The system I work with generates code and then compiles it into a shared library
to execute. Currently we choose to set -march=core2 which is easy enough but
quite silly because we don’t get any machine specific optimizations that could get
better execution times.
I can’t google how to find the appropriate arch type for the compiler. If I could do
that then I could check the hardware (somehow) and match it to the proper -march
tag, and have the resulting compile line include the target hardware thus generating
better performant code.
Anyone know how that could be done?
David
David Barto
barto@cambridgesemantics.com
Sometimes, my best code does nothing. Most of the rest of it has bugs.