Completed C-side of: Camera Transformation chapter. Now todo atom tape translation...

code/duffle/gen/macs.h

@@ -16,6 +16,7 @@
 //   source: C:\projects\Pikuma\ps1\code\duffle\lottes_tape.h
 //   source: C:\projects\Pikuma\ps1\code\duffle\bios.h
 //   source: C:\projects\Pikuma\ps1\code\duffle\psyq.h
+//   source: C:\projects\Pikuma\ps1\code\duffle\pad.c
 //   source: C:\projects\Pikuma\ps1\code\duffle\math.atom.c
 //   source: C:\projects\Pikuma\ps1\code\duffle\mips.atom.c
 //   source: C:\projects\Pikuma\ps1\code\duffle\gte.atom.c

code/duffle/gen/offsets.h

@@ -13,6 +13,7 @@
 //   source: C:\projects\Pikuma\ps1\code\duffle\lottes_tape.h
 //   source: C:\projects\Pikuma\ps1\code\duffle\bios.h
 //   source: C:\projects\Pikuma\ps1\code\duffle\psyq.h
+//   source: C:\projects\Pikuma\ps1\code\duffle\pad.c
 //   source: C:\projects\Pikuma\ps1\code\duffle\math.atom.c
 //   source: C:\projects\Pikuma\ps1\code\duffle\mips.atom.c
 //   source: C:\projects\Pikuma\ps1\code\duffle\gte.atom.c

code/duffle/gte.atom.c

@@ -53,16 +53,16 @@ FI_ Slice_MipsCode ac_gte_store_g4_p3(U4 r_primitive_cursor) atom_dbg_skip MipsA
 
 #pragma region Bsked Atoms
 
-typedef Struct_(Binds_SetGteWorld) {
-	M3_S2* transform;
+typedef Struct_(Binds_SetGteMT3S2S4) {
+	MT3_S2S4* transform;
 };
-internal MipsAtom_(set_gte_world) atom_info(
-		atom_bind(Binds_SetGteWorld)
+internal MipsAtom_(set_gte_mt3s2s4) atom_info(
+		atom_bind(Binds_SetGteMT3S2S4)
 	,	atom_reads(R_TapePtr)
 ){
 	/* Pop matrix address from tape into R_T3 ($11) */
-	load_word(R_T3, R_TapePtr, O_(Binds_SetGteWorld,transform)),
-	add_ui_self(    R_TapePtr, S_(Binds_SetGteWorld)),
+	load_word(R_T3, R_TapePtr, O_(Binds_SetGteMT3S2S4,transform)),
+	add_ui_self(    R_TapePtr, S_(Binds_SetGteMT3S2S4)),
 	/* Load 3x3 Rotation + 3x1 Translation from R_T3 into GTE CONTROL Regs (ctc2) */
 	load_word(R_T0, R_T3, 0),  load_word(R_T1, R_T3,  4),
 	gte_mv_to_ctrl_r(R_T0, gte_cr_RT11), gte_mv_to_ctrl_r(R_T1, gte_cr_RT12),

code/duffle/math.h

@@ -26,8 +26,8 @@ typedef Struct_(Extent2_S4) { S4 width; S4 height; };
 typedef Struct_(V2_U1)      { U1 x; U1 y; };
 typedef Struct_(V2_S2)      { S2 x; S2 y; };
 typedef Struct_(V2_S4)      { S4 x; S4 y; };
-typedef Struct_(V3_S2)      { S2 x; S2 y; S2 z; S2 pad; };
-typedef Struct_(V3_S4)      { S4 x; S4 y; S4 z; S4 pad; };
+typedef Struct_(V3_S2)      { S2 x; S2 y; S2 z; S2 pad; }; // PSY-Q: SVECTOR
+typedef Struct_(V3_S4)      { S4 x; S4 y; S4 z; S4 pad; }; // PSY-Q: VECTOR
 typedef Struct_(V4_S2)      { S2 x; S2 y; S2 z; S2 w; };
 typedef Struct_(V4_S4)      { S4 x; S4 y; S4 z; S4 w; };
 
@@ -37,13 +37,19 @@ typedef Struct_(R2_S4)      { V2_S4 p0; V2_S4 p1; };
 typedef Struct_(Rect_S2)    { S2 x; S2 y; S2 width; S2 height; };
 typedef Struct_(Rect_S4)    { S4 x; S4 y; S4 width; S4 height; };
 
-typedef Struct_(M3_S2)      { A3x3_S2 m; A3_S4 t; };
+typedef Struct_(MT3_S2S4)   { A3x3_S2 m; A3_S4 t; }; // PSY-Q: MATRIX
 
 typedef Array_(V2_U1, 2);
 typedef Array_(V2_S2, 2);
 typedef Array_(V2_S2, 3);
 typedef Array_(V2_S2, 4);
 
+enum {
+	fp_one = (1 << 12),
+};
+
+#define v3s4_fp_one() v3s4(fp_one, fp_one, fp_one)
+
 #define v2s2(x,y)     (V2_S2){x,y}
 #define v3s2(x,y,z)   (V3_S2){x,y,z,0}
 #define v3s4(x,y,z)   (V3_S4){x,y,z,0}
@@ -62,5 +68,30 @@ FI_ void add_a3s4_fp(A3_S4_R out_a, A3_S4 b) {
 	(out_a[0])[2] += b[2] >> 1;
 }
 
+FI_ void sub_a3s4(A3_S4_R out_a, A3_S4 b) {
+	(out_a[0])[0] -= b[0];
+	(out_a[0])[1] -= b[1];
+	(out_a[0])[2] -= b[2];
+}
+
+FI_ void sub_a3s4_fp(A3_S4_R out_a, A3_S4 b) {
+	(out_a[0])[0] -= b[0] >> 1;
+	(out_a[0])[1] -= b[1] >> 1;
+	(out_a[0])[2] -= b[2] >> 1;
+}
+
+FI_ void mul_a3s4(A3_S4_R out_a, A3_S4 b) {
+	(out_a[0])[0] *= b[0];
+	(out_a[0])[1] *= b[1];
+	(out_a[0])[2] *= b[2];
+}
+
 FI_ void add_v3s4   (V3_S4_R out_a, V3_S4 b) {  add_a3s4   (pcast(A3_S4_R, out_a), pcast(A3_S4, b));  }
 FI_ void add_v3s4_fp(V3_S4_R out_a, V3_S4 b) {  add_a3s4_fp(pcast(A3_S4_R, out_a), pcast(A3_S4, b));  }
+
+FI_ void sub_v3s4   (V3_S4_R out_a, V3_S4 b) {  sub_a3s4   (pcast(A3_S4_R, out_a), pcast(A3_S4, b));  }
+FI_ void sub_v3s4_fp(V3_S4_R out_a, V3_S4 b) {  sub_a3s4_fp(pcast(A3_S4_R, out_a), pcast(A3_S4, b));  }
+
+FI_ void mul_v3s4   (V3_S4_R out_a, V3_S4 b) {  mul_a3s4   (pcast(A3_S4_R, out_a), pcast(A3_S4, b));  }
+
+

code/duffle/pad.c

@@ -0,0 +1,78 @@
+#ifdef INTELLISENSE_DIRECTIVES
+#	include "dsl.h"
+#	include "gcc_asm.h"
+#	include "mips.h"
+#	include "bios.h"
+#	include "pad.h"
+#endif
+
+/* Uses ONE 8-byte frame allocated via the compiler's standard prologue.
+ * 4 wasted-arg words for B(12h) InitPAD2 are at [SP+0..15] but are not explicitly allocated.
+ * Compiler handles the MIPS O32 "wasted stack" convention for us by treating the B-call as a 4-arg call.
+ *
+ * The buffer pointers are passed as arguments so the compiler keeps them in callee-saved registers;
+ * The B(12h) asm volatile block does NOT clobber those registers (it clobbers only the volatile GPRs + B-table arg registers explicitly).
+ * The C-level writes after the call re-load the pointers from their callee-saved homes.
+ *
+ * The clobber list for both B-calls names the full BIOS destroy set documented in kernelbios.md:167-174 (R1..R15, R24..R25, R31, HI/LO).
+ * The kernel-ABI "volatile GPRs" subset is clb_mem_drain; the rest of the destroy set is enumerated explicitly here. */
+NI_ void pad_bios_init_start(PadBiosRaw* raw0, PadBiosRaw* raw1)
+{
+	/* Pin raw0 + raw1 to $a0 + $a1 via rgcc; the B(12h) call uses these directly.
+	 * The `(void)` casts mark them as unread after the call so the compiler doesn't need to move them back. */
+	register PadBiosRaw* p0 rgcc(R_A0) = raw0;
+	register PadBiosRaw* p1 rgcc(R_A1) = raw1;
+	(void)p0; (void)p1;
+
+	// TODO(Ed): Properly annotate the raw values in the inline asm instructions.
+	// Use enums.
+
+	/* B(12h) InitPAD2(raw0, 0x22, raw1, 0x22)
+	 *   $a0 = raw0 (rgcc-bound; survives the sequence below)
+	 *   $a1 = raw1 (preserved into $a2 before $a1 is overwritten)
+	 *   $a2 = raw1 (moved from $a1; survives $a1's overwrite)
+	 *   $a3 = 0x22 (immediate)
+	 *   $t1 = 0x12 (function number)
+	 *   $t2 = 0xB0 (BIOS B-table address) */
+	asm volatile(
+		asm_words(
+			or_u(    rarg_2, rarg_1, rdiscard),               /* $a2 = $a1 = raw1 */
+			add_ui(  rarg_1, rdiscard, bios_pad_buffer_size), /* $a1 = 0x22 */
+			add_ui(  rarg_3, rdiscard, bios_pad_buffer_size), /* $a3 = 0x22 */
+			add_ui(  rtmp_1, rdiscard, bios_init_pad_2),      /* $t1 = 0x12 */
+			add_ui(  rtmp_2, rdiscard, bios_btable_addr),     /* $t2 = 0xB0 */
+			call_reg(rtmp_2),                                 /* jalr $t2, $ra */
+			nop                                               /* BD slot */
+		)
+		asm_rpins, r_use(p0), r_use(p1)
+		asm_clobber: 
+			rlit(R_AT), 
+			rlit(R_V0), rlit(R_V1), 
+			rlit(R_T0), rlit(R_T1), rlit(R_T2), rlit(R_T3), rlit(R_T4),
+			rlit(R_T5), rlit(R_T6), rlit(R_T7), rlit(R_T8), rlit(R_T9),
+			rlit(R_RA),
+			clb_mem_drain
+	);
+
+	/* The C-level writes re-load the pointers via the parameter names and write 0xFF to each 
+	 * buffer's status byte to mark the initial-state hazard documented in kernelbios.md:1621-1624. */
+	u1_v(raw0)[0] = 0xFF;
+	u1_v(raw1)[0] = 0xFF;
+
+	/* B(13h) StartPAD2() — no args. The BIOS preserves $sp. */
+	asm volatile(
+		asm_words(
+			add_ui(  rtmp_1, rdiscard, bios_start_pad_2), /* $t1 = 0x13 */
+			add_ui(  rtmp_2, rdiscard, bios_btable_addr), /* $t2 = 0xB0 (re-load) */
+			call_reg(rtmp_2),                             /* jalr $t2, $ra */
+			nop                                           /* BD slot */
+		)
+		asm_clobber:
+			rlit(R_AT),
+			rlit(R_V0), rlit(R_V1),
+			rlit(R_T0), rlit(R_T1), rlit(R_T2), rlit(R_T3), rlit(R_T4),
+			rlit(R_T5), rlit(R_T6), rlit(R_T7), rlit(R_T8), rlit(R_T9),
+			rlit(R_RA),
+			clb_mem_drain
+	);
+}

code/duffle/pad.h

@@ -33,9 +33,6 @@ enum {
 	Pad1 = 1 << PadId_Offset,
 };
 
-#define pad0_(btn_id) (btn_id << Pad0)
-#define pad1_(btn_id) (btn_id << Pad1)
-
 /* =============================================================================
  * BIOS pad-buffer subsystem: docs/psx-spx/docs/kernelbios.md (B(12h) + B(13h))
  * ============================================================================= */
@@ -113,3 +110,5 @@ typedef Struct_(PadState) {
 		};
 	};
 };
+
+internal void pad_bios_init_start(PadBiosRaw* raw0, PadBiosRaw* raw1);

code/duffle/psyq.h

@@ -64,9 +64,9 @@ typedef Struct_(Tile) {
 	Linear Algebra
 */
 
-M3_S2* m3s2_rotation   (V3_S2* vec, M3_S2* mat) asm("RotMatrix");
-M3_S2* m3s2_translation(M3_S2* mat, V3_S4* vec) asm("TransMatrix");
-M3_S2* m3s2_scale      (M3_S2* mat, V3_S4* vec) asm("ScaleMatrix");
+MT3_S2S4* mt3s2s4_rotation   (V3_S2*    vec, MT3_S2S4* mat) asm("RotMatrix");
+MT3_S2S4* mt3s2s4_translation(MT3_S2S4* mat, V3_S4*    vec) asm("TransMatrix");
+MT3_S2S4* mt3s2s4_scale      (MT3_S2S4* mat, V3_S4*    vec) asm("ScaleMatrix");
 
 // Rotation, Translation, Perspective
 
@@ -99,5 +99,17 @@ FI_ S4 rtp_avg_nclip_a4_v3s2(
 	);
 }
 
-void gte_matrix_set_rotation   (M3_S2* mat) asm("SetRotMatrix");
-void gte_matrix_set_translation(M3_S2* mat) asm("SetTransMatrix");
+void gte_matrix_set_rotation   (MT3_S2S4* mat) asm("SetRotMatrix");
+void gte_matrix_set_translation(MT3_S2S4* mat) asm("SetTransMatrix");
+
+// Einheit, Metrication to unit vector. "Normalization", not Orthogonal "Normal, Normalis". Directionalization.
+S4 normalize_v3s4(V3_S4* v0, V3_S4* v1) asm("VectorNormal");
+
+V3_S4* mul_m3s2_v3s4(MT3_S2S4* m, V3_S4* v, V3_S4* result) asm("ApplyMatrixLV");
+
+MT3_S2S4* trans_m3s2(MT3_S2S4* m, V3_S4* off) asm("TransMatrix");
+
+MT3_S2S4* gte_comp_coord_m3s2(MT3_S2S4* m0, MT3_S2S4* m1, MT3_S2S4* result) asm("CompMatrixLV");
+
+// TODO(Ed): Want to interpret this under the lens of Eric Lengyel's geometric algebra
+void cross_v3s4(V3_S4* v0, V3_S4* v1, V3_S4* result) asm("OuterProduct12");

code/hello_camera/gen/offsets.h

@@ -8,7 +8,7 @@
 #pragma region hello_camera
 
 
-// --- atom: pad_apply_input (60 words) ---
+// --- atom: pad_input_cube_rotation (60 words) ---
 
 #define _atom_offset_dpad_left_exit_dpad_left         6
 #define _atom_offset_dpad_right_exit_dpad_right       6

code/hello_camera/hello_camera.atom.c

@@ -210,7 +210,7 @@ enum {
 	R_CubeRot    = R_T1 atom_reg,
 	R_FloorRot   = R_T2 atom_reg,
 };
-internal MipsAtom_(pad_apply_input) atom_info(atom_bind(Binds_PadApplyInput)
+internal MipsAtom_(pad_input_cube_rotation) atom_info(atom_bind(Binds_PadApplyInput)
 , atom_reads(R_T0, R_CubeRot, R_FloorRot, R_T3, R_T4, R_PadStateT5, R_TapePtr)
 , atom_writes(     R_CubeRot, R_FloorRot)
 ) {
@@ -225,7 +225,7 @@ internal MipsAtom_(pad_apply_input) atom_info(atom_bind(Binds_PadApplyInput)
 	// Note(Ed): Potential op with delay slot?
 
 	/* D-pad Left: cube_rot.y += 30, floor_rot.y += 5.  */
-	and_i(R_T3, R_T0, pad0_(Pad_Left)), branch_le_zero(R_T3, atom_offset(dpad_left, exit_dpad_left)),
+	and_i(R_T3, R_T0, Pad_Left), branch_le_zero(R_T3, atom_offset(dpad_left, exit_dpad_left)),
 		load_half( R_T4, R_CubeRot,  O_(V3_S2,y)),   /* BD-slot */
 		load_half( R_T3, R_FloorRot, O_(V3_S2,y)),
 		add_si(    R_T4, R_T4, 30),
@@ -235,7 +235,7 @@ internal MipsAtom_(pad_apply_input) atom_info(atom_bind(Binds_PadApplyInput)
 	atom_label(exit_dpad_left)
 
 	/* D-pad Right: cube_rot.y -= 30, floor_rot.y -= 5. */
-	and_i(R_T3, R_T0, pad0_(Pad_Right)), branch_le_zero(R_T3, atom_offset(dpad_right, exit_dpad_right)),
+	and_i(R_T3, R_T0, Pad_Right), branch_le_zero(R_T3, atom_offset(dpad_right, exit_dpad_right)),
 		load_half( R_T4, R_CubeRot,  O_(V3_S2,y)),   /* BD-slot */
 		load_half( R_T3, R_FloorRot, O_(V3_S2,y)),
 		add_si(    R_T4, R_T4, -30),
@@ -312,6 +312,32 @@ atom_label(exit_stick)
 	mac_yield_tail(),
 };
 
+enum {
+	R_Cam = R_T4 atom_reg,
+};
+typedef Struct_(Binds_PadInputCam) {
+	Camera* cam;
+};
+internal MipsAtom_(pad_input_cam) atom_info(atom_bind(Binds_PadInputCam)) {
+	load_word(R_Cam, R_TapePtr, O_(Binds_PadInputCam,cam)),
+	add_ui_self(     R_TapePtr, S_(Binds_PadInputCam)),
+	// TODO(Ed): Implement.
+
+	mac_yield(),
+};
+
+enum {
+	_LookAt_WIP,
+};
+typedef Struct_(Binds_ResolveLookAt) {
+	U1 bla;
+};
+internal MipsAtom_(resolve_look_at) atom_info(atom_bind(Binds_ResolveLookAt)) {
+	add_ui_self(R_TapePtr, S_(Binds_ResolveLookAt)),
+
+	mac_yield(),
+};
+
 enum {
 	R_PrimCursor = R_T7 atom_reg atom_type(U4*),    /* VRAM output cursor (primitive buffer) */
 	R_FaceCursor = R_T4 atom_reg atom_type(V4_S2*), /* Cube face-index cursor (V4_S2*); floor context switches to V3_S2* via atom_phase */

code/hello_camera/hello_camera.c

@@ -31,6 +31,7 @@
 #pragma endregion Duffle Headers
 
 #pragma region Duffle TUs
+#include "duffle/pad.c"
 #include "duffle/math.atom.c"
 #include "duffle/mips.atom.c"
 #include "duffle/gte.atom.c"
@@ -62,7 +63,10 @@ typedef Struct_(SMemory) {
 
 	U4 MemTape[MemTape_Len];
 
-	M3_S2 tform_world;
+	MT3_S2S4 tform_world;
+	MT3_S2S4 tform_view;
+
+	Camera cam;
 
 	Ent_Cube  cube;
 	Ent_Floor floor;
@@ -75,6 +79,9 @@ typedef Struct_(SMemory) {
 global SMemory smem;
 extern SMemory smem;
 
+#define pad0_btn_(btn) btn & smem.pad[0].buttons
+#define pad1_btn_(btn) btn & smem.pad[1].buttons
+
 I_ B1* prim__alloc(U4 type_width, Str8 type_name) {
 	gknown PrimitiveArena* pa  = & smem.primitives;
 	gknown B1*             buf = (B1*) r_(smem.primitives.buf)[smem.active_buf_id];
@@ -85,77 +92,31 @@ I_ B1* prim__alloc(U4 type_width, Str8 type_name) {
 }
 #define prim_alloc(type) (type*)prim__alloc(S_(type), slit( stringify(type)))
 
-/* Uses ONE 8-byte frame allocated via the compiler's standard prologue.
- * The 4 wasted-arg words for B(12h) InitPAD2 live at [SP+0..15] but are not explicitly allocated.
- * The compiler handles the MIPS O32 "wasted stack" convention for us by treating the B-call as a 4-arg call.
- *
- * The buffer pointers are passed as arguments so the compiler keeps them in callee-saved registers;
- * The B(12h) asm volatile block does NOT clobber those registers (it clobbers only the volatile GPRs + the B-table arg registers explicitly).
- * The C-level writes after the call re-load the pointers from their callee-saved homes.
- *
- * The clobber list for both B-calls names the full BIOS destroy set documented in kernelbios.md:167-174 (R1..R15, R24..R25, R31, HI/LO).
- * The kernel-ABI "volatile GPRs" subset is clb_system; the rest of the destroy set is enumerated explicitly here. */
-NI_ void pad_bios_init_start(PadBiosRaw* raw0, PadBiosRaw* raw1)
-{
-	/* Pin raw0 + raw1 to $a0 + $a1 via rgcc; the B(12h) call uses these directly.
-	 * The `(void)` casts mark them as unread after the call so the compiler doesn't need to move them back. */
-	register PadBiosRaw* p0 rgcc(R_A0) = raw0;
-	register PadBiosRaw* p1 rgcc(R_A1) = raw1;
-	(void)p0; (void)p1;
+void
+resolve_look_at_c11(MT3_S2S4* look_at, V3_S4* eye, V3_S4* target, V3_S4* up_in) {
+// TODO(Ed): Want to interpret this under the lens of Eric Lengyel's geometric algebra
+	V3_S4 right, up, forward;
+	V3_S4 ux, uy, uz;
+	V3_S4 pos, off;
 
-	// TODO(Ed): Properly annotate the raw values in the inline asm instructions.
-	// Use enums.
+	forward = target[0]; sub_v3s4(& forward, eye[0]);
+	normalize_v3s4(& forward, & uz);
 
-	/* B(12h) InitPAD2(raw0, 0x22, raw1, 0x22)
-	 *   $a0 = raw0 (rgcc-bound; survives the sequence below)
-	 *   $a1 = raw1 (preserved into $a2 before $a1 is overwritten)
-	 *   $a2 = raw1 (moved from $a1; survives $a1's overwrite)
-	 *   $a3 = 0x22 (immediate)
-	 *   $t1 = 0x12 (function number)
-	 *   $t2 = 0xB0 (BIOS B-table address) */
-	asm volatile(
-		asm_words(
-			or_u(    rarg_2, rarg_1, rdiscard), /* $a2 = $a1 = raw1 */
-			add_ui(  rarg_1, rdiscard, bios_pad_buffer_size), /* $a1 = 0x22 */
-			add_ui(  rarg_3, rdiscard, bios_pad_buffer_size), /* $a3 = 0x22 */
-			add_ui(  rtmp_1, rdiscard, bios_init_pad_2),      /* $t1 = 0x12 */
-			add_ui(  rtmp_2, rdiscard, bios_btable_addr),     /* $t2 = 0xB0 */
-			call_reg(rtmp_2),                   /* jalr $t2, $ra */
-			nop                                 /* BD slot */
-		)
-		asm_rpins, r_use(p0), r_use(p1)
-		asm_clobber: 
-			rlit(R_AT), 
-			rlit(R_V0), rlit(R_V1), 
-			rlit(R_T0), rlit(R_T1), rlit(R_T2), rlit(R_T3), rlit(R_T4),
-			rlit(R_T5), rlit(R_T6), rlit(R_T7), rlit(R_T8), rlit(R_T9),
-			rlit(R_RA),
-			clb_mem_drain
-	);
+	cross_v3s4(& uz,   up_in, & right); normalize_v3s4(& right, & ux);
+	cross_v3s4(& uz, & ux,    & up);    normalize_v3s4(& up,    & uy);
 
-	/* The C-level writes re-load the pointers via the parameter names and write 0xFF to each 
-	 * buffer's status byte to mark the initial-state hazard documented in kernelbios.md:1621-1624. */
-	u1_v(raw0)[0] = 0xFF;
-	u1_v(raw1)[0] = 0xFF;
+	look_at->m[0][0] = ux.x; look_at->m[0][1] = ux.y; look_at->m[0][2] = ux.z;
+	look_at->m[1][0] = uy.x; look_at->m[1][1] = uy.y; look_at->m[1][2] = uy.z;
+	look_at->m[2][0] = uz.x; look_at->m[2][1] = uz.y; look_at->m[2][2] = uz.z;
+	
+	pos = eye[0]; mul_v3s4(& pos, v3s4(-1,-1,-1));
 
-	/* B(13h) StartPAD2() — no args. The BIOS preserves $sp. */
-	asm volatile(
-		asm_words(
-			add_ui(  rtmp_1, rdiscard, bios_start_pad_2), /* $t1 = 0x13 */
-			add_ui(  rtmp_2, rdiscard, bios_btable_addr), /* $t2 = 0xB0 (re-load) */
-			call_reg(rtmp_2),                 /* jalr $t2, $ra */
-			nop                               /* BD slot */
-		)
-		asm_clobber:
-			rlit(R_AT),
-			rlit(R_V0), rlit(R_V1),
-			rlit(R_T0), rlit(R_T1), rlit(R_T2), rlit(R_T3), rlit(R_T4),
-			rlit(R_T5), rlit(R_T6), rlit(R_T7), rlit(R_T8), rlit(R_T9),
-			rlit(R_RA),
-			clb_mem_drain
-	);
+	mul_m3s2_v3s4(look_at, & pos, & off);
+	trans_m3s2(look_at, & off);
 }
 
+FI_ void camera_look_at_c11(Camera* c, V3_S4* target, V3_S4* up_in) { resolve_look_at_c11(& c->look_at, & c->pos, target, up_in); }
+
 GCC_OPTIMIZATION_DISABLE
 void update(PrimitiveArena* pa, U4* ordering_buf) 
 {
@@ -171,11 +132,36 @@ void update(PrimitiveArena* pa, U4* ordering_buf)
 			tb_emit_(pad_bios_snapshot);
 				tb_data_(raw,   & smem.pad_raw[1]);
 				tb_data_(state, & smem.pad[1]);
-			// Demo input
-			tb_emit_(pad_apply_input);
-				tb_data_(state,     & smem.pad[0]);
-				tb_data_(cube_rot,  & smem.cube.rot);
-				tb_data_(floor_rot, & smem.floor.rot);
+
+			// TODO(Ed): Implement based on below.
+			// tb_emit_(pad_input_cam);
+			// 	tb_data_(cam, & smem.cam);
+
+			if (pad0_btn_(Pad_Left)) {
+				smem.cam.pos.x -= 50;
+			}
+			if (pad0_btn_(Pad_Right)) {
+				smem.cam.pos.x += 50;
+			}
+			if (pad0_btn_(Pad_Up)) {
+				smem.cam.pos.y -= 50;
+			}
+			if (pad0_btn_(Pad_Down)) {
+				smem.cam.pos.y += 50;
+			}
+			if (pad0_btn_(Pad_Cross)) {
+				smem.cam.pos.z -= 50;
+			}
+			if (pad0_btn_(Pad_Circle)) {
+				smem.cam.pos.z += 50;
+			}
+
+
+			// Demo input (not longer using)
+			// tb_emit_(pad_input_cube_rotation);
+			// 	tb_data_(state,     & smem.pad[0]);
+			// 	tb_data_(cube_rot,  & smem.cube.rot);
+			// 	tb_data_(floor_rot, & smem.floor.rot);
 		}
 	}
 
@@ -202,15 +188,34 @@ void update(PrimitiveArena* pa, U4* ordering_buf)
 	A2_S2 p;    //???
 	S4 flag; //????
 
+	// Camera Look at
+	if (1)
+	{
+		camera_look_at_c11(& smem.cam, & smem.cube.pos, & v3s4(0, -fp_one, 0));
+	}
+	// Camera look at (Tape)
+	if (0)
+	{
+		tb.used = 0; tb_scope_run(& tb) {
+			tb_emit_(resolve_look_at);
+				// tb_data_();
+		}
+	}
 
 	// Draw cube
 	if (1)
 	{
-		m3s2_rotation   (& smem.cube.rot,    & smem.tform_world);
-		m3s2_translation(& smem.tform_world, & smem.cube.pos);
-		m3s2_scale      (& smem.tform_world, & smem.cube.scale);
-		gte_matrix_set_rotation   (& smem.tform_world);
-		gte_matrix_set_translation(& smem.tform_world);
+		mt3s2s4_rotation   (& smem.cube.rot,    & smem.tform_world);
+		mt3s2s4_translation(& smem.tform_world, & smem.cube.pos);
+		mt3s2s4_scale      (& smem.tform_world, & smem.cube.scale);
+
+		// Combine world and look_at matrix.
+		gte_comp_coord_m3s2(& smem.cam.look_at, & smem.tform_world, & smem.tform_view);
+		gte_matrix_set_rotation   (& smem.tform_view);
+		gte_matrix_set_translation(& smem.tform_view);
+
+		// gte_matrix_set_rotation   (& smem.tform_world);
+		// gte_matrix_set_translation(& smem.tform_world);
 
 		U4 prim_base   = u4_(pa->buf[smem.active_buf_id]);
 		U4 prim_cursor = prim_base + pa->used;
@@ -238,9 +243,15 @@ void update(PrimitiveArena* pa, U4* ordering_buf)
 	// Draw floor
 	if (1)
 	{
-		m3s2_rotation   (& smem.floor.rot,   & smem.tform_world);
-		m3s2_translation(& smem.tform_world, & smem.floor.pos);
-		m3s2_scale      (& smem.tform_world, & smem.floor.scale);
+		mt3s2s4_rotation   (& smem.floor.rot,   & smem.tform_world);
+		mt3s2s4_translation(& smem.tform_world, & smem.floor.pos);
+		mt3s2s4_scale      (& smem.tform_world, & smem.floor.scale);
+
+		// Combine world and look_at matrix.
+		gte_comp_coord_m3s2(& smem.cam.look_at, & smem.tform_world, & smem.tform_view);
+
+		gte_matrix_set_rotation   (& smem.tform_view);
+		gte_matrix_set_translation(& smem.tform_view);
 
 		U4 prim_base   = u4_(pa->buf[smem.active_buf_id]);
 		U4 prim_cursor = prim_base + pa->used;
@@ -250,8 +261,8 @@ void update(PrimitiveArena* pa, U4* ordering_buf)
 
 		// Prepare the tape. (Push protocol to tape)
 		tb.used = 0; tb_scope(& tb) {
-			tb_emit(& tb, set_gte_world);
-				tb_data(& tb, u4_(& smem.tform_world));
+			// tb_emit(& tb, set_gte_mt3s2s4);
+			// 	tb_data(& tb, u4_(& smem.tform_view));
 
 			tb_emit(& tb, rbind_floor_f3_face);
 			// TODO(Ed): Just use a single context struct ref?
@@ -297,6 +308,7 @@ int main(void)
 	smem.scratchpad = C_(U4_V, 0x1F800000);
 	// smem.primitives.used = 0;
 	// smem.active_buf_id   = 0;
+	smem.cam.pos = v3s4(500, -1000, -1500);
 	/*Persistent Entity Setup*/{
 		ent_cube128_init(& smem.cube.verts, & smem.cube.faces); {
 			Ent_Cube* cube = & smem.cube;

code/hello_camera/hello_camera.h

@@ -21,12 +21,6 @@ enum {
 	ScreenRes_CenterY = (ScreenRes_Y >> 1),
 };
 
-enum {
-	fp_one = (1 << 12),
-};
-
-#define v3s4_fp_one() v3s4(fp_one, fp_one, fp_one)
-
 typedef U4 OrderingTable_Buffer[OrderingTbl_Len];
 typedef Array_(OrderingTable_Buffer, 2);
 
@@ -100,3 +94,9 @@ typedef Struct_(Ent_Floor) {
 	A4_V3_S2 verts;
 	A2_V3_S2 faces;
 };
+
+typedef Struct_(Camera) {
+	V3_S4    pos;
+	V3_S2    rot;
+	MT3_S2S4 look_at;
+};