4 ms·
Well, really the consequences would have been virtually the same either way, since the number of OpenSSL deployments where the system malloc uses guard pages is
by caf 13y ago
Well, really the consequences would have been virtually the same either way, since the number of OpenSSL deployments where the system malloc uses guard pages is a rounding error.
- Sanddancer 13y agoYes, but because of the caching malloc, the OpenBSD implementations are just as affected as the rest of the world's. Irresponsible software design sucks, plain and simple.
- caf 13y agoIt is a stretch to call this irresponsible. Is there any evidence that the OpenBSD people ever drew upstream's attention to this?
- usefulcat 13y agoNot necessarily. It's probably still more likely that the bug would have been discovered sooner, even if only via a relatively small number of crashes.