From mboxrd@z Thu Jan 1 00:00:00 1970 From: Zhang Rui Subject: Re: [PATCH V4 13/13] Thermal: Introduce locking for cdev.thermal_instances list. Date: Tue, 21 Aug 2012 08:53:27 +0800 Message-ID: <1345510407.1682.930.camel@rui.sh.intel.com> References: <1343292083-2047-1-git-send-email-rui.zhang@intel.com> <1343292083-2047-14-git-send-email-rui.zhang@intel.com> <20120820154518.GB9833@besouro> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from mga11.intel.com ([192.55.52.93]:2916 "EHLO mga11.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752744Ab2HUAw0 (ORCPT ); Mon, 20 Aug 2012 20:52:26 -0400 In-Reply-To: <20120820154518.GB9833@besouro> Sender: linux-acpi-owner@vger.kernel.org List-Id: linux-acpi@vger.kernel.org To: eduardo.valentin@ti.com Cc: "Rafael J. Wysocki" , Matthew Garrett , Len Brown , R Durgadoss , Amit Kachhap , Wei Ni , linux-acpi@vger.kernel.org, linux-pm@vger.kernel.org On =E4=B8=80, 2012-08-20 at 18:45 +0300, Eduardo Valentin wrote: > Hello Rui, >=20 > On Thu, Jul 26, 2012 at 04:41:23PM +0800, Zhang Rui wrote: > > we need to go over all the thermal_instance list of a cooling devic= e > > to decide which cooling state to put the cooling device to. > >=20 > > But at this time, as a cooling device may be referenced in multiple > > thermal zones, we need to lock the list first in case > > another thermal zone is updating this cooling device. > >=20 > > Signed-off-by: Zhang Rui >=20 > Reviewed-by: Eduardo Valentin >=20 > A minor comment bellow. >=20 > > --- > > drivers/thermal/thermal_sys.c | 8 ++++++++ > > include/linux/thermal.h | 1 + > > 2 files changed, 9 insertions(+) > >=20 > > diff --git a/drivers/thermal/thermal_sys.c b/drivers/thermal/therma= l_sys.c > > index 7f3a891..356a59d 100644 > > --- a/drivers/thermal/thermal_sys.c > > +++ b/drivers/thermal/thermal_sys.c > > @@ -799,6 +799,7 @@ int thermal_zone_bind_cooling_device(struct the= rmal_zone_device *tz, > > goto remove_symbol_link; > > =20 > > mutex_lock(&tz->lock); > > + mutex_lock(&cdev->lock); >=20 > Why do you need to lock while going through the tz thermal_instances? >=20 > > list_for_each_entry(pos, &tz->thermal_instances, tz_node) > > if (pos->tz =3D=3D tz && pos->trip =3D=3D trip && pos->cdev =3D= =3D cdev) { > > result =3D -EEXIST; > > @@ -808,6 +809,7 @@ int thermal_zone_bind_cooling_device(struct the= rmal_zone_device *tz, > > list_add_tail(&dev->tz_node, &tz->thermal_instances); > > list_add_tail(&dev->cdev_node, &cdev->thermal_instances); >=20 > Locking the above operation should be enough, no? >=20 we need to add this thermal_instance to the cooling device thermal_instance list. say, what if another thermal zone that references this cooling device i= s being unregistered at the same time? thanks, rui > > } > > + mutex_unlock(&cdev->lock); > > mutex_unlock(&tz->lock); > > =20 > > if (!result) > > @@ -840,14 +842,17 @@ int thermal_zone_unbind_cooling_device(struct= thermal_zone_device *tz, > > struct thermal_instance *pos, *next; > > =20 > > mutex_lock(&tz->lock); > > + mutex_lock(&cdev->lock); > > list_for_each_entry_safe(pos, next, &tz->thermal_instances, tz_no= de) { > > if (pos->tz =3D=3D tz && pos->trip =3D=3D trip && pos->cdev =3D=3D= cdev) { > > list_del(&pos->tz_node); > > list_del(&pos->cdev_node); > > + mutex_unlock(&cdev->lock); > > mutex_unlock(&tz->lock); > > goto unbind; > > } > > } > > + mutex_unlock(&cdev->lock); > > mutex_unlock(&tz->lock); > > =20 > > return -ENODEV; > > @@ -913,6 +918,7 @@ thermal_cooling_device_register(char *type, voi= d *devdata, > > } > > =20 > > strcpy(cdev->type, type); > > + mutex_init(&cdev->lock); > > INIT_LIST_HEAD(&cdev->thermal_instances); > > cdev->ops =3D ops; > > cdev->updated =3D true; > > @@ -1016,6 +1022,7 @@ static void thermal_cdev_do_update(struct the= rmal_cooling_device *cdev) > > if (cdev->updated) > > return; > > =20 > > + mutex_lock(&cdev->lock); > > /* Make sure cdev enters the deepest cooling state */ > > list_for_each_entry(instance, &cdev->thermal_instances, cdev_node= ) { > > if (instance->target =3D=3D THERMAL_NO_TARGET) > > @@ -1023,6 +1030,7 @@ static void thermal_cdev_do_update(struct the= rmal_cooling_device *cdev) > > if (instance->target > target) > > target =3D instance->target; > > } > > + mutex_unlock(&cdev->lock); > > cdev->ops->set_cur_state(cdev, target); > > cdev->updated =3D true; > > } > > diff --git a/include/linux/thermal.h b/include/linux/thermal.h > > index 06fd04d..d02d06d 100644 > > --- a/include/linux/thermal.h > > +++ b/include/linux/thermal.h > > @@ -95,6 +95,7 @@ struct thermal_cooling_device { > > void *devdata; > > const struct thermal_cooling_device_ops *ops; > > bool updated; /* true if the cooling device does not need update = */ > > + struct mutex lock; /* protect thermal_instances list */ > > struct list_head thermal_instances; > > struct list_head node; > > }; > > --=20 > > 1.7.9.5 > >=20 -- To unsubscribe from this list: send the line "unsubscribe linux-acpi" i= n the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html