[Ecls-list] Improving CLOS?

[Ram Krishnan]

On Thu, Mar 18, 2010 at 3:30 PM, Juan Jose Garcia-Ripoll
<juanjose.garciaripoll at googlemail.com> wrote:
> On Thu, Mar 18, 2010 at 11:06 PM, Matthew Mondor <mm_lists at pulsar-zone.net>
> wrote:
>>
>> I'm however still relatively new at CL and have only read a bit about
>> MOP (too few to fully understand its implications, I've not finished
>> AMOP).  This surprised me, but one of the points which the MOP
>> introduction stressed was how it was possible to have MOP even enhance
>> the performance of a system, but I assume that it was in comparison to
>> some heavy and slow object-oriented implementations of the time?
>
> The MOP allows coding dispatch functions that are different from the default
> ones. They can be faster, not relying on multiple classes dispatch, using
> hashes, or even defaulting to something like C++'s indexed arrays of
> methods. In principle this can be done, but it implies a huge effort.
>
> Juanjo
>
> --
> Instituto de Física Fundamental, CSIC
> c/ Serrano, 113b, Madrid 28006 (Spain)
> http://tream.dreamhosters.com
>
> ------------------------------------------------------------------------------
> Download Intel® Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> Ecls-list mailing list
> Ecls-list at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/ecls-list
>
>

Juan,

The boot time initialization of CLOS in ECL is of particular interest
to me. Do you have you any thoughts or suggestions on how to reduce or
eliminate the need for the CONVERT-ONE-CLASS call in the fixup pass?
While the time cost of doing this is mostly negligible in a desktop
environment, it isn't so cheap in a lower powered embedded
environment.

Cheers,

-ram