linux-kernel - Re: [regression] Bug 216989 - since 6.1 systems with AMD Ryzen stutter when fTPM is enabled

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]

Date:   Tue, 7 Feb 2023 19:13:46 -0300
From:   "Jason A. Donenfeld" <Jason@...c4.com>
To:     Linux regressions mailing list <regressions@...ts.linux.dev>
Cc:     Dominik Brodowski <linux@...inikbrodowski.net>,
        Herbert Xu <herbert@...dor.apana.org.au>,
        linux-integrity@...r.kernel.org,
        LKML <linux-kernel@...r.kernel.org>, reach622@...lcuk.com,
        1138267643@...com, Jarkko Sakkinen <jarkko@...nel.org>,
        Peter Huewe <peterhuewe@....de>, Jason Gunthorpe <jgg@...pe.ca>
Subject: Re: [regression] Bug 216989 - since 6.1 systems with AMD Ryzen
 stutter when fTPM is enabled

On Sun, Feb 05, 2023 at 06:38:16PM +0100, Linux kernel regression tracking (Thorsten Leemhuis) wrote:
> [ccing Dominik (who authored the culprit) and Herbert (who merged it)]
> 
> On 02.02.23 11:28, Linux kernel regression tracking (Thorsten Leemhuis)
> wrote:
> > 
> > I noticed a regression report in bugzilla.kernel.org. As many (most?)
> > kernel developer don't keep an eye on it, I decided to forward it by
> > mail. Quoting from https://bugzilla.kernel.org/show_bug.cgi?id=216989 :
> 
> Turns out according to a bisection from one of the reporters that
> b006c439d58d ("hwrng: core - start hwrng kthread also for untrusted
> sources") (merged for 6.1) apparently makes this hardware issue occur
> quicker/more frequently a lot quicker on any board that didn't get the
> firmware update yet. So it could be argued that from the point of the
> kernel it *might* be considered a regression.

Finally replying without HTML, now that I'm at my laptop.

This isn't a bug with the commit you mentioned. Rather, this is a bug in
the TPM hardware and/or in the TPM driver. Probably the TPM driver
should quirk around the faulty BIOS to disable whatever functionality is
broken, or it should notice these latency spikes and warn, or something.
But I'll leave that to James; that's his wheelhouse.

Jason