gnupic: Re: [gnupic] gpsim --"invalid file register" error


Previous by date: 5 Feb 2007 19:32:36 +0000 Re: [gnupic] TMR1H:TMR1L dont's clear, Nestor A. Marchesini
Next by date: 5 Feb 2007 19:32:36 +0000 Re: [gnupic] in-circuit debugger, Byron A Jeff
Previous in thread: 5 Feb 2007 19:32:36 +0000 Re: [gnupic] gpsim --"invalid file register" error, David Barnett
Next in thread: 5 Feb 2007 19:32:36 +0000 Re: [gnupic] gpsim --"invalid file register" error, Ralph Corderoy

Subject: Re: [gnupic] gpsim --"invalid file register" error
From: Robert Pearce ####@####.####
Date: 5 Feb 2007 19:32:36 +0000
Message-Id: <20070205193151.bffb4fd3.rob@bdt-home.demon.co.uk>

Hi Ralph,

On Mon, 05 Feb 2007 12:03:48 +0000 you wrote:
> But I've been arguing for the coder to give that information from my
> first post of this subject!  Have you come into the thread part way
> through?  Here it is in a web archive:
> 
>     http://www.linuxhacker.org/cgi-bin/ezmlm-cgi?1:mss:6252:fdkhhdmkjajgjmpjbcll

No, I joined the thread right at the beginning, when you said :

>couldn't the assembler/compiler
>make available a set of constants.

and I pointed out that it already does. However, you also said in a later e-mail :

>I'd want the simulator to perform the check regardless of where the
>executable came from,

which rather implies you want it to work on HEX files. So I pointed out that it can't, and now you seem to be getting uppity about that. I'm sorry if my reading of what you wrote has caused any offence.

Anyway, in that first posting you also wrote :

>Similarly, compile time checks that you're not writing to the read-only
>REG_FOO_VERSION could be useful. 

And I maintain that this is not generically possible without the compiler doing code path analysis on the source.

So you're left with putting "assert BANKSEL=3" before every register access in your code, letting the assembler pass that through to the COD file, and then letting GPSim use that information. But wouldn't it just be simpler to let GPSim use the existing cross-reference to source code and show what you originally wrote, complete with the (already existing) symbol for the specific register on the specific page you meant? Oh, hang on, that's what the source browser does.

I'm sorry, I just seem to be confused. On the one hand, you think there's something insufficient about having your source code available, on the other hand you're not asking for the completely OTT option, but the gripping hand is I'm failing to see what intermediate option you're thinking of.

Previous by date: 5 Feb 2007 19:32:36 +0000 Re: [gnupic] TMR1H:TMR1L dont's clear, Nestor A. Marchesini
Next by date: 5 Feb 2007 19:32:36 +0000 Re: [gnupic] in-circuit debugger, Byron A Jeff
Previous in thread: 5 Feb 2007 19:32:36 +0000 Re: [gnupic] gpsim --"invalid file register" error, David Barnett
Next in thread: 5 Feb 2007 19:32:36 +0000 Re: [gnupic] gpsim --"invalid file register" error, Ralph Corderoy


Powered by ezmlm-browse 0.20.