From mboxrd@z Thu Jan 1 00:00:00 1970 From: Keir Fraser Subject: Re: [PATCH v2 1/3] x86/HPET: obtain proper lock for changing IRQ affinity Date: Wed, 17 Oct 2012 12:58:36 +0100 Message-ID: References: <507EB69302000078000A1F46@nat28.tlf.novell.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <507EB69302000078000A1F46@nat28.tlf.novell.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Jan Beulich Cc: xen-devel List-Id: xen-devel@lists.xenproject.org On 17/10/2012 12:45, "Jan Beulich" wrote: > The IRQ descriptor lock should be held while adjusting the affinity of > any IRQ; the HPET channel lock isn't sufficient to protect namely > against races with moving the IRQ to a different CPU. > > Signed-off-by: Jan Beulich Acked-by: Keir Fraser > --- > v2: rename new helper function > > --- a/xen/arch/x86/hpet.c > +++ b/xen/arch/x86/hpet.c > @@ -436,6 +436,16 @@ static struct hpet_event_channel *hpet_g > return ch; > } > > +static void set_channel_irq_affinity(const struct hpet_event_channel *ch) > +{ > + struct irq_desc *desc = irq_to_desc(ch->irq); > + > + ASSERT(!local_irq_is_enabled()); > + spin_lock(&desc->lock); > + hpet_msi_set_affinity(desc, cpumask_of(ch->cpu)); > + spin_unlock(&desc->lock); > +} > + > static void hpet_attach_channel(unsigned int cpu, > struct hpet_event_channel *ch) > { > @@ -450,7 +460,7 @@ static void hpet_attach_channel(unsigned > if ( ch->cpu != cpu ) > return; > > - hpet_msi_set_affinity(irq_to_desc(ch->irq), cpumask_of(ch->cpu)); > + set_channel_irq_affinity(ch); > } > > static void hpet_detach_channel(unsigned int cpu, > @@ -472,7 +482,7 @@ static void hpet_detach_channel(unsigned > } > > ch->cpu = cpumask_first(ch->cpumask); > - hpet_msi_set_affinity(irq_to_desc(ch->irq), cpumask_of(ch->cpu)); > + set_channel_irq_affinity(ch); > } > > #include > > >