Re: [RFC PATCH] crypto: arc4: Implement a version optimized for memory usage

From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
To: Eric Biggers <ebiggers@kernel.org>
Cc: herbert@gondor.apana.org.au, davem@davemloft.net,
	linux-crypto@vger.kernel.org, linux-kernel@vger.kernel.org,
	kernel-janitors@vger.kernel.org
Subject: Re: [RFC PATCH] crypto: arc4: Implement a version optimized for memory usage
Date: Tue, 4 May 2021 19:59:38 +0200	[thread overview]
Message-ID: <d523902e-744c-1291-aee8-9be734f2a3ce@wanadoo.fr> (raw)
In-Reply-To: <YJF8/oaWUqZsWfOb@gmail.com>

Le 04/05/2021 à 18:57, Eric Biggers a écrit :
> On Sun, May 02, 2021 at 09:29:46PM +0200, Christophe JAILLET wrote:
>> +#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS)
>> +#define S_type	u8
>> +#else
>> +#define S_type	u32
>> +#endif
>> +
>>   struct arc4_ctx {
>> -	u32 S[256];
>> +	S_type S[256];
>>   	u32 x, y;
>>   };
> 
> Is it actually useful to keep both versions?  It seems we could just use the u8
> version everywhere.  Note that there aren't actually any unaligned memory
> accesses, so choosing the version conditionally on
> CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS seems odd.  What are you trying to
> determine by checking that?

Hi, this is a bad interpretation from me.

I thought that S[1] would likely use an odd address and would trigger an 
unaligned access. But as we would read only 1 byte, this is not the case.

Looking at [1], we have : "At this point, it should be clear that 
accessing a single byte (u8 or char) will never cause an unaligned 
access, because all memory addresses are evenly divisible by one."

I wanted to avoid potential performance cost related to using char (i.e 
u8) instead of int (i.e. u32).
On some architecture this could require some shift or masking or 
whatever to "unpack" the values of S.

[1]: 
https://www.kernel.org/doc/html/latest/core-api/unaligned-memory-access.html

CJ

> 
> - Eric
> 