linux-kernel - Re: [PATCH v2] PCI/MSI: Don't touch MSI bits when the PCI device is disconnected

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: <1f7776f2901b40368c847353b74a8cf7@ausx13mps321.AMER.DELL.COM>
Date:   Wed, 14 Nov 2018 00:31:13 +0000
From:   <Alex_Gagniuc@...lteam.com>
To:     <keith.busch@...el.com>
Cc:     <helgaas@...nel.org>, <oohall@...il.com>,
        <gregkh@...uxfoundation.org>, <mr.nuke.me@...il.com>,
        <linux-pci@...r.kernel.org>, <Austin.Bolen@...l.com>,
        <Shyam.Iyer@...l.com>, <linux-kernel@...r.kernel.org>,
        <jonathan.derrick@...el.com>, <lukas@...ner.de>,
        <ruscur@...sell.cc>, <sbobroff@...ux.ibm.com>,
        <linuxppc-dev@...ts.ozlabs.org>
Subject: Re: [PATCH v2] PCI/MSI: Don't touch MSI bits when the PCI device is
 disconnected

On 11/13/2018 04:56 PM, Keith Busch wrote:
> On Tue, Nov 13, 2018 at 10:39:15PM +0000, Alex_Gagniuc@...lteam.com wrote:
>> On 11/12/2018 11:02 PM, Bjorn Helgaas wrote:
>>> The whole issue of firmware-first, the mechanism by which firmware
>>> gets control, the System Error enables in Root Port Root Control
>>> registers, etc., is very murky to me.  Jon has a sort of similar issue
>>> with VMD where he needs to leave System Errors enabled instead of
>>> disabling them as we currently do.
>>
>> Well, OS gets control via _OSC method, and based on that it should
>> touch/not touch the AER bits. The bits that get set/cleared come from
>> _HPX method, and there's a more about the FFS described in ACPI spec. It
>> seems that if platform, wants to enable VMD, it should pass the correct
>> bits via _HPX. I'm curious to know in what new twisted way FFS doesn't
>> work as intended.
> 
> When VMD is enabled, the platform sees a VMD endpoint. It doesn't see
> any of the root ports on that domain, so ACPI can't provide policies for
> them nor AER registers for the platform to consider controlling.

I'm not understanding the interdependency between RP AER settings and 
VMD. My understanding of VMD is quite rudimentary though, so I'll take 
your word for it.

Alex