NEWS.txt 2.89 KB
Newer Older
COULAUD Olivier's avatar
COULAUD Olivier committed
1
Copyright (c) 2011-2014 Inria,   All rights reserved.
PIACIBELLO Cyrille's avatar
PIACIBELLO Cyrille committed
2 3


COULAUD Olivier's avatar
COULAUD Olivier committed
4 5 6 7
===========================================================================

This file contains the main features as well as overviews of specific
bug fixes (and other actions) for each version of ScalFMM since
COULAUD Olivier's avatar
COULAUD Olivier committed
8
version 1.1
9

10 11
2.0
-----
COULAUD Olivier's avatar
COULAUD Olivier committed
12
- New API for Kernels (The same kernel can be used witth all Octree (Classical, block, Adaptive)
13
- Adaptive FMM is available in shared memory paradigm (OpenMP and StarPU)
COULAUD Olivier's avatar
COULAUD Olivier committed
14
- scoped enum (FParticleTYpe (target,source), OPM4_Priorities, ...) is used
15
- StarPU works with Duplicated and Let trees with MPI (STF approach)
16
- Remove all warnings with gcc 7.2
17
- std::complex is used for equispaced points interpolation
18
- Fix bug with intel compiler and target/source 
COULAUD Olivier's avatar
COULAUD Olivier committed
19 20
- Improvement in CmakeLists  - FUSE is working again -  
- Compile with Intel and CLang (with a lot of warnings)
21
- Now morse_cmake is a git submodule 
22
-
COULAUD Olivier's avatar
COULAUD Olivier committed
23

COULAUD Olivier's avatar
COULAUD Olivier committed
24

COULAUD Olivier's avatar
COULAUD Olivier committed
25 26
1.5
-----
COULAUD Olivier's avatar
COULAUD Olivier committed
27
- improvements of the MPI version of ScalFMM 
28 29
- ScalFMM on StarPU (runtime system) is available 
- Add GPU version for ufmm (interpolation FMM based on equispaced grid points)
COULAUD Olivier's avatar
COULAUD Olivier committed
30
- Fix bug in distribution generation
COULAUD Olivier's avatar
COULAUD Olivier committed
31 32
-

33 34
1.4
-----
35
- Add interpolation FMM based on equispaced grid points (ufmm)
36
- Add the group tree and the things related to starpu (runtime system)
37
- Change the P2P (it is now done by tree functions in the abstract kernel) refer to the P2P page of the doc
38 39
- Update the P2P/M2L prototype, the position of the elements is known from an array position
- Fix bug in SSE, AVX, ...  kernel 1/r 
40 41

1.3
COULAUD Olivier's avatar
COULAUD Olivier committed
42
-----
COULAUD Olivier's avatar
COULAUD Olivier committed
43 44 45
- The choice of floating point values is done by template (by passing template to all classes) and no more at configure time
- There is no more ScalFMM variables and defines but only SCALFMM (all letters in upper case)
- We start to use unique ptr (and advise to use it) in the main files
COULAUD Olivier's avatar
COULAUD Olivier committed
46
- Add blocked version of the algorithm to increase the granularity (task-based approach)
COULAUD Olivier's avatar
.  
COULAUD Olivier committed
47 48
- Fix bug with Chebyschev kernel and multi RHS.
- Improvements in unit tests (works with Linux and MacOs X for intel and gcc)
COULAUD Olivier's avatar
COULAUD Olivier committed
49 50
- Use MORSE distribution to detect BLAS, FFTW, ....
- BLAS and FFTW are turned on by default, and the compilation continue even if they are not found
COULAUD Olivier's avatar
COULAUD Olivier committed
51

COULAUD Olivier's avatar
COULAUD Olivier committed
52 53 54 55
1.2.1
-----
- Bug fix : Support for huge MPI message in tree construction and Parallel QuickSort (count can be greater than Int32.MaxValue)
- Bug fix : Data sharing attribute clauses for omp in Core/FAlgorithmThreadProc.hpp
COULAUD Olivier's avatar
COULAUD Olivier committed
56

COULAUD Olivier's avatar
COULAUD Olivier committed
57
1.2
COULAUD Olivier's avatar
COULAUD Olivier committed
58
-----
COULAUD Olivier's avatar
COULAUD Olivier committed
59 60
- New FMA format to read/write particles
- Add examples repository
PIACIBELLO Cyrille's avatar
PIACIBELLO Cyrille committed
61 62
- Add distributions generator
- Fix bugs in periodic model
COULAUD Olivier's avatar
COULAUD Olivier committed
63
- Fix bug and performance improvements in MPI thread algorithms
PIACIBELLO Cyrille's avatar
PIACIBELLO Cyrille committed
64
- Now we use MPI_THREAD_SERIALIZED rather than MPI_THREAD_MULTIPLE in MPI+OpenMP algorithm.
COULAUD Olivier's avatar
COULAUD Olivier committed
65
- Improve tests -> now we use CTest and CDash (Thanks to  F Rue  (Inria))
COULAUD Olivier's avatar
COULAUD Olivier committed
66
- Add support for Intel icl++ compiler (Intel 2015 compiler) on MacOS X and Clang on linux system
COULAUD Olivier's avatar
COULAUD Olivier committed
67
- Add SSE and AVX support in CMake
PIACIBELLO Cyrille's avatar
PIACIBELLO Cyrille committed
68
- Add SSE and AVX code for 1/r kernel
COULAUD Olivier's avatar
COULAUD Olivier committed
69
- CMake improvements