Re: [RFA] wrong language used when re-setting breakpoint

[Tom Tromey <tromey@redhat.com>]

>>>>> "Joel" == Joel Brobecker <brobecker@adacore.com> writes:

Joel> The reason behind the error is that the linespec code that re-sets
Joel> the breakpoints uses the current_language global when iterating
Joel> over a symtab's symbols. However, the that global gets switched from
Joel> Ada to C during the startup phase, probably as a side-effect of stopping
Joel> in some system code for which debugging info is available. The fix
Joel> is to make sure that we use the correct language.

I think the breakpoint re-setting code should be temporarily setting
current_language.  breakpoint_re_set does this.

I see that linespec now is trying to treat the current language as a
parameter, not a global.  I think that's a good idea, but I wonder
whether it is done consistently enough.  For instance, does linespec
really not call anything that might implicitly use the global?  (I
expect at least expression parsing...).  And offhand I see a use of
current_language in get_search_block.

I don't have a problem with your patch.  I think it is an improvement.
But I think it is important to understand why the current mechanism
isn't working, since I think a failure here may have other bad effects
as well.

Tom