Electronic – deserializing high speed data

No, you don't have to worry about meta-stability, but you do need to make sure you constrain the clocks appropriately. Many modern FPGA Static Timing Analysis (STA) tools have commands to help constrain related clocks.

You didn't mention which vendor you're using. For the sake of example, the process for Altera would be to specify a clock constraint on the 54MHz source clock as your normally would, and then use "create_generated_clock" for the slower clock as shown below.

create_clock -name {clk} -period 18.518 [get_ports {clk}]
create_generated_clock -name {clk4} -source [get_ports {clk}] -divide-by 4 -master_clock {clk} [get_keepers {clk_div_4}]

Note that we did not specify the frequency for the slower clock. We only told the STA tool that it is the frequency of the source clock divided by 4. If you follow this approach the STA tool will properly account for the latencies of both the source and derived clocks.

To verify that things are working as you desire it's often helpful simply just to use the STA tool to report the timing between registers that cross the domain and see if the results match your expectations. Again, not sure which vendor you're using but you'll probably want to look for commands like "report clock transfers" or "report clock tree".

As for whether or not your approach is a good one vs running everything off the 54MHz, that depends on the details of your design and there is no one right answer. Either approach is acceptable in various situations. For example, if you run the entire design at the higher frequency then you don't have to worry with clock crossings, but your code may be significantly more complicated because of pipelining, or the use of multi-cycle paths. However, if it's a just a small portion of your design that you want to run on a slower clock then perhaps you'd be better off doing the extra work to pipeline it instead of dealing with the second clock domain.

Xilinx's, for example, have very fast giga-bit transceivers that can operate at multi-gigabit/sec rates, but they also have simple serial and parallel shift registers out at the IO edges of their dies that are typically used for the serial to parallel shifting in both directions. In addition, to get the fastest speeds with synchronous external clocks, I typically use half-rate clocks and DDR (double datarate serializer or de-serializer settings). Its pretty straightfoward in a series-7 Xilinx to get 1Gb/sec on each LVDS I/O pin when applying a 500Mhz external clock. This is all done without using the gigabit transceivers. This also involves no PLLs so it can operate smoothly from low speeds to high speeds continuously.

If you have 640MHz clock today, you may be after 640Mbps on your data lines. My recommendation would be to use a 320Mhz clock and a 1:4 or 4:1 internal serial-to-parallel or parallel-to-serial converter in DDR mode (See Xilinx OSERDESE2 and ISERDESE2 built-in components). This would leave your parallel clocking rates in your main logic down at 160Mhz which is OK.

There are good application notes on setting this up. You can also use the built-in IP features of their development tools to create this component for you.

Hope that helps.