org.lwjgl.opencl.AMDBusAddressableMemory Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of lwjgl-opencl Show documentation
Show all versions of lwjgl-opencl Show documentation
An open, royalty-free standard for cross-platform, parallel programming of diverse processors found in personal computers, servers, mobile devices and embedded platforms.
/*
* Copyright LWJGL. All rights reserved.
* License terms: https://www.lwjgl.org/license
* MACHINE GENERATED FILE, DO NOT EDIT
*/
package org.lwjgl.opencl;
import javax.annotation.*;
import org.lwjgl.*;
import org.lwjgl.system.*;
import static org.lwjgl.system.Checks.*;
import static org.lwjgl.system.JNI.*;
import static org.lwjgl.system.MemoryUtil.*;
/**
* Native bindings to the amd_bus_addressable_memory extension.
*
* This extension defines an API that allows improved control of the physical memory used by the graphics device.
*
* It allows to share a memory allocated by the Graphics driver to be used by other device on the bus by exposing a write-only bus address. One example of
* application would be a video capture device which would DMA into the GPU memory.
*
* It also offers the reverse operation of specifying a buffer allocated on another device to be used for write access by the GPU.
*/
public class AMDBusAddressableMemory {
/**
* Accepted by the {@code flags} parameter of {@link CL10#clCreateBuffer CreateBuffer}.
*
* This flag specifies that the application wants the OpenCL implementation to create a buffer that can be accessed by remote device DMA.
*
* {@link #CL_MEM_BUS_ADDRESSABLE_AMD MEM_BUS_ADDRESSABLE_AMD}, {@link CL10#CL_MEM_ALLOC_HOST_PTR MEM_ALLOC_HOST_PTR} and {@link CL10#CL_MEM_USE_HOST_PTR MEM_USE_HOST_PTR} are mutually exclusive.
*/
public static final int CL_MEM_BUS_ADDRESSABLE_AMD = 1<<30;
/**
* Accepted by the {@code flags} parameter of {@link CL10#clCreateBuffer CreateBuffer}.
*
* This flag specifies that the application wants the OpenCL implementation to create a buffer from an already allocated memory on remote device.
*
* {@link #CL_MEM_EXTERNAL_PHYSICAL_AMD MEM_EXTERNAL_PHYSICAL_AMD}, {@link CL10#CL_MEM_ALLOC_HOST_PTR MEM_ALLOC_HOST_PTR}, {@link CL10#CL_MEM_COPY_HOST_PTR MEM_COPY_HOST_PTR} and {@link CL10#CL_MEM_USE_HOST_PTR MEM_USE_HOST_PTR} are mutually exclusive.
*
* {@link #CL_MEM_EXTERNAL_PHYSICAL_AMD MEM_EXTERNAL_PHYSICAL_AMD}, {@link CL10#CL_MEM_READ_WRITE MEM_READ_WRITE} and {@link CL10#CL_MEM_READ_ONLY MEM_READ_ONLY} are mutually exclusive.
*/
public static final int CL_MEM_EXTERNAL_PHYSICAL_AMD = 1<<31;
/** New command types for the events returned by the amd_bus_addressable_memory functions. */
public static final int
CL_COMMAND_WAIT_SIGNAL_AMD = 0x4080,
CL_COMMAND_WRITE_SIGNAL_AMD = 0x4081,
CL_COMMAND_MAKE_BUFFERS_RESIDENT_AMD = 0x4082;
protected AMDBusAddressableMemory() {
throw new UnsupportedOperationException();
}
static boolean isAvailable(CLCapabilities caps) {
return checkFunctions(
caps.clEnqueueWaitSignalAMD, caps.clEnqueueWriteSignalAMD, caps.clEnqueueMakeBuffersResidentAMD
);
}
// --- [ clEnqueueWaitSignalAMD ] ---
/**
* Unsafe version of: {@link #clEnqueueWaitSignalAMD EnqueueWaitSignalAMD}
*
* @param num_events_in_wait_list the number of events in {@code event_wait_list}
*/
public static int nclEnqueueWaitSignalAMD(long command_queue, long mem_object, int value, int num_events_in_wait_list, long event_wait_list, long event) {
long __functionAddress = CL.getICD().clEnqueueWaitSignalAMD;
if (CHECKS) {
check(__functionAddress);
check(command_queue);
check(mem_object);
}
return callPPPPI(command_queue, mem_object, value, num_events_in_wait_list, event_wait_list, event, __functionAddress);
}
/**
* Instructs the OpenCL to wait until {@code value} is written to {@code buffer} before issuing the next command.
*
* @param command_queue a command-queue
* @param mem_object a memory object
* @param value the signal value
* @param event_wait_list a list of events that need to complete before this particular command can be executed. If {@code event_wait_list} is {@code NULL}, then this particular command
* does not wait on any event to complete. The events specified in {@code event_wait_list} act as synchronization points. The context associated with events in
* {@code event_wait_list} and {@code command_queue} must be the same.
* @param event Returns an event object that identifies this particular command and can be used to query or queue a wait for this particular command to complete.
* {@code event} can be {@code NULL} in which case it will not be possible for the application to query the status of this command or queue a wait for this command to
* complete. If the {@code event_wait_list} and the {@code event} arguments are not {@code NULL}, the event argument should not refer to an element of the
* {@code event_wait_list} array.
*
* @return {@link CL10#CL_SUCCESS SUCCESS} if the function is executed successfully. Otherwise, it returns one of the following errors:
*
*
* - {@link CL10#CL_INVALID_MEM_OBJECT INVALID_MEM_OBJECT} is generated if the {@code buffer} parameter of clEnqueueWaitSignalAMD is not a valid buffer.
* - {@link CL10#CL_INVALID_COMMAND_QUEUE INVALID_COMMAND_QUEUE} is generated if the {@code command_queue} parameter of clEnqueueWaitSignalAMD is not a valid command queue.
* - {@link CL10#CL_INVALID_MEM_OBJECT INVALID_MEM_OBJECT} is generated if the {@code buffer} parameter of clEnqueueWaitSignalAMD does not represent a buffer allocated with
* {@link #CL_MEM_BUS_ADDRESSABLE_AMD MEM_BUS_ADDRESSABLE_AMD}.
* - {@link CL10#CL_INVALID_VALUE INVALID_VALUE} is generated if the signal address used by clEnqueueWaitSignalAMD of {@code bufffer} is invalid (for example 0).
*
*/
@NativeType("cl_int")
public static int clEnqueueWaitSignalAMD(@NativeType("cl_command_queue") long command_queue, @NativeType("cl_mem") long mem_object, @NativeType("cl_uint") int value, @Nullable @NativeType("cl_event const *") PointerBuffer event_wait_list, @Nullable @NativeType("cl_event *") PointerBuffer event) {
if (CHECKS) {
checkSafe(event, 1);
}
return nclEnqueueWaitSignalAMD(command_queue, mem_object, value, remainingSafe(event_wait_list), memAddressSafe(event_wait_list), memAddressSafe(event));
}
// --- [ clEnqueueWriteSignalAMD ] ---
/**
* Unsafe version of: {@link #clEnqueueWriteSignalAMD EnqueueWriteSignalAMD}
*
* @param num_events_in_wait_list the number of events in {@code event_wait_list}
*/
public static int nclEnqueueWriteSignalAMD(long command_queue, long mem_object, int value, long offset, int num_events_in_wait_list, long event_wait_list, long event) {
long __functionAddress = CL.getICD().clEnqueueWriteSignalAMD;
if (CHECKS) {
check(__functionAddress);
check(command_queue);
check(mem_object);
}
return callPPJPPI(command_queue, mem_object, value, offset, num_events_in_wait_list, event_wait_list, event, __functionAddress);
}
/**
* This command instructs the OpenCL to write {@code value} to the signal address + {@code offset} of {@code buffer} (which must be a buffer created with
* {@link #CL_MEM_EXTERNAL_PHYSICAL_AMD MEM_EXTERNAL_PHYSICAL_AMD}). This should be done after a write operation by the device into that buffer is complete. Consecutive marker values must
* keep increasing.
*
* @param command_queue a command-queue
* @param mem_object a memory object
* @param value the signal value
* @param offset the write offset
* @param event_wait_list a list of events that need to complete before this particular command can be executed. If {@code event_wait_list} is {@code NULL}, then this particular command
* does not wait on any event to complete. The events specified in {@code event_wait_list} act as synchronization points. The context associated with events in
* {@code event_wait_list} and {@code command_queue} must be the same.
* @param event Returns an event object that identifies this particular command and can be used to query or queue a wait for this particular command to complete.
* {@code event} can be {@code NULL} in which case it will not be possible for the application to query the status of this command or queue a wait for this command to
* complete. If the {@code event_wait_list} and the {@code event} arguments are not {@code NULL}, the event argument should not refer to an element of the
* {@code event_wait_list} array.
*
* @return {@link CL10#CL_SUCCESS SUCCESS} if the function is executed successfully. Otherwise, it returns one of the following errors:
*
*
* - {@link CL10#CL_INVALID_MEM_OBJECT INVALID_MEM_OBJECT} is generated if the {@code buffer} parameter of clEnqueueWriteSignalAMD is not a valid buffer.
* - {@link CL10#CL_INVALID_COMMAND_QUEUE INVALID_COMMAND_QUEUE} is generated if the {@code command_queue} parameter of clEnqueueWriteSignalAMD is not a valid command queue.
* - {@link CL10#CL_INVALID_MEM_OBJECT INVALID_MEM_OBJECT} is generated if the {@code buffer} parameter of clEnqueueWriteSignalAMD does not represent a buffer defined as
* {@link #CL_MEM_EXTERNAL_PHYSICAL_AMD MEM_EXTERNAL_PHYSICAL_AMD}.
* - {@link CL10#CL_INVALID_BUFFER_SIZE INVALID_BUFFER_SIZE} is generated if the {@code offset} parameter of clEnqueueWriteSignalAMD would lead to a write beyond the size of
* {@code buffer}.
* - {@link CL10#CL_INVALID_VALUE INVALID_VALUE} is generated if the signal address used by clEnqueueWriteSignalAMD of {@code bufffer} is invalid (for example 0).
*
*/
@NativeType("cl_int")
public static int clEnqueueWriteSignalAMD(@NativeType("cl_command_queue") long command_queue, @NativeType("cl_mem") long mem_object, @NativeType("cl_uint") int value, @NativeType("cl_ulong") long offset, @Nullable @NativeType("cl_event const *") PointerBuffer event_wait_list, @Nullable @NativeType("cl_event *") PointerBuffer event) {
if (CHECKS) {
checkSafe(event, 1);
}
return nclEnqueueWriteSignalAMD(command_queue, mem_object, value, offset, remainingSafe(event_wait_list), memAddressSafe(event_wait_list), memAddressSafe(event));
}
// --- [ clEnqueueMakeBuffersResidentAMD ] ---
/**
* Unsafe version of: {@link #clEnqueueMakeBuffersResidentAMD EnqueueMakeBuffersResidentAMD}
*
* @param num_mem_objs the number of memory objects in {@code mem_objects}
* @param num_events_in_wait_list the number of events in {@code event_wait_list}
*/
public static int nclEnqueueMakeBuffersResidentAMD(long command_queue, int num_mem_objs, long mem_objects, int blocking_make_resident, long bus_addresses, int num_events_in_wait_list, long event_wait_list, long event) {
long __functionAddress = CL.getICD().clEnqueueMakeBuffersResidentAMD;
if (CHECKS) {
check(__functionAddress);
check(command_queue);
}
return callPPPPPI(command_queue, num_mem_objs, mem_objects, blocking_make_resident, bus_addresses, num_events_in_wait_list, event_wait_list, event, __functionAddress);
}
/**
* The application requires the bus address in order to access the buffers from a remote device. As the OS may rearrange buffers to make space for other
* memory allocation, we must make the buffers resident before trying to access them on remote device.
*
* This function is used to make buffers resident.
*
* @param command_queue a command-queue
* @param mem_objects a pointer to a list of memory objects created with {@link #CL_MEM_BUS_ADDRESSABLE_AMD MEM_BUS_ADDRESSABLE_AMD} flag
* @param blocking_make_resident indicates if read operation is blocking or non-blocking
* @param bus_addresses a pointer to a list of {@link CLBusAddressAMD} structures
* @param event_wait_list a list of events that need to complete before this particular command can be executed. If {@code event_wait_list} is {@code NULL}, then this particular command
* does not wait on any event to complete. The events specified in {@code event_wait_list} act as synchronization points. The context associated with events in
* {@code event_wait_list} and {@code command_queue} must be the same.
* @param event Returns an event object that identifies this particular command and can be used to query or queue a wait for this particular command to complete.
* {@code event} can be {@code NULL} in which case it will not be possible for the application to query the status of this command or queue a wait for this command to
* complete. If the {@code event_wait_list} and the {@code event} arguments are not {@code NULL}, the event argument should not refer to an element of the
* {@code event_wait_list} array.
*
* @return {@link CL10#CL_SUCCESS SUCCESS} if the function is executed successfully. Otherwise, it returns one of the following errors:
*
*
* - {@link CL10#CL_INVALID_OPERATION INVALID_OPERATION} is generated if any of the pointer parameters of clEnqueueMakeBuffersResidentAMD are {@code NULL} (and count is > 0).
* - {@link CL10#CL_INVALID_OPERATION INVALID_OPERATION} is generated if any of the mem_objects passed to clEnqueueMakeBuffersResidentAMD was not a valid cl_mem object created with
* {@link #CL_MEM_BUS_ADDRESSABLE_AMD MEM_BUS_ADDRESSABLE_AMD} flag.
* - {@link CL10#CL_OUT_OF_HOST_MEMORY OUT_OF_HOST_MEMORY} is generated if any of the mem_objects passed to clEnqueueMakeBuffersResidentAMD could not be made resident so that the buffer
* or signal bus addresses will be returned as 0.
*
*/
@NativeType("cl_int")
public static int clEnqueueMakeBuffersResidentAMD(@NativeType("cl_command_queue") long command_queue, @NativeType("cl_mem const *") PointerBuffer mem_objects, @NativeType("cl_bool") boolean blocking_make_resident, @NativeType("cl_bus_address_amd *") CLBusAddressAMD.Buffer bus_addresses, @Nullable @NativeType("cl_event const *") PointerBuffer event_wait_list, @Nullable @NativeType("cl_event *") PointerBuffer event) {
if (CHECKS) {
check(bus_addresses, mem_objects.remaining());
checkSafe(event, 1);
}
return nclEnqueueMakeBuffersResidentAMD(command_queue, mem_objects.remaining(), memAddress(mem_objects), blocking_make_resident ? 1 : 0, bus_addresses.address(), remainingSafe(event_wait_list), memAddressSafe(event_wait_list), memAddressSafe(event));
}
}