At 14:10 + on 21 Jan (1421845837), Jan Beulich wrote:
> >>> On 21.01.15 at 13:21, wrote:
> > On 19/01/15 15:58, Jan Beulich wrote:
> >> --- a/xen/common/core_parking.c
> >> +++ b/xen/common/core_parking.c
> >> @@ -75,11 +75,10 @@ static unsigned int core_parking_perform
> >> if (
On 01/21/2015 03:06 PM, Jan Beulich wrote:
On 21.01.15 at 15:42, wrote:
>> On 01/21/2015 02:35 PM, Jan Beulich wrote:
>> On 21.01.15 at 15:28, wrote:
On 01/19/2015 03:58 PM, Jan Beulich wrote:
> @@ -780,10 +780,7 @@ rt_schedule(const struct scheduler *ops,
> }
>
>>> On 21.01.15 at 15:42, wrote:
> On 01/21/2015 02:35 PM, Jan Beulich wrote:
> On 21.01.15 at 15:28, wrote:
>>> On 01/19/2015 03:58 PM, Jan Beulich wrote:
@@ -780,10 +780,7 @@ rt_schedule(const struct scheduler *ops,
}
else
{
-cpumask_t cur_cpu
On 01/21/2015 02:35 PM, Jan Beulich wrote:
On 21.01.15 at 15:28, wrote:
>> On 01/19/2015 03:58 PM, Jan Beulich wrote:
>>> Using atomic (LOCKed on x86) bitops for certain of the operations on
>>> cpumask_t is overkill when the variables aren't concurrently accessible
>>> (e.g. local function v
>>> On 21.01.15 at 15:28, wrote:
> On 01/19/2015 03:58 PM, Jan Beulich wrote:
>> Using atomic (LOCKed on x86) bitops for certain of the operations on
>> cpumask_t is overkill when the variables aren't concurrently accessible
>> (e.g. local function variables, or due to explicit locking). Introduce
On 01/19/2015 03:58 PM, Jan Beulich wrote:
> Using atomic (LOCKed on x86) bitops for certain of the operations on
> cpumask_t is overkill when the variables aren't concurrently accessible
> (e.g. local function variables, or due to explicit locking). Introduce
> alternatives using non-atomic bitops
>>> On 21.01.15 at 13:21, wrote:
> On 19/01/15 15:58, Jan Beulich wrote:
>> --- a/xen/common/core_parking.c
>> +++ b/xen/common/core_parking.c
>> @@ -75,11 +75,10 @@ static unsigned int core_parking_perform
>> if ( core_weight < core_tmp )
>> {
>> core_we
On 19/01/15 15:58, Jan Beulich wrote:
> --- a/xen/common/core_parking.c
> +++ b/xen/common/core_parking.c
> @@ -75,11 +75,10 @@ static unsigned int core_parking_perform
> if ( core_weight < core_tmp )
> {
> core_weight = core_tmp;
> -cpumas
Using atomic (LOCKed on x86) bitops for certain of the operations on
cpumask_t is overkill when the variables aren't concurrently accessible
(e.g. local function variables, or due to explicit locking). Introduce
alternatives using non-atomic bitops and use them where appropriate.
Signed-off-by: Ja
