ed/pikuma_ps1
1
0
Fork 0
mirror of https://github.com/Ed94/pikuma_ps1.git synced 2026-08-05 23:28:47 +00:00
Code Issues Packages Projects Releases Wiki Activity

add ac_yield_load and ac_yield_tail for delay slot optimization opportunities.

Browse Source
This commit is contained in:
ed
2026-08-04 17:25:02 -04:00
parent 888ffce859
commit b5953a723b
7 changed files with 185 additions and 19 deletions
Download Patch File Download Diff File Expand all files Collapse all files
code/duffle/gen/duffle.macs.h
+17 -1
View File
@@ -2,7 +2,7 @@
#pragma once
#endif
// Auto-generated by ps1_meta.lua — DO NOT EDIT
// Source: C:\projects\Pikuma\ps1\code\duffle\lottes_tape.h
// Source: C:\projects\pikuma\ps1\code\duffle\lottes_tape.h
// Component atoms (MipsAtomComp_(ac_*)) -> macro variants (mac_*)
#ifndef WORD_COUNT
@@ -15,6 +15,10 @@
* These do NOT yield. They are expanded inline inside Tape Atoms.
* ---------------------------------------------------------------------------*/
// The 'Yield' sequence for Tape Atoms (mac_yield).
// - mac_yield() is the safe default for atom-endings: 4 words, BD-slot of jr is mandatory nop.
// - mac_yield_load() + mac_yield_tail():
// - unconditional branch: mac_yield_load fills the branch's BD-slot (replaces a nop);
// - mac_yield_tail runs at the branch target (does NOT re-load R_AtomJmp).
#define mac_yield(...) \
load_word(R_AtomJmp, R_TapePtr, 0) \
, add_ui_self( R_TapePtr, S_(MipsCode)) \
@@ -22,6 +26,18 @@
, nop
WORD_COUNT(mac_yield, 4)
/* atom_dbg_skip */
#define mac_yield_load(...) \
load_word(R_AtomJmp, R_TapePtr, 0)
WORD_COUNT(mac_yield_load, 1)
/* atom_dbg_skip */
#define mac_yield_tail(...) \
add_ui_self(R_TapePtr, S_(MipsCode)) \
, jump_reg( R_AtomJmp) \
, nop
WORD_COUNT(mac_yield_tail, 3)
/* atom_dbg_skip */
/* Words: 3; Loads 3 S2 indices from the face array */
#define mac_load_tri_indices(...) \
code/duffle/gen/duffle.offsets.h
+1 -1
View File
@@ -1,5 +1,5 @@
// Auto-generated by ps1_meta.lua (passes/offsets.lua) — DO NOT EDIT
// Source: C:\projects\Pikuma\ps1\code\duffle\lottes_tape.h
// Source: C:\projects\pikuma\ps1\code\duffle\lottes_tape.h
#pragma once
#pragma region lottes_tape
code/duffle/lottes_tape.h
+15 -1
View File
@@ -122,13 +122,27 @@ FI_ void tb_scope_run_end(TapeBuilder* tb) { tb_emit(tb,tape_exit); tape_run(tb_
* These do NOT yield. They are expanded inline inside Tape Atoms.
* ---------------------------------------------------------------------------*/
// The 'Yield' sequence for Tape Atoms (mac_yield).
// The 'Yield' sequence for Tape Atoms (mac_yield).
// - mac_yield() is the safe default for atom-endings: 4 words, BD-slot of jr is mandatory nop.
// - mac_yield_load() + mac_yield_tail():
// - unconditional branch: mac_yield_load fills the branch's BD-slot (replaces a nop);
// - mac_yield_tail runs at the branch target (does NOT re-load R_AtomJmp).
atom_dbg_skip MipsAtomComp_(ac_yield) {
load_word(R_AtomJmp, R_TapePtr, 0),
add_ui_self( R_TapePtr, S_(MipsCode)),
jump_reg( R_AtomJmp), nop,
};
atom_dbg_skip MipsAtomComp_(ac_yield_load) {
load_word(R_AtomJmp, R_TapePtr, 0),
};
atom_dbg_skip MipsAtomComp_(ac_yield_tail) {
add_ui_self(R_TapePtr, S_(MipsCode)),
jump_reg( R_AtomJmp), nop,
};
enum {
R_PrimCursor = R_T7 atom_reg atom_type(U4*), /* VRAM output cursor (primitive buffer) */
R_FaceCursor = R_T4 atom_reg atom_type(V4_S2*), /* Cube face-index cursor (V4_S2*); floor context switches to V3_S2* via atom_phase */