Each nontiled interconnect transaction that reaches the DMM on a TILER port is subject to the same processing.
The TILER blocks consider separated request, data, and response paths. An overview of each path follows and more detailed information is in Section 15.2.3.5.3, DMM Internal Macro-Architecture.
On the request path, the flow consists of:
- Allocating a TILER response context for the timely generation of the appropriate responses
- Splitting 2D requests in a collection of 1D requests – TILER ports
- Splitting requests at DMM unit boundaries; the split granularity is provided by the LISA mapping registers
- Allocating an available buffer in the appropriate ROBIN, for both read and write requests
- In case of a write request, allocating and updating a TILER write context to direct the incoming write data into the relevant reordering buffer
- Generating the initiator-indexed priority extension by use of the PEG block
On the write data path, the flow consists of forwarding incoming data to the relevant reordering buffer in accordance with the corresponding TILER write context.
On the response path, responses are returned when:
- At least one response has entered each related buffer, in case of read requests.
- All related responses have returned from the SDRC, in case of write requests.
- No other previous pending response with the same tag exists.