> Most importantly, select() isn't critical path code!!! If the execution
> time of select gets increased by (say) 20 or 40 cycles, no one is going
> to be able to notice the difference. We're talking nanoseconds
> here.....
A quick profile of my kernel with a mass of straces shows that almost
every select done is passed 256 as the table size, and everyone seems to
be using the fd table size as an argument
Well, all of *my* code uses maxfds+1 --- I didn't realize everyone was
quite that broken. :-)
What set of programs were in your sample set?
- Ted