> > I do not see ... > > port x3F6 DeviceControl ... > > especially nIEN ... > > Is it required ? ... > Can't we manage with 1f7 ?
Sorry, somehow we're not communicating.
Two ports exist at address x1F7, the x1F7 Command OUT port and the x1F7 Status IN port. The host cannot read the Command port, the host cannot write the Status port, thus the two ports can rudely share the same address.
Similarly, two ports exist at address x3F6, the x3F6 Device Control OUT port and the x3F6 Alternate Status IN port.
Your traces do not show the last write to x3F6 OUT i.e. your traces do not show the whole command.
> > > x50 ... > > > > x20 Read ... > > On xEC identify it gives me 0x50 ...
Consistent observations! My guess remains your system has a parallel thread, such as a BIOS interrupt, fetching the data for you.
> if that was the case > then 0x50 shouldn't be returned
No. The drive cannot electronically distinguish your code from BIOS code running on the same cpu. The drive reports x50 to say the drive has completed the command without error. The drive is speaking truth no matter if one thread or two in the host actually helped the drive finish.
> > x20/21 or xA0/A1 or exactly what - so we can > > see you clear your own INTRQ. That exercise > > you can go thru with x00 Nop, no DRQ > > involved. > > Can you tell me more about this
Eventually I will google, since you have kindly expressed interest: I imagine we can find a list of Dos i/o ports on the web.
At the moment maybe x20/21 or xA0/A1 is all I remember of which Dos i/o ports show if the Primary and Secondary INTRQ are hi, which are lo, which are masked. I remember they aren't especially easy to use, you have to write something meaningful before you can read meaningful bits back, and some other thread running between the write and the read can mess you up.
> Polling x1F7 Status ... > change it to x3f6 ... > how is it going to affect finally ?
Yes in theory this change should have no effect.
If in practice you actually try this experiment and see an effect, then we will have begun to learn something new about your host and/or your drive.
A similarly "should not matter but might inform" experiment would be to try both 0 and 1 in nIEN of x3F6 Device Control.
> Can you tell me more about this , > if not in this forum atleast in a mail ?
On the one hand, me, I greatly prefer speaking in public archive places over person-to-person e-mail. On the other hand, although it's now 2004, rumour tells me still a significant fraction of subscribers here open every individual email sent here (!).
If you'd rather speak offline, you can speak at:
--- unexpectedly no data --- http://ide-byte-counting.blog-city.com/read/476118.htm
t13 folk have lately spoken of devices unexpectedly appearing to ask for no data to be copied ...
I've created this blog entry to serve as a place we can say more without having to disturb every t13 subscriber with every individual comment.
---
You can copy the previous posts from you and me there, if you like. I can copy your posts only if you give me permission or you find me an already public archive of this thread of t13.
Pat LaVarre
