Uses of Class
at.dms.kjc.backendSupport.SchedulingPhase

Packages that use SchedulingPhase

at.dms.kjc.backendSupport	Package documentation for at.dms.kjc.backendSupport
at.dms.kjc.cell
at.dms.kjc.spacetime
at.dms.kjc.vanillaSlice	This package was written to provide a test for the classes in backendSupport.

Uses of SchedulingPhase in at.dms.kjc.backendSupport

Fields in at.dms.kjc.backendSupport declared as SchedulingPhase

protected SchedulingPhase	ProcessOutputSliceNode.whichPhase
protected SchedulingPhase	ProcessInputSliceNode.whichPhase
protected SchedulingPhase	ProcessFilterSliceNode.whichPhase

Methods in at.dms.kjc.backendSupport that return SchedulingPhase

static SchedulingPhase	SchedulingPhase.valueOf(String name) Returns the enum constant of this type with the specified name.
static SchedulingPhase[]	SchedulingPhase.values() Returns an array containing the constants of this enum type, in the order they're declared.

Methods in at.dms.kjc.backendSupport with parameters of type SchedulingPhase

int	FilterInfo.getMult(SchedulingPhase whichPhase)
static boolean	SchedulingPhase.isInitOrPrimepump(SchedulingPhase whichPhase) to determine if INIT || PRIMEPUMP.
protected void	BackEndScaffold.iterateInorder(Slice[] slices, SchedulingPhase whichPhase, ComputeNodesI computeNodes) Iterate over the schedule of slices and over each node of each slice and generate the code necessary to fire the schedule.
abstract void	BackEndFactory.processFilterSliceNode(FilterSliceNode filter, SchedulingPhase whichPhase, ComputeNodesType computeNodes) Process a filter slice node: find the correct ProcElement(s) and add code and buffers.
abstract void	BackEndFactory.processFilterSlices(Slice slice, SchedulingPhase whichPhase, ComputeNodesType computeNodes) Process all filter slice nodes in a Slice (just one in a SimpleSlice): find the correct ProcElement(s) and add filter code.
abstract void	BackEndFactory.processInputSliceNode(InputSliceNode input, SchedulingPhase whichPhase, ComputeNodesType computeNodes) Process an input slice node: find the correct ProcElement(s) and add joiner code, and buffers.
abstract void	BackEndFactory.processOutputSliceNode(OutputSliceNode output, SchedulingPhase whichPhase, ComputeNodesType computeNodes) Process an output slice node: find the correct ProcElement(s) and add splitter code, and buffers.
int	FilterInfo.totalItemsReceived(SchedulingPhase whichPhase) get the total number of items received during the execution of the stage we are in (based on whichPhase.
int	FilterInfo.totalItemsSent(SchedulingPhase whichPhase) get the total number of itmes sent during the execution of the stage we are in (based on whichPhase.

Constructors in at.dms.kjc.backendSupport with parameters of type SchedulingPhase

ProcessFilterSliceNode(FilterSliceNode filterNode, SchedulingPhase whichPhase, BackEndFactory backEndBits)

ProcessInputSliceNode(InputSliceNode inputNode, SchedulingPhase whichPhase, BackEndFactory backEndBits)
Constructor

ProcessOutputSliceNode(OutputSliceNode outputNode, SchedulingPhase whichPhase, BackEndFactory backEndBits)
Constructor

Uses of SchedulingPhase in at.dms.kjc.cell

Methods in at.dms.kjc.cell with parameters of type SchedulingPhase

void	CellComputeCodeStore.attachInputChannelArray(int filterId, LinkedList<Integer> inputIds, SchedulingPhase whichPhase) Initialize an array of input buffers for this slice and attaches each input buffer to the appropriate upstream channel.
void	CellBackendFactory.processFilterSliceNode(FilterSliceNode filter, SchedulingPhase whichPhase, CellChip computeNodes)
void	CellBackendFactory.processFilterSlices(Slice slice, SchedulingPhase whichPhase, CellChip computeNodes)
void	CellBackendFactory.processInputSliceNode(InputSliceNode input, SchedulingPhase whichPhase, CellChip computeNodes)
void	CellBackendFactory.processOutputSliceNode(OutputSliceNode output, SchedulingPhase whichPhase, CellChip computeNodes)
JBlock	CellComputeCodeStore.setupFilterDescriptionWorkFunc(SliceNode sliceNode, SchedulingPhase whichPhase) Set the filter description's work func to be the appropriate work func depending on the phase (init or steady)
JBlock	CellComputeCodeStore.setupPSPIOBytes(SliceNode sliceNode, SchedulingPhase whichPhase) Set up the [pop/peek_extra/push]_bytes for the PSP params.
JBlock	CellComputeCodeStore.setupPSPNumIOputs(SliceNode sliceNode, SchedulingPhase whichPhase)

Constructors in at.dms.kjc.cell with parameters of type SchedulingPhase

CellProcessFilterSliceNode(FilterSliceNode filterNode, SchedulingPhase whichPhase, CellBackendFactory backEndBits)

CellProcessInputSliceNode(InputSliceNode inputNode, SchedulingPhase whichPhase, CellBackendFactory backEndBits)

CellProcessOutputSliceNode(OutputSliceNode outputNode, SchedulingPhase whichPhase, CellBackendFactory backEndBits)

Uses of SchedulingPhase in at.dms.kjc.spacetime

Methods in at.dms.kjc.spacetime with parameters of type SchedulingPhase

void	RawComputeCodeStore.addDRAMCommand(boolean read, SchedulingPhase whichPhase, int bytes, OffChipBuffer buffer, boolean staticNet) Add a dram command and possibly a rotation of the buffer at the current position of this code compute code for either the init stage (including primepump) or the steady state.
void	RawComputeCodeStore.addDRAMGDNReadCommand(SchedulingPhase whichPhase, int bytes, OffChipBuffer buffer, boolean presynched, RawTile dest) Add a dram read command at the current position of this code compute code for either the init stage (including primepump) or the steady state over the gdn and send it to the
void	RawComputeCodeStore.addFileGDNReadCommand(SchedulingPhase whichPhase, int words, OffChipBuffer buffer, RawTile dest) Generate a command to read from a file daisy-chained to a i/o port on the gdn and send the data to
void	RawComputeCodeStore.addGDNStoreCommandWithSynch(SchedulingPhase whichPhase, int bytes, OffChipBuffer buffer) If we are generating a DRAM command for a store over the gdn and the tile that is writing the data is different from the home tile, and we cannot use any other synchronization, we must send a word over the static network to the tile writing the data for the store from this tile to tell it that we have sent the dram command and it can begin the store.
void	RawBackEndFactory.processFilterSliceNode(FilterSliceNode filter, SchedulingPhase whichPhase, RawChip rawChip) Process a filter slice node: find the correct ProcElement(s) and add code and buffers.
void	RawBackEndFactory.processFilterSlices(Slice slice, SchedulingPhase whichPhase, RawChip rawChip) Process all filter slice nodes in a Slice (just one in a SimpleSlice): find the correct ProcElement(s) and add filter code.
static void	Rawify.processFilterSlices(Slice slice, SchedulingPhase whichPhase, RawProcElements rawChip) First generate the dram commands for input/output to/from the first/last filters of the trace.
void	RawBackEndFactory.processInputSliceNode(InputSliceNode input, SchedulingPhase whichPhase, RawChip rawChip) Process an input slice node: find the correct ProcElement(s) and add joiner code, and buffers.
static void	Rawify.processInputSliceNode(InputSliceNode traceNode, SchedulingPhase whichPhase, RawProcElements rawChip) Create dram commands and switch code to implement the joining described by the input trace node.
void	RawBackEndFactory.processOutputSliceNode(OutputSliceNode output, SchedulingPhase whichPhase, RawChip rawChip) Process an output slice node: find the correct ProcElement(s) and add splitter code and buffers.
static void	Rawify.processOutputSliceNode(OutputSliceNode traceNode, SchedulingPhase whichPhase, RawProcElements rawChip) Create the dram commands and the switch code to implement the splitting described by the output trace node.
protected JStatement	RawExecutionCode.setupGDNStore(SchedulingPhase whichPhase)
boolean	RawComputeCodeStore.shouldPresynch(boolean read, SchedulingPhase whichPhase, OffChipBuffer buffer) Decide if a dram command should be presynched at the given time based on the command and the buffer.

Uses of SchedulingPhase in at.dms.kjc.vanillaSlice

Methods in at.dms.kjc.vanillaSlice with parameters of type SchedulingPhase

void	UniBackEndFactory.processFilterSliceNode(FilterSliceNode filter, SchedulingPhase whichPhase, UniProcessors computeNodes) Process a filter slice node: find the correct ProcElement(s) and add code and buffers.
void	UniBackEndFactory.processFilterSlices(Slice slice, SchedulingPhase whichPhase, UniProcessors computeNodes) Process all filter slice nodes in a Slice (just one in a SimpleSlice): find the correct ProcElement(s) and add filter code.
void	UniBackEndFactory.processInputSliceNode(InputSliceNode input, SchedulingPhase whichPhase, UniProcessors computeNodes) Process an input slice node: find the correct ProcElement(s) and add joiner code, and buffers.
void	UniBackEndFactory.processOutputSliceNode(OutputSliceNode output, SchedulingPhase whichPhase, UniProcessors computeNodes) Process an output slice node: find the correct ProcElement(s) and add splitter code, and buffers.