Re: AIM7 40% regression with 2.6.26-rc1

[Linus Torvalds]

On Wed, 7 May 2008, Ingo Molnar wrote:
> 
> There's far more normal mutex fastpath use during an AIM7 run than any 
> BKL use. So if it's due to any direct fastpath overhead and the 
> resulting widening of the window for the real slowdown, we should see a 
> severe slowdown on AIM7 with CONFIG_MUTEX_DEBUG=y. Agreed?

Not agreed.

The BKL is special because it is a *single* lock.

All the "normal" mutex code use fine-grained locking, so even if you slow 
down the fast path, that won't cause the same kind of fastpath->slowpath 
increase.

In order to see the fastpath->slowpath thing, you do need to have many 
threads hitting the same lock: ie the slowdown has to result in real 
contention. 

Almost no mutexes have any potential for contention what-so-ever, except 
for things that very consciously try to hit it (multiple threads doing 
readdir and/or file creation on the *same* directory etc).

The BKL really is special. 

		Linus
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/