Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1210091

Re: [Patch V3 2/9] kernel/profile.c: Replace cpu_to_mem() with cpu_to_node()

Show all headers | View raw

On Wed, 19 Aug 2015, Jiang Liu wrote:

> On 2015/8/18 8:31, David Rientjes wrote:
> > On Mon, 17 Aug 2015, Jiang Liu wrote:
> > 
> >> Function profile_cpu_callback() allocates memory without specifying
> >> __GFP_THISNODE flag, so replace cpu_to_mem() with cpu_to_node()
> >> because cpu_to_mem() may cause suboptimal memory allocation if
> >> there's no free memory on the node returned by cpu_to_mem().
> >>
> > 
> > Why is cpu_to_node() better with regard to free memory and NUMA locality?
> Hi David,
> 	Thanks for review. This is a special case pointed out by Tejun.
> For the imagined topology, A<->B<->X<->C<->D, where A, B, C, D has
> memory and X is memoryless.
> Possible fallback lists are:
> B: [ B, A, C, D]
> X: [ B, C, A, D]
> C: [ C, D, B, A]
> 
> cpu_to_mem(X) will either return B or C. Let's assume it returns B.
> Then we will use "B: [ B, A, C, D]" to allocate memory for X, which
> is not the optimal fallback list for X. And cpu_to_node(X) returns
> X, and "X: [ B, C, A, D]" is the optimal fallback list for X.

Ok, that makes sense, but I would prefer that this 
alloc_pages_exact_node() change to alloc_pages_node() since, as you 
mention in your commit message, __GFP_THISNODE is not set.

In the longterm, if we setup both zonelists correctly (no __GFP_THISNODE 
and with __GFP_THISNODE), then I'm not sure there's any reason to ever use 
cpu_to_mem() for alloc_pages().
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[Patch V3 0/9] Enable memoryless node support for x86 Jiang Liu <jiang.liu@linux.intel.com> - 2015-08-17 05:20 +0200
  [Patch V3 6/9] i40evf: Use numa_mem_id() to better support memoryless node Jiang Liu <jiang.liu@linux.intel.com> - 2015-08-17 05:20 +0200
    RE: [Intel-wired-lan] [Patch V3 6/9] i40evf: Use numa_mem_id() to  better support memoryless node "Patil, Kiran" <kiran.patil@intel.com> - 2015-08-17 21:10 +0200
      Re: [Intel-wired-lan] [Patch V3 6/9] i40evf: Use numa_mem_id() to  better support memoryless node Jeff Kirsher <jeffrey.t.kirsher@intel.com> - 2015-08-18 23:40 +0200
  [Patch V3 9/9] mm, x86: Enable memoryless node support to better support CPU/memory hotplug Jiang Liu <jiang.liu@linux.intel.com> - 2015-08-17 05:20 +0200
    Re: [Patch V3 9/9] mm, x86: Enable memoryless node support to better  support CPU/memory hotplug Tang Chen <tangchen@cn.fujitsu.com> - 2015-08-18 08:20 +0200
      Re: [Patch V3 9/9] mm, x86: Enable memoryless node support to better  support CPU/memory hotplug Tang Chen <tangchen@cn.fujitsu.com> - 2015-08-18 13:40 +0200
    Re: [Patch V3 9/9] mm, x86: Enable memoryless node support to better  support CPU/memory hotplug Ingo Molnar <mingo@kernel.org> - 2015-08-18 09:40 +0200
  [Patch V3 1/9] x86, NUMA, ACPI: Online node earlier when doing CPU hot-addition Jiang Liu <jiang.liu@linux.intel.com> - 2015-08-17 05:20 +0200
  [Patch V3 2/9] kernel/profile.c: Replace cpu_to_mem() with cpu_to_node() Jiang Liu <jiang.liu@linux.intel.com> - 2015-08-17 05:20 +0200
    Re: [Patch V3 2/9] kernel/profile.c: Replace cpu_to_mem() with  cpu_to_node() David Rientjes <rientjes@google.com> - 2015-08-18 02:40 +0200
      Re: [Patch V3 2/9] kernel/profile.c: Replace cpu_to_mem() with  cpu_to_node() Jiang Liu <jiang.liu@linux.intel.com> - 2015-08-19 09:20 +0200
        Re: [Patch V3 2/9] kernel/profile.c: Replace cpu_to_mem() with  cpu_to_node() David Rientjes <rientjes@google.com> - 2015-08-20 02:10 +0200
  [Patch V3 4/9] openvswitch: Replace cpu_to_node() with cpu_to_mem() to support memoryless node Jiang Liu <jiang.liu@linux.intel.com> - 2015-08-17 05:20 +0200
    Re: [Patch V3 4/9] openvswitch: Replace cpu_to_node() with  cpu_to_mem() to support memoryless node Pravin Shelar <pshelar@nicira.com> - 2015-08-18 02:20 +0200
  [Patch V3 7/9] x86, numa: Kill useless code to improve code readability Jiang Liu <jiang.liu@linux.intel.com> - 2015-08-17 05:20 +0200
  [Patch V3 3/9] sgi-xp: Replace cpu_to_node() with cpu_to_mem() to support memoryless node Jiang Liu <jiang.liu@linux.intel.com> - 2015-08-17 05:20 +0200
    Re: [Patch V3 3/9] sgi-xp: Replace cpu_to_node() with cpu_to_mem()  to support memoryless node David Rientjes <rientjes@google.com> - 2015-08-18 02:30 +0200
      Re: [Patch V3 3/9] sgi-xp: Replace cpu_to_node() with cpu_to_mem() to  support memoryless node Jiang Liu <jiang.liu@linux.intel.com> - 2015-08-19 10:30 +0200
        Re: [Patch V3 3/9] sgi-xp: Replace cpu_to_node() with cpu_to_mem()  to support memoryless node David Rientjes <rientjes@google.com> - 2015-08-20 02:10 +0200
          Re: [Patch V3 3/9] sgi-xp: Replace cpu_to_node() with cpu_to_mem() to  support memoryless node Jiang Liu <jiang.liu@linux.intel.com> - 2015-08-20 08:40 +0200
    Re: [Patch V3 3/9] sgi-xp: Replace cpu_to_node() with cpu_to_mem() to  support memoryless node Robin Holt <robinmholt@gmail.com> - 2015-08-19 14:00 +0200
      Re: [Patch V3 3/9] sgi-xp: Replace cpu_to_node() with cpu_to_mem() to  support memoryless node Jiang Liu <jiang.liu@linux.intel.com> - 2015-08-19 14:50 +0200
  [Patch V3 5/9] i40e: Use numa_mem_id() to better support memoryless node Jiang Liu <jiang.liu@linux.intel.com> - 2015-08-17 05:20 +0200
    Re: [Patch V3 5/9] i40e: Use numa_mem_id() to better support memoryless  node David Rientjes <rientjes@google.com> - 2015-08-18 02:40 +0200
    RE: [Intel-wired-lan] [Patch V3 5/9] i40e: Use numa_mem_id() to  better support memoryless node "Patil, Kiran" <kiran.patil@intel.com> - 2015-08-20 00:40 +0200
      RE: [Intel-wired-lan] [Patch V3 5/9] i40e: Use numa_mem_id() to  better support memoryless node David Rientjes <rientjes@google.com> - 2015-08-20 02:20 +0200
  [Patch V3 8/9] mm: Update _mem_id_[] for every possible CPU when memory configuration changes Jiang Liu <jiang.liu@linux.intel.com> - 2015-08-17 05:20 +0200
  Re: [Patch V3 0/9] Enable memoryless node support for x86 Andrew Morton <akpm@linux-foundation.org> - 2015-08-17 23:40 +0200
  Re: [Patch V3 0/9] Enable memoryless node support for x86 Tang Chen <tangchen@cn.fujitsu.com> - 2015-08-18 12:10 +0200
    Re: [Patch V3 0/9] Enable memoryless node support for x86 Jiang Liu <jiang.liu@linux.intel.com> - 2015-08-19 10:10 +0200

csiph-web