Re: Release 1.6 critical TODO items.

[Rob Browning]

Evan Prodromou <address@hidden> writes:

> No, they don't.
>
> It *is* a common thing to do, though.

Looking in ice-9/threads.scm, we do have (with-mutex (mutex . body))
which uses a dynamic wind to make sure the mutex is always locked
within the body.

Oh, and Marius pointed out that 1.5.6 is broken if you don't specify
--with-threads.  That's because I presumed (without thinking about it)
that the mutex ops would be no-ops when threads weren't available, but
they're actually undefined.  It looked like checking (defined?
'make-mutex) was how I'd need to conditionalize the code, or is there
something like (threads-available?) that I overlooked?  (In the long
run, this seems like a good candidate for cond-expand or similar).

(BTW: did anything ever come of that long argument about dynamic-wind
 and call/cc on comp.lang.scheme?  If nothing else, we should document
 exactly how guile handles those issues unless we already have...)

-- 
Rob Browning
rlb @defaultvalue.org, @linuxdevel.com, and @debian.org
Previously @cs.utexas.edu
GPG=1C58 8B2C FB5E 3F64 EA5C  64AE 78FE E5FE F0CB A0AD