Re: [PATCH net-next v2 3/3] reciprocal_divide: correction/update of the algorithm

2014-01-16 Thread Eric Dumazet
On Fri, 2014-01-17 at 05:29 +0100, Hannes Frederic Sowa wrote: > Also I doubt the performance drop for SLAB will be that massive. Also it was > already replaced by SLUB as the default SLAB allocator, which doesn't use > reciprocal_divide. Google servers use SLAB, not SLUB, for various reasons, an

Re: [PATCH net-next v2 3/3] reciprocal_divide: correction/update of the algorithm

2014-01-16 Thread Hannes Frederic Sowa
On Thu, Jan 16, 2014 at 06:33:37PM -0800, Eric Dumazet wrote: > On Fri, 2014-01-17 at 01:28 +0100, Hannes Frederic Sowa wrote: > > Jakub Zawadzki noticed that some divisions by reciprocal_divide() > > were not correct [1][2], which he could also show with BPF code after > > divisions are transforme

Re: [PATCH net-next v2 3/3] reciprocal_divide: correction/update of the algorithm

2014-01-16 Thread Eric Dumazet
On Fri, 2014-01-17 at 01:28 +0100, Hannes Frederic Sowa wrote: > Jakub Zawadzki noticed that some divisions by reciprocal_divide() > were not correct [1][2], which he could also show with BPF code after > divisions are transformed into reciprocal_value() for runtime invariant > which can be passed