From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-return-156580-listarch-gcc=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 6618 invoked by alias); 16 Sep 2009 08:28:40 -0000
Received: (qmail 6603 invoked by uid 22791); 16 Sep 2009 08:28:38 -0000
X-SWARE-Spam-Status: No, hits=-1.8 required=5.0 	tests=AWL,BAYES_00,SARE_MSGID_LONG40,SPF_PASS
X-Spam-Check-By: sourceware.org
Received: from mail-vw0-f174.google.com (HELO mail-vw0-f174.google.com) (209.85.212.174)     by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Wed, 16 Sep 2009 08:28:31 +0000
Received: by vws4 with SMTP id 4so1714964vws.0         for <gcc@gcc.gnu.org>; Wed, 16 Sep 2009 01:28:30 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.220.88.100 with SMTP id z36mr12054067vcl.114.1253089709907;  	Wed, 16 Sep 2009 01:28:29 -0700 (PDT)
In-Reply-To: <1253054340.7094.70.camel@tidbit.site>
References: <1253054340.7094.70.camel@tidbit.site>
Date: Wed, 16 Sep 2009 08:28:00 -0000
Message-ID: <84fc9c000909160128h5b2ea7bfpd6186fea488c6ee0@mail.gmail.com>
Subject: Re: Types with power-of-two size in ggc-page.c extra_order_size_table[]
From: Richard Guenther <richard.guenther@gmail.com>
To: okellogg@users.sourceforge.net
Cc: gcc@gcc.gnu.org
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
X-IsSubscribed: yes
Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc/>
List-Post: <mailto:gcc@gcc.gnu.org>
List-Help: <http://gcc.gnu.org/ml/>
Sender: gcc-owner@gcc.gnu.org
X-SW-Source: 2009-09/txt/msg00289.txt.bz2

On Wed, Sep 16, 2009 at 12:39 AM, Oliver Kellogg
<oliver.kellogg@t-online.de> wrote:
> Hi,
>
> Looking at ggc-page.c:175ff.,
>
> static const size_t extra_order_size_table[] =3D {
> =A0sizeof (struct var_ann_d),
> =A0sizeof (struct tree_decl_non_common),
> =A0sizeof (struct tree_field_decl),
> =A0sizeof (struct tree_parm_decl),
> =A0sizeof (struct tree_var_decl),
> =A0sizeof (struct tree_list),
> =A0sizeof (struct tree_ssa_name),
> =A0sizeof (struct function),
> =A0sizeof (struct basic_block_def),
> =A0sizeof (bitmap_element),
> =A0sizeof (bitmap_head),
> =A0TREE_EXP_SIZE (2),
> =A0RTL_SIZE (2), =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 /* MEM, PLUS, etc. =A0*/
> =A0RTL_SIZE (9), =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 /* INSN */
> };
>
> On my openSuSE-11.1 i686-pc-linux-gnu, sizeof (struct basic_block_def)
> is 64 and sizeof (bitmap_head) is 16.
> Why are these types entered into extra_order_size_table?

They likely changed in size and/or the 64bit host-wide-int variants are not
power-of-two sized.

>
> ggc-page.c (init_ggc):1567ff.
> =A0for (order =3D HOST_BITS_PER_PTR; order < NUM_ORDERS; ++order)
> =A0 =A0{
> =A0 =A0 =A0int o;
> =A0 =A0 =A0int i;
>
> =A0 =A0 =A0i =3D OBJECT_SIZE (order);
> =A0 =A0 =A0if (i >=3D NUM_SIZE_LOOKUP)
> =A0 =A0 =A0 =A0continue;
>
> =A0 =A0 =A0for (o =3D size_lookup[i]; o =3D=3D size_lookup [i]; --i)
> =A0 =A0 =A0 =A0size_lookup[i] =3D order;
> =A0 =A0}
>
> At order 42, OBJECT_SIZE returns 16 (from extra_order_size_table[10])
> and size_lookup[16] down to [9] is changed to 42.
> That seems strange to me, given the documentation of size_lookup[],

I don't remember exactly but the last time I spent with the code it did
the right thing.  Giving more context might help to understand the problem
you are seeing.

Richard.

> /* This table provides a fast way to determine ceil(log_2(size)) for
> =A0 allocation requests. =A0The minimum allocation size is eight bytes. =
=A0*/
>
>
> Thanks,
>
> Oliver
>
>
>
>
>