Re: a recursive lock prototype

[Thomas Bushnell, BSG]

Marcus Brinkmann <Marcus.Brinkmann@ruhr-uni-bochum.de> writes:

> We can talk about this for the Hurd (I think we can do fine with a small
> interface change), but at least in the dynamic linker there has to be a
> recursive lock (there is a XXX in the code for that, we will deadlock on the
> second invocation.  But I don't know under which circumstances this will
> happen.)

Perhaps we can fix the problem in the dynamic linker; I don't know
what the interface looks like.

If an interface is designed poorly, it is often the case that it can
only be implemented through using a global variable: and the need to
use recursive locks is just another kind of global variable
reference.  

It would depend on the details of the interface, and the dynamic
linker interface is the sort of thing that is more-or-less fixed.  I
don't know whether it really needs recursive locks or not without
understanding the case in more detail.  Perhaps Roland can shed light
on the issue?

However, in the Hurd I'd like to keep recursive locks out of the works
entirely; we should always be able to write interfaces such that they
aren't necessary.  (As indeed the Hurd support code in libc also is
careful to do.)

Thomas