AMD cpu bug update -- AMD confirms!

ejc eric.j.christeson at gmail.com
Mon Mar 5 11:52:26 PST 2012


Congratulations!
You spent a heck of a lot of time working on that -- nice to see that
the work paid off.

Eric

On Mon, Mar 5, 2012 at 1:27 PM, Matthew Dillon
<dillon at apollo.backplane.com> wrote:
>    AMD has confirmed the cpu bug that I analyzed.  I am going to quote
>    the first paragraph.  They gave me some more info and will be updating
>    their revision guide.  To avoid any confusion (so they can finalize their
>    wording for the entry they will be putting in the revision guide) I'm
>    just including the confirmation here.
>
>    quote>>
>
>    "AMD has taken your example and also analyzed the segmentation fault and
>    the fill_sons_in_loop code. We confirm that you have found an erratum
>    with som e AMD processor families. The specific compiled version of
>    the fill_sons_in_loop code, through a very specific sequence of
>    consecutive back-to-back pops and (near) return instructions, can
>    create a condition where the process or incorrectly updates the
>    stack pointer."
>
>    <<endquote
>
>    We exchanged a few emails to try to come up with a good test case.
>    Owing to the difficulty of reproducing the bug I constructed a
>    fully bootable DFly operating system & test case USB image and
>    verified that the bug was present on my test box using that image.
>    AMD was then able to reproduce the bug using that image on their own
>    machines.  Over the last few months they have been working through
>    the possibilities and today emailed me the confirmation that it was,
>    indeed, a cpu bug.
>
>    I'm pretty stoked... it isn't every day that a guy like me gets to
>    find an honest-to-god hardware bug in a major cpu!
>
>                                        -Matt
>                                        Matthew Dillon
>                                        <dillon at backplane.com>





More information about the Kernel mailing list