Re: [Qemu-devel] [PATCH 1/5] Add target memory mapping API

[Avi Kivity]

Anthony Liguori wrote:
> Scenario 2 (callbacks)
0) Register minimum required bounce memory at init time
> 1) Register callback and begin data mapping
> 2) Succeed in mapping first part of packet and 1 page of bounced buffer
> 3) Bounce buffer is exhausted, callback will not be invoked until the 
> next unmap
> 4) Deadlock
>
> You could add some sort of call to determine if the bounce memory is 
> exhausted and try to take some corrective action in #2 but it gets ugly.
>
> So as I said earlier, I think the callback model makes sense when you 
> have streaming data (where there are no packet boundaries).  I think 
> it's pretty difficult to avoid deadlocking though with callbacks in a 
> packet model (where boundaries must be preserved).

If you specify the boundaries (an additional input array, or maybe a 
function that says whether a particular point is a "good enough" 
boundary) then I think it can be workable.

--
error compiling committee.c: too many arguments to function