From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:33429) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1ejWED-00038a-VH for qemu-devel@nongnu.org; Wed, 07 Feb 2018 15:22:03 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1ejWE9-0007lF-0G for qemu-devel@nongnu.org; Wed, 07 Feb 2018 15:22:01 -0500 Received: from 1.mo2.mail-out.ovh.net ([46.105.63.121]:46724) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1ejWE8-0007kC-P7 for qemu-devel@nongnu.org; Wed, 07 Feb 2018 15:21:56 -0500 Received: from player157.ha.ovh.net (gw6.ovh.net [213.251.189.206]) by mo2.mail-out.ovh.net (Postfix) with ESMTP id 39CBB118E86 for ; Wed, 7 Feb 2018 21:21:55 +0100 (CET) Date: Wed, 7 Feb 2018 21:21:48 +0100 From: Greg Kurz Message-ID: <20180207212148.73c77de1@bahia.lan> In-Reply-To: <20180207161730.17194-1-lvivier@redhat.com> References: <20180207161730.17194-1-lvivier@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [Qemu-ppc] [PATCH] spapr: check smp_threads < vsmt List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Laurent Vivier Cc: qemu-devel@nongnu.org, qemu-ppc@nongnu.org, David Gibson On Wed, 7 Feb 2018 17:17:30 +0100 Laurent Vivier wrote: > We ignore silently the value of smp_threads when we set > the VSMT value, and if smp_threads is greater than VSMT > kernel is going into trouble later. > > Fixes: 8904e5a750 > ("spapr: Adjust default VSMT value for better migration compatibility") > > Signed-off-by: Laurent Vivier > --- > hw/ppc/spapr.c | 17 ++++++++--------- > 1 file changed, 8 insertions(+), 9 deletions(-) > > diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c > index 32a876be56..638b3cafd0 100644 > --- a/hw/ppc/spapr.c > +++ b/hw/ppc/spapr.c > @@ -2294,15 +2294,7 @@ static void spapr_set_vsmt_mode(sPAPRMachineState *spapr, Error **errp) > } > > /* Detemine the VSMT mode to use: */ > - if (vsmt_user) { > - if (spapr->vsmt < smp_threads) { > - error_setg(&local_err, "Cannot support VSMT mode %d" > - " because it must be >= threads/core (%d)", > - spapr->vsmt, smp_threads); In this case, we'd error out because the user requested two settings that are known to be incompatible. > - goto out; > - } > - /* In this case, spapr->vsmt has been set by the command line */ > - } else { > + if (!vsmt_user) { > /* > * Default VSMT value is tricky, because we need it to be as > * consistent as possible (for migration), but this requires > @@ -2313,6 +2305,13 @@ static void spapr_set_vsmt_mode(sPAPRMachineState *spapr, Error **errp) > spapr->vsmt = 8; In this case, we'd error out because the user requested a setting that is incompatible with our default. Maybe worth a separate error message ? "Cannot support %d threads/core because it must be <= to default VSMT mode (8)" Anyway, the fix is good: Reviewed-by: Greg Kurz > } > > + if (spapr->vsmt < smp_threads) { > + error_setg(&local_err, "Cannot support VSMT mode %d" > + " because it must be >= threads/core (%d)", > + spapr->vsmt, smp_threads); > + goto out; > + } > + > /* KVM: If necessary, set the SMT mode: */ > if (kvm_enabled() && (spapr->vsmt != kvm_smt)) { > ret = kvmppc_set_smt_threads(spapr->vsmt);