On 11/19/2018 04:58 AM, Thomas Gleixner wrote:
>> +DECLARE_STATIC_KEY_TRUE(cpu_smt_enabled);
>
> And here it is declared unconditionally which allows code to use it despite
> CONFIG_HOTPLUG_SMT=n and subsequently breaks the build.
Will put this under #ifdef CONFIG_HOTPLUG_SMT
Tim
On 11/19/2018 04:58 AM, Thomas Gleixner wrote:
>> +DECLARE_STATIC_KEY_TRUE(cpu_smt_enabled);
>
> And here it is declared unconditionally which allows code to use it despite
> CONFIG_HOTPLUG_SMT=n and subsequently breaks the build.
Will put this under #ifdef CONFIG_HOTPLUG_SMT
Tim
Thomas,
>>
>> +#ifdef CONFIG_SCHED_SMT
>> +
>> +extern struct static_key_false sched_smt_present;
>> +
>> +static inline bool cpu_smt_present(void)
>> +{
>> +if (static_branch_unlikely(_smt_present))
>> +return true;
>> +else
>> +return false;
>
> What's wrong
Thomas,
>>
>> +#ifdef CONFIG_SCHED_SMT
>> +
>> +extern struct static_key_false sched_smt_present;
>> +
>> +static inline bool cpu_smt_present(void)
>> +{
>> +if (static_branch_unlikely(_smt_present))
>> +return true;
>> +else
>> +return false;
>
> What's wrong
Tim,
On Mon, 19 Nov 2018, Tim Chen wrote:
> On 11/19/2018 06:57 AM, Peter Zijlstra wrote:
> > In particular, the SMT topology bits are set before we enable
> > interrrupts and similarly, are cleared after we disable interrupts for
> > the last time and die.
>
>
> Peter & Thomas,
>
> Any
Tim,
On Mon, 19 Nov 2018, Tim Chen wrote:
> On 11/19/2018 06:57 AM, Peter Zijlstra wrote:
> > In particular, the SMT topology bits are set before we enable
> > interrrupts and similarly, are cleared after we disable interrupts for
> > the last time and die.
>
>
> Peter & Thomas,
>
> Any
On 11/19/2018 06:57 AM, Peter Zijlstra wrote:
> On Fri, Nov 16, 2018 at 05:53:51PM -0800, Tim Chen wrote:
>> In later code, STIBP will be turned on/off in the context switch code
>> path when SMT is enabled. Checks for SMT is best
>> avoided on such hot paths.
>>
>> Create cpu_smt_enabled static
On 11/19/2018 06:57 AM, Peter Zijlstra wrote:
> On Fri, Nov 16, 2018 at 05:53:51PM -0800, Tim Chen wrote:
>> In later code, STIBP will be turned on/off in the context switch code
>> path when SMT is enabled. Checks for SMT is best
>> avoided on such hot paths.
>>
>> Create cpu_smt_enabled static
On Fri, Nov 16, 2018 at 05:53:51PM -0800, Tim Chen wrote:
> In later code, STIBP will be turned on/off in the context switch code
> path when SMT is enabled. Checks for SMT is best
> avoided on such hot paths.
>
> Create cpu_smt_enabled static key to turn on such SMT specific code
> statically.
On Fri, Nov 16, 2018 at 05:53:51PM -0800, Tim Chen wrote:
> In later code, STIBP will be turned on/off in the context switch code
> path when SMT is enabled. Checks for SMT is best
> avoided on such hot paths.
>
> Create cpu_smt_enabled static key to turn on such SMT specific code
> statically.
On Fri, 16 Nov 2018, Tim Chen wrote:
> In later code, STIBP will be turned on/off in the context switch code
> path when SMT is enabled. Checks for SMT is best
> avoided on such hot paths.
>
> Create cpu_smt_enabled static key to turn on such SMT specific code
> statically.
>
> This key is set
On Fri, 16 Nov 2018, Tim Chen wrote:
> In later code, STIBP will be turned on/off in the context switch code
> path when SMT is enabled. Checks for SMT is best
> avoided on such hot paths.
>
> Create cpu_smt_enabled static key to turn on such SMT specific code
> statically.
>
> This key is set
In later code, STIBP will be turned on/off in the context switch code
path when SMT is enabled. Checks for SMT is best
avoided on such hot paths.
Create cpu_smt_enabled static key to turn on such SMT specific code
statically.
This key is set in code under hot plug, so it is limited in
scope to
In later code, STIBP will be turned on/off in the context switch code
path when SMT is enabled. Checks for SMT is best
avoided on such hot paths.
Create cpu_smt_enabled static key to turn on such SMT specific code
statically.
This key is set in code under hot plug, so it is limited in
scope to
14 matches
Mail list logo