transfer /opt/local to another machine
René J.V. Bertin
rjvbertin at gmail.com
Mon Apr 13 03:06:06 PDT 2015
On Monday April 13 2015 11:43:51 j. van den hoff wrote:
>> If the core-i7 build is using CPU instructions your core 2 due does not
>> have, then the performance hit will be 'significant' (i.e. it simply
>> will not run...).
>OK, _that_ I would note and accordingly would recompile ;-). I'm not
>trying to avoid that at all costs, anyway. being able to transfer an
And it's more than likely that it will happen. It would surprise me somewhat if Atlas doesn't check for `-march=native` support, and that would give you AVX (and later) support if your i7 supports that.
You ought to ask the question the other way round: how much of a (real-world) performance hit would you see when running Atlas built on a C2Duo on an i7.
BTW: I do feel that "base" could add -march=native in user installations (with an optional override and not on the build bots) since AFAIU locally built packages are not intended to be distributed. The only times I've seen this flag cause issues is when running in a VM hosted on an i7, where the VM doesn't support the AVX instruction set and the compiler presumes it to be available.
More information about the macports-users