We've just updated MediaWiki and its underlying software. If anything doesn't look or work quite right, please mention it to us. --RanAS

LMULT (Super FX): Difference between revisions

From SnesLab
Jump to: navigation, search
(speed depends on CFGR)
(made flags affected more prominent)
 
(28 intermediate revisions by the same user not shown)
Line 2: Line 2:
!colspan="8"|Basic Info
!colspan="8"|Basic Info
|+
|+
|'''Addressing Mode'''
|'''Opcode'''
|'''Opcode'''
|'''Length'''
|'''Length'''
Line 8: Line 9:
|'''Cache Speed'''
|'''Cache Speed'''
|+
|+
|[[Implied]] (type 1)
|3D9F
|3D9F
|2 bytes
|2 bytes
Line 16: Line 18:


{| class="wikitable" style="float:right;clear:right;width:30%"
{| class="wikitable" style="float:right;clear:right;width:30%"
!colspan="9"|Flags Clobbered
!colspan="9"|Flags Affected
|+
|+
|B
|[[B Flag|B]]
|ALT1
|[[ALT1]]
|ALT2
|[[ALT2]]
|O/V
|[[O/V]]
|S
|[[Sign Flag|S]]
|CY
|[[CY]]
|Z
|[[Zero Flag|Z]]
|+
|+
|0
|0
Line 30: Line 32:
|0
|0
|.
|.
|
|S
|
|CY
|
|Z
|}
|}


'''LMULT''' is a signed multiplication [[Super FX]] instruction.  The exact speed depends on the state of the [[CFGR]] register.
'''LMULT''' is a signed multiplication [[Super FX]] instruction.  The two 16-bit factors are the [[source register]] and R<sub>6</sub>.  The exact speed depends on the state of the [[CFGR]] register.
 
The upper 16 bits of the product are stored in the [[destination register]].  The lower 16 bits of the product are stored in R<sub>4</sub>.
 
LMULT utilizes the 8-bit multiplier four times.<sup>[3]</sup>
 
LMULT shares its multiplication circuit with [[FMULT]].
 
If R<sub>4</sub> is specified as the destination register, the product will be invalid.<sup>[1]</sup>
 
The [[ALT0]] state is restored.
 
The source and destination registers should be specified in advance using [[WITH]], [[FROM]], or [[TO]].  Otherwise, R<sub>0</sub> serves as the default.
 
==== Syntax ====
LMULT
 
==== Example ====
Let:
S<sub>reg</sub> : R<sub>9</sub>
D<sub>reg</sub> : R<sub>8</sub>
R<sub>9</sub> = b556h
R<sub>6</sub> = daabh
After LMULT is executed:
R<sub>8</sub> = 0ae3h
R<sub>4</sub> = 5c72h


=== See Also ===
=== See Also ===
* [[FMULT]]
* [[UMULT]]
* [[UMULT]]
* [[MULT]]
* [[Partial Product Buffer]]
* [[DIV2]]
* [[ALT1]]


=== External Links ===
=== External Links ===
* Official Nintendo documentation on LMULT: [https://archive.org/details/SNESDevManual/book2/page/n230 page 2-9-73 of Book II]
# Official Nintendo documentation on LMULT: 9.52 on [https://archive.org/details/SNESDevManual/book2/page/n229 page 2-9-73 of Book II]
# example: [https://archive.org/details/SNESDevManual/book2/page/n230 page 2-9-74 of Book II], lbid.
# [https://archive.org/details/SNESDevManual/book2/page/n155 page 2-8-16 of Book II], lbid.


[[Category:ASM]]
[[Category:ASM]]
[[Category:Super FX]]
[[Category:Super FX]]
[[Category:Expects Sreg/Dreg Prearranged]]

Latest revision as of 19:27, 30 July 2024

Basic Info
Addressing Mode Opcode Length ROM Speed RAM Speed Cache Speed
Implied (type 1) 3D9F 2 bytes 10 or 14 cycles 10 or 14 cycles 5 or 9 cycles
Flags Affected
B ALT1 ALT2 O/V S CY Z
0 0 0 . S CY Z

LMULT is a signed multiplication Super FX instruction. The two 16-bit factors are the source register and R6. The exact speed depends on the state of the CFGR register.

The upper 16 bits of the product are stored in the destination register. The lower 16 bits of the product are stored in R4.

LMULT utilizes the 8-bit multiplier four times.[3]

LMULT shares its multiplication circuit with FMULT.

If R4 is specified as the destination register, the product will be invalid.[1]

The ALT0 state is restored.

The source and destination registers should be specified in advance using WITH, FROM, or TO. Otherwise, R0 serves as the default.

Syntax

LMULT

Example

Let:

Sreg : R9
Dreg : R8
R9 = b556h
R6 = daabh

After LMULT is executed:

R8 = 0ae3h
R4 = 5c72h

See Also

External Links

  1. Official Nintendo documentation on LMULT: 9.52 on page 2-9-73 of Book II
  2. example: page 2-9-74 of Book II, lbid.
  3. page 2-8-16 of Book II, lbid.