linux-kernel - Re: [PATCH] crypto: essiv - Replace memcpy() + NUL-termination with strscpy()

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: <99BDF560-307D-4F94-BC14-44B58ED579CC@linux.dev>
Date: Sun, 23 Mar 2025 14:08:38 +0100
From: Thorsten Blum <thorsten.blum@...ux.dev>
To: David Laight <david.laight.linux@...il.com>
Cc: Herbert Xu <herbert@...dor.apana.org.au>,
 "David S. Miller" <davem@...emloft.net>,
 linux-crypto@...r.kernel.org,
 linux-kernel@...r.kernel.org
Subject: Re: [PATCH] crypto: essiv - Replace memcpy() + NUL-termination with
 strscpy()

Hi David,

On 23. Mar 2025, at 11:20, David Laight wrote:
> On Sun, 16 Mar 2025 22:15:04 +0100 Thorsten Blum wrote:
> 
>> Use strscpy() to copy the NUL-terminated string 'p' to the destination
>> buffer instead of using memcpy() followed by a manual NUL-termination.
>> 
>> No functional changes intended.
>> 
>> Signed-off-by: Thorsten Blum <thorsten.blum@...ux.dev>
>> ---
> 
> The 'len' argument to strscpy() is supposed to be the length of the
> buffer (in order to avoid overflow) not the number of characters.

Not necessarily, see linux/string.h:

/**
 * strscpy - Copy a C-string into a sized buffer
 * ...
 * The size argument @... is only required when @dst is not an array, or
 * when the copy needs to be smaller than sizeof(@dst).
 * ...
 */

> In this case the bound check is before the copy (and the buffer assumed
> to be the right size!)
> So memcpy() + terminate is exactly correct.

Yes, this is simply a refactoring, there's nothing wrong with memcpy()
followed by a manual NUL-termination.

However, strscpy() is shorter and semantically better imo because we're
copying C-strings and not just raw bytes. strscpy() also has additional
compile-time checks regarding C-strings that memcpy() doesn't.

Thanks,
Thorsten