Re: PATCH: Fix ll/sc for mips (take 3)

On Sat, Feb 02, 2002 at 01:53:23PM -0500, Justin Carlson wrote:

2)  More importantly, most implementations don't use any sort of dynamic
branch prediction on branch likelies.  They predict taken, always, since
that's the specified intent (it's a branch *likely* to be taken).

CPU guys hate branch likely and would probably love if whoever invented
them hires at Intel ;-)

For most spin locks, the normal behaviour is a fall through, not taking
that branch, so you're inflicting a branch mispredict penalty on every
lock grabbed without contention.  Even for locks which the general case
is contention, giving the processor branch predictor a chance to learn
that is a Good Idea.

I was thinking about spinlocks like

retry:	la	addrreg, retry
	ll	reg, lockvar
	...
	sc	reg, lockvar
	teq	$0, reg

That would depend on the price of a trap instruction when it's not taken
and the probability of the lock being congested.

  Ralf