linux-kernel - Re: [PATCH 1/3] Input: matrix_keypad - avoid repeatedly converting GPIO to IRQ

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]

Message-ID: <ZbqV_EqBqppL5pNF@google.com>
Date: Wed, 31 Jan 2024 10:48:28 -0800
From: Dmitry Torokhov <dmitry.torokhov@...il.com>
To: Bartosz Golaszewski <bartosz.golaszewski@...aro.org>
Cc: linux-input@...r.kernel.org, Dan Carpenter <dan.carpenter@...aro.org>,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/3] Input: matrix_keypad - avoid repeatedly converting
 GPIO to IRQ

On Mon, Jan 22, 2024 at 09:19:56AM +0100, Bartosz Golaszewski wrote:
> On Sun, 21 Jan 2024 at 06:32, Dmitry Torokhov <dmitry.torokhov@...il.com> wrote:
> >
> > gpio_to_irq() is getting more expensive and may require sleeping.
> 
> This is no longer true, the relevant commit was reverted and I have
> now a working series where I'm using SRCU for lockless list traversal
> so if anything it will actually become cheaper.
> 
> > Convert row GPIOs to interrupt numbers once in probe() and use
> > this information when the driver needs to enable or disable given
> > interrupt line.
> 
> That being said, I'm not against this change, though this may not be
> very useful after all.

OK, then I'll tone down the patch description and apply.

Thanks.

-- 
Dmitry