ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/OpenCL/OpenCL.pm
(Generate patch)

Comparing OpenCL/OpenCL.pm (file contents):
Revision 1.34 by root, Thu Apr 19 13:06:55 2012 UTC vs.
Revision 1.62 by root, Sat Apr 28 13:33:04 2012 UTC

43 43
44OpenCL::Event objects are used to signal when something is complete. 44OpenCL::Event objects are used to signal when something is complete.
45 45
46=head2 HELPFUL RESOURCES 46=head2 HELPFUL RESOURCES
47 47
48The OpenCL spec used to develop this module (1.2 spec was available, but 48The OpenCL specs used to develop this module:
49no implementation was available to me :).
50 49
51 http://www.khronos.org/registry/cl/specs/opencl-1.1.pdf 50 http://www.khronos.org/registry/cl/specs/opencl-1.1.pdf
51 http://www.khronos.org/registry/cl/specs/opencl-1.2.pdf
52 http://www.khronos.org/registry/cl/specs/opencl-1.2-extensions.pdf
52 53
53OpenCL manpages: 54OpenCL manpages:
54 55
55 http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/ 56 http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/
57 http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/
56 58
57If you are into UML class diagrams, the following diagram might help - if 59If you are into UML class diagrams, the following diagram might help - if
58not, it will be mildly cobfusing: 60not, it will be mildly confusing (also, the class hierarchy of this module
61is much more fine-grained):
59 62
60 http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/classDiagram.html 63 http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/classDiagram.html
61 64
62Here's a tutorial from AMD (very AMD-centric, too), not sure how useful it 65Here's a tutorial from AMD (very AMD-centric, too), not sure how useful it
63is, but at least it's free of charge: 66is, but at least it's free of charge:
64 67
65 http://developer.amd.com/zones/OpenCLZone/courses/Documents/Introduction_to_OpenCL_Programming%20Training_Guide%20%28201005%29.pdf 68 http://developer.amd.com/zones/OpenCLZone/courses/Documents/Introduction_to_OpenCL_Programming%20Training_Guide%20%28201005%29.pdf
138=head2 Create a buffer with some predefined data, read it back synchronously, 141=head2 Create a buffer with some predefined data, read it back synchronously,
139then asynchronously. 142then asynchronously.
140 143
141 my $buf = $ctx->buffer_sv (OpenCL::MEM_COPY_HOST_PTR, "helmut"); 144 my $buf = $ctx->buffer_sv (OpenCL::MEM_COPY_HOST_PTR, "helmut");
142 145
143 $queue->enqueue_read_buffer ($buf, 1, 1, 3, my $data); 146 $queue->read_buffer ($buf, 1, 1, 3, my $data);
144 print "$data\n"; 147 print "$data\n";
145 148
146 my $ev = $queue->enqueue_read_buffer ($buf, 0, 1, 3, my $data); 149 my $ev = $queue->read_buffer ($buf, 0, 1, 3, my $data);
147 $ev->wait; 150 $ev->wait;
148 print "$data\n"; # prints "elm" 151 print "$data\n"; # prints "elm"
149 152
150=head2 Create and build a program, then create a kernel out of one of its 153=head2 Create and build a program, then create a kernel out of one of its
151functions. 154functions.
157 $id = get_global_id (0); 160 $id = get_global_id (0);
158 output [id] = input [id] * input [id]; 161 output [id] = input [id] * input [id];
159 } 162 }
160 '; 163 ';
161 164
162 my $prog = $ctx->program_with_source ($src); 165 my $prog = $ctx->build_program ($src);
163
164 # build croaks on compile errors, so catch it and print the compile errors
165 eval { $prog->build ($dev); 1 }
166 or die $prog->build_log;
167
168 my $kernel = $prog->kernel ("squareit"); 166 my $kernel = $prog->kernel ("squareit");
169 167
170=head2 Create some input and output float buffers, then call the 168=head2 Create some input and output float buffers, then call the
171'squareit' kernel on them. 169'squareit' kernel on them.
172 170
176 # set buffer 174 # set buffer
177 $kernel->set_buffer (0, $input); 175 $kernel->set_buffer (0, $input);
178 $kernel->set_buffer (1, $output); 176 $kernel->set_buffer (1, $output);
179 177
180 # execute it for all 4 numbers 178 # execute it for all 4 numbers
181 $queue->enqueue_nd_range_kernel ($kernel, undef, [4], undef); 179 $queue->nd_range_kernel ($kernel, undef, [4], undef);
182 180
183 # enqueue a synchronous read 181 # enqueue a synchronous read
184 $queue->enqueue_read_buffer ($output, 1, 0, OpenCL::SIZEOF_FLOAT * 4, my $data); 182 $queue->read_buffer ($output, 1, 0, OpenCL::SIZEOF_FLOAT * 4, my $data);
185 183
186 # print the results: 184 # print the results:
187 printf "%s\n", join ", ", unpack "f*", $data; 185 printf "%s\n", join ", ", unpack "f*", $data;
188 186
189=head2 The same enqueue operations as before, but assuming an out-of-order queue, 187=head2 The same enqueue operations as before, but assuming an out-of-order queue,
190showing off barriers. 188showing off barriers.
191 189
192 # execute it for all 4 numbers 190 # execute it for all 4 numbers
193 $queue->enqueue_nd_range_kernel ($kernel, undef, [4], undef); 191 $queue->nd_range_kernel ($kernel, undef, [4], undef);
194 192
195 # enqueue a barrier to ensure in-order execution 193 # enqueue a barrier to ensure in-order execution
196 $queue->enqueue_barrier; 194 $queue->barrier;
197 195
198 # enqueue an async read 196 # enqueue an async read
199 $queue->enqueue_read_buffer ($output, 0, 0, OpenCL::SIZEOF_FLOAT * 4, my $data); 197 $queue->read_buffer ($output, 0, 0, OpenCL::SIZEOF_FLOAT * 4, my $data);
200 198
201 # wait for all requests to finish 199 # wait for all requests to finish
202 $queue->finish; 200 $queue->finish;
203 201
204=head2 The same enqueue operations as before, but assuming an out-of-order queue, 202=head2 The same enqueue operations as before, but assuming an out-of-order queue,
205showing off event objects and wait lists. 203showing off event objects and wait lists.
206 204
207 # execute it for all 4 numbers 205 # execute it for all 4 numbers
208 my $ev = $queue->enqueue_nd_range_kernel ($kernel, undef, [4], undef); 206 my $ev = $queue->nd_range_kernel ($kernel, undef, [4], undef);
209 207
210 # enqueue an async read 208 # enqueue an async read
211 $ev = $queue->enqueue_read_buffer ($output, 0, 0, OpenCL::SIZEOF_FLOAT * 4, my $data, $ev); 209 $ev = $queue->read_buffer ($output, 0, 0, OpenCL::SIZEOF_FLOAT * 4, my $data, $ev);
212 210
213 # wait for the last event to complete 211 # wait for the last event to complete
214 $ev->wait; 212 $ev->wait;
213
214=head2 Use the OpenGL module to share a texture between OpenCL and OpenGL and draw some julia
215set tunnel effect.
216
217This is quite a long example to get you going.
218
219 use OpenGL ":all";
220 use OpenCL;
221
222 # open a window and create a gl texture
223 OpenGL::glpOpenWindow width => 256, height => 256;
224 my $texid = glGenTextures_p 1;
225 glBindTexture GL_TEXTURE_2D, $texid;
226 glTexImage2D_c GL_TEXTURE_2D, 0, GL_RGBA8, 256, 256, 0, GL_RGBA, GL_UNSIGNED_BYTE, 0;
227
228 # find and use the first opencl device that let's us get a shared opengl context
229 my $platform;
230 my $dev;
231 my $ctx;
232
233 for (OpenCL::platforms) {
234 $platform = $_;
235 for ($platform->devices) {
236 $dev = $_;
237 $ctx = $platform->context ([OpenCL::GLX_DISPLAY_KHR, undef, OpenCL::GL_CONTEXT_KHR, undef], [$dev])
238 and last;
239 }
240 }
241
242 $ctx
243 or die "cannot find suitable OpenCL device\n";
244
245 my $queue = $ctx->queue ($dev);
246
247 # now attach an opencl image2d object to the opengl texture
248 my $tex = $ctx->gl_texture2d (OpenCL::MEM_WRITE_ONLY, GL_TEXTURE_2D, 0, $texid);
249
250 # now the boring opencl code
251 my $src = <<EOF;
252 kernel void
253 juliatunnel (write_only image2d_t img, float time)
254 {
255 float2 p = (float2)(get_global_id (0), get_global_id (1)) / 256.f * 2.f - 1.f;
256
257 float2 m = (float2)(1.f, p.y) / fabs (p.x);
258 m.x = fabs (fmod (m.x + time * 0.05f, 4.f)) - 2.f;
259
260 float2 z = m;
261 float2 c = (float2)(sin (time * 0.05005), cos (time * 0.06001));
262
263 for (int i = 0; i < 25 && dot (z, z) < 4.f; ++i)
264 z = (float2)(z.x * z.x - z.y * z.y, 2.f * z.x * z.y) + c;
265
266 float3 colour = (float3)(z.x, z.y, z.x * z.y);
267 write_imagef (img, (int2)(get_global_id (0), get_global_id (1)), (float4)(colour * p.x * p.x, 1.));
268 }
269 EOF
270
271 my $prog = $ctx->build_program ($src);
272 my $kernel = $prog->kernel ("juliatunnel");
273
274 # program compiled, kernel ready, now draw and loop
275
276 for (my $time; ; ++$time) {
277 # acquire objects from opengl
278 $queue->acquire_gl_objects ([$tex]);
279
280 # configure and run our kernel
281 $kernel->set_image2d (0, $tex);
282 $kernel->set_float (1, $time);
283 $queue->nd_range_kernel ($kernel, undef, [256, 256], undef);
284
285 # release objects to opengl again
286 $queue->release_gl_objects ([$tex]);
287
288 # wait
289 $queue->finish;
290
291 # now draw the texture, the defaults should be all right
292 glTexParameterf GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST;
293
294 glEnable GL_TEXTURE_2D;
295 glBegin GL_QUADS;
296 glTexCoord2f 0, 1; glVertex3i -1, -1, -1;
297 glTexCoord2f 0, 0; glVertex3i 1, -1, -1;
298 glTexCoord2f 1, 0; glVertex3i 1, 1, -1;
299 glTexCoord2f 1, 1; glVertex3i -1, 1, -1;
300 glEnd;
301
302 glXSwapBuffers;
303
304 select undef, undef, undef, 1/60;
305 }
215 306
216=head1 DOCUMENTATION 307=head1 DOCUMENTATION
217 308
218=head2 BASIC CONVENTIONS 309=head2 BASIC CONVENTIONS
219 310
241=item * Structures are often specified by flattening out their components 332=item * Structures are often specified by flattening out their components
242as with short vectors, and returned as arrayrefs. 333as with short vectors, and returned as arrayrefs.
243 334
244=item * When enqueuing commands, the wait list is specified by adding 335=item * When enqueuing commands, the wait list is specified by adding
245extra arguments to the function - anywhere a C<$wait_events...> argument 336extra arguments to the function - anywhere a C<$wait_events...> argument
246is documented this can be any number of event objects. 337is documented this can be any number of event objects. As an extsnion
338implemented by this module, C<undef> values will be ignored in the event
339list.
247 340
248=item * When enqueuing commands, if the enqueue method is called in void 341=item * When enqueuing commands, if the enqueue method is called in void
249context, no event is created. In all other contexts an event is returned 342context, no event is created. In all other contexts an event is returned
250by the method. 343by the method.
251 344
271 ulong IV - Q 364 ulong IV - Q
272 float NV float f 365 float NV float f
273 half IV ushort S 366 half IV ushort S
274 double NV double d 367 double NV double d
275 368
369=head2 GLX SUPPORT
370
371Due to the sad state that OpenGL support is in in Perl (mostly the OpenGL
372module, which has little to no documentation and has little to no support
373for glX), this module, as a special extension, treats context creation
374properties C<OpenCL::GLX_DISPLAY_KHR> and C<OpenCL::GL_CONTEXT_KHR>
375specially: If either or both of these are C<undef>, then the OpenCL
376module tries to dynamically resolve C<glXGetCurrentDisplay> and
377C<glXGetCurrentContext>, call these functions and use their return values
378instead.
379
380For this to work, the OpenGL library must be loaded, a GLX context must
381have been created and be made current, and C<dlsym> must be available and
382capable of finding the function via C<RTLD_DEFAULT>.
383
384=head2 EVENT SYSTEM
385
386OpenCL can generate a number of (potentially) asynchronous events, for
387example, after compiling a program, to signal a context-related error or,
388perhaps most important, to signal completion of queued jobs (by setting
389callbacks on OpenCL::Event objects).
390
391To facilitate this, this module maintains an event queue - each
392time an asynchronous event happens, it is queued, and perl will be
393interrupted. This is implemented via the L<Async::Interrupt> module. In
394addition, this module has L<AnyEvent> support, so it can seamlessly
395integrate itself into many event loops.
396
397Since this module is a bit hard to understand, here are some case examples:
398
399=head3 Don't use callbacks.
400
401When your program never uses any callbacks, then there will never be any
402notifications you need to take care of, and therefore no need to worry
403about all this.
404
405You can achieve a great deal by explicitly waiting for events, or using
406barriers and flush calls. In many programs, there is no need at all to
407tinker with asynchronous events.
408
409=head3 Use AnyEvent
410
411This module automatically registers a watcher that invokes all outstanding
412event callbacks when AnyEvent is initialised (and block asynchronous
413interruptions). Using this mode of operations is the safest and most
414recommended one.
415
416To use this, simply use AnyEvent and this module normally, make sure you
417have an event loop running:
418
419 use Gtk2 -init;
420 use AnyEvent;
421
422 # initialise AnyEvent, by creating a watcher, or:
423 AnyEvent::detect;
424
425 my $e = $queue->marker;
426 $e->cb (sub {
427 warn "opencl is finished\n";
428 })
429
430 main Gtk2;
431
432Note that this module will not initialise AnyEvent for you. Before
433AnyEvent is initialised, the module will asynchronously interrupt perl
434instead. To avoid any surprises, it's best to explicitly initialise
435AnyEvent.
436
437You can temporarily enable asynchronous interruptions (see next paragraph)
438by calling C<$OpenCL::INTERRUPT->unblock> and disable them again by
439calling C<$OpenCL::INTERRUPT->block>.
440
441=head3 Let yourself be interrupted at any time
442
443This mode is the default unless AnyEvent is loaded and initialised. In
444this mode, OpenCL asynchronously interrupts a running perl program. The
445emphasis is on both I<asynchronously> and I<running> here.
446
447Asynchronously means that perl might execute your callbacks at any
448time. For example, in the following code (I<THAT YOU SHOULD NOT COPY>),
449the C<until> loop following the marker call will be interrupted by the
450callback:
451
452 my $e = $queue->marker;
453 my $flag;
454 $e->cb (sub { $flag = 1 });
455 1 until $flag;
456 # $flag is now 1
457
458The reason why you shouldn't blindly copy the above code is that
459busy waiting is a really really bad thing, and really really bad for
460performance.
461
462While at first this asynchronous business might look exciting, it can be
463really hard, because you need to be prepared for the callback code to be
464executed at any time, which limits the amount of things the callback code
465can do safely.
466
467This can be mitigated somewhat by using C<<
468$OpenCL::INTERRUPT->scope_block >> (see the L<Async::Interrupt>
469documentation for details).
470
471The other problem is that your program must be actively I<running> to be
472interrupted. When you calculate stuff, your program is running. When you
473hang in some C functions or other block execution (by calling C<sleep>,
474C<select>, running an event loop and so on), your program is waiting, not
475running.
476
477One way around that would be to attach a read watcher to your event loop,
478listening for events on C<< $OpenCL::INTERRUPT->pipe_fileno >>, using a
479dummy callback (C<sub { }>) to temporarily execute some perl code.
480
481That is then awfully close to using the built-in AnyEvent support above,
482though, so consider that one instead.
483
484=head3 Be creative
485
486OpenCL exports the L<Async::Interrupt> object it uses in the global
487variable C<$OpenCL::INTERRUPT>. You can configure it in any way you like.
488
489So if you want to feel like a real pro, err, wait, if you feel no risk
490menas no fun, you can experiment by implementing your own mode of
491operations.
492
493=cut
494
495package OpenCL;
496
497use common::sense;
498use Carp ();
499use Async::Interrupt ();
500
501our $POLL_FUNC; # set by XS
502
503BEGIN {
504 our $VERSION = '0.98';
505
506 require XSLoader;
507 XSLoader::load (__PACKAGE__, $VERSION);
508
509 @OpenCL::Platform::ISA =
510 @OpenCL::Device::ISA =
511 @OpenCL::Context::ISA =
512 @OpenCL::Queue::ISA =
513 @OpenCL::Memory::ISA =
514 @OpenCL::Sampler::ISA =
515 @OpenCL::Program::ISA =
516 @OpenCL::Kernel::ISA =
517 @OpenCL::Event::ISA = OpenCL::Object::;
518
519 @OpenCL::Buffer::ISA =
520 @OpenCL::Image::ISA = OpenCL::Memory::;
521
522 @OpenCL::BufferObj::ISA = OpenCL::Buffer::;
523
524 @OpenCL::Image2D::ISA =
525 @OpenCL::Image3D::ISA =
526 @OpenCL::Image2DArray::ISA =
527 @OpenCL::Image1D::ISA =
528 @OpenCL::Image1DArray::ISA =
529 @OpenCL::Image1DBuffer::ISA = OpenCL::Image::;
530
531 @OpenCL::UserEvent::ISA = OpenCL::Event::;
532}
533
276=head2 THE OpenCL PACKAGE 534=head2 THE OpenCL PACKAGE
277 535
278=over 4 536=over 4
279 537
280=item $int = OpenCL::errno 538=item $int = OpenCL::errno
281 539
282The last error returned by a function - it's only valid after an error occured 540The last error returned by a function - it's only valid after an error occured
283and before calling another OpenCL function. 541and before calling another OpenCL function.
284 542
285=item $str = OpenCL::err2str $errval 543=item $str = OpenCL::err2str [$errval]
286 544
287Comverts an error value into a human readable string. 545Converts an error value into a human readable string. IF no error value is
546given, then the last error will be used (as returned by OpenCL::errno).
288 547
289=item $str = OpenCL::enum2str $enum 548=item $str = OpenCL::enum2str $enum
290 549
291Converts most enum values (of parameter names, image format constants, 550Converts most enum values (of parameter names, image format constants,
292object types, addressing and filter modes, command types etc.) into a 551object types, addressing and filter modes, command types etc.) into a
298 557
299Returns all available OpenCL::Platform objects. 558Returns all available OpenCL::Platform objects.
300 559
301L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetPlatformIDs.html> 560L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetPlatformIDs.html>
302 561
303=item $ctx = OpenCL::context_from_type $properties, $type = OpenCL::DEVICE_TYPE_DEFAULT, $notify = undef 562=item $ctx = OpenCL::context_from_type $properties, $type = OpenCL::DEVICE_TYPE_DEFAULT, $callback->($err, $pvt) = $print_stderr
304 563
305Tries to create a context from a default device and platform - never worked for me. 564Tries to create a context from a default device and platform type - never worked for me.
306 565
307L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html> 566L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html>
308 567
568=item $ctx = OpenCL::context $properties, \@devices, $callback->($err, $pvt) = $print_stderr)
569
570Create a new OpenCL::Context object using the given device object(s). This
571function isn't implemented yet, use C<< $platform->context >> instead.
572
573L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContext.html>
574
309=item OpenCL::wait_for_events $wait_events... 575=item OpenCL::wait_for_events $wait_events...
310 576
311Waits for all events to complete. 577Waits for all events to complete.
312 578
313L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clWaitForEvents.html> 579L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clWaitForEvents.html>
314 580
581=item OpenCL::poll
582
583Checks if there are any outstanding events (see L<EVENT SYSTEM>) and
584invokes their callbacks.
585
586=item $OpenCL::INTERRUPT
587
588The L<Async::Interrupt> object used to signal asynchronous events (see
589L<EVENT SYSTEM>).
590
591=cut
592
593our $INTERRUPT = new Async::Interrupt c_cb => [$POLL_FUNC, 0];
594
595&_eq_initialise ($INTERRUPT->signal_func);
596
597=item $OpenCL::WATCHER
598
599The L<AnyEvent> watcher object used to watch for asynchronous events (see
600L<EVENT SYSTEM>). This variable is C<undef> until L<AnyEvent> has been
601loaded I<and> initialised (e.g. by calling C<AnyEvent::detect>).
602
603=cut
604
605our $WATCHER;
606
607sub _init_anyevent {
608 $INTERRUPT->block;
609 $WATCHER = AE::io ($INTERRUPT->pipe_fileno, 0, sub { $INTERRUPT->handle });
610}
611
612if (defined $AnyEvent::MODEL) {
613 _init_anyevent;
614} else {
615 push @AnyEvent::post_detect, \&_init_anyevent;
616}
617
315=back 618=back
316 619
620=head2 THE OpenCL::Object CLASS
621
622This is the base class for all objects in the OpenCL module. The only
623method it implements is the C<id> method, which is only useful if you want
624to interface to OpenCL on the C level.
625
626=over 4
627
628=item $iv = $obj->id
629
630OpenCL objects are represented by pointers or integers on the C level. If
631you want to interface to an OpenCL object directly on the C level, then
632you need this value, which is returned by this method. You should use an
633C<IV> type in your code and cast that to the correct type.
634
635=cut
636
637sub OpenCL::Object::id {
638 ref $_[0] eq "SCALAR"
639 ? ${ $_[0] }
640 : $_[0][0]
641}
642
643=back
644
317=head2 THE OpenCL::Platform CLASS 645=head2 THE OpenCL::Platform CLASS
318 646
319=over 4 647=over 4
320 648
321=item @devices = $platform->devices ($type = OpenCL::DEVICE_TYPE_ALL) 649=item @devices = $platform->devices ($type = OpenCL::DEVICE_TYPE_ALL)
322 650
323Returns a list of matching OpenCL::Device objects. 651Returns a list of matching OpenCL::Device objects.
324 652
325=item $ctx = $platform->context_from_type ($properties, $type = OpenCL::DEVICE_TYPE_DEFAULT, $notify = undef) 653=item $ctx = $platform->context_from_type ($properties, $type = OpenCL::DEVICE_TYPE_DEFAULT, $callback->($err, $pvt) = $print_stderr)
326 654
327Tries to create a context. Never worked for me, and you need devices explicitly anyway. 655Tries to create a context. Never worked for me, and you need devices explicitly anyway.
328 656
329L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html> 657L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html>
330 658
331=item $ctx = $platform->context ($properties = undef, @$devices, $notify = undef) 659=item $ctx = $platform->context ($properties, \@devices, $callback->($err, $pvt) = $print_stderr)
332 660
333Create a new OpenCL::Context object using the given device object(s)- a 661Create a new OpenCL::Context object using the given device object(s)- a
334CL_CONTEXT_PLATFORM property is supplied automatically. 662CL_CONTEXT_PLATFORM property is supplied automatically.
335 663
336L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContext.html> 664L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContext.html>
344It's best to avoid this method and use one of the following convenience 672It's best to avoid this method and use one of the following convenience
345wrappers. 673wrappers.
346 674
347L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetPlatformInfo.html> 675L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetPlatformInfo.html>
348 676
677=item $platform->unload_compiler
678
679Attempts to unload the compiler for this platform, for endless
680profit. Does nothing on OpenCL 1.1.
681
682L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clUnloadPlatformCompiler.html>
683
349=for gengetinfo begin platform 684=for gengetinfo begin platform
350 685
351=item $string = $platform->profile 686=item $string = $platform->profile
352 687
353Calls C<clGetPlatformInfo> with C<CL_PLATFORM_PROFILE> and returns the result. 688Calls C<clGetPlatformInfo> with C<CL_PLATFORM_PROFILE> and returns the result.
638 973
639=item @device_partition_property_exts = $device->affinity_domains_ext 974=item @device_partition_property_exts = $device->affinity_domains_ext
640 975
641Calls C<clGetDeviceInfo> with C<CL_DEVICE_AFFINITY_DOMAINS_EXT> and returns the result. 976Calls C<clGetDeviceInfo> with C<CL_DEVICE_AFFINITY_DOMAINS_EXT> and returns the result.
642 977
643=item $uint = $device->reference_count_ext 978=item $uint = $device->reference_count_ext
644 979
645Calls C<clGetDeviceInfo> with C<CL_DEVICE_REFERENCE_COUNT_EXT > and returns the result. 980Calls C<clGetDeviceInfo> with C<CL_DEVICE_REFERENCE_COUNT_EXT> and returns the result.
646 981
647=item @device_partition_property_exts = $device->partition_style_ext 982=item @device_partition_property_exts = $device->partition_style_ext
648 983
649Calls C<clGetDeviceInfo> with C<CL_DEVICE_PARTITION_STYLE_EXT> and returns the result. 984Calls C<clGetDeviceInfo> with C<CL_DEVICE_PARTITION_STYLE_EXT> and returns the result.
650 985
654 989
655=head2 THE OpenCL::Context CLASS 990=head2 THE OpenCL::Context CLASS
656 991
657=over 4 992=over 4
658 993
994=item $prog = $ctx->build_program ($program, $options = "")
995
996This convenience function tries to build the program on all devices in
997the context. If the build fails, then the function will C<croak> with the
998build log. Otherwise ti returns the program object.
999
1000The C<$program> can either be a C<OpenCL::Program> object or a string
1001containing the program. In the latter case, a program objetc will be
1002created automatically.
1003
1004=cut
1005
1006sub OpenCL::Context::build_program {
1007 my ($self, $prog, $options) = @_;
1008
1009 $prog = $self->program_with_source ($prog)
1010 unless ref $prog;
1011
1012 eval { $prog->build (undef, $options); 1 }
1013 or errno == BUILD_PROGRAM_FAILURE
1014 or Carp::croak "OpenCL::Context->build_program: " . err2str;
1015
1016 # we check status for all devices
1017 for my $dev ($self->devices) {
1018 $prog->build_status ($dev) == BUILD_SUCCESS
1019 or Carp::croak "Building OpenCL program for device '" . $dev->name . "' failed:\n"
1020 . $prog->build_log ($dev);
1021 }
1022
1023 $prog
1024}
1025
659=item $queue = $ctx->queue ($device, $properties) 1026=item $queue = $ctx->queue ($device, $properties)
660 1027
661Create a new OpenCL::Queue object from the context and the given device. 1028Create a new OpenCL::Queue object from the context and the given device.
662 1029
663L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateCommandQueue.html> 1030L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateCommandQueue.html>
1031
1032Example: create an out-of-order queue.
1033
1034 $queue = $ctx->queue ($device, OpenCL::QUEUE_OUT_OF_ORDER_EXEC_MODE_ENABLE);
664 1035
665=item $ev = $ctx->user_event 1036=item $ev = $ctx->user_event
666 1037
667Creates a new OpenCL::UserEvent object. 1038Creates a new OpenCL::UserEvent object.
668 1039
678=item $buf = $ctx->buffer_sv ($flags, $data) 1049=item $buf = $ctx->buffer_sv ($flags, $data)
679 1050
680Creates a new OpenCL::Buffer (actually OpenCL::BufferObj) object and 1051Creates a new OpenCL::Buffer (actually OpenCL::BufferObj) object and
681initialise it with the given data values. 1052initialise it with the given data values.
682 1053
1054=item $img = $ctx->image ($self, $flags, $channel_order, $channel_type, $type, $width, $height, $depth = 0, $array_size = 0, $row_pitch = 0, $slice_pitch = 0, $num_mip_level = 0, $num_samples = 0, $*data = &PL_sv_undef)
1055
1056Creates a new OpenCL::Image object and optionally initialises it with
1057the given data values.
1058
1059L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clCreateImage.html>
1060
683=item $img = $ctx->image2d ($flags, $channel_order, $channel_type, $width, $height, $row_pitch = 0, $data = undef) 1061=item $img = $ctx->image2d ($flags, $channel_order, $channel_type, $width, $height, $row_pitch = 0, $data = undef)
684 1062
685Creates a new OpenCL::Image2D object and optionally initialises it with 1063Creates a new OpenCL::Image2D object and optionally initialises it with
686the given data values. 1064the given data values.
687 1065
699Creates a new OpenCL::Buffer (actually OpenCL::BufferObj) object that refers to the given 1077Creates a new OpenCL::Buffer (actually OpenCL::BufferObj) object that refers to the given
700OpenGL buffer object. 1078OpenGL buffer object.
701 1079
702http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLBuffer.html 1080http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLBuffer.html
703 1081
1082=item $img = $ctx->gl_texture ($flags, $target, $miplevel, $texture)
1083
1084Creates a new OpenCL::Image object that refers to the given OpenGL
1085texture object or buffer.
1086
1087http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clCreateFromGLTexture.html
1088
704=item $ctx->gl_texture2d ($flags, $target, $miplevel, $texture) 1089=item $img = $ctx->gl_texture2d ($flags, $target, $miplevel, $texture)
705 1090
706Creates a new OpenCL::Image2D object that refers to the given OpenGL 1091Creates a new OpenCL::Image2D object that refers to the given OpenGL
7072D texture object. 10922D texture object.
708 1093
709http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLTexture2D.html 1094http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLTexture2D.html
710 1095
711=item $ctx->gl_texture3d ($flags, $target, $miplevel, $texture) 1096=item $img = $ctx->gl_texture3d ($flags, $target, $miplevel, $texture)
712 1097
713Creates a new OpenCL::Image3D object that refers to the given OpenGL 1098Creates a new OpenCL::Image3D object that refers to the given OpenGL
7143D texture object. 10993D texture object.
715 1100
716http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLTexture3D.html 1101http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLTexture3D.html
770=back 1155=back
771 1156
772=head2 THE OpenCL::Queue CLASS 1157=head2 THE OpenCL::Queue CLASS
773 1158
774An OpenCL::Queue represents an execution queue for OpenCL. You execute 1159An OpenCL::Queue represents an execution queue for OpenCL. You execute
775requests by calling their respective C<enqueue_xxx> method and waitinf for 1160requests by calling their respective method and waiting for it to complete
776it to complete in some way. 1161in some way.
777 1162
778All the enqueue methods return an event object that can be used to wait 1163Most methods that enqueue some request return an event object that can
779for completion, unless the method is called in void context, in which case 1164be used to wait for completion (optionally using a callback), unless
780no event object is created. 1165the method is called in void context, in which case no event object is
1166created.
781 1167
782They also allow you to specify any number of other event objects that this 1168They also allow you to specify any number of other event objects that this
783request has to wait for before it starts executing, by simply passing the 1169request has to wait for before it starts executing, by simply passing the
784event objects as extra parameters to the enqueue methods. 1170event objects as extra parameters to the enqueue methods. To simplify
1171program design, this module ignores any C<undef> values in the list of
1172events. This makes it possible to code operations such as this, without
1173having to put a valid event object into C<$event> first:
1174
1175 $event = $queue->xxx (..., $event);
785 1176
786Queues execute in-order by default, without any parallelism, so in most 1177Queues execute in-order by default, without any parallelism, so in most
787cases (i.e. you use only one queue) it's not necessary to wait for or 1178cases (i.e. you use only one queue) it's not necessary to wait for or
788create event objects. 1179create event objects, althoguh an our of order queue is often a bit
1180faster.
789 1181
790=over 4 1182=over 4
791 1183
792=item $ev = $queue->enqueue_read_buffer ($buffer, $blocking, $offset, $len, $data, $wait_events...) 1184=item $ev = $queue->read_buffer ($buffer, $blocking, $offset, $len, $data, $wait_events...)
793 1185
794Reads data from buffer into the given string. 1186Reads data from buffer into the given string.
795 1187
796L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReadBuffer.html> 1188L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReadBuffer.html>
797 1189
798=item $ev = $queue->enqueue_write_buffer ($buffer, $blocking, $offset, $data, $wait_events...) 1190=item $ev = $queue->write_buffer ($buffer, $blocking, $offset, $data, $wait_events...)
799 1191
800Writes data to buffer from the given string. 1192Writes data to buffer from the given string.
801 1193
802L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteBuffer.html> 1194L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteBuffer.html>
803 1195
804=item $ev = $queue->enqueue_copy_buffer ($src, $dst, $src_offset, $dst_offset, $len, $wait_events...) 1196=item $ev = $queue->copy_buffer ($src, $dst, $src_offset, $dst_offset, $len, $wait_events...)
805 1197
806L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBuffer.html> 1198L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBuffer.html>
807 1199
808=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...) 1200=item $ev = $queue->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...)
809 1201
810http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReadBufferRect.html 1202http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReadBufferRect.html
811 1203
812=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...) 1204=item $ev = $queue->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...)
813 1205
814http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteBufferRect.html 1206http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteBufferRect.html
815 1207
816=item $ev = $queue->enqueue_read_image ($src, $blocking, $x, $y, $z, $width, $height, $depth, $row_pitch, $slice_pitch, $data, $wait_events...) 1208=item $ev = $queue->read_image ($src, $blocking, $x, $y, $z, $width, $height, $depth, $row_pitch, $slice_pitch, $data, $wait_events...)
817 1209
818L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBufferRect.html> 1210L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBufferRect.html>
819 1211
820=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...) 1212=item $ev = $queue->copy_buffer_to_image ($src_buffer, $dst_image, $src_offset, $dst_x, $dst_y, $dst_z, $width, $height, $depth, $wait_events...)
821 1213
822L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReadImage.html> 1214L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReadImage.html>
823 1215
824=item $ev = $queue->enqueue_write_image ($src, $blocking, $x, $y, $z, $width, $height, $depth, $row_pitch, $slice_pitch, $data, $wait_events...) 1216=item $ev = $queue->write_image ($src, $blocking, $x, $y, $z, $width, $height, $depth, $row_pitch, $slice_pitch, $data, $wait_events...)
825 1217
826L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteImage.html> 1218L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteImage.html>
827 1219
828=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...) 1220=item $ev = $queue->copy_image ($src_image, $dst_image, $src_x, $src_y, $src_z, $dst_x, $dst_y, $dst_z, $width, $height, $depth, $wait_events...)
829 1221
830L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyImage.html> 1222L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyImage.html>
831 1223
832=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...) 1224=item $ev = $queue->copy_image_to_buffer ($src_image, $dst_image, $src_x, $src_y, $src_z, $width, $height, $depth, $dst_offset, $wait_events...)
833 1225
834L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyImageToBuffer.html> 1226L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyImageToBuffer.html>
835 1227
836=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...) 1228=item $ev = $queue->copy_buffer_rect ($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...)
837 1229
838Yeah. 1230Yeah.
839 1231
840L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBufferToImage.html>. 1232L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBufferToImage.html>.
841 1233
1234=item $ev = $queue->fill_buffer ($mem, $pattern, $offset, $size, ...)
1235
1236Fills the given buffer object with repeated applications of C<$pattern>,
1237starting at C<$offset> for C<$size> octets.
1238
1239L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clEnqueueFillBuffer.html>
1240
1241=item $ev = $queue->fill_image ($img, $r, $g, $b, $a, $x, $y, $z, $width, $height, $depth, ...)
1242
1243Fills the given image area with the given rgba colour components. The
1244components are normally floating point values between C<0> and C<1>,
1245except when the image channel data type is a signe dor unsigned
1246unnormalised format, in which case the range is determined by the format.
1247
1248L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clEnqueueFillImage.html>
1249
842=item $ev = $queue->enqueue_task ($kernel, $wait_events...) 1250=item $ev = $queue->task ($kernel, $wait_events...)
843 1251
844L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueTask.html> 1252L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueTask.html>
845 1253
846=item $ev = $queue->enqueue_nd_range_kernel ($kernel, @$global_work_offset, @$global_work_size, @$local_work_size, $wait_events...) 1254=item $ev = $queue->nd_range_kernel ($kernel, \@global_work_offset, \@global_work_size, \@local_work_size, $wait_events...)
847 1255
848Enqueues a kernel execution. 1256Enqueues a kernel execution.
849 1257
850@$global_work_size must be specified as a reference to an array of 1258\@global_work_size must be specified as a reference to an array of
851integers specifying the work sizes (element counts). 1259integers specifying the work sizes (element counts).
852 1260
853@$global_work_offset must be either C<undef> (in which case all offsets 1261\@global_work_offset must be either C<undef> (in which case all offsets
854are C<0>), or a reference to an array of work offsets, with the same number 1262are C<0>), or a reference to an array of work offsets, with the same number
855of elements as @$global_work_size. 1263of elements as \@global_work_size.
856 1264
857@$local_work_size must be either C<undef> (in which case the 1265\@local_work_size must be either C<undef> (in which case the
858implementation is supposed to choose good local work sizes), or a 1266implementation is supposed to choose good local work sizes), or a
859reference to an array of local work sizes, with the same number of 1267reference to an array of local work sizes, with the same number of
860elements as @$global_work_size. 1268elements as \@global_work_size.
861 1269
862L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueNDRangeKernel.html> 1270L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueNDRangeKernel.html>
863 1271
864=item $ev = $queue->enqueue_marker 1272=item $ev = $queue->acquire_gl_objects ([object, ...], $wait_events...)
865 1273
1274Enqueues a list (an array-ref of OpenCL::Memory objects) to be acquired
1275for subsequent OpenCL usage.
1276
866L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueMarker.html> 1277L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueAcquireGLObjects.html>
867 1278
1279=item $ev = $queue->release_gl_objects ([object, ...], $wait_events...)
1280
1281Enqueues a list (an array-ref of OpenCL::Memory objects) to be released
1282for subsequent OpenGL usage.
1283
1284L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReleaseGLObjects.html>
1285
868=item $ev = $queue->enqueue_wait_for_events ($wait_events...) 1286=item $ev = $queue->wait_for_events ($wait_events...)
869 1287
870L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWaitForEvents.html> 1288L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWaitForEvents.html>
871 1289
872=item $queue->enqueue_barrier 1290=item $ev = $queue->marker ($wait_events...)
873 1291
1292L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clEnqueueMarkerWithWaitList.html>
1293
1294=item $ev = $queue->barrier ($wait_events...)
1295
874L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueBarrier.html> 1296L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clEnqueueBarrierWithWaitList.html>
875 1297
876=item $queue->flush 1298=item $queue->flush
877 1299
878L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clFlush.html> 1300L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clFlush.html>
879 1301
994 1416
995=back 1417=back
996 1418
997=head2 THE OpenCL::Image CLASS 1419=head2 THE OpenCL::Image CLASS
998 1420
999This is the superclass of all image objects - OpenCL::Image2D and OpenCL::Image3D. 1421This is the superclass of all image objects - OpenCL::Image1D,
1422OpenCL::Image1DArray, OpenCL::Image1DBuffer, OpenCL::Image2D,
1423OpenCL::Image2DArray and OpenCL::Image3D.
1000 1424
1001=over 4 1425=over 4
1002 1426
1003=item $packed_value = $ev->image_info ($name) 1427=item $packed_value = $image->image_info ($name)
1004 1428
1005See C<< $platform->info >> for details. 1429See C<< $platform->info >> for details.
1006 1430
1007The reason this method is not called C<info> is that there already is an 1431The reason this method is not called C<info> is that there already is an
1008C<< ->info >> method inherited from C<OpenCL::Memory>. 1432C<< ->info >> method inherited from C<OpenCL::Memory>.
1009 1433
1010L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetImageInfo.html> 1434L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetImageInfo.html>
1011 1435
1436=item ($channel_order, $channel_data_type) = $image->format
1437
1438Returns the channel order and type used to create the image by calling
1439C<clGetImageInfo> with C<CL_IMAGE_FORMAT>.
1440
1012=for gengetinfo begin image 1441=for gengetinfo begin image
1013 1442
1014=item $int = $image->element_size 1443=item $int = $image->element_size
1015 1444
1016Calls C<clGetImageInfo> with C<CL_IMAGE_ELEMENT_SIZE> and returns the result. 1445Calls C<clGetImageInfo> with C<CL_IMAGE_ELEMENT_SIZE> and returns the result.
1039 1468
1040=for gengetinfo begin gl_texture 1469=for gengetinfo begin gl_texture
1041 1470
1042=item $GLenum = $gl_texture->target 1471=item $GLenum = $gl_texture->target
1043 1472
1044Calls C<clGetGlTextureInfo> with C<CL_GL_TEXTURE_TARGET> and returns the result. 1473Calls C<clGetGLTextureInfo> with C<CL_GL_TEXTURE_TARGET> and returns the result.
1045 1474
1046=item $GLint = $gl_texture->gl_mipmap_level 1475=item $GLint = $gl_texture->gl_mipmap_level
1047 1476
1048Calls C<clGetGlTextureInfo> with C<CL_GL_MIPMAP_LEVEL> and returns the result. 1477Calls C<clGetGLTextureInfo> with C<CL_GL_MIPMAP_LEVEL> and returns the result.
1049 1478
1050=for gengetinfo end gl_texture 1479=for gengetinfo end gl_texture
1051 1480
1052=back 1481=back
1053 1482
1089 1518
1090=head2 THE OpenCL::Program CLASS 1519=head2 THE OpenCL::Program CLASS
1091 1520
1092=over 4 1521=over 4
1093 1522
1094=item $program->build ($device, $options = "") 1523=item $program->build (\@devices = undef, $options = "", $cb->($program) = undef)
1095 1524
1096Tries to build the program with the givne options. 1525Tries to build the program with the given options. See also the
1526C<$ctx->build> convenience function.
1527
1528If a callback is specified, then it will be called when compilation is
1529finished. Note that many OpenCL implementations block your program while
1530compiling whether you use a callback or not. See C<build_async> if you
1531want to make sure the build is done in the background.
1532
1533Note that some OpenCL implementations atc up badly, and don't call the
1534callback in some error cases (but call it in others). This implementation
1535assumes the callback will always be called, and leaks memory if this is
1536not so. So best make sure you don't pass in invalid values.
1097 1537
1098L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clBuildProgram.html> 1538L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clBuildProgram.html>
1539
1540=item $program->build_async (\@devices = undef, $options = "", $cb->($program) = undef)
1541
1542Similar to C<< ->build >>, except it starts a thread, and never fails (you
1543need to check the compilation status form the callback, or by polling).
1099 1544
1100=item $packed_value = $program->build_info ($device, $name) 1545=item $packed_value = $program->build_info ($device, $name)
1101 1546
1102Similar to C<< $platform->info >>, but returns build info for a previous 1547Similar to C<< $platform->info >>, but returns build info for a previous
1103build attempt for the given device. 1548build attempt for the given device.
1108 1553
1109Creates an OpenCL::Kernel object out of the named C<__kernel> function in 1554Creates an OpenCL::Kernel object out of the named C<__kernel> function in
1110the program. 1555the program.
1111 1556
1112L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateKernel.html> 1557L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateKernel.html>
1558
1559=item @kernels = $program->kernels_in_program
1560
1561Returns all kernels successfully compiled for all devices in program.
1562
1563http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateKernelsInProgram.html
1113 1564
1114=for gengetinfo begin program_build 1565=for gengetinfo begin program_build
1115 1566
1116=item $build_status = $program->build_status ($device) 1567=item $build_status = $program->build_status ($device)
1117 1568
1239 1690
1240Calls C<clGetKernelWorkGroupInfo> with C<CL_KERNEL_PRIVATE_MEM_SIZE> and returns the result. 1691Calls C<clGetKernelWorkGroupInfo> with C<CL_KERNEL_PRIVATE_MEM_SIZE> and returns the result.
1241 1692
1242=for gengetinfo end kernel_work_group 1693=for gengetinfo end kernel_work_group
1243 1694
1695=item $kernel->setf ($format, ...)
1696
1697Sets the arguments of a kernel. Since OpenCL 1.1 doesn't have a generic
1698way to set arguments (and with OpenCL 1.2 it might be rather slow), you
1699need to specify a format argument, much as with C<printf>, to tell OpenCL
1700what type of argument it is.
1701
1702The format arguments are single letters:
1703
1704 c char
1705 C unsigned char
1706 s short
1707 S unsigned short
1708 i int
1709 I unsigned int
1710 l long
1711 L unsigned long
1712
1713 h half float (0..65535)
1714 f float
1715 d double
1716
1717 z local (octet size)
1718
1719 m memory object (buffer or image)
1720 a sampler
1721 e event
1722
1723Space characters in the format string are ignored.
1724
1725Example: set the arguments for a kernel that expects an int, two floats, a buffer and an image.
1726
1727 $kernel->setf ("i ff mm", 5, 0.5, 3, $buffer, $image);
1728
1244=item $kernel->set_TYPE ($index, $value) 1729=item $kernel->set_TYPE ($index, $value)
1245 1730
1731=item $kernel->set_char ($index, $value)
1732
1733=item $kernel->set_uchar ($index, $value)
1734
1735=item $kernel->set_short ($index, $value)
1736
1737=item $kernel->set_ushort ($index, $value)
1738
1739=item $kernel->set_int ($index, $value)
1740
1741=item $kernel->set_uint ($index, $value)
1742
1743=item $kernel->set_long ($index, $value)
1744
1745=item $kernel->set_ulong ($index, $value)
1746
1747=item $kernel->set_half ($index, $value)
1748
1749=item $kernel->set_float ($index, $value)
1750
1751=item $kernel->set_double ($index, $value)
1752
1753=item $kernel->set_memory ($index, $value)
1754
1755=item $kernel->set_buffer ($index, $value)
1756
1757=item $kernel->set_image ($index, $value)
1758
1759=item $kernel->set_sampler ($index, $value)
1760
1761=item $kernel->set_local ($index, $value)
1762
1763=item $kernel->set_event ($index, $value)
1764
1246This is a family of methods to set the kernel argument with the number C<$index> to the give C<$value>. 1765This is a family of methods to set the kernel argument with the number
1247 1766C<$index> to the give C<$value>.
1248TYPE is one of C<char>, C<uchar>, C<short>, C<ushort>, C<int>, C<uint>,
1249C<long>, C<ulong>, C<half>, C<float>, C<double>, C<memory>, C<buffer>,
1250C<image2d>, C<image3d>, C<sampler> or C<event>.
1251 1767
1252Chars and integers (including the half type) are specified as integers, 1768Chars and integers (including the half type) are specified as integers,
1253float and double as floating point values, memory/buffer/image2d/image3d 1769float and double as floating point values, memory/buffer/image must be
1254must be an object of that type or C<undef>, and sampler and event must be 1770an object of that type or C<undef>, local-memory arguments are set by
1255objects of that type. 1771specifying the size, and sampler and event must be objects of that type.
1772
1773Note that C<set_memory> works for all memory objects (all types of buffers
1774and images) - the main purpose of the more specific C<set_TYPE> functions
1775is type checking.
1776
1777Setting an argument for a kernel does NOT keep a reference to the object -
1778for example, if you set an argument to some image object, free the image,
1779and call the kernel, you will run into undefined behaviour.
1256 1780
1257L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetKernelArg.html> 1781L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetKernelArg.html>
1258 1782
1259=back 1783=back
1260 1784
1269 1793
1270Waits for the event to complete. 1794Waits for the event to complete.
1271 1795
1272L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clWaitForEvents.html> 1796L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clWaitForEvents.html>
1273 1797
1798=item $ev->cb ($exec_callback_type, $callback->($event, $event_command_exec_status))
1799
1800Adds a callback to the callback stack for the given event type. There is
1801no way to remove a callback again.
1802
1803L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetEventCallback.html>
1804
1274=item $packed_value = $ev->info ($name) 1805=item $packed_value = $ev->info ($name)
1275 1806
1276See C<< $platform->info >> for details. 1807See C<< $platform->info >> for details.
1277 1808
1278L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetEventInfo.html> 1809L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetEventInfo.html>
1338 1869
1339=over 4 1870=over 4
1340 1871
1341=item $ev->set_status ($execution_status) 1872=item $ev->set_status ($execution_status)
1342 1873
1874Sets the execution status of the user event. Can only be called once,
1875either with OpenCL::COMPLETE or a negative number as status.
1876
1343L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetUserEventStatus.html> 1877L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetUserEventStatus.html>
1344 1878
1345=back 1879=back
1346 1880
1347=cut 1881=cut
1348
1349package OpenCL;
1350
1351use common::sense;
1352
1353BEGIN {
1354 our $VERSION = '0.92';
1355
1356 require XSLoader;
1357 XSLoader::load (__PACKAGE__, $VERSION);
1358
1359 @OpenCL::Buffer::ISA =
1360 @OpenCL::Image::ISA = OpenCL::Memory::;
1361
1362 @OpenCL::BufferObj::ISA = OpenCL::Buffer::;
1363
1364 @OpenCL::Image2D::ISA =
1365 @OpenCL::Image3D::ISA = OpenCL::Image::;
1366
1367 @OpenCL::UserEvent::ISA = OpenCL::Event::;
1368}
1369 1882
13701; 18831;
1371 1884
1372=head1 AUTHOR 1885=head1 AUTHOR
1373 1886

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines