ts-7000
[Top] [All Lists]

[ts-7000] Re: TS-7550 XUARTs

To:
Subject: [ts-7000] Re: TS-7550 XUARTs
From: "jesseoff" <>
Date: Thu, 01 Sep 2011 22:01:21 -0000







--- In  mike ingle <> wrote:
>
> Hi Jesse,
> 
> thank you for the detailed response.  I am a fan of user space drivers and 
> big hardware buffers, for the same reasons you outlined, I just wanted to 
> make sure that there weren't any gotcha's present.   While I maybe have your 
> attention, what is the usability of the FPGAs for user reconfiguration on:
> 
> ts 7500   sd on fpga
> ts 7550   xnand no sd on fpga 
> ts 4500   xnand sd on fpga
> 
> i am considering all of the above as possible solutions.  I tend to want to 
> add tools to my toolbox, and would like to settle on a board I can use for 
> future projects as well.  With the lattice fpga can you offer encrypted 
> modules, allowing the user to have their own IP and still have the sd and 
> xuart functionality?  

All those three boards have what we call "open cores" available for download:

ftp://ftp.embeddedarm.com/ts-arm-sbc/ts-7500-linux/sources/ts75xx_opencores_src_diamond_aug052011.tar.gz

The above tarball includes pre-configured Lattice Diamond FPGA project files 
for each of the types of boards listed.  The projects uses a few binary "black 
box" .ngo modules for the SD, NAND, and XUART cores, but everything else has 
full Verilog source code.  These projects compile right out of the box and will 
create a .jed JEDEC file that can be loaded on board startup via the command 
"ts7500ctl --loadfpga foo.jed".  Although it can use the .jed, we recommend 
first converting it to a .vme.gz with the "jed2vme" utility we've written as 
then the fpga reload file is only 90-120kbyte and can fit on the initrd.  The 
"ts7500ctl --loadfpga" command can take a .jed, .vme, or .vme.gz.

Reloading the FPGA on startup requires about 3 extra seconds of boot time, and 
can be done at any time post-boot also.  I've optimized this as far as I can 
and while I could feed the FPGA its new configuration faster, it seems the 
Lattice XP2 FPGA's don't like it.  All of the userspace drivers are "fpga 
reload safe" since "ts7500ctl --loadfpga" will first acquire the SBUS lock 
thereby suspending any in progress  FPGA communication and hold onto it while 
it lobotomizes the FPGA and reprograms it.  After reprogramming, the SBUS lock 
is released and a SIGHUP is sent to all the userspace drivers which instructs 
them to reinitialize hardware state.  If the number of XUART ports have grown 
or shrank, connections may be dropped or new listener sockets created.

It is important to note that the FPGA reload is only a SRAM cell update and it 
does not update the XP2 on chip flash configuration.  Updating the flash FPGA 
program can render a board unbootable, so we only do that here.  During 
development, I've gotten my share of boards here in unbootable/unreprogrammable 
states where I actually had have a technician desolder some pins on the FPGA or 
CPU to bring the board back to life-- its not something anybody on our 
 mailing list would want to walk a customer through.

Also, although we provide these FPGA projects for free download, we have to 
draw the line on answering questions on them or supporting them.   FPGA's are 
just too complicated and we're not wanting to sign up for helping engineering 
departments learn about FPGA's or understand our Verilog on a $89 board.  This 
is not to say we won't help period, just we won't help for free.  There is only 
a few of us here that can help besides the authors.

>Also, in the past I have had issues using the 0.1" dual row headers, where the 
>female connector was damaged by prying the module out at an angle (hard to do 
>any other way),  have you had issues on the ts75xx modules with that?


Yeah -- the .1" male header is not the most high-tech style connector available 
these days.  However, it is easy and cheap to repair or replace.  I've talked 
to a few customers that have desoldered and pulled it off and soldered wires in 
instead.  Its also common to lay out another PCB with .1" holes and solder the 
TS-7550 to another board.  We've always liked through-hole connectors because 
they hold up to a lot more abuse before damaging the PCB -- a lot of the 
surface mount connectors will peel the copper traces from the FR4 substrate 
under stress.  We haven't had a lot of problems with the female side, although 
I do remember experimenting with a lower cost female PC104 connector 5 years 
ago which did not use the same internal mating mechanism and had problems with 
male pins not from the same manufacturer.

//Jesse Off



------------------------------------

Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/ts-7000/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/ts-7000/join
    (Yahoo! ID required)

<*> To change settings via email:
     
    

<*> To unsubscribe from this group, send an email to:
    

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/

<Prev in Thread] Current Thread [Next in Thread>
Admin

Disclaimer: Neither Andrew Taylor nor the University of NSW School of Computer and Engineering take any responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the birding-aus mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU