[MLton-devel] icfp?

[Stephen Weeks]

> Sounds like a good idea.  My only reservation on using the profiling info
> is that the code generated when profiling is on can be quite different than
> the code generated when profiling is off. 

On the contrary, the code is almost identical, due to the profiling C
calls being inserted in the very last phase of the backend.  I agree
that the effect on time can be large.  However, the time effect can
also be measured, since one can compare to a run without profiling.
Also, note that the effect on allocation can also be precisely
measured, and is typically very small.


-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
_______________________________________________
MLton-devel mailing list
MLton-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mlton-devel