Home | History | Annotate | Line # | Download | only in pa32
sub_n.asm revision 1.1 
      1  1.1  mrg dnl  HP-PA mpn_sub_n -- Subtract two limb vectors of the same length > 0 and
      2  1.1  mrg dnl  store difference in a third limb vector.
      3  1.1  mrg 
      4  1.1  mrg dnl  Copyright 1992, 1994, 2000, 2001, 2002 Free Software Foundation, Inc.
      5  1.1  mrg 
      6  1.1  mrg dnl  This file is part of the GNU MP Library.
      7  1.1  mrg 
      8  1.1  mrg dnl  The GNU MP Library is free software; you can redistribute it and/or modify
      9  1.1  mrg dnl  it under the terms of the GNU Lesser General Public License as published
     10  1.1  mrg dnl  by the Free Software Foundation; either version 3 of the License, or (at
     11  1.1  mrg dnl  your option) any later version.
     12  1.1  mrg 
     13  1.1  mrg dnl  The GNU MP Library is distributed in the hope that it will be useful, but
     14  1.1  mrg dnl  WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
     15  1.1  mrg dnl  or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public
     16  1.1  mrg dnl  License for more details.
     17  1.1  mrg 
     18  1.1  mrg dnl  You should have received a copy of the GNU Lesser General Public License
     19  1.1  mrg dnl  along with the GNU MP Library.  If not, see http://www.gnu.org/licenses/.
     20  1.1  mrg 
     21  1.1  mrg include(`../config.m4')
     22  1.1  mrg 
     23  1.1  mrg C INPUT PARAMETERS
     24  1.1  mrg C res_ptr	gr26
     25  1.1  mrg C s1_ptr	gr25
     26  1.1  mrg C s2_ptr	gr24
     27  1.1  mrg C size		gr23
     28  1.1  mrg 
     29  1.1  mrg C One might want to unroll this as for other processors, but it turns out that
     30  1.1  mrg C the data cache contention after a store makes such unrolling useless.  We
     31  1.1  mrg C can't come under 5 cycles/limb anyway.
     32  1.1  mrg 
     33  1.1  mrg ASM_START()
     34  1.1  mrg PROLOGUE(mpn_sub_n)
     35  1.1  mrg 	ldws,ma		4(0,%r25),%r20
     36  1.1  mrg 	ldws,ma		4(0,%r24),%r19
     37  1.1  mrg 
     38  1.1  mrg 	addib,=		-1,%r23,L(end)	C check for (SIZE == 1)
     39  1.1  mrg 	 sub		%r20,%r19,%r28	C subtract first limbs ignoring cy
     40  1.1  mrg 
     41  1.1  mrg LDEF(loop)
     42  1.1  mrg 	ldws,ma		4(0,%r25),%r20
     43  1.1  mrg 	ldws,ma		4(0,%r24),%r19
     44  1.1  mrg 	stws,ma		%r28,4(0,%r26)
     45  1.1  mrg 	addib,<>	-1,%r23,L(loop)
     46  1.1  mrg 	 subb		%r20,%r19,%r28
     47  1.1  mrg 
     48  1.1  mrg LDEF(end)
     49  1.1  mrg 	stws		%r28,0(0,%r26)
     50  1.1  mrg 	addc		%r0,%r0,%r28
     51  1.1  mrg 	bv		0(%r2)
     52  1.1  mrg 	 subi		1,%r28,%r28
     53  1.1  mrg EPILOGUE()
     54