linux-kernel - Re: Linux 6.5 speed regression, boot VERY slow with anything systemd related

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: <ZO4RzCr/Ugwi70bZ@google.com>
Date:   Tue, 29 Aug 2023 08:42:04 -0700
From:   Sean Christopherson <seanjc@...gle.com>
To:     Marc Haber <mh+linux-kernel@...schlus.de>
Cc:     Bagas Sanjaya <bagasdotme@...il.com>, linux-kernel@...r.kernel.org,
        Linux Regressions <regressions@...ts.linux.dev>,
        Linux KVM <kvm@...r.kernel.org>,
        Paolo Bonzini <pbonzini@...hat.com>,
        Tony Lindgren <tony@...mide.com>
Subject: Re: Linux 6.5 speed regression, boot VERY slow with anything systemd related

On Tue, Aug 29, 2023, Marc Haber wrote:
> On Tue, Aug 29, 2023 at 07:53:45AM -0700, Sean Christopherson wrote:
> > What is different between the bad host(s) and the good host(s)?  E.g. kernel, QEMU,
> 
> The bad host is an APU ("AMD GX-412TC SOC") with 4 GB of RAM, one of the
> good hosts is a "Xeon(R) CPU E3-1246 v3" with 32 GB of RAM.

I don't expect it to help, but can you try booting the bad host with
"spec_rstack_overflow=off"?

> system configuration is from the same ansible playbook, but of course there
> are differences.

Can you capture the QEMU command lines for the good and bad hosts?  KVM doesn't
get directly involved in serial port emulation; if the blamed commit in 6.5 is
triggering unexpected behavior then QEMU is a better starting point than KVM.