[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <YHidc5Xdjo5LK0D5@gmail.com>
Date: Thu, 15 Apr 2021 13:09:23 -0700
From: Eric Biggers <ebiggers@...nel.org>
To: Chris von Recklinghausen <crecklin@...hat.com>
Cc: ardb@...nel.org, simo@...hat.com, rafael@...nel.org,
decui@...rosoft.com, linux-pm@...r.kernel.org,
linux-crypto@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v8 1/1] use crc32 instead of md5 for hibernation e820
integrity check
On Thu, Apr 15, 2021 at 03:46:46PM -0400, Chris von Recklinghausen wrote:
> Hibernation fails on a system in fips mode because md5 is used for the e820
> integrity check and is not available. Use crc32 instead.
>
> This patch changes the integrity check algorithm from md5 to crc32.
The second paragraph is redundant with the first.
> /**
> - * get_e820_md5 - calculate md5 according to given e820 table
> + * compute_e820_crc32 - calculate md5 according to given e820 table
> *
> * @table: the e820 table to be calculated
> - * @buf: the md5 result to be stored to
> */
This comment still mentions MD5.
Also, this isn't a well-formed kerneldoc comment, since it doesn't document the
return value.
Also, this function is calculating the checksum *of* the table, not calculating
a checksum "according to" it (whatever that means).
Something like this would be good, I think:
/**
* compute_e820_crc32 - compute the CRC-32 of the given e820 table
*
* @table: the e820 table to be checksummed
*
* Return: the resulting checksum
*/
Also, please try 'git grep -i md5 arch/x86/kernel/'. There is still another
reference to MD5 that should be updated, in arch/x86/kernel/e820.c.
- Eric
Powered by blists - more mailing lists