From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([209.51.188.92]:44420) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1gti4o-00024w-2l for qemu-devel@nongnu.org; Tue, 12 Feb 2019 19:07:01 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1gti49-0008Lt-7i for qemu-devel@nongnu.org; Tue, 12 Feb 2019 19:06:18 -0500 Message-ID: <8adeacf3681e03343d64fac82b73505ffa122bfd.camel@kernel.crashing.org> From: Benjamin Herrenschmidt Date: Wed, 13 Feb 2019 11:04:27 +1100 In-Reply-To: <20190212060227.GO1884@umbus.fritz.box> References: <20190128094625.4428-1-clg@kaod.org> <20190128094625.4428-9-clg@kaod.org> <20190212060227.GO1884@umbus.fritz.box> Content-Type: text/plain; charset="UTF-8" Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PATCH 08/19] target/ppc: Fix nip on power management instructions List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: David Gibson , =?ISO-8859-1?Q?C=E9dric?= Le Goater Cc: qemu-ppc@nongnu.org, qemu-devel@nongnu.org On Tue, 2019-02-12 at 17:02 +1100, David Gibson wrote: > On Mon, Jan 28, 2019 at 10:46:14AM +0100, C=C3=A9dric Le Goater wrote: > > From: Benjamin Herrenschmidt > >=20 > > Those instructions currently raise an exception from within > > the helper. This tends to result in a bogus nip value in > > the env context (typically the beginning of the TB). Such > > a helper needs a gen_update_nip() first. > >=20 > > This fixes it with a different approach which is to throw > > the exception from translate.c instead of the helper using > > gen_exception_nip() which does the right thing. > >=20 > > Signed-off-by: Benjamin Herrenschmidt > > Signed-off-by: C=C3=A9dric Le Goater > > --- > > target/ppc/excp_helper.c | 1 - > > target/ppc/translate.c | 12 ++++++++---- > > 2 files changed, 8 insertions(+), 5 deletions(-) > >=20 > > diff --git a/target/ppc/excp_helper.c b/target/ppc/excp_helper.c > > index 751d759fcc1d..8407e0ade938 100644 > > --- a/target/ppc/excp_helper.c > > +++ b/target/ppc/excp_helper.c > > @@ -958,7 +958,6 @@ void helper_pminsn(CPUPPCState *env, powerpc_pm_i= nsn_t insn) > > * but this doesn't seem to be a problem. > > */ > > env->msr |=3D (1ull << MSR_EE); > > - raise_exception(env, EXCP_HLT); > > } > > #endif /* defined(TARGET_PPC64) */ > > =20 > > diff --git a/target/ppc/translate.c b/target/ppc/translate.c > > index 7d40a1fbe6bd..55281a8975e0 100644 > > --- a/target/ppc/translate.c > > +++ b/target/ppc/translate.c > > @@ -3571,7 +3571,8 @@ static void gen_doze(DisasContext *ctx) > > t =3D tcg_const_i32(PPC_PM_DOZE); > > gen_helper_pminsn(cpu_env, t); > > tcg_temp_free_i32(t); > > - gen_stop_exception(ctx); > > + /* Stop translation, as the CPU is supposed to sleep from now */ > > + gen_exception_nip(ctx, EXCP_HLT, ctx->base.pc_next); >=20 > IIUC this also changes from EXCP_STOP to EXCP_HLT, is that intentional? Off the top of my head, it might be to break out of the outer exec loop, but I don't remember off hand. > > #endif /* defined(CONFIG_USER_ONLY) */ > > } > > =20 > > @@ -3586,7 +3587,8 @@ static void gen_nap(DisasContext *ctx) > > t =3D tcg_const_i32(PPC_PM_NAP); > > gen_helper_pminsn(cpu_env, t); > > tcg_temp_free_i32(t); > > - gen_stop_exception(ctx); > > + /* Stop translation, as the CPU is supposed to sleep from now */ > > + gen_exception_nip(ctx, EXCP_HLT, ctx->base.pc_next); > > #endif /* defined(CONFIG_USER_ONLY) */ > > } > > =20 > > @@ -3606,7 +3608,8 @@ static void gen_sleep(DisasContext *ctx) > > t =3D tcg_const_i32(PPC_PM_SLEEP); > > gen_helper_pminsn(cpu_env, t); > > tcg_temp_free_i32(t); > > - gen_stop_exception(ctx); > > + /* Stop translation, as the CPU is supposed to sleep from now */ > > + gen_exception_nip(ctx, EXCP_HLT, ctx->base.pc_next); > > #endif /* defined(CONFIG_USER_ONLY) */ > > } > > =20 > > @@ -3621,7 +3624,8 @@ static void gen_rvwinkle(DisasContext *ctx) > > t =3D tcg_const_i32(PPC_PM_RVWINKLE); > > gen_helper_pminsn(cpu_env, t); > > tcg_temp_free_i32(t); > > - gen_stop_exception(ctx); > > + /* Stop translation, as the CPU is supposed to sleep from now */ > > + gen_exception_nip(ctx, EXCP_HLT, ctx->base.pc_next); > > #endif /* defined(CONFIG_USER_ONLY) */ > > } > > #endif /* #if defined(TARGET_PPC64) */