linux-kernel - Re: [PATCH v4] gpio: add Intel WhiskeyCove GPIO driver

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:	Mon, 11 Jul 2016 17:12:33 -0700
From:	Bin Gao <bin.gao@...ux.intel.com>
To:	Linus Walleij <linus.walleij@...aro.org>
Cc:	Mika Westerberg <mika.westerberg@...ux.intel.com>,
	Mathias Nyman <mathias.nyman@...ux.intel.com>,
	Alexandre Courbot <gnurou@...il.com>,
	"linux-gpio@...r.kernel.org" <linux-gpio@...r.kernel.org>,
	Andy Shevchenko <andy.shevchenko@...il.com>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	Ajay Thomas <ajay.thomas.david.rajamanickam@...el.com>,
	Yegnesh S Iyer <yegnesh.s.iyer@...el.com>,
	Bin Gao <bin.gao@...el.com>
Subject: Re: [PATCH v4] gpio: add Intel WhiskeyCove GPIO driver

On Wed, Jul 06, 2016 at 10:57:19AM +0200, Linus Walleij wrote:
> > +       gpiochip_irqchip_add(&wg->chip, &wcove_irqchip, 0,
> > +                            handle_simple_irq, IRQ_TYPE_NONE);
> 
> Reexamine the use of handle_simple_irq() here. We have two kinds of
> irq hardware: those with one register for ACKing and reading the status
> of an IRQ, and those with two registers for it: one where you ACK the
> IRQ (so it can immediately re-trigger) and one to read the status of
> whether it happened. Sometimes different handling is needed for
> levek and edge IRQs even (c.f. gpio-pl061.c).
> 
> Only the hardware with just one register for both things should use
> handle_simple_irq(). This seems to be the case here but I want you
> to verify.

Yes, our case is handle_simple_irq(), not handle_edge_irq(), handle_level_irq() or
handle_fasteoi_irq(), etc. because there is no ACK mechanism inside the
GPIO controller's interrupt logic - all we need to do is read the status
register to get the status and write-to-clear the status register so that
a new interrupt can be triggered, i.e. there is only one register for both.

> 
> Yours,
> Linus Walleij