Menu

Show posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Show posts Menu

Messages - Jan Waclawek

#121
Feature Requests / Re: ISP software in linux
April 20, 2006, 01:49:35 PM
While I have only marginal experience with Linux and zero experience with using programming tools under it, I often use a plain terminal emulator for programming the RD2s (Temic/Atmel and Philips, both 'C' and 'V' - I recommend the latter). Requires perhaps a bit of self-discipline, but the Linux users are used to it, dont' they? :-)

Have a look here: http://www.8052.com/users/redtommo/

... and perhaps let us know about your findings...

Good luck!

Jan Waclawek
#122
Bonny,

> Ofcourse I read the manual, and I have found all info I need already.

Then what was the purpose of posting the question here? And, if you found out the answers you needed already, why didn't you post those here, or at least a note saying you don't need us to answer your questions any more?


> It is just that when I am searching for infos, I don't like to waste much time,
> so I put outstanding queries as much as I can, on different fora and groups.

That's great. You don't want to waste YOUR time, so you don't hesitate to ask others to waste THEIR time; and when they DO so, PLUS point out you might have found the information yourself with a little effort, you feel offended.

Next time, please, consider the value of time of the participants in these fora to be of the same value as your own time and post your requests accordingly. Thanks.

Have a nice day!

Jan Waclawek.
#123
> Where exactly is the orig bootloader located

In the boot block, block1, 8kB of Flash located at 0-1FFFh and accesses by the '51 core depending on BSEL(FCF.0)and SWR(FCF.1) bits. But you should (forgive me) RTFM first - this is chapter 7.1.1.
The bootloader itself (including IAP support) occupies the first 2kB of the boot block, plus some space at the top (IAP jump vector, security key and some other data)

> Is it possible to execute it by "jumping" to it?

Yes, why not. Place a code above 2000h, which switches to Block1 (clear both BSEL and SWR) and jumps to 0000h. Note that by this time transmission of the 'U' character for autobauding should have started, otherwise the bootloader will restart into the application code within the watchdog timeout.


> What about bootvector registers or statusbytes?

None of them (of course - have you seen them in the manual?) But there are two things of a similar kind, affecting the behaviour of bootloader, the security key ("serial number" as they call it in the datasheet), and the SoftICE mode flag. This is in addition to two "hardware" bits, the security lock bit and the x2 mode bit.
For details, again, refer to the datasheet.

Jan Waclawek
#124
General / Re: Question about backdoor.doc by eric
March 09, 2006, 12:19:48 AM
Sorry, tired a bit...
But it would be nice to have a common tool, wouldn't it? I'd guess it would be a piece of cake for Andy to support all of them, just the sponsor of FM would not like it at all...
;)

Jan Waclawek



Post Edited (03-09-06 01:20)
#125
General / Re: Question about backdoor.doc by eric
March 08, 2006, 09:13:20 AM
Yes, but they erase - contrary to the rest of FLASH - to 00h, so if you have the chip running the application from 0000h, you have at least the status byte in the erased state...

Not anything you will read out from the datasheets but you can find it out yourself experimenting with IAP using a terminal program rather than [inserted]FlashMagic[/inserted][strikethrough]FLIP[/strikethrough], if you are brave enough (or if you have a parallel programmer at hand :-)

Jan Waclawek



Post Edited (03-09-06 01:17)
#126
Playing with a P89V51RD2-based gadget and FlashMagic I realized an another reason for the "Unable to connect at the specified baud rate...." message: there was an application running, which echoed back the characters, but worked at a different baudrate (19200) than I set FlashMagic (38400). So I had a perfectly working connection and everything, but FM appeared to fail.

Maybe it is trivial, but maybe it would help to somebody.

Jan Waclawek

PS. Just for the record: IMHO the proper operation sequence is: switch the device OFF, press "Start" in FM, window "Reset device now" appears, switch the device ON. Alternatively, the device can be held in RESET all the time, if a RESET button is available.



Post Edited (02-28-06 10:16)
#127
P89V51Rx2/P89LV51Rx2 / Re: 89V51RD2
February 28, 2006, 09:10:47 AM
The best '51 disassembler around is on the 8052.com page, where this question belongs anyway.

Jan Waclawek
#128
- run a terminal program (I prefer terminal by Bray rather than hyperterminal) 9600/8N1, no handshake (and no local echo o hyperterminal)
- remove the P89C51RA2 from socket (I hope it is socketed) and short Rx and Tx pins in the socket. Type anything, it should echo back. This checks the serial chain (COM port in PC, cable, MAX232, PCB etc.)
- put the chip back. Make sure the hardware conditions (/EA, /PSEN, RESET) are met for starting the bootloader. I hope you have a proper reset generator (invalid reset may result in "wonders"). Make sure the oscillator works (using oscilloscope or counter).  Type "U", it should echo back "U"
- if it does not, it may be the infamous "boot vector lost". Go back to the shop where you bought it and ask (politely) the owner for help.

Jan Waclawek
#130
The general recommendation is use a lower baudrate (what is the baudrate you are attempting now?). 9600 should be generally fine. There is a hint on my website on which baudrates are suitable for autobauding with particular oscillator frequencies.

But I'd try the integrity of the serial line first - see e.g. http://www.8052.com/forum/read.phtml?id=95101

Jan Waclawek
#131
Dear Rajani (& Naga),

What is the value of crystal you are using? With certain values of clock it is better to use a lower baudrate (although 9600 should work pretty well with most of the common crystal values...)

Also, you should rule out the usual sources for programming errors, such as faulty/insufficient power supply, insufficient decoupling of supply pins, faulty reset etc.

Jan Waclawek

PS.: Alternatively, you can try downloading the P89V51RD2 from a suitable terminal emulator (hyperterm, tera term) by simply sending out the hexfile to it. Set a small delay after line (e.g. 100ms) and don't forget to erase beforehand (:0100000301FB).
For Tera Term, I have written a macro which enables to use it as a command line tool for downloading P89V51RD2 ( http://www.efton.sk/t0t1/sendhex.ttl ).



Post Edited (01-02-06 06:00)
#132
IMHO the setting of high-medium-low frequency crystal modifies only the drive and/or feedback of the oscillator circuit and not the input, so to recover a chip with low frequency crystal selected, remove the crystal and apply an appropriate high-frequency clock to XTAL1.

Usual disclaimers apply...

Jan Waclawek
#133
P89V51Rx2/P89LV51Rx2 / Re: pictiva
November 21, 2005, 11:29:05 AM
That might also be a customized (e.g. crypting) bootloader, which was developed originally for the "non-L" version and simply burned into the "L" version by the supplier of the application.

Jan Waclawek
#134
> ...tripods...

...such as MAX810 and the uncountable clones of it...


Jan Waclawek
#135
LPC9xx/LPC9xxx / Re: Security Bit
October 20, 2005, 06:40:06 AM
In which chip?