Speedster7t GDDR6 User Guide (UG091)
26
The following figure shows how the master logic in the FPGA fabric interacts with the GDDR6 interface utilizing
the NoC.
Figure 15:
Data Flow from the FPGA Fabric to the GDDR6 Subsystem Through the
NoC
Connectivity Through the Beachfront
There are only four GDDR6 subsystems, the middle two, on the east and west sides of the chip that enable the
beachfront interface connection directly to the fabric. This connection is a 512-bit AXI4 Interface (one per
controller per subsystem), capable of running up to 500 MHz supporting data rates of up to 16 Gbps:
AXI4 512b is a 512-bit slave interface which is connected to fabric master AXI
Asynchronous read and write clock. The write clock is half of the read clock. An asynchronous FIFO
handles clock domain crossing and 512-bit AXI to 256-bit local data conversion.
The write clock is provided by one of the global clocks. The read clock is the controller clock which is
synchronous to controller, AXI1 and PHY interfaces. The selection of clocks are controlled via the
IPCNTRL register.
AXI4 converts AXI transactions to the local bus transactions.
The figure below shows the I/O diagram of an AXI4-512b Interface: