This message is from the T13 list server.

At 12:46 PM -0800 3/11/02, Marushak, Nathan wrote:
>This message is from the T13 list server.
>
>
>Hi all,
>
>I had a few questions I was hoping someone could answer.
>*      Why was the Write Verify command removed from the specification in
>ATA/ATAPI-4?

I don't know specifically - some of the long-term members might 
remember the discussion. Perhaps you'll find it in the minutes of 
some of the meetings of that time.

>*      ATA drives reassign bad blocks internally.

True.

>*      Any ideas for how many sectors are reserved, by the drive, for
>reassignment?

No, this is vendor and device specific.

>*      Any ideas concerning the frequency with which drives must perform a
>reassignment?

Again, it is up to the drive vendor.

>*      Since reassign is done internally, how do ATA drives detect/record
>ECC errors?  Do the drives check on every write?

Vendor.

>*      When the drives reassignment pool has been exhausted, what is the
>reporting mechanism back to the device on the next subsequent bad block?

The behavior would be a error reported from the drive, if it happened 
during the course of the command. If a block was accepted into the 
drive's cache and later the drive could not commit the data to media 
(deferred write error), the drive should lock-up reporting busy 
forever, until power is removed. I have witnessed this on several 
brands of devices, but I cannot verify that all behave this way. They 
should though.

>I greatly appreciate any information you may have.  Thanks for your time.
>
>Regards,
>Nathan Marushak
>
>Senior Software Engineer
>intel* Corporation
>I/O & Bridges Division
>WORK: 480.554.0934
>PAGER: 888.650.4273
>EMAIL PAGE: [EMAIL PROTECTED]


-- 

---------------------
I make stuff go.
---------------------

Larry Barras
Apple Computer Inc.
1 Infinite Loop
MS:  306-2TC
Cupertino, CA  95014
(408) 974-3220

Reply via email to