It is no longer overly relevant what someone might think is best or not. It is a tenet of the assembler to try to continue forward if there is a reasonable approach that can be taken, and in this case the approach taken was eminently reasonable. Perhaps that came from 50+ years ago when it was really important to avoid excessive CPU consumption as multiple re-assemblies might take.
We live with a lot of the default design choices we have made, in order to avoid breaking existing application (including breaking a re-assembly). That's a major reason there is still a mainframe OS about which this assembler-list can comment. A change to make this case an error would be incompatible and would not therefore be viewed as acceptable since the benefit of a change in this area would be minuscule A good build process would probably be set up to fail upon RC=4 from the assembly regardless of the reason for the RC. So if you don't like the behavior of warning upon label-on-DROP then consider changing your build process. Peter Relson z/OS Core Technology Design
