Re: [Xen-devel] [PATCH v2 4/5] tools: xenpm: continue to support {set, get}-vcpu-migration-delay

[George Dunlap <george.dunlap@xxxxxxxxxx>]

On 02/28/2018 03:17 PM, Dario Faggioli wrote:
> On Wed, 2018-02-28 at 15:02 +0000, George Dunlap wrote:
>> On 02/23/2018 04:41 PM, Dario Faggioli wrote:
>>> From this commit on, `xenpm {set,get}-vcpu-migration-delay'
>>>
>>> commands work again. But that is only for the sake of
>>> backward compatibility, and their use is deprecated, in
>>> favour of 'xl sched-credit -s [-c <poolid>] -m <delay>'.
>>>
>>> Signed-off-by: Dario Faggioli <dfaggioli@xxxxxxxx>
>>> Acked-by: Wei Liu <wei.liu2@xxxxxxxxxx>
>>
>> If we cared about strict behavioral compatibility, we would modify
>> the
>> xenpm function to set the migration delay for all cpupools; but I
>> think
>> we can worry about that if anyone complains. :-)
>>
> Yes, I thought about that. But, considering what would be required to
> implement such behavior, as compared to how many people (as far as we
> can tell, of course) use this feature, and use xenpm to change it, I
> also thought we are indeed fine "waiting and seeing".
> 
>>> @@ -1087,13 +1097,17 @@ void set_vcpu_migration_delay_func(int
>>> argc, char *argv[])
>>>  
>>>  void get_vcpu_migration_delay_func(int argc, char *argv[])
>>>  {
>>> -    uint32_t value;
>>> +    struct xen_sysctl_credit_schedule sparam;
>>> +
>>> +    printf("WARNING: using xenpm for this purpose is deprecated."
>>> +           " Check out `xl sched-credit -s'\n");
>>
>> Should these warnings be to stderr rather than stdout, so that if
>> anything is manually parsing stout it will continue to work?
>>
> Again, I considered this, for that same reason. The reason why I
> decided for stdout is for maximizing the chances that anyone actually
> using this will see the warning (even if, e.g., they're redirecting
> stderr).

If they're redirecting stderr then they have explicitly decided to
ignore warnings and errors -- if they want to miss messages like this
they should be allowed to.  I would think it much more likely that
stdout would be redirected somewhere than stderr.

 -George

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel