Re: [microblaze-uclinux] Jtag Debug

[Paul Hartke <phartke@xxxxxxxxxxxx>]

Sounds like everything is working as it should then.

Check to make sure the caches are enabled and also note that the Microblaze
FPU is single-precision only.

Sounds like you are in the middle of the normal
figuring-out-why-the-app-is-running-so-slow-and-doing-something-about-it
loop....

:-)

Paul

Quoting Abot Botbot <dejanigma@xxxxxxxxxxx>:
> Paul,
>
>   Yea the design was compiled in EDK using BSB. I didn't want to get into
> specifics right away, but I can give you a little perspective. The
> algorithm
> I'm using computes a floating point FFT on a 2x64-point array (real,
> imag).
> This was originally running on a TI DSP chip, but we want to reduce the
> form
> factor and eliminate the a lot of the other hardware with one FPGA. This
> operation is run as emulated floating point math on the 150 MHz DSP, it
> does
> several thousand FFTs per second. On the SP3 and the ML401 designs
> running
> the same code (at 50 Mhz and 100 MHz respectively), the same algorithm
> results in about one FFT per second, on both platforms. Granted we're
> comparing apples and oranges, but the difference in speed is phenominal,
> several orders of magnitude. Now that you have a little perspective,
> maybe
> you have a better idea where the problem is. The past month has been
> spent
> just trying to get around these massive performance hinderances. Adding
> the
> FPU didn't give any noticable increase in speed. The designs are running
> out
> of external memory (SDRAM) with cachelink over FSL to the microblaze.
> Granted it takes dozens of clock cycles for every read and write, or at
> least in the worst cases, but that still doesn't add up to the
> performance I
> see. Some people have asked why we are using the mircoblaze for DSP at
> all,
> but I don't see any architecture restrictions to doing emulated floating
> point math at a decent speed. Since the clock speeds are different and
> the
> memory access times are longer, I would immedaitely expect to have maybe
> 10%
> of the speed of the 150 MHz DSP, but i get .001% or worse... If you need
> any
> further info I can go on all day :D.. Any insight you can give would be
> invaluable. In the final implementation I expect to have dedicated
> hardware
> for doing the FFTs, but the microblaze should still be able to do it
> decently, right?
>
> Thanks,
> Andy
>
>
> >From: Paul Hartke <phartke@xxxxxxxxxxxx>
> >To: microblaze-uclinux@xxxxxxxxxxxxxx, Abot Botbot
> <dejanigma@xxxxxxxxxxx>
> >CC: microblaze-uclinux@xxxxxxxxxxxxxx
> >Subject: Re: [microblaze-uclinux] Jtag Debug
> >Date: Wed, 26 Oct 2005 07:53:35 -0700
> >
> >How do you know it runs "slow"--that's really a value judgement?  What
> >makes
> >you think it should run faster--or even slower for that matter?  The
> design
> >appears to be created by Base System Builder which lots of folks have
> used
> >on ml401.
> >
> >Paul
> >
> >Quoting Abot Botbot <dejanigma@xxxxxxxxxxx>:
> > > Hey Friends,
> > >
> > >    This might be considered off topic, but I think I'm missing a
> > > fundamantal
> > > explanation for the performance that I get. Here's my basic design
> flow,
> > > which doesn't use uCLinux yet, I was just trying to verify that the
> DSP
> > > functionality I want is possible on the microblaze... I download the
> XMD
> > > stub to the board, then run "dow executable.elf" from XMD and type
> > > "run"...
> > > This sets my code in motion, but it seems to be going extremely slow.
> > > It's
> > > running entirely out of SRAM, but that shouldn't be the only factor
> in
> > > the
> > > performance I'm seeing... I'm wondering if my whole system is running
> > > over
> > > the JTAG clock and putting out debug symbols to the PC, even though
> I'm
> > > not
> > > observing them, and I took the -g option out of the compiler flags..
> So
> > > two
> > > questions... Am I running in some debug mode where it is running much
> > > slower
> > > than it could be... How do I get the hardware to run independantly,
> if
> >it
> > > is
> > > in some debug mode.. After I finish a few things for my graduate
> classes
> > > (mid terms, blech!) I will get back on the uCLinux train and get that
> > > running on my ML401 board. Thanks for any help you can give! I'll
> >attatch
> > > some files in case anyone thinks looking at them will help them to
> > > understand.
> > >
> > > Thanks,
> > > Andy
> > >
___________________________
microblaze-uclinux mailing list
microblaze-uclinux@xxxxxxxxxxxxxx
Project Home Page : http://www.itee.uq.edu.au/~jwilliams/mblaze-uclinux
Mailing List Archive : http://www.itee.uq.edu.au/~listarch/microblaze-uclinux/