Re: [PATCH 03/11] x86/mm/hotplug: Don't remove PGD entries in remove_pagetable()

2016-03-10 Thread Andy Lutomirski
On Thu, Mar 10, 2016 at 1:56 AM, Ingo Molnar wrote: > > * Andy Lutomirski wrote: > >> On Fri, Feb 12, 2016 at 11:04 AM, Andy Lutomirski >> wrote: >> > On Mon, Sep 21, 2015 at 11:23 PM, Ingo Molnar wrote: >> >> So when memory hotplug removes a piece of physical memory from pagetable >> >> mappi

Re: [PATCH 03/11] x86/mm/hotplug: Don't remove PGD entries in remove_pagetable()

2016-03-10 Thread Ingo Molnar
* Andy Lutomirski wrote: > On Fri, Feb 12, 2016 at 11:04 AM, Andy Lutomirski wrote: > > On Mon, Sep 21, 2015 at 11:23 PM, Ingo Molnar wrote: > >> So when memory hotplug removes a piece of physical memory from pagetable > >> mappings, it also frees the underlying PGD entry. > >> > >> This compl

Re: [PATCH 03/11] x86/mm/hotplug: Don't remove PGD entries in remove_pagetable()

2016-03-09 Thread Andy Lutomirski
On Fri, Feb 12, 2016 at 11:04 AM, Andy Lutomirski wrote: > On Mon, Sep 21, 2015 at 11:23 PM, Ingo Molnar wrote: >> So when memory hotplug removes a piece of physical memory from pagetable >> mappings, it also frees the underlying PGD entry. >> >> This complicates PGD management, so don't do this.

Re: [PATCH 03/11] x86/mm/hotplug: Don't remove PGD entries in remove_pagetable()

2016-02-12 Thread Andy Lutomirski
On Mon, Sep 21, 2015 at 11:23 PM, Ingo Molnar wrote: > So when memory hotplug removes a piece of physical memory from pagetable > mappings, it also frees the underlying PGD entry. > > This complicates PGD management, so don't do this. We can keep the > PGD mapped and the PUD table all clear - it's

Re: [PATCH 03/11] x86/mm/hotplug: Don't remove PGD entries in remove_pagetable()

2015-10-05 Thread Kamezawa Hiroyuki
On 2015/09/22 15:23, Ingo Molnar wrote: > So when memory hotplug removes a piece of physical memory from pagetable > mappings, it also frees the underlying PGD entry. > > This complicates PGD management, so don't do this. We can keep the > PGD mapped and the PUD table all clear - it's only a singl

[PATCH 03/11] x86/mm/hotplug: Don't remove PGD entries in remove_pagetable()

2015-09-21 Thread Ingo Molnar
So when memory hotplug removes a piece of physical memory from pagetable mappings, it also frees the underlying PGD entry. This complicates PGD management, so don't do this. We can keep the PGD mapped and the PUD table all clear - it's only a single 4K page per 512 GB of memory hotplugged. Cc: An