---------------------<snip>----------------------
<>So there are zero defects now in the nucleus? I doubt it. A disabled
page
fault is easy to handle - just ABEND the current piece of work and
move right
along. What happens next depends on what the current piece of work was.
Even if the nucleus refuses to page fault, nevertheless disabled
nucleus code
can be interrupted and, if the interrupt status is not preserved, the
nucleus
resumes execution with interrupts now enabled, and I posit that the
results
of this situation would likely be far worse than if the nucleus
suffered a
disabled page-fault.
---------------------<unsnip>------------------------
IIRC, the parts of the nucleus that run disabled are in page-fixed
storage, starting at IPL time,
and are NEVER allowed to page. And the amount of disabled code seems to
be VERY
small and getting smaller.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
