slub: move kmem_cache_node into it's own cacheline
in [Kernel]
|
From: Pekka Enberg on 22 May 2010 04:40 Alexander Duyck wrote: > This patch is meant to improve the performance of SLUB by moving the local > kmem_cache_node lock into it's own cacheline separate from kmem_cache. > This is accomplished by simply removing the local_node when NUMA is enabled. > > On my system with 2 nodes I saw around a 5% performance increase w/ > hackbench times dropping from 6.2 seconds to 5.9 seconds on average. I > suspect the performance gain would increase as the number of nodes > increases, but I do not have the data to currently back that up. > > Signed-off-by: Alexander Duyck <alexander.h.duyck(a)intel.com> Thanks for the fix, Alexander! Yanmin and Alex, can I have your Tested-by or Acked-by please so we can close "[Bug #15713] hackbench regression due to commit 9dfc6e68bfe6e" after this patch is merged? Pekka -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Shi, Alex on 23 May 2010 09:20 sure. glad to a fix on this! -----Original Message----- From: Pekka Enberg [mailto:penberg(a)cs.helsinki.fi] Sent: 2010Äê5ÔÂ22ÈÕ 16:36 To: Duyck, Alexander H Cc: cl(a)linux.com; linux-mm(a)kvack.org; Shi, Alex; yanmin_zhang(a)linux.intel.com; akpm(a)linux-foundation.org; linux-kernel(a)vger.kernel.org; rjw(a)sisk.pl Subject: Re: [PATCH v2] slub: move kmem_cache_node into it's own cacheline Alexander Duyck wrote: > This patch is meant to improve the performance of SLUB by moving the > local kmem_cache_node lock into it's own cacheline separate from kmem_cache. > This is accomplished by simply removing the local_node when NUMA is enabled. > > On my system with 2 nodes I saw around a 5% performance increase w/ > hackbench times dropping from 6.2 seconds to 5.9 seconds on average. > I suspect the performance gain would increase as the number of nodes > increases, but I do not have the data to currently back that up. > > Signed-off-by: Alexander Duyck <alexander.h.duyck(a)intel.com> Thanks for the fix, Alexander! Yanmin and Alex, can I have your Tested-by or Acked-by please so we can close "[Bug #15713] hackbench regression due to commit 9dfc6e68bfe6e" after this patch is merged? Pekka
|
Pages: 1 Prev: [PATCH 26/27] drivers/isdn: Use memdup_user Next: interrupt.h: fix fatal kernel-doc error |