Alan DuBoff wrote:
> On Wednesday 07 March 2007 01:44 pm, Bruno Jargot wrote:
> > You could start creating a solution based on ksh93. I think it may be
> > a good candidate for a Google Summer of Code project.
> 
> And how would this help us reduce the size of the current footprint?

Depends on what "footprint" means. ksh93 looks _huge_ from the size of
the binaries but the majority of this size is code, tables and the
builtin manual pages which are all read-only data (e.g. this is
|mmap()|'ed and consumes memory only once). The shell could be further
stripped, for example the seperate codepaths for
singlebyte+multibyte-locales could be folded together etc. which would
reduce it a litte bit further (some of the size reduction work is
already on our ToDo list).

The real advantage of ksh93 happens if it _runs_ - it's very fast
(unlike other shells ksh93 avoids |fork()| at all costs, e.g things like
$ ( cmd1 ; ( cmd2 ; cmd3 ) ) # no longer trigger the |fork()| of a
process per subshell (nor do builtin commands create a seperate process)
# , resource usage is quite low (lower than the old ksh88 and we're
going to work on that after the initial putback), many commands are
already builtin or can be replaced by builtin ksh language
functionality, adding more builtins like "grep" require little work (for
example in "grep"'s case most of the functionlity is already present in
libast) and more builtins can be loaded at runtime on demand.

----

Bye,
Roland

-- 
  __ .  . __
 (o.\ \/ /.o) [EMAIL PROTECTED]
  \__\/\/__/  MPEG specialist, C&&JAVA&&Sun&&Unix programmer
  /O /==\ O\  TEL +49 641 7950090
 (;O/ \/ \O;)
_______________________________________________
opensolaris-code mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/opensolaris-code

Reply via email to