Re: Benchmarking temporary Lisp objects [Was: Re: [RFC] temporary Lisp_Strings]

[Paul Eggert <eggert@cs.ucla.edu>]

Dmitry Antipov wrote:
> I'm voting for simpler and cleaner version with
> alloca/fallback to regular GC for vectors and strings.

I like the idea of simpler and cleaner, but your patch's implementations 
of build_local_vector and build_local_string use alloca expressions as 
function arguments, and the GNU API for alloca doesn't support that. 
You can work around the problem by using statement expressions, falling 
back on plain Fmake_vector and/or make_string for compilers that don't 
support statement expressions.

To address Stefan's concern about alloca-allocated thingies escaping to 
Elisp, is there some way you can include a runtime test for that, when 
checking is enabled?  It wouldn't have to catch every escapee, just a 
high percentage of them.