Split off libsmd from n00bdemo into separate library

14 files changed, 1086 insertions, 1053 deletions

diff --git a/doc/cmake_reference.md b/doc/cmake_reference.md
index ef3f5dc..169a7e6 100644
--- a/doc/cmake_reference.md
+++ b/doc/cmake_reference.md
@@ -76,6 +76,7 @@ the following targets are defined:
 - `psxsio`
 - `psxetc`
 - `psxapi`
+- `smd`
 - `lzp`
 - `c`
 
@@ -394,4 +395,4 @@ CMake's `add_custom_command()` and `add_custom_target()` to convert models and
 generate LZP archives as part of the build pipeline.
 
 -----------------------------------------
-_Last updated on 2022-12-18 by spicyjpeg_
+_Last updated on 2023-06-20 by spicyjpeg_
diff --git a/examples/demos/n00bdemo/disp.c b/examples/demos/n00bdemo/disp.c
index a43a1d6..2e02a9d 100644
--- a/examples/demos/n00bdemo/disp.c
+++ b/examples/demos/n00bdemo/disp.c
@@ -3,7 +3,7 @@
 #include <psxgpu.h>
 #include <psxgte.h>
 #include <inline_c.h>
-#include "smd.h"
+#include <smd/smd.h>
 #include "disp.h"
 
 DISPENV disp;
diff --git a/examples/demos/n00bdemo/lightdemo.c b/examples/demos/n00bdemo/lightdemo.c
index 5afd5c8..26da9ef 100644
--- a/examples/demos/n00bdemo/lightdemo.c
+++ b/examples/demos/n00bdemo/lightdemo.c
@@ -3,8 +3,8 @@
 #include <psxgte.h>
 #include <psxgpu.h>
 #include <inline_c.h>
+#include <smd/smd.h>
 #include "disp.h"
-#include "smd.h"
 
 extern MATRIX lgt_colmtx;
 
@@ -14,227 +14,227 @@ extern SMD *o_lightbulb;
 void sort_overlay(int showlotl);
 
 void lightdemo() {
-	
+
 	/*
 		The point lighting demo is perhaps the most impressive part of
 		n00bDEMO. A more streamlined version of this demo where you control
 		various attributes of the light source such as position, intensity
 		and color might be made as a dedicated example program in the future.
-		
+
 		The point lighting trick is actually not that too complicated. You
 		basically calculate the distance and direction vector of two points
 		which are the light source and the vertex of a polygon.
-		
+
 		Calculating the normal whose result can later be used to calculate
 		the distance between two points is achieved with:
-		
+
 		vec_dir.vx = lgt_point.vx - pri_vert.vx;
 		vec_dir.vy = lgt_point.vy - pri_vert.vy;
 		vec_dir.vz = lgt_point.vz - pri_vert.vz;
-		
+
 		The intensity is calculated with (this might not be accurate but this
 		is faster than applying a square root):
-		
+
 		i = 4096 - ( (
-			(vec_dir.vx*vec_dir.vx) + 
-			(vec_dir.vy*vec_dir.vy) + 
+			(vec_dir.vx*vec_dir.vx) +
+			(vec_dir.vy*vec_dir.vy) +
 			(vec_dir.vz*vec_dir.vz) ) >> 7 );
-			
+
 		// Clip minimum intensity
 		if( i < 0 )
 			i = 0;
-		
+
 		This intensity value is then used to set the color of the light source
 		through the light color matrix.
-		
+
 		col_mtx.m[0][0] = i;
 		col_mtx.m[1][0] = i;
 		col_mtx.m[2][0] = i;
 		gte_SetColorMatrix( &col_mtx );
-		
+
 		The direction vector can then be used as the direction of the light
 		source. It is recommended to normalize it first to prevent possible
 		overflow related issues.
-		
+
 		VectorNormalS( &vec_dir, &vec_norm );
-		
+
 		lgt_mtx.m[0][0] = vec_norm.vx;
 		lgt_mtx.m[0][1] = vec_norm.vy;
 		lgt_mtx.m[0][2] = vec_norm.vz;
-		
+
 		gte_SetLightMatrix( &lgt_mtx );
-		
+
 		This operation is then performed for each point of a polygon to
 		achieve a nice smooth shaded point lighting effect. The macros used
 		are still the same as doing light source calculation with the GTE
 		the normal way.
-		
+
 		3D geometry still requires normal data as with most lighting
 		processing operations. 'Flat' normals (faces with a single normal
 		vector) work best on flat surfaces while 'smooth' normals (faces with
 		normals on each point) work best on round or curved surfaces.
-		
+
 	*/
-	
+
 	int			i,p_ang;
-	
+
 	SC_OT		s_ot;
-	
+
 	SVECTOR		rot;
 	VECTOR		pos;
 	SMD_PRIM	s_pri;
-	
+
 	VECTOR		l_point;
 	SVECTOR		nrm;
-	
+
 	MATRIX		lmtx,llmtx,omtx;
-	
+
 	SVECTOR orot = { 0 };
-	
+
 	int timeout = SCENE_TIME;
 
-	
+
 	// Set clear color to black
 	setRGB0( &draw, 0, 0, 0 );
-	
+
 	// Base values for the environment geometry
 	setVector( &pos, 0, 0, 600 );
 	setVector( &rot, 512, 0, 0 );
-	
+
 	// Set base tpage value for the SMD drawing routines
 	smdSetBaseTPage( 0x200 );
-	
+
 	// Set back or ambient color to black for pure darkness
 	gte_SetBackColor( 0, 0, 0 );
-	
+
 	memset( &llmtx, 0, sizeof(MATRIX) );
-	
-	
+
+
 	// demo loop
 	while( 1 ) {
-		
+
 		char buff[32];
-		
+
 		RotMatrix( &rot, &mtx );
 		TransMatrix( &mtx, &pos );
-		
+
 		rot.vy += 4;
-		
+
 		gte_SetRotMatrix( &mtx );
 		gte_SetTransMatrix( &mtx );
-		
-		
+
+
 		setVector( &l_point, (icos( p_ang )>>2)>>2, -350+(icos( p_ang<<1 )>>4), (isin( p_ang )>>2)>>2 );
 		p_ang += 16;
-		
-		
+
+
 		// Begin parsing the SMD data of the environment
 		OpenSMD( o_world );
-		
+
 		// Prototype point lighting renderer
 		while( ReadSMD( &s_pri ) ) {
-			
+
 			VECTOR	v_dir;
 			SVECTOR v_nrm;
 			VECTOR	v_sqr;
-			
+
 			int		flg;
-			
+
 			if( s_pri.prim_id.texture ) {
-				
+
 				POLY_GT4	*pri;
-				
+
 				// Perform standard rotate, translate and perspective
 				// transformation of the geometry
 				pri = (POLY_GT4*)nextpri;
-				
-				gte_ldv3( 
+
+				gte_ldv3(
 					&o_world->p_verts[s_pri.v0],
 					&o_world->p_verts[s_pri.v1],
 					&o_world->p_verts[s_pri.v2] );
-				
+
 				gte_rtpt();
-				
+
 				gte_nclip();			// Backface culling
-				
+
 				gte_stopz( &flg );
-				
+
 				if( flg < 0 )
 					continue;
-				
+
 				gte_stsxy3( &pri->x0, &pri->x1, &pri->x2 );
-				
+
 				gte_ldv0( &o_world->p_verts[s_pri.v3] );
 				gte_rtps();
-				
+
 				gte_avsz4();			// Depth sort
 				gte_stotz( &flg );
-				
+
 				if( (flg>>2) >= OT_LEN )
 					continue;
-				
+
 				gte_stsxy( &pri->x3 );
-				
-				
+
+
 				// Load base color of polygon to GTE
 				gte_ldrgb( &s_pri.r0 );
-				
+
 				// Load normal of polygon
 				gte_ldv0( &o_world->p_norms[s_pri.n0] );
-				
+
 				// Calculate the direction between the vertex of the
 				// polygon and the light source
 				v_dir.vx = l_point.vx - o_world->p_verts[s_pri.v0].vx;
 				v_dir.vy = l_point.vy - o_world->p_verts[s_pri.v0].vy;
 				v_dir.vz = l_point.vz - o_world->p_verts[s_pri.v0].vz;
-				
+
 				// Calculate distance and light intensity using square
 				i = 4096 - ( (
-					(v_dir.vx*v_dir.vx) + 
-					(v_dir.vy*v_dir.vy) + 
+					(v_dir.vx*v_dir.vx) +
+					(v_dir.vy*v_dir.vy) +
 					(v_dir.vz*v_dir.vz) ) >> 7 );
-					
+
 				// Clip minimum intensity
 				if( i < 0 )
 					i = 0;
-				
+
 				// Set intensity to color matrix
 				llmtx.m[0][0] = i;
 				llmtx.m[1][0] = i;
 				llmtx.m[2][0] = i;
 				gte_SetColorMatrix( &llmtx );
-				
+
 				// Normalize light direction and set it to light matrix
 				VectorNormalS( &v_dir, &v_nrm );
 				lmtx.m[0][0] = v_nrm.vx;
 				lmtx.m[0][1] = v_nrm.vy;
 				lmtx.m[0][2] = v_nrm.vz;
 				gte_SetLightMatrix( &lmtx );
-				
+
 				// Calculate (output is retrieved through gte_strgb)
 				gte_nccs();
-				
-				
+
+
 				// Repeat process for the next 3 vertices
 				v_dir.vx = l_point.vx - o_world->p_verts[s_pri.v1].vx;
 				v_dir.vy = l_point.vy - o_world->p_verts[s_pri.v1].vy;
 				v_dir.vz = l_point.vz - o_world->p_verts[s_pri.v1].vz;
-				
+
 				i = 4096 - ( (
-					(v_dir.vx*v_dir.vx) + 
-					(v_dir.vy*v_dir.vy) + 
+					(v_dir.vx*v_dir.vx) +
+					(v_dir.vy*v_dir.vy) +
 					(v_dir.vz*v_dir.vz) ) >> 7 );
-					
+
 				if( i < 0 )
 					i = 0;
-				
-				
+
+
 				llmtx.m[0][0] = i;
 				llmtx.m[1][0] = i;
 				llmtx.m[2][0] = i;
-				
+
 				gte_strgb( &pri->r0 );
-				
+
 				gte_SetColorMatrix( &llmtx );
 				VectorNormalS( &v_dir, &v_nrm );
 				lmtx.m[0][0] = v_nrm.vx;
@@ -242,25 +242,25 @@ void lightdemo() {
 				lmtx.m[0][2] = v_nrm.vz;
 				gte_SetLightMatrix( &lmtx );
 				gte_nccs();
-				
+
 				v_dir.vx = l_point.vx - o_world->p_verts[s_pri.v2].vx;
 				v_dir.vy = l_point.vy - o_world->p_verts[s_pri.v2].vy;
 				v_dir.vz = l_point.vz - o_world->p_verts[s_pri.v2].vz;
-				
+
 				i = 4096 - ( (
-					(v_dir.vx*v_dir.vx) + 
-					(v_dir.vy*v_dir.vy) + 
+					(v_dir.vx*v_dir.vx) +
+					(v_dir.vy*v_dir.vy) +
 					(v_dir.vz*v_dir.vz) ) >> 7 );
-					
+
 				if( i < 0 )
 					i = 0;
-				
+
 				llmtx.m[0][0] = i;
 				llmtx.m[1][0] = i;
 				llmtx.m[2][0] = i;
-				
+
 				gte_strgb( &pri->r1 );
-				
+
 				gte_SetColorMatrix( &llmtx );
 				VectorNormalS( &v_dir, &v_nrm );
 				lmtx.m[0][0] = v_nrm.vx;
@@ -268,25 +268,25 @@ void lightdemo() {
 				lmtx.m[0][2] = v_nrm.vz;
 				gte_SetLightMatrix( &lmtx );
 				gte_nccs();
-				
+
 				v_dir.vx = l_point.vx - o_world->p_verts[s_pri.v3].vx;
 				v_dir.vy = l_point.vy - o_world->p_verts[s_pri.v3].vy;
 				v_dir.vz = l_point.vz - o_world->p_verts[s_pri.v3].vz;
-				
+
 				i = 4096 - ( (
-					(v_dir.vx*v_dir.vx) + 
-					(v_dir.vy*v_dir.vy) + 
+					(v_dir.vx*v_dir.vx) +
+					(v_dir.vy*v_dir.vy) +
 					(v_dir.vz*v_dir.vz) ) >> 7 );
-					
+
 				if( i < 0 )
 					i = 0;
-				
+
 				llmtx.m[0][0] = i;
 				llmtx.m[1][0] = i;
 				llmtx.m[2][0] = i;
-				
+
 				gte_strgb( &pri->r2 );
-				
+
 				gte_SetColorMatrix( &llmtx );
 				VectorNormalS( &v_dir, &v_nrm );
 				lmtx.m[0][0] = v_nrm.vx;
@@ -294,70 +294,70 @@ void lightdemo() {
 				lmtx.m[0][2] = v_nrm.vz;
 				gte_SetLightMatrix( &lmtx );
 				gte_nccs();
-				
-				setUV4( pri, 
+
+				setUV4( pri,
 					s_pri.tu0, s_pri.tv0,
-					s_pri.tu1, s_pri.tv1, 
+					s_pri.tu1, s_pri.tv1,
 					s_pri.tu2, s_pri.tv2,
 					s_pri.tu3, s_pri.tv3 );
-				
+
 				pri->tpage = s_pri.tpage;
 				pri->clut = s_pri.clut;
-				
+
 				setPolyGT4( pri );
 				addPrim( ot[db]+(flg>>2), pri );
 				nextpri += sizeof(POLY_GT4);
-				
+
 				gte_strgb( &pri->r3 );
-				
+
 			} else {
-				
+
 				POLY_G4	*pri;
-				
+
 				pri = (POLY_G4*)nextpri;
-				
-				gte_ldv3( 
+
+				gte_ldv3(
 					&o_world->p_verts[s_pri.v0],
 					&o_world->p_verts[s_pri.v1],
 					&o_world->p_verts[s_pri.v2] );
-				
+
 				gte_rtpt();
-				
+
 				gte_nclip();
-				
+
 				gte_stopz( &flg );
-				
+
 				if( flg < 0 )
 					continue;
-				
+
 				gte_stsxy3( &pri->x0, &pri->x1, &pri->x2 );
-				
+
 				gte_ldv0( &o_world->p_verts[s_pri.v3] );
 				gte_rtps();
-				
+
 				gte_avsz4();
 				gte_stotz( &flg );
-				
+
 				if( (flg>>2) >= OT_LEN )
 					continue;
-				
+
 				gte_stsxy( &pri->x3 );
-				
+
 				gte_ldrgb( &s_pri.r0 );
 				gte_ldv0( &o_world->p_norms[s_pri.n0] );
-				
+
 				v_dir.vx = l_point.vx - o_world->p_verts[s_pri.v0].vx;
 				v_dir.vy = l_point.vy - o_world->p_verts[s_pri.v0].vy;
 				v_dir.vz = l_point.vz - o_world->p_verts[s_pri.v0].vz;
-				
+
 				i = 4096 - ( (
-					(v_dir.vx*v_dir.vx) + 
-					(v_dir.vy*v_dir.vy) + 
+					(v_dir.vx*v_dir.vx) +
+					(v_dir.vy*v_dir.vy) +
 					(v_dir.vz*v_dir.vz) ) >> 7 );
-					
+
 				if( i < 0 )
 					i = 0;
-				
+
 				llmtx.m[0][0] = i;
 				llmtx.m[1][0] = i;
 				llmtx.m[2][0] = i;
@@ -368,26 +368,26 @@ void lightdemo() {
 				lmtx.m[0][2] = v_nrm.vz;
 				gte_SetLightMatrix( &lmtx );
 				gte_nccs();
-				
-				
+
+
 				v_dir.vx = l_point.vx - o_world->p_verts[s_pri.v1].vx;
 				v_dir.vy = l_point.vy - o_world->p_verts[s_pri.v1].vy;
 				v_dir.vz = l_point.vz - o_world->p_verts[s_pri.v1].vz;
-				
+
 				i = 4096 - ( (
-					(v_dir.vx*v_dir.vx) + 
-					(v_dir.vy*v_dir.vy) + 
+					(v_dir.vx*v_dir.vx) +
+					(v_dir.vy*v_dir.vy) +
 					(v_dir.vz*v_dir.vz) ) >> 7 );
-					
+
 				if( i < 0 )
 					i = 0;
-				
+
 				llmtx.m[0][0] = i;
 				llmtx.m[1][0] = i;
 				llmtx.m[2][0] = i;
-				
+
 				gte_strgb( &pri->r0 );
-				
+
 				gte_SetColorMatrix( &llmtx );
 				VectorNormalS( &v_dir, &v_nrm );
 				lmtx.m[0][0] = v_nrm.vx;
@@ -395,25 +395,25 @@ void lightdemo() {
 				lmtx.m[0][2] = v_nrm.vz;
 				gte_SetLightMatrix( &lmtx );
 				gte_nccs();
-				
+
 				v_dir.vx = l_point.vx - o_world->p_verts[s_pri.v2].vx;
 				v_dir.vy = l_point.vy - o_world->p_verts[s_pri.v2].vy;
 				v_dir.vz = l_point.vz - o_world->p_verts[s_pri.v2].vz;
-				
+
 				i = 4096 - ( (
-					(v_dir.vx*v_dir.vx) + 
-					(v_dir.vy*v_dir.vy) + 
+					(v_dir.vx*v_dir.vx) +
+					(v_dir.vy*v_dir.vy) +
 					(v_dir.vz*v_dir.vz) ) >> 7 );
-					
+
 				if( i < 0 )
 					i = 0;
-				
+
 				llmtx.m[0][0] = i;
 				llmtx.m[1][0] = i;
 				llmtx.m[2][0] = i;
-				
+
 				gte_strgb( &pri->r1 );
-				
+
 				gte_SetColorMatrix( &llmtx );
 				VectorNormalS( &v_dir, &v_nrm );
 				lmtx.m[0][0] = v_nrm.vx;
@@ -421,25 +421,25 @@ void lightdemo() {
 				lmtx.m[0][2] = v_nrm.vz;
 				gte_SetLightMatrix( &lmtx );
 				gte_nccs();
-				
+
 				v_dir.vx = l_point.vx - o_world->p_verts[s_pri.v3].vx;
 				v_dir.vy = l_point.vy - o_world->p_verts[s_pri.v3].vy;
 				v_dir.vz = l_point.vz - o_world->p_verts[s_pri.v3].vz;
-				
+
 				i = 4096 - ( (
-					(v_dir.vx*v_dir.vx) + 
-					(v_dir.vy*v_dir.vy) + 
+					(v_dir.vx*v_dir.vx) +
+					(v_dir.vy*v_dir.vy) +
 					(v_dir.vz*v_dir.vz) ) >> 7 );
-					
+
 				if( i < 0 )
 					i = 0;
-				
+
 				llmtx.m[0][0] = i;
 				llmtx.m[1][0] = i;
 				llmtx.m[2][0] = i;
-				
+
 				gte_strgb( &pri->r2 );
-				
+
 				gte_SetColorMatrix( &llmtx );
 				VectorNormalS( &v_dir, &v_nrm );
 				lmtx.m[0][0] = v_nrm.vx;
@@ -447,49 +447,49 @@ void lightdemo() {
 				lmtx.m[0][2] = v_nrm.vz;
 				gte_SetLightMatrix( &lmtx );
 				gte_nccs();
-				
+
 				setPolyG4( pri );
 				addPrim( ot[db]+(flg>>2), pri );
-				
+
 				nextpri += sizeof(POLY_G4);
-				
+
 				gte_strgb( &pri->r3 );
-				
+
 			}
-			
+
 		}
-		
-		
+
+
 		// Sort the light bulb to represent the position of the light source
 		orot.vx += 32;
 		orot.vy += 32;
 		orot.vz += 32;
-		
+
 		RotMatrix( &orot, &omtx );
 		TransMatrix( &omtx, &l_point );
-		
+
 		CompMatrixLV( &mtx, &omtx, &mtx );
 
 		gte_SetRotMatrix( &mtx );
 		gte_SetTransMatrix( &mtx );
-		
+
 		s_ot.ot = ot[db];
 		s_ot.otlen = OT_LEN;
 		s_ot.zdiv = 2;
 		s_ot.zoff = 0;
-	
+
 		nextpri = smdSortModel( &s_ot, nextpri, o_lightbulb );
 
-		
+
 		// Sort overlay and display
 		sort_overlay( 1 );
-		
+
 		display();
-		
+
 		timeout--;
 		if( timeout < 0 )
 			break;
-		
+
 	}
-	
+
 }
\ No newline at end of file
diff --git a/examples/demos/n00bdemo/logo.c b/examples/demos/n00bdemo/logo.c
index 784c9e1..c6faf51 100644
--- a/examples/demos/n00bdemo/logo.c
+++ b/examples/demos/n00bdemo/logo.c
@@ -5,7 +5,7 @@
 #include <psxgte.h>
 #include <psxgpu.h>
 #include <inline_c.h>
-#include "smd.h"
+#include <smd/smd.h>
 #include <lzp/lzp.h>
 
 #include "disp.h"
@@ -54,29 +54,29 @@ typedef struct {
 extern NODE _end[];
 
 void DumpHeap() {
-	
+
 	NODE *n = _end;
-	
+
 	printf( "--\n" );
-	
+
 	while( 1 ) {
-		
-		printf( "B:%08p  P:%08p  N:%08p  SZ:%d  BS:%d\n", 
-			n, n->prev, n->next, n->size, 
+
+		printf( "B:%08p  P:%08p  N:%08p  SZ:%d  BS:%d\n",
+			n, n->prev, n->next, n->size,
 			((unsigned int)n->next - (unsigned int)n) );
-		
+
 		if ( !n->next )
 			break;
-		
+
 		n = (NODE*)n->next;
-		
+
 	}
-	
+
 }
 
 
 void intro() {
-	
+
 	SVECTOR		quad_coords[] = {
 		{ -100, -100, 0, 0 },
 		{  100, -100, 0, 0 },
@@ -85,18 +85,18 @@ void intro() {
 	};
 
 	PARTICLE	stars[MAX_STARS];
-	
+
 	VECTOR		mpos;
 	SVECTOR		mrot,trot;
 	SC_OT		s_ot;
-	
+
 	SMD			*o_disk, *o_star, *o_text, *o_psn00b, *o_n00blogo;
-	
+
 	int i,count = 0;
-	
+
 	int logo_scale;
 	int logo_rot;
-	
+
 	int logo_yvel;
 	int logo_ypos;
 	int logo_step;
@@ -104,10 +104,10 @@ void intro() {
 	int logo_svel;
 	int logo_count = 0;
 	int logo_tscale;
-	
+
 	int logo_n00b_spin;
 	int logo_fade = 0;
-	
+
 	struct {
 		int step;
 		int yvel;
@@ -129,7 +129,7 @@ void intro() {
 		3072,
 		0
 	};
-	
+
 	struct {
 		VECTOR pos;
 		int scale;
@@ -139,180 +139,180 @@ void intro() {
 		ONE,
 		1024
 	};
-	
-	
+
+
 	i = lzpSearchFile( "mtekdisk", lz_resources );
 	o_disk = (SMD*)malloc( lzpFileSize( lz_resources, i ) );
 	lzpUnpackFile( o_disk, lz_resources, i );
 	smdInitData( o_disk );
-	
+
 	i = lzpSearchFile( "mtektext", lz_resources );
 	o_text = (SMD*)malloc( lzpFileSize( lz_resources, i ) );
 	lzpUnpackFile( o_text, lz_resources, i );
 	smdInitData( o_text );
-	
+
 	i = lzpSearchFile( "starsprite", lz_resources );
 	o_star = (SMD*)malloc( lzpFileSize( lz_resources, i ) );
 	lzpUnpackFile( o_star, lz_resources, i );
 	smdInitData( o_star );
-	
+
 	i = lzpSearchFile( "psn00blogo", lz_resources );
 	o_psn00b = (SMD*)malloc( lzpFileSize( lz_resources, i ) );
 	lzpUnpackFile( o_psn00b, lz_resources, i );
 	smdInitData( o_psn00b );
-	
+
 	i = lzpSearchFile( "n00blogo", lz_resources );
 	o_n00blogo = (SMD*)malloc( lzpFileSize( lz_resources, i ) );
 	lzpUnpackFile( o_n00blogo, lz_resources, i );
 	smdInitData( o_n00blogo );
-	
+
 	//DumpHeap();
-	
-	
+
+
 	// Set some Scarlet global parameters
 	smdSetBaseTPage( 0x200 );
-	
+
 	// Change clear color
 	setRGB0( &draw, 255, 255, 255 );
-	
+
 	setVector( &trot, 0, 0, 0 );
-	
-	
+
+
 	// Initialize stars
 	for( i=0; i<MAX_STARS; i++ ) {
-	
+
 		stars[i].x = ONE*(-200+(rand()%400));
 		stars[i].y = ONE*(80+(rand()%100));
-		
+
 		stars[i].yvel = -(rand()%32768);
 		stars[i].xvel = -(ONE*(10-(rand()%20)));
-		
+
 		stars[i].scale = 2048+(rand()%2048);
 		stars[i].rot = ONE*(rand()%4096);
 		stars[i].rotv = ONE*(-40+(rand()%80));
-		
+
 	}
-	
+
 	logo_n00b_spin = ONE+1024;
-	
+
 	while( logo_count < 1320 ) {
-		
+
 		if( logo_count < 700 ) {
-			
+
 			if( logo_count > 360 ) {
 				mtek.spin -= mtek.spinvel;
 				mtek.spinvel += 1280;
 			}
-			
+
 			setVector( &mpos, 0, -(mtek.ypos>>12)-20, 400 );
 			setVector( &mrot, 0, mtek.spin>>12, mtek.rot>>12 );
-			
+
 			RotMatrix( &mrot, &mtx );
 			TransMatrix( &mtx, &mpos );
 			setVector( &mpos, mtek.scale, mtek.scale, 4096 );
 			ScaleMatrix( &mtx, &mpos );
-			
+
 			gte_SetRotMatrix( &mtx );
 			gte_SetTransMatrix( &mtx );
-			
+
 			nextpri = smdSortModelFlat( ot[db]+10, nextpri, o_disk );
 
-			
+
 			if( mtek.step > 0 ) {
-				
+
 				setVector( &mpos, 0, 180, 400 );
 				TransMatrix( &mtx, &mpos );
 				gte_SetTransMatrix( &mtx );
-				
+
 				nextpri = smdSortModelFlat( ot[db]+8, nextpri, o_text );
-				
+
 				if( logo_count < 480 ) {
-					
+
 					SVECTOR srot;
-					
+
 					for( i=0; i<MAX_STARS; i++ ) {
-						
+
 						setVector( &mpos, stars[i].x>>12, stars[i].y>>12, 320 );
 						setVector( &srot, 0, 0, stars[i].rot>>12 );
-						
+
 						stars[i].x += stars[i].xvel;
 						stars[i].y += stars[i].yvel;
-						
+
 						stars[i].yvel += 512;
 						stars[i].rot += stars[i].rotv;
-						
+
 						RotMatrix( &srot, &mtx );
 						TransMatrix( &mtx, &mpos );
-						
+
 						setVector( &mpos, stars[i].scale, stars[i].scale, 4096 );
 						ScaleMatrix( &mtx, &mpos );
-						
+
 						gte_SetRotMatrix( &mtx );
 						gte_SetTransMatrix( &mtx );
-						
+
 						nextpri = smdSortModelFlat( ot[db]+6, nextpri, o_star );
-						
+
 					}
-					
+
 				}
-				
+
 			}
-			
+
 			if( !mtek.step ) {
 				mtek.rot -= ONE*20;
 			} else if( mtek.step == 1 ) {
 				mtek.rot += ( -mtek.rot )>>4;
 			}
-			
-			
+
+
 			if( mtek.step < 2 ) {
 				mtek.scale	+= (ONE-mtek.scale )>>5;
 				mtek.yvel += 512;
 				mtek.ypos -= mtek.yvel;
 			}
-			
-			
+
+
 			if( ( mtek.ypos <= 0 ) && ( mtek.yvel > 0 ) ) {
 				mtek.yvel = -ONE*6;
 				mtek.step++;
 			}
-			
+
 			// Does the transition effect
 			if( logo_count > 480 ) {
-				
+
 				TILE *rect = (TILE*)nextpri;
-				
+
 				mtek.trot += 16;
-				
+
 				setTile( rect );
 				setXY0( rect, 0, 0 );
 				setWH( rect, 640, 511 );
 				setRGB0( rect, 0, 0, 0 );
 				addPrim( ot[db]+4, rect );
 				nextpri += sizeof(TILE);
-				
+
 				setVector( &mrot, 0, 0, mtek.trot );
 				setVector( &mpos, mtek.tscale, mtek.tscale, 4096 );
-				
+
 				RotMatrix( &mrot, &mtx );
 				ScaleMatrix( &mtx, &mpos );
-				
+
 				gte_SetRotMatrix( &mtx );
-				
+
 				for( i=0; i<6; i++ ) {
-					
+
 					MASKP_F4 *pol4 = (MASKP_F4*)nextpri;
-					
+
 					setVector( &mpos, -100+(100*(i%3)), -60+(120*(i/3)), 160 );
 					TransMatrix( &mtx, &mpos );
 					gte_SetTransMatrix( &mtx );
-					
+
 					pol4->tag = 0x08000000;
 					pol4->tpage = 0xe1000020;
 					pol4->mask1 = 0xe6000001;
 					pol4->mask2 = 0xe6000002;
 					pol4->code = 0x2A;
-					
+
 					gte_ldv3( &quad_coords[0], &quad_coords[1], &quad_coords[2] );
 					gte_rtpt();
 					gte_stsxy0( &pol4->x0 );
@@ -320,111 +320,111 @@ void intro() {
 					gte_rtps();
 					gte_stsxy3( &pol4->x1, &pol4->x2, &pol4->x3 );
 					setRGB0( pol4, 0, 0, 0 );
-					
+
 					addPrim( ot[db]+4, pol4 );
 					nextpri += sizeof(MASKP_F4);
-				
+
 				}
-				
+
 				mtek.tscale += ( -mtek.tscale )>>6;
-				
+
 			}
-			
+
 		} else {
-			
+
 			setRGB0( &draw, 0, 0, 0 );
-			
+
 		}
-		
+
 		if( logo_count >= 480 ) {
-			
+
 			if( logo_count > 840 ) {
-			
+
 				psn00b.pos.vx += ((ONE*-450)-psn00b.pos.vx)>>4;
 				psn00b.pos.vy += ((ONE*350)-psn00b.pos.vy)>>4;
-				
+
 				setVector( &mpos, 0, 0, 600 );
 				mpos.vx = psn00b.pos.vx>>12;
 				mpos.vy = psn00b.pos.vy>>12;
-				
+
 				psn00b.scale += ( 1536-psn00b.scale )>>4;
-				
+
 				setVector( &mrot, 0, 0, 0 );
-			
+
 				RotMatrix( &mrot, &mtx );
 				TransMatrix( &mtx, &mpos );
-				
+
 				setVector( &mpos, psn00b.scale, psn00b.scale, ONE );
-				
+
 				ScaleMatrix( &mtx, &mpos );
-				
+
 			} else {
-				
+
 				setVector( &mpos, 0, 0, 600 );
 				setVector( &mrot, 0, logo_n00b_spin, 0 );
-			
+
 				RotMatrix( &mrot, &mtx );
 				TransMatrix( &mtx, &mpos );
-				
+
 				logo_n00b_spin += ( -logo_n00b_spin )>>6;
-				
+
 			}
-			
+
 			gte_SetRotMatrix( &mtx );
 			gte_SetTransMatrix( &mtx );
-			
+
 			nextpri = smdSortModelFlat( ot[db]+3, nextpri, o_psn00b );
-			
+
 			if( logo_count > 900 ) {
-				
+
 				setVector( &mpos, 0, 0, 250 );
 				setVector( &mrot, psn00b.scarlet_spin, 0, 0 );
-				
+
 				RotMatrix( &mrot, &mtx );
 				TransMatrix( &mtx, &mpos );
-				
+
 				gte_SetRotMatrix( &mtx );
 				gte_SetTransMatrix( &mtx );
-				
+
 				nextpri = smdSortModelFlat( ot[db]+3, nextpri, o_n00blogo );
-				
+
 				psn00b.scarlet_spin += ( -psn00b.scarlet_spin )>>6;
-				
+
 			}
-			
+
 		}
-		
+
 		if( logo_count > 1200 ) {
-			
+
 			FADERECT *fade = (FADERECT*)nextpri;
-			
+
 			fade->tag = 0x04000000;
 			fade->tpage = 0xe1000040;
 			fade->code = 0x62;
-			
+
 			setRGB0( fade, logo_fade, logo_fade, logo_fade );
 			setXY0( fade, 0, 0 );
 			setWH( fade, 640, 480 );
 			addPrim( ot[db]+2, fade );
-			
+
 			if( logo_fade < 250 ) {
 				logo_fade += 4;
 			}
-			
+
 			nextpri += sizeof(FADERECT);
-			
+
 		}
-		
+
 		display();
-		
+
 		logo_count++;
-		
+
 	}
-	
+
 	free( o_disk );
 	free( o_text );
 	free( o_star );
 	free( o_psn00b );
 	free( o_n00blogo );
-	
+
 }
diff --git a/examples/demos/n00bdemo/main.c b/examples/demos/n00bdemo/main.c
index 55dbbc4..d607495 100644
--- a/examples/demos/n00bdemo/main.c
+++ b/examples/demos/n00bdemo/main.c
@@ -1,4 +1,4 @@
-/* 
+/*
  * LibPSn00b Example Programs
  *
  * n00bDEMO Source Code
@@ -41,8 +41,8 @@
 #include <string.h>
 #include <lzp/lzp.h>
 #include <lzp/lzqlp.h>
+#include <smd/smd.h>
 
-#include "smd.h"
 #include "data.h"
 #include "disp.h"
 #include "logo.h"
@@ -119,16 +119,16 @@ void UploadTIM(TIM_IMAGE *tim) {
 	/*
 		Very simple texture upload function
 	*/
-	
+
 	LoadImage( tim->prect, tim->paddr );
-	
+
 	if( tim->mode & 0x8 )
 		LoadImage( tim->crect, tim->caddr );
-	
+
 }
 
 void loadTextures() {
-	
+
 	/*
 		Unpack textures from an embedded LZP archive and upload them to VRAM.
 	*/
@@ -136,241 +136,241 @@ void loadTextures() {
 	int *ttim,j;
 	QLP_HEAD *tex_buff;
 	TIM_IMAGE tim;
-	
+
 	i = lzpSearchFile( "textures", lz_resources );
 	tex_buff = (QLP_HEAD*)malloc( lzpFileSize( lz_resources, i ) );
 	lzpUnpackFile( tex_buff, lz_resources, i );
 
-	
+
 	for( j=0; j<qlpFileCount( tex_buff )-4; j++ ) {
-		
+
 		if( !GetTimInfo( (u_long*)qlpFileAddr( j, tex_buff ), &tim ) ) {
-		
+
 			UploadTIM( &tim );
-			
+
 		}
-		
+
 	}
-	
-	
-	GetTimInfo( (u_long*)qlpFileAddr( 
+
+
+	GetTimInfo( (u_long*)qlpFileAddr(
 		qlpFindFile( "n00blogo", tex_buff ), tex_buff ), &tim );
-	
+
 	UploadTIM( &tim );
-	
+
 	psn00b_tpage = getTPage( 0, 0, tim.prect->x, tim.prect->y )|0x200;
 	setSprt( &psn00b_sprite );
 	setClut( &psn00b_sprite, tim.crect->x, tim.crect->y );
 	setWH( &psn00b_sprite, tim.prect->w<<2, tim.prect->h );
 	setUV0( &psn00b_sprite, (tim.prect->x%64)<<2, tim.prect->y );
 	setRGB0( &psn00b_sprite, 128, 128, 128 );
-	
-	
-	GetTimInfo( (u_long*)qlpFileAddr( 
+
+
+	GetTimInfo( (u_long*)qlpFileAddr(
 		qlpFindFile( "lamelotl", tex_buff ), tex_buff ), &tim );
-	
+
 	UploadTIM( &tim );
-	
+
 	lamelotl_tpage = getTPage( 0, 0, tim.prect->x, tim.prect->y )|0x200;
 	setSprt( &llotl_sprite );
 	setClut( &llotl_sprite, tim.crect->x, tim.crect->y );
 	setWH( &llotl_sprite, tim.prect->w<<2, tim.prect->h );
 	setUV0( &llotl_sprite, 0, 0 );
 	setRGB0( &llotl_sprite, 128, 128, 128 );
-	
-	
-	GetTimInfo( (u_long*)qlpFileAddr( 
+
+
+	GetTimInfo( (u_long*)qlpFileAddr(
 		qlpFindFile( "celmap", tex_buff ), tex_buff ), &tim );
-		
+
 	UploadTIM( &tim );
-	
+
 	smdSetCelTex( getTPage( 0, 2, tim.prect->x, tim.prect->y )|0x200,
 		getClut( tim.crect->x, tim.crect->y ) );
 	smdSetCelParam( 3, 3, 0x4f4f4f );
-	
-	
-	GetTimInfo( (u_long*)qlpFileAddr( 
+
+
+	GetTimInfo( (u_long*)qlpFileAddr(
 		qlpFindFile( "font", tex_buff ), tex_buff ), &tim );
-		
+
 	UploadTIM( &tim );
-	
+
 	font_tpage	= getTPage( 0, 1, tim.prect->x, tim.prect->y )|0x200;
 	font_clut	= getClut( tim.crect->x, tim.crect->y );
-	
+
 	free( tex_buff );
 }
-	
+
 void unpackModels() {
-	
+
 	/*
 		Unpack model data from an embedded LZP archive.
 	*/
-	
+
 	int i;
-	
+
 	i = lzpSearchFile( "rbowshade", lz_resources );
 	o_rbowshade = (SMD*)malloc( lzpFileSize( lz_resources, i ) );
 	lzpUnpackFile( o_rbowshade, lz_resources, i );
 	smdInitData( o_rbowshade );
-	
+
 	i = lzpSearchFile( "bungirl", lz_resources );
 	o_bungirl = (SMD*)malloc( lzpFileSize( lz_resources, i ) );
 	lzpUnpackFile( o_bungirl, lz_resources, i );
 	smdInitData( o_bungirl );
-	
+
 	i = lzpSearchFile( "lightworld", lz_resources );
 	o_world = (SMD*)malloc( lzpFileSize( lz_resources, i ) );
 	lzpUnpackFile( o_world, lz_resources, i );
 	smdInitData( o_world );
-	
+
 	i = lzpSearchFile( "lightbulb", lz_resources );
 	o_lightbulb = (SMD*)malloc( lzpFileSize( lz_resources, i ) );
 	lzpUnpackFile( o_lightbulb, lz_resources, i );
 	smdInitData( o_lightbulb );
-	
+
 	i = lzpSearchFile( "timerift", lz_resources );
 	o_timerift = (SMD*)malloc( lzpFileSize( lz_resources, i ) );
 	lzpUnpackFile( o_timerift, lz_resources, i );
 	smdInitData( o_timerift );
-	
+
 	i = lzpSearchFile( "starmask", lz_resources );
 	o_star = (SMD*)malloc( lzpFileSize( lz_resources, i ) );
 	lzpUnpackFile( o_star, lz_resources, i );
 	smdInitData( o_star );
-	
+
 	i = lzpSearchFile( "hatkid", lz_resources );
 	o_hatkid = (SMD*)malloc( lzpFileSize( lz_resources, i ) );
 	lzpUnpackFile( o_hatkid, lz_resources, i );
 	smdInitData( o_hatkid );
-	
+
 }
 
 void init() {
 	// Init display
 	initDisplay();
 	timerSetup();
-	
+
 	FntLoad( 960, 0 );
-	
+
 	// Just to remove all sound
 	SpuInit();
-	
+
 	// Load all textures
 	loadTextures();
-	
+
 }
 
 // Bungirl stuff
 void bungirldemo() {
-	
+
 	/*
 	The bunny girl demo is just a simple demonstration of rendering a 1390
 	polygon 3D model of a bunny girl with lighting using SMD drawing routines
 	from Scarlet Engine.
 	*/
-	
+
 	SC_OT s_ot;
-	
+
 	MATRIX lmtx;
-	
+
 	SVECTOR rot,brot,srot;
 	VECTOR pos;
-	
+
 	int timeout = SCENE_TIME;
-	
+
 	// Set clear color
 	setRGB0( &draw, 63, 0, 127 );
-	
+
 	// Set tpage base value for SMD drawing routines
 	smdSetBaseTPage( 0x200 );
-	
+
 	// Set initial rotation values
 	setVector( &rot, 192, 0, 0 );
 	setVector( &brot, 0, 0, 128 );
 	setVector( &srot, 0, 0, 0 );
-	
+
 	// Set lighting color matrix
 	gte_SetColorMatrix( &lgt_colmtx );
-	
+
 	// Set 'backside' or ambient light color
 	gte_SetBackColor( 113, 113, 113 );
-	
-	
+
+
 	// Demo loop
 	while( 1 ) {
-		
+
 		// Bungirl position
 		setVector( &pos, 0, 50, 350 );
-		
+
 		TransMatrix( &mtx, &pos );
 		RotMatrix( &brot, &mtx );
-		
+
 		// Calculate light matrix
 		lmtx.m[0][0] = isin( -brot.vy<<2 );
 		lmtx.m[0][1] = -2048;
 		lmtx.m[0][2] = icos( -brot.vy<<2 );
-		
+
 		MulMatrix0( &lmtx, &mtx, &lmtx );
-		
+
 		// Set matrices
 		gte_SetRotMatrix( &mtx );
 		gte_SetTransMatrix( &mtx );
 		gte_SetLightMatrix( &lmtx );
-		
+
 		// Sort the bungirl model
 		s_ot.ot		= ot[db];
 		s_ot.otlen	= OT_LEN;
 		s_ot.zdiv	= 1;
 		s_ot.zoff	= 0;
 		nextpri = smdSortModel( &s_ot, nextpri, o_bungirl );
-		
-		
+
+
 		// Sort the rotating rainbow background
 		setVector( &pos, 0, 0, 200 );
-		
+
 		TransMatrix( &mtx, &pos );
 		RotMatrix( &srot, &mtx );
-		
+
 		gte_SetRotMatrix( &mtx );
 		gte_SetTransMatrix( &mtx );
 		nextpri = smdSortModelFlat( ot[db]+(OT_LEN-1), nextpri, o_rbowshade );
-		
+
 		brot.vy += 8;
 		srot.vz += 4;
-		
-		
+
+
 		// Sort overlay then display
 		sort_overlay( 0 );
-		
+
 		display();
-		
+
 		timeout--;
 		if( timeout < 0 )
 			break;
-		
+
 	}
 
 }
 
 // Stencil effect stuff
 void stencilstuff() {
-	
-	/* 
+
+	/*
 	The stencil demo is achieved by utilizing the mask bit setting
 	primitive GP0(E6h). The structure of this primitive is defined as
 	DR_STP initialized and set by setDrawStp().
-	
+
 	The DR_STP primitive controls mask bit operations for drawing
 	primitives such as setting mask bits on every pixel drawn or mask
 	bit test where pixels won't be drawn on pixels with the mask bit set.
 	It applies to most graphics drawing primitives except VRAM fill.
 	The mask bits are stored in the 16th bit of each pixel drawn.
-	
+
 	The semi-transparency bits of a texture always carry over as mask
 	bits in textured primitives. The only way to clear mask bits is by
 	using VRAM fill commands or drawing primitives with the set mask
 	bit operation disabled.
-	
+
 	The stencil effect featured in this demo is achieved by enabling set
 	mask bit with DR_STP, drawing semi-transparent primitives using
 	additive blending but color is all zero to make it completely invisible
@@ -380,96 +380,96 @@ void stencilstuff() {
 	drawing the scene that will be 'below' the mask layer.
 	*/
 	int timeout = SCENE_TIME;
-	
+
 	int spin=0;
-	
+
 	DR_STP *mask;
 	TILE *rect;
-	
+
 	SC_OT s_ot;
-	
+
 	SVECTOR rot;
 	SVECTOR srot;
 	VECTOR pos;
-	
+
 	// Set clear color
 	setRGB0( &draw, 127, 0, 63 );
-	
+
 	// Set tpage base value for SMD drawing routines
 	smdSetBaseTPage( 0x200 );
-	
+
 	// Base rotation coordinates
 	setVector( &rot, 0, 0, 0 );
 	setVector( &srot, 0, 0, 0 );
-	
+
 	// Set perspective
 	gte_SetGeomScreen( 320 );
-	
-	
+
+
 	// Demo loop
 	while( timeout > 0 ) {
-	
-	
+
+
 		// Draw the timerift background
 		setVector( &rot, 0, spin, 0 );
 		setVector( &pos, 0, 0, 0 );
-		
+
 		TransMatrix( &mtx, &pos );
 		RotMatrix( &rot, &mtx );
-		
+
 		gte_SetRotMatrix( &mtx );
 		gte_SetTransMatrix( &mtx );
-		
+
 		s_ot.ot		= ot[db];
 		s_ot.otlen 	= 32;
 		s_ot.zdiv	= 2;
 		s_ot.zoff	= 20;
-		
+
 		nextpri = smdSortModelFlat( ot[db]+(OT_LEN-1), nextpri, o_timerift );
-		
-		
+
+
 		// Sort mask primitive that enables setting mask bits
 		mask = (DR_STP*)nextpri;
 		setDrawStp( mask, 1, 0 );
 		addPrim( ot[db]+20, mask );
 		nextpri += sizeof(DR_STP);
-		
-		
+
+
 		// Sort the stars
 		setVector( &rot, 0, 0, spin<<2 );
-		setVector( &pos, 
-			(isin(spin<<2)*icos(spin))>>16, 
+		setVector( &pos,
+			(isin(spin<<2)*icos(spin))>>16,
 			(icos(spin<<2)*isin(spin))>>16, 400 );
-		
+
 		TransMatrix( &mtx, &pos );
 		RotMatrix( &rot, &mtx );
-		
+
 		gte_SetRotMatrix( &mtx );
 		gte_SetTransMatrix( &mtx );
-		
+
 		nextpri = smdSortModelFlat( ot[db]+19, nextpri, o_star );
-		
+
 		setVector( &rot, 0, 0, -spin<<2 );
-		setVector( &pos, 
-			(isin(-spin<<2)*icos(-spin))>>16, 
+		setVector( &pos,
+			(isin(-spin<<2)*icos(-spin))>>16,
 			(icos(-spin<<2)*isin(-spin))>>16, 400 );
-		
+
 		TransMatrix( &mtx, &pos );
 		RotMatrix( &rot, &mtx );
-		
+
 		gte_SetRotMatrix( &mtx );
 		gte_SetTransMatrix( &mtx );
-		
+
 		nextpri = smdSortModelFlat( ot[db]+19, nextpri, o_star );
-		
-		
+
+
 		// Sort mask primitive that enables mask bit test
 		mask = (DR_STP*)nextpri;
 		setDrawStp( mask, 0, 1 );
 		addPrim( ot[db]+18, mask );
 		nextpri += sizeof(DR_STP);
 
-		
+
 		// Sort rectangle that fills the screen
 		rect = (TILE*)nextpri;
 		setTile( rect );
@@ -478,119 +478,119 @@ void stencilstuff() {
 		setRGB0( rect, 128, 0, 255 );
 		addPrim( ot[db]+17, rect );
 		nextpri += sizeof(TILE);
-		
-		
+
+
 		// Clear all mask settings
 		mask = (DR_STP*)nextpri;
 		setDrawStp( mask, 0, 0 );
 		addPrim( ot[db]+15, mask );
 		nextpri += sizeof(DR_STP);
-		
+
 
 		// Sort overlay then display
 		sort_overlay( 0 );
-		
+
 		display();
-		
+
 		spin += 4;
 		timeout--;
 	}
-	
+
 }
 
 // Orbiting around cel-shaded hatkid stuff
 void hatkidstuff() {
-	
+
 	/*
 		The cel-shading effect works in a similar manner as conventional
 		per-vertex light source calculation except the color values are used
-		in a different manner. The resulting color values are divided down 
+		in a different manner. The resulting color values are divided down
 		using bit shifts to fit as texture coordinates (usually within the
 		range of 0-31) which are then used to map to a shading map which is a
 		simple texture of 3 to 4 shading levels. This technique is very similar
 		to how cel-shading is achieved on the Gamecube except all the vertex to
 		texture coordinate conversion is all done on the GX itself.
-		
+
 		To achieve this effect on untextured and textured polygons seamlessly
 		the shading map is drawn over the original unshaded polygons of the
 		model as semi-transparent polygons with subtractive blending. The
 		shading map had to be inverted for this effect to work properly since
 		blending is subtractive and not multiplicative after all.
 	*/
-	
+
 	int timeout = SCENE_TIME;
 	int spin=0;
-	
+
 	SC_OT s_ot;
-	
+
 	MATRIX lmtx;
-	
+
 	SVECTOR rot;
 	VECTOR pos;
-	
+
 	// Set clear color
 	setRGB0( &draw, 127, 0, 63 );
-	
+
 	// Sets base TPage value for SMD drawing routines
 	smdSetBaseTPage( 0x200 );
-	
+
 	setVector( &rot, 0, 0, 0 );
-	
+
 	// Set light color matrix
 	gte_SetColorMatrix( &lgt_colmtx );
-	
+
 	// Set back/ambient color to black (required for good cel-shading)
 	gte_SetBackColor( 0, 0, 0 );
-	
+
 	gte_SetGeomScreen( 320 );
-	
+
 	// Demo loop
 	while( timeout > 0 ) {
-		
+
 		// Sort 3D timerift background
 		setVector( &rot, spin, spin, spin>>1 );
 		setVector( &pos, 0, 0, 0 );
-		
+
 		TransMatrix( &mtx, &pos );
 		RotMatrix( &rot, &mtx );
-		
+
 		// Fixed light source direction
 		lmtx.m[0][0] = -2048;
 		lmtx.m[0][1] = -2048;
 		lmtx.m[0][2] = -2048;
-		
+
 		// Multiply by rotation matrix to make it relative to the camera rotation
 		MulMatrix0( &lmtx, &mtx, &lmtx );
-		
+
 		gte_SetLightMatrix( &lmtx );
 		gte_SetRotMatrix( &mtx );
 		gte_SetTransMatrix( &mtx );
-		
+
 		nextpri = smdSortModelFlat( ot[db]+(OT_LEN-1), nextpri, o_timerift );
-		
-		
+
+
 		// Sort the cel-shaded hatkid
 		setVector( &pos, 0, 0, 600 );
 		TransMatrix( &mtx, &pos );
 		gte_SetTransMatrix( &mtx );
-		
+
 		s_ot.ot		= ot[db];
 		s_ot.otlen 	= 250;
 		s_ot.zdiv	= 1;
 		s_ot.zoff	= 0;
-		
+
 		nextpri = smdSortModelCel( &s_ot, nextpri, o_hatkid );
 
-		
+
 		// Sort overlay then display
 		sort_overlay( 1 );
 		display();
-		
+
 		spin += 8;
 		timeout--;
-		
+
 	}
-	
+
 }
 
 // Plasma stuff
@@ -598,101 +598,101 @@ void genPlasma(char *out, int count);
 char *sortPlasma(u_long *ot, char *pri, char *map);
 
 void plasmastuff() {
-	
+
 	DR_TPAGE* tp;
 	char plasbuff[1271];
-	
+
 	int pcount = 0;
 	int timeout = SCENE_TIME;
-	
+
 	while( timeout > 0 ) {
-		
+
 		genPlasma( plasbuff, pcount );
 		nextpri = sortPlasma( ot[db]+1, nextpri, plasbuff );
-	
+
 		sort_overlay( 1 );
-		
+
 		display();
 		timeout--;
 		pcount++;
-		
+
 	}
-	
+
 }
 
 // Simple stripe transition effect
 void transition() {
-	
+
 	int count = 0;
 	int bheight[16] = { 0 };
-	
+
 	TILE *tile = (TILE*)nextpri;
 	draw.isbg = 0;
-	
+
 	while( 1 ) {
-		
+
 		int comp = 0;
-		
+
 		for( int i=0; i<16; i++ ) {
-			
+
 			if( bheight[i] > 0 ) {
-				
-				
+
+
 				setTile( tile );
 				setXY0( tile, 0, 32*i );
 				setRGB0( tile, 0, 151, 255 );
 				setWH( tile, 640, bheight[i] );
 				addPrim( ot[db], tile );
 				tile++;
-				
+
 				if( bheight[i] < 32 )
 					bheight[i]++;
 				else
 					comp++;
-				
+
 			}
-			
+
 		}
-		
+
 		if( bheight[count>>1] == 0 )
 			bheight[count>>1] = 1;
 		display();
 		count++;
-		
+
 		if( comp >= 16 )
 			break;
 	}
-	
+
 	DrawSync(0);
-	
+
 	draw.isbg = 1;
-	
+
 }
 
 int main(int argc, const char *argv[]) {
-	
+
 	// Init
 	init();
-	
+
 	// Do transition
 	transition();
-	
+
 	// Do Meido-Tek, PSn00bSDK and n00bDEMO logo intros
 	intro();
-	
+
 	// Unpack model data
 	unpackModels();
-	
+
 	// Demo sequence loop
 	timer_counter = 100;
 	while( 1 ) {
-		
+
 		lightdemo();
 		bungirldemo();
 		stencilstuff();
 		hatkidstuff();
 		plasmastuff();
-		
+
 	}
 
 	return 0;
diff --git a/examples/demos/n00bdemo/smd.h b/examples/demos/n00bdemo/smd.h
deleted file mode 100644
index 4306534..0000000
--- a/examples/demos/n00bdemo/smd.h
+++ /dev/null
@@ -1,68 +0,0 @@
-#ifndef _SMD_H
-#define _SMD_H
-
-typedef struct {
-	u_long			*ot;
-	short			otlen;
-	unsigned char	zdiv,zoff;
-} SC_OT;
-
-typedef struct {
-	char			id[3];
-	unsigned char	version;
-	unsigned short	flags;
-	unsigned short	n_verts;
-	unsigned short	n_norms;
-	unsigned short	n_prims;
-	SVECTOR			*p_verts;
-	SVECTOR			*p_norms;
-	void			*p_prims;
-} SMD;
-
-typedef struct {
-	unsigned char type:2;
-	unsigned char l_type:2;
-	unsigned char c_type:1;
-    unsigned char texture:1;
-	unsigned char blend:2;
-	unsigned char zoff:4;
-	unsigned char nocull:1;
-	unsigned char mask:1;
-	unsigned char texwin:2;
-	unsigned char texoff:2;
-	unsigned char reserved:6;
-	unsigned char len;
-} SMD_PRI_TYPE;
-
-typedef struct {
-	SMD_PRI_TYPE	prim_id;
-	unsigned short	v0,v1,v2,v3;		// Vertex indices
-	unsigned short	n0,n1,n2,n3;		// Normal indices
-	unsigned char	r0,g0,b0,code;		// RGB0
-	unsigned char	r1,g1,b1,p0;		// RGB1
-	unsigned char	r2,g2,b2,p1;		// RGB2
-	unsigned char	r3,g3,b3,p2;		// RGB3
-	unsigned char	tu0,tv0;
-	unsigned char	tu1,tv1;
-	unsigned char	tu2,tv2;
-	unsigned char	tu3,tv3;
-	unsigned short	tpage,clut;
-} SMD_PRIM;
-
-
-int OpenSMD(void *smd);
-SMD_PRIM *ReadSMD(SMD_PRIM *pri);
-
-void scSetClipRect(int x0, int y0, int x1, int y1);
-
-SMD *smdInitData(void *data);
-void smdSetBaseTPage(unsigned short tpage);
-
-char *smdSortModel(SC_OT *ot, char* pribuff, SMD *smd);
-char *smdSortModelFlat(u_long *ot, char* pribuff, SMD *smd);
-
-void smdSetCelTex(unsigned short tpage, unsigned short clut);
-void smdSetCelParam(int udiv, int vdiv, unsigned int col);
-char *smdSortModelCel(SC_OT *ot, char* pribuff, SMD *smd);
-
-#endif
\ No newline at end of file
diff --git a/libpsn00b/cmake/internal_setup.cmake b/libpsn00b/cmake/internal_setup.cmake
index e24595a..c3bb2dc 100644
--- a/libpsn00b/cmake/internal_setup.cmake
+++ b/libpsn00b/cmake/internal_setup.cmake
@@ -51,6 +51,7 @@ set(
 		psxsio
 		psxetc
 		psxapi
+		smd
 		lzp
 		c
 )
diff --git a/libpsn00b/include/psxapi.h b/libpsn00b/include/psxapi.h
index 5bb2b6f..67fa51a 100644
--- a/libpsn00b/include/psxapi.h
+++ b/libpsn00b/include/psxapi.h
@@ -310,7 +310,7 @@ int LoadExec(const char *path, int argc, const char **argv);
 void FlushCache(void);
 
 void ResetEntryInt(void);
-void HookEntryInt(jmp_buf buf);
+void HookEntryInt(const JumpBuffer *buf);
 void ReturnFromException(void);
 
 int SetConf(int evcb, int tcb, uint32_t sp);
diff --git a/libpsn00b/include/smd/smd.h b/libpsn00b/include/smd/smd.h
new file mode 100644
index 0000000..658f7fa
--- /dev/null
+++ b/libpsn00b/include/smd/smd.h
@@ -0,0 +1,85 @@
+/*
+ * PSn00bSDK .SMD model parser library
+ * (C) 2019-2023 Lameguy64, spicyjpeg - MPL licensed
+ */
+
+#pragma once
+
+#include <stdint.h>
+#include <psxgte.h>
+
+/* Structure definitions */
+
+typedef struct {
+	uint32_t *ot;
+	int16_t  otlen;
+	uint8_t  zdiv,zoff;
+} SC_OT;
+
+typedef struct {
+	char     id[3];
+	uint8_t  version;
+	uint16_t flags;
+	uint16_t n_verts;
+	uint16_t n_norms;
+	uint16_t n_prims;
+	SVECTOR  *p_verts;
+	SVECTOR  *p_norms;
+	void     *p_prims;
+} SMD;
+
+typedef struct {
+	uint8_t type     : 2;
+	uint8_t l_type   : 2;
+	uint8_t c_type   : 1;
+    uint8_t texture  : 1;
+	uint8_t blend    : 2;
+	uint8_t zoff     : 4;
+	uint8_t nocull   : 1;
+	uint8_t mask     : 1;
+	uint8_t texwin   : 2;
+	uint8_t texoff   : 2;
+	uint8_t reserved : 6;
+	uint8_t len;
+} SMD_PRI_TYPE;
+
+typedef struct {
+	SMD_PRI_TYPE prim_id;
+
+	uint16_t v0,v1,v2,v3;		// Vertex indices
+	uint16_t n0,n1,n2,n3;		// Normal indices
+	uint8_t  r0,g0,b0,code;		// RGB0
+	uint8_t  r1,g1,b1,p0;		// RGB1
+	uint8_t  r2,g2,b2,p1;		// RGB2
+	uint8_t  r3,g3,b3,p2;		// RGB3
+	uint8_t  tu0,tv0;
+	uint8_t  tu1,tv1;
+	uint8_t  tu2,tv2;
+	uint8_t  tu3,tv3;
+	uint16_t tpage,clut;
+} SMD_PRIM;
+
+/* API */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+int OpenSMD(const void *smd);
+SMD_PRIM *ReadSMD(SMD_PRIM *pri);
+
+void scSetClipRect(int x0, int y0, int x1, int y1);
+
+SMD *smdInitData(const void *data);
+void smdSetBaseTPage(uint16_t tpage);
+
+uint8_t *smdSortModel(SC_OT *ot, uint8_t *pribuff, SMD *smd);
+uint8_t *smdSortModelFlat(uint32_t *ot, uint8_t *pribuff, SMD *smd);
+
+void smdSetCelTex(uint16_t tpage, uint16_t clut);
+void smdSetCelParam(int udiv, int vdiv, unsigned int col);
+uint8_t *smdSortModelCel(SC_OT *ot, uint8_t *pribuff, SMD *smd);
+
+#ifdef __cplusplus
+}
+#endif
diff --git a/examples/demos/n00bdemo/smd_s.h b/libpsn00b/include/smd/smd_s.inc
index 4dca4dd..827e9e3 100644
--- a/examples/demos/n00bdemo/smd_s.h
+++ b/libpsn00b/include/smd/smd_s.inc
@@ -1,3 +1,6 @@
+# PSn00bSDK .SMD model parser library
+# (C) 2019-2023 Lameguy64, spicyjpeg - MPL licensed
+
 .set OT_ADDR,			0
 .set OT_LEN,			4
 .set OT_ZDIV,			8
@@ -332,5 +335,5 @@
 	nop
 	ori		$s3, CLIP_BOTTOM
 .no_clip_b_y3_q:
-	
+
 .endm
diff --git a/examples/demos/n00bdemo/smd.s b/libpsn00b/smd/smd.s
index d25f760..3c87a5e 100644
--- a/examples/demos/n00bdemo/smd.s
+++ b/libpsn00b/smd/smd.s
@@ -1,35 +1,37 @@
+# PSn00bSDK .SMD model parser library
+# (C) 2019-2023 Lameguy64, spicyjpeg - MPL licensed
+
 .set noreorder
 
 .include "gtereg.inc"
 .include "inline_s.inc"
-.include "smd_s.h"
-
-
-.section .text
+.include "smd/smd_s.inc"
 
 # Currently does not do header checks
+.section .text.smdInitData, "ax", @progbits
 .global smdInitData
 .type smdInitData, @function
+
 smdInitData:
-	
 	lw		$a1, SMD_HEAD_PVERTS($a0)	# Initialize header pointers
 	lw		$a2, SMD_HEAD_PNORMS($a0)
 	lw		$a3, SMD_HEAD_PPRIMS($a0)
-	
+
 	addu	$a1, $a0
 	addu	$a2, $a0
 	addu	$a3, $a0
-	
+
 	sw		$a1, SMD_HEAD_PVERTS($a0)
 	sw		$a2, SMD_HEAD_PNORMS($a0)
 	sw		$a3, SMD_HEAD_PPRIMS($a0)
-	
+
 	jr		$ra
 	move	$v0, $a0
-	
-	
+
+.section .text.scSetClipRect, "ax", @progbits
 .global scSetClipRect
 .type scSetClipRect, @function
+
 scSetClipRect:
 	sll		$a1, 16
 	or		$a0, $a1
@@ -39,50 +41,51 @@ scSetClipRect:
 	sw		$a0, 0( $a1 )
 	jr		$ra
 	sw		$a2, 4( $a1 )
-	
-	
+
+.section .text.smdSetBaseTPage, "ax", @progbits
 .global smdSetBaseTPage
 .type smdSetBaseTPage, @function
+
 smdSetBaseTPage:
 	la		$v0, _smd_tpage_base
 	jr		$ra
 	sw		$a0, 0($v0)
-	
-	
+
+.section .text.smdSortModel, "ax", @progbits
 .global smdSortModel
 .type smdSortModel, @function
+
 smdSortModel:
-	
 	# a0 - Pointer SC_OT structure
 	# a1 - Pointer to next primitive
 	# a2 - Pointer to SMD data address
 	# v0 - New pointer of primitive buffer (return)
-	
+
 	addiu	$sp, -16
 	sw		$s0, 0($sp)
 	sw		$s1, 4($sp)
 	sw		$s2, 8($sp)
 	sw		$s3, 12($sp)
-	
+
 	la		$v0, _sc_clip
 	lw		$t8, 0($v0)
 	lw		$t9, 4($v0)
-	
+
 	lw		$t0, OT_LEN($a0)
 	lw		$a0, OT_ADDR($a0)
 	lw		$t1, SMD_HEAD_PVERTS($a2)
 	lw		$t2, SMD_HEAD_PNORMS($a2)
 	lw		$t3, SMD_HEAD_PPRIMS($a2)
-	
+
 .sort_loop:
 
 	nop
 	lw		$a3, 0($t3)				# Get primitive ID word
 	move	$t4, $t3
-	
+
 	beqz	$a3, .exit				# Check if terminator (just zero)
 	addiu	$t4, 4
-	
+
 	lhu		$t5, 0( $t4 )			# Load vertices
 	lhu		$t6, 2( $t4 )
 	lhu		$t7, 4( $t4 )
@@ -98,38 +101,38 @@ smdSortModel:
 	lwc2	C2_VZ1 , 4( $t6 )
 	lwc2	C2_VXY2, 0( $t7 )
 	lwc2	C2_VZ2 , 4( $t7 )
-	
+
 	srl		$v1, $a3, 24			# Get primitive size
 	addu	$t3, $v1				# Step main pointer to next primitive
-	
+
 	RTPT
-	
+
 	cfc2	$v0, C2_FLAG			# Get GTE flag value
 	nop
-	
+
 	bltz	$v0, .skip_prim			# Skip primitive if Z overflow
 	andi	$v0, $a3, 0x3
-	
+
 	NCLIP							# Backface culling
-	
+
 	srl		$v1, $a3, 12
 	andi	$v1, 1
-	
+
 	bnez	$v1, .no_culling
 	nop
-	
+
 	mfc2	$v1, C2_MAC0
 	nop
 	bltz	$v1, .skip_prim
 	nop
-	
+
 .no_culling:
 
 	beq		$v0, 0x1, .prim_tri		# If primitive is a triangle
 	nop
 	beq		$v0, 0x2, .prim_quad	# If primitive is a quad
 	nop
-	
+
 	b		.skip_prim
 	nop
 
@@ -140,25 +143,25 @@ smdSortModel:
 	addiu	$t4, 8					# Advance from indices
 
 	AVSZ3							# Calculate average Z
-	
+
 	srl		$v0, $t0, 16			# Get Z divisor from OT_LEN value
 	andi	$v0, 0xff
-	
+
 	mfc2	$t5, C2_OTZ				# Get AVSZ3 result
-	
+
 	sra		$v1, $t0, 24			# Get Z offset from OT_LEN value
-	
+
 	srl		$t5, $v0				# Apply divisor and offset
 	sub		$t5, $v1
-	
+
 	blez	$t5, .skip_prim			# Skip primitive if less than zero
 	andi	$v1, $t0, 0xffff
 	bge		$t5, $v1, .skip_prim	# Skip primitive if greater than OT length
 	sll		$t5, 2
 	addu	$t5, $a0				# Append OTZ to OT address
-	
+
 	ClipTestTri
-	
+
 	and		$v0, $s0, $s1			# v0 & v1
 	beqz	$v0, .do_draw
 	and		$v0, $s1, $s2			# v1 & v2
@@ -168,10 +171,10 @@ smdSortModel:
 	nop
 	b		.skip_prim
 	nop
-	
+
 .do_draw:
-	
-	
+
+
 	srl		$v0, $a3, 2					# Lighting enabled?
 	andi	$v0, 0x3
 	bnez	$v0, .F3_light
@@ -180,18 +183,18 @@ smdSortModel:
 	andi	$v0, $a3, 0x10				# Gouraud shaded
 	bnez	$v0, .F3_gouraud
 	nop
-	
+
 	andi	$v0, $a3, 0x20				# Textured triangle
 	bnez	$v0, .F3_textured
 	nop
-	
+
 	lw		$v0, 0( $t4 )				# Flat color, no lighting
 	lui		$v1, 0x2000
 	or		$v0, $v1
-	
+
 	b		.sort_F3_pri
 	sw		$v0, POLYF3_rgbc( $a1 )
-	
+
 .F3_gouraud:
 
 	lw		$v0, 0($t4)
@@ -213,30 +216,30 @@ smdSortModel:
 	or		$v0, $v1
 	sw		$v0, POLYFT3_rgbc( $a1 )
 	addiu	$t4, 4
-	
+
 	lhu		$v0, 0( $t4 )				# Load texture coordinates
 	lhu		$v1, 2( $t4 )
 	sh		$v0, POLYFT3_uv0( $a1 )
 	lhu		$v0, 4( $t4 )
 	sh		$v1, POLYFT3_uv1( $a1 )
 	sh		$v0, POLYFT3_uv2( $a1 )
-	
+
 	lw		$v0, 8( $t4 )				# Tpage + CLUT
 	nop
 	andi	$v1, $v0, 0xffff
 	sh		$v1, POLYFT3_tpage( $a1 )
 	srl		$v0, 16
-	
+
 	b		.sort_FT3_pri
 	sh		$v0, POLYFT3_clut( $a1 )
-	
+
 .F3_light:
-	
+
 	lhu		$v0, 0( $t4 )				# Load normal 0
-	
+
 	srl		$v1, $a3, 2
 	andi	$v1, $v1, 0x3
-	
+
 	sll		$v0, 3
 	addu	$v0, $t2
 	lwc2	C2_VXY0, 0( $v0 )
@@ -244,26 +247,26 @@ smdSortModel:
 
 	beq		$v1, 0x2, .F3_light_smt
 	nop
-	
+
 	lw		$v0, 4( $t4 )
 	lui		$v1, 0x2000
 	or		$v0, $v1
 	mtc2	$v0, C2_RGB
-	
+
 	addiu	$t4, 8
 	nop
-	
+
 	NCCS
 
 	andi	$v0, $a3, 0x20				# Textured triangle
 	bnez	$v0, .F3_light_tex
 	nop
-	
+
 	swc2	C2_RGB2, POLYF3_rgbc( $a1 )
-	
+
 	b		.sort_F3_pri
 	nop
-	
+
 .F3_light_tex:
 
 	lhu		$v0, 0( $t4 )				# Load texture coordinates
@@ -272,23 +275,23 @@ smdSortModel:
 	lhu		$v0, 4( $t4 )
 	sh		$v1, POLYFT3_uv1( $a1 )
 	sh		$v0, POLYFT3_uv2( $a1 )
-	
+
 	lw		$v1, 8( $t4 )
 	nop
 	andi	$v0, $v1, 0xffff
 	sh		$v0, POLYFT3_tpage( $a1 )
 	srl		$v0, $v1, 16
 	sh		$v0, POLYFT3_clut( $a1 )
-	
+
 	mfc2	$v0, C2_RGB2
 	lui		$v1, 0x2400
 	or		$v0, $v1
-	
+
 	b		.sort_FT3_pri
 	sw		$v0, POLYFT3_rgbc( $a1 )
-	
+
 .F3_light_smt:
-	
+
 	lhu		$v0, 2( $t4 )			# Load normals 1 and 2
 	lhu		$v1, 4( $t4 )
 	sll		$v0, 3
@@ -303,20 +306,20 @@ smdSortModel:
 	lui		$v1, 0x3000				# Load color
 	or		$v0, $v1
 	mtc2	$v0, C2_RGB
-	
+
 	addiu	$t4, 12
 	nop
-	
+
 	NCCT
-	
+
 	andi	$v0, $a3, 0x20				# Textured triangle
 	bnez	$v0, .F3_light_tex_smt
 	nop
-	
+
 	swc2	C2_RGB0, POLYG3_rgbc0( $a1 )
 	swc2	C2_RGB1, POLYG3_rgbc1( $a1 )
 	swc2	C2_RGB2, POLYG3_rgbc2( $a1 )
-	
+
 	b		.sort_G3_pri
 	nop
 
@@ -328,21 +331,21 @@ smdSortModel:
 	lhu		$v0, 4( $t4 )
 	sh		$v1, POLYGT3_uv1( $a1 )
 	sh		$v0, POLYGT3_uv2( $a1 )
-	
+
 	lw		$v1, 8( $t4 )
 	nop
 	andi	$v0, $v1, 0xffff
 	sh		$v0, POLYGT3_tpage( $a1 )
 	srl		$v0, $v1, 16
 	sh		$v0, POLYGT3_clut( $a1 )
-	
+
 	mfc2	$v0, C2_RGB0
 	lui		$v1, 0x3400
 	or		$v0, $v1
-	
+
 	swc2	C2_RGB1, POLYGT3_rgbc1( $a1 )
 	swc2	C2_RGB2, POLYGT3_rgbc2( $a1 )
-	
+
 	b		.sort_GT3_pri
 	sw		$v0, POLYGT3_rgbc0( $a1 )
 
@@ -351,7 +354,7 @@ smdSortModel:
 	swc2	C2_SXY0, POLYF3_xy0($a1)
 	swc2	C2_SXY1, POLYF3_xy1($a1)
 	swc2	C2_SXY2, POLYF3_xy2($a1)
-	
+
 	la		$v0, _smd_tpage_base
 	lhu		$v0, 0($v0)
 	srl		$v1, $a3, 6				# Get blend mode
@@ -361,7 +364,7 @@ smdSortModel:
 	lui		$v1, 0xe100
 	or		$v0, $v1
 	sw		$v0, POLYF3_tpage($a1)	# Store TPage
-	
+
 	.set noat
 
 	lui		$v1, 0x0500
@@ -379,21 +382,21 @@ smdSortModel:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($t5)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYF3_len
-	
+
 .sort_FT3_pri:
 
 	swc2	C2_SXY0, POLYFT3_xy0( $a1 )
 	swc2	C2_SXY1, POLYFT3_xy1( $a1 )
 	swc2	C2_SXY2, POLYFT3_xy2( $a1 )
-	
+
 	.set noat
 
 	lui		$v1, 0x0700
@@ -411,21 +414,21 @@ smdSortModel:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($t5)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYFT3_len
-	
+
 .sort_G3_pri:
 
 	swc2	C2_SXY0, POLYG3_xy0( $a1 )
 	swc2	C2_SXY1, POLYG3_xy1( $a1 )
 	swc2	C2_SXY2, POLYG3_xy2( $a1 )
-	
+
 	la		$v0, _smd_tpage_base
 	lhu		$v0, 0($v0)
 	srl		$v1, $a3, 6				# Get blend mode
@@ -435,7 +438,7 @@ smdSortModel:
 	lui		$v1, 0xe100
 	or		$v0, $v1
 	sw		$v0, POLYG3_tpage($a1)	# Store TPage
-	
+
 	.set noat
 
 	lui		$v1, 0x0700
@@ -453,21 +456,21 @@ smdSortModel:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($t5)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYG3_len
-	
+
 .sort_GT3_pri:
 
 	swc2	C2_SXY0, POLYGT3_xy0( $a1 )
 	swc2	C2_SXY1, POLYGT3_xy1( $a1 )
 	swc2	C2_SXY2, POLYGT3_xy2( $a1 )
-	
+
 	.set noat
 
 	lui		$v1, 0x0900
@@ -485,17 +488,17 @@ smdSortModel:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($t5)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYGT3_len
 
 ## Quads
-	
+
 .prim_quad:							# Quad processing
 
 	mfc2	$t6, C2_SXY0			# Retrieve first projected vertex
@@ -506,38 +509,38 @@ smdSortModel:
 	addu	$t5, $t1
 	lwc2	C2_VXY0, 0( $t5 )
 	lwc2	C2_VZ0 , 4( $t5 )
-	
+
 	nRTPS
-	
+
 	cfc2	$v1, C2_FLAG			# Get GTE flag value
-	
+
 	srl		$v0, $t0, 16			# Get Z divisor from OT_LEN value
-	
+
 	bltz	$v1, .skip_prim
 	nop
-	
+
 	AVSZ4
-	
+
 	andi	$v0, 0xff
-	
+
 	mfc2	$t5, C2_OTZ
-	
+
 	sra		$v1, $t0, 24				# Get Z offset from OT_LEN value
-	
+
 	srl		$t5, $v0					# Apply divisor and offset
 	sub		$t5, $v1
-	
+
 	blez	$t5, .skip_prim				# Skip primitive if less than zero
 	andi	$v1, $t0, 0xffff
 	bge		$t5, $v1, .skip_prim		# Skip primitive if greater than OT length
 	sll		$t5, 2
 	addu	$t5, $a0					# Append OTZ to OT address
-	
+
 	# no touch:
 	# a0, a1, a2, a3, t0, t1, t2, t3, t4, t5(ot), t6(sxy0)
-	
+
 	ClipTestQuad
-	
+
 	and		$v0, $s0, $s1				# v0 & v1
 	beqz	$v0, .do_draw_q
 	and		$v0, $s1, $s2				# v1 & v2
@@ -553,29 +556,29 @@ smdSortModel:
 	nop
 	b		.skip_prim
 	nop
-	
+
 .do_draw_q:
 
 	srl		$v0, $a3, 2					# Lighting enabled?
 	andi	$v0, 0x3
 	bnez	$v0, .F4_light
 	nop
-	
+
 	andi	$v0, $a3, 0x10				# Gouraud quad
 	bnez	$v0, .F4_gouraud
 	nop
-	
+
 	andi	$v0, $a3, 0x20				# Textured quad
 	bnez	$v0, .F4_textured
 	nop
-	
+
 	lw		$v0, 0($t4)
 	lui		$v1, 0x2800
 	or		$v0, $v1
-	
+
 	b		.sort_F4_pri
 	sw		$v0, POLYF4_rgbc($a1)
-	
+
 .F4_gouraud:
 
 	lw		$v0, 0($t4)
@@ -591,15 +594,15 @@ smdSortModel:
 	sw		$v0, POLYG4_rgbc2($a1)
 	b		.sort_G4_pri
 	sw		$v1, POLYG4_rgbc3($a1)
-	
+
 .F4_textured:
-	
+
 	lw		$v0, 0($t4)
 	lui		$v1, 0x2c00
 	or		$v0, $v1
 	sw		$v0, POLYFT4_rgbc( $a1 )
 	addiu	$t4, 4
-	
+
 	lhu		$v0, 0($t4)					# Load texture coordinates
 	lhu		$v1, 2($t4)
 	sh		$v0, POLYFT4_uv0( $a1 )
@@ -608,23 +611,23 @@ smdSortModel:
 	lhu		$v1, 6( $t4 )
 	sh		$v0, POLYFT4_uv2( $a1 )
 	sh		$v1, POLYFT4_uv3( $a1 )
-	
+
 	lw		$v1, 8( $t4 )
 	nop
 	andi	$v0, $v1, 0xffff
 	sh		$v0, POLYFT4_tpage( $a1 )
 	srl		$v0, $v1, 16
-	
+
 	b		.sort_FT4_pri
 	sh		$v0, POLYFT4_clut($a1)
-	
+
 .F4_light:
 
 	lhu		$v0, 0( $t4 )				# Load normal 0
-	
+
 	srl		$v1, $a3, 2
 	andi	$v1, $v1, 0x3
-	
+
 	sll		$v0, 3
 	addu	$v0, $t2
 	lwc2	C2_VXY0, 0( $v0 )
@@ -632,26 +635,26 @@ smdSortModel:
 
 	beq		$v1, 0x2, .F4_light_smt
 	nop
-	
+
 	lw		$v0, 4( $t4 )
 	lui		$v1, 0x2800
 	or		$v0, $v1
 	mtc2	$v0, C2_RGB
-	
+
 	addiu	$t4, 8
 	nop
-	
+
 	NCCS
-	
+
 	andi	$v0, $a3, 0x20				# Textured triangle
 	bnez	$v0, .F4_light_tex
 	nop
 
 	swc2	C2_RGB2, POLYF4_rgbc( $a1 )
-	
+
 	b		.sort_F4_pri
 	nop
-	
+
 .F4_light_tex:
 
 	lhu		$v0, 0( $t4 )				# Load texture coordinates
@@ -662,21 +665,21 @@ smdSortModel:
 	lhu		$v1, 6( $t4 )
 	sh		$v0, POLYFT4_uv2( $a1 )
 	sh		$v1, POLYFT4_uv3( $a1 )
-	
+
 	lw		$v1, 8( $t4 )
 	nop
 	andi	$v0, $v1, 0xffff
 	sh		$v0, POLYFT4_tpage( $a1 )
 	srl		$v0, $v1, 16
 	sh		$v0, POLYFT4_clut( $a1 )
-	
+
 	mfc2	$v0, C2_RGB2
 	lui		$v1, 0x2c00
 	or		$v0, $v1
-	
+
 	b		.sort_FT4_pri
 	sw		$v0, POLYFT4_rgbc( $a1 )
-	
+
 .F4_light_smt:
 
 	lhu		$v0, 2( $t4 )			# Load normals 1 and 2
@@ -689,38 +692,38 @@ smdSortModel:
 	lwc2	C2_VZ1 , 4( $v0 )
 	lwc2	C2_VXY2, 0( $v1 )
 	lwc2	C2_VZ2 , 4( $v1 )
-	
+
 	lw		$v0, 8( $t4 )
 	lui		$v1, 0x3800				# Load color
 	or		$v0, $v1
 	mtc2	$v0, C2_RGB
-	
+
 	nNCCT
-	
+
 	lhu		$v0, 6( $t4 )			# Load normal 3
-	
+
 	addiu	$t4, 12
-	
+
 	sll		$v0, 3
 	addu	$v0, $t2
 	lwc2	C2_VXY0, 0( $v0 )
 	lwc2	C2_VZ0 , 4( $v0 )
-	
+
 	andi	$v0, $a3, 0x20				# Textured triangle
 	bnez	$v0, .F4_light_tex_smt
 	nop
-	
+
 	swc2	C2_RGB0, POLYG4_rgbc0( $a1 )
 	swc2	C2_RGB1, POLYG4_rgbc1( $a1 )
 	swc2	C2_RGB2, POLYG4_rgbc2( $a1 )
-	
+
 	nNCCS
-	
+
 	swc2	C2_RGB2, POLYG4_rgbc3( $a1 )
-	
+
 	b		.sort_G4_pri
 	nop
-	
+
 .F4_light_tex_smt:
 
 	mfc2	$v0, C2_RGB0
@@ -729,9 +732,9 @@ smdSortModel:
 	sw		$v0, POLYGT4_rgbc0( $a1 )
 	swc2	C2_RGB1, POLYGT4_rgbc1( $a1 )
 	swc2	C2_RGB2, POLYGT4_rgbc2( $a1 )
-	
+
 	NCCS
-	
+
 	lhu		$v0, 0( $t4 )				# Load texture coordinates
 	lhu		$v1, 2( $t4 )
 	sh		$v0, POLYGT4_uv0( $a1 )
@@ -743,21 +746,21 @@ smdSortModel:
 
 	lw		$v1, 8( $t4 )
 	swc2	C2_RGB2, POLYGT4_rgbc3( $a1 )
-	
+
 	andi	$v0, $v1, 0xffff
 	sh		$v0, POLYGT4_tpage( $a1 )
 	srl		$v0, $v1, 16
-	
+
 	b		.sort_GT4_pri
 	sh		$v0, POLYGT4_clut( $a1 )
-	
+
 .sort_F4_pri:
 
 	sw		$t6, POLYF4_xy0($a1)
 	swc2	C2_SXY0, POLYF4_xy1($a1)
 	swc2	C2_SXY1, POLYF4_xy2($a1)
 	swc2	C2_SXY2, POLYF4_xy3($a1)
-	
+
 	la		$v0, _smd_tpage_base
 	lhu		$v0, 0($v0)
 	srl		$v1, $a3, 6				# Get blend mode
@@ -767,7 +770,7 @@ smdSortModel:
 	lui		$v1, 0xe100
 	or		$v0, $v1
 	sw		$v0, POLYF4_tpage($a1)	# Store TPage
-	
+
 	.set noat
 
 	lui		$v1, 0x0600
@@ -785,22 +788,22 @@ smdSortModel:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($t5)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYF4_len
-	
+
 .sort_FT4_pri:
 
 	sw		$t6, POLYFT4_xy0($a1)
 	swc2	C2_SXY0, POLYFT4_xy1($a1)
 	swc2	C2_SXY1, POLYFT4_xy2($a1)
 	swc2	C2_SXY2, POLYFT4_xy3($a1)
-	
+
 	.set noat
 
 	lui		$v1, 0x0900
@@ -818,22 +821,22 @@ smdSortModel:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($t5)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYFT4_len
-	
+
 .sort_G4_pri:
 
 	sw		$t6, POLYG4_xy0($a1)
 	swc2	C2_SXY0, POLYG4_xy1($a1)
 	swc2	C2_SXY1, POLYG4_xy2($a1)
 	swc2	C2_SXY2, POLYG4_xy3($a1)
-	
+
 	la		$v0, _smd_tpage_base
 	lhu		$v0, 0($v0)
 	srl		$v1, $a3, 6				# Get blend mode
@@ -843,7 +846,7 @@ smdSortModel:
 	lui		$v1, 0xe100
 	or		$v0, $v1
 	sw		$v0, POLYG4_tpage($a1)	# Store TPage
-	
+
 	.set noat
 
 	lui		$v1, 0x0900
@@ -861,22 +864,22 @@ smdSortModel:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($t5)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYG4_len
-	
+
 .sort_GT4_pri:
 
 	sw		$t6, POLYGT4_xy0($a1)
 	swc2	C2_SXY0, POLYGT4_xy1($a1)
 	swc2	C2_SXY1, POLYGT4_xy2($a1)
 	swc2	C2_SXY2, POLYGT4_xy3($a1)
-	
+
 	.set noat
 
 	lui		$v1, 0x0c00
@@ -894,20 +897,20 @@ smdSortModel:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($t5)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYGT4_len
-	
+
 .skip_prim:
 
 	b		.sort_loop
 	nop
-	
+
 .exit:
 
 	lw		$s0, 0( $sp )
@@ -917,8 +920,9 @@ smdSortModel:
 	addiu	$sp, 16
 	jr		$ra
 	move	$v0, $a1
-	
-	
+
+.section .bss._smd_tpage_base, "w"
 .comm _smd_tpage_base, 4, 4
 
+.section .bss._sc_clip, "w"
 .comm _sc_clip, 8, 4
diff --git a/examples/demos/n00bdemo/smd_cel.s b/libpsn00b/smd/smd_cel.s
index 383f043..8a39f01 100644
--- a/examples/demos/n00bdemo/smd_cel.s
+++ b/libpsn00b/smd/smd_cel.s
@@ -1,14 +1,16 @@
+# PSn00bSDK .SMD model parser library
+# (C) 2019-2023 Lameguy64, spicyjpeg - MPL licensed
+
 .set noreorder
 
 .include "gtereg.inc"
 .include "inline_s.inc"
-.include "smd_s.h"
-
+.include "smd/smd_s.inc"
 
-.section .text
-	
+.section .text.smdSetCelTex, "ax", @progbits
 .global smdSetCelTex
 .type smdSetCelTex, @function
+
 smdSetCelTex:
 	# a0 - TPage value
 	# a1 - CLUT value
@@ -18,11 +20,11 @@ smdSetCelTex:
 	or		$a0, $a1
 	jr		$ra
 	sw		$a0, 0($v0)
-	
-	
-	
+
+.section .text.smdSetCelParam, "ax", @progbits
 .global smdSetCelParam
 .type smdSetCelParam, @function
+
 smdSetCelParam:
 	# a0 - Shading texture U offset
 	# a1 - Shading texture V offset
@@ -38,42 +40,42 @@ smdSetCelParam:
 	or		$a3, $v1
 	jr		$ra
 	sw		$a2, 0($v0)
-	
-	
+
+.section .text.smdSortModelCel, "ax", @progbits
 .global smdSortModelCel
 .type smdSortModelCel, @function
+
 smdSortModelCel:
-	
 	# a0 - Pointer SC_OT structure
 	# a1 - Pointer to next primitive
 	# a2 - Pointer to SMD data address
 	# v0 - New pointer of primitive buffer (return)
-	
+
 	addiu	$sp, -16
 	sw		$s0, 0($sp)
 	sw		$s1, 4($sp)
 	sw		$s2, 8($sp)
 	sw		$s3, 12($sp)
-	
+
 	la		$v0, _sc_clip
 	lw		$t8, 0($v0)
 	lw		$t9, 4($v0)
-	
+
 	lw		$t0, OT_LEN($a0)
 	lw		$a0, OT_ADDR($a0)
 	lw		$t1, SMD_HEAD_PVERTS($a2)
 	lw		$t2, SMD_HEAD_PNORMS($a2)
 	lw		$t3, SMD_HEAD_PPRIMS($a2)
-	
+
 .sort_loop:
 
 	nop
 	lw		$a3, 0($t3)				# Get primitive ID word
 	move	$t4, $t3
-	
+
 	beqz	$a3, .exit				# Check if terminator (just zero)
 	addiu	$t4, 4
-	
+
 	lhu		$t5, 0( $t4 )			# Load vertices
 	lhu		$t6, 2( $t4 )
 	lhu		$t7, 4( $t4 )
@@ -89,38 +91,38 @@ smdSortModelCel:
 	lwc2	C2_VZ1 , 4( $t6 )
 	lwc2	C2_VXY2, 0( $t7 )
 	lwc2	C2_VZ2 , 4( $t7 )
-	
+
 	srl		$v1, $a3, 24			# Get primitive size
 	addu	$t3, $v1				# Step main pointer to next primitive
-	
+
 	RTPT
-	
+
 	cfc2	$v0, C2_FLAG			# Get GTE flag value
 	nop
-	
+
 	bltz	$v0, .skip_prim			# Skip primitive if Z overflow
 	andi	$v0, $a3, 0x3
-	
+
 	NCLIP							# Backface culling
-	
+
 	srl		$v1, $a3, 12
 	andi	$v1, 1
-	
+
 	bnez	$v1, .no_culling
 	nop
-	
+
 	mfc2	$v1, C2_MAC0
 	nop
 	bltz	$v1, .skip_prim
 	nop
-	
+
 .no_culling:
 
 	beq		$v0, 0x1, .prim_tri		# If primitive is a triangle
 	nop
 	beq		$v0, 0x2, .prim_quad	# If primitive is a quad
 	nop
-	
+
 	b		.skip_prim
 	nop
 
@@ -131,25 +133,25 @@ smdSortModelCel:
 	addiu	$t4, 8					# Advance from indices
 
 	AVSZ3							# Calculate average Z
-	
+
 	srl		$v0, $t0, 16			# Get Z divisor from OT_LEN value
 	andi	$v0, 0xff
-	
+
 	mfc2	$t5, C2_OTZ				# Get AVSZ3 result
-	
+
 	sra		$v1, $t0, 24			# Get Z offset from OT_LEN value
-	
+
 	srl		$t5, $v0				# Apply divisor and offset
 	sub		$t5, $v1
-	
+
 	blez	$t5, .skip_prim			# Skip primitive if less than zero
 	andi	$v1, $t0, 0xffff
 	bge		$t5, $v1, .skip_prim	# Skip primitive if greater than OT length
 	sll		$t5, 2
 	addu	$t5, $a0				# Append OTZ to OT address
-	
+
 	ClipTestTri
-	
+
 	and		$v0, $s0, $s1			# v0 & v1
 	beqz	$v0, .do_draw
 	and		$v0, $s1, $s2			# v1 & v2
@@ -159,10 +161,10 @@ smdSortModelCel:
 	nop
 	b		.skip_prim
 	nop
-	
+
 .do_draw:
-	
-	
+
+
 	srl		$v0, $a3, 2					# Lighting enabled?
 	andi	$v0, 0x3
 	bnez	$v0, .F3_light
@@ -171,18 +173,18 @@ smdSortModelCel:
 	andi	$v0, $a3, 0x10				# Gouraud shaded
 	bnez	$v0, .F3_gouraud
 	nop
-	
+
 	andi	$v0, $a3, 0x20				# Textured triangle
 	bnez	$v0, .F3_textured
 	nop
-	
+
 	lw		$v0, 0( $t4 )				# Flat color, no lighting
 	lui		$v1, 0x2000
 	or		$v0, $v1
-	
+
 	b		.sort_F3_pri
 	sw		$v0, POLYF3_rgbc( $a1 )
-	
+
 .F3_gouraud:
 
 	lw		$v0, 0($t4)
@@ -204,30 +206,30 @@ smdSortModelCel:
 	or		$v0, $v1
 	sw		$v0, POLYFT3_rgbc( $a1 )
 	addiu	$t4, 4
-	
+
 	lhu		$v0, 0( $t4 )				# Load texture coordinates
 	lhu		$v1, 2( $t4 )
 	sh		$v0, POLYFT3_uv0( $a1 )
 	lhu		$v0, 4( $t4 )
 	sh		$v1, POLYFT3_uv1( $a1 )
 	sh		$v0, POLYFT3_uv2( $a1 )
-	
+
 	lw		$v0, 8( $t4 )				# Tpage + CLUT
 	nop
 	andi	$v1, $v0, 0xffff
 	sh		$v1, POLYFT3_tpage( $a1 )
 	srl		$v0, 16
-	
+
 	b		.sort_FT3_pri
 	sh		$v0, POLYFT3_clut( $a1 )
-	
+
 .F3_light:
-	
+
 	lhu		$v0, 0( $t4 )				# Load normal 0
-	
+
 	srl		$v1, $a3, 2
 	andi	$v1, $v1, 0x3
-	
+
 	sll		$v0, 3
 	addu	$v0, $t2
 	lwc2	C2_VXY0, 0( $v0 )
@@ -235,26 +237,26 @@ smdSortModelCel:
 
 	beq		$v1, 0x2, .F3_light_smt
 	nop
-	
+
 	lw		$v0, 4( $t4 )
 	lui		$v1, 0x2000
 	or		$v0, $v1
 	mtc2	$v0, C2_RGB
-	
+
 	addiu	$t4, 8
 	nop
-	
+
 	NCCS
 
 	andi	$v0, $a3, 0x20				# Textured triangle
 	bnez	$v0, .F3_light_tex
 	nop
-	
+
 	swc2	C2_RGB2, POLYF3_rgbc( $a1 )
-	
+
 	b		.sort_F3_pri
 	nop
-	
+
 .F3_light_tex:
 
 	lhu		$v0, 0( $t4 )				# Load texture coordinates
@@ -263,23 +265,23 @@ smdSortModelCel:
 	lhu		$v0, 4( $t4 )
 	sh		$v1, POLYFT3_uv1( $a1 )
 	sh		$v0, POLYFT3_uv2( $a1 )
-	
+
 	lw		$v1, 8( $t4 )
 	nop
 	andi	$v0, $v1, 0xffff
 	sh		$v0, POLYFT3_tpage( $a1 )
 	srl		$v0, $v1, 16
 	sh		$v0, POLYFT3_clut( $a1 )
-	
+
 	mfc2	$v0, C2_RGB2
 	lui		$v1, 0x2400
 	or		$v0, $v1
-	
+
 	b		.sort_FT3_pri
 	sw		$v0, POLYFT3_rgbc( $a1 )
-	
+
 .F3_light_smt:
-	
+
 	lhu		$v0, 2($t4)			# Load normals 1 and 2
 	lhu		$v1, 4($t4)
 	sll		$v0, 3
@@ -293,32 +295,32 @@ smdSortModelCel:
 	lwc2	C2_VXY2, 0($v1)
 	lwc2	C2_VZ2 , 4($v1)
 	mtc2	$v0, C2_RGB
-	
+
 	swc2	C2_SXY0, POLYFT3_xy0($a1)
 	swc2	C2_SXY1, POLYFT3_xy1($a1)
 	swc2	C2_SXY2, POLYFT3_xy2($a1)
-	
+
 	la		$v0, _smd_cel_tpage		# Load cel shader TPage and CLUT values
 	lw		$v0, 0($v0)
-	
+
 	NCT
-	
+
 	andi	$v1, $v0, 0xffff
 	sh		$v1, POLYFT3_tpage($a1)
 	srl		$v1, $v0, 16
 	sh		$v1, POLYFT3_clut($a1)
-	
+
 	# Usable regs: v0, v1, at, t6, t7
-	
+
 	.set noat
-	
+
 	la		$at, _smd_cel_param			# Load cel shader parameters
 	lhu		$at, 0($at)
-	
+
 	mfc2	$t7, C2_RGB0				# Get first shaded color
 	andi	$v1, $at, 0xff				# Get U divisor value
 	andi	$t7, 0xffff					# Only keep R and G colors
-	
+
 	andi	$v0, $t7, 0xff				# U0
 	srl		$v0, $v1
 	sb		$v0, POLYFT3_uv0($a1)
@@ -326,7 +328,7 @@ smdSortModelCel:
 	srl		$v1, $at, 8
 	srl		$v0, $v1
 	sb		$v0, POLYFT3_uv0+1($a1)
-	
+
 	mfc2	$t7, C2_RGB1
 	andi	$v1, $at, 0xff
 	andi	$t7, 0xffff
@@ -337,7 +339,7 @@ smdSortModelCel:
 	srl		$v1, $at, 8
 	srl		$v0, $v1
 	sb		$v0, POLYFT3_uv1+1($a1)
-	
+
 	mfc2	$t7, C2_RGB2
 	andi	$v1, $at, 0xff
 	andi	$t7, 0xffff
@@ -348,16 +350,16 @@ smdSortModelCel:
 	srl		$v1, $at, 8
 	srl		$v0, $v1
 	sb		$v0, POLYFT3_uv2+1($a1)
-	
+
 	la		$v0, _smd_cel_col
 	lw		$v0, 0($v0)
 	lui		$v1, 0x2600
 	or		$v0, $v1
 	sw		$v0, POLYFT3_rgbc($a1)
-	
+
 	lw		$t7, 8($t4)
 	addiu	$t4, 12
-	
+
 	lui		$v1, 0x0700
 	lw		$v0, 0($t5)
 	lui		$at, 0xff00
@@ -373,21 +375,21 @@ smdSortModelCel:
 	and		$v0, $at
 	or		$v0, $v1
 	sw		$v0, 0($t5)
-	
+
 	lui		$v0, 0x8000
 	or		$a1, $v0
 	addiu	$a1, POLYFT3_len
-	
+
 	.set at
-	
+
 	andi	$v0, $a3, 0x20				# Textured triangle
 	bnez	$v0, .F3_light_tex_smt
 	nop
-	
+
 	#swc2	C2_RGB0, POLYG3_rgbc0( $a1 )
 	#swc2	C2_RGB1, POLYG3_rgbc1( $a1 )
 	#swc2	C2_RGB2, POLYG3_rgbc2( $a1 )
-	
+
 	lui		$v0, 0x2000
 	or		$t7, $v0
 	b		.sort_F3_pri
@@ -401,17 +403,17 @@ smdSortModelCel:
 	lhu		$v0, 4( $t4 )
 	sh		$v1, POLYFT3_uv1( $a1 )
 	sh		$v0, POLYFT3_uv2( $a1 )
-	
+
 	lw		$v1, 8( $t4 )
 	nop
 	andi	$v0, $v1, 0xffff
 	sh		$v0, POLYFT3_tpage( $a1 )
 	srl		$v0, $v1, 16
 	sh		$v0, POLYFT3_clut( $a1 )
-	
+
 	lui		$v1, 0x2400
 	or		$t7, $v1
-	
+
 	b		.sort_FT3_pri
 	sw		$t7, POLYFT3_rgbc( $a1 )
 
@@ -420,7 +422,7 @@ smdSortModelCel:
 	swc2	C2_SXY0, POLYF3_xy0($a1)
 	swc2	C2_SXY1, POLYF3_xy1($a1)
 	swc2	C2_SXY2, POLYF3_xy2($a1)
-	
+
 	la		$v0, _smd_tpage_base
 	lhu		$v0, 0($v0)
 	srl		$v1, $a3, 6				# Get blend mode
@@ -430,7 +432,7 @@ smdSortModelCel:
 	lui		$v1, 0xe100
 	or		$v0, $v1
 	sw		$v0, POLYF3_tpage($a1)	# Store TPage
-	
+
 	.set noat
 
 	lui		$v1, 0x0500
@@ -448,21 +450,21 @@ smdSortModelCel:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($t5)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYF3_len
-	
+
 .sort_FT3_pri:
 
 	swc2	C2_SXY0, POLYFT3_xy0( $a1 )
 	swc2	C2_SXY1, POLYFT3_xy1( $a1 )
 	swc2	C2_SXY2, POLYFT3_xy2( $a1 )
-	
+
 	.set noat
 
 	lui		$v1, 0x0700
@@ -480,21 +482,21 @@ smdSortModelCel:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($t5)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYFT3_len
-	
+
 .sort_G3_pri:
 
 	swc2	C2_SXY0, POLYG3_xy0( $a1 )
 	swc2	C2_SXY1, POLYG3_xy1( $a1 )
 	swc2	C2_SXY2, POLYG3_xy2( $a1 )
-	
+
 	la		$v0, _smd_tpage_base
 	lhu		$v0, 0($v0)
 	srl		$v1, $a3, 6				# Get blend mode
@@ -504,7 +506,7 @@ smdSortModelCel:
 	lui		$v1, 0xe100
 	or		$v0, $v1
 	sw		$v0, POLYG3_tpage($a1)	# Store TPage
-	
+
 	.set noat
 
 	lui		$v1, 0x0700
@@ -522,21 +524,21 @@ smdSortModelCel:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($t5)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYG3_len
-	
+
 .sort_GT3_pri:
 
 	swc2	C2_SXY0, POLYGT3_xy0( $a1 )
 	swc2	C2_SXY1, POLYGT3_xy1( $a1 )
 	swc2	C2_SXY2, POLYGT3_xy2( $a1 )
-	
+
 	.set noat
 
 	lui		$v1, 0x0900
@@ -554,17 +556,17 @@ smdSortModelCel:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($t5)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYGT3_len
 
 ## Quads
-	
+
 .prim_quad:							# Quad processing
 
 	mfc2	$t6, C2_SXY0			# Retrieve first projected vertex
@@ -575,38 +577,38 @@ smdSortModelCel:
 	addu	$t5, $t1
 	lwc2	C2_VXY0, 0( $t5 )
 	lwc2	C2_VZ0 , 4( $t5 )
-	
+
 	nRTPS
-	
+
 	cfc2	$v1, C2_FLAG			# Get GTE flag value
-	
+
 	srl		$v0, $t0, 16			# Get Z divisor from OT_LEN value
-	
+
 	bltz	$v1, .skip_prim
 	nop
-	
+
 	AVSZ4
-	
+
 	andi	$v0, 0xff
-	
+
 	mfc2	$t5, C2_OTZ
-	
+
 	sra		$v1, $t0, 24				# Get Z offset from OT_LEN value
-	
+
 	srl		$t5, $v0					# Apply divisor and offset
 	sub		$t5, $v1
-	
+
 	blez	$t5, .skip_prim				# Skip primitive if less than zero
 	andi	$v1, $t0, 0xffff
 	bge		$t5, $v1, .skip_prim		# Skip primitive if greater than OT length
 	sll		$t5, 2
 	addu	$t5, $a0					# Append OTZ to OT address
-	
+
 	# no touch:
 	# a0, a1, a2, a3, t0, t1, t2, t3, t4, t5(ot), t6(sxy0)
-	
+
 	ClipTestQuad
-	
+
 	and		$v0, $s0, $s1				# v0 & v1
 	beqz	$v0, .do_draw_q
 	and		$v0, $s1, $s2				# v1 & v2
@@ -622,29 +624,29 @@ smdSortModelCel:
 	nop
 	b		.skip_prim
 	nop
-	
+
 .do_draw_q:
 
 	srl		$v0, $a3, 2					# Lighting enabled?
 	andi	$v0, 0x3
 	bnez	$v0, .F4_light
 	nop
-	
+
 	andi	$v0, $a3, 0x10				# Gouraud quad
 	bnez	$v0, .F4_gouraud
 	nop
-	
+
 	andi	$v0, $a3, 0x20				# Textured quad
 	bnez	$v0, .F4_textured
 	nop
-	
+
 	lw		$v0, 0($t4)
 	lui		$v1, 0x2800
 	or		$v0, $v1
-	
+
 	b		.sort_F4_pri
 	sw		$v0, POLYF4_rgbc($a1)
-	
+
 .F4_gouraud:
 
 	lw		$v0, 0($t4)
@@ -660,15 +662,15 @@ smdSortModelCel:
 	sw		$v0, POLYG4_rgbc2($a1)
 	b		.sort_G4_pri
 	sw		$v1, POLYG4_rgbc3($a1)
-	
+
 .F4_textured:
-	
+
 	lw		$v0, 0($t4)
 	lui		$v1, 0x2c00
 	or		$v0, $v1
 	sw		$v0, POLYFT4_rgbc( $a1 )
 	addiu	$t4, 4
-	
+
 	lhu		$v0, 0($t4)					# Load texture coordinates
 	lhu		$v1, 2($t4)
 	sh		$v0, POLYFT4_uv0( $a1 )
@@ -677,23 +679,23 @@ smdSortModelCel:
 	lhu		$v1, 6( $t4 )
 	sh		$v0, POLYFT4_uv2( $a1 )
 	sh		$v1, POLYFT4_uv3( $a1 )
-	
+
 	lw		$v1, 8( $t4 )
 	nop
 	andi	$v0, $v1, 0xffff
 	sh		$v0, POLYFT4_tpage( $a1 )
 	srl		$v0, $v1, 16
-	
+
 	b		.sort_FT4_pri
 	sh		$v0, POLYFT4_clut($a1)
-	
+
 .F4_light:
 
 	lhu		$v0, 0( $t4 )				# Load normal 0
-	
+
 	srl		$v1, $a3, 2
 	andi	$v1, $v1, 0x3
-	
+
 	sll		$v0, 3
 	addu	$v0, $t2
 	lwc2	C2_VXY0, 0( $v0 )
@@ -701,26 +703,26 @@ smdSortModelCel:
 
 	beq		$v1, 0x2, .F4_light_smt
 	nop
-	
+
 	lw		$v0, 4( $t4 )
 	lui		$v1, 0x2800
 	or		$v0, $v1
 	mtc2	$v0, C2_RGB
-	
+
 	addiu	$t4, 8
 	nop
-	
+
 	NCS
-	
+
 	andi	$v0, $a3, 0x20				# Textured triangle
 	bnez	$v0, .F4_light_tex
 	nop
 
 	swc2	C2_RGB2, POLYF4_rgbc( $a1 )
-	
+
 	b		.sort_F4_pri
 	nop
-	
+
 .F4_light_tex:
 
 	lhu		$v0, 0( $t4 )				# Load texture coordinates
@@ -731,22 +733,22 @@ smdSortModelCel:
 	lhu		$v1, 6( $t4 )
 	sh		$v0, POLYFT4_uv2( $a1 )
 	sh		$v1, POLYFT4_uv3( $a1 )
-	
+
 	lw		$v1, 8( $t4 )
 	nop
 	andi	$v0, $v1, 0xffff
 	sh		$v0, POLYFT4_tpage( $a1 )
 	srl		$v0, $v1, 16
 	sh		$v0, POLYFT4_clut( $a1 )
-	
+
 	mfc2	$v0, C2_RGB2
 	lui		$v1, 0x2c00
 	or		$v0, $v1
-	
+
 	b		.sort_FT4_pri
 	nop
 	sw		$v0, POLYFT4_rgbc( $a1 )
-	
+
 .F4_light_smt:
 
 	lhu		$v0, 2( $t4 )			# Load normals 1 and 2
@@ -759,33 +761,33 @@ smdSortModelCel:
 	lwc2	C2_VZ1 , 4( $v0 )
 	lwc2	C2_VXY2, 0( $v1 )
 	lwc2	C2_VZ2 , 4( $v1 )
-	
+
 	sw		$t6, POLYFT4_xy0($a1)
 	swc2	C2_SXY0, POLYFT4_xy1($a1)
 	swc2	C2_SXY1, POLYFT4_xy2($a1)
 	swc2	C2_SXY2, POLYFT4_xy3($a1)
-	
+
 	la		$v0, _smd_cel_tpage		# Load cel shader TPage and CLUT values
 	lw		$v0, 0($v0)
-	
+
 	NCT
-	
+
 	andi	$v1, $v0, 0xffff
 	sh		$v1, POLYFT4_tpage($a1)
 	srl		$v1, $v0, 16
 	sh		$v1, POLYFT4_clut($a1)
-	
+
 	# Usable regs: v0, v1, at, t7
-	
+
 	.set noat
-	
+
 	la		$at, _smd_cel_param			# Load cel shader parameters
 	lhu		$at, 0($at)
-	
+
 	mfc2	$t7, C2_RGB0
 	andi	$v1, $at, 0xff				# Get U divisor value
 	andi	$t7, 0xffff					# Only keep R and G colors
-	
+
 	andi	$v0, $t7, 0xff				# U0
 	srl		$v0, $v1
 	sb		$v0, POLYFT4_uv0($a1)
@@ -793,7 +795,7 @@ smdSortModelCel:
 	srl		$v1, $at, 8
 	srl		$v0, $v1
 	sb		$v0, POLYFT4_uv0+1($a1)
-	
+
 	mfc2	$t7, C2_RGB1
 	andi	$v1, $at, 0xff
 	andi	$t7, 0xffff
@@ -804,7 +806,7 @@ smdSortModelCel:
 	srl		$v1, $at, 8
 	srl		$v0, $v1
 	sb		$v0, POLYFT4_uv1+1($a1)
-	
+
 	mfc2	$t7, C2_RGB2
 	andi	$v1, $at, 0xff
 	andi	$t7, 0xffff
@@ -815,24 +817,24 @@ smdSortModelCel:
 	srl		$v1, $at, 8
 	srl		$v0, $v1
 	sb		$v0, POLYFT4_uv2+1($a1)
-	
+
 	la		$v0, _smd_cel_col
 	lw		$v0, 0($v0)
 	lui		$v1, 0x2E00
 	or		$v0, $v1
 	sw		$v0, POLYFT4_rgbc($a1)
-	
+
 	lw		$t7, 8($t4)
-	
+
 	lhu		$v0, 6($t4)			# Load normal 3
 	addiu	$t4, 12
 	sll		$v0, 3
 	addu	$v0, $t2
 	lwc2	C2_VXY0, 0( $v0 )
 	lwc2	C2_VZ0 , 4( $v0 )
-	
+
 	nNCS
-	
+
 	mfc2	$s0, C2_RGB2
 	andi	$v1, $at, 0xff
 	andi	$s0, 0xffff
@@ -843,7 +845,7 @@ smdSortModelCel:
 	srl		$v1, $at, 8
 	srl		$v0, $v1
 	sb		$v0, POLYFT4_uv3+1($a1)
-	
+
 	lui		$v1, 0x0900
 	lw		$v0, 0($t5)
 	lui		$at, 0xff00
@@ -859,24 +861,24 @@ smdSortModelCel:
 	and		$v0, $at
 	or		$v0, $v1
 	sw		$v0, 0($t5)
-	
+
 	lui		$v0, 0x8000
 	or		$a1, $v0
 	addiu	$a1, POLYFT4_len
-	
+
 	.set at
-	
+
 	andi	$v0, $a3, 0x20				# Textured quad
 	bnez	$v0, .F4_light_tex_smt
 	nop
-	
+
 	lui		$v0, 0x2800
 	or		$t7, $v0
 	b		.sort_F4_pri
 	sw		$t7, POLYF4_rgbc($a1)
-	
+
 .F4_light_tex_smt:
-	
+
 	lhu		$v0, 0($t4)				# Load texture coordinates
 	lhu		$v1, 2($t4)
 	sh		$v0, POLYFT4_uv0($a1)
@@ -887,25 +889,25 @@ smdSortModelCel:
 	sh		$v1, POLYFT4_uv3($a1)
 
 	lw		$v1, 8($t4)
-	
+
 	lui		$v0, 0x2E00
 	or		$t7, $v0
 	sw		$t7, POLYFT4_rgbc($a1)
-	
+
 	andi	$v0, $v1, 0xffff
 	sh		$v0, POLYFT4_tpage($a1)
 	srl		$v0, $v1, 16
-	
+
 	b		.sort_FT4_pri
 	sh		$v0, POLYFT4_clut($a1)
-	
+
 .sort_F4_pri:
 
 	sw		$t6, POLYF4_xy0($a1)
 	swc2	C2_SXY0, POLYF4_xy1($a1)
 	swc2	C2_SXY1, POLYF4_xy2($a1)
 	swc2	C2_SXY2, POLYF4_xy3($a1)
-	
+
 	la		$v0, _smd_tpage_base
 	lhu		$v0, 0($v0)
 	srl		$v1, $a3, 6				# Get blend mode
@@ -915,7 +917,7 @@ smdSortModelCel:
 	lui		$v1, 0xe100
 	or		$v0, $v1
 	sw		$v0, POLYF4_tpage($a1)	# Store TPage
-	
+
 	.set noat
 
 	lui		$v1, 0x0600
@@ -933,22 +935,22 @@ smdSortModelCel:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($t5)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYF4_len
-	
+
 .sort_FT4_pri:
 
 	sw		$t6, POLYFT4_xy0($a1)
 	swc2	C2_SXY0, POLYFT4_xy1($a1)
 	swc2	C2_SXY1, POLYFT4_xy2($a1)
 	swc2	C2_SXY2, POLYFT4_xy3($a1)
-	
+
 	.set noat
 
 	lui		$v1, 0x0900
@@ -966,22 +968,22 @@ smdSortModelCel:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($t5)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYFT4_len
-	
+
 .sort_G4_pri:
 
 	sw		$t6, POLYG4_xy0($a1)
 	swc2	C2_SXY0, POLYG4_xy1($a1)
 	swc2	C2_SXY1, POLYG4_xy2($a1)
 	swc2	C2_SXY2, POLYG4_xy3($a1)
-	
+
 	la		$v0, _smd_tpage_base
 	lhu		$v0, 0($v0)
 	srl		$v1, $a3, 6				# Get blend mode
@@ -991,7 +993,7 @@ smdSortModelCel:
 	lui		$v1, 0xe100
 	or		$v0, $v1
 	sw		$v0, POLYG4_tpage($a1)	# Store TPage
-	
+
 	.set noat
 
 	lui		$v1, 0x0900
@@ -1009,22 +1011,22 @@ smdSortModelCel:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($t5)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYG4_len
-	
+
 .sort_GT4_pri:
 
 	sw		$t6, POLYGT4_xy0($a1)
 	swc2	C2_SXY0, POLYGT4_xy1($a1)
 	swc2	C2_SXY1, POLYGT4_xy2($a1)
 	swc2	C2_SXY2, POLYGT4_xy3($a1)
-	
+
 	.set noat
 
 	lui		$v1, 0x0c00
@@ -1042,20 +1044,20 @@ smdSortModelCel:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($t5)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYGT4_len
-	
+
 .skip_prim:
 
 	b		.sort_loop
 	nop
-	
+
 .exit:
 
 	lw		$s0, 0( $sp )
@@ -1065,9 +1067,12 @@ smdSortModelCel:
 	addiu	$sp, 16
 	jr		$ra
 	move	$v0, $a1
-	
 
+.section .bss._smd_cel_col, "w"
 .comm _smd_cel_col, 4, 4	# STP shading polygon color
+
+.section .bss._smd_cel_param, "w"
 .comm _smd_cel_param, 4, 4	# U divisor, V divisor, shading clip
-.comm _smd_cel_tpage, 4, 4	# CEL shader texture page & CLUT
 
+.section .bss._smd_cel_tpage, "w"
+.comm _smd_cel_tpage, 4, 4	# CEL shader texture page & CLUT
diff --git a/examples/demos/n00bdemo/smd_flat.s b/libpsn00b/smd/smd_flat.s
index 4b47be4..843b8d3 100644
--- a/examples/demos/n00bdemo/smd_flat.s
+++ b/libpsn00b/smd/smd_flat.s
@@ -1,43 +1,45 @@
+# PSn00bSDK .SMD model parser library
+# (C) 2019-2023 Lameguy64, spicyjpeg - MPL licensed
+
 .set noreorder
 
 .include "gtereg.inc"
 .include "inline_s.inc"
-.include "smd_s.h"
-
-
-.section .text
+.include "smd/smd_s.inc"
 
+.section .text.smdSortModelFlat, "ax", @progbits
 .global smdSortModelFlat
 .type smdSortModelFlat, @function
+
 smdSortModelFlat:
 	# a0 - Pointer SC_OT structure
 	# a1 - Pointer to next primitive
 	# a2 - Pointer to SMD data address
 	# v0 - New pointer of primitive buffer (return)
-	
+
 	addiu	$sp, -16
 	sw		$s0, 0( $sp )
 	sw		$s1, 4( $sp )
 	sw		$s2, 8( $sp )
 	sw		$s3, 12( $sp )
-	
+
 	la		$v0, _sc_clip
 	lw		$t8, 0($v0)
 	lw		$t9, 4($v0)
-	
+
 	lw		$t1, SMD_HEAD_PVERTS( $a2 )
 	lw		$t2, SMD_HEAD_PNORMS( $a2 )
 	lw		$t3, SMD_HEAD_PPRIMS( $a2 )
-	
+
 .sort_loop:
 
 	nop
 	lw		$a3, 0($t3)				# Get primitive ID word
 	move	$t4, $t3
-	
+
 	beqz	$a3, .exit				# Check if terminator (just zero)
 	addiu	$t4, 4
-	
+
 	lhu		$t5, 0( $t4 )			# Load vertices
 	lhu		$t6, 2( $t4 )
 	lhu		$t7, 4( $t4 )
@@ -53,32 +55,32 @@ smdSortModelFlat:
 	lwc2	C2_VZ1 , 4( $t6 )
 	lwc2	C2_VXY2, 0( $t7 )
 	lwc2	C2_VZ2 , 4( $t7 )
-	
+
 	srl		$v1, $a3, 24			# Get primitive size
 	addu	$t3, $v1				# Step main pointer to next primitive
-	
+
 	RTPT
-	
+
 	cfc2	$v0, C2_FLAG			# Get GTE flag value
 	nop
-	
+
 	bltz	$v0, .skip_prim			# Skip primitive if Z overflow
 	nop
-	
+
 	#NCLIP							# Backface culling
-	
+
 	#mfc2	$v1, C2_MAC0
-	
+
 	andi	$v0, $a3, 0x3
-	
+
 	#bltz	$v1, .skip_prim
 	#nop
-	
+
 	beq		$v0, 0x1, .prim_tri		# If primitive is a triangle
 	nop
 	beq		$v0, 0x2, .prim_quad	# If primitive is a quad
 	nop
-	
+
 	b		.skip_prim
 	nop
 
@@ -89,9 +91,9 @@ smdSortModelFlat:
 	addiu	$t4, 8						# Advance from indices
 
 	#AVSZ3								# Calculate average Z
-	
+
 	ClipTestTri
-	
+
 	and		$v0, $s0, $s1				# v0 & v1
 	beqz	$v0, .do_draw
 	and		$v0, $s1, $s2				# v1 & v2
@@ -101,9 +103,9 @@ smdSortModelFlat:
 	nop
 	b		.skip_prim
 	nop
-	
+
 .do_draw:
-	
+
 	#srl		$v0, $a3, 2					# Lighting enabled?
 	#andi	$v0, 0x3
 	#bnez	$v0, .F3_light
@@ -112,18 +114,18 @@ smdSortModelFlat:
 	andi	$v0, $a3, 0x20				# Textured triangle
 	bnez	$v0, .F3_textured
 	nop
-	
+
 	andi	$v0, $a3, 0x10				# Gouraud shaded
 	bnez	$v0, .F3_gouraud
 	nop
-	
+
 	lw		$v0, 0( $t4 )				# Flat color, no lighting
 	lui		$v1, 0x2000
 	or		$v0, $v1
-	
+
 	b		.sort_F3_pri
 	sw		$v0, POLYF3_rgbc( $a1 )
-	
+
 .F3_textured:
 
 	lw		$v0, 0( $t4 )				# Flat color, no lighting
@@ -131,23 +133,23 @@ smdSortModelFlat:
 	or		$v0, $v1
 	sw		$v0, POLYFT3_rgbc( $a1 )
 	addiu	$t4, 4
-	
+
 	lhu		$v0, 0( $t4 )				# Load texture coordinates
 	lhu		$v1, 2( $t4 )
 	sh		$v0, POLYFT3_uv0( $a1 )
 	lhu		$v0, 4( $t4 )
 	sh		$v1, POLYFT3_uv1( $a1 )
 	sh		$v0, POLYFT3_uv2( $a1 )
-	
+
 	lw		$v0, 8( $t4 )				# Tpage + CLUT
 	nop
 	andi	$v1, $v0, 0xffff
 	sh		$v1, POLYFT3_tpage( $a1 )
 	srl		$v0, 16
-	
+
 	b		.sort_FT3_pri
 	sh		$v0, POLYFT3_clut( $a1 )
-	
+
 .F3_gouraud:
 	lw		$v0, 0($t4)
 	lw		$v1, 4($t4)
@@ -160,14 +162,14 @@ smdSortModelFlat:
 	sw		$v1, POLYG3_rgbc1($a1)
 	b		.sort_G3_pri
 	sw		$v0, POLYG3_rgbc2($a1)
-	
+
 #.F3_light:
-	
+
 	#lhu		$v0, 0( $t4 )				# Load normal 0
-	
+
 	#srl		$v1, $a3, 2
 	#andi	$v1, $v1, 0x3
-	
+
 	#sll		$v0, 3
 	#addu	$v0, $t2
 	#lwc2	C2_VXY0, 0( $v0 )
@@ -175,26 +177,26 @@ smdSortModelFlat:
 
 	#beq		$v1, 0x2, .F3_light_smt
 	#nop
-	
+
 	#lw		$v0, 4( $t4 )
 	#lui		$v1, 0x2000
 	#or		$v0, $v1
 	#mtc2	$v0, C2_RGB
-	
+
 	#addiu	$t4, 8
 	#nop
-	
+
 	#NCCS
 
 	#andi	$v0, $a3, 0x20				# Textured triangle
 	#bnez	$v0, .F3_light_tex
 	#nop
-	
+
 	#swc2	C2_RGB2, POLYF3_rgbc( $a1 )
-	
+
 	#b		.sort_F3_pri
 	#nop
-	
+
 #.F3_light_tex:
 
 	#lhu		$v0, 0( $t4 )				# Load texture coordinates
@@ -203,23 +205,23 @@ smdSortModelFlat:
 	#lhu		$v0, 4( $t4 )
 	#sh		$v1, POLYFT3_uv1( $a1 )
 	#sh		$v0, POLYFT3_uv2( $a1 )
-	
+
 	#lw		$v1, 8( $t4 )
 	#nop
 	#andi	$v0, $v1, 0xffff
 	#sh		$v0, POLYFT3_tpage( $a1 )
 	#srl		$v0, $v1, 16
 	#sh		$v0, POLYFT3_clut( $a1 )
-	
+
 	#mfc2	$v0, C2_RGB2
 	#lui		$v1, 0x2400
 	#or		$v0, $v1
-	
+
 	#b		.sort_FT3_pri
 	#sw		$v0, POLYFT3_rgbc( $a1 )
-	
+
 #.F3_light_smt:
-	
+
 	#lhu		$v0, 2( $t4 )			# Load normals 1 and 2
 	#lhu		$v1, 4( $t4 )
 	#sll		$v0, 3
@@ -234,20 +236,20 @@ smdSortModelFlat:
 	#lui		$v1, 0x3000				# Load color
 	#or		$v0, $v1
 	#mtc2	$v0, C2_RGB
-	
+
 	#addiu	$t4, 12
 	#nop
-	
+
 	#NCCT
-	
+
 	#andi	$v0, $a3, 0x20				# Textured triangle
 	#bnez	$v0, .F3_light_tex_smt
 	#nop
-	
+
 	#swc2	C2_RGB0, POLYG3_rgbc0( $a1 )
 	#swc2	C2_RGB1, POLYG3_rgbc1( $a1 )
 	#swc2	C2_RGB2, POLYG3_rgbc2( $a1 )
-	
+
 	#b		.sort_G3_pri
 	#nop
 
@@ -259,21 +261,21 @@ smdSortModelFlat:
 	# lhu		$v0, 4( $t4 )
 	# sh		$v1, POLYGT3_uv1( $a1 )
 	# sh		$v0, POLYGT3_uv2( $a1 )
-	
+
 	# lw		$v1, 8( $t4 )
 	# nop
 	# andi	$v0, $v1, 0xffff
 	# sh		$v0, POLYGT3_tpage( $a1 )
 	# srl		$v0, $v1, 16
 	# sh		$v0, POLYGT3_clut( $a1 )
-	
+
 	# mfc2	$v0, C2_RGB0
 	# lui		$v1, 0x3400
 	# or		$v0, $v1
-	
+
 	# swc2	C2_RGB1, POLYGT3_rgbc1( $a1 )
 	# swc2	C2_RGB2, POLYGT3_rgbc2( $a1 )
-	
+
 	# b		.sort_GT3_pri
 	# sw		$v0, POLYGT3_rgbc0( $a1 )
 
@@ -282,7 +284,7 @@ smdSortModelFlat:
 	swc2	C2_SXY0, POLYF3_xy0($a1)
 	swc2	C2_SXY1, POLYF3_xy1($a1)
 	swc2	C2_SXY2, POLYF3_xy2($a1)
-	
+
 	la		$v0, _smd_tpage_base
 	lhu		$v0, 0($v0)
 	srl		$v1, $a3, 6				# Get blend mode
@@ -292,7 +294,7 @@ smdSortModelFlat:
 	lui		$v1, 0xe100
 	or		$v0, $v1
 	sw		$v0, POLYF3_tpage($a1)	# Store TPage
-	
+
 	.set noat
 
 	lui		$v1, 0x0500
@@ -310,21 +312,21 @@ smdSortModelFlat:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($a0)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYF3_len
-	
+
 .sort_FT3_pri:
 
 	swc2	C2_SXY0, POLYFT3_xy0( $a1 )
 	swc2	C2_SXY1, POLYFT3_xy1( $a1 )
 	swc2	C2_SXY2, POLYFT3_xy2( $a1 )
-	
+
 	.set noat
 
 	lui		$v1, 0x0700
@@ -342,21 +344,21 @@ smdSortModelFlat:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($a0)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYFT3_len
-	
+
 .sort_G3_pri:
 
 	swc2	C2_SXY0, POLYG3_xy0( $a1 )
 	swc2	C2_SXY1, POLYG3_xy1( $a1 )
 	swc2	C2_SXY2, POLYG3_xy2( $a1 )
-	
+
 	la		$v0, _smd_tpage_base
 	lhu		$v0, 0($v0)
 	srl		$v1, $a3, 6				# Get blend mode
@@ -366,7 +368,7 @@ smdSortModelFlat:
 	lui		$v1, 0xe100
 	or		$v0, $v1
 	sw		$v0, POLYG3_tpage($a1)	# Store TPage
-	
+
 	.set noat
 
 	lui		$v1, 0x0700
@@ -384,21 +386,21 @@ smdSortModelFlat:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($a0)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYG3_len
-	
+
 .sort_GT3_pri:
 
 	swc2	C2_SXY0, POLYGT3_xy0( $a1 )
 	swc2	C2_SXY1, POLYGT3_xy1( $a1 )
 	swc2	C2_SXY2, POLYGT3_xy2( $a1 )
-	
+
 	.set noat
 
 	lui		$v1, 0x0900
@@ -416,17 +418,17 @@ smdSortModelFlat:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($a0)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYGT3_len
 
 ## Quads
-	
+
 .prim_quad:							# Quad processing
 
 	mfc2	$t6, C2_SXY0			# Retrieve first projected vertex
@@ -437,18 +439,18 @@ smdSortModelFlat:
 	addu	$t5, $t1
 	lwc2	C2_VXY0, 0( $t5 )
 	lwc2	C2_VZ0 , 4( $t5 )
-	
+
 	nRTPS
-	
+
 	cfc2	$v1, C2_FLAG			# Get GTE flag value
-	
+
 	nop
-	
+
 	bltz	$v1, .skip_prim
 	nop
-	
+
 	ClipTestQuad
-	
+
 	and		$v0, $s0, $s1				# v0 & v1
 	beqz	$v0, .do_draw_q
 	and		$v0, $s1, $s2				# v1 & v2
@@ -464,37 +466,37 @@ smdSortModelFlat:
 	nop
 	b		.skip_prim
 	nop
-	
+
 .do_draw_q:
 
 	srl		$v0, $a3, 2					# Lighting enabled?
 	andi	$v0, 0x3
 	bnez	$v0, .F4_light
 	nop
-	
+
 	andi	$v0, $a3, 0x10				# Gouraud quad
 	bnez	$v0, .F4_gouraud
 	nop
-	
+
 	andi	$v0, $a3, 0x20				# Textured quad
 	bnez	$v0, .F4_textured
 	nop
-	
+
 	lw		$v0, 0($t4)
 	lui		$v1, 0x2800
 	or		$v0, $v1
-	
+
 	b		.sort_F4_pri
 	sw		$v0, POLYF4_rgbc($a1)
 
 .F4_textured:
-	
+
 	lw		$v0, 0($t4)
 	lui		$v1, 0x2c00
 	or		$v0, $v1
 	sw		$v0, POLYFT4_rgbc( $a1 )
 	addiu	$t4, 4
-	
+
 	lhu		$v0, 0($t4)					# Load texture coordinates
 	lhu		$v1, 2($t4)
 	sh		$v0, POLYFT4_uv0( $a1 )
@@ -503,16 +505,16 @@ smdSortModelFlat:
 	lhu		$v1, 6( $t4 )
 	sh		$v0, POLYFT4_uv2( $a1 )
 	sh		$v1, POLYFT4_uv3( $a1 )
-	
+
 	lw		$v1, 8( $t4 )
 	nop
 	andi	$v0, $v1, 0xffff
 	sh		$v0, POLYFT4_tpage( $a1 )
 	srl		$v0, $v1, 16
-	
+
 	b		.sort_FT4_pri
 	sh		$v0, POLYFT4_clut($a1)
-	
+
 .F4_gouraud:
 
 	lw		$v0, 0($t4)
@@ -528,15 +530,15 @@ smdSortModelFlat:
 	sw		$v0, POLYG4_rgbc2($a1)
 	b		.sort_G4_pri
 	sw		$v1, POLYG4_rgbc3($a1)
-	
+
 
 .F4_light:
 
 	lhu		$v0, 0( $t4 )				# Load normal 0
-	
+
 	srl		$v1, $a3, 2
 	andi	$v1, $v1, 0x3
-	
+
 	sll		$v0, 3
 	addu	$v0, $t2
 	lwc2	C2_VXY0, 0( $v0 )
@@ -544,26 +546,26 @@ smdSortModelFlat:
 
 	beq		$v1, 0x2, .F4_light_smt
 	nop
-	
+
 	lw		$v0, 4( $t4 )
 	lui		$v1, 0x2800
 	or		$v0, $v1
 	mtc2	$v0, C2_RGB
-	
+
 	addiu	$t4, 8
 	nop
-	
+
 	NCCS
-	
+
 	andi	$v0, $a3, 0x20				# Textured triangle
 	bnez	$v0, .F4_light_tex
 	nop
 
 	swc2	C2_RGB2, POLYF4_rgbc( $a1 )
-	
+
 	b		.sort_F4_pri
 	nop
-	
+
 .F4_light_tex:
 
 	lhu		$v0, 0( $t4 )				# Load texture coordinates
@@ -574,21 +576,21 @@ smdSortModelFlat:
 	lhu		$v1, 6( $t4 )
 	sh		$v0, POLYFT4_uv2( $a1 )
 	sh		$v1, POLYFT4_uv3( $a1 )
-	
+
 	lw		$v1, 8( $t4 )
 	nop
 	andi	$v0, $v1, 0xffff
 	sh		$v0, POLYFT4_tpage( $a1 )
 	srl		$v0, $v1, 16
 	sh		$v0, POLYFT4_clut( $a1 )
-	
+
 	mfc2	$v0, C2_RGB2
 	lui		$v1, 0x2c00
 	or		$v0, $v1
-	
+
 	b		.sort_FT4_pri
 	sw		$v0, POLYFT4_rgbc( $a1 )
-	
+
 .F4_light_smt:
 
 	lhu		$v0, 2( $t4 )			# Load normals 1 and 2
@@ -601,38 +603,38 @@ smdSortModelFlat:
 	lwc2	C2_VZ1 , 4( $v0 )
 	lwc2	C2_VXY2, 0( $v1 )
 	lwc2	C2_VZ2 , 4( $v1 )
-	
+
 	lw		$v0, 8( $t4 )
 	lui		$v1, 0x3800				# Load color
 	or		$v0, $v1
 	mtc2	$v0, C2_RGB
-	
+
 	nNCCT
-	
+
 	lhu		$v0, 6( $t4 )			# Load normal 3
-	
+
 	addiu	$t4, 12
-	
+
 	sll		$v0, 3
 	addu	$v0, $t2
 	lwc2	C2_VXY0, 0( $v0 )
 	lwc2	C2_VZ0 , 4( $v0 )
-	
+
 	andi	$v0, $a3, 0x20				# Textured triangle
 	bnez	$v0, .F4_light_tex_smt
 	nop
-	
+
 	swc2	C2_RGB0, POLYG4_rgbc0( $a1 )
 	swc2	C2_RGB1, POLYG4_rgbc1( $a1 )
 	swc2	C2_RGB2, POLYG4_rgbc2( $a1 )
-	
+
 	nNCCS
-	
+
 	swc2	C2_RGB2, POLYG4_rgbc3( $a1 )
-	
+
 	b		.sort_G4_pri
 	nop
-	
+
 .F4_light_tex_smt:
 
 	mfc2	$v0, C2_RGB0
@@ -641,9 +643,9 @@ smdSortModelFlat:
 	sw		$v0, POLYGT4_rgbc0( $a1 )
 	swc2	C2_RGB1, POLYGT4_rgbc1( $a1 )
 	swc2	C2_RGB2, POLYGT4_rgbc2( $a1 )
-	
+
 	NCCS
-	
+
 	lhu		$v0, 0( $t4 )				# Load texture coordinates
 	lhu		$v1, 2( $t4 )
 	sh		$v0, POLYGT4_uv0( $a1 )
@@ -655,21 +657,21 @@ smdSortModelFlat:
 
 	lw		$v1, 8( $t4 )
 	swc2	C2_RGB2, POLYGT4_rgbc3( $a1 )
-	
+
 	andi	$v0, $v1, 0xffff
 	sh		$v0, POLYGT4_tpage( $a1 )
 	srl		$v0, $v1, 16
-	
+
 	b		.sort_GT4_pri
 	sh		$v0, POLYGT4_clut( $a1 )
-	
+
 .sort_F4_pri:
 
 	sw		$t6, POLYF4_xy0($a1)
 	swc2	C2_SXY0, POLYF4_xy1($a1)
 	swc2	C2_SXY1, POLYF4_xy2($a1)
 	swc2	C2_SXY2, POLYF4_xy3($a1)
-	
+
 	la		$v0, _smd_tpage_base
 	lhu		$v0, 0($v0)
 	srl		$v1, $a3, 6				# Get blend mode
@@ -679,7 +681,7 @@ smdSortModelFlat:
 	lui		$v1, 0xe100
 	or		$v0, $v1
 	sw		$v0, POLYF4_tpage($a1)	# Store TPage
-	
+
 	.set noat
 
 	lui		$v1, 0x0600
@@ -697,22 +699,22 @@ smdSortModelFlat:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($a0)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYF4_len
-	
+
 .sort_FT4_pri:
 
 	sw		$t6, POLYFT4_xy0($a1)
 	swc2	C2_SXY0, POLYFT4_xy1($a1)
 	swc2	C2_SXY1, POLYFT4_xy2($a1)
 	swc2	C2_SXY2, POLYFT4_xy3($a1)
-	
+
 	.set noat
 
 	lui		$v1, 0x0900
@@ -730,22 +732,22 @@ smdSortModelFlat:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($a0)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYFT4_len
-	
+
 .sort_G4_pri:
 
 	sw		$t6, POLYG4_xy0($a1)
 	swc2	C2_SXY0, POLYG4_xy1($a1)
 	swc2	C2_SXY1, POLYG4_xy2($a1)
 	swc2	C2_SXY2, POLYG4_xy3($a1)
-	
+
 	la		$v0, _smd_tpage_base
 	lhu		$v0, 0($v0)
 	srl		$v1, $a3, 6				# Get blend mode
@@ -755,7 +757,7 @@ smdSortModelFlat:
 	lui		$v1, 0xe100
 	or		$v0, $v1
 	sw		$v0, POLYG4_tpage($a1)	# Store TPage
-	
+
 	.set noat
 
 	lui		$v1, 0x0900
@@ -773,22 +775,22 @@ smdSortModelFlat:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($a0)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYG4_len
-	
+
 .sort_GT4_pri:
 
 	sw		$t6, POLYGT4_xy0($a1)
 	swc2	C2_SXY0, POLYGT4_xy1($a1)
 	swc2	C2_SXY1, POLYGT4_xy2($a1)
 	swc2	C2_SXY2, POLYGT4_xy3($a1)
-	
+
 	.set noat
 
 	lui		$v1, 0x0c00
@@ -806,20 +808,20 @@ smdSortModelFlat:
 	and		$v0, $at
 	or		$v0, $a1
 	sw		$v0, 0($a0)
-	
+
 	.set at
 
 	lui		$v0, 0x8000
 	or		$a1, $v0
-	
+
 	b		.sort_loop
 	addiu	$a1, POLYGT4_len
-	
+
 .skip_prim:
 
 	b		.sort_loop
 	nop
-	
+
 .exit:
 
 	lw		$s0, 0( $sp )
@@ -829,4 +831,3 @@ smdSortModelFlat:
 	addiu	$sp, 16
 	jr		$ra
 	move	$v0, $a1
-	
\ No newline at end of file
diff --git a/examples/demos/n00bdemo/smdparser.s b/libpsn00b/smd/smdparser.s
index 379690f..656e509 100644
--- a/examples/demos/n00bdemo/smdparser.s
+++ b/libpsn00b/smd/smdparser.s
@@ -1,6 +1,9 @@
+# PSn00bSDK .SMD model parser library
+# (C) 2019-2023 Lameguy64, spicyjpeg - MPL licensed
+
 .set noreorder
 
-.include "smd_s.h"
+.include "smd/smd_s.inc"
 
 .set SMD_PRI_ID,		0
 .set SMD_PRI_v0,		4
@@ -22,108 +25,106 @@
 .set SMD_PRI_tpage,		44
 .set SMD_PRI_clut,		46
 
-.section .text
-
-
-.comm _smd_parse_addr, 4, 4
-
-
+.section .text.OpenSMD, "ax", @progbits
 .global OpenSMD
 .type OpenSMD, @function
+
 OpenSMD:
 	lw		$v0, SMD_HEAD_ID($a0)
 	li		$v1, 0x01444d53
-	
+
 	bne		$v0, $v1, .not_smd
 	nop
-	
+
 	lw		$v0, SMD_HEAD_PPRIMS($a0)
 	la		$v1, _smd_parse_addr
 	sw		$v0, 0($v1)
-	
+
 	jr		$ra
 	lhu		$v0, SMD_HEAD_NPRIMS($a0)
-	
+
 .not_smd:
 	jr		$ra
 	move	$v0, $0
-	
-	
+
+.section .text.ReadSMD, "ax", @progbits
 .global ReadSMD
 .type ReadSMD, @function
-ReadSMD:
 
+ReadSMD:
 	la		$v0, _smd_parse_addr
 	lw		$v0, 0($v0)
 	nop
-	
+
 	lw		$a2, 0($v0)				# Load primitive ID
 	addiu	$a1, $v0, 4
-	
+
 	sw		$a2, SMD_PRI_ID($a0)
-	
+
 	beqz	$a2, $end_prim
 	nop
-	
+
 	srl		$v1, $a2, 24			# Get primitive size
 	addu	$v0, $v1
 	la		$v1, _smd_parse_addr
 	sw		$v0, 0($v1)
-	
+
 	lw		$v0, 0($a1)				# Copy vertex coords
 	lw		$v1, 4($a1)
 	sw		$v0, SMD_PRI_v0($a0)
 	sw		$v1, SMD_PRI_v2($a0)
 	addiu	$a1, 8
-	
+
 	srl		$v0, $a2, 2				# Lighting enabled?
 	andi	$v0, 0x3
 	bnez	$v0, $light
 	nop
-	
+
 	b		$no_light
 	nop
-	
+
 $light:
 	srl		$v1, $a2, 2
 	lw		$v0, 0($a1)				# Copy vertex coords
 	andi	$v1, 0x3
 	sw		$v0, SMD_PRI_n0($a0)
-	
+
 	bne		$v1, 0x2, $light_flat
 	addiu	$a1, 4
-	
+
 	lw		$v1, 0($a1)
 	addiu	$a1, 4
 	sw		$v1, SMD_PRI_n2($a0)
 
 $light_flat:
 $no_light:
-	
+
 	lw		$v0, 0($a1)
 	nop
 	sw		$v0, SMD_PRI_rgbc0($a0)
 	addiu	$a1, 4
-	
+
 	srl		$v0, $a2, 5
 	andi	$v0, 0x1
 	beqz	$v0, $not_textured
 	nop
-	
+
 	lw		$v0, 0($a1)
 	lw		$v1, 4($a1)
 	sw		$v0, SMD_PRI_tuv0($a0)
 	lw		$v0, 8($a1)
 	sw		$v1, SMD_PRI_tuv2($a0)
 	sw		$v0, SMD_PRI_tpage($a0)
-	
+
 $not_textured:
-	
+
 	jr		$ra
 	move	$v0, $a0
-	
+
 $end_prim:
 
 	jr		$ra
 	move	$v0, $0
-	
\ No newline at end of file
+
+.section .bss._smd_parse_addr, "w"
+.comm _smd_parse_addr, 4, 4