The input DATA channel and output DATA channels optionally support a tuser field. For each, the tuser field can be used to convey a User Field and/or a Channel ID field. When both are selected, they are concatenated, with Channel ID in the least significant bit positions. When User Field is selected on the input channel it is automatically selected for the output channel, as this User Field, like ‘packet-based’ tlast is a facility whereby the User Field is passed through the core, but subject to the same latency delay as the tdata path from input to output. This is intended to ease system design. The User Field has user-selected width.
The Channel ID field has the minimum width required to describe the number of channels in a time-division multiplex cycle (log2roundup(number_of_channels)), for example, with 13 channels, channel ID is 4 bits wide.
The output DATA channel also includes a Data Valid field when aresetn has been selected without Data vector reset being selected. This field can be used for additional validation of the m_axis_data_tdata bus. See Resets for more details. The Data Valid field occupies the LSB of m_axis_data_tuser with the other tuser fields, when selected, being shifted up the bus.
|
X-Ref Target - Figure 3-3 
|
When the core has been configured to implement a rate change the following rules are applied to tuser and tlast .
• When the core is configured with no rate change tuser and tlast propagate through the core unmodified.
• When the core is configured to up convert by X the input tuser and tlast are duplicated on the last sample of the corresponding block of X output samples. tuser is undefined for the other X-1 output samples.
• When the core is configured to down convert by X the tuser value for a given output sample is taken from the tuser value of the first input sample of the corresponding X input samples. tlast is OR’d over X input samples with the result being used for the tlast of the corresponding output sample.