linux-kernel - Re: [PATCH] x86/intel_rdt: reinitialize cbm for new group allocation

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: <alpine.DEB.2.20.1701092244110.3534@nanos>
Date:   Mon, 9 Jan 2017 22:58:13 +0100 (CET)
From:   Thomas Gleixner <tglx@...utronix.de>
To:     Shaohua Li <shli@...com>
cc:     LKML <linux-kernel@...r.kernel.org>,
        Fenghua Yu <fenghua.yu@...el.com>,
        Peter Zijlstra <peterz@...radead.org>
Subject: Re: [PATCH] x86/intel_rdt: reinitialize cbm for new group
 allocation

On Fri, 6 Jan 2017, Shaohua Li wrote:

> rdtgroup_rmdir deletes a group and frees the closid, but doesn't
> reiniaialize domain's cbm to default for the closid. Next time the
> closid is allocated again, 'schemata' will show old cbm.

Rightfully so. Because that's the value which is in the configuration
MSR. That's perfectly fine because nothing can use that closid and therefor
the configuration MSR is completely irrelevant until the closid is reused.

> We can do the reinitialization at free/alloc closid, but sounds doing it
> at alloc is more nartual.

What you are doing is not natural, it's just wrong.

Assume the admin wants to keep the 'max_cbm' value for a domain and
therefor does not update the schemata file, but the MSR still contains the
old value. Not what you really want, right?

If you really want to have the CBM values reset to max_cbm then this needs
to be done when the directory is removed and this needs to be written to
the hardware as well on all affected packages. Not sure if that's worth the
effort.

Thanks,

	tglx