Apple-2-SW/prog8
1
0
Fork 0
mirror of https://github.com/irmen/prog8.git synced 2024-12-26 14:29:35 +00:00
Code Issues Projects Releases Wiki Activity

added verafx.mult16()

Browse Source
This commit is contained in:
Irmen de Jong 2024-09-28 01:00:28 +02:00
parent 9c7a645e18
commit c78b7b1a24
3 changed files with 45 additions and 17 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
compiler/res/prog8lib/cx16/verafx.p8
View File

@ -111,7 +111,7 @@ verafx {
; unsigned multiplication just passes the values as signed to muls
; if you do this yourself in your call to muls, it will save a few instructions.
; TODO fix this: verafx.muls doesn't support unsigned values like this
; TODO fix this: verafx.muls doesn't support unsigned values like this for full 32 bit result
; inline asmsub mult(uword value1 @R0, uword value2 @R1) clobbers(X) -> uword @AY, uword @R0 {
; ; Returns the 32 bits unsigned result in AY and R0 (lower word, upper word).
; %asm {{
@ -119,6 +119,23 @@ verafx {
; }}
; }
asmsub mult16(uword value1 @R0, uword value2 @R1) clobbers(X) -> uword @AY {
; Returns the 16 bits unsigned result of R0*R1 in AY.
; Note: only the lower 16 bits! (the upper 16 bits are not valid for unsigned word multiplications, only for signed)
%asm {{
lda cx16.r0
sta P8ZP_SCRATCH_W1
lda cx16.r0+1
sta P8ZP_SCRATCH_W1+1
jsr verafx.muls
ldx P8ZP_SCRATCH_W1
stx cx16.r0
ldx P8ZP_SCRATCH_W1+1
stx cx16.r0+1
rts
}}
}
asmsub muls(word value1 @R0, word value2 @R1) clobbers(X) -> word @AY, word @R0 {
; Returns the 32 bits signed result in AY and R0 (lower word, upper word).
; Vera Fx multiplication support only works on signed values!

21
docs/source/libraries.rst
View File

@ -982,15 +982,22 @@ the emulators already support it).
``available``
Returns true if Vera FX is available, false if not (that would be an older Vera chip)
``mult`` , ``muls``
The hardware 16*16 multiplier is exposed via ``mult`` and ``muls`` routines (unsigned and signed respectively).
They are about 4 to 5 times faster as the default 6502 cpu routine for word multiplication.
But they depend on some Vera manipulation and 4 bytes in vram just below the PSG registers for storage.
``muls``
The VeraFX signed word 16*16 to 32 multiplier is accessible via the ``muls`` routine.
It is about 4 to 5 times faster than the default 6502 cpu routine for word multiplication.
But it depends on some Vera manipulation and 4 bytes in vram just below the PSG registers for storage.
Note: there is a block level %option "verafxmuls" that automatically replaces all word multiplications in that block
by calls to verafx.muls/mult, but be careful with it because it may interfere with other Vera operations or IRQs.
by calls to verafx, but be careful with it because it may interfere with other Vera operations or IRQs.
The full 32 bits result value is returned in two result values: lower word, upper word.
Note: the lower 16 bits of the 32 bits result is returned as the normal subroutine's returnvalue,
but the upper 16 bits is returned in cx16.r0 so you can still access those separately.
``mult16``
VeraFX hardware multiplication of two unsigned words.
NOTE: it only returns the lower 16 bits of the full 32 bits result, because the upper 16 bits are not valid for unsigned word multiplications here
(the signed word multiplier ``muls`` does return the full 32 bits result).
It is about 4 to 5 times faster than the default 6502 cpu routine for word multiplication.
But it depends on some Vera manipulation and 4 bytes in vram just below the PSG registers for storage.
Note: there is a block level %option "verafxmuls" that automatically replaces all word multiplications in that block
by calls to verafx, but be careful with it because it may interfere with other Vera operations or IRQs.
``clear``
Very quickly clear a piece of vram to a given byte value (it writes 4 bytes at a time).

22
examples/test.p8
View File

@ -1,16 +1,20 @@
%import textio
%zeropage dontuse
%output raw
%launcher none
%address $2000
%import verafx
%zeropage basicsafe
%option no_sysinit
main {
uword @shared variable
word @shared w1 = -30
word @shared w2 = -40
uword @shared uw1 = 9999
uword @shared uw2 = 4
sub start() {
txt.print("hello!\n")
txt.print_uw(variable)
cx16.r0 = 12345
txt.print_uw(verafx.mult16(uw1, uw2))
txt.spc()
txt.print_uw(uw1 * uw2)
txt.nl()
sys.exit3(1,2,3,false)
}
}