.skip 32 // makes the loop body aligned at 64-byte boundary
bn_add_words:
.prologue
- .fframe 0
.save ar.pfs,r2
{ .mii; alloc r2=ar.pfs,4,12,0,16
cmp4.le p6,p0=r35,r0 };;
{ .mfb; mov r8=r0 // return value
(p6) br.ret.spnt.many b0 };;
- .save ar.lc,r3
{ .mib; sub r10=r35,r0,1
+ .save ar.lc,r3
mov r3=ar.lc
brp.loop.imp .L_bn_add_words_ctop,.L_bn_add_words_cend-16
}
.skip 32 // makes the loop body aligned at 64-byte boundary
bn_sub_words:
.prologue
- .fframe 0
.save ar.pfs,r2
{ .mii; alloc r2=ar.pfs,4,12,0,16
cmp4.le p6,p0=r35,r0 };;
{ .mfb; mov r8=r0 // return value
(p6) br.ret.spnt.many b0 };;
- .save ar.lc,r3
{ .mib; sub r10=r35,r0,1
+ .save ar.lc,r3
mov r3=ar.lc
brp.loop.imp .L_bn_sub_words_ctop,.L_bn_sub_words_cend-16
}
.skip 32 // makes the loop body aligned at 64-byte boundary
bn_mul_words:
.prologue
- .fframe 0
.save ar.pfs,r2
#ifdef XMA_TEMPTATION
{ .mfi; alloc r2=ar.pfs,4,0,0,0 };;
cmp4.le p6,p0=r34,r0
(p6) br.ret.spnt.many b0 };;
- .save ar.lc,r3
{ .mii; sub r10=r34,r0,1
+ .save ar.lc,r3
mov r3=ar.lc
mov r9=pr };;
.skip 48 // makes the loop body aligned at 64-byte boundary
bn_mul_add_words:
.prologue
- .fframe 0
.save ar.pfs,r2
- .save ar.lc,r3
- .save pr,r9
{ .mmi; alloc r2=ar.pfs,4,4,0,8
cmp4.le p6,p0=r34,r0
+ .save ar.lc,r3
mov r3=ar.lc };;
{ .mib; mov r8=r0 // return value
sub r10=r34,r0,1
.body
{ .mib; setf.sig f8=r35 // w
+ .save pr,r9
mov r9=pr
brp.loop.imp .L_bn_mul_add_words_ctop,.L_bn_mul_add_words_cend-16
}
.skip 32 // makes the loop body aligned at 64-byte boundary
bn_sqr_words:
.prologue
- .fframe 0
.save ar.pfs,r2
{ .mii; alloc r2=ar.pfs,3,0,0,0
sxt4 r34=r34 };;
nop.f 0x0
(p6) br.ret.spnt.many b0 };;
- .save ar.lc,r3
{ .mii; sub r10=r34,r0,1
+ .save ar.lc,r3
mov r3=ar.lc
+ .save pr,r9
mov r9=pr };;
.body
.align 64
bn_sqr_comba8:
.prologue
- .fframe 0
.save ar.pfs,r2
#if defined(_HPUX_SOURCE) && !defined(_LP64)
{ .mii; alloc r2=ar.pfs,2,1,0,0
.align 64
bn_mul_comba8:
.prologue
- .fframe 0
.save ar.pfs,r2
#if defined(_HPUX_SOURCE) && !defined(_LP64)
{ .mii; alloc r2=ar.pfs,3,0,0,0
.align 64
bn_sqr_comba4:
.prologue
- .fframe 0
.save ar.pfs,r2
#if defined(_HPUX_SOURCE) && !defined(_LP64)
{ .mii; alloc r2=ar.pfs,2,1,0,0
.align 64
bn_mul_comba4:
.prologue
- .fframe 0
.save ar.pfs,r2
#if defined(_HPUX_SOURCE) && !defined(_LP64)
{ .mii; alloc r2=ar.pfs,3,0,0,0
.align 64
bn_div_words:
.prologue
- .fframe 0
.save ar.pfs,r2
- .save b0,r3
{ .mii; alloc r2=ar.pfs,3,5,0,8
+ .save b0,r3
mov r3=b0
mov r10=pr };;
{ .mmb; cmp.eq p6,p0=r34,r0