[PATCH] Consistent whitespace in asm
Whitespace changes only.
diff -urN a/common/d_parta.S head/common/d_parta.S
--- a/common/d_parta.S 2004-07-25 15:59:59.000000000 +0930
+++ head/common/d_parta.S 2006-02-19 07:29:50.000000000 +1030
@@ -46,8 +46,8 @@
.align 4
.globl C(D_DrawParticle)
C(D_DrawParticle):
- pushl %ebp // preserve caller's stack frame
- pushl %edi // preserve register variables
+ pushl %ebp // preserve caller's stack frame
+ pushl %edi // preserve register variables
pushl %ebx
movl P(%esp),%edi
@@ -62,10 +62,10 @@
fsubs C(r_origin)+4
flds pt_org+8(%edi)
fsubs C(r_origin)+8
- fxch %st(2) // local[0] | local[1] | local[2]
+ fxch %st(2) // local[0] | local[1] | local[2]
// transformed[2] = DotProduct(local, r_ppn);
- flds C(r_ppn) // r_ppn[0] | local[0] | local[1] | local[2]
+ flds C(r_ppn) // r_ppn[0] | local[0] | local[1] | local[2]
fmul %st(1),%st(0) // dot0 | local[0] | local[1] | local[2]
flds C(r_ppn)+4 // r_ppn[1] | dot0 | local[0] | local[1] | local[2]
fmul %st(3),%st(0) // dot1 | dot0 | local[0] | local[1] | local[2]
@@ -73,39 +73,39 @@
// local[1] | local[2]
fmul %st(5),%st(0) // dot2 | dot1 | dot0 | local[0] | local[1] | local[2]
fxch %st(2) // dot0 | dot1 | dot2 | local[0] | local[1] | local[2]
- faddp %st(0),%st(1) // dot0 + dot1 | dot2 | local[0] | local[1] |
- // local[2]
- faddp %st(0),%st(1) // z | local[0] | local[1] | local[2]
- fld %st(0) // z | z | local[0] | local[1] |
- // local[2]
+ faddp %st(0),%st(1) // dot0 + dot1 | dot2 | local[0] | local[1] |
+ // local[2]
+ faddp %st(0),%st(1) // z | local[0] | local[1] | local[2]
+ fld %st(0) // z | z | local[0] | local[1] |
+ // local[2]
fdivrs float_1 // 1/z | z | local[0] | local[1] | local[2]
fxch %st(1) // z | 1/z | local[0] | local[1] | local[2]
// if (transformed[2] < PARTICLE_Z_CLIP)
// return;
fcomps float_particle_z_clip // 1/z | local[0] | local[1] | local[2]
- fxch %st(3) // local[2] | local[0] | local[1] | 1/z
+ fxch %st(3) // local[2] | local[0] | local[1] | 1/z
flds C(r_pup) // r_pup[0] | local[2] | local[0] | local[1] | 1/z
fmul %st(2),%st(0) // dot0 | local[2] | local[0] | local[1] | 1/z
flds C(r_pup)+4 // r_pup[1] | dot0 | local[2] | local[0] |
- // local[1] | 1/z
+ // local[1] | 1/z
fnstsw %ax
testb $1,%ah
- jnz LPop6AndDone
+ jnz LPop6AndDone
// transformed[1] = DotProduct(local, r_pup);
fmul %st(4),%st(0) // dot1 | dot0 | local[2] | local[0] | local[1] | 1/z
flds C(r_pup)+8 // r_pup[2] | dot1 | dot0 | local[2] |
- // local[0] | local[1] | 1/z
+ // local[0] | local[1] | 1/z
fmul %st(3),%st(0) // dot2 | dot1 | dot0 | local[2] | local[0] |
- // local[1] | 1/z
+ // local[1] | 1/z
fxch %st(2) // dot0 | dot1 | dot2 | local[2] | local[0] |
- // local[1] | 1/z
- faddp %st(0),%st(1) // dot0 + dot1 | dot2 | local[2] | local[0] |
- // local[1] | 1/z
- faddp %st(0),%st(1) // y | local[2] | local[0] | local[1] | 1/z
+ // local[1] | 1/z
+ faddp %st(0),%st(1) // dot0 + dot1 | dot2 | local[2] | local[0] |
+ // local[1] | 1/z
+ faddp %st(0),%st(1) // y | local[2] | local[0] | local[1] | 1/z
fxch %st(3) // local[1] | local[2] | local[0] | y | 1/z
// transformed[0] = DotProduct(local, r_pright);
@@ -115,32 +115,32 @@
fxch %st(1) // local[2] | dot0 | dot1 | y | 1/z
fmuls C(r_pright)+8 // dot2 | dot0 | dot1 | y | 1/z
fxch %st(2) // dot1 | dot0 | dot2 | y | 1/z
- faddp %st(0),%st(1) // dot1 + dot0 | dot2 | y | 1/z
+ faddp %st(0),%st(1) // dot1 + dot0 | dot2 | y | 1/z
faddp %st(0),%st(1) // x | y | 1/z
- fxch %st(1) // y | x | 1/z
+ fxch %st(1) // y | x | 1/z
// project the point
fmul %st(2),%st(0) // y/z | x | 1/z
- fxch %st(1) // x | y/z | 1/z
+ fxch %st(1) // x | y/z | 1/z
fmul %st(2),%st(0) // x/z | y/z | 1/z
- fxch %st(1) // y/z | x/z | 1/z
- fsubrs C(ycenter) // v | x/z | 1/z
- fxch %st(1) // x/z | v | 1/z
- fadds C(xcenter) // u | v | 1/z
+ fxch %st(1) // y/z | x/z | 1/z
+ fsubrs C(ycenter) // v | x/z | 1/z
+ fxch %st(1) // x/z | v | 1/z
+ fadds C(xcenter) // u | v | 1/z
// FIXME: preadjust xcenter and ycenter
- fxch %st(1) // v | u | 1/z
+ fxch %st(1) // v | u | 1/z
fadds float_point5 // v | u | 1/z
- fxch %st(1) // u | v | 1/z
+ fxch %st(1) // u | v | 1/z
fadds float_point5 // u | v | 1/z
- fxch %st(2) // 1/z | v | u
- fmuls DP_32768 // 1/z * 0x8000 | v | u
- fxch %st(2) // u | v | 1/z * 0x8000
+ fxch %st(2) // 1/z | v | u
+ fmuls DP_32768 // 1/z * 0x8000 | v | u
+ fxch %st(2) // u | v | 1/z * 0x8000
// FIXME: use Terje's fp->int trick here?
// FIXME: check we're getting proper rounding here
- fistpl DP_u // v | 1/z * 0x8000
- fistpl DP_v // 1/z * 0x8000
+ fistpl DP_u // v | 1/z * 0x8000
+ fistpl DP_v // 1/z * 0x8000
movl DP_u,%eax
movl DP_v,%edx
@@ -156,25 +156,25 @@
movl C(d_vrectbottom_particle),%ebx
movl C(d_vrectright_particle),%ecx
cmpl %ebx,%edx
- jg LPop1AndDone
+ jg LPop1AndDone
cmpl %ecx,%eax
- jg LPop1AndDone
+ jg LPop1AndDone
movl C(d_vrecty),%ebx
movl C(d_vrectx),%ecx
cmpl %ebx,%edx
- jl LPop1AndDone
+ jl LPop1AndDone
cmpl %ecx,%eax
- jl LPop1AndDone
+ jl LPop1AndDone
flds pt_color(%edi) // color | 1/z * 0x8000
// FIXME: use Terje's fast fp->int trick?
- fistpl DP_Color // 1/z * 0x8000
+ fistpl DP_Color // 1/z * 0x8000
movl C(d_viewbuffer),%ebx
addl %eax,%ebx
- movl C(d_scantable)(,%edx,4),%edi // point to the pixel
+ movl C(d_scantable)(,%edx,4),%edi // point to the pixel
imull C(d_zrowbytes),%edx // point to the z pixel
@@ -194,42 +194,42 @@
movl izi,%ebp
// if (pix < d_pix_min)
-// pix = d_pix_min;
+// pix = d_pix_min;
// else if (pix > d_pix_max)
// pix = d_pix_max;
movl C(d_pix_min),%ebx
movl C(d_pix_max),%ecx
cmpl %ebx,%eax
- jnl LTestPixMax
+ jnl LTestPixMax
movl %ebx,%eax
- jmp LTestDone
+ jmp LTestDone
LTestPixMax:
cmpl %ecx,%eax
- jng LTestDone
+ jng LTestDone
movl %ecx,%eax
-LTestDone:
+LTestDone:
movb DP_Color,%ch
movl C(d_y_aspect_shift),%ebx
testl %ebx,%ebx
- jnz LDefault
+ jnz LDefault
cmpl $4,%eax
- ja LDefault
+ ja LDefault
- jmp DP_EntryTable-4(,%eax,4)
+ jmp DP_EntryTable-4(,%eax,4)
// 1x1
.globl DP_1x1
DP_1x1:
- cmpw %bp,(%edx) // just one pixel to do
- jg LDone
+ cmpw %bp,(%edx) // just one pixel to do
+ jg LDone
movw %bp,(%edx)
movb %ch,(%edi)
- jmp LDone
+ jmp LDone
// 2x2
.globl DP_2x2
@@ -239,28 +239,28 @@
movl C(d_zrowbytes),%esi
cmpw %bp,(%edx)
- jg L2x2_1
+ jg L2x2_1
movw %bp,(%edx)
movb %ch,(%edi)
L2x2_1:
cmpw %bp,2(%edx)
- jg L2x2_2
+ jg L2x2_2
movw %bp,2(%edx)
movb %ch,1(%edi)
L2x2_2:
cmpw %bp,(%edx,%esi,1)
- jg L2x2_3
+ jg L2x2_3
movw %bp,(%edx,%esi,1)
movb %ch,(%edi,%ebx,1)
L2x2_3:
cmpw %bp,2(%edx,%esi,1)
- jg L2x2_4
+ jg L2x2_4
movw %bp,2(%edx,%esi,1)
movb %ch,1(%edi,%ebx,1)
L2x2_4:
popl %esi
- jmp LDone
+ jmp LDone
// 3x3
.globl DP_3x3
@@ -270,55 +270,55 @@
movl C(d_zrowbytes),%esi
cmpw %bp,(%edx)
- jg L3x3_1
+ jg L3x3_1
movw %bp,(%edx)
movb %ch,(%edi)
L3x3_1:
cmpw %bp,2(%edx)
- jg L3x3_2
+ jg L3x3_2
movw %bp,2(%edx)
movb %ch,1(%edi)
L3x3_2:
cmpw %bp,4(%edx)
- jg L3x3_3
+ jg L3x3_3
movw %bp,4(%edx)
movb %ch,2(%edi)
L3x3_3:
cmpw %bp,(%edx,%esi,1)
- jg L3x3_4
+ jg L3x3_4
movw %bp,(%edx,%esi,1)
movb %ch,(%edi,%ebx,1)
L3x3_4:
cmpw %bp,2(%edx,%esi,1)
- jg L3x3_5
+ jg L3x3_5
movw %bp,2(%edx,%esi,1)
movb %ch,1(%edi,%ebx,1)
L3x3_5:
cmpw %bp,4(%edx,%esi,1)
- jg L3x3_6
+ jg L3x3_6
movw %bp,4(%edx,%esi,1)
movb %ch,2(%edi,%ebx,1)
L3x3_6:
cmpw %bp,(%edx,%esi,2)
- jg L3x3_7
+ jg L3x3_7
movw %bp,(%edx,%esi,2)
movb %ch,(%edi,%ebx,2)
L3x3_7:
cmpw %bp,2(%edx,%esi,2)
- jg L3x3_8
+ jg L3x3_8
movw %bp,2(%edx,%esi,2)
movb %ch,1(%edi,%ebx,2)
L3x3_8:
cmpw %bp,4(%edx,%esi,2)
- jg L3x3_9
+ jg L3x3_9
movw %bp,4(%edx,%esi,2)
movb %ch,2(%edi,%ebx,2)
L3x3_9:
popl %esi
- jmp LDone
+ jmp LDone
// 4x4
@@ -329,43 +329,43 @@
movl C(d_zrowbytes),%esi
cmpw %bp,(%edx)
- jg L4x4_1
+ jg L4x4_1
movw %bp,(%edx)
movb %ch,(%edi)
L4x4_1:
cmpw %bp,2(%edx)
- jg L4x4_2
+ jg L4x4_2
movw %bp,2(%edx)
movb %ch,1(%edi)
L4x4_2:
cmpw %bp,4(%edx)
- jg L4x4_3
+ jg L4x4_3
movw %bp,4(%edx)
movb %ch,2(%edi)
L4x4_3:
cmpw %bp,6(%edx)
- jg L4x4_4
+ jg L4x4_4
movw %bp,6(%edx)
movb %ch,3(%edi)
L4x4_4:
cmpw %bp,(%edx,%esi,1)
- jg L4x4_5
+ jg L4x4_5
movw %bp,(%edx,%esi,1)
movb %ch,(%edi,%ebx,1)
L4x4_5:
cmpw %bp,2(%edx,%esi,1)
- jg L4x4_6
+ jg L4x4_6
movw %bp,2(%edx,%esi,1)
movb %ch,1(%edi,%ebx,1)
L4x4_6:
cmpw %bp,4(%edx,%esi,1)
- jg L4x4_7
+ jg L4x4_7
movw %bp,4(%edx,%esi,1)
movb %ch,2(%edi,%ebx,1)
L4x4_7:
cmpw %bp,6(%edx,%esi,1)
- jg L4x4_8
+ jg L4x4_8
movw %bp,6(%edx,%esi,1)
movb %ch,3(%edi,%ebx,1)
L4x4_8:
@@ -374,49 +374,49 @@
leal (%edi,%ebx,2),%edi
cmpw %bp,(%edx)
- jg L4x4_9
+ jg L4x4_9
movw %bp,(%edx)
movb %ch,(%edi)
L4x4_9:
cmpw %bp,2(%edx)
- jg L4x4_10
+ jg L4x4_10
movw %bp,2(%edx)
movb %ch,1(%edi)
L4x4_10:
cmpw %bp,4(%edx)
- jg L4x4_11
+ jg L4x4_11
movw %bp,4(%edx)
movb %ch,2(%edi)
L4x4_11:
cmpw %bp,6(%edx)
- jg L4x4_12
+ jg L4x4_12
movw %bp,6(%edx)
movb %ch,3(%edi)
L4x4_12:
cmpw %bp,(%edx,%esi,1)
- jg L4x4_13
+ jg L4x4_13
movw %bp,(%edx,%esi,1)
movb %ch,(%edi,%ebx,1)
L4x4_13:
cmpw %bp,2(%edx,%esi,1)
- jg L4x4_14
+ jg L4x4_14
movw %bp,2(%edx,%esi,1)
movb %ch,1(%edi,%ebx,1)
L4x4_14:
cmpw %bp,4(%edx,%esi,1)
- jg L4x4_15
+ jg L4x4_15
movw %bp,4(%edx,%esi,1)
movb %ch,2(%edi,%ebx,1)
L4x4_15:
cmpw %bp,6(%edx,%esi,1)
- jg L4x4_16
+ jg L4x4_16
movw %bp,6(%edx,%esi,1)
movb %ch,3(%edi,%ebx,1)
L4x4_16:
popl %esi
- jmp LDone
+ jmp LDone
// default case, handling any size particle
LDefault:
@@ -445,23 +445,23 @@
LGenColLoop:
cmpw %bp,-2(%edx,%eax,2)
- jg LGSkip
+ jg LGSkip
movw %bp,-2(%edx,%eax,2)
movb %ch,-1(%edi,%eax,1)
LGSkip:
- decl %eax // --pix
- jnz LGenColLoop
+ decl %eax // --pix
+ jnz LGenColLoop
addl C(d_zrowbytes),%edx
addl C(screenwidth),%edi
- decl %ebx // --count
- jnz LGenRowLoop
+ decl %ebx // --count
+ jnz LGenRowLoop
LDone:
- popl %ebx // restore register variables
+ popl %ebx // restore register variables
popl %edi
- popl %ebp // restore the caller's stack frame
+ popl %ebp // restore the caller's stack frame
ret
LPop6AndDone:
@@ -472,6 +472,6 @@
fstp %st(0)
LPop1AndDone:
fstp %st(0)
- jmp LDone
+ jmp LDone
#endif // id386
diff -urN a/common/d_polysa.S head/common/d_polysa.S
--- a/common/d_polysa.S 2004-08-08 16:37:41.000000000 +0930
+++ head/common/d_polysa.S 2006-02-19 07:37:21.000000000 +1030
@@ -36,22 +36,20 @@
//#define SPAN_SIZE (((DPS_MAXSPANS + 1 + ((CACHE_SIZE - 1) / spanpackage_t_size)) + 1) * spanpackage_t_size)
#define SPAN_SIZE ((1024+1+1+1)*32)
-
.data
-
.align 4
-p10_minus_p20: .single 0
-p01_minus_p21: .single 0
+
+p10_minus_p20: .single 0
+p01_minus_p21: .single 0
temp0: .single 0
temp1: .single 0
Ltemp: .single 0
aff8entryvec_table: .long LDraw8, LDraw7, LDraw6, LDraw5
- .long LDraw4, LDraw3, LDraw2, LDraw1
+ .long LDraw4, LDraw3, LDraw2, LDraw1
lzistepx: .long 0
-
.text
.extern C(D_PolysetSetEdgeTable)
@@ -76,39 +74,39 @@
//
// ystepdenominv = -xstepdenominv;
- fildl C(r_p0)+0 // r_p0[0]
- fildl C(r_p2)+0 // r_p2[0] | r_p0[0]
- fildl C(r_p0)+4 // r_p0[1] | r_p2[0] | r_p0[0]
- fildl C(r_p2)+4 // r_p2[1] | r_p0[1] | r_p2[0] | r_p0[0]
- fildl C(r_p1)+0 // r_p1[0] | r_p2[1] | r_p0[1] | r_p2[0] | r_p0[0]
- fildl C(r_p1)+4 // r_p1[1] | r_p1[0] | r_p2[1] | r_p0[1] |
- // r_p2[0] | r_p0[0]
- fxch %st(3) // r_p0[1] | r_p1[0] | r_p2[1] | r_p1[1] |
- // r_p2[0] | r_p0[0]
+ fildl C(r_p0)+0 // r_p0[0]
+ fildl C(r_p2)+0 // r_p2[0] | r_p0[0]
+ fildl C(r_p0)+4 // r_p0[1] | r_p2[0] | r_p0[0]
+ fildl C(r_p2)+4 // r_p2[1] | r_p0[1] | r_p2[0] | r_p0[0]
+ fildl C(r_p1)+0 // r_p1[0] | r_p2[1] | r_p0[1] | r_p2[0] | r_p0[0]
+ fildl C(r_p1)+4 // r_p1[1] | r_p1[0] | r_p2[1] | r_p0[1] |
+ // r_p2[0] | r_p0[0]
+ fxch %st(3) // r_p0[1] | r_p1[0] | r_p2[1] | r_p1[1] |
+ // r_p2[0] | r_p0[0]
fsub %st(2),%st(0) // p01_minus_p21 | r_p1[0] | r_p2[1] | r_p1[1] |
- // r_p2[0] | r_p0[0]
- fxch %st(1) // r_p1[0] | p01_minus_p21 | r_p2[1] | r_p1[1] |
- // r_p2[0] | r_p0[0]
+ // r_p2[0] | r_p0[0]
+ fxch %st(1) // r_p1[0] | p01_minus_p21 | r_p2[1] | r_p1[1] |
+ // r_p2[0] | r_p0[0]
fsub %st(4),%st(0) // p10_minus_p20 | p01_minus_p21 | r_p2[1] |
- // r_p1[1] | r_p2[0] | r_p0[0]
- fxch %st(5) // r_p0[0] | p01_minus_p21 | r_p2[1] |
- // r_p1[1] | r_p2[0] | p10_minus_p20
+ // r_p1[1] | r_p2[0] | r_p0[0]
+ fxch %st(5) // r_p0[0] | p01_minus_p21 | r_p2[1] |
+ // r_p1[1] | r_p2[0] | p10_minus_p20
fsubp %st(0),%st(4) // p01_minus_p21 | r_p2[1] | r_p1[1] |
- // p00_minus_p20 | p10_minus_p20
- fxch %st(2) // r_p1[1] | r_p2[1] | p01_minus_p21 |
- // p00_minus_p20 | p10_minus_p20
+ // p00_minus_p20 | p10_minus_p20
+ fxch %st(2) // r_p1[1] | r_p2[1] | p01_minus_p21 |
+ // p00_minus_p20 | p10_minus_p20
fsubp %st(0),%st(1) // p11_minus_p21 | p01_minus_p21 |
- // p00_minus_p20 | p10_minus_p20
- fxch %st(1) // p01_minus_p21 | p11_minus_p21 |
- // p00_minus_p20 | p10_minus_p20
- flds C(d_xdenom) // d_xdenom | p01_minus_p21 | p11_minus_p21 |
- // p00_minus_p20 | p10_minus_p20
- fxch %st(4) // p10_minus_p20 | p01_minus_p21 | p11_minus_p21 |
- // p00_minus_p20 | d_xdenom
+ // p00_minus_p20 | p10_minus_p20
+ fxch %st(1) // p01_minus_p21 | p11_minus_p21 |
+ // p00_minus_p20 | p10_minus_p20
+ flds C(d_xdenom) // d_xdenom | p01_minus_p21 | p11_minus_p21 |
+ // p00_minus_p20 | p10_minus_p20
+ fxch %st(4) // p10_minus_p20 | p01_minus_p21 | p11_minus_p21 |
+ // p00_minus_p20 | d_xdenom
fstps p10_minus_p20 // p01_minus_p21 | p11_minus_p21 |
- // p00_minus_p20 | d_xdenom
+ // p00_minus_p20 | d_xdenom
fstps p01_minus_p21 // p11_minus_p21 | p00_minus_p20 | xstepdenominv
- fxch %st(2) // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fxch %st(2) // xstepdenominv | p00_minus_p20 | p11_minus_p21
//// ceil () for light so positive steps are exaggerated, negative steps
//// diminished, pushing us away from underflow toward overflow. Underflow is
@@ -116,331 +114,331 @@
// t0 = r_p0[4] - r_p2[4];
// t1 = r_p1[4] - r_p2[4];
- fildl C(r_p2)+16 // r_p2[4] | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
- fildl C(r_p0)+16 // r_p0[4] | r_p2[4] | xstepdenominv |
- // p00_minus_p20 | p11_minus_p21
- fildl C(r_p1)+16 // r_p1[4] | r_p0[4] | r_p2[4] | xstepdenominv |
- // p00_minus_p20 | p11_minus_p21
- fxch %st(2) // r_p2[4] | r_p0[4] | r_p1[4] | xstepdenominv |
- // p00_minus_p20 | p11_minus_p21
- fld %st(0) // r_p2[4] | r_p2[4] | r_p0[4] | r_p1[4] |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fildl C(r_p2)+16 // r_p2[4] | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
+ fildl C(r_p0)+16 // r_p0[4] | r_p2[4] | xstepdenominv |
+ // p00_minus_p20 | p11_minus_p21
+ fildl C(r_p1)+16 // r_p1[4] | r_p0[4] | r_p2[4] | xstepdenominv |
+ // p00_minus_p20 | p11_minus_p21
+ fxch %st(2) // r_p2[4] | r_p0[4] | r_p1[4] | xstepdenominv |
+ // p00_minus_p20 | p11_minus_p21
+ fld %st(0) // r_p2[4] | r_p2[4] | r_p0[4] | r_p1[4] |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
fsubrp %st(0),%st(2) // r_p2[4] | t0 | r_p1[4] | xstepdenominv |
- // p00_minus_p20 | p11_minus_p21
+ // p00_minus_p20 | p11_minus_p21
fsubrp %st(0),%st(2) // t0 | t1 | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
+ // p11_minus_p21
// r_lstepx = (int)
// ceil((t1 * p01_minus_p21 - t0 * p11_minus_p21) * xstepdenominv);
// r_lstepy = (int)
// ceil((t1 * p00_minus_p20 - t0 * p10_minus_p20) * ystepdenominv);
- fld %st(0) // t0 | t0 | t1 | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
+ fld %st(0) // t0 | t0 | t1 | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
fmul %st(5),%st(0) // t0*p11_minus_p21 | t0 | t1 | xstepdenominv |
- // p00_minus_p20 | p11_minus_p21
- fxch %st(2) // t1 | t0 | t0*p11_minus_p21 | xstepdenominv |
- // p00_minus_p20 | p11_minus_p21
- fld %st(0) // t1 | t1 | t0 | t0*p11_minus_p21 |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ // p00_minus_p20 | p11_minus_p21
+ fxch %st(2) // t1 | t0 | t0*p11_minus_p21 | xstepdenominv |
+ // p00_minus_p20 | p11_minus_p21
+ fld %st(0) // t1 | t1 | t0 | t0*p11_minus_p21 |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
fmuls p01_minus_p21 // t1*p01_minus_p21 | t1 | t0 | t0*p11_minus_p21 |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fxch %st(2) // t0 | t1 | t1*p01_minus_p21 | t0*p11_minus_p21 |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fxch %st(2) // t0 | t1 | t1*p01_minus_p21 | t0*p11_minus_p21 |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
fmuls p10_minus_p20 // t0*p10_minus_p20 | t1 | t1*p01_minus_p21 |
- // t0*p11_minus_p21 | xstepdenominv |
- // p00_minus_p20 | p11_minus_p21
- fxch %st(1) // t1 | t0*p10_minus_p20 | t1*p01_minus_p21 |
- // t0*p11_minus_p21 | xstepdenominv |
- // p00_minus_p20 | p11_minus_p21
+ // t0*p11_minus_p21 | xstepdenominv |
+ // p00_minus_p20 | p11_minus_p21
+ fxch %st(1) // t1 | t0*p10_minus_p20 | t1*p01_minus_p21 |
+ // t0*p11_minus_p21 | xstepdenominv |
+ // p00_minus_p20 | p11_minus_p21
fmul %st(5),%st(0) // t1*p00_minus_p20 | t0*p10_minus_p20 |
- // t1*p01_minus_p21 | t0*p11_minus_p21 |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fxch %st(2) // t1*p01_minus_p21 | t0*p10_minus_p20 |
- // t1*p00_minus_p20 | t0*p11_minus_p21 |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ // t1*p01_minus_p21 | t0*p11_minus_p21 |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fxch %st(2) // t1*p01_minus_p21 | t0*p10_minus_p20 |
+ // t1*p00_minus_p20 | t0*p11_minus_p21 |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
fsubp %st(0),%st(3) // t0*p10_minus_p20 | t1*p00_minus_p20 |
- // t1*p01_minus_p21 - t0*p11_minus_p21 |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ // t1*p01_minus_p21 - t0*p11_minus_p21 |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
fsubrp %st(0),%st(1) // t1*p00_minus_p20 - t0*p10_minus_p20 |
- // t1*p01_minus_p21 - t0*p11_minus_p21 |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fld %st(2) // xstepdenominv |
- // t1*p00_minus_p20 - t0*p10_minus_p20 |
- // t1*p01_minus_p21 - t0*p11_minus_p21 |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ // t1*p01_minus_p21 - t0*p11_minus_p21 |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fld %st(2) // xstepdenominv |
+ // t1*p00_minus_p20 - t0*p10_minus_p20 |
+ // t1*p01_minus_p21 - t0*p11_minus_p21 |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
fmuls float_minus_1 // ystepdenominv |
- // t1*p00_minus_p20 - t0*p10_minus_p20 |
- // t1*p01_minus_p21 - t0*p11_minus_p21 |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fxch %st(2) // t1*p01_minus_p21 - t0*p11_minus_p21 |
- // t1*p00_minus_p20 - t0*p10_minus_p20 |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
+ // t1*p00_minus_p20 - t0*p10_minus_p20 |
+ // t1*p01_minus_p21 - t0*p11_minus_p21 |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fxch %st(2) // t1*p01_minus_p21 - t0*p11_minus_p21 |
+ // t1*p00_minus_p20 - t0*p10_minus_p20 |
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
fmul %st(3),%st(0) // (t1*p01_minus_p21 - t0*p11_minus_p21)*
- // xstepdenominv |
- // t1*p00_minus_p20 - t0*p10_minus_p20 |
- // | ystepdenominv | xstepdenominv |
- // p00_minus_p20 | p11_minus_p21
- fxch %st(1) // t1*p00_minus_p20 - t0*p10_minus_p20 |
- // (t1*p01_minus_p21 - t0*p11_minus_p21)*
- // xstepdenominv | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ // xstepdenominv |
+ // t1*p00_minus_p20 - t0*p10_minus_p20 |
+ // | ystepdenominv | xstepdenominv |
+ // p00_minus_p20 | p11_minus_p21
+ fxch %st(1) // t1*p00_minus_p20 - t0*p10_minus_p20 |
+ // (t1*p01_minus_p21 - t0*p11_minus_p21)*
+ // xstepdenominv | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
fmul %st(2),%st(0) // (t1*p00_minus_p20 - t0*p10_minus_p20)*
- // ystepdenominv |
- // (t1*p01_minus_p21 - t0*p11_minus_p21)*
- // xstepdenominv | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ // ystepdenominv |
+ // (t1*p01_minus_p21 - t0*p11_minus_p21)*
+ // xstepdenominv | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
fldcw ceil_cw
- fistpl C(r_lstepy) // r_lstepx | ystepdenominv | xstepdenominv |
- // p00_minus_p20 | p11_minus_p21
- fistpl C(r_lstepx) // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
+ fistpl C(r_lstepy) // r_lstepx | ystepdenominv | xstepdenominv |
+ // p00_minus_p20 | p11_minus_p21
+ fistpl C(r_lstepx) // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
fldcw single_cw
// t0 = r_p0[2] - r_p2[2];
// t1 = r_p1[2] - r_p2[2];
- fildl C(r_p2)+8 // r_p2[2] | ystepdenominv | xstepdenominv |
- // p00_minus_p20 | p11_minus_p21
- fildl C(r_p0)+8 // r_p0[2] | r_p2[2] | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fildl C(r_p1)+8 // r_p1[2] | r_p0[2] | r_p2[2] | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fxch %st(2) // r_p2[2] | r_p0[2] | r_p1[2] | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fld %st(0) // r_p2[2] | r_p2[2] | r_p0[2] | r_p1[2] |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
+ fildl C(r_p2)+8 // r_p2[2] | ystepdenominv | xstepdenominv |
+ // p00_minus_p20 | p11_minus_p21
+ fildl C(r_p0)+8 // r_p0[2] | r_p2[2] | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fildl C(r_p1)+8 // r_p1[2] | r_p0[2] | r_p2[2] | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fxch %st(2) // r_p2[2] | r_p0[2] | r_p1[2] | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fld %st(0) // r_p2[2] | r_p2[2] | r_p0[2] | r_p1[2] |
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
fsubrp %st(0),%st(2) // r_p2[2] | t0 | r_p1[2] | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
fsubrp %st(0),%st(2) // t0 | t1 | ystepdenominv | xstepdenominv |
- // p00_minus_p20 | p11_minus_p21
+ // p00_minus_p20 | p11_minus_p21
// r_sstepx = (int)((t1 * p01_minus_p21 - t0 * p11_minus_p21) *
// xstepdenominv);
// r_sstepy = (int)((t1 * p00_minus_p20 - t0 * p10_minus_p20) *
// ystepdenominv);
- fld %st(0) // t0 | t0 | t1 | ystepdenominv | xstepdenominv
+ fld %st(0) // t0 | t0 | t1 | ystepdenominv | xstepdenominv
fmul %st(6),%st(0) // t0*p11_minus_p21 | t0 | t1 | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fxch %st(2) // t1 | t0 | t0*p11_minus_p21 | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fld %st(0) // t1 | t1 | t0 | t0*p11_minus_p21 |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fxch %st(2) // t1 | t0 | t0*p11_minus_p21 | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fld %st(0) // t1 | t1 | t0 | t0*p11_minus_p21 |
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
fmuls p01_minus_p21 // t1*p01_minus_p21 | t1 | t0 | t0*p11_minus_p21 |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
- fxch %st(2) // t0 | t1 | t1*p01_minus_p21 | t0*p11_minus_p21 |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
+ fxch %st(2) // t0 | t1 | t1*p01_minus_p21 | t0*p11_minus_p21 |
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
fmuls p10_minus_p20 // t0*p10_minus_p20 | t1 | t1*p01_minus_p21 |
- // t0*p11_minus_p21 | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fxch %st(1) // t1 | t0*p10_minus_p20 | t1*p01_minus_p21 |
- // t0*p11_minus_p21 | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ // t0*p11_minus_p21 | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fxch %st(1) // t1 | t0*p10_minus_p20 | t1*p01_minus_p21 |
+ // t0*p11_minus_p21 | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
fmul %st(6),%st(0) // t1*p00_minus_p20 | t0*p10_minus_p20 |
- // t1*p01_minus_p21 | t0*p11_minus_p21 |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
- fxch %st(2) // t1*p01_minus_p21 | t0*p10_minus_p20 |
- // t1*p00_minus_p20 | t0*p11_minus_p21 |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
+ // t1*p01_minus_p21 | t0*p11_minus_p21 |
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
+ fxch %st(2) // t1*p01_minus_p21 | t0*p10_minus_p20 |
+ // t1*p00_minus_p20 | t0*p11_minus_p21 |
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
fsubp %st(0),%st(3) // t0*p10_minus_p20 | t1*p00_minus_p20 |
- // t1*p01_minus_p21 - t0*p11_minus_p21 |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
+ // t1*p01_minus_p21 - t0*p11_minus_p21 |
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
fsubrp %st(0),%st(1) // t1*p00_minus_p20 - t0*p10_minus_p20 |
- // t1*p01_minus_p21 - t0*p11_minus_p21 |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
+ // t1*p01_minus_p21 - t0*p11_minus_p21 |
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
fmul %st(2),%st(0) // (t1*p00_minus_p20 - t0*p10_minus_p20)*
- // ystepdenominv |
- // t1*p01_minus_p21 - t0*p11_minus_p21 |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
- fxch %st(1) // t1*p01_minus_p21 - t0*p11_minus_p21 |
- // (t1*p00_minus_p20 - t0*p10_minus_p20)*
- // ystepdenominv | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ // ystepdenominv |
+ // t1*p01_minus_p21 - t0*p11_minus_p21 |
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
+ fxch %st(1) // t1*p01_minus_p21 - t0*p11_minus_p21 |
+ // (t1*p00_minus_p20 - t0*p10_minus_p20)*
+ // ystepdenominv | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
fmul %st(3),%st(0) // (t1*p01_minus_p21 - t0*p11_minus_p21)*
- // xstepdenominv |
- // (t1*p00_minus_p20 - t0*p10_minus_p20)*
- // ystepdenominv | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fxch %st(1) // (t1*p00_minus_p20 - t0*p10_minus_p20)*
- // ystepdenominv |
- // (t1*p01_minus_p21 - t0*p11_minus_p21)*
- // xstepdenominv | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fistpl C(r_sstepy) // r_sstepx | ystepdenominv | xstepdenominv |
- // p00_minus_p20 | p11_minus_p21
- fistpl C(r_sstepx) // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
+ // xstepdenominv |
+ // (t1*p00_minus_p20 - t0*p10_minus_p20)*
+ // ystepdenominv | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fxch %st(1) // (t1*p00_minus_p20 - t0*p10_minus_p20)*
+ // ystepdenominv |
+ // (t1*p01_minus_p21 - t0*p11_minus_p21)*
+ // xstepdenominv | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fistpl C(r_sstepy) // r_sstepx | ystepdenominv | xstepdenominv |
+ // p00_minus_p20 | p11_minus_p21
+ fistpl C(r_sstepx) // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
// t0 = r_p0[3] - r_p2[3];
// t1 = r_p1[3] - r_p2[3];
- fildl C(r_p2)+12 // r_p2[3] | ystepdenominv | xstepdenominv |
- // p00_minus_p20 | p11_minus_p21
- fildl C(r_p0)+12 // r_p0[3] | r_p2[3] | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fildl C(r_p1)+12 // r_p1[3] | r_p0[3] | r_p2[3] | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fxch %st(2) // r_p2[3] | r_p0[3] | r_p1[3] | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fld %st(0) // r_p2[3] | r_p2[3] | r_p0[3] | r_p1[3] |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
+ fildl C(r_p2)+12 // r_p2[3] | ystepdenominv | xstepdenominv |
+ // p00_minus_p20 | p11_minus_p21
+ fildl C(r_p0)+12 // r_p0[3] | r_p2[3] | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fildl C(r_p1)+12 // r_p1[3] | r_p0[3] | r_p2[3] | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fxch %st(2) // r_p2[3] | r_p0[3] | r_p1[3] | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fld %st(0) // r_p2[3] | r_p2[3] | r_p0[3] | r_p1[3] |
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
fsubrp %st(0),%st(2) // r_p2[3] | t0 | r_p1[3] | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
fsubrp %st(0),%st(2) // t0 | t1 | ystepdenominv | xstepdenominv |
- // p00_minus_p20 | p11_minus_p21
+ // p00_minus_p20 | p11_minus_p21
// r_tstepx = (int)((t1 * p01_minus_p21 - t0 * p11_minus_p21) *
// xstepdenominv);
// r_tstepy = (int)((t1 * p00_minus_p20 - t0 * p10_minus_p20) *
// ystepdenominv);
- fld %st(0) // t0 | t0 | t1 | ystepdenominv | xstepdenominv |
- // p00_minus_p20 | p11_minus_p21
+ fld %st(0) // t0 | t0 | t1 | ystepdenominv | xstepdenominv |
+ // p00_minus_p20 | p11_minus_p21
fmul %st(6),%st(0) // t0*p11_minus_p21 | t0 | t1 | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fxch %st(2) // t1 | t0 | t0*p11_minus_p21 | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fld %st(0) // t1 | t1 | t0 | t0*p11_minus_p21 |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fxch %st(2) // t1 | t0 | t0*p11_minus_p21 | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fld %st(0) // t1 | t1 | t0 | t0*p11_minus_p21 |
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
fmuls p01_minus_p21 // t1*p01_minus_p21 | t1 | t0 | t0*p11_minus_p21 |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
- fxch %st(2) // t0 | t1 | t1*p01_minus_p21 | t0*p11_minus_p21 |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
+ fxch %st(2) // t0 | t1 | t1*p01_minus_p21 | t0*p11_minus_p21 |
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
fmuls p10_minus_p20 // t0*p10_minus_p20 | t1 | t1*p01_minus_p21 |
- // t0*p11_minus_p21 | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fxch %st(1) // t1 | t0*p10_minus_p20 | t1*p01_minus_p21 |
- // t0*p11_minus_p21 | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ // t0*p11_minus_p21 | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fxch %st(1) // t1 | t0*p10_minus_p20 | t1*p01_minus_p21 |
+ // t0*p11_minus_p21 | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
fmul %st(6),%st(0) // t1*p00_minus_p20 | t0*p10_minus_p20 |
- // t1*p01_minus_p21 | t0*p11_minus_p21 |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
- fxch %st(2) // t1*p01_minus_p21 | t0*p10_minus_p20 |
- // t1*p00_minus_p20 | t0*p11_minus_p21 |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
+ // t1*p01_minus_p21 | t0*p11_minus_p21 |
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
+ fxch %st(2) // t1*p01_minus_p21 | t0*p10_minus_p20 |
+ // t1*p00_minus_p20 | t0*p11_minus_p21 |
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
fsubp %st(0),%st(3) // t0*p10_minus_p20 | t1*p00_minus_p20 |
- // t1*p01_minus_p21 - t0*p11_minus_p21 |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
+ // t1*p01_minus_p21 - t0*p11_minus_p21 |
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
fsubrp %st(0),%st(1) // t1*p00_minus_p20 - t0*p10_minus_p20 |
- // t1*p01_minus_p21 - t0*p11_minus_p21 |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
+ // t1*p01_minus_p21 - t0*p11_minus_p21 |
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
fmul %st(2),%st(0) // (t1*p00_minus_p20 - t0*p10_minus_p20)*
- // ystepdenominv |
- // t1*p01_minus_p21 - t0*p11_minus_p21 |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
- fxch %st(1) // t1*p01_minus_p21 - t0*p11_minus_p21 |
- // (t1*p00_minus_p20 - t0*p10_minus_p20)*
- // ystepdenominv | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ // ystepdenominv |
+ // t1*p01_minus_p21 - t0*p11_minus_p21 |
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
+ fxch %st(1) // t1*p01_minus_p21 - t0*p11_minus_p21 |
+ // (t1*p00_minus_p20 - t0*p10_minus_p20)*
+ // ystepdenominv | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
fmul %st(3),%st(0) // (t1*p01_minus_p21 - t0*p11_minus_p21)*
- // xstepdenominv |
- // (t1*p00_minus_p20 - t0*p10_minus_p20)*
- // ystepdenominv | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fxch %st(1) // (t1*p00_minus_p20 - t0*p10_minus_p20)*
- // ystepdenominv |
- // (t1*p01_minus_p21 - t0*p11_minus_p21)*
- // xstepdenominv | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fistpl C(r_tstepy) // r_tstepx | ystepdenominv | xstepdenominv |
- // p00_minus_p20 | p11_minus_p21
- fistpl C(r_tstepx) // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
+ // xstepdenominv |
+ // (t1*p00_minus_p20 - t0*p10_minus_p20)*
+ // ystepdenominv | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fxch %st(1) // (t1*p00_minus_p20 - t0*p10_minus_p20)*
+ // ystepdenominv |
+ // (t1*p01_minus_p21 - t0*p11_minus_p21)*
+ // xstepdenominv | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fistpl C(r_tstepy) // r_tstepx | ystepdenominv | xstepdenominv |
+ // p00_minus_p20 | p11_minus_p21
+ fistpl C(r_tstepx) // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
// t0 = r_p0[5] - r_p2[5];
// t1 = r_p1[5] - r_p2[5];
- fildl C(r_p2)+20 // r_p2[5] | ystepdenominv | xstepdenominv |
- // p00_minus_p20 | p11_minus_p21
- fildl C(r_p0)+20 // r_p0[5] | r_p2[5] | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fildl C(r_p1)+20 // r_p1[5] | r_p0[5] | r_p2[5] | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fxch %st(2) // r_p2[5] | r_p0[5] | r_p1[5] | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
- fld %st(0) // r_p2[5] | r_p2[5] | r_p0[5] | r_p1[5] |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // p11_minus_p21
+ fildl C(r_p2)+20 // r_p2[5] | ystepdenominv | xstepdenominv |
+ // p00_minus_p20 | p11_minus_p21
+ fildl C(r_p0)+20 // r_p0[5] | r_p2[5] | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fildl C(r_p1)+20 // r_p1[5] | r_p0[5] | r_p2[5] | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fxch %st(2) // r_p2[5] | r_p0[5] | r_p1[5] | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ fld %st(0) // r_p2[5] | r_p2[5] | r_p0[5] | r_p1[5] |
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // p11_minus_p21
fsubrp %st(0),%st(2) // r_p2[5] | t0 | r_p1[5] | ystepdenominv |
- // xstepdenominv | p00_minus_p20 | p11_minus_p21
+ // xstepdenominv | p00_minus_p20 | p11_minus_p21
fsubrp %st(0),%st(2) // t0 | t1 | ystepdenominv | xstepdenominv |
- // p00_minus_p20 | p11_minus_p21
+ // p00_minus_p20 | p11_minus_p21
// r_zistepx = (int)((t1 * p01_minus_p21 - t0 * p11_minus_p21) *
// xstepdenominv);
// r_zistepy = (int)((t1 * p00_minus_p20 - t0 * p10_minus_p20) *
// ystepdenominv);
- fld %st(0) // t0 | t0 | t1 | ystepdenominv | xstepdenominv |
- // p00_minus_p20 | p11_minus_p21
+ fld %st(0) // t0 | t0 | t1 | ystepdenominv | xstepdenominv |
+ // p00_minus_p20 | p11_minus_p21
fmulp %st(0),%st(6) // t0 | t1 | ystepdenominv | xstepdenominv |
- // p00_minus_p20 | t0*p11_minus_p21
- fxch %st(1) // t1 | t0 | ystepdenominv | xstepdenominv |
- // p00_minus_p20 | t0*p11_minus_p21
- fld %st(0) // t1 | t1 | t0 | ystepdenominv | xstepdenominv |
- // p00_minus_p20 | t0*p11_minus_p21
+ // p00_minus_p20 | t0*p11_minus_p21
+ fxch %st(1) // t1 | t0 | ystepdenominv | xstepdenominv |
+ // p00_minus_p20 | t0*p11_minus_p21
+ fld %st(0) // t1 | t1 | t0 | ystepdenominv | xstepdenominv |
+ // p00_minus_p20 | t0*p11_minus_p21
fmuls p01_minus_p21 // t1*p01_minus_p21 | t1 | t0 | ystepdenominv |
- // xstepdenominv | p00_minus_p20 |
- // t0*p11_minus_p21
- fxch %st(2) // t0 | t1 | t1*p01_minus_p21 | ystepdenominv |
- // xstepdenominv | p00_minus_p20 |
- // t0*p11_minus_p21
+ // xstepdenominv | p00_minus_p20 |
+ // t0*p11_minus_p21
+ fxch %st(2) // t0 | t1 | t1*p01_minus_p21 | ystepdenominv |
+ // xstepdenominv | p00_minus_p20 |
+ // t0*p11_minus_p21
fmuls p10_minus_p20 // t0*p10_minus_p20 | t1 | t1*p01_minus_p21 |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // t0*p11_minus_p21
- fxch %st(1) // t1 | t0*p10_minus_p20 | t1*p01_minus_p21 |
- // ystepdenominv | xstepdenominv | p00_minus_p20 |
- // t0*p11_minus_p21
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // t0*p11_minus_p21
+ fxch %st(1) // t1 | t0*p10_minus_p20 | t1*p01_minus_p21 |
+ // ystepdenominv | xstepdenominv | p00_minus_p20 |
+ // t0*p11_minus_p21
fmulp %st(0),%st(5) // t0*p10_minus_p20 | t1*p01_minus_p21 |
- // ystepdenominv | xstepdenominv |
- // t1*p00_minus_p20 | t0*p11_minus_p21
- fxch %st(5) // t0*p11_minus_p21 | t1*p01_minus_p21 |
- // ystepdenominv | xstepdenominv |
- // t1*p00_minus_p20 | t0*p10_minus_p20
+ // ystepdenominv | xstepdenominv |
+ // t1*p00_minus_p20 | t0*p11_minus_p21
+ fxch %st(5) // t0*p11_minus_p21 | t1*p01_minus_p21 |
+ // ystepdenominv | xstepdenominv |
+ // t1*p00_minus_p20 | t0*p10_minus_p20
fsubrp %st(0),%st(1) // t1*p01_minus_p21 - t0*p11_minus_p21 |
- // ystepdenominv | xstepdenominv |
- // t1*p00_minus_p20 | t0*p10_minus_p20
- fxch %st(3) // t1*p00_minus_p20 | ystepdenominv |
- // xstepdenominv |
- // t1*p01_minus_p21 - t0*p11_minus_p21 |
- // t0*p10_minus_p20
+ // ystepdenominv | xstepdenominv |
+ // t1*p00_minus_p20 | t0*p10_minus_p20
+ fxch %st(3) // t1*p00_minus_p20 | ystepdenominv |
+ // xstepdenominv |
+ // t1*p01_minus_p21 - t0*p11_minus_p21 |
+ // t0*p10_minus_p20
fsubp %st(0),%st(4) // ystepdenominv | xstepdenominv |
- // t1*p01_minus_p21 - t0*p11_minus_p21 |
- // t1*p00_minus_p20 - t0*p10_minus_p20
- fxch %st(1) // xstepdenominv | ystepdenominv |
- // t1*p01_minus_p21 - t0*p11_minus_p21 |
- // t1*p00_minus_p20 - t0*p10_minus_p20
+ // t1*p01_minus_p21 - t0*p11_minus_p21 |
+ // t1*p00_minus_p20 - t0*p10_minus_p20
+ fxch %st(1) // xstepdenominv | ystepdenominv |
+ // t1*p01_minus_p21 - t0*p11_minus_p21 |
+ // t1*p00_minus_p20 - t0*p10_minus_p20
fmulp %st(0),%st(2) // ystepdenominv |
- // (t1*p01_minus_p21 - t0*p11_minus_p21) *
- // xstepdenominv |
- // t1*p00_minus_p20 - t0*p10_minus_p20
+ // (t1*p01_minus_p21 - t0*p11_minus_p21) *
+ // xstepdenominv |
+ // t1*p00_minus_p20 - t0*p10_minus_p20
fmulp %st(0),%st(2) // (t1*p01_minus_p21 - t0*p11_minus_p21) *
- // xstepdenominv |
- // (t1*p00_minus_p20 - t0*p10_minus_p20) *
- // ystepdenominv
+ // xstepdenominv |
+ // (t1*p00_minus_p20 - t0*p10_minus_p20) *
+ // ystepdenominv
fistpl C(r_zistepx) // (t1*p00_minus_p20 - t0*p10_minus_p20) *
- // ystepdenominv
+ // ystepdenominv
fistpl C(r_zistepy)
// a_sstepxfrac = r_sstepx << 16;
@@ -479,16 +477,16 @@
.globl C(D_PolysetRecursiveTriangle)
C(D_PolysetRecursiveTriangle):
- pushl %ebp // preserve caller stack frame pointer
- pushl %esi // preserve register variables
+ pushl %ebp // preserve caller stack frame pointer
+ pushl %esi // preserve register variables
pushl %edi
pushl %ebx
-// int *temp;
-// int d;
-// int new[6];
-// int i;
-// int z;
+// int *temp;
+// int d;
+// int new[6];
+// int i;
+// int z;
// short *zbuf;
movl lp2(%esp),%esi
movl lp1(%esp),%ebx
@@ -509,7 +507,7 @@
incl %eax
cmpl $2,%eax
- ja LSplit
+ ja LSplit
// d = lp2[1] - lp1[1];
// if (d < -1 || d > 1)
@@ -518,7 +516,7 @@
incl %ebp
cmpl $2,%ebp
- ja LSplit
+ ja LSplit
// d = lp3[0] - lp2[0];
// if (d < -1 || d > 1)
@@ -533,7 +531,7 @@
incl %eax
cmpl $2,%eax
- ja LSplit2
+ ja LSplit2
// d = lp3[1] - lp2[1];
// if (d < -1 || d > 1)
@@ -542,7 +540,7 @@
incl %ebp
cmpl $2,%ebp
- ja LSplit2
+ ja LSplit2
// d = lp1[0] - lp3[0];
// if (d < -1 || d > 1)
@@ -560,7 +558,7 @@
movl %ebx,%edx
cmpl $2,%eax
- ja LSplit3
+ ja LSplit3
// d = lp1[1] - lp3[1];
// if (d < -1 || d > 1)
@@ -576,13 +574,13 @@
// return; // entire tri is filled
//
cmpl $2,%ebp
- jna LDone
+ jna LDone
LSplit3:
movl %edi,%ebx
movl %esi,%edi
movl %edx,%esi
- jmp LSplit
+ jmp LSplit
//split2:
LSplit2:
@@ -649,15 +647,15 @@
// if (lp2[1] > lp1[1])
// goto nodraw;
cmpl %eax,4(%esi)
- jg LNoDraw
+ jg LNoDraw
// if ((lp2[1] == lp1[1]) && (lp2[0] < lp1[0]))
// goto nodraw;
movl 0(%esi),%edx
- jnz LDraw
+ jnz LDraw
cmpl %ecx,%edx
- jl LNoDraw
+ jl LNoDraw
LDraw:
@@ -730,7 +728,7 @@
popl %edi
popl %esi
popl %ebp // restore caller stack frame pointer
- ret $12
+ ret $12
//----------------------------------------------------------------------
@@ -745,16 +743,16 @@
.globl C(D_PolysetDrawSpans8)
C(D_PolysetDrawSpans8):
- pushl %esi // preserve register variables
+ pushl %esi // preserve register variables
pushl %ebx
movl pspans(%esp),%esi // point to the first span descriptor
movl C(r_zistepx),%ecx
- pushl %ebp // preserve caller's stack frame
+ pushl %ebp // preserve caller's stack frame
pushl %edi
- rorl $16,%ecx // put high 16 bits of 1/z step in low word
+ rorl $16,%ecx // put high 16 bits of 1/z step in low word
movl spanpackage_t_count(%esi),%edx
movl %ecx,lzistepx
@@ -779,7 +777,7 @@
movl C(erroradjustup),%edx
movl C(errorterm),%ebx
addl %edx,%ebx
- js LNoTurnover
+ js LNoTurnover
movl C(erroradjustdown),%edx
movl C(d_countextrastep),%edi
@@ -788,7 +786,7 @@
movl %ebx,C(errorterm)
addl %edi,%ebp
movl %ebp,C(d_aspancount)
- jmp LRightEdgeStepped
+ jmp LRightEdgeStepped
LNoTurnover:
movl C(d_aspancount),%edi
@@ -800,8 +798,8 @@
LRightEdgeStepped:
cmpl $1,%eax
- jl LNextSpan
- jz LExactlyOneLong
+ jl LNextSpan
+ jz LExactlyOneLong
//
// set up advancetable
@@ -845,7 +843,7 @@
pushl %esi
movl spanpackage_t_ptex(%esi),%esi
- jmp aff8entryvec_table(,%eax,4)
+ jmp aff8entryvec_table(,%eax,4)
// %bx = count of full and partial loops
// %ebx high word = sfrac
@@ -867,7 +865,7 @@
LDraw8:
cmpw (%ecx),%bp
- jl Lp1
+ jl Lp1
xorl %eax,%eax
movb %dh,%ah
movb (%esi),%al
@@ -885,7 +883,7 @@
LDraw7:
cmpw 2(%ecx),%bp
- jl Lp2
+ jl Lp2
xorl %eax,%eax
movb %dh,%ah
movb (%esi),%al
@@ -903,7 +901,7 @@
LDraw6:
cmpw 4(%ecx),%bp
- jl Lp3
+ jl Lp3
xorl %eax,%eax
movb %dh,%ah
movb (%esi),%al
@@ -921,7 +919,7 @@
LDraw5:
cmpw 6(%ecx),%bp
- jl Lp4
+ jl Lp4
xorl %eax,%eax
movb %dh,%ah
movb (%esi),%al
@@ -939,7 +937,7 @@
LDraw4:
cmpw 8(%ecx),%bp
- jl Lp5
+ jl Lp5
xorl %eax,%eax
movb %dh,%ah
movb (%esi),%al
@@ -957,7 +955,7 @@
LDraw3:
cmpw 10(%ecx),%bp
- jl Lp6
+ jl Lp6
xorl %eax,%eax
movb %dh,%ah
movb (%esi),%al
@@ -975,7 +973,7 @@
LDraw2:
cmpw 12(%ecx),%bp
- jl Lp7
+ jl Lp7
xorl %eax,%eax
movb %dh,%ah
movb (%esi),%al
@@ -993,7 +991,7 @@
LDraw1:
cmpw 14(%ecx),%bp
- jl Lp8
+ jl Lp8
xorl %eax,%eax
movb %dh,%ah
movb (%esi),%al
@@ -1013,15 +1011,15 @@
addl $16,%ecx
decw %bx
- jnz LDrawLoop
+ jnz LDrawLoop
popl %esi // restore spans pointer
LNextSpan:
addl $(spanpackage_t_size),%esi // point to next span
LNextSpanESISet:
movl spanpackage_t_count(%esi),%edx
- cmpl $-999999,%edx // any more spans?
- jnz LSpanLoop // yes
+ cmpl $-999999,%edx // any more spans?
+ jnz LSpanLoop // yes
popl %edi
popl %ebp // restore the caller's stack frame
@@ -1041,7 +1039,7 @@
movl spanpackage_t_ptex(%esi),%ebx
cmpw (%ecx),%bp
- jl LNextSpan
+ jl LNextSpan
xorl %eax,%eax
movl spanpackage_t_pdest(%esi),%edi
movb spanpackage_t_light+1(%esi),%ah
@@ -1052,7 +1050,7 @@
LPatch9:
movb %al,(%edi)
- jmp LNextSpanESISet
+ jmp LNextSpanESISet
.globl C(D_PolysetAff8End)
C(D_PolysetAff8End):
@@ -1102,7 +1100,7 @@
// D_DrawNonSubdiv ();
movl C(r_affinetridesc)+atd_drawtype,%eax
testl %eax,%eax
- jz C(D_DrawNonSubdiv)
+ jz C(D_DrawNonSubdiv)
pushl %ebp // preserve caller stack frame pointer
@@ -1161,15 +1159,15 @@
fildl fv_v+4(%esi) // i1v1 | i0v1
fildl fv_v+0(%ecx) // i0v0 | i1v1 | i0v1
fildl fv_v+0(%edx) // i2v0 | i0v0 | i1v1 | i0v1
- fxch %st(2) // i1v1 | i0v0 | i2v0 | i0v1
+ fxch %st(2) // i1v1 | i0v0 | i2v0 | i0v1
fsubr %st(3),%st(0) // i0v1-i1v1 | i0v0 | i2v0 | i0v1
fildl fv_v+0(%esi) // i1v0 | i0v1-i1v1 | i0v0 | i2v0 | i0v1
- fxch %st(2) // i0v0 | i0v1-i1v1 | i1v0 | i2v0 | i0v1
+ fxch %st(2) // i0v0 | i0v1-i1v1 | i1v0 | i2v0 | i0v1
fsub %st(0),%st(3) // i0v0 | i0v1-i1v1 | i1v0 | i0v0-i2v0 | i0v1
fildl fv_v+4(%edx) // i2v1 | i0v0 | i0v1-i1v1 | i1v0 | i0v0-i2v0| i0v1
- fxch %st(1) // i0v0 | i2v1 | i0v1-i1v1 | i1v0 | i0v0-i2v0| i0v1
+ fxch %st(1) // i0v0 | i2v1 | i0v1-i1v1 | i1v0 | i0v0-i2v0| i0v1
fsubp %st(0),%st(3) // i2v1 | i0v1-i1v1 | i0v0-i1v0 | i0v0-i2v0 | i0v1
- fxch %st(1) // i0v1-i1v1 | i2v1 | i0v0-i1v0 | i0v0-i2v0 | i0v1
+ fxch %st(1) // i0v1-i1v1 | i2v1 | i0v0-i1v0 | i0v0-i2v0 | i0v1
fmulp %st(0),%st(3) // i2v1 | i0v0-i1v0 | i0v1-i1v1*i0v0-i2v0 | i0v1
fsubrp %st(0),%st(3) // i0v0-i1v0 | i0v1-i1v1*i0v0-i2v0 | i0v1-i2v1
movl fv_v+16(%ecx),%eax
@@ -1181,14 +1179,14 @@
fstps Ltemp
movl Ltemp,%eax
subl $0x80000001,%eax
- jc Lskip
+ jc Lskip
// if (ptri[i].facesfront)
// {
// D_PolysetRecursiveTriangle(index0->v, index1->v, index2->v);
movl mtri_facesfront-16(%ebx,%ebp,),%eax
testl %eax,%eax
- jz Lfacesback
+ jz Lfacesback
pushl %edx
pushl %esi
@@ -1196,8 +1194,8 @@
call C(D_PolysetRecursiveTriangle)
subl $16,%ebp
- jnz Llooptop
- jmp Ldone2
+ jnz Llooptop
+ jmp Ldone2
// }
// else
@@ -1220,21 +1218,21 @@
// index0->v[2] += r_affinetridesc.seamfixupX16;
movl C(r_affinetridesc)+atd_seamfixupX16,%eax
testl $(ALIAS_ONSEAM),fv_flags(%ecx)
- jz Lp11
+ jz Lp11
addl %eax,fv_v+8(%ecx)
Lp11:
// if (index1->flags & ALIAS_ONSEAM)
// index1->v[2] += r_affinetridesc.seamfixupX16;
testl $(ALIAS_ONSEAM),fv_flags(%esi)
- jz Lp12
+ jz Lp12
addl %eax,fv_v+8(%esi)
Lp12:
// if (index2->flags & ALIAS_ONSEAM)
// index2->v[2] += r_affinetridesc.seamfixupX16;
testl $(ALIAS_ONSEAM),fv_flags(%edx)
- jz Lp13
+ jz Lp13
addl %eax,fv_v+8(%edx)
Lp13:
@@ -1260,7 +1258,7 @@
// }
Lskip:
subl $16,%ebp
- jnz Llooptop
+ jnz Llooptop
Ldone2:
popl %edi // restore the caller's stack frame
@@ -1290,7 +1288,7 @@
movl C(d_sfrac),%ecx
andl $0xFFFF,%eax
movl C(d_ptex),%ebx
- orl %eax,%ecx
+ orl %eax,%ecx
movl C(d_pedgespanpackage),%esi
movl C(d_tfrac),%edx
movl C(d_light),%edi
@@ -1344,7 +1342,7 @@
// if (errorterm >= 0)
// {
- js LNoLeftEdgeTurnover
+ js LNoLeftEdgeTurnover
// errorterm -= erroradjustdown;
// d_pdest += d_pdestextrastep;
@@ -1373,7 +1371,7 @@
// if (d_tfrac & 0x10000)
// {
- jnc LSkip1
+ jnc LSkip1
// d_ptex += r_affinetridesc.skinwidth;
// d_tfrac &= 0xFFFF;
@@ -1392,7 +1390,7 @@
movl C(d_pedgespanpackage),%esi
decl %ecx
testl $0xFFFF,%ecx
- jnz LScanLoop
+ jnz LScanLoop
popl %ebx
popl %edi
@@ -1431,7 +1429,7 @@
// if (d_tfrac & 0x10000)
// {
- jnc LSkip2
+ jnc LSkip2
// d_ptex += r_affinetridesc.skinwidth;
// d_tfrac &= 0xFFFF;
@@ -1451,7 +1449,7 @@
movl C(d_pedgespanpackage),%esi
decl %ecx
testl $0xFFFF,%ecx
- jnz LScanLoop
+ jnz LScanLoop
popl %ebx
popl %edi
@@ -1493,11 +1491,11 @@
movl fv_v+0(%ebx),%eax
movl C(r_refdef)+rd_vrectright,%edx
cmpl %edx,%eax
- jge LNextVert
+ jge LNextVert
movl fv_v+4(%ebx),%esi
movl C(r_refdef)+rd_vrectbottom,%edx
cmpl %edx,%esi
- jge LNextVert
+ jge LNextVert
// zbuf = zspantable[fv->v[1]] + fv->v[0];
movl C(zspantable)(,%esi,4),%edi
@@ -1510,7 +1508,7 @@
// {
// int pix;
cmpw (%edi,%eax,2),%dx
- jl LNextVert
+ jl LNextVert
// *zbuf = z;
movw %dx,(%edi,%eax,2)
@@ -1543,7 +1541,7 @@
LNextVert:
addl $(fv_size),%ebx
decl %ecx
- jnz LFVLoop
+ jnz LFVLoop
popl %edi
popl %esi
@@ -1560,18 +1558,18 @@
.globl C(D_DrawNonSubdiv)
C(D_DrawNonSubdiv):
- pushl %ebp // preserve caller stack frame pointer
+ pushl %ebp // preserve caller stack frame pointer
movl C(r_affinetridesc)+atd_numtriangles,%ebp
pushl %ebx
shll $(mtri_shift),%ebp
- pushl %esi // preserve register variables
+ pushl %esi // preserve register variables
movl C(r_affinetridesc)+atd_ptriangles,%esi
pushl %edi
-// mtriangle_t *ptri;
-// finalvert_t *pfv, *index0, *index1, *index2;
-// int i;
-// int lnumtriangles;
+// mtriangle_t *ptri;
+// finalvert_t *pfv, *index0, *index1, *index2;
+// int i;
+// int lnumtriangles;
// pfv = r_affinetridesc.pfinalverts;
// ptri = r_affinetridesc.ptriangles;
@@ -1596,8 +1594,9 @@
addl %edi,%ebx
// d_xdenom = (index0->v[1]-index1->v[1]) *
-// (index0->v[0]-index2->v[0]) -
-// (index0->v[0]-index1->v[0])*(index0->v[1]-index2->v[1]);
+// (index0->v[0]-index2->v[0]) -
+// (index0->v[0]-index1->v[0]) *
+// (index0->v[1]-index2->v[1]);
movl fv_v+4(%ecx),%eax
movl fv_v+0(%ecx),%esi
subl fv_v+4(%edx),%eax
@@ -1613,7 +1612,7 @@
// if (d_xdenom >= 0)
// {
// continue;
- jns LNextTri
+ jns LNextTri
// }
@@ -1684,7 +1683,7 @@
// if (!ptri->facesfront)
// {
testl %eax,%eax
- jnz LFacesFront
+ jnz LFacesFront
// if (index0->flags & ALIAS_ONSEAM)
// r_p0[2] += r_affinetridesc.seamfixupX16;
@@ -1693,21 +1692,21 @@
movl fv_flags(%ebx),%edi
testl $(ALIAS_ONSEAM),%eax
movl C(r_affinetridesc)+atd_seamfixupX16,%eax
- jz LOnseamDone0
+ jz LOnseamDone0
addl %eax,C(r_p0)+8
LOnseamDone0:
// if (index1->flags & ALIAS_ONSEAM)
// r_p1[2] += r_affinetridesc.seamfixupX16;
testl $(ALIAS_ONSEAM),%esi
- jz LOnseamDone1
+ jz LOnseamDone1
addl %eax,C(r_p1)+8
LOnseamDone1:
// if (index2->flags & ALIAS_ONSEAM)
// r_p2[2] += r_affinetridesc.seamfixupX16;
testl $(ALIAS_ONSEAM),%edi
- jz LOnseamDone2
+ jz LOnseamDone2
addl %eax,C(r_p2)+8
LOnseamDone2:
@@ -1719,13 +1718,13 @@
// D_PolysetSetEdgeTable ();
// D_RasterizeAliasPolySmooth ();
- call C(D_PolysetSetEdgeTable)
- call C(D_RasterizeAliasPolySmooth)
+ call C(D_PolysetSetEdgeTable)
+ call C(D_RasterizeAliasPolySmooth)
LNextTri:
- movl C(r_affinetridesc)+atd_ptriangles,%esi
- subl $16,%ebp
- jnz LNDLoop
+ movl C(r_affinetridesc)+atd_ptriangles,%esi
+ subl $16,%ebp
+ jnz LNDLoop
// }
popl %edi
@@ -1739,4 +1738,3 @@
#endif // id386
-