linux-kernel - Re: [PATCH] power: add an API to log wakeup reasons

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-prev] [day] [month] [year] [list]

Date:	Tue, 11 Mar 2014 17:23:57 -0700
From:	Joe Perches <joe@...ches.com>
To:	Ruchi Kandoi <kandoiruchi@...gle.com>
Cc:	"Rafael J. Wysocki" <rjw@...ysocki.net>,
	linux-kernel@...r.kernel.org, linux-pm@...r.kernel.org,
	ghackmann@...gle.com, john.stultz@...aro.org,
	Todd Poynor <toddpoynor@...gle.com>
Subject: Re: [PATCH] power: add an API to log wakeup reasons

On Tue, 2014-03-11 at 17:02 -0700, Ruchi Kandoi wrote:
> This API would be called from the platform specific code,
[]
> This is already in use on some Android devices. We are trying to make
> this a generic API which could be called by other platforms as well,
> standardizing the format in which the info is
> logged in dmesg and the format of the info exported to userspace for
> collecting power management statistics.


logging comments:

> >> diff --git a/kernel/power/wakeup_reason.c b/kernel/power/wakeup_reason.c
[]
> >> @@ -0,0 +1,140 @@

Please add the standard

#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt

before any #include

to prefix this with either "power: " or "wakeup_reason: "
(it depends on how the makefile was written, I don't see it)

> >> +#include <linux/wakeup_reason.h>
> >> +#include <linux/kernel.h>

[]

> >> +void log_wakeup_reason(int irq)
> >> +{
> >> +     struct irq_desc *desc;
> >> +     desc = irq_to_desc(irq);
> >> +     if (desc && desc->action && desc->action->name)
> >> +             printk(KERN_INFO "Resume caused by IRQ %d, %s\n", irq,
> >> +                             desc->action->name);
> >> +     else
> >> +             printk(KERN_INFO "Resume caused by IRQ %d\n", irq);

	if (desc && desc->action && desc->action->name)
		pr_info("Resume caused by IRQ %d, %s\n",
			irq, desc->action->name);
	else
		pr_info("Resume caused by IRQ %d\n", irq);

> >> +     if (irqcount == MAX_WAKEUP_REASON_IRQS) {
> >> +             spin_unlock(&resume_reason_lock);
> >> +             printk(KERN_WARNING "Resume caused by more than %d IRQs\n",
> >> +                             MAX_WAKEUP_REASON_IRQS);

		pr_warn("Resume caused by more than %d IRQs\n",
			MAX_WAKEUP_REASON_IRQS);

> >> +             return;
> >> +     }
> >> +
> >> +     irq_list[irqcount++] = irq;
> >> +     spin_unlock(&resume_reason_lock);
> >> +}

I'd probably write this with:

	if (irqcount >= MAX_REASON_IRQS) {

[]

> >> +int __init wakeup_reason_init(void)
> >> +{
> >> +     int retval;
> >> +     spin_lock_init(&resume_reason_lock);
> >> +     retval = register_pm_notifier(&wakeup_reason_pm_notifier_block);
> >> +     if (retval)
> >> +             printk(KERN_WARNING "[%s] failed to register PM notifier %d\n",
> >> +                             __func__, retval);

Kernel style generally uses function names emitted as "%s: ", __func__
not putting function name in brackets.  I suggest using that style:

		pr_warn("%s: failed to register PM notifier %d\n",
			__func__, retval);

> >> +
> >> +     wakeup_reason = kobject_create_and_add("wakeup_reasons", kernel_kobj);
> >> +     if (!wakeup_reason) {
> >> +             printk(KERN_WARNING "[%s] failed to create a sysfs kobject\n",
> >> +                             __func__);

		pr_warn("%s: failed to create a sysfs kobject\n", __func___);

> >> +             return 1;
> >> +     }
> >> +     retval = sysfs_create_group(wakeup_reason, &attr_group);
> >> +     if (retval) {
> >> +             kobject_put(wakeup_reason);
> >> +             printk(KERN_WARNING "[%s] failed to create a sysfs group %d\n",
> >> +                             __func__, retval);

		pr_warn("%s: failed to create a sysfs group %d\n",
			__func___, retval);


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/