mbox series

[v3,0/4] examples/power: allow use of more than 64 cores

Message ID 20181214133137.23516-1-david.hunt@intel.com (mailing list archive)
Headers
Series examples/power: allow use of more than 64 cores |

Message

Hunt, David Dec. 14, 2018, 1:31 p.m. UTC 
  First of all we convert all the relevant uint64_t's to char arrays. Then
we remove the unneeded mask functions that were limited to 64 cores. Also
extend the guest functionality and finally rause the number of supported
cores to something more sensible, i.e. 256.

v2 changes:
  * Updates out of Review by Anatoly.
  * Indentation fixes
  * resolved missing spinlock around config updates
  * simplified some loops
  * fixed bug in non-equivalent code

v3 changes:
  * Moved a spinlock to a more appropriate place. Was in a for-loop, made
    more sense to have it outside.

[1/4] examples/power: change 64-bit masks to arrays
[2/4] examples/power: remove mask functions
[3/4] examples/power: allow vms to use lcores over 63
[4/4] examples/power: increase max cores to 256

Comments

Burakov, Anatoly Dec. 14, 2018, 1:37 p.m. UTC | #1 
On 14-Dec-18 1:31 PM, David Hunt wrote:
> First of all we convert all the relevant uint64_t's to char arrays. Then
> we remove the unneeded mask functions that were limited to 64 cores. Also
> extend the guest functionality and finally rause the number of supported
> cores to something more sensible, i.e. 256.
> 
> v2 changes:
>    * Updates out of Review by Anatoly.
>    * Indentation fixes
>    * resolved missing spinlock around config updates
>    * simplified some loops
>    * fixed bug in non-equivalent code
> 
> v3 changes:
>    * Moved a spinlock to a more appropriate place. Was in a for-loop, made
>      more sense to have it outside.
> 
> [1/4] examples/power: change 64-bit masks to arrays
> [2/4] examples/power: remove mask functions
> [3/4] examples/power: allow vms to use lcores over 63
> [4/4] examples/power: increase max cores to 256
> 
> 

Reviewed-by: Anatoly Burakov <anatoly.burakov@intel.com>
Thomas Monjalon Dec. 19, 2018, 9:25 p.m. UTC | #2 
14/12/2018 14:37, Burakov, Anatoly:
> On 14-Dec-18 1:31 PM, David Hunt wrote:
> > First of all we convert all the relevant uint64_t's to char arrays. Then
> > we remove the unneeded mask functions that were limited to 64 cores. Also
> > extend the guest functionality and finally rause the number of supported
> > cores to something more sensible, i.e. 256.
> > 
> > v2 changes:
> >    * Updates out of Review by Anatoly.
> >    * Indentation fixes
> >    * resolved missing spinlock around config updates
> >    * simplified some loops
> >    * fixed bug in non-equivalent code
> > 
> > v3 changes:
> >    * Moved a spinlock to a more appropriate place. Was in a for-loop, made
> >      more sense to have it outside.
> > 
> > [1/4] examples/power: change 64-bit masks to arrays
> > [2/4] examples/power: remove mask functions
> > [3/4] examples/power: allow vms to use lcores over 63
> > [4/4] examples/power: increase max cores to 256
> 
> Reviewed-by: Anatoly Burakov <anatoly.burakov@intel.com>

Applied, thanks