On 8/7/2020 4:24 PM, Martin Kletzander wrote:
On Fri, Aug 07, 2020 at 01:27:59PM +0800, Zhong, Luyao wrote:
On 8/3/2020 7:00 PM, Martin Kletzander wrote:
On Mon, Aug 03, 2020 at 05:31:56PM +0800, Luyao Zhong wrote:
Hi Libvirt experts,
I would like enhence the numatune snippet configuration. Given a
example snippet:
<domain>
 ...
 <numatune>
  <memory mode="strict" nodeset="1-4,^3"/>
  <memnode cellid="0" mode="strict" nodeset="1"/>
  <memnode cellid="2" mode="preferred" nodeset="2"/>
 </numatune>
 ...
</domain>
Currently, attribute mode is either 'interleave', 'strict', or
'preferred',
I propose to add a new 'default'Â option. I give the reason as
following.
Presume we are using cgroups v1, Libvirt sets cpuset.mems for all vcpu
threads
according to 'nodeset' in memory element. And translate the memnode
element to
qemu config options (--object memory-backend-ram) for per numa cell,
which
invoking mbind() system call at the end.[1]
But what if we want using default memory policy and request each guest
numa cell
pinned to different host memory nodes? We can't use mbind via qemu
config options,
because (I quoto here) "For MPOL_DEFAULT, the nodemask and maxnode
arguments must
be specify the empty set of nodes." [2]
So my solution is introducing a new 'default' option for attribute
mode. e.g.
<domain>
 ...
 <numatune>
  <memory mode="default" nodeset="1-2"/>
  <memnode cellid="0" mode="default" nodeset="1"/>
  <memnode cellid="1" mode="default" nodeset="2"/>
 </numatune>
 ...
</domain>
If the mode is 'default', libvirt should avoid generating qemu command
line
'--object memory-backend-ram', and invokes cgroups to set cpuset.mems
for per guest numa
combining with numa topology config. Presume the numa topology is :
<cpu>
 ...
 <numa>
  <cell id='0' cpus='0-3' memory='512000' unit='KiB' />
  <cell id='1' cpus='4-7' memory='512000' unit='KiB' />
 </numa>
 ...
</cpu>
Then libvirt should set cpuset.mems to '1' for vcpus 0-3, and '2' for
vcpus 4-7.
Is this reasonable and feasible? Welcome any comments.
There are couple of problems here. The memory is not (always)
allocated
by the
vCPU threads. I also remember it to not be allocated by the process,
but in KVM
in a way that was not affected by the cgroup settings.
Thanks for your reply. Maybe I don't get what you mean, could you give
me more context? But what I proposed will have no effect on other memory
allocation.
Check how cgroups work. We can set the memory nodes that a process will
allocate from. However to set the node for the process (thread) QEMU
needs to
be started with the vCPU threads already spawned (albeit stopped). And
for that
QEMU already allocates some memory. Moreover if extra memory was allocated
after we set the cpuset.mems it is not guaranteed that it will be
allocated by
the vCPU in that NUMA cell, it might be done in the emulator instead or
the KVM
module in the kernel in which case it might not be accounted for the
process
actually causing the allocation (as we've already seen with Linux). In all
these cases cgroups will not do what you want them to do. The last case
might
be fixed, the first ones are by default not going to work.
That might be
fixed now,
however.
But basically what we have against is all the reasons why we started
using
QEMU's command line arguments for all that.
I'm not proposing use QEMU's command line arguments, on contrary I want
using cgroups setting to support a new config/requirement. I give a
solution about if we require default memory policy and memory numa
pinning.
And I'm suggesting you look at the commit log to see why we *had* to add
these
command line arguments, even though I think I managed to describe most
of them
above already (except for one that _might_ already be fixed in the
kernel). I
understand the git log is huge and the code around NUMA memory
allocation was
changing a lot, so I hope my explanation will be enough.
Thank you for detailed explanation, I think I get it now. We can't
guarantee memory allocation matching requirement since there is a time
slot before setting cpuset.mems.
Thanks,
Luyao
Sorry, but I think it will more likely break rather than fix stuff.
Maybe this
could be dealt with by a switch in `qemu.conf` with a huge warning above
it.
I'm not trying to fix something, I propose how to support a new
requirement just like I stated above.
I guess we should take a couple of steps back, I don't get what you are
trying
to achieve. Maybe if you describe your use case it will be easier to
reach a
conclusion.
Yeah, I do have a usecase I didn't mention before. It's a feature in
kernel but not merged yet, we call it memory tiering.
(https://lwn.net/Articles/802544/)
If memory tiering is enabled on host, DRAM is top tier memory, and
PMEM(persistent memory) is second tier memory, PMEM is shown as numa
node without cpu. For short, pages can be migrated between DRAM and PMEM
based on DRAM pressure and how cold/hot they are.
We could configure multiple memory migrating path. For example,
node 0: DRAM, node 1: DRAM, node 2: PMEM, node 3: PMEM
we can make 0+2 to a group, and 1+3 to a group. In each group, page is
allowed to migrated down(demotion) and up(promotion).
If **we want our VMs utilizing memory tiering and with NUMA topology**,
we need handle the guest memory mapping to host memory, that means we
need bind each guest numa node to a memory nodes group(DRAM node + PMEM
node) on host. For example, guest node 0 -> host node 0+2.
However, only cgroups setting can make the memory tiering work, if we
use mbind() system call, demoted pages will never go back to DRAM.
That's why I propose to add 'default' option and bypass mbind in QEMU.
I hope I make myself understandable. I'll appreciate if you could give
some suggestion.
Regards,
Luyao
Have a nice day,
Martin
Regards,
Luyao
[1]https://github.com/qemu/qemu/blob/f2a1cf9180f63e88bb38ff21c169da97c3f2bad5/backends/hostmem.c#L379
[2]https://man7.org/linux/man-pages/man2/mbind.2.html
--
2.25.1
