Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 4 May 2005 21:02:44 +0200
From:      Marc Olzheim <marcolz@stack.nl>
To:        Daniel Eischen <deischen@freebsd.org>
Cc:        freebsd-threads@freebsd.org
Subject:   Re: Floating point exceptions with -pthread
Message-ID:  <20050504190244.GB76171@stack.nl>
In-Reply-To: <20050504183132.GA76171@stack.nl>
References:  <20050504103144.GA74093@stack.nl> <Pine.GSO.4.43.0505041057040.28975-100000@sea.ntplx.net> <20050504183132.GA76171@stack.nl>

next in thread | previous in thread | raw e-mail | index | archive | help

--b5gNqxB1S1yM7hjW
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline

On Wed, May 04, 2005 at 08:31:32PM +0200, Marc Olzheim wrote:
> It seems to me like something about the floorf() assembly might have
> something to do with it.

One of the reasons I think it has to do with specific functions, is that
we've got one system that runs heavily threaded floating point heavy
code (a search engine ranker) and that hasn't had this problem in the
past seven days. The database builder is the only piece of software
giving us trouble, which happens the only software we have that uses
floorf().

So after some testing, I could remove all code but the floorf() and
still get the exception, but without it, I'm until now unsuccesful.

Marc

--b5gNqxB1S1yM7hjW
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (FreeBSD)

iD8DBQFCeRxUezjnobFOgrERAh+rAJ9kJLbpsWkKIJntjeJHbvFGRyFsRgCglm1B
KS/BbYLUjCLyKT6G6CkZi0k=
=Jltc
-----END PGP SIGNATURE-----

--b5gNqxB1S1yM7hjW--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050504190244.GB76171>