This is the mail archive of the
mailing list for the GCC project.
Re: incremental compiler project
- From: Manuel LÃpez-IbÃÃez <lopezibanez at gmail dot com>
- To: David Kunsman <dmkunsman at gmail dot com>, GCC Development <gcc at gcc dot gnu dot org>
- Cc: Tom Tromey <"\"tom"@tromey.com>
- Date: Thu, 3 Sep 2015 18:36:24 +0200
- Subject: Re: incremental compiler project
- Authentication-results: sourceware.org; auth=none
- References: <CAPVyUPD0eAyAuPOktpjqpY=UKx0t2YEvFQxMUMY2Lv2pwUAjLA at mail dot gmail dot com>
On 02/09/15 22:44, David Kunsman wrote:
Hello, I just read over the incremental compiler project on the gcc
wiki...and I am excited to try to finish it. I am just wondering if
it is even wanted anymore because it is 7-8 years old. Does anybody
know if this project is wanted anymore?
The overall goal of the project is worthwhile, however, it is unclear whether
the approach envisioned in the wiki page will lead to the desired benefits. See
http://tromey.com/blog/?p=420 which is the last status report that I am aware
of. In addition to that, the implementation itself would already be incredibly
challenging for anyone with many years of experience in GCC.
I do not wish to discourage you, perhaps you will manage to pull it off. On the
bright side, GCC has improved a lot since 2008 and there are many new
improvements that will probably make the incremental compiler easier to
implement today, such as binding-oracle in libcc1, global state removal by
libgccjit, recent modularization work, etc.
If you have never submitted any changes to GCC, I will strongly recommend that
you start with something small and immediately useful such as any of the
EasyHacks (https://gcc.gnu.org/wiki/EasyHacks). Then, move to something more
complex and related to the incremental compiler, but still self-contained such
as "The C front end has been modified to lex the entire file at once". This
would be useful on its own since it is likely to speed-up the C FE (and it is
what C++ already does).