mirror/oot
1
0
Fork 0
mirror of https://github.com/zeldaret/oot synced 2026-09-01 09:43:06 -04:00
Code Issues Packages Projects Releases Wiki Activity

Use IDO for assembling handwritten asm files in src (#2390)

Browse Source 
* as0

* Fix ASOPTFLAGS for src/libc, remove unnecessary noreorder region in kanread

* Suggested changes

* Use %half to load the boot bss size for matching

Co-authored-by: cadmic <cadmic24@gmail.com>

* Wrap all of __osProbeTLB in noreorder

---------

Co-authored-by: cadmic <cadmic24@gmail.com>
This commit is contained in:
Tharo
2024-12-28 20:18:45 +00:00
committed by GitHub
parent ba6a83533a
commit 7e082f0c4f
54 changed files with 2612 additions and 2653 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/boot/mio0.s
+46 -46
View File
@@ -1,10 +1,7 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
.set noreorder
.section .text
.balign 16
.text
/**
* void Mio0_Decompress(void* src, void* dst);
@@ -12,53 +9,56 @@
* Decompress Mio0 chunk
*/
LEAF(Mio0_Decompress)
lw $a3, 8($a0) // compressed offset
lw $t9, 0xC($a0) // uncompressed offset
lw $t8, 4($a0) // decompressed length
add $a3, $a3, $a0 // compressed start
add $t9, $t9, $a0 // uncompressed start
move $a2, $zero // 0
addi $a0, $a0, 0x10 // move past header
add $t8, $t8, $a1 // dst + decompressed length = end
lw a3, 0x08(a0) /* compressed offset */
lw t9, 0x0C(a0) /* uncompressed offset */
lw t8, 0x04(a0) /* decompressed length */
add a3, a3, a0 /* compressed start */
add t9, t9, a0 /* uncompressed start */
move a2, zero /* 0 */
addi a0, a0, 0x10 /* move past header */
add t8, t8, a1 /* dst + decompressed length = end */
mainloop:
bnez $a2, 1f
nop
lw $t0, ($a0)
li $a2, 32
addi $a0, $a0, 4
bnez a2, 1f
lw t0, (a0)
li a2, 32
addi a0, a0, 4
1:
slt $t1, $t0, $zero
beqz $t1, read_comp
nop
lb $t2, ($t9) // read 1 byte from uncompressed data
addi $t9, $t9, 1 // advance uncompressed start
addi $a1, $a1, 1
slt t1, t0, zero
beqz t1, read_comp
lb t2, (t9) /* read 1 byte from uncompressed data */
sb t2, (a1) /* store uncompressed byte */
addi t9, t9, 1 /* advance uncompressed start */
addi a1, a1, 1
b next_iter
sb $t2, -1($a1) // store uncompressed byte
read_comp:
lhu $t2, ($a3) // read 2 bytes from compressed data
addi $a3, $a3, 2 // advance compressed start
srl $t3, $t2, 0xC
andi $t2, $t2, 0xFFF
beqz $t3, 3f
sub $t1, $a1, $t2
addi $t3, $t3, 2
lhu t2, (a3) /* read 2 bytes from compressed data */
addi a3, a3, 2 /* advance compressed start */
srl t3, t2, 0xC
andi t2, t2, 0xFFF
sub t1, a1, t2
beqz t3, 3f
addi t3, t3, 2
2:
lb $t2, -1($t1)
addi $t3, $t3, -1
addi $t1, $t1, 1
addi $a1, $a1, 1
bnez $t3, 2b
sb $t2, -1($a1)
lb t2, -1(t1)
addi t3, t3, -1
addi t1, t1, 1
sb t2, (a1)
addi a1, a1, 1
bnez t3, 2b
next_iter:
sll $t0, $t0, 1
bne $a1, $t8, mainloop // continue until decompressed length is reached
addi $a2, $a2, -1
jr $ra
nop
sll t0, t0, 1
addi a2, a2, -1
bne a1, t8, mainloop /* continue until decompressed length is reached */
jr ra
3:
lbu $t3, ($t9)
addi $t9, $t9, 1
lbu t3, (t9)
addi t9, t9, 1
addi t3, t3, 0x12
b 2b
addi $t3, $t3, 0x12
END(Mio0_Decompress)
src/code/kanread.s
+57 -68
View File
@@ -1,11 +1,7 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
.set noat
.set noreorder
.section .text
.balign 16
.text
/**
* s32 Kanji_OffsetFromShiftJIS(s32 sjis);
@@ -26,64 +22,57 @@
* @remark Almost identical to "LeoGetKadr" from libleo.
*/
LEAF(Kanji_OffsetFromShiftJIS)
// Characters with codepoints >= 0x8800 are kanji. Arrangement is regular,
// so convert index directly.
li $at, 0x8800
slt $at, $a0, $at
bnez $at, .nonkanji
// 0xBC is number of glyphs in one block in the `kanji` file:
// 0x100 possible codepoints with the same byte1
// - 0x40 unused at beginning
// - 1 unused at 0x7F
// - 3 unused at 0xFD, 0xFE, 0xFF
li $a2, 0xBC
// Get byte1 and adjust so starts at 0
srl $a1, $a0, 8
addi $a1, $a1, -0x88
multu $a2, $a1
// Get byte2 and adjust so starts at 0
andi $a3, $a0, 0xFF
addi $a3, $a3, -0x40
slti $at, $a3, (0x80 - 0x40)
mflo $a2
// 0x__7F is always empty and elided in the file, so decrement if larger
bnezl $at, .kanji_lower_halfblock
mflo $a2
addi $a3, $a3, -1
mflo $a2
.kanji_lower_halfblock:
addi $a3, $a3, 0x30A
add $a3, $a3, $a2
jr $ra
sll $v0, $a3, 7
// returns (0x30A + (adjusted byte2) + (adjusted byte1) * 0xBC) * FONT_CHAR_TEX_SIZE
/* Characters with codepoints >= 0x8800 are kanji. Arrangement is regular, */
/* so convert index directly. */
li a2, 0xBC
blt a0, 0x8800, .nonkanji
/* 0xBC is number of glyphs in one block in the `kanji` file: */
/* 0x100 possible codepoints with the same byte1 */
/* - 0x40 unused at beginning */
/* - 1 unused at 0x7F */
/* - 3 unused at 0xFD, 0xFE, 0xFF */
/* Get byte1 and adjust so starts at 0 */
srl a1, a0, 8
addi a1, a1, -0x88
/* Get byte2 and adjust so starts at 0 */
andi a3, a0, 0xFF
addi a3, a3, -0x40
mul a2, a2, a1
blt a3, 0x40, .kanji_lower_halfblock
// Non-kanji are arranged with irregular gaps, use the lookup table.
/* 0x__7F is always empty and elided in the file, so decrement if larger */
addi a3, a3, -1
.kanji_lower_halfblock:
mflo a2 /* Unncessary mflo: the pseudo-op `mul` already performs mflo */
addi a3, a3, 0x30A
add a3, a3, a2
/* returns (0x30A + (adjusted byte2) + (adjusted byte1) * 0xBC) * FONT_CHAR_TEX_SIZE */
sll v0, a3, 7
jr ra
/* Non-kanji are arranged with irregular gaps, use the lookup table. */
.nonkanji:
// Get byte1 and adjust so starts at 0
srl $a1, $a0, 8
addi $a1, $a1, -0x81
multu $a2, $a1
// Get byte2 and adjust so starts at 0
andi $a3, $a0, 0xFF
addi $a3, $a3, -0x40
slti $at, $a3, (0x80 - 0x40)
mflo $a2
// 0x__7F is always empty and elided in the file, so decrement if larger
bnezl $at, .nonkanji_lower_halfblock
mflo $a2
addi $a3, $a3, -1
mflo $a2
/* Get byte1 and adjust so starts at 0 */
srl a1, a0, 8
addi a1, a1, -0x81
/* Get byte2 and adjust so starts at 0 */
andi a3, a0, 0xFF
addi a3, a3, -0x40
mul a2, a2, a1
blt a3, 0x40, .nonkanji_lower_halfblock
/* 0x__7F is always empty and elided in the file, so decrement if larger */
addi a3, a3, -1
.nonkanji_lower_halfblock:
add $a3, $a3, $a2
lui $a2, %hi(sNonKanjiIndices)
sll $a3, $a3, 1
addiu $a2, %lo(sNonKanjiIndices)
add $a3, $a3, $a2
lh $a2, ($a3)
jr $ra
sll $v0, $a2, 7
// returns sNonKanjiIndices[(adjusted byte2) + (adjusted byte1) * 0xBC] * FONT_CHAR_TEX_SIZE
mflo a2 /* Unncessary mflo: the pseudo-op `mul` already performs mflo */
add a3, a3, a2
sll a3, a3, 1
la a2, sNonKanjiIndices
add a3, a3, a2
lh a2, (a3)
/* returns sNonKanjiIndices[(adjusted byte2) + (adjusted byte1) * 0xBC] * FONT_CHAR_TEX_SIZE */
sll v0, a2, 7
jr ra
END(Kanji_OffsetFromShiftJIS)
/**
@@ -116,7 +105,7 @@ END(Kanji_OffsetFromShiftJIS)
* into blocks by high byte.
*/
DATA(sNonKanjiIndices)
// 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F
/* 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F */
/* 0x814_ */ .half 0x0000, 0x0001, 0x0002, 0x0003, 0x0004, 0x0005, 0x0006, 0x0007, 0x0008, 0x0009, 0x000A, 0x000B, 0x000C, 0x000D, 0x000E, 0x000F
/* 0x815_ */ .half 0x0010, 0x0011, 0x0012, 0x0013, 0x0014, 0x0015, 0x0016, 0x0017, 0x0018, 0x0019, 0x001A, 0x001B, 0x001C, 0x001D, 0x001E, 0x001F
/* 0x816_ */ .half 0x0020, 0x0021, 0x0022, 0x0023, 0x0024, 0x0025, 0x0026, 0x0027, 0x0028, 0x0029, 0x002A, 0x002B, 0x002C, 0x002D, 0x002E, 0x002F
@@ -130,7 +119,7 @@ DATA(sNonKanjiIndices)
/* 0x81E_ */ .half 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087, 0x0088, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
/* 0x81F_ */ .half 0x0089, 0x008A, 0x008B, 0x008C, 0x008D, 0x008E, 0x008F, 0x0090, 0x0000, 0x0000, 0x0000, 0x0000, 0x0091
// 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F
/* 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F */
/* 0x824_ */ .half 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0092
/* 0x825_ */ .half 0x0093, 0x0094, 0x0095, 0x0096, 0x0097, 0x0098, 0x0099, 0x009A, 0x009B, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
/* 0x826_ */ .half 0x009C, 0x009D, 0x009E, 0x009F, 0x00A0, 0x00A1, 0x00A2, 0x00A3, 0x00A4, 0x00A5, 0x00A6, 0x00A7, 0x00A8, 0x00A9, 0x00AA, 0x00AB
@@ -144,7 +133,7 @@ DATA(sNonKanjiIndices)
/* 0x82E_ */ .half 0x0111, 0x0112, 0x0113, 0x0114, 0x0115, 0x0116, 0x0117, 0x0118, 0x0119, 0x011A, 0x011B, 0x011C, 0x011D, 0x011E, 0x011F, 0x0120
/* 0x82F_ */ .half 0x0121, 0x0122, 0x0123, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
// 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F
/* 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F */
/* 0x834_ */ .half 0x0124, 0x0125, 0x0126, 0x0127, 0x0128, 0x0129, 0x012A, 0x012B, 0x012C, 0x012D, 0x012E, 0x012F, 0x0130, 0x0131, 0x0132, 0x0133
/* 0x835_ */ .half 0x0134, 0x0135, 0x0136, 0x0137, 0x0138, 0x0139, 0x013A, 0x013B, 0x013C, 0x013D, 0x013E, 0x013F, 0x0140, 0x0141, 0x0142, 0x0143
/* 0x836_ */ .half 0x0144, 0x0145, 0x0146, 0x0147, 0x0148, 0x0149, 0x014A, 0x014B, 0x014C, 0x014D, 0x014E, 0x014F, 0x0150, 0x0151, 0x0152, 0x0153
@@ -158,7 +147,7 @@ DATA(sNonKanjiIndices)
/* 0x83E_ */ .half 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
/* 0x83F_ */ .half 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
// 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F
/* 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F */
/* 0x844_ */ .half 0x01AB, 0x01AC, 0x01AD, 0x01AE, 0x01AF, 0x01B0, 0x01B1, 0x01B2, 0x01B3, 0x01B4, 0x01B5, 0x01B6, 0x01B7, 0x01B8, 0x01B9, 0x01BA
/* 0x845_ */ .half 0x01BB, 0x01BC, 0x01BD, 0x01BE, 0x01BF, 0x01C0, 0x01C1, 0x01C2, 0x01C3, 0x01C4, 0x01C5, 0x01C6, 0x01C7, 0x01C8, 0x01C9, 0x01CA
/* 0x846_ */ .half 0x01CB, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
@@ -172,7 +161,7 @@ DATA(sNonKanjiIndices)
/* 0x84E_ */ .half 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
/* 0x84F_ */ .half 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
// 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F
/* 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F */
/* 0x854_ */ .half 0x020E, 0x020F, 0x0210, 0x0211, 0x0212, 0x0213, 0x0214, 0x0215, 0x0216, 0x0217, 0x0218, 0x0219, 0x021A, 0x021B, 0x021C, 0x021D
/* 0x855_ */ .half 0x021E, 0x021F, 0x0220, 0x0221, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0222, 0x0223
/* 0x856_ */ .half 0x0224, 0x0225, 0x0226, 0x0227, 0x0228, 0x0229, 0x022A, 0x022B, 0x022C, 0x022D, 0x022E, 0x022F, 0x0230, 0x0231, 0x0232, 0x0233
@@ -186,7 +175,7 @@ DATA(sNonKanjiIndices)
/* 0x85E_ */ .half 0x026C, 0x026D, 0x026E, 0x026F, 0x0270, 0x0271, 0x0272, 0x0273, 0x0274, 0x0275, 0x0276, 0x0277, 0x0278, 0x0279, 0x027A, 0x027B
/* 0x85F_ */ .half 0x027C, 0x027D, 0x027E, 0x027F, 0x0280, 0x0281, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
// 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F
/* 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F */
/* 0x864_ */ .half 0x0282, 0x0283, 0x0284, 0x0285, 0x0286, 0x0287, 0x0288, 0x0289, 0x028A, 0x028B, 0x028C, 0x028D, 0x028E, 0x028F, 0x0290, 0x0291
/* 0x865_ */ .half 0x0292, 0x0293, 0x0294, 0x0295, 0x0296, 0x0297, 0x0298, 0x0299, 0x029A, 0x029B, 0x029C, 0x029D, 0x029E, 0x029F, 0x0000, 0x0000
/* 0x866_ */ .half 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
@@ -200,7 +189,7 @@ DATA(sNonKanjiIndices)
/* 0x86E_ */ .half 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
/* 0x86F_ */ .half 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
// 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F
/* 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F */
/* 0x874_ */ .half 0x02C0, 0x02C1, 0x02C2, 0x02C3, 0x02C4, 0x02C5, 0x02C6, 0x02C7, 0x02C8, 0x02C9, 0x02CA, 0x02CB, 0x02CC, 0x02CD, 0x02CE, 0x02CF
/* 0x875_ */ .half 0x02D0, 0x02D1, 0x02D2, 0x02D3, 0x02D4, 0x02D5, 0x02D6, 0x02D7, 0x02D8, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
/* 0x876_ */ .half 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
src/code/z_vimode.c
+1 -1
View File
@@ -86,7 +86,7 @@ void ViMode_Configure(ViMode* viMode, s32 type, s32 tvType, s32 loRes, s32 antia
yScaleHiOddField = modeF ? (loResInterlaced ? (F210(0.75) << 16) : (F210(0.5) << 16)) : 0;
viMode->customViMode.type = type;
viMode->customViMode.comRegs.ctrl = VI_CTRL_PIXEL_ADV_3 | VI_CTRL_GAMMA_ON | VI_CTRL_GAMMA_DITHER_ON |
viMode->customViMode.comRegs.ctrl = VI_CTRL_PIXEL_ADV(3) | VI_CTRL_GAMMA_ON | VI_CTRL_GAMMA_DITHER_ON |
(!loResDeinterlaced ? VI_CTRL_SERRATE_ON : 0) |
(antialiasOn ? VI_CTRL_DIVOT_ON : 0) |
(fb32Bit ? VI_CTRL_TYPE_32 : VI_CTRL_TYPE_16);
src/libc/absf.s
+4 -10
View File
@@ -1,15 +1,9 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#if DEBUG_FEATURES
.set noreorder
#endif
.section .text
.balign 16
.text
LEAF(absf)
abs.s $f0, $f12
jr $ra
nop
abs.s fv0, fa0
jr ra
END(absf)
src/libc/sqrt.s
+4 -10
View File
@@ -1,15 +1,9 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#if DEBUG_FEATURES
.set noreorder
#endif
.section .text
.balign 16
.text
LEAF(sqrt)
sqrt.d $f0, $f12
jr $ra
nop
sqrt.d fv0, fa0
jr ra
END(sqrt)
src/libc64/fp.s
+73 -90
View File
@@ -1,12 +1,10 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
.set noreorder
.section .data
.balign 16
.data
#if !PLATFORM_N64
DATA(gPositiveInfinity)
.word 0x7F800000
ENDDATA(gPositiveInfinity)
@@ -34,150 +32,135 @@ ENDDATA(qNaN0x10000)
DATA(sNaN0x3FFFFF)
.word 0x7FFFFFFF
ENDDATA(sNaN0x3FFFFF)
#endif
.section .text
.balign 16
.text
LEAF(floorf)
floor.w.s $f12, $f12
jr $ra
cvt.s.w $f0, $f12
floor.w.s fa0, fa0
cvt.s.w fv0, fa0
jr ra
END(floorf)
LEAF(floor)
floor.w.d $f12, $f12
jr $ra
cvt.d.w $f0, $f12
floor.w.d fa0, fa0
cvt.d.w fv0, fa0
jr ra
END(floor)
LEAF(lfloorf)
floor.w.s $f4, $f12
mfc1 $v0, $f4
jr $ra
nop
floor.w.s ft0, fa0
mfc1 v0, ft0
jr ra
END(lfloorf)
LEAF(lfloor)
floor.w.d $f4, $f12
mfc1 $v0, $f4
jr $ra
nop
floor.w.d ft0, fa0
mfc1 v0, ft0
jr ra
END(lfloor)
LEAF(ceilf)
ceil.w.s $f12, $f12
jr $ra
cvt.s.w $f0, $f12
ceil.w.s fa0, fa0
cvt.s.w fv0, fa0
jr ra
END(ceilf)
LEAF(ceil)
ceil.w.d $f12, $f12
jr $ra
cvt.d.w $f0, $f12
ceil.w.d fa0, fa0
cvt.d.w fv0, fa0
jr ra
END(ceil)
LEAF(lceilf)
ceil.w.s $f4, $f12
mfc1 $v0, $f4
jr $ra
nop
ceil.w.s ft0, fa0
mfc1 v0, ft0
jr ra
END(lceilf)
LEAF(lceil)
ceil.w.d $f4, $f12
mfc1 $v0, $f4
jr $ra
nop
ceil.w.d ft0, fa0
mfc1 v0, ft0
jr ra
END(lceil)
LEAF(truncf)
trunc.w.s $f12, $f12
jr $ra
cvt.s.w $f0, $f12
trunc.w.s fa0, fa0
cvt.s.w fv0, fa0
jr ra
END(truncf)
LEAF(trunc)
trunc.w.d $f12, $f12
jr $ra
cvt.d.w $f0, $f12
trunc.w.d fa0, fa0
cvt.d.w fv0, fa0
jr ra
END(trunc)
LEAF(ltruncf)
trunc.w.s $f4, $f12
mfc1 $v0, $f4
jr $ra
nop
trunc.w.s ft0, fa0
mfc1 v0, ft0
jr ra
END(ltruncf)
LEAF(ltrunc)
trunc.w.d $f4, $f12
mfc1 $v0, $f4
jr $ra
nop
trunc.w.d ft0, fa0
mfc1 v0, ft0
jr ra
END(ltrunc)
LEAF(nearbyintf)
round.w.s $f12, $f12
jr $ra
cvt.s.w $f0, $f12
round.w.s fa0, fa0
cvt.s.w fv0, fa0
jr ra
END(nearbyintf)
LEAF(nearbyint)
round.w.d $f12, $f12
jr $ra
cvt.d.w $f0, $f12
round.w.d fa0, fa0
cvt.d.w fv0, fa0
jr ra
END(nearbyint)
LEAF(lnearbyintf)
round.w.s $f4, $f12
mfc1 $v0, $f4
jr $ra
nop
round.w.s ft0, fa0
mfc1 v0, ft0
jr ra
END(lnearbyintf)
LEAF(lnearbyint)
round.w.d $f4, $f12
mfc1 $v0, $f4
jr $ra
nop
round.w.d ft0, fa0
mfc1 v0, ft0
jr ra
END(lnearbyint)
LEAF(roundf)
li.s $f4, 0.5
nop
add.s $f0, $f12, $f4
floor.w.s $f0, $f0
jr $ra
cvt.s.w $f0, $f0
li.s ft0, 0.5
add.s fv0, fa0, ft0
floor.w.s fv0, fv0
cvt.s.w fv0, fv0
jr ra
END(roundf)
LEAF(round)
li.d $f4, 0.5
nop
add.d $f0, $f12, $f4
floor.w.d $f0, $f0
jr $ra
cvt.d.w $f0, $f0
li.d ft0, 0.5
add.d fv0, fa0, ft0
floor.w.d fv0, fv0
cvt.d.w fv0, fv0
jr ra
END(round)
LEAF(lroundf)
li.s $f4, 0.5
nop
add.s $f0, $f12, $f4
floor.w.s $f0, $f0
mfc1 $v0, $f0
jr $ra
nop
li.s ft0, 0.5
add.s fv0, fa0, ft0
floor.w.s fv0, fv0
mfc1 v0, fv0
jr ra
END(lroundf)
LEAF(lround)
li.d $f4, 0.5
nop
add.d $f0, $f12, $f4
floor.w.d $f0, $f0
mfc1 $v0, $f0
jr $ra
nop
li.d ft0, 0.5
add.d fv0, fa0, ft0
floor.w.d fv0, fv0
mfc1 v0, fv0
jr ra
END(lround)
src/libleo/api/getaadr.s
+25 -34
View File
@@ -1,42 +1,33 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
.set noat
.set noreorder
.section .text
.balign 16
.text
LEAF(LeoGetAAdr)
bltz $a0, .out_of_range
slti $at, $a0, 0x908
beqz $at, .out_of_range
nop
lui $v1, %hi(asc2tbl)
sll $t0, $a0, 2
addiu $v1, %lo(asc2tbl)
add $t1, $t0, $v1
lbu $t8, 2($t1)
lhu $t9, 0($t1)
li $at, 0x70000
andi $t2, $t8, 0xF
addi $t3, $t2, 1
sw $t3, ($a2)
lb $t0, 3($t1)
srl $t4, $t8, 4
ori $at, $at, 0xEE80
andi $t5, $t0, 1
sll $t6, $t5, 4
or $t7, $t6, $t4
sw $t7, ($a1)
sll $v0, $t9, 1
sra $v1, $t0, 1
sw $v1, ($a3)
jr $ra
add $v0, $v0, $at
bltz a0, .out_of_range
bge a0, 0x908, .out_of_range
sll t0, a0, 2
la v1, asc2tbl
add t1, t0, v1
lbu t8, 2(t1)
lhu t9, 0(t1)
andi t2, t8, 0xF
addi t3, t2, 1
sw t3, (a2)
lb t0, 3(t1)
srl t4, t8, 4
andi t5, t0, 1
sll t6, t5, 4
or t7, t6, t4
sw t7, (a1)
sll v0, t9, 1
sra v1, t0, 1
sw v1, (a3)
add v0, v0, 0x7EE80
jr ra
.out_of_range:
jr $ra
li $v0, -1
li v0, -1
jr ra
END(LeoGetAAdr)
DATA(asc2tbl)
src/libleo/api/getaadr2.s
+20 -28
View File
@@ -1,33 +1,25 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
.set noat
.set noreorder
.section .text
.balign 16
.text
LEAF(LeoGetAAdr2)
li $at, 0x10000
srl $t8, $a0, 0xF
ori $at, $at, 0xFFFE
and $v1, $t8, $at
li $at, 0x70000
ori $at, $at, 0xEE80
srl $t2, $a0, 0x8
add $v0, $v1, $at
andi $t3, $t2, 0xF
andi $t5, $a0, 0xFE
addi $t4, $t3, 1
sll $t6, $t5, 0x18
andi $t9, $a0, 0x1
srl $v1, $a0, 0xC
sw $t4, ($a2)
sra $t7, $t6, 0x19
sll $t8, $t9, 0x4
andi $t1, $v1, 0xF
sw $t7, ($a3)
or $t2, $t1, $t8
jr $ra
sw $t2, ($a1)
srl t8, a0, 15
and v1, t8, 0x1FFFE
add v0, v1, 0x7EE80
srl t2, a0, 8
andi t3, t2, 0xF
andi t5, a0, 0xFE
addi t4, t3, 1
sll t6, t5, 24
andi t9, a0, 1
srl v1, a0, 12
sw t4, (a2)
sra t7, t6, 25
sll t8, t9, 4
andi t1, v1, 0xF
sw t7, (a3)
or t2, t1, t8
sw t2, (a1)
jr ra
END(LeoGetAAdr2)
src/libleo/api/getkadr.s
+59 -76
View File
@@ -1,11 +1,7 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
.set noat
.set noreorder
.section .text
.balign 16
.text
/**
* int LeoGetKAdr(int sjis);
@@ -22,74 +18,61 @@
* @return int Storage offset address into the N64 Disk Drive mask ROM.
*/
LEAF(LeoGetKAdr)
// Check if the codepoint is in the range 0x8140 to 0x9872.
li $at, 0x8140
slt $at, $a0, $at
bnez $at, .out_of_range
li $at, 0x9873
slt $at, $a0, $at
beqz $at, .out_of_range
// Characters with codepoints >= 0x8800 are kanji. Arrangement is regular,
// so convert index directly.
li $at, 0x8800
slt $at, $a0, $at
bnez $at, .nonkanji
// 0xBC is number of glyphs in one block in the `kanji` file:
// 0x100 possible codepoints with the same byte1
// - 0x40 unused at beginning
// - 1 unused at 0x7F
// - 3 unused at 0xFD, 0xFE, 0xFF
li $a2, 0xBC
// Get byte1 and adjust so starts at 0
srl $a1, $a0, 8
addi $a1, $a1, -0x88
multu $a2, $a1
// Get byte2 and adjust so starts at 0
andi $a3, $a0, 0xFF
addi $a3, $a3, -0x40
slti $at, $a3, (0x80 - 0x40)
mflo $a2
// 0x__7F is always empty and elided in the file, so decrement if larger
bnezl $at, .kanji_lower_halfblock
mflo $a2
addi $a3, $a3, -1
mflo $a2
/* Check if the codepoint is in the range 0x8140 to 0x9872. */
blt a0, 0x8140, .out_of_range
bge a0, 0x9873, .out_of_range
/* Characters with codepoints >= 0x8800 are kanji. Arrangement is regular,
* so convert index directly. */
li a2, 0xBC
blt a0, 0x8800, .nonkanji
/* 0xBC is number of glyphs in one block in the `kanji` file:
* 0x100 possible codepoints with the same byte1
* - 0x40 unused at beginning
* - 1 unused at 0x7F
* - 3 unused at 0xFD, 0xFE, 0xFF */
/* Get byte1 and adjust so starts at 0 */
srl a1, a0, 8
addi a1, a1, -0x88
mul a2, a2, a1
/* Get byte2 and adjust so starts at 0 */
andi a3, a0, 0xFF
addi a3, a3, -0x40
/* 0x__7F is always empty and elided in the file, so decrement if larger */
blt a3, 0x40, .kanji_lower_halfblock
addi a3, a3, -1
.kanji_lower_halfblock:
addi $a3, $a3, 0x30A
add $a3, $a3, $a2
jr $ra
sll $v0, $a3, 7
// returns (0x30A + (adjusted byte2) + (adjusted byte1) * 0xBC) * FONT_CHAR_TEX_SIZE
mflo a2
addi a3, a3, 0x30A
add a3, a3, a2
sll v0, a3, 7
jr ra
/* returns (0x30A + (adjusted byte2) + (adjusted byte1) * 0xBC) * FONT_CHAR_TEX_SIZE */
// Non-kanji are arranged with irregular gaps, use the lookup table.
/* Non-kanji are arranged with irregular gaps, use the lookup table. */
.nonkanji:
// Get byte1 and adjust so starts at 0
srl $a1, $a0, 8
addi $a1, $a1, -0x81
multu $a2, $a1
// Get byte2 and adjust so starts at 0
andi $a3, $a0, 0xFF
addi $a3, $a3, -0x40
slti $at, $a3, (0x80 - 0x40)
mflo $a2
// 0x__7F is always empty and elided in the file, so decrement if larger
bnezl $at, .nonkanji_lower_halfblock
mflo $a2
addi $a3, $a3, -1
mflo $a2
/* Get byte1 and adjust so starts at 0 */
srl a1, a0, 8
addi a1, a1, -0x81
mul a2, a2, a1
/* Get byte2 and adjust so starts at 0 */
andi a3, a0, 0xFF
addi a3, a3, -0x40
/* 0x__7F is always empty and elided in the file, so decrement if larger */
blt a3, 0x40, .nonkanji_lower_halfblock
addi a3, a3, -1
.nonkanji_lower_halfblock:
add $a3, $a3, $a2
lui $a2, %hi(kantbl)
sll $a3, $a3, 1
addiu $a2, %lo(kantbl)
add $a3, $a3, $a2
lh $a2, ($a3)
jr $ra
sll $v0, $a2, 7
// returns kantbl[(adjusted byte2) + (adjusted byte1) * 0xBC] * FONT_CHAR_TEX_SIZE
mflo a2
add a3, a3, a2
sll a3, a3, 1
la a2, kantbl
add a3, a3, a2
lh a2, (a3)
sll v0, a2, 7
jr ra
/* returns kantbl[(adjusted byte2) + (adjusted byte1) * 0xBC] * FONT_CHAR_TEX_SIZE */
.out_of_range:
jr $ra
li $v0, -1
li v0, -1
jr ra
END(LeoGetKAdr)
/**
@@ -129,7 +112,7 @@ END(LeoGetKAdr)
* into blocks by high byte.
*/
DATA(kantbl)
// 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F
/* 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F */
/* 0x814_ */ .half 0x0000, 0x0001, 0x0002, 0x0003, 0x0004, 0x0005, 0x0006, 0x0007, 0x0008, 0x0009, 0x000A, 0x000B, 0x000C, 0x000D, 0x000E, 0x000F
/* 0x815_ */ .half 0x0010, 0x0011, 0x0012, 0x0013, 0x0014, 0x0015, 0x0016, 0x0017, 0x0018, 0x0019, 0x001A, 0x001B, 0x001C, 0x001D, 0x001E, 0x001F
/* 0x816_ */ .half 0x0020, 0x0021, 0x0022, 0x0023, 0x0024, 0x0025, 0x0026, 0x0027, 0x0028, 0x0029, 0x002A, 0x002B, 0x002C, 0x002D, 0x002E, 0x002F
@@ -143,7 +126,7 @@ DATA(kantbl)
/* 0x81E_ */ .half 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087, 0x0088, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
/* 0x81F_ */ .half 0x0089, 0x008A, 0x008B, 0x008C, 0x008D, 0x008E, 0x008F, 0x0090, 0x0000, 0x0000, 0x0000, 0x0000, 0x0091
// 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F
/* 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F */
/* 0x824_ */ .half 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0092
/* 0x825_ */ .half 0x0093, 0x0094, 0x0095, 0x0096, 0x0097, 0x0098, 0x0099, 0x009A, 0x009B, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
/* 0x826_ */ .half 0x009C, 0x009D, 0x009E, 0x009F, 0x00A0, 0x00A1, 0x00A2, 0x00A3, 0x00A4, 0x00A5, 0x00A6, 0x00A7, 0x00A8, 0x00A9, 0x00AA, 0x00AB
@@ -157,7 +140,7 @@ DATA(kantbl)
/* 0x82E_ */ .half 0x0111, 0x0112, 0x0113, 0x0114, 0x0115, 0x0116, 0x0117, 0x0118, 0x0119, 0x011A, 0x011B, 0x011C, 0x011D, 0x011E, 0x011F, 0x0120
/* 0x82F_ */ .half 0x0121, 0x0122, 0x0123, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
// 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F
/* 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F */
/* 0x834_ */ .half 0x0124, 0x0125, 0x0126, 0x0127, 0x0128, 0x0129, 0x012A, 0x012B, 0x012C, 0x012D, 0x012E, 0x012F, 0x0130, 0x0131, 0x0132, 0x0133
/* 0x835_ */ .half 0x0134, 0x0135, 0x0136, 0x0137, 0x0138, 0x0139, 0x013A, 0x013B, 0x013C, 0x013D, 0x013E, 0x013F, 0x0140, 0x0141, 0x0142, 0x0143
/* 0x836_ */ .half 0x0144, 0x0145, 0x0146, 0x0147, 0x0148, 0x0149, 0x014A, 0x014B, 0x014C, 0x014D, 0x014E, 0x014F, 0x0150, 0x0151, 0x0152, 0x0153
@@ -171,7 +154,7 @@ DATA(kantbl)
/* 0x83E_ */ .half 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
/* 0x83F_ */ .half 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
// 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F
/* 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F */
/* 0x844_ */ .half 0x01AB, 0x01AC, 0x01AD, 0x01AE, 0x01AF, 0x01B0, 0x01B1, 0x01B2, 0x01B3, 0x01B4, 0x01B5, 0x01B6, 0x01B7, 0x01B8, 0x01B9, 0x01BA
/* 0x845_ */ .half 0x01BB, 0x01BC, 0x01BD, 0x01BE, 0x01BF, 0x01C0, 0x01C1, 0x01C2, 0x01C3, 0x01C4, 0x01C5, 0x01C6, 0x01C7, 0x01C8, 0x01C9, 0x01CA
/* 0x846_ */ .half 0x01CB, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
@@ -185,7 +168,7 @@ DATA(kantbl)
/* 0x84E_ */ .half 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
/* 0x84F_ */ .half 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
// 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F
/* 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F */
/* 0x854_ */ .half 0x020E, 0x020F, 0x0210, 0x0211, 0x0212, 0x0213, 0x0214, 0x0215, 0x0216, 0x0217, 0x0218, 0x0219, 0x021A, 0x021B, 0x021C, 0x021D
/* 0x855_ */ .half 0x021E, 0x021F, 0x0220, 0x0221, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0222, 0x0223
/* 0x856_ */ .half 0x0224, 0x0225, 0x0226, 0x0227, 0x0228, 0x0229, 0x022A, 0x022B, 0x022C, 0x022D, 0x022E, 0x022F, 0x0230, 0x0231, 0x0232, 0x0233
@@ -199,7 +182,7 @@ DATA(kantbl)
/* 0x85E_ */ .half 0x026C, 0x026D, 0x026E, 0x026F, 0x0270, 0x0271, 0x0272, 0x0273, 0x0274, 0x0275, 0x0276, 0x0277, 0x0278, 0x0279, 0x027A, 0x027B
/* 0x85F_ */ .half 0x027C, 0x027D, 0x027E, 0x027F, 0x0280, 0x0281, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
// 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F
/* 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F */
/* 0x864_ */ .half 0x0282, 0x0283, 0x0284, 0x0285, 0x0286, 0x0287, 0x0288, 0x0289, 0x028A, 0x028B, 0x028C, 0x028D, 0x028E, 0x028F, 0x0290, 0x0291
/* 0x865_ */ .half 0x0292, 0x0293, 0x0294, 0x0295, 0x0296, 0x0297, 0x0298, 0x0299, 0x029A, 0x029B, 0x029C, 0x029D, 0x029E, 0x029F, 0x0000, 0x0000
/* 0x866_ */ .half 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
@@ -213,7 +196,7 @@ DATA(kantbl)
/* 0x86E_ */ .half 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
/* 0x86F_ */ .half 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
// 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F
/* 0x___0 0x___1 0x___2 0x___3 0x___4 0x___5 0x___6 0x___7 0x___8 0x___9 0x___A 0x___B 0x___C 0x___D 0x___E 0x___F */
/* 0x874_ */ .half 0x02C0, 0x02C1, 0x02C2, 0x02C3, 0x02C4, 0x02C5, 0x02C6, 0x02C7, 0x02C8, 0x02C9, 0x02CA, 0x02CB, 0x02CC, 0x02CD, 0x02CE, 0x02CF
/* 0x875_ */ .half 0x02D0, 0x02D1, 0x02D2, 0x02D3, 0x02D4, 0x02D5, 0x02D6, 0x02D7, 0x02D8, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
/* 0x876_ */ .half 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000
src/libultra/gu/libm_vals.s
+1 -3
View File
@@ -1,8 +1,6 @@
#include "ultra64/asm.h"
.section .rodata
.balign 16
.rdata
DATA(__libm_qnan_f)
.word 0x7F810000
src/libultra/io/vimodefpallan1.c
+9 -9
View File
@@ -16,15 +16,15 @@ OSViMode osViModeFpalLan1 = {
{
// comRegs
VI_CTRL_TYPE_16 | VI_CTRL_GAMMA_DITHER_ON | VI_CTRL_GAMMA_ON | VI_CTRL_DIVOT_ON | VI_CTRL_ANTIALIAS_MODE_1 |
VI_CTRL_PIXEL_ADV_3, // ctrl
WIDTH(320), // width
BURST(58, 30, 4, 69), // burst
VSYNC(625), // vSync
HSYNC(3177, 23), // hSync
LEAP(3183, 3181), // leap
HSTART(128, 768), // hStart
SCALE(2, 0), // xScale
VCURRENT(0), // vCurrent
VI_CTRL_PIXEL_ADV(3), // ctrl
WIDTH(320), // width
BURST(58, 30, 4, 69), // burst
VSYNC(625), // vSync
HSYNC(3177, 23), // hSync
LEAP(3183, 3181), // leap
HSTART(128, 768), // hStart
SCALE(2, 0), // xScale
VCURRENT(0), // vCurrent
},
{ // fldRegs
{
src/libultra/io/vimodempallan1.c
+9 -9
View File
@@ -16,15 +16,15 @@ OSViMode osViModeMpalLan1 = {
{
// comRegs
VI_CTRL_TYPE_16 | VI_CTRL_GAMMA_DITHER_ON | VI_CTRL_GAMMA_ON | VI_CTRL_DIVOT_ON | VI_CTRL_ANTIALIAS_MODE_1 |
VI_CTRL_PIXEL_ADV_3, // ctrl
WIDTH(320), // width
BURST(57, 30, 5, 70), // burst
VSYNC(525), // vSync
HSYNC(3089, 4), // hSync
LEAP(3097, 3098), // leap
HSTART(108, 748), // hStart
SCALE(2, 0), // xScale
VCURRENT(0), // vCurrent
VI_CTRL_PIXEL_ADV(3), // ctrl
WIDTH(320), // width
BURST(57, 30, 5, 70), // burst
VSYNC(525), // vSync
HSYNC(3089, 4), // hSync
LEAP(3097, 3098), // leap
HSTART(108, 748), // hStart
SCALE(2, 0), // xScale
VCURRENT(0), // vCurrent
},
{ // fldRegs
{
src/libultra/io/vimodentsclan1.c
+9 -9
View File
@@ -16,15 +16,15 @@ OSViMode osViModeNtscLan1 = {
{
// comRegs
VI_CTRL_TYPE_16 | VI_CTRL_GAMMA_DITHER_ON | VI_CTRL_GAMMA_ON | VI_CTRL_DIVOT_ON | VI_CTRL_ANTIALIAS_MODE_1 |
VI_CTRL_PIXEL_ADV_3, // ctrl
WIDTH(320), // width
BURST(57, 34, 5, 62), // burst
VSYNC(525), // vSync
HSYNC(3093, 0), // hSync
LEAP(3093, 3093), // leap
HSTART(108, 748), // hStart
SCALE(2, 0), // xScale
VCURRENT(0), // vCurrent
VI_CTRL_PIXEL_ADV(3), // ctrl
WIDTH(320), // width
BURST(57, 34, 5, 62), // burst
VSYNC(525), // vSync
HSYNC(3093, 0), // hSync
LEAP(3093, 3093), // leap
HSTART(108, 748), // hStart
SCALE(2, 0), // xScale
VCURRENT(0), // vCurrent
},
{ // fldRegs
{
src/libultra/io/vimodepallan1.c
+9 -9
View File
@@ -16,15 +16,15 @@ OSViMode osViModePalLan1 = {
{
// comRegs
VI_CTRL_TYPE_16 | VI_CTRL_GAMMA_DITHER_ON | VI_CTRL_GAMMA_ON | VI_CTRL_DIVOT_ON | VI_CTRL_ANTIALIAS_MODE_1 |
VI_CTRL_PIXEL_ADV_3, // ctrl
WIDTH(320), // width
BURST(58, 30, 4, 69), // burst
VSYNC(625), // vSync
HSYNC(3177, 23), // hSync
LEAP(3183, 3181), // leap
HSTART(128, 768), // hStart
SCALE(2, 0), // xScale
VCURRENT(0), // vCurrent
VI_CTRL_PIXEL_ADV(3), // ctrl
WIDTH(320), // width
BURST(58, 30, 4, 69), // burst
VSYNC(625), // vSync
HSYNC(3177, 23), // hSync
LEAP(3183, 3181), // leap
HSTART(128, 768), // hStart
SCALE(2, 0), // xScale
VCURRENT(0), // vCurrent
},
{ // fldRegs
{
src/libultra/libc/bcmp.s
+72 -76
View File
@@ -1,93 +1,89 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
.set noat
.set noreorder
.section .text
.balign 16
.text
LEAF(bcmp)
slti $at, $a2, 0x10
bnez $at, bytecmp
xor $v0, $a0, $a1
andi $v0, $v0, 3
bnez $v0, unaligncmp
negu $t8, $a0
andi $t8, $t8, 3
beqz $t8, wordcmp
subu $a2, $a2, $t8
move $v0, $v1
lwl $v0, ($a0)
lwl $v1, ($a1)
addu $a0, $a0, $t8
addu $a1, $a1, $t8
bne $v0, $v1, cmpne
xor v0, a0, a1
blt a2, 0x10, bytecmp
and v0, v0, 3
negu t8, a0
bnez v0, unaligncmp
and t8, t8, 3
subu a2, a2, t8
beqz t8, wordcmp
move v0, v1
lwl v0, (a0)
lwl v1, (a1)
addu a0, a0, t8
addu a1, a1, t8
bne v0, v1, cmpne
wordcmp:
li $at, ~3
and $a3, $a2, $at
beqz $a3, bytecmp
subu $a2, $a2, $a3
addu $a3, $a3, $a0
lw $v0, ($a0)
and a3, a2, ~3
subu a2, a2, a3
beqz a3, bytecmp
addu a3, a3, a0
1:
lw $v1, ($a1)
addiu $a0, $a0, 4
addiu $a1, $a1, 4
bne $v0, $v1, cmpne
nop
bnel $a0, $a3, 1b
lw $v0, ($a0)
lw v0, (a0)
lw v1, (a1)
addu a0, a0, 4
addu a1, a1, 4
bne v0, v1, cmpne
bne a0, a3, 1b
b bytecmp
nop
unaligncmp:
negu $a3, $a1
andi $a3, $a3, 3
beqz $a3, partaligncmp
subu $a2, $a2, $a3
addu $a3, $a3, $a0
lbu $v0, ($a0)
negu a3, a1
and a3, a3, 3
subu a2, a2, a3
beqz a3, partaligncmp
addu a3, a3, a0
1:
lbu $v1, ($a1)
addiu $a0, $a0, 1
addiu $a1, $a1, 1
bne $v0, $v1, cmpne
nop
bnel $a0, $a3, 1b
lbu $v0, ($a0)
lbu v0, (a0)
lbu v1, (a1)
addu a0, a0, 1
addu a1, a1, 1
bne v0, v1, cmpne
bne a0, a3, 1b
partaligncmp:
li $at, ~3
and $a3, $a2, $at
beqz $a3, bytecmp
subu $a2, $a2, $a3
addu $a3, $a3, $a0
lwl $v0, ($a0)
and a3, a2, ~3
subu a2, a2, a3
beqz a3, bytecmp
addu a3, a3, a0
1:
lw $v1, ($a1)
lwr $v0, 3($a0)
addiu $a0, $a0, 4
addiu $a1, $a1, 4
bne $v0, $v1, cmpne
nop
bnel $a0, $a3, 1b
lwl $v0, ($a0)
lwl v0, (a0)
lw v1, (a1)
lwr v0, 3(a0)
addu a0, a0, 4
addu a1, a1, 4
bne v0, v1, cmpne
bne a0, a3, 1b
bytecmp:
blez $a2, cmpdone
addu $a3, $a2, $a0
lbu $v0, ($a0)
addu a3, a2, a0
blez a2, cmpdone
1:
lbu $v1, ($a1)
addiu $a0, $a0, 1
addiu $a1, $a1, 1
bne $v0, $v1, cmpne
nop
bnel $a0, $a3, 1b
lbu $v0, ($a0)
lbu v0, (a0)
lbu v1, (a1)
addu a0, a0, 1
addu a1, a1, 1
bne v0, v1, cmpne
bne a0, a3, 1b
cmpdone:
jr $ra
move $v0, $zero
move v0, zero
jr ra
cmpne:
jr $ra
li $v0, 1
li v0, 1
jr ra
END(bcmp)
src/libultra/libc/bcopy.s
+172 -194
View File
@@ -1,233 +1,211 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
.set noat
.set noreorder
.section .text
.balign 16
.text
LEAF(bcopy)
beqz $a2, ret
move $a3, $a1
beq $a0, $a1, ret
slt $at, $a1, $a0
bnezl $at, goforwards
slti $at, $a2, 0x10
add $v0, $a0, $a2
slt $at, $a1, $v0
beql $at, $zero, goforwards
slti $at, $a2, 0x10
move a3, a1
beqz a2, ret
beq a0, a1, ret
blt a1, a0, goforwards
add v0, a0, a2
bge a1, v0, goforwards
b gobackwards
slti $at, $a2, 0x10
slti $at, $a2, 0x10
goforwards:
bnez $at, forwards_bytecopy
nop
andi $v0, $a0, 3
andi $v1, $a1, 3
beq $v0, $v1, forwalignable
nop
blt a2, 0x10, forwards_bytecopy
and v0, a0, 3
and v1, a1, 3
beq v0, v1, forwalignable
forwards_bytecopy:
beqz $a2, ret
nop
addu $v1, $a0, $a2
beqz a2, ret
addu v1, a0, a2
99:
lb $v0, ($a0)
addiu $a0, $a0, 1
addiu $a1, $a1, 1
bne $a0, $v1, 99b
sb $v0, -1($a1)
lb v0, (a0)
addu a0, a0, 1
sb v0, (a1)
addu a1, a1, 1
bne a0, v1, 99b
ret:
jr $ra
move $v0, $a3
move v0, a3
jr ra
forwalignable:
beqz $v0, forwards_32
li $at, 1
beq $v0, $at, forw_copy3
li $at, 2
beql $v0, $at, forw_copy2
lh $v0, ($a0)
lb $v0, ($a0)
addiu $a0, $a0, 1
addiu $a1, $a1, 1
addiu $a2, $a2, -1
beqz v0, forwards_32
beq v0, 1, forw_copy3
beq v0, 2, forw_copy2
lb v0, (a0)
addu a0, a0, 1
sb v0, (a1)
addu a1, a1, 1
addu a2, a2, -1
b forwards_32
sb $v0, -1($a1)
lh $v0, ($a0)
forw_copy2:
addiu $a0, $a0, 2
addiu $a1, $a1, 2
addiu $a2, $a2, -2
lh v0, (a0)
addu a0, a0, 2
sh v0, (a1)
addu a1, a1, 2
addu a2, a2, -2
b forwards_32
sh $v0, -2($a1)
forw_copy3:
lb $v0, ($a0)
lh $v1, 1($a0)
addiu $a0, $a0, 3
addiu $a1, $a1, 3
addiu $a2, $a2, -3
sb $v0, -3($a1)
sh $v1, -2($a1)
lb v0, (a0)
lh v1, 1(a0)
addiu a0, a0, 3
sb v0, (a1)
sh v1, 1(a1)
addiu a1, a1, 3
addiu a2, a2, -3
forwards:
forwards_32:
slti $at, $a2, 0x20
bnezl $at, forwards_16_
slti $at, $a2, 0x10
lw $v0, ($a0)
lw $v1, 4($a0)
lw $t0, 8($a0)
lw $t1, 0xC($a0)
lw $t2, 0x10($a0)
lw $t3, 0x14($a0)
lw $t4, 0x18($a0)
lw $t5, 0x1C($a0)
addiu $a0, $a0, 0x20
addiu $a1, $a1, 0x20
addiu $a2, $a2, -0x20
sw $v0, -0x20($a1)
sw $v1, -0x1C($a1)
sw $t0, -0x18($a1)
sw $t1, -0x14($a1)
sw $t2, -0x10($a1)
sw $t3, -0xC($a1)
sw $t4, -8($a1)
blt a2, 32, forwards_16
lw v0, 0(a0)
lw v1, 4(a0)
lw t0, 8(a0)
lw t1, 12(a0)
lw t2, 16(a0)
lw t3, 20(a0)
lw t4, 24(a0)
lw t5, 28(a0)
addiu a0, a0, 32
sw v0, 0(a1)
sw v1, 4(a1)
sw t0, 8(a1)
sw t1, 12(a1)
sw t2, 16(a1)
sw t3, 20(a1)
sw t4, 24(a1)
sw t5, 28(a1)
addiu a1, a1, 32
addiu a2, a2, -32
b forwards_32
sw $t5, -4($a1)
forwards_16:
slti $at, $a2, 0x10
forwards_16_: // fake label due to branch likely optimization
bnezl $at, forwards_4_
slti $at, $a2, 4
lw $v0, ($a0)
lw $v1, 4($a0)
lw $t0, 8($a0)
lw $t1, 0xC($a0)
addiu $a0, $a0, 0x10
addiu $a1, $a1, 0x10
addiu $a2, $a2, -0x10
sw $v0, -0x10($a1)
sw $v1, -0xC($a1)
sw $t0, -8($a1)
blt a2, 16, forwards_4
lw v0, 0(a0)
lw v1, 4(a0)
lw t0, 8(a0)
lw t1, 12(a0)
addiu a0, a0, 16
sw v0, 0(a1)
sw v1, 4(a1)
sw t0, 8(a1)
sw t1, 12(a1)
addiu a1, a1, 16
addiu a2, a2, -16
b forwards_16
sw $t1, -4($a1)
forwards_4:
slti $at, $a2, 4
forwards_4_: // fake label due to branch likely optimization
bnez $at, forwards_bytecopy
nop
lw $v0, ($a0)
addiu $a0, $a0, 4
addiu $a1, $a1, 4
addiu $a2, $a2, -4
blt a2, 4, forwards_bytecopy
lw v0, 0(a0)
addiu a0, a0, 4
sw v0, 0(a1)
addiu a1, a1, 4
addiu a2, a2, -4
b forwards_4
sw $v0, -4($a1)
slti $at, $a2, 0x10
gobackwards:
add $a0, $a0, $a2
bnez $at, backwards_bytecopy
add $a1, $a1, $a2
andi $v0, $a0, 3
andi $v1, $a1, 3
beq $v0, $v1, backalignable
nop
add a0, a0,a2
add a1, a1,a2
blt a2, 16, backwards_bytecopy
andi v0, a0, 0x3
andi v1, a1, 0x3
beq v0, v1, backalignable
backwards_bytecopy:
beqz $a2, ret
nop
addiu $a0, $a0, -1
addiu $a1, $a1, -1
subu $v1, $a0, $a2
beqz a2, ret
addiu a0, a0, -1
addiu a1, a1, -1
subu v1, a0,a2
99:
lb $v0, ($a0)
addiu $a0, $a0, -1
addiu $a1, $a1, -1
bne $a0, $v1, 99b
sb $v0, 1($a1)
jr $ra
move $v0, $a3
lb v0, 0(a0)
addiu a0, a0, -1
sb v0, 0(a1)
addiu a1, a1, -1
bne a0, v1,99b
move v0, a3
jr ra
backalignable:
beqz $v0, backwards_32
li $at, 3
beq $v0, $at, back_copy3
li $at, 2
beql $v0, $at, back_copy2
lh $v0, -2($a0)
lb $v0, -1($a0)
addiu $a0, $a0, -1
addiu $a1, $a1, -1
addiu $a2, $a2, -1
b backwards_32
sb $v0, ($a1)
lh $v0, -2($a0)
beqz v0, backwards
beq v0, 3, back_copy3
beq v0, 2, back_copy2
lb v0, -1(a0)
addiu a0, a0, -1
sb v0, -1(a1)
addiu a1, a1, -1
addiu a2, a2, -1
b backwards
back_copy2:
addiu $a0, $a0, -2
addiu $a1, $a1, -2
addiu $a2, $a2, -2
b backwards_32
sh $v0, ($a1)
lh v0, -2(a0)
addiu a0, a0, -2
sh v0, -2(a1)
addiu a1, a1, -2
addiu a2, a2, -2
b backwards
back_copy3:
lb $v0, -1($a0)
lh $v1, -3($a0)
addiu $a0, $a0, -3
addiu $a1, $a1, -3
addiu $a2, $a2, -3
sb $v0, 2($a1)
sh $v1, ($a1)
lb v0, -1(a0)
lh v1, -3(a0)
addiu a0, a0, -3
sb v0, -1(a1)
sh v1, -3(a1)
addiu a1, a1, -3
addiu a2, a2, -3
backwards:
backwards_32:
slti $at, $a2, 0x20
bnezl $at, backwards_16_
slti $at, $a2, 0x10
lw $v0, -4($a0)
lw $v1, -8($a0)
lw $t0, -0xc($a0)
lw $t1, -0x10($a0)
lw $t2, -0x14($a0)
lw $t3, -0x18($a0)
lw $t4, -0x1c($a0)
lw $t5, -0x20($a0)
addiu $a0, $a0, -0x20
addiu $a1, $a1, -0x20
addiu $a2, $a2, -0x20
sw $v0, 0x1C($a1)
sw $v1, 0x18($a1)
sw $t0, 0x14($a1)
sw $t1, 0x10($a1)
sw $t2, 0xC($a1)
sw $t3, 8($a1)
sw $t4, 4($a1)
blt a2, 32, backwards_16
lw v0, -4(a0)
lw v1, -8(a0)
lw t0, -12(a0)
lw t1, -16(a0)
lw t2, -20(a0)
lw t3, -24(a0)
lw t4, -28(a0)
lw t5, -32(a0)
addiu a0, a0, -32
sw v0, -4(a1)
sw v1, -8(a1)
sw t0, -12(a1)
sw t1, -16(a1)
sw t2, -20(a1)
sw t3, -24(a1)
sw t4, -28(a1)
sw t5, -32(a1)
addiu a1, a1, -32
addiu a2, a2, -32
b backwards_32
sw $t5, ($a1)
backwards_16:
slti $at, $a2, 0x10
backwards_16_: // fake label due to branch likely optimization
bnezl $at, backwards_4_
slti $at, $a2, 4
lw $v0, -4($a0)
lw $v1, -8($a0)
lw $t0, -0xC($a0)
lw $t1, -0x10($a0)
addiu $a0, $a0, -0x10
addiu $a1, $a1, -0x10
addiu $a2, $a2, -0x10
sw $v0, 0xC($a1)
sw $v1, 8($a1)
sw $t0, 4($a1)
blt a2, 16, backwards_4
lw v0, -4(a0)
lw v1, -8(a0)
lw t0, -12(a0)
lw t1, -16(a0)
addiu a0, a0, -16
sw v0, -4(a1)
sw v1, -8(a1)
sw t0, -12(a1)
sw t1, -16(a1)
addiu a1, a1, -16
addiu a2, a2, -16
b backwards_16
sw $t1, ($a1)
backwards_4:
slti $at, $a2, 4
backwards_4_: // fake label due to branch likely optimization
bnez $at, backwards_bytecopy
nop
lw $v0, -4($a0)
addiu $a0, $a0, -4
addiu $a1, $a1, -4
addiu $a2, $a2, -4
blt a2, 4, backwards_bytecopy
lw v0, -4(a0)
addiu a0, a0, -4
sw v0, -4(a1)
addiu a1, a1, -4
addiu a2, a2, -4
b backwards_4
sw $v0, ($a1)
END(bcopy)
src/libultra/libc/bzero.s
+47 -53
View File
@@ -1,65 +1,59 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
.set noat
.set noreorder
.section .text
.balign 16
.text
LEAF(bzero)
slti $at, $a1, 0xC
bnez $at, bytezero
negu $v1, $a0
andi $v1, $v1, 3
beqz $v1, blkzero
subu $a1, $a1, $v1
swl $zero, ($a0)
addu $a0, $a0, $v1
negu v1, a0
blt a1, 0xC, bytezero
and v1, v1, 3
subu a1, a1, v1
beqz v1, blkzero
swl zero, (a0)
addu a0, a0, v1
blkzero:
// align backwards to 0x20
li $at, ~0x1F
and $a3, $a1, $at
// If the result is zero, the amount to zero is less than 0x20 bytes
beqz $a3, wordzero
subu $a1, $a1, $a3
// zero in blocks of 0x20 at a time
addu $a3, $a3, $a0
/* align backwards to 0x20 */
and a3, a1, ~(0x20 - 1)
/* If the result is zero, the amount to zero is less than 0x20 bytes */
subu a1, a1, a3
beqz a3, wordzero
/* zero in blocks of 0x20 at a time */
addu a3, a3, a0
1:
addiu $a0, $a0, 0x20
sw $zero, -0x20($a0)
sw $zero, -0x1C($a0)
sw $zero, -0x18($a0)
sw $zero, -0x14($a0)
sw $zero, -0x10($a0)
sw $zero, -0xC($a0)
sw $zero, -8($a0)
bne $a0, $a3, 1b
sw $zero, -4($a0)
sw zero, 0(a0)
sw zero, 4(a0)
sw zero, 8(a0)
sw zero, 12(a0)
addiu a0, a0, 0x20
sw zero, -16(a0)
sw zero, -12(a0)
sw zero, -8(a0)
sw zero, -4(a0)
bne a0, a3, 1b
wordzero:
// align backwards to 0x4
li $at, ~3
and $a3, $a1, $at
// If the result is zero, the amount to zero is less than 0x4 bytes
beqz $a3, bytezero
subu $a1, $a1, $a3
// zero one word at a time
addu $a3, $a3, $a0
/* align backwards to 0x4 */
and a3, a1, ~3
/* If the result is zero, the amount to zero is less than 0x4 bytes */
subu a1, a1, a3
beqz a3, bytezero
/* zero one word at a time */
addu a3, a3, a0
1:
addiu $a0, $a0, 4
bne $a0, $a3, 1b
sw $zero, -4($a0)
addu a0, a0, 4
sw zero, -4(a0)
bne a0, a3, 1b
bytezero:
// test if nothing left to zero
blez $a1, zerodone
nop
// zero one byte at a time
addu $a1, $a1, $a0
/* test if nothing left to zero */
blez a1, zerodone
/* zero one byte at a time */
addu a1, a1, a0
1:
addiu $a0, $a0, 1
bne $a0, $a1, 1b
sb $zero, -1($a0)
sb zero, (a0)
addiu a0, a0, 1
bne a0, a1, 1b
zerodone:
jr $ra
nop
jr ra
END(bzero)
src/libultra/mgu/mtxf2l.s
+27 -31
View File
@@ -1,40 +1,36 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
.set noat
.set noreorder
.section .text
.balign 32
.text
.align 5
#define MTX_INTPART 0
#define MTX_FRACPART 0x20
LEAF(guMtxF2L)
li $at, 0x47800000 // 65536.0f
mtc1 $at, $f0
li $t9, 0xFFFF0000
addiu $t8, $a1, MTX_FRACPART
li.s fv0, 65536.0
li t9, 0xFFFF0000
addu t8, a1, MTX_FRACPART
1:
lwc1 $f4, ($a0)
lwc1 $f10, 4($a0)
addiu $a1, $a1, 4
mul.s $f6, $f4, $f0
addiu $a0, $a0, 8
mul.s $f16, $f10, $f0
trunc.w.s $f8, $f6
trunc.w.s $f18, $f16
mfc1 $t0, $f8
mfc1 $t1, $f18
and $t2, $t0, $t9
sll $t5, $t0, 0x10
srl $t3, $t1, 0x10
andi $t6, $t1, 0xFFFF
or $t4, $t2, $t3
or $t7, $t5, $t6
sw $t4, (MTX_INTPART-4)($a1)
bne $a1, $t8, 1b
sw $t7, (MTX_FRACPART-4)($a1)
jr $ra
nop
lwc1 ft0, 0(a0)
lwc1 ft3, 4(a0)
mul.s ft1, ft0, fv0
mul.s ft4, ft3, fv0
trunc.w.s ft2, ft1
trunc.w.s ft5, ft4
mfc1 t0, ft2
mfc1 t1, ft5
srl t3, t1, 0x10
and t6, t1, 0xFFFF
and t2, t0, t9
sll t5, t0, 0x10
or t4, t2, t3
or t7, t5, t6
sw t4, (MTX_INTPART)(a1)
sw t7, (MTX_FRACPART)(a1)
addu a0, a0, 8
addu a1, a1, 4
bne a1, t8, 1b
jr ra
END(guMtxF2L)
src/libultra/mgu/mtxident.s
+22 -24
View File
@@ -1,29 +1,27 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
.set noreorder
.section .text
.balign 32
.text
.align 5
LEAF(guMtxIdent)
addi $t0, $zero, 1
sll $t1, $t0, 0x10
sw $t1, ($a0)
sw $zero, 4($a0)
sw $t0, 8($a0)
sw $zero, 0xc($a0)
sw $zero, 0x10($a0)
sw $t1, 0x14($a0)
sw $zero, 0x18($a0)
sw $t0, 0x1C($a0)
sw $zero, 0x20($a0)
sw $zero, 0x24($a0)
sw $zero, 0x28($a0)
sw $zero, 0x2c($a0)
sw $zero, 0x30($a0)
sw $zero, 0x34($a0)
sw $zero, 0x38($a0)
jr $ra
sw $zero, 0x3C($a0)
add t0, zero, 1
sll t1, t0, 0x10
sw t1, 0x00(a0)
sw zero, 0x04(a0)
sw t0, 0x08(a0)
sw zero, 0x0C(a0)
sw zero, 0x10(a0)
sw t1, 0x14(a0)
sw zero, 0x18(a0)
sw t0, 0x1C(a0)
sw zero, 0x20(a0)
sw zero, 0x24(a0)
sw zero, 0x28(a0)
sw zero, 0x2c(a0)
sw zero, 0x30(a0)
sw zero, 0x34(a0)
sw zero, 0x38(a0)
sw zero, 0x3C(a0)
jr ra
END(guMtxIdent)
src/libultra/mgu/mtxidentf.s
+21 -23
View File
@@ -1,28 +1,26 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
.set noreorder
.section .text
.balign 32
.text
.align 5
LEAF(guMtxIdentF)
li $t0, 0x3F800000 // 1.0f
sw $t0, ($a0)
sw $zero, 4($a0)
sw $zero, 8($a0)
sw $zero, 0xC($a0)
sw $zero, 0x10($a0)
sw $t0, 0x14($a0)
sw $zero, 0x18($a0)
sw $zero, 0x1C($a0)
sw $zero, 0x20($a0)
sw $zero, 0x24($a0)
sw $t0, 0x28($a0)
sw $zero, 0x2C($a0)
sw $zero, 0x30($a0)
sw $zero, 0x34($a0)
sw $zero, 0x38($a0)
jr $ra
sw $t0, 0x3C($a0)
li.s t0, 1.0
sw t0, 0x00(a0)
sw zero, 0x04(a0)
sw zero, 0x08(a0)
sw zero, 0x0C(a0)
sw zero, 0x10(a0)
sw t0, 0x14(a0)
sw zero, 0x18(a0)
sw zero, 0x1C(a0)
sw zero, 0x20(a0)
sw zero, 0x24(a0)
sw t0, 0x28(a0)
sw zero, 0x2C(a0)
sw zero, 0x30(a0)
sw zero, 0x34(a0)
sw zero, 0x38(a0)
sw t0, 0x3C(a0)
jr ra
END(guMtxIdentF)
src/libultra/mgu/mtxl2f.s
+34 -32
View File
@@ -1,41 +1,43 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
.set noat
.set noreorder
.section .text
.balign 32
.text
.align 5
#define MTX_INTPART 0
#define MTX_FRACPART 0x20
#define FIXTOF 0.0000152587890625 /* 1.0f / 65536.0f */
LEAF(guMtxL2F)
li $at, 0x37800000 // 1.0f / 65536.0f
mtc1 $at, $f0
li $t9, 0xFFFF0000
addiu $t8, $a1, MTX_FRACPART
li.s fv0, FIXTOF
li t9, 0xFFFF0000
addiu t8, a1, MTX_FRACPART
1:
lw $t0, MTX_INTPART($a1)
lw $t1, MTX_FRACPART($a1)
addiu $a1, $a1, 4
and $t2, $t0, $t9
srl $t3, $t1, 0x10
or $t4, $t2, $t3
mtc1 $t4, $f4
sll $t5, $t0, 0x10
andi $t6, $t1, 0xFFFF
or $t7, $t5, $t6
cvt.s.w $f6, $f4
mtc1 $t7, $f10
addiu $a0, $a0, 8
cvt.s.w $f16, $f10
mul.s $f8, $f6, $f0
nop
mul.s $f18, $f16, $f0
swc1 $f8, -8($a0)
bne $a1, $t8, 1b
swc1 $f18, -4($a0)
jr $ra
nop
lw t0, MTX_INTPART(a1)
lw t1, MTX_FRACPART(a1)
and t2, t0, t9
srl t3, t1, 0x10
or t4, t2, t3
sll t5, t0, 0x10
and t6, t1, 0xFFFF
or t7, t5, t6
mtc1 t4, ft0
cvt.s.w ft1, ft0
mul.s ft2, ft1, fv0
mtc1 t7, ft3
cvt.s.w ft4, ft3
mul.s ft5, ft4, fv0
swc1 ft2, 0(a0)
swc1 ft5, 4(a0)
addu a0, a0, 8
addu a1, a1, 4
bne a1, t8, 1b
jr ra
END(guMtxL2F)
src/libultra/mgu/normalize.s
+22 -26
View File
@@ -1,31 +1,27 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
.set noreorder
.section .text
.balign 32
.text
.align 5
LEAF(guNormalize)
lwc1 $f4, ($a0)
lwc1 $f6, ($a1)
lwc1 $f8, ($a2)
mul.s $f10, $f4, $f4
li $t0, 0x3F800000 // 1.0f
mul.s $f16, $f6, $f6
add.s $f18, $f10, $f16
mul.s $f16, $f8, $f8
add.s $f10, $f16, $f18
mtc1 $t0, $f18
sqrt.s $f16, $f10
div.s $f10, $f18, $f16
mul.s $f16, $f4, $f10
nop
mul.s $f18, $f6, $f10
nop
mul.s $f4, $f8, $f10
swc1 $f16, ($a0)
swc1 $f18, ($a1)
jr $ra
swc1 $f4, ($a2)
lwc1 ft0, (a0)
lwc1 ft1, (a1)
lwc1 ft2, (a2)
li.s t0, 1.0
mul.s ft3, ft0, ft0
mul.s ft4, ft1, ft1
add.s ft5, ft3, ft4
mul.s ft4, ft2, ft2
add.s ft3, ft4, ft5
mtc1 t0, ft5
sqrt.s ft4, ft3
div.s ft3, ft5, ft4
mul.s ft4, ft0, ft3
mul.s ft5, ft1, ft3
mul.s ft0, ft2, ft3
swc1 ft4, (a0)
swc1 ft5, (a1)
swc1 ft0, (a2)
jr ra
END(guNormalize)
src/libultra/mgu/scale.s
+51 -47
View File
@@ -1,52 +1,56 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
.set noat
.set noreorder
.section .text
.balign 32
.text
.align 5
LEAF(guScale)
li $at, 0x47800000 // 65536.0f
mtc1 $at, $f4
mtc1 $a1, $f6
sw $zero, 4($a0)
sw $zero, 0xC($a0)
mul.s $f8, $f6, $f4
mtc1 $a2, $f6
sw $zero, 0x10($a0)
sw $zero, 0x18($a0)
sw $zero, 0x24($a0)
sw $zero, 0x2C($a0)
sw $zero, 0x30($a0)
trunc.w.s $f10, $f8
mul.s $f8, $f6, $f4
mtc1 $a3, $f6
sw $zero, 0x38($a0)
mfc1 $t1, $f10
sw $zero, 0x3C($a0)
srl $t2, $t1, 0x10
trunc.w.s $f10, $f8
mul.s $f8, $f6, $f4
sll $t0, $t2, 0x10
sll $t2, $t1, 0x10
mfc1 $t1, $f10
sw $t0, ($a0)
sw $t2, 0x20($a0)
srl $t0, $t1, 0x10
trunc.w.s $f10, $f8
andi $t2, $t1, 0xFFFF
sw $t2, 0x28($a0)
sw $t0, 8($a0)
mfc1 $t1, $f10
nop
srl $t2, $t1, 0x10
sll $t0, $t2, 0x10
sw $t0, 0x14($a0)
li $t0, 1
sll $t2, $t1, 0x10
sw $t2, 0x34($a0)
jr $ra
sw $t0, 0x1C($a0)
li.s ft0, 65536.0
mtc1 a1, ft1
mul.s ft2, ft1, ft0
trunc.w.s ft3, ft2
mfc1 t1, ft3
srl t2, t1, 0x10
sll t0, t2, 0x10
sw t0, 0x00(a0)
sll t2, t1, 0x10
sw t2, 0x20(a0)
mtc1 a2, ft1
mul.s ft2, ft1, ft0
trunc.w.s ft3, ft2
mfc1 t1, ft3
srl t0, t1, 0x10
sw t0, 0x08(a0)
andi t2, t1, 0xFFFF
sw t2, 0x28(a0)
mtc1 a3, ft1
mul.s ft2, ft1, ft0
trunc.w.s ft3, ft2
mfc1 t1, ft3
srl t2, t1, 0x10
sll t0, t2, 0x10
sw t0, 0x14(a0)
sll t2, t1, 0x10
sw t2, 0x34(a0)
li t0, 1
sw t0, 0x1C(a0)
sw zero, 0x04(a0)
sw zero, 0x0C(a0)
sw zero, 0x10(a0)
sw zero, 0x18(a0)
sw zero, 0x24(a0)
sw zero, 0x2C(a0)
sw zero, 0x30(a0)
sw zero, 0x38(a0)
sw zero, 0x3C(a0)
jr ra
END(guScale)
src/libultra/mgu/translate.s
+63 -56
View File
@@ -1,61 +1,68 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
.set noat
.set noreorder
.section .text
.balign 32
.text
.align 5
LEAF(guTranslate)
li $at, 0x47800000 // 65536.0f
mtc1 $at, $f4
mtc1 $a1, $f6
sw $zero, ($a0)
sw $zero, 0x14($a0)
mul.s $f8, $f6, $f4
mtc1 $a2, $f6
sw $zero, 8($a0)
sw $zero, 4($a0)
sw $zero, 0xC($a0)
sw $zero, 0x10($a0)
sw $zero, 0x20($a0)
trunc.w.s $f10, $f8
mul.s $f8, $f6, $f4
mtc1 $a3, $f6
sw $zero, 0x24($a0)
mfc1 $t1, $f10
sw $zero, 0x28($a0)
sw $zero, 0x2C($a0)
srl $t2, $t1, 0x10
trunc.w.s $f10, $f8
mul.s $f8, $f6, $f4
sll $t0, $t2, 0x10
sw $zero, 0x30($a0)
mfc1 $t3, $f10
sw $zero, 0x34($a0)
srl $t2, $t3, 0x10
trunc.w.s $f10, $f8
or $t0, $t0, $t2
sw $t0, 0x18($a0)
sll $t0, $t1, 0x10
sll $t2, $t3, 0x10
mfc1 $t1, $f10
srl $t2, $t2, 0x10
or $t0, $t0, $t2
sw $t0, 0x38($a0)
srl $t2, $t1, 0x10
sll $t0, $t2, 0x10
addiu $t0, $t0, 1
sw $t0, 0x1C($a0)
lui $t0, 1
ori $t0, $t0, 0
sw $t0, ($a0)
sw $t0, 0x14($a0)
lui $t0, (0x00000001 >> 16)
ori $t0, (0x00000001 & 0xFFFF)
sll $t2, $t1, 0x10
sw $t2, 0x3C($a0)
jr $ra
sw $t0, 8($a0)
li.s ft0, 65536.0
mtc1 a1, ft1
mul.s ft2, ft1, ft0
trunc.w.s ft3, ft2
mfc1 t1, ft3
mtc1 a2, ft1
mul.s ft2, ft1, ft0
trunc.w.s ft3, ft2
mfc1 t3, ft3
srl t2, t1, 0x10
sll t0, t2, 0x10
srl t2, t3, 0x10
or t0, t0, t2
sw t0, 0x18(a0)
sll t2, t3, 0x10
sll t0, t1, 0x10
srl t2, t2, 0x10
or t0, t0, t2
sw t0, 0x38(a0)
mtc1 a3, ft1
mul.s ft2, ft1, ft0
trunc.w.s ft3, ft2
mfc1 t1, ft3
srl t2, t1, 0x10
sll t0, t2, 0x10
addiu t0, t0, 1
sw t0, 0x1C(a0)
sll t2, t1, 0x10
sw t2, 0x3C(a0)
sw zero, 0x00(a0)
sw zero, 0x04(a0)
sw zero, 0x08(a0)
sw zero, 0x0C(a0)
sw zero, 0x10(a0)
sw zero, 0x14(a0)
sw zero, 0x20(a0)
sw zero, 0x24(a0)
sw zero, 0x28(a0)
sw zero, 0x2C(a0)
sw zero, 0x30(a0)
sw zero, 0x34(a0)
lui t0, (0x00010000 >> 16)
ori t0, (0x00010000 & 0xFFFF)
sw t0, (a0)
sw t0, 0x14(a0)
lui t0, (0x00000001 >> 16)
ori t0, (0x00000001 & 0xFFFF)
sw t0, 8(a0)
jr ra
END(guTranslate)
src/libultra/os/exceptasm.s
+593 -699
View File
File diff suppressed because it is too large Load Diff
src/libultra/os/getcause.s
+4 -8
View File
@@ -1,14 +1,10 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#include "ultra64/R4300.h"
.set noreorder
.section .text
.balign 16
.text
LEAF(__osGetCause)
mfc0 $v0, C0_CAUSE
jr $ra
nop
MFC0( v0, C0_CAUSE)
jr ra
END(__osGetCause)
src/libultra/os/getcount.s
+4 -8
View File
@@ -1,14 +1,10 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#include "ultra64/R4300.h"
.set noreorder
.section .text
.balign 16
.text
LEAF(osGetCount)
mfc0 $v0, C0_COUNT
jr $ra
nop
MFC0( v0, C0_COUNT)
jr ra
END(osGetCount)
src/libultra/os/getfpccsr.s
+4 -8
View File
@@ -1,14 +1,10 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#include "ultra64/R4300.h"
.set noreorder
.section .text
.balign 16
.text
LEAF(__osGetFpcCsr)
cfc1 $v0, C1_FPCSR
jr $ra
nop
CFC1( v0, C1_FPCSR)
jr ra
END(__osGetFpcCsr)
src/libultra/os/getintmask.s
+31 -38
View File
@@ -1,14 +1,10 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#include "ultra64/R4300.h"
#include "ultra64/rcp.h"
#include "ultra64/exception.h"
.set noat
.set noreorder
.section .text
.balign 16
.text
/**
* OSIntMask osGetIntMask(void);
@@ -23,38 +19,35 @@
* See the comment for osSetIntMask for more details.
*/
LEAF(osGetIntMask)
// Extract interrupt enable bits from current SR
mfc0 $v0, C0_SR
andi $v0, $v0, (SR_IMASK | SR_IE)
// Get value of __OSGlobalIntMask
lui $t0, %hi(__OSGlobalIntMask)
addiu $t0, %lo(__OSGlobalIntMask)
lw $t1, 0($t0)
// Bitwise-OR in the disabled CPU bits of __OSGlobalIntMask
li $at, ~0
xor $t0, $t1, $at
andi $t0, $t0, SR_IMASK
or $v0, $v0, $t0
// Fetch MI_INTR_MASK_REG
lui $t1, %hi(PHYS_TO_K1(MI_INTR_MASK_REG))
lw $t1, %lo(PHYS_TO_K1(MI_INTR_MASK_REG))($t1)
// If there are RCP interrupts masked
beqz $t1, 1f
// Get value of __OSGlobalIntMask
lui $t0, %hi(__OSGlobalIntMask)
addiu $t0, %lo(__OSGlobalIntMask)
lw $t0, 0($t0)
// Bitwise-OR in the disabled RCP bits of __OSGlobalIntMask
srl $t0, $t0, RCP_IMASKSHIFT
li $at, ~0
xor $t0, $t0, $at
andi $t0, $t0, (RCP_IMASK >> RCP_IMASKSHIFT)
or $t1, $t1, $t0
.set noreorder
/* Extract interrupt enable bits from current SR */
mfc0 v0, C0_SR
andi v0, v0, (SR_IMASK | SR_IE)
/* Get value of __OSGlobalIntMask */
la t0, __OSGlobalIntMask
lw t1, (t0)
/* Bitwise-OR in the disabled CPU bits of __OSGlobalIntMask */
xor t0, t1, ~0
andi t0, t0, SR_IMASK
or v0, v0, t0
/* Fetch MI_INTR_MASK_REG */
lw t1, PHYS_TO_K1(MI_INTR_MASK_REG)
/* If there are RCP interrupts masked */
beqz t1, 1f
/* Get value of __OSGlobalIntMask */
la t0, __OSGlobalIntMask /* Note: macro expansion in delay slot */
lw t0, (t0)
/* Bitwise-OR in the disabled RCP bits of __OSGlobalIntMask */
srl t0, t0, RCP_IMASKSHIFT
xor t0, t0, ~0
andi t0, t0, (RCP_IMASK >> RCP_IMASKSHIFT)
or t1, t1, t0
1:
// Shift the RCP bits to not conflict with the CPU bits
sll $t2, $t1, RCP_IMASKSHIFT
// OR the CPU and RCP bits together
or $v0, $v0, $t2
jr $ra
/* Shift the RCP bits to not conflict with the CPU bits */
sll t2, t1, RCP_IMASKSHIFT
/* OR the CPU and RCP bits together */
or v0, v0, t2
jr ra
nop
.set reorder
END(osGetIntMask)
src/libultra/os/getsr.s
+4 -8
View File
@@ -1,14 +1,10 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#include "ultra64/R4300.h"
.set noreorder
.section .text
.balign 16
.text
LEAF(__osGetSR)
mfc0 $v0, C0_SR
jr $ra
nop
MFC0( v0, C0_SR)
jr ra
END(__osGetSR)
src/libultra/os/interrupt.s
+34 -42
View File
@@ -1,53 +1,45 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#include "ultra64/R4300.h"
#include "ultra64/thread.h"
.set noat
.set noreorder
.section .text
.balign 16
.text
LEAF(__osDisableInt)
lui $t2, %hi(__OSGlobalIntMask)
addiu $t2, $t2, %lo(__OSGlobalIntMask)
lw $t3, ($t2)
andi $t3, $t3, SR_IMASK
mfc0 $t0, C0_SR
li $at, ~SR_IE
and $t1, $t0, $at
mtc0 $t1, C0_SR
andi $v0, $t0, SR_IE
lw $t0, ($t2)
andi $t0, $t0, SR_IMASK
beq $t0, $t3, No_Change_Global_Int
lui $t2, %hi(__osRunningThread)
//! @bug this addiu should be lw, it may never come up in practice as to reach this code
//! the CPU bits of __OSGlobalIntMask must have changed while this function is running.
addiu $t2, $t2, %lo(__osRunningThread)
lw $t1, THREAD_SR($t2)
andi $t2, $t1, SR_IMASK
and $t2, $t2, $t0
li $at, ~SR_IMASK
and $t1, $t1, $at
or $t1, $t1, $t2
li $at, ~SR_IE
and $t1, $t1, $at
mtc0 $t1, C0_SR
nop
nop
la t2, __OSGlobalIntMask
lw t3, (t2)
and t3, t3, SR_IMASK
MFC0( t0, C0_SR)
and t1, t0, ~SR_IE
MTC0( t1, C0_SR)
and v0, t0, SR_IE
lw t0, (t2)
and t0, t0, SR_IMASK
.set noreorder
beq t0, t3, No_Change_Global_Int
/*! @bug this la should be lw, it may never come up in practice as to reach this code
*! the CPU bits of __OSGlobalIntMask must have changed while this function is running.
*/
la t2, __osRunningThread
lw t1, THREAD_SR(t2)
and t2, t1, SR_IMASK
and t2, t2, t0
.set reorder
and t1, t1, ~SR_IMASK
or t1, t1, t2
and t1, t1, ~SR_IE
MTC0( t1, C0_SR)
NOP
NOP
No_Change_Global_Int:
jr $ra
nop
jr ra
END(__osDisableInt)
LEAF(__osRestoreInt)
mfc0 $t0, C0_SR
or $t0, $t0, $a0
mtc0 $t0, C0_SR
nop
nop
jr $ra
nop
MFC0( t0, C0_SR)
or t0, t0, a0
MTC0( t0, C0_SR)
NOP
NOP
jr ra
END(__osRestoreInt)
src/libultra/os/invaldcache.s
+51 -59
View File
@@ -1,12 +1,8 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#include "ultra64/R4300.h"
.set noat
.set noreorder
.section .text
.balign 16
.text
/**
* void osInvalDCache(void* vaddr, s32 nbytes);
@@ -24,71 +20,67 @@
* the entire data cache is invalidated.
*/
LEAF(osInvalDCache)
// If the amount to invalidate is less than or equal to 0, return immediately
blez $a1, 3f
.set noreorder
/* If the amount to invalidate is less than or equal to 0, return immediately */
blez a1, 3f
nop
// If the amount to invalidate is as large as or larger than
// the data cache size, invalidate all
li $t3, DCACHE_SIZE
sltu $at, $a1, $t3
beqz $at, 4f
/* If the amount to invalidate is as large as or larger than
* the data cache size, invalidate all */
li t3, DCACHE_SIZE
bgeu a1, t3, 4f
nop
// Ensure end address doesn't wrap around and end up smaller
// than the start address
move $t0, $a0
addu $t1, $a0, $a1
sltu $at, $t0, $t1
beqz $at, 3f
/* Ensure end address doesn't wrap around and end up smaller
* than the start address */
move t0, a0
addu t1, a0, a1
bgeu t0, t1, 3f
nop
// Mask start with cache line
andi $t2, $t0, DCACHE_LINEMASK
// If mask is not zero, the start is not cache aligned
beqz $t2, 1f
addiu $t1, $t1, -DCACHE_LINESIZE
// Subtract mask result to align to cache line
subu $t0, $t0, $t2
// Hit-Writeback-Invalidate unaligned part
cache (CACH_PD | C_HWBINV), ($t0)
sltu $at, $t0, $t1
// If that's all there is to do, return early
beqz $at, 3f
/* Mask start with cache line */
andi t2, t0, DCACHE_LINEMASK
/* If mask is not zero, the start is not cache aligned */
beqz t2, 1f
addiu t1, t1, -DCACHE_LINESIZE
/* Subtract mask result to align to cache line */
subu t0, t0, t2
/* Hit-Writeback-Invalidate unaligned part */
cache (CACH_PD | C_HWBINV), (t0)
/* If that's all there is to do, return early */
bgeu t0, t1, 3f
nop
addiu $t0, $t0, DCACHE_LINESIZE
addiu t0, t0, DCACHE_LINESIZE
1:
// Mask end with cache line
andi $t2, $t1, DCACHE_LINEMASK
// If mask is not zero, the end is not cache aligned
beqz $t2, 1f
/* Mask end with cache line */
andi t2, t1, DCACHE_LINEMASK
/* If mask is not zero, the end is not cache aligned */
beqz t2, 1f
nop
// Subtract mask result to align to cache line
subu $t1, $t1, $t2
// Hit-Writeback-Invalidate unaligned part
cache (CACH_PD | C_HWBINV), DCACHE_LINESIZE($t1)
sltu $at, $t1, $t0
// If that's all there is to do, return early
bnez $at, 3f
/* Subtract mask result to align to cache line */
subu t1, t1, t2
/* Hit-Writeback-Invalidate unaligned part */
cache (CACH_PD | C_HWBINV), DCACHE_LINESIZE(t1)
/* If that's all there is to do, return early */
bltu t1, t0, 3f
nop
// Invalidate the rest
/* Invalidate the rest */
1:
// Hit-Invalidate
cache (CACH_PD | C_HINV), ($t0)
sltu $at, $t0, $t1
bnez $at, 1b
addiu $t0, $t0, DCACHE_LINESIZE
/* Hit-Invalidate */
cache (CACH_PD | C_HINV), (t0)
bltu t0, t1, 1b
addiu t0, t0, DCACHE_LINESIZE
3:
jr $ra
jr ra
nop
4:
li $t0, K0BASE
addu $t1, $t0, $t3
addiu $t1, $t1, -DCACHE_LINESIZE
li t0, K0BASE
addu t1, t0, t3
addiu t1, t1, -DCACHE_LINESIZE
5:
// Index-Writeback-Invalidate
cache (CACH_PD | C_IWBINV), ($t0)
sltu $at, $t0, $t1
bnez $at, 5b
addiu $t0, DCACHE_LINESIZE
jr $ra
/* Index-Writeback-Invalidate */
cache (CACH_PD | C_IWBINV), (t0)
bltu t0, t1, 5b
addiu t0, DCACHE_LINESIZE
jr ra
nop
.set reorder
END(osInvalDCache)
src/libultra/os/invalicache.s
+30 -36
View File
@@ -1,52 +1,46 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#include "ultra64/R4300.h"
.set noat
.set noreorder
.section .text
.balign 16
.text
LEAF(osInvalICache)
// If the amount to invalidate is less than or equal to 0, return immediately
blez $a1, 2f
.set noreorder
/* If the amount to invalidate is less than or equal to 0, return immediately */
blez a1, 2f
nop
// If the amount to invalidate is as large as or larger than
// the instruction cache size, invalidate all
li $t3, ICACHE_SIZE
sltu $at, $a1, $t3
beqz $at, 3f
/* If the amount to invalidate is as large as or larger than */
/* the instruction cache size, invalidate all */
li t3, ICACHE_SIZE
bgeu a1, t3, 3f
nop
// ensure end address doesn't wrap around and end up smaller
// than the start address
move $t0, $a0
addu $t1, $a0, $a1
sltu $at, $t0, $t1
beqz $at, 2f
/* ensure end address doesn't wrap around and end up smaller */
/* than the start address */
move t0, a0
addu t1, a0, a1
bgeu t0, t1, 2f
nop
// Mask and subtract to align to cache line
andi $t2, $t0, ICACHE_LINEMASK
addiu $t1, $t1, -ICACHE_LINESIZE
subu $t0, $t0, $t2
/* Mask and subtract to align to cache line */
andi t2, t0, ICACHE_LINEMASK
addiu t1, t1, -ICACHE_LINESIZE
subu t0, t0, t2
1:
cache (CACH_PI | C_HINV), ($t0)
sltu $at, $t0, $t1
bnez $at, 1b
addiu $t0, $t0, ICACHE_LINESIZE
cache (CACH_PI | C_HINV), (t0)
bltu t0, t1, 1b
addiu t0, t0, ICACHE_LINESIZE
2:
jr $ra
jr ra
nop
3:
li $t0, K0BASE
addu $t1, $t0, $t3
addiu $t1, $t1, -ICACHE_LINESIZE
li t0, K0BASE
addu t1, t0, t3
addiu t1, t1, -ICACHE_LINESIZE
4:
cache (CACH_PI | C_IINV), ($t0)
sltu $at, $t0, $t1
bnez $at, 4b
addiu $t0, ICACHE_LINESIZE
jr $ra
cache (CACH_PI | C_IINV), (t0)
bltu t0, t1, 4b
addiu t0, ICACHE_LINESIZE
jr ra
nop
.set reorder
END(osInvalICache)
src/libultra/os/maptlbrdb.s
+24 -29
View File
@@ -1,36 +1,31 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#include "ultra64/R4300.h"
#include "ultra64/rdb.h"
.set noreorder
.section .text
.balign 16
.text
LEAF(osMapTLBRdb)
mfc0 $t0, C0_ENTRYHI
li $t1, NTLBENTRIES
mtc0 $t1, C0_INX
mtc0 $zero, C0_PAGEMASK
li $t2, (TLBLO_UNCACHED | TLBLO_D | TLBLO_V | TLBLO_G)
li $t1, (RDB_BASE_REG & TLBHI_VPN2MASK)
mtc0 $t1, C0_ENTRYHI
// Possible bug? Virtual address instead of physical address
// set as page frame number
li $t1, RDB_BASE_VIRTUAL_ADDR
srl $t3, $t1, TLBLO_PFNSHIFT
or $t3, $t3, $t2
mtc0 $t3, C0_ENTRYLO0
li $t1, TLBLO_G
mtc0 $t1, C0_ENTRYLO1
nop
tlbwi
nop
nop
nop
nop
mtc0 $t0, C0_ENTRYHI
jr $ra
nop
MFC0( t0, C0_ENTRYHI)
li t1, NTLBENTRIES
MTC0( t1, C0_INX)
MTC0( zero, C0_PAGEMASK)
li t2, (TLBLO_UNCACHED | TLBLO_D | TLBLO_V | TLBLO_G)
li t1, (RDB_BASE_REG & TLBHI_VPN2MASK)
MTC0( t1, C0_ENTRYHI)
/* Possible bug? Virtual address instead of physical address set as page frame number */
li t1, RDB_BASE_VIRTUAL_ADDR
srl t3, t1, TLBLO_PFNSHIFT
or t3, t3, t2
MTC0( t3, C0_ENTRYLO0)
li t1, TLBLO_G
MTC0( t1, C0_ENTRYLO1)
NOP
TLBWI
NOP
NOP
NOP
NOP
MTC0( t0, C0_ENTRYHI)
jr ra
END(osMapTLBRdb)
src/libultra/os/parameters.s
+24 -17
View File
@@ -1,22 +1,29 @@
#include "ultra64/asm.h"
.section .text
.text
.macro IPL_SYMBOL name, address, size
.global \name
.set \name, \address
.type \name, @object
.size \name, \size
.endm
#ifdef __sgi
#define IPL_SYMBOL(name, address, size) \
ABS(name, address)
#else
#define IPL_SYMBOL(name, address, sz) \
ABS(name, address) ;\
.type name, @object ;\
.size name, sz
#endif
IPL_SYMBOL leoBootID, 0x800001A0, 4
IPL_SYMBOL osTvType, 0x80000300, 4
IPL_SYMBOL osRomType, 0x80000304, 4
IPL_SYMBOL osRomBase, 0x80000308, 4
IPL_SYMBOL osResetType, 0x8000030C, 4
IPL_SYMBOL osCicId, 0x80000310, 4
IPL_SYMBOL osVersion, 0x80000314, 4
IPL_SYMBOL osMemSize, 0x80000318, 4
IPL_SYMBOL osAppNMIBuffer, 0x8000031C, 0x40
.fill 0x60
IPL_SYMBOL(leoBootID, 0x800001A0, 4)
IPL_SYMBOL(osTvType, 0x80000300, 4)
IPL_SYMBOL(osRomType, 0x80000304, 4)
IPL_SYMBOL(osRomBase, 0x80000308, 4)
IPL_SYMBOL(osResetType, 0x8000030C, 4)
IPL_SYMBOL(osCicId, 0x80000310, 4)
IPL_SYMBOL(osVersion, 0x80000314, 4)
IPL_SYMBOL(osMemSize, 0x80000318, 4)
IPL_SYMBOL(osAppNMIBuffer, 0x8000031C, 0x40)
.repeat 0x60/4
NOP
.endr
src/libultra/os/probetlb.s
+48 -51
View File
@@ -1,12 +1,8 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#include "ultra64/R4300.h"
.set noat
.set noreorder
.section .text
.balign 16
.text
/**
* u32 __osProbeTLB(void* vaddr);
@@ -17,70 +13,71 @@
* Returns the physical address if found, or -1 if not found.
*/
LEAF(__osProbeTLB)
// Set C0_ENTRYHI based on supplied vaddr
mfc0 $t0, C0_ENTRYHI
andi $t1, $t0, TLBHI_PIDMASK
li $at, TLBHI_VPN2MASK
and $t2, $a0, $at
or $t1, $t1, $t2
mtc0 $t1, C0_ENTRYHI
.set noreorder
/* Set C0_ENTRYHI based on supplied vaddr */
mfc0 t0, C0_ENTRYHI
and t1, t0, TLBHI_PIDMASK
and t2, a0, TLBHI_VPN2MASK
or t1, t1, t2
mtc0 t1, C0_ENTRYHI
nop
nop
nop
// TLB probe, sets C0_INX to a value matching C0_ENTRYHI.
// If no match is found the TLBINX_PROBE bit is set to indicate this.
/* TLB probe, sets C0_INX to a value matching C0_ENTRYHI. */
/* If no match is found the TLBINX_PROBE bit is set to indicate this. */
tlbp
nop
nop
// Read result
mfc0 $t3, C0_INX
li $at, TLBINX_PROBE
and $t3, $t3, $at
// Branch if no match was found
bnez $t3, 3f
/* Read result */
mfc0 t3, C0_INX
and t3, t3, TLBINX_PROBE
/* Branch if no match was found */
bnez t3, 3f
nop
// Read TLB, sets C0_ENTRYHI, C0_ENTRYLO0, C0_ENTRYLO1 and C0_PAGEMASK for the TLB
// entry indicated by C0_INX
/* Read TLB, sets C0_ENTRYHI, C0_ENTRYLO0, C0_ENTRYLO1 and C0_PAGEMASK for the TLB */
/* entry indicated by C0_INX */
tlbr
nop
nop
nop
// Calculate page size = (page mask + 0x2000) >> 1
mfc0 $t3, C0_PAGEMASK
addi $t3, $t3, 0x2000
srl $t3, $t3, 1
// & with vaddr
and $t4, $t3, $a0
// Select C0_ENTRYLO0 or C0_ENTRYLO1
bnez $t4, 1f
addi $t3, $t3, -1 // make bitmask out of page size
mfc0 $v0, C0_ENTRYLO0
/* Calculate page size = (page mask + 0x2000) >> 1 */
mfc0 t3, C0_PAGEMASK
add t3, t3, 0x2000
srl t3, t3, 1
/* & with vaddr */
and t4, t3, a0
/* Select C0_ENTRYLO0 or C0_ENTRYLO1 */
bnez t4, 1f
add t3, t3, -1 /* make bitmask out of page size */
mfc0 v0, C0_ENTRYLO0
b 2f
nop
1:
mfc0 $v0, C0_ENTRYLO1
mfc0 v0, C0_ENTRYLO1
2:
// Check valid bit and branch if not valid
andi $t5, $v0, TLBLO_V
beqz $t5, 3f
/* Check valid bit and branch if not valid */
and t5, v0, TLBLO_V
beqz t5, 3f
nop
// Extract the Page Frame Number from the entry
li $at, TLBLO_PFNMASK
and $v0, $v0, $at
sll $v0, $v0, TLBLO_PFNSHIFT
// Mask vaddr with page size mask
and $t5, $a0, $t3
// Add masked vaddr to pfn to obtain the physical address
add $v0, $v0, $t5
/* Extract the Page Frame Number from the entry */
and v0, v0, TLBLO_PFNMASK
sll v0, v0, TLBLO_PFNSHIFT
/* Mask vaddr with page size mask */
and t5, a0, t3
/* Add masked vaddr to pfn to obtain the physical address */
add v0, v0, t5
b 4f
nop
3:
// No physical address for the supplied virtual address was found,
// return -1
li $v0, -1
/* No physical address for the supplied virtual address was found, */
/* return -1 */
li v0, -1
4:
// Restore original C0_ENTRYHI value before returning
mtc0 $t0, C0_ENTRYHI
jr $ra
/* Restore original C0_ENTRYHI value before returning */
mtc0 t0, C0_ENTRYHI
jr ra
nop
.set reorder
END(__osProbeTLB)
src/libultra/os/setcompare.s
+4 -8
View File
@@ -1,14 +1,10 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#include "ultra64/R4300.h"
.set noreorder
.section .text
.balign 16
.text
LEAF(__osSetCompare)
mtc0 $a0, C0_COMPARE
jr $ra
nop
MTC0( a0, C0_COMPARE)
jr ra
END(__osSetCompare)
src/libultra/os/setfpccsr.s
+5 -9
View File
@@ -1,15 +1,11 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#include "ultra64/R4300.h"
.set noreorder
.section .text
.balign 16
.text
LEAF(__osSetFpcCsr)
cfc1 $v0, C1_FPCSR
ctc1 $a0, C1_FPCSR
jr $ra
nop
CFC1( v0, C1_FPCSR)
CTC1( a0, C1_FPCSR)
jr ra
END(__osSetFpcCsr)
src/libultra/os/setintmask.s
+46 -60
View File
@@ -1,14 +1,11 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#include "ultra64/R4300.h"
#include "ultra64/rcp.h"
#include "ultra64/exception.h"
.set noat
.set noreorder
.section .rodata
.balign 16
.rdata
.align 2
/**
* LUT to convert between an interrupt mask value and a value for MI_INTR_MASK_REG.
@@ -82,9 +79,7 @@ DATA(__osRcpImTable)
.half MI_INTR_MASK_SET_SP | MI_INTR_MASK_SET_SI | MI_INTR_MASK_SET_AI | MI_INTR_MASK_SET_VI | MI_INTR_MASK_SET_PI | MI_INTR_MASK_SET_DP
ENDDATA(__osRcpImTable)
.section .text
.balign 16
.text
/**
* OSIntMask osSetIntMask(OSIntMask);
@@ -113,56 +108,47 @@ ENDDATA(__osRcpImTable)
* OS_IM_ALL, so the operation is usually simply (SR | 0).
*/
LEAF(osSetIntMask)
// Extract interrupt enable bits from current SR
mfc0 $t4, C0_SR
andi $v0, $t4, (SR_IMASK | SR_IE)
// Get value of __OSGlobalIntMask
lui $t0, %hi(__OSGlobalIntMask)
addiu $t0, %lo(__OSGlobalIntMask)
lw $t3, ($t0)
// Bitwise-OR in the disabled CPU bits of __OSGlobalIntMask
li $at, ~0
xor $t0, $t3, $at
andi $t0, $t0, SR_IMASK
or $v0, $v0, $t0
// Fetch MI_INTR_MASK_REG
lui $t2, %hi(PHYS_TO_K1(MI_INTR_MASK_REG))
lw $t2, %lo(PHYS_TO_K1(MI_INTR_MASK_REG))($t2)
// If there are RCP interrupts masked
beqz $t2, 1f
srl $t1, $t3, RCP_IMASKSHIFT
// Bitwise-OR in the disabled RCP bits of __OSGlobalIntMask
li $at, ~0
xor $t1, $t1, $at
andi $t1, $t1, (RCP_IMASK >> RCP_IMASKSHIFT)
or $t2, $t2, $t1
/* Extract interrupt enable bits from current SR */
MFC0( t4, C0_SR)
.set noreorder
and v0, t4, (SR_IMASK | SR_IE)
/* Get value of __OSGlobalIntMask */
la t0, __OSGlobalIntMask
lw t3, (t0)
/* Bitwise-OR in the disabled CPU bits of __OSGlobalIntMask */
xor t0, t3, ~0
and t0, t0, SR_IMASK
or v0, v0, t0
/* Fetch MI_INTR_MASK_REG */
lw t2, PHYS_TO_K1(MI_INTR_MASK_REG)
/* If there are RCP interrupts masked */
beqz t2, 1f
srl t1, t3, RCP_IMASKSHIFT
/* Bitwise-OR in the disabled RCP bits of __OSGlobalIntMask */
xor t1, t1, ~0
and t1, t1, (RCP_IMASK >> RCP_IMASKSHIFT)
or t2, t2, t1
1:
// Shift the RCP bits to not conflict with the CPU bits
sll $t2, $t2, RCP_IMASKSHIFT
// OR the CPU and RCP bits together
or $v0, $v0, $t2
// Extract RCP interrupt enable bits from requested mask and mask with __OSGlobalIntMask
li $at, RCP_IMASK
and $t0, $a0, $at
and $t0, $t0, $t3
// Convert to a value for MI_INTR_MASK_REG and set it
srl $t0, $t0, (RCP_IMASKSHIFT-1)
lui $t2, %hi(__osRcpImTable)
addu $t2, $t2, $t0
lhu $t2, %lo(__osRcpImTable)($t2)
lui $at, %hi(PHYS_TO_K1(MI_INTR_MASK_REG))
sw $t2, %lo(PHYS_TO_K1(MI_INTR_MASK_REG))($at)
// Extract CPU interrupt enable bits from requested mask and mask with __OSGlobalIntMask
andi $t0, $a0, OS_IM_CPU
andi $t1, $t3, SR_IMASK
and $t0, $t0, $t1
li $at, ~SR_IMASK
and $t4, $t4, $at
// Bitwise OR in the remaining bits of SR and set new SR
or $t4, $t4, $t0
mtc0 $t4, C0_SR
nop
nop
jr $ra
nop
/* Shift the RCP bits to not conflict with the CPU bits */
sll t2, t2, RCP_IMASKSHIFT
/* OR the CPU and RCP bits together */
or v0, v0, t2
/* Extract RCP interrupt enable bits from requested mask and mask with __OSGlobalIntMask */
and t0, a0, RCP_IMASK
and t0, t0, t3
/* Convert to a value for MI_INTR_MASK_REG and set it */
srl t0, t0, (RCP_IMASKSHIFT-1)
lhu t2, __osRcpImTable(t0)
sw t2, PHYS_TO_K1(MI_INTR_MASK_REG)
/* Extract CPU interrupt enable bits from requested mask and mask with __OSGlobalIntMask */
and t0, a0, OS_IM_CPU
and t1, t3, SR_IMASK
and t0, t0, t1
and t4, t4, ~SR_IMASK
/* Bitwise OR in the remaining bits of SR and set new SR */
or t4, t4, t0
MTC0( t4, C0_SR)
NOP
NOP
jr ra
END(osSetIntMask)
src/libultra/os/setsr.s
+5 -9
View File
@@ -1,15 +1,11 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#include "ultra64/R4300.h"
.set noreorder
.section .text
.balign 16
.text
LEAF(__osSetSR)
mtc0 $a0, C0_SR
nop
jr $ra
nop
MTC0( a0, C0_SR)
NOP
jr ra
END(__osSetSR)
src/libultra/os/setwatchlo.s
+5 -9
View File
@@ -1,15 +1,11 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#include "ultra64/R4300.h"
.set noreorder
.section .text
.balign 16
.text
LEAF(__osSetWatchLo)
mtc0 $a0, C0_WATCHLO
nop
jr $ra
nop
MTC0( a0, C0_WATCHLO)
NOP
jr ra
END(__osSetWatchLo)
src/libultra/os/unmaptlball.s
+18 -22
View File
@@ -1,29 +1,25 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#include "ultra64/R4300.h"
.set noreorder
.section .text
.balign 16
.text
LEAF(osUnmapTLBAll)
mfc0 $t0, C0_ENTRYHI
li $t1, (NTLBENTRIES - 1)
li $t2, (K0BASE & TLBHI_VPN2MASK)
mtc0 $t2, C0_ENTRYHI
mtc0 $zero, C0_ENTRYLO0
mtc0 $zero, C0_ENTRYLO1
MFC0( t0, C0_ENTRYHI)
li t1, (NTLBENTRIES - 1)
li t2, (K0BASE & TLBHI_VPN2MASK)
MTC0( t2, C0_ENTRYHI)
MTC0( zero, C0_ENTRYLO0)
MTC0( zero, C0_ENTRYLO1)
1:
mtc0 $t1, C0_INX
nop
tlbwi
nop
nop
addi $t1, $t1, -1
bgez $t1, 1b
nop
mtc0 $t0, C0_ENTRYHI
jr $ra
nop
MTC0( t1, C0_INX)
NOP
TLBWI
NOP
NOP
addi t1, t1, -1
bgez t1, 1b
MTC0( t0, C0_ENTRYHI)
jr ra
END(osUnmapTLBAll)
src/libultra/os/writebackdcache.s
+39 -45
View File
@@ -1,12 +1,8 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#include "ultra64/R4300.h"
.set noat
.set noreorder
.section .text
.balign 16
.text
/**
* void osWritebackDCache(void* vaddr, s32 nbytes);
@@ -16,45 +12,43 @@
* written back.
*/
LEAF(osWritebackDCache)
// If the amount to write back is less than or equal to 0, return immediately
blez $a1, .ret
nop
// If the amount to write back is as large as or larger than
// the data cache size, write back all
li $t3, DCACHE_SIZE
sltu $at, $a1, $t3
beqz $at, .all
nop
// ensure end address doesn't wrap around and end up smaller
// than the start address
move $t0, $a0
addu $t1, $a0, $a1
sltu $at, $t0, $t1
beqz $at, .ret
nop
// Mask and subtract to align to cache line
andi $t2, $t0, DCACHE_LINEMASK
addiu $t1, $t1, -DCACHE_LINESIZE
subu $t0, $t0, $t2
1:
cache (CACH_PD | C_HWB), ($t0)
sltu $at, $t0, $t1
bnez $at, 1b
addiu $t0, $t0, DCACHE_LINESIZE
.ret:
jr $ra
nop
/* If the amount to write back is less than or equal to 0, return immediately */
blez a1, .ret
// same as osWritebackDCacheAll in operation
.all:
li $t0, K0BASE
addu $t1, $t0, $t3
addiu $t1, $t1, -DCACHE_LINESIZE
/* If the amount to write back is as large as or larger than */
/* the data cache size, write back all */
li t3, DCACHE_SIZE
bgeu a1, t3, .all
/* ensure end address doesn't wrap around and end up smaller */
/* than the start address */
move t0, a0
addu t1, a0, a1
bgeu t0, t1, .ret
/* Mask and subtract to align to cache line */
andi t2, t0, DCACHE_LINEMASK
addiu t1, t1, -DCACHE_LINESIZE
subu t0, t0, t2
1:
cache (CACH_PD | C_IWBINV), ($t0)
sltu $at, $t0, $t1
bnez $at, 1b
addiu $t0, DCACHE_LINESIZE
jr $ra
nop
.set noreorder
cache (CACH_PD | C_HWB), (t0)
bltu t0, t1, 1b
addiu t0, t0, DCACHE_LINESIZE
.set reorder
.ret:
jr ra
/* same as osWritebackDCacheAll in operation */
.all:
li t0, K0BASE
addu t1, t0, t3
addiu t1, t1, -DCACHE_LINESIZE
1:
.set noreorder
cache (CACH_PD | C_IWBINV), (t0)
bltu t0, t1, 1b
addiu t0, DCACHE_LINESIZE
.set reorder
jr ra
END(osWritebackDCache)
src/libultra/os/writebackdcacheall.s
+12 -16
View File
@@ -1,23 +1,19 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#include "ultra64/R4300.h"
.set noat
.set noreorder
.section .text
.balign 16
.text
LEAF(osWritebackDCacheAll)
li $t0, K0BASE
li $t2, DCACHE_SIZE
addu $t1, $t0, $t2
addiu $t1, $t1, -DCACHE_LINESIZE
li t0, K0BASE
li t2, DCACHE_SIZE
addu t1, t0, t2
addiu t1, t1, -DCACHE_LINESIZE
1:
cache (CACH_PD | C_IWBINV), ($t0)
sltu $at, $t0, $t1
bnez $at, 1b
addiu $t0, DCACHE_LINESIZE
jr $ra
nop
.set noreorder
cache (CACH_PD | C_IWBINV), (t0)
bltu t0, t1, 1b
addiu t0, DCACHE_LINESIZE
.set reorder
jr ra
END(osWritebackDCacheAll)
src/makerom/entry.s
+34 -24
View File
@@ -1,34 +1,44 @@
#include "ultra64/asm.h"
#include "ultra64/regdef.h"
#include "boot.h"
.set noreorder
.text
.section .text
.balign 16
#if defined(__sgi) && !defined(AVOID_UB)
/* IDO assembler workaround: The makerom tool in the N64 SDK was given the bss segment size as a const
* literal, and since this literal was < 0x10000 it was loaded in one instruction. We don't have access
* to the bss segment size until we link everything so we cannot do the same thing. Instead we must load
* only the lower 16 bits of the bss size for matching.
* When AVOID_UB is enabled, don't do this and instead load the full symbol value, otherwise not all of
* bss may be cleared. */
#define LOAD_BSS_SIZE(reg) li reg, %half(_bootSegmentBssSize)
#else
#define LOAD_BSS_SIZE(reg) la reg, _bootSegmentBssSize
#endif
LEAF(entrypoint)
// Clear boot segment .bss
la $t0, _bootSegmentBssStart
#ifndef AVOID_UB
// UB: li only loads the lower 16 bits of _bootSegmentBssSize when it may be larger than this,
// so not all of bss may be cleared if it is too large
li $t1, _bootSegmentBssSize
#else
la $t1, _bootSegmentBssSize
#endif
/* Clear boot segment .bss */
la t0, _bootSegmentBssStart
LOAD_BSS_SIZE(t1)
.clear_bss:
addi $t1, $t1, -8
sw $zero, ($t0)
sw $zero, 4($t0)
bnez $t1, .clear_bss
addi $t0, $t0, 8
// Set up stack and enter program code
lui $t2, %hi(bootproc)
lui $sp, %hi(sBootThreadStack + BOOT_STACK_SIZE)
addiu $t2, %lo(bootproc)
jr $t2
addiu $sp, %lo(sBootThreadStack + BOOT_STACK_SIZE)
sw zero, 0(t0)
sw zero, 4(t0)
addi t0, t0, 8
addi t1, t1, -8
bnez t1, .clear_bss
/* Set up stack and enter program code */
la sp, sBootThreadStack + BOOT_STACK_SIZE
la t2, bootproc
jr t2
END(entrypoint)
#ifdef __GNUC__
/* Pad to a total size of 0x60 */
.fill 0x60 - (. - entrypoint)
#else
/* IDO can't take absolute differences of symbols.. */
.repeat (0x60 - 0x34)
.byte 0
.endr
#endif
src/makerom/ipl3.s
-4
View File
@@ -1,4 +0,0 @@
.section .text
.incbin "incbin/ipl3"