[ViewVC] Diff of: cvs/OpenCL/OpenCL.pm

Comparing OpenCL/OpenCL.pm (file contents):
Revision 1.30 by root, Thu Apr 5 10:08:59 2012 UTC vs.
Revision 1.45 by root, Sat Apr 21 17:50:27 2012 UTC

…		…
149		149
150	=head2 Create and build a program, then create a kernel out of one of its	150	=head2 Create and build a program, then create a kernel out of one of its
151	functions.	151	functions.
152		152
153	my $src = '	153	my $src = '
154	__kernel void	154	kernel void
155	squareit (__global float input, __global float output)	155	squareit (global float input, global float output)
156	{	156	{
157	$id = get_global_id (0);	157	$id = get_global_id (0);
158	output [id] = input [id] * input [id];	158	output [id] = input [id] * input [id];
159	}	159	}
160	';	160	';
161		161
162	my $prog = $ctx->program_with_source ($src);	162	my $prog = $ctx->program_with_source ($src);
163		163
164	# build croaks on compile errors, so catch it and print the compile errors	164	# build croaks on compile errors, so catch it and print the compile errors
165	eval { $prog->build ($dev); 1 }	165	eval { $prog->build ($dev, "-cl-fast-relaxed-math"); 1 }
166	or die $prog->build_log;	166	or die $prog->build_log;
167		167
168	my $kernel = $prog->kernel ("squareit");	168	my $kernel = $prog->kernel ("squareit");
169		169
170	=head2 Create some input and output float buffers, then call the	170	=head2 Create some input and output float buffers, then call the
…		…
210	# enqueue an async read	210	# enqueue an async read
211	$ev = $queue->enqueue_read_buffer ($output, 0, 0, OpenCL::SIZEOF_FLOAT * 4, my $data, $ev);	211	$ev = $queue->enqueue_read_buffer ($output, 0, 0, OpenCL::SIZEOF_FLOAT * 4, my $data, $ev);
212		212
213	# wait for the last event to complete	213	# wait for the last event to complete
214	$ev->wait;	214	$ev->wait;
		215
		216	=head2 Use the OpenGL module to share a texture between OpenCL and OpenGL and draw some julia
		217	set tunnel effect.
		218
		219	This is quite a long example to get you going.
		220
		221	use OpenGL ":all";
		222	use OpenCL;
		223
		224	# open a window and create a gl texture
		225	OpenGL::glpOpenWindow width => 256, height => 256;
		226	my $texid = glGenTextures_p 1;
		227	glBindTexture GL_TEXTURE_2D, $texid;
		228	glTexImage2D_c GL_TEXTURE_2D, 0, GL_RGBA8, 256, 256, 0, GL_RGBA, GL_UNSIGNED_BYTE, 0;
		229
		230	# find and use the first opencl device that let's us get a shared opengl context
		231	my $platform;
		232	my $dev;
		233	my $ctx;
		234
		235	for (OpenCL::platforms) {
		236	$platform = $_;
		237	for ($platform->devices) {
		238	$dev = $_;
		239	$ctx = $platform->context ([OpenCL::GLX_DISPLAY_KHR, undef, OpenCL::GL_CONTEXT_KHR, undef], [$dev])
		240	and last;
		241	}
		242	}
		243
		244	$ctx
		245	or die "cannot find suitable OpenCL device\n";
		246
		247	my $queue = $ctx->queue ($dev);
		248
		249	# now attach an opencl image2d object to the opengl texture
		250	my $tex = $ctx->gl_texture2d (OpenCL::MEM_WRITE_ONLY, GL_TEXTURE_2D, 0, $texid);
		251
		252	# now the boring opencl code
		253	my $src = <<EOF;
		254	kernel void
		255	juliatunnel (write_only image2d_t img, float time)
		256	{
		257	float2 p = (float2)(get_global_id (0), get_global_id (1)) / 256.f * 2.f - 1.f;
		258
		259	float2 m = (float2)(1.f, p.y) / fabs (p.x);
		260	m.x = fabs (fmod (m.x + time * 0.05f, 4.f)) - 2.f;
		261
		262	float2 z = m;
		263	float2 c = (float2)(sin (time * 0.05005), cos (time * 0.06001));
		264
		265	for (int i = 0; i < 25 && dot (z, z) < 4.f; ++i)
		266	z = (float2)(z.x * z.x - z.y * z.y, 2.f * z.x * z.y) + c;
		267
		268	float3 colour = (float3)(z.x, z.y, z.x * z.y);
		269	write_imagef (img, (int2)(get_global_id (0), get_global_id (1)), (float4)(colour * p.x * p.x, 1.));
		270	}
		271	EOF
		272	my $prog = $ctx->program_with_source ($src);
		273	eval { $prog->build ($dev); 1 }
		274	or die $prog->build_log ($dev);
		275
		276	my $kernel = $prog->kernel ("juliatunnel");
		277
		278	# program compiled, kernel ready, now draw and loop
		279
		280	for (my $time; ; ++$time) {
		281	# acquire objects from opengl
		282	$queue->enqueue_acquire_gl_objects ([$tex]);
		283
		284	# configure and run our kernel
		285	$kernel->set_image2d (0, $tex);
		286	$kernel->set_float (1, $time);
		287	$queue->enqueue_nd_range_kernel ($kernel, undef, [256, 256], undef);
		288
		289	# release objects to opengl again
		290	$queue->enqueue_release_gl_objects ([$tex]);
		291
		292	# wait
		293	$queue->finish;
		294
		295	# now draw the texture, the defaults should be all right
		296	glTexParameterf GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST;
		297
		298	glEnable GL_TEXTURE_2D;
		299	glBegin GL_QUADS;
		300	glTexCoord2f 0, 1; glVertex3i -1, -1, -1;
		301	glTexCoord2f 0, 0; glVertex3i 1, -1, -1;
		302	glTexCoord2f 1, 0; glVertex3i 1, 1, -1;
		303	glTexCoord2f 1, 1; glVertex3i -1, 1, -1;
		304	glEnd;
		305
		306	glXSwapBuffers;
		307
		308	select undef, undef, undef, 1/60;
		309	}
215		310
216	=head1 DOCUMENTATION	311	=head1 DOCUMENTATION
217		312
218	=head2 BASIC CONVENTIONS	313	=head2 BASIC CONVENTIONS
219		314
…		…
241	=item * Structures are often specified by flattening out their components	336	=item * Structures are often specified by flattening out their components
242	as with short vectors, and returned as arrayrefs.	337	as with short vectors, and returned as arrayrefs.
243		338
244	=item * When enqueuing commands, the wait list is specified by adding	339	=item * When enqueuing commands, the wait list is specified by adding
245	extra arguments to the function - anywhere a C<$wait_events...> argument	340	extra arguments to the function - anywhere a C<$wait_events...> argument
246	is documented this can be any number of event objects.	341	is documented this can be any number of event objects. As an extsnion
		342	implemented by this module, C<undef> values will be ignored in the event
		343	list.
247		344
248	=item * When enqueuing commands, if the enqueue method is called in void	345	=item * When enqueuing commands, if the enqueue method is called in void
249	context, no event is created. In all other contexts an event is returned	346	context, no event is created. In all other contexts an event is returned
250	by the method.	347	by the method.
251		348
…		…
271	ulong IV - Q	368	ulong IV - Q
272	float NV float f	369	float NV float f
273	half IV ushort S	370	half IV ushort S
274	double NV double d	371	double NV double d
275		372
		373	=head2 GLX SUPPORT
		374
		375	Due to the sad state that OpenGL support is in in Perl (mostly the OpenGL
		376	module, which has little to no documentation and has little to no support
		377	for glX), this module, as a special extension, treats context creation
		378	properties C<OpenCL::GLX_DISPLAY_KHR> and C<OpenCL::GL_CONTEXT_KHR>
		379	specially: If either or both of these are C<undef>, then the OpenCL
		380	module tries to dynamically resolve C<glXGetCurrentDisplay> and
		381	C<glXGetCurrentContext>, call these functions and use their return values
		382	instead.
		383
		384	For this to work, the OpenGL library must be loaded, a GLX context must
		385	have been created and be made current, and C<dlsym> must be available and
		386	capable of finding the function via C<RTLD_DEFAULT>.
		387
276	=head2 THE OpenCL PACKAGE	388	=head2 THE OpenCL PACKAGE
277		389
278	=over 4	390	=over 4
279		391
280	=item $int = OpenCL::errno	392	=item $int = OpenCL::errno
…		…
638		750
639	=item @device_partition_property_exts = $device->affinity_domains_ext	751	=item @device_partition_property_exts = $device->affinity_domains_ext
640		752
641	Calls C<clGetDeviceInfo> with C<CL_DEVICE_AFFINITY_DOMAINS_EXT> and returns the result.	753	Calls C<clGetDeviceInfo> with C<CL_DEVICE_AFFINITY_DOMAINS_EXT> and returns the result.
642		754
643	=item $uint = $device->reference_count_ext	755	=item $uint = $device->reference_count_ext
644		756
645	Calls C<clGetDeviceInfo> with C<CL_DEVICE_REFERENCE_COUNT_EXT > and returns the result.	757	Calls C<clGetDeviceInfo> with C<CL_DEVICE_REFERENCE_COUNT_EXT > and returns the result.
646		758
647	=item @device_partition_property_exts = $device->partition_style_ext	759	=item @device_partition_property_exts = $device->partition_style_ext
648		760
…		…
659	=item $queue = $ctx->queue ($device, $properties)	771	=item $queue = $ctx->queue ($device, $properties)
660		772
661	Create a new OpenCL::Queue object from the context and the given device.	773	Create a new OpenCL::Queue object from the context and the given device.
662		774
663	L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateCommandQueue.html>	775	L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateCommandQueue.html>
		776
		777	Example: create an out-of-order queue.
		778
		779	$queue = $ctx->queue ($device, OpenCL::QUEUE_OUT_OF_ORDER_EXEC_MODE_ENABLE);
664		780
665	=item $ev = $ctx->user_event	781	=item $ev = $ctx->user_event
666		782
667	Creates a new OpenCL::UserEvent object.	783	Creates a new OpenCL::UserEvent object.
668		784
…		…
691		807
692	Creates a new OpenCL::Image3D object and optionally initialises it with	808	Creates a new OpenCL::Image3D object and optionally initialises it with
693	the given data values.	809	the given data values.
694		810
695	L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateImage3D.html>	811	L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateImage3D.html>
		812
		813	=item $buffer = $ctx->gl_buffer ($flags, $bufobj)
		814
		815	Creates a new OpenCL::Buffer (actually OpenCL::BufferObj) object that refers to the given
		816	OpenGL buffer object.
		817
		818	http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLBuffer.html
		819
		820	=item $ctx->gl_texture2d ($flags, $target, $miplevel, $texture)
		821
		822	Creates a new OpenCL::Image2D object that refers to the given OpenGL
		823	2D texture object.
		824
		825	http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLTexture2D.html
		826
		827	=item $ctx->gl_texture3d ($flags, $target, $miplevel, $texture)
		828
		829	Creates a new OpenCL::Image3D object that refers to the given OpenGL
		830	3D texture object.
		831
		832	http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLTexture3D.html
		833
		834	=item $ctx->gl_renderbuffer ($flags, $renderbuffer)
		835
		836	Creates a new OpenCL::Image2D object that refers to the given OpenGL
		837	render buffer.
		838
		839	http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clCreateFromGLRenderbuffer.html
696		840
697	=item @formats = $ctx->supported_image_formats ($flags, $image_type)	841	=item @formats = $ctx->supported_image_formats ($flags, $image_type)
698		842
699	Returns a list of matching image formats - each format is an arrayref with	843	Returns a list of matching image formats - each format is an arrayref with
700	two values, $channel_order and $channel_type, in it.	844	two values, $channel_order and $channel_type, in it.
…		…
751	for completion, unless the method is called in void context, in which case	895	for completion, unless the method is called in void context, in which case
752	no event object is created.	896	no event object is created.
753		897
754	They also allow you to specify any number of other event objects that this	898	They also allow you to specify any number of other event objects that this
755	request has to wait for before it starts executing, by simply passing the	899	request has to wait for before it starts executing, by simply passing the
756	event objects as extra parameters to the enqueue methods.	900	event objects as extra parameters to the enqueue methods. To simplify
		901	program design, this module ignores any C<undef> values in the list of
		902	events. This makes it possible to code operations such as this, without
		903	having to put a valid event object into C<$event> first:
		904
		905	$event = $queue->enqueue_xxx (..., $event);
757		906
758	Queues execute in-order by default, without any parallelism, so in most	907	Queues execute in-order by default, without any parallelism, so in most
759	cases (i.e. you use only one queue) it's not necessary to wait for or	908	cases (i.e. you use only one queue) it's not necessary to wait for or
760	create event objects.	909	create event objects, althoguh an our of order queue is often a bit
		910	faster.
761		911
762	=over 4	912	=over 4
763		913
764	=item $ev = $queue->enqueue_read_buffer ($buffer, $blocking, $offset, $len, $data, $wait_events...)	914	=item $ev = $queue->enqueue_read_buffer ($buffer, $blocking, $offset, $len, $data, $wait_events...)
765		915
…		…
831	reference to an array of local work sizes, with the same number of	981	reference to an array of local work sizes, with the same number of
832	elements as @$global_work_size.	982	elements as @$global_work_size.
833		983
834	L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueNDRangeKernel.html>	984	L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueNDRangeKernel.html>
835		985
836	=item $ev = $queue->enqueue_marker	986	=item $ev = $queue->enqueue_marker ($wait_events...)
837		987
838	L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueMarker.html>	988	L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueMarker.html>
		989
		990	=item $ev = $queue->enqueue_acquire_gl_objects ([object, ...], $wait_events...)
		991
		992	Enqueues a list (an array-ref of OpenCL::Memory objects) to be acquired
		993	for subsequent OpenCL usage.
		994
		995	L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueAcquireGLObjects.html>
		996
		997	=item $ev = $queue->enqueue_release_gl_objects ([object, ...], $wait_events...)
		998
		999	Enqueues a list (an array-ref of OpenCL::Memory objects) to be released
		1000	for subsequent OpenGL usage.
		1001
		1002	L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueReleaseGLObjects.html>
839		1003
840	=item $ev = $queue->enqueue_wait_for_events ($wait_events...)	1004	=item $ev = $queue->enqueue_wait_for_events ($wait_events...)
841		1005
842	L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWaitForEvents.html>	1006	L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clEnqueueWaitForEvents.html>
843		1007
…		…
931	=item $int = $mem->offset	1095	=item $int = $mem->offset
932		1096
933	Calls C<clGetMemObjectInfo> with C<CL_MEM_OFFSET> and returns the result.	1097	Calls C<clGetMemObjectInfo> with C<CL_MEM_OFFSET> and returns the result.
934		1098
935	=for gengetinfo end mem	1099	=for gengetinfo end mem
		1100
		1101	=item ($type, $name) = $mem->gl_object_info
		1102
		1103	Returns the OpenGL object type (e.g. OpenCL::GL_OBJECT_TEXTURE2D) and the
		1104	object "name" (e.g. the texture name) used to create this memory object.
		1105
		1106	L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clGetGLObjectInfo.html>
936		1107
937	=back	1108	=back
938		1109
939	=head2 THE OpenCL::Buffer CLASS	1110	=head2 THE OpenCL::Buffer CLASS
940		1111
…		…
1000		1171
1001	Calls C<clGetImageInfo> with C<CL_IMAGE_DEPTH> and returns the result.	1172	Calls C<clGetImageInfo> with C<CL_IMAGE_DEPTH> and returns the result.
1002		1173
1003	=for gengetinfo end image	1174	=for gengetinfo end image
1004		1175
		1176	=for gengetinfo begin gl_texture
		1177
		1178	=item $GLenum = $gl_texture->target
		1179
		1180	Calls C<clGetGLTextureInfo> with C<CL_GL_TEXTURE_TARGET> and returns the result.
		1181
		1182	=item $GLint = $gl_texture->gl_mipmap_level
		1183
		1184	Calls C<clGetGLTextureInfo> with C<CL_GL_MIPMAP_LEVEL> and returns the result.
		1185
		1186	=for gengetinfo end gl_texture
		1187
1005	=back	1188	=back
1006		1189
1007	=head2 THE OpenCL::Sampler CLASS	1190	=head2 THE OpenCL::Sampler CLASS
1008		1191
1009	=over 4	1192	=over 4
…		…
1198		1381
1199	This is a family of methods to set the kernel argument with the number C<$index> to the give C<$value>.	1382	This is a family of methods to set the kernel argument with the number C<$index> to the give C<$value>.
1200		1383
1201	TYPE is one of C<char>, C<uchar>, C<short>, C<ushort>, C<int>, C<uint>,	1384	TYPE is one of C<char>, C<uchar>, C<short>, C<ushort>, C<int>, C<uint>,
1202	C<long>, C<ulong>, C<half>, C<float>, C<double>, C<memory>, C<buffer>,	1385	C<long>, C<ulong>, C<half>, C<float>, C<double>, C<memory>, C<buffer>,
1203	C<image2d>, C<image3d>, C<sampler> or C<event>.	1386	C<image2d>, C<image3d>, C<sampler>, C<local> or C<event>.
1204		1387
1205	Chars and integers (including the half type) are specified as integers,	1388	Chars and integers (including the half type) are specified as integers,
1206	float and double as floating point values, memory/buffer/image2d/image3d	1389	float and double as floating point values, memory/buffer/image2d/image3d
1207	must be an object of that type or C<undef>, and sampler and event must be	1390	must be an object of that type or C<undef>, local-memory arguments are
1208	objects of that type.	1391	set by specifying the size, and sampler and event must be objects of that
		1392	type.
1209		1393
1210	L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetKernelArg.html>	1394	L<http://www.khronos.org/registry/cl/sdk/1.1/docs/man/xhtml/clSetKernelArg.html>
1211		1395
1212	=back	1396	=back
1213		1397
…		…
1302	package OpenCL;	1486	package OpenCL;
1303		1487
1304	use common::sense;	1488	use common::sense;
1305		1489
1306	BEGIN {	1490	BEGIN {
1307	our $VERSION = '0.91';	1491	our $VERSION = '0.96';
1308		1492
1309	require XSLoader;	1493	require XSLoader;
1310	XSLoader::load (__PACKAGE__, $VERSION);	1494	XSLoader::load (__PACKAGE__, $VERSION);
1311		1495
1312	@OpenCL::Buffer::ISA =	1496	@OpenCL::Buffer::ISA =

Diff Legend

-–
+Removed lines
-+
+Added lines
-<
+Changed lines
->
+Changed lines

Comparing OpenCL/OpenCL.pm (file contents): Revision 1.30 by root, Thu Apr 5 10:08:59 2012 UTC vs. Revision 1.45 by root, Sat Apr 21 17:50:27 2012 UTC

Diff Legend

Comparing OpenCL/OpenCL.pm (file contents):
Revision 1.30 by root, Thu Apr 5 10:08:59 2012 UTC vs.
Revision 1.45 by root, Sat Apr 21 17:50:27 2012 UTC