On Mon, Aug 13, 2001 at 04:01:43PM -0700, richard offer wrote: > My applications == your applications. Richard, don't forget that the point of this thread is to discuss the possible methods of multiplexing one system call for several modules. The system call exists so that applications written for a specific module can communicate with the module, using some agreement between the module and the application. So, the applications that are going to be calling the new syscall are by definition *not* the applications that other people may run. There are two cases: module-specific applications and general applications. The first case is trivial. The second case, well, it *was* forked to use the new syscall, so don't get too upset about the fork. (If not forking is important to you, don't use a new system call that uses features specific to any module. :) > In the red corner we have people who think it should be done via /proc > Greg, Jesse, In the grand tradition of putting words into the mouths of other people, I think Greg is weighing in on the idea of making a brand new filesystem similar to /proc. (He has been saying this for months, and until today I always thought he meant a new directory in /proc. Heheh.) (In case anyone cares, my opinion is that a syscall error return should not be the primary method of module identification. /proc or special filesystems are both OK by me. Detailed reasoning is available if anyone cares enough to ask. :) _______________________________________________ linux-security-module mailing list linux-security-moduleat_private http://mail.wirex.com/mailman/listinfo/linux-security-module
This archive was generated by hypermail 2b30 : Mon Aug 13 2001 - 16:31:33 PDT