linux-kernel - Re: [PATCH v3 17/20] ASoC: dt-bindings: tas2770: add flags for SDOUT pulldown and zero-fill

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: <20250305132239.GA1415729-robh@kernel.org>
Date: Wed, 5 Mar 2025 07:22:39 -0600
From: Rob Herring <robh@...nel.org>
To: James Calligeros <jcalligeros99@...il.com>
Cc: Liam Girdwood <lgirdwood@...il.com>, Mark Brown <broonie@...nel.org>,
	Jaroslav Kysela <perex@...ex.cz>, Takashi Iwai <tiwai@...e.com>,
	Shenghao Ding <shenghao-ding@...com>, Kevin Lu <kevin-lu@...com>,
	Baojun Xu <baojun.xu@...com>, Dan Murphy <dmurphy@...com>,
	Krzysztof Kozlowski <krzk+dt@...nel.org>,
	Conor Dooley <conor+dt@...nel.org>,
	Shi Fu <shifu0704@...ndersoft.com>,
	Jean Delvare <jdelvare@...e.com>,
	Guenter Roeck <linux@...ck-us.net>,
	Alyssa Rosenzweig <alyssa@...enzweig.io>,
	Martin Povišer <povik+lin@...ebit.org>,
	Hector Martin <marcan@...can.st>, linux-sound@...r.kernel.org,
	linux-kernel@...r.kernel.org, devicetree@...r.kernel.org,
	asahi@...ts.linux.dev, linux-hwmon@...r.kernel.org
Subject: Re: [PATCH v3 17/20] ASoC: dt-bindings: tas2770: add flags for SDOUT
 pulldown and zero-fill

On Wed, Mar 05, 2025 at 01:19:15AM +0000, James Calligeros wrote:
> On Tue, Mar 4, 2025 at 1:50 PM Rob Herring <robh@...nel.org> wrote:
> > Can't you align this with the other property you added? Or extend the
> > existing TDM properties we have.
> 
> I don't think either option makes sense given the functionality. This chip
> behaves differently to TAS2764, and instead of using a bitmask to determine
> which slots to ignore, we only get a single bit to tell the chip whether we want
> it to fill or pull down *all* inactive slots. The property being a u32 mask
> therefore does not make sense here.

If there's a single bit control, then that just means there's only 1 
valid value for a mask in that case.

Or maybe a mask is overkill. What's the usecase for fill or pulldown 
*some* inactive slots?

> Building the logic off the existing generic TDM slot properties would alter
> behaviour of existing implementations where zero-fill and pulldown may not be
> required or even wanted. This may continue to be the case going forward so I'd
> rather make it an explicit opt-in rather than some unconditional thing we try to
> turn on heuristically.

Existing implementations would not have the new/extra properties and 
would continue to operate as before. Or those drivers could simply 
ignore the properties.

> I gave some thought to flipping these bits if a TDM slot mask is passed to the
> driver, however it can still be the case that we don't want both zero-fill *and*
> pulldown active at the same time, or as above some implementations may want
> neither, so we still need to be able to specify them individually.

This just feels like something common because any TDM interface may need 
to control this. It's not really a property of the chip, but requirement 
of the TDM interface.

Rob