public inbox for gcc@gcc.gnu.org
 help / color / mirror / Atom feed
From: Jim Wilson <wilson@cygnus.com>
To: Peter Barada <pbarada@wavemark.com>
Cc: mrs@wrs.com, egcs@cygnus.com
Subject: Re: m68k structure packing
Date: Thu, 02 Oct 1997 20:09:00 -0000	[thread overview]
Message-ID: <199710030309.UAA16166@cygnus.com> (raw)
In-Reply-To: <199710012255.SAA11990@miacomet.wavemark.com>

	If you pack the structure on any hardware with strict alignment then
	you have to access any non-byte element using ONLY byte accesses since
	you can not assume the alignment of the address of the start of the structure.
	Unfortunately on the 68k this produces some really slow code. Ex: 

The solution is simple.  If you care about performance, then don't pack your
structures.

	If you want packed structures then they should be sized up to a short, and
	anything that is not a char in a packed struct should be aligned on a
	short boundary(BIGGEST_ALIGNMENT). But this is what happens on the 68k
	anyway.

It makes no sense to disable structure packing because of performance
concerns.  People who are packing structures are doing it because they
actually need a particular structure layout.  If the programmer requests
packed structures, then the compiler should obey, and the compiler should
make structure packing work the same way for all targets.  Mike Stump's
patch is a step in this direction.

Jim

  parent reply	other threads:[~1997-10-02 20:09 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
1997-10-01 15:08 Mike Stump
1997-10-01 15:56 ` Peter Barada
1997-10-01 16:16   ` Per Bothner
1997-10-02 20:14     ` Jim Wilson
1997-10-02  6:49   ` Paul Koning
1997-10-02 20:09   ` Jim Wilson [this message]
1997-10-02 20:01 ` Jim Wilson
1997-10-02 21:40   ` Richard Henderson
  -- strict thread matches above, loose matches on Subject: below --
1997-10-01 12:16 Mike Stump
1997-10-01 12:39 ` Joel Sherrill
1997-10-02 18:37 ` Jim Wilson
1997-09-30 21:42 Jim Wilson
1997-10-01  5:44 ` Kamil Iskra
1997-09-30 20:16 Mike Stump
1997-09-30 22:03 ` Jeffrey A Law
1997-09-30 16:58 Mike Stump
1997-09-30 18:20 ` Jim Wilson
1997-10-01  9:02   ` Peter Barada
1997-09-30 13:32 Mike Stump
1997-09-30 14:56 ` Jim Wilson
1997-09-30 12:08 Mike Stump
1997-09-30 12:57 ` Charles M. Hannum
1997-09-30 19:58 ` Jim Wilson
1997-09-30 21:13   ` Richard Henderson
1997-09-30 21:22   ` Jim Wilson
1997-10-01 15:14 ` Jim Wilson

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=199710030309.UAA16166@cygnus.com \
    --to=wilson@cygnus.com \
    --cc=egcs@cygnus.com \
    --cc=mrs@wrs.com \
    --cc=pbarada@wavemark.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).