Date: Wed, 17 Sep 1997 19:05:52 +0000 (GMT) From: Terry Lambert <tlambert@primenet.com> To: julian@whistle.com (Julian Elischer) Cc: Shimon@i-connect.net, FreeBSD-Hackers@FreeBSD.ORG Subject: Re: Fast Encryption (in kernel) seeked Message-ID: <199709171905.MAA04926@usr02.primenet.com> In-Reply-To: <Pine.BSF.3.95.970916003201.13496B-100000@current1.whistle.com> from "Julian Elischer" at Sep 16, 97 00:38:42 am
next in thread | previous in thread | raw e-mail | index | archive | help
> To answer this question correctly, > We need to know more details.. > every algorythm has its day. It depends on what the situation is. Heh. > For example, if there will be less than 100 of these 'pointers' > outstanding at a time, different schemes would be used to the case where > there are 100000 outstanding.. Oh, for example, we could call the "100 case"... "file descriptors"? 8-). Using a pool that you know of, you can easily check if the address is within the bounds of the pool, and whethr the address is pointing to a structure boundry or not (((addr - base) % size) == 0), and whether the struct is valid (ad a "valid" flag to it). > can you guarentee that each pointer given out will be 'returned'? ...if you can't, you will have to use range restrictions in order to validate pointers as if they were cookies. Probably you will need to include validators co the caller can indicate them to the kernel. > will it be returned only once? ...if so, then allocate a context struct that is freed when the pointer is returned, since it's proper N:N reflexive. > what if a process dies while owning a pointer? ...if you allocate anything in kernel space on a per process basis, it will have to be freed by _exit. > when does the buffer become free? ...something to consider, in case you want to avoid filling up all of kernel memory. 8-) 8-). > who allocates the buffers? ...whoever it is will need to free them for the interface to be reflexive and not bubject to races. > I have code that does all this but it's only useful to you if > what you want to do matches the goals of what I was designing for.. > > tell me more.. Especially tell us how a kernel "encryption" whose source is know can fool a user program that is aware of your algorithm with anthing more than statistical protection. 8-) 8-). Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199709171905.MAA04926>