researchv9/jerq/src/lib/j/texture32.os - view

File: [Research Unix] / researchv9 / jerq / src / lib / j / texture32.os
Revision 1.1.1.1 (vendor branch): download - view: text, annotated - select for diffs
Tue Apr 24 17:21:59 2018 UTC (8 years, 1 month ago) by root
Branches: belllabs, MAIN
CVS tags: researchv9-SUN3_old, researchv9-SUN3, HEAD

researchv9-SUN3(old)

.text .globl texture32 texture32: cmph 8(%ap),4(%ap) # width in bits BGB dxpos # width positive ?? RET # NO!, return dxpos: SAVE %r3 # lets save all the registers addw2 &8,%sp # local vars PUSHW 0(%ap) # push pointer to bitmap PUSHW 4(%ap) # push rect.origin call &2,addr # calculate the address of the point MOVW %r0,%r8 # %r8 = addr(bitmap,rect.origin) movw 0(%ap),%r7 # pointer to bitmap llsw3 &2,4(%r7),%r7 # width of bitmap in bytes -> r7 movw 12(%ap),%r6 # r6 = pointer to texture subh3 6(%ap),10(%ap),%r4 # dy andh3 &0x1f,6(%ap),%r0 llsw3 &2,%r0,%r3 # bit index into texture = r3 movh 4(%ap),%r2 # rec.origin.x -> r2 lrsw3 %r2,&-1,0(%fp) # generate mask1 -> 0(%fp) subh3 &1,8(%ap),%r5 arsw3 %r5,&0x80000000,4(%fp) # generate mask2 -> 4(%fp) lrsw3 &5,%r5,%r5 # (x + dx -1) >> WORDSHIFT lrsw3 &5,%r2,%r2 # x >> WORDSHIFT subw2 %r2,%r5 DECW %r5 # inner word count ii BGEB notnarrow # if ii == -1 then narrow andw2 4(%fp),0(%fp) # narrow, combine both masks here mcomw 0(%fp),4(%fp) # mask2 = ~mask1 subw2 &4,%r7 # amount to get to next line BRB restcode notnarrow: addw3 &1,%r5,%r0 # offset in words llsw3 &2,%r0,%r0 # offset in bytes subw2 %r0,%r7 # amount to get to next line restcode: ALSW3 &2,16(%ap),%r0 # code * 4 -> r0 BLB end # bad if code less than 0 cmpw %r0,&12 # see if code greater than 3 jle *jmptbl(%r0) # jump to correct routine if ok end: subw2 &8,%sp RESTORE %r3 RET .data jmptbl: .word FSTORE .word FOR .word FCLEAR .word FXOR .text FXOR: DECW %r4 # number of lines jl end # if neg. then done addw3 %r6,%r3,%r2 movw 0(%r2),%r2 # bits = map[bitindex] addw2 &4,%r3 #bitindex += 1; andw2 &0x7f,%r3 # bitindex = ( bitindex + 1) & WORDMASK andw3 0(%fp),%r2,%r0 # xorw2 %r0,0(%r8) #screenp ^= bits & mask1 addw2 &4,%r8 # screenp++ movw %r5,%r1 # i = ii BLB restloop # if i = -1 then narrow BEB mask2ok # if i = 0 then no inner words lrsw3 &2,%r1,%r0 # i/4 BLEB less4 # less than 4 inner words more4: xorw2 %r2,0(%r8) # screenp ^= bits xorw2 %r2,4(%r8) # screenp ^= bits xorw2 %r2,8(%r8) # screenp ^= bits xorw2 %r2,12(%r8) # screenp ^= bits addw2 &16,%r8 # screenp += 4 DECW %r0 BGB more4 # more inner words andw2 &3,%r1 # just how many are left BLEB mask2ok # no more to do less4: xorw2 %r2,0(%r8) # screenp ^= bits addw2 &4,%r8 # screenp++ DECW %r1 BGB less4 # more inner words mask2ok: andw3 4(%fp),%r2,%r0 xorw2 %r0,0(%r8) # screenp ^= (bits & mask2) restloop: addw2 %r7,%r8 # screenp += width BRB FXOR FCLEAR: mcomw 0(%fp),0(%fp) mcomw 4(%fp),4(%fp) FCLR: DECW %r4 # number of lines jl end # if neg. then done addw3 %r6,%r3,%r2 mcomw 0(%r2),%r2 # bits = ~map[bitindex] addw2 &4,%r3 #bitindex += 1; andw2 &0x7f,%r3 # bitindex = ( bitindex + 1) & WORDMASK orw3 0(%fp),%r2,%r0 # andw2 %r0,0(%r8) #screenp &= bits & mask1 addw2 &4,%r8 # screenp++ movw %r5,%r1 # i = ii BLB restclr # if i = -1 then narrow BEB msk2clr # if i = 0 then no inner words lrsw3 &2,%r1,%r0 # i/4 BLEB less4clr # less than 4 inner words Fclr_4: andw2 %r2,0(%r8) # screenp ^= bits andw2 %r2,4(%r8) # screenp ^= bits andw2 %r2,8(%r8) # screenp ^= bits andw2 %r2,12(%r8) # screenp ^= bits addw2 &16,%r8 # screenp += 4 DECW %r0 BGB Fclr_4 # more inner words andw2 &3,%r1 # just how many are left BLEB msk2clr # no more to do less4clr: andw2 %r2,0(%r8) # screenp ^= bits addw2 &4,%r8 # screenp++ DECW %r1 BGB less4clr # more inner words msk2clr: orw3 4(%fp),%r2,%r0 andw2 %r0,0(%r8) # screenp ^= (bits & mask2) restclr: addw2 %r7,%r8 # screenp += width BRB FCLR FOR: DECW %r4 # number of lines jl end # if neg. then done addw3 %r6,%r3,%r2 movw 0(%r2),%r2 # bits = map[bitindex] addw2 &4,%r3 #bitindex += 1; andw2 &0x7f,%r3 # bitindex = ( bitindex + 1) & WORDMASK andw3 0(%fp),%r2,%r0 # orw2 %r0,0(%r8) #screenp ^= bits & mask1 addw2 &4,%r8 # screenp++ movw %r5,%r1 # i = ii BLB rest_or # if i = -1 then narrow BEB mask2or # if i = 0 then no inner words lrsw3 &2,%r1,%r0 # i/4 BLEB less4or # less than 4 inner words For_4: orw2 %r2,0(%r8) # screenp ^= bits orw2 %r2,4(%r8) # screenp ^= bits orw2 %r2,8(%r8) # screenp ^= bits orw2 %r2,12(%r8) # screenp ^= bits addw2 &16,%r8 # screenp += 4 DECW %r0 BGB For_4 # more inner words andw2 &3,%r1 # just how many are left BLEB mask2or # no more to do less4or: orw2 %r2,0(%r8) # screenp ^= bits addw2 &4,%r8 # screenp++ DECW %r1 BGB less4or # more inner words mask2or: andw3 4(%fp),%r2,%r0 orw2 %r0,0(%r8) # screenp ^= (bits & mask2) rest_or: addw2 %r7,%r8 # screenp += width BRB FOR FSTORE: DECW %r4 # number of lines jl end # if neg. then done addw3 %r6,%r3,%r2 movw 0(%r2),%r2 # bits = map[bitindex] addw2 &4,%r3 #bitindex += 1; andw2 &0x7f,%r3 # bitindex = ( bitindex + 1) & WORDMASK andw3 0(%fp),%r2,%r0 # xorw3 0(%r8),%r2,%r0 #screenp = bits & mask1 andw2 0(%fp),%r0 xorw2 %r0,0(%r8) # addw2 &4,%r8 # screenp++ movw %r5,%r1 # i = ii BLB stoloop # if i = -1 then narrow BEB sto2ok # if i = 0 then no inner words lrsw3 &2,%r1,%r0 # i/4 BLEB st4less # less than 4 inner words store4: movw %r2,0(%r8) # screenp = bits movw %r2,4(%r8) # screenp = bits movw %r2,8(%r8) # screenp = bits movw %r2,12(%r8) # screenp = bits addw2 &16,%r8 # screenp += 4 DECW %r0 BGB store4 # more inner words andw2 &3,%r1 # just how many are left BLEB sto2ok # no more to do st4less: movw %r2,0(%r8) # screenp ^= bits addw2 &4,%r8 # screenp++ DECW %r1 BGB st4less # more inner words sto2ok: xorw3 0(%r8),%r2,%r0 #screenp = bits & mask2 andw2 4(%fp),%r0 xorw2 %r0,0(%r8) # stoloop: addw2 %r7,%r8 # screenp += width BRB FSTORE

This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.