[ag-automation] Adding OSADL patch to 2.6.33.7 increases latency and CPU usage?

[Grant Edwards]

On Thu, May 10, 2012 at 07:57:53AM +0200, Robert Schwebel wrote:
> On Wed, May 09, 2012 at 06:27:41PM -0500, Grant Edwards wrote:
> > X is somewhere around 20. The really annoying thing is that the
> > "thing" that requires its interrupt serviced so fast is a blob of IP
> > in an FPGA. The blob asserts an IRQ line, and I have to toggle a bit
> > in a register inside that same blob within 25us (5 of 25 is already
> > allocated to another bit of intervening hardware). Why somebody
> > couldn't spend a few gates and do that in the FPGA itself is beyond
> > me. I can't mention who's responsible for that chumk of IP (sometimes
> > I wonder if that's what NDA's are really for).
> 
> Well, it is an FPGA, so you could put a little state machine around your
> IP core which does not much more than the servicing of the hot path.

Yes, we've discussed that.  The problem with that approach is that the
ISR, weather in hardware or software, has to read a value from a
register, and then do two writes to a different register.  These reads
and writes have to be done using an address/data bus that's also
connected to the ARM9 CPU.  In addition to the state machine, we would
have to add arbitration for the address/data bus.  It could be done,
but it would be a fair bit of work.

-- 
Grant