From: Karl B. <ka...@tu...> - 2002-01-08 05:15:14
|
I started to use the P89C51RD2 at work as well. Nice chip. I added some support for it in my terminal program ttty which you can download on my web site. The support isn't polished yet, and it should be because its easy to break these chips so you can't get access to the ISP bootloader. In order to run your app you need to re-program the STATUS flash config byte to a zero. Only way to do this is to flash erase the STATUS and BOOT-VECTOR byte together. But the BOOT-VECTOR needs to be re-programmed to point at the built-in bootloader, otherwise the bootloader(ISP) becomes inaccessable. They don't highlight this in the doc. I've got one dead chip I hope to resurect by hooking up to an eprom system and writing some code to call into the bootloader to re-program it. Once you re-program these config bytes, you don't really have to do it again, you strap PSEN to ground for bootloader or let it float for the application. I don't think Dynamic C runs out of the GUI, which is a shame. And I'm afraid what you see is what you get with Dynamic C, the only difference you will get with the full-product is some more source code for all the boards.(You won't get a console program with multiple .c capability). The non-standard behavior is annoying. I found a way to work around it. I had some existing code that I take pride in keeping portable, and I certainly wasn't going to bastardize it to Dynamic C format. So I made up some scripts to copy the .c and .h files to .lib files, and to play games with the .h file format so that I could compile it with standard C compilers and generate the odd-ball Dynamic C format. It worked, and afterwards I was happy with the compiled and running results. But it would be nice to get SDCC(a real C compiler :-) running with rabbit. cheers, Karl. Rod Boyce wrote: >Karl and all, > >If you/we get the compiler working. I'll easily get the flash drivers, >serial drivers and a better BIOS than the Rabbit BIOS working simply as this >is what I do in my day job. I'm currently working on embedded Linux and the >power PC but I also do HC11, 8051 and lots of other chips. I am currently >using the SDCC compiler on a 8051 project using the P89C51RD2 micro from >Philips a very sexy micro my intention is to write a Windows first (hey it >is what I know best) then a Linux downloader. All of these will be >available free most likely on the knowledge site. > >Back to the Rabbit. I have bought a development kit for the Rabbit and was >very disappointed when I discovered the limitations in the compiler, only >one file no includes, must compiler in the crap(IMHO) GUI, this is basically >a toy / teaser. If you want the real thing you have to buy another $250USD >CD with the compiler on it and I have yet to find out if it supports >makefiles and multiple file compilation. The Rabbit micro looks very >interesting but I want a decent compiler to try it out with. >After we get the compiler I could create a BIOS with all the drivers you can >imagine under the sun. But I want a decent compiler like SDCC. > > >Regards, >Rod Boyce > > -----Original Message----- > From: Karl Bongers [mailto:ka...@tu...] > Sent: Monday, 7 January 2002 20:16 > To: Rod Boyce > Subject: Re: [Sdcc-user] Porting SDCC to the Rabbit >processor? > > There is some interest here. I started investigating, and >found a nice > GPL project on sourceforge(z2k tool suite) which is an > assembler/preprocessor, download app, and startup >code/kernel > for the rabbit. I talked to the author Stephen Hardy, and >he is > having a look at SDCC. > > I have a goal of getting a small bit of SDCC code running on >a rabbit > board. A rabbit is mostly z80 compatible, so I should be >able > to generate a small application with Michael Hope's z80/SDCC > compiler. > > I'm working on adding z80 support to ucsim, so I should know > more than I want about z80 soon. I'm checking this > in to the ucsim-034-pre3 branch that Bernhard mentioned > a few days ago. This will be a work in progress for >sometime. > > Some of the tough issues that would need to be solved >include > code bank switching seamlessly to allow for apps larger than > 64K. Also, basic hardware support like flash-burning, > serial drivers, can be a lot of work to generate. > > For now, stick with the Dynamic C, but hopefully in the >future > we could adapt SDCC for the rabbit. > > karl. > > > Rod Boyce wrote: > > >All, > > > >Is any one planning on porting SDCC to the Rabbit >processor? I'm an > >embedded developer not a compiler developer and wouldn't >not where to start > >but am happy to be a guinea-pig and help with the port if I >can. > > > > > >Regards, > >Rod Boyce. > > > >_______________________________________________ > >Sdcc-user mailing list > >Sdc...@li... > >https://lists.sourceforge.net/lists/listinfo/sdcc-user > > > > > > >_______________________________________________ >Sdcc-user mailing list >Sdc...@li... >https://lists.sourceforge.net/lists/listinfo/sdcc-user > > |