caml-list - the Caml user's mailing list
 help / color / mirror / Atom feed
From: whitequark <whitequark@whitequark.org>
To: Raoul Duke <raould@gmail.com>
Cc: OCaml <caml-list@inria.fr>
Subject: Re: [Caml-list] whither portability?
Date: Sun, 27 Sep 2015 17:33:00 +0300	[thread overview]
Message-ID: <1acefe77c30ce33c0adba49605bd9325@whitequark.org> (raw)
In-Reply-To: <CAJ7XQb5Xmd0TW_91_8gqeHptFPAuaYpeNCuP4piZx-rsQ2X9yA@mail.gmail.com>

On 2015-09-27 03:55, Raoul Duke wrote:
>> Both ocamlc and ocamlopt work for iOS and Android; there is not really
>> a good reason to use the bytecode compiler for deployment to mobile.
>> 
>> There is no such thing as "LLVM bytecode" and LLVM bitcode (sic)
>> is merely an intermediate format for native executable code. Shipping
>> LLVM bitcode offers no conceptual improvement over shipping
>> machine code and in any case, there is neither an LLVM backend
>> for ocamlopt nor much need for such a backend.
> 
> Yes, "bitcode" is the right term. It is now the format Apple wants. So
> I dunno why people would actively discount it here.

Right now, LLVM IR (of which the bitcode is a serialization) is not
flexible enough to express the invariants and metadata required
for the OCaml GC to function[1]. Even if someone released the necessary,
highly nontrivial changes to both LLVM and OCaml today, it would take
many months for them to be reviewed, merged & propagated into Apple's
LLVM fork--all for a small improvement in a select few numeric workloads
that can benefit from using the instruction selector tailored for
the particular device, and a substantial reduction in being able to
debug your code[2].

[1]: 
https://github.com/whitequark/ocaml-llvm-ng/issues/1#issuecomment-141844579
[2]: 
https://medium.com/@FredericJacobs/why-i-m-not-enabling-bitcode-f35cd8fbfcc5

-- 
whitequark

  reply	other threads:[~2015-09-27 14:33 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-09-25 21:13 Raoul Duke
2015-09-26  9:28 ` Gerd Stolpmann
2015-09-26 12:48   ` Raoul Duke
2015-09-26 15:59     ` whitequark
2015-09-26 16:21       ` Gerd Stolpmann
2015-09-26 23:29       ` Raoul Duke
2015-09-27  0:47         ` Spiros Eliopoulos
2015-09-27  0:51           ` Raoul Duke
2015-09-27  0:20       ` Oliver Bandel
2015-09-27  0:27         ` Yotam Barnoy
2015-09-27 14:10           ` Oliver Bandel
2015-09-27  0:55       ` Raoul Duke
2015-09-27 14:33         ` whitequark [this message]
2015-09-27 17:19           ` Raoul Duke
2015-09-27 17:41             ` whitequark
2015-09-26 17:10     ` Gerd Stolpmann

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1acefe77c30ce33c0adba49605bd9325@whitequark.org \
    --to=whitequark@whitequark.org \
    --cc=caml-list@inria.fr \
    --cc=raould@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).