An Open Source Hardware Development Tool

LuxuryYacht writes "The PLAICE is an open source hardware and software project developing a powerful in-circuit development tool that combines in one device the features of a FLASH Programmer, Memory Emulator, and High Speed Multi-Channel Logic Analyzer. It runs uClinux. The logic analyzer features up to 200MHz sampling rates and up to 32 input channels. The logic analyzer Java client supports up to 200MHz sampling rates, user-controlled filtering operations, time line in diagrams, transfer rates, and user configurable drawing modes. The Java client supports access via almost any PC with a serial port and uses the RXTX serial library with support for 34 platforms including Linux, Windows, and Solaris. Java client plugins include an SPI and I2C bus protocol analyzer, conversion of timing analysis to state analysis, and post-processing functions."

Comments

Pretty sparse Web site

[Wesley Felter]

An open-source logic analyzer for $150 sounds nice, but the site is seriously lacking in screen shots.

That's nice

[tftp]

But logic analyzers are history. If you want to debug logic today you use ChipScope. That's not just because it is easier, but because breakout connectors (Mictor etc.) are expensive, large, and they disrupt the timing of the circuit.

As memory emulator this device may be useful sometimes, but many MCUs today come with internal RAM, and those that don't - they expect DDR2 speeds, and you can't emulate that.

This can be a full-featured Microblaze development system, though, with tons of samples. I think that's its best value. MicroBlaze was always poorly supported by Linux, as opposed to Nios (which Altera itself supports.) If we have, finally, a working [uc] Linux port to MB that alone is a great achievement. When I looked a year or two ago there was only one, non-functioning, port to a hardware that did not exist.

Re:That's nice

[evanbd]

History? Since when? I've been doing a bit of PIC microcontroller programming lately, and for debugging the various communications between a PIC and various peripherals this sounds perfect. I'll be seriously investigating getting one of these. Mostly the timing of such circuits isn't the issue, it's seeing the data actually on the channel without adding lots of debug code I don't have time or space for. This sounds like a great tool for the serious hobbyist.

Re:That's nice

[tftp]

Since when? Well, for some designs (like yours) they are still useful, and I am glad that there is still someone out there who knows which end of the soldering iron to hold onto.

But in an industrial setting they are quickly replaced by JTAG-connected tools; ChipScope in particular (if you are a Xilinx slave) is great because it captures the signals into the local, very fast RAM, and then sends you the snapshot over a slower JTAG connection. The snapshot is true to what is really happening, and if you design a DDR controller (or faster) then just forget the external wires, they are useless at those speeds. And most of modern commercial designs push the devices to the limit. That's what makes standalone logic analyzers less appealing to a mass manufacturer. Logic analyzers in such conditions become tools of last resort, just like ICEs, where you have to spend a day just preparing your board for testing.

Myself, if I do not have an FPGA in between (and so ChipScope is not an option) then I just use an oscilloscope. I have a 4-channel, inexpensive Infiniium model, and 3 probes is the most I ever needed; staring at the schematic does the rest :-)

Re:That's nice

[LuxuryYacht]

The logic analyzer design is targeted at debugging logic outside of the FPGA and the board itself. ChipScope is supported by the board for debug of logic inside the onboard Spartan-3E FPGA.

The memory emulator is currently targeted at FLASH devices.

Re:That's nice

[networkBoy]

You do realize that tektronix doesn't use mictor any more. They use compression probes that land on micro pads on your system board and are perfectly happy at DDR and DDR2 speeds.
LAs are alive and well. Your tools only work once the board is up and running, till then you still need to see raw I/O and such.
-nB

Re:That's nice

[Alioth]

There are plenty of embedded microprocessors around that do NOT expect DDR2 speeds. Many digital circuits simply don't need a 2GHz space heater and run just fine with a 2MHz Z80 or equivalent (indeed, the Z80 is still manufactured and popular in its 'classic' 40 pin DIL form. I have one on my work table that was manufactured less than 6 months ago). Many many products use chips of this sort of class.

Just because there's a clamour for ever faster (and hotter) chips in PCs and servers, it does not follow that the same is true of an embedded computer. If a 4MHz processor works for a particular application, there is absolutely no benefit in using something that 'expects DDR2'. Normal 70ns static RAM and flash chips are sold by the millions because they are cheap, electronically simple to interface, and low speed circuits are much cheaper and easier to lay out on a PCB. You don't need DDR2 on a weather station embedded computer or washing machine.

Many microcontrollers like the Atmega can interface with external memory (even though they have some internal flash and RAM).

Re:That's nice

[c_oflynn]

I got a Logicport (500 MHz, 32 channel logic analyzer for $350) some time ago, and never looked back. I use it for FPGA designs where I want to capture a lot more data.

I always bring out some spare pins to a header, that becomes my debug port. Then you can route any internal signals to this header. Provided you're not trying to debug anything insanely high-speed, it works great.

It also supports the SPI/I2C/Serial decoding like this project. Only downside is the software is Windows only, and it uses USB so not sure how well it would work on Wine. I've got VMWare to run Windows though so don't mind.

  -Colin

Re:That's nice

[wpiman]

I cannot say a bad word about chipscope, it is a great tool- but I must say that Mictor connectors do not disrupt the timing of the circuit if implemented correctly. Mictor means matched impedance connector. And often times, you need to look outside the device. Chipscope and its Altera equivalent are for internal signals, and the amount of data you can view depends on how much memory you have left in your device. An external logic analyzer has infinite more samples.

Re:That's nice

[acarguy]

Well, logic analyzers aren't dead just yet. I work for Agilent (used to be HP), and we still make quite a few of 'em. Although the JTAG-based tools for internal FPGA logic analysis like Chip Scope are very useful and relatively inexpensive, they present some limitations (like consuming the internal block RAM of the chip) that traditional LAs don't have. Like anything, there are tradeoffs. Agilent has some tools that customize our logic analyzers to make internal FPGA measurements more easily and bridge the gap between the internal tools and the traditional LA. For example, we have a tool for MicroBlaze debug that lets you easily probe the internal MB signals and bring them out to an Agilent LA for trace, inverse assembly and correlation with the original source code. There is a tool called FPGA Dynamic Probe that works well for 'generic' logic debug as well. Traditional LAs can be very useful for tracking down the complex interactions between parts of designs implemented in FPGAs, discrete processors, memory systems and peripherals.

This could eventually be important

[Anonymous Coward]

If Microsoft gets what it wants, it will be hard to get hardware that runs Linux. Well that's Microsoft's dream anyway. In order to protect precious DRM Microsoft has ordained that only 'bullet proof' hardware will be allowed to run in HD mode with Vista. Their spec even says that unencrypted signals must run only on inner layers of pc boards.

Being able to create Linux friendly hardware could, if Microsoft succeeds, be necessary if we are to have high performance video and audio.

This project is not alone as open source hardware. My current favorite is the Arduino board using an Atmel microcontroller. www.arduino.cc I am also playing with the Make controller that uses an Arm. www.makezine.com/controller

Re:RS-232?

[tftp]

I see that the board has Ethernet transceiver installed, and the connector. However the SoftTEMAC IP from Xilinx is not free, and because of that you can't use Ethernet. Virtex-4 (and 5) FPGAs have HardTEMAC which is not just free, it is a hard core in the FPGA, so it is ready to use, and it can do Gigabit Ethernet as well. Because of that I may question the wisdom of picking a S3 platform that is some $ cheaper than V4 but requires a $5,000 IP to do something really useful (Ethernet connectivity is not too much to ask for these days.) Or, alternatively, write your own [T]EMAC module, it's not impossible but you need to be a decent FPGA coder to even get started.

Re:RS-232?

[plasmoidia]

If it doesn't have to run all the time, you can use the Xilinx soft-core EMAC (10/100) for free. The catch is that it times out after about 8-9 hours, at which time it stops functioning. It is limiting, but it will get the job done if you cannot afford to buy the full core.

Not open source hardware...

[plasmoidia]

The hardware is not open source. Actually, the hardware is a Spartan-3E Starter Kit board. Nothing special there. What will be open source is the *firmware* (as well as the software running on top). Semantics aside, this should be an interesting project. This seems to be an attempt to build an entire system in an FPGA with open source firmware/software. As others have expressed, I am not sure how useful it will be as a logic analyzer, but perhaps this could be a start for more open source firmware projects.

Re:RS-232?

[jcgf]

Yeah, the only problem is those never work ;)

Re:RS-232?

[fractoid]

It depends a lot on what adapter you use. I too used FTDI-based USB-to-RS232 adapters at my last job and some of them (can't remember brands, sorry, it was a while back) used to wedge themselves after a couple of hours continuous use. Fine for copying stuff off a device but not good for automation. Then again, we did find one brand that was solid even after a weekend of hammering data back and forth. The FTDI chips seem solid, I'd tend towards blaming the low-budget boards they get used in.

Re:Huh...

[Square Snow Man]

That you are going to be used to make this software populair, so they can start making money of it in the future (think MySQL).

Re:Huh...

[stewwy]

Well I can think of several uses in the HD=dvd and Bluray area :), and its nice that the code is available

Based on a project at sump.org?

[kent.dickey]

The listed site links to it itself--http://www.sump.org/projects/analyzer/ is where it is getting the logic analyzer hardware design from. It's not clear to me what flash-plaice.wikispaces.com is adding, other than porting the design to the Spartan-3E proto board (from a Spartan-3 proto board).

A commercial LA system has carefully designed probes to reduce the load on the signals being probed. I made a home-made PCI data capture "card" by soldering stubs to a blank PCI connector, and connecting directly to an HP analyzer I bought on ebay. I needed the probes to not be a large load on the PCI bus, and using the HP probes solved that problem for me.

If you want cheap logic analyzers, you can get a very nice HP analyzer on ebay for about $500--and have it capture at least 96 bits wide at 100MHz, and have a usable interface. You can upgrade to capturing a million states at 135MHz for a few hundred more. And you can add in a correlated scope for another few hundred bucks. Although this is a bit beyond what folks would want to pay, it's pretty reasonable for hardware that used to cost $10,000+ and is less than a new decked out PC.