Re: AIM7 40% regression with 2.6.26-rc1

[Linus Torvalds]

On Thu, 8 May 2008, Zhang, Yanmin wrote:
>
> I collected oprofile data. It looks not useful, as cpu idle is more than 50%.

Ahh, so it's probably still the BKL that is the problem, it's just not in 
the file locking code. The changes to fs/locks.c probably didn't matter 
all that much, and the additional regression was likely just some 
perturbation.

So it's probably fasync that AIM7 tests. Quite possibly coupled with 
/dev/tty etc. No file locking.

			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/