further disable optimizations

dreamcast/Makefile

@@ -83,24 +83,25 @@ OBJS_TEXCONV += \
 # Add compilation units to this list to explicity compile them with
 # -O3 optimizations, while the rest get the default (-Os) treatment
 # to conserve RAM.
-OBJS_O3 = \
-	../vendor/librw/src/dc/rwdc.o \
-	../src/core/World.o \
-	../src/collision/Collision.o \
-	../src/math/math.o \
-	../src/math/Matrix.o \
-	../src/math/Quaternion.o \
-	../src/math/Rect.o \
-	../src/math/Vector.o \
-	../vendor/librw/src/base.o \
-	../src/renderer/Shadows.o
+OBJS_O3 = 
+# \
+# 	../vendor/librw/src/dc/rwdc.o \
+# 	../src/core/World.o \
+# 	../src/collision/Collision.o \
+# 	../src/math/math.o \
+# 	../src/math/Matrix.o \
+# 	../src/math/Quaternion.o \
+# 	../src/math/Rect.o \
+# 	../src/math/Vector.o \
+# 	../vendor/librw/src/base.o \
+# 	../src/renderer/Shadows.o
 
 OBJS_NO_FAST_MATH = \
 	../src/core/Cam.o \
 	../src/core/Camera.o
 
-KOS_CPPFLAGS += -fbuiltin -ffast-math -ffp-contract=fast \
-	-mfsrra -mfsca
+KOS_CPPFLAGS += -mfsrra -mfsca  -fbuiltin #-ffast-math -ffp-contract=fast \
+	
 
 ifdef KOS_BASE
 include $(KOS_BASE)/Makefile.rules
@@ -113,9 +114,9 @@ DEPS = $(OBJS:.o=.d) $(OBJS_TEXCONV:.o:.d)
 CXXFLAGS += $(if $(WITH_32MB),-O3,-Os) \
 	$(if $(WITH_IDE),-DWITH_IDE) \
 	$(if $(WITH_PROF),-DWITH_PROF=\"$(WITH_PROF)\") \
-	-MMD -MP -ffunction-sections -fdata-sections -ffast-math \
+	-MMD -MP -ffunction-sections -fdata-sections \
 	-fmerge-all-constants -fomit-frame-pointer -ml -std=gnu++20 \
-	-fno-exceptions -fno-rtti -flto=auto -fipa-pta -Wno-write-strings \
+	-fno-exceptions -fno-rtti  -fipa-pta -Wno-write-strings \
 	-Wno-deprecated-enum-enum-conversion -Wno-deprecated-enum-float-conversion \
 	-Wno-multichar -Wno-unused-value -Wno-char-subscripts -Wno-reorder \
 	-Wno-unused-function -Wno-class-memaccess -fno-permissive \
@@ -155,7 +156,7 @@ $(OBJS_NO_FAST_MATH): %.o: %.cpp
 
 $(TARGET): $(OBJS)
 	kos-c++ -o $(TARGET) $(OBJS) -Wl,--gc-sections -Wl,--as-needed -Wl,-Map,output.map \
-		-flto=auto $(if $(WITH_IDE),-lkosfat) $(if $(WITH_SD),-lkosfat) -Wl,--build-id=sha1 
+		 $(if $(WITH_IDE),-lkosfat) $(if $(WITH_SD),-lkosfat) -Wl,--build-id=sha1 
 	@echo && echo && echo "*** Build Completed Successfully ***" && echo && echo
 
 run: $(TARGET)

src/math/Vector.cpp

@@ -27,7 +27,7 @@ CrossProduct(const CVector &v1, const CVector &v2)
 CVector
 Multiply3x3(const CMatrix &mat, const CVector &vec)
 {
-#ifdef DC_SH4
+#if defined(DC_SH4) && 0
     register float __x __asm__("fr12") = vec.x;
     register float __y __asm__("fr13") = vec.y;
     register float __z __asm__("fr14") = vec.z;
@@ -58,7 +58,7 @@ Multiply3x3(const CVector &vec, const CMatrix &mat)
 CVector
 operator*(const CMatrix &mat, const CVector &vec)
 {
-#ifdef DC_SH4
+#if defined(DC_SH4) && 0
 	CVector out;
 	mat_load(reinterpret_cast<matrix_t *>(const_cast<CMatrix *>(&mat)));
 	mat_trans_single3_nodiv_nomod(vec.x, vec.y, vec.z, out.x, out.y, out.z);

src/math/Vector.h

@@ -20,7 +20,7 @@ public:
   // (0,1,0) means no rotation. So get right vector and its atan
   __always_inline float Heading(void) const { return Atan2(-x, y); }
   __always_inline float Magnitude(void) const { 
-#ifdef DC_SH4
+#if defined(DC_SH4) && 0
     float w;
     vec3f_length(x, y, z, w);
     return w;
@@ -29,7 +29,7 @@ public:
 #endif
 }
   __always_inline float MagnitudeSqr(void) const { 
-#ifdef DC_SH4
+#if defined(DC_SH4) && 0
     return fipr_magnitude_sqr(x, y,z, 0.0f); 
 #else
     return x*x + y*y + z*z;
@@ -118,7 +118,7 @@ inline CVector operator/(const CVector &left, float right)
 __always_inline float
 DotProduct(const CVector &v1, const CVector &v2)
 {
-#ifdef DC_SH4
+#if defined(DC_SH4) && 0
   return fipr(v1.x, v1.y, v1.z, 0.0f, v2.x, v2.y, v2.z, 0.0f);
 #else
   return v1.x*v2.x + v1.y*v2.y + v1.z*v2.z;
@@ -131,7 +131,7 @@ inline float
 Distance(const CVector &v1, const CVector &v2)
 {
   float w;
-#ifdef DC_SH4
+#if defined(DC_SH4) && 0
   vec3f_distance(v1.x, v1.y, v1.z, v2.x, v2.y, v2.z, w);
   return w;
 #else

src/math/VuVector.h

@@ -43,7 +43,7 @@ __always_inline void TransformPoint(CVuVector &out, const CMatrix &mat, const CV
 		vmaddw.xyz	vf06,vf05,vf00\n\
 		sqc2    vf06,0x0(%0)\n\
 		": : "r" (&out) , "r" (&mat) ,"r" (&in): "memory");
-#elif defined(DC_SH4)
+#elif defined(DC_SH4) && 0
 	mat_load(reinterpret_cast<matrix_t *>(const_cast<CMatrix *>(&mat)));
 	mat_trans_nodiv_nomod(in.x, in.y, in.z, out.x, out.y, out.z, out.y);
 #else
@@ -70,7 +70,7 @@ __always_inline void TransformPoint(CVuVector &out, const CMatrix &mat, const Rw
 		vmaddw.xyz	vf06,vf05,vf00\n\
 		sqc2    vf06,0x0(%0)\n\
 		": : "r" (&out) , "r" (&mat) ,"r" (&in): "memory");
-#elif defined(DC_SH4)
+#elif  defined(DC_SH4) && 0
 	mat_load(reinterpret_cast<matrix_t *>(const_cast<CMatrix *>(&mat)));
 	mat_trans_nodiv_nomod(in.x, in.y, in.z, out.x, out.y, out.z, out.y);
 #else
@@ -108,7 +108,7 @@ __always_inline void TransformPoints(CVuVector *out, int n, const CMatrix &mat,
 		sqc2    vf06,-0x10(%0)\n\
 		bnez	%1,1b\n\
 		": : "r" (out) , "r" (n), "r" (&mat), "r" (in), "r" (stride): "memory");
-#elif defined(DC_SH4)
+#elif  defined(DC_SH4) && 0
     mat_load(reinterpret_cast<matrix_t *>(const_cast<CMatrix *>(&mat)));
 	mat_transform(reinterpret_cast<vector_t *>(const_cast<RwV3d *>(in)),
 	              reinterpret_cast<vector_t *>(out), 

vendor/librw/src/dc/rwdc.cpp

@@ -2462,8 +2462,7 @@ void* interpolateAndSubmit(void* dst, const void* src1, const void* src2, uint32
 		float16 v_u = v1_u + t * (v2_u - v1_u);
 		float16 v_v = v1_v + t * (v2_v - v1_v);
 
-		v->u = v_u.raw;
-		v->v = v_v.raw;
+		v->uv = v_v.raw | (v_u.raw << 16);
 
 		pvr_dr_commit(v);
 		v ++;