FPU: Simplify IDLE state code
authorPaul Mackerras <paulus@ozlabs.org>
Sat, 21 May 2022 09:36:17 +0000 (19:36 +1000)
committerPaul Mackerras <paulus@ozlabs.org>
Tue, 9 Aug 2022 09:51:48 +0000 (19:51 +1000)
commit932da4c114b518504ea9863415159af01bd412b4
treeab967694e0577a50b8c3a8b600db0bcc5129912e
parent8da05e5331914674bd3e66a5728d435228d3ebf7
FPU: Simplify IDLE state code

Do more decoding of the instruction ahead of the IDLE state
processing so that the IDLE state code becomes much simpler.
To make the decoding easier, we now use four insn_type_t codes for
floating-point operations rather than two.  This also rearranges the
insn_type_t values a little to get the 4 FP opcode values to differ
only in the bottom 2 bits, and put OP_DIV, OP_DIVE and OP_MOD next to
them.

Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
common.vhdl
decode1.vhdl
decode2.vhdl
decode_types.vhdl
execute1.vhdl
fpu.vhdl
scripts/fmt_log/fmt_log.c