>>> On 10.02.15 at 19:03, <tamas.leng...@zentific.com> wrote:
> On Tue, Feb 10, 2015 at 6:39 PM, Jan Beulich <jbeul...@suse.com> wrote:
>>>>> Tamas K Lengyel <tamas.leng...@zentific.com> 02/10/15 5:38 PM >>>
>>>On Tue, Feb 10, 2015 at 5:17 PM, Jan Beulich <jbeul...@suse.com> wrote:
>>>>>>> Tamas K Lengyel <tamas.leng...@zentific.com> 02/10/15 2:51 PM >>>
>>>> On Tue, Feb 10, 2015 at 1:52 PM, Jan Beulich <jbeul...@suse.com> wrote:
>>>>>>>> On 09.02.15 at 19:53, <tamas.leng...@zentific.com> wrote:
>>>>>>> @@ -598,6 +600,12 @@ int mem_sharing_sharing_resume(struct domain *d)
>>>>>>> {
>>>>>>> struct vcpu *v;
>>>>>>>
>>>>>>> + if ( rsp.version != MEM_EVENT_INTERFACE_VERSION )
>>>>>>> + {
>>>>>>> + gdprintk(XENLOG_WARNING, "mem_event interface version
> mismatch!\n");
>>>>>>
>>>>>> Why gdprintk()?
>>>>>
>>>>>Is that only for debug cases?
>>>>
>>>> I'm intending to propose compiling out alll dprintk() and gdprintk()
> instance in
>>>> non-debug builds. Right now they're preferable when the message is so terse
>>>> that identifying its origin without file name and line number is
>>>> difficult.
> Clearly
>>>> any non-debug messages shouldn't be of such poor quality.
>>>
>>>I willwrap it into #ifndef NDEBUG as it is really only for debugging.
>>
>> That'll make the code even uglier, and won't address the question I
> originally asked.
>
> I just reused the function since I've seen it being used for printing
> warnings. What would be the preffered print function to be used here?
printk(XENLOG_G_* ...);
Jan
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
