Steel Bank Common Lisp

On Fri, 25 Jun 2004, Paolo Amoroso wrote:

> I heave read in the past about problems with SB-THREAD and x86 Linux
> 2.4.x kernels, and the suggestion to use 2.6.x.  But I am still a bit
> confused.  I don't understand whether such problems and issues make
> threads undesirable with 2.4.x for production quality work, or also
> for playing and experimenting at a hobby level.  I am interested in
> the latter (I don't have a nuclear power plant, if that helps).

Dan'b can confirm or disconfirm this, but AFAIK:

 Building SBCL with threading enabled under 2.4 should not
 affect the stability of the system as long as threading isn't
 actually used.

 Threads under 2.4 do have problems, and fixing has a priority of "none"
 or thereabouts. For "simple" use this may be ok, but at least
 signal delivery can get seriously confused fairly easy.

 In other words: ok for experimenting and playing, but when/if things go
 wierd you're on your own (well, in good company actually, but without
 any support to speak of). For production quality work definitely
 discouraged.

Cheers,

 -- Nikodemus                   "Not as clumsy or random as a C++ or Java.
                             An elegant weapon for a more civilized time."

Nikodemus Siivola <tsiivola@...> writes:

>  Threads under 2.4 do have problems, and fixing has a priority of "none"
>  or thereabouts. For "simple" use this may be ok, but at least

Yes, that's the only thing about thread that was clear to me :)

Thanks for the explanation,


Paolo
-- 
Why Lisp? http://alu.cliki.net/RtL%20Highlight%20Film

Daniel Barlow <dan@...> writes:

> Despite my own advice I have actually been running a low traffic web
> site on threaded sbcl for the last six to nine months, and I think
> only had one crash (had a couple of other restarts for unrelated
> reasons).  But it's just a web site; it does the same thing day in day

Do you mean with kernel 2.6?


Paolo
-- 
Why Lisp? http://alu.cliki.net/RtL%20Highlight%20Film