This message is from the T13 list server.
James, If a drive capacity is less than 137Gb, why should it report capacity in words 100-103? IIRC, there were dicussions on the usage. I am lost in the various "shalls" right now. For me I really do not care about the issue, except if the vendor stuffs any capacity in words 100-103, it best do all of 48-bit. The host driver issues results from hosts-controllers whom are badly broken when attached a drive (48bit capable regardless of capacity) and a host driver which is aware of the drive capablities and attempts to setup up init parameters properly. The host-controller will choke and die if issued a 48bit command set, period. Now if the drive's capacity is 28bit or smaller the bios will do just fine, but I want to see a bios recover from issuing a 48bit command to a 48bit capable drive whose capacity <= 28bit and the host-controller does not report command set limitations. Oh the fun and the deadlocks, there are many VERY BAD BAD host-contollers out there in the market. Back on topic, so if the drive does not report capacity in words 100-103 is a generally safe assumption one is to use 28-bit commands. As a secondary check, one gones to peek at the rest of the identify data, and sees the device is 48bit capable. See above for the deadlock joys. Cheers, Andre Hedrick LAD Storage Consulting Group If you want a list of HBA's that choke and die, do it offline. On Wed, 25 Sep 2002 [EMAIL PROTECTED] wrote: > This message is from the T13 list server. > > > > Please clarify what you mean by: > "There are drive vendors who are stuffing the > identify page in the techincal wrong place." > > What, specifically, is incorrect in the IDENTIFY data ? > > Thanx !!! > ----------------------------------------------------------------- > Jim Hatfield > ATA Interface Firmware > Seagate Technology - PSG > e-mail: [EMAIL PROTECTED] > s-mail: 389 Disc Drive; Longmont, CO 80501 USA > voice: 720-684-2120 > fax : 720-684-2711 > ==================================================== > > > > > Andre Hedrick > > <t13@linux-id To: Gary Laatsch ><[EMAIL PROTECTED]> > e.org> cc: "T13 (E-mail)" <[EMAIL PROTECTED]> > > Sent by: Subject: Re: [t13] 48BIT Supported Poll > > owner-forum@t > > 13.org > > No Phone Info > > Available > > > > 09/24/2002 > > 05:53 PM > > > > > > > > > > This message is from the T13 list server. > > > > Gary, > > If you are not checking the DCO identify page, the presences of any HPA > regions, while looking at words 100-103 you are toast. This is kind of > overkill, but what needs to happens is to restrict the usage of words > 100-103 to devices addressing beyond 137GB. There are drive vendors who > are stuffing the identify page in the techincal wrong place. However, > this is expected so one deals with it and moves on. > > SET_MAX_EXT needs to be tested against words 100-103, ideally. > > Comments? > > Cheers, > > Andre Hedrick > LAD Storage Consulting Group > > On Tue, 24 Sep 2002, Gary Laatsch wrote: > > > This message is from the T13 list server. > > > > > > Question to all the drive folks....... > > > > There seems to be a fuzzy area about use of the 48-bit addressing > > supported bit in the IDENTIFY DATA (bit 10 of WORD 83). I guess some > > are setting this bit regardless of the drive capacity and some are only > > setting it if the capacity is over 137GB. I am hearing "rumors" that > > this might be creating some driver issues because of the SET MAX and SET > > MAX EXT commands. > > > > Just wondering how you drive folks are setting this? > > > > Gary Laatsch > > Principal Engineer > > Mass Storage > > Phoenix Technologies Ltd. > > Office: 949-790-2107 > > Fax: 949-790-2001 > > [EMAIL PROTECTED] > > > > "Working code is good code, good code is not always working code...." > > - Craig Llewellyn (my first firmware manager) circa 1978 > > > > > > > > > > >
