forked from AccelerateHS/accelerate
-
Notifications
You must be signed in to change notification settings - Fork 0
/
accelerate.cabal
246 lines (205 loc) · 9.77 KB
/
accelerate.cabal
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
Name: accelerate
Version: 0.9.0.0
Cabal-version: >= 1.6
Tested-with: GHC >= 6.12.3
Build-type: Simple
Synopsis: An embedded language for accelerated array processing
Description: This library defines an embedded language for
regular, multi-dimensional array computations with
multiple backends to facilitate high-performance
implementations. Currently, there are two backends:
(1) an interpreter that serves as a reference
implementation of the intended semantics of the
language and (2) a CUDA backend generating code for
CUDA-capable NVIDIA GPUs.
.
To use the CUDA backend, you need to have CUDA version 3.x
installed. The CUDA backend currently doesn't support 'Char'
and 'Bool' arrays.
.
Known bugs in this version:
<http://trac.haskell.org/accelerate/query?status=new&status=assigned&status=reopened&status=closed&version=0.8.1.0&order=priority>
.
* New in 0.8.1.0: bug fixes and some performance tweaks
.
* New in 0.8.0.0: 'replicate', 'slice' and 'foldSeg' supported in the
CUDA backend; frontend and interpreter support for 'stencil'; bug fixes
.
* New in 0.7.1.0: the CUDA backend and a number of scalar functions
.
License: BSD3
License-file: LICENSE
Author: Manuel M T Chakravarty,
Gabriele Keller,
Sean Lee,
Trevor L. McDonell
Maintainer: Manuel M T Chakravarty <chak@cse.unsw.edu.au>
Homepage: http://www.cse.unsw.edu.au/~chak/project/accelerate/
Bug-reports: http://trac.haskell.org/accelerate
Category: Compilers/Interpreters, Concurrency, Data
Stability: Experimental
-- Should be in the Library stanza, and only enabled for the CUDA backend,
-- but Cabal does not support that.
Data-files: cubits/accelerate_cuda_extras.h
cubits/accelerate_cuda_function.h
cubits/accelerate_cuda_shape.h
cubits/accelerate_cuda_stencil.h
cubits/accelerate_cuda_texture.h
cubits/accelerate_cuda_util.h
cubits/generate.inl
cubits/backpermute.inl
cubits/fold.inl
cubits/foldAll.inl
cubits/foldSeg.inl
cubits/map.inl
cubits/stencil.inl
cubits/stencil2.inl
cubits/permute.inl
cubits/reduce.inl
cubits/replicate.inl
cubits/scan.inl
cubits/scan1.inl
cubits/slice.inl
cubits/zipWith.inl
cubits/thrust/safe_scan_intervals.inl
cubits/thrust/inclusive_scan.inl
cubits/thrust/exclusive_scan.inl
Extra-source-files: INSTALL
include/accelerate.h
utils/README
utils/Paths_accelerate.hs
utils/dot_ghci
Flag llvm
Description: Enable the LLVM backend (sequential)
Default: False
Flag cuda
Description: Enable the CUDA parallel backend for NVIDIA GPUs
Default: True
Flag more-pp
Description: Enable HTML and Graphviz pretty printing.
Default: False
Flag pcache
Description: Enable the persistent caching of the compiled CUDA modules (experimental)
Default: False
Flag test-suite
Description: Export extra test modules
Default: False
Flag bounds-checks
Description: Enable bounds checking
Default: True
Flag unsafe-checks
Description: Enable bounds checking in unsafe operations
Default: False
Flag internal-checks
Description: Enable internal consistency checks
Default: False
Flag io
Description: Provide access to the block copy I/O functionality
Default: False
Flag inplace
Default: False
Library
Build-depends: array == 0.3.*,
base == 4.*,
containers >= 0.3 && < 0.5,
directory >= 1.0 && < 1.2,
ghc-prim == 0.2.*,
mtl == 2.0.*,
pretty == 1.0.*
Include-Dirs: include
if flag(llvm)
Build-depends: llvm >= 0.6.8
if flag(cuda)
Build-depends: binary == 0.5.*,
bytestring == 0.9.*,
cuda >= 0.2.2,
fclabels >= 0.9 && < 1.0,
filepath >= 1.0 && < 1.3,
language-c == 0.3.*,
transformers == 0.2.*,
unix == 2.4.*,
zlib == 0.5.*
if flag(io)
Build-depends: bytestring == 0.9.*
-- if flag(test-suite)
-- Build-depends: QuickCheck == 2.*
if flag(more-pp)
Build-depends: bytestring == 0.9.*,
blaze-html == 0.3.*,
text == 0.10.*
if flag(inplace)
hs-source-dirs: . utils
Exposed-modules: Data.Array.Accelerate
Data.Array.Accelerate.Interpreter
Data.Array.Accelerate.Array.Sugar
Data.Array.Accelerate.Array.Representation
Data.Array.Accelerate.Smart
Data.Array.Accelerate.AST
Data.Array.Accelerate.Array.Data
Data.Array.Accelerate.Tuple
Data.Array.Accelerate.Type
-- If flag(llvm)
-- Exposed-modules: Data.Array.Accelerate.LLVM
if flag(cuda)
Exposed-modules: Data.Array.Accelerate.CUDA
if flag(io)
Other-modules: Data.Array.Accelerate.IO.BlockCopy
Exposed-modules: Data.Array.Accelerate.IO
Data.Array.Accelerate.IO.Ptr
Data.Array.Accelerate.IO.ByteString
-- If flag(test-suite)
-- Exposed-modules: Data.Array.Accelerate.Test
-- Other-modules: Data.Array.Accelerate.Test.QuickCheck
-- Data.Array.Accelerate.Test.QuickCheck.Arbitrary
Other-modules: Data.Array.Accelerate.Internal.Check
Data.Array.Accelerate.Array.Delayed
Data.Array.Accelerate.Analysis.Type
Data.Array.Accelerate.Analysis.Shape
Data.Array.Accelerate.Analysis.Stencil
Data.Array.Accelerate.Debug
Data.Array.Accelerate.Language
Data.Array.Accelerate.Prelude
Data.Array.Accelerate.Pretty
Data.Array.Accelerate.Pretty.Print
Data.Array.Accelerate.Pretty.Traverse
Paths_accelerate
if flag(more-pp)
Other-modules: Data.Array.Accelerate.Pretty.HTML
Data.Array.Accelerate.Pretty.Graphviz
-- If flag(llvm)
-- Other-modules: Data.Array.Accelerate.LLVM.CodeGen
if flag(cuda)
CPP-options: -DACCELERATE_CUDA_BACKEND
Other-modules: Data.Array.Accelerate.CUDA.Analysis.Device
Data.Array.Accelerate.CUDA.Analysis.Hash
Data.Array.Accelerate.CUDA.Analysis.Launch
Data.Array.Accelerate.CUDA.Array.Data
Data.Array.Accelerate.CUDA.CodeGen.Data
Data.Array.Accelerate.CUDA.CodeGen.Skeleton
Data.Array.Accelerate.CUDA.CodeGen.Stencil
Data.Array.Accelerate.CUDA.CodeGen.Tuple
Data.Array.Accelerate.CUDA.CodeGen.Util
Data.Array.Accelerate.CUDA.CodeGen
Data.Array.Accelerate.CUDA.Compile
Data.Array.Accelerate.CUDA.Execute
Data.Array.Accelerate.CUDA.State
if flag(pcache)
CPP-options: -DACCELERATE_CUDA_PERSISTENT_CACHE
if flag(bounds-checks)
cpp-options: -DACCELERATE_BOUNDS_CHECKS
if flag(unsafe-checks)
cpp-options: -DACCELERATE_UNSAFE_CHECKS
if flag(internal-checks)
cpp-options: -DACCELERATE_INTERNAL_CHECKS
ghc-options: -O2 -Wall -funbox-strict-fields -fno-warn-name-shadowing
if impl(ghc >= 7.0)
ghc-options: -fspec-constr-count=25
Extensions: BangPatterns, CPP, DeriveDataTypeable, EmptyDataDecls,
FlexibleContexts, FlexibleInstances, GADTs, MagicHash,
MultiParamTypeClasses, PatternGuards, RankNTypes,
ScopedTypeVariables, StandaloneDeriving,
TemplateHaskell, TupleSections, TypeFamilies,
TypeOperators, TypeSynonymInstances, UnboxedTuples
source-repository head
type: darcs
location: http://code.haskell.org/accelerate