Re: [PATCH 1/5] x86/cpu: Newline on 'invalid siblings' warning

[Andrew Cooper <andrew.cooper3@citrix.com>]

On 29/07/14 16:10, Jan Beulich wrote:
>>>> On 29.07.14 at 16:29, <andrew.cooper3@citrix.com> wrote:
>> --- a/xen/arch/x86/cpu/common.c
>> +++ b/xen/arch/x86/cpu/common.c
>> @@ -444,7 +444,9 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c)
>>  	} else if (c->x86_num_siblings > 1 ) {
>>  
>>  		if (c->x86_num_siblings > nr_cpu_ids) {
>> -			printk(KERN_WARNING "CPU: Unsupported number of the siblings %d", c->x86_num_siblings);
>> +			printk(KERN_WARNING
>> +			       "CPU: Unsupported number of the siblings %d\n",
>> +			       c->x86_num_siblings);
> Since you have to fiddle with this anyway, can you limit the printing
> to BP or opt_cpu_info? That said, it's rather bogus a check anyway -
> if one built a hypervisor for just 2 CPUs (let's assume for a minute
> that nr_cpu_ids then won't be larger than 2) and ran on a 4-fold
> HT CPU, this would trigger for no good reason. Hence the whole
> conditional could as well go away imo (and current Linux indeed
> doesn't have it or any equivalent anymore).
>
> As to indentation - if you want to keep it, how about moving it up
> as a middle "else if" prior to the enclosing one?
>
> Jan
>

Hmm - I thought it was somewhat silly but opted for not changing what
was there.

Given that it is clearly bogus, I will just drop the entire clause.

~Andrew