Allows to replay selected review data instantaneously

Post Proc Replay

Overview

The Post-Proc Replay operator allows to process data from a data source instantaneously.

If the Post-Proc Replay operator is used in connection with a signal reviewer, then data selected in the reviewer can be re-played instantaneously.

The operator should be connected to a data source that provides Stored Data, such as a File Replay ("File Replay") or a Storage ("Storage"). It is often connected to the same source a Y-T Reviewer ("Y-T Reviewer") is connected to, in order to allow the user to select data for post-processing in the Y-T Reviewer.

Please note that for the Post-Proc Replay to work, the system should have been started (press the Start button, or issue a START action).

Operator ports

Input Review data: Any sample type. The connection is not limited to one type of signal.

Output Instantaneous: Any sample type. The connection is not limited to one type of signal.

Properties

Find more information about changing properties here: "Properties Viewer"

PostProcInfo address

type: Address, starts with '$' sign (see "Address - how to address objects and variable parameters")
Select the address of a Reviewer Information List, that will controls this replayer. Look for a variable parameter called PostProcInformation. Press the small button to open the Address Dialog, or copy an address by hand.

You may type or paste the address, or you may press the small button on the right to open the Address Constructor ("Address Constructor")

A typical viewer that can control this Post-Proc Replay, is the Y-T Reviewer ("Y-T Reviewer"). It has a variable parameter called PostProcInformation. You should find the address of this variable of one specific Y-T Reviewer.

MaxBlockSize

type: Integer value
Maximum size of a block of samples that is processed at once. By default this should be 100. Changing this value may degradate performance.

The Post-Proc Replay operator monitors signal selections the user makes in a reviewer. As soon as the user has started to make a selection, the selected samples are processed by this operator at once.
Because it is more efficient to read a block of multiple samples from the source at once, this property is greater than 1. Do not make this value too large, because then this Post Proc Replay operator may influence performance of other components in the system.

ResetEventCode

type: See description
Optional event code that is sent to the output on the first sample of a new block of samples that is processed. This event can be used to reset any calculations at the output of the Post-Proc Replay operator.

You may leave this property empty, if you don't need a processing start mark.
Read more about event markers here: "Events and Markers".

ReadyEventCode

type: See description
Optional event code that is sent to the output on the last sample that is processed. This event can be used to determine if post-processing is ready.

You may leave this property empty, if you don't need a processing end mark.
Read more about event markers here: "Events and Markers".

SyncID

type: See description
A synchronization identifier that determines if this operator should react to a START, STOP or RESET action.

If you have specified a SyncID, and a START, STOP or RESET Action is issued with another SyncID, then the Post-Proc Replay will not react to that action.
For more information about the behavior of the START, STOP and RESET actions in relation to the SyncID setting, see here: "SyncID in relation to START, STOP and RESET".

Caption

type: Word or phrase
The name of the object in the project. This name must not contain '.', '$' nor '@' characters.

For more information about the rules and usage of the Caption property, please refer to "Caption property - background and usage".

Documentation

type: See description
Optional documentation of this object. If this object is an operator, the Documentation text is displayed below the operator symbol.

Variable Parameters

Find more information about Variable Parameters here: "Variable Parameters"

MaxBlockSize

type: See description
Maximum size of a block of samples that is processed at once. The larger this value, the quicker the post-processing runs, but the more other processes in Polybench have to wait.

For more information, see the property MaxBlockSize.

ResetEventCode

Read more about event markers here: "Events and Markers".

ReadyEventCode

type: See description
Optional event code that is sent to the output on the last sample that is processed. This event can be used to determine if post-processing is ready.

Read more about event markers here: "Events and Markers".

SourceTimeStamp

type: Parameter list (see "Parameter List")
Start date and time of the data source

SourceInfo

type: Parameter list (see "Parameter List")
List with information about the current source stream

Functions

Find more information about Functions here: "Operator Functions"

Evaluate

Reads information about the data source that is connected to the input, and shows the information in the variable parameters.

Process

Start processing the specified sample range of the data source.

Parameter:
Specify either the start and end time as [time 1]-[time 2], or the start time and duration as [time];[duration]. For example: '20.4-3:23' or '4:09:25; 30', or '10:2.3;5:00'. Leave empty to re-process the current selection in the referenced reviewer. If no reviewer is specified, then all data from the source is processed.

Details

The Post-Proc Replay does not try to output data in a real-time fashion, but in stead re-plays it as fast as it can. Using this operator, you can simulate an instantaneous calculation on a selected range of data in a data source, and show the result at once.

Use the Post-Proc Replay in the same way you use a normal File Replay. You can connect any calculation operators, viewers and storages at the output of this operator. Optionally, you can specify an event to mark the first sample outputted, and an event to mark the last sample of a selection, so that your configuration can determine one complete stream of data. See the example below for a demonstration.

The Post-Proc Replay may process samples very fast, or less fast if the computer is currently busy doing other things. In general, the Post-Proc Replay tries to fill up the processing power of the computer, so that other Polybench processes are not disturbed, but in a way the the Post-Proc Replay is as fast as possible.

The Post-Proc Replay can also be used without a Y-T Reviewer. It is then controlled by its function Process. The Process function either specifies a start and end time in the data source, or a start time and a duration. If no data range is specified, then either the complete data source is played by the Post-Proc Replay, or in case a reviewer is linked, the last selection is re-processed.

Limitations

Please note that you cannot connect the outputs of two Post Proc Replay operators, nor the outputs of Post Proc Replay and normal File Replay operators to the same calculation operator. Because the Post Proc Replay processes data at a different speed than the other operators, Polybench cannot combine the data streams of those sources.

Be careful using Actions in a processing network behind a Post-Proc Replay. Data is replayed so quickly, that any actions may be called even after processing is already done. Trying to set a parameter using the SET_PARAMETER action in an operator that is connected to the Post-Proc Replay will probably result in unexpected behavior, because you never know when the parameter will be changed in relation to data processing.

Examples

Example: DF0800236 Post-Proc Replay Demo

Demonstrates how a Post-Proc Replay operator is used to calculate the mean of the selected signals in a Y-T Reviewer.
Examples\DF0800236_001_Post-Proc_Replay_Demo.xmc