Difference between revisions of "Instruction Set/mulsfwe"
From Mill Computing Wiki
m (Protected "Instruction Set/mulsfwe": generated ([Edit=<protect-level-bot>] (indefinite) [Move=<protect-level-bot>] (indefinite))) | |
(No difference) | |
Revision as of 01:33, 3 January 2015
realizing exu stream exu block compute phase operation in the signed fixed point value domain using widening overflow behavior that produces condition codes and rounds to nearest, ties toward even adjacent value
native on: all
Signed Fixed Point multiply. Rounds towards nearest even. Widening.
mulsfwe(sf x, sf y, bit dot) → sf r0
operands: like Identity [xx:x]
| Core | In Slots | Latencies |
|---|---|---|
| Tin | E0 | b,b:b=3 bv,bv:bv=3 h,h:h=3 hv,hv:hv=3 w,w:w=3 wv,wv:wv=3 d,d:d=4 dv,dv:dv=4 q,q:q=4 qv,qv:qv=4 |
| Copper | E0 | b,b:b=3 bv,bv:bv=3 h,h:h=3 hv,hv:hv=3 w,w:w=3 wv,wv:wv=3 d,d:d=4 dv,dv:dv=4 q,q:q=4 qv,qv:qv=4 |
| Silver | E0 E1 | b,b:b=3 bv,bv:bv=3 h,h:h=3 hv,hv:hv=3 w,w:w=3 wv,wv:wv=3 d,d:d=4 dv,dv:dv=4 q,q:q=4 qv,qv:qv=4 |
| Gold | E0 E1 E2 E3 | b,b:b=3 bv,bv:bv=3 h,h:h=3 hv,hv:hv=3 w,w:w=3 wv,wv:wv=3 d,d:d=4 dv,dv:dv=4 q,q:q=4 qv,qv:qv=4 |
| Decimal8 | E0 E1 | b,b:b=3 bv,bv:bv=3 h,h:h=3 hv,hv:hv=3 w,w:w=3 wv,wv:wv=3 d,d:d=4 dv,dv:dv=4 q,q:q=4 qv,qv:qv=4 |
| Decimal16 | E0 E1 | b,b:b=3 bv,bv:bv=3 h,h:h=3 hv,hv:hv=3 w,w:w=3 wv,wv:wv=3 d,d:d=4 dv,dv:dv=4 q,q:q=4 qv,qv:qv=4 |
Instruction Set, alphabetical, Instruction Set by Category, Instruction Set, sortable, filterable
