linux-kernel - Re: [PATCH RFC v1] random: use static branch for crng

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] [thread-next>] [day] [month] [year] [list]

Date:   Thu, 12 May 2022 12:35:51 +0800
From:   Herbert Xu <herbert@...dor.apana.org.au>
To:     "Jason A. Donenfeld" <Jason@...c4.com>
Cc:     linux-kernel@...r.kernel.org, linux-crypto@...r.kernel.org,
        tytso@....edu, linux@...inikbrodowski.net, rostedt@...dmis.org,
        ardb@...nel.org
Subject: Re: [PATCH RFC v1] random: use static branch for crng_ready()

Jason A. Donenfeld <Jason@...c4.com> wrote:
> On Tue, May 03, 2022 at 03:40:52PM +0200, Jason A. Donenfeld wrote:
>> +static bool crng_ready_slowpath(void)
>> +{
>> +     if (crng_init <= 1)
>> +             return false;
>> +     if (in_atomic() || irqs_disabled() || cmpxchg(&crng_init, 2, 3) != 2)
>> +             return true;
> 
> Nobody chimed in here, but for posterity I thought I should point out
> that this approach actually won't work, since in_atomic() doesn't work
> with CONFIG_PREEMPT_COUNT=n kernels.
> 
> So back to the drawing board in trying to figure out the best way to do
> this...

Well the standard solution to code paths that require sleeping is
to use a work queue.  So any reason why you can't just schedule
a work to do the static_branch_enable?

Cheers,
-- 
Email: Herbert Xu <herbert@...dor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt