Skip to main content
Topic: Test Release 2.03 and 2.04 - SPI mode. (Read 37438 times) previous topic - next topic

Re: Test Release 2.03 and 2.04 - SPI mode.

Reply #30
@robots, yes that fixes it.

The attached works for me. Well, read (-r) and write (-w) work, verify (-v) doesn't, but I upgraded to FW 0.3 without any difficulty.

Re: Test Release 2.03 and 2.04 - SPI mode.

Reply #31
i am using a vpn
i am actually from serbia
it's no problem, just sucks that i have to do extra work
i will make a program for my ft2232 test board to do the pic programming

Re: Test Release 2.03 and 2.04 - SPI mode.

Reply #32
OK heres the lowdown with my faulty board sans bootloader

had to look for some header pins to solder to the board,( always in the last place to find them)
I uploaded the bootloader using pickit2, went smoothly no problems

I tried firmware 2 with with the latest bitstream following the procedures, everything went well
trying the client, as others found an exception error was raised, since its sitting in the same filepath I'm surprised, I expect there is a hardcoded attribute in the jar thats throwing things off. Trying with th eoriginal client still generated 'device not found'. A bit confused by this????

I then tried firmware3 with the latest bitstream, this worked with the original client.

I then followed the same procedure to update the working board. This too was successful. i.e. running firmware 3, the device is found time after time.

In both cases I have not tried to connect the probes to a signal


/Mac

Re: Test Release 2.03 and 2.04 - SPI mode.

Reply #33
I have no bootloader, maybe overwritten by testing.
I made some tests and I have still the problem with the wrong PID.

Here my test:

Erased the PIC.
Don't know about the programs/datas inside the the PROM and the FPGA.
Program the PIC with ICD2 with the bootloader from the 1.03 Packet.

There's no Serial Device because of no fw.

Jumpered PGC/PGD

HID Device with PID FC90 is there.

Update fw with OLS-program.bat from 2.04TestRelease.

In  both modes the PID is still 000A.


Just for information.

Re: Test Release 2.03 and 2.04 - SPI mode.

Reply #34
Thanks for the good work, I am waiting until I hear from you before doing anything more here, other that pinging friends for a pic programmer if they have one.

Re: Test Release 2.03 and 2.04 - SPI mode.

Reply #35
Quote
In  both modes the PID is still 000A.

Both ROM update and SUMP mode have the same PID. That isn't the correct PID (compiler error), but it is correct that those two modes share the same PID. They work from the same driver, have the same connection type, the update button just branches to a different loop.
Got a question? Please ask in the forum for the fastest answers.

Re: Test Release 2.03 and 2.04 - SPI mode.

Reply #36
@crawford - Thanks for the report. Can you please try a few things and let us know the results. We might be able to pin this on the hardware or the software with a little testing.

First, could you please post the output of the SUMP Java debug screen from a failed capture.

Second, could you please try using a program like herculese (http://www.hw-group.com/products/hercules/index_en.html) to send 00 00 00 00 00 02 to the OLS serial port. It should reply 1ALS every time. (please see the attached screenshot).

Finally, can you please try running portmon (http://technet.microsoft.com/en-us/sysi ... 96644.aspx) on the OLS serial port, then start SUMP and log port traffic during a failed capture. Please post the log here.

This debugging info could help us figure out what's going on.
Got a question? Please ask in the forum for the fastest answers.

Re: Test Release 2.03 and 2.04 - SPI mode.

Reply #37
Hi,
here are my results with the two Test packages:
The 2.03 Version works fine. With the original Firmware/Bitstreams that came with the board preprogrammed I had to reset the board manually sometimes to make it work. Now it works all the time. (But I never had any problems capturing data with the original firmware/bitstreams, so I don't seem to have that timing problem with the FPGA)

The 2.04 Version does not work at all. I tried this 4 times: switching between version 2.03 and 2.04.
2.03 always worked -
2.04 never even answered to the 5 RESET commands followed by GET-ID.

Eberhard

Re: Test Release 2.03 and 2.04 - SPI mode.

Reply #38
@Wayoda: did you also try the new firmware (0.3) Ian posted?

I had simular error (however both SPI bitstreams didn't work) but it was solved with the new firmware.

Re: Test Release 2.03 and 2.04 - SPI mode.

Reply #39
Regarding the bitstreams,
Can I run the old bitstreams with firmware 3?
Or do they all need to be ported to the 2.04 release format???

/mac

Re: Test Release 2.03 and 2.04 - SPI mode.

Reply #40
@swissMac - No, the old bitstreams have a UART connection between the PIC and FPGA, the new ones use an SPI connection. The FPGA won't respond with a mis-matched interface type.
Got a question? Please ask in the forum for the fastest answers.

Re: Test Release 2.03 and 2.04 - SPI mode.

Reply #41
[quote author="Sjaak"]
@Wayoda: did you also try the new firmware (0.3) Ian posted?

I had simular error (however both SPI bitstreams didn't work) but it was solved with the new firmware.
[/quote]
No, I don't think that would help in any way. I never had any problems with my OLS boards and my powered hub.
There are also no problems with the USB part of the firmware. The  device is enumerated by the USB system without any problems.
The linux kernel is pretty good at reporting errors in the USB-subsystem, but there is no trace of it. 
The 2.04 firmware simply never answers the getID request, connection times out (I set the timeout to 5 seconds) and that's the whole problem.
(I also tried this with the hercules-serial-utility, the device simply does not answer).

But I lost track about how this went for others, did anybody else get the 2.04 version to work?

Since the 2.03 version works fine on the first board, I will now try 2.04 on the second board.
Eberhard

Re: Test Release 2.03 and 2.04 - SPI mode.

Reply #42
@wayoda,

With firmware 0.2 my OLS didn't respond to the getID request (USB device did connect OK), when I upgraded to firmware 0.3, it automagically got working and the OLS respond to the getID:

Quote
Seems like I had  the same problem as wayoda. After upgrading the firmware to 0.3 it worked again. It seem to work ervery time and not intermittingly like crawford (not tested much yet). Haven't had a go at the integrated testpattern, since upgrading took me longer then expected ;) I'm sorry but tonight i haven't got any time to do anymore testing, sorry.

When it didn't work I tried to do it with the hercules util but no reaction from the OLS, even tried extra leading 0x00, but nothing.

I know that the changes Ian described about the 0.3 firmware don't mention fixing this, but it did work after the upgrade.

Re: Test Release 2.03 and 2.04 - SPI mode.

Reply #43
[quote author="Sjaak"]
@wayoda,

With firmware 0.2 my OLS didn't respond to the getID request (USB device did connect OK), when I upgraded to firmware 0.3, it automagically got working and the OLS respond to the getID:

I know that the changes Ian described about the 0.3 firmware don't mention fixing this, but it did work after the upgrade.
[/quote]
Confirmed. The boards responds and collects data.
Up to 4k of data the device seems to work good. All 4 trigger-steps work fine. 
If I switch to 8k of data with groups 2+3 disabled the results becomes unpredictable.
The device triggers sometimes on the right data but most of time it returns the wrong data (all zero).

So my Test status is now:
Release 2.03 works
Release 2.04 works with firmware revision 0.3 up to 4k of data

Eberhard

Re: Test Release 2.03 and 2.04 - SPI mode.

Reply #44
Here is a firmware v0.4 to test. It has some additional changes and an additional self test.

I hope maybe the extra attention and delays to the startup sequence will eliminate some erratic behavior, especially on the first run.

This is an 'SPI' link firmware, and will only work with the 2.03 or 2.04 FPGA bitstreams posted at the top of this thread.

You will need to use the updated .INF file included to reinstall the OLS after updating. The corrected PID is used, and the device now appears as 'Logic Sniffer CDC-232'.

v0.4
*updated the USB descriptor file for power, current, PID, says 'Logic Sniffer CDC-232'
*extra care to reset FPGA after SPI CS pin is setup
*ADC configured for lower power use
*ADC on PROG_B pin used to measure 2.5volt power supply as extra self test
*change to central variable for timers

Edit: it looks like the gadget factory website is down, I'll commit these changes to SVN when I can.
Got a question? Please ask in the forum for the fastest answers.