READ LONG Conundrum or Command
Ralph Weber -- VMS -- ZKO3-4/U14
weber at star.enet.dec.com
Mon Oct 3 09:44:59 PDT 1994
I represent a company that has a product that is totally dependent on READ
LONG (and WRITE LONG). The product makes us lots of money. Everyday,
thousands of customers world-wide execute READ LONG/WRITE LONG commands,
as they use the OpenVMS Host-Based Shadowing product.
Needless to say, making the READ LONG command obsolete is something that
Digital must vote against. If READ LONG were to be made obsolete, I would
have to bring forward alternative proposals that will really make you wretch.
On the other hand, decoupling READ LONG from the READ-WRITE Error Recovery
mode page is just fine with me. I really like the fact that I don't have to
cross check Gene's list of bits where the READ LONG and READ-WRITE Error
Recovery mode page interfere with each other.
The only down side I can see in decoupling is a possible inability to do
something useful where the right combination of READ LONG and READ-WRITE Error
Recovery bits currently will work. I would propose adding more function bits
to the READ LONG CDB to handle these cases. I can see at least 8 convenient
new function bits that could be defined in the READ LONG CDB.
More information about the T10