linux-kernel - Re: [RFC PATCH V2 3/8] genirq: Add runtime power management support for IRQ chips

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

Date:	Thu, 21 Jan 2016 13:40:58 +0100
From:	Ulf Hansson <ulf.hansson@...aro.org>
To:	Thomas Gleixner <tglx@...utronix.de>
Cc:	Jon Hunter <jonathanh@...dia.com>,
	Jason Cooper <jason@...edaemon.net>,
	Marc Zyngier <marc.zyngier@....com>,
	Jiang Liu <jiang.liu@...ux.intel.com>,
	Stephen Warren <swarren@...dotorg.org>,
	Thierry Reding <thierry.reding@...il.com>,
	Kevin Hilman <khilman@...nel.org>,
	Geert Uytterhoeven <geert@...ux-m68k.org>,
	Grygorii Strashko <grygorii.strashko@...com>,
	Lars-Peter Clausen <lars@...afoo.de>,
	Linus Walleij <linus.walleij@...aro.org>,
	Soren Brinkmann <soren.brinkmann@...inx.com>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"linux-tegra@...r.kernel.org" <linux-tegra@...r.kernel.org>,
	"linux-pm@...r.kernel.org" <linux-pm@...r.kernel.org>,
	"Rafael J. Wysocki" <rjw@...ysocki.net>
Subject: Re: [RFC PATCH V2 3/8] genirq: Add runtime power management support
 for IRQ chips

[...]

>> > I don't think using __free_irq() is the correct place to decrease the
>> > runtime PM usage count. It will keep the irqchip runtime resumed even
>> > if there are no irqs enabled for it.
>> >
>> > Instead I would rather allow the irqchip to be runtime suspended, when
>> > there are no irqs enabled on it.
>
> Which is a no no, as you might lose interrupts that way. We disable interrupts
> lazy, i.e. we do not mask them. So no, you cannot do that from
> enable/disable_irq().

Thanks for the input!

My main point around the approach suggested in $subject patch, is that
I don't think it's *enough* fine grained.

>From a runtime PM perspective, we should allow the irqchip to enter a
low power state when there are no IRQs enabled for it.

As enable|disable_irq() isn't the place to do runtime PM reference
counting from, perhaps there is another option? Maybe the irqchip
driver itself is better suited to take these decisions?

[...]

Kind regards
Uffe