Irmen de Jong
78f9df203c
inline math trig routines
2026-10-10 20:46:10 +02:00
Irmen de Jong
c4679bd8af
m68k: optimize lerp/interpolate/CRC routines; add interpolatew
2026-10-10 18:28:12 +02:00
Irmen de Jong
822cfd4f35
tweak textelite
2026-10-07 02:24:17 +02:00
Irmen de Jong
38db571f20
implemented -o / --output to specify specific output file
2026-10-02 22:33:43 +02:00
Irmen de Jong
cff7732189
fix unsound index-reload peephole and extend it to two indexed accesses
2026-09-28 00:48:00 +02:00
Irmen de Jong
5222562493
move sqr range check into prog8_math.square instead of inlining it at each call site
2026-09-27 11:49:49 +02:00
Irmen de Jong
c8cda7947a
6502 optimization for uword * ubyte
2026-09-26 05:15:41 +02:00
Irmen de Jong
b8c975704a
6502 byte shift optimization for const multiplications
2026-09-26 03:25:25 +02:00
Irmen de Jong
2ef0ddd42c
6502 word shift optimization for const multiplications
2026-09-26 02:45:57 +02:00
Irmen de Jong
4eaba62f1e
6502 codegen: inline clamp() on word/uword when having compile-time constant bounds
2026-09-23 22:54:57 +02:00
Irmen de Jong
815f4109a5
cx16/gfx_lores: optimize text() by setting VERA address once per character
2026-09-19 22:06:05 +02:00
Irmen de Jong
457d55b392
fix potential endless loop in wavfile library, also made a correct implementation of it for m68k targets (little/big endianness problem resolved)
2026-09-17 00:04:34 +02:00
Irmen de Jong
b21a984dbb
even more optimized adpcm decoding routine
2026-09-16 19:04:49 +02:00
Irmen de Jong
7427e6ddad
changed adpcm decoder by a table driven one, almost twice as fast now
2026-09-16 05:03:13 +02:00
Irmen de Jong
9621552056
add 'arena' allocator library
...
6502 targets use uword sizes with no alignment overhead; m68k/virtual use
long sizes with even-address alignment. Updated the six pointer examples that used to have their own arena implementation.
2026-09-13 05:43:57 +02:00
Irmen de Jong
5685a1f229
6502: tighter for loop code for word and long loop vars
2026-09-13 04:29:59 +02:00
Irmen de Jong
619a215a87
fix test size expectation
2026-09-07 19:14:58 +02:00
Irmen de Jong
c12e27dacc
cx16: added verafx-accelerated line routine, added cobramk3 example that uses it next to the cpu only version
2026-09-06 22:51:12 +02:00
Irmen de Jong
58f544fe14
Implement counted-loop IR (IRLoopChunk)
...
Replace the early lowering of constant repeat/for loops into p8_regfile
counter sequences with a first-class IRLoopChunk that carries the trip
count and body. This removes the fragile REGFILE-DEPENDENT m68k dbra
peephole and gives each backend explicit control.
2026-09-04 00:01:00 +02:00
Irmen de Jong
7dc05688bf
Add peephole optimization to fold array[variable+const] index offsets
...
Adds three new passes to the 6502 assembly optimizer:
- optimizeFoldIndexOffset (byte/bool arrays)
- optimizeFoldIndexOffsetSplitWord (split word arrays)
- optimizeFoldIndexOffsetNonSplitWord (@nosplit word arrays)
fixes github issue #234
2026-09-03 21:36:29 +02:00
Irmen de Jong
59c41ad129
ir: add scaled indexing (S=) to LOADX/STOREX/STOREZX
...
Remove byte-offset pre-scaling from IR generation and encode
element index + explicit scale in the IR. Effective address is
base + index*scale + disp (disp = symbolOffset).
2026-09-02 00:34:23 +02:00
Irmen de Jong
deae2e6422
Unify variable and memory slab relocation with vars address configuration
2026-08-29 01:17:44 +02:00
Irmen de Jong
277e9c09bb
making defer program-wide: also unwinds at program exit()/return from start()
2026-08-28 05:10:09 +02:00
Irmen de Jong
250ddda11b
ir,m68k: add LOADP_INC/STOREP_INC for pointer post-inc (a0)+
2026-08-27 02:57:54 +02:00
Irmen de Jong
8647b136bd
6502: fix constant byte store via pointer+const offset storing garbage
...
fixes regression from commit 102992911
2026-08-25 23:59:41 +02:00
Irmen de Jong
4199ab4200
fix broken unit test dependency on vasm
2026-08-23 16:07:27 +02:00
Irmen de Jong
dfda43e747
more 6502 optimizations
2026-08-23 10:18:05 +02:00
Irmen de Jong
1029929111
6502 optimizations
2026-08-23 04:15:47 +02:00
Irmen de Jong
ae55109a18
m68k: peepholes
2026-08-23 03:24:20 +02:00
Irmen de Jong
d17c978892
Merge branch 'master' into new-codegens
...
# Conflicts:
# compiler/test/TestCompilerOnExamples.kt
# examples/c64/cube3d.p8
# examples/cube3d-gfx.p8
# examples/cx16/cobramk3-gfx.p8
# examples/cx16/cube3d.p8
2026-08-22 19:12:06 +02:00
Irmen de Jong
2cd2bb0121
optimized the 3d rotation examples
2026-08-22 19:03:14 +02:00
Irmen de Jong
82828606e6
removed some string routines from shared because virtual target versions now need a 32 bit compatible implementation
2026-08-17 01:39:31 +02:00
Irmen de Jong
4fdd12929a
vm: switched to m68k like 32 bits registers
2026-08-16 21:14:15 +02:00
Irmen de Jong
fa1ceab78f
M68K codegen: extend immediate call-argument forwarding and dead-store removal to floating-point registers and `p8_fregfile`.
2026-08-16 08:08:02 +02:00
Irmen de Jong
d870c39aa4
m68k: use moveq and movea for call argument loading, drop +0 offset
...
- use moveq for small immediate call args (bytes 128-255 map to signed -128..-1)
- use movea.l for pointer/long args into address registers, dropping +0 suffix
- loadPointerToA0 also emits movea.l without the redundant +0 offset
- update stale C64 multiplexer expected output size
2026-08-15 18:49:28 +02:00
Irmen de Jong
22035077af
clean up for loop vardecls in examples
2026-08-09 17:05:34 +02:00
Irmen de Jong
41c2b9c039
fix bit shift code for shifts by 16/32 bits or more
2026-08-09 14:26:45 +02:00
Irmen de Jong
b5dfc0cd97
fix debug message and tests
2026-08-09 00:32:58 +02:00
Irmen de Jong
e5d9609db1
IR: added new multi-bit shifting opcodes
2026-08-05 01:42:49 +02:00
Irmen de Jong
b5fd07bb91
IR: added STOREIM instruction for immediate store constant to memory
2026-08-04 23:11:07 +02:00
Irmen de Jong
4097d51514
IR: more efficient code for comparisons against constants
2026-08-04 22:28:59 +02:00
Irmen de Jong
af80ec6502
fix invalid optimization of signed division into a shift (wrong rounding result) - IR codegen
2026-08-04 17:42:12 +02:00
Irmen de Jong
b9f65186cb
fix expression operand swap bug when there are side effects
2026-08-04 14:49:01 +02:00
Irmen de Jong
3e03980478
tweaking ast optimizations for m68k
2026-08-03 12:45:48 +02:00
Irmen de Jong
247559858f
fix bitwise instructions on m68k
2026-07-11 23:40:46 +02:00
Irmen de Jong
cf2a5ae4f9
fix test sizes
2026-07-11 23:06:26 +02:00
Irmen de Jong
57714efb40
added pushp() and popp() builtin functions
2026-07-11 21:49:22 +02:00
Irmen de Jong
57254233e4
Merge branch 'master' into new-codegens
...
# Conflicts:
# codeCore/src/prog8/code/target/NormalMemSizer.kt
2026-07-11 18:22:11 +02:00
Irmen de Jong
5b5f553d37
improved error message when attempting to alias types,
...
fix size test
2026-07-11 17:04:52 +02:00
Irmen de Jong
cb240f6a4b
Merge branch 'master' into new-codegens
2026-07-08 21:02:17 +02:00
Irmen de Jong