Nathan Lynch writes:
> Hi Scott, I've owed you a follow-up on this for a couple weeks, sorry.
I see v2 was posted as I was writing this, so I'll follow up on that
thread.
Hi Scott, I've owed you a follow-up on this for a couple weeks, sorry.
Beyond the issue of whether to remove the drmem_info->lmbs array, there
are some other things to address.
Scott Cheloha writes:
> diff --git a/arch/powerpc/include/asm/drmem.h
> b/arch/powerpc/include/asm/drmem.h
> index 3d7
On 2/3/2020 2:13 PM, Scott Cheloha wrote:
> On Thu, Jan 30, 2020 at 10:09:32AM -0600, Fontenot, Nathan wrote:
>> On 1/29/2020 12:10 PM, Scott Cheloha wrote:
>>> On Tue, Jan 28, 2020 at 05:56:55PM -0600, Nathan Lynch wrote:
Scott Cheloha writes:
> LMB lookup is currently an O(n) linear sea
On Tue, Jan 28, 2020 at 05:56:55PM -0600, Nathan Lynch wrote:
> Scott Cheloha writes:
> > LMB lookup is currently an O(n) linear search. This scales poorly when
> > there are many LMBs.
> >
> > If we cache each LMB by both its base address and its DRC index
> > in an xarray we can cut lookups to
On Thu, Jan 30, 2020 at 10:09:32AM -0600, Fontenot, Nathan wrote:
> On 1/29/2020 12:10 PM, Scott Cheloha wrote:
> > On Tue, Jan 28, 2020 at 05:56:55PM -0600, Nathan Lynch wrote:
> >> Scott Cheloha writes:
> >>> LMB lookup is currently an O(n) linear search. This scales poorly when
> >>> there are
On 1/29/2020 12:10 PM, Scott Cheloha wrote:
> On Tue, Jan 28, 2020 at 05:56:55PM -0600, Nathan Lynch wrote:
>> Scott Cheloha writes:
>>> LMB lookup is currently an O(n) linear search. This scales poorly when
>>> there are many LMBs.
>>>
>>> If we cache each LMB by both its base address and its DR
On Tue, Jan 28, 2020 at 05:56:55PM -0600, Nathan Lynch wrote:
> Scott Cheloha writes:
> > LMB lookup is currently an O(n) linear search. This scales poorly when
> > there are many LMBs.
> >
> > If we cache each LMB by both its base address and its DRC index
> > in an xarray we can cut lookups to
Scott Cheloha writes:
> LMB lookup is currently an O(n) linear search. This scales poorly when
> there are many LMBs.
>
> If we cache each LMB by both its base address and its DRC index
> in an xarray we can cut lookups to O(log n), greatly accelerating
> drmem initialization and memory hotplug.
LMB lookup is currently an O(n) linear search. This scales poorly when
there are many LMBs.
If we cache each LMB by both its base address and its DRC index
in an xarray we can cut lookups to O(log n), greatly accelerating
drmem initialization and memory hotplug.
This patch introduces two xarrays