Re: [PATCH] alloc_percpu() fails to allocate percpu data

2008-02-23 Thread Nick Piggin
On Friday 22 February 2008 09:26, Peter Zijlstra wrote: > On Thu, 2008-02-21 at 19:00 +0100, Eric Dumazet wrote: > > Some oprofile results obtained while using tbench on a 2x2 cpu machine > > were very surprising. > > > > For example, loopback_xmit() function was using high number of cpu > > cycles

Re: [PATCH] alloc_percpu() fails to allocate percpu data

2008-02-23 Thread Andrew Morton
On Thu, 21 Feb 2008 19:00:03 +0100 Eric Dumazet <[EMAIL PROTECTED]> wrote: > +#ifndef cache_line_size > +#define cache_line_size()L1_CACHE_BYTES > +#endif argh, you made me look. Really cache_line_size() should be implemented in include/linux/cache.h. Then we tromp the stupid private implem

Re: [PATCH] alloc_percpu() fails to allocate percpu data

2008-02-21 Thread Peter Zijlstra
On Thu, 2008-02-21 at 19:00 +0100, Eric Dumazet wrote: > Some oprofile results obtained while using tbench on a 2x2 cpu machine > were very surprising. > > For example, loopback_xmit() function was using high number of cpu > cycles to perform the statistic updates, supposed to be real cheap > s

[PATCH] alloc_percpu() fails to allocate percpu data

2008-02-21 Thread Eric Dumazet
Some oprofile results obtained while using tbench on a 2x2 cpu machine were very surprising. For example, loopback_xmit() function was using high number of cpu cycles to perform the statistic updates, supposed to be real cheap since they use percpu data pcpu_lstats = netdev_priv(dev);