netdev - Re: [PATCH v3] random: use expired per-cpu timer rather than wq for mixing fast pool

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:   Wed, 28 Sep 2022 14:06:45 +0200
From:   Sebastian Andrzej Siewior <bigeasy@...utronix.de>
To:     "Jason A. Donenfeld" <Jason@...c4.com>
Cc:     netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
        Sherry Yang <sherry.yang@...cle.com>,
        Paul Webb <paul.x.webb@...cle.com>,
        Phillip Goerl <phillip.goerl@...cle.com>,
        Jack Vogel <jack.vogel@...cle.com>,
        Nicky Veitch <nicky.veitch@...cle.com>,
        Colm Harrington <colm.harrington@...cle.com>,
        Ramanan Govindarajan <ramanan.govindarajan@...cle.com>,
        Dominik Brodowski <linux@...inikbrodowski.net>,
        Tejun Heo <tj@...nel.org>,
        Sultan Alsawaf <sultan@...neltoast.com>, stable@...r.kernel.org
Subject: Re: [PATCH v3] random: use expired per-cpu timer rather than wq for
 mixing fast pool

On 2022-09-27 12:42:33 [+0200], Jason A. Donenfeld wrote:
…
> This is an ordinary pattern done all over the kernel. However, Sherry
> noticed a 10% performance regression in qperf TCP over a 40gbps
> InfiniBand card. Quoting her message:
> 
> > MT27500 Family [ConnectX-3] cards:
> > Infiniband device 'mlx4_0' port 1 status:
…

While looking at the mlx4 driver, it looks like they don't use any NAPI
handling in their interrupt handler which _might_ be the case that they
handle more than 1k interrupts a second. I'm still curious to get that
ACKed from Sherry's side.

Jason, from random's point of view: deferring until 1k interrupts + 1sec
delay is not desired due to low entropy, right?

> Rather than incur the scheduling latency from queue_work_on, we can
> instead switch to running on the next timer tick, on the same core. This
> also batches things a bit more -- once per jiffy -- which is okay now
> that mix_interrupt_randomness() can credit multiple bits at once.

Hmmm. Do you see higher contention on input_pool.lock? Just asking
because if more than once CPUs invokes this timer callback aligned, then
they block on the same lock.

Sebastian