Yes, the old wish_block stuff was kind of incompatible with the
new way of doing things, and if we are going to do something along these
lines it would need to be redone.
The first question is whether there is still a need for such
functionality?
I have to admit I don't quite remember why it was added in the
first place. Wasn't it to deal with buggy DMA chipsets that screwed up
when there was more than one bus mastering ISA DMA card active at the
same time? I realize that there was a specific problem you were trying
to solve, but I was never wild about having mid-level scsi having to be
aware that some DMA chips were faulty.
If we end up having to re-implement this functionality, I am
thinking that we probably want a separate version of the queue handler
function (perhaps a wrapper for scsi_request_fn, perhaps not). The
functionality would probably be similar to what we would need for the
single_lun case, which I am still not completely satisfied with.
-Eric
"The world was a library, and its books were the stones, leaves,
brooks, grass, and the birds of the earth. We learned to do what only
a student of nature ever learns, and that was to feel beauty."
Chief Luther Standing Bear - Teton Sioux
On Sun, 2 Jan 2000, Dario_Ballabio wrote:
> %UATTACH
> Hi Eric. I noticed that you removed the "wish_block" flag and almost
> all the associated blocking list functionality from the new
> scsi implementation. The requirement of having a single scsi host
> active at a time is still there on my system with multiple
> isa boards.
> Do you plan to reintroduce an equivalemt functionality on th new
> implementation?. If not I have to build an equivalent infrastructure
> inside the low level drivers, but this looks backward to me.
> Cheers,
>
> -db
>
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to [EMAIL PROTECTED]
