[<prev] [next>] [day] [month] [year] [list]
Date: Tue, 6 Jun 2017 21:18:15 +0200
From: "Jason A. Donenfeld" <Jason@...c4.com>
To: David Howells <dhowells@...hat.com>
Cc: keyrings@...r.kernel.org, Eric Biggers <ebiggers3@...il.com>,
Herbert Xu <herbert@...dor.apana.org.au>,
Kirill Marinushkin <k.marinushkin@...il.com>,
security@...nel.org, stable@...r.kernel.org,
"Theodore Ts'o" <tytso@....edu>,
kernel-hardening@...ts.openwall.com,
LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] security/keys: rewrite all of big_key crypto
On Tue, Jun 6, 2017 at 8:45 PM, David Howells <dhowells@...hat.com> wrote:
> Jason A. Donenfeld <Jason@...c4.com> wrote:
>
>> Thank you! Perfect instructions, I'll play around with this and make
>> sure it works.
>
> Don't forget that it's bimodal. You need to give it sufficient data to
> trigger storage in swappable space.
Somewhat incredibly, it works perfectly.
First I tried the instructions you noted, and things worked, both for
big files and small ones.
Then I modified the source to print to dmesg the data buffer and the
key before and after the encryption/decryption function. I verified
with a small python script that indeed standard aes-gcm is being used
successfully.
Thus, pending Ted's approval of the new random API, this patch should
be ready for merging.
Regards,
Jason
Powered by blists - more mailing lists