Does anyone know of a gcc compiler flag or option that would enable us to
compile and successfully run 'C' code that will access mis-aligned integers
or dioubles which runs on a SPARC workstation running Solaris.
A trap handler that does the unaligned load might be best, gcc generally doesn't support doing this without typing information, and even when that is used, you cannot violate the type system and expect it to work.
This could be done be changing your port to generate the unaligned loads and store all the time instead of the aligned versions, but, this would cause generated code to suffer significant performance degradation, which is why in general, people don't do this.
