Mayuresh <mayuresh%acm.org@localhost> writes: > Can somebody throw some light on this change in behavior of malloc, > whether it was intentional and its purpose etc. At the risk of being duplicative with Joerg and David's comments: While it's progress that Daniel and you understand what's going wrong, the key question is whether NetBSD's malloc conforms to the relevant standards. If it does (and so far there is not even a hint that it does not), and some program relies on the behavior being more narrowly defined than the standard (e.g., not using b31), then that program is incorrect. Also, if gprolog assumes that b31 is clear, it should check and assert this, because the failure you encountered was very non-obvious.
Description: PGP signature