Re: 2.6.24-rc2: Network commit causes SLUB performance regression with tbench

!MAILaRCHIVE_VOTE_RePLACE
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
To: David Miller <davem@...>
Cc: <clameter@...>, <netdev@...>, <herbert@...>, <linux-kernel@...>
Date: Tuesday, November 13, 2007 - 6:55 pm

On Wednesday 14 November 2007 09:27, Nick Piggin wrote:


Doesn't help (with vanilla kernel -- Herbert's patch applied).

data_len histogram drops to 0 and goes to len (I guess that's not
surprising).

Performance is pretty similar (ie. not good).

I'll look at allocator patterns next.
-
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
2.6.24-rc2 slab vs slob tbench numbers, Nick Piggin, (Fri Nov 9, 8:36 am)
Re: 2.6.24-rc2 slab vs slob tbench numbers, Matt Mackall, (Mon Nov 12, 4:13 pm)
Re: 2.6.24-rc2 slab vs slob tbench numbers, Nick Piggin, (Tue Nov 13, 7:44 am)
Re: 2.6.24-rc2 slab vs slob tbench numbers, Christoph Lameter, (Fri Nov 9, 11:15 am)
Re: 2.6.24-rc2 slab vs slob tbench numbers, Christoph Lameter, (Fri Nov 9, 1:49 pm)
Re: 2.6.24-rc2: Network commit causes SLUB performance regre..., Christoph Lameter, (Mon Nov 12, 3:44 pm)
Re: 2.6.24-rc2: Network commit causes SLUB performance regre..., Christoph Lameter, (Wed Nov 14, 9:03 pm)
Re: 2.6.24-rc2: Network commit causes SLUB performance regre..., Christoph Lameter, (Wed Nov 14, 2:33 pm)
Re: 2.6.24-rc2: Network commit causes SLUB performance regre..., Nick Piggin, (Tue Nov 13, 6:55 pm)