Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 22 Jul 2003 09:05:11 +1000
From:      Peter Jeremy <peter.jeremy@alcatel.com.au>
To:        ticso@cicely.de
Cc:        freebsd-alpha@freebsd.org
Subject:   Re: Problem building Mozilla
Message-ID:  <20030721230511.GN430@gsmx07.alcatel.com.au>
In-Reply-To: <20030721105042.GD19386@cicely12.cicely.de>
References:  <20030716215658.GA43409@gsmx07.alcatel.com.au> <20030717132611.GJ14530@cicely12.cicely.de> <20030718031240.GF430@gsmx07.alcatel.com.au> <20030718094556.GE19386@cicely12.cicely.de> <20030721032508.GH430@gsmx07.alcatel.com.au> <20030721105042.GD19386@cicely12.cicely.de>

next in thread | previous in thread | raw e-mail | index | archive | help
On 2003-Jul-21 12:50:43 +0200, Bernd Walter <ticso@cicely12.cicely.de> wrote:
>On Mon, Jul 21, 2003 at 01:25:08PM +1000, Peter Jeremy wrote:
>> On 2003-Jul-18 11:45:56 +0200, Bernd Walter <ticso@cicely12.cicely.de> wrote:
>> >I compiled myself with:
>> >CFLAGS=-O2 -pipe
>> >CPUTYPE?=ev56
>> 
>> I'll try that and see what happens.  Did you specify 'WITH_OPTIMIZED_CFLAGS'
>> or any other tunables?

CFLAGS=-O2 -pipe -g
CPUTYPE?=ev56

helped - it now spits out reams of unaligned accesses and a large
number of "junk pointer in free() - too high" messages and finally
dies with a segmentation violation in FcCharSetDestroy() trying to
load a long from an address 4 bytes below the end of a segment.

It looks like the situation with gcc has changed so that -O2 is
generating more correct code than -O (at least for mozilla).

>The only special flag beside compiler flags is the following:
>WITHOUT_XFT=yes

I'll try that.

>Well looks like the current port needs maintenance anyway, so I will
>have to try it myself.

Thanks for that.  1.3.1 seems fairly fragile and I can't get 1.4 to
compile at all.  

Peter



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030721230511.GN430>