[coreboot] x86 rowhammer vulnerability

Timothy Pearson tpearson at raptorengineeringinc.com
Wed Aug 12 15:52:16 CET 2015


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 08/12/2015 09:36 AM, Francis Rowe wrote:
> Hi,
> 
> Further to my previous question about the sinkhole attack, I'm also
> interested in whether this vulnerability affects coreboot systems:
> http://www.rowhammer.com/
> 
> Specifically, in my case, I am interested in the following coreboot
> systems:
> * i945 platforms (Lenovo X60/T60, Macbook2,1)
> * GM45 platforms (Lenovo X200/T400/T500/R400/R500)
> * fam10h AMD platforms (ASUS KFSN4-DRE, ASUS KGPE-D16)
> 
> Could someone shed a light on this?
> 
> Regards,
> Francis Rowe.
> 

This is highly DIMM and controller specific.  Basically you need to test
your own hardware to know one way or the other.

That being said, even though ECC does not offer much theoretical
resistance to this attack, from what I gather servers with ECC have not
yet been found vulnerable.  It is possible that manufacturers simply
chose higher binned chips when assembling the ECC DIMMs, or that some
other process variable is different.

Most likely to be affected would be the GM45 platforms in laptops (DDR3,
and corner cutting is highly likely in the SO-DIMM format).

- -- 
Timothy Pearson
Raptor Engineering
+1 (415) 727-8645
http://www.raptorengineeringinc.com
c
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJVy2urAAoJEK+E3vEXDOFbhMsH/2mgKdgmvxroSqP+VpEqRalB
0GQVRrUzBv0myXb9e5bvnJzjphTXKCeQ/Je+rsNPPM0Xf2jqsKMiHwlQYr4kElb9
+kPavBw+QCdXOE8c2poUXgFF2eI3tMUHptrRidfYCoYl2rwOyT3z4SgBMYUVn821
T/4ALSaEXEdKUovRZ1pMa0g3ACtuNtfsfJa+8dLJrsxJMIc4hegFSHDnZPazpjog
uMiAx4vfjnxdZt/oA/2f/oewwteqEOjDPUiwn1IOUGWyLZSzXhNW01zyAF6erZpk
VHAP3WR8daMuqWHsspuHUeh5IgP3NWcLIxL+/KKBPXl3K1iMnAXoUB+mOzFY6gs=
=7gxb
-----END PGP SIGNATURE-----



More information about the coreboot mailing list