linux-kernel - Re: [PATCH] KVM: x86: Ensure liveliness of nested VM-Enter fail tracepoint message

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: <20210609130153.42b3ca81@oasis.local.home>
Date:   Wed, 9 Jun 2021 13:01:53 -0400
From:   Steven Rostedt <rostedt@...dmis.org>
To:     Sean Christopherson <seanjc@...gle.com>
Cc:     Wanpeng Li <kernellwp@...il.com>,
        Paolo Bonzini <pbonzini@...hat.com>,
        Vitaly Kuznetsov <vkuznets@...hat.com>,
        Wanpeng Li <wanpengli@...cent.com>,
        Jim Mattson <jmattson@...gle.com>,
        Joerg Roedel <joro@...tes.org>, kvm <kvm@...r.kernel.org>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] KVM: x86: Ensure liveliness of nested VM-Enter fail
 tracepoint message

On Wed, 9 Jun 2021 09:12:40 -0700
Sean Christopherson <seanjc@...gle.com> wrote:

> > I can observe tons of other kvm tracepoints warning like this after
> > commit 9a6944fee68e25 (tracing: Add a verifier to check string
> > pointers for trace events), just echo 1 >
> > /sys/kernel/tracing/events/kvm/enable and boot a linux guest.  
> 
> Can you provide your .config?  With all of events/kvm and events/kvmmmu enabled
> I don't get any warnings running a Linux guest, a nested Linux guest, and
> kvm-unit-tests.
> 
> Do you see the behavior with other tracepoints?  E.g. enabling all events on my
> systems yields warnings for a USB module, but everything else is clean.

Right, I tested this with running KVM guests when it was added. I was
actually surprised with Sean's report saying it was a kvm trace event,
until I noticed that the problem event is only triggered on error paths.

-- Steve