linux-kernel - Re: 5.13-rt1 + KVM = WARNING: at fs/eventfd.c:74 eventfd

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>] [day] [month] [year] [list]

Message-ID: <8759e324-418b-493a-adee-236738cc3a4f@redhat.com>
Date:   Fri, 16 Jul 2021 09:59:15 +0200
From:   Paolo Bonzini <pbonzini@...hat.com>
To:     Hillf Danton <hdanton@...a.com>
Cc:     Jason Wang <jasowang@...hat.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Sebastian Andrzej Siewior <bigeasy@...utronix.de>,
        "Michael S. Tsirkin" <mst@...hat.com>,
        Juri Lelli <jlelli@...hat.com>,
        LKML <linux-kernel@...r.kernel.org>,
        Al Viro <viro@...iv.linux.org.uk>,
        He Zhe <zhe.he@...driver.com>
Subject: Re: 5.13-rt1 + KVM = WARNING: at fs/eventfd.c:74 eventfd_signal()

On 16/07/21 09:55, Hillf Danton wrote:
> On Fri, 16 Jul 2021 08:54:58 +0200 Paolo Bonzini wrote:
>> On 16/07/21 04:06, Hillf Danton wrote:
>>>> With the patch:
>>>> 	- no warn
>>>> 	- continue using the VM normally...
>>> Well with the patch applied, the VM works fine without the stuff protected
>>> by the spin_lock_irqsave(), then without the patch why simply printing a
>>> warning makes the VM dumb, given the warning is there actually also
>>> preventing you from touching the lock.
>>
>> If the warning is triggered, eventfd_signal will not do the wakeup.
> 
> [I am assuming we are not talking about the deadlock in the comment.]
> 
> No preemption occured without the warning printed.
> Why will the wakeup behavior change without peemption?

Sorry, I don't follow.  What I'm saying is that without the patch:

* the warning only occurs if preemption occurs during the 
spin_lock_irqsave critical section (and therefore it can only occur in 
PREEMPT_RT kernels)

* the warning causes an early return 0 that messes up the VM's networking

Paolo