On Tue, Mar 28, 2017 at 2:03 PM, Michael Ellerman wrote:
> Oliver O'Halloran writes:
>> On Wed, Mar 15, 2017 at 10:18 PM, Michael Ellerman
>> wrote:
>>> Oliver O'Halloran writes:
+ /* threads that share a chip-id are considered siblings (same die) */
>>>
>>> Also "Threads" :)
>>
>> Th
Oliver O'Halloran writes:
> On Wed, Mar 15, 2017 at 10:18 PM, Michael Ellerman
> wrote:
>> Oliver O'Halloran writes:
>>> + /* threads that share a chip-id are considered siblings (same die) */
>>
>> Also "Threads" :)
>
> The cpus masks are all built in terms of threads, so this is
> technic
On Wed, Mar 15, 2017 at 10:18 PM, Michael Ellerman wrote:
> Oliver O'Halloran writes:
>
>> To determine which logical CPUs are on the same core the kernel uses the
>> ibm,chipid property from the device tree node associated with that cpu.
>> The lookup for this this information is currently open
Oliver O'Halloran writes:
> To determine which logical CPUs are on the same core the kernel uses the
> ibm,chipid property from the device tree node associated with that cpu.
> The lookup for this this information is currently open coded in both
> traverse_siblings() and traverse_siblings_chip_id
On Thu, Mar 02, 2017 at 11:49:16AM +1100, Oliver O'Halloran wrote:
> To determine which logical CPUs are on the same core the kernel uses the
> ibm,chipid property from the device tree node associated with that cpu.
> The lookup for this this information is currently open coded in both
> traverse_s
To determine which logical CPUs are on the same core the kernel uses the
ibm,chipid property from the device tree node associated with that cpu.
The lookup for this this information is currently open coded in both
traverse_siblings() and traverse_siblings_chip_id(). This patch replaces
these manual