Commit Graph

87 Commits

Author SHA1 Message Date
Gael Guennebaud
b3adc4face Add missing pconj specializations 2013-05-17 17:25:29 +02:00
Gael Guennebaud
d63712163c Add SSE4 min/max for integers 2013-03-20 18:28:40 +01:00
Gael Guennebaud
e8aa1f00c5 add SSE pexp function for double, make use of _mm_floor_p* for pexp with SSE4.1 2012-07-27 23:40:04 +02:00
Benoit Jacob
69124cfca2 Automatic relicensing to MPL2 using Keirs script. Manual fixup follows. 2012-07-13 14:42:47 -04:00
Jitse Niesen
3c412183b2 Get rid of include directives inside namespace blocks (bug #339). 2012-04-15 11:06:28 +01:00
Gael Guennebaud
634fedaf68 proper C++ casting 2012-01-31 18:56:25 +01:00
Gael Guennebaud
9c86ee2695 fix static inline versus inline static issues (the former is the correct order) 2012-01-31 12:58:52 +01:00
Gael Guennebaud
c331c092d5 no comment 2011-09-21 14:20:41 +02:00
Gael Guennebaud
7301f4345c quick workaround of MSVC9' ICE in pset1 2011-09-21 14:18:41 +02:00
Gael Guennebaud
c8e1b679fa re-enable fast pset1-pstore by introducing a new higher level pstore1 function 2011-03-02 10:55:44 +01:00
Benoit Jacob
eef03525b8 fix bug #203: revert to using _mm_set1_p[sd] 2011-02-28 00:04:05 -05:00
Benoit Jacob
9be2712bf7 remove now-useless comments 2011-02-27 22:35:17 -05:00
Benoit Jacob
0612768c1c fix bug #201: Clang too has intrinsics bugs preventing us to use custom unaligned loads 2011-02-27 21:59:07 -05:00
Benoit Jacob
b3544ce2ae bug #195 - fix this once and for all: just never use _mm_load_sd on gcc/i386, it generates redundant x87 ops 2011-02-27 17:26:59 -05:00
Benoit Jacob
5dfae4524b fix bug #195: fast unaligned load for integer using _mm_load_sd failed when the value interpreted as a NaN 2011-02-24 10:31:57 -05:00
Hauke Heibel
1a6597b8e4 MSVC does not like using uninitialized SSE variables, so we have to pass all zeros. 2011-02-12 21:29:16 +01:00
Gael Guennebaud
9d2bf35a05 implement optimized ploadu for MSVC10: this also fix bad code generation in gebp_kernel :) 2011-02-12 16:40:09 +01:00
Benoit Jacob
6a5a13e394 The pfirst hack is needed also on msvc 2010 as it gets completely nuts, even though it doesnt segfault as msvc 2008 did 2011-02-09 15:13:23 -05:00
Hauke Heibel
7bc8e3ac09 Initial fixes for bug #85.
Renamed meta_{true|false} to {true|false}_type, meta_if to conditional, is_same_type to is_same, un{ref|pointer|const} to remove_{reference|pointer|const} and makeconst to add_const.
Changed boolean type 'ret' member to 'value'.
Changed 'ret' members refering to types to 'type'.
Adapted all code occurences.
2010-10-25 22:13:49 +02:00
Benoit Jacob
4716040703 bug #86 : use internal:: namespace instead of ei_ prefix 2010-10-25 10:15:22 -04:00
Benoit Jacob
b80d9dd42e fix determination of number of registers on sse:
__i386__ was not defined by MSVC 2010.
fixed as (2*sizeof(void*)).
also move that to SSE/ and let the default for unknown arch's be just 8.
2010-08-13 13:55:28 -04:00
Gael Guennebaud
c2ee454df4 * fix compilation of mixed scalar product
* optimize mixed scalar products
2010-07-19 16:49:09 +02:00
Gael Guennebaud
f8aae7a908 * _mm_loaddup_pd is slow
* optimize SSE ei_ploaddup<Packet4f>
2010-07-19 15:43:27 +02:00
Gael Guennebaud
cd0e5dca9b wip: extend the gebp kernel to optimize complex and mixed products 2010-07-19 08:50:59 +02:00
Gael Guennebaud
ff96c94043 mixing types in product step 2:
* pload* and pset1 are now templated on the packet type
* gemv routines are now embeded into a structure with
  a consistent API with respect to gemm
* some configurations of vector * matrix and matrix * matrix works fine,
  some need more work...
2010-07-11 15:48:30 +02:00
Gael Guennebaud
4161b8be67 sync 2010-07-10 22:58:51 +02:00
Benoit Jacob
6dcd373b9d let ei_pset1 use _mm_loaddup_pd. Not a significant speed improvement, but also not a speed regression, and replaces 3 instructions by 1 single instruction. 2010-07-09 18:51:17 -04:00
Gael Guennebaud
96f9015807 disable MSVC optimization when the underlying compiler is ICC 2010-07-09 19:33:43 +02:00
Gael Guennebaud
300a226ffa scalars fitting in a single packet requires more work, step 1
* add a, Alignable trait
* update LinearVectorization assignment
2010-07-08 14:27:47 +02:00
Gael Guennebaud
dd18b22f0b optimize pmul for complex<double> 2010-07-07 15:29:04 +02:00
Gael Guennebaud
b0896382a3 s/IsVectorized/Vectorizable 2010-07-07 11:10:46 +02:00
Gael Guennebaud
bfa606d16f * add a IsVectorized mechanism (instead of packet-size>1...)
* vectorize complex<double>
2010-07-06 23:36:00 +02:00
Gael Guennebaud
d6454788d9 add support for vectorized conjugated products 2010-07-06 19:10:24 +02:00
Gael Guennebaud
c69a226192 * extend the Has* packet traits and makes all functor use it
* extend the packing routines to support conjugation
2010-07-05 23:27:54 +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
28e64b0da3 email change 2010-06-24 23:21:58 +02:00
Konstantinos Margaritis
9337f371d2 (proper commit this time)
replaced _mm_prefetch in GeneralBlockPanelKernel.h, with ei_prefetch() inline function.
Implemented NEON and AltiVec versions, copied SSE version over from GeneralBlockPanelKernel.h.
Also in GCC case (or rather !_MSC_VER) it's implemented using __builtin_prefetch().
NEON managed to give a small but welcome boost, 0.88GFLOPS -> 0.91GFLOPS.
2010-04-24 00:58:44 +03:00
Konstantinos Margaritis
5acf46bd12 Backed out changeset 6972c140f7 2010-04-24 00:57:10 +03:00
oem
6972c140f7 replaced _mm_prefetch in GeneralBlockPanelKernel.h, with ei_prefetch() inline function.
Implemented NEON and AltiVec versions, copied SSE version over from GeneralBlockPanelKernel.h.
Also in GCC case (or rather !_MSC_VER) it's implemented using __builtin_prefetch().
NEON managed to give a small but welcome boost, 0.88GFLOPS -> 0.91GFLOPS.
2010-04-24 00:44:14 +03:00
Hauke Heibel
9fe040ad29 Reintroduced the if-clause for MSVC ei_ploadu via _loadu_. 2010-03-07 14:05:26 +01:00
Gael Guennebaud
cefd9b8888 merge with default branch 2010-03-04 18:47:52 +01:00
Thomas Capricelli
0f3d69b65e Provide "eigen" defines to decide which instruction set is used
(sse3, ssse3 and sse4), independantly from the compiler.
Only those defines should be used in other places, and the user can
rely on those to know which sets are used.
2010-02-24 21:43:30 +01:00
Gael Guennebaud
eb905500b6 significant speedup in the matrix-matrix products 2010-02-23 13:06:49 +01:00
Hauke Heibel
4365a48748 Added an ei_linspaced_op to create linearly spaced vectors.
Added setLinSpaced/LinSpaced functionality to DenseBase.
Improved vectorized assignment - overcomes MSVC optimization issues.
CwiseNullaryOp is now requiring functors to offer 1D and 2D operators.
Adapted existing functors to the new CwiseNullaryOp requirements.
Added ei_plset to create packages as [a, a+1, ..., a+size].
Added more nullaray unit tests.
2010-01-26 19:42:17 +01:00
Hauke Heibel
325da2ea3c Fixed conservativeResize.
Fixed multiple overloads for operator=.
Removed debug output.
2010-01-11 13:57:50 +01:00
Gael Guennebaud
eaaba30cac merge with default branch 2009-12-22 22:51:08 +01:00
Gael Guennebaud
6db6774c46 * fix aliasing checks when the lhs is also transposed. At the same time,
significantly simplify the code of these checks while extending them
  to catch much more expressions!
* move the enabling/disabling of vectorized sin/cos to the architecture traits
2009-12-16 11:41:16 +01:00
Benoit Jacob
684d76eba3 add SSE4 support, start with integer multiplication 2009-11-24 15:12:43 -05:00
Gael Guennebaud
eb8f450071 Hey, finally the copyCoeff stuff is not only used to implement swap anymore :)
Add an internal pseudo expression allowing to optimize operators like +=, *= using
the copyCoeff stuff.
This allows to easily enforce aligned load for the destination matrix everywhere.
2009-11-20 15:39:38 +01:00
Benoit Jacob
92749eed11 * merge
* remove a ctor in QuaternionBase as it gives a strange error with GCC 4.4.2.
2009-11-09 09:08:03 -05:00