From mboxrd@z Thu Jan  1 00:00:00 1970
MIME-version: 1.0
Content-transfer-encoding: 7BIT
Content-type: text/plain; delsp=yes; format=flowed; charset=US-ASCII
Date: Wed, 25 Mar 2009 19:01:01 -0700
From: Roman Shaposhnik <rvs@sun.com>
In-reply-to: <7A7AC405-BEBA-4395-8611-54DE6F54E4F3@telus.net>
To: Fans of the OS Plan 9 from Bell Labs <9fans@9fans.net>
Message-id: <207E772D-A81C-432B-A12A-EF9ADFCEBFA9@sun.com>
References: <9bab1584ddc2c1730933f3047c024018@terzarima.net>
	<7A7AC405-BEBA-4395-8611-54DE6F54E4F3@telus.net>
Subject: Re: [9fans] request for more GSoC project suggestions
Topicbox-Message-UUID: c69b4eea-ead4-11e9-9d60-3106f5b1d025

On Mar 25, 2009, at 6:51 PM, Paul Lalonde wrote:
> A cfront-ish approach to templates leads to hellish duplication of
> template-generated code in each module, and thence to even worse
> code bloat.

That's not the case, really. The compiler (well, at least the
conventional one, not the one like
we have on Plan9) has very little tricks up its sleeves that can help
with that. The best
case scenario is to generate everything into .os and hope that "de-
duping" will be
done by the linker. That's how COMDAT works. cfront-ish approach can
do exactly
the same.

Thanks,
Roman.