[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <ZB11jq3qnZUpmMrm@gondor.apana.org.au>
Date: Fri, 24 Mar 2023 18:03:58 +0800
From: Herbert Xu <herbert@...dor.apana.org.au>
To: Ard Biesheuvel <ardb@...nel.org>
Cc: Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>,
"David S. Miller" <davem@...emloft.net>,
Nicolas Ferre <nicolas.ferre@...rochip.com>,
Alexandre Belloni <alexandre.belloni@...tlin.com>,
Claudiu Beznea <claudiu.beznea@...rochip.com>,
linux-crypto@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/2] crypto - img-hash: Drop of_match_ptr for ID table
On Thu, Mar 23, 2023 at 09:43:58AM +0100, Ard Biesheuvel wrote:
>
> No. PRP0001 support in ACPI does not depend on OF, so drivers that may
> be bound to such devices should not either.
>
> If you are concerned about the memory used by such tables, you can
> always propose making PRP0001 support configurable in the ACPI core,
> but making individual devices depend on OF for PRP0001 matching seems
> wrong to me.
What I am against is removing of_match_ptr by doing a million
tiny patches. Either we should keep it, in which case the
of_match_table field should be made conditional on OF or perhaps
a new Kconfig option if there is no way to reconcile this with
ACPI, or we decide to get rid of it and you should do one giant
patch to remove of_match_ptr across the kernel tree.
We should not be doing a patch for every single driver that has
of_match_table based on whether it can or cannot be used through
ACPI.
Cheers,
--
Email: Herbert Xu <herbert@...dor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
Powered by blists - more mailing lists