On 08/02/2013 09:48 PM, Tom Tromey wrote:
Pedro> These new target-resume.h, target-wait.h, target-waitstatus.h,
Pedro> target-waitstatus.c files might be looked at as actually something
Pedro> else. This is code defining the interface between GDB core and
Pedro> target_ops, and as such is used by all sort of targets on the
Pedro> GDB side (remote.c, etc.). I'm not sure they should go in the same
Pedro> directory as the *-nat.c, etc. files.
These seem like classic "target" bits to me.
Yep. So, if we move the classic "target" bits to a "target/"
module directory, and put the native bits in their own dir, we
have:
target/resume.h
target/waitstatus.[c|h]
target/wait.h
nat/i386-nat.c
nat/linux-nat.c
nat/linux-ptrace.c
nat/linux-waitpid.c
etc.
Is this what you're thinking of? _This_, I'm fine with.
No mix of native bits with generic "target" bits, which was
my main worry.
It's actually very similar to something else I suggested on IRC,
but forgot to put in email form: "IMO, the interfaces themselves would be
in an include dir. e.g., gdb/include/target-waitstatus.h or some such,
and then we'd have gdb/nat/linux-nat.c, etc."
What else goes in "target/" ? remote.c, corelow.c, etc.?
Do we move things into subdirectories beneath it too, for better
submodule partitioning? I didn't want to suggest starting a
mass move, that's easy to overdo. (That was the point at which I
suggested that someone thinks this through, and comes up with an
initial design/guide of what things will look like in the end, so
that we can then discuss and hash it out.)