Date: Sun, 25 Mar 2007 11:43:25 +0000 From: Geoff Buckingham <geoffb@chuggalug.clues.com> To: Alexander Leidinger <Alexander@leidinger.net> Cc: multimedia@freebsd.org Subject: Re: Updated mythtv-fixes port Message-ID: <20070325114325.GD46175@chuggalug.clues.com> In-Reply-To: <20070324185725.91d9auipu88gs8w4@webmail.leidinger.net> References: <20070323181558.GA46175@chuggalug.clues.com> <20070324145558.2e0f462b@Magellan.Leidinger.net> <20070324151536.GB46175@chuggalug.clues.com> <20070324185725.91d9auipu88gs8w4@webmail.leidinger.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, Mar 24, 2007 at 06:57:25PM +0100, Alexander Leidinger wrote: > Quoting Geoff Buckingham <geoffb@chuggalug.clues.com> (from Sat, 24 > Mar 2007 15:15:36 +0000): > > >The CPU detection only gets used if the mythtv configure isn't given the > >relevent arguments. It should be possible to get the port's makefile > >to add these if required. > > Note: The port should not assume any specific processor in the > PACAKGE_BUILDING case (maybe i486 or better). And I'm not sure if it > should do the CPU detection by default in all other cases. To be on > the safe side, make it an option. > While I agree the PACAKAGE_BUILDING case must produce a genuinly portable binary. I can not see the point in deliberatly making it difficult to get equivalent performance to the various Linux distros. Realisticly most folk tring to run myth are going to have i686 or better with MMX.(Having a working via unichrome kernel module may change this) My incliniation is the port should self tune, unless PACKAGE_BUILDING or there a optimisation flags from make.conf.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20070325114325.GD46175>