Re: [microblaze-uclinux] Boa + SSL hanging after certificate download

[Matthew Rubenstein <email@xxxxxxxxxxxx>]

	Indeed. Not just for the techniques of HW-accelerated crypto that
starts with a running Linux app. More broadly, as examples of porting
Linux app components from SW instructions to parallel HW. Such
techniques are some of the most valuable research derived from the
MB/uCL platform. Learning some best practices and development patterns
from these tasks will amplify the productivity of developers on the
platform, and other "FPGA OS" projects. Please help us learn from your
porting work on this list.


On Tue, 2005-10-25 at 08:46 -0700, Paul Hartke wrote:
> Ivan,
> Sounds like great work.  Can you point us to more info on the project?
> 
> Thanks.
> Paul
> 
> Quoting Ivan <salorankatu@xxxxxxxx>:
> > El lun, 24-10-2005 a las 23:15 -0700, Jonathan Jung escribió:
> > > Good news (sort of),
> > >
> > > I was able to run openssl encryption methods using the openssl app
> > > (RC4-40 worked at least ;).
> > >
> > > Bad news: SSH didn't compile on first run, so I'll have to see why
> > > that's freaking out on me.  (don't have the output at the moment).
> >
> > Hi Jonathan,
> >
> > about your question to execute 'openssl speed' command on my
> > environment, it is not possible because now I am in pre-doctoral stay in
> > Switzerland :(
> >
> > About your problem SSH, I had some problems too, but nothing important.
> >
> > > Still haven't found out why Boa + SSL is hanging, but openssl might
> > > be a good substitute for me.
> > >
> > > BTW, the purpose of getting an encryption/decryption algorithm
> > > working is to move the fun parts in to hardware, and connect that
> > > hardware via FSL links (and use John's driver)!
> >
> > My last work was the same that you proposed. I implemented a hardware
> > accelerated version of SSH using MicroBlaze-uClinux. The connection
> > between MicroBlaze and the cipher cores was made using the FSL
> > interface. I tested only the AES-128 and 3DES, but I have implemented
> > the IDEA, Blowfish, DES, RC4, MD5 and SHA-128/256 algorithms. In this
> > work I measured the performance obtained using the pure-software version
> > and the hardware-based version using the scp application. Also, the
> > different cipher cores are partial reconfigured at run-time, that is,
> > the MicroBlaze downloads from Internet the negotiated cipher core
> > (partial bitstream), configures the FPGA (self-reconfiguration) and uses
> > it. If the next time the algorithm changes, then MicroBlaze can
> > reconfigure the core again with the new cipher :P
> > If you are interested, we can talk about it out of the mailing-list
> > (because I think that it is an off-topic). I can give some suggestions
> > in order to improve my results :)
> >
> > Regards,
> >
> > Ivan
> >
> ___________________________
> 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/
--

(C) Matthew Rubenstein

___________________________
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/