hackern64/F3DEX3
0
0
Fork 0
mirror of https://github.com/HackerN64/F3DEX3.git synced 2026-01-21 10:37:45 -08:00
Code Issues Packages Projects Releases Wiki Activity

Integrated changes from Tharo

Browse Source
This commit is contained in:
Sauraen
2024-01-28 16:40:14 -08:00
parent 36c2500ef9
commit d7f33fea92
4 changed files with 110 additions and 88 deletions
Download Patch File Download Diff File Expand all files Collapse all files

28
README.md
View File

@@ -83,6 +83,11 @@ you should expect crashes and graphical issues.**
commands in the next DL to be fetched, rather than always fetching full
buffers, **saving some DRAM traffic** (maybe around 100 us per frame). The
bits used for this are ignored by HLE.
- Segment addresses are now resolved relative to other segments (feature by
Tharo). This enables a strategy for skipping repeated material DLs: call
a segment to run the material, remap the segment in the material to a
display list that immediately returns, and so if the material is called again
it won't run.
- Clipped triangles are drawn by minimal overlapping scanlines algorithm; this
**slightly improves RDP draw time** for large tris (max of about 500 us per
frame, usually much less or zero).
@@ -299,8 +304,7 @@ and commands except:
MVP matrix in F3DEX3.
- `G_MV_POINT` has been removed. This was not used in any command; it would have
likely been used for debugging to copy vertices from DMEM to examine them.
This does not affect `SPModifyVertex`, which is still supported, though this
is moved to Overlay 4 (see below) so it will be slower than in F3DEX2.
This does not affect `SPModifyVertex`, which is still supported.
- `G_MW_PERSPNORM` has been removed; `SPPerspNormalize` is still supported but
is encoded differently, no longer using this define.
- `G_MVO_LOOKATX` and `G_MVO_LOOKATY` have been removed, and `SPLookAtX` and
@@ -336,6 +340,12 @@ them binary incompatible. The lighting data structures, e.g. `Light_t`,
`PosLight_t`, `LookAt_t`, `Lightsn`, `Lights*`, `PosLights*`, etc., have also
changed--generally only slightly, so most code is compatible with no changes.
`SPSegment` has been given a different command id (`G_RELSEGMENT` vs.
`G_MOVEWORD`) to facilitate relative segmented address translation. The
original binary encoding is still valid, but does not support relative
translation like the new encoding. However, recompiling with the C GBI will
always use the new encoding.
## What are the tradeoffs for all these new features?
@@ -368,7 +378,6 @@ This overlay contains handlers for:
discussed below
- The codepath for `SPMatrix` with `G_MTX_MUL` set
- `SPBranchLessZ*`
- `SPModifyVertex`
- `SPDma_io`
Whenever any of these features is needed, the RSP has to swap to Overlay 4. The
@@ -475,6 +484,17 @@ the FIFO is occupied by full-size tris, so the buffers are effectively only two
tris in size because a third tri can't fit. So, their size has been reduced to
two tris, saving a substantial amount of DMEM.
### Segment 0
Segment 0 is now reserved: ensure segment 0 is never set to anything but
0x00000000. In F3DEX2 and prior this was only a good idea (and SM64 and OoT
always follow this); in F3DEX3 segmented addresses are now resolved relative to
other segments. That is, `gsSPSegment(0x08, 0x07001000)` sets segment 8 to the
base address of segment 7 with an additional offset of 0x1000. So for correct
behavior when supplying a direct-mapped or physical address such as 0x80101000,
segment 0 must always be 0x00000000 so that this address resolves to e.g.
0x101000 as expected in this example.
### Obscure semantic differences from F3DEX2 that should never matter in practice
- `SPLoadUcode*` corrupts the current M inverse transpose matrix state. If using
@@ -534,7 +554,7 @@ are credited.
Other credits:
- Wiseguy: large chunk of F3DEX2 disassembly documentation and first version of
build system
- Tharo: relative segment resolution feature, other feature discussions
- Kaze Emanuar: several feature suggestions, testing
- thecozies: Fresnel feature suggestion
- Tharo: feature discussions
- neoshaman: feature discussions

162
f3dex3.s
View File

@@ -415,7 +415,7 @@ jumpTableEntry G_TEXRECT_handler
jumpTableEntry G_TEXRECTFLIP_handler
cmdJumpTable:
jumpTableEntry G_VTX_handler
jumpTableEntry ovl234_ovl4_entrypoint // G_MODIFYVTX
jumpTableEntry G_MODIFYVTX_handler
jumpTableEntry G_CULLDL_handler
jumpTableEntry ovl234_ovl4_entrypoint // G_BRANCH_WZ
jumpTableEntry G_TRI1_handler
@@ -424,6 +424,7 @@ jumpTableEntry G_QUAD_handler
jumpTableEntry G_TRISTRIP_handler
jumpTableEntry G_TRIFAN_handler
jumpTableEntry G_LIGHTTORDP_handler
jumpTableEntry G_RELSEGMENT_handler
// The maximum number of generated vertices in a clip polygon. In reality, this
// is equal to MAX_CLIP_POLY_VERTS, but for testing we can change them separately.
@@ -811,34 +812,32 @@ call_ret_common:
j displaylist_dma_with_count
sb $1, displayListStackLength
G_CULLDL_handler:
j vtx_addrs_from_cmd // Load start vtx addr in $12, end vtx in $3
li $11, culldl_return_from_addrs
culldl_return_from_addrs:
/*
CLIP_OCCLUDED can't be included here because: Suppose the list consists of N-1
verts which are behind the occlusion plane, and 1 vert which is behind the camera
plane and therefore randomly erroneously also set as behind the occlusion plane.
However, the convex hull of all the verts goes through visible area. This will be
incorrectly culled here. We can't afford the extra few instructions to disable
the occlusion plane if the vert is behind the camera, because this only matters for
G_CULLDL and not for tris.
*/
li $1, (CLIP_SCRN_NPXY | CLIP_CAMPLANE)
lhu $11, VTX_CLIP($12)
culldl_loop:
and $1, $1, $11
beqz $1, run_next_DL_command // Some vertex is on the screen-side of all clipping planes; have to render
lhu $11, (vtxSize + VTX_CLIP)($12) // next vertex clip flags
bne $12, $3, culldl_loop // loop until reaching the last vertex
addi $12, $12, vtxSize // advance to the next vertex
li cmd_w0, 0 // Clear count of DL cmds to skip loading
G_ENDDL_handler:
lbu $1, displayListStackLength // Load the DL stack index; if end stack,
beqz $1, load_overlay_0_and_enter // load overlay 0; $1 < 0 signals end
addi $1, $1, -4 // Decrement the DL stack index
j call_ret_common // has a different version in ovl1
lw taskDataPtr, (displayListStack)($1) // Load addr of DL to return to
.if !CFG_GCLK_SAMPLE
G_LIGHTTORDP_handler:
lbu $11, numLightsxSize // Ambient light
lbu $1, (inputBufferEnd - 0x6)(inputBufferPos) // Byte 2 = light count from end * size
andi $2, cmd_w0, 0x00FF // Byte 3 = alpha
sub $1, $11, $1 // Light address; byte 2 counts from end
lw $3, (lightBufferMain-1)($1) // Load light RGB into lower 3 bytes
move cmd_w0, cmd_w1_dram // Move second word to first (cmd byte, prim level)
sll $3, $3, 8 // Shift light RGB to upper 3 bytes and clear alpha byte
j G_RDP_handler // Send to RDP
or cmd_w1_dram, $3, $2 // Combine RGB and alpha in second word
.endif
G_SETxIMG_handler:
beqz $7, G_RDP_handler // Don't do any of this for G_NOOP
lb $3, materialCullMode // Get current mode
jal segmented_to_physical // Convert image to physical address
lw $2, lastMatDLPhyAddr // Get last material physical addr
bnez $3, G_RDP_handler // If not in normal mode (0), exit
add $12, taskDataPtr, inputBufferPos // Current material physical addr
beq $12, $2, @@skip // Branch if we are executing the same mat again
sw $12, lastMatDLPhyAddr // Store material physical addr
li $7, 1 // > 0: in material first time
@@skip: // Otherwise $7 was < 0: cull mode (in mat second time)
j G_RDP_handler
sb $7, materialCullMode
refine_cmd_further:
addi $12, $7, -(0xFF00 | G_SETSCISSOR) // Relative to G_SETSCISSOR = 0
@@ -1492,6 +1491,10 @@ vtx_skip_fog:
jr $ra
sh $12, (VTX_CLIP )(outputVtxPos) // Store first vertex results
G_MODIFYVTX_handler:
// Command byte 3 = vtx being modified; its addr -> $12
li $11, do_moveword // Moveword adds cmd_w0 to $12 for final addr
lbu cmd_w0, (inputBufferEnd - 0x07)(inputBufferPos) // offset in vtx
vtx_addrs_from_cmd:
// Treat eight bytes of last command each as vertex indices << 1
// inputBufferEnd is close enough to the end of DMEM to fit in signed offset
@@ -1841,30 +1844,31 @@ tDaDeI equ $v9
vmadh tV1AtI, tDaDeI, $v26[1]
sdv tDaDeI[8], 0x0020($1) // Store DsDe, DtDe, DwDe texture coefficients (integer)
tV1AtFF equ $v10
vmudn tV1AtFF, tDaDeF, $v4[1] // Super-frac (frac * frac) part; assumes v4 factor >= 0
vmudn tDaDeF, tDaDeF, $v30[7] // 0x0020
vmadh tDaDeI, tDaDeI, $v30[7] // 0x0020
// All values start in element 7. "a", attribute, is Z. Need
// tV1AtI, tV1AtF, tDaDxI, tDaDxF, tDaDeI, tDaDeF, tDaDyI, tDaDyF
vmov tDaDyF[5], tDaDeF[7] // DaDy already in elem 7; DaDe to elem 5
sdv tV1AtF[0], 0x0010($2) // Store RGBA shade color (fractional)
vmudn tDaDxF, tDaDxF, $v30[7] // 0x0020
vmov tDaDyI[5], tDaDeI[7]
sdv tV1AtI[0], 0x0000($2) // Store RGBA shade color (integer)
vmadh tDaDxI, tDaDxI, $v30[7] // 0x0020
vmov tDaDyF[3], tDaDxF[7] // DaDx to elem 3
sdv tV1AtF[8], 0x0010($1) // Store S, T, W texture coefficients (fractional)
vmudn tDaDyF, tDaDyF, $v30[7] // 0x0020
vmov tDaDyI[3], tDaDxI[7]
sdv tV1AtI[8], 0x0000($1) // Store S, T, W texture coefficients (integer)
vmudn tV1AtFF, tDaDeF, $v4[1] // Super-frac (frac * frac) part; assumes v4 factor >= 0
beqz $6, check_rdp_buffer_full // see below
sdv tV1AtI[8], 0x0000($1) // Store S, T, W texture coefficients (integer)
veq $v29, $v31, $v31[1q] // Set VCC to 01010101
vmudn tDaDyF, tDaDyF, $v30[7] // 0x0020
vmadh tDaDyI, tDaDyI, $v30[7] // 0x0020
ssv tDaDeF[14], -0x0006(rdpCmdBufPtr)
vmudl $v29, tV1AtFF, $v30[7] // 0x0020
ssv tDaDeI[14], -0x0008(rdpCmdBufPtr)
vmadn tV1AtF, tV1AtF, $v30[7] // 0x0020
ssv tDaDxF[14], -0x000A(rdpCmdBufPtr)
vmadh tV1AtI, tV1AtI, $v30[7] // 0x0020
ssv tDaDxI[14], -0x000C(rdpCmdBufPtr)
ssv tDaDyF[14], -0x0002(rdpCmdBufPtr)
ssv tDaDyI[14], -0x0004(rdpCmdBufPtr)
ssv tV1AtF[14], -0x000E(rdpCmdBufPtr)
vmrg tDaDyF, tDaDyF, tDaDyI[1q] // Move int elems 3, 5, 7 to result 2, 4, 6
ssv tV1AtF[14], -0x0E(rdpCmdBufPtr)
ssv tV1AtI[14], -0x10(rdpCmdBufPtr)
slv tDaDyF[4], -0x0C(rdpCmdBufPtr) // DaDx i/f
j check_rdp_buffer_full // eventually returns to $ra, which is next cmd, second tri in TRI2, or middle of clipping
ssv tV1AtI[14], -0x10(rdpCmdBufPtr)
sdv tDaDyF[8], -0x08(rdpCmdBufPtr) // DaDe i/f, DaDy i/f
load_overlay_0_and_enter:
G_LOAD_UCODE_handler:
@@ -2097,8 +2101,10 @@ G_RDPHALF_2_handler:
j G_RDP_handler
sdv $v29[0], -8(rdpCmdBufPtr)
G_RELSEGMENT_handler:
jal segmented_to_physical // Resolve new segment address relative to existing segment
G_MOVEWORD_handler:
srl $2, cmd_w0, 16 // load the moveword command and word index into $2 (e.g. 0xDB06 for G_MW_SEGMENT)
srl $2, cmd_w0, 16 // load the moveword command and word index into $2 (e.g. 0xDB06 for G_MW_SEGMENT)
lhu $12, (movewordTable - (G_MOVEWORD << 8))($2) // subtract the moveword label and offset the word table by the word index (e.g. 0xDB06 becomes 0x0304)
do_moveword:
sll $11, cmd_w0, 16 // Sign bit = upper bit of offset
@@ -2118,32 +2124,35 @@ segmented_to_physical:
jr $ra
add cmd_w1_dram, cmd_w1_dram, $11 // Add the segment's address to the masked input address, resulting in the virtual address
G_SETxIMG_handler:
beqz $7, G_RDP_handler // Don't do any of this for G_NOOP
lb $3, materialCullMode // Get current mode
jal segmented_to_physical // Convert image to physical address
lw $2, lastMatDLPhyAddr // Get last material physical addr
bnez $3, G_RDP_handler // If not in normal mode (0), exit
add $12, taskDataPtr, inputBufferPos // Current material physical addr
beq $12, $2, @@skip // Branch if we are executing the same mat again
sw $12, lastMatDLPhyAddr // Store material physical addr
li $7, 1 // > 0: in material first time
@@skip: // Otherwise $7 was < 0: cull mode (in mat second time)
j G_RDP_handler
sb $7, materialCullMode
G_CULLDL_handler:
j vtx_addrs_from_cmd // Load start vtx addr in $12, end vtx in $3
li $11, culldl_return_from_addrs
culldl_return_from_addrs:
/*
CLIP_OCCLUDED can't be included here because: Suppose the list consists of N-1
verts which are behind the occlusion plane, and 1 vert which is behind the camera
plane and therefore randomly erroneously also set as behind the occlusion plane.
However, the convex hull of all the verts goes through visible area. This will be
incorrectly culled here. We can't afford the extra few instructions to disable
the occlusion plane if the vert is behind the camera, because this only matters for
G_CULLDL and not for tris.
*/
li $1, (CLIP_SCRN_NPXY | CLIP_CAMPLANE)
lhu $11, VTX_CLIP($12)
culldl_loop:
and $1, $1, $11
beqz $1, run_next_DL_command // Some vertex is on the screen-side of all clipping planes; have to render
lhu $11, (vtxSize + VTX_CLIP)($12) // next vertex clip flags
bne $12, $3, culldl_loop // loop until reaching the last vertex
addi $12, $12, vtxSize // advance to the next vertex
li cmd_w0, 0 // Clear count of DL cmds to skip loading
G_ENDDL_handler:
lbu $1, displayListStackLength // Load the DL stack index; if end stack,
beqz $1, load_overlay_0_and_enter // load overlay 0; $1 < 0 signals end
addi $1, $1, -4 // Decrement the DL stack index
j call_ret_common // has a different version in ovl1
lw taskDataPtr, (displayListStack)($1) // Load addr of DL to return to
.if !CFG_GCLK_SAMPLE
G_LIGHTTORDP_handler:
lbu $11, numLightsxSize // Ambient light
lbu $1, (inputBufferEnd - 0x6)(inputBufferPos) // Byte 2 = light count from end * size
andi $2, cmd_w0, 0x00FF // Byte 3 = alpha
sub $1, $11, $1 // Light address; byte 2 counts from end
lw $3, (lightBufferMain-1)($1) // Load light RGB into lower 3 bytes
move cmd_w0, cmd_w1_dram // Move second word to first (cmd byte, prim level)
sll $3, $3, 8 // Shift light RGB to upper 3 bytes and clear alpha byte
j G_RDP_handler // Send to RDP
or cmd_w1_dram, $3, $2 // Combine RGB and alpha in second word
.endif
ovl1_end:
.align 8
@@ -2496,10 +2505,8 @@ ovl4_select_instr:
beq $2, $7, calc_mit // otherwise $7 = command byte
li $3, G_BRANCH_WZ
beq $3, $7, G_BRANCH_WZ_handler
li $2, G_MODIFYVTX
beq $2, $7, G_MODIFYVTX_handler
li $3, (0xFF00 | G_DMA_IO)
beq $3, $7, G_DMA_IO_handler
li $2, (0xFF00 | G_DMA_IO)
beq $2, $7, G_DMA_IO_handler
// Otherwise G_MTX_end, which starts with a harmless instruction
G_MTX_end: // Multiplies the temp loaded matrix into the M or VP matrix
@@ -2547,13 +2554,6 @@ G_DMA_IO_handler:
j dma_read_write // Trigger a DMA read or write, depending on the G_DMA_IO flag (which will occupy the sign bit of dmemAddr)
li $ra, wait_for_dma_and_run_next_command // Setup the return address for running the next DL command
G_MODIFYVTX_handler:
j vtx_addrs_from_cmd // byte 3 = vtx being modified; addr -> $12
li $11, modifyvtx_return_from_addrs
modifyvtx_return_from_addrs:
j do_moveword // Moveword adds cmd_w0 to $12 for final addr
lbu cmd_w0, (inputBufferEnd - 0x07)(inputBufferPos)
G_BRANCH_WZ_handler:
j vtx_addrs_from_cmd // byte 3 = vtx being tested; addr -> $12
li $11, branchwz_return_from_addrs

7
gbi.h
View File

@@ -97,6 +97,7 @@ of warnings if you use -Wpedantic. */
#define G_TRISTRIP 0x08 /* = G_LINE3D was a no-op in F3DEX2, has been removed */
#define G_TRIFAN 0x09
#define G_LIGHTTORDP 0x0A
#define G_RELSEGMENT 0x0B
/* names differ between F3DEX2 and F3DZEX */
#define G_BRANCH_Z G_BRANCH_WZ
@@ -2600,11 +2601,11 @@ _DW({ \
/*
* Moveword commands
*/
/* not strictly a moveword command anymore */
#define gSPSegment(pkt, segment, base) \
gMoveWd(pkt, G_MW_SEGMENT, (segment) * 4, (base))
gDma1p((pkt), G_RELSEGMENT, (base), ((segment) * 4) & 0xFFF, G_MW_SEGMENT)
#define gsSPSegment(segment, base) \
gsMoveWd( G_MW_SEGMENT, (segment) * 4, (base))
gsDma1p( G_RELSEGMENT, (base), ((segment) * 4) & 0xFFF, G_MW_SEGMENT)
#define gSPPerspNormalize(pkt, s) gMoveHalfwd(pkt, G_MW_FX, G_MWO_PERSPNORM, (s))
#define gsSPPerspNormalize(s) gsMoveHalfwd( G_MW_FX, G_MWO_PERSPNORM, (s))

1
rsp/gbi.inc
View File

@@ -88,6 +88,7 @@ G_QUAD equ 0x07
G_TRISTRIP equ 0x08
G_TRIFAN equ 0x09
G_LIGHTTORDP equ 0x0a
G_RELSEGMENT equ 0x0b
G_BRANCH_Z equ G_BRANCH_WZ