linux-kernel - Re: [PATCH 3/5] x86/mcheck/AMD: Reduce number of blocks scanned per bank

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: <56982A6E.10000@amd.com>
Date:	Thu, 14 Jan 2016 17:08:30 -0600
From:	Aravind Gopalakrishnan <aravind.gopalakrishnan@....com>
To:	Borislav Petkov <bp@...en8.de>
CC:	<tony.luck@...el.com>, <tglx@...utronix.de>, <mingo@...hat.com>,
	<hpa@...or.com>, <x86@...nel.org>, <linux-edac@...r.kernel.org>,
	<linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 3/5] x86/mcheck/AMD: Reduce number of blocks scanned per
 bank

On 1/14/2016 4:53 PM, Borislav Petkov wrote:
> On Thu, Jan 14, 2016 at 04:48:22PM -0600, Aravind Gopalakrishnan wrote:
>> True. But that BlkPtr logic also will undergo changes as it's interpretation
>> for future processors is different.
> But there still must be a bit there which says "this register is valid",
> like MCi_MISC[63].

There is a bit to say if it's valid or not.

> And so I'd very much prefer checking a bit (or bits) instead of relying
> on defines.
>
>

But we'd still need to know the last available MISC register for a bank 
to know when to end the loop right?

Currently we loop over all the possible blocks-
  for (block = 0; block < NR_BLOCKS; ++block) {
  <code>...

  increment block address;

  <code>..
  }

Here, we know we have to stop at block number 8 as that is the last MISC 
register that is present for a bank.
In the same manner, we'd still have to know the last possible MISC 
register for future processors..

Thanks,
-Aravind.