David Wagner wrote: > > Steve Beattie wrote: > >It seems unlikely to me that using /proc or a private fs can NOT be > >slower. > > The important question is not whether it is slower. The real question > is whether it is too slow. Those are *very* different questions. Interesting results of my "semi-scientific" relative performance tests of a proc system read of uid_t size compared to getuid(): 2.0.x kernels: (PROC:GETUID) 1.2:1 on MIPS 1:1 on i586 2.2.x kernels: (PROC:GETUID) 6:1 on i686 Not sure why that should be, except it appears that 10million reads of /proc fs data take approximately the same time for all the systems (24000ms) (bogomips ranging from 40 to 1000) and getuid is MUCH FASTER on the 2.2 kernel with 1000 bogomips. (caching?) All systems were at >0.10 load (and similarly loaded) during the test. Ergo, my simple test (performed from userspace as root) suggests the impact is "highly variable", but COULD be relatively insignificant if, as Mr. Wagner suggests, it's used sparingly. For What It's Worth, J. Melvin Jones |>------------------------------------------------------ || J. MELVIN JONES jmjonesat_private |>------------------------------------------------------ || Microcomputer Systems Consultant || Software Developer || Web Site Design, Hosting, and Administration || Network and Systems Administration |>------------------------------------------------------ || http://www.jmjones.com/ |>------------------------------------------------------ _______________________________________________ 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 : Wed Apr 25 2001 - 05:45:36 PDT