Re: New breakpoint_re_set call vs remote targets

[Daniel Jacobowitz <drow@false.org>]

On Thu, Jun 25, 2009 at 09:52:36AM -0700, Doug Evans wrote:
> I wonder if one useful step is to reassess post_create_inferior, and
> maybe split it up or something.

Adding it was a huge boon for consolidation.  I don't really want to
split it up again if we don't have to.  Lots of actions used to be
missed when using a particular target, such as remote or remote-mips
or even core.

Converting watchpoints from hardware to software (the point of the new
call) is a sensible thing to do at the point of connection.  It's just
the prologue skipping that's bitten us - I like Pedro's idea of doing
this at load.

> From my perhaps ancient point of view, gdb is for debugging two kinds
> of programs: hosted and freestanding (to borrow jargon from C - though
> non-bare-metal and bare-metal may be more accurate. 1/2 :-)), and I
> wonder if they're being inadvertently fused.

I've been wondering about this too... I fear that if we introduce any
switch between these two modes, we'll find it's not granular enough.

-- 
Daniel Jacobowitz
CodeSourcery