From mboxrd@z Thu Jan  1 00:00:00 1970
From: erik quanstrom <quanstro@labs.coraid.com>
Date: Wed, 30 Jan 2013 22:50:49 -0500
To: 9fans@9fans.net
Message-ID: <e3167c3256b3ba2a465fdbb4055318ad@coraid.com>
In-Reply-To: <20130130221845.CB4BAB82A@mail.bitblocks.com>
References: <87c109280f2fc2bdaa0c68867f849daf@hamnavoe.com>
	<d1329b9babc534669c4b183f34394a01@kw.quanstro.net>
	<20130130221845.CB4BAB82A@mail.bitblocks.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Subject: Re: [9fans] emulated fp on arm
Topicbox-Message-UUID: 11fe0f28-ead8-11e9-9d60-3106f5b1d025

> As ieee754 (1985):
>=20
>   When the sum of two operands with opposite signs (or the
>   difference of two operands with like signs) is exactly zero,
>   the sign of that sum (or difference) shall be + in all
>   rounding modes except round toward -INFINITY, in which mode
>   that sign shall be -.

due to the emulator interface, both cases call fpisub.
i've run a number of tests on this with relatively small
numbers that are equal.  and this seems to solve the
problems without creating new ones.

- erik

; diff -c fpi.c /sys/src/9/omap
fpi.c:137,143 - /sys/src/9/omap/fpi.c:137,142
  void
  fpisub(Internal *x, Internal *y, Internal *i)
  {
- 	int exact;
  	Internal *t;
 =20
  	if(y->e < x->e
fpi.c:158,164 - /sys/src/9/omap/fpi.c:157,162
  			SetInfinity(i);
  		return;
  	}
- 	exact =3D x->e =3D=3D y->e;
  	matchexponents(x, y);
  	i->e =3D y->e;
  	i->h =3D y->h - x->h;
fpi.c:167,179 - /sys/src/9/omap/fpi.c:165,172
  		i->l +=3D CarryBit;
  		i->h--;
  	}
- 	if(i->h =3D=3D 0 && i->l =3D=3D 0){
+ 	if(i->h =3D=3D 0 && i->l =3D=3D 0)
  		SetZero(i);
-=20
- 		/* canceling rule */
- 		if(x->s !=3D y->s && exact)
- 			i->s =3D 0;	/* 1 in round to -=E2=88=9E mode */
- 	}
  	else while(i->e > 1 && (i->h & HiddenBit) =3D=3D 0)
  		shift(i);
  }