Re: [PATCH] ALSA: control: Verify put() result when in debug mode

[Cezary Rojewski <cezary.rojewski@intel.com>]

On 2026-01-30 4:13 PM, Mark Brown wrote:
> On Fri, Jan 30, 2026 at 04:00:10PM +0100, Cezary Rojewski wrote:

>> Wouldn't memset(0) as a preparation-step solve the issue? That is, each
>> element - instance of struct snd_ctl_elem_value - has up to 512 bytes of
>> data, regardless of type. Not seeing usefulness of ->info() here. Perhaps
>> I'm missing something.
> 
> A memset() on the original value before reading would get it into a
> known state but wouldn't solve the problem that userpace can pass any
> old junk in the unused parts of the new value it passes into put().

Thank you for pointing this out. Indeed, "proper userspace" is not 
always the case, which my v1 clearly forgot about.

I did few tests with ->info, namely info->type and info->count. Reused 
the existing fill_remaining_elem_value() to be exact, with '0' as a 
'pattern'. Hopefully that method is enough to deem whatever is provided 
to ->put() worthy for the verification.

Kind regards,
Czarek