Re: [RFC] [PATCH] Improve range stepping efficiency (ensure GDB steps over entire line, at once)

[Tom Tromey <tom@tromey.com>]

>>>>> "Nav" == Nav Mohammed <nav@oscillate.io> writes:

Nav> First time contributor to GDB, please go easy on me.

Hi, and welcome to gdb.

Thank you for the patch.

Nav> I've attached a patch to address this issue, but in all honesty, it's a bit
Nav> of a hacky workaround where I've tried to keep the number of changed lines
Nav> to an absolute minimum. This is because I'm totally new to the GDB codebase
Nav> and I don't want to introduce more bugs by making too many changes. And
Nav> given that find_pc_sect_line() is called in quite a few places, I'm
Nav> concerned that some of those may *expect* the old behaviour from that
Nav> function (return a PC range for a single statement).

I have this same fear.  I don't really understand this function too
well, and I tend to think that somebody else ought to review it.

One thought that occurred to me, though, is that it would be good to
have a test case.  A unit test seems maybe simplest/best, but that seems
like a real pain, though, since the function reaches into the minimal
symbols, blockvector, etc.

Maybe the body of the central loop could be pulled into a separate
function and then that could be unit-tested?  It would only rely on line
tables, and making test line tables seems pretty easy.

I wonder if you have any thoughts on how to test it.

Tom