[PATCH] [DOC] Mapping High-Level Language Constructs to LLVM IR

[David Blaikie]

On Tue, Dec 17, 2013 at 12:36 PM, Mikael Lyngvig <mikael at lyngvig.org> wrote:

> I agree it is not suited for publication.  I'll be happy to spend some
> hours trying to make it ready to go live.  But I cannot document most of
> the things in such a short time.
>

Sorry for the confusion - I was trying to suggest a low-investment way to
get this to a publishable state. I didn't mean to suggest you should just
'finish everything'.


>  What I can do is to move the to-do's to an appendix at the end of the
> document and try to clean up the existing things.
>

Something like that would make sense to me - a "coming soon" section,
perhaps. (or it could be kept in non-user-visible comments inside the doc,
or in a bug/bugs in the llvm.org/bugs database)


>  I'd hate to loose the to-do list as it is a very important indication of
> what direction the document is heading - and also a great way to ensure
> that all the great suggestions that people have made actually make it into
> the final document.
>
> I am open to any and all suggestions as to how to do this, though.
>
>
> -- Mikael
>
>
> 2013/12/17 David Blaikie <dblaikie at gmail.com>
>
>> Would it be helpful to find more incremental forms of this documentation
>> so we could make it 'live' (linked from places, indexed, etc) and improve
>> it from there - rather than having an incomplete/unusable (I get the
>> impression it's unusable in its current state - is that correct?) document?
>>
>>
>> On Tue, Dec 17, 2013 at 10:33 AM, Mikael Lyngvig <mikael at lyngvig.org>wrote:
>>
>>> Hi guys,
>>>
>>> I am a bit short on time for the time being so I thought it would be
>>> better to move the quite incomplete mapping document into the LLVM docs
>>> folder so that others are more likely to join in the fun of documenting
>>> every detail of LLVM IR from the opposite direction of what the existing
>>> documentation does (the existing documentation is very developer-centric
>>> whereas the Mapping document is very user-centric).  Also, I had expected
>>> many more to join in than actually did join in so I figure it is better to
>>> store the document with the rest of the LLVM documentation.
>>>
>>> I am by no means done working on this document, I just want to make it
>>> easier for others to contribute.  And I'd like to encourage everybody to
>>> add their tips, tricks, observations, experiences, and so forth so that we
>>> get a gigantic document that will help all newbies a lot.
>>>
>>> The end goal is to eliminate any and all TODO's in the document as well
>>> as the first chapter of to-do items.  I must also confess that we are
>>> reaching the limits of what I currently know about compilers and LLVM IR.
>>>  I figure somebody with actual hands-on experience of how to implement
>>> exception handing using the Itanium ABI can document this in a matter of
>>> hours where it would take me weeks because I'd have to learn it all from
>>> scratch, experiment with it, and then finally document my findings.
>>>
>>>
>>> -- Mikael
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> llvm-commits mailing list
>>> llvm-commits at cs.uiuc.edu
>>> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20131217/907d18e8/attachment.html>