Mercurial > hg > truffle

annotate src/gpu/ptx/vm/ptxKernelArguments.hpp @ 13843:f5f81bc9c9f0

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Align successor when deleting aligned empty block.
author Roland Schatz <roland.schatz@oracle.com>
date Fri, 31 Jan 2014 11:38:32 +0100
parents 220ed109bf77
children
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
rev   line source
11485
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
1 /*
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
2 * Copyright (c) 2013, Oracle and/or its affiliates. All rights reserved.
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
3 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
4 *
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
5 * This code is free software; you can redistribute it and/or modify it
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
6 * under the terms of the GNU General Public License version 2 only, as
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
7 * published by the Free Software Foundation.
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
8 *
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
9 * This code is distributed in the hope that it will be useful, but WITHOUT
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
10 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
11 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
12 * version 2 for more details (a copy is included in the LICENSE file that
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
13 * accompanied this code).
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
14 *
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
15 * You should have received a copy of the GNU General Public License version
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
16 * 2 along with this work; if not, write to the Free Software Foundation,
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
17 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
18 *
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
19 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
20 * or visit www.oracle.com if you need additional information or have any
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
21 * questions.
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
22 *
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
23 */
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
24
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
25 #ifndef KERNEL_ARGUMENTS_PTX_HPP
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
26 #define KERNEL_ARGUMENTS_PTX_HPP
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
27
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
28 #include "runtime/gpu.hpp"
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
29 #include "runtime/signature.hpp"
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
30
11902
67a1e27a8dbb PTX initial float and double
Morris Meyer <morris.meyer@oracle.com>
parents: 11901
diff changeset
31 #define T_BYTE_SIZE 1
67a1e27a8dbb PTX initial float and double
Morris Meyer <morris.meyer@oracle.com>
parents: 11901
diff changeset
32 #define T_BOOLEAN_SIZE 4
67a1e27a8dbb PTX initial float and double
Morris Meyer <morris.meyer@oracle.com>
parents: 11901
diff changeset
33 #define T_INT_BYTE_SIZE 4
67a1e27a8dbb PTX initial float and double
Morris Meyer <morris.meyer@oracle.com>
parents: 11901
diff changeset
34 #define T_FLOAT_BYTE_SIZE 4
67a1e27a8dbb PTX initial float and double
Morris Meyer <morris.meyer@oracle.com>
parents: 11901
diff changeset
35 #define T_DOUBLE_BYTE_SIZE 8
67a1e27a8dbb PTX initial float and double
Morris Meyer <morris.meyer@oracle.com>
parents: 11901
diff changeset
36 #define T_LONG_BYTE_SIZE 8
13624
220ed109bf77 initial code for calling PTX kernel code from Java with parameter marshaling and return value unmarshaling performed by a wrapper specified via manual graph construction
Doug Simon <doug.simon@oracle.com>
parents: 12653
diff changeset
37 #define T_OBJECT_BYTE_SIZE 8
11902
67a1e27a8dbb PTX initial float and double
Morris Meyer <morris.meyer@oracle.com>
parents: 11901
diff changeset
38 #define T_ARRAY_BYTE_SIZE 8
11485
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
39
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
40 class PTXKernelArguments : public SignatureIterator {
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
41 public:
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
42 // Buffer holding CUdeviceptr values that represent the kernel arguments
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
43 char _kernelArgBuffer[1024];
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
44 // Current offset into _kernelArgBuffer
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
45 size_t _bufferOffset;
12519
f020e149c1b6 PTX codegen enhancements; fixes to PTX test regressions.
S.Bharadwaj Yadavalli <bharadwaj.yadavalli@oracle.com>
parents: 11902
diff changeset
46 // Device pointer holding return value
f020e149c1b6 PTX codegen enhancements; fixes to PTX test regressions.
S.Bharadwaj Yadavalli <bharadwaj.yadavalli@oracle.com>
parents: 11902
diff changeset
47 gpu::Ptx::CUdeviceptr _dev_return_value;
f020e149c1b6 PTX codegen enhancements; fixes to PTX test regressions.
S.Bharadwaj Yadavalli <bharadwaj.yadavalli@oracle.com>
parents: 11902
diff changeset
48
11485
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
49 private:
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
50 // Array of java argument oops
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
51 arrayOop _args;
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
52 // Current index into _args
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
53 int _index;
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
54 // Flag to indicate successful creation of kernel argument buffer
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
55 bool _success;
11821
d8659ad83fcc PTX single-threaded array store, Warp annotation
Morris Meyer <morris.meyer@oracle.com>
parents: 11596
diff changeset
56
11485
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
57 // Get next java argument
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
58 oop next_arg(BasicType expectedType);
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
59
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
60 public:
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
61 PTXKernelArguments(Symbol* signature, arrayOop args, bool is_static) : SignatureIterator(signature) {
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
62 this->_return_type = T_ILLEGAL;
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
63 _index = 0;
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
64 _args = args;
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
65 _success = true;
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
66 _bufferOffset = 0;
12519
f020e149c1b6 PTX codegen enhancements; fixes to PTX test regressions.
S.Bharadwaj Yadavalli <bharadwaj.yadavalli@oracle.com>
parents: 11902
diff changeset
67 _dev_return_value = 0;
11485
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
68 if (!is_static) {
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
69 // TODO : Create a device argument for receiver object and add it to _kernelBuffer
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
70 tty->print_cr("{CUDA] ****** TODO: Support for execution of non-static java methods not implemented yet.");
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
71 }
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
72 // Iterate over the entire signature
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
73 iterate();
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
74 assert((_success && (_index == args->length())), "arg count mismatch with signature");
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
75 }
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
76
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
77 inline char* device_argument_buffer() {
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
78 return _kernelArgBuffer;
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
79 }
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
80
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
81 inline size_t device_argument_buffer_size() {
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
82 return _bufferOffset;
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
83 }
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
84
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
85 // Get the return oop value
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
86 oop get_return_oop();
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
87
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
88 // get device return value ptr
12519
f020e149c1b6 PTX codegen enhancements; fixes to PTX test regressions.
S.Bharadwaj Yadavalli <bharadwaj.yadavalli@oracle.com>
parents: 11902
diff changeset
89 gpu::Ptx::CUdeviceptr get_dev_return_value() {
f020e149c1b6 PTX codegen enhancements; fixes to PTX test regressions.
S.Bharadwaj Yadavalli <bharadwaj.yadavalli@oracle.com>
parents: 11902
diff changeset
90 return _dev_return_value;
11485
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
91 }
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
92
12653
1a7e7011a341 * PTX kernel argument buffer now has naturally aligned arguments as required by PTX JIT compiler.
S.Bharadwaj Yadavalli <bharadwaj.yadavalli@oracle.com>
parents: 12519
diff changeset
93 /*
1a7e7011a341 * PTX kernel argument buffer now has naturally aligned arguments as required by PTX JIT compiler.
S.Bharadwaj Yadavalli <bharadwaj.yadavalli@oracle.com>
parents: 12519
diff changeset
94 * Pad kernel argument buffer to naturally align for given size.
1a7e7011a341 * PTX kernel argument buffer now has naturally aligned arguments as required by PTX JIT compiler.
S.Bharadwaj Yadavalli <bharadwaj.yadavalli@oracle.com>
parents: 12519
diff changeset
95 */
1a7e7011a341 * PTX kernel argument buffer now has naturally aligned arguments as required by PTX JIT compiler.
S.Bharadwaj Yadavalli <bharadwaj.yadavalli@oracle.com>
parents: 12519
diff changeset
96 void pad_kernel_argument_buffer(size_t);
11596
91e5f927af63 Initial implementation of PTXRuntime (RegisterConfig, PTX description etc); guarded with new flag UseGPU. Specify -XX:+UseGPU to exercise this new implementation.
bharadwaj
parents: 11527
diff changeset
97
11485
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
98 void do_byte();
11901
61767ccd4600 PTX boolean return value, emitIntegerTestMove, warnings
Morris Meyer <morris.meyer@oracle.com>
parents: 11894
diff changeset
99 void do_bool();
11485
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
100 void do_int();
11894
c7abc8411011 Fixed BasicPTXTest and IntegerPTXTest
Morris Meyer <morris.meyer@oracle.com>
parents: 11821
diff changeset
101 void do_float();
11902
67a1e27a8dbb PTX initial float and double
Morris Meyer <morris.meyer@oracle.com>
parents: 11901
diff changeset
102 void do_double();
11485
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
103 void do_long();
11821
d8659ad83fcc PTX single-threaded array store, Warp annotation
Morris Meyer <morris.meyer@oracle.com>
parents: 11596
diff changeset
104 void do_array(int begin, int end);
d8659ad83fcc PTX single-threaded array store, Warp annotation
Morris Meyer <morris.meyer@oracle.com>
parents: 11596
diff changeset
105 void do_void();
11485
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
106
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
107 inline void do_char() {
11596
91e5f927af63 Initial implementation of PTXRuntime (RegisterConfig, PTX description etc); guarded with new flag UseGPU. Specify -XX:+UseGPU to exercise this new implementation.
bharadwaj
parents: 11527
diff changeset
108 /* TODO : To be implemented */
11821
d8659ad83fcc PTX single-threaded array store, Warp annotation
Morris Meyer <morris.meyer@oracle.com>
parents: 11596
diff changeset
109 guarantee(false, "do_char:NYI");
11485
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
110 }
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
111 inline void do_short() {
11596
91e5f927af63 Initial implementation of PTXRuntime (RegisterConfig, PTX description etc); guarded with new flag UseGPU. Specify -XX:+UseGPU to exercise this new implementation.
bharadwaj
parents: 11527
diff changeset
112 /* TODO : To be implemented */
11821
d8659ad83fcc PTX single-threaded array store, Warp annotation
Morris Meyer <morris.meyer@oracle.com>
parents: 11596
diff changeset
113 guarantee(false, "do_short:NYI");
11485
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
114 }
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
115 inline void do_object() {
11596
91e5f927af63 Initial implementation of PTXRuntime (RegisterConfig, PTX description etc); guarded with new flag UseGPU. Specify -XX:+UseGPU to exercise this new implementation.
bharadwaj
parents: 11527
diff changeset
116 /* TODO : To be implemented */
11821
d8659ad83fcc PTX single-threaded array store, Warp annotation
Morris Meyer <morris.meyer@oracle.com>
parents: 11596
diff changeset
117 guarantee(false, "do_object:NYI");
11485
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
118 }
11821
d8659ad83fcc PTX single-threaded array store, Warp annotation
Morris Meyer <morris.meyer@oracle.com>
parents: 11596
diff changeset
119
11485
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
120 inline void do_object(int begin, int end) {
11596
91e5f927af63 Initial implementation of PTXRuntime (RegisterConfig, PTX description etc); guarded with new flag UseGPU. Specify -XX:+UseGPU to exercise this new implementation.
bharadwaj
parents: 11527
diff changeset
121 /* TODO : To be implemented */
11821
d8659ad83fcc PTX single-threaded array store, Warp annotation
Morris Meyer <morris.meyer@oracle.com>
parents: 11596
diff changeset
122 guarantee(false, "do_object(II):NYI");
11485
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
123 }
11821
d8659ad83fcc PTX single-threaded array store, Warp annotation
Morris Meyer <morris.meyer@oracle.com>
parents: 11596
diff changeset
124
11485
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
125 };
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
126
49bb1bc983c6 Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
bharadwaj
parents:
diff changeset
127 #endif // KERNEL_ARGUMENTS_HPP