[LLVMdev] seeking advice
mats at planetcatfish.com
Mon Aug 3 10:57:01 PDT 2015
When I set out on my "Pascal Compiler project", I had never written a
compiler before - I had SOME experience in writing interpreters and other
parser/parser-like projects. That's about 14 months ago. I think I started
around end of February, and by the beginning of march, when I pushed the
first commit to github, I already had small programs working.
I only spend (some of my) evenings and weekends on this project, and there
have been periods where I didn't do any work for several weeks for one
reason or another. I'm now at the a stage where the parser and semantic
analysis is parsing the iso7185pat.pas ("pat = Pascal Acceptance Test").
It's not yet passing the code-gen phase - the current stumbling block is
trampolining for nested function pointer calls, and I think there are going
to be a few other problems along the way. Given that this file is "trying
to do everything the ISO7185 standard allows", it's not surprising it finds
a few interesting issues...
I have a few decent size home-grown tests, such as a sudoku solver that
"works" - and because LLVM provides good optimisation, my compiler beats
the "regular" Free Pascal compiler on all like-for-like comparisons I've
made so far - in particular, the Sudoku solver is about 2x faster with my
compiler than Free Pascal (ok, I cheat, some 5% of that comes from having
implemented a "popcount" intrinsic that can take a "set" and give the
number of elements in the set, which standard pascal or Free Pascal doesn't
Of course, like Renato says, for every small step towards "full
functionality" gets harder - and the good thing with a hobby project is the
ability to choose how much time you spend on what features, and how you
I thoroughly recommend "try it, see how it goes". I've found it
surprisingly easy... [Mostly because other people have done the REALLY
hard part of making the LLVM IR to machine code phase! - Thanks guys!]
On 3 August 2015 at 18:07, Renato Golin <renato.golin at linaro.org> wrote:
> On 3 August 2015 at 16:29, Arno Bastenhof <arnobastenhof at gmail.com> wrote:
> > Based on these
> > readings, as a matter of exercise I can take what I learnt from LCC to
> > a small compiler in C targeting the LLVM IR for, say, a subset of Prolog.
> Hi Arno,
> This is not a monumental task. There are many tutorials in the LLVM
> tree, and producing IR from a simple AST is actually simple and
> So many languages can target LLVM IR, that it's very hard that you'll
> find something that cannot be represented in IR. At least something
> that other implementations haven't found already, and worked around.
> With your broad knowledge in CS, I wouldn't expect you to take more
> than a month or two to get a front-end working with the JIT compiler
> on a reasonable subset of Prolog. Of course, the deeper you go, the
> more time it takes to progress the same 1%, but you can always stop
> when productivity drops as fast as enjoyment. :)
> But you won't need to understand the back-end at all, modulo some
> quirks on your platform of choice. The hint is to write the same logic
> in Prolog and C, compile the C with Clang and see what they do. If you
> spot hackery, the least resistance path is to replicate the hackery.
> Asking on the list is also a good way to understand why the hackery,
> and maybe even give you an entry point into changing the middle/back
> end to support a proper construct in IR for better codegen.
> From there, I'd recommend you to try different targets. Get some
> board, device, phone you have at your disposal (RaspberryPi2 is a
> great inexpensive kit), and make it work. You'll take a lot less
> effort than you're imagining...
> LLVM Developers mailing list
> LLVMdev at cs.uiuc.edu http://llvm.cs.uiuc.edu
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the llvm-dev