[Bug middle-end/63273] atomic operations lead to inefficient code

[amacleod at redhat dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63273

Andrew Macleod <amacleod at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |amacleod at redhat dot com

--- Comment #3 from Andrew Macleod <amacleod at redhat dot com> ---
GCC atomic implementation is intended to be correct first, and expose efficient
implementations for each of the memory models for the builtins.   No one has
had time to venture into the optimization of these atomic operations, although
I drew up plans once upon a time.
(https://gcc.gnu.org/wiki/Atomic/GCCMM/Optimizations)

Its simply not clear given the nature of how atomics are normally used that its
worth the time to heavily optimize in and around them for real world test
cases.  Its on the back-burner until such time that it is deemed to be
worthwhile.  

Anyone who wished to work on this is welcome to have at it...