Re: Rombug?

[Elena Zannoni <ezannoni at cygnus dot com>]

I forgot to mention that another rombug hack is present in
symbol_file_command(). Same story, it gets the .text address from the
target side.  Can I get rid of that too? It would simplify the code
quite a lot.

Elena


Stan Shebs writes:
 > Elena Zannoni wrote:
 > > 
 > > While cleaning up the syntax for add-symbol-file, I saw that that
 > > command has a special syntax just for the rombug case.  For rombug, one
 > > doesn't need to specify the address of the .text section to load the
 > > file at. For all the other cases, such address must instead be specified as
 > > argument to the command.  Needless to say, this little variation
 > > complicates things quite a bit.
 > > 
 > > When is rombug used and what exactly is it? There is a brief mention of
 > > it in the gdb manual, but not enough for me to understand much.
 > > Is it still in use?
 > 
 > Rombug is the boot monitor for Microware's OS-9 / OS-9000 (yes, these are
 > the guys that are suing Apple over having a version 9 of MacOS, believe
 > it or not), so its role is similar to that of CygMon in eCos.
 > remote-os9k.c is just another cheesy monitor interface, and probably
 > hasn't been used in a long time, so I don't anybody will notice if it
 > crumbles a little more.  From surfing around Microware's web site
 > (www.microware.com), it looks like they recommend their own proprietary
 > debug protocol anyway.
 > 
 > I'd recommend leaving out the special rombug hack, but putting extra
 > notes in the sources and ChangeLog, so that if someone goes looking for
 > it in the future, they won't wonder if they were hallucinating when they
 > remembered some sort of rombug support in add-symbol-file.
 > 
 > Stan
 >