Configuring Comma Patterns

Versal Adaptive SoC GTY and GTYP Transceivers Architecture Manual (AM002)

Document ID
Release Date
1.3 English

To set the comma pattern that the block searches for in the incoming data stream, the RX_ALIGN_MCOMMA_VAL, RX_ALIGN_PCOMMA_VAL, RX_PCOMMA_ALIGEN, and RX_MCOMMA_ALIGNEN attributes are used. The comma lengths depend on RX_DATA_WIDTH. The following figure shows how RX_ALIGN_COMMA_ENABLE masks each of the comma values to allow partial pattern matching.

Figure 1. Comma Pattern Masking

The following figure shows how the commas are combined when RX_ALIGN_COMMA_DOUBLE is TRUE.

Figure 2. Extended Comma Pattern Definition

The following figure shows how a comma is combined with RX_ALIGN_COMMA_ENABLE to make a wild-carded comma for a 20-bit internal comma. If RX_ALIGN_COMMA_DOUBLE is TRUE, the MCOMMA and PCOMMA patterns are combined so that the block searches for two commas in a row. The number of bits in the comma depends on RX_DATA_WIDTH. Either a 16-bit or a 20-bit comma alignment mode is possible. A double comma is only detected when the received data has a PCOMMA defined by RX_ALIGN_PCOMMA_VAL, followed by an MCOMMA defined by RX_ALIGN_MCOMMA_VAL with no extra bits in between. For commas in RAW data mode with 32- or 16-bit data widths, only the lower eight of the ten comma bits are available, even though the Wizard displays all ten.

Figure 3. Extended Comma Pattern Masking