SSP vs x86 kernels

[matthew green <mrg%eterna.com.au@localhost>]

hi folks.


as others have reported, the slowdown from enabling SSP in x86 kernels is
not really measureable.  i've just finished compiling a bunch of timings for
various things i care about and i can't see any slowdown.  infact, for a lot
of the tests, the SSP enabled kernels got faster averages or faster best-score
times.  it's all noise.


we are going to re-enable this for x86 kernels.


speaking for core,

.mrg.