| Message ID | 20201209163950.8494-3-will@kernel.org (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | Create 'old' ptes for faultaround mappings on arm64 with hardware access flag | expand |
On Wed, Dec 09, 2020 at 04:39:50PM +0000, Will Deacon wrote: > diff --git a/arch/arm64/include/asm/cpufeature.h b/arch/arm64/include/asm/cpufeature.h > index da250e4741bd..3424f5881390 100644 > --- a/arch/arm64/include/asm/cpufeature.h > +++ b/arch/arm64/include/asm/cpufeature.h > @@ -764,6 +764,18 @@ static inline bool cpu_has_hw_af(void) > ID_AA64MMFR1_HADBS_SHIFT); > } > > +static inline bool system_has_hw_af(void) > +{ > + u64 mmfr1; > + > + if (!IS_ENABLED(CONFIG_ARM64_HW_AFDBM)) > + return false; > + > + mmfr1 = read_sanitised_ftr_reg(SYS_ID_AA64MMFR1_EL1); > + return cpuid_feature_extract_unsigned_field(mmfr1, > + ID_AA64MMFR1_HADBS_SHIFT); > +} Could we not add a new system-wide cpu feature that checks for hardware AF? This read_sanitised_ftr_reg() does a binary search on each invocation.
On Wed, Dec 09, 2020 at 06:35:09PM +0000, Catalin Marinas wrote: > On Wed, Dec 09, 2020 at 04:39:50PM +0000, Will Deacon wrote: > > diff --git a/arch/arm64/include/asm/cpufeature.h b/arch/arm64/include/asm/cpufeature.h > > index da250e4741bd..3424f5881390 100644 > > --- a/arch/arm64/include/asm/cpufeature.h > > +++ b/arch/arm64/include/asm/cpufeature.h > > @@ -764,6 +764,18 @@ static inline bool cpu_has_hw_af(void) > > ID_AA64MMFR1_HADBS_SHIFT); > > } > > > > +static inline bool system_has_hw_af(void) > > +{ > > + u64 mmfr1; > > + > > + if (!IS_ENABLED(CONFIG_ARM64_HW_AFDBM)) > > + return false; > > + > > + mmfr1 = read_sanitised_ftr_reg(SYS_ID_AA64MMFR1_EL1); > > + return cpuid_feature_extract_unsigned_field(mmfr1, > > + ID_AA64MMFR1_HADBS_SHIFT); > > +} > > Could we not add a new system-wide cpu feature that checks for hardware > AF? This read_sanitised_ftr_reg() does a binary search on each > invocation. I posted a diff [1] which would allow removing the binary search for cases where we can pass the register coding as a constant (like this), but honestly, it's not like we have many ID registers so I doubt it really matters in the grand scheme of things. That said, I'm spinning a v2 anyway so I can include it for comments since I haven't posted it as a proper patch before. Will [1] https://lore.kernel.org/r/20201202172727.GC29813@willie-the-truck
diff --git a/arch/arm64/include/asm/cpufeature.h b/arch/arm64/include/asm/cpufeature.h index da250e4741bd..3424f5881390 100644 --- a/arch/arm64/include/asm/cpufeature.h +++ b/arch/arm64/include/asm/cpufeature.h @@ -764,6 +764,18 @@ static inline bool cpu_has_hw_af(void) ID_AA64MMFR1_HADBS_SHIFT); } +static inline bool system_has_hw_af(void) +{ + u64 mmfr1; + + if (!IS_ENABLED(CONFIG_ARM64_HW_AFDBM)) + return false; + + mmfr1 = read_sanitised_ftr_reg(SYS_ID_AA64MMFR1_EL1); + return cpuid_feature_extract_unsigned_field(mmfr1, + ID_AA64MMFR1_HADBS_SHIFT); +} + #ifdef CONFIG_ARM64_AMU_EXTN /* Check whether the cpu supports the Activity Monitors Unit (AMU) */ extern bool cpu_has_amu_feat(int cpu); diff --git a/arch/arm64/include/asm/pgtable.h b/arch/arm64/include/asm/pgtable.h index 5628289b9d5e..d5c2a7625e9a 100644 --- a/arch/arm64/include/asm/pgtable.h +++ b/arch/arm64/include/asm/pgtable.h @@ -974,7 +974,13 @@ static inline bool arch_faults_on_old_pte(void) return !cpu_has_hw_af(); } -#define arch_faults_on_old_pte arch_faults_on_old_pte +#define arch_faults_on_old_pte arch_faults_on_old_pte + +/* + * Experimentally, it's cheap to set the access flag in hardware and we + * benefit from prefaulting mappings as 'old' to start with. + */ +#define arch_wants_old_faultaround_pte system_has_hw_af #endif /* !__ASSEMBLY__ */
On CPUs with hardware AF/DBM, initialising prefaulted PTEs as 'old' improves vmscan behaviour and does not appear to introduce any overhead. Implement arch_wants_old_faultaround_pte() to return 'true' if we detect hardware access flag support at runtime. This can be extended in future based on MIDR matching if necessary. Cc: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Will Deacon <will@kernel.org> --- arch/arm64/include/asm/cpufeature.h | 12 ++++++++++++ arch/arm64/include/asm/pgtable.h | 8 +++++++- 2 files changed, 19 insertions(+), 1 deletion(-)