;
; Mesa 3-D graphics library
; Version:  2.5
; Copyright (C) 1995-1997  Brian Paul
;
; This library is free software; you can redistribute it and/or
; modify it under the terms of the GNU Library General Public
; License as published by the Free Software Foundation; either
; version 2 of the License, or (at your option) any later version.
;
; This library is distributed in the hope that it will be useful,
; but WITHOUT ANY WARRANTY; without even the implied warranty of
; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
; Library General Public License for more details.
;
; You should have received a copy of the GNU Library General Public
; License along with this library; if not, write to the Free
; Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
;

;   vbxform_asmppc.p
;   16.5.1998 by Sam Jordan
;
;   PowerPC assembler optimizations of several functions in vbxform.c
;   Originally written for AMIGA OS/PowerOpen. To use this source on other
;   PowerPC based platforms or with other programming models, some
;   modifications might be needed.

CLIP_RIGHT_BIT  =       $01
CLIP_LEFT_BIT   =       $02
CLIP_TOP_BIT    =       $04
CLIP_BOTTOM_BIT =       $08
CLIP_NEAR_BIT   =       $10
CLIP_FAR_BIT    =       $20


                XDEF    _asm_transform_points3_general
                XDEF    _asm_transform_points3_identity
                XDEF    _asm_transform_points3_2d
                XDEF    _asm_transform_points3_2d_no_rot
                XDEF    _asm_transform_points3_3d
                XDEF    _asm_transform_points4_general
                XDEF    _asm_transform_points4_identity
                XDEF    _asm_transform_points4_2d
                XDEF    _asm_transform_points4_2d_no_rot
                XDEF    _asm_transform_points4_3d
                XDEF    _asm_project_and_cliptest_general
                XDEF    _asm_project_and_cliptest_identity
                XDEF    _asm_project_and_cliptest_ortho
                XDEF    _asm_project_and_cliptest_perspective

                XDEF    _asm_vp_map_vertices_now
                XDEF    _asm_vp_map_vertices



_asm_transform_points3_general

;            const GLfloat *m = ctx->ModelViewMatrix;
;            GLfloat m0 = m[0],  m4 = m[4],  m8 = m[8],  m12 = m[12];
;            GLfloat m1 = m[1],  m5 = m[5],  m9 = m[9],  m13 = m[13];
;            GLfloat m2 = m[2],  m6 = m[6],  m10 = m[10],  m14 = m[14];
;            GLfloat m3 = m[3],  m7 = m[7],  m11 = m[11],  m15 = m[15];
;            GLuint i;
;            for (i=0;i<n;i++) {
;               GLfloat ox = vObj[i][0], oy = vObj[i][1], oz = vObj[i][2];
;               vEye[i][0] = m0 * ox + m4 * oy + m8  * oz + m12;
;               vEye[i][1] = m1 * ox + m5 * oy + m9  * oz + m13;
;               vEye[i][2] = m2 * ox + m6 * oy + m10 * oz + m14;
;               vEye[i][3] = m3 * ox + m7 * oy + m11 * oz + m15;
;            }


                stfd    f14,-1*8(r1)
                stfd    f15,-2*8(r1)
                stfd    f16,-3*8(r1)
                stfd    f17,-4*8(r1)
                stfd    f18,-5*8(r1)
                stfd    f19,-6*8(r1)
                stfd    f20,-7*8(r1)
                stfd    f21,-8*8(r1)
                stfd    f22,-9*8(r1)
                mr.     r3,r3
                beq     .end
                subi    r5,r5,4
                lfsu    f0,4(r5)
                lfsu    f1,4(r5)
                lfsu    f2,4(r5)
                lfsu    f3,4(r5)
                lfsu    f4,4(r5)
                lfsu    f5,4(r5)
                lfsu    f6,4(r5)
                lfsu    f7,4(r5)
                lfsu    f8,4(r5)
                lfsu    f9,4(r5)
                lfsu    f10,4(r5)
                lfsu    f11,4(r5)
                lfsu    f12,4(r5)
                lfsu    f13,4(r5)
                lfsu    f14,4(r5)
                lfsu    f15,4(r5)
                mtctr   r3
                subi    r6,r6,4
                subi    r4,r4,4
.loop
                lfsu    f16,4(r6)               ;f16 = ox
                fmadds  f19,f0,f16,f12
                fmadds  f20,f1,f16,f13
                lfsu    f17,4(r6)               ;f17 = oy
                fmadds  f21,f2,f16,f14
                fmadds  f22,f3,f16,f15
                fmadds  f19,f4,f17,f19
                fmadds  f20,f5,f17,f20
                lfsu    f18,4(r6)               ;f18 = oz
                fmadds  f21,f6,f17,f21
                fmadds  f22,f7,f17,f22
                addi    r6,r6,4
                fmadds  f19,f8,f18,f19
                stfsu   f19,4(r4)
                fmadds  f20,f9,f18,f20
                stfsu   f20,4(r4)
                fmadds  f21,f10,f18,f21
                stfsu   f21,4(r4)
                fmadds  f22,f11,f18,f22
                stfsu   f22,4(r4)
                bdnz    .loop
.end
                lfd     f22,-9*8(r1)
                lfd     f21,-8*8(r1)
                lfd     f20,-7*8(r1)
                lfd     f19,-6*8(r1)
                lfd     f18,-5*8(r1)
                lfd     f17,-4*8(r1)
                lfd     f16,-3*8(r1)
                lfd     f15,-2*8(r1)
                lfd     f14,-1*8(r1)
                blr





_asm_transform_points3_identity

;            GLuint i;
;            for (i=0;i<n;i++) {
;               vEye[i][0] = vObj[i][0];
;               vEye[i][1] = vObj[i][1];
;               vEye[i][2] = vObj[i][2];
;               vEye[i][3] = 1.0F;
;            }

                mr.     r3,r3
                beq     .end
                subi    r4,r4,4
                subi    r5,r5,4
                mtctr   r3
                lis     r6,$3f80
.loop
                lwzu    r0,4(r5)
                stwu    r0,4(r4)
                lwzu    r0,4(r5)
                stwu    r0,4(r4)
                lwzu    r0,4(r5)
                stwu    r0,4(r4)
                addi    r5,r5,4
                stwu    r6,4(r4)
                bdnz    .loop
.end
                blr





_asm_transform_points3_2d


;            const GLfloat *m = ctx->ModelViewMatrix;
;            GLfloat m0 = m[0], m1 = m[1], m4 = m[4], m5 = m[5];
;            GLfloat m12 = m[12], m13 = m[13];
;            GLuint i;
;            for (i=0;i<n;i++) {
;               GLfloat ox = vObj[i][0], oy = vObj[i][1], oz = vObj[i][2];
;               vEye[i][0] = m0 * ox + m4 * oy            + m12       ;
;               vEye[i][1] = m1 * ox + m5 * oy            + m13       ;
;               vEye[i][2] =                   +       oz             ;
;               vEye[i][3] =                                      1.0F;
;            }

                mr.     r3,r3
                beq     .end
                lfs     f0,0*4(r5)
                lfs     f1,1*4(r5)
                lfs     f4,4*4(r5)
                lfs     f5,5*4(r5)
                lfs     f12,12*4(r5)
                lfs     f13,13*4(r5)
                mtctr   r3
                subi    r6,r6,4
                subi    r4,r4,4
                lis     r7,$3f80
.loop
                lfsu    f6,4(r6)                ;f6 = ox
                fmadds  f2,f0,f6,f12
                fmadds  f3,f1,f6,f13
                lfsu    f7,4(r6)                ;f7 = oy
                fmadds  f2,f4,f7,f2
                lfsu    f8,4(r6)                ;f8 = oz
                fmadds  f3,f5,f7,f3
                stfsu   f2,4(r4)
                stfsu   f3,4(r4)
                stfsu   f8,4(r4)
                addi    r6,r6,4
                stwu    r7,4(r4)
                bdnz    .loop
.end
                blr





_asm_transform_points3_2d_no_rot

;            const GLfloat *m = ctx->ModelViewMatrix;
;            GLfloat m0 = m[0], m5 = m[5], m12 = m[12], m13 = m[13];
;            GLuint i;
;            for (i=0;i<n;i++) {
;               GLfloat ox = vObj[i][0], oy = vObj[i][1], oz = vObj[i][2];
;               vEye[i][0] = m0 * ox                      + m12       ;
;               vEye[i][1] =           m5 * oy            + m13       ;
;               vEye[i][2] =                   +       oz             ;
;               vEye[i][3] =                                      1.0F;
;            }

                mr.     r3,r3
                beq     .end
                lfs     f0,0*4(r5)
                lfs     f5,5*4(r5)
                lfs     f12,12*4(r5)
                lfs     f13,13*4(r5)
                mtctr   r3
                subi    r6,r6,4
                subi    r4,r4,4
                lis     r7,$3f80
.loop
                lfsu    f6,4(r6)                ;f6 = ox
                fmadds  f2,f0,f6,f12
                lfsu    f7,4(r6)                ;f7 = oy
                fmadds  f3,f5,f7,f13
                lfsu    f8,4(r6)                ;f7 = oy
                stfsu   f2,4(r4)
                stfsu   f3,4(r4)
                stfsu   f8,4(r4)
                addi    r6,r6,4
                stwu    r7,4(r4)
                bdnz    .loop
.end
                blr





_asm_transform_points3_3d


;            const GLfloat *m = ctx->ModelViewMatrix;
;            GLfloat m0 = m[0], m1 = m[1], m2 = m[2], m4 = m[4], m5 = m[5];
;            GLfloat m6 = m[6], m8 = m[8], m9 = m[9], m10 = m[10];
;            GLfloat m12 = m[12], m13 = m[13], m14 = m[14];
;            GLuint i;
;            for (i=0;i<n;i++) {
;               GLfloat ox = vObj[i][0], oy = vObj[i][1], oz = vObj[i][2];
;               vEye[i][0] = m0 * ox + m4 * oy +  m8 * oz + m12       ;
;               vEye[i][1] = m1 * ox + m5 * oy +  m9 * oz + m13       ;
;               vEye[i][2] = m2 * ox + m6 * oy + m10 * oz + m14       ;
;               vEye[i][3] =                                      1.0F;
;            }

                stfd    f14,-1*8(r1)
                stfd    f15,-2*8(r1)
                stfd    f16,-3*8(r1)
                stfd    f17,-4*8(r1)
                mr.     r3,r3
                beq     .end
                subi    r5,r5,4
                lfsu    f0,4(r5)
                lfsu    f1,4(r5)
                lfsu    f2,4(r5)
                addi    r5,r5,4
                lfsu    f4,4(r5)
                lfsu    f5,4(r5)
                lfsu    f6,4(r5)
                addi    r5,r5,4
                lfsu    f8,4(r5)
                lfsu    f9,4(r5)
                lfsu    f10,4(r5)
                addi    r5,r5,4
                lfsu    f12,4(r5)
                lfsu    f13,4(r5)
                lfsu    f14,4(r5)
                mtctr   r3
                subi    r6,r6,4
                subi    r4,r4,4
                lis     r7,$3f80
.loop
                lfsu    f3,4(r6)                ;f3 = ox
                fmadds  f15,f0,f3,f12
                fmadds  f16,f1,f3,f13
                lfsu    f7,4(r6)                ;f7 = oy
                fmadds  f17,f2,f3,f14
                fmadds  f15,f4,f7,f15
                lfsu    f11,4(r6)               ;f11 = oz
                fmadds  f16,f5,f7,f16
                fmadds  f17,f6,f7,f17
                fmadds  f15,f8,f11,f15
                stfsu   f15,4(r4)
                fmadds  f16,f9,f11,f16
                addi    r6,r6,4
                stfsu   f16,4(r4)
                fmadds  f17,f10,f11,f17
                stfsu   f17,4(r4)
                stwu    r7,4(r4)
                bdnz    .loop
.end
                lfd     f17,-4*8(r1)
                lfd     f16,-3*8(r1)
                lfd     f15,-2*8(r1)
                lfd     f14,-1*8(r1)
                blr







_asm_transform_points4_general

;            const GLfloat *m = ctx->ModelViewMatrix;
;            GLfloat m0 = m[0],  m4 = m[4],  m8 = m[8],  m12 = m[12];
;            GLfloat m1 = m[1],  m5 = m[5],  m9 = m[9],  m13 = m[13];
;            GLfloat m2 = m[2],  m6 = m[6],  m10 = m[10],  m14 = m[14];
;            GLfloat m3 = m[3],  m7 = m[7],  m11 = m[11],  m15 = m[15];
;            GLuint i;
;            for (i=0;i<n;i++) {
;               GLfloat ox = vObj[i][0], oy = vObj[i][1];
;               GLfloat oz = vObj[i][2], ow = vObj[i][3];
;               vEye[i][0] = m0 * ox + m4 * oy + m8  * oz + m12 * ow;
;               vEye[i][1] = m1 * ox + m5 * oy + m9  * oz + m13 * ow;
;               vEye[i][2] = m2 * ox + m6 * oy + m10 * oz + m14 * ow;
;               vEye[i][3] = m3 * ox + m7 * oy + m11 * oz + m15 * ow;
;            }

                stfd    f14,-1*8(r1)
                stfd    f15,-2*8(r1)
                stfd    f16,-3*8(r1)
                stfd    f17,-4*8(r1)
                stfd    f18,-5*8(r1)
                stfd    f19,-6*8(r1)
                stfd    f20,-7*8(r1)
                stfd    f21,-8*8(r1)
                stfd    f22,-9*8(r1)
                stfd    f23,-10*8(r1)
                mr.     r3,r3
                beq     .end
                subi    r5,r5,4
                lfsu    f0,4(r5)
                lfsu    f1,4(r5)
                lfsu    f2,4(r5)
                lfsu    f3,4(r5)
                lfsu    f4,4(r5)
                lfsu    f5,4(r5)
                lfsu    f6,4(r5)
                lfsu    f7,4(r5)
                lfsu    f8,4(r5)
                lfsu    f9,4(r5)
                lfsu    f10,4(r5)
                lfsu    f11,4(r5)
                lfsu    f12,4(r5)
                lfsu    f13,4(r5)
                lfsu    f14,4(r5)
                lfsu    f15,4(r5)
                mtctr   r3
                subi    r6,r6,4
                subi    r4,r4,4
.loop
                lfsu    f16,4(r6)               ;f16 = ox
                fmuls   f19,f0,f16
                fmuls   f20,f1,f16
                lfsu    f17,4(r6)               ;f17 = oy
                fmuls   f21,f2,f16
                fmuls   f22,f3,f16
                fmadds  f19,f4,f17,f19
                lfsu    f18,4(r6)               ;f18 = oz
                fmadds  f20,f5,f17,f20
                fmadds  f21,f6,f17,f21
                fmadds  f22,f7,f17,f22
                lfsu    f23,4(r6)               ;f23 = ow
                fmadds  f19,f8,f18,f19
                fmadds  f20,f9,f18,f20
                fmadds  f21,f10,f18,f21
                fmadds  f22,f11,f18,f22
                fmadds  f19,f12,f23,f19
                stfsu   f19,4(r4)
                fmadds  f20,f13,f23,f20
                stfsu   f20,4(r4)
                fmadds  f21,f14,f23,f21
                stfsu   f21,4(r4)
                fmadds  f22,f15,f23,f22
                stfsu   f22,4(r4)
                bdnz    .loop
.end
                lfd     f23,-10*8(r1)
                lfd     f22,-9*8(r1)
                lfd     f21,-8*8(r1)
                lfd     f20,-7*8(r1)
                lfd     f19,-6*8(r1)
                lfd     f18,-5*8(r1)
                lfd     f17,-4*8(r1)
                lfd     f16,-3*8(r1)
                lfd     f15,-2*8(r1)
                lfd     f14,-1*8(r1)
                blr





_asm_transform_points4_identity

;            GLuint i;
;            for (i=0;i<n;i++) {
;               vEye[i][0] = vObj[i][0];
;               vEye[i][1] = vObj[i][1];
;               vEye[i][2] = vObj[i][2];
;               vEye[i][3] = vObj[i][3];
;            }

                mr.     r3,r3
                beq     .end
                subi    r4,r4,4
                subi    r5,r5,4
                mtctr   r3
.loop
                lwzu    r0,4(r5)
                stwu    r0,4(r4)
                lwzu    r0,4(r5)
                stwu    r0,4(r4)
                lwzu    r0,4(r5)
                stwu    r0,4(r4)
                lwzu    r0,4(r5)
                stwu    r0,4(r4)
                bdnz    .loop
.end
                blr





_asm_transform_points4_2d


;            const GLfloat *m = ctx->ModelViewMatrix;
;            GLfloat m0 = m[0], m1 = m[1], m4 = m[4], m5 = m[5];
;            GLfloat m12 = m[12], m13 = m[13];
;            GLuint i;
;            for (i=0;i<n;i++) {
;               GLfloat ox = vObj[i][0], oy = vObj[i][1];
;               GLfloat oz = vObj[i][2], ow = vObj[i][3];
;               vEye[i][0] = m0 * ox + m4 * oy            + m12 * ow;
;               vEye[i][1] = m1 * ox + m5 * oy            + m13 * ow;
;               vEye[i][2] =                   +       oz           ;
;               vEye[i][3] =                                      ow;

                mr.     r3,r3
                beq     .end
                lfs     f0,0*4(r5)
                lfs     f1,1*4(r5)
                lfs     f4,4*4(r5)
                lfs     f5,5*4(r5)
                lfs     f12,12*4(r5)
                lfs     f13,13*4(r5)
                mtctr   r3
                subi    r6,r6,4
                subi    r4,r4,4
.loop
                lfsu    f6,4(r6)                ;f6 = ox
                fmuls   f2,f0,f6
                lfsu    f7,4(r6)                ;f7 = oy
                fmuls   f3,f1,f6
                lfsu    f8,4(r6)                ;f8 = oz
                fmadds  f2,f4,f7,f2
                lfsu    f9,4(r6)                ;f9 = ow
                fmadds  f3,f5,f7,f3
                fmadds  f2,f12,f9,f2
                stfsu   f2,4(r4)
                fmadds  f3,f13,f9,f3
                stfsu   f3,4(r4)
                stfsu   f8,4(r4)
                stfsu   f9,4(r4)
                bdnz    .loop
.end
                blr





_asm_transform_points4_2d_no_rot

;            const GLfloat *m = ctx->ModelViewMatrix;
;            GLfloat m0 = m[0], m5 = m[5], m12 = m[12], m13 = m[13];
;            GLuint i;
;            for (i=0;i<n;i++) {
;               GLfloat ox = vObj[i][0], oy = vObj[i][1];
;               GLfloat oz = vObj[i][2], ow = vObj[i][3];
;               vEye[i][0] = m0 * ox                      + m12 * ow;
;               vEye[i][1] =           m5 * oy            + m13 * ow;
;               vEye[i][2] =                   +       oz           ;
;               vEye[i][3] =                                      ow;
;            }

                mr.     r3,r3
                beq     .end
                lfs     f0,0*4(r5)
                lfs     f5,5*4(r5)
                lfs     f12,12*4(r5)
                lfs     f13,13*4(r5)
                mtctr   r3
                subi    r6,r6,4
                subi    r4,r4,4
.loop
                lfsu    f6,4(r6)                ;f6 = ox
                lfsu    f7,4(r6)                ;f7 = oy
                fmuls   f2,f0,f6
                lfsu    f8,4(r6)                ;f7 = oy
                fmuls   f3,f5,f7
                lfsu    f9,4(r6)                ;f9 = ow
                fmadds  f2,f12,f9,f2
                stfsu   f2,4(r4)
                fmadds  f3,f13,f9,f3
                stfsu   f3,4(r4)
                stfsu   f8,4(r4)
                stfsu   f9,4(r4)
                bdnz    .loop
.end
                blr





_asm_transform_points4_3d

;            const GLfloat *m = ctx->ModelViewMatrix;
;            GLfloat m0 = m[0], m1 = m[1], m2 = m[2], m4 = m[4], m5 = m[5];
;            GLfloat m6 = m[6], m8 = m[8], m9 = m[9], m10 = m[10];
;            GLfloat m12 = m[12], m13 = m[13], m14 = m[14];
;            GLuint i;
;            for (i=0;i<n;i++) {
;               GLfloat ox = vObj[i][0], oy = vObj[i][1];
;               GLfloat oz = vObj[i][2], ow = vObj[i][3];
;               vEye[i][0] = m0 * ox + m4 * oy +  m8 * oz + m12 * ow;
;               vEye[i][1] = m1 * ox + m5 * oy +  m9 * oz + m13 * ow;
;               vEye[i][2] = m2 * ox + m6 * oy + m10 * oz + m14 * ow;
;               vEye[i][3] =                                      ow;
;            }

                stfd    f14,-1*8(r1)
                stfd    f15,-2*8(r1)
                stfd    f16,-3*8(r1)
                stfd    f17,-4*8(r1)
                stfd    f18,-5*8(r1)
                mr.     r3,r3
                beq     .end
                subi    r5,r5,4
                lfsu    f0,4(r5)
                lfsu    f1,4(r5)
                lfsu    f2,4(r5)
                addi    r5,r5,4
                lfsu    f4,4(r5)
                lfsu    f5,4(r5)
                lfsu    f6,4(r5)
                addi    r5,r5,4
                lfsu    f8,4(r5)
                lfsu    f9,4(r5)
                lfsu    f10,4(r5)
                addi    r5,r5,4
                lfsu    f12,4(r5)
                lfsu    f13,4(r5)
                lfsu    f14,4(r5)
                mtctr   r3
                subi    r6,r6,4
                subi    r4,r4,4
.loop
                lfsu    f3,4(r6)                ;f3 = ox
                fmuls   f15,f0,f3
                fmuls   f16,f1,f3
                lfsu    f7,4(r6)                ;f7 = oy
                fmuls   f17,f2,f3
                fmadds  f15,f4,f7,f15
                lfsu    f11,4(r6)               ;f11 = oz
                fmadds  f16,f5,f7,f16
                fmadds  f17,f6,f7,f17
                fmadds  f15,f8,f11,f15
                lfsu    f18,4(r6)               ;f18 = ow
                fmadds  f16,f9,f11,f16
                fmadds  f17,f10,f11,f17
                fmadds  f15,f12,f18,f15
                stfsu   f15,4(r4)
                fmadds  f16,f13,f18,f16
                stfsu   f16,4(r4)
                fmadds  f17,f14,f18,f17
                stfsu   f17,4(r4)
                stfsu   f18,4(r4)
                bdnz    .loop
.end
                lfd     f18,-5*8(r1)
                lfd     f17,-4*8(r1)
                lfd     f16,-3*8(r1)
                lfd     f15,-2*8(r1)
                lfd     f14,-1*8(r1)
                blr




_asm_project_and_cliptest_general

;            const GLfloat *m = ctx->ProjectionMatrix;
;            GLfloat m0 = m[0],  m4 = m[4],  m8 = m[8],  m12 = m[12];
;            GLfloat m1 = m[1],  m5 = m[5],  m9 = m[9],  m13 = m[13];
;            GLfloat m2 = m[2],  m6 = m[6],  m10 = m[10],  m14 = m[14];
;            GLfloat m3 = m[3],  m7 = m[7],  m11 = m[11],  m15 = m[15];
;            GLuint i;
;            for (i=0;i<n;i++) {
;               GLfloat ex = vEye[i][0], ey = vEye[i][1];
;               GLfloat ez = vEye[i][2], ew = vEye[i][3];
;               GLfloat cx = m0 * ex + m4 * ey + m8  * ez + m12 * ew;
;               GLfloat cy = m1 * ex + m5 * ey + m9  * ez + m13 * ew;
;               GLfloat cz = m2 * ex + m6 * ey + m10 * ez + m14 * ew;
;               GLfloat cw = m3 * ex + m7 * ey + m11 * ez + m15 * ew;
;               GLubyte mask = 0;
;               vClip[i][0] = cx;
;               vClip[i][1] = cy;
;               vClip[i][2] = cz;
;               vClip[i][3] = cw;
;               if (cx >  cw)       mask |= CLIP_RIGHT_BIT;
;               else if (cx < -cw)  mask |= CLIP_LEFT_BIT;
;               if (cy >  cw)       mask |= CLIP_TOP_BIT;
;               else if (cy < -cw)  mask |= CLIP_BOTTOM_BIT;
;               if (cz >  cw)       mask |= CLIP_FAR_BIT;
;               else if (cz < -cw)  mask |= CLIP_NEAR_BIT;
;               if (mask) {
;                  clipMask[i] |= mask;
;                  tmpOrMask |= mask;
;               }
;               tmpAndMask &= mask;
;            }


                IFNE    1
                stfd    f14,-1*8(r1)
                stfd    f15,-2*8(r1)
                stfd    f16,-3*8(r1)
                stfd    f17,-4*8(r1)
                stfd    f18,-5*8(r1)
                stfd    f19,-6*8(r1)
                stfd    f20,-7*8(r1)
                stfd    f21,-8*8(r1)
                stfd    f22,-9*8(r1)
                stfd    f23,-10*8(r1)
                mr.     r3,r3
                beq     .end
                subi    r5,r5,4
                lfsu    f0,4(r5)
                lfsu    f1,4(r5)
                lfsu    f2,4(r5)
                lfsu    f3,4(r5)
                lfsu    f4,4(r5)
                lfsu    f5,4(r5)
                lfsu    f6,4(r5)
                lfsu    f7,4(r5)
                lfsu    f8,4(r5)
                lfsu    f9,4(r5)
                lfsu    f10,4(r5)
                lfsu    f11,4(r5)
                lfsu    f12,4(r5)
                lfsu    f13,4(r5)
                lfsu    f14,4(r5)
                lfsu    f15,4(r5)
                mtctr   r3
                subi    r6,r6,4
                subi    r4,r4,4
                lbz     r11,0(r8)
                lbz     r12,0(r9)
.loop
                li      r10,0
                lfsu    f16,4(r6)               ;f16 = ex
                fmuls   f19,f0,f16
                fmuls   f20,f1,f16
                lfsu    f17,4(r6)               ;f17 = ey
                fmuls   f21,f2,f16
                fmuls   f22,f3,f16
                fmadds  f19,f4,f17,f19
                lfsu    f18,4(r6)               ;f18 = ez
                fmadds  f20,f5,f17,f20
                fmadds  f21,f6,f17,f21
                fmadds  f22,f7,f17,f22
                lfsu    f23,4(r6)               ;f23 = ew
                fmadds  f19,f8,f18,f19
                fmadds  f20,f9,f18,f20
                fmadds  f21,f10,f18,f21
                fmadds  f22,f11,f18,f22
                fmadds  f19,f12,f23,f19         ;f19 = cx
                stfsu   f19,4(r4)
                fmadds  f20,f13,f23,f20         ;f20 = cy
                stfsu   f20,4(r4)
                fmadds  f21,f14,f23,f21         ;f21 = cz
                stfsu   f21,4(r4)
                fmadds  f22,f15,f23,f22         ;f22 = cw
                stfsu   f22,4(r4)

                fabs    f16,f19
                fabs    f17,f20
                fabs    f18,f21
                fsubs   f23,f16,f17
                fsel    f16,f23,f16,f17
                fsubs   f17,f16,f18
                fsel    f18,f17,f16,f18
                fcmpu   f18,f22
                bgt     .check
                li      r10,0
                b       .cont7
.check



                fneg    f16,f22                 ;f16 = -cw
                fcmpu   f19,f22
                ble     .cont1
                ori     r10,r10,CLIP_RIGHT_BIT
                b       .cont2
.cont1
                fcmpu   f19,f16
                bge     .cont2
                ori     r10,r10,CLIP_LEFT_BIT
.cont2
                fcmpu   f20,f22
                ble     .cont3
                ori     r10,r10,CLIP_TOP_BIT
                b       .cont4
.cont3
                fcmpu   f20,f16
                bge     .cont4
                ori     r10,r10,CLIP_BOTTOM_BIT
.cont4
                fcmpu   f21,f22
                ble     .cont5
                ori     r10,r10,CLIP_FAR_BIT
                b       .cont6
.cont5
                fcmpu   f21,f16
                bge     .cont6
                ori     r10,r10,CLIP_NEAR_BIT
.cont6
                mr.     r10,r10
                beq     .cont7
                lbz     r0,0(r7)
                or      r0,r0,r10
                stb     r0,0(r7)
                or      r11,r11,r10
.cont7
                and     r12,r12,r10
                addi    r7,r7,1
                bdnz    .loop
                stb     r11,0(r8)
                stb     r12,0(r9)
.end
                lfd     f23,-10*8(r1)
                lfd     f22,-9*8(r1)
                lfd     f21,-8*8(r1)
                lfd     f20,-7*8(r1)
                lfd     f19,-6*8(r1)
                lfd     f18,-5*8(r1)
                lfd     f17,-4*8(r1)
                lfd     f16,-3*8(r1)
                lfd     f15,-2*8(r1)
                lfd     f14,-1*8(r1)
                blr
                ELSEIF
                stfd    f14,-1*8(r1)
                stfd    f15,-2*8(r1)
                stfd    f16,-3*8(r1)
                stfd    f17,-4*8(r1)
                stfd    f18,-5*8(r1)
                stfd    f19,-6*8(r1)
                stfd    f20,-7*8(r1)
                stfd    f21,-8*8(r1)
                stfd    f22,-9*8(r1)
                stfd    f23,-10*8(r1)
                mr.     r3,r3
                beq     .end
                subi    r5,r5,4
                lfsu    f0,4(r5)
                lfsu    f1,4(r5)
                lfsu    f2,4(r5)
                lfsu    f3,4(r5)
                lfsu    f4,4(r5)
                lfsu    f5,4(r5)
                lfsu    f6,4(r5)
                lfsu    f7,4(r5)
                lfsu    f8,4(r5)
                lfsu    f9,4(r5)
                lfsu    f10,4(r5)
                lfsu    f11,4(r5)
                lfsu    f12,4(r5)
                lfsu    f13,4(r5)
                lfsu    f14,4(r5)
                lfsu    f15,4(r5)
                mtctr   r3
                subi    r6,r6,4
                subi    r4,r4,4
                lbz     r11,0(r8)
                lbz     r12,0(r9)
.loop
                li      r10,0
                lfsu    f16,4(r6)               ;f16 = ex
                fmuls   f19,f0,f16
                fmuls   f20,f1,f16
                lfsu    f17,4(r6)               ;f17 = ey
                fmuls   f21,f2,f16
                fmuls   f22,f3,f16
                fmadds  f19,f4,f17,f19
                lfsu    f18,4(r6)               ;f18 = ez
                fmadds  f20,f5,f17,f20
                fmadds  f21,f6,f17,f21
                fmadds  f22,f7,f17,f22
                lfsu    f23,4(r6)               ;f23 = ew
                fmadds  f19,f8,f18,f19
                fmadds  f20,f9,f18,f20
                fmadds  f21,f10,f18,f21
                fmadds  f22,f11,f18,f22
                fmadds  f19,f12,f23,f19         ;f19 = cx
                stfsu   f19,4(r4)
                fmadds  f20,f13,f23,f20         ;f20 = cy
                stfsu   f20,4(r4)
                fmadds  f21,f14,f23,f21         ;f21 = cz
                stfsu   f21,4(r4)
                fmadds  f22,f15,f23,f22         ;f22 = cw
                stfsu   f22,4(r4)
                fneg    f16,f22                 ;f16 = -cw

                fcmpu   f19,f22
                ble     .cont1
                ori     r10,r10,CLIP_RIGHT_BIT
                b       .cont2
.cont1
                fcmpu   f19,f16
                bge     .cont2
                ori     r10,r10,CLIP_LEFT_BIT
.cont2
                fcmpu   f20,f22
                ble     .cont3
                ori     r10,r10,CLIP_TOP_BIT
                b       .cont4
.cont3
                fcmpu   f20,f16
                bge     .cont4
                ori     r10,r10,CLIP_BOTTOM_BIT
.cont4
                fcmpu   f21,f22
                ble     .cont5
                ori     r10,r10,CLIP_FAR_BIT
                b       .cont6
.cont5
                fcmpu   f21,f16
                bge     .cont6
                ori     r10,r10,CLIP_NEAR_BIT
.cont6
                mr.     r10,r10
                beq     .cont7
                lbz     r0,0(r7)
                or      r0,r0,r10
                stb     r0,0(r7)
                or      r11,r11,r10
.cont7
                and     r12,r12,r10
                addi    r7,r7,1
                bdnz    .loop
                stb     r11,0(r8)
                stb     r12,0(r9)
.end
                lfd     f23,-10*8(r1)
                lfd     f22,-9*8(r1)
                lfd     f21,-8*8(r1)
                lfd     f20,-7*8(r1)
                lfd     f19,-6*8(r1)
                lfd     f18,-5*8(r1)
                lfd     f17,-4*8(r1)
                lfd     f16,-3*8(r1)
                lfd     f15,-2*8(r1)
                lfd     f14,-1*8(r1)
                blr
                ENDC





_asm_project_and_cliptest_identity

;            GLuint i;
;            for (i=0;i<n;i++) {
;               GLfloat cx = vClip[i][0] = vEye[i][0];
;               GLfloat cy = vClip[i][1] = vEye[i][1];
;               GLfloat cz = vClip[i][2] = vEye[i][2];
;               GLfloat cw = vClip[i][3] = vEye[i][3];
;               GLubyte mask = 0;
;               if (cx >  cw)       mask |= CLIP_RIGHT_BIT;
;               else if (cx < -cw)  mask |= CLIP_LEFT_BIT;
;               if (cy >  cw)       mask |= CLIP_TOP_BIT;
;               else if (cy < -cw)  mask |= CLIP_BOTTOM_BIT;
;               if (cz >  cw)       mask |= CLIP_FAR_BIT;
;               else if (cz < -cw)  mask |= CLIP_NEAR_BIT;
;               if (mask) {
;                  clipMask[i] |= mask;
;                  tmpOrMask |= mask;
;               }
;               tmpAndMask &= mask;
;            }

                IFNE    1
                mr.     r3,r3
                beq     .end
                mtctr   r3
                subi    r5,r5,4
                subi    r4,r4,4
                lbz     r11,0(r7)
                lbz     r12,0(r8)
.loop
                li      r10,0
                lfsu    f0,4(r5)
                stfsu   f0,4(r4)
                lfsu    f1,4(r5)
                stfsu   f1,4(r4)
                lfsu    f2,4(r5)
                stfsu   f2,4(r4)
                lfsu    f3,4(r5)
                stfsu   f3,4(r4)

                fabs    f6,f1
                fabs    f7,f2
                fabs    f8,f3
                fsubs   f11,f6,f7
                fsel    f6,f11,f6,f7
                fsubs   f7,f6,f8
                fsel    f8,f7,f6,f8
                fcmpu   f8,f3
                bgt     .check
                li      r10,0
                b       .cont7
.check

                fneg    f4,f3
                fcmpu   f0,f3
                ble     .cont1
                ori     r10,r10,CLIP_RIGHT_BIT
                b       .cont2
.cont1
                fcmpu   f0,f4
                bge     .cont2
                ori     r10,r10,CLIP_LEFT_BIT
.cont2
                fcmpu   f1,f3
                ble     .cont3
                ori     r10,r10,CLIP_TOP_BIT
                b       .cont4
.cont3
                fcmpu   f1,f4
                bge     .cont4
                ori     r10,r10,CLIP_BOTTOM_BIT
.cont4
                fcmpu   f2,f3
                ble     .cont5
                ori     r10,r10,CLIP_FAR_BIT
                b       .cont6
.cont5
                fcmpu   f2,f4
                bge     .cont6
                ori     r10,r10,CLIP_NEAR_BIT
.cont6
                mr.     r10,r10
                beq     .cont7
                lbz     r0,0(r6)
                or      r0,r0,r10
                stb     r0,0(r6)
                or      r11,r11,r10
.cont7
                and     r12,r12,r10
                addi    r6,r6,1
                bdnz    .loop
                stb     r11,0(r7)
                stb     r12,0(r8)
.end
                blr
                ELSEIF
                mr.     r3,r3
                beq     .end
                mtctr   r3
                subi    r5,r5,4
                subi    r4,r4,4
                lbz     r11,0(r7)
                lbz     r12,0(r8)
.loop
                li      r10,0
                lfsu    f0,4(r5)
                stfsu   f0,4(r4)
                lfsu    f1,4(r5)
                stfsu   f1,4(r4)
                lfsu    f2,4(r5)
                stfsu   f2,4(r4)
                lfsu    f3,4(r5)
                stfsu   f3,4(r4)
                fneg    f4,f3
                fcmpu   f0,f3
                ble     .cont1
                ori     r10,r10,CLIP_RIGHT_BIT
                b       .cont2
.cont1
                fcmpu   f0,f4
                bge     .cont2
                ori     r10,r10,CLIP_LEFT_BIT
.cont2
                fcmpu   f1,f3
                ble     .cont3
                ori     r10,r10,CLIP_TOP_BIT
                b       .cont4
.cont3
                fcmpu   f1,f4
                bge     .cont4
                ori     r10,r10,CLIP_BOTTOM_BIT
.cont4
                fcmpu   f2,f3
                ble     .cont5
                ori     r10,r10,CLIP_FAR_BIT
                b       .cont6
.cont5
                fcmpu   f2,f4
                bge     .cont6
                ori     r10,r10,CLIP_NEAR_BIT
.cont6
                mr.     r10,r10
                beq     .cont7
                lbz     r0,0(r6)
                or      r0,r0,r10
                stb     r0,0(r6)
                or      r11,r11,r10
.cont7
                and     r12,r12,r10
                addi    r6,r6,1
                bdnz    .loop
                stb     r11,0(r7)
                stb     r12,0(r8)
.end
                blr
                ENDC





_asm_project_and_cliptest_ortho

;            const GLfloat *m = ctx->ProjectionMatrix;
;            GLfloat m0 = m[0], m5 = m[5], m10 = m[10], m12 = m[12];
;            GLfloat m13 = m[13], m14 = m[14];
;            GLuint i;
;            for (i=0;i<n;i++) {
;               GLfloat ex = vEye[i][0], ey = vEye[i][1];
;               GLfloat ez = vEye[i][2], ew = vEye[i][3];
;               GLfloat cx = m0 * ex                      + m12 * ew;
;               GLfloat cy =           m5 * ey            + m13 * ew;
;               GLfloat cz =                     m10 * ez + m14 * ew;
;               GLfloat cw =                                      ew;
;               GLubyte mask = 0;
;               vClip[i][0] = cx;
;               vClip[i][1] = cy;
;               vClip[i][2] = cz;
;               vClip[i][3] = cw;
;               if (cx >  cw)       mask |= CLIP_RIGHT_BIT;
;               else if (cx < -cw)  mask |= CLIP_LEFT_BIT;
;               if (cy >  cw)       mask |= CLIP_TOP_BIT;
;               else if (cy < -cw)  mask |= CLIP_BOTTOM_BIT;
;               if (cz >  cw)       mask |= CLIP_FAR_BIT;
;               else if (cz < -cw)  mask |= CLIP_NEAR_BIT;
;               if (mask) {
;                  clipMask[i] |= mask;
;                  tmpOrMask |= mask;
;               }
;               tmpAndMask &= mask;
;            }

                IFNE    1
                mr.     r3,r3
                beq     .end
                lfs     f0,0*4(r5)
                lfs     f5,5*4(r5)
                lfs     f10,10*4(r5)
                lfs     f12,12*4(r5)
                lfs     f13,13*4(r5)
                lfs     f4,14*4(r5)
                mtctr   r3
                subi    r6,r6,4
                subi    r4,r4,4
                lbz     r11,0(r8)
                lbz     r12,0(r9)
.loop
                li      r10,0
                lfsu    f6,4(r6)                ;f6 = ex
                lfsu    f7,4(r6)                ;f7 = ey
                fmuls   f1,f0,f6
                lfsu    f8,4(r6)                ;f8 = ez
                fmuls   f2,f5,f7
                lfsu    f9,4(r6)                ;f9 = ew
                fmuls   f3,f10,f8
                fmadds  f1,f12,f9,f1            ;f1 = cx
                stfsu   f1,4(r4)
                fmadds  f2,f13,f9,f2            ;f2 = cy
                stfsu   f2,4(r4)
                fmadds  f3,f4,f9,f3             ;f3 = cz
                stfsu   f3,4(r4)
                stfsu   f9,4(r4)                ;f9 = cw

                fabs    f6,f1
                fabs    f7,f2
                fabs    f8,f3
                fsubs   f11,f6,f7
                fsel    f6,f11,f6,f7
                fsubs   f7,f6,f8
                fsel    f8,f7,f6,f8
                fcmpu   f8,f9
                bgt     .check
                li      r10,0
                b       .cont7
.check

                fneg    f11,f9                  ;f11 = -cw
                fcmpu   f1,f9
                ble     .cont1
                ori     r10,r10,CLIP_RIGHT_BIT
                b       .cont2
.cont1
                fcmpu   f1,f11
                bge     .cont2
                ori     r10,r10,CLIP_LEFT_BIT
.cont2
                fcmpu   f2,f9
                ble     .cont3
                ori     r10,r10,CLIP_TOP_BIT
                b       .cont4
.cont3
                fcmpu   f2,f11
                bge     .cont4
                ori     r10,r10,CLIP_BOTTOM_BIT
.cont4
                fcmpu   f3,f9
                ble     .cont5
                ori     r10,r10,CLIP_FAR_BIT
                b       .cont6
.cont5
                fcmpu   f3,f11
                bge     .cont6
                ori     r10,r10,CLIP_NEAR_BIT
.cont6
                mr.     r10,r10
                beq     .cont7
                lbz     r0,0(r7)
                or      r0,r0,r10
                stb     r0,0(r7)
                or      r11,r11,r10
.cont7
                and     r12,r12,r10
                addi    r7,r7,1
                bdnz    .loop
                stb     r11,0(r8)
                stb     r12,0(r9)
.end
                blr
                ELSEIF
                mr.     r3,r3
                beq     .end
                lfs     f0,0*4(r5)
                lfs     f5,5*4(r5)
                lfs     f10,10*4(r5)
                lfs     f12,12*4(r5)
                lfs     f13,13*4(r5)
                lfs     f4,14*4(r5)
                mtctr   r3
                subi    r6,r6,4
                subi    r4,r4,4
                lbz     r11,0(r8)
                lbz     r12,0(r9)
.loop
                li      r10,0
                lfsu    f6,4(r6)                ;f6 = ex
                lfsu    f7,4(r6)                ;f7 = ey
                fmuls   f1,f0,f6
                lfsu    f8,4(r6)                ;f8 = ez
                fmuls   f2,f5,f7
                lfsu    f9,4(r6)                ;f9 = ew
                fmuls   f3,f10,f8
                fmadds  f1,f12,f9,f1            ;f1 = cx
                stfsu   f1,4(r4)
                fmadds  f2,f13,f9,f2            ;f2 = cy
                stfsu   f2,4(r4)
                fmadds  f3,f4,f9,f3             ;f3 = cz
                stfsu   f3,4(r4)
                stfsu   f9,4(r4)                ;f9 = cw
                fneg    f11,f9                  ;f11 = -cw

                fcmpu   f1,f9
                ble     .cont1
                ori     r10,r10,CLIP_RIGHT_BIT
                b       .cont2
.cont1
                fcmpu   f1,f11
                bge     .cont2
                ori     r10,r10,CLIP_LEFT_BIT
.cont2
                fcmpu   f2,f9
                ble     .cont3
                ori     r10,r10,CLIP_TOP_BIT
                b       .cont4
.cont3
                fcmpu   f2,f11
                bge     .cont4
                ori     r10,r10,CLIP_BOTTOM_BIT
.cont4
                fcmpu   f3,f9
                ble     .cont5
                ori     r10,r10,CLIP_FAR_BIT
                b       .cont6
.cont5
                fcmpu   f3,f11
                bge     .cont6
                ori     r10,r10,CLIP_NEAR_BIT
.cont6
                mr.     r10,r10
                beq     .cont7
                lbz     r0,0(r7)
                or      r0,r0,r10
                stb     r0,0(r7)
                or      r11,r11,r10
.cont7
                and     r12,r12,r10
                addi    r7,r7,1
                bdnz    .loop
                stb     r11,0(r8)
                stb     r12,0(r9)
.end
                blr
                ENDC

_asm_project_and_cliptest_perspective

;            const GLfloat *m = ctx->ProjectionMatrix;
;            GLfloat m0 = m[0], m5 = m[5], m8 = m[8], m9 = m[9];
;            GLfloat m10 = m[10], m14 = m[14];
;            GLuint i;
;            for (i=0;i<n;i++) {
;               GLfloat ex = vEye[i][0], ey = vEye[i][1];
;               GLfloat ez = vEye[i][2], ew = vEye[i][3];
;               GLfloat cx = m0 * ex           + m8  * ez           ;
;               GLfloat cy =           m5 * ey + m9  * ez           ;
;               GLfloat cz =                     m10 * ez + m14 * ew;
;               GLfloat cw =                          -ez           ;
;               GLubyte mask = 0;
;               vClip[i][0] = cx;
;               vClip[i][1] = cy;
;               vClip[i][2] = cz;
;               vClip[i][3] = cw;
;               if (cx >  cw)       mask |= CLIP_RIGHT_BIT;
;               else if (cx < -cw)  mask |= CLIP_LEFT_BIT;
;               if (cy >  cw)       mask |= CLIP_TOP_BIT;
;               else if (cy < -cw)  mask |= CLIP_BOTTOM_BIT;
;               if (cz >  cw)       mask |= CLIP_FAR_BIT;
;               else if (cz < -cw)  mask |= CLIP_NEAR_BIT;
;               if (mask) {
;                  clipMask[i] |= mask;
;                  tmpOrMask |= mask;
;               }
;               tmpAndMask &= mask;
;            }
                IFNE    1
                mr.     r3,r3
                beq     .end
                lfs     f0,0*4(r5)
                lfs     f5,5*4(r5)
                lfs     f8,8*4(r5)
                lfs     f9,9*4(r5)
                lfs     f10,10*4(r5)
                lfs     f4,14*4(r5)
                mtctr   r3
                subi    r6,r6,4
                subi    r4,r4,4
                lbz     r11,0(r8)
                lbz     r12,0(r9)
.loop
                li      r10,0
                lfsu    f6,4(r6)                ;f6 = ex
                lfsu    f7,4(r6)                ;f7 = ey
                fmuls   f1,f0,f6
                lfsu    f12,4(r6)               ;f12 = ez
                fmuls   f2,f5,f7
                lfsu    f13,4(r6)               ;f13 = ew
                fmuls   f3,f10,f12
                fmadds  f1,f8,f12,f1            ;f1 = cx
                stfsu   f1,4(r4)
                fmadds  f2,f9,f12,f2            ;f2 = cy
                stfsu   f2,4(r4)
                fmadds  f3,f4,f13,f3            ;f3 = cz
                stfsu   f3,4(r4)
                fneg    f13,f12                 ;f13 = cw
                stfsu   f13,4(r4)

                fabs    f6,f1
                fabs    f7,f2
                fabs    f12,f3
                fsubs   f11,f6,f7
                fsel    f6,f11,f6,f7
                fsubs   f7,f6,f12
                fsel    f12,f7,f6,f12
                fcmpu   f12,f13
                bgt     .check
                li      r10,0
                b       .cont7
.check


                fneg    f11,f13                 ;f11 = -cw
                fcmpu   f1,f13
                ble     .cont1
                ori     r10,r10,CLIP_RIGHT_BIT
                b       .cont2
.cont1
                fcmpu   f1,f11
                bge     .cont2
                ori     r10,r10,CLIP_LEFT_BIT
.cont2
                fcmpu   f2,f13
                ble     .cont3
                ori     r10,r10,CLIP_TOP_BIT
                b       .cont4
.cont3
                fcmpu   f2,f11
                bge     .cont4
                ori     r10,r10,CLIP_BOTTOM_BIT
.cont4
                fcmpu   f3,f13
                ble     .cont5
                ori     r10,r10,CLIP_FAR_BIT
                b       .cont6
.cont5
                fcmpu   f3,f11
                bge     .cont6
                ori     r10,r10,CLIP_NEAR_BIT
.cont6
                mr.     r10,r10
                beq     .cont7
                lbz     r0,0(r7)
                or      r0,r0,r10
                stb     r0,0(r7)
                or      r11,r11,r10
.cont7
                and     r12,r12,r10
                addi    r7,r7,1
                bdnz    .loop
                stb     r11,0(r8)
                stb     r12,0(r9)
.end
                blr
                ELSEIF
                mr.     r3,r3
                beq     .end
                lfs     f0,0*4(r5)
                lfs     f5,5*4(r5)
                lfs     f8,8*4(r5)
                lfs     f9,9*4(r5)
                lfs     f10,10*4(r5)
                lfs     f4,14*4(r5)
                mtctr   r3
                subi    r6,r6,4
                subi    r4,r4,4
                lbz     r11,0(r8)
                lbz     r12,0(r9)
.loop
                li      r10,0
                lfsu    f6,4(r6)                ;f6 = ex
                lfsu    f7,4(r6)                ;f7 = ey
                fmuls   f1,f0,f6
                lfsu    f12,4(r6)               ;f12 = ez
                fmuls   f2,f5,f7
                lfsu    f13,4(r6)               ;f13 = ew
                fmuls   f3,f10,f12
                fmadds  f1,f8,f12,f1            ;f1 = cx
                stfsu   f1,4(r4)
                fmadds  f2,f9,f12,f2            ;f2 = cy
                stfsu   f2,4(r4)
                fmadds  f3,f4,f13,f3            ;f3 = cz
                stfsu   f3,4(r4)
                fneg    f13,f12                 ;f13 = cw
                stfsu   f13,4(r4)
                fneg    f11,f13                 ;f11 = -cw
                fcmpu   f1,f13
                ble     .cont1
                ori     r10,r10,CLIP_RIGHT_BIT
                b       .cont2
.cont1
                fcmpu   f1,f11
                bge     .cont2
                ori     r10,r10,CLIP_LEFT_BIT
.cont2
                fcmpu   f2,f13
                ble     .cont3
                ori     r10,r10,CLIP_TOP_BIT
                b       .cont4
.cont3
                fcmpu   f2,f11
                bge     .cont4
                ori     r10,r10,CLIP_BOTTOM_BIT
.cont4
                fcmpu   f3,f13
                ble     .cont5
                ori     r10,r10,CLIP_FAR_BIT
                b       .cont6
.cont5
                fcmpu   f3,f11
                bge     .cont6
                ori     r10,r10,CLIP_NEAR_BIT
.cont6
                mr.     r10,r10
                beq     .cont7
                lbz     r0,0(r7)
                or      r0,r0,r10
                stb     r0,0(r7)
                or      r11,r11,r10
.cont7
                and     r12,r12,r10
                addi    r7,r7,1
                bdnz    .loop
                stb     r11,0(r8)
                stb     r12,0(r9)
.end
                blr
                ENDC

_asm_vp_map_vertices_now

;      if (clipMask) {
;         /* one or more vertices are clipped */
;         GLuint i;
;         for (i=0;i<n;i++) {
;            if (clipMask[i]==0) {
;               vWin[i][0] = vClip[i][0] * sx + tx;
;               vWin[i][1] = vClip[i][1] * sy + ty;
;               vWin[i][2] = vClip[i][2] * sz + tz;
;            }
;         }
;      }
;      else {
;         /* no vertices are clipped */
;         GLuint i;
;         for (i=0;i<n;i++) {
;            vWin[i][0] = vClip[i][0] * sx + tx;
;            vWin[i][1] = vClip[i][1] * sy + ty;
;            vWin[i][2] = vClip[i][2] * sz + tz;
;         }
;      }

                mr.     r4,r4
                beq     .done
                subi    r5,r5,4
                mtctr   r4
                mr.     r3,r3
                beq     .loop2
                subi    r3,r3,1
.loop1
                lbzu    r0,1(r3)
                mr.     r0,r0
                bne     .inc
                lfs     f7,0(r6)
                fmadds  f10,f1,f7,f4
                lfs     f8,4(r6)
                fmadds  f11,f2,f8,f5
                lfs     f9,8(r6)
                fmadds  f12,f3,f9,f6
                stfsu   f10,4(r5)
                stfsu   f11,4(r5)
                stfsu   f12,4(r5)
                b       .next
.inc
                addi    r5,r5,12
.next
                addi    r6,r6,16
                bdnz    .loop1
                blr
.loop2
                lfs     f7,0(r6)
                fmadds  f10,f1,f7,f4
                lfs     f8,4(r6)
                fmadds  f11,f2,f8,f5
                lfs     f9,8(r6)
                fmadds  f12,f3,f9,f6
                stfsu   f10,4(r5)
                stfsu   f11,4(r5)
                addi    r6,r6,16
                stfsu   f12,4(r5)
                bdnz    .loop2
.done
                blr

_asm_vp_map_vertices

;      if (clipMask) {
;         /* one or more vertices are clipped */
;         GLuint i;
;         for (i=0;i<n;i++) {
;            if (clipMask[i] == 0) {
;               if (vClip[i][3] != 0.0F) {
;                  GLfloat wInv = 1.0F / vClip[i][3];
;                  vWin[i][0] = vClip[i][0] * wInv * sx + tx;
;                  vWin[i][1] = vClip[i][1] * wInv * sy + ty;
;                  vWin[i][2] = vClip[i][2] * wInv * sz + tz;
;               }
;               else {
;                  /* Div by zero!  Can't set window coords to infinity, so...*/
;                  vWin[i][0] = 0.0F;
;                  vWin[i][1] = 0.0F;
;                  vWin[i][2] = 0.0F;
;               }
;            }
;         }
;      }
;      else {
;         /* no vertices are clipped */
;         GLuint i;
;         for (i=0;i<n;i++) {
;            if (vClip[i][3] != 0.0F) {
;               GLfloat wInv = 1.0F / vClip[i][3];
;               vWin[i][0] = vClip[i][0] * wInv * sx + tx;
;               vWin[i][1] = vClip[i][1] * wInv * sy + ty;
;               vWin[i][2] = vClip[i][2] * wInv * sz + tz;
;            }
;            else {
;               /* Divide by zero!  Can't set window coords to infinity, so...*/
;               vWin[i][0] = 0.0F;
;               vWin[i][1] = 0.0F;
;               vWin[i][2] = 0.0F;
;            }
;         }
;      }

                stfd    f31,-8(r1)
                stfd    f30,-16(r1)
                stfd    f29,-24(r1)
                stfd    f28,-32(r1)
                stfd    f27,-40(r1)
                stfd    f26,-48(r1)
                mr.     r4,r4
                beq     .done
                subi    r5,r5,4
                mr.     r3,r3
                lfs     f12,fp_0(r2)
                lfs     f11,fp_2(r2)
                beq     .loop2
.loop1
                lbz     r0,0(r3)
                mr.     r0,r0
                bne     .inc
                lfs     f13,12(r6)
                fcmpu   f13,f12
                beq     .zero
                fres    f13,f13
                lfs     f7,0(r6)
                fmuls   f7,f7,f13
                lfs     f8,4(r6)
                fmadds  f7,f1,f7,f4
                fmuls   f8,f8,f13
                lfs     f9,8(r6)
                fmadds  f8,f2,f8,f5
                stfsu   f7,4(r5)
                fmuls   f9,f9,f13
                stfsu   f8,4(r5)
                fmadds  f9,f3,f9,f6
                stfsu   f9,4(r5)
                b       .next
.zero
                li      r0,0
                stwu    r0,4(r5)
                stwu    r0,4(r5)
                stwu    r0,4(r5)
                b       .next
.inc
                addi    r5,r5,12
.next
                addi    r6,r6,16
                addi    r3,r3,1
                subic.  r4,r4,1
                bne     .loop1
                b       .done
.loop2

                cmplwi  r4,4
                blt     .normal

/* quad parallel turbo division */
                lfs     f8,12(r6)
                lfs     f9,12+16(r6)
                lfs     f10,12+2*16(r6)
                lfs     f13,12+3*16(r6)

                fmuls   f0,f8,f8
                fmuls   f31,f9,f9
                fmuls   f29,f10,f10
                fmuls   f27,f13,f13
                fcmpu   f0,f12
                beq     .skip1
                frsqrte f0,f0
.skip1
                fcmpu   f31,f12
                beq     .skip2
                frsqrte f31,f31
.skip2
                fcmpu   f29,f12
                beq     .skip3
                frsqrte f29,f29
.skip3
                fcmpu   f27,f12
                beq     .skip4
                frsqrte f27,f27
.skip4
                fnmsubs f7,f0,f8,f11
                fnmsubs f30,f31,f9,f11
                fnmsubs f28,f29,f10,f11
                fnmsubs f26,f27,f13,f11
                fmuls   f0,f0,f7
                fmuls   f31,f31,f30
                fmuls   f29,f29,f28
                fmuls   f27,f27,f26
                fnmsubs f7,f0,f8,f11
                fnmsubs f30,f31,f9,f11
                fnmsubs f28,f29,f10,f11
                fnmsubs f26,f27,f13,f11
                fmuls   f0,f0,f7
                fmuls   f31,f31,f30
                fmuls   f29,f29,f28
                fmuls   f27,f27,f26
                fnmsubs f7,f0,f8,f11
                fnmsubs f30,f31,f9,f11
                fnmsubs f28,f29,f10,f11
                fnmsubs f26,f27,f13,f11
                fmuls   f0,f0,f7
                fmuls   f31,f31,f30
                fmuls   f29,f29,f28
                fmuls   f27,f27,f26

                lfs     f7,0(r6)
                fmuls   f7,f7,f0
                lfs     f8,4(r6)
                fmadds  f7,f1,f7,f4
                fmuls   f8,f8,f0
                lfs     f9,8(r6)
                fmadds  f8,f2,f8,f5
                stfsu   f7,4(r5)
                fmuls   f9,f9,f0
                stfsu   f8,4(r5)
                fmadds  f9,f3,f9,f6
                addi    r6,r6,16
                stfsu   f9,4(r5)

                lfs     f7,0(r6)
                fmuls   f7,f7,f31
                lfs     f8,4(r6)
                fmadds  f7,f1,f7,f4
                fmuls   f8,f8,f31
                lfs     f9,8(r6)
                fmadds  f8,f2,f8,f5
                stfsu   f7,4(r5)
                fmuls   f9,f9,f31
                stfsu   f8,4(r5)
                fmadds  f9,f3,f9,f6
                addi    r6,r6,16
                stfsu   f9,4(r5)

                lfs     f7,0(r6)
                fmuls   f7,f7,f29
                lfs     f8,4(r6)
                fmadds  f7,f1,f7,f4
                fmuls   f8,f8,f29
                lfs     f9,8(r6)
                fmadds  f8,f2,f8,f5
                stfsu   f7,4(r5)
                fmuls   f9,f9,f29
                stfsu   f8,4(r5)
                fmadds  f9,f3,f9,f6
                addi    r6,r6,16
                stfsu   f9,4(r5)

                lfs     f7,0(r6)
                fmuls   f7,f7,f27
                lfs     f8,4(r6)
                fmadds  f7,f1,f7,f4
                fmuls   f8,f8,f27
                lfs     f9,8(r6)
                fmadds  f8,f2,f8,f5
                stfsu   f7,4(r5)
                fmuls   f9,f9,f27
                stfsu   f8,4(r5)
                fmadds  f9,f3,f9,f6
                addi    r6,r6,16
                stfsu   f9,4(r5)

                subic.  r4,r4,4
                bne     .loop2
                b       .done

.normal


                lfs     f13,12(r6)
                fcmpu   f13,f12
                beq     .zero2
                fres    f13,f13
                lfs     f7,0(r6)
                fmuls   f7,f7,f13
                lfs     f8,4(r6)
                fmadds  f7,f1,f7,f4
                fmuls   f8,f8,f13
                lfs     f9,8(r6)
                fmadds  f8,f2,f8,f5
                stfsu   f7,4(r5)
                fmuls   f9,f9,f13
                stfsu   f8,4(r5)
                fmadds  f9,f3,f9,f6
                addi    r6,r6,16
                stfsu   f9,4(r5)
                b       .next2
.zero2
                li      r0,0
                stwu    r0,4(r5)
                stwu    r0,4(r5)
                stwu    r0,4(r5)
                addi    r6,r6,16
.next2
                subic.  r4,r4,1
                bne     .loop2
.done
                lfd     f26,-48(r1)
                lfd     f27,-40(r1)
                lfd     f28,-32(r1)
                lfd     f29,-24(r1)
                lfd     f30,-16(r1)
                lfd     f31,-8(r1)
                blr

                section data
fp_0            dc.s    0
fp_2            dc.s    2