From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([140.186.70.92]:47460)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <blauwirbel@gmail.com>) id 1QMQP1-0005XV-Ao
	for qemu-devel@nongnu.org; Tue, 17 May 2011 15:58:00 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <blauwirbel@gmail.com>) id 1QMQP0-0004qB-AE
	for qemu-devel@nongnu.org; Tue, 17 May 2011 15:57:59 -0400
Received: from mail-qy0-f180.google.com ([209.85.216.180]:54295)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <blauwirbel@gmail.com>) id 1QMQP0-0004q5-1R
	for qemu-devel@nongnu.org; Tue, 17 May 2011 15:57:58 -0400
Received: by qyk10 with SMTP id 10so513016qyk.4
	for <qemu-devel@nongnu.org>; Tue, 17 May 2011 12:57:57 -0700 (PDT)
MIME-Version: 1.0
In-Reply-To: <BANLkTi=Bv=+6QgcHC0=jGiRFOc3D3ug2JA@mail.gmail.com>
References: <BANLkTikeSfsp537pf-unOyR4ftxctmfP2w@mail.gmail.com>
	<BANLkTi=Bv=+6QgcHC0=jGiRFOc3D3ug2JA@mail.gmail.com>
From: Blue Swirl <blauwirbel@gmail.com>
Date: Tue, 17 May 2011 22:57:37 +0300
Message-ID: <BANLkTinHBB6pg9E-YeiPQS3ny1WRyHdPHw@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Subject: Re: [Qemu-devel] [PATCH 2/2][SPARC] Fix TA0_Shutdown feature
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: Julien Grall <julien.grall@gmail.com>
Cc: qemu-devel@nongnu.org, Fabien Chouteau <chouteau@adacore.com>

On Tue, May 17, 2011 at 6:32 PM, Julien Grall <julien.grall@gmail.com> wrot=
e:
> Fix TA0_SHUTDOWN feature

But what would be the bug?

> Signed-off-by: Julien Grall <julien.grall@gmail.com>
> ---
> =C2=A0target-sparc/op_helper.c | =C2=A0 13 +++++++++++--
> =C2=A0target-sparc/translate.c | =C2=A0 =C2=A09 +--------
> =C2=A02 files changed, 12 insertions(+), 10 deletions(-)
>
> diff --git a/target-sparc/op_helper.c b/target-sparc/op_helper.c
> index a6fabad..cb775f5 100644
> --- a/target-sparc/op_helper.c
> +++ b/target-sparc/op_helper.c
> @@ -326,8 +326,17 @@ void HELPER(raise_exception)(int tt)
>
> =C2=A0void HELPER(trap_always)(int tt)
> =C2=A0{
> - =C2=A0 =C2=A0env->exception_index =3D tt;
> - =C2=A0 =C2=A0do_interrupt(env);
> + =C2=A0 =C2=A0if (tt =3D=3D TT_TRAP
> + =C2=A0 =C2=A0 =C2=A0 =C2=A0&& env->def->features & CPU_FEATURE_TA0_SHUT=
DOWN
> +#ifndef TARGET_SPARC64
> + =C2=A0 =C2=A0 =C2=A0 =C2=A0&& env->psret =3D=3D 0
> +#endif
> + =C2=A0 =C2=A0 =C2=A0 =C2=A0) {
> + =C2=A0 =C2=A0 =C2=A0 =C2=A0helper_shutdown();
> + =C2=A0 =C2=A0} else {
> + =C2=A0 =C2=A0 =C2=A0 =C2=A0env->exception_index =3D tt;
> + =C2=A0 =C2=A0 =C2=A0 =C2=A0do_interrupt(env);
> + =C2=A0 =C2=A0}
> =C2=A0}
>
> =C2=A0void helper_shutdown(void)
> diff --git a/target-sparc/translate.c b/target-sparc/translate.c
> index b30003b..a47a2de 100644
> --- a/target-sparc/translate.c
> +++ b/target-sparc/translate.c
> @@ -2009,14 +2009,7 @@ static void disas_sparc_insn(DisasContext * dc)
> =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 tcg=
_gen_addi_tl(cpu_dst, cpu_dst, TT_TRAP);
> =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 tcg=
_gen_trunc_tl_i32(cpu_tmp32, cpu_dst);
>
> - =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0if=
 (rs2 =3D=3D 0 &&
> - =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0dc->def->features & CPU_FEATURE_TA0_SHUTDOWN) {
> -
> - =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0gen_helper_shutdown();
> -
> - =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0} =
else {
> - =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 =C2=A0gen_helper_trap_always(cpu_tmp32);
> - =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0}
> + =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0ge=
n_helper_trap_always(cpu_tmp32);

No, this would actually be just opposite to how QEMU works.
Performance comes from doing more work at translation time in order to
save time during executing the generated code.