From kde-devel  Fri Oct 29 15:03:04 2004
From: Waldo Bastian <bastian () kde ! org>
Date: Fri, 29 Oct 2004 15:03:04 +0000
To: kde-devel
Subject: Re: How many processes?
Message-Id: <200410291703.08952.bastian () kde ! org>
X-MARC-Message: https://marc.info/?l=kde-devel&m=109906226203622
MIME-Version: 1
Content-Type: multipart/mixed; boundary="--===============2106272734=="

--===============2106272734==
Content-Type: multipart/signed;
  boundary="nextPart3143431.6JhtT8jKsp";
  protocol="application/pgp-signature";
  micalg=pgp-sha1
Content-Transfer-Encoding: 7bit

--nextPart3143431.6JhtT8jKsp
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

> Wouldn't it be a better idea to have multiple processes but the majority =
of
> code in a shared library?  That way you are conserving memory,  but if one
> crashes the rest are unaffected.

The memory used by the code itself is already shared between multiple=20
instances of the same executable. Whether code is part of a shared library =
or=20
part of an executable makes no difference in that regard.

However the non-shared data used by the average KDE process is substantial =
(in=20
the order of 1Mb =B9) and by having multiple processes this data gets=20
duplicated in ram.=20

Cheers,
Waldo

=B9) For a significant part, but not only, due to XIM and Xft.
=2D-=20
bastian@kde.org   |   SUSE LINUX 9.2: Order now!   |   bastian@suse.com
  http://www.suse.de/us/private/products/suse_linux/preview/index.html

--nextPart3143431.6JhtT8jKsp
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (GNU/Linux)

iD8DBQBBglusN4pvrENfboIRAlnWAJ964/akOT6OCd4pJSMVsi1fY2vH4gCgpWhr
U6EyIMk+T9AI/DaqutC81JA=
=rRdL
-----END PGP SIGNATURE-----

--nextPart3143431.6JhtT8jKsp--

--===============2106272734==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

 
>> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<

--===============2106272734==--