Re: [patch] lockless, scalable get_pid(), for_each_process() elimination, 2.5.35-BK

[Martin J. Bligh]

>> Nobody's trying to screw the desktop users, we're being mind- bogglingly
>> careful not to, in fact. If you have specific objections to a particular
>> patch, raise them as technical arguments. Saying "we shouldn't do that
>> because I'm not interested in it" is far less useful.
> 
> i fully agree with your points, but it does not hold in this specific
> case. Eliminating for_each_task loops (the ones completely unrelated to
> the get_pid() issue) is an improvement even for desktop setups, which have
> at most 1000 processes running.

Right - which is exactly why I was saying we should stick to technical 
debates rather than whether some people were interested in a particular
market segment or not ;-)

M.