Re: Direct io on block device has performance regression on 2.6.xkernel

From: Andrew Morton
Date: Wed Mar 09 2005 - 23:54:57 EST

"Chen, Kenneth W" <kenneth.w.chen@xxxxxxxxx> wrote:
> > Did you generate a kernel profile?
> Top 40 kernel hot functions, percentage is normalized to kernel utilization.
> _spin_unlock_irqrestore 23.54%
> _spin_unlock_irq 19.27%


Is that with CONFIG_PREEMPT? If so, and if you disable CONFIG_PREEMPT,
this cost should be accounting the the spin_unlock() caller and we can see
who the culprit is. Perhaps dio->bio_lock.

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at