> the REAL gain will be from clever
> compression.  Even more of a gain will come from not blindly
> downloading the whole trace.

If we can't make the raw transfer fast enough, then yes, look at
compression.  And yes, look at having the display program figure out
which section of data it needs for the initial display and upload
that section first.  Then while the developer is staring at the
screen, the display program can be uploading additional data in
case it is needed.  Similar to demand paged virtual memory.

> > Can we safely assume that the board under test will not screw up the
> > PCI bus until we talk to the board?
> 
> No.  Why does it matter?

Run test.
Board under test screws up PCI bus.
Press reset button.  (OGD RAM is protected from this, somehow)
Test machine boots up.
Transfer OGD data through test machine's PCI bus, and then send it to
the support machine via Ethernet.

If you don't mind hot-unplugging the board under test, you could do
it this way even if the board under test does screw up the PCI bus
before you talk to it.  Or if the OGD RAM data survives a complete
power off of the test machine, you can remove the board under test
without hot-unplugging it.

> Maybe what we need to do is physically cut all power traces from PCI
> to the rest of the PCB so that the board can be powered ONLY through
> external means.

Perhaps a short PCI extender card and cut the power traces on that?

> I'm not an analog engineer, but I THINK that as long
> as the ground is common, we'll be able to measure signals properly.

Normal test equipment does not share power with the device under test.
Voltmeters, oscilloscopes, logic analyzers, etc.  You can even do a
differential measurement with two 'scope probes and no ground connection
to the device under test.  I don't see why this would be any different
just because it is a PCI card.

> Doing it this way, you could indeed physically remove the tracer form
> the test machine and still retain a trace in memory.  But with the
> power traces cut, I'm not sure if we still have the option of
> injecting signals or using PCI to download a trace.

You should still be able to inject signals or use PCI to transfer a trace.
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)

Reply via email to