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

Comparing OpenCL/OpenCL.pm (file contents):
Revision 1.4 by root, Tue Nov 15 21:16:19 2011 UTC vs.
Revision 1.53 by root, Tue Apr 24 14:30:57 2012 UTC

1=head1 NAME 1=head1 NAME
2 2
3OpenCL - bindings to, well, OpenCL 3OpenCL - Open Computing Language Bindings
4 4
5=head1 SYNOPSIS 5=head1 SYNOPSIS
6 6
7 use OpenCL; 7 use OpenCL;
8 8
9=head1 DESCRIPTION 9=head1 DESCRIPTION
10 10
11This is an early release which is not useful yet. 11This is an early release which might be useful, but hasn't seen much testing.
12 12
13=head2 OpenCL FROM 10000 FEET HEIGHT
14
15Here is a high level overview of OpenCL:
16
17First you need to find one or more OpenCL::Platforms (kind of like
18vendors) - usually there is only one.
19
20Each platform gives you access to a number of OpenCL::Device objects, e.g.
21your graphics card.
22
23From a platform and some device(s), you create an OpenCL::Context, which is
24a very central object in OpenCL: Once you have a context you can create
25most other objects:
26
27OpenCL::Program objects, which store source code and, after building for a
28specific device ("compiling and linking"), also binary programs. For each
29kernel function in a program you can then create an OpenCL::Kernel object
30which represents basically a function call with argument values.
31
32OpenCL::Memory objects of various flavours: OpenCL::Buffer objects (flat
33memory areas, think arrays or structs) and OpenCL::Image objects (think 2d
34or 3d array) for bulk data and input and output for kernels.
35
36OpenCL::Sampler objects, which are kind of like texture filter modes in
37OpenGL.
38
39OpenCL::Queue objects - command queues, which allow you to submit memory
40reads, writes and copies, as well as kernel calls to your devices. They
41also offer a variety of methods to synchronise request execution, for
42example with barriers or OpenCL::Event objects.
43
44OpenCL::Event objects are used to signal when something is complete.
45
13=head1 HELPFUL RESOURCES 46=head2 HELPFUL RESOURCES
14 47
15The OpenCL spec used to dveelop this module (1.2 spec was available, but 48The OpenCL spec used to develop this module (1.2 spec was available, but
16no implementation was available to me :). 49no implementation was available to me :).
17 50
18 http://www.khronos.org/registry/cl/specs/opencl-1.1.pdf 51 http://www.khronos.org/registry/cl/specs/opencl-1.1.pdf
19 52
20OpenCL manpages: 53OpenCL manpages:
21 54
22 http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/ 55 http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/
23 56
57If you are into UML class diagrams, the following diagram might help - if
58not, it will be mildly cobfusing:
59
60 http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/classDiagram.html
61
62Here's a tutorial from AMD (very AMD-centric, too), not sure how useful it
63is, but at least it's free of charge:
64
65 http://developer.amd.com/zones/OpenCLZone/courses/Documents/Introduction_to_OpenCL_Programming%20Training_Guide%20%28201005%29.pdf
66
67And here's NVIDIA's OpenCL Best Practises Guide:
68
69 http://developer.download.nvidia.com/compute/cuda/3_2/toolkit/docs/OpenCL_Best_Practices_Guide.pdf
70
71=head1 BASIC WORKFLOW
72
73To get something done, you basically have to do this once (refer to the
74examples below for actual code, this is just a high-level description):
75
76Find some platform (e.g. the first one) and some device(s) (e.g. the first
77device of the platform), and create a context from those.
78
79Create program objects from your OpenCL source code, then build (compile)
80the programs for each device you want to run them on.
81
82Create kernel objects for all kernels you want to use (surprisingly, these
83are not device-specific).
84
85Then, to execute stuff, you repeat these steps, possibly resuing or
86sharing some buffers:
87
88Create some input and output buffers from your context. Set these as
89arguments to your kernel.
90
91Enqueue buffer writes to initialise your input buffers (when not
92initialised at creation time).
93
94Enqueue the kernel execution.
95
96Enqueue buffer reads for your output buffer to read results.
97
24=head1 EXAMPLES 98=head1 EXAMPLES
25 99
26Enumerate all devices and get contexts for them; 100=head2 Enumerate all devices and get contexts for them.
101
102Best run this once to get a feel for the platforms and devices in your
103system.
27 104
28 for my $platform (OpenCL::platforms) { 105 for my $platform (OpenCL::platforms) {
29 warn $platform->info (OpenCL::PLATFORM_NAME); 106 printf "platform: %s\n", $platform->name;
30 warn $platform->info (OpenCL::PLATFORM_EXTENSIONS); 107 printf "extensions: %s\n", $platform->extensions;
31 for my $device ($platform->devices) { 108 for my $device ($platform->devices) {
32 warn $device->info (OpenCL::DEVICE_NAME); 109 printf "+ device: %s\n", $device->name;
33 my $ctx = $device->context_simple; 110 my $ctx = $platform->context (undef, [$device]);
34 # do stuff 111 # do stuff
35 } 112 }
36 } 113 }
37 114
38Get a useful context and a command queue: 115=head2 Get a useful context and a command queue.
39 116
40 my $dev = ((OpenCL::platforms)[0]->devices)[0]; 117This is a useful boilerplate for any OpenCL program that only wants to use
41 my $ctx = $dev->context_simple; 118one device,
42 my $queue = $ctx->command_queue_simple ($dev);
43 119
44Create a buffer with some predefined data, read it back synchronously, 120 my ($platform) = OpenCL::platforms; # find first platform
45then asynchronously: 121 my ($dev) = $platform->devices; # find first device of platform
122 my $ctx = $platform->context (undef, [$dev]); # create context out of those
123 my $queue = $ctx->queue ($dev); # create a command queue for the device
46 124
47 my $buf = $ctx->buffer_sv (OpenCL::MEM_COPY_HOST_PTR, "helmut");
48
49 $queue->enqueue_read_buffer ($buf, 1, 1, 3, my $data);
50 warn $data;
51
52 my $ev = $queue->enqueue_read_buffer ($buf, 0, 1, 3, my $data);
53 $ev->wait;
54 warn $data;
55
56Print all supported image formats: 125=head2 Print all supported image formats of a context.
126
127Best run this once for your context, to see whats available and how to
128gather information.
57 129
58 for my $type (OpenCL::MEM_OBJECT_IMAGE2D, OpenCL::MEM_OBJECT_IMAGE3D) { 130 for my $type (OpenCL::MEM_OBJECT_IMAGE2D, OpenCL::MEM_OBJECT_IMAGE3D) {
59 say "supported image formats for ", OpenCL::enum2str $type; 131 print "supported image formats for ", OpenCL::enum2str $type, "\n";
60 132
61 for my $f ($ctx->supported_image_formats (0, $type)) { 133 for my $f ($ctx->supported_image_formats (0, $type)) {
62 printf " %-10s %-20s\n", OpenCL::enum2str $f->[0], OpenCL::enum2str $f->[1]; 134 printf " %-10s %-20s\n", OpenCL::enum2str $f->[0], OpenCL::enum2str $f->[1];
63 } 135 }
64 } 136 }
65 137
138=head2 Create a buffer with some predefined data, read it back synchronously,
139then asynchronously.
140
141 my $buf = $ctx->buffer_sv (OpenCL::MEM_COPY_HOST_PTR, "helmut");
142
143 $queue->enqueue_read_buffer ($buf, 1, 1, 3, my $data);
144 print "$data\n";
145
146 my $ev = $queue->enqueue_read_buffer ($buf, 0, 1, 3, my $data);
147 $ev->wait;
148 print "$data\n"; # prints "elm"
149
66Create and build a program, then create a kernel out of one of its 150=head2 Create and build a program, then create a kernel out of one of its
67functions: 151functions.
68 152
69 my $src = ' 153 my $src = '
70 __kernel void 154 kernel void
71 squareit (__global float *input, __global float *output) 155 squareit (global float *input, global float *output)
72 { 156 {
73 size_t id = get_global_id (0); 157 $id = get_global_id (0);
74 output [id] = input [id] * input [id]; 158 output [id] = input [id] * input [id];
75 } 159 }
76 '; 160 ';
77 161
78 my $prog = $ctx->program_with_source ($src); 162 my $prog = $ctx->build_program ($src);
79
80 eval { $prog->build ($dev); 1 }
81 or die $prog->build_info ($dev, OpenCL::PROGRAM_BUILD_LOG);
82
83 my $kernel = $prog->kernel ("squareit"); 163 my $kernel = $prog->kernel ("squareit");
84 164
85Create some input and output float buffers, then call squareit on them: 165=head2 Create some input and output float buffers, then call the
166'squareit' kernel on them.
86 167
87 my $input = $ctx->buffer_sv (OpenCL::MEM_COPY_HOST_PTR, pack "f*", 1, 2, 3, 4.5); 168 my $input = $ctx->buffer_sv (OpenCL::MEM_COPY_HOST_PTR, pack "f*", 1, 2, 3, 4.5);
88 my $output = $ctx->buffer (0, OpenCL::SIZEOF_FLOAT * 5); 169 my $output = $ctx->buffer (0, OpenCL::SIZEOF_FLOAT * 5);
89 170
90 # set buffer 171 # set buffer
92 $kernel->set_buffer (1, $output); 173 $kernel->set_buffer (1, $output);
93 174
94 # execute it for all 4 numbers 175 # execute it for all 4 numbers
95 $queue->enqueue_nd_range_kernel ($kernel, undef, [4], undef); 176 $queue->enqueue_nd_range_kernel ($kernel, undef, [4], undef);
96 177
97 # enqueue a barrier ot ensure in-order execution (not really needed in this case) 178 # enqueue a synchronous read
179 $queue->enqueue_read_buffer ($output, 1, 0, OpenCL::SIZEOF_FLOAT * 4, my $data);
180
181 # print the results:
182 printf "%s\n", join ", ", unpack "f*", $data;
183
184=head2 The same enqueue operations as before, but assuming an out-of-order queue,
185showing off barriers.
186
187 # execute it for all 4 numbers
188 $queue->enqueue_nd_range_kernel ($kernel, undef, [4], undef);
189
190 # enqueue a barrier to ensure in-order execution
98 $queue->enqueue_barrier; 191 $queue->enqueue_barrier;
99 192
100 # enqueue an async read (could easily be blocking here though), then wait for it: 193 # enqueue an async read
194 $queue->enqueue_read_buffer ($output, 0, 0, OpenCL::SIZEOF_FLOAT * 4, my $data);
195
196 # wait for all requests to finish
197 $queue->finish;
198
199=head2 The same enqueue operations as before, but assuming an out-of-order queue,
200showing off event objects and wait lists.
201
202 # execute it for all 4 numbers
203 my $ev = $queue->enqueue_nd_range_kernel ($kernel, undef, [4], undef);
204
205 # enqueue an async read
101 my $ev = $queue->enqueue_read_buffer ($output, 0, 0, OpenCL::SIZEOF_FLOAT * 4, my $data); 206 $ev = $queue->enqueue_read_buffer ($output, 0, 0, OpenCL::SIZEOF_FLOAT * 4, my $data, $ev);
207
208 # wait for the last event to complete
102 $ev->wait; 209 $ev->wait;
103 210
104 # print the results: 211=head2 Use the OpenGL module to share a texture between OpenCL and OpenGL and draw some julia
105 say join ", ", unpack "f*", $data; 212set tunnel effect.
213
214This is quite a long example to get you going.
215
216 use OpenGL ":all";
217 use OpenCL;
218
219 # open a window and create a gl texture
220 OpenGL::glpOpenWindow width => 256, height => 256;
221 my $texid = glGenTextures_p 1;
222 glBindTexture GL_TEXTURE_2D, $texid;
223 glTexImage2D_c GL_TEXTURE_2D, 0, GL_RGBA8, 256, 256, 0, GL_RGBA, GL_UNSIGNED_BYTE, 0;
224
225 # find and use the first opencl device that let's us get a shared opengl context
226 my $platform;
227 my $dev;
228 my $ctx;
229
230 for (OpenCL::platforms) {
231 $platform = $_;
232 for ($platform->devices) {
233 $dev = $_;
234 $ctx = $platform->context ([OpenCL::GLX_DISPLAY_KHR, undef, OpenCL::GL_CONTEXT_KHR, undef], [$dev])
235 and last;
236 }
237 }
238
239 $ctx
240 or die "cannot find suitable OpenCL device\n";
241
242 my $queue = $ctx->queue ($dev);
243
244 # now attach an opencl image2d object to the opengl texture
245 my $tex = $ctx->gl_texture2d (OpenCL::MEM_WRITE_ONLY, GL_TEXTURE_2D, 0, $texid);
246
247 # now the boring opencl code
248 my $src = <<EOF;
249 kernel void
250 juliatunnel (write_only image2d_t img, float time)
251 {
252 float2 p = (float2)(get_global_id (0), get_global_id (1)) / 256.f * 2.f - 1.f;
253
254 float2 m = (float2)(1.f, p.y) / fabs (p.x);
255 m.x = fabs (fmod (m.x + time * 0.05f, 4.f)) - 2.f;
256
257 float2 z = m;
258 float2 c = (float2)(sin (time * 0.05005), cos (time * 0.06001));
259
260 for (int i = 0; i < 25 && dot (z, z) < 4.f; ++i)
261 z = (float2)(z.x * z.x - z.y * z.y, 2.f * z.x * z.y) + c;
262
263 float3 colour = (float3)(z.x, z.y, z.x * z.y);
264 write_imagef (img, (int2)(get_global_id (0), get_global_id (1)), (float4)(colour * p.x * p.x, 1.));
265 }
266 EOF
267
268 my $prog = $ctx->build_program ($src);
269 my $kernel = $prog->kernel ("juliatunnel");
270
271 # program compiled, kernel ready, now draw and loop
272
273 for (my $time; ; ++$time) {
274 # acquire objects from opengl
275 $queue->enqueue_acquire_gl_objects ([$tex]);
276
277 # configure and run our kernel
278 $kernel->set_image2d (0, $tex);
279 $kernel->set_float (1, $time);
280 $queue->enqueue_nd_range_kernel ($kernel, undef, [256, 256], undef);
281
282 # release objects to opengl again
283 $queue->enqueue_release_gl_objects ([$tex]);
284
285 # wait
286 $queue->finish;
287
288 # now draw the texture, the defaults should be all right
289 glTexParameterf GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST;
290
291 glEnable GL_TEXTURE_2D;
292 glBegin GL_QUADS;
293 glTexCoord2f 0, 1; glVertex3i -1, -1, -1;
294 glTexCoord2f 0, 0; glVertex3i 1, -1, -1;
295 glTexCoord2f 1, 0; glVertex3i 1, 1, -1;
296 glTexCoord2f 1, 1; glVertex3i -1, 1, -1;
297 glEnd;
298
299 glXSwapBuffers;
300
301 select undef, undef, undef, 1/60;
302 }
303
304=head1 DOCUMENTATION
305
306=head2 BASIC CONVENTIONS
307
308This is not a one-to-one C-style translation of OpenCL to Perl - instead
309I attempted to make the interface as type-safe as possible by introducing
310object syntax where it makes sense. There are a number of important
311differences between the OpenCL C API and this module:
106 312
107=over 4 313=over 4
108 314
315=item * Object lifetime managament is automatic - there is no need
316to free objects explicitly (C<clReleaseXXX>), the release function
317is called automatically once all Perl references to it go away.
318
319=item * OpenCL uses CamelCase for function names
320(e.g. C<clGetPlatformIDs>, C<clGetPlatformInfo>), while this module
321uses underscores as word separator and often leaves out prefixes
322(C<OpenCL::platforms>, C<< $platform->info >>).
323
324=item * OpenCL often specifies fixed vector function arguments as short
325arrays (C<size_t origin[3]>), while this module explicitly expects the
326components as separate arguments (C<$orig_x, $orig_y, $orig_z>) in
327function calls.
328
329=item * Structures are often specified by flattening out their components
330as with short vectors, and returned as arrayrefs.
331
332=item * When enqueuing commands, the wait list is specified by adding
333extra arguments to the function - anywhere a C<$wait_events...> argument
334is documented this can be any number of event objects. As an extsnion
335implemented by this module, C<undef> values will be ignored in the event
336list.
337
338=item * When enqueuing commands, if the enqueue method is called in void
339context, no event is created. In all other contexts an event is returned
340by the method.
341
342=item * This module expects all functions to return C<CL_SUCCESS>. If any
343other status is returned the function will throw an exception, so you
344don't normally have to to any error checking.
345
346=back
347
348=head2 PERL AND OPENCL TYPES
349
350This handy(?) table lists OpenCL types and their perl, PDL and pack/unpack
351format equivalents:
352
353 OpenCL perl PDL pack/unpack
354 char IV - c
355 uchar IV byte C
356 short IV short s
357 ushort IV ushort S
358 int IV long? l
359 uint IV - L
360 long IV longlong q
361 ulong IV - Q
362 float NV float f
363 half IV ushort S
364 double NV double d
365
366=head2 GLX SUPPORT
367
368Due to the sad state that OpenGL support is in in Perl (mostly the OpenGL
369module, which has little to no documentation and has little to no support
370for glX), this module, as a special extension, treats context creation
371properties C<OpenCL::GLX_DISPLAY_KHR> and C<OpenCL::GL_CONTEXT_KHR>
372specially: If either or both of these are C<undef>, then the OpenCL
373module tries to dynamically resolve C<glXGetCurrentDisplay> and
374C<glXGetCurrentContext>, call these functions and use their return values
375instead.
376
377For this to work, the OpenGL library must be loaded, a GLX context must
378have been created and be made current, and C<dlsym> must be available and
379capable of finding the function via C<RTLD_DEFAULT>.
380
109=cut 381=cut
110 382
111package OpenCL; 383package OpenCL;
112 384
113use common::sense; 385use common::sense;
114 386
115BEGIN { 387BEGIN {
116 our $VERSION = '0.02'; 388 our $VERSION = '0.96';
117 389
118 require XSLoader; 390 require XSLoader;
119 XSLoader::load (__PACKAGE__, $VERSION); 391 XSLoader::load (__PACKAGE__, $VERSION);
120 392
393 @OpenCL::Platform::ISA =
394 @OpenCL::Device::ISA =
395 @OpenCL::Context::ISA =
396 @OpenCL::Queue::ISA =
397 @OpenCL::Memory::ISA =
398 @OpenCL::Sampler::ISA =
399 @OpenCL::Program::ISA =
400 @OpenCL::Kernel::ISA =
401 @OpenCL::Event::ISA = OpenCL::Object::;
402
121 @OpenCL::Buffer::ISA = 403 @OpenCL::Buffer::ISA =
122 @OpenCL::Image::ISA = OpenCL::Memory::; 404 @OpenCL::Image::ISA = OpenCL::Memory::;
123 405
406 @OpenCL::BufferObj::ISA = OpenCL::Buffer::;
407
124 @OpenCL::Image2D::ISA = 408 @OpenCL::Image2D::ISA =
409 @OpenCL::Image3D::ISA =
410 @OpenCL::Image2DArray::ISA =
411 @OpenCL::Image1D::ISA =
412 @OpenCL::Image1DArray::ISA =
125 @OpenCL::Image3D::ISA = OpenCL::Image::; 413 @OpenCL::Image1DBuffer::ISA = OpenCL::Image::;
414
415 @OpenCL::UserEvent::ISA = OpenCL::Event::;
126} 416}
127 417
418=head2 THE OpenCL PACKAGE
419
420=over 4
421
422=item $int = OpenCL::errno
423
424The last error returned by a function - it's only valid after an error occured
425and before calling another OpenCL function.
426
427=item $str = OpenCL::err2str $errval
428
429Comverts an error value into a human readable string.
430
431=item $str = OpenCL::enum2str $enum
432
433Converts most enum values (of parameter names, image format constants,
434object types, addressing and filter modes, command types etc.) into a
435human readable string. When confronted with some random integer it can be
436very helpful to pass it through this function to maybe get some readable
437string out of it.
438
439=item @platforms = OpenCL::platforms
440
441Returns all available OpenCL::Platform objects.
442
443L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetPlatformIDs.html>
444
445=item $ctx = OpenCL::context_from_type $properties, $type = OpenCL::DEVICE_TYPE_DEFAULT, $notify = undef
446
447Tries to create a context from a default device and platform - never worked for me.
448
449L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html>
450
451=item OpenCL::wait_for_events $wait_events...
452
453Waits for all events to complete.
454
455L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clWaitForEvents.html>
456
457=back
458
459=head2 THE OpenCL::Object CLASS
460
461This is the base class for all objects in the OpenCL module. The only
462method it implements is the C<id> method, which is only useful if you want
463to interface to OpenCL on the C level.
464
465=over 4
466
467=item $iv = $obj->id
468
469OpenCL objects are represented by pointers or integers on the C level. If
470you want to interface to an OpenCL object directly on the C level, then
471you need this value, which is returned by this method. You should use an
472C<IV> type in your code and cast that to the correct type.
473
474=cut
475
476sub OpenCL::Object::id {
477 ${$_[0]}
478}
479
480=back
481
482=head2 THE OpenCL::Platform CLASS
483
484=over 4
485
486=item @devices = $platform->devices ($type = OpenCL::DEVICE_TYPE_ALL)
487
488Returns a list of matching OpenCL::Device objects.
489
490=item $ctx = $platform->context_from_type ($properties, $type = OpenCL::DEVICE_TYPE_DEFAULT, $notify = undef)
491
492Tries to create a context. Never worked for me, and you need devices explicitly anyway.
493
494L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html>
495
496=item $ctx = $platform->context ($properties = undef, @$devices, $notify = undef)
497
498Create a new OpenCL::Context object using the given device object(s)- a
499CL_CONTEXT_PLATFORM property is supplied automatically.
500
501L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContext.html>
502
503=item $packed_value = $platform->info ($name)
504
505Calls C<clGetPlatformInfo> and returns the packed, raw value - for
506strings, this will be the string (possibly including terminating \0), for
507other values you probably need to use the correct C<unpack>.
508
509It's best to avoid this method and use one of the following convenience
510wrappers.
511
512L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetPlatformInfo.html>
513
514=item $platform->unload_compiler
515
516Attempts to unload the compiler for this platform, for endless
517profit. Does nothing on OpenCL 1.1.
518
519L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clUnloadPlatformCompiler.html>
520
521=for gengetinfo begin platform
522
523=item $string = $platform->profile
524
525Calls C<clGetPlatformInfo> with C<CL_PLATFORM_PROFILE> and returns the result.
526
527=item $string = $platform->version
528
529Calls C<clGetPlatformInfo> with C<CL_PLATFORM_VERSION> and returns the result.
530
531=item $string = $platform->name
532
533Calls C<clGetPlatformInfo> with C<CL_PLATFORM_NAME> and returns the result.
534
535=item $string = $platform->vendor
536
537Calls C<clGetPlatformInfo> with C<CL_PLATFORM_VENDOR> and returns the result.
538
539=item $string = $platform->extensions
540
541Calls C<clGetPlatformInfo> with C<CL_PLATFORM_EXTENSIONS> and returns the result.
542
543=for gengetinfo end platform
544
545=back
546
547=head2 THE OpenCL::Device CLASS
548
549=over 4
550
551=item $packed_value = $device->info ($name)
552
553See C<< $platform->info >> for details.
554
555L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetDeviceInfo.html>
556
557=for gengetinfo begin device
558
559=item $device_type = $device->type
560
561Calls C<clGetDeviceInfo> with C<CL_DEVICE_TYPE> and returns the result.
562
563=item $uint = $device->vendor_id
564
565Calls C<clGetDeviceInfo> with C<CL_DEVICE_VENDOR_ID> and returns the result.
566
567=item $uint = $device->max_compute_units
568
569Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_COMPUTE_UNITS> and returns the result.
570
571=item $uint = $device->max_work_item_dimensions
572
573Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_WORK_ITEM_DIMENSIONS> and returns the result.
574
575=item $int = $device->max_work_group_size
576
577Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_WORK_GROUP_SIZE> and returns the result.
578
579=item @ints = $device->max_work_item_sizes
580
581Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_WORK_ITEM_SIZES> and returns the result.
582
583=item $uint = $device->preferred_vector_width_char
584
585Calls C<clGetDeviceInfo> with C<CL_DEVICE_PREFERRED_VECTOR_WIDTH_CHAR> and returns the result.
586
587=item $uint = $device->preferred_vector_width_short
588
589Calls C<clGetDeviceInfo> with C<CL_DEVICE_PREFERRED_VECTOR_WIDTH_SHORT> and returns the result.
590
591=item $uint = $device->preferred_vector_width_int
592
593Calls C<clGetDeviceInfo> with C<CL_DEVICE_PREFERRED_VECTOR_WIDTH_INT> and returns the result.
594
595=item $uint = $device->preferred_vector_width_long
596
597Calls C<clGetDeviceInfo> with C<CL_DEVICE_PREFERRED_VECTOR_WIDTH_LONG> and returns the result.
598
599=item $uint = $device->preferred_vector_width_float
600
601Calls C<clGetDeviceInfo> with C<CL_DEVICE_PREFERRED_VECTOR_WIDTH_FLOAT> and returns the result.
602
603=item $uint = $device->preferred_vector_width_double
604
605Calls C<clGetDeviceInfo> with C<CL_DEVICE_PREFERRED_VECTOR_WIDTH_DOUBLE> and returns the result.
606
607=item $uint = $device->max_clock_frequency
608
609Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_CLOCK_FREQUENCY> and returns the result.
610
611=item $bitfield = $device->address_bits
612
613Calls C<clGetDeviceInfo> with C<CL_DEVICE_ADDRESS_BITS> and returns the result.
614
615=item $uint = $device->max_read_image_args
616
617Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_READ_IMAGE_ARGS> and returns the result.
618
619=item $uint = $device->max_write_image_args
620
621Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_WRITE_IMAGE_ARGS> and returns the result.
622
623=item $ulong = $device->max_mem_alloc_size
624
625Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_MEM_ALLOC_SIZE> and returns the result.
626
627=item $int = $device->image2d_max_width
628
629Calls C<clGetDeviceInfo> with C<CL_DEVICE_IMAGE2D_MAX_WIDTH> and returns the result.
630
631=item $int = $device->image2d_max_height
632
633Calls C<clGetDeviceInfo> with C<CL_DEVICE_IMAGE2D_MAX_HEIGHT> and returns the result.
634
635=item $int = $device->image3d_max_width
636
637Calls C<clGetDeviceInfo> with C<CL_DEVICE_IMAGE3D_MAX_WIDTH> and returns the result.
638
639=item $int = $device->image3d_max_height
640
641Calls C<clGetDeviceInfo> with C<CL_DEVICE_IMAGE3D_MAX_HEIGHT> and returns the result.
642
643=item $int = $device->image3d_max_depth
644
645Calls C<clGetDeviceInfo> with C<CL_DEVICE_IMAGE3D_MAX_DEPTH> and returns the result.
646
647=item $uint = $device->image_support
648
649Calls C<clGetDeviceInfo> with C<CL_DEVICE_IMAGE_SUPPORT> and returns the result.
650
651=item $int = $device->max_parameter_size
652
653Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_PARAMETER_SIZE> and returns the result.
654
655=item $uint = $device->max_samplers
656
657Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_SAMPLERS> and returns the result.
658
659=item $uint = $device->mem_base_addr_align
660
661Calls C<clGetDeviceInfo> with C<CL_DEVICE_MEM_BASE_ADDR_ALIGN> and returns the result.
662
663=item $uint = $device->min_data_type_align_size
664
665Calls C<clGetDeviceInfo> with C<CL_DEVICE_MIN_DATA_TYPE_ALIGN_SIZE> and returns the result.
666
667=item $device_fp_config = $device->single_fp_config
668
669Calls C<clGetDeviceInfo> with C<CL_DEVICE_SINGLE_FP_CONFIG> and returns the result.
670
671=item $device_mem_cache_type = $device->global_mem_cache_type
672
673Calls C<clGetDeviceInfo> with C<CL_DEVICE_GLOBAL_MEM_CACHE_TYPE> and returns the result.
674
675=item $uint = $device->global_mem_cacheline_size
676
677Calls C<clGetDeviceInfo> with C<CL_DEVICE_GLOBAL_MEM_CACHELINE_SIZE> and returns the result.
678
679=item $ulong = $device->global_mem_cache_size
680
681Calls C<clGetDeviceInfo> with C<CL_DEVICE_GLOBAL_MEM_CACHE_SIZE> and returns the result.
682
683=item $ulong = $device->global_mem_size
684
685Calls C<clGetDeviceInfo> with C<CL_DEVICE_GLOBAL_MEM_SIZE> and returns the result.
686
687=item $ulong = $device->max_constant_buffer_size
688
689Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_CONSTANT_BUFFER_SIZE> and returns the result.
690
691=item $uint = $device->max_constant_args
692
693Calls C<clGetDeviceInfo> with C<CL_DEVICE_MAX_CONSTANT_ARGS> and returns the result.
694
695=item $device_local_mem_type = $device->local_mem_type
696
697Calls C<clGetDeviceInfo> with C<CL_DEVICE_LOCAL_MEM_TYPE> and returns the result.
698
699=item $ulong = $device->local_mem_size
700
701Calls C<clGetDeviceInfo> with C<CL_DEVICE_LOCAL_MEM_SIZE> and returns the result.
702
703=item $boolean = $device->error_correction_support
704
705Calls C<clGetDeviceInfo> with C<CL_DEVICE_ERROR_CORRECTION_SUPPORT> and returns the result.
706
707=item $int = $device->profiling_timer_resolution
708
709Calls C<clGetDeviceInfo> with C<CL_DEVICE_PROFILING_TIMER_RESOLUTION> and returns the result.
710
711=item $boolean = $device->endian_little
712
713Calls C<clGetDeviceInfo> with C<CL_DEVICE_ENDIAN_LITTLE> and returns the result.
714
715=item $boolean = $device->available
716
717Calls C<clGetDeviceInfo> with C<CL_DEVICE_AVAILABLE> and returns the result.
718
719=item $boolean = $device->compiler_available
720
721Calls C<clGetDeviceInfo> with C<CL_DEVICE_COMPILER_AVAILABLE> and returns the result.
722
723=item $device_exec_capabilities = $device->execution_capabilities
724
725Calls C<clGetDeviceInfo> with C<CL_DEVICE_EXECUTION_CAPABILITIES> and returns the result.
726
727=item $command_queue_properties = $device->properties
728
729Calls C<clGetDeviceInfo> with C<CL_DEVICE_QUEUE_PROPERTIES> and returns the result.
730
731=item $ = $device->platform
732
733Calls C<clGetDeviceInfo> with C<CL_DEVICE_PLATFORM> and returns the result.
734
735=item $string = $device->name
736
737Calls C<clGetDeviceInfo> with C<CL_DEVICE_NAME> and returns the result.
738
739=item $string = $device->vendor
740
741Calls C<clGetDeviceInfo> with C<CL_DEVICE_VENDOR> and returns the result.
742
743=item $string = $device->driver_version
744
745Calls C<clGetDeviceInfo> with C<CL_DRIVER_VERSION> and returns the result.
746
747=item $string = $device->profile
748
749Calls C<clGetDeviceInfo> with C<CL_DEVICE_PROFILE> and returns the result.
750
751=item $string = $device->version
752
753Calls C<clGetDeviceInfo> with C<CL_DEVICE_VERSION> and returns the result.
754
755=item $string = $device->extensions
756
757Calls C<clGetDeviceInfo> with C<CL_DEVICE_EXTENSIONS> and returns the result.
758
759=item $uint = $device->preferred_vector_width_half
760
761Calls C<clGetDeviceInfo> with C<CL_DEVICE_PREFERRED_VECTOR_WIDTH_HALF> and returns the result.
762
763=item $uint = $device->native_vector_width_char
764
765Calls C<clGetDeviceInfo> with C<CL_DEVICE_NATIVE_VECTOR_WIDTH_CHAR> and returns the result.
766
767=item $uint = $device->native_vector_width_short
768
769Calls C<clGetDeviceInfo> with C<CL_DEVICE_NATIVE_VECTOR_WIDTH_SHORT> and returns the result.
770
771=item $uint = $device->native_vector_width_int
772
773Calls C<clGetDeviceInfo> with C<CL_DEVICE_NATIVE_VECTOR_WIDTH_INT> and returns the result.
774
775=item $uint = $device->native_vector_width_long
776
777Calls C<clGetDeviceInfo> with C<CL_DEVICE_NATIVE_VECTOR_WIDTH_LONG> and returns the result.
778
779=item $uint = $device->native_vector_width_float
780
781Calls C<clGetDeviceInfo> with C<CL_DEVICE_NATIVE_VECTOR_WIDTH_FLOAT> and returns the result.
782
783=item $uint = $device->native_vector_width_double
784
785Calls C<clGetDeviceInfo> with C<CL_DEVICE_NATIVE_VECTOR_WIDTH_DOUBLE> and returns the result.
786
787=item $uint = $device->native_vector_width_half
788
789Calls C<clGetDeviceInfo> with C<CL_DEVICE_NATIVE_VECTOR_WIDTH_HALF> and returns the result.
790
791=item $device_fp_config = $device->double_fp_config
792
793Calls C<clGetDeviceInfo> with C<CL_DEVICE_DOUBLE_FP_CONFIG> and returns the result.
794
795=item $device_fp_config = $device->half_fp_config
796
797Calls C<clGetDeviceInfo> with C<CL_DEVICE_HALF_FP_CONFIG> and returns the result.
798
799=item $boolean = $device->host_unified_memory
800
801Calls C<clGetDeviceInfo> with C<CL_DEVICE_HOST_UNIFIED_MEMORY> and returns the result.
802
803=item $device = $device->parent_device_ext
804
805Calls C<clGetDeviceInfo> with C<CL_DEVICE_PARENT_DEVICE_EXT> and returns the result.
806
807=item @device_partition_property_exts = $device->partition_types_ext
808
809Calls C<clGetDeviceInfo> with C<CL_DEVICE_PARTITION_TYPES_EXT> and returns the result.
810
811=item @device_partition_property_exts = $device->affinity_domains_ext
812
813Calls C<clGetDeviceInfo> with C<CL_DEVICE_AFFINITY_DOMAINS_EXT> and returns the result.
814
815=item $uint = $device->reference_count_ext
816
817Calls C<clGetDeviceInfo> with C<CL_DEVICE_REFERENCE_COUNT_EXT> and returns the result.
818
819=item @device_partition_property_exts = $device->partition_style_ext
820
821Calls C<clGetDeviceInfo> with C<CL_DEVICE_PARTITION_STYLE_EXT> and returns the result.
822
823=for gengetinfo end device
824
825=back
826
827=head2 THE OpenCL::Context CLASS
828
829=over 4
830
831=item $prog = $ctx->build_program ($program, $options = "")
832
833This convenience function tries to build the program on all devices in
834the context. If the build fails, then the function will C<croak> with the
835build log. Otherwise ti returns the program object.
836
837The C<$program> can either be a C<OpenCL::Program> object or a string
838containing the program. In the latter case, a program objetc will be
839created automatically.
840
841=cut
842
843sub OpenCL::Context::build_program {
844 my ($self, $prog, $options) = @_;
845
846 $prog = $self->program_with_source ($prog)
847 unless ref $prog;
848
849 for my $dev ($self->devices) {
850 eval { $prog->build ($dev, $options); 1 }
851 or Carp::croak "Building OpenCL program for device '" . $dev->name . "' failed:\n"
852 . $prog->build_log ($dev);
853 }
854
855 $prog
856}
857
858=item $queue = $ctx->queue ($device, $properties)
859
860Create a new OpenCL::Queue object from the context and the given device.
861
862L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateCommandQueue.html>
863
864Example: create an out-of-order queue.
865
866 $queue = $ctx->queue ($device, OpenCL::QUEUE_OUT_OF_ORDER_EXEC_MODE_ENABLE);
867
868=item $ev = $ctx->user_event
869
870Creates a new OpenCL::UserEvent object.
871
872L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateUserEvent.html>
873
874=item $buf = $ctx->buffer ($flags, $len)
875
876Creates a new OpenCL::Buffer (actually OpenCL::BufferObj) object with the
877given flags and octet-size.
878
879L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateBuffer.html>
880
881=item $buf = $ctx->buffer_sv ($flags, $data)
882
883Creates a new OpenCL::Buffer (actually OpenCL::BufferObj) object and
884initialise it with the given data values.
885
886=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)
887
888Creates a new OpenCL::Image object and optionally initialises it with
889the given data values.
890
891L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clCreateImage.html>
892
893=item $img = $ctx->image2d ($flags, $channel_order, $channel_type, $width, $height, $row_pitch = 0, $data = undef)
894
895Creates a new OpenCL::Image2D object and optionally initialises it with
896the given data values.
897
898L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateImage2D.html>
899
900=item $img = $ctx->image3d ($flags, $channel_order, $channel_type, $width, $height, $depth, $row_pitch = 0, $slice_pitch = 0, $data = undef)
901
902Creates a new OpenCL::Image3D object and optionally initialises it with
903the given data values.
904
905L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateImage3D.html>
906
907=item $buffer = $ctx->gl_buffer ($flags, $bufobj)
908
909Creates a new OpenCL::Buffer (actually OpenCL::BufferObj) object that refers to the given
910OpenGL buffer object.
911
912http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLBuffer.html
913
914=item $img = $ctx->gl_texture ($flags, $target, $miplevel, $texture)
915
916Creates a new OpenCL::Image object that refers to the given OpenGL
917texture object or buffer.
918
919http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clCreateFromGLTexture.html
920
921=item $img = $ctx->gl_texture2d ($flags, $target, $miplevel, $texture)
922
923Creates a new OpenCL::Image2D object that refers to the given OpenGL
9242D texture object.
925
926http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLTexture2D.html
927
928=item $img = $ctx->gl_texture3d ($flags, $target, $miplevel, $texture)
929
930Creates a new OpenCL::Image3D object that refers to the given OpenGL
9313D texture object.
932
933http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLTexture3D.html
934
935=item $ctx->gl_renderbuffer ($flags, $renderbuffer)
936
937Creates a new OpenCL::Image2D object that refers to the given OpenGL
938render buffer.
939
940http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLRenderbuffer.html
941
942=item @formats = $ctx->supported_image_formats ($flags, $image_type)
943
944Returns a list of matching image formats - each format is an arrayref with
945two values, $channel_order and $channel_type, in it.
946
947L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetSupportedImageFormats.html>
948
949=item $sampler = $ctx->sampler ($normalized_coords, $addressing_mode, $filter_mode)
950
951Creates a new OpenCL::Sampler object.
952
953L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateSampler.html>
954
955=item $program = $ctx->program_with_source ($string)
956
957Creates a new OpenCL::Program object from the given source code.
958
959L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateProgramWithSource.html>
960
961=item $packed_value = $ctx->info ($name)
962
963See C<< $platform->info >> for details.
964
965L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetContextInfo.html>
966
967=for gengetinfo begin context
968
969=item $uint = $context->reference_count
970
971Calls C<clGetContextInfo> with C<CL_CONTEXT_REFERENCE_COUNT> and returns the result.
972
973=item @devices = $context->devices
974
975Calls C<clGetContextInfo> with C<CL_CONTEXT_DEVICES> and returns the result.
976
977=item @property_ints = $context->properties
978
979Calls C<clGetContextInfo> with C<CL_CONTEXT_PROPERTIES> and returns the result.
980
981=item $uint = $context->num_devices
982
983Calls C<clGetContextInfo> with C<CL_CONTEXT_NUM_DEVICES> and returns the result.
984
985=for gengetinfo end context
986
987=back
988
989=head2 THE OpenCL::Queue CLASS
990
991An OpenCL::Queue represents an execution queue for OpenCL. You execute
992requests by calling their respective C<enqueue_xxx> method and waitinf for
993it to complete in some way.
994
995All the enqueue methods return an event object that can be used to wait
996for completion, unless the method is called in void context, in which case
997no event object is created.
998
999They also allow you to specify any number of other event objects that this
1000request has to wait for before it starts executing, by simply passing the
1001event objects as extra parameters to the enqueue methods. To simplify
1002program design, this module ignores any C<undef> values in the list of
1003events. This makes it possible to code operations such as this, without
1004having to put a valid event object into C<$event> first:
1005
1006 $event = $queue->enqueue_xxx (..., $event);
1007
1008Queues execute in-order by default, without any parallelism, so in most
1009cases (i.e. you use only one queue) it's not necessary to wait for or
1010create event objects, althoguh an our of order queue is often a bit
1011faster.
1012
1013=over 4
1014
1015=item $ev = $queue->enqueue_read_buffer ($buffer, $blocking, $offset, $len, $data, $wait_events...)
1016
1017Reads data from buffer into the given string.
1018
1019L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReadBuffer.html>
1020
1021=item $ev = $queue->enqueue_write_buffer ($buffer, $blocking, $offset, $data, $wait_events...)
1022
1023Writes data to buffer from the given string.
1024
1025L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteBuffer.html>
1026
1027=item $ev = $queue->enqueue_copy_buffer ($src, $dst, $src_offset, $dst_offset, $len, $wait_events...)
1028
1029L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBuffer.html>
1030
1031=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...)
1032
1033http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReadBufferRect.html
1034
1035=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...)
1036
1037http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteBufferRect.html
1038
1039=item $ev = $queue->enqueue_read_image ($src, $blocking, $x, $y, $z, $width, $height, $depth, $row_pitch, $slice_pitch, $data, $wait_events...)
1040
1041L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBufferRect.html>
1042
1043=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...)
1044
1045L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReadImage.html>
1046
1047=item $ev = $queue->enqueue_write_image ($src, $blocking, $x, $y, $z, $width, $height, $depth, $row_pitch, $slice_pitch, $data, $wait_events...)
1048
1049L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteImage.html>
1050
1051=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...)
1052
1053L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyImage.html>
1054
1055=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...)
1056
1057L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyImageToBuffer.html>
1058
1059=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...)
1060
1061Yeah.
1062
1063L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBufferToImage.html>.
1064
1065=item $ev = $queue->enqueue_fill_buffer ($mem, $pattern, $offset, $size, ...)
1066
1067Fills the given buffer object with repeated applications of C<$pattern>,
1068starting at C<$offset> for C<$size> octets.
1069
1070L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clEnqueueFillBuffer.html>
1071
1072=item $ev = $queue->enqueue_fill_image ($img, $r, $g, $b, $a, $x, $y, $z, $width, $height, $depth, ...)
1073
1074Fills the given image area with the given rgba colour components. The
1075components are normally floating point values between C<0> and C<1>,
1076except when the image channel data type is a signe dor unsigned
1077unnormalised format, in which case the range is determined by the format.
1078
1079L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clEnqueueFillImage.html>
1080
1081=item $ev = $queue->enqueue_task ($kernel, $wait_events...)
1082
1083L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueTask.html>
1084
1085=item $ev = $queue->enqueue_nd_range_kernel ($kernel, @$global_work_offset, @$global_work_size, @$local_work_size, $wait_events...)
1086
1087Enqueues a kernel execution.
1088
1089@$global_work_size must be specified as a reference to an array of
1090integers specifying the work sizes (element counts).
1091
1092@$global_work_offset must be either C<undef> (in which case all offsets
1093are C<0>), or a reference to an array of work offsets, with the same number
1094of elements as @$global_work_size.
1095
1096@$local_work_size must be either C<undef> (in which case the
1097implementation is supposed to choose good local work sizes), or a
1098reference to an array of local work sizes, with the same number of
1099elements as @$global_work_size.
1100
1101L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueNDRangeKernel.html>
1102
1103=item $ev = $queue->enqueue_acquire_gl_objects ([object, ...], $wait_events...)
1104
1105Enqueues a list (an array-ref of OpenCL::Memory objects) to be acquired
1106for subsequent OpenCL usage.
1107
1108L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueAcquireGLObjects.html>
1109
1110=item $ev = $queue->enqueue_release_gl_objects ([object, ...], $wait_events...)
1111
1112Enqueues a list (an array-ref of OpenCL::Memory objects) to be released
1113for subsequent OpenGL usage.
1114
1115L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReleaseGLObjects.html>
1116
1117=item $ev = $queue->enqueue_wait_for_events ($wait_events...)
1118
1119L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWaitForEvents.html>
1120
1121=item $ev = $queue->enqueue_marker ($wait_events...)
1122
1123L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clEnqueueMarkerWithWaitList.html>
1124
1125=item $ev = $queue->enqueue_barrier ($wait_events...)
1126
1127L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clEnqueueBarrierWithWaitList.html>
1128
1129=item $queue->flush
1130
1131L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clFlush.html>
1132
1133=item $queue->finish
1134
1135L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clFinish.html>
1136
1137=item $packed_value = $queue->info ($name)
1138
1139See C<< $platform->info >> for details.
1140
1141L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetCommandQueueInfo.html>
1142
1143=for gengetinfo begin command_queue
1144
1145=item $ctx = $command_queue->context
1146
1147Calls C<clGetCommandQueueInfo> with C<CL_QUEUE_CONTEXT> and returns the result.
1148
1149=item $device = $command_queue->device
1150
1151Calls C<clGetCommandQueueInfo> with C<CL_QUEUE_DEVICE> and returns the result.
1152
1153=item $uint = $command_queue->reference_count
1154
1155Calls C<clGetCommandQueueInfo> with C<CL_QUEUE_REFERENCE_COUNT> and returns the result.
1156
1157=item $command_queue_properties = $command_queue->properties
1158
1159Calls C<clGetCommandQueueInfo> with C<CL_QUEUE_PROPERTIES> and returns the result.
1160
1161=for gengetinfo end command_queue
1162
1163=back
1164
1165=head2 THE OpenCL::Memory CLASS
1166
1167This the superclass of all memory objects - OpenCL::Buffer, OpenCL::Image,
1168OpenCL::Image2D and OpenCL::Image3D.
1169
1170=over 4
1171
1172=item $packed_value = $memory->info ($name)
1173
1174See C<< $platform->info >> for details.
1175
1176L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetMemObjectInfo.html>
1177
1178=for gengetinfo begin mem
1179
1180=item $mem_object_type = $mem->type
1181
1182Calls C<clGetMemObjectInfo> with C<CL_MEM_TYPE> and returns the result.
1183
1184=item $mem_flags = $mem->flags
1185
1186Calls C<clGetMemObjectInfo> with C<CL_MEM_FLAGS> and returns the result.
1187
1188=item $int = $mem->size
1189
1190Calls C<clGetMemObjectInfo> with C<CL_MEM_SIZE> and returns the result.
1191
1192=item $ptr_value = $mem->host_ptr
1193
1194Calls C<clGetMemObjectInfo> with C<CL_MEM_HOST_PTR> and returns the result.
1195
1196=item $uint = $mem->map_count
1197
1198Calls C<clGetMemObjectInfo> with C<CL_MEM_MAP_COUNT> and returns the result.
1199
1200=item $uint = $mem->reference_count
1201
1202Calls C<clGetMemObjectInfo> with C<CL_MEM_REFERENCE_COUNT> and returns the result.
1203
1204=item $ctx = $mem->context
1205
1206Calls C<clGetMemObjectInfo> with C<CL_MEM_CONTEXT> and returns the result.
1207
1208=item $mem = $mem->associated_memobject
1209
1210Calls C<clGetMemObjectInfo> with C<CL_MEM_ASSOCIATED_MEMOBJECT> and returns the result.
1211
1212=item $int = $mem->offset
1213
1214Calls C<clGetMemObjectInfo> with C<CL_MEM_OFFSET> and returns the result.
1215
1216=for gengetinfo end mem
1217
1218=item ($type, $name) = $mem->gl_object_info
1219
1220Returns the OpenGL object type (e.g. OpenCL::GL_OBJECT_TEXTURE2D) and the
1221object "name" (e.g. the texture name) used to create this memory object.
1222
1223L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetGLObjectInfo.html>
1224
1225=back
1226
1227=head2 THE OpenCL::Buffer CLASS
1228
1229This is a subclass of OpenCL::Memory, and the superclass of
1230OpenCL::BufferObj. Its purpose is simply to distinguish between buffers
1231and sub-buffers.
1232
1233=head2 THE OpenCL::BufferObj CLASS
1234
1235This is a subclass of OpenCL::Buffer and thus OpenCL::Memory. It exists
1236because one cna create sub buffers of OpenLC::BufferObj objects, but not
1237sub buffers from these sub buffers.
1238
1239=over 4
1240
1241=item $subbuf = $buf_obj->sub_buffer_region ($flags, $origin, $size)
1242
1243Creates an OpenCL::Buffer objects from this buffer and returns it. The
1244C<buffer_create_type> is assumed to be C<CL_BUFFER_CREATE_TYPE_REGION>.
1245
1246L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateSubBuffer.html>
1247
1248=back
1249
1250=head2 THE OpenCL::Image CLASS
1251
1252This is the superclass of all image objects - OpenCL::Image1D,
1253OpenCL::Image1DArray, OpenCL::Image1DBuffer, OpenCL::Image2D,
1254OpenCL::Image2DArray and OpenCL::Image3D.
1255
1256=over 4
1257
1258=item $packed_value = $image->image_info ($name)
1259
1260See C<< $platform->info >> for details.
1261
1262The reason this method is not called C<info> is that there already is an
1263C<< ->info >> method inherited from C<OpenCL::Memory>.
1264
1265L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetImageInfo.html>
1266
1267=item ($channel_order, $channel_data_type) = $image->format
1268
1269Returns the channel order and type used to create the image by calling
1270C<clGetImageInfo> with C<CL_IMAGE_FORMAT>.
1271
1272=for gengetinfo begin image
1273
1274=item $int = $image->element_size
1275
1276Calls C<clGetImageInfo> with C<CL_IMAGE_ELEMENT_SIZE> and returns the result.
1277
1278=item $int = $image->row_pitch
1279
1280Calls C<clGetImageInfo> with C<CL_IMAGE_ROW_PITCH> and returns the result.
1281
1282=item $int = $image->slice_pitch
1283
1284Calls C<clGetImageInfo> with C<CL_IMAGE_SLICE_PITCH> and returns the result.
1285
1286=item $int = $image->width
1287
1288Calls C<clGetImageInfo> with C<CL_IMAGE_WIDTH> and returns the result.
1289
1290=item $int = $image->height
1291
1292Calls C<clGetImageInfo> with C<CL_IMAGE_HEIGHT> and returns the result.
1293
1294=item $int = $image->depth
1295
1296Calls C<clGetImageInfo> with C<CL_IMAGE_DEPTH> and returns the result.
1297
1298=for gengetinfo end image
1299
1300=for gengetinfo begin gl_texture
1301
1302=item $GLenum = $gl_texture->target
1303
1304Calls C<clGetGLTextureInfo> with C<CL_GL_TEXTURE_TARGET> and returns the result.
1305
1306=item $GLint = $gl_texture->gl_mipmap_level
1307
1308Calls C<clGetGLTextureInfo> with C<CL_GL_MIPMAP_LEVEL> and returns the result.
1309
1310=for gengetinfo end gl_texture
1311
1312=back
1313
1314=head2 THE OpenCL::Sampler CLASS
1315
1316=over 4
1317
1318=item $packed_value = $sampler->info ($name)
1319
1320See C<< $platform->info >> for details.
1321
1322L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetSamplerInfo.html>
1323
1324=for gengetinfo begin sampler
1325
1326=item $uint = $sampler->reference_count
1327
1328Calls C<clGetSamplerInfo> with C<CL_SAMPLER_REFERENCE_COUNT> and returns the result.
1329
1330=item $ctx = $sampler->context
1331
1332Calls C<clGetSamplerInfo> with C<CL_SAMPLER_CONTEXT> and returns the result.
1333
1334=item $addressing_mode = $sampler->normalized_coords
1335
1336Calls C<clGetSamplerInfo> with C<CL_SAMPLER_NORMALIZED_COORDS> and returns the result.
1337
1338=item $filter_mode = $sampler->addressing_mode
1339
1340Calls C<clGetSamplerInfo> with C<CL_SAMPLER_ADDRESSING_MODE> and returns the result.
1341
1342=item $boolean = $sampler->filter_mode
1343
1344Calls C<clGetSamplerInfo> with C<CL_SAMPLER_FILTER_MODE> and returns the result.
1345
1346=for gengetinfo end sampler
1347
1348=back
1349
1350=head2 THE OpenCL::Program CLASS
1351
1352=over 4
1353
1354=item $program->build ($device, $options = "")
1355
1356Tries to build the program with the given options. See also the
1357C<$ctx->build> convenience function.
1358
1359L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clBuildProgram.html>
1360
1361=item $packed_value = $program->build_info ($device, $name)
1362
1363Similar to C<< $platform->info >>, but returns build info for a previous
1364build attempt for the given device.
1365
1366L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetBuildInfo.html>
1367
1368=item $kernel = $program->kernel ($function_name)
1369
1370Creates an OpenCL::Kernel object out of the named C<__kernel> function in
1371the program.
1372
1373L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateKernel.html>
1374
1375=item @kernels = $program->kernels_in_program
1376
1377Returns all kernels successfully compiled for all devices in program.
1378
1379http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateKernelsInProgram.html
1380
1381=for gengetinfo begin program_build
1382
1383=item $build_status = $program->build_status ($device)
1384
1385Calls C<clGetProgramBuildInfo> with C<CL_PROGRAM_BUILD_STATUS> and returns the result.
1386
1387=item $string = $program->build_options ($device)
1388
1389Calls C<clGetProgramBuildInfo> with C<CL_PROGRAM_BUILD_OPTIONS> and returns the result.
1390
1391=item $string = $program->build_log ($device)
1392
1393Calls C<clGetProgramBuildInfo> with C<CL_PROGRAM_BUILD_LOG> and returns the result.
1394
1395=for gengetinfo end program_build
1396
1397=item $packed_value = $program->info ($name)
1398
1399See C<< $platform->info >> for details.
1400
1401L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetProgramInfo.html>
1402
1403=for gengetinfo begin program
1404
1405=item $uint = $program->reference_count
1406
1407Calls C<clGetProgramInfo> with C<CL_PROGRAM_REFERENCE_COUNT> and returns the result.
1408
1409=item $ctx = $program->context
1410
1411Calls C<clGetProgramInfo> with C<CL_PROGRAM_CONTEXT> and returns the result.
1412
1413=item $uint = $program->num_devices
1414
1415Calls C<clGetProgramInfo> with C<CL_PROGRAM_NUM_DEVICES> and returns the result.
1416
1417=item @devices = $program->devices
1418
1419Calls C<clGetProgramInfo> with C<CL_PROGRAM_DEVICES> and returns the result.
1420
1421=item $string = $program->source
1422
1423Calls C<clGetProgramInfo> with C<CL_PROGRAM_SOURCE> and returns the result.
1424
1425=item @ints = $program->binary_sizes
1426
1427Calls C<clGetProgramInfo> with C<CL_PROGRAM_BINARY_SIZES> and returns the result.
1428
1429=for gengetinfo end program
1430
1431=item @blobs = $program->binaries
1432
1433Returns a string for the compiled binary for every device associated with
1434the program, empty strings indicate missing programs, and an empty result
1435means no program binaries are available.
1436
1437These "binaries" are often, in fact, informative low-level assembly
1438sources.
1439
1440L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetProgramInfo.html>
1441
1442=back
1443
1444=head2 THE OpenCL::Kernel CLASS
1445
1446=over 4
1447
1448=item $packed_value = $kernel->info ($name)
1449
1450See C<< $platform->info >> for details.
1451
1452L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetKernelInfo.html>
1453
1454=for gengetinfo begin kernel
1455
1456=item $string = $kernel->function_name
1457
1458Calls C<clGetKernelInfo> with C<CL_KERNEL_FUNCTION_NAME> and returns the result.
1459
1460=item $uint = $kernel->num_args
1461
1462Calls C<clGetKernelInfo> with C<CL_KERNEL_NUM_ARGS> and returns the result.
1463
1464=item $uint = $kernel->reference_count
1465
1466Calls C<clGetKernelInfo> with C<CL_KERNEL_REFERENCE_COUNT> and returns the result.
1467
1468=item $ctx = $kernel->context
1469
1470Calls C<clGetKernelInfo> with C<CL_KERNEL_CONTEXT> and returns the result.
1471
1472=item $program = $kernel->program
1473
1474Calls C<clGetKernelInfo> with C<CL_KERNEL_PROGRAM> and returns the result.
1475
1476=for gengetinfo end kernel
1477
1478=item $packed_value = $kernel->work_group_info ($device, $name)
1479
1480See C<< $platform->info >> for details.
1481
1482The reason this method is not called C<info> is that there already is an
1483C<< ->info >> method.
1484
1485L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetKernelWorkGroupInfo.html>
1486
1487=for gengetinfo begin kernel_work_group
1488
1489=item $int = $kernel->work_group_size ($device)
1490
1491Calls C<clGetKernelWorkGroupInfo> with C<CL_KERNEL_WORK_GROUP_SIZE> and returns the result.
1492
1493=item @ints = $kernel->compile_work_group_size ($device)
1494
1495Calls C<clGetKernelWorkGroupInfo> with C<CL_KERNEL_COMPILE_WORK_GROUP_SIZE> and returns the result.
1496
1497=item $ulong = $kernel->local_mem_size ($device)
1498
1499Calls C<clGetKernelWorkGroupInfo> with C<CL_KERNEL_LOCAL_MEM_SIZE> and returns the result.
1500
1501=item $int = $kernel->preferred_work_group_size_multiple ($device)
1502
1503Calls C<clGetKernelWorkGroupInfo> with C<CL_KERNEL_PREFERRED_WORK_GROUP_SIZE_MULTIPLE> and returns the result.
1504
1505=item $ulong = $kernel->private_mem_size ($device)
1506
1507Calls C<clGetKernelWorkGroupInfo> with C<CL_KERNEL_PRIVATE_MEM_SIZE> and returns the result.
1508
1509=for gengetinfo end kernel_work_group
1510
1511=item $kernel->set_TYPE ($index, $value)
1512
1513This is a family of methods to set the kernel argument with the number C<$index> to the give C<$value>.
1514
1515TYPE is one of C<char>, C<uchar>, C<short>, C<ushort>, C<int>, C<uint>,
1516C<long>, C<ulong>, C<half>, C<float>, C<double>, C<memory>, C<buffer>,
1517C<image2d>, C<image3d>, C<sampler>, C<local> or C<event>.
1518
1519Chars and integers (including the half type) are specified as integers,
1520float and double as floating point values, memory/buffer/image2d/image3d
1521must be an object of that type or C<undef>, local-memory arguments are
1522set by specifying the size, and sampler and event must be objects of that
1523type.
1524
1525Setting an argument for a kernel does NOT keep a reference to the object -
1526for example, if you set an argument to some image object, free the image,
1527and call the kernel, you will run into undefined behaviour.
1528
1529L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetKernelArg.html>
1530
1531=back
1532
1533=head2 THE OpenCL::Event CLASS
1534
1535This is the superclass for all event objects (including OpenCL::UserEvent
1536objects).
1537
1538=over 4
1539
1540=item $ev->wait
1541
1542Waits for the event to complete.
1543
1544L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clWaitForEvents.html>
1545
1546=item $packed_value = $ev->info ($name)
1547
1548See C<< $platform->info >> for details.
1549
1550L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetEventInfo.html>
1551
1552=for gengetinfo begin event
1553
1554=item $queue = $event->command_queue
1555
1556Calls C<clGetEventInfo> with C<CL_EVENT_COMMAND_QUEUE> and returns the result.
1557
1558=item $command_type = $event->command_type
1559
1560Calls C<clGetEventInfo> with C<CL_EVENT_COMMAND_TYPE> and returns the result.
1561
1562=item $uint = $event->reference_count
1563
1564Calls C<clGetEventInfo> with C<CL_EVENT_REFERENCE_COUNT> and returns the result.
1565
1566=item $uint = $event->command_execution_status
1567
1568Calls C<clGetEventInfo> with C<CL_EVENT_COMMAND_EXECUTION_STATUS> and returns the result.
1569
1570=item $ctx = $event->context
1571
1572Calls C<clGetEventInfo> with C<CL_EVENT_CONTEXT> and returns the result.
1573
1574=for gengetinfo end event
1575
1576=item $packed_value = $ev->profiling_info ($name)
1577
1578See C<< $platform->info >> for details.
1579
1580The reason this method is not called C<info> is that there already is an
1581C<< ->info >> method.
1582
1583L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetProfilingInfo.html>
1584
1585=for gengetinfo begin profiling
1586
1587=item $ulong = $event->profiling_command_queued
1588
1589Calls C<clGetEventProfilingInfo> with C<CL_PROFILING_COMMAND_QUEUED> and returns the result.
1590
1591=item $ulong = $event->profiling_command_submit
1592
1593Calls C<clGetEventProfilingInfo> with C<CL_PROFILING_COMMAND_SUBMIT> and returns the result.
1594
1595=item $ulong = $event->profiling_command_start
1596
1597Calls C<clGetEventProfilingInfo> with C<CL_PROFILING_COMMAND_START> and returns the result.
1598
1599=item $ulong = $event->profiling_command_end
1600
1601Calls C<clGetEventProfilingInfo> with C<CL_PROFILING_COMMAND_END> and returns the result.
1602
1603=for gengetinfo end profiling
1604
1605=back
1606
1607=head2 THE OpenCL::UserEvent CLASS
1608
1609This is a subclass of OpenCL::Event.
1610
1611=over 4
1612
1613=item $ev->set_status ($execution_status)
1614
1615L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetUserEventStatus.html>
1616
1617=back
1618
1619=cut
1620
1281; 16211;
129
130=back
131 1622
132=head1 AUTHOR 1623=head1 AUTHOR
133 1624
134 Marc Lehmann <schmorp@schmorp.de> 1625 Marc Lehmann <schmorp@schmorp.de>
135 http://home.schmorp.de/ 1626 http://home.schmorp.de/

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines