lkml.org 
[lkml]   [2021]   [Apr]   [8]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH v5 1/1] use crc32 instead of md5 for hibernation e820 integrity check
On Thu, Apr 08, 2021 at 09:15:06AM -0400, Chris von Recklinghausen wrote:
> Suspend 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. This integrity check is used only to verify accidental
> corruption of the hybernation data and is not intended as a
> cryptographic integrity check.
> Md5 is overkill in this case and also disabled in FIPS mode because it
> is known to be broken for cryptographic purposes.
>
> Fixes: 62a03defeabd ("PM / hibernate: Verify the consistent of e820 memory map
> by md5 digest")
>
> Tested-by: Dexuan Cui <decui@microsoft.com>
> Reviewed-by: Dexuan Cui <decui@microsoft.com>
> Signed-off-by: Chris von Recklinghausen <crecklin@redhat.com>
> ---
> v1 -> v2
> bump up RESTORE_MAGIC
> v2 -> v3
> move embelishment from cover letter to commit comments (no code change)
> v3 -> v4
> add note to comments that md5 isn't used for encryption here.
> v4 -> v5
> reword comment per Simo's suggestion
>
> arch/x86/power/hibernate.c | 35 +++++++++++++++++++----------------
> 1 file changed, 19 insertions(+), 16 deletions(-)
>
> diff --git a/arch/x86/power/hibernate.c b/arch/x86/power/hibernate.c
> index cd3914fc9f3d..b56172553275 100644
> --- a/arch/x86/power/hibernate.c
> +++ b/arch/x86/power/hibernate.c
> @@ -55,31 +55,31 @@ int pfn_is_nosave(unsigned long pfn)
> }
>
>
> -#define MD5_DIGEST_SIZE 16
> +#define CRC32_DIGEST_SIZE 16
>
> struct restore_data_record {
> unsigned long jump_address;
> unsigned long jump_address_phys;
> unsigned long cr3;
> unsigned long magic;
> - u8 e820_digest[MD5_DIGEST_SIZE];
> + u8 e820_digest[CRC32_DIGEST_SIZE];
> };
>
> -#if IS_BUILTIN(CONFIG_CRYPTO_MD5)
> +#if IS_BUILTIN(CONFIG_CRYPTO_CRC32)

Should CONFIG_CRYPTO_CRC32 be getting selected from somewhere?

If that is too hard because it would pull in too much of the crypto API, maybe
using the library interface to CRC-32 (lib/crc32.c) would be a better fit?

- Eric

\
 
 \ /
  Last update: 2021-04-08 17:30    [W:0.090 / U:0.620 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site