linux-kernel - Re: Latency spikes on V6.15.1 Preempt RT and maybe related to intel? IGB

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]

Message-ID: <20250617101507.jU00mupE@linutronix.de>
Date: Tue, 17 Jun 2025 12:15:07 +0200
From: Sebastian Andrzej Siewior <bigeasy@...utronix.de>
To: Marc Strämke <marc.straemke@...ropuls.de>
Cc: linux-kernel@...r.kernel.org, linux-rt-users@...r.kernel.org
Subject: Re: Latency spikes on V6.15.1 Preempt RT and maybe related to intel?
 IGB

On 2025-06-17 12:03:31 [+0200], Marc Strämke wrote:
> Hi Sebastian,
Hi,

> On 17.06.25 12:00, Sebastian Andrzej Siewior wrote:
> > Even if CPU1 would handle CPU0's timers then it would wake cyclictest on
> > CPU0 but that thread would have to wake until CPU0 is done with the PCI
> > bus. CPU1 knows nothing about it.
> 
> Okay then the latency I see on the other CPU must be from a PCI access done
> by the second CPU which stall on the same shared bus.

Okay.
> 
> Anyway: Thanks for your help Sebastian! I can probably live well with these
> spikes in latency. I was more concerned that there is a deeper issue with my
> config and the response time could be unbounded.

You don't have to live with it. You could add a read after the writes in
the loop (wrfl()). This should help.
Some Intel CPUs have a MSR bit to disable this kind of caching on the
PCI bus. Maybe the AMD CPU has this, too. Or there might be a switch in
the BIOS.

> Marc

Sebastian