Re: [PATCH v3 0/5] mm: Enable CONFIG_NODES_SPAN_OTHER_NODES by default for NUMA

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 03/30/20 at 09:44am, Michal Hocko wrote:
> On Sun 29-03-20 08:19:24, Baoquan He wrote:
> > On 03/28/20 at 11:31am, Hoan Tran wrote:
> > > In NUMA layout which nodes have memory ranges that span across other nodes,
> > > the mm driver can detect the memory node id incorrectly.
> > > 
> > > For example, with layout below
> > > Node 0 address: 0000 xxxx 0000 xxxx
> > > Node 1 address: xxxx 1111 xxxx 1111
> > 
> > Sorry, I read this example several times, but still don't get what it
> > means. Can it be given with real hex number address as an exmaple? I
> > mean just using the memory layout you have seen from some systems. The
> > change looks interesting though.
> 
> Does this make it more clear?
>            physical address range and its node associaion
>          [0 0 0 0 1 1 1 1 0 0 0 0 1 1 1 1 0 0 0 0 1 1 1 1]

I later read it again, have got what Hoan is trying to say, thanks.

I think the change in this patchset makes sense, still have some concern
though, let me add comment in other thread.





[Index of Archives]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Bugtraq]     [Linux OMAP]     [Linux MIPS]     [eCos]     [Asterisk Internet PBX]     [Linux API]

  Powered by Linux