1 |
=head1 NAME |
2 |
|
3 |
OpenCL - Open Computing Language Bindings |
4 |
|
5 |
=head1 SYNOPSIS |
6 |
|
7 |
use OpenCL; |
8 |
|
9 |
=head1 DESCRIPTION |
10 |
|
11 |
This is an early release which might be useful, but hasn't seen much testing. |
12 |
|
13 |
=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 |
From a platform and some device(s), you create an OpenCL::Context, which is |
24 |
a very central object in OpenCL: Once you have a context you can create |
25 |
most other objects: |
26 |
|
27 |
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 |
|
32 |
OpenCL::Memory objects of various flavours: OpenCL::Buffer objects (flat |
33 |
memory areas, think arrays or structs) and OpenCL::Image objects (think 2d |
34 |
or 3d 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 |
|
48 |
The OpenCL spec used to develop this module (1.2 spec was available, but |
49 |
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 |
If you are into UML class diagrams, the following diagram might help - if |
58 |
not, it will be mildly cobfusing: |
59 |
|
60 |
http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/classDiagram.html |
61 |
|
62 |
Here's a tutorial from AMD (very AMD-centric, too), not sure how useful it |
63 |
is, but at least it's free of charge: |
64 |
|
65 |
http://developer.amd.com/zones/OpenCLZone/courses/Documents/Introduction_to_OpenCL_Programming%20Training_Guide%20%28201005%29.pdf |
66 |
|
67 |
And here's NVIDIA's OpenCL Best Practises Guide: |
68 |
|
69 |
http://developer.download.nvidia.com/compute/cuda/3_2/toolkit/docs/OpenCL_Best_Practices_Guide.pdf |
70 |
|
71 |
=head1 BASIC WORKFLOW |
72 |
|
73 |
To get something done, you basically have to do this once (refer to the |
74 |
examples below for actual code, this is just a high-level description): |
75 |
|
76 |
Find some platform (e.g. the first one) and some device(s) (e.g. the first |
77 |
device of the platform), and create a context from those. |
78 |
|
79 |
Create program objects from your OpenCL source code, then build (compile) |
80 |
the programs for each device you want to run them on. |
81 |
|
82 |
Create kernel objects for all kernels you want to use (surprisingly, these |
83 |
are not device-specific). |
84 |
|
85 |
Then, to execute stuff, you repeat these steps, possibly resuing or |
86 |
sharing some buffers: |
87 |
|
88 |
Create some input and output buffers from your context. Set these as |
89 |
arguments to your kernel. |
90 |
|
91 |
Enqueue buffer writes to initialise your input buffers (when not |
92 |
initialised at creation time). |
93 |
|
94 |
Enqueue the kernel execution. |
95 |
|
96 |
Enqueue buffer reads for your output buffer to read results. |
97 |
|
98 |
=head1 EXAMPLES |
99 |
|
100 |
=head2 Enumerate all devices and get contexts for them. |
101 |
|
102 |
Best run this once to get a feel for the platforms and devices in your |
103 |
system. |
104 |
|
105 |
for my $platform (OpenCL::platforms) { |
106 |
printf "platform: %s\n", $platform->name; |
107 |
printf "extensions: %s\n", $platform->extensions; |
108 |
for my $device ($platform->devices) { |
109 |
printf "+ device: %s\n", $device->name; |
110 |
my $ctx = $platform->context (undef, [$device]); |
111 |
# do stuff |
112 |
} |
113 |
} |
114 |
|
115 |
=head2 Get a useful context and a command queue. |
116 |
|
117 |
This is a useful boilerplate for any OpenCL program that only wants to use |
118 |
one device, |
119 |
|
120 |
my ($platform) = OpenCL::platforms; # find first platform |
121 |
my ($dev) = $platform->devices; # find first device of platform |
122 |
my $ctx = $platform->context (undef, [$dev]); # create context out of those |
123 |
my $queue = $ctx->queue ($dev); # create a command queue for the device |
124 |
|
125 |
=head2 Print all supported image formats of a context. |
126 |
|
127 |
Best run this once for your context, to see whats available and how to |
128 |
gather information. |
129 |
|
130 |
for my $type (OpenCL::MEM_OBJECT_IMAGE2D, OpenCL::MEM_OBJECT_IMAGE3D) { |
131 |
print "supported image formats for ", OpenCL::enum2str $type, "\n"; |
132 |
|
133 |
for my $f ($ctx->supported_image_formats (0, $type)) { |
134 |
printf " %-10s %-20s\n", OpenCL::enum2str $f->[0], OpenCL::enum2str $f->[1]; |
135 |
} |
136 |
} |
137 |
|
138 |
=head2 Create a buffer with some predefined data, read it back synchronously, |
139 |
then asynchronously. |
140 |
|
141 |
my $buf = $ctx->buffer_sv (OpenCL::MEM_COPY_HOST_PTR, "helmut"); |
142 |
|
143 |
$queue->enqueue_read_buffer ($buf, 1, 1, 3, my $data); |
144 |
print "$data\n"; |
145 |
|
146 |
my $ev = $queue->enqueue_read_buffer ($buf, 0, 1, 3, my $data); |
147 |
$ev->wait; |
148 |
print "$data\n"; # prints "elm" |
149 |
|
150 |
=head2 Create and build a program, then create a kernel out of one of its |
151 |
functions. |
152 |
|
153 |
my $src = ' |
154 |
kernel void |
155 |
squareit (global float *input, global float *output) |
156 |
{ |
157 |
$id = get_global_id (0); |
158 |
output [id] = input [id] * input [id]; |
159 |
} |
160 |
'; |
161 |
|
162 |
my $prog = $ctx->build_program ($src); |
163 |
my $kernel = $prog->kernel ("squareit"); |
164 |
|
165 |
=head2 Create some input and output float buffers, then call the |
166 |
'squareit' kernel on them. |
167 |
|
168 |
my $input = $ctx->buffer_sv (OpenCL::MEM_COPY_HOST_PTR, pack "f*", 1, 2, 3, 4.5); |
169 |
my $output = $ctx->buffer (0, OpenCL::SIZEOF_FLOAT * 5); |
170 |
|
171 |
# set buffer |
172 |
$kernel->set_buffer (0, $input); |
173 |
$kernel->set_buffer (1, $output); |
174 |
|
175 |
# execute it for all 4 numbers |
176 |
$queue->enqueue_nd_range_kernel ($kernel, undef, [4], undef); |
177 |
|
178 |
# enqueue a synchronous read |
179 |
$queue->enqueue_read_buffer ($output, 1, 0, OpenCL::SIZEOF_FLOAT * 4, my $data); |
180 |
|
181 |
# print the results: |
182 |
printf "%s\n", join ", ", unpack "f*", $data; |
183 |
|
184 |
=head2 The same enqueue operations as before, but assuming an out-of-order queue, |
185 |
showing off barriers. |
186 |
|
187 |
# execute it for all 4 numbers |
188 |
$queue->enqueue_nd_range_kernel ($kernel, undef, [4], undef); |
189 |
|
190 |
# enqueue a barrier to ensure in-order execution |
191 |
$queue->enqueue_barrier; |
192 |
|
193 |
# enqueue an async read |
194 |
$queue->enqueue_read_buffer ($output, 0, 0, OpenCL::SIZEOF_FLOAT * 4, my $data); |
195 |
|
196 |
# wait for all requests to finish |
197 |
$queue->finish; |
198 |
|
199 |
=head2 The same enqueue operations as before, but assuming an out-of-order queue, |
200 |
showing off event objects and wait lists. |
201 |
|
202 |
# execute it for all 4 numbers |
203 |
my $ev = $queue->enqueue_nd_range_kernel ($kernel, undef, [4], undef); |
204 |
|
205 |
# enqueue an async read |
206 |
$ev = $queue->enqueue_read_buffer ($output, 0, 0, OpenCL::SIZEOF_FLOAT * 4, my $data, $ev); |
207 |
|
208 |
# wait for the last event to complete |
209 |
$ev->wait; |
210 |
|
211 |
=head2 Use the OpenGL module to share a texture between OpenCL and OpenGL and draw some julia |
212 |
set tunnel effect. |
213 |
|
214 |
This is quite a long example to get you going. |
215 |
|
216 |
use OpenGL ":all"; |
217 |
use OpenCL; |
218 |
|
219 |
# open a window and create a gl texture |
220 |
OpenGL::glpOpenWindow width => 256, height => 256; |
221 |
my $texid = glGenTextures_p 1; |
222 |
glBindTexture GL_TEXTURE_2D, $texid; |
223 |
glTexImage2D_c GL_TEXTURE_2D, 0, GL_RGBA8, 256, 256, 0, GL_RGBA, GL_UNSIGNED_BYTE, 0; |
224 |
|
225 |
# find and use the first opencl device that let's us get a shared opengl context |
226 |
my $platform; |
227 |
my $dev; |
228 |
my $ctx; |
229 |
|
230 |
for (OpenCL::platforms) { |
231 |
$platform = $_; |
232 |
for ($platform->devices) { |
233 |
$dev = $_; |
234 |
$ctx = $platform->context ([OpenCL::GLX_DISPLAY_KHR, undef, OpenCL::GL_CONTEXT_KHR, undef], [$dev]) |
235 |
and last; |
236 |
} |
237 |
} |
238 |
|
239 |
$ctx |
240 |
or die "cannot find suitable OpenCL device\n"; |
241 |
|
242 |
my $queue = $ctx->queue ($dev); |
243 |
|
244 |
# now attach an opencl image2d object to the opengl texture |
245 |
my $tex = $ctx->gl_texture2d (OpenCL::MEM_WRITE_ONLY, GL_TEXTURE_2D, 0, $texid); |
246 |
|
247 |
# now the boring opencl code |
248 |
my $src = <<EOF; |
249 |
kernel void |
250 |
juliatunnel (write_only image2d_t img, float time) |
251 |
{ |
252 |
float2 p = (float2)(get_global_id (0), get_global_id (1)) / 256.f * 2.f - 1.f; |
253 |
|
254 |
float2 m = (float2)(1.f, p.y) / fabs (p.x); |
255 |
m.x = fabs (fmod (m.x + time * 0.05f, 4.f)) - 2.f; |
256 |
|
257 |
float2 z = m; |
258 |
float2 c = (float2)(sin (time * 0.05005), cos (time * 0.06001)); |
259 |
|
260 |
for (int i = 0; i < 25 && dot (z, z) < 4.f; ++i) |
261 |
z = (float2)(z.x * z.x - z.y * z.y, 2.f * z.x * z.y) + c; |
262 |
|
263 |
float3 colour = (float3)(z.x, z.y, z.x * z.y); |
264 |
write_imagef (img, (int2)(get_global_id (0), get_global_id (1)), (float4)(colour * p.x * p.x, 1.)); |
265 |
} |
266 |
EOF |
267 |
|
268 |
my $prog = $ctx->build_program ($src); |
269 |
my $kernel = $prog->kernel ("juliatunnel"); |
270 |
|
271 |
# program compiled, kernel ready, now draw and loop |
272 |
|
273 |
for (my $time; ; ++$time) { |
274 |
# acquire objects from opengl |
275 |
$queue->enqueue_acquire_gl_objects ([$tex]); |
276 |
|
277 |
# configure and run our kernel |
278 |
$kernel->set_image2d (0, $tex); |
279 |
$kernel->set_float (1, $time); |
280 |
$queue->enqueue_nd_range_kernel ($kernel, undef, [256, 256], undef); |
281 |
|
282 |
# release objects to opengl again |
283 |
$queue->enqueue_release_gl_objects ([$tex]); |
284 |
|
285 |
# wait |
286 |
$queue->finish; |
287 |
|
288 |
# now draw the texture, the defaults should be all right |
289 |
glTexParameterf GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST; |
290 |
|
291 |
glEnable GL_TEXTURE_2D; |
292 |
glBegin GL_QUADS; |
293 |
glTexCoord2f 0, 1; glVertex3i -1, -1, -1; |
294 |
glTexCoord2f 0, 0; glVertex3i 1, -1, -1; |
295 |
glTexCoord2f 1, 0; glVertex3i 1, 1, -1; |
296 |
glTexCoord2f 1, 1; glVertex3i -1, 1, -1; |
297 |
glEnd; |
298 |
|
299 |
glXSwapBuffers; |
300 |
|
301 |
select undef, undef, undef, 1/60; |
302 |
} |
303 |
|
304 |
=head1 DOCUMENTATION |
305 |
|
306 |
=head2 BASIC CONVENTIONS |
307 |
|
308 |
This is not a one-to-one C-style translation of OpenCL to Perl - instead |
309 |
I attempted to make the interface as type-safe as possible by introducing |
310 |
object syntax where it makes sense. There are a number of important |
311 |
differences between the OpenCL C API and this module: |
312 |
|
313 |
=over 4 |
314 |
|
315 |
=item * Object lifetime managament is automatic - there is no need |
316 |
to free objects explicitly (C<clReleaseXXX>), the release function |
317 |
is called automatically once all Perl references to it go away. |
318 |
|
319 |
=item * OpenCL uses CamelCase for function names |
320 |
(e.g. C<clGetPlatformIDs>, C<clGetPlatformInfo>), while this module |
321 |
uses underscores as word separator and often leaves out prefixes |
322 |
(C<OpenCL::platforms>, C<< $platform->info >>). |
323 |
|
324 |
=item * OpenCL often specifies fixed vector function arguments as short |
325 |
arrays (C<size_t origin[3]>), while this module explicitly expects the |
326 |
components as separate arguments (C<$orig_x, $orig_y, $orig_z>) in |
327 |
function calls. |
328 |
|
329 |
=item * Structures are often specified by flattening out their components |
330 |
as with short vectors, and returned as arrayrefs. |
331 |
|
332 |
=item * When enqueuing commands, the wait list is specified by adding |
333 |
extra arguments to the function - anywhere a C<$wait_events...> argument |
334 |
is documented this can be any number of event objects. As an extsnion |
335 |
implemented by this module, C<undef> values will be ignored in the event |
336 |
list. |
337 |
|
338 |
=item * When enqueuing commands, if the enqueue method is called in void |
339 |
context, no event is created. In all other contexts an event is returned |
340 |
by the method. |
341 |
|
342 |
=item * This module expects all functions to return C<CL_SUCCESS>. If any |
343 |
other status is returned the function will throw an exception, so you |
344 |
don't normally have to to any error checking. |
345 |
|
346 |
=back |
347 |
|
348 |
=head2 PERL AND OPENCL TYPES |
349 |
|
350 |
This handy(?) table lists OpenCL types and their perl, PDL and pack/unpack |
351 |
format equivalents: |
352 |
|
353 |
OpenCL perl PDL pack/unpack |
354 |
char IV - c |
355 |
uchar IV byte C |
356 |
short IV short s |
357 |
ushort IV ushort S |
358 |
int IV long? l |
359 |
uint IV - L |
360 |
long IV longlong q |
361 |
ulong IV - Q |
362 |
float NV float f |
363 |
half IV ushort S |
364 |
double NV double d |
365 |
|
366 |
=head2 GLX SUPPORT |
367 |
|
368 |
Due to the sad state that OpenGL support is in in Perl (mostly the OpenGL |
369 |
module, which has little to no documentation and has little to no support |
370 |
for glX), this module, as a special extension, treats context creation |
371 |
properties C<OpenCL::GLX_DISPLAY_KHR> and C<OpenCL::GL_CONTEXT_KHR> |
372 |
specially: If either or both of these are C<undef>, then the OpenCL |
373 |
module tries to dynamically resolve C<glXGetCurrentDisplay> and |
374 |
C<glXGetCurrentContext>, call these functions and use their return values |
375 |
instead. |
376 |
|
377 |
For this to work, the OpenGL library must be loaded, a GLX context must |
378 |
have been created and be made current, and C<dlsym> must be available and |
379 |
capable of finding the function via C<RTLD_DEFAULT>. |
380 |
|
381 |
=head2 EVENT SYSTEM |
382 |
|
383 |
OpenCL can generate a number of (potentially) asynchronous events, for |
384 |
example, after compiling a program, to signal a context-related error or, |
385 |
perhaps most important, to signal completion of queued jobs (by setting |
386 |
callbacks on OpenCL::Event objects). |
387 |
|
388 |
To facilitate this, this module maintains an event queue - each |
389 |
time an asynchronous event happens, it is queued, and perl will be |
390 |
interrupted. This is implemented via the L<Async::Interrupt> module. In |
391 |
addition, this module has L<AnyEvent> support, so it can seamlessly |
392 |
integrate itself into many event loops. |
393 |
|
394 |
Since this module is a bit hard to understand, here are some case examples: |
395 |
|
396 |
=head3 Don't use callbacks. |
397 |
|
398 |
When your program never uses any callbacks, then there will never be any |
399 |
notifications you need to take care of, and therefore no need to worry |
400 |
about all this. |
401 |
|
402 |
You can achieve a great deal by explicitly waiting for events, or using |
403 |
barriers and flush calls. In many programs, there is no need at all to |
404 |
tinker with asynchronous events. |
405 |
|
406 |
=head3 Use AnyEvent |
407 |
|
408 |
This module automatically registers a watcher that invokes all outstanding |
409 |
event callbacks when AnyEvent is initialised (and block asynchronous |
410 |
interruptions). Using this mode of operations is the safest and most |
411 |
recommended one. |
412 |
|
413 |
To use this, simply use AnyEvent and this module normally, make sure you |
414 |
have an event loop running: |
415 |
|
416 |
use Gtk2 -init; |
417 |
use AnyEvent; |
418 |
|
419 |
# initialise AnyEvent, by creating a watcher, or: |
420 |
AnyEvent::detect; |
421 |
|
422 |
my $e = $queue->enqueue_marker; |
423 |
$e->cb (sub { |
424 |
warn "opencl is finished\n"; |
425 |
}) |
426 |
|
427 |
main Gtk2; |
428 |
|
429 |
Note that this module will not initialise AnyEvent for you. Before |
430 |
AnyEvent is initialised, the module will asynchronously interrupt perl |
431 |
instead. To avoid any surprises, it's best to explicitly initialise |
432 |
AnyEvent. |
433 |
|
434 |
You can temporarily enable asynchronous interruptions (see next paragraph) |
435 |
by calling C<$OpenCL::INTERRUPT->unblock> and disable them again by |
436 |
calling C<$OpenCL::INTERRUPT->block>. |
437 |
|
438 |
=head3 Let yourself be interrupted at any time |
439 |
|
440 |
This mode is the default unless AnyEvent is loaded and initialised. In |
441 |
this mode, OpenCL asynchronously interrupts a running perl program. The |
442 |
emphasis is on both I<asynchronously> and I<running> here. |
443 |
|
444 |
Asynchronously means that perl might execute your callbacks at any |
445 |
time. For example, in the following code (I<THAT YOU SHOULD NOT COPY>), |
446 |
the C<until> loop following the marker call will be interrupted by the |
447 |
callback: |
448 |
|
449 |
my $e = $queue->enqueue_marker; |
450 |
my $flag; |
451 |
$e->cb (sub { $flag = 1 }); |
452 |
1 until $flag; |
453 |
# $flag is now 1 |
454 |
|
455 |
The reason why you shouldn't blindly copy the above code is that |
456 |
busy waiting is a really really bad thing, and really really bad for |
457 |
performance. |
458 |
|
459 |
While at first this asynchronous business might look exciting, it can be |
460 |
really hard, because you need to be prepared for the callback code to be |
461 |
executed at any time, which limits the amount of things the callback code |
462 |
can do safely. |
463 |
|
464 |
This can be mitigated somewhat by using C<< |
465 |
$OpenCL::INTERRUPT->scope_block >> (see the L<Async::Interrupt> |
466 |
documentation for details). |
467 |
|
468 |
The other problem is that your program must be actively I<running> to be |
469 |
interrupted. When you calculate stuff, your program is running. When you |
470 |
hang in some C functions or other block execution (by calling C<sleep>, |
471 |
C<select>, running an event loop and so on), your program is waiting, not |
472 |
running. |
473 |
|
474 |
One way around that would be to attach a read watcher to your event loop, |
475 |
listening for events on C<< $OpenCL::INTERRUPT->pipe_fileno >>, using a |
476 |
dummy callback (C<sub { }>) to temporarily execute some perl code. |
477 |
|
478 |
That is then awfully close to using the built-in AnyEvent support above, |
479 |
though, so consider that one instead. |
480 |
|
481 |
=head3 Be creative |
482 |
|
483 |
OpenCL exports the L<Async::Interrupt> object it uses in the global |
484 |
variable C<$OpenCL::INTERRUPT>. You can configure it in any way you like. |
485 |
|
486 |
So if you want to feel like a real pro, err, wait, if you feel no risk |
487 |
menas no fun, you can experiment by implementing your own mode of |
488 |
operations. |
489 |
|
490 |
=cut |
491 |
|
492 |
package OpenCL; |
493 |
|
494 |
use common::sense; |
495 |
use Async::Interrupt (); |
496 |
|
497 |
our $POLL_FUNC; # set by XS |
498 |
|
499 |
BEGIN { |
500 |
our $VERSION = '0.97'; |
501 |
|
502 |
require XSLoader; |
503 |
XSLoader::load (__PACKAGE__, $VERSION); |
504 |
|
505 |
@OpenCL::Platform::ISA = |
506 |
@OpenCL::Device::ISA = |
507 |
@OpenCL::Context::ISA = |
508 |
@OpenCL::Queue::ISA = |
509 |
@OpenCL::Memory::ISA = |
510 |
@OpenCL::Sampler::ISA = |
511 |
@OpenCL::Program::ISA = |
512 |
@OpenCL::Kernel::ISA = |
513 |
@OpenCL::Event::ISA = OpenCL::Object::; |
514 |
|
515 |
@OpenCL::Buffer::ISA = |
516 |
@OpenCL::Image::ISA = OpenCL::Memory::; |
517 |
|
518 |
@OpenCL::BufferObj::ISA = OpenCL::Buffer::; |
519 |
|
520 |
@OpenCL::Image2D::ISA = |
521 |
@OpenCL::Image3D::ISA = |
522 |
@OpenCL::Image2DArray::ISA = |
523 |
@OpenCL::Image1D::ISA = |
524 |
@OpenCL::Image1DArray::ISA = |
525 |
@OpenCL::Image1DBuffer::ISA = OpenCL::Image::; |
526 |
|
527 |
@OpenCL::UserEvent::ISA = OpenCL::Event::; |
528 |
} |
529 |
|
530 |
=head2 THE OpenCL PACKAGE |
531 |
|
532 |
=over 4 |
533 |
|
534 |
=item $int = OpenCL::errno |
535 |
|
536 |
The last error returned by a function - it's only valid after an error occured |
537 |
and before calling another OpenCL function. |
538 |
|
539 |
=item $str = OpenCL::err2str $errval |
540 |
|
541 |
Comverts an error value into a human readable string. |
542 |
|
543 |
=item $str = OpenCL::enum2str $enum |
544 |
|
545 |
Converts most enum values (of parameter names, image format constants, |
546 |
object types, addressing and filter modes, command types etc.) into a |
547 |
human readable string. When confronted with some random integer it can be |
548 |
very helpful to pass it through this function to maybe get some readable |
549 |
string out of it. |
550 |
|
551 |
=item @platforms = OpenCL::platforms |
552 |
|
553 |
Returns all available OpenCL::Platform objects. |
554 |
|
555 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetPlatformIDs.html> |
556 |
|
557 |
=item $ctx = OpenCL::context_from_type $properties, $type = OpenCL::DEVICE_TYPE_DEFAULT, $notify = $print_stderr |
558 |
|
559 |
Tries to create a context from a default device and platform - never worked for me. |
560 |
|
561 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html> |
562 |
|
563 |
=item OpenCL::wait_for_events $wait_events... |
564 |
|
565 |
Waits for all events to complete. |
566 |
|
567 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clWaitForEvents.html> |
568 |
|
569 |
=item OpenCL::poll |
570 |
|
571 |
Checks if there are any outstanding events (see L<EVENT SYSTEM>) and |
572 |
invokes their callbacks. |
573 |
|
574 |
=item $OpenCL::INTERRUPT |
575 |
|
576 |
The L<Async::Interrupt> object used to signal asynchronous events (see |
577 |
L<EVENT SYSTEM>). |
578 |
|
579 |
=cut |
580 |
|
581 |
our $INTERRUPT = new Async::Interrupt c_cb => [$POLL_FUNC, 0]; |
582 |
|
583 |
&_eq_initialise ($INTERRUPT->signal_func); |
584 |
|
585 |
=item $OpenCL::WATCHER |
586 |
|
587 |
The L<AnyEvent> watcher object used to watch for asynchronous events (see |
588 |
L<EVENT SYSTEM>). This variable is C<undef> until L<AnyEvent> has been |
589 |
loaded I<and> initialised (e.g. by calling C<AnyEvent::detect>). |
590 |
|
591 |
=cut |
592 |
|
593 |
our $WATCHER; |
594 |
|
595 |
sub _init_anyevent { |
596 |
$INTERRUPT->block; |
597 |
$WATCHER = AE::io ($INTERRUPT->pipe_fileno, 0, sub { $INTERRUPT->handle }); |
598 |
} |
599 |
|
600 |
if (defined $AnyEvent::MODEL) { |
601 |
_init_anyevent; |
602 |
} else { |
603 |
push @AnyEvent::post_detect, \&_init_anyevent; |
604 |
} |
605 |
|
606 |
=back |
607 |
|
608 |
=head2 THE OpenCL::Object CLASS |
609 |
|
610 |
This is the base class for all objects in the OpenCL module. The only |
611 |
method it implements is the C<id> method, which is only useful if you want |
612 |
to interface to OpenCL on the C level. |
613 |
|
614 |
=over 4 |
615 |
|
616 |
=item $iv = $obj->id |
617 |
|
618 |
OpenCL objects are represented by pointers or integers on the C level. If |
619 |
you want to interface to an OpenCL object directly on the C level, then |
620 |
you need this value, which is returned by this method. You should use an |
621 |
C<IV> type in your code and cast that to the correct type. |
622 |
|
623 |
=cut |
624 |
|
625 |
sub OpenCL::Object::id { |
626 |
ref $_[0] eq "SCALAR" |
627 |
? ${ $_[0] } |
628 |
: $_[0][0] |
629 |
} |
630 |
|
631 |
=back |
632 |
|
633 |
=head2 THE OpenCL::Platform CLASS |
634 |
|
635 |
=over 4 |
636 |
|
637 |
=item @devices = $platform->devices ($type = OpenCL::DEVICE_TYPE_ALL) |
638 |
|
639 |
Returns a list of matching OpenCL::Device objects. |
640 |
|
641 |
=item $ctx = $platform->context_from_type ($properties, $type = OpenCL::DEVICE_TYPE_DEFAULT, $notify = $print_stderr) |
642 |
|
643 |
Tries to create a context. Never worked for me, and you need devices explicitly anyway. |
644 |
|
645 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html> |
646 |
|
647 |
=item $ctx = $platform->context ($properties, @$devices, $notify = $print_stderr) |
648 |
|
649 |
Create a new OpenCL::Context object using the given device object(s)- a |
650 |
CL_CONTEXT_PLATFORM property is supplied automatically. |
651 |
|
652 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContext.html> |
653 |
|
654 |
=item $packed_value = $platform->info ($name) |
655 |
|
656 |
Calls C<clGetPlatformInfo> and returns the packed, raw value - for |
657 |
strings, this will be the string (possibly including terminating \0), for |
658 |
other values you probably need to use the correct C<unpack>. |
659 |
|
660 |
It's best to avoid this method and use one of the following convenience |
661 |
wrappers. |
662 |
|
663 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetPlatformInfo.html> |
664 |
|
665 |
=item $platform->unload_compiler |
666 |
|
667 |
Attempts to unload the compiler for this platform, for endless |
668 |
profit. Does nothing on OpenCL 1.1. |
669 |
|
670 |
L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clUnloadPlatformCompiler.html> |
671 |
|
672 |
=for gengetinfo begin platform |
673 |
|
674 |
=item $string = $platform->profile |
675 |
|
676 |
Calls C<clGetPlatformInfo> with C<CL_PLATFORM_PROFILE> and returns the result. |
677 |
|
678 |
=item $string = $platform->version |
679 |
|
680 |
Calls C<clGetPlatformInfo> with C<CL_PLATFORM_VERSION> and returns the result. |
681 |
|
682 |
=item $string = $platform->name |
683 |
|
684 |
Calls C<clGetPlatformInfo> with C<CL_PLATFORM_NAME> and returns the result. |
685 |
|
686 |
=item $string = $platform->vendor |
687 |
|
688 |
Calls C<clGetPlatformInfo> with C<CL_PLATFORM_VENDOR> and returns the result. |
689 |
|
690 |
=item $string = $platform->extensions |
691 |
|
692 |
Calls C<clGetPlatformInfo> with C<CL_PLATFORM_EXTENSIONS> and returns the result. |
693 |
|
694 |
=for gengetinfo end platform |
695 |
|
696 |
=back |
697 |
|
698 |
=head2 THE OpenCL::Device CLASS |
699 |
|
700 |
=over 4 |
701 |
|
702 |
=item $packed_value = $device->info ($name) |
703 |
|
704 |
See C<< $platform->info >> for details. |
705 |
|
706 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetDeviceInfo.html> |
707 |
|
708 |
=for gengetinfo begin device |
709 |
|
710 |
=item $device_type = $device->type |
711 |
|
712 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_TYPE> and returns the result. |
713 |
|
714 |
=item $uint = $device->vendor_id |
715 |
|
716 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_VENDOR_ID> and returns the result. |
717 |
|
718 |
=item $uint = $device->max_compute_units |
719 |
|
720 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_COMPUTE_UNITS> and returns the result. |
721 |
|
722 |
=item $uint = $device->max_work_item_dimensions |
723 |
|
724 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_WORK_ITEM_DIMENSIONS> and returns the result. |
725 |
|
726 |
=item $int = $device->max_work_group_size |
727 |
|
728 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_WORK_GROUP_SIZE> and returns the result. |
729 |
|
730 |
=item @ints = $device->max_work_item_sizes |
731 |
|
732 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_WORK_ITEM_SIZES> and returns the result. |
733 |
|
734 |
=item $uint = $device->preferred_vector_width_char |
735 |
|
736 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_PREFERRED_VECTOR_WIDTH_CHAR> and returns the result. |
737 |
|
738 |
=item $uint = $device->preferred_vector_width_short |
739 |
|
740 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_PREFERRED_VECTOR_WIDTH_SHORT> and returns the result. |
741 |
|
742 |
=item $uint = $device->preferred_vector_width_int |
743 |
|
744 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_PREFERRED_VECTOR_WIDTH_INT> and returns the result. |
745 |
|
746 |
=item $uint = $device->preferred_vector_width_long |
747 |
|
748 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_PREFERRED_VECTOR_WIDTH_LONG> and returns the result. |
749 |
|
750 |
=item $uint = $device->preferred_vector_width_float |
751 |
|
752 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_PREFERRED_VECTOR_WIDTH_FLOAT> and returns the result. |
753 |
|
754 |
=item $uint = $device->preferred_vector_width_double |
755 |
|
756 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_PREFERRED_VECTOR_WIDTH_DOUBLE> and returns the result. |
757 |
|
758 |
=item $uint = $device->max_clock_frequency |
759 |
|
760 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_CLOCK_FREQUENCY> and returns the result. |
761 |
|
762 |
=item $bitfield = $device->address_bits |
763 |
|
764 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_ADDRESS_BITS> and returns the result. |
765 |
|
766 |
=item $uint = $device->max_read_image_args |
767 |
|
768 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_READ_IMAGE_ARGS> and returns the result. |
769 |
|
770 |
=item $uint = $device->max_write_image_args |
771 |
|
772 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_WRITE_IMAGE_ARGS> and returns the result. |
773 |
|
774 |
=item $ulong = $device->max_mem_alloc_size |
775 |
|
776 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_MEM_ALLOC_SIZE> and returns the result. |
777 |
|
778 |
=item $int = $device->image2d_max_width |
779 |
|
780 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_IMAGE2D_MAX_WIDTH> and returns the result. |
781 |
|
782 |
=item $int = $device->image2d_max_height |
783 |
|
784 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_IMAGE2D_MAX_HEIGHT> and returns the result. |
785 |
|
786 |
=item $int = $device->image3d_max_width |
787 |
|
788 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_IMAGE3D_MAX_WIDTH> and returns the result. |
789 |
|
790 |
=item $int = $device->image3d_max_height |
791 |
|
792 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_IMAGE3D_MAX_HEIGHT> and returns the result. |
793 |
|
794 |
=item $int = $device->image3d_max_depth |
795 |
|
796 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_IMAGE3D_MAX_DEPTH> and returns the result. |
797 |
|
798 |
=item $uint = $device->image_support |
799 |
|
800 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_IMAGE_SUPPORT> and returns the result. |
801 |
|
802 |
=item $int = $device->max_parameter_size |
803 |
|
804 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_PARAMETER_SIZE> and returns the result. |
805 |
|
806 |
=item $uint = $device->max_samplers |
807 |
|
808 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_SAMPLERS> and returns the result. |
809 |
|
810 |
=item $uint = $device->mem_base_addr_align |
811 |
|
812 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_MEM_BASE_ADDR_ALIGN> and returns the result. |
813 |
|
814 |
=item $uint = $device->min_data_type_align_size |
815 |
|
816 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_MIN_DATA_TYPE_ALIGN_SIZE> and returns the result. |
817 |
|
818 |
=item $device_fp_config = $device->single_fp_config |
819 |
|
820 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_SINGLE_FP_CONFIG> and returns the result. |
821 |
|
822 |
=item $device_mem_cache_type = $device->global_mem_cache_type |
823 |
|
824 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_GLOBAL_MEM_CACHE_TYPE> and returns the result. |
825 |
|
826 |
=item $uint = $device->global_mem_cacheline_size |
827 |
|
828 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_GLOBAL_MEM_CACHELINE_SIZE> and returns the result. |
829 |
|
830 |
=item $ulong = $device->global_mem_cache_size |
831 |
|
832 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_GLOBAL_MEM_CACHE_SIZE> and returns the result. |
833 |
|
834 |
=item $ulong = $device->global_mem_size |
835 |
|
836 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_GLOBAL_MEM_SIZE> and returns the result. |
837 |
|
838 |
=item $ulong = $device->max_constant_buffer_size |
839 |
|
840 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_CONSTANT_BUFFER_SIZE> and returns the result. |
841 |
|
842 |
=item $uint = $device->max_constant_args |
843 |
|
844 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_CONSTANT_ARGS> and returns the result. |
845 |
|
846 |
=item $device_local_mem_type = $device->local_mem_type |
847 |
|
848 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_LOCAL_MEM_TYPE> and returns the result. |
849 |
|
850 |
=item $ulong = $device->local_mem_size |
851 |
|
852 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_LOCAL_MEM_SIZE> and returns the result. |
853 |
|
854 |
=item $boolean = $device->error_correction_support |
855 |
|
856 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_ERROR_CORRECTION_SUPPORT> and returns the result. |
857 |
|
858 |
=item $int = $device->profiling_timer_resolution |
859 |
|
860 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_PROFILING_TIMER_RESOLUTION> and returns the result. |
861 |
|
862 |
=item $boolean = $device->endian_little |
863 |
|
864 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_ENDIAN_LITTLE> and returns the result. |
865 |
|
866 |
=item $boolean = $device->available |
867 |
|
868 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_AVAILABLE> and returns the result. |
869 |
|
870 |
=item $boolean = $device->compiler_available |
871 |
|
872 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_COMPILER_AVAILABLE> and returns the result. |
873 |
|
874 |
=item $device_exec_capabilities = $device->execution_capabilities |
875 |
|
876 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_EXECUTION_CAPABILITIES> and returns the result. |
877 |
|
878 |
=item $command_queue_properties = $device->properties |
879 |
|
880 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_QUEUE_PROPERTIES> and returns the result. |
881 |
|
882 |
=item $ = $device->platform |
883 |
|
884 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_PLATFORM> and returns the result. |
885 |
|
886 |
=item $string = $device->name |
887 |
|
888 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_NAME> and returns the result. |
889 |
|
890 |
=item $string = $device->vendor |
891 |
|
892 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_VENDOR> and returns the result. |
893 |
|
894 |
=item $string = $device->driver_version |
895 |
|
896 |
Calls C<clGetDeviceInfo> with C<CL_DRIVER_VERSION> and returns the result. |
897 |
|
898 |
=item $string = $device->profile |
899 |
|
900 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_PROFILE> and returns the result. |
901 |
|
902 |
=item $string = $device->version |
903 |
|
904 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_VERSION> and returns the result. |
905 |
|
906 |
=item $string = $device->extensions |
907 |
|
908 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_EXTENSIONS> and returns the result. |
909 |
|
910 |
=item $uint = $device->preferred_vector_width_half |
911 |
|
912 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_PREFERRED_VECTOR_WIDTH_HALF> and returns the result. |
913 |
|
914 |
=item $uint = $device->native_vector_width_char |
915 |
|
916 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_NATIVE_VECTOR_WIDTH_CHAR> and returns the result. |
917 |
|
918 |
=item $uint = $device->native_vector_width_short |
919 |
|
920 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_NATIVE_VECTOR_WIDTH_SHORT> and returns the result. |
921 |
|
922 |
=item $uint = $device->native_vector_width_int |
923 |
|
924 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_NATIVE_VECTOR_WIDTH_INT> and returns the result. |
925 |
|
926 |
=item $uint = $device->native_vector_width_long |
927 |
|
928 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_NATIVE_VECTOR_WIDTH_LONG> and returns the result. |
929 |
|
930 |
=item $uint = $device->native_vector_width_float |
931 |
|
932 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_NATIVE_VECTOR_WIDTH_FLOAT> and returns the result. |
933 |
|
934 |
=item $uint = $device->native_vector_width_double |
935 |
|
936 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_NATIVE_VECTOR_WIDTH_DOUBLE> and returns the result. |
937 |
|
938 |
=item $uint = $device->native_vector_width_half |
939 |
|
940 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_NATIVE_VECTOR_WIDTH_HALF> and returns the result. |
941 |
|
942 |
=item $device_fp_config = $device->double_fp_config |
943 |
|
944 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_DOUBLE_FP_CONFIG> and returns the result. |
945 |
|
946 |
=item $device_fp_config = $device->half_fp_config |
947 |
|
948 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_HALF_FP_CONFIG> and returns the result. |
949 |
|
950 |
=item $boolean = $device->host_unified_memory |
951 |
|
952 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_HOST_UNIFIED_MEMORY> and returns the result. |
953 |
|
954 |
=item $device = $device->parent_device_ext |
955 |
|
956 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_PARENT_DEVICE_EXT> and returns the result. |
957 |
|
958 |
=item @device_partition_property_exts = $device->partition_types_ext |
959 |
|
960 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_PARTITION_TYPES_EXT> and returns the result. |
961 |
|
962 |
=item @device_partition_property_exts = $device->affinity_domains_ext |
963 |
|
964 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_AFFINITY_DOMAINS_EXT> and returns the result. |
965 |
|
966 |
=item $uint = $device->reference_count_ext |
967 |
|
968 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_REFERENCE_COUNT_EXT> and returns the result. |
969 |
|
970 |
=item @device_partition_property_exts = $device->partition_style_ext |
971 |
|
972 |
Calls C<clGetDeviceInfo> with C<CL_DEVICE_PARTITION_STYLE_EXT> and returns the result. |
973 |
|
974 |
=for gengetinfo end device |
975 |
|
976 |
=back |
977 |
|
978 |
=head2 THE OpenCL::Context CLASS |
979 |
|
980 |
=over 4 |
981 |
|
982 |
=item $prog = $ctx->build_program ($program, $options = "") |
983 |
|
984 |
This convenience function tries to build the program on all devices in |
985 |
the context. If the build fails, then the function will C<croak> with the |
986 |
build log. Otherwise ti returns the program object. |
987 |
|
988 |
The C<$program> can either be a C<OpenCL::Program> object or a string |
989 |
containing the program. In the latter case, a program objetc will be |
990 |
created automatically. |
991 |
|
992 |
=cut |
993 |
|
994 |
sub OpenCL::Context::build_program { |
995 |
my ($self, $prog, $options) = @_; |
996 |
|
997 |
require Carp; |
998 |
|
999 |
$prog = $self->program_with_source ($prog) |
1000 |
unless ref $prog; |
1001 |
|
1002 |
# we build separately per device so we instantly know which one failed |
1003 |
for my $dev ($self->devices) { |
1004 |
eval { $prog->build ([$dev], $options); 1 } |
1005 |
or Carp::croak ("Building OpenCL program for device '" . $dev->name . "' failed:\n" |
1006 |
. $prog->build_log ($dev)); |
1007 |
} |
1008 |
|
1009 |
$prog |
1010 |
} |
1011 |
|
1012 |
=item $queue = $ctx->queue ($device, $properties) |
1013 |
|
1014 |
Create a new OpenCL::Queue object from the context and the given device. |
1015 |
|
1016 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateCommandQueue.html> |
1017 |
|
1018 |
Example: create an out-of-order queue. |
1019 |
|
1020 |
$queue = $ctx->queue ($device, OpenCL::QUEUE_OUT_OF_ORDER_EXEC_MODE_ENABLE); |
1021 |
|
1022 |
=item $ev = $ctx->user_event |
1023 |
|
1024 |
Creates a new OpenCL::UserEvent object. |
1025 |
|
1026 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateUserEvent.html> |
1027 |
|
1028 |
=item $buf = $ctx->buffer ($flags, $len) |
1029 |
|
1030 |
Creates a new OpenCL::Buffer (actually OpenCL::BufferObj) object with the |
1031 |
given flags and octet-size. |
1032 |
|
1033 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateBuffer.html> |
1034 |
|
1035 |
=item $buf = $ctx->buffer_sv ($flags, $data) |
1036 |
|
1037 |
Creates a new OpenCL::Buffer (actually OpenCL::BufferObj) object and |
1038 |
initialise it with the given data values. |
1039 |
|
1040 |
=item $img = $ctx->image ($self, $flags, $channel_order, $channel_type, $type, $width, $height, $depth, $array_size = 0, $row_pitch = 0, $slice_pitch = 0, $num_mip_level = 0, $num_samples = 0, $*data = &PL_sv_undef) |
1041 |
|
1042 |
Creates a new OpenCL::Image object and optionally initialises it with |
1043 |
the given data values. |
1044 |
|
1045 |
L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clCreateImage.html> |
1046 |
|
1047 |
=item $img = $ctx->image2d ($flags, $channel_order, $channel_type, $width, $height, $row_pitch = 0, $data = undef) |
1048 |
|
1049 |
Creates a new OpenCL::Image2D object and optionally initialises it with |
1050 |
the given data values. |
1051 |
|
1052 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateImage2D.html> |
1053 |
|
1054 |
=item $img = $ctx->image3d ($flags, $channel_order, $channel_type, $width, $height, $depth, $row_pitch = 0, $slice_pitch = 0, $data = undef) |
1055 |
|
1056 |
Creates a new OpenCL::Image3D object and optionally initialises it with |
1057 |
the given data values. |
1058 |
|
1059 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateImage3D.html> |
1060 |
|
1061 |
=item $buffer = $ctx->gl_buffer ($flags, $bufobj) |
1062 |
|
1063 |
Creates a new OpenCL::Buffer (actually OpenCL::BufferObj) object that refers to the given |
1064 |
OpenGL buffer object. |
1065 |
|
1066 |
http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLBuffer.html |
1067 |
|
1068 |
=item $img = $ctx->gl_texture ($flags, $target, $miplevel, $texture) |
1069 |
|
1070 |
Creates a new OpenCL::Image object that refers to the given OpenGL |
1071 |
texture object or buffer. |
1072 |
|
1073 |
http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clCreateFromGLTexture.html |
1074 |
|
1075 |
=item $img = $ctx->gl_texture2d ($flags, $target, $miplevel, $texture) |
1076 |
|
1077 |
Creates a new OpenCL::Image2D object that refers to the given OpenGL |
1078 |
2D texture object. |
1079 |
|
1080 |
http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLTexture2D.html |
1081 |
|
1082 |
=item $img = $ctx->gl_texture3d ($flags, $target, $miplevel, $texture) |
1083 |
|
1084 |
Creates a new OpenCL::Image3D object that refers to the given OpenGL |
1085 |
3D texture object. |
1086 |
|
1087 |
http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLTexture3D.html |
1088 |
|
1089 |
=item $ctx->gl_renderbuffer ($flags, $renderbuffer) |
1090 |
|
1091 |
Creates a new OpenCL::Image2D object that refers to the given OpenGL |
1092 |
render buffer. |
1093 |
|
1094 |
http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLRenderbuffer.html |
1095 |
|
1096 |
=item @formats = $ctx->supported_image_formats ($flags, $image_type) |
1097 |
|
1098 |
Returns a list of matching image formats - each format is an arrayref with |
1099 |
two values, $channel_order and $channel_type, in it. |
1100 |
|
1101 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetSupportedImageFormats.html> |
1102 |
|
1103 |
=item $sampler = $ctx->sampler ($normalized_coords, $addressing_mode, $filter_mode) |
1104 |
|
1105 |
Creates a new OpenCL::Sampler object. |
1106 |
|
1107 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateSampler.html> |
1108 |
|
1109 |
=item $program = $ctx->program_with_source ($string) |
1110 |
|
1111 |
Creates a new OpenCL::Program object from the given source code. |
1112 |
|
1113 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateProgramWithSource.html> |
1114 |
|
1115 |
=item $packed_value = $ctx->info ($name) |
1116 |
|
1117 |
See C<< $platform->info >> for details. |
1118 |
|
1119 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetContextInfo.html> |
1120 |
|
1121 |
=for gengetinfo begin context |
1122 |
|
1123 |
=item $uint = $context->reference_count |
1124 |
|
1125 |
Calls C<clGetContextInfo> with C<CL_CONTEXT_REFERENCE_COUNT> and returns the result. |
1126 |
|
1127 |
=item @devices = $context->devices |
1128 |
|
1129 |
Calls C<clGetContextInfo> with C<CL_CONTEXT_DEVICES> and returns the result. |
1130 |
|
1131 |
=item @property_ints = $context->properties |
1132 |
|
1133 |
Calls C<clGetContextInfo> with C<CL_CONTEXT_PROPERTIES> and returns the result. |
1134 |
|
1135 |
=item $uint = $context->num_devices |
1136 |
|
1137 |
Calls C<clGetContextInfo> with C<CL_CONTEXT_NUM_DEVICES> and returns the result. |
1138 |
|
1139 |
=for gengetinfo end context |
1140 |
|
1141 |
=back |
1142 |
|
1143 |
=head2 THE OpenCL::Queue CLASS |
1144 |
|
1145 |
An OpenCL::Queue represents an execution queue for OpenCL. You execute |
1146 |
requests by calling their respective C<enqueue_xxx> method and waitinf for |
1147 |
it to complete in some way. |
1148 |
|
1149 |
All the enqueue methods return an event object that can be used to wait |
1150 |
for completion, unless the method is called in void context, in which case |
1151 |
no event object is created. |
1152 |
|
1153 |
They also allow you to specify any number of other event objects that this |
1154 |
request has to wait for before it starts executing, by simply passing the |
1155 |
event objects as extra parameters to the enqueue methods. To simplify |
1156 |
program design, this module ignores any C<undef> values in the list of |
1157 |
events. This makes it possible to code operations such as this, without |
1158 |
having to put a valid event object into C<$event> first: |
1159 |
|
1160 |
$event = $queue->enqueue_xxx (..., $event); |
1161 |
|
1162 |
Queues execute in-order by default, without any parallelism, so in most |
1163 |
cases (i.e. you use only one queue) it's not necessary to wait for or |
1164 |
create event objects, althoguh an our of order queue is often a bit |
1165 |
faster. |
1166 |
|
1167 |
=over 4 |
1168 |
|
1169 |
=item $ev = $queue->enqueue_read_buffer ($buffer, $blocking, $offset, $len, $data, $wait_events...) |
1170 |
|
1171 |
Reads data from buffer into the given string. |
1172 |
|
1173 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReadBuffer.html> |
1174 |
|
1175 |
=item $ev = $queue->enqueue_write_buffer ($buffer, $blocking, $offset, $data, $wait_events...) |
1176 |
|
1177 |
Writes data to buffer from the given string. |
1178 |
|
1179 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteBuffer.html> |
1180 |
|
1181 |
=item $ev = $queue->enqueue_copy_buffer ($src, $dst, $src_offset, $dst_offset, $len, $wait_events...) |
1182 |
|
1183 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBuffer.html> |
1184 |
|
1185 |
=item $ev = $queue->enqueue_read_buffer_rect (OpenCL::Memory buf, cl_bool blocking, $buf_x, $buf_y, $buf_z, $host_x, $host_y, $host_z, $width, $height, $depth, $buf_row_pitch, $buf_slice_pitch, $host_row_pitch, $host_slice_pitch, $data, $wait_events...) |
1186 |
|
1187 |
http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReadBufferRect.html |
1188 |
|
1189 |
=item $ev = $queue->enqueue_write_buffer_rect (OpenCL::Memory buf, cl_bool blocking, $buf_x, $buf_y, $buf_z, $host_x, $host_y, $host_z, $width, $height, $depth, $buf_row_pitch, $buf_slice_pitch, $host_row_pitch, $host_slice_pitch, $data, $wait_events...) |
1190 |
|
1191 |
http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteBufferRect.html |
1192 |
|
1193 |
=item $ev = $queue->enqueue_read_image ($src, $blocking, $x, $y, $z, $width, $height, $depth, $row_pitch, $slice_pitch, $data, $wait_events...) |
1194 |
|
1195 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBufferRect.html> |
1196 |
|
1197 |
=item $ev = $queue->enqueue_copy_buffer_to_image ($src_buffer, $dst_image, $src_offset, $dst_x, $dst_y, $dst_z, $width, $height, $depth, $wait_events...) |
1198 |
|
1199 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReadImage.html> |
1200 |
|
1201 |
=item $ev = $queue->enqueue_write_image ($src, $blocking, $x, $y, $z, $width, $height, $depth, $row_pitch, $slice_pitch, $data, $wait_events...) |
1202 |
|
1203 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteImage.html> |
1204 |
|
1205 |
=item $ev = $queue->enqueue_copy_image ($src_image, $dst_image, $src_x, $src_y, $src_z, $dst_x, $dst_y, $dst_z, $width, $height, $depth, $wait_events...) |
1206 |
|
1207 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyImage.html> |
1208 |
|
1209 |
=item $ev = $queue->enqueue_copy_image_to_buffer ($src_image, $dst_image, $src_x, $src_y, $src_z, $width, $height, $depth, $dst_offset, $wait_events...) |
1210 |
|
1211 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyImageToBuffer.html> |
1212 |
|
1213 |
=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, $dst_row_pitch, $dst_slice_pitch, $wait_event...) |
1214 |
|
1215 |
Yeah. |
1216 |
|
1217 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBufferToImage.html>. |
1218 |
|
1219 |
=item $ev = $queue->enqueue_fill_buffer ($mem, $pattern, $offset, $size, ...) |
1220 |
|
1221 |
Fills the given buffer object with repeated applications of C<$pattern>, |
1222 |
starting at C<$offset> for C<$size> octets. |
1223 |
|
1224 |
L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clEnqueueFillBuffer.html> |
1225 |
|
1226 |
=item $ev = $queue->enqueue_fill_image ($img, $r, $g, $b, $a, $x, $y, $z, $width, $height, $depth, ...) |
1227 |
|
1228 |
Fills the given image area with the given rgba colour components. The |
1229 |
components are normally floating point values between C<0> and C<1>, |
1230 |
except when the image channel data type is a signe dor unsigned |
1231 |
unnormalised format, in which case the range is determined by the format. |
1232 |
|
1233 |
L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clEnqueueFillImage.html> |
1234 |
|
1235 |
=item $ev = $queue->enqueue_task ($kernel, $wait_events...) |
1236 |
|
1237 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueTask.html> |
1238 |
|
1239 |
=item $ev = $queue->enqueue_nd_range_kernel ($kernel, @$global_work_offset, @$global_work_size, @$local_work_size, $wait_events...) |
1240 |
|
1241 |
Enqueues a kernel execution. |
1242 |
|
1243 |
@$global_work_size must be specified as a reference to an array of |
1244 |
integers specifying the work sizes (element counts). |
1245 |
|
1246 |
@$global_work_offset must be either C<undef> (in which case all offsets |
1247 |
are C<0>), or a reference to an array of work offsets, with the same number |
1248 |
of elements as @$global_work_size. |
1249 |
|
1250 |
@$local_work_size must be either C<undef> (in which case the |
1251 |
implementation is supposed to choose good local work sizes), or a |
1252 |
reference to an array of local work sizes, with the same number of |
1253 |
elements as @$global_work_size. |
1254 |
|
1255 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueNDRangeKernel.html> |
1256 |
|
1257 |
=item $ev = $queue->enqueue_acquire_gl_objects ([object, ...], $wait_events...) |
1258 |
|
1259 |
Enqueues a list (an array-ref of OpenCL::Memory objects) to be acquired |
1260 |
for subsequent OpenCL usage. |
1261 |
|
1262 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueAcquireGLObjects.html> |
1263 |
|
1264 |
=item $ev = $queue->enqueue_release_gl_objects ([object, ...], $wait_events...) |
1265 |
|
1266 |
Enqueues a list (an array-ref of OpenCL::Memory objects) to be released |
1267 |
for subsequent OpenGL usage. |
1268 |
|
1269 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReleaseGLObjects.html> |
1270 |
|
1271 |
=item $ev = $queue->enqueue_wait_for_events ($wait_events...) |
1272 |
|
1273 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWaitForEvents.html> |
1274 |
|
1275 |
=item $ev = $queue->enqueue_marker ($wait_events...) |
1276 |
|
1277 |
L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clEnqueueMarkerWithWaitList.html> |
1278 |
|
1279 |
=item $ev = $queue->enqueue_barrier ($wait_events...) |
1280 |
|
1281 |
L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clEnqueueBarrierWithWaitList.html> |
1282 |
|
1283 |
=item $queue->flush |
1284 |
|
1285 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clFlush.html> |
1286 |
|
1287 |
=item $queue->finish |
1288 |
|
1289 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clFinish.html> |
1290 |
|
1291 |
=item $packed_value = $queue->info ($name) |
1292 |
|
1293 |
See C<< $platform->info >> for details. |
1294 |
|
1295 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetCommandQueueInfo.html> |
1296 |
|
1297 |
=for gengetinfo begin command_queue |
1298 |
|
1299 |
=item $ctx = $command_queue->context |
1300 |
|
1301 |
Calls C<clGetCommandQueueInfo> with C<CL_QUEUE_CONTEXT> and returns the result. |
1302 |
|
1303 |
=item $device = $command_queue->device |
1304 |
|
1305 |
Calls C<clGetCommandQueueInfo> with C<CL_QUEUE_DEVICE> and returns the result. |
1306 |
|
1307 |
=item $uint = $command_queue->reference_count |
1308 |
|
1309 |
Calls C<clGetCommandQueueInfo> with C<CL_QUEUE_REFERENCE_COUNT> and returns the result. |
1310 |
|
1311 |
=item $command_queue_properties = $command_queue->properties |
1312 |
|
1313 |
Calls C<clGetCommandQueueInfo> with C<CL_QUEUE_PROPERTIES> and returns the result. |
1314 |
|
1315 |
=for gengetinfo end command_queue |
1316 |
|
1317 |
=back |
1318 |
|
1319 |
=head2 THE OpenCL::Memory CLASS |
1320 |
|
1321 |
This the superclass of all memory objects - OpenCL::Buffer, OpenCL::Image, |
1322 |
OpenCL::Image2D and OpenCL::Image3D. |
1323 |
|
1324 |
=over 4 |
1325 |
|
1326 |
=item $packed_value = $memory->info ($name) |
1327 |
|
1328 |
See C<< $platform->info >> for details. |
1329 |
|
1330 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetMemObjectInfo.html> |
1331 |
|
1332 |
=for gengetinfo begin mem |
1333 |
|
1334 |
=item $mem_object_type = $mem->type |
1335 |
|
1336 |
Calls C<clGetMemObjectInfo> with C<CL_MEM_TYPE> and returns the result. |
1337 |
|
1338 |
=item $mem_flags = $mem->flags |
1339 |
|
1340 |
Calls C<clGetMemObjectInfo> with C<CL_MEM_FLAGS> and returns the result. |
1341 |
|
1342 |
=item $int = $mem->size |
1343 |
|
1344 |
Calls C<clGetMemObjectInfo> with C<CL_MEM_SIZE> and returns the result. |
1345 |
|
1346 |
=item $ptr_value = $mem->host_ptr |
1347 |
|
1348 |
Calls C<clGetMemObjectInfo> with C<CL_MEM_HOST_PTR> and returns the result. |
1349 |
|
1350 |
=item $uint = $mem->map_count |
1351 |
|
1352 |
Calls C<clGetMemObjectInfo> with C<CL_MEM_MAP_COUNT> and returns the result. |
1353 |
|
1354 |
=item $uint = $mem->reference_count |
1355 |
|
1356 |
Calls C<clGetMemObjectInfo> with C<CL_MEM_REFERENCE_COUNT> and returns the result. |
1357 |
|
1358 |
=item $ctx = $mem->context |
1359 |
|
1360 |
Calls C<clGetMemObjectInfo> with C<CL_MEM_CONTEXT> and returns the result. |
1361 |
|
1362 |
=item $mem = $mem->associated_memobject |
1363 |
|
1364 |
Calls C<clGetMemObjectInfo> with C<CL_MEM_ASSOCIATED_MEMOBJECT> and returns the result. |
1365 |
|
1366 |
=item $int = $mem->offset |
1367 |
|
1368 |
Calls C<clGetMemObjectInfo> with C<CL_MEM_OFFSET> and returns the result. |
1369 |
|
1370 |
=for gengetinfo end mem |
1371 |
|
1372 |
=item ($type, $name) = $mem->gl_object_info |
1373 |
|
1374 |
Returns the OpenGL object type (e.g. OpenCL::GL_OBJECT_TEXTURE2D) and the |
1375 |
object "name" (e.g. the texture name) used to create this memory object. |
1376 |
|
1377 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetGLObjectInfo.html> |
1378 |
|
1379 |
=back |
1380 |
|
1381 |
=head2 THE OpenCL::Buffer CLASS |
1382 |
|
1383 |
This is a subclass of OpenCL::Memory, and the superclass of |
1384 |
OpenCL::BufferObj. Its purpose is simply to distinguish between buffers |
1385 |
and sub-buffers. |
1386 |
|
1387 |
=head2 THE OpenCL::BufferObj CLASS |
1388 |
|
1389 |
This is a subclass of OpenCL::Buffer and thus OpenCL::Memory. It exists |
1390 |
because one cna create sub buffers of OpenLC::BufferObj objects, but not |
1391 |
sub buffers from these sub buffers. |
1392 |
|
1393 |
=over 4 |
1394 |
|
1395 |
=item $subbuf = $buf_obj->sub_buffer_region ($flags, $origin, $size) |
1396 |
|
1397 |
Creates an OpenCL::Buffer objects from this buffer and returns it. The |
1398 |
C<buffer_create_type> is assumed to be C<CL_BUFFER_CREATE_TYPE_REGION>. |
1399 |
|
1400 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateSubBuffer.html> |
1401 |
|
1402 |
=back |
1403 |
|
1404 |
=head2 THE OpenCL::Image CLASS |
1405 |
|
1406 |
This is the superclass of all image objects - OpenCL::Image1D, |
1407 |
OpenCL::Image1DArray, OpenCL::Image1DBuffer, OpenCL::Image2D, |
1408 |
OpenCL::Image2DArray and OpenCL::Image3D. |
1409 |
|
1410 |
=over 4 |
1411 |
|
1412 |
=item $packed_value = $image->image_info ($name) |
1413 |
|
1414 |
See C<< $platform->info >> for details. |
1415 |
|
1416 |
The reason this method is not called C<info> is that there already is an |
1417 |
C<< ->info >> method inherited from C<OpenCL::Memory>. |
1418 |
|
1419 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetImageInfo.html> |
1420 |
|
1421 |
=item ($channel_order, $channel_data_type) = $image->format |
1422 |
|
1423 |
Returns the channel order and type used to create the image by calling |
1424 |
C<clGetImageInfo> with C<CL_IMAGE_FORMAT>. |
1425 |
|
1426 |
=for gengetinfo begin image |
1427 |
|
1428 |
=item $int = $image->element_size |
1429 |
|
1430 |
Calls C<clGetImageInfo> with C<CL_IMAGE_ELEMENT_SIZE> and returns the result. |
1431 |
|
1432 |
=item $int = $image->row_pitch |
1433 |
|
1434 |
Calls C<clGetImageInfo> with C<CL_IMAGE_ROW_PITCH> and returns the result. |
1435 |
|
1436 |
=item $int = $image->slice_pitch |
1437 |
|
1438 |
Calls C<clGetImageInfo> with C<CL_IMAGE_SLICE_PITCH> and returns the result. |
1439 |
|
1440 |
=item $int = $image->width |
1441 |
|
1442 |
Calls C<clGetImageInfo> with C<CL_IMAGE_WIDTH> and returns the result. |
1443 |
|
1444 |
=item $int = $image->height |
1445 |
|
1446 |
Calls C<clGetImageInfo> with C<CL_IMAGE_HEIGHT> and returns the result. |
1447 |
|
1448 |
=item $int = $image->depth |
1449 |
|
1450 |
Calls C<clGetImageInfo> with C<CL_IMAGE_DEPTH> and returns the result. |
1451 |
|
1452 |
=for gengetinfo end image |
1453 |
|
1454 |
=for gengetinfo begin gl_texture |
1455 |
|
1456 |
=item $GLenum = $gl_texture->target |
1457 |
|
1458 |
Calls C<clGetGLTextureInfo> with C<CL_GL_TEXTURE_TARGET> and returns the result. |
1459 |
|
1460 |
=item $GLint = $gl_texture->gl_mipmap_level |
1461 |
|
1462 |
Calls C<clGetGLTextureInfo> with C<CL_GL_MIPMAP_LEVEL> and returns the result. |
1463 |
|
1464 |
=for gengetinfo end gl_texture |
1465 |
|
1466 |
=back |
1467 |
|
1468 |
=head2 THE OpenCL::Sampler CLASS |
1469 |
|
1470 |
=over 4 |
1471 |
|
1472 |
=item $packed_value = $sampler->info ($name) |
1473 |
|
1474 |
See C<< $platform->info >> for details. |
1475 |
|
1476 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetSamplerInfo.html> |
1477 |
|
1478 |
=for gengetinfo begin sampler |
1479 |
|
1480 |
=item $uint = $sampler->reference_count |
1481 |
|
1482 |
Calls C<clGetSamplerInfo> with C<CL_SAMPLER_REFERENCE_COUNT> and returns the result. |
1483 |
|
1484 |
=item $ctx = $sampler->context |
1485 |
|
1486 |
Calls C<clGetSamplerInfo> with C<CL_SAMPLER_CONTEXT> and returns the result. |
1487 |
|
1488 |
=item $addressing_mode = $sampler->normalized_coords |
1489 |
|
1490 |
Calls C<clGetSamplerInfo> with C<CL_SAMPLER_NORMALIZED_COORDS> and returns the result. |
1491 |
|
1492 |
=item $filter_mode = $sampler->addressing_mode |
1493 |
|
1494 |
Calls C<clGetSamplerInfo> with C<CL_SAMPLER_ADDRESSING_MODE> and returns the result. |
1495 |
|
1496 |
=item $boolean = $sampler->filter_mode |
1497 |
|
1498 |
Calls C<clGetSamplerInfo> with C<CL_SAMPLER_FILTER_MODE> and returns the result. |
1499 |
|
1500 |
=for gengetinfo end sampler |
1501 |
|
1502 |
=back |
1503 |
|
1504 |
=head2 THE OpenCL::Program CLASS |
1505 |
|
1506 |
=over 4 |
1507 |
|
1508 |
=item $program->build (\@devices = undef, $options = "", $cb->($program) = undef) |
1509 |
|
1510 |
Tries to build the program with the given options. See also the |
1511 |
C<$ctx->build> convenience function. |
1512 |
|
1513 |
If a callback is specified, then it will be called when compilation is |
1514 |
finished. Note that many OpenCL implementations block your program while |
1515 |
compiling whether you use a callback or not. See C<build_async> if you |
1516 |
want to make sure the build is done in the background. |
1517 |
|
1518 |
Note that some OpenCL implementations atc up badly, and don't call the |
1519 |
callback in some error cases (but call it in others). This implementation |
1520 |
assumes the callback will always be called, and leaks memory if this is |
1521 |
not so. So best make sure you don't pass in invalid values. |
1522 |
|
1523 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clBuildProgram.html> |
1524 |
|
1525 |
=item $program->build_async (\@devices = undef, $options = "", $cb->($program) = undef) |
1526 |
|
1527 |
Similar to C<< ->build >>, except it starts a thread, and never fails (you |
1528 |
need to check the compilation status form the callback, or by polling). |
1529 |
|
1530 |
=item $packed_value = $program->build_info ($device, $name) |
1531 |
|
1532 |
Similar to C<< $platform->info >>, but returns build info for a previous |
1533 |
build attempt for the given device. |
1534 |
|
1535 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetBuildInfo.html> |
1536 |
|
1537 |
=item $kernel = $program->kernel ($function_name) |
1538 |
|
1539 |
Creates an OpenCL::Kernel object out of the named C<__kernel> function in |
1540 |
the program. |
1541 |
|
1542 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateKernel.html> |
1543 |
|
1544 |
=item @kernels = $program->kernels_in_program |
1545 |
|
1546 |
Returns all kernels successfully compiled for all devices in program. |
1547 |
|
1548 |
http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateKernelsInProgram.html |
1549 |
|
1550 |
=for gengetinfo begin program_build |
1551 |
|
1552 |
=item $build_status = $program->build_status ($device) |
1553 |
|
1554 |
Calls C<clGetProgramBuildInfo> with C<CL_PROGRAM_BUILD_STATUS> and returns the result. |
1555 |
|
1556 |
=item $string = $program->build_options ($device) |
1557 |
|
1558 |
Calls C<clGetProgramBuildInfo> with C<CL_PROGRAM_BUILD_OPTIONS> and returns the result. |
1559 |
|
1560 |
=item $string = $program->build_log ($device) |
1561 |
|
1562 |
Calls C<clGetProgramBuildInfo> with C<CL_PROGRAM_BUILD_LOG> and returns the result. |
1563 |
|
1564 |
=for gengetinfo end program_build |
1565 |
|
1566 |
=item $packed_value = $program->info ($name) |
1567 |
|
1568 |
See C<< $platform->info >> for details. |
1569 |
|
1570 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetProgramInfo.html> |
1571 |
|
1572 |
=for gengetinfo begin program |
1573 |
|
1574 |
=item $uint = $program->reference_count |
1575 |
|
1576 |
Calls C<clGetProgramInfo> with C<CL_PROGRAM_REFERENCE_COUNT> and returns the result. |
1577 |
|
1578 |
=item $ctx = $program->context |
1579 |
|
1580 |
Calls C<clGetProgramInfo> with C<CL_PROGRAM_CONTEXT> and returns the result. |
1581 |
|
1582 |
=item $uint = $program->num_devices |
1583 |
|
1584 |
Calls C<clGetProgramInfo> with C<CL_PROGRAM_NUM_DEVICES> and returns the result. |
1585 |
|
1586 |
=item @devices = $program->devices |
1587 |
|
1588 |
Calls C<clGetProgramInfo> with C<CL_PROGRAM_DEVICES> and returns the result. |
1589 |
|
1590 |
=item $string = $program->source |
1591 |
|
1592 |
Calls C<clGetProgramInfo> with C<CL_PROGRAM_SOURCE> and returns the result. |
1593 |
|
1594 |
=item @ints = $program->binary_sizes |
1595 |
|
1596 |
Calls C<clGetProgramInfo> with C<CL_PROGRAM_BINARY_SIZES> and returns the result. |
1597 |
|
1598 |
=for gengetinfo end program |
1599 |
|
1600 |
=item @blobs = $program->binaries |
1601 |
|
1602 |
Returns a string for the compiled binary for every device associated with |
1603 |
the program, empty strings indicate missing programs, and an empty result |
1604 |
means no program binaries are available. |
1605 |
|
1606 |
These "binaries" are often, in fact, informative low-level assembly |
1607 |
sources. |
1608 |
|
1609 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetProgramInfo.html> |
1610 |
|
1611 |
=back |
1612 |
|
1613 |
=head2 THE OpenCL::Kernel CLASS |
1614 |
|
1615 |
=over 4 |
1616 |
|
1617 |
=item $packed_value = $kernel->info ($name) |
1618 |
|
1619 |
See C<< $platform->info >> for details. |
1620 |
|
1621 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetKernelInfo.html> |
1622 |
|
1623 |
=for gengetinfo begin kernel |
1624 |
|
1625 |
=item $string = $kernel->function_name |
1626 |
|
1627 |
Calls C<clGetKernelInfo> with C<CL_KERNEL_FUNCTION_NAME> and returns the result. |
1628 |
|
1629 |
=item $uint = $kernel->num_args |
1630 |
|
1631 |
Calls C<clGetKernelInfo> with C<CL_KERNEL_NUM_ARGS> and returns the result. |
1632 |
|
1633 |
=item $uint = $kernel->reference_count |
1634 |
|
1635 |
Calls C<clGetKernelInfo> with C<CL_KERNEL_REFERENCE_COUNT> and returns the result. |
1636 |
|
1637 |
=item $ctx = $kernel->context |
1638 |
|
1639 |
Calls C<clGetKernelInfo> with C<CL_KERNEL_CONTEXT> and returns the result. |
1640 |
|
1641 |
=item $program = $kernel->program |
1642 |
|
1643 |
Calls C<clGetKernelInfo> with C<CL_KERNEL_PROGRAM> and returns the result. |
1644 |
|
1645 |
=for gengetinfo end kernel |
1646 |
|
1647 |
=item $packed_value = $kernel->work_group_info ($device, $name) |
1648 |
|
1649 |
See C<< $platform->info >> for details. |
1650 |
|
1651 |
The reason this method is not called C<info> is that there already is an |
1652 |
C<< ->info >> method. |
1653 |
|
1654 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetKernelWorkGroupInfo.html> |
1655 |
|
1656 |
=for gengetinfo begin kernel_work_group |
1657 |
|
1658 |
=item $int = $kernel->work_group_size ($device) |
1659 |
|
1660 |
Calls C<clGetKernelWorkGroupInfo> with C<CL_KERNEL_WORK_GROUP_SIZE> and returns the result. |
1661 |
|
1662 |
=item @ints = $kernel->compile_work_group_size ($device) |
1663 |
|
1664 |
Calls C<clGetKernelWorkGroupInfo> with C<CL_KERNEL_COMPILE_WORK_GROUP_SIZE> and returns the result. |
1665 |
|
1666 |
=item $ulong = $kernel->local_mem_size ($device) |
1667 |
|
1668 |
Calls C<clGetKernelWorkGroupInfo> with C<CL_KERNEL_LOCAL_MEM_SIZE> and returns the result. |
1669 |
|
1670 |
=item $int = $kernel->preferred_work_group_size_multiple ($device) |
1671 |
|
1672 |
Calls C<clGetKernelWorkGroupInfo> with C<CL_KERNEL_PREFERRED_WORK_GROUP_SIZE_MULTIPLE> and returns the result. |
1673 |
|
1674 |
=item $ulong = $kernel->private_mem_size ($device) |
1675 |
|
1676 |
Calls C<clGetKernelWorkGroupInfo> with C<CL_KERNEL_PRIVATE_MEM_SIZE> and returns the result. |
1677 |
|
1678 |
=for gengetinfo end kernel_work_group |
1679 |
|
1680 |
=item $kernel->set_TYPE ($index, $value) |
1681 |
|
1682 |
This is a family of methods to set the kernel argument with the number C<$index> to the give C<$value>. |
1683 |
|
1684 |
TYPE is one of C<char>, C<uchar>, C<short>, C<ushort>, C<int>, C<uint>, |
1685 |
C<long>, C<ulong>, C<half>, C<float>, C<double>, C<memory>, C<buffer>, |
1686 |
C<image2d>, C<image3d>, C<sampler>, C<local> or C<event>. |
1687 |
|
1688 |
Chars and integers (including the half type) are specified as integers, |
1689 |
float and double as floating point values, memory/buffer/image2d/image3d |
1690 |
must be an object of that type or C<undef>, local-memory arguments are |
1691 |
set by specifying the size, and sampler and event must be objects of that |
1692 |
type. |
1693 |
|
1694 |
Setting an argument for a kernel does NOT keep a reference to the object - |
1695 |
for example, if you set an argument to some image object, free the image, |
1696 |
and call the kernel, you will run into undefined behaviour. |
1697 |
|
1698 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetKernelArg.html> |
1699 |
|
1700 |
=back |
1701 |
|
1702 |
=head2 THE OpenCL::Event CLASS |
1703 |
|
1704 |
This is the superclass for all event objects (including OpenCL::UserEvent |
1705 |
objects). |
1706 |
|
1707 |
=over 4 |
1708 |
|
1709 |
=item $ev->wait |
1710 |
|
1711 |
Waits for the event to complete. |
1712 |
|
1713 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clWaitForEvents.html> |
1714 |
|
1715 |
=item $ev->cb ($exec_callback_type, $callback->($event, $event_command_exec_status)) |
1716 |
|
1717 |
Adds a callback to the callback stack for the given event type. There is |
1718 |
no way to remove a callback again. |
1719 |
|
1720 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetEventCallback.html> |
1721 |
|
1722 |
=item $packed_value = $ev->info ($name) |
1723 |
|
1724 |
See C<< $platform->info >> for details. |
1725 |
|
1726 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetEventInfo.html> |
1727 |
|
1728 |
=for gengetinfo begin event |
1729 |
|
1730 |
=item $queue = $event->command_queue |
1731 |
|
1732 |
Calls C<clGetEventInfo> with C<CL_EVENT_COMMAND_QUEUE> and returns the result. |
1733 |
|
1734 |
=item $command_type = $event->command_type |
1735 |
|
1736 |
Calls C<clGetEventInfo> with C<CL_EVENT_COMMAND_TYPE> and returns the result. |
1737 |
|
1738 |
=item $uint = $event->reference_count |
1739 |
|
1740 |
Calls C<clGetEventInfo> with C<CL_EVENT_REFERENCE_COUNT> and returns the result. |
1741 |
|
1742 |
=item $uint = $event->command_execution_status |
1743 |
|
1744 |
Calls C<clGetEventInfo> with C<CL_EVENT_COMMAND_EXECUTION_STATUS> and returns the result. |
1745 |
|
1746 |
=item $ctx = $event->context |
1747 |
|
1748 |
Calls C<clGetEventInfo> with C<CL_EVENT_CONTEXT> and returns the result. |
1749 |
|
1750 |
=for gengetinfo end event |
1751 |
|
1752 |
=item $packed_value = $ev->profiling_info ($name) |
1753 |
|
1754 |
See C<< $platform->info >> for details. |
1755 |
|
1756 |
The reason this method is not called C<info> is that there already is an |
1757 |
C<< ->info >> method. |
1758 |
|
1759 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetProfilingInfo.html> |
1760 |
|
1761 |
=for gengetinfo begin profiling |
1762 |
|
1763 |
=item $ulong = $event->profiling_command_queued |
1764 |
|
1765 |
Calls C<clGetEventProfilingInfo> with C<CL_PROFILING_COMMAND_QUEUED> and returns the result. |
1766 |
|
1767 |
=item $ulong = $event->profiling_command_submit |
1768 |
|
1769 |
Calls C<clGetEventProfilingInfo> with C<CL_PROFILING_COMMAND_SUBMIT> and returns the result. |
1770 |
|
1771 |
=item $ulong = $event->profiling_command_start |
1772 |
|
1773 |
Calls C<clGetEventProfilingInfo> with C<CL_PROFILING_COMMAND_START> and returns the result. |
1774 |
|
1775 |
=item $ulong = $event->profiling_command_end |
1776 |
|
1777 |
Calls C<clGetEventProfilingInfo> with C<CL_PROFILING_COMMAND_END> and returns the result. |
1778 |
|
1779 |
=for gengetinfo end profiling |
1780 |
|
1781 |
=back |
1782 |
|
1783 |
=head2 THE OpenCL::UserEvent CLASS |
1784 |
|
1785 |
This is a subclass of OpenCL::Event. |
1786 |
|
1787 |
=over 4 |
1788 |
|
1789 |
=item $ev->set_status ($execution_status) |
1790 |
|
1791 |
Sets the execution status of the user event. Can only be called once, |
1792 |
either with OpenCL::COMPLETE or a negative number as status. |
1793 |
|
1794 |
L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetUserEventStatus.html> |
1795 |
|
1796 |
=back |
1797 |
|
1798 |
=cut |
1799 |
|
1800 |
1; |
1801 |
|
1802 |
=head1 AUTHOR |
1803 |
|
1804 |
Marc Lehmann <schmorp@schmorp.de> |
1805 |
http://home.schmorp.de/ |
1806 |
|
1807 |
=cut |
1808 |
|