/* SPARC v9 __mpn_submul_1 -- Multiply a limb vector with a single limb and
   subtract the product from a second limb vector.

   Copyright (C) 1996-2023 Free Software Foundation, Inc.

   This file is part of the GNU MP Library.

   The GNU MP Library is free software; you can redistribute it and/or modify
   it under the terms of the GNU Lesser General Public License as published by
   the Free Software Foundation; either version 2.1 of the License, or (at your
   option) any later version.

   The GNU MP Library 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 Lesser General Public
   License for more details.

   You should have received a copy of the GNU Lesser General Public License
   along with the GNU MP Library; see the file COPYING.LIB.  If not,
   see <https://www.gnu.org/licenses/>.  */

#include <sysdep.h>


/* INPUT PARAMETERS
   res_ptr	o0
   s1_ptr	o1
   size		o2
   s2_limb	o3  */

ENTRY(__mpn_submul_1)
	save	%sp,-192,%sp

	sub	%g0,%i2,%o7
	mov	0,%o0			! zero cy_limb
	sllx	%o7,3,%o7
	sethi	%hi(0x80000000),%o2
	srl	%i3,0,%o1		! extract low 32 bits of s2_limb
	sub	%i1,%o7,%o3
	srlx	%i3,32,%i3		! extract high 32 bits of s2_limb
	sub	%i0,%o7,%o4
	add	%o2,%o2,%o2		! o2 = 0x100000000

	!   hi   !
             !  mid-1 !
             !  mid-2 !
		 !   lo   !
1:
	ldx	[%o3+%o7],%g5
	srl	%g5,0,%i0		! zero hi bits
	ldx	[%o4+%o7],%l1
	srlx	%g5,32,%g5
	mulx	%o1,%i0,%i4		! lo product
	mulx	%i3,%i0,%i1		! mid-1 product
	mulx	%o1,%g5,%l2		! mid-2 product
	mulx	%i3,%g5,%i5		! hi product
	srlx	%i4,32,%i0		! extract high 32 bits of lo product...
	add	%i1,%i0,%i1		! ...and add it to the mid-1 product
	addcc	%i1,%l2,%i1		! add mid products
	mov	0,%l0			! we need the carry from that add...
	movcs	%xcc,%o2,%l0		! ...compute it and...
	sllx	%i1,32,%i0		!  align low bits of mid product
	add	%i5,%l0,%i5		! ...add to bit 32 of the hi product
	srl	%i4,0,%g5		! zero high 32 bits of lo product
	add	%i0,%g5,%i0		! combine into low 64 bits of result
	srlx	%i1,32,%i1		! extract high bits of mid product...
	addcc	%i0,%o0,%i0		!  add cy_limb to low 64 bits of result
	add	%i5,%i1,%i1		! ...and add them to the high result
	mov	0,%g5
	movcs	%xcc,1,%g5
	subcc	%l1,%i0,%i0
	stx	%i0,[%o4+%o7]
	add	%g5,1,%l1
	movcs	%xcc,%l1,%g5
	addcc	%o7,8,%o7
	bne,pt	%xcc,1b
	 add	%i1,%g5,%o0		! compute new cy_limb

	jmpl	%i7+8, %g0
	 restore %o0,%g0,%o0

END(__mpn_submul_1)