Actually I don't think it's buggy. But it probably triggers existing bugs, I suspect that
some xprs have LinearAccessBit and shouldn't have it.
Also this fixes the "bugs" with JacobiSVD ---> now it works again
- rename EvalBeforeAssignBit to MayAliasBit
- make .lazy() remove the MayAliasBit only, and mark it as deprecated
- add a NoAlias pseudo expression, and MatrixBase::noalias() function
Todo:
- we have to decide whether += and -= assume no aliasing by default ?
- once we agree on the API: update the Sparse module and the unit tests respectively.
* fix issues in Product revealed by this test
* in Dot.h forbid mixing of different types (at least for now, might allow real.dot(complex) in the future).
* use _mm_malloc/_mm_free on other platforms than linux of MSVC (eg., cygwin, OSX)
* replace a lot of inline keywords by EIGEN_STRONG_INLINE to compensate for
poor MSVC inlining
- in matrix-matrix product, static assert on the two scalar types to be the same.
- Similarly in CwiseBinaryOp. POTENTIALLY CONTROVERSIAL: we don't allow anymore binary
ops to take two different scalar types. The functors that we defined take two args
of the same type anyway; also we still allow the return type to be different.
Again the reason is that different scalar types are incompatible with vectorization.
Better have the user realize explicitly what mixing different numeric types costs him
in terms of performance.
See comment in CwiseBinaryOp constructor.
- This allowed to fix a little mistake in test/regression.cpp, mixing float and double
- Remove redundant semicolon (;) after static asserts
- added a MapBase base xpr on top of which Map and the specialization
of Block are implemented
- MapBase forces both aligned loads (and aligned stores, see below) in expressions
such as "x.block(...) += other_expr"
* Significant vectorization improvement:
- added a AlignedBit flag meaning the first coeff/packet is aligned,
this allows to not generate extra code to deal with the first unaligned part
- removed all unaligned stores when no unrolling
- removed unaligned loads in Sum when the input as the DirectAccessBit flag
* Some code simplification in CacheFriendly product
* Some minor documentation improvements
- added explicit enum to int conversion where needed
- if a function is not defined as declared and the return type is "tricky"
then the type must be typedefined somewhere. A "tricky return type" can be:
* a template class with a default parameter which depends on another template parameter
* a nested template class, or type of a nested template class
and vector * row-major products. Currently, it is enabled only is the matrix
has DirectAccessBit flag and the product is "large enough".
Added the respective unit tests in test/product/cpp.
* added complete implementation of sparse matrix product
(with a little glue in Eigen/Core)
* added an exhaustive bench of sparse products including GMM++ and MTL4
=> Eigen outperforms in all transposed/density configurations !
* rework PacketMath and DummyPacketMath, make these actual template
specializations instead of just overriding by non-template inline
functions
* introduce ei_ploadt and ei_pstoret, make use of them in Map and Matrix
* remove Matrix::map() methods, use Map constructors instead.
* introduce packet(int), make use of it in linear vectorized paths
--> completely fixes the slowdown noticed in benchVecAdd.
* generalize coeff(int) to linear-access xprs
* clarify the access flag bits
* rework api dox in Coeffs.h and util/Constants.h
* improve certain expressions's flags, allowing more vectorization
* fix bug in Block: start(int) and end(int) returned dyn*dyn size
* fix bug in Block: just because the Eval type has packet access
doesn't imply the block xpr should have it too.
(could come back to redux after it has been vectorized,
and could serve as a starting point for that)
also make the abs2 functor vectorizable (for real types).
packet access, it is not certain that it will bring a performance
improvement: benchmarking needed.
* improve logic choosing slice vectorization.
* fix typo in SSE packet math, causing crash in unaligned case.
* fix bug in Product, causing crash in unaligned case.
* add TEST_SSE3 CMake option.
* make Matrix2f (and similar) vectorized using linear path
* fix a couple of warnings and compilation issues with ICC and gcc 3.3/3.4
(cannot get Transform compiles with gcc 3.3/3.4, see the FIXME)
** Much better organization
** Fix a few bugs
** Add the ability to unroll only the inner loop
** Add an unrolled path to the Like1D vectorization. Not well tested.
** Add placeholder for sliced vectorization. Unimplemented.
* Rework of corrected_flags:
** improve rules determining vectorizability
** for vectors, the storage-order is indifferent, so we tweak it
to allow vectorization of row-vectors.
* fix compilation in benchmark, and a warning in Transpose.
Triangular class
- full meta-unrolling in Part
- move inverseProduct() to MatrixBase
- compilation fix in ProductWIP: introduce a meta-selector to only do
direct access on types that support it.
- phase out the old Product, remove the WIP_DIRTY stuff.
- misc renaming and fixes
