On 03/18/2014 11:21 PM, Stefan Weil wrote: > Are there test cases which fail with the old code?
Only when running under analysis tools looking for technical violations of the standard. Under normal circumstances it would never crash since, as with the native backends, the underlying cpu is going to treat this some rational way. But I sincerely doubt that we could measure any performance loss in TCI. The single extra arithmetic operation should be swallowed by the loop back to interpret the next opcode. r~
