From: Ira Weiny ira.weiny@intel.com
Protection Keys for Supervisor pages (PKS) enables fast, hardware thread specific, manipulation of permission restrictions on supervisor page mappings. It uses the same mechanism of Protection Keys as those on User mappings but applies that mechanism to supervisor mappings using a supervisor specific MSR.
Add setup code and the lowest level of PKS MSR write support. The write value is cached per-cpu to avoid the overhead of the MSR write if the value has not changed.
That said, it should be noted that the underlying WRMSR(MSR_IA32_PKRS) is not serializing but still maintains ordering properties similar to WRPKRU. The current SDM section on PKRS needs updating but should be the same as that of WRPKRU. So to quote from the WRPKRU text:
WRPKRU will never execute transiently. Memory accesses affected by PKRU register will not execute (even transiently) until all prior executions of WRPKRU have completed execution and updated the PKRU register.
write_pkrs() contributed by Peter Zijlstra.
Introduce asm/pks.h to declare setup_pks() as an internal function call. Later patches will also need this new header as a place to declare internal structures and functions.
Reviewed-by: Dan Williams dan.j.williams@intel.com Co-developed-by: Peter Zijlstra peterz@infradead.org Signed-off-by: Peter Zijlstra peterz@infradead.org Co-developed-by: Fenghua Yu fenghua.yu@intel.com Signed-off-by: Fenghua Yu fenghua.yu@intel.com Signed-off-by: Ira Weiny ira.weiny@intel.com
--- Changes from V3: From Dan Williams: Update commit message Add pks.h to hold ifdefery out of *.c files s/ARCH_HAS.../SUPERVISOR_PKEYS move setup_pks to pkeys.c (remove more ifdefery) Remove 'domain' language from commit message Clarify comment in fault handler Move the removal of the WARN_ON_ONCE in the fault path to this patch. Previously it was in: [07/10] x86/fault: Report the PKRS state on fault
Changes from V2 From Thomas: Make this patch last so PKS is not enabled until all the PKS mechanisms are in place. Specifically: 1) Modify setup_pks() to call write_pkrs() to properly set up the initial value when enabled.
2) Split this patch into two. 1) a precursor patch with the required defines/config options and 2) this patch which actually enables feature on CPUs which support it.
Changes since RFC V3 Per Dave Hansen Update comment Add X86_FEATURE_PKS to disabled-features.h Rebase based on latest TIP tree --- arch/x86/include/asm/pks.h | 15 +++++++++++ arch/x86/kernel/cpu/common.c | 2 ++ arch/x86/mm/pkeys.c | 48 ++++++++++++++++++++++++++++++++++++ 3 files changed, 65 insertions(+) create mode 100644 arch/x86/include/asm/pks.h
diff --git a/arch/x86/include/asm/pks.h b/arch/x86/include/asm/pks.h new file mode 100644 index 000000000000..5d7067ada8fb --- /dev/null +++ b/arch/x86/include/asm/pks.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _ASM_X86_PKS_H +#define _ASM_X86_PKS_H + +#ifdef CONFIG_ARCH_ENABLE_SUPERVISOR_PKEYS + +void setup_pks(void); + +#else /* !CONFIG_ARCH_ENABLE_SUPERVISOR_PKEYS */ + +static inline void setup_pks(void) { } + +#endif /* CONFIG_ARCH_ENABLE_SUPERVISOR_PKEYS */ + +#endif /* _ASM_X86_PKS_H */ diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index ab640abe26b6..de49d0c0f4e0 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -58,6 +58,7 @@ #include <asm/intel-family.h> #include <asm/cpu_device_id.h> #include <asm/uv/uv.h> +#include <asm/pks.h>
#include "cpu.h"
@@ -1594,6 +1595,7 @@ static void identify_cpu(struct cpuinfo_x86 *c)
x86_init_rdrand(c); setup_pku(c); + setup_pks();
/* * Clear/Set all flags overridden by options, need do it diff --git a/arch/x86/mm/pkeys.c b/arch/x86/mm/pkeys.c index fc8c7e2bb21b..f6a3a54b8d7d 100644 --- a/arch/x86/mm/pkeys.c +++ b/arch/x86/mm/pkeys.c @@ -229,3 +229,51 @@ u32 update_pkey_val(u32 pk_reg, int pkey, unsigned int flags)
return pk_reg; } + +#ifdef CONFIG_ARCH_ENABLE_SUPERVISOR_PKEYS + +static DEFINE_PER_CPU(u32, pkrs_cache); + +/* + * write_pkrs() optimizes MSR writes by maintaining a per cpu cache which can + * be checked quickly. + * + * It should also be noted that the underlying WRMSR(MSR_IA32_PKRS) is not + * serializing but still maintains ordering properties similar to WRPKRU. + * The current SDM section on PKRS needs updating but should be the same as + * that of WRPKRU. So to quote from the WRPKRU text: + * + * WRPKRU will never execute transiently. Memory accesses + * affected by PKRU register will not execute (even transiently) + * until all prior executions of WRPKRU have completed execution + * and updated the PKRU register. + */ +void write_pkrs(u32 new_pkrs) +{ + u32 *pkrs; + + if (!static_cpu_has(X86_FEATURE_PKS)) + return; + + pkrs = get_cpu_ptr(&pkrs_cache); + if (*pkrs != new_pkrs) { + *pkrs = new_pkrs; + wrmsrl(MSR_IA32_PKRS, new_pkrs); + } + put_cpu_ptr(pkrs); +} + +/* + * PKS is independent of PKU and either or both may be supported on a CPU. + * Configure PKS if the CPU supports the feature. + */ +void setup_pks(void) +{ + if (!cpu_feature_enabled(X86_FEATURE_PKS)) + return; + + write_pkrs(INIT_PKRS_VALUE); + cr4_set_bits(X86_CR4_PKS); +} + +#endif