5.1, Data Corruption, Intel, Oh my! [patch]

[Bosko Milekic]

  All of the Email I have received from the many people testing this
  patch (THANKS!!!) has been positive so far with respect to the
  data corruption problem.

  There remains one outstanding issue to be resolved prior to committing
  this and due to the nature of the problem, it's been taking a little
  longer for me to fix it.  The issue has to do with running the patch
  in combination with APM (I am told that the APM problem actually
  existed in the past as well but was somehow only apparent when running
  with DISABLE_PSE).  I *think* that the problem has to do with our bios
  code switching to the page 0 stack which somehow leads to problems
  when this patch is applied (or when stock -current is used but with
  DISABLE_PSE).  Because I don't have any machine that use the APM code,
  debugging the problem consists of me changing something, then providing
  a patch to someone with APM, and then waiting for the results.  This
  is a time-consuming process, but as soon as the problem is addressed,
  I will commit the changes.

  Thanks to everyone who has helped test this so far.

Regards,
-- 
Bosko Milekic  *  bmilekic at technokratis.com  *  bmilekic at FreeBSD.org
TECHNOkRATIS Consulting Services  *  http://www.technokratis.com/