Eugene Zhulenev
eb21bab769
Parallelize tensor contraction only by sharding dimension and use 'thread-local' memory for packing
2019-02-04 10:43:16 -08:00
Gael Guennebaud
871e2e5339
bug #1674 : disable GCC's unsafe-math-optimizations in sin/cos vectorization (results are completely wrong otherwise)
2019-02-03 08:54:47 +01:00
Rasmus Larsen
e7b481ea74
Merged in rmlarsen/eigen (pull request PR-578)
...
Speed up Eigen matrix*vector and vector*matrix multiplication.
Approved-by: Eugene Zhulenev <ezhulenev@google.com>
2019-02-02 01:53:44 +00:00
Sameer Agarwal
b55b5c7280
Speed up row-major matrix-vector product on ARM
...
The row-major matrix-vector multiplication code uses a threshold to
check if processing 8 rows at a time would thrash the cache.
This change introduces two modifications to this logic.
1. A smaller threshold for ARM and ARM64 devices.
The value of this threshold was determined empirically using a Pixel2
phone, by benchmarking a large number of matrix-vector products in the
range [1..4096]x[1..4096] and measuring performance separately on
small and little cores with frequency pinning.
On big (out-of-order) cores, this change has little to no impact. But
on the small (in-order) cores, the matrix-vector products are up to
700% faster. Especially on large matrices.
The motivation for this change was some internal code at Google which
was using hand-written NEON for implementing similar functionality,
processing the matrix one row at a time, which exhibited substantially
better performance than Eigen.
With the current change, Eigen handily beats that code.
2. Make the logic for choosing number of simultaneous rows apply
unifiormly to 8, 4 and 2 rows instead of just 8 rows.
Since the default threshold for non-ARM devices is essentially
unchanged (32000 -> 32 * 1024), this change has no impact on non-ARM
performance. This was verified by running the same set of benchmarks
on a Xeon desktop.
2019-02-01 15:23:53 -08:00
Rasmus Munk Larsen
4c0fa6ce0f
Speed up Eigen matrix*vector and vector*matrix multiplication.
...
This change speeds up Eigen matrix * vector and vector * matrix multiplication for dynamic matrices when it is known at runtime that one of the factors is a vector.
The benchmarks below test
c.noalias()= n_by_n_matrix * n_by_1_matrix;
c.noalias()= 1_by_n_matrix * n_by_n_matrix;
respectively.
Benchmark measurements:
SSE:
Run on *** (72 X 2992 MHz CPUs); 2019-01-28T17:51:44.452697457-08:00
CPU: Intel Skylake Xeon with HyperThreading (36 cores) dL1:32KB dL2:1024KB dL3:24MB
Benchmark Base (ns) New (ns) Improvement
------------------------------------------------------------------
BM_MatVec/64 1096 312 +71.5%
BM_MatVec/128 4581 1464 +68.0%
BM_MatVec/256 18534 5710 +69.2%
BM_MatVec/512 118083 24162 +79.5%
BM_MatVec/1k 704106 173346 +75.4%
BM_MatVec/2k 3080828 742728 +75.9%
BM_MatVec/4k 25421512 4530117 +82.2%
BM_VecMat/32 352 130 +63.1%
BM_VecMat/64 1213 425 +65.0%
BM_VecMat/128 4640 1564 +66.3%
BM_VecMat/256 17902 5884 +67.1%
BM_VecMat/512 70466 24000 +65.9%
BM_VecMat/1k 340150 161263 +52.6%
BM_VecMat/2k 1420590 645576 +54.6%
BM_VecMat/4k 8083859 4364327 +46.0%
AVX2:
Run on *** (72 X 2993 MHz CPUs); 2019-01-28T17:45:11.508545307-08:00
CPU: Intel Skylake Xeon with HyperThreading (36 cores) dL1:32KB dL2:1024KB dL3:24MB
Benchmark Base (ns) New (ns) Improvement
------------------------------------------------------------------
BM_MatVec/64 619 120 +80.6%
BM_MatVec/128 9693 752 +92.2%
BM_MatVec/256 38356 2773 +92.8%
BM_MatVec/512 69006 12803 +81.4%
BM_MatVec/1k 443810 160378 +63.9%
BM_MatVec/2k 2633553 646594 +75.4%
BM_MatVec/4k 16211095 4327148 +73.3%
BM_VecMat/64 925 227 +75.5%
BM_VecMat/128 3438 830 +75.9%
BM_VecMat/256 13427 2936 +78.1%
BM_VecMat/512 53944 12473 +76.9%
BM_VecMat/1k 302264 157076 +48.0%
BM_VecMat/2k 1396811 675778 +51.6%
BM_VecMat/4k 8962246 4459010 +50.2%
AVX512:
Run on *** (72 X 2993 MHz CPUs); 2019-01-28T17:35:17.239329863-08:00
CPU: Intel Skylake Xeon with HyperThreading (36 cores) dL1:32KB dL2:1024KB dL3:24MB
Benchmark Base (ns) New (ns) Improvement
------------------------------------------------------------------
BM_MatVec/64 401 111 +72.3%
BM_MatVec/128 1846 513 +72.2%
BM_MatVec/256 36739 1927 +94.8%
BM_MatVec/512 54490 9227 +83.1%
BM_MatVec/1k 487374 161457 +66.9%
BM_MatVec/2k 2016270 643824 +68.1%
BM_MatVec/4k 13204300 4077412 +69.1%
BM_VecMat/32 324 106 +67.3%
BM_VecMat/64 1034 246 +76.2%
BM_VecMat/128 3576 802 +77.6%
BM_VecMat/256 13411 2561 +80.9%
BM_VecMat/512 58686 10037 +82.9%
BM_VecMat/1k 320862 163750 +49.0%
BM_VecMat/2k 1406719 651397 +53.7%
BM_VecMat/4k 7785179 4124677 +47.0%
Currently watchingStop watching
2019-01-31 14:24:08 -08:00
Gael Guennebaud
7ef879f6bf
GEBP: improves pipelining in the 1pX4 path with FMA.
...
Prior to this change, a product with a LHS having 8 rows was faster with AVX-only than with AVX+FMA.
With AVX+FMA I measured a speed up of about x1.25 in such cases.
2019-01-30 23:45:12 +01:00
Gael Guennebaud
de77bf5d6c
Fix compilation with ARM64.
2019-01-30 16:48:20 +01:00
Gael Guennebaud
d586686924
Workaround lack of support for arbitrary packet-type in Tensor by manually loading half/quarter packets in tensor contraction mapper.
2019-01-30 16:48:01 +01:00
Gael Guennebaud
eb4c6bb22d
Fix conflicts and merge
2019-01-30 15:57:08 +01:00
Gael Guennebaud
e3622a0396
Slightly extend discussions on auto and move the content of the Pit falls wiki page here.
...
http://eigen.tuxfamily.org/index.php?title=Pit_Falls
2019-01-30 13:09:21 +01:00
Gael Guennebaud
df12fae8b8
According to https://gcc.gnu.org/bugzilla/show_bug.cgi?id=89101 , the previous GCC issue is fixed in GCC trunk (will be gcc 9).
2019-01-30 11:52:28 +01:00
Gael Guennebaud
3775926bba
ARM64 & GEBP: add specialization for double +30% speed up
2019-01-30 11:49:06 +01:00
Gael Guennebaud
be5b0f664a
ARM64 & GEBP: Make use of vfmaq_laneq_f32 and workaround GCC's issue in generating good ASM
2019-01-30 11:48:25 +01:00
Christoph Hertzberg
a7779a9b42
Hide some annoying unused variable warnings in g++8.1
2019-01-29 16:48:21 +01:00
Gael Guennebaud
efe02292a6
Add recent gemm related changesets and various cleanups in perf-monitoring
2019-01-29 11:53:47 +01:00
Gael Guennebaud
8a06c699d0
bug #1669 : fix PartialPivLU/inverse with zero-sized matrices.
2019-01-29 10:27:13 +01:00
Gael Guennebaud
a2a07e62b9
Fix compilation with c++03 (local class cannot be template arguments), and make SparseMatrix::assignDiagonal truly protected.
2019-01-29 10:10:07 +01:00
Gael Guennebaud
f489f44519
bug #1574 : implement "sparse_matrix =,+=,-= diagonal_matrix" with smart insertion strategies of missing diagonal coeffs.
2019-01-28 17:29:50 +01:00
Gael Guennebaud
803fa79767
Move evaluator<SparseCompressedBase>::find(i,j) to a more general and reusable SparseCompressedBase::lower_bound(i,j) functiion
2019-01-28 17:24:44 +01:00
Gael Guennebaud
53560f9186
bug #1672 : fix unit test compilation with MSVC by adding overloads of test_is* for long long (and factorize copy/paste code through a macro)
2019-01-28 13:47:28 +01:00
Christoph Hertzberg
c9825b967e
Renaming even more I identifiers
2019-01-26 13:22:13 +01:00
Christoph Hertzberg
5a52e35f9a
Renaming some more I identifiers
2019-01-26 13:18:21 +01:00
Rasmus Munk Larsen
71429883ee
Fix compilation error in NEON GEBP specializaition of madd.
2019-01-25 17:00:21 -08:00
Christoph Hertzberg
934b8a1304
Avoid I as an identifier, since it may clash with the C-header complex.h
2019-01-25 14:54:39 +01:00
Gael Guennebaud
ec8a387972
cleanup
2019-01-24 10:24:45 +01:00
Gael Guennebaud
6908ce2a15
More thoroughly check variadic template ctor of fixed-size vectors
2019-01-24 10:24:28 +01:00
David Tellenbach
237b03b372
PR 574: use variadic template instead of initializer_list to implement fixed-size vector ctor from coefficients.
2019-01-23 00:07:19 +01:00
Christoph Hertzberg
bd6dadcda8
Tell doxygen that cxx11 math is available
2019-01-24 00:14:02 +01:00
Gael Guennebaud
c64d5d3827
Bypass inline asm for non compatible compilers.
2019-01-23 23:43:13 +01:00
Christoph Hertzberg
e16913a45f
Fix name of tutorial snippet.
2019-01-23 10:35:06 +01:00
Gael Guennebaud
80f81f9c4b
Cleanup SFINAE in Array/Matrix(initializer_list) ctors and minor doc editing.
2019-01-22 17:08:47 +01:00
David Tellenbach
db152b9ee6
PR 572: Add initializer list constructors to Matrix and Array (include unit tests and doc)
...
- {1,2,3,4,5,...} for fixed-size vectors only
- {{1,2,3},{4,5,6}} for the general cases
- {{1,2,3,4,5,....}} is allowed for both row and column-vector
2019-01-21 16:25:57 +01:00
Gael Guennebaud
543529da6a
Add more extensive tests of Array ctors, including {} variants
2019-01-22 15:30:50 +01:00
nluehr
92774f0275
Replace host_define.h with cuda_runtime_api.h
2019-01-18 16:10:09 -06:00
Gael Guennebaud
d18f49cbb3
Fix compilation of unit tests with gcc and c++17
2019-01-18 11:12:42 +01:00
Christoph Hertzberg
da0a41b9ce
Mask unused-parameter warnings, when building with NDEBUG
2019-01-18 10:41:14 +01:00
Rasmus Munk Larsen
2eccbaf3f7
Add missing logical packet ops for GPU and NEON.
2019-01-17 17:45:08 -08:00
Christoph Hertzberg
d575505d25
After fixing bug #1557 , boostmultiprec_7 failed with NumericalIssue instead of NoConvergence (all that matters here is no Success)
2019-01-17 19:14:07 +01:00
Gael Guennebaud
ee3662abc5
Remove some useless const_cast
2019-01-17 18:27:49 +01:00
Gael Guennebaud
0fe6b7d687
Make nestByValue works again (broken since 3.3) and add unit tests.
2019-01-17 18:27:25 +01:00
Gael Guennebaud
4b7cf7ff82
Extend reshaped unit tests and remove useless const_cast
2019-01-17 17:35:32 +01:00
Gael Guennebaud
b57c9787b1
Cleanup useless const_cast and add missing broadcast assignment tests
2019-01-17 16:55:42 +01:00
Gael Guennebaud
be05d0030d
Make FullPivLU use conjugateIf<>
2019-01-17 12:01:00 +01:00
Patrick Peltzer
bba2f05064
Boosttest only available for Boost version >= 1.53.0
2019-01-17 11:54:37 +01:00
Patrick Peltzer
15e53d5d93
PR 567: makes all dense solvers inherit SoverBase (LU,Cholesky,QR,SVD).
...
This changeset also includes:
* add HouseholderSequence::conjugateIf
* define int as the StorageIndex type for all dense solvers
* dedicated unit tests, including assertion checking
* _check_solve_assertion(): this method can be implemented in derived solver classes to implement custom checks
* CompleteOrthogonalDecompositions: add applyZOnTheLeftInPlace, fix scalar type in applyZAdjointOnTheLeftInPlace(), add missing assertions
* Cholesky: add missing assertions
* FullPivHouseholderQR: Corrected Scalar type in _solve_impl()
* BDCSVD: Unambiguous return type for ternary operator
* SVDBase: Corrected Scalar type in _solve_impl()
2019-01-17 01:17:39 +01:00
Gael Guennebaud
7f32109c11
Add conjugateIf<bool> members to DesneBase, TriangularView, SelfadjointView, and make PartialPivLU use it.
2019-01-17 11:33:43 +01:00
Gael Guennebaud
7b35c26b1c
Doc: remove link to porting guide
2019-01-17 10:35:50 +01:00
Gael Guennebaud
4759d9e86d
Doc: add manual page on STL iterators
2019-01-17 10:35:14 +01:00
Gael Guennebaud
562985bac4
bug #1646 : fix false aliasing detection for A.row(0) = A.col(0);
...
This changeset completely disable the detection for vectors for which are current mechanism cannot detect any positive aliasing anyway.
2019-01-17 00:14:27 +01:00
Rasmus Munk Larsen
7401e2541d
Fix compilation error for logical packet ops with older compilers.
2019-01-16 14:43:33 -08:00
