* [PATCH v3] x86, sched: Fix the AMD CPPC maximum perf on some specific generations
@ 2021-04-23 2:39 Huang Rui
2021-04-23 12:09 ` Rafael J. Wysocki
0 siblings, 1 reply; 7+ messages in thread
From: Huang Rui @ 2021-04-23 2:39 UTC (permalink / raw)
To: linux-pm, linux-kernel
Cc: Alex Deucher, Jason Bagavatsingham, Pierre-Loup A . Griffais,
Huang Rui, Nathan Fontenot, Rafael J . Wysocki, Borislav Petkov,
x86, stable
Some AMD Ryzen generations has different calculation method on maximum
perf. 255 is not for all asics, some specific generations should use 166
as the maximum perf. Otherwise, it will report incorrect frequency value
like below:
~ → lscpu | grep MHz
CPU MHz: 3400.000
CPU max MHz: 7228.3198
CPU min MHz: 2200.0000
Fixes: 41ea667227ba ("x86, sched: Calculate frequency invariance for AMD systems")
Fixes: 3c55e94c0ade ("cpufreq: ACPI: Extend frequency tables to cover boost frequencies")
Reported-by: Jason Bagavatsingham <jason.bagavatsingham@gmail.com>
Tested-by: Jason Bagavatsingham <jason.bagavatsingham@gmail.com>
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=211791
Signed-off-by: Huang Rui <ray.huang@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Nathan Fontenot <nathan.fontenot@amd.com>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: x86@kernel.org
Cc: stable@vger.kernel.org
---
Changes from V1 -> V2:
- Enhance the commit message.
- Move amd_get_highest_perf() into amd.c.
- Refine the implementation of switch-case.
- Cc stable mail list.
Changes from V2 -> V3:
- Move the update into cppc_get_perf_caps() to correct the highest perf value in
the API.
---
arch/x86/include/asm/processor.h | 2 ++
arch/x86/kernel/cpu/amd.c | 22 ++++++++++++++++++++++
drivers/acpi/cppc_acpi.c | 8 ++++++--
3 files changed, 30 insertions(+), 2 deletions(-)
diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
index f1b9ed5efaa9..908bcaea1361 100644
--- a/arch/x86/include/asm/processor.h
+++ b/arch/x86/include/asm/processor.h
@@ -804,8 +804,10 @@ DECLARE_PER_CPU(u64, msr_misc_features_shadow);
#ifdef CONFIG_CPU_SUP_AMD
extern u32 amd_get_nodes_per_socket(void);
+extern u32 amd_get_highest_perf(void);
#else
static inline u32 amd_get_nodes_per_socket(void) { return 0; }
+static inline u32 amd_get_highest_perf(void) { return 0; }
#endif
static inline uint32_t hypervisor_cpuid_base(const char *sig, uint32_t leaves)
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index 347a956f71ca..aadb691d9357 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -1170,3 +1170,25 @@ void set_dr_addr_mask(unsigned long mask, int dr)
break;
}
}
+
+u32 amd_get_highest_perf(void)
+{
+ struct cpuinfo_x86 *c = &boot_cpu_data;
+ u32 cppc_max_perf = 225;
+
+ switch (c->x86) {
+ case 0x17:
+ if ((c->x86_model >= 0x30 && c->x86_model < 0x40) ||
+ (c->x86_model >= 0x70 && c->x86_model < 0x80))
+ cppc_max_perf = 166;
+ break;
+ case 0x19:
+ if ((c->x86_model >= 0x20 && c->x86_model < 0x30) ||
+ (c->x86_model >= 0x40 && c->x86_model < 0x70))
+ cppc_max_perf = 166;
+ break;
+ }
+
+ return cppc_max_perf;
+}
+EXPORT_SYMBOL_GPL(amd_get_highest_perf);
diff --git a/drivers/acpi/cppc_acpi.c b/drivers/acpi/cppc_acpi.c
index 69057fcd2c04..58e72b6e222f 100644
--- a/drivers/acpi/cppc_acpi.c
+++ b/drivers/acpi/cppc_acpi.c
@@ -1107,8 +1107,12 @@ int cppc_get_perf_caps(int cpunum, struct cppc_perf_caps *perf_caps)
}
}
- cpc_read(cpunum, highest_reg, &high);
- perf_caps->highest_perf = high;
+ if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) {
+ perf_caps->highest_perf = amd_get_highest_perf();
+ } else {
+ cpc_read(cpunum, highest_reg, &high);
+ perf_caps->highest_perf = high;
+ }
cpc_read(cpunum, lowest_reg, &low);
perf_caps->lowest_perf = low;
--
2.25.1
^ permalink raw reply related [flat|nested] 7+ messages in thread* Re: [PATCH v3] x86, sched: Fix the AMD CPPC maximum perf on some specific generations 2021-04-23 2:39 [PATCH v3] x86, sched: Fix the AMD CPPC maximum perf on some specific generations Huang Rui @ 2021-04-23 12:09 ` Rafael J. Wysocki 2021-04-23 12:52 ` Huang Rui 0 siblings, 1 reply; 7+ messages in thread From: Rafael J. Wysocki @ 2021-04-23 12:09 UTC (permalink / raw) To: Huang Rui Cc: Linux PM, Linux Kernel Mailing List, Alex Deucher, Jason Bagavatsingham, Pierre-Loup A . Griffais, Nathan Fontenot, Rafael J . Wysocki, Borislav Petkov, the arch/x86 maintainers, Stable On Fri, Apr 23, 2021 at 4:40 AM Huang Rui <ray.huang@amd.com> wrote: > > Some AMD Ryzen generations has different calculation method on maximum > perf. 255 is not for all asics, some specific generations should use 166 > as the maximum perf. Otherwise, it will report incorrect frequency value > like below: > > ~ → lscpu | grep MHz > CPU MHz: 3400.000 > CPU max MHz: 7228.3198 > CPU min MHz: 2200.0000 > > Fixes: 41ea667227ba ("x86, sched: Calculate frequency invariance for AMD systems") > Fixes: 3c55e94c0ade ("cpufreq: ACPI: Extend frequency tables to cover boost frequencies") > > Reported-by: Jason Bagavatsingham <jason.bagavatsingham@gmail.com> > Tested-by: Jason Bagavatsingham <jason.bagavatsingham@gmail.com> > Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=211791 > Signed-off-by: Huang Rui <ray.huang@amd.com> > Cc: Alex Deucher <alexander.deucher@amd.com> > Cc: Nathan Fontenot <nathan.fontenot@amd.com> > Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> > Cc: Borislav Petkov <bp@suse.de> > Cc: x86@kernel.org > Cc: stable@vger.kernel.org > --- > > Changes from V1 -> V2: > - Enhance the commit message. > - Move amd_get_highest_perf() into amd.c. > - Refine the implementation of switch-case. > - Cc stable mail list. > > Changes from V2 -> V3: > - Move the update into cppc_get_perf_caps() to correct the highest perf value in > the API. > > --- > arch/x86/include/asm/processor.h | 2 ++ > arch/x86/kernel/cpu/amd.c | 22 ++++++++++++++++++++++ > drivers/acpi/cppc_acpi.c | 8 ++++++-- > 3 files changed, 30 insertions(+), 2 deletions(-) > > diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h > index f1b9ed5efaa9..908bcaea1361 100644 > --- a/arch/x86/include/asm/processor.h > +++ b/arch/x86/include/asm/processor.h > @@ -804,8 +804,10 @@ DECLARE_PER_CPU(u64, msr_misc_features_shadow); > > #ifdef CONFIG_CPU_SUP_AMD > extern u32 amd_get_nodes_per_socket(void); > +extern u32 amd_get_highest_perf(void); > #else > static inline u32 amd_get_nodes_per_socket(void) { return 0; } > +static inline u32 amd_get_highest_perf(void) { return 0; } > #endif > > static inline uint32_t hypervisor_cpuid_base(const char *sig, uint32_t leaves) > diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c > index 347a956f71ca..aadb691d9357 100644 > --- a/arch/x86/kernel/cpu/amd.c > +++ b/arch/x86/kernel/cpu/amd.c > @@ -1170,3 +1170,25 @@ void set_dr_addr_mask(unsigned long mask, int dr) > break; > } > } > + > +u32 amd_get_highest_perf(void) > +{ > + struct cpuinfo_x86 *c = &boot_cpu_data; > + u32 cppc_max_perf = 225; > + > + switch (c->x86) { > + case 0x17: > + if ((c->x86_model >= 0x30 && c->x86_model < 0x40) || > + (c->x86_model >= 0x70 && c->x86_model < 0x80)) > + cppc_max_perf = 166; > + break; > + case 0x19: > + if ((c->x86_model >= 0x20 && c->x86_model < 0x30) || > + (c->x86_model >= 0x40 && c->x86_model < 0x70)) > + cppc_max_perf = 166; > + break; > + } > + > + return cppc_max_perf; > +} > +EXPORT_SYMBOL_GPL(amd_get_highest_perf); > diff --git a/drivers/acpi/cppc_acpi.c b/drivers/acpi/cppc_acpi.c > index 69057fcd2c04..58e72b6e222f 100644 > --- a/drivers/acpi/cppc_acpi.c > +++ b/drivers/acpi/cppc_acpi.c > @@ -1107,8 +1107,12 @@ int cppc_get_perf_caps(int cpunum, struct cppc_perf_caps *perf_caps) > } > } > > - cpc_read(cpunum, highest_reg, &high); > - perf_caps->highest_perf = high; > + if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) { This is a generic arch-independent file. Can we avoid adding the x86-specific check here? > + perf_caps->highest_perf = amd_get_highest_perf(); > + } else { > + cpc_read(cpunum, highest_reg, &high); > + perf_caps->highest_perf = high; > + } > > cpc_read(cpunum, lowest_reg, &low); > perf_caps->lowest_perf = low; > -- > 2.25.1 > ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v3] x86, sched: Fix the AMD CPPC maximum perf on some specific generations 2021-04-23 12:09 ` Rafael J. Wysocki @ 2021-04-23 12:52 ` Huang Rui 2021-04-23 13:53 ` Rafael J. Wysocki 0 siblings, 1 reply; 7+ messages in thread From: Huang Rui @ 2021-04-23 12:52 UTC (permalink / raw) To: Rafael J. Wysocki Cc: Linux PM, Linux Kernel Mailing List, Deucher, Alexander, Jason Bagavatsingham, Pierre-Loup A . Griffais, Fontenot, Nathan, Rafael J . Wysocki, Borislav Petkov, the arch/x86 maintainers, Stable On Fri, Apr 23, 2021 at 08:09:49PM +0800, Rafael J. Wysocki wrote: > On Fri, Apr 23, 2021 at 4:40 AM Huang Rui <ray.huang@amd.com> wrote: > > > > Some AMD Ryzen generations has different calculation method on maximum > > perf. 255 is not for all asics, some specific generations should use 166 > > as the maximum perf. Otherwise, it will report incorrect frequency value > > like below: > > > > ~ ^[$B"*^[(B lscpu | grep MHz > > CPU MHz: 3400.000 > > CPU max MHz: 7228.3198 > > CPU min MHz: 2200.0000 > > > > Fixes: 41ea667227ba ("x86, sched: Calculate frequency invariance for AMD systems") > > Fixes: 3c55e94c0ade ("cpufreq: ACPI: Extend frequency tables to cover boost frequencies") > > > > Reported-by: Jason Bagavatsingham <jason.bagavatsingham@gmail.com> > > Tested-by: Jason Bagavatsingham <jason.bagavatsingham@gmail.com> > > Bugzilla: https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fbugzilla.kernel.org%2Fshow_bug.cgi%3Fid%3D211791&data=04%7C01%7Cray.huang%40amd.com%7Ce9ed877387fc4b7431e108d90650b98f%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637547766057950380%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=R%2FSBLaYOhTjrli%2BT054EytKeh8VmN7ryOQuQW4mgz6M%3D&reserved=0 > > Signed-off-by: Huang Rui <ray.huang@amd.com> > > Cc: Alex Deucher <alexander.deucher@amd.com> > > Cc: Nathan Fontenot <nathan.fontenot@amd.com> > > Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> > > Cc: Borislav Petkov <bp@suse.de> > > Cc: x86@kernel.org > > Cc: stable@vger.kernel.org > > --- > > > > Changes from V1 -> V2: > > - Enhance the commit message. > > - Move amd_get_highest_perf() into amd.c. > > - Refine the implementation of switch-case. > > - Cc stable mail list. > > > > Changes from V2 -> V3: > > - Move the update into cppc_get_perf_caps() to correct the highest perf value in > > the API. > > > > --- > > arch/x86/include/asm/processor.h | 2 ++ > > arch/x86/kernel/cpu/amd.c | 22 ++++++++++++++++++++++ > > drivers/acpi/cppc_acpi.c | 8 ++++++-- > > 3 files changed, 30 insertions(+), 2 deletions(-) > > > > diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h > > index f1b9ed5efaa9..908bcaea1361 100644 > > --- a/arch/x86/include/asm/processor.h > > +++ b/arch/x86/include/asm/processor.h > > @@ -804,8 +804,10 @@ DECLARE_PER_CPU(u64, msr_misc_features_shadow); > > > > #ifdef CONFIG_CPU_SUP_AMD > > extern u32 amd_get_nodes_per_socket(void); > > +extern u32 amd_get_highest_perf(void); > > #else > > static inline u32 amd_get_nodes_per_socket(void) { return 0; } > > +static inline u32 amd_get_highest_perf(void) { return 0; } > > #endif > > > > static inline uint32_t hypervisor_cpuid_base(const char *sig, uint32_t leaves) > > diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c > > index 347a956f71ca..aadb691d9357 100644 > > --- a/arch/x86/kernel/cpu/amd.c > > +++ b/arch/x86/kernel/cpu/amd.c > > @@ -1170,3 +1170,25 @@ void set_dr_addr_mask(unsigned long mask, int dr) > > break; > > } > > } > > + > > +u32 amd_get_highest_perf(void) > > +{ > > + struct cpuinfo_x86 *c = &boot_cpu_data; > > + u32 cppc_max_perf = 225; > > + > > + switch (c->x86) { > > + case 0x17: > > + if ((c->x86_model >= 0x30 && c->x86_model < 0x40) || > > + (c->x86_model >= 0x70 && c->x86_model < 0x80)) > > + cppc_max_perf = 166; > > + break; > > + case 0x19: > > + if ((c->x86_model >= 0x20 && c->x86_model < 0x30) || > > + (c->x86_model >= 0x40 && c->x86_model < 0x70)) > > + cppc_max_perf = 166; > > + break; > > + } > > + > > + return cppc_max_perf; > > +} > > +EXPORT_SYMBOL_GPL(amd_get_highest_perf); > > diff --git a/drivers/acpi/cppc_acpi.c b/drivers/acpi/cppc_acpi.c > > index 69057fcd2c04..58e72b6e222f 100644 > > --- a/drivers/acpi/cppc_acpi.c > > +++ b/drivers/acpi/cppc_acpi.c > > @@ -1107,8 +1107,12 @@ int cppc_get_perf_caps(int cpunum, struct cppc_perf_caps *perf_caps) > > } > > } > > > > - cpc_read(cpunum, highest_reg, &high); > > - perf_caps->highest_perf = high; > > + if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) { > > This is a generic arch-independent file. > > Can we avoid adding the x86-specific check here? OK, I see, it will be used by ARM as well. Can I rollback to implementation of V2: https://lore.kernel.org/r/20210421023807.1540290-1-ray.huang@amd.com If stick to add quirk in cppc_acpi.c and avoid x86-specific check at the same time here, the code will not be straight forward. Or will you have any other good idea? Thanks, Ray ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v3] x86, sched: Fix the AMD CPPC maximum perf on some specific generations 2021-04-23 12:52 ` Huang Rui @ 2021-04-23 13:53 ` Rafael J. Wysocki 2021-04-23 15:07 ` Huang Rui 0 siblings, 1 reply; 7+ messages in thread From: Rafael J. Wysocki @ 2021-04-23 13:53 UTC (permalink / raw) To: Huang Rui Cc: Rafael J. Wysocki, Linux PM, Linux Kernel Mailing List, Deucher, Alexander, Jason Bagavatsingham, Pierre-Loup A . Griffais, Fontenot, Nathan, Rafael J . Wysocki, Borislav Petkov, the arch/x86 maintainers, Stable On Fri, Apr 23, 2021 at 2:52 PM Huang Rui <ray.huang@amd.com> wrote: > > On Fri, Apr 23, 2021 at 08:09:49PM +0800, Rafael J. Wysocki wrote: > > On Fri, Apr 23, 2021 at 4:40 AM Huang Rui <ray.huang@amd.com> wrote: > > > > > > Some AMD Ryzen generations has different calculation method on maximum > > > perf. 255 is not for all asics, some specific generations should use 166 > > > as the maximum perf. Otherwise, it will report incorrect frequency value > > > like below: > > > > > > ~ $B"* (B lscpu | grep MHz > > > CPU MHz: 3400.000 > > > CPU max MHz: 7228.3198 > > > CPU min MHz: 2200.0000 > > > > > > Fixes: 41ea667227ba ("x86, sched: Calculate frequency invariance for AMD systems") > > > Fixes: 3c55e94c0ade ("cpufreq: ACPI: Extend frequency tables to cover boost frequencies") > > > > > > Reported-by: Jason Bagavatsingham <jason.bagavatsingham@gmail.com> > > > Tested-by: Jason Bagavatsingham <jason.bagavatsingham@gmail.com> > > > Bugzilla: https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fbugzilla.kernel.org%2Fshow_bug.cgi%3Fid%3D211791&data=04%7C01%7Cray.huang%40amd.com%7Ce9ed877387fc4b7431e108d90650b98f%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637547766057950380%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=R%2FSBLaYOhTjrli%2BT054EytKeh8VmN7ryOQuQW4mgz6M%3D&reserved=0 > > > Signed-off-by: Huang Rui <ray.huang@amd.com> > > > Cc: Alex Deucher <alexander.deucher@amd.com> > > > Cc: Nathan Fontenot <nathan.fontenot@amd.com> > > > Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> > > > Cc: Borislav Petkov <bp@suse.de> > > > Cc: x86@kernel.org > > > Cc: stable@vger.kernel.org > > > --- > > > > > > Changes from V1 -> V2: > > > - Enhance the commit message. > > > - Move amd_get_highest_perf() into amd.c. > > > - Refine the implementation of switch-case. > > > - Cc stable mail list. > > > > > > Changes from V2 -> V3: > > > - Move the update into cppc_get_perf_caps() to correct the highest perf value in > > > the API. > > > > > > --- > > > arch/x86/include/asm/processor.h | 2 ++ > > > arch/x86/kernel/cpu/amd.c | 22 ++++++++++++++++++++++ > > > drivers/acpi/cppc_acpi.c | 8 ++++++-- > > > 3 files changed, 30 insertions(+), 2 deletions(-) > > > > > > diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h > > > index f1b9ed5efaa9..908bcaea1361 100644 > > > --- a/arch/x86/include/asm/processor.h > > > +++ b/arch/x86/include/asm/processor.h > > > @@ -804,8 +804,10 @@ DECLARE_PER_CPU(u64, msr_misc_features_shadow); > > > > > > #ifdef CONFIG_CPU_SUP_AMD > > > extern u32 amd_get_nodes_per_socket(void); > > > +extern u32 amd_get_highest_perf(void); > > > #else > > > static inline u32 amd_get_nodes_per_socket(void) { return 0; } > > > +static inline u32 amd_get_highest_perf(void) { return 0; } > > > #endif > > > > > > static inline uint32_t hypervisor_cpuid_base(const char *sig, uint32_t leaves) > > > diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c > > > index 347a956f71ca..aadb691d9357 100644 > > > --- a/arch/x86/kernel/cpu/amd.c > > > +++ b/arch/x86/kernel/cpu/amd.c > > > @@ -1170,3 +1170,25 @@ void set_dr_addr_mask(unsigned long mask, int dr) > > > break; > > > } > > > } > > > + > > > +u32 amd_get_highest_perf(void) > > > +{ > > > + struct cpuinfo_x86 *c = &boot_cpu_data; > > > + u32 cppc_max_perf = 225; > > > + > > > + switch (c->x86) { > > > + case 0x17: > > > + if ((c->x86_model >= 0x30 && c->x86_model < 0x40) || > > > + (c->x86_model >= 0x70 && c->x86_model < 0x80)) > > > + cppc_max_perf = 166; > > > + break; > > > + case 0x19: > > > + if ((c->x86_model >= 0x20 && c->x86_model < 0x30) || > > > + (c->x86_model >= 0x40 && c->x86_model < 0x70)) > > > + cppc_max_perf = 166; > > > + break; > > > + } > > > + > > > + return cppc_max_perf; > > > +} > > > +EXPORT_SYMBOL_GPL(amd_get_highest_perf); > > > diff --git a/drivers/acpi/cppc_acpi.c b/drivers/acpi/cppc_acpi.c > > > index 69057fcd2c04..58e72b6e222f 100644 > > > --- a/drivers/acpi/cppc_acpi.c > > > +++ b/drivers/acpi/cppc_acpi.c > > > @@ -1107,8 +1107,12 @@ int cppc_get_perf_caps(int cpunum, struct cppc_perf_caps *perf_caps) > > > } > > > } > > > > > > - cpc_read(cpunum, highest_reg, &high); > > > - perf_caps->highest_perf = high; > > > + if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) { > > > > This is a generic arch-independent file. > > > > Can we avoid adding the x86-specific check here? > > OK, I see, it will be used by ARM as well. > > Can I rollback to implementation of V2: > > https://lore.kernel.org/r/20210421023807.1540290-1-ray.huang@amd.com This would work IMO, but it can be simplified somewhat AFAICS. The obvious drawback is that amd_get_highest_perf() would need to be called directly wherever the CPPC highest perf is needed and the vendor may be AMD. > If stick to add quirk in cppc_acpi.c and avoid x86-specific check at the > same time here, the code will not be straight forward. Or will you have any > other good idea? Not at the moment, sorry. ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v3] x86, sched: Fix the AMD CPPC maximum perf on some specific generations 2021-04-23 13:53 ` Rafael J. Wysocki @ 2021-04-23 15:07 ` Huang Rui 2021-04-23 15:20 ` Rafael J. Wysocki 0 siblings, 1 reply; 7+ messages in thread From: Huang Rui @ 2021-04-23 15:07 UTC (permalink / raw) To: Rafael J. Wysocki Cc: Linux PM, Linux Kernel Mailing List, Deucher, Alexander, Jason Bagavatsingham, Pierre-Loup A . Griffais, Fontenot, Nathan, Rafael J . Wysocki, Borislav Petkov, the arch/x86 maintainers, Stable On Fri, Apr 23, 2021 at 09:53:37PM +0800, Rafael J. Wysocki wrote: > On Fri, Apr 23, 2021 at 2:52 PM Huang Rui <ray.huang@amd.com> wrote: > > > > On Fri, Apr 23, 2021 at 08:09:49PM +0800, Rafael J. Wysocki wrote: > > > On Fri, Apr 23, 2021 at 4:40 AM Huang Rui <ray.huang@amd.com> wrote: > > > > > > > > Some AMD Ryzen generations has different calculation method on maximum > > > > perf. 255 is not for all asics, some specific generations should use 166 > > > > as the maximum perf. Otherwise, it will report incorrect frequency value > > > > like below: > > > > > > > > ~ $B"* (B lscpu | grep MHz > > > > CPU MHz: 3400.000 > > > > CPU max MHz: 7228.3198 > > > > CPU min MHz: 2200.0000 > > > > > > > > Fixes: 41ea667227ba ("x86, sched: Calculate frequency invariance for AMD systems") > > > > Fixes: 3c55e94c0ade ("cpufreq: ACPI: Extend frequency tables to cover boost frequencies") > > > > > > > > Reported-by: Jason Bagavatsingham <jason.bagavatsingham@gmail.com> > > > > Tested-by: Jason Bagavatsingham <jason.bagavatsingham@gmail.com> > > > > Bugzilla: https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fbugzilla.kernel.org%2Fshow_bug.cgi%3Fid%3D211791&data=04%7C01%7Cray.huang%40amd.com%7C9c4d68e3c053401c4b4108d9065f38b7%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637547828334533410%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=AEMijLiBtz7Tf%2F8Uh1XEd4QUclZUfafyEy48yMf4JSw%3D&reserved=0 > > > > Signed-off-by: Huang Rui <ray.huang@amd.com> > > > > Cc: Alex Deucher <alexander.deucher@amd.com> > > > > Cc: Nathan Fontenot <nathan.fontenot@amd.com> > > > > Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> > > > > Cc: Borislav Petkov <bp@suse.de> > > > > Cc: x86@kernel.org > > > > Cc: stable@vger.kernel.org > > > > --- > > > > > > > > Changes from V1 -> V2: > > > > - Enhance the commit message. > > > > - Move amd_get_highest_perf() into amd.c. > > > > - Refine the implementation of switch-case. > > > > - Cc stable mail list. > > > > > > > > Changes from V2 -> V3: > > > > - Move the update into cppc_get_perf_caps() to correct the highest perf value in > > > > the API. > > > > > > > > --- > > > > arch/x86/include/asm/processor.h | 2 ++ > > > > arch/x86/kernel/cpu/amd.c | 22 ++++++++++++++++++++++ > > > > drivers/acpi/cppc_acpi.c | 8 ++++++-- > > > > 3 files changed, 30 insertions(+), 2 deletions(-) > > > > > > > > diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h > > > > index f1b9ed5efaa9..908bcaea1361 100644 > > > > --- a/arch/x86/include/asm/processor.h > > > > +++ b/arch/x86/include/asm/processor.h > > > > @@ -804,8 +804,10 @@ DECLARE_PER_CPU(u64, msr_misc_features_shadow); > > > > > > > > #ifdef CONFIG_CPU_SUP_AMD > > > > extern u32 amd_get_nodes_per_socket(void); > > > > +extern u32 amd_get_highest_perf(void); > > > > #else > > > > static inline u32 amd_get_nodes_per_socket(void) { return 0; } > > > > +static inline u32 amd_get_highest_perf(void) { return 0; } > > > > #endif > > > > > > > > static inline uint32_t hypervisor_cpuid_base(const char *sig, uint32_t leaves) > > > > diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c > > > > index 347a956f71ca..aadb691d9357 100644 > > > > --- a/arch/x86/kernel/cpu/amd.c > > > > +++ b/arch/x86/kernel/cpu/amd.c > > > > @@ -1170,3 +1170,25 @@ void set_dr_addr_mask(unsigned long mask, int dr) > > > > break; > > > > } > > > > } > > > > + > > > > +u32 amd_get_highest_perf(void) > > > > +{ > > > > + struct cpuinfo_x86 *c = &boot_cpu_data; > > > > + u32 cppc_max_perf = 225; > > > > + > > > > + switch (c->x86) { > > > > + case 0x17: > > > > + if ((c->x86_model >= 0x30 && c->x86_model < 0x40) || > > > > + (c->x86_model >= 0x70 && c->x86_model < 0x80)) > > > > + cppc_max_perf = 166; > > > > + break; > > > > + case 0x19: > > > > + if ((c->x86_model >= 0x20 && c->x86_model < 0x30) || > > > > + (c->x86_model >= 0x40 && c->x86_model < 0x70)) > > > > + cppc_max_perf = 166; > > > > + break; > > > > + } > > > > + > > > > + return cppc_max_perf; > > > > +} > > > > +EXPORT_SYMBOL_GPL(amd_get_highest_perf); > > > > diff --git a/drivers/acpi/cppc_acpi.c b/drivers/acpi/cppc_acpi.c > > > > index 69057fcd2c04..58e72b6e222f 100644 > > > > --- a/drivers/acpi/cppc_acpi.c > > > > +++ b/drivers/acpi/cppc_acpi.c > > > > @@ -1107,8 +1107,12 @@ int cppc_get_perf_caps(int cpunum, struct cppc_perf_caps *perf_caps) > > > > } > > > > } > > > > > > > > - cpc_read(cpunum, highest_reg, &high); > > > > - perf_caps->highest_perf = high; > > > > + if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) { > > > > > > This is a generic arch-independent file. > > > > > > Can we avoid adding the x86-specific check here? > > > > OK, I see, it will be used by ARM as well. > > > > Can I rollback to implementation of V2: > > > > https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Flore.kernel.org%2Fr%2F20210421023807.1540290-1-ray.huang%40amd.com&data=04%7C01%7Cray.huang%40amd.com%7C9c4d68e3c053401c4b4108d9065f38b7%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637547828334533410%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=Pk0VKl7iSaKz%2FYQx7YfT5D1XP%2FZRfQTW6moE%2F5sS1c0%3D&reserved=0 > > This would work IMO, but it can be simplified somewhat AFAICS. > > The obvious drawback is that amd_get_highest_perf() would need to be > called directly wherever the CPPC highest perf is needed and the > vendor may be AMD. Should I send V4 to continue review (fallback to V2 actually) or you can comment it on V2 directly? > > > If stick to add quirk in cppc_acpi.c and avoid x86-specific check at the > > same time here, the code will not be straight forward. Or will you have any > > other good idea? > > Not at the moment, sorry. No problem, thanks for your time and comments. :-) Thanks, Ray ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v3] x86, sched: Fix the AMD CPPC maximum perf on some specific generations 2021-04-23 15:07 ` Huang Rui @ 2021-04-23 15:20 ` Rafael J. Wysocki 2021-04-25 7:11 ` Huang Rui 0 siblings, 1 reply; 7+ messages in thread From: Rafael J. Wysocki @ 2021-04-23 15:20 UTC (permalink / raw) To: Huang Rui Cc: Rafael J. Wysocki, Linux PM, Linux Kernel Mailing List, Deucher, Alexander, Jason Bagavatsingham, Pierre-Loup A . Griffais, Fontenot, Nathan, Rafael J . Wysocki, Borislav Petkov, the arch/x86 maintainers, Stable On Fri, Apr 23, 2021 at 5:07 PM Huang Rui <ray.huang@amd.com> wrote: > > On Fri, Apr 23, 2021 at 09:53:37PM +0800, Rafael J. Wysocki wrote: > > On Fri, Apr 23, 2021 at 2:52 PM Huang Rui <ray.huang@amd.com> wrote: > > > > > > On Fri, Apr 23, 2021 at 08:09:49PM +0800, Rafael J. Wysocki wrote: > > > > On Fri, Apr 23, 2021 at 4:40 AM Huang Rui <ray.huang@amd.com> wrote: > > > > > > > > > > Some AMD Ryzen generations has different calculation method on maximum > > > > > perf. 255 is not for all asics, some specific generations should use 166 > > > > > as the maximum perf. Otherwise, it will report incorrect frequency value > > > > > like below: > > > > > > > > > > ~ $B"* (B lscpu | grep MHz > > > > > CPU MHz: 3400.000 > > > > > CPU max MHz: 7228.3198 > > > > > CPU min MHz: 2200.0000 > > > > > > > > > > Fixes: 41ea667227ba ("x86, sched: Calculate frequency invariance for AMD systems") > > > > > Fixes: 3c55e94c0ade ("cpufreq: ACPI: Extend frequency tables to cover boost frequencies") > > > > > > > > > > Reported-by: Jason Bagavatsingham <jason.bagavatsingham@gmail.com> > > > > > Tested-by: Jason Bagavatsingham <jason.bagavatsingham@gmail.com> > > > > > Bugzilla: https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fbugzilla.kernel.org%2Fshow_bug.cgi%3Fid%3D211791&data=04%7C01%7Cray.huang%40amd.com%7C9c4d68e3c053401c4b4108d9065f38b7%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637547828334533410%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=AEMijLiBtz7Tf%2F8Uh1XEd4QUclZUfafyEy48yMf4JSw%3D&reserved=0 > > > > > Signed-off-by: Huang Rui <ray.huang@amd.com> > > > > > Cc: Alex Deucher <alexander.deucher@amd.com> > > > > > Cc: Nathan Fontenot <nathan.fontenot@amd.com> > > > > > Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> > > > > > Cc: Borislav Petkov <bp@suse.de> > > > > > Cc: x86@kernel.org > > > > > Cc: stable@vger.kernel.org > > > > > --- > > > > > > > > > > Changes from V1 -> V2: > > > > > - Enhance the commit message. > > > > > - Move amd_get_highest_perf() into amd.c. > > > > > - Refine the implementation of switch-case. > > > > > - Cc stable mail list. > > > > > > > > > > Changes from V2 -> V3: > > > > > - Move the update into cppc_get_perf_caps() to correct the highest perf value in > > > > > the API. > > > > > > > > > > --- > > > > > arch/x86/include/asm/processor.h | 2 ++ > > > > > arch/x86/kernel/cpu/amd.c | 22 ++++++++++++++++++++++ > > > > > drivers/acpi/cppc_acpi.c | 8 ++++++-- > > > > > 3 files changed, 30 insertions(+), 2 deletions(-) > > > > > > > > > > diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h > > > > > index f1b9ed5efaa9..908bcaea1361 100644 > > > > > --- a/arch/x86/include/asm/processor.h > > > > > +++ b/arch/x86/include/asm/processor.h > > > > > @@ -804,8 +804,10 @@ DECLARE_PER_CPU(u64, msr_misc_features_shadow); > > > > > > > > > > #ifdef CONFIG_CPU_SUP_AMD > > > > > extern u32 amd_get_nodes_per_socket(void); > > > > > +extern u32 amd_get_highest_perf(void); > > > > > #else > > > > > static inline u32 amd_get_nodes_per_socket(void) { return 0; } > > > > > +static inline u32 amd_get_highest_perf(void) { return 0; } > > > > > #endif > > > > > > > > > > static inline uint32_t hypervisor_cpuid_base(const char *sig, uint32_t leaves) > > > > > diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c > > > > > index 347a956f71ca..aadb691d9357 100644 > > > > > --- a/arch/x86/kernel/cpu/amd.c > > > > > +++ b/arch/x86/kernel/cpu/amd.c > > > > > @@ -1170,3 +1170,25 @@ void set_dr_addr_mask(unsigned long mask, int dr) > > > > > break; > > > > > } > > > > > } > > > > > + > > > > > +u32 amd_get_highest_perf(void) > > > > > +{ > > > > > + struct cpuinfo_x86 *c = &boot_cpu_data; > > > > > + u32 cppc_max_perf = 225; > > > > > + > > > > > + switch (c->x86) { > > > > > + case 0x17: > > > > > + if ((c->x86_model >= 0x30 && c->x86_model < 0x40) || > > > > > + (c->x86_model >= 0x70 && c->x86_model < 0x80)) > > > > > + cppc_max_perf = 166; > > > > > + break; > > > > > + case 0x19: > > > > > + if ((c->x86_model >= 0x20 && c->x86_model < 0x30) || > > > > > + (c->x86_model >= 0x40 && c->x86_model < 0x70)) > > > > > + cppc_max_perf = 166; > > > > > + break; > > > > > + } > > > > > + > > > > > + return cppc_max_perf; > > > > > +} > > > > > +EXPORT_SYMBOL_GPL(amd_get_highest_perf); > > > > > diff --git a/drivers/acpi/cppc_acpi.c b/drivers/acpi/cppc_acpi.c > > > > > index 69057fcd2c04..58e72b6e222f 100644 > > > > > --- a/drivers/acpi/cppc_acpi.c > > > > > +++ b/drivers/acpi/cppc_acpi.c > > > > > @@ -1107,8 +1107,12 @@ int cppc_get_perf_caps(int cpunum, struct cppc_perf_caps *perf_caps) > > > > > } > > > > > } > > > > > > > > > > - cpc_read(cpunum, highest_reg, &high); > > > > > - perf_caps->highest_perf = high; > > > > > + if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) { > > > > > > > > This is a generic arch-independent file. > > > > > > > > Can we avoid adding the x86-specific check here? > > > > > > OK, I see, it will be used by ARM as well. > > > > > > Can I rollback to implementation of V2: > > > > > > https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Flore.kernel.org%2Fr%2F20210421023807.1540290-1-ray.huang%40amd.com&data=04%7C01%7Cray.huang%40amd.com%7C9c4d68e3c053401c4b4108d9065f38b7%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637547828334533410%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=Pk0VKl7iSaKz%2FYQx7YfT5D1XP%2FZRfQTW6moE%2F5sS1c0%3D&reserved=0 > > > > This would work IMO, but it can be simplified somewhat AFAICS. > > > > The obvious drawback is that amd_get_highest_perf() would need to be > > called directly wherever the CPPC highest perf is needed and the > > vendor may be AMD. > > Should I send V4 to continue review (fallback to V2 actually) or you can > comment it on V2 directly? Done, thanks! ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v3] x86, sched: Fix the AMD CPPC maximum perf on some specific generations 2021-04-23 15:20 ` Rafael J. Wysocki @ 2021-04-25 7:11 ` Huang Rui 0 siblings, 0 replies; 7+ messages in thread From: Huang Rui @ 2021-04-25 7:11 UTC (permalink / raw) To: Rafael J. Wysocki Cc: Linux PM, Linux Kernel Mailing List, Deucher, Alexander, Jason Bagavatsingham, Pierre-Loup A . Griffais, Fontenot, Nathan, Rafael J . Wysocki, Borislav Petkov, the arch/x86 maintainers, Stable On Fri, Apr 23, 2021 at 11:20:37PM +0800, Rafael J. Wysocki wrote: > On Fri, Apr 23, 2021 at 5:07 PM Huang Rui <ray.huang@amd.com> wrote: > > > > On Fri, Apr 23, 2021 at 09:53:37PM +0800, Rafael J. Wysocki wrote: > > > On Fri, Apr 23, 2021 at 2:52 PM Huang Rui <ray.huang@amd.com> wrote: > > > > > > > > On Fri, Apr 23, 2021 at 08:09:49PM +0800, Rafael J. Wysocki wrote: > > > > > On Fri, Apr 23, 2021 at 4:40 AM Huang Rui <ray.huang@amd.com> wrote: > > > > > > > > > > > > Some AMD Ryzen generations has different calculation method on maximum > > > > > > perf. 255 is not for all asics, some specific generations should use 166 > > > > > > as the maximum perf. Otherwise, it will report incorrect frequency value > > > > > > like below: > > > > > > > > > > > > ~ $B"* (B lscpu | grep MHz > > > > > > CPU MHz: 3400.000 > > > > > > CPU max MHz: 7228.3198 > > > > > > CPU min MHz: 2200.0000 > > > > > > > > > > > > Fixes: 41ea667227ba ("x86, sched: Calculate frequency invariance for AMD systems") > > > > > > Fixes: 3c55e94c0ade ("cpufreq: ACPI: Extend frequency tables to cover boost frequencies") > > > > > > > > > > > > Reported-by: Jason Bagavatsingham <jason.bagavatsingham@gmail.com> > > > > > > Tested-by: Jason Bagavatsingham <jason.bagavatsingham@gmail.com> > > > > > > Bugzilla: https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fbugzilla.kernel.org%2Fshow_bug.cgi%3Fid%3D211791&data=04%7C01%7Cray.huang%40amd.com%7C5e4ef90268394256017508d9066b6011%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637547880536218729%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=6UKHrjewjid2pnAFJTXen1xPVSwIAOkcDdiQMfORZmw%3D&reserved=0 > > > > > > Signed-off-by: Huang Rui <ray.huang@amd.com> > > > > > > Cc: Alex Deucher <alexander.deucher@amd.com> > > > > > > Cc: Nathan Fontenot <nathan.fontenot@amd.com> > > > > > > Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> > > > > > > Cc: Borislav Petkov <bp@suse.de> > > > > > > Cc: x86@kernel.org > > > > > > Cc: stable@vger.kernel.org > > > > > > --- > > > > > > > > > > > > Changes from V1 -> V2: > > > > > > - Enhance the commit message. > > > > > > - Move amd_get_highest_perf() into amd.c. > > > > > > - Refine the implementation of switch-case. > > > > > > - Cc stable mail list. > > > > > > > > > > > > Changes from V2 -> V3: > > > > > > - Move the update into cppc_get_perf_caps() to correct the highest perf value in > > > > > > the API. > > > > > > > > > > > > --- > > > > > > arch/x86/include/asm/processor.h | 2 ++ > > > > > > arch/x86/kernel/cpu/amd.c | 22 ++++++++++++++++++++++ > > > > > > drivers/acpi/cppc_acpi.c | 8 ++++++-- > > > > > > 3 files changed, 30 insertions(+), 2 deletions(-) > > > > > > > > > > > > diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h > > > > > > index f1b9ed5efaa9..908bcaea1361 100644 > > > > > > --- a/arch/x86/include/asm/processor.h > > > > > > +++ b/arch/x86/include/asm/processor.h > > > > > > @@ -804,8 +804,10 @@ DECLARE_PER_CPU(u64, msr_misc_features_shadow); > > > > > > > > > > > > #ifdef CONFIG_CPU_SUP_AMD > > > > > > extern u32 amd_get_nodes_per_socket(void); > > > > > > +extern u32 amd_get_highest_perf(void); > > > > > > #else > > > > > > static inline u32 amd_get_nodes_per_socket(void) { return 0; } > > > > > > +static inline u32 amd_get_highest_perf(void) { return 0; } > > > > > > #endif > > > > > > > > > > > > static inline uint32_t hypervisor_cpuid_base(const char *sig, uint32_t leaves) > > > > > > diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c > > > > > > index 347a956f71ca..aadb691d9357 100644 > > > > > > --- a/arch/x86/kernel/cpu/amd.c > > > > > > +++ b/arch/x86/kernel/cpu/amd.c > > > > > > @@ -1170,3 +1170,25 @@ void set_dr_addr_mask(unsigned long mask, int dr) > > > > > > break; > > > > > > } > > > > > > } > > > > > > + > > > > > > +u32 amd_get_highest_perf(void) > > > > > > +{ > > > > > > + struct cpuinfo_x86 *c = &boot_cpu_data; > > > > > > + u32 cppc_max_perf = 225; > > > > > > + > > > > > > + switch (c->x86) { > > > > > > + case 0x17: > > > > > > + if ((c->x86_model >= 0x30 && c->x86_model < 0x40) || > > > > > > + (c->x86_model >= 0x70 && c->x86_model < 0x80)) > > > > > > + cppc_max_perf = 166; > > > > > > + break; > > > > > > + case 0x19: > > > > > > + if ((c->x86_model >= 0x20 && c->x86_model < 0x30) || > > > > > > + (c->x86_model >= 0x40 && c->x86_model < 0x70)) > > > > > > + cppc_max_perf = 166; > > > > > > + break; > > > > > > + } > > > > > > + > > > > > > + return cppc_max_perf; > > > > > > +} > > > > > > +EXPORT_SYMBOL_GPL(amd_get_highest_perf); > > > > > > diff --git a/drivers/acpi/cppc_acpi.c b/drivers/acpi/cppc_acpi.c > > > > > > index 69057fcd2c04..58e72b6e222f 100644 > > > > > > --- a/drivers/acpi/cppc_acpi.c > > > > > > +++ b/drivers/acpi/cppc_acpi.c > > > > > > @@ -1107,8 +1107,12 @@ int cppc_get_perf_caps(int cpunum, struct cppc_perf_caps *perf_caps) > > > > > > } > > > > > > } > > > > > > > > > > > > - cpc_read(cpunum, highest_reg, &high); > > > > > > - perf_caps->highest_perf = high; > > > > > > + if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) { > > > > > > > > > > This is a generic arch-independent file. > > > > > > > > > > Can we avoid adding the x86-specific check here? > > > > > > > > OK, I see, it will be used by ARM as well. > > > > > > > > Can I rollback to implementation of V2: > > > > > > > > https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Flore.kernel.org%2Fr%2F20210421023807.1540290-1-ray.huang%40amd.com&data=04%7C01%7Cray.huang%40amd.com%7C5e4ef90268394256017508d9066b6011%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637547880536218729%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=jytSyFAHiOc3JdkHyTkD86Vsp9%2FT9e2OENqOTkDtt5M%3D&reserved=0 > > > > > > This would work IMO, but it can be simplified somewhat AFAICS. > > > > > > The obvious drawback is that amd_get_highest_perf() would need to be > > > called directly wherever the CPPC highest perf is needed and the > > > vendor may be AMD. > > > > Should I send V4 to continue review (fallback to V2 actually) or you can > > comment it on V2 directly? > > Done, thanks! Thank you! Ray ^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2021-04-25 7:12 UTC | newest] Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2021-04-23 2:39 [PATCH v3] x86, sched: Fix the AMD CPPC maximum perf on some specific generations Huang Rui 2021-04-23 12:09 ` Rafael J. Wysocki 2021-04-23 12:52 ` Huang Rui 2021-04-23 13:53 ` Rafael J. Wysocki 2021-04-23 15:07 ` Huang Rui 2021-04-23 15:20 ` Rafael J. Wysocki 2021-04-25 7:11 ` Huang Rui
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).