| Message ID | 1457317732-18406-16-git-send-email-sjg@chromium.org |
|---|---|
| State | Superseded |
| Delegated to: | Bin Meng |
| Headers | show |
Hi Simon, On Mon, Mar 7, 2016 at 10:27 AM, Simon Glass <sjg@chromium.org> wrote: > These two identifiers can be useful for drivers which need to adjust their > behaviour depending on the CPU family or stepping (revision). > > Signed-off-by: Simon Glass <sjg@chromium.org> > --- > > arch/x86/cpu/cpu.c | 10 ++++++++++ > arch/x86/include/asm/cpu.h | 14 ++++++++++++++ > 2 files changed, 24 insertions(+) > > diff --git a/arch/x86/cpu/cpu.c b/arch/x86/cpu/cpu.c > index 8800e09..e2aad19 100644 > --- a/arch/x86/cpu/cpu.c > +++ b/arch/x86/cpu/cpu.c > @@ -333,6 +333,16 @@ static inline void get_fms(struct cpuinfo_x86 *c, uint32_t tfms) > c->x86_model += ((tfms >> 16) & 0xF) << 4; > } > > +u32 cpu_get_family_model(void) > +{ > + return gd->arch.x86_device & 0x0fff0ff0; The decrypted family and model are stored in gd->arch.x86 and gd->arch.x86_model. Returning raw data would still need the caller to parse it. Why not just return these directly? > +} > + > +u32 cpu_get_stepping(void) > +{ > + return gd->arch.x86_device & 0xf; This is gd->arch.x86_mask. > +} > + > int x86_cpu_init_f(void) > { > const u32 em_rst = ~X86_CR0_EM; > diff --git a/arch/x86/include/asm/cpu.h b/arch/x86/include/asm/cpu.h > index 18b0345..987dc65 100644 > --- a/arch/x86/include/asm/cpu.h > +++ b/arch/x86/include/asm/cpu.h > @@ -260,4 +260,18 @@ void cpu_call32(ulong code_seg32, ulong target, ulong table); > */ > int cpu_jump_to_64bit(ulong setup_base, ulong target); > > +/** > + * cpu_get_family_model() - Get the family and model for the CPU > + * > + * @return the CPU ID masked with 0x0fff0ff0 > + */ > +u32 cpu_get_family_model(void); > + > +/** > + * cpu_get_stepping() - Get the stepping value for the CPU > + * > + * @return the CPU ID masked with 0xf > + */ > +u32 cpu_get_stepping(void); > + > #endif > -- Regards, Bin
Hi Bin, On 10 March 2016 at 21:52, Bin Meng <bmeng.cn@gmail.com> wrote: > Hi Simon, > > On Mon, Mar 7, 2016 at 10:27 AM, Simon Glass <sjg@chromium.org> wrote: >> These two identifiers can be useful for drivers which need to adjust their >> behaviour depending on the CPU family or stepping (revision). >> >> Signed-off-by: Simon Glass <sjg@chromium.org> >> --- >> >> arch/x86/cpu/cpu.c | 10 ++++++++++ >> arch/x86/include/asm/cpu.h | 14 ++++++++++++++ >> 2 files changed, 24 insertions(+) >> >> diff --git a/arch/x86/cpu/cpu.c b/arch/x86/cpu/cpu.c >> index 8800e09..e2aad19 100644 >> --- a/arch/x86/cpu/cpu.c >> +++ b/arch/x86/cpu/cpu.c >> @@ -333,6 +333,16 @@ static inline void get_fms(struct cpuinfo_x86 *c, uint32_t tfms) >> c->x86_model += ((tfms >> 16) & 0xF) << 4; >> } >> >> +u32 cpu_get_family_model(void) >> +{ >> + return gd->arch.x86_device & 0x0fff0ff0; > > The decrypted family and model are stored in gd->arch.x86 and > gd->arch.x86_model. Returning raw data would still need the caller to > parse it. Why not just return these directly? I want to compare them in one shot to BROADWELL_FAMILY_ULT, etc. What exactly are you suggesting? > >> +} >> + >> +u32 cpu_get_stepping(void) >> +{ >> + return gd->arch.x86_device & 0xf; > > This is gd->arch.x86_mask. OK. That's a funny name for stepping. > >> +} >> + >> int x86_cpu_init_f(void) >> { >> const u32 em_rst = ~X86_CR0_EM; >> diff --git a/arch/x86/include/asm/cpu.h b/arch/x86/include/asm/cpu.h >> index 18b0345..987dc65 100644 >> --- a/arch/x86/include/asm/cpu.h >> +++ b/arch/x86/include/asm/cpu.h >> @@ -260,4 +260,18 @@ void cpu_call32(ulong code_seg32, ulong target, ulong table); >> */ >> int cpu_jump_to_64bit(ulong setup_base, ulong target); >> >> +/** >> + * cpu_get_family_model() - Get the family and model for the CPU >> + * >> + * @return the CPU ID masked with 0x0fff0ff0 >> + */ >> +u32 cpu_get_family_model(void); >> + >> +/** >> + * cpu_get_stepping() - Get the stepping value for the CPU >> + * >> + * @return the CPU ID masked with 0xf >> + */ >> +u32 cpu_get_stepping(void); >> + >> #endif >> -- > > Regards, > Bin
Hi Simon, On Sat, Mar 12, 2016 at 1:04 PM, Simon Glass <sjg@chromium.org> wrote: > Hi Bin, > > On 10 March 2016 at 21:52, Bin Meng <bmeng.cn@gmail.com> wrote: >> Hi Simon, >> >> On Mon, Mar 7, 2016 at 10:27 AM, Simon Glass <sjg@chromium.org> wrote: >>> These two identifiers can be useful for drivers which need to adjust their >>> behaviour depending on the CPU family or stepping (revision). >>> >>> Signed-off-by: Simon Glass <sjg@chromium.org> >>> --- >>> >>> arch/x86/cpu/cpu.c | 10 ++++++++++ >>> arch/x86/include/asm/cpu.h | 14 ++++++++++++++ >>> 2 files changed, 24 insertions(+) >>> >>> diff --git a/arch/x86/cpu/cpu.c b/arch/x86/cpu/cpu.c >>> index 8800e09..e2aad19 100644 >>> --- a/arch/x86/cpu/cpu.c >>> +++ b/arch/x86/cpu/cpu.c >>> @@ -333,6 +333,16 @@ static inline void get_fms(struct cpuinfo_x86 *c, uint32_t tfms) >>> c->x86_model += ((tfms >> 16) & 0xF) << 4; >>> } >>> >>> +u32 cpu_get_family_model(void) >>> +{ >>> + return gd->arch.x86_device & 0x0fff0ff0; >> >> The decrypted family and model are stored in gd->arch.x86 and >> gd->arch.x86_model. Returning raw data would still need the caller to >> parse it. Why not just return these directly? > > I want to compare them in one shot to BROADWELL_FAMILY_ULT, etc. > > What exactly are you suggesting? > I guess we can also break BROADWELL_FAMILY_ULT down to values and have them compared to gd->arch.x86 and gd->arch.x86_model. But I suppose you have your logic already working, so this is fine. >> >>> +} >>> + >>> +u32 cpu_get_stepping(void) >>> +{ >>> + return gd->arch.x86_device & 0xf; >> >> This is gd->arch.x86_mask. > > OK. That's a funny name for stepping. Yep, we can rename this to stepping in the future. [snip] Regards, Bin
diff --git a/arch/x86/cpu/cpu.c b/arch/x86/cpu/cpu.c index 8800e09..e2aad19 100644 --- a/arch/x86/cpu/cpu.c +++ b/arch/x86/cpu/cpu.c @@ -333,6 +333,16 @@ static inline void get_fms(struct cpuinfo_x86 *c, uint32_t tfms) c->x86_model += ((tfms >> 16) & 0xF) << 4; } +u32 cpu_get_family_model(void) +{ + return gd->arch.x86_device & 0x0fff0ff0; +} + +u32 cpu_get_stepping(void) +{ + return gd->arch.x86_device & 0xf; +} + int x86_cpu_init_f(void) { const u32 em_rst = ~X86_CR0_EM; diff --git a/arch/x86/include/asm/cpu.h b/arch/x86/include/asm/cpu.h index 18b0345..987dc65 100644 --- a/arch/x86/include/asm/cpu.h +++ b/arch/x86/include/asm/cpu.h @@ -260,4 +260,18 @@ void cpu_call32(ulong code_seg32, ulong target, ulong table); */ int cpu_jump_to_64bit(ulong setup_base, ulong target); +/** + * cpu_get_family_model() - Get the family and model for the CPU + * + * @return the CPU ID masked with 0x0fff0ff0 + */ +u32 cpu_get_family_model(void); + +/** + * cpu_get_stepping() - Get the stepping value for the CPU + * + * @return the CPU ID masked with 0xf + */ +u32 cpu_get_stepping(void); + #endif
These two identifiers can be useful for drivers which need to adjust their behaviour depending on the CPU family or stepping (revision). Signed-off-by: Simon Glass <sjg@chromium.org> --- arch/x86/cpu/cpu.c | 10 ++++++++++ arch/x86/include/asm/cpu.h | 14 ++++++++++++++ 2 files changed, 24 insertions(+)