Re: [PATCH v3] x86: correct socket_cpumask allocation

[Boris Ostrovsky <boris.ostrovsky@oracle.com>]

On 07/09/2015 11:16 AM, Andrew Cooper wrote:
> On 09/07/15 15:23, Chao Peng wrote:
>> For booting cpu, the socket number is not needed to be 0 so
>> it needs to be computed by cpu number.
>>
>> For secondary cpu, phys_proc_id is not valid in CPU_PREPARE
>> notifier(cpu_smpboot_alloc), so cpu_to_socket(cpu) can't be used.
>> Instead, pre-allocate secondary_cpu_mask in cpu_smpboot_alloc()
>> and later consume it in smp_store_cpu_info().
>>
>> This patch also change socket_cpumask type from 'cpumask_var_t *'
>> to 'cpumask_t **' so that smaller NR_CPUS works.
>>
>> Reported-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
>> Tested-by: Dario Faggioli <dario.faggioli@citrix.com>
>> Signed-off-by: Chao Peng <chao.p.peng@linux.intel.com>
> The v2 of this path seems to be holding up fine against the XenServer
> test pool.  No failures to boot encountered so far.
>
> Tested-by: Andrew Cooper <andrew.cooper3@citrix.com>
>
> As the differences between v2 and v3 appear largely mechanical and not
> functional in nature, this Tested-by can probably still stand.

Works for me as well. (You can add 'Tested-by: Boris Ostrovsky 
<boris.ostrovsky@oracle.com>' although I think by now there are enough 
of those)

-boris