summaryrefslogtreecommitdiffstatshomepage
path: root/src/devices/cpu/m88000/m88000d.cpp
diff options
context:
space:
mode:
Diffstat (limited to 'src/devices/cpu/m88000/m88000d.cpp')
-rw-r--r--src/devices/cpu/m88000/m88000d.cpp5
1 files changed, 4 insertions, 1 deletions
diff --git a/src/devices/cpu/m88000/m88000d.cpp b/src/devices/cpu/m88000/m88000d.cpp
index 2f718dba068..26ba54afc2a 100644
--- a/src/devices/cpu/m88000/m88000d.cpp
+++ b/src/devices/cpu/m88000/m88000d.cpp
@@ -23,6 +23,9 @@
byte ordering by setting bit 30 of PSR, this has no effect on
instruction decoding.
+ The MC88100 executes integer multiplication instructions by using the
+ floating-point multiply pipeline. This is not the case on the MC88110.
+
The extended register file added to the MC88110 is used mainly for
floating-point operations. Numbers of any precision can be stored in
any of these 80-bit registers except x0, which is fixed to contain
@@ -1011,7 +1014,7 @@ offs_t m88000_disassembler::dasm_fp(std::ostream &stream, const char *mnemonic,
if ((inst & 0xfc006000) == 0x84004000)
{
util::stream_format(stream, "%-12s%c%d,", mnemonic,
- /*(inst & 0xfc007c00) == 0x84004200 ? 'x' : */'r',
+ (inst & 0xfc007e00) == 0x84004200 ? 'x' : 'r',
(inst & 0x03e00000) >> 21);
}
else