Re: [PATCH v3 1/8] sched, x86: Add SD_ASYM_PACKING flags to x86 cpu topology for ITMT

[Tim Chen]

On Mon, 2016-09-19 at 11:11 +0200, Jiri Olsa wrote:
> On Tue, Sep 13, 2016 at 09:26:13AM +0200, Peter Zijlstra wrote:
> > 
> > On Mon, Sep 12, 2016 at 01:50:15PM -0700, Tim Chen wrote:
> > > 
> > > > 
> > > > No. We first rework the code so that the IMT stuff can be added in a 
> > > > later
> > > > patch easily.
> > > I'll move this patch to come after current patch 5.
> > Split the patch, one doing the topo cleanup, one adding the
> > x86_*_flags().
> hi,
> could you please CC me on the next version,
> I'm interested in that cleanup patch..

Sure. Will do.

Tim

Re: [PATCH v3 1/8] sched, x86: Add SD_ASYM_PACKING flags to x86 cpu topology for ITMT

[Tim Chen]

On Mon, 2016-09-19 at 11:11 +0200, Jiri Olsa wrote:
> On Tue, Sep 13, 2016 at 09:26:13AM +0200, Peter Zijlstra wrote:
> > 
> > On Mon, Sep 12, 2016 at 01:50:15PM -0700, Tim Chen wrote:
> > > 
> > > > 
> > > > No. We first rework the code so that the IMT stuff can be added in a 
> > > > later
> > > > patch easily.
> > > I'll move this patch to come after current patch 5.
> > Split the patch, one doing the topo cleanup, one adding the
> > x86_*_flags().
> hi,
> could you please CC me on the next version,
> I'm interested in that cleanup patch..

Sure. Will do.

Tim

Re: [PATCH v3 1/8] sched, x86: Add SD_ASYM_PACKING flags to x86 cpu topology for ITMT

[Jiri Olsa]

On Tue, Sep 13, 2016 at 09:26:13AM +0200, Peter Zijlstra wrote:
> On Mon, Sep 12, 2016 at 01:50:15PM -0700, Tim Chen wrote:
> > > No. We first rework the code so that the IMT stuff can be added in a later
> > > patch easily.
> > 
> > I'll move this patch to come after current patch 5.
> 
> Split the patch, one doing the topo cleanup, one adding the
> x86_*_flags().

hi,
could you please CC me on the next version,
I'm interested in that cleanup patch..

thanks a lot,
jirka

Re: [PATCH v3 1/8] sched, x86: Add SD_ASYM_PACKING flags to x86 cpu topology for ITMT

[Jiri Olsa]

On Tue, Sep 13, 2016 at 09:26:13AM +0200, Peter Zijlstra wrote:
> On Mon, Sep 12, 2016 at 01:50:15PM -0700, Tim Chen wrote:
> > > No. We first rework the code so that the IMT stuff can be added in a later
> > > patch easily.
> > 
> > I'll move this patch to come after current patch 5.
> 
> Split the patch, one doing the topo cleanup, one adding the
> x86_*_flags().

hi,
could you please CC me on the next version,
I'm interested in that cleanup patch..

thanks a lot,
jirka

Re: [PATCH v3 1/8] sched, x86: Add SD_ASYM_PACKING flags to x86 cpu topology for ITMT

[Peter Zijlstra]

On Mon, Sep 12, 2016 at 01:50:15PM -0700, Tim Chen wrote:
> > No. We first rework the code so that the IMT stuff can be added in a later
> > patch easily.
> 
> I'll move this patch to come after current patch 5.

Split the patch, one doing the topo cleanup, one adding the
x86_*_flags().

Re: [PATCH v3 1/8] sched, x86: Add SD_ASYM_PACKING flags to x86 cpu topology for ITMT

[Peter Zijlstra]

On Mon, Sep 12, 2016 at 01:50:15PM -0700, Tim Chen wrote:
> > No. We first rework the code so that the IMT stuff can be added in a later
> > patch easily.
> 
> I'll move this patch to come after current patch 5.

Split the patch, one doing the topo cleanup, one adding the
x86_*_flags().

Re: [PATCH v3 1/8] sched, x86: Add SD_ASYM_PACKING flags to x86 cpu topology for ITMT

[Tim Chen]

On Sat, 2016-09-10 at 15:10 +0200, Thomas Gleixner wrote:
> On Thu, 8 Sep 2016, Srinivas Pandruvada wrote:
> > 
> > From: Tim Chen 
> > 
> > We uses ASYM_PACKING feature in the scheduler to move tasks to more
> > capable cpus that can be boosted to higher frequency. This is enabled by
> > Intel Turbo Boost Max Technology 3.0 (ITMT).  We mark the sched domain
> > topology level with SD_ASYM_PACKING flag for such systems to indicate
> > scheduler can use the ASYM_PACKING feature to move load to the
> > more capable cpus.
> Sigh. This changelog does not tell anything about the nature of the patch,
> the rationale for it etc. It's just a meaningless blurb.

Okay, I can add more details about ITMT.  Will also be clearer if
the ITMT patch (patch 5) comes before this one.



> > +}
> > +#else
> > +#define x86_core_flags cpu_core_flags
> > +#define x86_smt_flags cpu_smt_flags
> > +#endif
> No. We first rework the code so that the IMT stuff can be added in a later
> patch easily.

I'll move this patch to come after current patch 5.

Thanks.

Tim

Re: [PATCH v3 1/8] sched, x86: Add SD_ASYM_PACKING flags to x86 cpu topology for ITMT

[Tim Chen]

On Sat, 2016-09-10 at 15:10 +0200, Thomas Gleixner wrote:
> On Thu, 8 Sep 2016, Srinivas Pandruvada wrote:
> > 
> > From: Tim Chen 
> > 
> > We uses ASYM_PACKING feature in the scheduler to move tasks to more
> > capable cpus that can be boosted to higher frequency. This is enabled by
> > Intel Turbo Boost Max Technology 3.0 (ITMT).  We mark the sched domain
> > topology level with SD_ASYM_PACKING flag for such systems to indicate
> > scheduler can use the ASYM_PACKING feature to move load to the
> > more capable cpus.
> Sigh. This changelog does not tell anything about the nature of the patch,
> the rationale for it etc. It's just a meaningless blurb.

Okay, I can add more details about ITMT.  Will also be clearer if
the ITMT patch (patch 5) comes before this one.



> > +}
> > +#else
> > +#define x86_core_flags cpu_core_flags
> > +#define x86_smt_flags cpu_smt_flags
> > +#endif
> No. We first rework the code so that the IMT stuff can be added in a later
> patch easily.

I'll move this patch to come after current patch 5.

Thanks.

Tim

Re: [PATCH v3 1/8] sched, x86: Add SD_ASYM_PACKING flags to x86 cpu topology for ITMT

[Thomas Gleixner]

On Thu, 8 Sep 2016, Srinivas Pandruvada wrote:
>  
>  void set_sched_topology(struct sched_domain_topology_level *tl)
>  {
> + if (WARN_ON_ONCE(sched_smp_initialized))
> + return;
> +
>   sched_domain_topology = tl;
>  }

This change has nothing to do with $subject. Seperate patch please.

Thanks,

tglx

Re: [PATCH v3 1/8] sched, x86: Add SD_ASYM_PACKING flags to x86 cpu topology for ITMT

[Thomas Gleixner]

On Thu, 8 Sep 2016, Srinivas Pandruvada wrote:
>  
>  void set_sched_topology(struct sched_domain_topology_level *tl)
>  {
> + if (WARN_ON_ONCE(sched_smp_initialized))
> + return;
> +
>   sched_domain_topology = tl;
>  }

This change has nothing to do with $subject. Seperate patch please.

Thanks,

tglx

Re: [PATCH v3 1/8] sched, x86: Add SD_ASYM_PACKING flags to x86 cpu topology for ITMT

[Thomas Gleixner]

On Thu, 8 Sep 2016, Srinivas Pandruvada wrote:
> From: Tim Chen 
> 
> We uses ASYM_PACKING feature in the scheduler to move tasks to more
> capable cpus that can be boosted to higher frequency. This is enabled by
> Intel Turbo Boost Max Technology 3.0 (ITMT).  We mark the sched domain
> topology level with SD_ASYM_PACKING flag for such systems to indicate
> scheduler can use the ASYM_PACKING feature to move load to the
> more capable cpus.

Sigh. This changelog does not tell anything about the nature of the patch,
the rationale for it etc. It's just a meaningless blurb.


> +unsigned int __read_mostly sysctl_sched_itmt_enabled;
> +
>  static inline void smpboot_setup_warm_reset_vector(unsigned long start_eip)
>  {
>   unsigned long flags;
> @@ -471,31 +473,57 @@ static bool match_die(struct cpuinfo_x86 *c, struct 
> cpuinfo_x86 *o)
>   return false;
>  }
>  
> -static struct sched_domain_topology_level numa_inside_package_topology[] = {
> +#ifdef CONFIG_SCHED_ITMT
> +static int x86_core_flags(void)
> +{
> + int flags = cpu_core_flags();
> +
> + if (sysctl_sched_itmt_enabled)
> + flags |= SD_ASYM_PACKING;
> +
> + return flags;
> +}
> +
> +static int x86_smt_flags(void)
> +{
> + int flags = cpu_smt_flags();
> +
> + if (sysctl_sched_itmt_enabled)
> + flags |= SD_ASYM_PACKING;
> +
> + return flags;
> +}
> +#else
> +#define x86_core_flags cpu_core_flags
> +#define x86_smt_flags cpu_smt_flags
> +#endif

No. We first rework the code so that the IMT stuff can be added in a later
patch easily.

Thanks,

tglx

Re: [PATCH v3 1/8] sched, x86: Add SD_ASYM_PACKING flags to x86 cpu topology for ITMT

[Thomas Gleixner]

On Thu, 8 Sep 2016, Srinivas Pandruvada wrote:
> From: Tim Chen 
> 
> We uses ASYM_PACKING feature in the scheduler to move tasks to more
> capable cpus that can be boosted to higher frequency. This is enabled by
> Intel Turbo Boost Max Technology 3.0 (ITMT).  We mark the sched domain
> topology level with SD_ASYM_PACKING flag for such systems to indicate
> scheduler can use the ASYM_PACKING feature to move load to the
> more capable cpus.

Sigh. This changelog does not tell anything about the nature of the patch,
the rationale for it etc. It's just a meaningless blurb.


> +unsigned int __read_mostly sysctl_sched_itmt_enabled;
> +
>  static inline void smpboot_setup_warm_reset_vector(unsigned long start_eip)
>  {
>   unsigned long flags;
> @@ -471,31 +473,57 @@ static bool match_die(struct cpuinfo_x86 *c, struct 
> cpuinfo_x86 *o)
>   return false;
>  }
>  
> -static struct sched_domain_topology_level numa_inside_package_topology[] = {
> +#ifdef CONFIG_SCHED_ITMT
> +static int x86_core_flags(void)
> +{
> + int flags = cpu_core_flags();
> +
> + if (sysctl_sched_itmt_enabled)
> + flags |= SD_ASYM_PACKING;
> +
> + return flags;
> +}
> +
> +static int x86_smt_flags(void)
> +{
> + int flags = cpu_smt_flags();
> +
> + if (sysctl_sched_itmt_enabled)
> + flags |= SD_ASYM_PACKING;
> +
> + return flags;
> +}
> +#else
> +#define x86_core_flags cpu_core_flags
> +#define x86_smt_flags cpu_smt_flags
> +#endif

No. We first rework the code so that the IMT stuff can be added in a later
patch easily.

Thanks,

tglx