linux-kernel - Re: x86_64 and AMD with C1E

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:	Sat, 15 Dec 2007 12:06:15 +0100 (CET)
From:	Thomas Gleixner <tglx@...utronix.de>
To:	Chuck Ebbert <cebbert@...hat.com>
cc:	Mikhail Kshevetskiy <mikhail.kshevetskiy@...il.com>,
	linux-kernel@...r.kernel.org
Subject: Re: x86_64 and AMD with C1E

On Wed, 24 Oct 2007, Chuck Ebbert wrote:

> On 10/24/2007 05:26 PM, Mikhail Kshevetskiy wrote:
> >>>
> >>> I fill something wrong here.
> >>> Is it possible to reduce the amount of timer interrupts?
> >>> Is it possible to force enable C1,C2 and C3 states when c1e disabled?
> >>>
> >> How are you disabling C1E?
> >>
> >>
> > dirty hack, i just follow the FreeBSD way and clear C1e bit in lapic
> > initialization code. I make it for test purpose only, so i do not produce a
> > patch.
> > 
> 
> Why does disabling C1E disable C1, C2 and C3?
> 
> Thomas, in the case of the machines where C1E is disabled on CPU 0 but
> enabled on CPU 1, could we just disable it? Maybe it's a BIOS bug and the
> vendor just forgot to disable CPU 1...

This whole C1E business is a lot of magic trickery in the guts of the
BIOS. I have an X2 box here with an extrem interesting behaviour. C1E
is not advertised at all, but the box freezes with the same symptoms
unless I disable the local apic timer on the kernel command line. I
have no clue how this works internally.

     tglx
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/