mirror/eigen
2
0
Fork 0
mirror of https://gitlab.com/libeigen/eigen.git synced 2025-01-06 14:14:46 +08:00
Code Issues Packages Projects Releases Wiki Activity
8,159 Commits 11 Branches 89 Tags 168 MiB
3d71d3918e
Commit Graph

4731 Commits

Author SHA1 Message Date
Benoit Steiner
ff4a289572 Cleaned up the fp16 code 2016-05-24 18:50:09 -07:00
Gael Guennebaud
e68e165a23 bug #256: enable vectorization with unaligned loads/stores. 
This concerns all architectures and all sizes.
This new behavior can be disabled by defining EIGEN_UNALIGNED_VECTORIZE=0
 2016-05-24 21:54:03 +02:00
Gael Guennebaud
78390e4189 Block<> should not disable vectorization based on inner-size, this is the responsibilty of the assignment logic. 2016-05-24 17:14:01 +02:00
Gael Guennebaud
64bb7576eb Clean propagation of Dest/Src alignments. 2016-05-24 17:12:12 +02:00
Benoit Jacob
40a16282c7 Remove now-unused protate PacketMath func 2016-05-24 11:01:18 -04:00
Benoit Jacob
6136f4fdd4 Remove the rotating kernel. It was only useful on some ARM CPUs (Qualcomm Krait) that are not as ubiquitous today as they were when I introduced it. 2016-05-24 10:00:32 -04:00
Benoit Steiner
e617711306 Don't attempt to use MMX instructions with visualstudio since they're only partially supported. 2016-05-24 06:43:58 -07:00
Benoit Steiner
334e76537f Worked around missing clang intrinsic 2016-05-24 00:29:28 -07:00
Benoit Steiner
b517ab349b Use the generic ploadquad intrinsics since it does the job 2016-05-24 00:11:17 -07:00
Benoit Steiner
646872cb3b Worked around missing clang intrinsics 2016-05-24 00:07:08 -07:00
Benoit Steiner
3dfc391a61 Added missing EIGEN_DEVICE_FUNC qualifier 2016-05-23 20:56:59 -07:00
Benoit Steiner
3d0741f027 Include mmintrin.h to make it possible to use mmx instructions when needed. For example, this will enable the definition of a half packet for the Packet4f type. 2016-05-23 20:43:48 -07:00
Benoit Steiner
33a94f5dc7 Use the Index type instead of integers to specify the strides in pgather/pscatter 2016-05-23 20:37:30 -07:00
Benoit Steiner
6bc684ab6a Added missing alignment in the fp16 packet traits 2016-05-23 20:32:30 -07:00
Benoit Steiner
283e33dea4 ptranspose is not a template. 2016-05-23 19:55:55 -07:00
Benoit Steiner
a5a3ba2b80 Avoid unnecessary float to double conversions 2016-05-23 17:16:09 -07:00
Benoit Steiner
5ba0ebe7c9 Avoid unnecessary float to double conversion. 2016-05-23 17:14:31 -07:00
Benoit Steiner
7d980d74e5 Started to vectorize the processing of 16bit floats on CPU. 2016-05-23 15:21:40 -07:00
Benoit Steiner
5d51a7f12c Don't optimize the processing of the last rows of a matrix matrix product in cases that violate the assumptions made by the optimized code path. 2016-05-23 15:13:16 -07:00
Christoph Hertzberg
88654762da Replace multiple constructors of half-type by a generic/templated constructor. This fixes an incompatibility with long double, exposed by the previous commit. 2016-05-23 10:03:03 +02:00
Christoph Hertzberg
718521d5cf Silenced several double-promotion warnings 2016-05-22 18:17:04 +02:00
Gael Guennebaud
ccaace03c9 Make EIGEN_HAS_CONSTEXPR user configurable 2016-05-20 15:10:08 +02:00
Gael Guennebaud
c3410804cd Make EIGEN_HAS_VARIADIC_TEMPLATES user configurable 2016-05-20 15:05:38 +02:00
Gael Guennebaud
abd1c1af7a Make EIGEN_HAS_STD_RESULT_OF user configurable 2016-05-20 15:01:27 +02:00
Gael Guennebaud
1395056fc0 Make EIGEN_HAS_C99_MATH user configurable 2016-05-20 14:58:19 +02:00
Gael Guennebaud
48bf5ec216 Make EIGEN_HAS_RVALUE_REFERENCES user configurable 2016-05-20 14:54:20 +02:00
Gael Guennebaud
f43ae88892 Rename EIGEN_HAVE_RVALUE_REFERENCES to EIGEN_HAS_RVALUE_REFERENCES 2016-05-20 14:48:51 +02:00
Gael Guennebaud
998f2efc58 Add a EIGEN_MAX_CPP_VER option to limit the C++ version to be used. 2016-05-20 14:44:28 +02:00
Gael Guennebaud
c028d96089 Improve doc of special math functions 2016-05-20 14:18:48 +02:00
Gael Guennebaud
0ba32f99bd Rename UniformRandom to UnitRandom. 2016-05-20 13:21:34 +02:00
Gael Guennebaud
7a9d9cde94 Fix coding practice in Quaternion::UniformRandom 2016-05-20 13:19:52 +02:00
Joseph Mirabel
eb0cc2573a bug #823: add static method to Quaternion for uniform random rotations. 2016-05-20 13:15:40 +02:00
Gael Guennebaud
6761c64d60 zeta and polygamma are not unary functions, but binary ones. 2016-05-19 18:34:16 +02:00
Gael Guennebaud
7a54032408 zeta and digamma do not require C++11/C99 2016-05-19 17:36:47 +02:00
Gael Guennebaud
ce12562710 Add some c++11 flags in documentation 2016-05-19 17:35:30 +02:00
Gael Guennebaud
b6ed8244b4 bug #1201: optimize affine*vector products 2016-05-19 16:09:15 +02:00
Gael Guennebaud
73693b5de6 bug #1221: disable gcc 6 warning: ignoring attributes on template argument 2016-05-19 15:21:53 +02:00
Gael Guennebaud
df9a5e13c6 Fix SelfAdjointEigenSolver for some input expression types, and add new regression unit tests for sparse and selfadjointview inputs. 2016-05-19 13:07:33 +02:00
Gael Guennebaud
6a2916df80 DiagonalWrapper is a vector, so it must expose the LinearAccessBit flag. 2016-05-19 13:06:21 +02:00
Gael Guennebaud
a226f6af6b Add support for SelfAdjointView::diagonal() 2016-05-19 13:05:33 +02:00
Gael Guennebaud
ee7da3c7c5 Fix SelfAdjointView::triangularView for complexes. 2016-05-19 13:01:51 +02:00
Gael Guennebaud
b6b8578a67 bug #1230: add support for SelfadjointView::triangularView. 2016-05-19 11:36:38 +02:00
Gael Guennebaud
84df9142e7 bug #1231: fix compilation regression regarding complex_array/=real_array and add respective unit tests 2016-05-18 23:00:13 +02:00
Gael Guennebaud
21d692d054 Use coeff(i,j) instead of operator(). 2016-05-18 17:09:20 +02:00
Gael Guennebaud
8456bbbadb bug #1224: fix regression in (dense*dense).sparseView() by specializing evaluator<SparseView<Product>> for sparse products only. 2016-05-18 16:53:28 +02:00
Gael Guennebaud
b507b82326 Use default sorting strategy for square products. 2016-05-18 16:51:54 +02:00
Gael Guennebaud
747e3290c0 bug #1213: rename some enums type for consistency. 2016-05-18 13:26:56 +02:00
Rasmus Munk Larsen
0dbd68145f Roll back changes to core. Move include of TensorFunctors.h up to satisfy dependence in TensorCostModel.h. 2016-05-17 10:25:19 -07:00
Rasmus Munk Larsen
e55deb21c5 Improvements to parallelFor. 
Move some scalar functors from TensorFunctors. to Eigen core.
 2016-05-12 14:07:22 -07:00
Benoit Steiner
fae0493f98 Fixed a couple of bugs related to the Pascalfamily of GPUs 
H: Enter commit message.  Lines beginning with 'HG:' are removed.
 2016-05-11 23:02:26 -07:00
Benoit Steiner
b6a517c47d Added the ability to load fp16 using the texture path. 
Improved the performance of some reductions on fp16
 2016-05-11 21:26:48 -07:00
Benoit Steiner
518149e868 Misc fixes for fp16 2016-05-11 20:11:14 -07:00
Benoit Steiner
56a1757d74 Made predux_min and predux_max on fp16 less noisy 2016-05-11 17:37:34 -07:00
Benoit Steiner
9091351dbe __ldg is only available with cuda architectures >= 3.5 2016-05-11 15:22:13 -07:00
Benoit Steiner
02f76dae2d Fixed a typo 2016-05-11 15:08:38 -07:00
Christoph Hertzberg
131e5a1a4a Do not copy for trivial 1x1 case. This also avoids a "maybe-uninitialized" warning in some situations. 2016-05-11 23:50:13 +02:00
Benoit Steiner
70195a5ff7 Added missing EIGEN_DEVICE_FUNC 2016-05-11 14:10:09 -07:00
Benoit Steiner
09a19c33a8 Added missing EIGEN_DEVICE_FUNC qualifiers 2016-05-11 14:07:43 -07:00
Christoph Hertzberg
33ca7e3c8d bug #1207: Add and fix logical-op warnings 2016-05-11 19:36:34 +02:00
Benoit Steiner
217d984abc Fixed a typo in my previous commit 2016-05-11 10:22:15 -07:00
Christoph Hertzberg
0f61343893 Workaround maybe-uninitialized warning 2016-05-11 09:00:18 +02:00
Christoph Hertzberg
3bfc9b47ca Workaround "misleading-indentation" warnings 2016-05-11 08:41:36 +02:00
Benoit Steiner
0b9e3dcd06 Added packet primitives to compute exp, log, sqrt and rsqrt on fp16. This improves the performance by 10 to 30%. 2016-05-10 11:05:33 -07:00
Benoit Steiner
8adf5cc70f Added support for packet processing of fp16 on kepler and maxwell gpus 2016-05-06 19:16:43 -07:00
Christoph Hertzberg
a11bd82dc3 bug #1213: Give names to anonymous enums 2016-05-06 11:31:56 +02:00
Benoit Steiner
0451940fa4 Relaxed the dummy precision for fp16 2016-05-05 15:40:01 -07:00
Christoph Hertzberg
dacb469bc9 Enable and fix -Wdouble-conversion warnings 2016-05-05 13:35:45 +02:00
Ola Røer Thorsen
be78aea6b3 fix double-promotion/float-conversion in Core/SpecialFunctions.h 2016-05-04 10:52:08 +02:00
Gael Guennebaud
75a94b9662 Improve documentation of BDCSVD 2016-05-04 12:53:14 +02:00
Gael Guennebaud
e2ca478485 bug #1214: consider denormals as zero in D&C SVD. This also workaround infinite binary search when compiling with ICC's unsafe optimizations. 2016-05-03 23:15:29 +02:00
Benoit Steiner
6c3e5b85bc Fixed compilation error with cuda >= 7.5 2016-05-03 09:38:42 -07:00
Benoit Steiner
da50419df8 Made a cast explicit 2016-05-02 19:50:22 -07:00
Gael Guennebaud
b1bd53aa6b Fix performance regression: with AVX, unaligned stores were emitted instead of aligned ones for fixed size assignement. 2016-05-01 23:25:06 +02:00
Benoit Steiner
2b890ae618 Fixed compilation errors generated by clang 2016-04-29 18:30:40 -07:00
Benoit Steiner
46bcb70969 Don't turn on const expressions when compiling with gcc >= 4.8 unless the -std=c++11 option has been used 2016-04-29 15:20:59 -07:00
Gael Guennebaud
0f3c4c8ff4 Fix compilation of sparse.cast<>().transpose(). 2016-04-29 18:26:08 +02:00
Benoit Steiner
dacb23277e Fixed the igamma and igammac implementations to make them callable from a gpu kernel. 2016-04-28 18:54:54 -07:00
Benoit Steiner
a5d4545083 Deleted unused variable 2016-04-28 14:14:48 -07:00
Justin Lebar
40d1e2f8c7 Eliminate mutual recursion in igamma{,c}_impl::Run. 
Presently, igammac_impl::Run calls igamma_impl::Run, which in turn calls
igammac_impl::Run.

This isn't actually mutual recursion; the calls are guarded such that we never
get into a loop.  Nonetheless, it's a stretch for clang to prove this.  As a
result, clang emits a recursive call in both igammac_impl::Run and
igamma_impl::Run.

That this is suboptimal code is bad enough, but it's particularly bad when
compiling for CUDA/nvptx.  nvptx allows recursion, but only begrudgingly: If
you have recursive calls in a kernel, it's on you to manually specify the
kernel's stack size.  Otherwise, ptxas will dump a warning, make a guess, and
who knows if it's right.

This change explicitly eliminates the mutual recursion in igammac_impl::Run and
igamma_impl::Run.
 2016-04-28 13:57:08 -07:00
Benoit Steiner
2b917291d9 Merged in rmlarsen/eigen2 (pull request PR-183) 
Detect cxx_constexpr support when compiling with clang.
 2016-04-27 15:19:54 -07:00
Rasmus Munk Larsen
09b9e951e3 Depend on the more extensive support for constexpr in clang: 
http://clang.llvm.org/docs/LanguageExtensions.html#c-1y-relaxed-constexpr
 2016-04-27 14:59:11 -07:00
Rasmus Munk Larsen
1a325ef71c Detect cxx_constexpr support when compiling with clang. 2016-04-27 14:33:51 -07:00
Benoit Steiner
c61170e87d fpclassify isn't portable enough. In particular, the return values of the function are not available on all the platforms Eigen supportes: remove it from Eigen. 2016-04-27 14:22:20 -07:00
Benoit Steiner
f629fe95c8 Made the index type a template parameter to evaluateProductBlockingSizes 
Use numext::mini and numext::maxi instead of std::min/std::max to compute blocking sizes.
 2016-04-27 13:11:19 -07:00
Benoit Steiner
25141b69d4 Improved support for min and max on 16 bit floats when running on recent cuda gpus 2016-04-27 12:57:21 -07:00
Benoit Steiner
6744d776ba Added support for fpclassify in Eigen::Numext 2016-04-27 12:10:25 -07:00
Benoit Steiner
5c372d19e3 Merged in rmlarsen/eigen (pull request PR-179) 
Prevent crash in CompleteOrthogonalDecomposition if object was default constructed.
 2016-04-21 18:06:36 -07:00
Rasmus Munk Larsen
a3256d78d8 Prevent crash in CompleteOrthogonalDecomposition if object was default constructed. 2016-04-21 16:49:28 -07:00
Benoit Steiner
80200a1828 Don't attempt to leverage the _cvtss_sh and _cvtsh_ss instructions when compiling with clang since it's unclear which versions of clang actually support these instruction. 2016-04-20 12:10:27 -07:00
Benoit Steiner
1d0238375d Made sure all the required header files are included when trying to use fp16 2016-04-19 17:44:12 -07:00
Gael Guennebaud
e4fe611e2c Enable lazy-coeff-based-product for vector*(1x1) products 2016-04-16 15:17:39 +02:00
Benoit Steiner
1a16fb1532 Deleted extraneous comma. 2016-04-15 15:50:13 -07:00
Gael Guennebaud
2a7115daca bug #1203: by-pass large stack-allocation in stableNorm if EIGEN_STACK_ALLOCATION_LIMIT is too small 2016-04-15 22:34:11 +02:00
Benoit Steiner
1d23430628 Improved the matrix multiplication blocking in the case where mr is not a power of 2 (e.g on Haswell CPUs). 2016-04-15 10:53:31 -07:00
Gael Guennebaud
1e80bddde3 Fix trmv for mixing types. 2016-04-15 17:58:36 +02:00
Benoit Steiner
a62e924656 Added ability to access the cache sizes from the tensor devices 2016-04-14 21:25:06 -07:00
Benoit Steiner
18e6f67426 Added support for exclusive or 2016-04-14 20:37:46 -07:00
Gael Guennebaud
20f387fafa Improve numerical robustness of JacoviSVD: 
- avoid noise amplification in complex to real conversion
 - compare off-diagonal entries to the current biggest diagonal entry: no need to bother about a 2x2 block containing ridiculously small entries compared to the rest of the matrix.
 2016-04-14 22:46:55 +02:00
Benoit Steiner
7718749fee Force the inlining of the << operator on half floats 2016-04-14 11:51:54 -07:00
Benoit Steiner
5379d2b594 Inline the << operator on half floats 2016-04-14 11:40:48 -07:00