I'm getting the following timing summary from the synthesis:
Timing Summary:
Speed Grade: -1
Minimum period: 9.982ns (Maximum Frequency: 100.180MHz)
Minimum input arrival time before clock: 4.597ns
Maximum output required time after clock: 4.364ns
Maximum combinational path delay: 2.788ns
I want to improve that, is there a way to find the critical path and maybe buffer it up a bit?
you'll have to look in detail at the timing. I cant remember the xilinx tools exactly but I know its there and should show you the slowest paths. Then you have to go back to the RTL to modify it - then re-simulate it to make sure it works with the new pipelining, then re-build it in ISE, and recheck timing.