2108 Commits

Author SHA1 Message Date
Gael Guennebaud
ea27678153 fix compilation of ei_tridiagonalization_inplace_selector for 1x1 matrix 2010-07-18 17:10:11 +02:00
Gael Guennebaud
2d78023815 fix hemm to not use the imaginary part of the diagonal entries 2010-07-17 11:57:54 +02:00
Gael Guennebaud
4c19024fbf re-enable writing to reversed objects 2010-07-16 22:26:07 +02:00
Gael Guennebaud
fb041c260c fix for empty matrices 2010-07-16 22:25:35 +02:00
Gael Guennebaud
883a8cbb2c disable the optimized 3x3 path for complexes which was not working at all 2010-07-16 18:22:00 +02:00
Gael Guennebaud
6ab9e8632f fix bad fuzzy comparison in 3x3 tridiagonalization 2010-07-16 16:38:58 +02:00
Gael Guennebaud
044424b0e2 fix sum()/prod() on empty matrix making sure this does not affect fixed sized object, extend related unit tests including partial reduction 2010-07-16 14:02:20 +02:00
Gael Guennebaud
b08c26aefa merge 2010-07-15 20:41:33 +02:00
Gael Guennebaud
84fdbded4d add support for strictly triangular matrix in trmm though it is not really useful 2010-07-15 20:39:20 +02:00
Gael Guennebaud
bfbe61454e merge 2010-07-15 09:54:31 +02:00
Gael Guennebaud
cf9edd9958 fix compilation for non trivial types 2010-07-14 23:31:38 +02:00
Gael Guennebaud
90d6fc0e28 fix ei_aligned_delete for null pointers and non trivial dtors 2010-07-14 22:49:34 +02:00
Christoph Hertzberg
6ba5d2c90c Implemented SSE optimized double-precision Quaternion multiplication 2010-07-12 23:30:47 +02:00
Gael Guennebaud
850c6d8a2b fix unused warning 2010-07-11 10:58:58 +02:00
Gael Guennebaud
931027f31b add a utilility to debug cpuid, and makes sure we get 0 if we query an unsupported cpuid function 2010-06-26 23:15:06 +02:00
Gael Guennebaud
ec07c4109d add default parameters for InnerStride/OuterStride to be
able to simply write OuterStride<> instead of OuterStride<Dynamic>
2010-06-25 14:48:16 +02:00
Gael Guennebaud
4056db01e7 use const Scalar& instead of Scalar for function arguments 2010-06-25 13:52:23 +02:00
Gael Guennebaud
686689e9cf comment all disabled MSVC warnings 2010-06-25 13:31:07 +02:00
Gael Guennebaud
75b6d2b2f8 fix very annoying warning (gcc 4.3): type qualifiers ignored on function return type 2010-06-25 13:20:34 +02:00
Gael Guennebaud
01553c419e fox blcok size computation for fixed size objects 2010-06-25 11:44:55 +02:00
Gael Guennebaud
e313826890 add mixed sparse-dense outer product 2010-06-25 11:36:38 +02:00
Gael Guennebaud
1927b4dff5 Fix use of nesting types in SparseTranspose and split the big SparseProduct.h file 2010-06-25 10:26:24 +02:00
Gael Guennebaud
28e64b0da3 email change 2010-06-24 23:21:58 +02:00
Gael Guennebaud
99e4afd43e makes SparseView a true sparse expression and fix use of nesting types 2010-06-24 22:48:48 +02:00
Gael Guennebaud
f3b875e434 fix infinite loop 2010-06-24 22:18:09 +02:00
Gael Guennebaud
566867428c - add a low level mechanism to provide preallocated memory to gemm
- ensure static allocation for the product of "large" fixed size matrix
2010-06-24 21:44:24 +02:00
Gael Guennebaud
e039edcb42 fix temporary creation rule 2010-06-24 19:21:25 +02:00
Gael Guennebaud
b22fc6cdc3 bug fix in gemv:
solution always use a temporary in dst.innerStride != 1
even though this is not needed when packet_size == 1....
2010-06-24 17:51:25 +02:00
Gael Guennebaud
6be06745df block householder : minor optimization 2010-06-24 17:48:38 +02:00
Gael Guennebaud
905beb0953 fix symm 2010-06-24 16:42:43 +02:00
Gael Guennebaud
af38bccd3d fix syrk 2010-06-24 16:26:27 +02:00
Gael Guennebaud
19f2f53e2c fix compilation when default to row major 2010-06-24 15:13:41 +02:00
Gael Guennebaud
d44fce501b fix computation of blocking sizes for small triangular matrices 2010-06-24 11:50:28 +02:00
Hauke Heibel
0068d3ccf6 Added version testing for MSVC. 2010-06-24 10:05:24 +02:00
Hauke Heibel
22a6cb2dc0 Fix compilation when the memory layout is RowMajor. 2010-06-24 09:56:59 +02:00
Hauke Heibel
83f1b747e7 Fixed MSVC cpuid. 2010-06-24 09:55:53 +02:00
Gael Guennebaud
0a42f8c876 fix compilation when EIGEN_CPUD is not defined 2010-06-24 09:45:17 +02:00
Gael Guennebaud
8beb60bf63 fix EIGEN_CPUID for i386 & PIC, still remains to fix the MSVC version 2010-06-24 09:29:43 +02:00
Gael Guennebaud
546587c7d3 default to Intel's API by default 2010-06-23 17:14:06 +02:00
Gael Guennebaud
e1a6bad087 fix cache queries for non core2 CPU ;) 2010-06-23 16:34:51 +02:00
Gael Guennebaud
b4fe53f561 * makes all product use the new API to set the blocking sizes
* fix an issue preventing multithreading (now Dynamic = -1 ...)
2010-06-22 16:08:35 +02:00
Gael Guennebaud
fd9a9fa0ae slightly optimize computeProductBlockingSizes by explicitely precomputing what is known at compile time 2010-06-22 11:10:38 +02:00
Hauke Heibel
d132b5b061 Correct the options computation for RowMajor matrices. 2010-06-22 09:30:08 +02:00
Gael Guennebaud
98686ab86c fix in case we don't know how to query the L1/L2 cache sizes 2010-06-21 23:44:20 +02:00
Gael Guennebaud
0212eec23f simplify and optimize block sizes computation for matrix products. They
are now automatically computed from the L1 and L2 cache sizes which are
themselves automatically determined at runtime.
2010-06-21 23:28:50 +02:00
Gael Guennebaud
4cd38b333c make bench_gemm print out the queried cache sizes 2010-06-21 12:07:05 +02:00
Gael Guennebaud
e54635da11 add functions to query the L1 and L2/L3 cache sizes 2010-06-21 11:59:37 +02:00
Hauke Heibel
dc6ad5e25b More Index related stuff. 2010-06-21 11:36:00 +02:00
Gael Guennebaud
52165ba55a compilation fix 2010-06-20 22:27:35 +02:00
Hauke Heibel
f34eaa2628 Next try - more Index fixes. 2010-06-20 21:44:25 +02:00