From: Geoffrey Keating <geoffk@apple.com>
To: gcc List <gcc@gcc.gnu.org>
Subject: attribute data structure rewrite
Date: Fri, 24 Sep 2004 00:15:00 -0000 [thread overview]
Message-ID: <FA388EC7-0DBB-11D9-8B78-000A95B1F520@apple.com> (raw)
[-- Attachment #1: Type: text/plain, Size: 4178 bytes --]
As a precondition for other compile speed improvements, I've been
rewriting the way that GCC represents attributes. The result is a
medium-size patch that touches every language and many targets (a
surprising number of targets don't define any special attributes at
all).
The basic principle is that DECL_ATTRIBUTES and TREE_ATTRIBUTES now
point to an attribute_list:
/* A structure representing 'attributes' on a DECL or TYPE node.
Each attribute has a NAME (an IDENTIFIER_NODE) and possibly a VALUE.
*/
struct one_attribute GTY(())
{
tree name;
tree value;
};
/* A counted list of attributes. */
struct attribute_list_s GTY(())
{
attribute_count n_attributes;
/* There are 16 bits free here. */
struct one_attribute GTY((length ("%h.n_attributes"))) attribs[1];
};
This has many benefits, mostly flowing from the fact that it's not a
TREE_LIST any more.
1. Do we think this would be acceptable for stage 3, or should I make a
branch?
2. Would anyone be interested in helping me test it, especially on the
more difficult architectures like alpha-vms or i386-netware?
diffstat on my current patch looks like:
ada/gigi.h | 2
ada/utils.c | 2
attribs.c | 347
++++++++++++++++++++++++++-------------------
builtin-attrs.def | 62 +++-----
c-common.c | 116 +++++++++------
c-common.h | 4
c-decl.c | 76 +++++----
c-format.c | 25 +--
c-objc-common.c | 4
c-parse.in | 98 ++++++------
c-pragma.c | 10 -
c-tree.h | 28 +--
c-typeck.c | 12 -
cgraphunit.c | 2
config/alpha/alpha.c | 6
config/arm/arm.c | 37 ++--
config/arm/pe.c | 12 -
config/avr/avr.c | 14 -
config/c4x/c4x.c | 6
config/frv/frv.c | 2
config/h8300/h8300.c | 35 ----
config/i386/i386.c | 42 ++---
config/i386/netware.c | 8 -
config/i386/winnt.c | 22 +-
config/ia64/ia64.c | 10 -
config/ip2k/ip2k.c | 5
config/m32r/m32r.c | 10 -
config/m68hc11/m68hc11.c | 28 +--
config/m68k/m68k.c | 5
config/mcore/mcore.c | 6
config/ns32k/ns32k.c | 4
config/rs6000/rs6000.c | 22 +-
config/sh/sh.c | 10 -
config/sh/symbian.c | 28 +--
config/sol2.c | 8 -
config/stormy16/stormy16.c | 6
config/v850/v850.c | 20 --
coretypes.h | 6
cp/call.c | 2
cp/class.c | 2
cp/cp-tree.h | 29 ++-
cp/decl.c | 59 +++----
cp/decl.h | 2
cp/decl2.c | 14 -
cp/friend.c | 2
cp/method.c | 4
cp/name-lookup.c | 10 -
cp/name-lookup.h | 2
cp/optimize.c | 2
cp/parser.c | 176 ++++++++++++----------
cp/pt.c | 2
cp/tree.c | 4
cp/typeck.c | 8 -
fold-const.c | 4
gengtype-lex.l | 6
integrate.c | 9 -
java/decl.c | 46 ++---
java/java-tree.h | 2
java/jcf-reader.c | 12 -
langhooks.c | 4
langhooks.h | 2
objc/objc-act.c | 90 ++++++-----
passes.c | 5
print-tree.c | 23 ++
system.h | 2
target-def.h | 4
target.h | 10 -
targhooks.c | 10 +
targhooks.h | 4
tree-browser.c | 9 -
tree-inline.c | 6
tree.c | 307
+++++++++++++++++++++++++--------------
tree.h | 81 ++++++++--
treelang/treetree.c | 6
varasm.c | 6
75 files changed, 1171 insertions, 935 deletions
[-- Attachment #2: smime.p7s --]
[-- Type: application/pkcs7-signature, Size: 2408 bytes --]
next reply other threads:[~2004-09-23 23:55 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-09-24 0:15 Geoffrey Keating [this message]
2004-09-24 0:34 ` Andrew Pinski
2004-09-24 1:30 ` Joseph S. Myers
2004-09-24 1:47 ` Zack Weinberg
2004-09-24 6:52 ` Geoffrey Keating
2004-09-24 1:40 ` Gabriel Dos Reis
2004-09-24 2:34 ` Giovanni Bajo
2004-09-24 2:59 ` Gabriel Dos Reis
2004-09-24 4:17 ` Giovanni Bajo
2004-09-24 4:18 ` Gabriel Dos Reis
2004-09-24 8:40 ` Andreas Schwab
2004-09-24 8:40 ` Andreas Schwab
2004-09-24 9:17 ` Joseph S. Myers
2004-09-24 14:45 ` Giovanni Bajo
2004-09-24 8:36 ` Geoffrey Keating
2004-09-24 14:03 ` Joseph S. Myers
2004-09-24 2:35 ` Giovanni Bajo
2004-09-24 8:33 ` Geoffrey Keating
2004-09-24 12:24 ` Nathan Sidwell
2004-09-24 22:29 ` Mark Mitchell
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=FA388EC7-0DBB-11D9-8B78-000A95B1F520@apple.com \
--to=geoffk@apple.com \
--cc=gcc@gcc.gnu.org \
/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).