[PATCH v10 08/10] PM / BLOCK: add PM_QOS_FLAG_NO_POLL to skip events checking
From: Aaron Lu <hidden>
Date: 2012-11-21 11:50:44
Also in:
linux-acpi
Subsystem:
block layer, power management core, the rest · Maintainers:
Jens Axboe, "Rafael J. Wysocki", Linus Torvalds
Some device may enter into a power state when events checking is no longer necessary, this patch introduces the PM_QOS_FLAG_NO_POLL flag and once set, events checking work function in block layer will skip invoke the disk's check_events callback. This will be used by ZPODD, when ODD is powered off, it doesn't make sense to poll for media change events anymore. Signed-off-by: Aaron Lu <redacted> --- block/genhd.c | 8 +++++++- include/linux/pm_qos.h | 1 + 2 files changed, 8 insertions(+), 1 deletion(-)
diff --git a/block/genhd.c b/block/genhd.c
index 6cace66..b0092bb 100644
--- a/block/genhd.c
+++ b/block/genhd.c@@ -18,6 +18,7 @@ #include <linux/mutex.h> #include <linux/idr.h> #include <linux/log2.h> +#include <linux/pm_qos.h> #include "blk.h"
@@ -1595,9 +1596,14 @@ static void disk_events_workfn(struct work_struct *work) unsigned int events; unsigned long intv; int nr_events = 0, i; + enum pm_qos_flags_status stat; /* check events */ - events = disk->fops->check_events(disk, clearing); + stat = dev_pm_qos_flags(disk->driverfs_dev, PM_QOS_FLAG_NO_POLL); + if (stat == PM_QOS_FLAGS_ALL) + events = 0; + else + events = disk->fops->check_events(disk, clearing); /* accumulate pending events and schedule next poll if necessary */ spin_lock_irq(&ev->lock);
diff --git a/include/linux/pm_qos.h b/include/linux/pm_qos.h
index 5a95013..318a422 100644
--- a/include/linux/pm_qos.h
+++ b/include/linux/pm_qos.h@@ -36,6 +36,7 @@ enum pm_qos_flags_status { #define PM_QOS_FLAG_NO_POWER_OFF (1 << 0) #define PM_QOS_FLAG_REMOTE_WAKEUP (1 << 1) +#define PM_QOS_FLAG_NO_POLL (1 << 3) struct pm_qos_request { struct plist_node node;
--
1.7.12.4