mirror of
https://github.com/AuxXxilium/linux_dsm_epyc7002.git
synced 2024-12-05 16:06:42 +07:00
1b35a57b1c
For the explicit calls to .udiv/.umul in assembler, I made a mechanical (read as: safe) transformation. I didn't attempt to make any simplifications. In particular, __ndelay and __udelay can be simplified significantly. Some of the %y reads are unnecessary and these routines have no need any longer for allocating a register window, they can be leaf functions. Signed-off-by: David S. Miller <davem@davemloft.net>
282 lines
5.5 KiB
ArmAsm
282 lines
5.5 KiB
ArmAsm
/* Copyright (C) 1989, 1992, 1993, 1994, 1995 Free Software Foundation, Inc.
|
|
|
|
This file is part of GNU CC.
|
|
|
|
GNU CC is free software; you can redistribute it and/or modify
|
|
it under the terms of the GNU General Public License as published by
|
|
the Free Software Foundation; either version 2, or (at your option)
|
|
any later version.
|
|
|
|
GNU CC 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 General Public License for more details.
|
|
|
|
You should have received a copy of the GNU General Public License
|
|
along with GNU CC; see the file COPYING. If not, write to
|
|
the Free Software Foundation, 59 Temple Place - Suite 330,
|
|
Boston, MA 02111-1307, USA. */
|
|
|
|
.text
|
|
.align 4
|
|
.globl __divdi3
|
|
__divdi3:
|
|
save %sp,-104,%sp
|
|
cmp %i0,0
|
|
bge .LL40
|
|
mov 0,%l4
|
|
mov -1,%l4
|
|
sub %g0,%i1,%o0
|
|
mov %o0,%o5
|
|
subcc %g0,%o0,%g0
|
|
sub %g0,%i0,%o0
|
|
subx %o0,0,%o4
|
|
mov %o4,%i0
|
|
mov %o5,%i1
|
|
.LL40:
|
|
cmp %i2,0
|
|
bge .LL84
|
|
mov %i3,%o4
|
|
xnor %g0,%l4,%l4
|
|
sub %g0,%i3,%o0
|
|
mov %o0,%o3
|
|
subcc %g0,%o0,%g0
|
|
sub %g0,%i2,%o0
|
|
subx %o0,0,%o2
|
|
mov %o2,%i2
|
|
mov %o3,%i3
|
|
mov %i3,%o4
|
|
.LL84:
|
|
cmp %i2,0
|
|
bne .LL45
|
|
mov %i1,%i3
|
|
cmp %o4,%i0
|
|
bleu .LL46
|
|
mov %i3,%o1
|
|
mov 32,%g1
|
|
subcc %i0,%o4,%g0
|
|
1: bcs 5f
|
|
addxcc %o1,%o1,%o1 ! shift n1n0 and a q-bit in lsb
|
|
sub %i0,%o4,%i0 ! this kills msb of n
|
|
addx %i0,%i0,%i0 ! so this cannot give carry
|
|
subcc %g1,1,%g1
|
|
2: bne 1b
|
|
subcc %i0,%o4,%g0
|
|
bcs 3f
|
|
addxcc %o1,%o1,%o1 ! shift n1n0 and a q-bit in lsb
|
|
b 3f
|
|
sub %i0,%o4,%i0 ! this kills msb of n
|
|
4: sub %i0,%o4,%i0
|
|
5: addxcc %i0,%i0,%i0
|
|
bcc 2b
|
|
subcc %g1,1,%g1
|
|
! Got carry from n. Subtract next step to cancel this carry.
|
|
bne 4b
|
|
addcc %o1,%o1,%o1 ! shift n1n0 and a 0-bit in lsb
|
|
sub %i0,%o4,%i0
|
|
3: xnor %o1,0,%o1
|
|
b .LL50
|
|
mov 0,%o2
|
|
.LL46:
|
|
cmp %o4,0
|
|
bne .LL85
|
|
mov %i0,%o2
|
|
mov 1,%o0
|
|
mov 0,%o1
|
|
wr %g0, 0, %y
|
|
udiv %o0, %o1, %o0
|
|
mov %o0,%o4
|
|
mov %i0,%o2
|
|
.LL85:
|
|
mov 0,%g3
|
|
mov 32,%g1
|
|
subcc %g3,%o4,%g0
|
|
1: bcs 5f
|
|
addxcc %o2,%o2,%o2 ! shift n1n0 and a q-bit in lsb
|
|
sub %g3,%o4,%g3 ! this kills msb of n
|
|
addx %g3,%g3,%g3 ! so this cannot give carry
|
|
subcc %g1,1,%g1
|
|
2: bne 1b
|
|
subcc %g3,%o4,%g0
|
|
bcs 3f
|
|
addxcc %o2,%o2,%o2 ! shift n1n0 and a q-bit in lsb
|
|
b 3f
|
|
sub %g3,%o4,%g3 ! this kills msb of n
|
|
4: sub %g3,%o4,%g3
|
|
5: addxcc %g3,%g3,%g3
|
|
bcc 2b
|
|
subcc %g1,1,%g1
|
|
! Got carry from n. Subtract next step to cancel this carry.
|
|
bne 4b
|
|
addcc %o2,%o2,%o2 ! shift n1n0 and a 0-bit in lsb
|
|
sub %g3,%o4,%g3
|
|
3: xnor %o2,0,%o2
|
|
mov %g3,%i0
|
|
mov %i3,%o1
|
|
mov 32,%g1
|
|
subcc %i0,%o4,%g0
|
|
1: bcs 5f
|
|
addxcc %o1,%o1,%o1 ! shift n1n0 and a q-bit in lsb
|
|
sub %i0,%o4,%i0 ! this kills msb of n
|
|
addx %i0,%i0,%i0 ! so this cannot give carry
|
|
subcc %g1,1,%g1
|
|
2: bne 1b
|
|
subcc %i0,%o4,%g0
|
|
bcs 3f
|
|
addxcc %o1,%o1,%o1 ! shift n1n0 and a q-bit in lsb
|
|
b 3f
|
|
sub %i0,%o4,%i0 ! this kills msb of n
|
|
4: sub %i0,%o4,%i0
|
|
5: addxcc %i0,%i0,%i0
|
|
bcc 2b
|
|
subcc %g1,1,%g1
|
|
! Got carry from n. Subtract next step to cancel this carry.
|
|
bne 4b
|
|
addcc %o1,%o1,%o1 ! shift n1n0 and a 0-bit in lsb
|
|
sub %i0,%o4,%i0
|
|
3: xnor %o1,0,%o1
|
|
b .LL86
|
|
mov %o1,%l1
|
|
.LL45:
|
|
cmp %i2,%i0
|
|
bleu .LL51
|
|
sethi %hi(65535),%o0
|
|
b .LL78
|
|
mov 0,%o1
|
|
.LL51:
|
|
or %o0,%lo(65535),%o0
|
|
cmp %i2,%o0
|
|
bgu .LL58
|
|
mov %i2,%o1
|
|
cmp %i2,256
|
|
addx %g0,-1,%o0
|
|
b .LL64
|
|
and %o0,8,%o2
|
|
.LL58:
|
|
sethi %hi(16777215),%o0
|
|
or %o0,%lo(16777215),%o0
|
|
cmp %i2,%o0
|
|
bgu .LL64
|
|
mov 24,%o2
|
|
mov 16,%o2
|
|
.LL64:
|
|
srl %o1,%o2,%o0
|
|
sethi %hi(__clz_tab),%o1
|
|
or %o1,%lo(__clz_tab),%o1
|
|
ldub [%o0+%o1],%o0
|
|
add %o0,%o2,%o0
|
|
mov 32,%o1
|
|
subcc %o1,%o0,%o3
|
|
bne,a .LL72
|
|
sub %o1,%o3,%o1
|
|
cmp %i0,%i2
|
|
bgu .LL74
|
|
cmp %i3,%o4
|
|
blu .LL78
|
|
mov 0,%o1
|
|
.LL74:
|
|
b .LL78
|
|
mov 1,%o1
|
|
.LL72:
|
|
sll %i2,%o3,%o2
|
|
srl %o4,%o1,%o0
|
|
or %o2,%o0,%i2
|
|
sll %o4,%o3,%o4
|
|
srl %i0,%o1,%o2
|
|
sll %i0,%o3,%o0
|
|
srl %i3,%o1,%o1
|
|
or %o0,%o1,%i0
|
|
sll %i3,%o3,%i3
|
|
mov %i0,%o1
|
|
mov 32,%g1
|
|
subcc %o2,%i2,%g0
|
|
1: bcs 5f
|
|
addxcc %o1,%o1,%o1 ! shift n1n0 and a q-bit in lsb
|
|
sub %o2,%i2,%o2 ! this kills msb of n
|
|
addx %o2,%o2,%o2 ! so this cannot give carry
|
|
subcc %g1,1,%g1
|
|
2: bne 1b
|
|
subcc %o2,%i2,%g0
|
|
bcs 3f
|
|
addxcc %o1,%o1,%o1 ! shift n1n0 and a q-bit in lsb
|
|
b 3f
|
|
sub %o2,%i2,%o2 ! this kills msb of n
|
|
4: sub %o2,%i2,%o2
|
|
5: addxcc %o2,%o2,%o2
|
|
bcc 2b
|
|
subcc %g1,1,%g1
|
|
! Got carry from n. Subtract next step to cancel this carry.
|
|
bne 4b
|
|
addcc %o1,%o1,%o1 ! shift n1n0 and a 0-bit in lsb
|
|
sub %o2,%i2,%o2
|
|
3: xnor %o1,0,%o1
|
|
mov %o2,%i0
|
|
wr %g0,%o1,%y ! SPARC has 0-3 delay insn after a wr
|
|
sra %o4,31,%g2 ! Do not move this insn
|
|
and %o1,%g2,%g2 ! Do not move this insn
|
|
andcc %g0,0,%g1 ! Do not move this insn
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,%o4,%g1
|
|
mulscc %g1,0,%g1
|
|
add %g1,%g2,%o0
|
|
rd %y,%o2
|
|
cmp %o0,%i0
|
|
bgu,a .LL78
|
|
add %o1,-1,%o1
|
|
bne,a .LL50
|
|
mov 0,%o2
|
|
cmp %o2,%i3
|
|
bleu .LL50
|
|
mov 0,%o2
|
|
add %o1,-1,%o1
|
|
.LL78:
|
|
mov 0,%o2
|
|
.LL50:
|
|
mov %o1,%l1
|
|
.LL86:
|
|
mov %o2,%l0
|
|
mov %l0,%i0
|
|
mov %l1,%i1
|
|
cmp %l4,0
|
|
be .LL81
|
|
sub %g0,%i1,%o0
|
|
mov %o0,%l3
|
|
subcc %g0,%o0,%g0
|
|
sub %g0,%i0,%o0
|
|
subx %o0,0,%l2
|
|
mov %l2,%i0
|
|
mov %l3,%i1
|
|
.LL81:
|
|
ret
|
|
restore
|