On Thu, Oct 16, 2003 at 02:47:10PM +1000, Chris Johns wrote:
>Christopher Faylor wrote:
>>Is anyone actively considering cleaning up the mingw patches, getting
>>assignments for them with the FSF, and contributing them to gdb so that
>>gdb will, like gcc and binutils, build right from CVS?
>I am. I have applied for an assignment (in the post) and I am looking
>over the GDB sources to get the patches into shape to post. I assume I
>post to gdb-patches.
I think I've already mentioned that you won't be able to submit patches
that you did not author. So, you will need to find the authors of all of
the gdb changes you submit and make sure that they have assignments as
well. Either that or you'll need to start from scratch.
(these aren't my rules, so please don't shoot the messenger)
>My approach is to sort out the gdb code changes that are generic, eg
>changing 'struct environ' to 'struct gdbenrivon' first.
That's one that I can't approve.
>The next are a range Windows specific changes. These are a little harder as
>the changes add Windows code in files such as top.c and event-top.c. This
>will also include ^C and ^BREAK support if I can figure out a solution. I
>am currently looking at this and how it will fit in.
>Finally, I kind of understood from Andrew Cagney that MinGW needs to be
>updated for newer target support (something about the target header files)
>for gdb. I am looking at the GDB doco and tring to understand this a little
I don't see why this is necessary. It's a goal for the windows target in
general but I don't see why mingw should force this change.
>The last part is readline. Not too sure what happens here as readline is
>not really a GDB problem. It would be nice to have readline patches
That would be an upstream activity. You'd have to submit the changes to
the FSF maintainer.