Re: [Xc3sprog-users] Programming XCFxxP flash PROMs

[Uwe B. <bo...@el...>]

>>>>> "Joris" == Joris van Rantwijk <jo...@sr...> writes:

    Joris> On Wednesday 07 July 2010 19:46:03 Uwe Bonnes wrote: I noticed
    Joris> that programming big files is quite slow (20 minutes for a 20
    Joris> Mbit image). I'm using a Xilinx DLC10 USB programmer.
    >>  What are Impact programming times?

    Joris> Much faster, about 140 seconds for erasing plus programming.
    Joris> With xc3sprog, programming takes between 320 and 980 seconds,
    Joris> depending on the Linux timer frequency (1000 or 100 Hz).

Could it be that IMPACT polls for erase/write reporting finished while your
approach uses fixed maximum timing?

I poll for these time in progalgspi.cpp.

For read request, also asynchronous read might improve speed/

    Joris> There is a slight change to the XCFP programming loop that I'd
    Joris> like to make (patch in sourceforge tracker). My previous code
    Joris> would actually sleep shorter than the recommended time between
    Joris> commands.

I don't see a patch in the tracker yet

    >> Xc3sprog used the reversed engineered devices handling from
    >> ixo.de. Perhaps impact uses not-yet documented APIs for those big
    >> chunks. Can you perhaps send some log like the Linux usbmon api
    >> delivers?

    Joris> I captured a usbmon log file. It looks like Impact uses only A6
    Joris> commands during programming, just like xc3sprog. However Impact
    Joris> always sends large packets, 9398 bytes.

    Joris> I'm thinking that my programming time may be mostly due the
    Joris> Usleep calls in the algorithm. Short sleeps (10 - 100 us) may be
    Joris> rounded up by the OS scheduler, so this is very inefficient.

    Joris> In fact I think xc3sprog could be a lot smarter about sleeping
    Joris> when a USB programmer is used. If the bit rate of the programmer
    Joris> is known, sleeping could be done much more accurately inside the
    Joris> programmer by inserting NOPs in the bit stream. I think this is
    Joris> how Impact handles it.

With infinite time writeing clever algorithms you can gains infinite amount
of programming time...


-- 
Uwe Bonnes                bo...@el...

Institut fuer Kernphysik  Schlossgartenstrasse 9  64289 Darmstadt
--------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------