The decode2 stage was spaghetti code and needed cleaning up.
Create a series of functions to pull fields from a ppc instruction
and also a series of helpers to extract values for the execution
units.
As suggested by Paul, we should pass all signals to the execution
units and only set the valid signal conditionally, which should
use less resources.
Signed-off-by: Anton Blanchard <anton@linux.ibm.com>
Some VHDL compilers like verific [1] don't like these, so let's remove
them. Lots of random code changes, but passes make check.
Also add basic script to run verific and generate verilog.
1. https://www.verific.com/
Signed-off-by: Michael Neuling <mikey@neuling.org>
- Switch to using ghdl/vunit:llvm, it's a smaller container
- We need to "apt update" before installing packages
Signed-off-by: Anton Blanchard <anton@linux.ibm.com>
Mikey points out that our stack grows down from 512kB and our
heap is below that too, so we can reduce our BRAM requirements,
which allowing some smaller FPGA boards to work. Not sure why
I thought we were using memory between 512kB and 1MB.
Signed-off-by: Anton Blanchard <anton@linux.ibm.com>
The synth target can be used to analyze the core after synthesis
without running P&R. Currently, the only edalize backends that
support synthesis without P&R are vivado and icestorm, and icestorm
needs yosys built with verific support to parse vhdl.
To run synthesis only for a part, run
fusesoc run --target=synth --tool=vivado microwatt --part=<part>
where part is a valid Xilinx part such as xc7a100tcsg324-1
Add a temporary gcc patch to remove hardware divide instructions.
Also add a firmware.hex file built with a gcc with the above patch.
Right now micropython assumes 1MB of BRAM, which limits the FPGAs
we can run on. We should be able to cut it down somewhat.
Signed-off-by: Anton Blanchard <anton@linux.ibm.com>