This message is from the T13 list server.
FWIW, I'd assume read-ahead means read-ahead (seq) and implies that the read cache is enabled. Agree that a separate bit is needed for read cache on/off if one doesn't exist. Cache could be on with no read ahead for random access. -- David F. TeraByte Unlimited http://www.terabyteunlimited.com > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On > Behalf Of Mark Overby > Sent: Wednesday, June 29, 2005 9:09 AM > To: Hale Landis; [email protected] > Subject: RE: [t13] Read Look Ahead > > This message is from the T13 list server. > > > That's interesting Hale, because as a host vendor that's NOT > what I expect to have happen. > > If the reality is that some very large % of the drives > implement it this way, then I'd have no problem changing it. > But, if it's the other way around then I'd have a problem > changing it because it doesn't reflect reality. > > The other way out of this going forward is to have a seperate > bit indicating if the read cache is completely disabled by > this set features or not. > > ::>-----Original Message----- > ::>From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On > ::>Behalf Of Hale Landis > ::>Sent: Tuesday, June 28, 2005 10:36 PM > ::>To: [email protected] > ::>Subject: RE: [t13] Read Look Ahead > ::> > ::>This message is from the T13 list server. > ::> > ::> > ::>On Tue, 28 Jun 2005 22:15:51 -0700, Mark Overby wrote: > ::>>This message is from the T13 list server. > ::>>I'm not sure we have consensus on this and I'm not sure > that was the ::>>original intent of that bit. > ::> > ::>Actually we are talking about the name of the SET FEATURES > ::>subcommands that has generally been called Enable/Disable > Read ::>Ahead. Probably for the last 10 years the function > actually performed ::>is to enable/disable Read Cache. > ::> > ::>>It would appear (based on the discussions at the plenary > and side ::>>discussions I've had with people) that some > drives ::>implement this as a ::>>read cache enable/disable > others implement it differently. > ::> > ::>I would agree there are many different implementations of > Read (and ::>Write Cache) but I think most people expect the > SET FEATURE ::>subcommands in question to enable/disable the > Read (or Write) cache ::>feature/function in the drive no > matter what the actual cache ::>implementation is. > ::> > ::>Hale > ::> > ::> > ::>*** Hale Landis *** www.ata-atapi.com *** ::> ::> ::> ::>
