALLOCA when generating Thumb with clang.
 2018-07-13 11:03:04 -07:00
Gael Guennebaud
20991c3203 bug #1571: fix is_convertible<from,to> with "from" a reference. 2018-07-13 17:47:28 +02:00
Gael Guennebaud
86d9c0255c Forward declaring std::array does not work with all std libs, so let's just include <array> 2018-07-13 13:06:44 +02:00
Alexey Frunze
3875fb05aa Add support for MIPS SIMD (MSA) 2018-07-06 16:04:30 -07:00
Gael Guennebaud
5c73c9223a Fix shadowing typedefs 2018-07-12 17:01:07 +02:00
Gael Guennebaud
98728312c8 Fix compilation regarding std::array 2018-07-12 17:00:37 +02:00
Gael Guennebaud
eb3d8f68bb fix unused warning 2018-07-12 16:59:47 +02:00
Gael Guennebaud
006e18e52b Cleanup the mess in Eigen/Core by moving CUDA/HIP stuff at more appropriate places (Macros.h), 
and alignment/vectorization logic is now in util/ConfigureVectorization.h
 2018-07-12 16:57:41 +02:00
Julian Kent
6d451cf2b6 Add missing consts for rows and cols functions in SparseLU 2018-02-10 13:44:05 +01:00
Gael Guennebaud
8bdb214fd0 remove double ;; 2018-07-12 11:17:53 +02:00
Gael Guennebaud
a9060378d3 bug #1570: fix warning 2018-07-12 11:07:09 +02:00
Gael Guennebaud
da0c604078 Merged in deven-amd/eigen (pull request PR-402) 
Adding support for using Eigen in HIP kernels.
 2018-07-12 08:07:16 +00:00
Gael Guennebaud
a4ea611ca7 Remove useless specialization thanks to is_convertible being more robust. 2018-07-12 09:59:44 +02:00
Gael Guennebaud
8ef267ccbd spellcheck 2018-07-12 09:58:29 +02:00
Gael Guennebaud
21cf4a1a8b Make is_convertible more robust and conformant to std::is_convertible 2018-07-12 09:57:19 +02:00
Gael Guennebaud
8a5955a052 Optimize the product of a householder-sequence with the identity, and optimize the evaluation of a HouseholderSequence to a dense matrix using faster blocked product. 2018-07-11 17:16:50 +02:00
Gael Guennebaud
d193cc87f4 Fix regression in 9357838f94 2018-07-11 17:09:23 +02:00
Gael Guennebaud
fb33687736 Fix double ;; 2018-07-11 17:08:30 +02:00
Deven Desai
876f392c39 Updates corresponding to the latest round of PR feedback 
The major changes are

1. Moving CUDA/PacketMath.h to GPU/PacketMath.h
2. Moving CUDA/MathFunctions.h to GPU/MathFunction.h
3. Moving CUDA/CudaSpecialFunctions.h to GPU/GpuSpecialFunctions.h
    The above three changes effectively enable the Eigen "Packet" layer for the HIP platform

4. Merging the "hip_basic" and "cuda_basic" unit tests into one ("gpu_basic")
5. Updating the "EIGEN_DEVICE_FUNC" marking in some places

The change has been tested on the HIP and CUDA platforms.
 2018-07-11 10:39:54 -04:00
Deven Desai
471cfe5ff7 renaming CUDA* to GPU* for some header files 2018-07-11 09:22:04 -04:00
Deven Desai
38807a2575 merging updates from upstream 2018-07-11 09:17:33 -04:00
Gael Guennebaud
f00d08cc0a Optimize extraction of Q in SparseQR by exploiting the structure of the identity matrix. 2018-07-11 14:01:47 +02:00
Gael Guennebaud
1625476091 Add internall::is_identity compile-time helper 2018-07-11 14:00:24 +02:00
Gael Guennebaud
fe723d6129 Fix conversion warning 2018-07-10 09:10:32 +02:00
Gael Guennebaud
9357838f94 bug #1543: improve linear indexing for general block expressions 2018-07-10 09:10:15 +02:00
Gael Guennebaud
de9e31a06d Introduce the macro ei_declare_local_nested_eval to help allocating on the stack local temporaries via alloca, and let outer-products makes a good use of it. 
If successful, we should use it everywhere nested_eval is used to declare local dense temporaries.
 2018-07-09 15:41:14 +02:00
Gael Guennebaud
ec323b7e66 Skip null numerators in triangular-vector-solve (as in BLAS TRSV). 2018-07-09 11:13:19 +02:00
Gael Guennebaud
359dd77ec3 Fix legitimate "declaration shadows a typedef" warning 2018-07-09 11:03:39 +02:00
Mark D Ryan
90a53ca6fd Fix the Packet16h version of ptranspose 
The AVX512 version of ptranpose for PacketBlock<Packet16h,16> was
reordering the PacketBlock argument incorrectly.  This lead to errors in
the multiplication of matrices composed of 16 bit floats on AVX512
machines, if at least of the matrices was using RowMajor order.  This
error is responsible for one tensorflow unit test failure on AVX512
machines:

//tensorflow/python/kernel_tests:batch_matmul_op_test
 2018-06-16 15:13:06 -07:00
Gael Guennebaud
1f54164eca Fix a few issues with Packet16h 2018-07-07 00:15:07 +02:00
Gael Guennebaud
f2dc048df9 complete implementation of Packet16h (AVX512) 2018-07-06 17:43:11 +02:00
Gael Guennebaud
f4d623ffa7 Complete Packet8h implementation and test it in packetmath unit test 2018-07-06 17:13:36 +02:00
Deven Desai
b6cc0961b1 updates based on PR feedback 
There are two major changes (and a few minor ones which are not listed here...see PR discussion for details)

1. Eigen::half implementations for HIP and CUDA have been merged.
This means that
- `CUDA/Half.h` and `HIP/hcc/Half.h` got merged to a new file `GPU/Half.h`
- `CUDA/PacketMathHalf.h` and `HIP/hcc/PacketMathHalf.h` got merged to a new file `GPU/PacketMathHalf.h`
- `CUDA/TypeCasting.h` and `HIP/hcc/TypeCasting.h` got merged to a new file `GPU/TypeCasting.h`

After this change the `HIP/hcc` directory only contains one file `math_constants.h`. That will go away too once that file becomes a part of the HIP install.

2. new macros EIGEN_GPUCC, EIGEN_GPU_COMPILE_PHASE and EIGEN_HAS_GPU_FP16 have been added and the code has been updated to use them where appropriate.
- `EIGEN_GPUCC` is the same as `(EIGEN_CUDACC || EIGEN_HIPCC)`
- `EIGEN_GPU_DEVICE_COMPILE` is the same as `(EIGEN_CUDA_ARCH || EIGEN_HIP_DEVICE_COMPILE)`
- `EIGEN_HAS_GPU_FP16` is the same as `(EIGEN_HAS_CUDA_FP16 or EIGEN_HAS_HIP_FP16)`
 2018-06-14 10:21:54 -04:00
Deven Desai
ba972fb6b4 moving Half headers from CUDA dir to GPU dir, removing the HIP versions 2018-06-13 12:26:18 -04:00
Deven Desai
d1d22ef0f4 syncing this fork with upstream 2018-06-13 12:09:52 -04:00
Benoit Steiner
d3a380af4d Merged in mfigurnov/eigen/gamma-der-a (pull request PR-403) 
Derivative of the incomplete Gamma function and the sample of a Gamma random variable

Approved-by: Benoit Steiner <benoit.steiner.goog@gmail.com>
 2018-06-11 17:57:47 +00:00
Andrea Bocci
f7124b3e46 Extend CUDA support to matrix inversion and selfadjointeigensolver 2018-06-11 18:33:24 +02:00
Gael Guennebaud
0537123953 bug #1565: help MSVC to generatenot too bad ASM in reductions. 2018-07-05 09:21:26 +02:00