linux-kernel - Re: [v2] HID: corsair-void: Add Corsair Void headset family 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] [thread-next>] [day] [month] [year] [list]

Message-ID: <f0aa2ca0-6256-48e4-8d2a-dfd5da072ad4@web.de>
Date: Wed, 21 Aug 2024 09:26:35 +0200
From: Markus Elfring <Markus.Elfring@....de>
To: Stuart Hayhurst <stuart.a.hayhurst@...il.com>,
 linux-input@...r.kernel.org, kernel-janitors@...r.kernel.org
Cc: LKML <linux-kernel@...r.kernel.org>,
 Benjamin Tissoires <bentiss@...nel.org>, Jiri Kosina <jikos@...nel.org>
Subject: Re: [v2] HID: corsair-void: Add Corsair Void headset family driver

>> This was the case for a while.
>>
>> Increasing applications of scope-based resource management provide
>> further opportunities for smaller scopes according to some local variables,
>> don't they?
>
> Personally I'd rather it just fits in with the rest of the kernel,
> but if the general consensus is that new drivers should use tighter
> scopes, I can do that instead.

There are the usual communication challenges to consider also especially
with collateral evolution in such software areas.


>> How do you think about to collaborate with other data structures
>> than character arrays?
>>
>> See also:
>> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/process/coding-style.rst?h=v6.11-rc4#n953
>
> Hm, I picked a character array since all it's doing is sending a
> buffer to the device.
> There's no published specification to follow, only "Well the Windows
> driver sends these bytes and this happens".
> So there isn't really a structure that really comes naturally,
> especially with all the magic numbers.

I imagine that further development concerns can be adjusted accordingly.


> Unless you're suggesting I just do `unsigned char send_buf[3] = {...}`?

Such a programming approach might also look promising.


> I checked the docs, apparently I misread somewhere that
> `hid_hw_raw_request` couldn't use stack allocated memory safely,
> whoops.

Will safer API usage be clarified further?

Can applications of advanced data structures become more appealing?

Regards,
Markus