From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([208.118.235.92]:50901)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <afaerber@suse.de>) id 1SRBPr-00082C-1U
	for qemu-devel@nongnu.org; Sun, 06 May 2012 20:03:04 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <afaerber@suse.de>) id 1SRBPp-0001uD-2i
	for qemu-devel@nongnu.org; Sun, 06 May 2012 20:03:02 -0400
Message-ID: <4FA71132.7080409@suse.de>
Date: Mon, 07 May 2012 02:02:58 +0200
From: =?ISO-8859-1?Q?Andreas_F=E4rber?= <afaerber@suse.de>
MIME-Version: 1.0
References: <4FA54951.90908@suse.de>
	<CAAu8pHtfhHOH0fCpKLYFer2JO3Ws8nTJMNQjRR6W+T-TCbmSVA@mail.gmail.com>
	<4FA68483.8090805@suse.de>
	<CAAu8pHvE9kc25WobGO0eE10MnB21ksMOyP2TSG+ZXZa9SLV4OA@mail.gmail.com>
	<4FA6CF58.6050201@suse.de>
	<alpine.LNX.2.00.1205062325170.2923@linmac>
In-Reply-To: <alpine.LNX.2.00.1205062325170.2923@linmac>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Subject: Re: [Qemu-devel] sparc-softmmu uninitialized memory read?
List-Id: <qemu-devel.nongnu.org>
List-Unsubscribe: <https://lists.nongnu.org/mailman/options/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.nongnu.org/archive/html/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <https://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: malc <av1474@comtv.ru>
Cc: Blue Swirl <blauwirbel@gmail.com>, qemu-ppc <qemu-ppc@nongnu.org>, Alexander Graf <agraf@suse.de>, qemu-devel <qemu-devel@nongnu.org>

Am 06.05.2012 21:27, schrieb malc:
> The attached patch is broken for non SysV calling conventions, would be
> nice if you could test things on Darwin (and, if your power5 box still =
has
> AIX, on AIX)

I replaced AIX 5.x with openSUSE, sorry. :)

> diff --git a/tcg/ppc/tcg-target.c b/tcg/ppc/tcg-target.c
> index dc40716..311af18 100644
> --- a/tcg/ppc/tcg-target.c
> +++ b/tcg/ppc/tcg-target.c
[...]
> @@ -810,6 +829,17 @@ static void tcg_out_qemu_st (TCGContext *s, const =
TCGArg *args, int opc)
>  #endif
> =20
>      /* slow path */
> +#ifdef CONFIG_TCG_PASS_AREG0
> +    tcg_out_mov (s, TCG_TYPE_I32, 3, TCG_AREG0);
> +#if TARGET_LONG_BITS =3D=3D 32
> +    tcg_out_mov (s, TCG_TYPE_I32, 4, addr_reg);
> +    ir =3D 5;
> +#else
> +    tcg_out_mov (s, TCG_TYPE_I32, 5, addr_reg2);
> +    tcg_out_mov (s, TCG_TYPE_I32, 6, addr_reg);

Here we should be using r4 + r5 for non-aligned targets. Alternative
patch sent that hopefully avoids such issues and the code duplication.

If you prefer two separate code paths for some reason, please at least
consider using a fool-proof alignment macro such as proposed.

/-F

> +    ir =3D 7;
> +#endif
> +#else
>  #if TARGET_LONG_BITS =3D=3D 32
>      tcg_out_mov (s, TCG_TYPE_I32, 3, addr_reg);
>      ir =3D 4;
[snip]

--=20
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 N=FCrnberg, Germany
GF: Jeff Hawn, Jennifer Guild, Felix Imend=F6rffer; HRB 16746 AG N=FCrnbe=
rg