[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Axiom-developer] Hash problems
From: |
Camm Maguire |
Subject: |
Re: [Axiom-developer] Hash problems |
Date: |
Sat, 28 Jun 2014 20:26:14 -0400 |
User-agent: |
Gnus/5.13 (Gnus v5.13) Emacs/23.4 (gnu/linux) |
Greetings! I've resolved it now, but we should be aware of this going
forward.
axiom currently relies on a broken GCL implementation of functionp,
which incorrectly returns t for fboundp symbols. axiom uses this to
decide whether to funcall, as in (if (functionp x) (funcall x ...)).
With a correct functionp, as in GCL master, it should read (if (if
(symbolp x) (fboundp x) (functionp x)) (funcall x ...)). I think you
already have something like this in |functionp|. Anyway, when I release
a fix to this eventually, axiom will need to be changed. Luckily, the
axiom change can occur now with existing gcl as well.
The forthcoming 2.6.11 will still have the broken functionp.
(see book5)
Take care,
address@hidden writes:
>> Hi Tim! Working on 2.6.11, and did something to prevent the
>> hashing/memoization from taking place in pascal.input, e.g. pn(50)
>> hangs. How can I see what functions are being generated and trace the
>> compiler? )lisp (trace compile) does not work.
>
> The function you want is |genMapCode| in src/interp/i-map.lisp.pamphlet
> This calls |reportFunctionCompilation| in src/interp/slam.lisp.pamphlet
>
> It appears to use the variable |$whereCacheList|
>
> Tim
>
>
>
>
--
Camm Maguire address@hidden
==========================================================================
"The earth is but one country, and mankind its citizens." -- Baha'u'llah