linux-kernel - [PATCH] PM: runtime: add RPM_IDLE_SUSPEND / RPM_IDLE_NO

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Date:   Sat, 30 May 2020 18:33:33 +0200
From:   Heiner Kallweit <hkallweit1@...il.com>
To:     "Rafael J. Wysocki" <rjw@...ysocki.net>,
        Pavel Machek <pavel@....cz>, Len Brown <len.brown@...el.com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Cc:     Linux PM <linux-pm@...r.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: [PATCH] PM: runtime: add RPM_IDLE_SUSPEND / RPM_IDLE_NO_SUSPEND
 constants

runtime_idle() callback implementations have to return a non-zero value
if they don't intend to suspend now. Several drivers use an errno like
-EBUSY for this purpose. This can be problematic because the return
value is propagated up the call chain, from rpm_idle() to
__pm_runtime_idle(), and from there to callers like
pm_runtime_put_sync(). A driver author checking the return value of
e.g. pm_runtime_put_sync() may as usual check for retval < 0 and
bail out.
Therefore a positive value should be returned. To facilitate this
add constants RPM_IDLE_SUSPEND and RPM_IDLE_NO_SUSPEND.

Signed-off-by: Heiner Kallweit <hkallweit1@...il.com>
---
 drivers/base/power/runtime.c |  6 +++++-
 include/linux/pm.h           | 10 ++++++++--
 2 files changed, 13 insertions(+), 3 deletions(-)

diff --git a/drivers/base/power/runtime.c b/drivers/base/power/runtime.c
index 9f62790f6..4f529075e 100644
--- a/drivers/base/power/runtime.c
+++ b/drivers/base/power/runtime.c
@@ -453,7 +453,11 @@ static int rpm_idle(struct device *dev, int rpmflags)
 
  out:
 	trace_rpm_return_int_rcuidle(dev, _THIS_IP_, retval);
-	return retval ? retval : rpm_suspend(dev, rpmflags | RPM_AUTO);
+
+	if (retval == RPM_IDLE_SUSPEND)
+		return rpm_suspend(dev, rpmflags | RPM_AUTO);
+
+	return retval;
 }
 
 /**
diff --git a/include/linux/pm.h b/include/linux/pm.h
index 121c104a4..971ed3d77 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -227,8 +227,9 @@ typedef struct pm_message {
  *
  * @runtime_idle: Device appears to be inactive and it might be put into a
  *	low-power state if all of the necessary conditions are satisfied.
- *	Check these conditions, and return 0 if it's appropriate to let the PM
- *	core queue a suspend request for the device.
+ *	Check these conditions, and return RPM_IDLE_SUSPEND if it's
+ *	appropriate to let the PM core queue a suspend request for the device.
+ *	Return RPM_IDLE_NO_SUSPEND if you don't want to suspend now.
  *
  * Several device power state transitions are externally visible, affecting
  * the state of pending I/O queues and (for drivers that touch hardware)
@@ -523,6 +524,11 @@ enum rpm_request {
 	RPM_REQ_RESUME,
 };
 
+enum rpm_idle {
+	RPM_IDLE_SUSPEND = 0,
+	RPM_IDLE_NO_SUSPEND,
+};
+
 struct wakeup_source;
 struct wake_irq;
 struct pm_domain_data;
-- 
2.26.2