[mythtv-users] Installing mythtv.25 from macports fails

Craig Treleaven ctreleaven at cogeco.ca
Sat Apr 20 15:13:16 UTC 2013


At 10:40 PM +1000 4/20/13, Jean-Yves Avenard wrote:
>On 20 April 2013 21:57, Craig Treleaven 
><<mailto:ctreleaven at cogeco.ca>ctreleaven at cogeco.ca> wrote:
>At 8:22 AM +0200 4/20/13, Ton Machielsen wrote:
>[...]Significant errors in 
>/opt/local/var/macports/logs/_opt_local_var_macports_sources_rsync.macports.org_release_tarballs_ports_multimedia_mythtv-core.25/mythtv-core.25/main.log
>[...]:info:build filter_yadif.c:372: error: can't find a register in 
>class <E2><80><98>GENERAL_REGS<E2><80><99> while reloading 
><E2><80><98>asm
><E2><80><99> [...]
>As I replied on the MacPorts-users mailing list, this appears to be 
>a CoreSolo or CoreDuo machine?  There is an unresolved problem 
>building Myth on 32-bit machines.  These machines are also woefully 
>underpowered for running Myth so I haven't spent a lot of time 
>trying to track down the problem.
>i would disable the yadif  filter when compiling on 32 bits machine 
>or get the filter source code from current master/0.26 ; the 
>makefile fixes some compilation issues with the filters.
>
>Or alternatively, use a pre-built frontend...

J-Y, to disable building yadif, is it as simple as adding 
'--disable-filter=yadif' to myth's configure?  Or is anything else 
required?

As far as frontend v. backend, a CoreDuo or CoreSolo Mac would really 
only suffice as a standard def frontend.  I suppose it could be a 
backend-only machine for HD as long as it wasn't doing any 
commflagging or transcoding...the advantage in this case is that it 
has pretty good power-management built-in.  The OP hasn't said what 
he was hoping to use it for.

Craig


More information about the mythtv-users mailing list