This message is from the T13 list server.
John Sherman wrote:
This message is from the T13 list server.
I am performing a WRITE_SECTORS command to write 0x10 sectors to a HDD.
After each sector of data (512 bytes), Intrq is asserted, the status register
is read, and the next sector begins transferring.
After the last sector has been written to the HDD, Intrq is NOT assertted but
my understanding of the spec is that it should be. Am I missing something in
the spec that makes it optional for a device to assert Intrq at the end of a
PIO data-out transfer? Or should I consider this HDD out of spec?
[note, your word wrap is broken]
What Hale said is certainly true, but I also wonder if you are having a
problem with your own device driver. Is there any chance that your OS
driver reads the Status register before it should, thereby clearing the
last interrupt condition?
I've often seen mistakes in early-stage software, where the Status
register was read with no care taken to its side effects (interrupt
clearing).
Jeff
