Actel PA3 with DirectC or SVF, anybody had any success?

Hi

I wonder if anyone had any success using either DirectC 2.2 or SVF files to program Actel PA3 devices. All my attempts are failing so far.

1) DirectC 2.2 compiled with VC2005 (adapted to Xilinx Cable III), had 1 time success from 100 attempts to program the part 2) playing SVF or XSVF files with Impact fail very quick, SVF doesnt even get as far as checking jtag IDCODE 3) custom SVF player also shows weird behaviour

Antti

Reply to
Antti
Loading thread data ...

Maybe try with the Amontec SVF Player + the Amontec JTAGkey (USB to JTAG Cable). Amontec SVF Player support Trailer Header and infinite Scan Length.

Note: if the IDCODE failed SOMETIMES this could mean you have a bad JTAG connection (between the Xilinx Cable III target board). Make sure to not route the TCK parallel to the TDI or TDO, we have the experience that this could infer to the stability of the JTAG FSM even if you run JTAG at very low speed. You could try to add a 100R resistor on TCK close to Target.

Almost sure your issue is not coming from the SVF file nor the Player, but from your JTAG signal integrity.

Regards, Laurent

formatting link

Reply to
job

Larry

normally I would agree with you, but:

it is not that the failures are some times or random:

1) TLR-->shift-DR ===> correct IDCODE ALWAYS 100% reliable !!!

2) TLR->RTI->shift-IR (IDCODE)->RTI->shift-DR

100% failure when trying with SVF file and impact 100% ok, when trying XSVF file and impact

100% failure when adding extra delay in RTI state (TCK IDLE NOT TOGGLING) in custom SVF player

100% ok when doing FAST playback in custom SVF player

3) DirectC a) once worked full programming cycle b) place of failure depends on the VC compile target (debug or release) the place of failure is constant, but rather soon in the file, ISP_entry command, after IDCODE check. the DirectC timing calibration value is dependant on the build target, so i assume some speed issue here

hardware currently in use

*) Amontec Chameleon with Xilinx JTAG fly-wire adapter *) A3P250 on proto board, wire length from IC to JTAG pin < 10 mm for all connections

I cant use Amontec svfplayer it does NOT SUPPORT amontec Chameleon : ( for some time ago I purchased 3 pcs of FT2232 mainly to rebuild the jtagkey, but didnt ever bother doing it, so I dont have any jtagkey to test with :(

Please look above my observations, could you still assume it is signal integrity issue? The pass-fail depends on the time spent in RTI when TCK is not toggling, meaning there are no signal toggling at all! I am puzzled at least.

Antti

Reply to
Antti

This could help.

What is your ENDIR and ENDDR ?

Maybe try to add RUNTEST when in RTI. But if this help, there could be a bug in the Actel JTAG FSM!

... anyway, why do you not provide us a part of the SVF you are running for the IDCODE?

Laurent

formatting link

Reply to
job

== cutout from Actel libero 8.1 generated SVF == FREQUENCY 4E6 HZ; STATE RESET; RUNTEST IDLE 5 TCK; ENDIR IRPAUSE; ENDDR DRPAUSE; SIR 8 TDI(0F); SDR 32 TDI(00000000); STATE IDLE; RUNTEST IDLE 1 TCK; SDR 32 TDI(00000000) TDO(02A141CF) MASK(0EFFFFFF); ===

the above fails 100% with impact if played back as SVF the same or modified one converted to XSVF works 100%

in my own code i tried ENDxR IDLE vs PAUSE but i did not work out any fully working solution

Antti

Reply to
Antti

Try the two following one and let us know the result:

== cutout from Actel libero 8.1 generated SVF == FREQUENCY 4E6 HZ; STATE RESET; RUNTEST IDLE 5 TCK; ENDIR IDLE; ENDDR IDLE; SIR 8 TDI(0F); SDR 32 TDI(00000000); // STATE IDLE; RUNTEST IDLE 1 TCK; SDR 32 TDI(00000000) TDO(02A141CF) MASK(0EFFFFFF); ===

== cutout from Actel libero 8.1 generated SVF == FREQUENCY 4E6 HZ; STATE RESET; STATE RESET; STATE RESET; STATE RESET; STATE RESET; RUNTEST RESET 5 TCK; ENDIR IDLE; ENDDR IDLE; SIR 8 TDI(0F); SDR 32 TDI(00000000); // STATE IDLE; RUNTEST IDLE 1 TCK; SDR 32 TDI(00000000) TDO(02A141CF) MASK(0EFFFFFF); ===

Laurent

formatting link

Reply to
job

had

I tried some similar things, and many more, with not much luck

Antti

Reply to
Antti

had

doesnt

the

even

Player,

Antti,

But did you have try the two ones I gave you? similar or similar or maybe similar ? ;-)

Laurent

formatting link

Reply to
job

SVF

so

III), had

doesnt

the

even

Player,

to

signal

am

maybe similar ;)

I will run your commented version when i have the system again online, but do not think they will make explanation visible for the problem

Antti

Reply to
Antti

I must answer solve myself, here are reports

PCB1_a3p250

- directc 2.2 programmed 100% one time out of 150, 149 times fail very quick (first commands already)

- flashpro - FAIL (bad response for enter isp)

PCB2_a3p060

- directc 2.2 DEBUG Build, works 100% each time

- directc 2.2 RELEASE Build works maybe 1 of 10, when fails the enter_isp mode

- flashpro - works

so it does look like some problem related to the chip or wiring or powersupply, still little strange that the release-debug setting while compiling directC makes so big difference

Antti

Reply to
Antti

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.