[PATCH 1/1] NEWS.rst: update with pSeries initial memory fix

[Daniel Henrique Barboza]

On 1/12/21 10:24 AM, Andrea Bolognani wrote:
> On Mon, 2021-01-11 at 23:03 -0300, Daniel Henrique Barboza wrote:
>> On 1/11/21 2:56 PM, Andrea Bolognani wrote:
>>> Daniel, did you test migration between libvirt 7.0.0 and earlier
>>> releases? Can you confirm it works?
>>
>> Just tested a migration scenario between 2 hosts (a Power 9 Boston running Libvirt
>> 6.8.0 and a Power 9 Mihawk running Libvirt 7.0.0), with a domain that reproduces
>> the bug.
>>
>> TL;DR: migration works both ways without issues.
> 
> Thanks for checking! Did you also test the following, less
> straighforward scenarios:
> 
>    * creating the domain on 6.8.0, migrating it to 7.0.0, then
>      migrating it back to 6.8.0;

Ping-ponging works with either scenario, just checked.

> 
>    * creating the domain on 6.8.0, migrating it to 7.0.0, shutting it
>      down, starting it back up, then migrating it back to 6.8.0.


It works. Details down below:


*** In the 6.8.0 server:


[danielhb at ltc-boston118 build]$ sudo ./run tools/virsh start vm1_6.8.0
Domain vm1_6.8.0 started

[danielhb at ltc-boston118 build]$ sudo ./run tools/virsh -c 'qemu:///system' migrate --live \
--domain vm1_6.8.0  --desturi qemu+ssh://ltcmihawk39/system --timeout 60 \
--persistent --undefinesource

[danielhb at ltc-boston118 build]$ sudo ./run tools/virsh list --all
  Id   Name   State
--------------------

<--- migration success --->


***  7.0.0 server:


[danielhb at ltcmihawk39 build]$ sudo ./run tools/virsh list --all
  Id   Name        State
----------------------------
  5    vm1_6.8.0   running
  -    vm1_7.0.0   shut off


[danielhb at ltcmihawk39 build]$ sudo ./run tools/virsh destroy vm1_6.8.0
Domain 'vm1_6.8.0' destroyed

[danielhb at ltcmihawk39 build]$ sudo ./run tools/virsh start vm1_6.8.0
Domain 'vm1_6.8.0' started

[danielhb at ltcmihawk39 build]$ sudo tail -n 30 ~/libvirt_build/var/log/libvirt/qemu/vm1_6.8.0.log
(...)
-cpu POWER9 \
-m size=1835008k,slots=16,maxmem=4194304k \
(...)
-object memory-backend-ram,id=memdimm0,size=536870912,host-nodes=8,policy=bind \
-device pc-dimm,memdev=memdimm0,id=dimm0,slot=0 \

<-- Note: same memory parameters defined in the 6.8.0 version. We can't
     distinguish whether this memory setting was user intention or a
     misalign in the source Libvirt, since we do not re-align during
     migration -->

[danielhb at ltcmihawk39 build]$ sudo ./run tools/virsh -c 'qemu:///system' migrate --live \
--domain vm1_6.8.0  --desturi qemu+ssh://ltc-boston118/system --timeout 60 \
--persistent --undefinesource

[danielhb at ltcmihawk39 build]$

<--- migration success --->


*** Back to the 6.8.0 server:

[danielhb at ltc-boston118 build]$ sudo ./run tools/virsh list --all
[sudo] password for danielhb:
  Id   Name        State
---------------------------
  8    vm1_6.8.0   running




Thanks,


DHB



> 
> I believe they should work correctly, because in neither case
> ABI_UPDATE should be involved, but it would be great to have direct
> confirmation.
>