| Message ID | 20240627-add-cpu-type-v2-2-f927bde83ad0@linux.intel.com (mailing list archive) |
|---|---|
| State | Handled Elsewhere, archived |
| Headers | show |
| Series | Add CPU-type to topology | expand |
On Thu, Jun 27, 2024 at 10:44 PM Pawan Gupta <pawan.kumar.gupta@linux.intel.com> wrote: > > Intel pstate driver uses hybrid_get_type() to get the cpu-type of a given > CPU. It uses smp_call_function_single() which is sub-optimal. Avoid it by > using topology_hw_cpu_type(cpu) that returns the cached cpu-type. > > Suggested-by: Dave Hansen <dave.hansen@linux.intel.com> > Signed-off-by: Pawan Gupta <pawan.kumar.gupta@linux.intel.com> > Acked-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com> Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> and I'm assuming that it will be routed along with the rest of the series. Thanks! > --- > drivers/cpufreq/intel_pstate.c | 14 +++----------- > 1 file changed, 3 insertions(+), 11 deletions(-) > > diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c > index 15de5e3d96fd..0a1e832c7536 100644 > --- a/drivers/cpufreq/intel_pstate.c > +++ b/drivers/cpufreq/intel_pstate.c > @@ -1956,24 +1956,16 @@ static int knl_get_turbo_pstate(int cpu) > return ret; > } > > -static void hybrid_get_type(void *data) > -{ > - u8 *cpu_type = data; > - > - *cpu_type = get_this_hybrid_cpu_type(); > -} > - > static int hwp_get_cpu_scaling(int cpu) > { > - u8 cpu_type = 0; > + u8 cpu_type = topology_hw_cpu_type(cpu); > > - smp_call_function_single(cpu, hybrid_get_type, &cpu_type, 1); > /* P-cores have a smaller perf level-to-freqency scaling factor. */ > - if (cpu_type == 0x40) > + if (cpu_type == X86_HW_CPU_TYPE_INTEL_CORE) > return hybrid_scaling_factor; > > /* Use default core scaling for E-cores */ > - if (cpu_type == 0x20) > + if (cpu_type == X86_HW_CPU_TYPE_INTEL_ATOM) > return core_get_scaling(); > > /* > > --
On Mon, Jul 01, 2024 at 07:08:41PM +0200, Rafael J. Wysocki wrote: > On Thu, Jun 27, 2024 at 10:44 PM Pawan Gupta > <pawan.kumar.gupta@linux.intel.com> wrote: > > > > Intel pstate driver uses hybrid_get_type() to get the cpu-type of a given > > CPU. It uses smp_call_function_single() which is sub-optimal. Avoid it by > > using topology_hw_cpu_type(cpu) that returns the cached cpu-type. > > > > Suggested-by: Dave Hansen <dave.hansen@linux.intel.com> > > Signed-off-by: Pawan Gupta <pawan.kumar.gupta@linux.intel.com> > > Acked-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com> > > Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Thanks. > and I'm assuming that it will be routed along with the rest of the series. I believe yes, as it has a dependency on the first patch.
diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c index 15de5e3d96fd..0a1e832c7536 100644 --- a/drivers/cpufreq/intel_pstate.c +++ b/drivers/cpufreq/intel_pstate.c @@ -1956,24 +1956,16 @@ static int knl_get_turbo_pstate(int cpu) return ret; } -static void hybrid_get_type(void *data) -{ - u8 *cpu_type = data; - - *cpu_type = get_this_hybrid_cpu_type(); -} - static int hwp_get_cpu_scaling(int cpu) { - u8 cpu_type = 0; + u8 cpu_type = topology_hw_cpu_type(cpu); - smp_call_function_single(cpu, hybrid_get_type, &cpu_type, 1); /* P-cores have a smaller perf level-to-freqency scaling factor. */ - if (cpu_type == 0x40) + if (cpu_type == X86_HW_CPU_TYPE_INTEL_CORE) return hybrid_scaling_factor; /* Use default core scaling for E-cores */ - if (cpu_type == 0x20) + if (cpu_type == X86_HW_CPU_TYPE_INTEL_ATOM) return core_get_scaling(); /*