… | |
… | |
157 | $id = get_global_id (0); |
157 | $id = get_global_id (0); |
158 | output [id] = input [id] * input [id]; |
158 | output [id] = input [id] * input [id]; |
159 | } |
159 | } |
160 | '; |
160 | '; |
161 | |
161 | |
162 | my $prog = $ctx->program_with_source ($src); |
162 | 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"); |
163 | my $kernel = $prog->kernel ("squareit"); |
169 | |
164 | |
170 | =head2 Create some input and output float buffers, then call the |
165 | =head2 Create some input and output float buffers, then call the |
171 | 'squareit' kernel on them. |
166 | 'squareit' kernel on them. |
172 | |
167 | |
… | |
… | |
260 | m.x = fabs (fmod (m.x + time * 0.05f, 4.f)) - 2.f; |
255 | m.x = fabs (fmod (m.x + time * 0.05f, 4.f)) - 2.f; |
261 | |
256 | |
262 | float2 z = m; |
257 | float2 z = m; |
263 | float2 c = (float2)(sin (time * 0.05005), cos (time * 0.06001)); |
258 | float2 c = (float2)(sin (time * 0.05005), cos (time * 0.06001)); |
264 | |
259 | |
265 | for (int i = 0; i < 100 && dot (z, z) < 4.f; ++i) |
260 | for (int i = 0; i < 25 && dot (z, z) < 4.f; ++i) |
266 | z = (float2)(z.x * z.x - z.y * z.y, 2.f * z.x * z.y) + c; |
261 | z = (float2)(z.x * z.x - z.y * z.y, 2.f * z.x * z.y) + c; |
267 | |
262 | |
268 | float3 colour = (float3)(z.x, z.y, z.x * z.y); |
263 | float3 colour = (float3)(z.x, z.y, z.x * z.y); |
269 | write_imagef (img, (int2)(get_global_id (0), get_global_id (1)), (float4)(colour * p.x * p.x, 1.)); |
264 | write_imagef (img, (int2)(get_global_id (0), get_global_id (1)), (float4)(colour * p.x * p.x, 1.)); |
270 | } |
265 | } |
271 | EOF |
266 | EOF |
|
|
267 | |
272 | my $prog = $ctx->program_with_source ($src); |
268 | my $prog = $ctx->build_program ($src); |
273 | eval { $prog->build ($dev); 1 } |
|
|
274 | or die $prog->build_log ($dev); |
|
|
275 | |
|
|
276 | my $kernel = $prog->kernel ("juliatunnel"); |
269 | my $kernel = $prog->kernel ("juliatunnel"); |
277 | |
270 | |
278 | # program compiled, kernel ready, now draw and loop |
271 | # program compiled, kernel ready, now draw and loop |
279 | |
272 | |
280 | for (my $time; ; ++$time) { |
273 | for (my $time; ; ++$time) { |
… | |
… | |
288 | |
281 | |
289 | # release objects to opengl again |
282 | # release objects to opengl again |
290 | $queue->enqueue_release_gl_objects ([$tex]); |
283 | $queue->enqueue_release_gl_objects ([$tex]); |
291 | |
284 | |
292 | # wait |
285 | # wait |
293 | $queue->flush; |
286 | $queue->finish; |
294 | |
287 | |
295 | # now draw the texture, the defaults should be all right |
288 | # now draw the texture, the defaults should be all right |
296 | glTexParameterf GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST; |
289 | glTexParameterf GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST; |
297 | |
290 | |
298 | glEnable GL_TEXTURE_2D; |
291 | glEnable GL_TEXTURE_2D; |
… | |
… | |
336 | =item * Structures are often specified by flattening out their components |
329 | =item * Structures are often specified by flattening out their components |
337 | as with short vectors, and returned as arrayrefs. |
330 | as with short vectors, and returned as arrayrefs. |
338 | |
331 | |
339 | =item * When enqueuing commands, the wait list is specified by adding |
332 | =item * When enqueuing commands, the wait list is specified by adding |
340 | extra arguments to the function - anywhere a C<$wait_events...> argument |
333 | extra arguments to the function - anywhere a C<$wait_events...> argument |
341 | is documented this can be any number of event objects. |
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. |
342 | |
337 | |
343 | =item * When enqueuing commands, if the enqueue method is called in void |
338 | =item * When enqueuing commands, if the enqueue method is called in void |
344 | context, no event is created. In all other contexts an event is returned |
339 | context, no event is created. In all other contexts an event is returned |
345 | by the method. |
340 | by the method. |
346 | |
341 | |
… | |
… | |
381 | |
376 | |
382 | For this to work, the OpenGL library must be loaded, a GLX context must |
377 | For this to work, the OpenGL library must be loaded, a GLX context must |
383 | have been created and be made current, and C<dlsym> must be available and |
378 | have been created and be made current, and C<dlsym> must be available and |
384 | capable of finding the function via C<RTLD_DEFAULT>. |
379 | capable of finding the function via C<RTLD_DEFAULT>. |
385 | |
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 | |
386 | =head2 THE OpenCL PACKAGE |
530 | =head2 THE OpenCL PACKAGE |
387 | |
531 | |
388 | =over 4 |
532 | =over 4 |
389 | |
533 | |
390 | =item $int = OpenCL::errno |
534 | =item $int = OpenCL::errno |
… | |
… | |
408 | |
552 | |
409 | Returns all available OpenCL::Platform objects. |
553 | Returns all available OpenCL::Platform objects. |
410 | |
554 | |
411 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetPlatformIDs.html> |
555 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetPlatformIDs.html> |
412 | |
556 | |
413 | =item $ctx = OpenCL::context_from_type $properties, $type = OpenCL::DEVICE_TYPE_DEFAULT, $notify = undef |
557 | =item $ctx = OpenCL::context_from_type $properties, $type = OpenCL::DEVICE_TYPE_DEFAULT, $notify = $print_stderr |
414 | |
558 | |
415 | Tries to create a context from a default device and platform - never worked for me. |
559 | Tries to create a context from a default device and platform - never worked for me. |
416 | |
560 | |
417 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html> |
561 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html> |
418 | |
562 | |
… | |
… | |
420 | |
564 | |
421 | Waits for all events to complete. |
565 | Waits for all events to complete. |
422 | |
566 | |
423 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clWaitForEvents.html> |
567 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clWaitForEvents.html> |
424 | |
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 | |
425 | =back |
606 | =back |
426 | |
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 | |
427 | =head2 THE OpenCL::Platform CLASS |
633 | =head2 THE OpenCL::Platform CLASS |
428 | |
634 | |
429 | =over 4 |
635 | =over 4 |
430 | |
636 | |
431 | =item @devices = $platform->devices ($type = OpenCL::DEVICE_TYPE_ALL) |
637 | =item @devices = $platform->devices ($type = OpenCL::DEVICE_TYPE_ALL) |
432 | |
638 | |
433 | Returns a list of matching OpenCL::Device objects. |
639 | Returns a list of matching OpenCL::Device objects. |
434 | |
640 | |
435 | =item $ctx = $platform->context_from_type ($properties, $type = OpenCL::DEVICE_TYPE_DEFAULT, $notify = undef) |
641 | =item $ctx = $platform->context_from_type ($properties, $type = OpenCL::DEVICE_TYPE_DEFAULT, $notify = $print_stderr) |
436 | |
642 | |
437 | Tries to create a context. Never worked for me, and you need devices explicitly anyway. |
643 | Tries to create a context. Never worked for me, and you need devices explicitly anyway. |
438 | |
644 | |
439 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html> |
645 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html> |
440 | |
646 | |
441 | =item $ctx = $platform->context ($properties = undef, @$devices, $notify = undef) |
647 | =item $ctx = $platform->context ($properties, @$devices, $notify = $print_stderr) |
442 | |
648 | |
443 | Create a new OpenCL::Context object using the given device object(s)- a |
649 | Create a new OpenCL::Context object using the given device object(s)- a |
444 | CL_CONTEXT_PLATFORM property is supplied automatically. |
650 | CL_CONTEXT_PLATFORM property is supplied automatically. |
445 | |
651 | |
446 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContext.html> |
652 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContext.html> |
… | |
… | |
454 | It's best to avoid this method and use one of the following convenience |
660 | It's best to avoid this method and use one of the following convenience |
455 | wrappers. |
661 | wrappers. |
456 | |
662 | |
457 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetPlatformInfo.html> |
663 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetPlatformInfo.html> |
458 | |
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 | |
459 | =for gengetinfo begin platform |
672 | =for gengetinfo begin platform |
460 | |
673 | |
461 | =item $string = $platform->profile |
674 | =item $string = $platform->profile |
462 | |
675 | |
463 | Calls C<clGetPlatformInfo> with C<CL_PLATFORM_PROFILE> and returns the result. |
676 | Calls C<clGetPlatformInfo> with C<CL_PLATFORM_PROFILE> and returns the result. |
… | |
… | |
748 | |
961 | |
749 | =item @device_partition_property_exts = $device->affinity_domains_ext |
962 | =item @device_partition_property_exts = $device->affinity_domains_ext |
750 | |
963 | |
751 | Calls C<clGetDeviceInfo> with C<CL_DEVICE_AFFINITY_DOMAINS_EXT> and returns the result. |
964 | Calls C<clGetDeviceInfo> with C<CL_DEVICE_AFFINITY_DOMAINS_EXT> and returns the result. |
752 | |
965 | |
753 | =item $uint = $device->reference_count_ext |
966 | =item $uint = $device->reference_count_ext |
754 | |
967 | |
755 | Calls C<clGetDeviceInfo> with C<CL_DEVICE_REFERENCE_COUNT_EXT > and returns the result. |
968 | Calls C<clGetDeviceInfo> with C<CL_DEVICE_REFERENCE_COUNT_EXT> and returns the result. |
756 | |
969 | |
757 | =item @device_partition_property_exts = $device->partition_style_ext |
970 | =item @device_partition_property_exts = $device->partition_style_ext |
758 | |
971 | |
759 | Calls C<clGetDeviceInfo> with C<CL_DEVICE_PARTITION_STYLE_EXT> and returns the result. |
972 | Calls C<clGetDeviceInfo> with C<CL_DEVICE_PARTITION_STYLE_EXT> and returns the result. |
760 | |
973 | |
… | |
… | |
764 | |
977 | |
765 | =head2 THE OpenCL::Context CLASS |
978 | =head2 THE OpenCL::Context CLASS |
766 | |
979 | |
767 | =over 4 |
980 | =over 4 |
768 | |
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 | |
769 | =item $queue = $ctx->queue ($device, $properties) |
1012 | =item $queue = $ctx->queue ($device, $properties) |
770 | |
1013 | |
771 | Create a new OpenCL::Queue object from the context and the given device. |
1014 | Create a new OpenCL::Queue object from the context and the given device. |
772 | |
1015 | |
773 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateCommandQueue.html> |
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); |
774 | |
1021 | |
775 | =item $ev = $ctx->user_event |
1022 | =item $ev = $ctx->user_event |
776 | |
1023 | |
777 | Creates a new OpenCL::UserEvent object. |
1024 | Creates a new OpenCL::UserEvent object. |
778 | |
1025 | |
… | |
… | |
788 | =item $buf = $ctx->buffer_sv ($flags, $data) |
1035 | =item $buf = $ctx->buffer_sv ($flags, $data) |
789 | |
1036 | |
790 | Creates a new OpenCL::Buffer (actually OpenCL::BufferObj) object and |
1037 | Creates a new OpenCL::Buffer (actually OpenCL::BufferObj) object and |
791 | initialise it with the given data values. |
1038 | initialise it with the given data values. |
792 | |
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 | |
793 | =item $img = $ctx->image2d ($flags, $channel_order, $channel_type, $width, $height, $row_pitch = 0, $data = undef) |
1047 | =item $img = $ctx->image2d ($flags, $channel_order, $channel_type, $width, $height, $row_pitch = 0, $data = undef) |
794 | |
1048 | |
795 | Creates a new OpenCL::Image2D object and optionally initialises it with |
1049 | Creates a new OpenCL::Image2D object and optionally initialises it with |
796 | the given data values. |
1050 | the given data values. |
797 | |
1051 | |
… | |
… | |
809 | Creates a new OpenCL::Buffer (actually OpenCL::BufferObj) object that refers to the given |
1063 | Creates a new OpenCL::Buffer (actually OpenCL::BufferObj) object that refers to the given |
810 | OpenGL buffer object. |
1064 | OpenGL buffer object. |
811 | |
1065 | |
812 | http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLBuffer.html |
1066 | http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLBuffer.html |
813 | |
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 | |
814 | =item $ctx->gl_texture2d ($flags, $target, $miplevel, $texture) |
1075 | =item $img = $ctx->gl_texture2d ($flags, $target, $miplevel, $texture) |
815 | |
1076 | |
816 | Creates a new OpenCL::Image2D object that refers to the given OpenGL |
1077 | Creates a new OpenCL::Image2D object that refers to the given OpenGL |
817 | 2D texture object. |
1078 | 2D texture object. |
818 | |
1079 | |
819 | http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLTexture2D.html |
1080 | http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLTexture2D.html |
820 | |
1081 | |
821 | =item $ctx->gl_texture3d ($flags, $target, $miplevel, $texture) |
1082 | =item $img = $ctx->gl_texture3d ($flags, $target, $miplevel, $texture) |
822 | |
1083 | |
823 | Creates a new OpenCL::Image3D object that refers to the given OpenGL |
1084 | Creates a new OpenCL::Image3D object that refers to the given OpenGL |
824 | 3D texture object. |
1085 | 3D texture object. |
825 | |
1086 | |
826 | http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLTexture3D.html |
1087 | http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLTexture3D.html |
… | |
… | |
889 | for completion, unless the method is called in void context, in which case |
1150 | for completion, unless the method is called in void context, in which case |
890 | no event object is created. |
1151 | no event object is created. |
891 | |
1152 | |
892 | They also allow you to specify any number of other event objects that this |
1153 | They also allow you to specify any number of other event objects that this |
893 | request has to wait for before it starts executing, by simply passing the |
1154 | request has to wait for before it starts executing, by simply passing the |
894 | event objects as extra parameters to the enqueue methods. |
1155 | event objects as extra parameters to the enqueue methods. To simplify |
|
|
1156 | program design, this module ignores any C<undef> values in the list of |
|
|
1157 | events. This makes it possible to code operations such as this, without |
|
|
1158 | having to put a valid event object into C<$event> first: |
|
|
1159 | |
|
|
1160 | $event = $queue->enqueue_xxx (..., $event); |
895 | |
1161 | |
896 | Queues execute in-order by default, without any parallelism, so in most |
1162 | Queues execute in-order by default, without any parallelism, so in most |
897 | cases (i.e. you use only one queue) it's not necessary to wait for or |
1163 | cases (i.e. you use only one queue) it's not necessary to wait for or |
898 | create event objects. |
1164 | create event objects, althoguh an our of order queue is often a bit |
|
|
1165 | faster. |
899 | |
1166 | |
900 | =over 4 |
1167 | =over 4 |
901 | |
1168 | |
902 | =item $ev = $queue->enqueue_read_buffer ($buffer, $blocking, $offset, $len, $data, $wait_events...) |
1169 | =item $ev = $queue->enqueue_read_buffer ($buffer, $blocking, $offset, $len, $data, $wait_events...) |
903 | |
1170 | |
… | |
… | |
946 | =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...) |
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...) |
947 | |
1214 | |
948 | Yeah. |
1215 | Yeah. |
949 | |
1216 | |
950 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBufferToImage.html>. |
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> |
951 | |
1234 | |
952 | =item $ev = $queue->enqueue_task ($kernel, $wait_events...) |
1235 | =item $ev = $queue->enqueue_task ($kernel, $wait_events...) |
953 | |
1236 | |
954 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueTask.html> |
1237 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueTask.html> |
955 | |
1238 | |
… | |
… | |
969 | reference to an array of local work sizes, with the same number of |
1252 | reference to an array of local work sizes, with the same number of |
970 | elements as @$global_work_size. |
1253 | elements as @$global_work_size. |
971 | |
1254 | |
972 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueNDRangeKernel.html> |
1255 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueNDRangeKernel.html> |
973 | |
1256 | |
974 | =item $ev = $queue->enqueue_marker ($wait_events...) |
|
|
975 | |
|
|
976 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueMarker.html> |
|
|
977 | |
|
|
978 | =item $ev = $queue->enqueue_acquire_gl_objects ([object, ...], $wait_events...) |
1257 | =item $ev = $queue->enqueue_acquire_gl_objects ([object, ...], $wait_events...) |
979 | |
1258 | |
980 | Enqueues a list (an array-ref of OpenCL::Memory objects) to be acquired |
1259 | Enqueues a list (an array-ref of OpenCL::Memory objects) to be acquired |
981 | for subsequent OpenCL usage. |
1260 | for subsequent OpenCL usage. |
982 | |
1261 | |
… | |
… | |
991 | |
1270 | |
992 | =item $ev = $queue->enqueue_wait_for_events ($wait_events...) |
1271 | =item $ev = $queue->enqueue_wait_for_events ($wait_events...) |
993 | |
1272 | |
994 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWaitForEvents.html> |
1273 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWaitForEvents.html> |
995 | |
1274 | |
996 | =item $queue->enqueue_barrier |
1275 | =item $ev = $queue->enqueue_marker ($wait_events...) |
997 | |
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 | |
998 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueBarrier.html> |
1281 | L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clEnqueueBarrierWithWaitList.html> |
999 | |
1282 | |
1000 | =item $queue->flush |
1283 | =item $queue->flush |
1001 | |
1284 | |
1002 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clFlush.html> |
1285 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clFlush.html> |
1003 | |
1286 | |
… | |
… | |
1118 | |
1401 | |
1119 | =back |
1402 | =back |
1120 | |
1403 | |
1121 | =head2 THE OpenCL::Image CLASS |
1404 | =head2 THE OpenCL::Image CLASS |
1122 | |
1405 | |
1123 | This is the superclass of all image objects - OpenCL::Image2D and OpenCL::Image3D. |
1406 | This is the superclass of all image objects - OpenCL::Image1D, |
|
|
1407 | OpenCL::Image1DArray, OpenCL::Image1DBuffer, OpenCL::Image2D, |
|
|
1408 | OpenCL::Image2DArray and OpenCL::Image3D. |
1124 | |
1409 | |
1125 | =over 4 |
1410 | =over 4 |
1126 | |
1411 | |
1127 | =item $packed_value = $ev->image_info ($name) |
1412 | =item $packed_value = $image->image_info ($name) |
1128 | |
1413 | |
1129 | See C<< $platform->info >> for details. |
1414 | See C<< $platform->info >> for details. |
1130 | |
1415 | |
1131 | The reason this method is not called C<info> is that there already is an |
1416 | The reason this method is not called C<info> is that there already is an |
1132 | C<< ->info >> method inherited from C<OpenCL::Memory>. |
1417 | C<< ->info >> method inherited from C<OpenCL::Memory>. |
1133 | |
1418 | |
1134 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetImageInfo.html> |
1419 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetImageInfo.html> |
1135 | |
1420 | |
|
|
1421 | =item ($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 | |
1136 | =for gengetinfo begin image |
1426 | =for gengetinfo begin image |
1137 | |
1427 | |
1138 | =item $int = $image->element_size |
1428 | =item $int = $image->element_size |
1139 | |
1429 | |
1140 | Calls C<clGetImageInfo> with C<CL_IMAGE_ELEMENT_SIZE> and returns the result. |
1430 | Calls C<clGetImageInfo> with C<CL_IMAGE_ELEMENT_SIZE> and returns the result. |
… | |
… | |
1213 | |
1503 | |
1214 | =head2 THE OpenCL::Program CLASS |
1504 | =head2 THE OpenCL::Program CLASS |
1215 | |
1505 | |
1216 | =over 4 |
1506 | =over 4 |
1217 | |
1507 | |
1218 | =item $program->build ($device, $options = "") |
1508 | =item $program->build (\@devices = undef, $options = "", $cb->($program) = undef) |
1219 | |
1509 | |
1220 | Tries to build the program with the givne options. |
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. |
1221 | |
1522 | |
1222 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clBuildProgram.html> |
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). |
1223 | |
1529 | |
1224 | =item $packed_value = $program->build_info ($device, $name) |
1530 | =item $packed_value = $program->build_info ($device, $name) |
1225 | |
1531 | |
1226 | Similar to C<< $platform->info >>, but returns build info for a previous |
1532 | Similar to C<< $platform->info >>, but returns build info for a previous |
1227 | build attempt for the given device. |
1533 | build attempt for the given device. |
… | |
… | |
1232 | |
1538 | |
1233 | Creates an OpenCL::Kernel object out of the named C<__kernel> function in |
1539 | Creates an OpenCL::Kernel object out of the named C<__kernel> function in |
1234 | the program. |
1540 | the program. |
1235 | |
1541 | |
1236 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateKernel.html> |
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 |
1237 | |
1549 | |
1238 | =for gengetinfo begin program_build |
1550 | =for gengetinfo begin program_build |
1239 | |
1551 | |
1240 | =item $build_status = $program->build_status ($device) |
1552 | =item $build_status = $program->build_status ($device) |
1241 | |
1553 | |
… | |
… | |
1369 | |
1681 | |
1370 | This is a family of methods to set the kernel argument with the number C<$index> to the give C<$value>. |
1682 | This is a family of methods to set the kernel argument with the number C<$index> to the give C<$value>. |
1371 | |
1683 | |
1372 | TYPE is one of C<char>, C<uchar>, C<short>, C<ushort>, C<int>, C<uint>, |
1684 | TYPE is one of C<char>, C<uchar>, C<short>, C<ushort>, C<int>, C<uint>, |
1373 | C<long>, C<ulong>, C<half>, C<float>, C<double>, C<memory>, C<buffer>, |
1685 | C<long>, C<ulong>, C<half>, C<float>, C<double>, C<memory>, C<buffer>, |
1374 | C<image2d>, C<image3d>, C<sampler> or C<event>. |
1686 | C<image2d>, C<image3d>, C<sampler>, C<local> or C<event>. |
1375 | |
1687 | |
1376 | Chars and integers (including the half type) are specified as integers, |
1688 | Chars and integers (including the half type) are specified as integers, |
1377 | float and double as floating point values, memory/buffer/image2d/image3d |
1689 | float and double as floating point values, memory/buffer/image2d/image3d |
1378 | must be an object of that type or C<undef>, and sampler and event must be |
1690 | must be an object of that type or C<undef>, local-memory arguments are |
1379 | objects of that type. |
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. |
1380 | |
1697 | |
1381 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetKernelArg.html> |
1698 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetKernelArg.html> |
1382 | |
1699 | |
1383 | =back |
1700 | =back |
1384 | |
1701 | |
… | |
… | |
1393 | |
1710 | |
1394 | Waits for the event to complete. |
1711 | Waits for the event to complete. |
1395 | |
1712 | |
1396 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clWaitForEvents.html> |
1713 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clWaitForEvents.html> |
1397 | |
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 | |
1398 | =item $packed_value = $ev->info ($name) |
1722 | =item $packed_value = $ev->info ($name) |
1399 | |
1723 | |
1400 | See C<< $platform->info >> for details. |
1724 | See C<< $platform->info >> for details. |
1401 | |
1725 | |
1402 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetEventInfo.html> |
1726 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetEventInfo.html> |
… | |
… | |
1462 | |
1786 | |
1463 | =over 4 |
1787 | =over 4 |
1464 | |
1788 | |
1465 | =item $ev->set_status ($execution_status) |
1789 | =item $ev->set_status ($execution_status) |
1466 | |
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 | |
1467 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetUserEventStatus.html> |
1794 | L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetUserEventStatus.html> |
1468 | |
1795 | |
1469 | =back |
1796 | =back |
1470 | |
1797 | |
1471 | =cut |
1798 | =cut |
1472 | |
|
|
1473 | package OpenCL; |
|
|
1474 | |
|
|
1475 | use common::sense; |
|
|
1476 | |
|
|
1477 | BEGIN { |
|
|
1478 | our $VERSION = '0.95'; |
|
|
1479 | |
|
|
1480 | require XSLoader; |
|
|
1481 | XSLoader::load (__PACKAGE__, $VERSION); |
|
|
1482 | |
|
|
1483 | @OpenCL::Buffer::ISA = |
|
|
1484 | @OpenCL::Image::ISA = OpenCL::Memory::; |
|
|
1485 | |
|
|
1486 | @OpenCL::BufferObj::ISA = OpenCL::Buffer::; |
|
|
1487 | |
|
|
1488 | @OpenCL::Image2D::ISA = |
|
|
1489 | @OpenCL::Image3D::ISA = OpenCL::Image::; |
|
|
1490 | |
|
|
1491 | @OpenCL::UserEvent::ISA = OpenCL::Event::; |
|
|
1492 | } |
|
|
1493 | |
1799 | |
1494 | 1; |
1800 | 1; |
1495 | |
1801 | |
1496 | =head1 AUTHOR |
1802 | =head1 AUTHOR |
1497 | |
1803 | |