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:	Mon, 13 Nov 2006 09:34:47 -0800
From:	"Siddha, Suresh B" <suresh.b.siddha@...el.com>
To:	Ingo Molnar <mingo@...e.hu>
Cc:	"Siddha, Suresh B" <suresh.b.siddha@...el.com>,
	Andi Kleen <ak@...e.de>, Andrew Morton <akpm@...l.org>,
	linux-kernel@...r.kernel.org, ashok.raj@...el.com
Subject: Re: [patch] genapic: optimize & fix APIC mode setup

On Mon, Nov 13, 2006 at 09:43:16AM +0100, Ingo Molnar wrote:
> 
> * Siddha, Suresh B <suresh.b.siddha@...el.com> wrote:
> 
> > There is an issue of using clustered mode along with cpu hotplug. More 
> > details are at the below link.
> > 
> > http://marc.theaimsgroup.com/?l=linux-kernel&m=113261865814107&w=2
> 
> ok, to make sure i understand this right: it is not safe to switch any 
> local APIC in the system into clustered APIC mode on the E850x chipset 
> /at all/, because if one of the CPUs gets an INIT/startup IPI message 
> its local APIC will default to logical flat mode and might confuse the 
> chipset?

"at all" is not quite correct. We are fine as long as all the cpus are up
in clustered APIC mode before the IO-APIC RTE's are programmed.

Once the IO-APIC subsystem is up and running and later if the cpu comes online
then we have a window between the INIT/startup IPI message and the place
where we program the DFR and LDR, with in which the IO-APIC will interpret
the logical mode in RTE as 'logical flat' and will probably result in missing
the interrupt.

thanks,
suresh
> 
> on large systems that have their APIC IDs set up to group CPUs amongst 
> different clusters and hence triggered cluster mode, the chipset does 
> not get confused by this, correct?
> 
> 	Ingo
-
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/

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ