Johnson Lam has posted a new DRIVERS.ZIP dated 24-Feb-2012 on
his website at <>.

In it, UIDE and UIDE2 are corrected so that they again handle
"64K UltraDMA boundaries" properly.   A user I-O buffer which
crosses-over a 64K address boundary will cause the drivers to
move output data to, or input data from, their own XMS buffer
which is aligned with NO boundary and will handle actual I-O.

Like I wrote before on this forum, note that I "suspect" this
problem may affect only "old" year-2000 and earlier chipsets,
since the previous UIDE/UIDE2 have not given any "big number"
of data errors for users!   But I want the UIDE drivers to be
compatible with ALL UltraDMA controllers, so they are fixed!

Also, UIDE/UIDEJR are a bit smaller and UIDE2 is MUCH smaller
due to my "experiments" in writing a 7K-byte version of UIDE2
for my "backup/restore" diskettes.   My run-time driver logic
is "good", but my driver initialization logic is often "quick
and dirty" since it disappears after the driver loads!   That
has now been remedied, very much so in UIDE2!

A minor error -- I "date stamped" XMGR/UIDEJR in the files on
Johnson's website to 2-24-2011, not 2012!    A "typing error"
by me, which Johnson shall correct.   XMGR/UIDEJR do show the
year "2012" in their title messages, and they do work fine!

Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
Freedos-user mailing list

Reply via email to