Re: REGRESSION: Performance regressions from switchinganon_vma->lock to mutex

[Andi Kleen]

> Good spot that certainly isn't plain .39.

True.

> 
> It looks like those (http://marc.info/?l=linux-mm&m=130533041726258) are
> similar to Linus' patch, except Linus takes the hard line that the root

One difference to Linus' patch is that it does batching for both fork+exit.
I suspect Linus' patch could probably too.

But the fork+exit patch only improved things slightly, it's very unlikely to 
recover 52%. Maybe the overall locking here needs to be revisited.

And in general it looks like blind conversion from spinlock to mutex
is a bad idea right now.

-Andi
--
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/