linux-kernel - Re: [PATCH v2] x86/PCI: Handle PIRQ routing tables with no router device given

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

Date:   Fri, 9 Jul 2021 01:35:47 +0200 (CEST)
From:   "Maciej W. Rozycki" <macro@...am.me.uk>
To:     Nikolai Zhubr <zhubr.2@...il.com>
cc:     Bjorn Helgaas <bhelgaas@...gle.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
        "H. Peter Anvin" <hpa@...or.com>, Arnd Bergmann <arnd@...nel.org>,
        x86@...nel.org, linux-pci@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2] x86/PCI: Handle PIRQ routing tables with no router
 device given

On Fri, 9 Jul 2021, Nikolai Zhubr wrote:

> >   Have you tried contacting Nvidia about your ALI chipset?  Back in the day
> > I tried to avoid undocumented stuff and Intel was reasonably open about
> > most of their chipsets.
> 
> Well, being neither their customer nor a kernel developer, I'm not sure my
> request would be considered serious. Anyway, probably I'll give it a try a bit
> later when I have an opportunity to dismount this board for more comfortable
> testing.

 It never hurts asking.  At worst you'll be ignored, and at the second 
worst they'll say nay.  They may have lost it too.

> I was also going to try to modify its BIOS to remove some unwanted
> behaviour unrelated to IRQs, and it might happen that I also discover
> something about PCI handling along with (It is just 64k size, after all, and I
> have a 8086 debugger)

 Umm, the board may be old enough not to play any tricks with the BIOS, 
but mind that at reset x86 starts in the flat mode from 0xfffffff0 and the 
contents of ROM there may not be what you see at 0xf000:0xfff0 later on.

 I once worked on a project where I had an opportunity to access the BIOS 
at the reset vector (and poke at CPU registers, run, stop, single-step it, 
place hardware breakpoints, etc.) using GDB over JTAG with an Intel Atom 
board.  It was an interesting experience, but sadly most x86 hardware does 
not have the capability let alone a JTAG connector (called XDP or eXtended 
Debug Port in Intel-speak) to attach a probe to.

 You may try disassembling the PCI BIOS 2.1 service however.

  Maciej