6 Sep
2005
6 Sep
'05
12:18 p.m.
Hamish Guthrie hamish@prodigi.ch writes:
I would like to apologise for my previous tirade, but I have fixed this problem a few times in the past and the SMP guys just ignore UNDERSTANDING THE CODE!!!
That is part of the point of the code review. So we can work out the issues.
In general it is much better to have something like cpu_relax conditionalized in the header than in the calling code. As it is the code is hard to read and it is not at all obvious why. As I recall cpu_relax simply expands to rep; nop which should be safe on any cpu.
In any event, the patch is crap!
How so?
The point is to communicate and if there is something non-obvious we need in a comment in the code so people do not forget what is going on.
Eric