apple.metal.protocol

Interface MTLResourceStateCommandEncoder

All Superinterfaces:

MTLCommandEncoder

public interface MTLResourceStateCommandEncoder
extends MTLCommandEncoder

Method Summary

Modifier and Type	Method and Description
void	updateFence(MTLFence fence) updateFence:
void	updateTextureMappingModeIndirectBufferIndirectBufferOffset(MTLTexture texture, long mode, MTLBuffer indirectBuffer, long indirectBufferOffset) updateTextureMapping:indirectBuffer:indirectBufferOffset:
void	updateTextureMappingModeRegionMipLevelSlice(MTLTexture texture, long mode, MTLRegion region, long mipLevel, long slice) updateTextureMapping:region:mipLevel:slice:mode:
void	updateTextureMappingsModeRegionsMipLevelsSlicesNumRegions(MTLTexture texture, long mode, MTLRegion regions, org.moe.natj.general.ptr.ConstNUIntPtr mipLevels, org.moe.natj.general.ptr.ConstNUIntPtr slices, long numRegions) updateTextureMappings:regions:mipLevels:slices:numRegions:mode:
void	waitForFence(MTLFence fence) waitForFence:

Methods inherited from interface apple.metal.protocol.MTLCommandEncoder

device, endEncoding, insertDebugSignpost, label, popDebugGroup, pushDebugGroup, setLabel

Method Detail

updateFence

void updateFence(MTLFence fence)

updateFence:

Update the fence to capture all GPU work so far enqueued by this encoder.

The fence is updated at kernel submission to maintain global order and prevent deadlock. Drivers may delay fence updates until the end of the encoder. Drivers may also wait on fences at the beginning of an encoder. It is therefore illegal to wait on a fence after it has been updated in the same encoder.

updateTextureMappingModeIndirectBufferIndirectBufferOffset

void updateTextureMappingModeIndirectBufferIndirectBufferOffset(MTLTexture texture,
                                                                long mode,
                                                                MTLBuffer indirectBuffer,
                                                                long indirectBufferOffset)

updateTextureMapping:indirectBuffer:indirectBufferOffset:

Updates mapping for given sparse texture. Updates are driven via a MTLBuffer with the structure format defined by MTLMapIndirectBufferFormat.

struct MTLMapIndirectBufferFormat{ uint32_t numMappings; MTLMapIndirectArguments mappings[numMappings]; }

updateTextureMappingModeRegionMipLevelSlice

void updateTextureMappingModeRegionMipLevelSlice(MTLTexture texture,
                                                 long mode,
                                                 MTLRegion region,
                                                 long mipLevel,
                                                 long slice)

updateTextureMapping:region:mipLevel:slice:mode:

Updates mapping for given sparse texture

updateTextureMappingsModeRegionsMipLevelsSlicesNumRegions

void updateTextureMappingsModeRegionsMipLevelsSlicesNumRegions(MTLTexture texture,
                                                               long mode,
                                                               MTLRegion regions,
                                                               org.moe.natj.general.ptr.ConstNUIntPtr mipLevels,
                                                               org.moe.natj.general.ptr.ConstNUIntPtr slices,
                                                               long numRegions)

updateTextureMappings:regions:mipLevels:slices:numRegions:mode:

Updates multiple regions within a sparse texture.

waitForFence

void waitForFence(MTLFence fence)

waitForFence:

Prevent further GPU work until the fence is reached.

The fence is evaluated at kernel submision to maintain global order and prevent deadlock. Drivers may delay fence updates until the end of the encoder. Drivers may also wait on fences at the beginning of an encoder. It is therefore illegal to wait on a fence after it has been updated in the same encoder.