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

Comparing OpenCL/OpenCL.pm (file contents):
Revision 1.6 by root, Wed Nov 16 00:36:40 2011 UTC vs.
Revision 1.79 by root, Sat May 5 13:30:07 2012 UTC

6 6
7 use OpenCL; 7 use OpenCL;
8 8
9=head1 DESCRIPTION 9=head1 DESCRIPTION
10 10
11This is an early release which might be useful, but hasn't seen any testing. 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 develop this module (1.2 spec was available, but 48The OpenCL specs used to develop this module - download these and keept
16no implementation was available to me :). 49hema round, they are required reference material:
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
52 http://www.khronos.org/registry/cl/specs/opencl-1.2.pdf
53 http://www.khronos.org/registry/cl/specs/opencl-1.2-extensions.pdf
19 54
20OpenCL manpages: 55OpenCL manpages:
21 56
22 http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/ 57 http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/
58 http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/
59
60If you are into UML class diagrams, the following diagram might help - if
61not, it will be mildly confusing (also, the class hierarchy of this module
62is much more fine-grained):
63
64 http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/classDiagram.html
65
66Here's a tutorial from AMD (very AMD-centric, too), not sure how useful it
67is, but at least it's free of charge:
68
69 http://developer.amd.com/zones/OpenCLZone/courses/Documents/Introduction_to_OpenCL_Programming%20Training_Guide%20%28201005%29.pdf
70
71And here's NVIDIA's OpenCL Best Practises Guide:
72
73 http://developer.download.nvidia.com/compute/cuda/3_2/toolkit/docs/OpenCL_Best_Practices_Guide.pdf
74
75=head1 BASIC WORKFLOW
76
77To get something done, you basically have to do this once (refer to the
78examples below for actual code, this is just a high-level description):
79
80Find some platform (e.g. the first one) and some device(s) (e.g. the first
81device of the platform), and create a context from those.
82
83Create program objects from your OpenCL source code, then build (compile)
84the programs for each device you want to run them on.
85
86Create kernel objects for all kernels you want to use (surprisingly, these
87are not device-specific).
88
89Then, to execute stuff, you repeat these steps, possibly resuing or
90sharing some buffers:
91
92Create some input and output buffers from your context. Set these as
93arguments to your kernel.
94
95Enqueue buffer writes to initialise your input buffers (when not
96initialised at creation time).
97
98Enqueue the kernel execution.
99
100Enqueue buffer reads for your output buffer to read results.
101
102=head1 OPENCL 1.1 VS. OPENCL 1.2
103
104This module supports both OpenCL version 1.1 and 1.2, although the OpenCL
1051.2 interface hasn't been tested much for lack of availability of an
106actual implementation.
107
108Every function or method in this manual page that interfaces to a
109particular OpenCL function has a link to the its C manual page.
110
111If the link contains a F<1.1>, then this function is an OpenCL 1.1
112function. Most but not all also exist in OpenCL 1.2, and this module
113tries to emulate the missing ones for you, when told to do so at
114compiletime. You cna check whether a function was removed in OpenCL 1.2 by
115replacing the F<1.1> component in the URL by F<1.2>.
116
117If the link contains a F<1.2>, then this is a OpenCL 1.2-only
118function. Even if the module was compiled with OpenCL 1.2 header files
119and has an 1.2 OpenCL library, calling such a function on a platform that
120doesn't implement 1.2 causes undefined behaviour, usually a crash (But
121this is not guaranteed).
122
123You can find out whether this module was compiled to prefer 1.1
124functionality by ooking at C<OpenCL::PREFER_1_1> - if it is true, then
1251.1 functions generally are implemented using 1.1 OpenCL functions. If it
126is false, then 1.1 functions missing from 1.2 are emulated by calling 1.2
127fucntions.
128
129This is a somewhat sorry state of affairs, but the Khronos group choose to
130make every release of OpenCL source and binary incompatible with previous
131releases.
23 132
24=head1 EXAMPLES 133=head1 EXAMPLES
25 134
26=head2 Enumerate all devices and get contexts for them. 135=head2 Enumerate all devices and get contexts for them.
27 136
137Best run this once to get a feel for the platforms and devices in your
138system.
139
28 for my $platform (OpenCL::platforms) { 140 for my $platform (OpenCL::platforms) {
29 warn $platform->info (OpenCL::PLATFORM_NAME); 141 printf "platform: %s\n", $platform->name;
30 warn $platform->info (OpenCL::PLATFORM_EXTENSIONS); 142 printf "extensions: %s\n", $platform->extensions;
31 for my $device ($platform->devices) { 143 for my $device ($platform->devices) {
32 warn $device->info (OpenCL::DEVICE_NAME); 144 printf "+ device: %s\n", $device->name;
33 my $ctx = $device->context_simple; 145 my $ctx = $platform->context (undef, [$device]);
34 # do stuff 146 # do stuff
35 } 147 }
36 } 148 }
37 149
38=head2 Get a useful context and a command queue. 150=head2 Get a useful context and a command queue.
39 151
40 my $dev = ((OpenCL::platforms)[0]->devices)[0]; 152This is a useful boilerplate for any OpenCL program that only wants to use
41 my $ctx = $dev->context_simple; 153one device,
42 my $queue = $ctx->command_queue_simple ($dev); 154
155 my ($platform) = OpenCL::platforms; # find first platform
156 my ($dev) = $platform->devices; # find first device of platform
157 my $ctx = $platform->context (undef, [$dev]); # create context out of those
158 my $queue = $ctx->queue ($dev); # create a command queue for the device
43 159
44=head2 Print all supported image formats of a context. 160=head2 Print all supported image formats of a context.
45 161
162Best run this once for your context, to see whats available and how to
163gather information.
164
46 for my $type (OpenCL::MEM_OBJECT_IMAGE2D, OpenCL::MEM_OBJECT_IMAGE3D) { 165 for my $type (OpenCL::MEM_OBJECT_IMAGE2D, OpenCL::MEM_OBJECT_IMAGE3D) {
47 say "supported image formats for ", OpenCL::enum2str $type; 166 print "supported image formats for ", OpenCL::enum2str $type, "\n";
48 167
49 for my $f ($ctx->supported_image_formats (0, $type)) { 168 for my $f ($ctx->supported_image_formats (0, $type)) {
50 printf " %-10s %-20s\n", OpenCL::enum2str $f->[0], OpenCL::enum2str $f->[1]; 169 printf " %-10s %-20s\n", OpenCL::enum2str $f->[0], OpenCL::enum2str $f->[1];
51 } 170 }
52 } 171 }
54=head2 Create a buffer with some predefined data, read it back synchronously, 173=head2 Create a buffer with some predefined data, read it back synchronously,
55then asynchronously. 174then asynchronously.
56 175
57 my $buf = $ctx->buffer_sv (OpenCL::MEM_COPY_HOST_PTR, "helmut"); 176 my $buf = $ctx->buffer_sv (OpenCL::MEM_COPY_HOST_PTR, "helmut");
58 177
59 $queue->enqueue_read_buffer ($buf, 1, 1, 3, my $data); 178 $queue->read_buffer ($buf, 1, 1, 3, my $data);
60 warn $data; 179 print "$data\n";
61 180
62 my $ev = $queue->enqueue_read_buffer ($buf, 0, 1, 3, my $data); 181 my $ev = $queue->read_buffer ($buf, 0, 1, 3, my $data);
63 $ev->wait; 182 $ev->wait;
64 warn $data; 183 print "$data\n"; # prints "elm"
65 184
66=head2 Create and build a program, then create a kernel out of one of its 185=head2 Create and build a program, then create a kernel out of one of its
67functions. 186functions.
68 187
69 my $src = ' 188 my $src = '
70 __kernel void 189 kernel void
71 squareit (__global float *input, __global float *output) 190 squareit (global float *input, global float *output)
72 { 191 {
73 size_t id = get_global_id (0); 192 $id = get_global_id (0);
74 output [id] = input [id] * input [id]; 193 output [id] = input [id] * input [id];
75 } 194 }
76 '; 195 ';
77 196
78 my $prog = $ctx->program_with_source ($src); 197 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"); 198 my $kernel = $prog->kernel ("squareit");
84 199
85=head2 Create some input and output float buffers, then call squareit on them. 200=head2 Create some input and output float buffers, then call the
201'squareit' kernel on them.
86 202
87 my $input = $ctx->buffer_sv (OpenCL::MEM_COPY_HOST_PTR, pack "f*", 1, 2, 3, 4.5); 203 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); 204 my $output = $ctx->buffer (0, OpenCL::SIZEOF_FLOAT * 5);
89 205
90 # set buffer 206 # set buffer
91 $kernel->set_buffer (0, $input); 207 $kernel->set_buffer (0, $input);
92 $kernel->set_buffer (1, $output); 208 $kernel->set_buffer (1, $output);
93 209
94 # execute it for all 4 numbers 210 # execute it for all 4 numbers
95 $queue->enqueue_nd_range_kernel ($kernel, undef, [4], undef); 211 $queue->nd_range_kernel ($kernel, undef, [4], undef);
96 212
97 # enqueue a synchronous read 213 # enqueue a synchronous read
98 $queue->enqueue_read_buffer ($output, 1, 0, OpenCL::SIZEOF_FLOAT * 4, my $data); 214 $queue->read_buffer ($output, 1, 0, OpenCL::SIZEOF_FLOAT * 4, my $data);
99 215
100 # print the results: 216 # print the results:
101 say join ", ", unpack "f*", $data; 217 printf "%s\n", join ", ", unpack "f*", $data;
102 218
103=head2 The same enqueue operations as before, but assuming an out-of-order queue, 219=head2 The same enqueue operations as before, but assuming an out-of-order queue,
104showing off barriers. 220showing off barriers.
105 221
106 # execute it for all 4 numbers 222 # execute it for all 4 numbers
107 $queue->enqueue_nd_range_kernel ($kernel, undef, [4], undef); 223 $queue->nd_range_kernel ($kernel, undef, [4], undef);
108 224
109 # enqueue a barrier to ensure in-order execution 225 # enqueue a barrier to ensure in-order execution
110 $queue->enqueue_barrier; 226 $queue->barrier;
111 227
112 # enqueue an async read 228 # enqueue an async read
113 $queue->enqueue_read_buffer ($output, 0, 0, OpenCL::SIZEOF_FLOAT * 4, my $data); 229 $queue->read_buffer ($output, 0, 0, OpenCL::SIZEOF_FLOAT * 4, my $data);
114 230
115 # wait for all requests to finish 231 # wait for all requests to finish
116 $queue->finish; 232 $queue->finish;
117 233
118=head2 The same enqueue operations as before, but assuming an out-of-order queue, 234=head2 The same enqueue operations as before, but assuming an out-of-order queue,
119showing off event objects and wait lists. 235showing off event objects and wait lists.
120 236
121 # execute it for all 4 numbers 237 # execute it for all 4 numbers
122 my $ev = $queue->enqueue_nd_range_kernel ($kernel, undef, [4], undef); 238 my $ev = $queue->nd_range_kernel ($kernel, undef, [4], undef);
123 239
124 # enqueue an async read 240 # enqueue an async read
125 $ev = $queue->enqueue_read_buffer ($output, 0, 0, OpenCL::SIZEOF_FLOAT * 4, my $data, $ev); 241 $ev = $queue->read_buffer ($output, 0, 0, OpenCL::SIZEOF_FLOAT * 4, my $data, $ev);
126 242
127 # wait for the last event to complete 243 # wait for the last event to complete
128 $ev->wait; 244 $ev->wait;
129 245
246=head2 Use the OpenGL module to share a texture between OpenCL and OpenGL and draw some julia
247set flight effect.
248
249This is quite a long example to get you going - you can download it from
250L<http://cvs.schmorp.de/OpenCL/examples/juliaflight>.
251
252 use OpenGL ":all";
253 use OpenCL;
254
255 my $S = $ARGV[0] || 256; # window/texture size, smaller is faster
256
257 # open a window and create a gl texture
258 OpenGL::glpOpenWindow width => $S, height => $S;
259 my $texid = glGenTextures_p 1;
260 glBindTexture GL_TEXTURE_2D, $texid;
261 glTexImage2D_c GL_TEXTURE_2D, 0, GL_RGBA8, $S, $S, 0, GL_RGBA, GL_UNSIGNED_BYTE, 0;
262
263 # find and use the first opencl device that let's us get a shared opengl context
264 my $platform;
265 my $dev;
266 my $ctx;
267
268 for (OpenCL::platforms) {
269 $platform = $_;
270 for ($platform->devices) {
271 $dev = $_;
272 $ctx = $platform->context ([OpenCL::GLX_DISPLAY_KHR, undef, OpenCL::GL_CONTEXT_KHR, undef], [$dev])
273 and last;
274 }
275 }
276
277 $ctx
278 or die "cannot find suitable OpenCL device\n";
279
280 my $queue = $ctx->queue ($dev);
281
282 # now attach an opencl image2d object to the opengl texture
283 my $tex = $ctx->gl_texture2d (OpenCL::MEM_WRITE_ONLY, GL_TEXTURE_2D, 0, $texid);
284
285 # now the boring opencl code
286 my $src = <<EOF;
287 kernel void
288 juliatunnel (write_only image2d_t img, float time)
289 {
290 int2 xy = (int2)(get_global_id (0), get_global_id (1));
291 float2 p = convert_float2 (xy) / $S.f * 2.f - 1.f;
292
293 float2 m = (float2)(1.f, p.y) / fabs (p.x); // tunnel
294 m.x = fabs (fmod (m.x + time * 0.05f, 4.f) - 2.f);
295
296 float2 z = m;
297 float2 c = (float2)(sin (time * 0.01133f), cos (time * 0.02521f));
298
299 for (int i = 0; i < 25 && dot (z, z) < 4.f; ++i) // standard julia
300 z = (float2)(z.x * z.x - z.y * z.y, 2.f * z.x * z.y) + c;
301
302 float3 colour = (float3)(z.x, z.y, atan2 (z.y, z.x));
303 write_imagef (img, xy, (float4)(colour * p.x * p.x, 1.));
304 }
305 EOF
306
307 my $prog = $ctx->build_program ($src);
308 my $kernel = $prog->kernel ("juliatunnel");
309
310 # program compiled, kernel ready, now draw and loop
311
312 for (my $time; ; ++$time) {
313 # acquire objects from opengl
314 $queue->acquire_gl_objects ([$tex]);
315
316 # configure and run our kernel
317 $kernel->setf ("mf", $tex, $time*2); # mf = memory object, float
318 $queue->nd_range_kernel ($kernel, undef, [$S, $S], undef);
319
320 # release objects to opengl again
321 $queue->release_gl_objects ([$tex]);
322
323 # wait
324 $queue->finish;
325
326 # now draw the texture, the defaults should be all right
327 glTexParameterf GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST;
328
329 glEnable GL_TEXTURE_2D;
330 glBegin GL_QUADS;
331 glTexCoord2f 0, 1; glVertex3i -1, -1, -1;
332 glTexCoord2f 0, 0; glVertex3i 1, -1, -1;
333 glTexCoord2f 1, 0; glVertex3i 1, 1, -1;
334 glTexCoord2f 1, 1; glVertex3i -1, 1, -1;
335 glEnd;
336
337 glXSwapBuffers;
338
339 select undef, undef, undef, 1/60;
340 }
341
342=head2 How to modify the previous example to not rely on GL sharing.
343
344For those poor souls with only a sucky CPU OpenCL implementation, you
345currently have to read the image into some perl scalar, and then modify a
346texture or use glDrawPixels or so).
347
348First, when you don't need gl sharing, you can create the context much simpler:
349
350 $ctx = $platform->context (undef, [$dev])
351
352To use a texture, you would modify the above example by creating an
353OpenCL::Image manually instead of deriving it from a texture:
354
355 my $tex = $ctx->image2d (OpenCL::MEM_WRITE_ONLY, OpenCL::RGBA, OpenCL::UNORM_INT8, $S, $S);
356
357And in the draw loop, intead of acquire_gl_objects/release_gl_objects, you
358would read the image2d after the kernel has written it:
359
360 $queue->read_image ($tex, 0, 0, 0, 0, $S, $S, 1, 0, 0, my $data);
361
362And then you would upload the pixel data to the texture (or use glDrawPixels):
363
364 glTexSubImage2D_s GL_TEXTURE_2D, 0, 0, 0, $S, $S, GL_RGBA, GL_UNSIGNED_BYTE, $data;
365
366The fully modified example can be found at
367L<http://cvs.schmorp.de/OpenCL/examples/juliaflight-nosharing>.
368
369=head2 Julia sets look soooo 80ies.
370
371Then colour them differently, e.g. using orbit traps! Replace the loop and
372colour calculation from the previous examples by this:
373
374 float2 dm = (float2)(1.f, 1.f);
375
376 for (int i = 0; i < 25; ++i)
377 {
378 z = (float2)(z.x * z.x - z.y * z.y, 2.f * z.x * z.y) + c;
379 dm = fmin (dm, (float2)(fabs (dot (z, z) - 1.f), fabs (z.x - 1.f)));
380 }
381
382 float3 colour = (float3)(dm.x * dm.y, dm.x * dm.y, dm.x);
383
384Also try C<-10.f> instead of C<-1.f>.
385
130=head1 DOCUMENTATION 386=head1 DOCUMENTATION
131 387
132=head2 BASIC CONVENTIONS 388=head2 BASIC CONVENTIONS
133 389
134This is not a 1:1 C-style translation of OpenCL to Perl - instead I 390This is not a one-to-one C-style translation of OpenCL to Perl - instead
135attempted to make the interface as type-safe as possible and introducing 391I attempted to make the interface as type-safe as possible by introducing
136object syntax where it makes sense. There are a number of important 392object syntax where it makes sense. There are a number of important
137differences between the OpenCL C API and this module: 393differences between the OpenCL C API and this module:
138 394
139=over 4 395=over 4
140 396
141=item * Object lifetime managament is automatic - there is no need 397=item * Object lifetime managament is automatic - there is no need
142to free objects explicitly (C<clReleaseXXX>), the release function 398to free objects explicitly (C<clReleaseXXX>), the release function
143is called automatically once all Perl references to it go away. 399is called automatically once all Perl references to it go away.
144 400
145=item * OpenCL uses CamelCase for function names (C<clGetPlatformInfo>), 401=item * OpenCL uses CamelCase for function names
402(e.g. C<clGetPlatformIDs>, C<clGetPlatformInfo>), while this module
146while this module uses underscores as word separator and often leaves out 403uses underscores as word separator and often leaves out prefixes
147prefixes (C<< $platform->info >>). 404(C<OpenCL::platforms>, C<< $platform->info >>).
148 405
149=item * OpenCL often specifies fixed vector function arguments as short 406=item * OpenCL often specifies fixed vector function arguments as short
150arrays (C<size_t origin[3]>), while this module explicitly expects the 407arrays (C<size_t origin[3]>), while this module explicitly expects the
151components as separate arguments- 408components as separate arguments (C<$orig_x, $orig_y, $orig_z>) in
409function calls.
152 410
153=item * Where possible, the row_pitch value is calculated from the perl 411=item * Structures are often specified by flattening out their components
154scalar length and need not be specified. 412as with short vectors, and returned as arrayrefs.
155 413
156=item * When enqueuing commands, the wait list is specified by adding 414=item * When enqueuing commands, the wait list is specified by adding
157extra arguments to the function - everywhere a C<$wait_events...> argument 415extra arguments to the function - anywhere a C<$wait_events...> argument
158is documented this can be any number of event objects. 416is documented this can be any number of event objects. As an extsnion
417implemented by this module, C<undef> values will be ignored in the event
418list.
159 419
160=item * When enqueuing commands, if the enqueue method is called in void 420=item * When enqueuing commands, if the enqueue method is called in void
161context, no event is created. In all other contexts an event is returned 421context, no event is created. In all other contexts an event is returned
162by the method. 422by the method.
163 423
164=item * This module expects all functions to return C<CL_SUCCESS>. If any 424=item * This module expects all functions to return C<OpenCL::SUCCESS>. If any
165other status is returned the function will throw an exception, so you 425other status is returned the function will throw an exception, so you
166don't normally have to to any error checking. 426don't normally have to to any error checking.
167 427
168=back 428=back
169 429
430=head2 PERL AND OPENCL TYPES
431
432This handy(?) table lists OpenCL types and their perl, PDL and pack/unpack
433format equivalents:
434
435 OpenCL perl PDL pack/unpack
436 char IV - c
437 uchar IV byte C
438 short IV short s
439 ushort IV ushort S
440 int IV long? l
441 uint IV - L
442 long IV longlong q
443 ulong IV - Q
444 float NV float f
445 half IV ushort S
446 double NV double d
447
448=head2 GLX SUPPORT
449
450Due to the sad state that OpenGL support is in in Perl (mostly the OpenGL
451module, which has little to no documentation and has little to no support
452for glX), this module, as a special extension, treats context creation
453properties C<OpenCL::GLX_DISPLAY_KHR> and C<OpenCL::GL_CONTEXT_KHR>
454specially: If either or both of these are C<undef>, then the OpenCL
455module tries to dynamically resolve C<glXGetCurrentDisplay> and
456C<glXGetCurrentContext>, call these functions and use their return values
457instead.
458
459For this to work, the OpenGL library must be loaded, a GLX context must
460have been created and be made current, and C<dlsym> must be available and
461capable of finding the function via C<RTLD_DEFAULT>.
462
463=head2 EVENT SYSTEM
464
465OpenCL can generate a number of (potentially) asynchronous events, for
466example, after compiling a program, to signal a context-related error or,
467perhaps most important, to signal completion of queued jobs (by setting
468callbacks on OpenCL::Event objects).
469
470The OpenCL module converts all these callbacks into events - you can
471still register callbacks, but they are not executed when your OpenCL
472implementation calls the actual callback, but only later. Therefore, none
473of the limitations of OpenCL callbacks apply to the perl implementation:
474it is perfectly safe to make blocking operations from event callbacks, and
475enqueued operations don't need to be flushed.
476
477To facilitate this, this module maintains an event queue - each
478time an asynchronous event happens, it is queued, and perl will be
479interrupted. This is implemented via the L<Async::Interrupt> module. In
480addition, this module has L<AnyEvent> support, so it can seamlessly
481integrate itself into many event loops.
482
483Since L<Async::Interrupt> is a bit hard to understand, here are some case examples:
484
485=head3 Don't use callbacks.
486
487When your program never uses any callbacks, then there will never be any
488notifications you need to take care of, and therefore no need to worry
489about all this.
490
491You can achieve a great deal by explicitly waiting for events, or using
492barriers and flush calls. In many programs, there is no need at all to
493tinker with asynchronous events.
494
495=head3 Use AnyEvent
496
497This module automatically registers a watcher that invokes all outstanding
498event callbacks when AnyEvent is initialised (and block asynchronous
499interruptions). Using this mode of operations is the safest and most
500recommended one.
501
502To use this, simply use AnyEvent and this module normally, make sure you
503have an event loop running:
504
505 use Gtk2 -init;
506 use AnyEvent;
507
508 # initialise AnyEvent, by creating a watcher, or:
509 AnyEvent::detect;
510
511 my $e = $queue->marker;
512 $e->cb (sub {
513 warn "opencl is finished\n";
514 })
515
516 main Gtk2;
517
518Note that this module will not initialise AnyEvent for you. Before
519AnyEvent is initialised, the module will asynchronously interrupt perl
520instead. To avoid any surprises, it's best to explicitly initialise
521AnyEvent.
522
523You can temporarily enable asynchronous interruptions (see next paragraph)
524by calling C<$OpenCL::INTERRUPT->unblock> and disable them again by
525calling C<$OpenCL::INTERRUPT->block>.
526
527=head3 Let yourself be interrupted at any time
528
529This mode is the default unless AnyEvent is loaded and initialised. In
530this mode, OpenCL asynchronously interrupts a running perl program. The
531emphasis is on both I<asynchronously> and I<running> here.
532
533Asynchronously means that perl might execute your callbacks at any
534time. For example, in the following code (I<THAT YOU SHOULD NOT COPY>),
535the C<until> loop following the marker call will be interrupted by the
536callback:
537
538 my $e = $queue->marker;
539 my $flag;
540 $e->cb (sub { $flag = 1 });
541 1 until $flag;
542 # $flag is now 1
543
544The reason why you shouldn't blindly copy the above code is that
545busy waiting is a really really bad thing, and really really bad for
546performance.
547
548While at first this asynchronous business might look exciting, it can be
549really hard, because you need to be prepared for the callback code to be
550executed at any time, which limits the amount of things the callback code
551can do safely.
552
553This can be mitigated somewhat by using C<<
554$OpenCL::INTERRUPT->scope_block >> (see the L<Async::Interrupt>
555documentation for details).
556
557The other problem is that your program must be actively I<running> to be
558interrupted. When you calculate stuff, your program is running. When you
559hang in some C functions or other block execution (by calling C<sleep>,
560C<select>, running an event loop and so on), your program is waiting, not
561running.
562
563One way around that would be to attach a read watcher to your event loop,
564listening for events on C<< $OpenCL::INTERRUPT->pipe_fileno >>, using a
565dummy callback (C<sub { }>) to temporarily execute some perl code.
566
567That is then awfully close to using the built-in AnyEvent support above,
568though, so consider that one instead.
569
570=head3 Be creative
571
572OpenCL exports the L<Async::Interrupt> object it uses in the global
573variable C<$OpenCL::INTERRUPT>. You can configure it in any way you like.
574
575So if you want to feel like a real pro, err, wait, if you feel no risk
576menas no fun, you can experiment by implementing your own mode of
577operations.
578
579=cut
580
581package OpenCL;
582
583use common::sense;
584use Carp ();
585use Async::Interrupt ();
586
587our $POLL_FUNC; # set by XS
588
589BEGIN {
590 our $VERSION = '0.99';
591
592 require XSLoader;
593 XSLoader::load (__PACKAGE__, $VERSION);
594
595 @OpenCL::Platform::ISA =
596 @OpenCL::Device::ISA =
597 @OpenCL::Context::ISA =
598 @OpenCL::Queue::ISA =
599 @OpenCL::Memory::ISA =
600 @OpenCL::Sampler::ISA =
601 @OpenCL::Program::ISA =
602 @OpenCL::Kernel::ISA =
603 @OpenCL::Event::ISA = OpenCL::Object::;
604
605 @OpenCL::SubDevice::ISA = OpenCL::Device::;
606
607 @OpenCL::Buffer::ISA =
608 @OpenCL::Image::ISA = OpenCL::Memory::;
609
610 @OpenCL::BufferObj::ISA = OpenCL::Buffer::;
611
612 @OpenCL::Image2D::ISA =
613 @OpenCL::Image3D::ISA =
614 @OpenCL::Image2DArray::ISA =
615 @OpenCL::Image1D::ISA =
616 @OpenCL::Image1DArray::ISA =
617 @OpenCL::Image1DBuffer::ISA = OpenCL::Image::;
618
619 @OpenCL::UserEvent::ISA = OpenCL::Event::;
620
621 @OpenCL::MappedBuffer::ISA =
622 @OpenCL::MappedImage::ISA = OpenCL::Mapped::;
623}
624
170=head2 THE OpenCL PACKAGE 625=head2 THE OpenCL PACKAGE
171 626
172=over 4 627=over 4
173 628
174=item $int = OpenCL::errno 629=item $int = OpenCL::errno
175 630
176The last error returned by a function - it's only changed on errors. 631The last error returned by a function - it's only valid after an error occured
632and before calling another OpenCL function.
177 633
178=item $str = OpenCL::err2str $errval 634=item $str = OpenCL::err2str [$errval]
179 635
180Comverts an error value into a human readable string. 636Converts an error value into a human readable string. IF no error value is
637given, then the last error will be used (as returned by OpenCL::errno).
181 638
182=item $str = OpenCL::err2str $enum 639=item $str = OpenCL::enum2str $enum
183 640
184Converts most enum values (inof parameter names, image format constants, 641Converts most enum values (of parameter names, image format constants,
185object types, addressing and filter modes, command types etc.) into a 642object types, addressing and filter modes, command types etc.) into a
186human readbale string. When confronted with some random integer it can be 643human readable string. When confronted with some random integer it can be
187very helpful to pass it through this function to maybe get some readable 644very helpful to pass it through this function to maybe get some readable
188string out of it. 645string out of it.
189 646
190=item @platforms = OpenCL::platforms 647=item @platforms = OpenCL::platforms
191 648
192Returns all available OpenCL::Platform objects. 649Returns all available OpenCL::Platform objects.
193 650
194L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetPlatformIDs.html> 651L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetPlatformIDs.html>
195 652
196=item $ctx = OpenCL::context_from_type_simple $type = OpenCL::DEVICE_TYPE_DEFAULT 653=item $ctx = OpenCL::context_from_type $properties, $type = OpenCL::DEVICE_TYPE_DEFAULT, $callback->($err, $pvt) = $print_stderr
197 654
198Tries to create a context from a default device and platform - never worked for me. 655Tries to create a context from a default device and platform type - never worked for me.
656Consider using C<< $platform->context_from_type >> instead.
657
658type: OpenCL::DEVICE_TYPE_DEFAULT, OpenCL::DEVICE_TYPE_CPU, OpenCL::DEVICE_TYPE_GPU,
659OpenCL::DEVICE_TYPE_ACCELERATOR, OpenCL::DEVICE_TYPE_CUSTOM, OpenCL::DEVICE_TYPE_ALL.
199 660
200L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html> 661L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html>
201 662
663=item $ctx = OpenCL::context $properties, \@devices, $callback->($err, $pvt) = $print_stderr)
664
665Create a new OpenCL::Context object using the given device object(s).
666Consider using C<< $platform->context >> instead.
667
668L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContext.html>
669
202=item OpenCL::wait_for_events $wait_events... 670=item OpenCL::wait_for_events $wait_events...
203 671
204Waits for all events to complete. 672Waits for all events to complete.
205 673
206L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clWaitForEvents.html> 674L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clWaitForEvents.html>
207 675
676=item OpenCL::poll
677
678Checks if there are any outstanding events (see L<EVENT SYSTEM>) and
679invokes their callbacks.
680
681=item $OpenCL::INTERRUPT
682
683The L<Async::Interrupt> object used to signal asynchronous events (see
684L<EVENT SYSTEM>).
685
686=cut
687
688our $INTERRUPT = new Async::Interrupt c_cb => [$POLL_FUNC, 0];
689
690&_eq_initialise ($INTERRUPT->signal_func);
691
692=item $OpenCL::WATCHER
693
694The L<AnyEvent> watcher object used to watch for asynchronous events (see
695L<EVENT SYSTEM>). This variable is C<undef> until L<AnyEvent> has been
696loaded I<and> initialised (e.g. by calling C<AnyEvent::detect>).
697
698=cut
699
700our $WATCHER;
701
702sub _init_anyevent {
703 $INTERRUPT->block;
704 $WATCHER = AE::io ($INTERRUPT->pipe_fileno, 0, sub { $INTERRUPT->handle });
705}
706
707if (defined $AnyEvent::MODEL) {
708 _init_anyevent;
709} else {
710 push @AnyEvent::post_detect, \&_init_anyevent;
711}
712
208=back 713=back
209 714
715=head2 THE OpenCL::Object CLASS
716
717This is the base class for all objects in the OpenCL module. The only
718method it implements is the C<id> method, which is only useful if you want
719to interface to OpenCL on the C level.
720
721=over 4
722
723=item $iv = $obj->id
724
725OpenCL objects are represented by pointers or integers on the C level. If
726you want to interface to an OpenCL object directly on the C level, then
727you need this value, which is returned by this method. You should use an
728C<IV> type in your code and cast that to the correct type.
729
730=cut
731
732sub OpenCL::Object::id {
733 ref $_[0] eq "SCALAR"
734 ? ${ $_[0] }
735 : $_[0][0]
736}
737
738=back
739
210=head2 THE OpenCL::Platform CLASS 740=head2 THE OpenCL::Platform CLASS
211 741
212=over 4 742=over 4
213 743
744=item @devices = $platform->devices ($type = OpenCL::DEVICE_TYPE_ALL)
745
746Returns a list of matching OpenCL::Device objects.
747
748=item $ctx = $platform->context_from_type ($properties, $type = OpenCL::DEVICE_TYPE_DEFAULT, $callback->($err, $pvt) = $print_stderr)
749
750Tries to create a context. Never worked for me, and you need devices explicitly anyway.
751
752L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html>
753
754=item $ctx = $platform->context ($properties, \@devices, $callback->($err, $pvt) = $print_stderr)
755
756Create a new OpenCL::Context object using the given device object(s)- a
757OpenCL::CONTEXT_PLATFORM property is supplied automatically.
758
759L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContext.html>
760
214=item $packed_value = $platform->info ($name) 761=item $packed_value = $platform->info ($name)
215 762
216Calls C<clGetPlatformInfo> and returns the packed, raw value - for 763Calls C<clGetPlatformInfo> and returns the packed, raw value - for
217strings, this will be the string, for other values you probably need to 764strings, this will be the string (possibly including terminating \0), for
218use the correct C<unpack>. This might get improved in the future. Hopefully. 765other values you probably need to use the correct C<unpack>.
766
767It's best to avoid this method and use one of the following convenience
768wrappers.
219 769
220L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetPlatformInfo.html> 770L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetPlatformInfo.html>
221 771
222=item @devices = $platform->devices ($type = OpenCL::DEVICE_TYPE_ALL) 772=item $platform->unload_compiler
223 773
224Returns a list of matching OpenCL::Device objects. 774Attempts to unload the compiler for this platform, for endless
775profit. Does nothing on OpenCL 1.1.
225 776
226=item $ctx = $platform->context_from_type_simple ($type = OpenCL::DEVICE_TYPE_DEFAULT)
227
228Tries to create a context. Never worked for me.
229
230L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContextFromType.html> 777L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clUnloadPlatformCompiler.html>
778
779=for gengetinfo begin platform
780
781=item $string = $platform->profile
782
783Calls C<clGetPlatformInfo> with C<OpenCL::PLATFORM_PROFILE> and returns the result.
784
785=item $string = $platform->version
786
787Calls C<clGetPlatformInfo> with C<OpenCL::PLATFORM_VERSION> and returns the result.
788
789=item $string = $platform->name
790
791Calls C<clGetPlatformInfo> with C<OpenCL::PLATFORM_NAME> and returns the result.
792
793=item $string = $platform->vendor
794
795Calls C<clGetPlatformInfo> with C<OpenCL::PLATFORM_VENDOR> and returns the result.
796
797=item $string = $platform->extensions
798
799Calls C<clGetPlatformInfo> with C<OpenCL::PLATFORM_EXTENSIONS> and returns the result.
800
801=for gengetinfo end platform
231 802
232=back 803=back
233 804
234=head2 THE OpenCL::Device CLASS 805=head2 THE OpenCL::Device CLASS
235 806
237 808
238=item $packed_value = $device->info ($name) 809=item $packed_value = $device->info ($name)
239 810
240See C<< $platform->info >> for details. 811See C<< $platform->info >> for details.
241 812
813type: OpenCL::DEVICE_TYPE_DEFAULT, OpenCL::DEVICE_TYPE_CPU,
814OpenCL::DEVICE_TYPE_GPU, OpenCL::DEVICE_TYPE_ACCELERATOR,
815OpenCL::DEVICE_TYPE_CUSTOM, OpenCL::DEVICE_TYPE_ALL.
816
817fp_config: OpenCL::FP_DENORM, OpenCL::FP_INF_NAN, OpenCL::FP_ROUND_TO_NEAREST,
818OpenCL::FP_ROUND_TO_ZERO, OpenCL::FP_ROUND_TO_INF, OpenCL::FP_FMA,
819OpenCL::FP_SOFT_FLOAT, OpenCL::FP_CORRECTLY_ROUNDED_DIVIDE_SQRT.
820
821mem_cache_type: OpenCL::NONE, OpenCL::READ_ONLY_CACHE, OpenCL::READ_WRITE_CACHE.
822
823local_mem_type: OpenCL::LOCAL, OpenCL::GLOBAL.
824
825exec_capabilities: OpenCL::EXEC_KERNEL, OpenCL::EXEC_NATIVE_KERNEL.
826
827command_queue_properties: OpenCL::QUEUE_OUT_OF_ORDER_EXEC_MODE_ENABLE,
828OpenCL::QUEUE_PROFILING_ENABLE.
829
830partition_properties: OpenCL::DEVICE_PARTITION_EQUALLY,
831OpenCL::DEVICE_PARTITION_BY_COUNTS, OpenCL::DEVICE_PARTITION_BY_COUNTS_LIST_END,
832OpenCL::DEVICE_PARTITION_BY_AFFINITY_DOMAIN.
833
834affinity_domain: OpenCL::DEVICE_AFFINITY_DOMAIN_NUMA,
835OpenCL::DEVICE_AFFINITY_DOMAIN_L4_CACHE, OpenCL::DEVICE_AFFINITY_DOMAIN_L3_CACHE,
836OpenCL::DEVICE_AFFINITY_DOMAIN_L2_CACHE, OpenCL::DEVICE_AFFINITY_DOMAIN_L1_CACHE,
837OpenCL::DEVICE_AFFINITY_DOMAIN_NEXT_PARTITIONABLE.
838
242L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetDeviceInfo.html> 839L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetDeviceInfo.html>
243 840
244=item $ctx = $device->context_simple 841=item @devices = $device->sub_devices (\@properties)
245 842
246Convenience function to create a new OpenCL::Context object. 843Creates OpencL::SubDevice objects by partitioning an existing device.
247 844
248L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateContext.html> 845L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clCreateSubDevices.html>
846
847=for gengetinfo begin device
848
849=item $device_type = $device->type
850
851Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_TYPE> and returns the result.
852
853=item $uint = $device->vendor_id
854
855Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_VENDOR_ID> and returns the result.
856
857=item $uint = $device->max_compute_units
858
859Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_MAX_COMPUTE_UNITS> and returns the result.
860
861=item $uint = $device->max_work_item_dimensions
862
863Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_MAX_WORK_ITEM_DIMENSIONS> and returns the result.
864
865=item $int = $device->max_work_group_size
866
867Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_MAX_WORK_GROUP_SIZE> and returns the result.
868
869=item @ints = $device->max_work_item_sizes
870
871Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_MAX_WORK_ITEM_SIZES> and returns the result.
872
873=item $uint = $device->preferred_vector_width_char
874
875Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_PREFERRED_VECTOR_WIDTH_CHAR> and returns the result.
876
877=item $uint = $device->preferred_vector_width_short
878
879Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_PREFERRED_VECTOR_WIDTH_SHORT> and returns the result.
880
881=item $uint = $device->preferred_vector_width_int
882
883Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_PREFERRED_VECTOR_WIDTH_INT> and returns the result.
884
885=item $uint = $device->preferred_vector_width_long
886
887Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_PREFERRED_VECTOR_WIDTH_LONG> and returns the result.
888
889=item $uint = $device->preferred_vector_width_float
890
891Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_PREFERRED_VECTOR_WIDTH_FLOAT> and returns the result.
892
893=item $uint = $device->preferred_vector_width_double
894
895Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_PREFERRED_VECTOR_WIDTH_DOUBLE> and returns the result.
896
897=item $uint = $device->max_clock_frequency
898
899Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_MAX_CLOCK_FREQUENCY> and returns the result.
900
901=item $bitfield = $device->address_bits
902
903Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_ADDRESS_BITS> and returns the result.
904
905=item $uint = $device->max_read_image_args
906
907Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_MAX_READ_IMAGE_ARGS> and returns the result.
908
909=item $uint = $device->max_write_image_args
910
911Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_MAX_WRITE_IMAGE_ARGS> and returns the result.
912
913=item $ulong = $device->max_mem_alloc_size
914
915Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_MAX_MEM_ALLOC_SIZE> and returns the result.
916
917=item $int = $device->image2d_max_width
918
919Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_IMAGE2D_MAX_WIDTH> and returns the result.
920
921=item $int = $device->image2d_max_height
922
923Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_IMAGE2D_MAX_HEIGHT> and returns the result.
924
925=item $int = $device->image3d_max_width
926
927Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_IMAGE3D_MAX_WIDTH> and returns the result.
928
929=item $int = $device->image3d_max_height
930
931Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_IMAGE3D_MAX_HEIGHT> and returns the result.
932
933=item $int = $device->image3d_max_depth
934
935Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_IMAGE3D_MAX_DEPTH> and returns the result.
936
937=item $uint = $device->image_support
938
939Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_IMAGE_SUPPORT> and returns the result.
940
941=item $int = $device->max_parameter_size
942
943Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_MAX_PARAMETER_SIZE> and returns the result.
944
945=item $uint = $device->max_samplers
946
947Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_MAX_SAMPLERS> and returns the result.
948
949=item $uint = $device->mem_base_addr_align
950
951Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_MEM_BASE_ADDR_ALIGN> and returns the result.
952
953=item $uint = $device->min_data_type_align_size
954
955Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_MIN_DATA_TYPE_ALIGN_SIZE> and returns the result.
956
957=item $device_fp_config = $device->single_fp_config
958
959Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_SINGLE_FP_CONFIG> and returns the result.
960
961=item $device_mem_cache_type = $device->global_mem_cache_type
962
963Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_GLOBAL_MEM_CACHE_TYPE> and returns the result.
964
965=item $uint = $device->global_mem_cacheline_size
966
967Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_GLOBAL_MEM_CACHELINE_SIZE> and returns the result.
968
969=item $ulong = $device->global_mem_cache_size
970
971Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_GLOBAL_MEM_CACHE_SIZE> and returns the result.
972
973=item $ulong = $device->global_mem_size
974
975Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_GLOBAL_MEM_SIZE> and returns the result.
976
977=item $ulong = $device->max_constant_buffer_size
978
979Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_MAX_CONSTANT_BUFFER_SIZE> and returns the result.
980
981=item $uint = $device->max_constant_args
982
983Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_MAX_CONSTANT_ARGS> and returns the result.
984
985=item $device_local_mem_type = $device->local_mem_type
986
987Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_LOCAL_MEM_TYPE> and returns the result.
988
989=item $ulong = $device->local_mem_size
990
991Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_LOCAL_MEM_SIZE> and returns the result.
992
993=item $boolean = $device->error_correction_support
994
995Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_ERROR_CORRECTION_SUPPORT> and returns the result.
996
997=item $int = $device->profiling_timer_resolution
998
999Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_PROFILING_TIMER_RESOLUTION> and returns the result.
1000
1001=item $boolean = $device->endian_little
1002
1003Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_ENDIAN_LITTLE> and returns the result.
1004
1005=item $boolean = $device->available
1006
1007Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_AVAILABLE> and returns the result.
1008
1009=item $boolean = $device->compiler_available
1010
1011Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_COMPILER_AVAILABLE> and returns the result.
1012
1013=item $device_exec_capabilities = $device->execution_capabilities
1014
1015Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_EXECUTION_CAPABILITIES> and returns the result.
1016
1017=item $command_queue_properties = $device->properties
1018
1019Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_QUEUE_PROPERTIES> and returns the result.
1020
1021=item $ = $device->platform
1022
1023Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_PLATFORM> and returns the result.
1024
1025=item $string = $device->name
1026
1027Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_NAME> and returns the result.
1028
1029=item $string = $device->vendor
1030
1031Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_VENDOR> and returns the result.
1032
1033=item $string = $device->driver_version
1034
1035Calls C<clGetDeviceInfo> with C<OpenCL::DRIVER_VERSION> and returns the result.
1036
1037=item $string = $device->profile
1038
1039Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_PROFILE> and returns the result.
1040
1041=item $string = $device->version
1042
1043Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_VERSION> and returns the result.
1044
1045=item $string = $device->extensions
1046
1047Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_EXTENSIONS> and returns the result.
1048
1049=item $uint = $device->preferred_vector_width_half
1050
1051Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_PREFERRED_VECTOR_WIDTH_HALF> and returns the result.
1052
1053=item $uint = $device->native_vector_width_char
1054
1055Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_NATIVE_VECTOR_WIDTH_CHAR> and returns the result.
1056
1057=item $uint = $device->native_vector_width_short
1058
1059Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_NATIVE_VECTOR_WIDTH_SHORT> and returns the result.
1060
1061=item $uint = $device->native_vector_width_int
1062
1063Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_NATIVE_VECTOR_WIDTH_INT> and returns the result.
1064
1065=item $uint = $device->native_vector_width_long
1066
1067Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_NATIVE_VECTOR_WIDTH_LONG> and returns the result.
1068
1069=item $uint = $device->native_vector_width_float
1070
1071Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_NATIVE_VECTOR_WIDTH_FLOAT> and returns the result.
1072
1073=item $uint = $device->native_vector_width_double
1074
1075Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_NATIVE_VECTOR_WIDTH_DOUBLE> and returns the result.
1076
1077=item $uint = $device->native_vector_width_half
1078
1079Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_NATIVE_VECTOR_WIDTH_HALF> and returns the result.
1080
1081=item $device_fp_config = $device->double_fp_config
1082
1083Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_DOUBLE_FP_CONFIG> and returns the result.
1084
1085=item $device_fp_config = $device->half_fp_config
1086
1087Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_HALF_FP_CONFIG> and returns the result.
1088
1089=item $boolean = $device->host_unified_memory
1090
1091Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_HOST_UNIFIED_MEMORY> and returns the result.
1092
1093=item $device = $device->parent_device_ext
1094
1095Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_PARENT_DEVICE_EXT> and returns the result.
1096
1097=item @device_partition_property_exts = $device->partition_types_ext
1098
1099Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_PARTITION_TYPES_EXT> and returns the result.
1100
1101=item @device_partition_property_exts = $device->affinity_domains_ext
1102
1103Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_AFFINITY_DOMAINS_EXT> and returns the result.
1104
1105=item $uint = $device->reference_count_ext
1106
1107Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_REFERENCE_COUNT_EXT> and returns the result.
1108
1109=item @device_partition_property_exts = $device->partition_style_ext
1110
1111Calls C<clGetDeviceInfo> with C<OpenCL::DEVICE_PARTITION_STYLE_EXT> and returns the result.
1112
1113=for gengetinfo end device
249 1114
250=back 1115=back
251 1116
252=head2 THE OpenCL::Context CLASS 1117=head2 THE OpenCL::Context CLASS
253 1118
1119An OpenCL::Context is basically a container, or manager, for a number of
1120devices of a platform. It is used to create all sorts of secondary objects
1121such as buffers, queues, programs and so on.
1122
1123All context creation functions and methods take a list of properties
1124(type-value pairs). All property values can be specified as integers -
1125some additionally support other types:
1126
254=over 4 1127=over 4
255 1128
256=item $packed_value = $ctx->info ($name) 1129=item OpenCL::CONTEXT_PLATFORM
257 1130
258See C<< $platform->info >> for details. 1131Also accepts OpenCL::Platform objects.
259 1132
260L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetContextInfo.html> 1133=item OpenCL::GLX_DISPLAY_KHR
261 1134
262=item $queue = $ctx->command_queue_simple ($device) 1135Also accepts C<undef>, in which case a deep and troubling hack is engaged
1136to find the current glx display (see L<GLX SUPPORT>).
263 1137
1138=item OpenCL::GL_CONTEXT_KHR
1139
1140Also accepts C<undef>, in which case a deep and troubling hack is engaged
1141to find the current glx context (see L<GLX SUPPORT>).
1142
1143=back
1144
1145=over 4
1146
1147=item $prog = $ctx->build_program ($program, $options = "")
1148
1149This convenience function tries to build the program on all devices in
1150the context. If the build fails, then the function will C<croak> with the
1151build log. Otherwise ti returns the program object.
1152
1153The C<$program> can either be a C<OpenCL::Program> object or a string
1154containing the program. In the latter case, a program objetc will be
1155created automatically.
1156
1157=cut
1158
1159sub OpenCL::Context::build_program {
1160 my ($self, $prog, $options) = @_;
1161
1162 $prog = $self->program_with_source ($prog)
1163 unless ref $prog;
1164
1165 eval { $prog->build (undef, $options); 1 }
1166 or errno == BUILD_PROGRAM_FAILURE
1167 or errno == INVALID_BINARY # workaround nvidia bug
1168 or Carp::croak "OpenCL::Context->build_program: " . err2str;
1169
1170 # we check status for all devices
1171 for my $dev ($self->devices) {
1172 $prog->build_status ($dev) == BUILD_SUCCESS
1173 or Carp::croak "Building OpenCL program for device '" . $dev->name . "' failed:\n"
1174 . $prog->build_log ($dev);
1175 }
1176
1177 $prog
1178}
1179
1180=item $queue = $ctx->queue ($device, $properties)
1181
264Convenience function to create a new OpenCL::Queue object from the context and the given device. 1182Create a new OpenCL::Queue object from the context and the given device.
265 1183
266L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateCommandQueue.html> 1184L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateCommandQueue.html>
267 1185
1186Example: create an out-of-order queue.
1187
1188 $queue = $ctx->queue ($device, OpenCL::QUEUE_OUT_OF_ORDER_EXEC_MODE_ENABLE);
1189
268=item $ev = $ctx->user_event 1190=item $ev = $ctx->user_event
269 1191
270Creates a new OpenCL::UserEvent object. 1192Creates a new OpenCL::UserEvent object.
271 1193
272L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateUserEvent.html> 1194L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateUserEvent.html>
273 1195
274=item $buf = $ctx->buffer ($flags, $len) 1196=item $buf = $ctx->buffer ($flags, $len)
275 1197
276Creates a new OpenCL::Buffer object with the given flags and octet-size. 1198Creates a new OpenCL::Buffer (actually OpenCL::BufferObj) object with the
1199given flags and octet-size.
1200
1201flags: OpenCL::MEM_READ_WRITE, OpenCL::MEM_WRITE_ONLY, OpenCL::MEM_READ_ONLY,
1202OpenCL::MEM_USE_HOST_PTR, OpenCL::MEM_ALLOC_HOST_PTR, OpenCL::MEM_COPY_HOST_PTR,
1203OpenCL::MEM_HOST_WRITE_ONLY, OpenCL::MEM_HOST_READ_ONLY, OpenCL::MEM_HOST_NO_ACCESS.
277 1204
278L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateBuffer.html> 1205L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateBuffer.html>
279 1206
280=item $buf = $ctx->buffer_sv ($flags, $data) 1207=item $buf = $ctx->buffer_sv ($flags, $data)
281 1208
282Creates a new OpenCL::Buffer object and initialise it with the given data values. 1209Creates a new OpenCL::Buffer (actually OpenCL::BufferObj) object and
1210initialise it with the given data values.
283 1211
1212=item $img = $ctx->image ($self, $flags, $channel_order, $channel_type, $type, $width, $height, $depth = 0, $array_size = 0, $row_pitch = 0, $slice_pitch = 0, $num_mip_level = 0, $num_samples = 0, $*data = &PL_sv_undef)
1213
1214Creates a new OpenCL::Image object and optionally initialises it with
1215the given data values.
1216
1217channel_order: OpenCL::R, OpenCL::A, OpenCL::RG, OpenCL::RA, OpenCL::RGB,
1218OpenCL::RGBA, OpenCL::BGRA, OpenCL::ARGB, OpenCL::INTENSITY, OpenCL::LUMINANCE,
1219OpenCL::Rx, OpenCL::RGx, OpenCL::RGBx.
1220
1221channel_type: OpenCL::SNORM_INT8, OpenCL::SNORM_INT16, OpenCL::UNORM_INT8,
1222OpenCL::UNORM_INT16, OpenCL::UNORM_SHORT_565, OpenCL::UNORM_SHORT_555,
1223OpenCL::UNORM_INT_101010, OpenCL::SIGNED_INT8, OpenCL::SIGNED_INT16,
1224OpenCL::SIGNED_INT32, OpenCL::UNSIGNED_INT8, OpenCL::UNSIGNED_INT16,
1225OpenCL::UNSIGNED_INT32, OpenCL::HALF_FLOAT, OpenCL::FLOAT.
1226
1227type: OpenCL::MEM_OBJECT_BUFFER, OpenCL::MEM_OBJECT_IMAGE2D,
1228OpenCL::MEM_OBJECT_IMAGE3D, OpenCL::MEM_OBJECT_IMAGE2D_ARRAY,
1229OpenCL::MEM_OBJECT_IMAGE1D, OpenCL::MEM_OBJECT_IMAGE1D_ARRAY,
1230OpenCL::MEM_OBJECT_IMAGE1D_BUFFER.
1231
1232L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clCreateImage.html>
1233
284=item $img = $ctx->image2d ($flags, $channel_order, $channel_type, $width, $height, $data) 1234=item $img = $ctx->image2d ($flags, $channel_order, $channel_type, $width, $height, $row_pitch = 0, $data = undef)
285 1235
286Creates a new OpenCL::Image2D object and optionally initialises it with the given data values. 1236Creates a new OpenCL::Image2D object and optionally initialises it with
1237the given data values.
287 1238
288L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateImage2D.html> 1239L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateImage2D.html>
289 1240
290=item $img = $ctx->image3d ($flags, $channel_order, $channel_type, $width, $height, $depth, $slice_pitch, $data) 1241=item $img = $ctx->image3d ($flags, $channel_order, $channel_type, $width, $height, $depth, $row_pitch = 0, $slice_pitch = 0, $data = undef)
291 1242
292Creates a new OpenCL::Image3D object and optionally initialises it with the given data values. 1243Creates a new OpenCL::Image3D object and optionally initialises it with
1244the given data values.
293 1245
294L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateImage3D.html> 1246L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateImage3D.html>
1247
1248=item $buffer = $ctx->gl_buffer ($flags, $bufobj)
1249
1250Creates a new OpenCL::Buffer (actually OpenCL::BufferObj) object that refers to the given
1251OpenGL buffer object.
1252
1253flags: OpenCL::MEM_READ_WRITE, OpenCL::MEM_READ_ONLY, OpenCL::MEM_WRITE_ONLY.
1254
1255http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLBuffer.html
1256
1257=item $img = $ctx->gl_texture ($flags, $target, $miplevel, $texture)
1258
1259Creates a new OpenCL::Image object that refers to the given OpenGL
1260texture object or buffer.
1261
1262target: GL_TEXTURE_1D, GL_TEXTURE_1D_ARRAY, GL_TEXTURE_BUFFER,
1263GL_TEXTURE_2D, GL_TEXTURE_2D_ARRAY, GL_TEXTURE_3D,
1264GL_TEXTURE_CUBE_MAP_POSITIVE_X, GL_TEXTURE_CUBE_MAP_POSITIVE_Y,
1265GL_TEXTURE_CUBE_MAP_POSITIVE_Z, GL_TEXTURE_CUBE_MAP_NEGATIVE_X,
1266GL_TEXTURE_CUBE_MAP_NEGATIVE_Y, GL_TEXTURE_CUBE_MAP_NEGATIVE_Z,
1267GL_TEXTURE_RECTANGLE/GL_TEXTURE_RECTANGLE_ARB.
1268
1269http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clCreateFromGLTexture.html
1270
1271=item $img = $ctx->gl_texture2d ($flags, $target, $miplevel, $texture)
1272
1273Creates a new OpenCL::Image2D object that refers to the given OpenGL
12742D texture object.
1275
1276http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLTexture2D.html
1277
1278=item $img = $ctx->gl_texture3d ($flags, $target, $miplevel, $texture)
1279
1280Creates a new OpenCL::Image3D object that refers to the given OpenGL
12813D texture object.
1282
1283http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLTexture3D.html
1284
1285=item $ctx->gl_renderbuffer ($flags, $renderbuffer)
1286
1287Creates a new OpenCL::Image2D object that refers to the given OpenGL
1288render buffer.
1289
1290http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLRenderbuffer.html
295 1291
296=item @formats = $ctx->supported_image_formats ($flags, $image_type) 1292=item @formats = $ctx->supported_image_formats ($flags, $image_type)
297 1293
298Returns a list of matching image formats - each format is an arrayref with 1294Returns a list of matching image formats - each format is an arrayref with
299two values, $channel_order and $channel_type, in it. 1295two values, $channel_order and $channel_type, in it.
302 1298
303=item $sampler = $ctx->sampler ($normalized_coords, $addressing_mode, $filter_mode) 1299=item $sampler = $ctx->sampler ($normalized_coords, $addressing_mode, $filter_mode)
304 1300
305Creates a new OpenCL::Sampler object. 1301Creates a new OpenCL::Sampler object.
306 1302
1303addressing_mode: OpenCL::ADDRESS_NONE, OpenCL::ADDRESS_CLAMP_TO_EDGE,
1304OpenCL::ADDRESS_CLAMP, OpenCL::ADDRESS_REPEAT, OpenCL::ADDRESS_MIRRORED_REPEAT.
1305
1306filter_mode: OpenCL::FILTER_NEAREST, OpenCL::FILTER_LINEAR.
1307
307L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateSampler.html> 1308L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateSampler.html>
308 1309
309=item $program = $ctx->program_with_source ($string) 1310=item $program = $ctx->program_with_source ($string)
310 1311
311Creates a new OpenCL::Program object from the given source code. 1312Creates a new OpenCL::Program object from the given source code.
312 1313
313L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateProgramWithSource.html> 1314L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateProgramWithSource.html>
314 1315
1316=item ($program, \@status) = $ctx->program_with_binary (\@devices, \@binaries)
1317
1318Creates a new OpenCL::Program object from the given binaries.
1319
1320L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateProgramWithBinary.html>
1321
1322Example: clone an existing program object that contains a successfully
1323compiled program, no matter how useless this is.
1324
1325 my $clone = $ctx->program_with_binary ([$prog->devices], [$prog->binaries]);
1326
1327=item $program = $ctx->program_with_built_in_kernels (\@devices, $kernel_names)
1328
1329Creates a new OpenCL::Program object from the given built-in kernel names.
1330
1331L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clCreateProgramWithBuiltInKernels.html>
1332
1333=item $program = $ctx->link_program (\@devices, $options, \@programs, $cb->($program) = undef)
1334
1335Links all (already compiled) program objects specified in C<@programs>
1336together and returns a new OpenCL::Program object with the result.
1337
1338L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clLinkProgram.html>
1339
1340=item $packed_value = $ctx->info ($name)
1341
1342See C<< $platform->info >> for details.
1343
1344L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetContextInfo.html>
1345
1346=for gengetinfo begin context
1347
1348=item $uint = $context->reference_count
1349
1350Calls C<clGetContextInfo> with C<OpenCL::CONTEXT_REFERENCE_COUNT> and returns the result.
1351
1352=item @devices = $context->devices
1353
1354Calls C<clGetContextInfo> with C<OpenCL::CONTEXT_DEVICES> and returns the result.
1355
1356=item @property_ints = $context->properties
1357
1358Calls C<clGetContextInfo> with C<OpenCL::CONTEXT_PROPERTIES> and returns the result.
1359
1360=item $uint = $context->num_devices
1361
1362Calls C<clGetContextInfo> with C<OpenCL::CONTEXT_NUM_DEVICES> and returns the result.
1363
1364=for gengetinfo end context
1365
315=back 1366=back
316 1367
317=head2 THE OpenCL::Queue CLASS 1368=head2 THE OpenCL::Queue CLASS
318 1369
319An OpenCL::Queue represents an execution queue for OpenCL. You execute 1370An OpenCL::Queue represents an execution queue for OpenCL. You execute
320requests by calling their respective C<enqueue_xxx> method and waitinf for 1371requests by calling their respective method and waiting for it to complete
321it to complete in some way. 1372in some way.
322 1373
323All the enqueue methods return an event object that can be used to wait 1374Most methods that enqueue some request return an event object that can
324for completion, unless the method is called in void context, in which case 1375be used to wait for completion (optionally using a callback), unless
325no event object is created. 1376the method is called in void context, in which case no event object is
1377created.
326 1378
327They also allow you to specify any number of other event objects that this 1379They also allow you to specify any number of other event objects that this
328request has to wait for before it starts executing, by simply passing the 1380request has to wait for before it starts executing, by simply passing the
329event objects as extra parameters to the enqueue methods. 1381event objects as extra parameters to the enqueue methods. To simplify
1382program design, this module ignores any C<undef> values in the list of
1383events. This makes it possible to code operations such as this, without
1384having to put a valid event object into C<$event> first:
1385
1386 $event = $queue->xxx (..., $event);
330 1387
331Queues execute in-order by default, without any parallelism, so in most 1388Queues execute in-order by default, without any parallelism, so in most
332cases (i.e. you use only one queue) it's not necessary to wait for or 1389cases (i.e. you use only one queue) it's not necessary to wait for or
333create event objects. 1390create event objects, althoguh an our of order queue is often a bit
1391faster.
334 1392
335=over 4 1393=over 4
336 1394
337=item $packed_value = $ctx->info ($name)
338
339See C<< $platform->info >> for details.
340
341L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetCommandQueueInfo.html>
342
343=item $ev = $queue->enqueue_read_buffer ($buffer, $blocking, $offset, $len, $data, $wait_events...) 1395=item $ev = $queue->read_buffer ($buffer, $blocking, $offset, $len, $data, $wait_events...)
344 1396
345Reads data from buffer into the given string. 1397Reads data from buffer into the given string.
346 1398
347L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReadBuffer.html> 1399L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReadBuffer.html>
348 1400
349=item $ev = $queue->enqueue_write_buffer ($buffer, $blocking, $offset, $data, $wait_events...) 1401=item $ev = $queue->write_buffer ($buffer, $blocking, $offset, $data, $wait_events...)
350 1402
351Writes data to buffer from the given string. 1403Writes data to buffer from the given string.
352 1404
353L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteBuffer.html> 1405L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteBuffer.html>
354 1406
355=item $ev = $queue->enqueue_copy_buffer ($src, $dst, $src_offset, $dst_offset, $len, $wait_events...) 1407=item $ev = $queue->copy_buffer ($src, $dst, $src_offset, $dst_offset, $len, $wait_events...)
356 1408
357L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBuffer.html> 1409L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBuffer.html>
358 1410
1411=item $ev = $queue->read_buffer_rect (OpenCL::Memory buf, cl_bool blocking, $buf_x, $buf_y, $buf_z, $host_x, $host_y, $host_z, $width, $height, $depth, $buf_row_pitch, $buf_slice_pitch, $host_row_pitch, $host_slice_pitch, $data, $wait_events...)
1412
1413http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReadBufferRect.html
1414
1415=item $ev = $queue->write_buffer_rect (OpenCL::Memory buf, cl_bool blocking, $buf_x, $buf_y, $buf_z, $host_x, $host_y, $host_z, $width, $height, $depth, $buf_row_pitch, $buf_slice_pitch, $host_row_pitch, $host_slice_pitch, $data, $wait_events...)
1416
1417http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteBufferRect.html
1418
1419=item $ev = $queue->copy_buffer_to_image ($src_buffer, $dst_image, $src_offset, $dst_x, $dst_y, $dst_z, $width, $height, $depth, $wait_events...)
1420
1421L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBufferToImage.html>
1422
359=item $ev = $queue->enqueue_read_image ($src, $blocking, $x, $y, $z, $width, $height, $depth, $row_pitch, $slice_pitch, $data, $wait_events...) 1423=item $ev = $queue->read_image ($src, $blocking, $x, $y, $z, $width, $height, $depth, $row_pitch, $slice_pitch, $data, $wait_events...)
1424
1425C<$row_pitch> (and C<$slice_pitch>) can be C<0>, in which case the OpenCL
1426module uses the image width (and height) to supply default values.
360 1427
361L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReadImage.html> 1428L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReadImage.html>
362 1429
363=item $ev = $queue->enqueue_write_image ($src, $blocking, $x, $y, $z, $width, $height, $depth, $row_pitch, $data, $wait_events...) 1430=item $ev = $queue->write_image ($src, $blocking, $x, $y, $z, $width, $height, $depth, $row_pitch, $slice_pitch, $data, $wait_events...)
364 1431
1432C<$row_pitch> (and C<$slice_pitch>) can be C<0>, in which case the OpenCL
1433module uses the image width (and height) to supply default values.
365L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteImage.html> 1434L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWriteImage.html>
366 1435
1436=item $ev = $queue->copy_image ($src_image, $dst_image, $src_x, $src_y, $src_z, $dst_x, $dst_y, $dst_z, $width, $height, $depth, $wait_events...)
1437
1438L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyImage.html>
1439
1440=item $ev = $queue->copy_image_to_buffer ($src_image, $dst_image, $src_x, $src_y, $src_z, $width, $height, $depth, $dst_offset, $wait_events...)
1441
1442L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyImageToBuffer.html>
1443
367=item $ev = $queue->enqueue_copy_buffer_rect ($src, $dst, $src_x, $src_y, $src_z, $dst_x, $dst_y, $dst_z, $width, $height, $depth, $src_row_pitch, $src_slice_pitch, 4dst_row_pitch, $dst_slice_pitch, $ait_event...) 1444=item $ev = $queue->copy_buffer_rect ($src, $dst, $src_x, $src_y, $src_z, $dst_x, $dst_y, $dst_z, $width, $height, $depth, $src_row_pitch, $src_slice_pitch, $dst_row_pitch, $dst_slice_pitch, $wait_event...)
368 1445
369Yeah. 1446Yeah.
370 1447
371L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBufferRect.html>
372
373=item $ev = $queue->enqueue_copy_buffer_to_image (OpenCL::Buffer src, OpenCL::Image dst, size_t src_offset, size_t dst_x, size_t dst_y, size_t dst_z, size_t width, size_t height, size_t depth, ...)
374
375L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBufferToImage.html>. 1448L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyBufferToImage.html>.
376 1449
377=item $ev = $queue->enqueue_copy_image (OpenCL::Image src, OpenCL::Buffer dst, size_t src_x, size_t src_y, size_t src_z, size_t dst_x, size_t dst_y, size_t dst_z, size_t width, size_t height, size_t depth, ...) 1450=item $ev = $queue->fill_buffer ($mem, $pattern, $offset, $size, ...)
378 1451
1452Fills the given buffer object with repeated applications of C<$pattern>,
1453starting at C<$offset> for C<$size> octets.
1454
1455L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clEnqueueFillBuffer.html>
1456
1457=item $ev = $queue->fill_image ($img, $r, $g, $b, $a, $x, $y, $z, $width, $height, $depth, ...)
1458
1459Fills the given image area with the given rgba colour components. The
1460components are normally floating point values between C<0> and C<1>,
1461except when the image channel data type is a signe dor unsigned
1462unnormalised format, in which case the range is determined by the format.
1463
379L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyImage.html> 1464L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clEnqueueFillImage.html>
380 1465
381=item $ev = $queue->enqueue_copy_image_to_buffer (OpenCL::Image src, OpenCL::Buffer dst, size_t src_x, size_t src_y, size_t src_z, size_t width, size_t height, size_t depth, size_t dst_offset, ...)
382
383L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueCopyImageToBuffer.html>
384
385=item $ev = $queue->enqueue_task ($kernel, $wait_events...) 1466=item $ev = $queue->task ($kernel, $wait_events...)
386 1467
387L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueTask.html> 1468L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueTask.html>
388 1469
389=item $ev = $queue->enqueue_nd_range_kernel ($kernel, @$global_work_offset, @$global_work_size, @$local_work_size, $wait_events...) 1470=item $ev = $queue->nd_range_kernel ($kernel, \@global_work_offset, \@global_work_size, \@local_work_size, $wait_events...)
390 1471
391Enqueues a kernel execution. 1472Enqueues a kernel execution.
392 1473
393@$global_work_size must be specified as a reference to an array of 1474\@global_work_size must be specified as a reference to an array of
394integers specifying the work sizes (element counts). 1475integers specifying the work sizes (element counts).
395 1476
396@$global_work_offset must be either C<undef> (in which case all offsets 1477\@global_work_offset must be either C<undef> (in which case all offsets
397are C<0>), or a reference to an array of work offsets, with the same number 1478are C<0>), or a reference to an array of work offsets, with the same number
398of elements as @$global_work_size. 1479of elements as \@global_work_size.
399 1480
400@$local_work_size must be either C<undef> (in which case the 1481\@local_work_size must be either C<undef> (in which case the
401implementation is supposed to choose good local work sizes), or a 1482implementation is supposed to choose good local work sizes), or a
402reference to an array of local work sizes, with the same number of 1483reference to an array of local work sizes, with the same number of
403elements as @$global_work_size. 1484elements as \@global_work_size.
404 1485
405L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueNDRangeKernel.html> 1486L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueNDRangeKernel.html>
406 1487
407=item $ev = $queue->enqueue_marker 1488=item $ev = $queue->migrate_mem_objects (\@mem_objects, $flags, $wait_events...)
408 1489
1490Migrates a number of OpenCL::Memory objects to or from the device.
1491
1492flags: OpenCL::MIGRATE_MEM_OBJECT_HOST, OpenCL::MIGRATE_MEM_OBJECT_CONTENT_UNDEFINED
1493
1494L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clEnqueueMigrateMemObjects.html>
1495
1496=item $ev = $queue->acquire_gl_objects ([object, ...], $wait_events...)
1497
1498Enqueues a list (an array-ref of OpenCL::Memory objects) to be acquired
1499for subsequent OpenCL usage.
1500
409L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueMarker.html> 1501L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueAcquireGLObjects.html>
410 1502
1503=item $ev = $queue->release_gl_objects ([object, ...], $wait_events...)
1504
1505Enqueues a list (an array-ref of OpenCL::Memory objects) to be released
1506for subsequent OpenGL usage.
1507
1508L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReleaseGLObjects.html>
1509
411=item $ev = $queue->enqueue_wait_for_events ($wait_events...) 1510=item $ev = $queue->wait_for_events ($wait_events...)
412 1511
413L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWaitForEvents.html> 1512L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWaitForEvents.html>
414 1513
415=item $queue->enqueue_barrier 1514=item $ev = $queue->marker ($wait_events...)
416 1515
1516L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clEnqueueMarkerWithWaitList.html>
1517
1518=item $ev = $queue->barrier ($wait_events...)
1519
417L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueBarrier.html> 1520L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clEnqueueBarrierWithWaitList.html>
418 1521
419=item $queue->flush 1522=item $queue->flush
420 1523
421L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clFlush.html> 1524L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clFlush.html>
422 1525
423=item $queue->finish 1526=item $queue->finish
424 1527
425L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clFinish.html> 1528L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clFinish.html>
426 1529
1530=item $packed_value = $queue->info ($name)
1531
1532See C<< $platform->info >> for details.
1533
1534L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetCommandQueueInfo.html>
1535
1536=for gengetinfo begin command_queue
1537
1538=item $ctx = $command_queue->context
1539
1540Calls C<clGetCommandQueueInfo> with C<OpenCL::QUEUE_CONTEXT> and returns the result.
1541
1542=item $device = $command_queue->device
1543
1544Calls C<clGetCommandQueueInfo> with C<OpenCL::QUEUE_DEVICE> and returns the result.
1545
1546=item $uint = $command_queue->reference_count
1547
1548Calls C<clGetCommandQueueInfo> with C<OpenCL::QUEUE_REFERENCE_COUNT> and returns the result.
1549
1550=item $command_queue_properties = $command_queue->properties
1551
1552Calls C<clGetCommandQueueInfo> with C<OpenCL::QUEUE_PROPERTIES> and returns the result.
1553
1554=for gengetinfo end command_queue
1555
427=back 1556=back
428 1557
1558=head3 MEMORY MAPPED BUFFERS
1559
1560OpenCL allows you to map buffers and images to host memory (read: perl
1561scalars). This is done much like reading or copying a buffer, by enqueuing
1562a map or unmap operation on the command queue.
1563
1564The map operations return an C<OpenCL::Mapped> object - see L<THE
1565OpenCL::Mapped CLASS> section for details on what to do with these
1566objects.
1567
1568The object will be unmapped automatically when the mapped object is
1569destroyed (you can use a barrier to make sure the unmap has finished,
1570before using the buffer in a kernel), but you can also enqueue an unmap
1571operation manually.
1572
1573=over 4
1574
1575=item $mapped_buffer = $queue->map_buffer ($buf, $blocking=1, $map_flags=OpenCL::MAP_READ|OpenCL::MAP_WRITE, $offset=0, $size=undef, $wait_events...)
1576
1577Maps the given buffer into host memory and returns an
1578C<OpenCL::MappedBuffer> object. If C<$size> is specified as undef, then
1579the map will extend to the end of the buffer.
1580
1581map_flags: OpenCL::MAP_READ, OpenCL::MAP_WRITE, OpenCL::MAP_WRITE_INVALIDATE_REGION.
1582
1583L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueMapBuffer.html>
1584
1585Example: map the buffer $buf fully and replace the first 4 bytes by "abcd", then unmap.
1586
1587 {
1588 my $mapped = $queue->map_buffer ($buf, 1, OpenCL::MAP_WRITE);
1589 substr $$mapped, 0, 4, "abcd";
1590 } # asynchronously unmap because $mapped is destroyed
1591
1592=item $mapped_image = $queue->map_image ($img, $blocking=1, $map_flags=OpenCL::MAP_READ|OpenCL::MAP_WRITE, $x=0, $y=0, $z=0, $width=undef, $height=undef, $depth=undef, $wait_events...)
1593
1594Maps the given image area into host memory and return an
1595C<OpenCL::MappedImage> object.
1596
1597If any of C<$width>, C<$height> and/or C<$depth> are C<undef> then they
1598will be replaced by the maximum possible value.
1599
1600L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueMapImage.html>
1601
1602Example: map an image (with OpenCL::UNSIGNED_INT8 channel type) and set
1603the first channel of the leftmost column to 5, then explicitly unmap
1604it. You are not necessarily meant to do it this way, this example just
1605shows you the accessors to use :)
1606
1607 my $mapped = $queue->map_image ($image, 1, OpenCL::MAP_WRITE);
1608
1609 $mapped->set ($_ * $mapped->row_pitch, pack "C", 5)
1610 for 0..$image->height;
1611
1612 $mapped->unmap;.
1613 $mapped->wait; # only needed for out of order queues normally
1614
1615=item $ev = $queue->unmap ($mapped, $wait_events...)
1616
1617Unmaps the data from host memory. You must not call any methods that
1618modify the data, or modify the data scalar directly, after calling this
1619method.
1620
1621The mapped event object will always be passed as part of the
1622$wait_events. The mapped event object will be replaced by the new event
1623object that this request creates.
1624
1625=back
1626
429=head2 THE OpenCL::Memory CLASS 1627=head2 THE OpenCL::Memory CLASS
430 1628
431This the superclass of all memory objects - OpenCL::Buffer, OpenCL::Image, 1629This the superclass of all memory objects - OpenCL::Buffer, OpenCL::Image,
432OpenCL::Image2D and OpenCL::Image3D. The subclasses of this class 1630OpenCL::Image2D and OpenCL::Image3D.
433currently only exist to allow type-checking.
434 1631
435=over 4 1632=over 4
436 1633
437=item $packed_value = $memory->info ($name) 1634=item $packed_value = $memory->info ($name)
438 1635
439See C<< $platform->info >> for details. 1636See C<< $platform->info >> for details.
440 1637
441L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetMemObjectInfo.html> 1638L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetMemObjectInfo.html>
442 1639
1640=for gengetinfo begin mem
1641
1642=item $mem_object_type = $mem->type
1643
1644Calls C<clGetMemObjectInfo> with C<OpenCL::MEM_TYPE> and returns the result.
1645
1646=item $mem_flags = $mem->flags
1647
1648Calls C<clGetMemObjectInfo> with C<OpenCL::MEM_FLAGS> and returns the result.
1649
1650=item $int = $mem->size
1651
1652Calls C<clGetMemObjectInfo> with C<OpenCL::MEM_SIZE> and returns the result.
1653
1654=item $ptr_value = $mem->host_ptr
1655
1656Calls C<clGetMemObjectInfo> with C<OpenCL::MEM_HOST_PTR> and returns the result.
1657
1658=item $uint = $mem->map_count
1659
1660Calls C<clGetMemObjectInfo> with C<OpenCL::MEM_MAP_COUNT> and returns the result.
1661
1662=item $uint = $mem->reference_count
1663
1664Calls C<clGetMemObjectInfo> with C<OpenCL::MEM_REFERENCE_COUNT> and returns the result.
1665
1666=item $ctx = $mem->context
1667
1668Calls C<clGetMemObjectInfo> with C<OpenCL::MEM_CONTEXT> and returns the result.
1669
1670=item $mem = $mem->associated_memobject
1671
1672Calls C<clGetMemObjectInfo> with C<OpenCL::MEM_ASSOCIATED_MEMOBJECT> and returns the result.
1673
1674=item $int = $mem->offset
1675
1676Calls C<clGetMemObjectInfo> with C<OpenCL::MEM_OFFSET> and returns the result.
1677
1678=for gengetinfo end mem
1679
1680=item ($type, $name) = $mem->gl_object_info
1681
1682Returns the OpenGL object type (e.g. OpenCL::GL_OBJECT_TEXTURE2D) and the
1683object "name" (e.g. the texture name) used to create this memory object.
1684
1685L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetGLObjectInfo.html>
1686
443=back 1687=back
444 1688
1689=head2 THE OpenCL::Buffer CLASS
1690
1691This is a subclass of OpenCL::Memory, and the superclass of
1692OpenCL::BufferObj. Its purpose is simply to distinguish between buffers
1693and sub-buffers.
1694
1695=head2 THE OpenCL::BufferObj CLASS
1696
1697This is a subclass of OpenCL::Buffer and thus OpenCL::Memory. It exists
1698because one cna create sub buffers of OpenLC::BufferObj objects, but not
1699sub buffers from these sub buffers.
1700
1701=over 4
1702
1703=item $subbuf = $buf_obj->sub_buffer_region ($flags, $origin, $size)
1704
1705Creates an OpenCL::Buffer objects from this buffer and returns it. The
1706C<buffer_create_type> is assumed to be C<OpenCL::BUFFER_CREATE_TYPE_REGION>.
1707
1708L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateSubBuffer.html>
1709
1710=back
1711
1712=head2 THE OpenCL::Image CLASS
1713
1714This is the superclass of all image objects - OpenCL::Image1D,
1715OpenCL::Image1DArray, OpenCL::Image1DBuffer, OpenCL::Image2D,
1716OpenCL::Image2DArray and OpenCL::Image3D.
1717
1718=over 4
1719
1720=item $packed_value = $image->image_info ($name)
1721
1722See C<< $platform->info >> for details.
1723
1724The reason this method is not called C<info> is that there already is an
1725C<< ->info >> method inherited from C<OpenCL::Memory>.
1726
1727L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetImageInfo.html>
1728
1729=item ($channel_order, $channel_data_type) = $image->format
1730
1731Returns the channel order and type used to create the image by calling
1732C<clGetImageInfo> with C<OpenCL::IMAGE_FORMAT>.
1733
1734=for gengetinfo begin image
1735
1736=item $int = $image->element_size
1737
1738Calls C<clGetImageInfo> with C<OpenCL::IMAGE_ELEMENT_SIZE> and returns the result.
1739
1740=item $int = $image->row_pitch
1741
1742Calls C<clGetImageInfo> with C<OpenCL::IMAGE_ROW_PITCH> and returns the result.
1743
1744=item $int = $image->slice_pitch
1745
1746Calls C<clGetImageInfo> with C<OpenCL::IMAGE_SLICE_PITCH> and returns the result.
1747
1748=item $int = $image->width
1749
1750Calls C<clGetImageInfo> with C<OpenCL::IMAGE_WIDTH> and returns the result.
1751
1752=item $int = $image->height
1753
1754Calls C<clGetImageInfo> with C<OpenCL::IMAGE_HEIGHT> and returns the result.
1755
1756=item $int = $image->depth
1757
1758Calls C<clGetImageInfo> with C<OpenCL::IMAGE_DEPTH> and returns the result.
1759
1760=for gengetinfo end image
1761
1762=for gengetinfo begin gl_texture
1763
1764=item $GLenum = $gl_texture->target
1765
1766Calls C<clGetGLTextureInfo> with C<OpenCL::GL_TEXTURE_TARGET> and returns the result.
1767
1768=item $GLint = $gl_texture->gl_mipmap_level
1769
1770Calls C<clGetGLTextureInfo> with C<OpenCL::GL_MIPMAP_LEVEL> and returns the result.
1771
1772=for gengetinfo end gl_texture
1773
1774=back
1775
445=head2 THE OpenCL::Sampler CLASS 1776=head2 THE OpenCL::Sampler CLASS
446 1777
447=over 4 1778=over 4
448 1779
449=item $packed_value = $sampler->info ($name) 1780=item $packed_value = $sampler->info ($name)
450 1781
451See C<< $platform->info >> for details. 1782See C<< $platform->info >> for details.
452 1783
453L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetSamplerInfo.html> 1784L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetSamplerInfo.html>
454 1785
1786=for gengetinfo begin sampler
1787
1788=item $uint = $sampler->reference_count
1789
1790Calls C<clGetSamplerInfo> with C<OpenCL::SAMPLER_REFERENCE_COUNT> and returns the result.
1791
1792=item $ctx = $sampler->context
1793
1794Calls C<clGetSamplerInfo> with C<OpenCL::SAMPLER_CONTEXT> and returns the result.
1795
1796=item $addressing_mode = $sampler->normalized_coords
1797
1798Calls C<clGetSamplerInfo> with C<OpenCL::SAMPLER_NORMALIZED_COORDS> and returns the result.
1799
1800=item $filter_mode = $sampler->addressing_mode
1801
1802Calls C<clGetSamplerInfo> with C<OpenCL::SAMPLER_ADDRESSING_MODE> and returns the result.
1803
1804=item $boolean = $sampler->filter_mode
1805
1806Calls C<clGetSamplerInfo> with C<OpenCL::SAMPLER_FILTER_MODE> and returns the result.
1807
1808=for gengetinfo end sampler
1809
455=back 1810=back
456 1811
457=head2 THE OpenCL::Program CLASS 1812=head2 THE OpenCL::Program CLASS
458 1813
459=over 4 1814=over 4
460 1815
461=item $packed_value = $program->info ($name) 1816=item $program->build (\@devices = undef, $options = "", $cb->($program) = undef)
462 1817
463See C<< $platform->info >> for details.
464
465L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetProgramInfo.html>
466
467=item $program->build ($device, $options = "")
468
469Tries to build the program with the givne options. 1818Tries to build the program with the given options. See also the
1819C<$ctx->build> convenience function.
1820
1821If a callback is specified, then it will be called when compilation is
1822finished. Note that many OpenCL implementations block your program while
1823compiling whether you use a callback or not. See C<build_async> if you
1824want to make sure the build is done in the background.
1825
1826Note that some OpenCL implementations act up badly, and don't call the
1827callback in some error cases (but call it in others). This implementation
1828assumes the callback will always be called, and leaks memory if this is
1829not so. So best make sure you don't pass in invalid values.
1830
1831Some implementations fail with C<OpenCL::INVALID_BINARY> when the
1832compilation state is successful but some later stage fails.
1833
1834options: C<-D name>, C<-D name=definition>, C<-I dir>,
1835C<-cl-single-precision-constant>, C<-cl-denorms-are-zero>,
1836C<-cl-fp32-correctly-rounded-divide-sqrt>, C<-cl-opt-disable>,
1837C<-cl-mad-enable>, C<-cl-no-signed-zeros>, C<-cl-unsafe-math-optimizations>,
1838C<-cl-finite-math-only>, C<-cl-fast-relaxed-math>,
1839C<-w>, C<-Werror>, C<-cl-std=CL1.1/CL1.2>, C<-cl-kernel-arg-info>,
1840C<-create-library>, C<-enable-link-options>.
1841
1842build_status: OpenCL::BUILD_SUCCESS, OpenCL::BUILD_NONE,
1843OpenCL::BUILD_ERROR, OpenCL::BUILD_IN_PROGRESS.
470 1844
471L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clBuildProgram.html> 1845L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clBuildProgram.html>
1846
1847=item $program->build_async (\@devices = undef, $options = "", $cb->($program) = undef)
1848
1849Similar to C<< ->build >>, except it starts a thread, and never fails (you
1850need to check the compilation status form the callback, or by polling).
1851
1852=item $program->compile (\@devices = undef, $options = "", \%headers = undef, $cb->($program) = undef)
1853
1854Compiles the given program for the given devices (or all devices if
1855undef). If C<$headers> is given, it must be a hashref with include name =>
1856OpenCL::Program pairs.
1857
1858L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clCompileProgram.html>
472 1859
473=item $packed_value = $program->build_info ($device, $name) 1860=item $packed_value = $program->build_info ($device, $name)
474 1861
475Similar to C<< $platform->info >>, but returns build info for a previous 1862Similar to C<< $platform->info >>, but returns build info for a previous
476build attempt for the given device. 1863build attempt for the given device.
477 1864
1865binary_type: OpenCL::PROGRAM_BINARY_TYPE_NONE,
1866OpenCL::PROGRAM_BINARY_TYPE_COMPILED_OBJECT,
1867OpenCL::PROGRAM_BINARY_TYPE_LIBRARY,
1868OpenCL::PROGRAM_BINARY_TYPE_EXECUTABLE.
1869
478L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetBuildInfo.html> 1870L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetBuildInfo.html>
479 1871
480=item $kernel = $program->kernel ($function_name) 1872=item $kernel = $program->kernel ($function_name)
481 1873
482Creates an OpenCL::Kernel object out of the named C<__kernel> function in 1874Creates an OpenCL::Kernel object out of the named C<__kernel> function in
483the program. 1875the program.
484 1876
485L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateKernel.html> 1877L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateKernel.html>
486 1878
1879=item @kernels = $program->kernels_in_program
1880
1881Returns all kernels successfully compiled for all devices in program.
1882
1883http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateKernelsInProgram.html
1884
1885=for gengetinfo begin program_build
1886
1887=item $build_status = $program->build_status ($device)
1888
1889Calls C<clGetProgramBuildInfo> with C<OpenCL::PROGRAM_BUILD_STATUS> and returns the result.
1890
1891=item $string = $program->build_options ($device)
1892
1893Calls C<clGetProgramBuildInfo> with C<OpenCL::PROGRAM_BUILD_OPTIONS> and returns the result.
1894
1895=item $string = $program->build_log ($device)
1896
1897Calls C<clGetProgramBuildInfo> with C<OpenCL::PROGRAM_BUILD_LOG> and returns the result.
1898
1899=item $binary_type = $program->binary_type ($device)
1900
1901Calls C<clGetProgramBuildInfo> with C<OpenCL::PROGRAM_BINARY_TYPE> and returns the result.
1902
1903=for gengetinfo end program_build
1904
1905=item $packed_value = $program->info ($name)
1906
1907See C<< $platform->info >> for details.
1908
1909L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetProgramInfo.html>
1910
1911=for gengetinfo begin program
1912
1913=item $uint = $program->reference_count
1914
1915Calls C<clGetProgramInfo> with C<OpenCL::PROGRAM_REFERENCE_COUNT> and returns the result.
1916
1917=item $ctx = $program->context
1918
1919Calls C<clGetProgramInfo> with C<OpenCL::PROGRAM_CONTEXT> and returns the result.
1920
1921=item $uint = $program->num_devices
1922
1923Calls C<clGetProgramInfo> with C<OpenCL::PROGRAM_NUM_DEVICES> and returns the result.
1924
1925=item @devices = $program->devices
1926
1927Calls C<clGetProgramInfo> with C<OpenCL::PROGRAM_DEVICES> and returns the result.
1928
1929=item $string = $program->source
1930
1931Calls C<clGetProgramInfo> with C<OpenCL::PROGRAM_SOURCE> and returns the result.
1932
1933=item @ints = $program->binary_sizes
1934
1935Calls C<clGetProgramInfo> with C<OpenCL::PROGRAM_BINARY_SIZES> and returns the result.
1936
1937=for gengetinfo end program
1938
1939=item @blobs = $program->binaries
1940
1941Returns a string for the compiled binary for every device associated with
1942the program, empty strings indicate missing programs, and an empty result
1943means no program binaries are available.
1944
1945These "binaries" are often, in fact, informative low-level assembly
1946sources.
1947
1948L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetProgramInfo.html>
1949
487=back 1950=back
488 1951
489=head2 THE OpenCL::Kernel CLASS 1952=head2 THE OpenCL::Kernel CLASS
490 1953
491=over 4 1954=over 4
494 1957
495See C<< $platform->info >> for details. 1958See C<< $platform->info >> for details.
496 1959
497L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetKernelInfo.html> 1960L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetKernelInfo.html>
498 1961
1962=for gengetinfo begin kernel
1963
1964=item $string = $kernel->function_name
1965
1966Calls C<clGetKernelInfo> with C<OpenCL::KERNEL_FUNCTION_NAME> and returns the result.
1967
1968=item $uint = $kernel->num_args
1969
1970Calls C<clGetKernelInfo> with C<OpenCL::KERNEL_NUM_ARGS> and returns the result.
1971
1972=item $uint = $kernel->reference_count
1973
1974Calls C<clGetKernelInfo> with C<OpenCL::KERNEL_REFERENCE_COUNT> and returns the result.
1975
1976=item $ctx = $kernel->context
1977
1978Calls C<clGetKernelInfo> with C<OpenCL::KERNEL_CONTEXT> and returns the result.
1979
1980=item $program = $kernel->program
1981
1982Calls C<clGetKernelInfo> with C<OpenCL::KERNEL_PROGRAM> and returns the result.
1983
1984=for gengetinfo end kernel
1985
1986=item $packed_value = $kernel->work_group_info ($device, $name)
1987
1988See C<< $platform->info >> for details.
1989
1990L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetKernelWorkGroupInfo.html>
1991
1992=for gengetinfo begin kernel_work_group
1993
1994=item $int = $kernel->work_group_size ($device)
1995
1996Calls C<clGetKernelWorkGroupInfo> with C<OpenCL::KERNEL_WORK_GROUP_SIZE> and returns the result.
1997
1998=item @ints = $kernel->compile_work_group_size ($device)
1999
2000Calls C<clGetKernelWorkGroupInfo> with C<OpenCL::KERNEL_COMPILE_WORK_GROUP_SIZE> and returns the result.
2001
2002=item $ulong = $kernel->local_mem_size ($device)
2003
2004Calls C<clGetKernelWorkGroupInfo> with C<OpenCL::KERNEL_LOCAL_MEM_SIZE> and returns the result.
2005
2006=item $int = $kernel->preferred_work_group_size_multiple ($device)
2007
2008Calls C<clGetKernelWorkGroupInfo> with C<OpenCL::KERNEL_PREFERRED_WORK_GROUP_SIZE_MULTIPLE> and returns the result.
2009
2010=item $ulong = $kernel->private_mem_size ($device)
2011
2012Calls C<clGetKernelWorkGroupInfo> with C<OpenCL::KERNEL_PRIVATE_MEM_SIZE> and returns the result.
2013
2014=for gengetinfo end kernel_work_group
2015
2016=item $packed_value = $kernel->arg_info ($idx, $name)
2017
2018See C<< $platform->info >> for details.
2019
2020L<http://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clGetKernelArgInfo.html>
2021
2022=for gengetinfo begin kernel_arg
2023
2024=item $kernel_arg_address_qualifier = $kernel->arg_address_qualifier ($idx)
2025
2026Calls C<clGetKernelArgInfo> with C<OpenCL::KERNEL_ARG_ADDRESS_QUALIFIER> and returns the result.
2027
2028=item $kernel_arg_access_qualifier = $kernel->arg_access_qualifier ($idx)
2029
2030Calls C<clGetKernelArgInfo> with C<OpenCL::KERNEL_ARG_ACCESS_QUALIFIER> and returns the result.
2031
2032=item $string = $kernel->arg_type_name ($idx)
2033
2034Calls C<clGetKernelArgInfo> with C<OpenCL::KERNEL_ARG_TYPE_NAME> and returns the result.
2035
2036=item $kernel_arg_type_qualifier = $kernel->arg_type_qualifier ($idx)
2037
2038Calls C<clGetKernelArgInfo> with C<OpenCL::KERNEL_ARG_TYPE_QUALIFIER> and returns the result.
2039
2040=item $string = $kernel->arg_name ($idx)
2041
2042Calls C<clGetKernelArgInfo> with C<OpenCL::KERNEL_ARG_NAME> and returns the result.
2043
2044=for gengetinfo end kernel_arg
2045
2046=item $kernel->setf ($format, ...)
2047
2048Sets the arguments of a kernel. Since OpenCL 1.1 doesn't have a generic
2049way to set arguments (and with OpenCL 1.2 it might be rather slow), you
2050need to specify a format argument, much as with C<printf>, to tell OpenCL
2051what type of argument it is.
2052
2053The format arguments are single letters:
2054
2055 c char
2056 C unsigned char
2057 s short
2058 S unsigned short
2059 i int
2060 I unsigned int
2061 l long
2062 L unsigned long
2063
2064 h half float (0..65535)
2065 f float
2066 d double
2067
2068 z local (octet size)
2069
2070 m memory object (buffer or image)
2071 a sampler
2072 e event
2073
2074Space characters in the format string are ignored.
2075
2076Example: set the arguments for a kernel that expects an int, two floats, a buffer and an image.
2077
2078 $kernel->setf ("i ff mm", 5, 0.5, 3, $buffer, $image);
2079
499=item $kernel->set_TYPE ($index, $value) 2080=item $kernel->set_TYPE ($index, $value)
500 2081
2082=item $kernel->set_char ($index, $value)
2083
2084=item $kernel->set_uchar ($index, $value)
2085
2086=item $kernel->set_short ($index, $value)
2087
2088=item $kernel->set_ushort ($index, $value)
2089
2090=item $kernel->set_int ($index, $value)
2091
2092=item $kernel->set_uint ($index, $value)
2093
2094=item $kernel->set_long ($index, $value)
2095
2096=item $kernel->set_ulong ($index, $value)
2097
2098=item $kernel->set_half ($index, $value)
2099
2100=item $kernel->set_float ($index, $value)
2101
2102=item $kernel->set_double ($index, $value)
2103
2104=item $kernel->set_memory ($index, $value)
2105
2106=item $kernel->set_buffer ($index, $value)
2107
2108=item $kernel->set_image ($index, $value)
2109
2110=item $kernel->set_sampler ($index, $value)
2111
2112=item $kernel->set_local ($index, $value)
2113
2114=item $kernel->set_event ($index, $value)
2115
501This is a family of methods to set the kernel argument with the number C<$index> to the give C<$value>. 2116This is a family of methods to set the kernel argument with the number
502 2117C<$index> to the give C<$value>.
503TYPE is one of C<char>, C<uchar>, C<short>, C<ushort>, C<int>, C<uint>,
504C<long>, C<ulong>, C<half>, C<float>, C<double>, C<memory>, C<buffer>,
505C<image2d>, C<image3d>, C<sampler> or C<event>.
506 2118
507Chars and integers (including the half type) are specified as integers, 2119Chars and integers (including the half type) are specified as integers,
508float and double as floating point values, memory/buffer/image2d/image3d 2120float and double as floating point values, memory/buffer/image must be
509must be an object of that type or C<undef>, and sampler and event must be 2121an object of that type or C<undef>, local-memory arguments are set by
510objects of that type. 2122specifying the size, and sampler and event must be objects of that type.
2123
2124Note that C<set_memory> works for all memory objects (all types of buffers
2125and images) - the main purpose of the more specific C<set_TYPE> functions
2126is type checking.
2127
2128Setting an argument for a kernel does NOT keep a reference to the object -
2129for example, if you set an argument to some image object, free the image,
2130and call the kernel, you will run into undefined behaviour.
511 2131
512L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetKernelArg.html> 2132L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetKernelArg.html>
513 2133
514=back 2134=back
515 2135
518This is the superclass for all event objects (including OpenCL::UserEvent 2138This is the superclass for all event objects (including OpenCL::UserEvent
519objects). 2139objects).
520 2140
521=over 4 2141=over 4
522 2142
2143=item $ev->wait
2144
2145Waits for the event to complete.
2146
2147L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clWaitForEvents.html>
2148
2149=item $ev->cb ($exec_callback_type, $callback->($event, $event_command_exec_status))
2150
2151Adds a callback to the callback stack for the given event type. There is
2152no way to remove a callback again.
2153
2154L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetEventCallback.html>
2155
523=item $packed_value = $ev->info ($name) 2156=item $packed_value = $ev->info ($name)
524 2157
525See C<< $platform->info >> for details. 2158See C<< $platform->info >> for details.
526 2159
527L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetEventInfo.html> 2160L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetEventInfo.html>
528 2161
529=item $ev->wait 2162=for gengetinfo begin event
530 2163
531Waits for the event to complete. 2164=item $queue = $event->command_queue
532 2165
2166Calls C<clGetEventInfo> with C<OpenCL::EVENT_COMMAND_QUEUE> and returns the result.
2167
2168=item $command_type = $event->command_type
2169
2170Calls C<clGetEventInfo> with C<OpenCL::EVENT_COMMAND_TYPE> and returns the result.
2171
2172=item $uint = $event->reference_count
2173
2174Calls C<clGetEventInfo> with C<OpenCL::EVENT_REFERENCE_COUNT> and returns the result.
2175
2176=item $uint = $event->command_execution_status
2177
2178Calls C<clGetEventInfo> with C<OpenCL::EVENT_COMMAND_EXECUTION_STATUS> and returns the result.
2179
2180=item $ctx = $event->context
2181
2182Calls C<clGetEventInfo> with C<OpenCL::EVENT_CONTEXT> and returns the result.
2183
2184=for gengetinfo end event
2185
2186=item $packed_value = $ev->profiling_info ($name)
2187
2188See C<< $platform->info >> for details.
2189
2190The reason this method is not called C<info> is that there already is an
2191C<< ->info >> method.
2192
533L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clWaitForEvents.html> 2193L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetProfilingInfo.html>
2194
2195=for gengetinfo begin profiling
2196
2197=item $ulong = $event->profiling_command_queued
2198
2199Calls C<clGetEventProfilingInfo> with C<OpenCL::PROFILING_COMMAND_QUEUED> and returns the result.
2200
2201=item $ulong = $event->profiling_command_submit
2202
2203Calls C<clGetEventProfilingInfo> with C<OpenCL::PROFILING_COMMAND_SUBMIT> and returns the result.
2204
2205=item $ulong = $event->profiling_command_start
2206
2207Calls C<clGetEventProfilingInfo> with C<OpenCL::PROFILING_COMMAND_START> and returns the result.
2208
2209=item $ulong = $event->profiling_command_end
2210
2211Calls C<clGetEventProfilingInfo> with C<OpenCL::PROFILING_COMMAND_END> and returns the result.
2212
2213=for gengetinfo end profiling
534 2214
535=back 2215=back
536 2216
537=head2 THE OpenCL::UserEvent CLASS 2217=head2 THE OpenCL::UserEvent CLASS
538 2218
540 2220
541=over 4 2221=over 4
542 2222
543=item $ev->set_status ($execution_status) 2223=item $ev->set_status ($execution_status)
544 2224
2225Sets the execution status of the user event. Can only be called once,
2226either with OpenCL::COMPLETE or a negative number as status.
2227
2228execution_status: OpenCL::COMPLETE or a negative integer.
2229
545L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetUserEventStatus.html> 2230L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetUserEventStatus.html>
546 2231
547=back 2232=back
548 2233
2234=head2 THE OpenCL::Mapped CLASS
2235
2236This class represents objects mapped into host memory. They are
2237represented by a blessed string scalar. The string data is the mapped
2238memory area, that is, if you read or write it, then the mapped object is
2239accessed directly.
2240
2241You must only ever use operations that modify the string in-place - for
2242example, a C<substr> that doesn't change the length, or maybe a regex that
2243doesn't change the length. Any other operation might cause the data to be
2244copied.
2245
2246When the object is destroyed it will enqueue an implicit unmap operation
2247on the queue that was used to create it.
2248
2249Keep in mind that you I<need> to unmap (or destroy) mapped objects before
2250OpenCL sees the changes, even if some implementations don't need this
2251sometimes.
2252
2253Example, replace the first two floats in the mapped buffer by 1 and 2.
2254
2255 my $mapped = $queue->map_buffer ($buf, ...
2256 $mapped->event->wait; # make sure it's there
2257
2258 # now replace first 8 bytes by new data, which is exactly 8 bytes long
2259 # we blindly assume device endianness to equal host endianness
2260 # (and of course, we assume iee 754 single precision floats :)
2261 substr $$mapped, 0, 8, pack "f*", 1, 2;
2262
2263=over 4
2264
2265=item $ev = $mapped->unmap ($wait_events...)
2266
2267Unmaps the mapped memory object, using the queue originally used to create
2268it, quite similarly to C<< $queue->unmap ($mapped, ...) >>.
2269
2270=item $bool = $mapped->mapped
2271
2272Returns whether the object is still mapped - true before an C<unmap> is
2273enqueued, false afterwards.
2274
2275=item $ev = $mapped->event
2276
2277Return the event object associated with the mapped object. Initially, this
2278will be the event object created when mapping the object, and after an
2279unmap, this will be the event object that the unmap operation created.
2280
2281=item $mapped->wait
2282
2283Same as C<< $mapped->event->wait >> - makes sure no operations on this
2284mapped object are outstanding.
2285
2286=item $bytes = $mapped->size
2287
2288Returns the size of the mapped area, in bytes. Same as C<length $$mapped>.
2289
2290=item $ptr = $mapped->ptr
2291
2292Returns the raw memory address of the mapped area.
2293
2294=item $mapped->set ($offset, $data)
2295
2296Replaces the data at the given C<$offset> in the memory area by the new
2297C<$data>. This method is safer than direct manipulation of C<$mapped>
2298because it does bounds-checking, but also slower.
2299
2300=item $data = $mapped->get ($offset, $length)
2301
2302Returns (without copying) a scalar representing the data at the given
2303C<$offset> and C<$length> in the mapped memory area. This is the same as
2304the following substr, except much slower;
2305
2306 $data = substr $$mapped, $offset, $length
2307
549=cut 2308=cut
550 2309
551package OpenCL; 2310sub OpenCL::Mapped::get {
552 2311 substr ${$_[0]}, $_[1], $_[2]
553use common::sense;
554
555BEGIN {
556 our $VERSION = '0.03';
557
558 require XSLoader;
559 XSLoader::load (__PACKAGE__, $VERSION);
560
561 @OpenCL::Buffer::ISA =
562 @OpenCL::Image::ISA = OpenCL::Memory::;
563
564 @OpenCL::Image2D::ISA =
565 @OpenCL::Image3D::ISA = OpenCL::Image::;
566
567 @OpenCL::UserEvent::ISA = OpenCL::Event::;
568} 2312}
2313
2314=back
2315
2316=head2 THE OpenCL::MappedBuffer CLASS
2317
2318This is a subclass of OpenCL::Mapped, representing mapped buffers.
2319
2320=head2 THE OpenCL::MappedImage CLASS
2321
2322This is a subclass of OpenCL::Mapped, representing mapped images.
2323
2324=over 4
2325
2326=item $bytes = $mapped->row_pitch
2327
2328=item $bytes = $mapped->slice_pitch
2329
2330Return the row or slice pitch of the image that has been mapped.
2331
2332=back
2333
2334
2335=cut
569 2336
5701; 23371;
571 2338
572=head1 AUTHOR 2339=head1 AUTHOR
573 2340

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines