2554 Commits

Author SHA1 Message Date
Benoit Steiner
2dde63499c Generalized the matrix vector product code. 2014-10-31 16:33:51 -07:00
Benoit Steiner
bc99c5f7db fixed some potential alignment issues. 2014-10-30 18:09:53 -07:00
Benoit Steiner
1946cc4478 Added missing packet primitives for CUDA. 2014-10-30 17:52:32 -07:00
Christoph Hertzberg
883168ed94 Make select CUDA compatible (comparison operators aren't yet, so no test case yet) 2014-10-30 20:16:16 +01:00
Christoph Hertzberg
e5f134006b EIGEN_UNUSED_VARIABLE works better than casting to void. Make this also usable from CUDA code 2014-10-30 19:59:09 +01:00
Gael Guennebaud
21c0a2ce0c Move D&C SVD to official SVD module. 2014-10-29 11:29:33 +01:00
Christoph Hertzberg
e2e7ba9f85 bug #898: add inline hint to const_cast_ptr 2014-10-28 14:49:44 +01:00
Konstantinos Margaritis
fcb3573d17 Merged eigen/eigen into default 2014-10-22 10:42:18 +03:00
Konstantinos Margaritis
fae4fd7a26 Added ARMv8 support 2014-10-22 07:39:49 +00:00
Christoph Hertzberg
cf09c5f687 Prevent CUDA calling a __host__ function from a __host__ __device__ function is not allowed error. 2014-10-21 20:40:09 +02:00
Konstantinos Margaritis
b508619392 working 64-bit support in PacketMath.h, Complex.h needed 2014-10-21 18:10:33 +00:00
Gael Guennebaud
fe57b2f963 bug #701: workaround (min) and (max) blocking ADL by introducing numext::mini and numext::maxi internal functions and a EIGEN_NOT_A_MACRO macro. 2014-10-20 15:55:32 +02:00
Christoph Hertzberg
84aaa03182 Addendum to bug #859: pexp(NaN) for double did not return NaN, also, plog(NaN) did not return NaN.
psqrt(NaN) and psqrt(-1) shall return NaN if EIGEN_FAST_MATH==0
2014-10-20 13:13:43 +02:00
Gael Guennebaud
aa5f79206f Fix bug #859: pexp(NaN) returned Inf instead of NaN 2014-10-20 11:38:51 +02:00
Gael Guennebaud
8472e697ca Add lapack interface to JacobiSVD and BDCSVD 2014-10-17 15:31:11 +02:00
Benoit Steiner
bfdd9f3ac9 Made the blocking computation aware of the l3 cache
Also optimized the blocking parameters to take into account the number of threads used for a computation
2014-10-15 15:32:59 -07:00
Benoit Steiner
99d75235a9 Misc improvements and cleanups 2014-10-13 17:02:09 -07:00
Christoph Hertzberg
d3f52debc6 Make cuda_basic test compile again by adding lots of EIGEN_DEVICE_FUNC.
Although the test passes now, there might still be some missing.
2014-10-13 17:18:26 +02:00
Gael Guennebaud
48d537f59f Fix indentation 2014-10-09 23:35:26 +02:00
Gael Guennebaud
a48b82eece Add a scoped_array helper class to handle locally allocated/used arrays 2014-10-09 23:34:05 +02:00
Gael Guennebaud
5741349294 bug #882: fix various const-correctness issues with *View classes. 2014-10-07 18:29:28 +02:00
Gael Guennebaud
d44d432baa Re-enable products with triangular views of sparse matrices: we simply have to treat them as a sparse matrix. 2014-10-06 16:11:26 +02:00
Gael Guennebaud
893bfcf95f bug #887: use ei_declare_aligned_stack_constructed_variable instead of manual new[]/delete[] pairs in AMD and Paralellizer 2014-10-06 11:54:30 +02:00
Benoit Steiner
95a430a2ca Vector primitives for CUDA 2014-10-03 19:45:19 -07:00
Benoit Steiner
b7271dffb5 Generalized the gebp apis 2014-10-02 16:51:57 -07:00
Benoit Steiner
5cc23199be More tests to validate the const-correctness of the tensor code. 2014-10-02 10:30:44 -07:00
Konstantinos Margaritis
9d3c69952b fixed to make big-endian VSX work as well 2014-10-01 09:43:56 +00:00
Christoph Hertzberg
81517eebc1 Missing explicit 2014-09-30 16:42:04 +02:00
Christoph Hertzberg
12d59465cb bug #884: Copy constructor of Ref shall never malloc, constructing from other RefBase shall only malloc if the memory layout is incompatible. 2014-09-30 14:57:54 +02:00
Gael Guennebaud
56a0bbbbee Fix compilation with GCC 2014-09-29 18:28:18 +02:00
Gael Guennebaud
abd3502e9e Introduce a generic InnerIterator classes compatible with evaluators. 2014-09-29 13:36:57 +02:00
Gael Guennebaud
f0a62c90bc Avoid comparisons between different index types. 2014-09-29 10:27:51 +02:00
Georg Drenkhahn
2946992ad4 Using StorageIndexType for loop assigning initial permutation. Adding assert for index overflow. 2014-09-22 17:59:02 +02:00
Georg Drenkhahn
821ff0ecfb Using Index instead of hard coded int type to prevent potential implicit integer conversion 2014-09-22 16:12:35 +02:00
Georg Drenkhahn
2c4cace56c Using Index instead of hard coded int type to prevent potential implicit integer conversion 2014-09-22 15:54:34 +02:00
Georg Drenkhahn
b2755edcdd Replaced hard coded int types with Index types preventing implicit integer conversions. 2014-09-21 23:15:35 +02:00
Georg Drenkhahn
d1ef3c3546 Changed Diagonal::index() to return an Index type instead of int to prevent possible implicit conversion from long to int.
Added inline keyword to member methods.
2014-09-21 10:21:20 +02:00
Georg Drenkhahn
edaefeb978 Using Kernel::Index type instead of int to prevent possible implicit conversion from long to int. 2014-09-21 10:01:12 +02:00
Gael Guennebaud
74cde0c925 Add missing return derived() in ArrayBase::operator= 2014-09-28 09:16:13 +02:00
Christoph Hertzberg
4ba8aa1482 Fix bug #884: No malloc for zero-sized matrices or for Ref without temporaries 2014-09-25 16:05:17 +02:00
Gael Guennebaud
446001ef51 Fix nested_eval<Product<> > which wrongly returned a Product<> expression 2014-09-24 09:39:09 +02:00
Christoph Hertzberg
421feea3b2 member_redux constructor is explicit too. Renamed some typedefs for more consistency. 2014-09-23 18:55:42 +02:00
Christoph Hertzberg
36448c9e28 Make constructors explicit if they could lead to unintended implicit conversion 2014-09-23 14:28:23 +02:00
Gael Guennebaud
f9d6d3780f bug #879: fix compilation of tri1=mat*tri2 by copying tri2 into a full temporary. 2014-09-22 17:34:17 +02:00
Christoph Hertzberg
d9e0336a78 Merged in kmargar/eigen (pull request PR-84)
Add VSX support
2014-09-22 12:57:06 +02:00
Konstantinos Margaritis
de38ff2499 prefetch are noops on VSX, actually disable the prefetch trait 2014-09-21 11:56:07 +00:00
Konstantinos Margaritis
60e093a9dc Merged eigen/eigen into default 2014-09-21 14:02:51 +03:00
Konstantinos Margaritis
56408504e4 fix compile error on big endian altivec 2014-09-21 13:59:30 +03:00
Konstantinos Margaritis
974fe38ca3 prefetch are noops on VSX 2014-09-21 11:24:30 +00:00
Konstantinos Margaritis
c0205ca4af VSX supports vec_div, implement where appropriate (float/doubles) 2014-09-21 08:12:22 +00:00