linux-kernel - Re: [PATCH] PCI: hv: Move completion variable from stack to heap in hv_compose_msi

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: <20210601231339.GA1391@anparri>
Date:   Wed, 2 Jun 2021 01:13:39 +0200
From:   Andrea Parri <parri.andrea@...il.com>
To:     Long Li <longli@...rosoft.com>
Cc:     Michael Kelley <mikelley@...rosoft.com>,
        "longli@...uxonhyperv.com" <longli@...uxonhyperv.com>,
        KY Srinivasan <kys@...rosoft.com>,
        Haiyang Zhang <haiyangz@...rosoft.com>,
        Stephen Hemminger <sthemmin@...rosoft.com>,
        Wei Liu <wei.liu@...nel.org>,
        Lorenzo Pieralisi <lorenzo.pieralisi@....com>,
        Rob Herring <robh@...nel.org>,
        Bjorn Helgaas <bhelgaas@...gle.com>,
        "linux-hyperv@...r.kernel.org" <linux-hyperv@...r.kernel.org>,
        "linux-pci@...r.kernel.org" <linux-pci@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        Andrea Parri <Andrea.Parri@...rosoft.com>
Subject: Re: [PATCH] PCI: hv: Move completion variable from stack to heap in
 hv_compose_msi_msg()

> I agree if the intent is to deal with a untrusted host, I can follow the same principle to add this support to all requests to VSP. But this is a different problem to what this patch intends to address. I can see they may share the same design principle and common code. My question on a untrusted host is: If a host is untrusted and is misbehaving on purpose, what's the point of keep the VM running and not crashing the PCI driver?

I think the principle can be summarized with "keep the VM _running, if you can
handle the misbehaviour (possibly, warning on "something wrong/unexpected just
happened"); crash, otherwise".

Of course, this is just a principle: the exact meaning of that 'handle' should
be leverage case by case (which I admittedly haven't here); I'm thinking, e.g.,
at corresponding complexity/performance impacts and risks of 'mis-assessments'.

Thanks,
  Andrea