From: Jeff Law <law@redhat.com>
To: Ian Lance Taylor <iant@google.com>
Cc: David Malcolm <dmalcolm@redhat.com>,
gcc-patches <gcc-patches@gcc.gnu.org>,
Andrew MacLeod <amacleod@redhat.com>
Subject: Re: [PATCH 0/6] Conversion of gimple types to C++ inheritance (v3)
Date: Wed, 06 Nov 2013 20:49:00 -0000 [thread overview]
Message-ID: <527AA8BF.8000301@redhat.com> (raw)
In-Reply-To: <CAKOQZ8wq7HGJ2k11N-+qooiXshwXJGiMNSAd4pY_0LktBDrX_A@mail.gmail.com>
On 11/05/13 16:23, Ian Lance Taylor wrote:
> On Tue, Nov 5, 2013 at 2:12 PM, Jeff Law <law@redhat.com> wrote:
>>
>> I can't speak for Andrew, but my experience with this kind of object type
>> casting in a large C++ project is that it's a red flag for a design problem.
>
> I'm going to chime in to say that I think it's a pretty typical way to
> represent a compiler IR in C++. There is a base type that a lot of
> code uses, but there is also a real type, and the way to get to that
> real type is to use a cast. We could do it all with virtual
> functions, but those carry a different cost. In effect, using virtual
> functions increases the size of the code field from 16 bits to 64
> bits. It adds up.
>
> Also this seems to be a pretty direct version of the data structures
> we already have.
Certainly and to some degree these casts in one form or another are a
necessity. My concern is we use them as a crutch (as I've seen in other
projects) and we end up not getting the level of compile time static
type checking that we want.
If the bulk are temporary in nature and we as a project agree to look at
the refactoring necessary to minimize the is-a/as-a uses to a few low
level idioms, then I'll hesitatingly go along. If that's the choice we
make, then folks should know I'll probably look at each is-a/as-a in
patches I review and ask for a justification for their use.
jeff
next prev parent reply other threads:[~2013-11-06 20:39 UTC|newest]
Thread overview: 116+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-10-31 5:45 RFC: gimple.[ch] break apart Andrew MacLeod
2013-10-31 6:15 ` Jeff Law
2013-10-31 16:41 ` [PATCH 0/6] Conversion of gimple types to C++ inheritance (v3) David Malcolm
2013-10-31 16:27 ` [PATCH 1/6] Convert gimple types from a union to C++ inheritance David Malcolm
2013-11-14 23:00 ` Jeff Law
2013-11-19 0:22 ` David Malcolm
2013-11-19 8:49 ` Jeff Law
2013-10-31 16:27 ` [PATCH 4/6] Implement is_a_helper <>::test specializations for various gimple types David Malcolm
2013-11-14 9:41 ` Jeff Law
2013-11-18 19:51 ` David Malcolm
2013-11-18 20:10 ` Andrew MacLeod
2013-11-18 20:15 ` Jeff Law
2013-10-31 16:27 ` [PATCH 2/6] Hand-written port of various accessors within gimple.h David Malcolm
2013-11-14 9:53 ` Jeff Law
2013-10-31 16:31 ` [PATCH 3/6] Automated part of conversion of gimple types to use C++ inheritance David Malcolm
2013-11-14 9:43 ` Jeff Law
2013-11-18 22:17 ` [PATCH] Updated automated patch (was Re: [PATCH 3/6] Automated part of conversion of gimple types to use C++ inheritance) David Malcolm
2013-11-19 8:49 ` Jeff Law
2013-11-19 16:36 ` David Malcolm
2013-11-22 0:27 ` Jakub Jelinek
2013-11-22 0:35 ` Jeff Law
2013-11-22 1:51 ` Jakub Jelinek
2013-11-22 2:52 ` Andrew MacLeod
2013-11-22 3:48 ` David Malcolm
2013-11-25 18:09 ` [PATCH] Fix checking of gimple types David Malcolm
2013-11-25 18:42 ` Michael Matz
2013-11-25 22:18 ` Jeff Law
2013-11-27 8:54 ` David Malcolm
2014-07-23 13:16 ` Thomas Schwinge
2014-07-24 1:50 ` David Malcolm
2014-07-29 8:11 ` Thomas Schwinge
2013-10-31 16:46 ` [PATCH 6/6] Update gdb hooks to reflect changes to " David Malcolm
2013-11-14 9:10 ` Jeff Law
2013-10-31 16:49 ` [PATCH 5/6] Port various places from union access to subclass access David Malcolm
2013-11-14 9:23 ` Jeff Law
2013-11-19 0:52 ` David Malcolm
2013-10-31 18:43 ` [PATCH 0/6] Conversion of gimple types to C++ inheritance (v3) Basile Starynkevitch
2013-11-01 21:36 ` Andrew MacLeod
2013-11-01 21:41 ` Jakub Jelinek
2013-11-01 21:47 ` Andrew MacLeod
2013-11-01 21:57 ` Jakub Jelinek
2013-11-01 22:58 ` David Malcolm
2013-11-04 13:23 ` Andrew MacLeod
2013-11-04 21:52 ` David Malcolm
2013-11-04 22:09 ` David Malcolm
2013-11-04 22:31 ` Andrew MacLeod
2013-11-05 21:27 ` Jeff Law
2013-11-04 22:43 ` Andrew MacLeod
2013-11-04 22:28 ` Jakub Jelinek
2013-11-04 22:49 ` Andrew MacLeod
2013-11-05 21:09 ` Jeff Law
2013-11-05 11:53 ` Richard Biener
2013-11-05 12:33 ` David Malcolm
2013-11-05 12:52 ` Richard Biener
2013-11-04 14:00 ` Andrew MacLeod
2013-11-04 14:01 ` Jakub Jelinek
2013-11-04 14:15 ` Andrew MacLeod
2013-11-04 18:23 ` Jeff Law
2013-11-01 22:43 ` David Malcolm
2013-11-01 23:43 ` Trevor Saunders
2013-11-04 13:15 ` Andrew MacLeod
2013-11-05 17:23 ` [PATCH] Add gimple subclasses for every gimple code (was Re: [PATCH 0/6] Conversion of gimple types to C++ inheritance (v3)) David Malcolm
2013-11-06 16:53 ` Michael Matz
2013-11-07 6:19 ` David Malcolm
2013-11-07 7:08 ` Jeff Law
2013-11-08 19:23 ` David Malcolm
2013-11-14 8:38 ` Jeff Law
2013-11-14 15:06 ` Michael Matz
2013-11-14 18:32 ` David Malcolm
2013-11-15 2:49 ` Jeff Law
2013-11-07 14:57 ` Michael Matz
2013-11-08 0:07 ` Alec Teal
2013-11-08 14:31 ` Michael Matz
2013-11-05 18:22 ` [PATCH 0/6] Conversion of gimple types to C++ inheritance (v3) Andrew MacLeod
2013-11-05 21:33 ` Jeff Law
2013-11-05 22:01 ` David Malcolm
2013-11-05 22:17 ` Jeff Law
2013-11-06 1:14 ` Ian Lance Taylor
2013-11-06 20:49 ` Jeff Law [this message]
2013-11-06 20:57 ` Trevor Saunders
2013-11-05 22:24 ` Andrew MacLeod
2013-11-05 22:12 ` Andrew MacLeod
2013-11-06 9:37 ` Richard Biener
2013-11-06 11:20 ` Bernd Schmidt
2013-11-06 11:43 ` Richard Biener
2013-11-06 11:53 ` Jakub Jelinek
2013-11-06 13:14 ` Richard Biener
2013-11-06 13:23 ` Jakub Jelinek
2013-11-06 16:42 ` David Malcolm
2013-11-06 16:55 ` Jakub Jelinek
2013-11-06 18:34 ` Tom Tromey
2013-11-06 19:15 ` Jeff Law
2013-11-06 20:05 ` Tom Tromey
2013-11-06 20:45 ` Jeff Law
2013-11-06 13:31 ` Joseph S. Myers
2013-11-06 21:25 ` Jeff Law
2013-11-06 21:09 ` Jeff Law
2013-11-06 12:42 ` Bernd Schmidt
2013-11-06 21:04 ` Jeff Law
2013-11-06 21:06 ` Andrew MacLeod
2013-11-06 21:52 ` Jeff Law
2013-11-07 10:29 ` Richard Biener
2013-11-07 14:01 ` Joseph S. Myers
2013-11-07 14:42 ` Richard Biener
2013-11-07 14:53 ` Andrew MacLeod
2013-11-10 12:35 ` Richard Sandiford
2013-11-10 15:27 ` Richard Biener
2013-11-06 11:56 ` Eric Botcazou
2013-11-06 20:51 ` Jeff Law
2013-11-06 21:26 ` Jeff Law
2013-11-14 8:40 ` Jeff Law
2013-11-05 16:58 ` [patch] Create gimple-expr..[ch] ... was Re: RFC: gimple.[ch] break apart Andrew MacLeod
2013-11-05 17:52 ` Jeff Law
2013-11-07 10:58 ` Basile Starynkevitch
2013-11-07 13:47 ` Andrew MacLeod
2013-11-07 18:13 ` Diego Novillo
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=527AA8BF.8000301@redhat.com \
--to=law@redhat.com \
--cc=amacleod@redhat.com \
--cc=dmalcolm@redhat.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=iant@google.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 read-only IMAP folder(s) and NNTP newsgroup(s).