On Tue, 2006-04-11 at 10:35 -0400, Peter Memishian wrote: > > It would be too late, and bad things could happen. I can think of two > > things to say about this: > > > > 1. I don't believe that putting binaries compiled on one version of > > Solaris onto an older version of Solaris is supported or guaranteed to > > work in any case. > > Unfortunately, it's fairly common practice to copy drivers around. If at > all possible, we should fail to load these gracefully, rather than > panicking the system or corrupting data.
Ok, fair enough. > > > 2. If this really needs to work (where "work" means not crash the > > system), then we could version mac_alloc() as well so that it can simply > > fail the allocation in this case. > > That could work. If we do that, I don't see a point in having the driver > still fill in the version before calling mac_register(), though -- if we > really need to support multiple versions concurrently, we can support that > by having mac_alloc() do the work. Right, this can easily be done. I'll update the materials, but to summarize: - mac_alloc() will take a version number as its sole argument. - mac_alloc() will do the version checking and also set the passed-in version number in the allocated mac_register_t. -Seb _______________________________________________ networking-discuss mailing list [email protected]
