1 |
root |
1.1 |
=head1 NAME |
2 |
|
|
|
3 |
root |
1.5 |
OpenCL - Open Computing Language Bindings |
4 |
root |
1.1 |
|
5 |
|
|
=head1 SYNOPSIS |
6 |
|
|
|
7 |
|
|
use OpenCL; |
8 |
|
|
|
9 |
|
|
=head1 DESCRIPTION |
10 |
|
|
|
11 |
root |
1.7 |
This is an early release which might be useful, but hasn't seen much testing. |
12 |
root |
1.1 |
|
13 |
root |
1.9 |
=head2 OpenCL FROM 10000 FEET HEIGHT |
14 |
|
|
|
15 |
|
|
Here is a high level overview of OpenCL: |
16 |
|
|
|
17 |
|
|
First you need to find one or more OpenCL::Platforms (kind of like |
18 |
|
|
vendors) - usually there is only one. |
19 |
|
|
|
20 |
|
|
Each platform gives you access to a number of OpenCL::Device objects, e.g. |
21 |
|
|
your graphics card. |
22 |
|
|
|
23 |
root |
1.11 |
From a platform and some device(s), you create an OpenCL::Context, which is |
24 |
root |
1.9 |
a very central object in OpenCL: Once you have a context you can create |
25 |
|
|
most other objects: |
26 |
|
|
|
27 |
root |
1.11 |
OpenCL::Program objects, which store source code and, after building for a |
28 |
|
|
specific device ("compiling and linking"), also binary programs. For each |
29 |
|
|
kernel function in a program you can then create an OpenCL::Kernel object |
30 |
|
|
which represents basically a function call with argument values. |
31 |
root |
1.9 |
|
32 |
|
|
OpenCL::Memory objects of various flavours: OpenCL::Buffers objects (flat |
33 |
|
|
memory areas, think array) and OpenCL::Image objects (think 2d or 3d |
34 |
|
|
array) for bulk data and input and output for kernels. |
35 |
|
|
|
36 |
|
|
OpenCL::Sampler objects, which are kind of like texture filter modes in |
37 |
|
|
OpenGL. |
38 |
|
|
|
39 |
|
|
OpenCL::Queue objects - command queues, which allow you to submit memory |
40 |
|
|
reads, writes and copies, as well as kernel calls to your devices. They |
41 |
|
|
also offer a variety of methods to synchronise request execution, for |
42 |
|
|
example with barriers or OpenCL::Event objects. |
43 |
|
|
|
44 |
|
|
OpenCL::Event objects are used to signal when something is complete. |
45 |
|
|
|
46 |
|
|
=head2 HELPFUL RESOURCES |
47 |
root |
1.3 |
|
48 |
root |
1.5 |
The OpenCL spec used to develop this module (1.2 spec was available, but |
49 |
root |
1.3 |
no implementation was available to me :). |
50 |
|
|
|
51 |
|
|
http://www.khronos.org/registry/cl/specs/opencl-1.1.pdf |
52 |
|
|
|
53 |
|
|
OpenCL manpages: |
54 |
|
|
|
55 |
|
|
http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/ |
56 |
|
|
|
57 |
root |
1.9 |
=head1 BASIC WORKFLOW |
58 |
|
|
|
59 |
root |
1.11 |
To get something done, you basically have to do this once (refer to the |
60 |
|
|
examples below for actual code, this is just a high-level description): |
61 |
root |
1.9 |
|
62 |
root |
1.11 |
Find some platform (e.g. the first one) and some device(s) (e.g. the first |
63 |
|
|
device of the platform), and create a context from those. |
64 |
root |
1.9 |
|
65 |
root |
1.11 |
Create program objects from your OpenCL source code, then build (compile) |
66 |
|
|
the programs for each device you want to run them on. |
67 |
root |
1.9 |
|
68 |
root |
1.11 |
Create kernel objects for all kernels you want to use (surprisingly, these |
69 |
|
|
are not device-specific). |
70 |
root |
1.9 |
|
71 |
root |
1.11 |
Then, to execute stuff, you repeat these steps, possibly resuing or |
72 |
|
|
sharing some buffers: |
73 |
root |
1.9 |
|
74 |
root |
1.11 |
Create some input and output buffers from your context. Set these as |
75 |
|
|
arguments to your kernel. |
76 |
|
|
|
77 |
|
|
Enqueue buffer writes to initialise your input buffers (when not |
78 |
|
|
initialised at creation time). |
79 |
root |
1.9 |
|
80 |
|
|
Enqueue the kernel execution. |
81 |
|
|
|
82 |
|
|
Enqueue buffer reads for your output buffer to read results. |
83 |
|
|
|
84 |
root |
1.3 |
=head1 EXAMPLES |
85 |
|
|
|
86 |
root |
1.5 |
=head2 Enumerate all devices and get contexts for them. |
87 |
root |
1.1 |
|
88 |
root |
1.11 |
Best run this once to get a feel for the platforms and devices in your |
89 |
|
|
system. |
90 |
|
|
|
91 |
root |
1.1 |
for my $platform (OpenCL::platforms) { |
92 |
root |
1.10 |
printf "platform: %s\n", $platform->info (OpenCL::PLATFORM_NAME); |
93 |
|
|
printf "extensions: %s\n", $platform->info (OpenCL::PLATFORM_EXTENSIONS); |
94 |
root |
1.1 |
for my $device ($platform->devices) { |
95 |
root |
1.10 |
printf "+ device: %s\n", $device->info (OpenCL::DEVICE_NAME); |
96 |
|
|
my $ctx = $device->context; |
97 |
root |
1.1 |
# do stuff |
98 |
|
|
} |
99 |
|
|
} |
100 |
|
|
|
101 |
root |
1.5 |
=head2 Get a useful context and a command queue. |
102 |
root |
1.1 |
|
103 |
root |
1.11 |
This is a useful boilerplate for any OpenCL program that only wants to use |
104 |
|
|
one device, |
105 |
|
|
|
106 |
|
|
my ($platform) = OpenCL::platforms; # find first platform |
107 |
|
|
my ($dev) = $platform->devices; # find first device of platform |
108 |
|
|
my $ctx = $platform->context (undef, [$dev]); # create context out of those |
109 |
|
|
my $queue = $ctx->queue ($dev); # create a command queue for the device |
110 |
root |
1.1 |
|
111 |
root |
1.5 |
=head2 Print all supported image formats of a context. |
112 |
|
|
|
113 |
root |
1.11 |
Best run this once for your context, to see whats available and how to |
114 |
|
|
gather information. |
115 |
|
|
|
116 |
root |
1.5 |
for my $type (OpenCL::MEM_OBJECT_IMAGE2D, OpenCL::MEM_OBJECT_IMAGE3D) { |
117 |
root |
1.10 |
print "supported image formats for ", OpenCL::enum2str $type, "\n"; |
118 |
root |
1.5 |
|
119 |
|
|
for my $f ($ctx->supported_image_formats (0, $type)) { |
120 |
|
|
printf " %-10s %-20s\n", OpenCL::enum2str $f->[0], OpenCL::enum2str $f->[1]; |
121 |
|
|
} |
122 |
|
|
} |
123 |
|
|
|
124 |
|
|
=head2 Create a buffer with some predefined data, read it back synchronously, |
125 |
|
|
then asynchronously. |
126 |
root |
1.3 |
|
127 |
|
|
my $buf = $ctx->buffer_sv (OpenCL::MEM_COPY_HOST_PTR, "helmut"); |
128 |
|
|
|
129 |
|
|
$queue->enqueue_read_buffer ($buf, 1, 1, 3, my $data); |
130 |
root |
1.10 |
print "$data\n"; |
131 |
root |
1.3 |
|
132 |
|
|
my $ev = $queue->enqueue_read_buffer ($buf, 0, 1, 3, my $data); |
133 |
|
|
$ev->wait; |
134 |
root |
1.10 |
print "$data\n"; # prints "elm" |
135 |
root |
1.3 |
|
136 |
root |
1.5 |
=head2 Create and build a program, then create a kernel out of one of its |
137 |
|
|
functions. |
138 |
root |
1.3 |
|
139 |
|
|
my $src = ' |
140 |
|
|
__kernel void |
141 |
|
|
squareit (__global float *input, __global float *output) |
142 |
|
|
{ |
143 |
|
|
size_t id = get_global_id (0); |
144 |
|
|
output [id] = input [id] * input [id]; |
145 |
|
|
} |
146 |
|
|
'; |
147 |
|
|
|
148 |
|
|
my $prog = $ctx->program_with_source ($src); |
149 |
|
|
|
150 |
root |
1.11 |
# build croaks on compile errors, so catch it and print the compile errors |
151 |
root |
1.3 |
eval { $prog->build ($dev); 1 } |
152 |
|
|
or die $prog->build_info ($dev, OpenCL::PROGRAM_BUILD_LOG); |
153 |
|
|
|
154 |
|
|
my $kernel = $prog->kernel ("squareit"); |
155 |
|
|
|
156 |
root |
1.11 |
=head2 Create some input and output float buffers, then call the |
157 |
|
|
'squareit' kernel on them. |
158 |
root |
1.4 |
|
159 |
|
|
my $input = $ctx->buffer_sv (OpenCL::MEM_COPY_HOST_PTR, pack "f*", 1, 2, 3, 4.5); |
160 |
|
|
my $output = $ctx->buffer (0, OpenCL::SIZEOF_FLOAT * 5); |
161 |
|
|
|
162 |
|
|
# set buffer |
163 |
|
|
$kernel->set_buffer (0, $input); |
164 |
|
|
$kernel->set_buffer (1, $output); |
165 |
|
|
|
166 |
|
|
# execute it for all 4 numbers |
167 |
|
|
$queue->enqueue_nd_range_kernel ($kernel, undef, [4], undef); |
168 |
|
|
|
169 |
root |
1.5 |
# enqueue a synchronous read |
170 |
|
|
$queue->enqueue_read_buffer ($output, 1, 0, OpenCL::SIZEOF_FLOAT * 4, my $data); |
171 |
|
|
|
172 |
|
|
# print the results: |
173 |
root |
1.10 |
printf "%s\n", join ", ", unpack "f*", $data; |
174 |
root |
1.5 |
|
175 |
|
|
=head2 The same enqueue operations as before, but assuming an out-of-order queue, |
176 |
|
|
showing off barriers. |
177 |
|
|
|
178 |
|
|
# execute it for all 4 numbers |
179 |
|
|
$queue->enqueue_nd_range_kernel ($kernel, undef, [4], undef); |
180 |
|
|
|
181 |
|
|
# enqueue a barrier to ensure in-order execution |
182 |
root |
1.4 |
$queue->enqueue_barrier; |
183 |
|
|
|
184 |
root |
1.5 |
# enqueue an async read |
185 |
|
|
$queue->enqueue_read_buffer ($output, 0, 0, OpenCL::SIZEOF_FLOAT * 4, my $data); |
186 |
|
|
|
187 |
|
|
# wait for all requests to finish |
188 |
|
|
$queue->finish; |
189 |
|
|
|
190 |
|
|
=head2 The same enqueue operations as before, but assuming an out-of-order queue, |
191 |
|
|
showing off event objects and wait lists. |
192 |
|
|
|
193 |
|
|
# execute it for all 4 numbers |
194 |
|
|
my $ev = $queue->enqueue_nd_range_kernel ($kernel, undef, [4], undef); |
195 |
|
|
|
196 |
|
|
# enqueue an async read |
197 |
|
|
$ev = $queue->enqueue_read_buffer ($output, 0, 0, OpenCL::SIZEOF_FLOAT * 4, my $data, $ev); |
198 |
|
|
|
199 |
|
|
# wait for the last event to complete |
200 |
root |
1.4 |
$ev->wait; |
201 |
|
|
|
202 |
root |
1.5 |
=head1 DOCUMENTATION |
203 |
|
|
|
204 |
|
|
=head2 BASIC CONVENTIONS |
205 |
|
|
|
206 |
root |
1.14 |
This is not a one-to-one C-style translation of OpenCL to Perl - instead |
207 |
|
|
I attempted to make the interface as type-safe as possible by introducing |
208 |
root |
1.5 |
object syntax where it makes sense. There are a number of important |
209 |
|
|
differences between the OpenCL C API and this module: |
210 |
|
|
|
211 |
|
|
=over 4 |
212 |
|
|
|
213 |
|
|
=item * Object lifetime managament is automatic - there is no need |
214 |
|
|
to free objects explicitly (C<clReleaseXXX>), the release function |
215 |
|
|
is called automatically once all Perl references to it go away. |
216 |
|
|
|
217 |
|
|
=item * OpenCL uses CamelCase for function names (C<clGetPlatformInfo>), |
218 |
|
|
while this module uses underscores as word separator and often leaves out |
219 |
|
|
prefixes (C<< $platform->info >>). |
220 |
|
|
|
221 |
|
|
=item * OpenCL often specifies fixed vector function arguments as short |
222 |
|
|
arrays (C<size_t origin[3]>), while this module explicitly expects the |
223 |
|
|
components as separate arguments- |
224 |
|
|
|
225 |
root |
1.13 |
=item * Structures are often specified with their components, and returned |
226 |
|
|
as arrayrefs. |
227 |
|
|
|
228 |
root |
1.9 |
=item * Where possible, one of the pitch values is calculated from the |
229 |
|
|
perl scalar length and need not be specified. |
230 |
root |
1.5 |
|
231 |
|
|
=item * When enqueuing commands, the wait list is specified by adding |
232 |
root |
1.9 |
extra arguments to the function - anywhere a C<$wait_events...> argument |
233 |
root |
1.5 |
is documented this can be any number of event objects. |
234 |
|
|
|
235 |
|
|
=item * When enqueuing commands, if the enqueue method is called in void |
236 |
|
|
context, no event is created. In all other contexts an event is returned |
237 |
|
|
by the method. |
238 |
|
|
|
239 |
|
|
=item * This module expects all functions to return C<CL_SUCCESS>. If any |
240 |
|
|
other status is returned the function will throw an exception, so you |
241 |
|
|
don't normally have to to any error checking. |
242 |
|
|
|
243 |
|
|
=back |
244 |
|
|
|
245 |
root |
1.7 |
=head2 PERL AND OPENCL TYPES |
246 |
|
|
|
247 |
root |
1.8 |
This handy(?) table lists OpenCL types and their perl, PDL and pack/unpack |
248 |
root |
1.7 |
format equivalents: |
249 |
|
|
|
250 |
root |
1.8 |
OpenCL perl PDL pack/unpack |
251 |
|
|
char IV - c |
252 |
|
|
uchar IV byte C |
253 |
|
|
short IV short s |
254 |
|
|
ushort IV ushort S |
255 |
|
|
int IV long? l |
256 |
|
|
uint IV - L |
257 |
|
|
long IV longlong q |
258 |
|
|
ulong IV - Q |
259 |
|
|
float NV float f |
260 |
|
|
half IV ushort S |
261 |
|
|
double NV double d |
262 |
root |
1.7 |
|
263 |
root |
1.5 |
=head2 THE OpenCL PACKAGE |
264 |
|
|
|
265 |
|
|
=over 4 |
266 |
|
|
|
267 |
|
|
=item $int = OpenCL::errno |
268 |
|
|
|
269 |
root |
1.11 |
The last error returned by a function - it's only valid after an error occured |
270 |
|
|
and before calling another OpenCL function. |
271 |
root |
1.5 |
|
272 |
|
|
=item $str = OpenCL::err2str $errval |
273 |
|
|
|
274 |
|
|
Comverts an error value into a human readable string. |
275 |
|
|
|
276 |
root |
1.9 |
=item $str = OpenCL::enum2str $enum |
277 |
root |
1.5 |
|
278 |
|
|
Converts most enum values (inof parameter names, image format constants, |
279 |
|
|
object types, addressing and filter modes, command types etc.) into a |
280 |
|
|
human readbale string. When confronted with some random integer it can be |
281 |
|
|
very helpful to pass it through this function to maybe get some readable |
282 |
|
|
string out of it. |
283 |
|
|
|
284 |
|
|
=item @platforms = OpenCL::platforms |
285 |
|
|
|
286 |
|
|
Returns all available OpenCL::Platform objects. |
287 |
|
|
|
288 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetPlatformIDs.html> |
289 |
|
|
|
290 |
root |
1.10 |
=item $ctx = OpenCL::context_from_type $properties, $type = OpenCL::DEVICE_TYPE_DEFAULT, $notify = undef |
291 |
root |
1.5 |
|
292 |
|
|
Tries to create a context from a default device and platform - never worked for me. |
293 |
|
|
|
294 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html> |
295 |
|
|
|
296 |
|
|
=item OpenCL::wait_for_events $wait_events... |
297 |
|
|
|
298 |
|
|
Waits for all events to complete. |
299 |
|
|
|
300 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clWaitForEvents.html> |
301 |
|
|
|
302 |
|
|
=back |
303 |
|
|
|
304 |
|
|
=head2 THE OpenCL::Platform CLASS |
305 |
|
|
|
306 |
|
|
=over 4 |
307 |
|
|
|
308 |
|
|
=item $packed_value = $platform->info ($name) |
309 |
|
|
|
310 |
|
|
Calls C<clGetPlatformInfo> and returns the packed, raw value - for |
311 |
|
|
strings, this will be the string, for other values you probably need to |
312 |
|
|
use the correct C<unpack>. This might get improved in the future. Hopefully. |
313 |
|
|
|
314 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetPlatformInfo.html> |
315 |
|
|
|
316 |
|
|
=item @devices = $platform->devices ($type = OpenCL::DEVICE_TYPE_ALL) |
317 |
|
|
|
318 |
|
|
Returns a list of matching OpenCL::Device objects. |
319 |
|
|
|
320 |
root |
1.10 |
=item $ctx = $platform->context_from_type ($properties, $type = OpenCL::DEVICE_TYPE_DEFAULT, $notify = undef) |
321 |
root |
1.5 |
|
322 |
root |
1.11 |
Tries to create a context. Never worked for me, and you need devices explitly anyway. |
323 |
root |
1.5 |
|
324 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html> |
325 |
|
|
|
326 |
root |
1.11 |
=item $ctx = $device->context ($properties = undef, @$devices, $notify = undef) |
327 |
|
|
|
328 |
|
|
Create a new OpenCL::Context object using the given device object(s)- a |
329 |
|
|
CL_CONTEXT_PLATFORM property is supplied automatically. |
330 |
|
|
|
331 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContext.html> |
332 |
|
|
|
333 |
root |
1.5 |
=back |
334 |
|
|
|
335 |
|
|
=head2 THE OpenCL::Device CLASS |
336 |
|
|
|
337 |
|
|
=over 4 |
338 |
|
|
|
339 |
|
|
=item $packed_value = $device->info ($name) |
340 |
|
|
|
341 |
|
|
See C<< $platform->info >> for details. |
342 |
|
|
|
343 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetDeviceInfo.html> |
344 |
|
|
|
345 |
|
|
=back |
346 |
|
|
|
347 |
|
|
=head2 THE OpenCL::Context CLASS |
348 |
|
|
|
349 |
|
|
=over 4 |
350 |
|
|
|
351 |
|
|
=item $packed_value = $ctx->info ($name) |
352 |
|
|
|
353 |
|
|
See C<< $platform->info >> for details. |
354 |
|
|
|
355 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetContextInfo.html> |
356 |
|
|
|
357 |
root |
1.9 |
=item $queue = $ctx->queue ($device, $properties) |
358 |
root |
1.5 |
|
359 |
root |
1.9 |
Create a new OpenCL::Queue object from the context and the given device. |
360 |
root |
1.5 |
|
361 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateCommandQueue.html> |
362 |
|
|
|
363 |
|
|
=item $ev = $ctx->user_event |
364 |
|
|
|
365 |
|
|
Creates a new OpenCL::UserEvent object. |
366 |
|
|
|
367 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateUserEvent.html> |
368 |
|
|
|
369 |
|
|
=item $buf = $ctx->buffer ($flags, $len) |
370 |
|
|
|
371 |
|
|
Creates a new OpenCL::Buffer object with the given flags and octet-size. |
372 |
|
|
|
373 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateBuffer.html> |
374 |
|
|
|
375 |
|
|
=item $buf = $ctx->buffer_sv ($flags, $data) |
376 |
|
|
|
377 |
|
|
Creates a new OpenCL::Buffer object and initialise it with the given data values. |
378 |
|
|
|
379 |
|
|
=item $img = $ctx->image2d ($flags, $channel_order, $channel_type, $width, $height, $data) |
380 |
|
|
|
381 |
|
|
Creates a new OpenCL::Image2D object and optionally initialises it with the given data values. |
382 |
|
|
|
383 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateImage2D.html> |
384 |
|
|
|
385 |
|
|
=item $img = $ctx->image3d ($flags, $channel_order, $channel_type, $width, $height, $depth, $slice_pitch, $data) |
386 |
|
|
|
387 |
|
|
Creates a new OpenCL::Image3D object and optionally initialises it with the given data values. |
388 |
|
|
|
389 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateImage3D.html> |
390 |
|
|
|
391 |
|
|
=item @formats = $ctx->supported_image_formats ($flags, $image_type) |
392 |
|
|
|
393 |
|
|
Returns a list of matching image formats - each format is an arrayref with |
394 |
|
|
two values, $channel_order and $channel_type, in it. |
395 |
|
|
|
396 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetSupportedImageFormats.html> |
397 |
|
|
|
398 |
|
|
=item $sampler = $ctx->sampler ($normalized_coords, $addressing_mode, $filter_mode) |
399 |
|
|
|
400 |
|
|
Creates a new OpenCL::Sampler object. |
401 |
|
|
|
402 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateSampler.html> |
403 |
|
|
|
404 |
|
|
=item $program = $ctx->program_with_source ($string) |
405 |
|
|
|
406 |
|
|
Creates a new OpenCL::Program object from the given source code. |
407 |
|
|
|
408 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateProgramWithSource.html> |
409 |
|
|
|
410 |
|
|
=back |
411 |
|
|
|
412 |
|
|
=head2 THE OpenCL::Queue CLASS |
413 |
|
|
|
414 |
|
|
An OpenCL::Queue represents an execution queue for OpenCL. You execute |
415 |
|
|
requests by calling their respective C<enqueue_xxx> method and waitinf for |
416 |
|
|
it to complete in some way. |
417 |
|
|
|
418 |
|
|
All the enqueue methods return an event object that can be used to wait |
419 |
|
|
for completion, unless the method is called in void context, in which case |
420 |
|
|
no event object is created. |
421 |
|
|
|
422 |
|
|
They also allow you to specify any number of other event objects that this |
423 |
|
|
request has to wait for before it starts executing, by simply passing the |
424 |
|
|
event objects as extra parameters to the enqueue methods. |
425 |
|
|
|
426 |
|
|
Queues execute in-order by default, without any parallelism, so in most |
427 |
root |
1.6 |
cases (i.e. you use only one queue) it's not necessary to wait for or |
428 |
|
|
create event objects. |
429 |
root |
1.5 |
|
430 |
|
|
=over 4 |
431 |
|
|
|
432 |
|
|
=item $packed_value = $ctx->info ($name) |
433 |
|
|
|
434 |
|
|
See C<< $platform->info >> for details. |
435 |
|
|
|
436 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetCommandQueueInfo.html> |
437 |
|
|
|
438 |
|
|
=item $ev = $queue->enqueue_read_buffer ($buffer, $blocking, $offset, $len, $data, $wait_events...) |
439 |
|
|
|
440 |
|
|
Reads data from buffer into the given string. |
441 |
|
|
|
442 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReadBuffer.html> |
443 |
|
|
|
444 |
|
|
=item $ev = $queue->enqueue_write_buffer ($buffer, $blocking, $offset, $data, $wait_events...) |
445 |
|
|
|
446 |
|
|
Writes data to buffer from the given string. |
447 |
|
|
|
448 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteBuffer.html> |
449 |
|
|
|
450 |
|
|
=item $ev = $queue->enqueue_copy_buffer ($src, $dst, $src_offset, $dst_offset, $len, $wait_events...) |
451 |
|
|
|
452 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBuffer.html> |
453 |
|
|
|
454 |
|
|
=item $ev = $queue->enqueue_read_image ($src, $blocking, $x, $y, $z, $width, $height, $depth, $row_pitch, $slice_pitch, $data, $wait_events...) |
455 |
|
|
|
456 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReadImage.html> |
457 |
|
|
|
458 |
|
|
=item $ev = $queue->enqueue_write_image ($src, $blocking, $x, $y, $z, $width, $height, $depth, $row_pitch, $data, $wait_events...) |
459 |
|
|
|
460 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteImage.html> |
461 |
|
|
|
462 |
|
|
=item $ev = $queue->enqueue_copy_buffer_rect ($src, $dst, $src_x, $src_y, $src_z, $dst_x, $dst_y, $dst_z, $width, $height, $depth, $src_row_pitch, $src_slice_pitch, 4dst_row_pitch, $dst_slice_pitch, $ait_event...) |
463 |
|
|
|
464 |
|
|
Yeah. |
465 |
|
|
|
466 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBufferRect.html> |
467 |
|
|
|
468 |
|
|
=item $ev = $queue->enqueue_copy_buffer_to_image (OpenCL::Buffer src, OpenCL::Image dst, size_t src_offset, size_t dst_x, size_t dst_y, size_t dst_z, size_t width, size_t height, size_t depth, ...) |
469 |
|
|
|
470 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBufferToImage.html>. |
471 |
|
|
|
472 |
|
|
=item $ev = $queue->enqueue_copy_image (OpenCL::Image src, OpenCL::Buffer dst, size_t src_x, size_t src_y, size_t src_z, size_t dst_x, size_t dst_y, size_t dst_z, size_t width, size_t height, size_t depth, ...) |
473 |
|
|
|
474 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyImage.html> |
475 |
|
|
|
476 |
|
|
=item $ev = $queue->enqueue_copy_image_to_buffer (OpenCL::Image src, OpenCL::Buffer dst, size_t src_x, size_t src_y, size_t src_z, size_t width, size_t height, size_t depth, size_t dst_offset, ...) |
477 |
|
|
|
478 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyImageToBuffer.html> |
479 |
|
|
|
480 |
|
|
=item $ev = $queue->enqueue_task ($kernel, $wait_events...) |
481 |
|
|
|
482 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueTask.html> |
483 |
|
|
|
484 |
|
|
=item $ev = $queue->enqueue_nd_range_kernel ($kernel, @$global_work_offset, @$global_work_size, @$local_work_size, $wait_events...) |
485 |
|
|
|
486 |
|
|
Enqueues a kernel execution. |
487 |
|
|
|
488 |
|
|
@$global_work_size must be specified as a reference to an array of |
489 |
|
|
integers specifying the work sizes (element counts). |
490 |
|
|
|
491 |
|
|
@$global_work_offset must be either C<undef> (in which case all offsets |
492 |
|
|
are C<0>), or a reference to an array of work offsets, with the same number |
493 |
|
|
of elements as @$global_work_size. |
494 |
|
|
|
495 |
|
|
@$local_work_size must be either C<undef> (in which case the |
496 |
|
|
implementation is supposed to choose good local work sizes), or a |
497 |
|
|
reference to an array of local work sizes, with the same number of |
498 |
|
|
elements as @$global_work_size. |
499 |
|
|
|
500 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueNDRangeKernel.html> |
501 |
|
|
|
502 |
|
|
=item $ev = $queue->enqueue_marker |
503 |
|
|
|
504 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueMarker.html> |
505 |
|
|
|
506 |
|
|
=item $ev = $queue->enqueue_wait_for_events ($wait_events...) |
507 |
|
|
|
508 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWaitForEvents.html> |
509 |
|
|
|
510 |
|
|
=item $queue->enqueue_barrier |
511 |
|
|
|
512 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueBarrier.html> |
513 |
|
|
|
514 |
|
|
=item $queue->flush |
515 |
|
|
|
516 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clFlush.html> |
517 |
|
|
|
518 |
|
|
=item $queue->finish |
519 |
|
|
|
520 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clFinish.html> |
521 |
|
|
|
522 |
|
|
=back |
523 |
|
|
|
524 |
|
|
=head2 THE OpenCL::Memory CLASS |
525 |
|
|
|
526 |
|
|
This the superclass of all memory objects - OpenCL::Buffer, OpenCL::Image, |
527 |
|
|
OpenCL::Image2D and OpenCL::Image3D. The subclasses of this class |
528 |
|
|
currently only exist to allow type-checking. |
529 |
|
|
|
530 |
|
|
=over 4 |
531 |
|
|
|
532 |
|
|
=item $packed_value = $memory->info ($name) |
533 |
|
|
|
534 |
|
|
See C<< $platform->info >> for details. |
535 |
|
|
|
536 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetMemObjectInfo.html> |
537 |
|
|
|
538 |
|
|
=back |
539 |
|
|
|
540 |
|
|
=head2 THE OpenCL::Sampler CLASS |
541 |
|
|
|
542 |
|
|
=over 4 |
543 |
|
|
|
544 |
|
|
=item $packed_value = $sampler->info ($name) |
545 |
|
|
|
546 |
|
|
See C<< $platform->info >> for details. |
547 |
|
|
|
548 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetSamplerInfo.html> |
549 |
|
|
|
550 |
|
|
=back |
551 |
|
|
|
552 |
|
|
=head2 THE OpenCL::Program CLASS |
553 |
|
|
|
554 |
|
|
=over 4 |
555 |
|
|
|
556 |
|
|
=item $packed_value = $program->info ($name) |
557 |
|
|
|
558 |
|
|
See C<< $platform->info >> for details. |
559 |
|
|
|
560 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetProgramInfo.html> |
561 |
|
|
|
562 |
|
|
=item $program->build ($device, $options = "") |
563 |
|
|
|
564 |
|
|
Tries to build the program with the givne options. |
565 |
|
|
|
566 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clBuildProgram.html> |
567 |
|
|
|
568 |
|
|
=item $packed_value = $program->build_info ($device, $name) |
569 |
|
|
|
570 |
|
|
Similar to C<< $platform->info >>, but returns build info for a previous |
571 |
|
|
build attempt for the given device. |
572 |
|
|
|
573 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetBuildInfo.html> |
574 |
|
|
|
575 |
|
|
=item $kernel = $program->kernel ($function_name) |
576 |
|
|
|
577 |
|
|
Creates an OpenCL::Kernel object out of the named C<__kernel> function in |
578 |
|
|
the program. |
579 |
|
|
|
580 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateKernel.html> |
581 |
|
|
|
582 |
|
|
=back |
583 |
|
|
|
584 |
|
|
=head2 THE OpenCL::Kernel CLASS |
585 |
|
|
|
586 |
|
|
=over 4 |
587 |
|
|
|
588 |
|
|
=item $packed_value = $kernel->info ($name) |
589 |
|
|
|
590 |
|
|
See C<< $platform->info >> for details. |
591 |
|
|
|
592 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetKernelInfo.html> |
593 |
|
|
|
594 |
|
|
=item $kernel->set_TYPE ($index, $value) |
595 |
|
|
|
596 |
|
|
This is a family of methods to set the kernel argument with the number C<$index> to the give C<$value>. |
597 |
|
|
|
598 |
|
|
TYPE is one of C<char>, C<uchar>, C<short>, C<ushort>, C<int>, C<uint>, |
599 |
|
|
C<long>, C<ulong>, C<half>, C<float>, C<double>, C<memory>, C<buffer>, |
600 |
|
|
C<image2d>, C<image3d>, C<sampler> or C<event>. |
601 |
|
|
|
602 |
|
|
Chars and integers (including the half type) are specified as integers, |
603 |
|
|
float and double as floating point values, memory/buffer/image2d/image3d |
604 |
|
|
must be an object of that type or C<undef>, and sampler and event must be |
605 |
|
|
objects of that type. |
606 |
|
|
|
607 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetKernelArg.html> |
608 |
|
|
|
609 |
|
|
=back |
610 |
|
|
|
611 |
|
|
=head2 THE OpenCL::Event CLASS |
612 |
|
|
|
613 |
|
|
This is the superclass for all event objects (including OpenCL::UserEvent |
614 |
|
|
objects). |
615 |
|
|
|
616 |
|
|
=over 4 |
617 |
|
|
|
618 |
|
|
=item $packed_value = $ev->info ($name) |
619 |
|
|
|
620 |
|
|
See C<< $platform->info >> for details. |
621 |
|
|
|
622 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetEventInfo.html> |
623 |
|
|
|
624 |
|
|
=item $ev->wait |
625 |
|
|
|
626 |
|
|
Waits for the event to complete. |
627 |
|
|
|
628 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clWaitForEvents.html> |
629 |
|
|
|
630 |
|
|
=back |
631 |
|
|
|
632 |
|
|
=head2 THE OpenCL::UserEvent CLASS |
633 |
|
|
|
634 |
|
|
This is a subclass of OpenCL::Event. |
635 |
root |
1.4 |
|
636 |
root |
1.1 |
=over 4 |
637 |
|
|
|
638 |
root |
1.5 |
=item $ev->set_status ($execution_status) |
639 |
|
|
|
640 |
|
|
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetUserEventStatus.html> |
641 |
|
|
|
642 |
|
|
=back |
643 |
|
|
|
644 |
root |
1.1 |
=cut |
645 |
|
|
|
646 |
|
|
package OpenCL; |
647 |
|
|
|
648 |
root |
1.2 |
use common::sense; |
649 |
|
|
|
650 |
root |
1.1 |
BEGIN { |
651 |
root |
1.12 |
our $VERSION = '0.14'; |
652 |
root |
1.1 |
|
653 |
|
|
require XSLoader; |
654 |
|
|
XSLoader::load (__PACKAGE__, $VERSION); |
655 |
root |
1.3 |
|
656 |
root |
1.5 |
@OpenCL::Buffer::ISA = |
657 |
|
|
@OpenCL::Image::ISA = OpenCL::Memory::; |
658 |
root |
1.3 |
|
659 |
root |
1.5 |
@OpenCL::Image2D::ISA = |
660 |
|
|
@OpenCL::Image3D::ISA = OpenCL::Image::; |
661 |
|
|
|
662 |
|
|
@OpenCL::UserEvent::ISA = OpenCL::Event::; |
663 |
root |
1.1 |
} |
664 |
|
|
|
665 |
|
|
1; |
666 |
|
|
|
667 |
|
|
=head1 AUTHOR |
668 |
|
|
|
669 |
|
|
Marc Lehmann <schmorp@schmorp.de> |
670 |
|
|
http://home.schmorp.de/ |
671 |
|
|
|
672 |
|
|
=cut |
673 |
|
|
|