diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 86c91be8647c..a0fd7c8052a1 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -2225,8 +2225,12 @@ bytes respectively. Such letter suffixes can also be entirely omitted. nointroute [IA-64] + noinvpcid [X86] Disable the INVPCID cpu feature. + nojitter [IA-64] Disables jitter checking for ITC timers. + nopti [X86-64] Disable KAISER isolation of kernel from user. + no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page @@ -2259,6 +2263,8 @@ bytes respectively. Such letter suffixes can also be entirely omitted. nopat [X86] Disable PAT (page attribute table extension of pagetables) support. + nopcid [X86-64] Disable the PCID cpu feature. + norandmaps Don't use address space randomization. Equivalent to echo 0 > /proc/sys/kernel/randomize_va_space @@ -2746,6 +2752,12 @@ bytes respectively. Such letter suffixes can also be entirely omitted. pt. [PARIDE] See Documentation/blockdev/paride.txt. + pti= [X86_64] + Control KAISER user/kernel address space isolation: + on - enable + off - disable + auto - default setting + pty.legacy_count= [KNL] Number of legacy pty's. Overwrites compiled-in default number. diff --git a/Documentation/x86/tlb.txt b/Documentation/x86/tlb.txt new file mode 100644 index 000000000000..2b3a82e69151 --- /dev/null +++ b/Documentation/x86/tlb.txt @@ -0,0 +1,75 @@ +When the kernel unmaps or modified the attributes of a range of +memory, it has two choices: + 1. Flush the entire TLB with a two-instruction sequence. This is + a quick operation, but it causes collateral damage: TLB entries + from areas other than the one we are trying to flush will be + destroyed and must be refilled later, at some cost. + 2. Use the invlpg instruction to invalidate a single page at a + time. This could potentialy cost many more instructions, but + it is a much more precise operation, causing no collateral + damage to other TLB entries. + +Which method to do depends on a few things: + 1. The size of the flush being performed. A flush of the entire + address space is obviously better performed by flushing the + entire TLB than doing 2^48/PAGE_SIZE individual flushes. + 2. The contents of the TLB. If the TLB is empty, then there will + be no collateral damage caused by doing the global flush, and + all of the individual flush will have ended up being wasted + work. + 3. The size of the TLB. The larger the TLB, the more collateral + damage we do with a full flush. So, the larger the TLB, the + more attrative an individual flush looks. Data and + instructions have separate TLBs, as do different page sizes. + 4. The microarchitecture. The TLB has become a multi-level + cache on modern CPUs, and the global flushes have become more + expensive relative to single-page flushes. + +There is obviously no way the kernel can know all these things, +especially the contents of the TLB during a given flush. The +sizes of the flush will vary greatly depending on the workload as +well. There is essentially no "right" point to choose. + +You may be doing too many individual invalidations if you see the +invlpg instruction (or instructions _near_ it) show up high in +profiles. If you believe that individual invalidations being +called too often, you can lower the tunable: + + /sys/debug/kernel/x86/tlb_single_page_flush_ceiling + +This will cause us to do the global flush for more cases. +Lowering it to 0 will disable the use of the individual flushes. +Setting it to 1 is a very conservative setting and it should +never need to be 0 under normal circumstances. + +Despite the fact that a single individual flush on x86 is +guaranteed to flush a full 2MB [1], hugetlbfs always uses the full +flushes. THP is treated exactly the same as normal memory. + +You might see invlpg inside of flush_tlb_mm_range() show up in +profiles, or you can use the trace_tlb_flush() tracepoints. to +determine how long the flush operations are taking. + +Essentially, you are balancing the cycles you spend doing invlpg +with the cycles that you spend refilling the TLB later. + +You can measure how expensive TLB refills are by using +performance counters and 'perf stat', like this: + +perf stat -e + cpu/event=0x8,umask=0x84,name=dtlb_load_misses_walk_duration/, + cpu/event=0x8,umask=0x82,name=dtlb_load_misses_walk_completed/, + cpu/event=0x49,umask=0x4,name=dtlb_store_misses_walk_duration/, + cpu/event=0x49,umask=0x2,name=dtlb_store_misses_walk_completed/, + cpu/event=0x85,umask=0x4,name=itlb_misses_walk_duration/, + cpu/event=0x85,umask=0x2,name=itlb_misses_walk_completed/ + +That works on an IvyBridge-era CPU (i5-3320M). Different CPUs +may have differently-named counters, but they should at least +be there in some form. You can use pmu-tools 'ocperf list' +(https://github.com/andikleen/pmu-tools) to find the right +counters for a given CPU. + +1. A footnote in Intel's SDM "4.10.4.2 Recommended Invalidation" + says: "One execution of INVLPG is sufficient even for a page + with size greater than 4 KBytes." diff --git a/Makefile b/Makefile index 5cf074eb43ed..4386eb2679ae 100644 --- a/Makefile +++ b/Makefile @@ -1,6 +1,6 @@ VERSION = 3 PATCHLEVEL = 16 -SUBLEVEL = 52 +SUBLEVEL = 53 EXTRAVERSION = NAME = Museum of Fishiegoodies diff --git a/arch/arm/include/asm/mmu_context.h b/arch/arm/include/asm/mmu_context.h index 9b32f76bb0dd..10f662498eb7 100644 --- a/arch/arm/include/asm/mmu_context.h +++ b/arch/arm/include/asm/mmu_context.h @@ -61,6 +61,7 @@ static inline void check_and_switch_context(struct mm_struct *mm, cpu_switch_mm(mm->pgd, mm); } +#ifndef MODULE #define finish_arch_post_lock_switch \ finish_arch_post_lock_switch static inline void finish_arch_post_lock_switch(void) @@ -82,6 +83,7 @@ static inline void finish_arch_post_lock_switch(void) preempt_enable_no_resched(); } } +#endif /* !MODULE */ #endif /* CONFIG_MMU */ diff --git a/arch/x86/boot/compressed/misc.h b/arch/x86/boot/compressed/misc.h index 124312be129b..d527b9b12dd4 100644 --- a/arch/x86/boot/compressed/misc.h +++ b/arch/x86/boot/compressed/misc.h @@ -9,6 +9,7 @@ */ #undef CONFIG_PARAVIRT #undef CONFIG_PARAVIRT_SPINLOCKS +#undef CONFIG_PAGE_TABLE_ISOLATION #include #include diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S index b74ac9c5710b..88ceb07b5a2b 100644 --- a/arch/x86/ia32/ia32entry.S +++ b/arch/x86/ia32/ia32entry.S @@ -15,6 +15,8 @@ #include #include #include +#include +#include #include #include @@ -121,6 +123,7 @@ ENTRY(ia32_sysenter_target) CFI_DEF_CFA rsp,0 CFI_REGISTER rsp,rbp SWAPGS_UNSAFE_STACK + SWITCH_KERNEL_CR3_NO_STACK movq PER_CPU_VAR(kernel_stack), %rsp addq $(KERNEL_STACK_OFFSET),%rsp /* @@ -192,6 +195,7 @@ ENTRY(ia32_sysenter_target) popq_cfi %rcx /* User %esp */ CFI_REGISTER rsp,rcx TRACE_IRQS_ON + SWITCH_USER_CR3 ENABLE_INTERRUPTS_SYSEXIT32 CFI_RESTORE_STATE @@ -296,6 +300,7 @@ ENTRY(ia32_cstar_target) CFI_REGISTER rip,rcx /*CFI_REGISTER rflags,r11*/ SWAPGS_UNSAFE_STACK + SWITCH_KERNEL_CR3_NO_STACK movl %esp,%r8d CFI_REGISTER rsp,r8 movq PER_CPU_VAR(kernel_stack),%rsp @@ -350,6 +355,7 @@ ENTRY(ia32_cstar_target) xorq %r9,%r9 xorq %r8,%r8 TRACE_IRQS_ON + SWITCH_USER_CR3 movl RSP-ARGOFFSET(%rsp),%esp CFI_RESTORE rsp USERGS_SYSRET32 @@ -424,6 +430,7 @@ ENTRY(ia32_syscall) PARAVIRT_ADJUST_EXCEPTION_FRAME ASM_CLAC /* Do this early to minimize exposure */ SWAPGS + SWITCH_KERNEL_CR3_NO_STACK /* * No need to follow this irqs on/off section: the syscall * disabled irqs and here we enable it straight after entry: diff --git a/arch/x86/include/asm/alternative-asm.h b/arch/x86/include/asm/alternative-asm.h index 372231c22a47..524bddce0b76 100644 --- a/arch/x86/include/asm/alternative-asm.h +++ b/arch/x86/include/asm/alternative-asm.h @@ -18,12 +18,53 @@ .endm #endif -.macro altinstruction_entry orig alt feature orig_len alt_len +.macro altinstruction_entry orig alt feature orig_len alt_len pad_len .long \orig - . .long \alt - . .word \feature .byte \orig_len .byte \alt_len + .byte \pad_len +.endm + +.macro ALTERNATIVE oldinstr, newinstr, feature +140: + \oldinstr +141: + .skip -(((144f-143f)-(141b-140b)) > 0) * ((144f-143f)-(141b-140b)),0x90 +142: + + .pushsection .altinstructions,"a" + altinstruction_entry 140b,143f,\feature,142b-140b,144f-143f,142b-141b + .popsection + + .pushsection .altinstr_replacement,"ax" +143: + \newinstr +144: + .popsection +.endm + +.macro ALTERNATIVE_2 oldinstr, newinstr1, feature1, newinstr2, feature2 +140: + \oldinstr +141: + .skip -(((144f-143f)-(141b-140b)) > 0) * ((144f-143f)-(141b-140b)),0x90 + .skip -(((145f-144f)-(144f-143f)-(141b-140b)) > 0) * ((145f-144f)-(144f-143f)-(141b-140b)),0x90 +142: + + .pushsection .altinstructions,"a" + altinstruction_entry 140b,143f,\feature1,142b-140b,144f-143f,142b-141b + altinstruction_entry 140b,144f,\feature2,142b-140b,145f-144f,142b-141b + .popsection + + .pushsection .altinstr_replacement,"ax" +143: + \newinstr1 +144: + \newinstr2 +145: + .popsection .endm #endif /* __ASSEMBLY__ */ diff --git a/arch/x86/include/asm/alternative.h b/arch/x86/include/asm/alternative.h index 0a3f9c9f98d5..34310c03708a 100644 --- a/arch/x86/include/asm/alternative.h +++ b/arch/x86/include/asm/alternative.h @@ -48,8 +48,9 @@ struct alt_instr { s32 repl_offset; /* offset to replacement instruction */ u16 cpuid; /* cpuid bit set for replacement */ u8 instrlen; /* length of original instruction */ - u8 replacementlen; /* length of new instruction, <= instrlen */ -}; + u8 replacementlen; /* length of new instruction */ + u8 padlen; /* length of build-time padding */ +} __packed; extern void alternative_instructions(void); extern void apply_alternatives(struct alt_instr *start, struct alt_instr *end); @@ -76,50 +77,61 @@ static inline int alternatives_text_reserved(void *start, void *end) } #endif /* CONFIG_SMP */ -#define OLDINSTR(oldinstr) "661:\n\t" oldinstr "\n662:\n" +#define b_replacement(num) "664"#num +#define e_replacement(num) "665"#num -#define b_replacement(number) "663"#number -#define e_replacement(number) "664"#number +#define alt_end_marker "663" +#define alt_slen "662b-661b" +#define alt_pad_len alt_end_marker"b-662b" +#define alt_total_slen alt_end_marker"b-661b" +#define alt_rlen(num) e_replacement(num)"f-"b_replacement(num)"f" -#define alt_slen "662b-661b" -#define alt_rlen(number) e_replacement(number)"f-"b_replacement(number)"f" +#define __OLDINSTR(oldinstr, num) \ + "661:\n\t" oldinstr "\n662:\n" \ + ".skip -(((" alt_rlen(num) ")-(" alt_slen ")) > 0) * " \ + "((" alt_rlen(num) ")-(" alt_slen ")),0x90\n" -#define ALTINSTR_ENTRY(feature, number) \ +#define OLDINSTR(oldinstr, num) \ + __OLDINSTR(oldinstr, num) \ + alt_end_marker ":\n" + +/* + * Pad the second replacement alternative with additional NOPs if it is + * additionally longer than the first replacement alternative. + */ +#define OLDINSTR_2(oldinstr, num1, num2) \ + __OLDINSTR(oldinstr, num1) \ + ".skip -(((" alt_rlen(num2) ")-(" alt_rlen(num1) ")-(662b-661b)) > 0) * " \ + "((" alt_rlen(num2) ")-(" alt_rlen(num1) ")-(662b-661b)),0x90\n" \ + alt_end_marker ":\n" + +#define ALTINSTR_ENTRY(feature, num) \ " .long 661b - .\n" /* label */ \ - " .long " b_replacement(number)"f - .\n" /* new instruction */ \ + " .long " b_replacement(num)"f - .\n" /* new instruction */ \ " .word " __stringify(feature) "\n" /* feature bit */ \ - " .byte " alt_slen "\n" /* source len */ \ - " .byte " alt_rlen(number) "\n" /* replacement len */ + " .byte " alt_total_slen "\n" /* source len */ \ + " .byte " alt_rlen(num) "\n" /* replacement len */ \ + " .byte " alt_pad_len "\n" /* pad len */ -#define DISCARD_ENTRY(number) /* rlen <= slen */ \ - " .byte 0xff + (" alt_rlen(number) ") - (" alt_slen ")\n" - -#define ALTINSTR_REPLACEMENT(newinstr, feature, number) /* replacement */ \ - b_replacement(number)":\n\t" newinstr "\n" e_replacement(number) ":\n\t" +#define ALTINSTR_REPLACEMENT(newinstr, feature, num) /* replacement */ \ + b_replacement(num)":\n\t" newinstr "\n" e_replacement(num) ":\n\t" /* alternative assembly primitive: */ #define ALTERNATIVE(oldinstr, newinstr, feature) \ - OLDINSTR(oldinstr) \ + OLDINSTR(oldinstr, 1) \ ".pushsection .altinstructions,\"a\"\n" \ ALTINSTR_ENTRY(feature, 1) \ ".popsection\n" \ - ".pushsection .discard,\"aw\",@progbits\n" \ - DISCARD_ENTRY(1) \ - ".popsection\n" \ ".pushsection .altinstr_replacement, \"ax\"\n" \ ALTINSTR_REPLACEMENT(newinstr, feature, 1) \ ".popsection" #define ALTERNATIVE_2(oldinstr, newinstr1, feature1, newinstr2, feature2)\ - OLDINSTR(oldinstr) \ + OLDINSTR_2(oldinstr, 1, 2) \ ".pushsection .altinstructions,\"a\"\n" \ ALTINSTR_ENTRY(feature1, 1) \ ALTINSTR_ENTRY(feature2, 2) \ ".popsection\n" \ - ".pushsection .discard,\"aw\",@progbits\n" \ - DISCARD_ENTRY(1) \ - DISCARD_ENTRY(2) \ - ".popsection\n" \ ".pushsection .altinstr_replacement, \"ax\"\n" \ ALTINSTR_REPLACEMENT(newinstr1, feature1, 1) \ ALTINSTR_REPLACEMENT(newinstr2, feature2, 2) \ @@ -146,6 +158,9 @@ static inline int alternatives_text_reserved(void *start, void *end) #define alternative(oldinstr, newinstr, feature) \ asm volatile (ALTERNATIVE(oldinstr, newinstr, feature) : : : "memory") +#define alternative_2(oldinstr, newinstr1, feature1, newinstr2, feature2) \ + asm volatile(ALTERNATIVE_2(oldinstr, newinstr1, feature1, newinstr2, feature2) ::: "memory") + /* * Alternative inline assembly with input. * diff --git a/arch/x86/include/asm/cmdline.h b/arch/x86/include/asm/cmdline.h index e01f7f7ccb0c..84ae170bc3d0 100644 --- a/arch/x86/include/asm/cmdline.h +++ b/arch/x86/include/asm/cmdline.h @@ -2,5 +2,7 @@ #define _ASM_X86_CMDLINE_H int cmdline_find_option_bool(const char *cmdline_ptr, const char *option); +int cmdline_find_option(const char *cmdline_ptr, const char *option, + char *buffer, int bufsize); #endif /* _ASM_X86_CMDLINE_H */ diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h index 51f75503541d..6b8fc931a973 100644 --- a/arch/x86/include/asm/cpufeature.h +++ b/arch/x86/include/asm/cpufeature.h @@ -186,6 +186,10 @@ #define X86_FEATURE_DTHERM (7*32+ 7) /* Digital Thermal Sensor */ #define X86_FEATURE_HW_PSTATE (7*32+ 8) /* AMD HW-PState */ #define X86_FEATURE_PROC_FEEDBACK (7*32+ 9) /* AMD ProcFeedbackInterface */ +#define X86_FEATURE_INVPCID_SINGLE (7*32+10) /* Effectively INVPCID && CR4.PCIDE=1 */ + +/* Because the ALTERNATIVE scheme is for members of the X86_FEATURE club... */ +#define X86_FEATURE_KAISER (7*32+31) /* "" CONFIG_PAGE_TABLE_ISOLATION w/o nokaiser */ /* Virtualization flags: Linux defined, word 8 */ #define X86_FEATURE_TPR_SHADOW (8*32+ 0) /* Intel TPR Shadow */ @@ -389,6 +393,7 @@ static __always_inline __pure bool __static_cpu_has(u16 bit) " .word %P0\n" /* 1: do replace */ " .byte 2b - 1b\n" /* source len */ " .byte 0\n" /* replacement len */ + " .byte 0\n" /* pad len */ ".previous\n" /* skipping size check since replacement size = 0 */ : : "i" (X86_FEATURE_ALWAYS) : : t_warn); @@ -403,6 +408,7 @@ static __always_inline __pure bool __static_cpu_has(u16 bit) " .word %P0\n" /* feature bit */ " .byte 2b - 1b\n" /* source len */ " .byte 0\n" /* replacement len */ + " .byte 0\n" /* pad len */ ".previous\n" /* skipping size check since replacement size = 0 */ : : "i" (bit) : : t_no); @@ -428,6 +434,7 @@ static __always_inline __pure bool __static_cpu_has(u16 bit) " .word %P1\n" /* feature bit */ " .byte 2b - 1b\n" /* source len */ " .byte 4f - 3f\n" /* replacement len */ + " .byte 0\n" /* pad len */ ".previous\n" ".section .discard,\"aw\",@progbits\n" " .byte 0xff + (4f-3f) - (2b-1b)\n" /* size check */ @@ -454,31 +461,30 @@ static __always_inline __pure bool __static_cpu_has(u16 bit) static __always_inline __pure bool _static_cpu_has_safe(u16 bit) { #ifdef CC_HAVE_ASM_GOTO -/* - * We need to spell the jumps to the compiler because, depending on the offset, - * the replacement jump can be bigger than the original jump, and this we cannot - * have. Thus, we force the jump to the widest, 4-byte, signed relative - * offset even though the last would often fit in less bytes. - */ - asm_volatile_goto("1: .byte 0xe9\n .long %l[t_dynamic] - 2f\n" + asm_volatile_goto("1: jmp %l[t_dynamic]\n" "2:\n" + ".skip -(((5f-4f) - (2b-1b)) > 0) * " + "((5f-4f) - (2b-1b)),0x90\n" + "3:\n" ".section .altinstructions,\"a\"\n" " .long 1b - .\n" /* src offset */ - " .long 3f - .\n" /* repl offset */ + " .long 4f - .\n" /* repl offset */ " .word %P1\n" /* always replace */ - " .byte 2b - 1b\n" /* src len */ - " .byte 4f - 3f\n" /* repl len */ + " .byte 3b - 1b\n" /* src len */ + " .byte 5f - 4f\n" /* repl len */ + " .byte 3b - 2b\n" /* pad len */ ".previous\n" ".section .altinstr_replacement,\"ax\"\n" - "3: .byte 0xe9\n .long %l[t_no] - 2b\n" - "4:\n" + "4: jmp %l[t_no]\n" + "5:\n" ".previous\n" ".section .altinstructions,\"a\"\n" " .long 1b - .\n" /* src offset */ " .long 0\n" /* no replacement */ " .word %P0\n" /* feature bit */ - " .byte 2b - 1b\n" /* src len */ + " .byte 3b - 1b\n" /* src len */ " .byte 0\n" /* repl len */ + " .byte 0\n" /* pad len */ ".previous\n" : : "i" (bit), "i" (X86_FEATURE_ALWAYS) : : t_dynamic, t_no); @@ -498,6 +504,7 @@ static __always_inline __pure bool _static_cpu_has_safe(u16 bit) " .word %P2\n" /* always replace */ " .byte 2b - 1b\n" /* source len */ " .byte 4f - 3f\n" /* replacement len */ + " .byte 0\n" /* pad len */ ".previous\n" ".section .discard,\"aw\",@progbits\n" " .byte 0xff + (4f-3f) - (2b-1b)\n" /* size check */ @@ -512,6 +519,7 @@ static __always_inline __pure bool _static_cpu_has_safe(u16 bit) " .word %P1\n" /* feature bit */ " .byte 4b - 3b\n" /* src len */ " .byte 6f - 5f\n" /* repl len */ + " .byte 0\n" /* pad len */ ".previous\n" ".section .discard,\"aw\",@progbits\n" " .byte 0xff + (6f-5f) - (4b-3b)\n" /* size check */ diff --git a/arch/x86/include/asm/desc.h b/arch/x86/include/asm/desc.h index 69126184c609..2022e4181742 100644 --- a/arch/x86/include/asm/desc.h +++ b/arch/x86/include/asm/desc.h @@ -43,7 +43,7 @@ struct gdt_page { struct desc_struct gdt[GDT_ENTRIES]; } __attribute__((aligned(PAGE_SIZE))); -DECLARE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page); +DECLARE_PER_CPU_PAGE_ALIGNED_USER_MAPPED(struct gdt_page, gdt_page); static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu) { diff --git a/arch/x86/include/asm/hardirq.h b/arch/x86/include/asm/hardirq.h index 230853da4ec0..07e1325ac8db 100644 --- a/arch/x86/include/asm/hardirq.h +++ b/arch/x86/include/asm/hardirq.h @@ -21,12 +21,8 @@ typedef struct { #ifdef CONFIG_SMP unsigned int irq_resched_count; unsigned int irq_call_count; - /* - * irq_tlb_count is double-counted in irq_call_count, so it must be - * subtracted from irq_call_count when displaying irq_call_count - */ - unsigned int irq_tlb_count; #endif + unsigned int irq_tlb_count; #ifdef CONFIG_X86_THERMAL_VECTOR unsigned int irq_thermal_count; #endif diff --git a/arch/x86/include/asm/hw_irq.h b/arch/x86/include/asm/hw_irq.h index 4615906d83df..848d6f946203 100644 --- a/arch/x86/include/asm/hw_irq.h +++ b/arch/x86/include/asm/hw_irq.h @@ -194,7 +194,7 @@ extern void (*__initconst interrupt[NR_VECTORS-FIRST_EXTERNAL_VECTOR])(void); #define VECTOR_RETRIGGERED (-2) typedef int vector_irq_t[NR_VECTORS]; -DECLARE_PER_CPU(vector_irq_t, vector_irq); +DECLARE_PER_CPU_USER_MAPPED(vector_irq_t, vector_irq); extern void setup_vector_irq(int cpu); #ifdef CONFIG_X86_IO_APIC diff --git a/arch/x86/include/asm/kaiser.h b/arch/x86/include/asm/kaiser.h new file mode 100644 index 000000000000..802bbbdfe143 --- /dev/null +++ b/arch/x86/include/asm/kaiser.h @@ -0,0 +1,141 @@ +#ifndef _ASM_X86_KAISER_H +#define _ASM_X86_KAISER_H + +#include /* For PCID constants */ + +/* + * This file includes the definitions for the KAISER feature. + * KAISER is a counter measure against x86_64 side channel attacks on + * the kernel virtual memory. It has a shadow pgd for every process: the + * shadow pgd has a minimalistic kernel-set mapped, but includes the whole + * user memory. Within a kernel context switch, or when an interrupt is handled, + * the pgd is switched to the normal one. When the system switches to user mode, + * the shadow pgd is enabled. By this, the virtual memory caches are freed, + * and the user may not attack the whole kernel memory. + * + * A minimalistic kernel mapping holds the parts needed to be mapped in user + * mode, such as the entry/exit functions of the user space, or the stacks. + */ + +#define KAISER_SHADOW_PGD_OFFSET 0x1000 + +#ifdef __ASSEMBLY__ +#ifdef CONFIG_PAGE_TABLE_ISOLATION + +.macro _SWITCH_TO_KERNEL_CR3 reg +movq %cr3, \reg +andq $(~(X86_CR3_PCID_ASID_MASK | KAISER_SHADOW_PGD_OFFSET)), \reg +/* If PCID enabled, set X86_CR3_PCID_NOFLUSH_BIT */ +ALTERNATIVE "", "bts $63, \reg", X86_FEATURE_PCID +movq \reg, %cr3 +.endm + +.macro _SWITCH_TO_USER_CR3 reg regb +/* + * regb must be the low byte portion of reg: because we have arranged + * for the low byte of the user PCID to serve as the high byte of NOFLUSH + * (0x80 for each when PCID is enabled, or 0x00 when PCID and NOFLUSH are + * not enabled): so that the one register can update both memory and cr3. + */ +movq %cr3, \reg +orq PER_CPU_VAR(x86_cr3_pcid_user), \reg +js 9f +/* If PCID enabled, FLUSH this time, reset to NOFLUSH for next time */ +movb \regb, PER_CPU_VAR(x86_cr3_pcid_user+7) +9: +movq \reg, %cr3 +.endm + +.macro SWITCH_KERNEL_CR3 +ALTERNATIVE "jmp 8f", "pushq %rax", X86_FEATURE_KAISER +_SWITCH_TO_KERNEL_CR3 %rax +popq %rax +8: +.endm + +.macro SWITCH_USER_CR3 +ALTERNATIVE "jmp 8f", "pushq %rax", X86_FEATURE_KAISER +_SWITCH_TO_USER_CR3 %rax %al +popq %rax +8: +.endm + +.macro SWITCH_KERNEL_CR3_NO_STACK +ALTERNATIVE "jmp 8f", \ + __stringify(movq %rax, PER_CPU_VAR(unsafe_stack_register_backup)), \ + X86_FEATURE_KAISER +_SWITCH_TO_KERNEL_CR3 %rax +movq PER_CPU_VAR(unsafe_stack_register_backup), %rax +8: +.endm + +#else /* CONFIG_PAGE_TABLE_ISOLATION */ + +.macro SWITCH_KERNEL_CR3 +.endm +.macro SWITCH_USER_CR3 +.endm +.macro SWITCH_KERNEL_CR3_NO_STACK +.endm + +#endif /* CONFIG_PAGE_TABLE_ISOLATION */ + +#else /* __ASSEMBLY__ */ + +#ifdef CONFIG_PAGE_TABLE_ISOLATION +/* + * Upon kernel/user mode switch, it may happen that the address + * space has to be switched before the registers have been + * stored. To change the address space, another register is + * needed. A register therefore has to be stored/restored. +*/ +DECLARE_PER_CPU_USER_MAPPED(unsigned long, unsafe_stack_register_backup); + +DECLARE_PER_CPU(unsigned long, x86_cr3_pcid_user); + +extern char __per_cpu_user_mapped_start[], __per_cpu_user_mapped_end[]; + +extern int kaiser_enabled; +extern void __init kaiser_check_boottime_disable(void); +#else +#define kaiser_enabled 0 +static inline void __init kaiser_check_boottime_disable(void) {} +#endif /* CONFIG_PAGE_TABLE_ISOLATION */ + +/* + * Kaiser function prototypes are needed even when CONFIG_PAGE_TABLE_ISOLATION is not set, + * so as to build with tests on kaiser_enabled instead of #ifdefs. + */ + +/** + * kaiser_add_mapping - map a virtual memory part to the shadow (user) mapping + * @addr: the start address of the range + * @size: the size of the range + * @flags: The mapping flags of the pages + * + * The mapping is done on a global scope, so no bigger + * synchronization has to be done. the pages have to be + * manually unmapped again when they are not needed any longer. + */ +extern int kaiser_add_mapping(unsigned long addr, unsigned long size, unsigned long flags); + +/** + * kaiser_remove_mapping - unmap a virtual memory part of the shadow mapping + * @addr: the start address of the range + * @size: the size of the range + */ +extern void kaiser_remove_mapping(unsigned long start, unsigned long size); + +/** + * kaiser_init - Initialize the shadow mapping + * + * Most parts of the shadow mapping can be mapped upon boot + * time. Only per-process things like the thread stacks + * or a new LDT have to be mapped at runtime. These boot- + * time mappings are permanent and never unmapped. + */ +extern void kaiser_init(void); + +#endif /* __ASSEMBLY */ + +#endif /* _ASM_X86_KAISER_H */ diff --git a/arch/x86/include/asm/mmu.h b/arch/x86/include/asm/mmu.h index b6b7bc3f5d26..3bc5c0b13cd7 100644 --- a/arch/x86/include/asm/mmu.h +++ b/arch/x86/include/asm/mmu.h @@ -20,12 +20,6 @@ typedef struct { void __user *vdso; } mm_context_t; -#ifdef CONFIG_SMP void leave_mm(int cpu); -#else -static inline void leave_mm(int cpu) -{ -} -#endif #endif /* _ASM_X86_MMU_H */ diff --git a/arch/x86/include/asm/mmu_context.h b/arch/x86/include/asm/mmu_context.h index 50f622dc0b1a..740e65890eaa 100644 --- a/arch/x86/include/asm/mmu_context.h +++ b/arch/x86/include/asm/mmu_context.h @@ -68,89 +68,16 @@ void destroy_context(struct mm_struct *mm); static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk) { -#ifdef CONFIG_SMP if (this_cpu_read(cpu_tlbstate.state) == TLBSTATE_OK) this_cpu_write(cpu_tlbstate.state, TLBSTATE_LAZY); -#endif } -static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next, - struct task_struct *tsk) -{ - unsigned cpu = smp_processor_id(); +extern void switch_mm(struct mm_struct *prev, struct mm_struct *next, + struct task_struct *tsk); - if (likely(prev != next)) { -#ifdef CONFIG_SMP - this_cpu_write(cpu_tlbstate.state, TLBSTATE_OK); - this_cpu_write(cpu_tlbstate.active_mm, next); -#endif - cpumask_set_cpu(cpu, mm_cpumask(next)); - - /* - * Re-load page tables. - * - * This logic has an ordering constraint: - * - * CPU 0: Write to a PTE for 'next' - * CPU 0: load bit 1 in mm_cpumask. if nonzero, send IPI. - * CPU 1: set bit 1 in next's mm_cpumask - * CPU 1: load from the PTE that CPU 0 writes (implicit) - * - * We need to prevent an outcome in which CPU 1 observes - * the new PTE value and CPU 0 observes bit 1 clear in - * mm_cpumask. (If that occurs, then the IPI will never - * be sent, and CPU 0's TLB will contain a stale entry.) - * - * The bad outcome can occur if either CPU's load is - * reordered before that CPU's store, so both CPUs must - * execute full barriers to prevent this from happening. - * - * Thus, switch_mm needs a full barrier between the - * store to mm_cpumask and any operation that could load - * from next->pgd. TLB fills are special and can happen - * due to instruction fetches or for no reason at all, - * and neither LOCK nor MFENCE orders them. - * Fortunately, load_cr3() is serializing and gives the - * ordering guarantee we need. - * - */ - load_cr3(next->pgd); - - /* Stop flush ipis for the previous mm */ - cpumask_clear_cpu(cpu, mm_cpumask(prev)); - - /* Load the LDT, if the LDT is different: */ - if (unlikely(prev->context.ldt != next->context.ldt)) - load_mm_ldt(next); - } -#ifdef CONFIG_SMP - else { - this_cpu_write(cpu_tlbstate.state, TLBSTATE_OK); - BUG_ON(this_cpu_read(cpu_tlbstate.active_mm) != next); - - if (!cpumask_test_cpu(cpu, mm_cpumask(next))) { - /* - * On established mms, the mm_cpumask is only changed - * from irq context, from ptep_clear_flush() while in - * lazy tlb mode, and here. Irqs are blocked during - * schedule, protecting us from simultaneous changes. - */ - cpumask_set_cpu(cpu, mm_cpumask(next)); - - /* - * We were in lazy tlb mode and leave_mm disabled - * tlb flush IPI delivery. We must reload CR3 - * to make sure to use no freed page tables. - * - * As above, load_cr3() is serializing and orders TLB - * fills with respect to the mm_cpumask write. - */ - load_cr3(next->pgd); - load_mm_ldt(next); - } - } -#endif -} +extern void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next, + struct task_struct *tsk); +#define switch_mm_irqs_off switch_mm_irqs_off #define activate_mm(prev, next) \ do { \ diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h index aa97a070f09f..1b5b34cba964 100644 --- a/arch/x86/include/asm/pgtable.h +++ b/arch/x86/include/asm/pgtable.h @@ -17,6 +17,12 @@ #ifndef __ASSEMBLY__ #include +#ifdef CONFIG_PAGE_TABLE_ISOLATION +extern int kaiser_enabled; +#else +#define kaiser_enabled 0 +#endif + void ptdump_walk_pgd_level(struct seq_file *m, pgd_t *pgd); /* @@ -641,7 +647,17 @@ static inline pud_t *pud_offset(pgd_t *pgd, unsigned long address) static inline int pgd_bad(pgd_t pgd) { - return (pgd_flags(pgd) & ~_PAGE_USER) != _KERNPG_TABLE; + pgdval_t ignore_flags = _PAGE_USER; + /* + * We set NX on KAISER pgds that map userspace memory so + * that userspace can not meaningfully use the kernel + * page table by accident; it will fault on the first + * instruction it tries to run. See native_set_pgd(). + */ + if (kaiser_enabled) + ignore_flags |= _PAGE_NX; + + return (pgd_flags(pgd) & ~ignore_flags) != _KERNPG_TABLE; } static inline int pgd_none(pgd_t pgd) @@ -843,7 +859,15 @@ static inline void pmdp_set_wrprotect(struct mm_struct *mm, */ static inline void clone_pgd_range(pgd_t *dst, pgd_t *src, int count) { - memcpy(dst, src, count * sizeof(pgd_t)); + memcpy(dst, src, count * sizeof(pgd_t)); +#ifdef CONFIG_PAGE_TABLE_ISOLATION + if (kaiser_enabled) { + /* Clone the shadow pgd part as well */ + memcpy(native_get_shadow_pgd(dst), + native_get_shadow_pgd(src), + count * sizeof(pgd_t)); + } +#endif } #define PTE_SHIFT ilog2(PTRS_PER_PTE) diff --git a/arch/x86/include/asm/pgtable_64.h b/arch/x86/include/asm/pgtable_64.h index 3874693c0e53..ada2cb8adbc2 100644 --- a/arch/x86/include/asm/pgtable_64.h +++ b/arch/x86/include/asm/pgtable_64.h @@ -106,9 +106,32 @@ static inline void native_pud_clear(pud_t *pud) native_set_pud(pud, native_make_pud(0)); } +#ifdef CONFIG_PAGE_TABLE_ISOLATION +extern pgd_t kaiser_set_shadow_pgd(pgd_t *pgdp, pgd_t pgd); + +static inline pgd_t *native_get_shadow_pgd(pgd_t *pgdp) +{ +#ifdef CONFIG_DEBUG_VM + /* linux/mmdebug.h may not have been included at this point */ + BUG_ON(!kaiser_enabled); +#endif + return (pgd_t *)((unsigned long)pgdp | (unsigned long)PAGE_SIZE); +} +#else +static inline pgd_t kaiser_set_shadow_pgd(pgd_t *pgdp, pgd_t pgd) +{ + return pgd; +} +static inline pgd_t *native_get_shadow_pgd(pgd_t *pgdp) +{ + BUILD_BUG_ON(1); + return NULL; +} +#endif /* CONFIG_PAGE_TABLE_ISOLATION */ + static inline void native_set_pgd(pgd_t *pgdp, pgd_t pgd) { - *pgdp = pgd; + *pgdp = kaiser_set_shadow_pgd(pgdp, pgd); } static inline void native_pgd_clear(pgd_t *pgd) diff --git a/arch/x86/include/asm/pgtable_types.h b/arch/x86/include/asm/pgtable_types.h index f216963760e5..c5d7703dc591 100644 --- a/arch/x86/include/asm/pgtable_types.h +++ b/arch/x86/include/asm/pgtable_types.h @@ -116,7 +116,7 @@ #endif #define _PAGE_FILE (_AT(pteval_t, 1) << _PAGE_BIT_FILE) -#define _PAGE_PROTNONE (_AT(pteval_t, 1) << _PAGE_BIT_PROTNONE) +#define _PAGE_PROTNONE (_AT(pteval_t, 1) << _PAGE_BIT_PROTNONE) #define _PAGE_TABLE (_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | \ _PAGE_ACCESSED | _PAGE_DIRTY) @@ -129,6 +129,33 @@ _PAGE_SOFT_DIRTY | _PAGE_NUMA) #define _HPAGE_CHG_MASK (_PAGE_CHG_MASK | _PAGE_PSE | _PAGE_NUMA) +/* The ASID is the lower 12 bits of CR3 */ +#define X86_CR3_PCID_ASID_MASK (_AC((1<<12)-1,UL)) + +/* Mask for all the PCID-related bits in CR3: */ +#define X86_CR3_PCID_MASK (X86_CR3_PCID_NOFLUSH | X86_CR3_PCID_ASID_MASK) +#define X86_CR3_PCID_ASID_KERN (_AC(0x0,UL)) + +#if defined(CONFIG_PAGE_TABLE_ISOLATION) && defined(CONFIG_X86_64) +/* Let X86_CR3_PCID_ASID_USER be usable for the X86_CR3_PCID_NOFLUSH bit */ +#define X86_CR3_PCID_ASID_USER (_AC(0x80,UL)) + +#define X86_CR3_PCID_KERN_FLUSH (X86_CR3_PCID_ASID_KERN) +#define X86_CR3_PCID_USER_FLUSH (X86_CR3_PCID_ASID_USER) +#define X86_CR3_PCID_KERN_NOFLUSH (X86_CR3_PCID_NOFLUSH | X86_CR3_PCID_ASID_KERN) +#define X86_CR3_PCID_USER_NOFLUSH (X86_CR3_PCID_NOFLUSH | X86_CR3_PCID_ASID_USER) +#else +#define X86_CR3_PCID_ASID_USER (_AC(0x0,UL)) +/* + * PCIDs are unsupported on 32-bit and none of these bits can be + * set in CR3: + */ +#define X86_CR3_PCID_KERN_FLUSH (0) +#define X86_CR3_PCID_USER_FLUSH (0) +#define X86_CR3_PCID_KERN_NOFLUSH (0) +#define X86_CR3_PCID_USER_NOFLUSH (0) +#endif + #define _PAGE_CACHE_MASK (_PAGE_PCD | _PAGE_PWT) #define _PAGE_CACHE_WB (0) #define _PAGE_CACHE_WC (_PAGE_PWT) diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h index a4ea02351f4d..4a2f66843288 100644 --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -72,7 +72,6 @@ extern u16 __read_mostly tlb_lld_4k[NR_INFO]; extern u16 __read_mostly tlb_lld_2m[NR_INFO]; extern u16 __read_mostly tlb_lld_4m[NR_INFO]; extern u16 __read_mostly tlb_lld_1g[NR_INFO]; -extern s8 __read_mostly tlb_flushall_shift; /* * CPU type and hardware bug flags. Kept separately for each CPU. @@ -283,7 +282,7 @@ struct tss_struct { } ____cacheline_aligned; -DECLARE_PER_CPU_SHARED_ALIGNED(struct tss_struct, init_tss); +DECLARE_PER_CPU_SHARED_ALIGNED_USER_MAPPED(struct tss_struct, init_tss); /* * Save the original ist values for checking stack pointers during debugging @@ -579,39 +578,6 @@ static inline void load_sp0(struct tss_struct *tss, #define set_iopl_mask native_set_iopl_mask #endif /* CONFIG_PARAVIRT */ -/* - * Save the cr4 feature set we're using (ie - * Pentium 4MB enable and PPro Global page - * enable), so that any CPU's that boot up - * after us can get the correct flags. - */ -extern unsigned long mmu_cr4_features; -extern u32 *trampoline_cr4_features; - -static inline void set_in_cr4(unsigned long mask) -{ - unsigned long cr4; - - mmu_cr4_features |= mask; - if (trampoline_cr4_features) - *trampoline_cr4_features = mmu_cr4_features; - cr4 = read_cr4(); - cr4 |= mask; - write_cr4(cr4); -} - -static inline void clear_in_cr4(unsigned long mask) -{ - unsigned long cr4; - - mmu_cr4_features &= ~mask; - if (trampoline_cr4_features) - *trampoline_cr4_features = mmu_cr4_features; - cr4 = read_cr4(); - cr4 &= ~mask; - write_cr4(cr4); -} - typedef struct { unsigned long seg; } mm_segment_t; @@ -930,7 +896,7 @@ extern unsigned long KSTK_ESP(struct task_struct *task); /* * User space RSP while inside the SYSCALL fast path */ -DECLARE_PER_CPU(unsigned long, old_rsp); +DECLARE_PER_CPU_USER_MAPPED(unsigned long, old_rsp); #endif /* CONFIG_X86_64 */ diff --git a/arch/x86/include/asm/smap.h b/arch/x86/include/asm/smap.h index 8d3120f4e270..c56cb4f37be9 100644 --- a/arch/x86/include/asm/smap.h +++ b/arch/x86/include/asm/smap.h @@ -33,7 +33,7 @@ 662: __ASM_CLAC ; \ .popsection ; \ .pushsection .altinstructions, "a" ; \ - altinstruction_entry 661b, 662b, X86_FEATURE_SMAP, 3, 3 ; \ + altinstruction_entry 661b, 662b, X86_FEATURE_SMAP, 3, 3, 0 ; \ .popsection #define ASM_STAC \ @@ -42,7 +42,7 @@ 662: __ASM_STAC ; \ .popsection ; \ .pushsection .altinstructions, "a" ; \ - altinstruction_entry 661b, 662b, X86_FEATURE_SMAP, 3, 3 ; \ + altinstruction_entry 661b, 662b, X86_FEATURE_SMAP, 3, 3, 0 ; \ .popsection #else /* CONFIG_X86_SMAP */ diff --git a/arch/x86/include/asm/tlbflush.h b/arch/x86/include/asm/tlbflush.h index 5e4b0cc54e43..bd10f6775dfd 100644 --- a/arch/x86/include/asm/tlbflush.h +++ b/arch/x86/include/asm/tlbflush.h @@ -6,6 +6,55 @@ #include #include +#include + +static inline void __invpcid(unsigned long pcid, unsigned long addr, + unsigned long type) +{ + struct { u64 d[2]; } desc = { { pcid, addr } }; + + /* + * The memory clobber is because the whole point is to invalidate + * stale TLB entries and, especially if we're flushing global + * mappings, we don't want the compiler to reorder any subsequent + * memory accesses before the TLB flush. + * + * The hex opcode is invpcid (%ecx), %eax in 32-bit mode and + * invpcid (%rcx), %rax in long mode. + */ + asm volatile (".byte 0x66, 0x0f, 0x38, 0x82, 0x01" + : : "m" (desc), "a" (type), "c" (&desc) : "memory"); +} + +#define INVPCID_TYPE_INDIV_ADDR 0 +#define INVPCID_TYPE_SINGLE_CTXT 1 +#define INVPCID_TYPE_ALL_INCL_GLOBAL 2 +#define INVPCID_TYPE_ALL_NON_GLOBAL 3 + +/* Flush all mappings for a given pcid and addr, not including globals. */ +static inline void invpcid_flush_one(unsigned long pcid, + unsigned long addr) +{ + __invpcid(pcid, addr, INVPCID_TYPE_INDIV_ADDR); +} + +/* Flush all mappings for a given PCID, not including globals. */ +static inline void invpcid_flush_single_context(unsigned long pcid) +{ + __invpcid(pcid, 0, INVPCID_TYPE_SINGLE_CTXT); +} + +/* Flush all mappings, including globals, for all PCIDs. */ +static inline void invpcid_flush_all(void) +{ + __invpcid(0, 0, INVPCID_TYPE_ALL_INCL_GLOBAL); +} + +/* Flush all mappings for all PCIDs except globals. */ +static inline void invpcid_flush_all_nonglobals(void) +{ + __invpcid(0, 0, INVPCID_TYPE_ALL_NON_GLOBAL); +} #ifdef CONFIG_PARAVIRT #include @@ -15,6 +64,61 @@ #define __flush_tlb_single(addr) __native_flush_tlb_single(addr) #endif +/* Set in this cpu's CR4. */ +static inline void cr4_set_bits(unsigned long mask) +{ + unsigned long cr4; + + cr4 = read_cr4(); + cr4 |= mask; + write_cr4(cr4); +} + +/* Clear in this cpu's CR4. */ +static inline void cr4_clear_bits(unsigned long mask) +{ + unsigned long cr4; + + cr4 = read_cr4(); + cr4 &= ~mask; + write_cr4(cr4); +} + +/* + * Save some of cr4 feature set we're using (e.g. Pentium 4MB + * enable and PPro Global page enable), so that any CPU's that boot + * up after us can get the correct flags. This should only be used + * during boot on the boot cpu. + */ +extern unsigned long mmu_cr4_features; +extern u32 *trampoline_cr4_features; + +static inline void cr4_set_bits_and_update_boot(unsigned long mask) +{ + mmu_cr4_features |= mask; + if (trampoline_cr4_features) + *trampoline_cr4_features = mmu_cr4_features; + cr4_set_bits(mask); +} + +/* + * Declare a couple of kaiser interfaces here for convenience, + * to avoid the need for asm/kaiser.h in unexpected places. + */ +#ifdef CONFIG_PAGE_TABLE_ISOLATION +extern int kaiser_enabled; +extern void kaiser_setup_pcid(void); +extern void kaiser_flush_tlb_on_return_to_user(void); +#else +#define kaiser_enabled 0 +static inline void kaiser_setup_pcid(void) +{ +} +static inline void kaiser_flush_tlb_on_return_to_user(void) +{ +} +#endif + static inline void __native_flush_tlb(void) { /* @@ -23,6 +127,8 @@ static inline void __native_flush_tlb(void) * back: */ preempt_disable(); + if (kaiser_enabled) + kaiser_flush_tlb_on_return_to_user(); native_write_cr3(native_read_cr3()); preempt_enable(); } @@ -32,39 +138,84 @@ static inline void __native_flush_tlb_global_irq_disabled(void) unsigned long cr4; cr4 = native_read_cr4(); - /* clear PGE */ - native_write_cr4(cr4 & ~X86_CR4_PGE); - /* write old PGE again and flush TLBs */ - native_write_cr4(cr4); + if (cr4 & X86_CR4_PGE) { + /* clear PGE and flush TLB of all entries */ + native_write_cr4(cr4 & ~X86_CR4_PGE); + /* restore PGE as it was before */ + native_write_cr4(cr4); + } else { + /* do it with cr3, letting kaiser flush user PCID */ + __native_flush_tlb(); + } } static inline void __native_flush_tlb_global(void) { unsigned long flags; + if (this_cpu_has(X86_FEATURE_INVPCID)) { + /* + * Using INVPCID is considerably faster than a pair of writes + * to CR4 sandwiched inside an IRQ flag save/restore. + * + * Note, this works with CR4.PCIDE=0 or 1. + */ + invpcid_flush_all(); + return; + } + /* * Read-modify-write to CR4 - protect it from preemption and * from interrupts. (Use the raw variant because this code can * be called from deep inside debugging code.) */ raw_local_irq_save(flags); - __native_flush_tlb_global_irq_disabled(); - raw_local_irq_restore(flags); } static inline void __native_flush_tlb_single(unsigned long addr) { - asm volatile("invlpg (%0)" ::"r" (addr) : "memory"); + /* + * SIMICS #GP's if you run INVPCID with type 2/3 + * and X86_CR4_PCIDE clear. Shame! + * + * The ASIDs used below are hard-coded. But, we must not + * call invpcid(type=1/2) before CR4.PCIDE=1. Just call + * invlpg in the case we are called early. + */ + + if (!this_cpu_has(X86_FEATURE_INVPCID_SINGLE)) { + if (kaiser_enabled) + kaiser_flush_tlb_on_return_to_user(); + asm volatile("invlpg (%0)" ::"r" (addr) : "memory"); + return; + } + /* Flush the address out of both PCIDs. */ + /* + * An optimization here might be to determine addresses + * that are only kernel-mapped and only flush the kernel + * ASID. But, userspace flushes are probably much more + * important performance-wise. + * + * Make sure to do only a single invpcid when KAISER is + * disabled and we have only a single ASID. + */ + if (kaiser_enabled) + invpcid_flush_one(X86_CR3_PCID_ASID_USER, addr); + invpcid_flush_one(X86_CR3_PCID_ASID_KERN, addr); } static inline void __flush_tlb_all(void) { - if (cpu_has_pge) - __flush_tlb_global(); - else - __flush_tlb(); + __flush_tlb_global(); + /* + * Note: if we somehow had PCID but not PGE, then this wouldn't work -- + * we'd end up flushing kernel translations for the current ASID but + * we might fail to flush kernel translations for other cached ASIDs. + * + * To avoid this issue, we force PCID off if PGE is off. + */ } static inline void __flush_tlb_one(unsigned long addr) @@ -78,7 +229,6 @@ static inline void __flush_tlb_one(unsigned long addr) /* * TLB flushing: * - * - flush_tlb() flushes the current mm struct TLBs * - flush_tlb_all() flushes all processes TLBs * - flush_tlb_mm(mm) flushes the specified mm context TLB's * - flush_tlb_page(vma, vmaddr) flushes one page @@ -90,84 +240,6 @@ static inline void __flush_tlb_one(unsigned long addr) * and page-granular flushes are available only on i486 and up. */ -#ifndef CONFIG_SMP - -/* "_up" is for UniProcessor. - * - * This is a helper for other header functions. *Not* intended to be called - * directly. All global TLB flushes need to either call this, or to bump the - * vm statistics themselves. - */ -static inline void __flush_tlb_up(void) -{ - count_vm_tlb_event(NR_TLB_LOCAL_FLUSH_ALL); - __flush_tlb(); -} - -static inline void flush_tlb_all(void) -{ - count_vm_tlb_event(NR_TLB_LOCAL_FLUSH_ALL); - __flush_tlb_all(); -} - -static inline void flush_tlb(void) -{ - __flush_tlb_up(); -} - -static inline void local_flush_tlb(void) -{ - __flush_tlb_up(); -} - -static inline void flush_tlb_mm(struct mm_struct *mm) -{ - if (mm == current->active_mm) - __flush_tlb_up(); -} - -static inline void flush_tlb_page(struct vm_area_struct *vma, - unsigned long addr) -{ - if (vma->vm_mm == current->active_mm) - __flush_tlb_one(addr); -} - -static inline void flush_tlb_range(struct vm_area_struct *vma, - unsigned long start, unsigned long end) -{ - if (vma->vm_mm == current->active_mm) - __flush_tlb_up(); -} - -static inline void flush_tlb_mm_range(struct mm_struct *mm, - unsigned long start, unsigned long end, unsigned long vmflag) -{ - if (mm == current->active_mm) - __flush_tlb_up(); -} - -static inline void native_flush_tlb_others(const struct cpumask *cpumask, - struct mm_struct *mm, - unsigned long start, - unsigned long end) -{ -} - -static inline void reset_lazy_tlbstate(void) -{ -} - -static inline void flush_tlb_kernel_range(unsigned long start, - unsigned long end) -{ - flush_tlb_all(); -} - -#else /* SMP */ - -#include - #define local_flush_tlb() __flush_tlb() #define flush_tlb_mm(mm) flush_tlb_mm_range(mm, 0UL, TLB_FLUSH_ALL, 0UL) @@ -176,13 +248,14 @@ static inline void flush_tlb_kernel_range(unsigned long start, flush_tlb_mm_range(vma->vm_mm, start, end, vma->vm_flags) extern void flush_tlb_all(void); -extern void flush_tlb_current_task(void); -extern void flush_tlb_page(struct vm_area_struct *, unsigned long); extern void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start, unsigned long end, unsigned long vmflag); extern void flush_tlb_kernel_range(unsigned long start, unsigned long end); -#define flush_tlb() flush_tlb_current_task() +static inline void flush_tlb_page(struct vm_area_struct *vma, unsigned long a) +{ + flush_tlb_mm_range(vma->vm_mm, a, a + PAGE_SIZE, VM_NONE); +} void native_flush_tlb_others(const struct cpumask *cpumask, struct mm_struct *mm, @@ -203,8 +276,6 @@ static inline void reset_lazy_tlbstate(void) this_cpu_write(cpu_tlbstate.active_mm, &init_mm); } -#endif /* SMP */ - #ifndef CONFIG_PARAVIRT #define flush_tlb_others(mask, mm, start, end) \ native_flush_tlb_others(mask, mm, start, end) diff --git a/arch/x86/include/asm/virtext.h b/arch/x86/include/asm/virtext.h index 5da71c27cc59..f41e19ca717b 100644 --- a/arch/x86/include/asm/virtext.h +++ b/arch/x86/include/asm/virtext.h @@ -19,6 +19,7 @@ #include #include +#include /* * VMX functions: @@ -40,7 +41,7 @@ static inline int cpu_has_vmx(void) static inline void cpu_vmxoff(void) { asm volatile (ASM_VMX_VMXOFF : : : "cc"); - write_cr4(read_cr4() & ~X86_CR4_VMXE); + cr4_clear_bits(X86_CR4_VMXE); } static inline int cpu_vmx_enabled(void) diff --git a/arch/x86/include/asm/vsyscall.h b/arch/x86/include/asm/vsyscall.h index 2874be9aef0a..cbab442445ae 100644 --- a/arch/x86/include/asm/vsyscall.h +++ b/arch/x86/include/asm/vsyscall.h @@ -10,6 +10,7 @@ /* kernel space (writeable) */ extern int vgetcpu_mode; extern struct timezone sys_tz; +extern unsigned long vsyscall_pgprot; #include @@ -20,6 +21,7 @@ extern void map_vsyscall(void); * Returns true if handled. */ extern bool emulate_vsyscall(struct pt_regs *regs, unsigned long address); +extern bool vsyscall_enabled(void); #ifdef CONFIG_X86_64 diff --git a/arch/x86/include/uapi/asm/processor-flags.h b/arch/x86/include/uapi/asm/processor-flags.h index 180a0c3c224d..bd4513b7b877 100644 --- a/arch/x86/include/uapi/asm/processor-flags.h +++ b/arch/x86/include/uapi/asm/processor-flags.h @@ -79,7 +79,8 @@ #define X86_CR3_PWT _BITUL(X86_CR3_PWT_BIT) #define X86_CR3_PCD_BIT 4 /* Page Cache Disable */ #define X86_CR3_PCD _BITUL(X86_CR3_PCD_BIT) -#define X86_CR3_PCID_MASK _AC(0x00000fff,UL) /* PCID Mask */ +#define X86_CR3_PCID_NOFLUSH_BIT 63 /* Preserve old PCID */ +#define X86_CR3_PCID_NOFLUSH _BITULL(X86_CR3_PCID_NOFLUSH_BIT) /* * Intel CPU features in CR4 diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index 703130f469ec..af397cc98d05 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -52,10 +52,25 @@ static int __init setup_noreplace_paravirt(char *str) __setup("noreplace-paravirt", setup_noreplace_paravirt); #endif -#define DPRINTK(fmt, ...) \ -do { \ - if (debug_alternative) \ - printk(KERN_DEBUG fmt, ##__VA_ARGS__); \ +#define DPRINTK(fmt, args...) \ +do { \ + if (debug_alternative) \ + printk(KERN_DEBUG "%s: " fmt "\n", __func__, ##args); \ +} while (0) + +#define DUMP_BYTES(buf, len, fmt, args...) \ +do { \ + if (unlikely(debug_alternative)) { \ + int j; \ + \ + if (!(len)) \ + break; \ + \ + printk(KERN_DEBUG fmt, ##args); \ + for (j = 0; j < (len) - 1; j++) \ + printk(KERN_CONT "%02hhx ", buf[j]); \ + printk(KERN_CONT "%02hhx\n", buf[j]); \ + } \ } while (0) /* @@ -243,12 +258,86 @@ extern struct alt_instr __alt_instructions[], __alt_instructions_end[]; extern s32 __smp_locks[], __smp_locks_end[]; void *text_poke_early(void *addr, const void *opcode, size_t len); -/* Replace instructions with better alternatives for this CPU type. - This runs before SMP is initialized to avoid SMP problems with - self modifying code. This implies that asymmetric systems where - APs have less capabilities than the boot processor are not handled. - Tough. Make sure you disable such features by hand. */ +/* + * Are we looking at a near JMP with a 1 or 4-byte displacement. + */ +static inline bool is_jmp(const u8 opcode) +{ + return opcode == 0xeb || opcode == 0xe9; +} + +static void __init_or_module +recompute_jump(struct alt_instr *a, u8 *orig_insn, u8 *repl_insn, u8 *insnbuf) +{ + u8 *next_rip, *tgt_rip; + s32 n_dspl, o_dspl; + int repl_len; + + if (a->replacementlen != 5) + return; + + o_dspl = *(s32 *)(insnbuf + 1); + + /* next_rip of the replacement JMP */ + next_rip = repl_insn + a->replacementlen; + /* target rip of the replacement JMP */ + tgt_rip = next_rip + o_dspl; + n_dspl = tgt_rip - orig_insn; + + DPRINTK("target RIP: %p, new_displ: 0x%x", tgt_rip, n_dspl); + + if (tgt_rip - orig_insn >= 0) { + if (n_dspl - 2 <= 127) + goto two_byte_jmp; + else + goto five_byte_jmp; + /* negative offset */ + } else { + if (((n_dspl - 2) & 0xff) == (n_dspl - 2)) + goto two_byte_jmp; + else + goto five_byte_jmp; + } + +two_byte_jmp: + n_dspl -= 2; + + insnbuf[0] = 0xeb; + insnbuf[1] = (s8)n_dspl; + add_nops(insnbuf + 2, 3); + + repl_len = 2; + goto done; + +five_byte_jmp: + n_dspl -= 5; + insnbuf[0] = 0xe9; + *(s32 *)&insnbuf[1] = n_dspl; + + repl_len = 5; + +done: + + DPRINTK("final displ: 0x%08x, JMP 0x%lx", + n_dspl, (unsigned long)orig_insn + n_dspl + repl_len); +} + +static void __init_or_module optimize_nops(struct alt_instr *a, u8 *instr) +{ + add_nops(instr + (a->instrlen - a->padlen), a->padlen); + + DUMP_BYTES(instr, a->instrlen, "%p: [%d:%d) optimized NOPs: ", + instr, a->instrlen - a->padlen, a->padlen); +} + +/* + * Replace instructions with better alternatives for this CPU type. This runs + * before SMP is initialized to avoid SMP problems with self modifying code. + * This implies that asymmetric systems where APs have less capabilities than + * the boot processor are not handled. Tough. Make sure you disable such + * features by hand. + */ void __init_or_module apply_alternatives(struct alt_instr *start, struct alt_instr *end) { @@ -256,10 +345,10 @@ void __init_or_module apply_alternatives(struct alt_instr *start, u8 *instr, *replacement; u8 insnbuf[MAX_PATCH_LEN]; - DPRINTK("%s: alt table %p -> %p\n", __func__, start, end); + DPRINTK("alt table %p -> %p", start, end); /* * The scan order should be from start to end. A later scanned - * alternative code can overwrite a previous scanned alternative code. + * alternative code can overwrite previously scanned alternative code. * Some kernel functions (e.g. memcpy, memset, etc) use this order to * patch code. * @@ -267,29 +356,54 @@ void __init_or_module apply_alternatives(struct alt_instr *start, * order. */ for (a = start; a < end; a++) { + int insnbuf_sz = 0; + instr = (u8 *)&a->instr_offset + a->instr_offset; replacement = (u8 *)&a->repl_offset + a->repl_offset; - BUG_ON(a->replacementlen > a->instrlen); BUG_ON(a->instrlen > sizeof(insnbuf)); BUG_ON(a->cpuid >= (NCAPINTS + NBUGINTS) * 32); - if (!boot_cpu_has(a->cpuid)) + if (!boot_cpu_has(a->cpuid)) { + if (a->padlen > 1) + optimize_nops(a, instr); + continue; + } + + DPRINTK("feat: %d*32+%d, old: (%p, len: %d), repl: (%p, len: %d)", + a->cpuid >> 5, + a->cpuid & 0x1f, + instr, a->instrlen, + replacement, a->replacementlen); + + DUMP_BYTES(instr, a->instrlen, "%p: old_insn: ", instr); + DUMP_BYTES(replacement, a->replacementlen, "%p: rpl_insn: ", replacement); memcpy(insnbuf, replacement, a->replacementlen); + insnbuf_sz = a->replacementlen; /* 0xe8 is a relative jump; fix the offset. */ - if (*insnbuf == 0xe8 && a->replacementlen == 5) - *(s32 *)(insnbuf + 1) += replacement - instr; + if (*insnbuf == 0xe8 && a->replacementlen == 5) { + *(s32 *)(insnbuf + 1) += replacement - instr; + DPRINTK("Fix CALL offset: 0x%x, CALL 0x%lx", + *(s32 *)(insnbuf + 1), + (unsigned long)instr + *(s32 *)(insnbuf + 1) + 5); + } + + if (a->replacementlen && is_jmp(replacement[0])) + recompute_jump(a, instr, replacement, insnbuf); - add_nops(insnbuf + a->replacementlen, - a->instrlen - a->replacementlen); + if (a->instrlen > a->replacementlen) { + add_nops(insnbuf + a->replacementlen, + a->instrlen - a->replacementlen); + insnbuf_sz += a->instrlen - a->replacementlen; + } + DUMP_BYTES(insnbuf, insnbuf_sz, "%p: final_insn: ", instr); - text_poke_early(instr, insnbuf, a->instrlen); + text_poke_early(instr, insnbuf, insnbuf_sz); } } #ifdef CONFIG_SMP - static void alternatives_smp_lock(const s32 *start, const s32 *end, u8 *text, u8 *text_end) { @@ -371,8 +485,8 @@ void __init_or_module alternatives_smp_module_add(struct module *mod, smp->locks_end = locks_end; smp->text = text; smp->text_end = text_end; - DPRINTK("%s: locks %p -> %p, text %p -> %p, name %s\n", - __func__, smp->locks, smp->locks_end, + DPRINTK("locks %p -> %p, text %p -> %p, name %s\n", + smp->locks, smp->locks_end, smp->text, smp->text_end, smp->name); list_add_tail(&smp->next, &smp_alt_modules); @@ -440,7 +554,7 @@ int alternatives_text_reserved(void *start, void *end) return 0; } -#endif +#endif /* CONFIG_SMP */ #ifdef CONFIG_PARAVIRT void __init_or_module apply_paravirt(struct paravirt_patch_site *start, diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 80e35e8522e4..6238499e51a2 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -762,11 +762,6 @@ static unsigned int amd_size_cache(struct cpuinfo_x86 *c, unsigned int size) } #endif -static void cpu_set_tlb_flushall_shift(struct cpuinfo_x86 *c) -{ - tlb_flushall_shift = 6; -} - static void cpu_detect_tlb_amd(struct cpuinfo_x86 *c) { u32 ebx, eax, ecx, edx; @@ -814,8 +809,6 @@ static void cpu_detect_tlb_amd(struct cpuinfo_x86 *c) tlb_lli_2m[ENTRIES] = eax & mask; tlb_lli_4m[ENTRIES] = tlb_lli_2m[ENTRIES] >> 1; - - cpu_set_tlb_flushall_shift(c); } static const struct cpu_dev amd_cpu_dev = { diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 03445346ee0a..4c7dd836304a 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -65,6 +65,14 @@ static void __init check_fpu(void) void __init check_bugs(void) { +#ifdef CONFIG_X86_32 + /* + * Regardless of whether PCID is enumerated, the SDM says + * that it can't be enabled in 32-bit mode. + */ + setup_clear_cpu_cap(X86_FEATURE_PCID); +#endif + identify_boot_cpu(); #ifndef CONFIG_SMP pr_info("CPU: "); diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index b0f4cfef81c4..274f5a7b27e7 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -90,7 +90,7 @@ static const struct cpu_dev default_cpu = { static const struct cpu_dev *this_cpu = &default_cpu; -DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = { +DEFINE_PER_CPU_PAGE_ALIGNED_USER_MAPPED(struct gdt_page, gdt_page) = { .gdt = { #ifdef CONFIG_X86_64 /* * We need valid kernel segments for data and code in long mode too @@ -163,6 +163,40 @@ static int __init x86_xsaveopt_setup(char *s) } __setup("noxsaveopt", x86_xsaveopt_setup); +#ifdef CONFIG_X86_64 +static int __init x86_pcid_setup(char *s) +{ + /* require an exact match without trailing characters */ + if (strlen(s)) + return 0; + + /* do not emit a message if the feature is not present */ + if (!boot_cpu_has(X86_FEATURE_PCID)) + return 1; + + setup_clear_cpu_cap(X86_FEATURE_PCID); + pr_info("nopcid: PCID feature disabled\n"); + return 1; +} +__setup("nopcid", x86_pcid_setup); +#endif + +static int __init x86_noinvpcid_setup(char *s) +{ + /* noinvpcid doesn't accept parameters */ + if (s) + return -EINVAL; + + /* do not emit a message if the feature is not present */ + if (!boot_cpu_has(X86_FEATURE_INVPCID)) + return 0; + + setup_clear_cpu_cap(X86_FEATURE_INVPCID); + pr_info("noinvpcid: INVPCID feature disabled\n"); + return 0; +} +early_param("noinvpcid", x86_noinvpcid_setup); + #ifdef CONFIG_X86_32 static int cachesize_override = -1; static int disable_x86_serial_nr = 1; @@ -270,7 +304,7 @@ __setup("nosmep", setup_disable_smep); static __always_inline void setup_smep(struct cpuinfo_x86 *c) { if (cpu_has(c, X86_FEATURE_SMEP)) - set_in_cr4(X86_CR4_SMEP); + cr4_set_bits(X86_CR4_SMEP); } static __init int setup_disable_smap(char *arg) @@ -289,13 +323,46 @@ static __always_inline void setup_smap(struct cpuinfo_x86 *c) if (cpu_has(c, X86_FEATURE_SMAP)) { #ifdef CONFIG_X86_SMAP - set_in_cr4(X86_CR4_SMAP); + cr4_set_bits(X86_CR4_SMAP); #else - clear_in_cr4(X86_CR4_SMAP); + cr4_clear_bits(X86_CR4_SMAP); #endif } } +static void setup_pcid(struct cpuinfo_x86 *c) +{ + if (cpu_has(c, X86_FEATURE_PCID)) { + if (cpu_has(c, X86_FEATURE_PGE) || kaiser_enabled) { + cr4_set_bits(X86_CR4_PCIDE); + /* + * INVPCID has two "groups" of types: + * 1/2: Invalidate an individual address + * 3/4: Invalidate all contexts + * + * 1/2 take a PCID, but 3/4 do not. So, 3/4 + * ignore the PCID argument in the descriptor. + * But, we have to be careful not to call 1/2 + * with an actual non-zero PCID in them before + * we do the above cr4_set_bits(). + */ + if (cpu_has(c, X86_FEATURE_INVPCID)) + set_cpu_cap(c, X86_FEATURE_INVPCID_SINGLE); + } else { + /* + * flush_tlb_all(), as currently implemented, won't + * work if PCID is on but PGE is not. Since that + * combination doesn't exist on real hardware, there's + * no reason to try to fully support it, but it's + * polite to avoid corrupting data if we're on + * an improperly configured VM. + */ + clear_cpu_cap(c, X86_FEATURE_PCID); + } + } + kaiser_setup_pcid(); +} + /* * Some CPU features depend on higher CPUID levels, which may not always * be available due to CPUID level capping or broken virtualization @@ -482,26 +549,17 @@ u16 __read_mostly tlb_lld_2m[NR_INFO]; u16 __read_mostly tlb_lld_4m[NR_INFO]; u16 __read_mostly tlb_lld_1g[NR_INFO]; -/* - * tlb_flushall_shift shows the balance point in replacing cr3 write - * with multiple 'invlpg'. It will do this replacement when - * flush_tlb_lines <= active_lines/2^tlb_flushall_shift. - * If tlb_flushall_shift is -1, means the replacement will be disabled. - */ -s8 __read_mostly tlb_flushall_shift = -1; - void cpu_detect_tlb(struct cpuinfo_x86 *c) { if (this_cpu->c_detect_tlb) this_cpu->c_detect_tlb(c); printk(KERN_INFO "Last level iTLB entries: 4KB %d, 2MB %d, 4MB %d\n" - "Last level dTLB entries: 4KB %d, 2MB %d, 4MB %d, 1GB %d\n" - "tlb_flushall_shift: %d\n", + "Last level dTLB entries: 4KB %d, 2MB %d, 4MB %d, 1GB %d\n", tlb_lli_4k[ENTRIES], tlb_lli_2m[ENTRIES], tlb_lli_4m[ENTRIES], tlb_lld_4k[ENTRIES], tlb_lld_2m[ENTRIES], tlb_lld_4m[ENTRIES], - tlb_lld_1g[ENTRIES], tlb_flushall_shift); + tlb_lld_1g[ENTRIES]); } void detect_ht(struct cpuinfo_x86 *c) @@ -886,6 +944,9 @@ static void identify_cpu(struct cpuinfo_x86 *c) setup_smep(c); setup_smap(c); + /* Set up PCID */ + setup_pcid(c); + /* * The vendor-specific functions might have changed features. * Now we do "generic changes." @@ -1008,6 +1069,11 @@ void identify_secondary_cpu(struct cpuinfo_x86 *c) BUG_ON(c == &boot_cpu_data); identify_cpu(c); #ifdef CONFIG_X86_32 + /* + * Regardless of whether PCID is enumerated, the SDM says + * that it can't be enabled in 32-bit mode. + */ + clear_cpu_cap(c, X86_FEATURE_PCID); enable_sep_cpu(); #endif mtrr_ap_init(); @@ -1114,7 +1180,7 @@ static __init int setup_disablecpuid(char *arg) } __setup("clearcpuid=", setup_disablecpuid); -DEFINE_PER_CPU(unsigned long, kernel_stack) = +DEFINE_PER_CPU_USER_MAPPED(unsigned long, kernel_stack) = (unsigned long)&init_thread_union - KERNEL_STACK_OFFSET + THREAD_SIZE; EXPORT_PER_CPU_SYMBOL(kernel_stack); @@ -1155,7 +1221,7 @@ static const unsigned int exception_stack_sizes[N_EXCEPTION_STACKS] = { [DEBUG_STACK - 1] = DEBUG_STKSZ }; -static DEFINE_PER_CPU_PAGE_ALIGNED(char, exception_stacks +DEFINE_PER_CPU_PAGE_ALIGNED_USER_MAPPED(char, exception_stacks [(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ + DEBUG_STKSZ]); /* May not be marked __init: used by software suspend */ @@ -1277,6 +1343,15 @@ void cpu_init(void) int cpu; int i; + if (!kaiser_enabled) { + /* + * secondary_startup_64() deferred setting PGE in cr4: + * probe_page_size_mask() sets it on the boot cpu, + * but it needs to be set on each secondary cpu. + */ + cr4_set_bits(X86_CR4_PGE); + } + /* * Load microcode on this cpu if a valid microcode is available. * This is early microcode loading procedure. @@ -1300,7 +1375,7 @@ void cpu_init(void) pr_debug("Initializing CPU#%d\n", cpu); - clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE); + cr4_clear_bits(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE); /* * Initialize the per-CPU GDT with the boot GDT, @@ -1385,7 +1460,7 @@ void cpu_init(void) printk(KERN_INFO "Initializing CPU#%d\n", cpu); if (cpu_has_vme || cpu_has_tsc || cpu_has_de) - clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE); + cr4_clear_bits(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE); load_current_idt(); switch_to_new_gdt(cpu); diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index df64630d8053..cfbdac5c601e 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -663,31 +663,6 @@ static void intel_tlb_lookup(const unsigned char desc) } } -static void intel_tlb_flushall_shift_set(struct cpuinfo_x86 *c) -{ - switch ((c->x86 << 8) + c->x86_model) { - case 0x60f: /* original 65 nm celeron/pentium/core2/xeon, "Merom"/"Conroe" */ - case 0x616: /* single-core 65 nm celeron/core2solo "Merom-L"/"Conroe-L" */ - case 0x617: /* current 45 nm celeron/core2/xeon "Penryn"/"Wolfdale" */ - case 0x61d: /* six-core 45 nm xeon "Dunnington" */ - tlb_flushall_shift = -1; - break; - case 0x63a: /* Ivybridge */ - tlb_flushall_shift = 2; - break; - case 0x61a: /* 45 nm nehalem, "Bloomfield" */ - case 0x61e: /* 45 nm nehalem, "Lynnfield" */ - case 0x625: /* 32 nm nehalem, "Clarkdale" */ - case 0x62c: /* 32 nm nehalem, "Gulftown" */ - case 0x62e: /* 45 nm nehalem-ex, "Beckton" */ - case 0x62f: /* 32 nm Xeon E7 */ - case 0x62a: /* SandyBridge */ - case 0x62d: /* SandyBridge, "Romely-EP" */ - default: - tlb_flushall_shift = 6; - } -} - static void intel_detect_tlb(struct cpuinfo_x86 *c) { int i, j, n; @@ -712,7 +687,6 @@ static void intel_detect_tlb(struct cpuinfo_x86 *c) for (j = 1 ; j < 16 ; j++) intel_tlb_lookup(desc[j]); } - intel_tlb_flushall_shift_set(c); } static const struct cpu_dev intel_cpu_dev = { diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 8ae7e3f27b1b..72cb777b4625 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -43,6 +43,7 @@ #include #include +#include #include #include @@ -1470,7 +1471,7 @@ static void __mcheck_cpu_init_generic(void) bitmap_fill(all_banks, MAX_NR_BANKS); machine_check_poll(MCP_UC | m_fl, &all_banks); - set_in_cr4(X86_CR4_MCE); + cr4_set_bits(X86_CR4_MCE); rdmsrl(MSR_IA32_MCG_CAP, cap); if (cap & MCG_CTL_P) diff --git a/arch/x86/kernel/cpu/mcheck/p5.c b/arch/x86/kernel/cpu/mcheck/p5.c index a3042989398c..30692ac88d1e 100644 --- a/arch/x86/kernel/cpu/mcheck/p5.c +++ b/arch/x86/kernel/cpu/mcheck/p5.c @@ -8,6 +8,7 @@ #include #include +#include #include #include @@ -59,7 +60,7 @@ void intel_p5_mcheck_init(struct cpuinfo_x86 *c) "Intel old style machine check architecture supported.\n"); /* Enable MCE: */ - set_in_cr4(X86_CR4_MCE); + cr4_set_bits(X86_CR4_MCE); printk(KERN_INFO "Intel old style machine check reporting enabled on CPU#%d.\n", smp_processor_id()); diff --git a/arch/x86/kernel/cpu/mcheck/winchip.c b/arch/x86/kernel/cpu/mcheck/winchip.c index 7dc5564d0cdf..590cc753ba8f 100644 --- a/arch/x86/kernel/cpu/mcheck/winchip.c +++ b/arch/x86/kernel/cpu/mcheck/winchip.c @@ -7,6 +7,7 @@ #include #include +#include #include #include @@ -31,7 +32,7 @@ void winchip_mcheck_init(struct cpuinfo_x86 *c) lo &= ~(1<<4); /* Enable MCE */ wrmsr(MSR_IDT_FCR1, lo, hi); - set_in_cr4(X86_CR4_MCE); + cr4_set_bits(X86_CR4_MCE); printk(KERN_INFO "Winchip machine check reporting enabled on CPU#0.\n"); diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index ec91ccee0661..33f549ccdcf1 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -31,6 +31,7 @@ #include #include #include +#include #include #include #include @@ -1340,7 +1341,7 @@ x86_pmu_notifier(struct notifier_block *self, unsigned long action, void *hcpu) case CPU_STARTING: if (x86_pmu.attr_rdpmc) - set_in_cr4(X86_CR4_PCE); + cr4_set_bits(X86_CR4_PCE); if (x86_pmu.cpu_starting) x86_pmu.cpu_starting(cpu); break; @@ -1855,9 +1856,9 @@ static void change_rdpmc(void *info) bool enable = !!(unsigned long)info; if (enable) - set_in_cr4(X86_CR4_PCE); + cr4_set_bits(X86_CR4_PCE); else - clear_in_cr4(X86_CR4_PCE); + cr4_clear_bits(X86_CR4_PCE); } static ssize_t set_attr_rdpmc(struct device *cdev, diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c index bae2bfa2af0e..711135df29d2 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_ds.c +++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c @@ -2,11 +2,15 @@ #include #include +#include #include #include #include "perf_event.h" +static +DEFINE_PER_CPU_SHARED_ALIGNED_USER_MAPPED(struct debug_store, cpu_debug_store); + /* The size of a BTS record in bytes: */ #define BTS_RECORD_SIZE 24 @@ -256,6 +260,39 @@ void fini_debug_store_on_cpu(int cpu) static DEFINE_PER_CPU(void *, insn_buffer); +static void *dsalloc(size_t size, gfp_t flags, int node) +{ +#ifdef CONFIG_PAGE_TABLE_ISOLATION + unsigned int order = get_order(size); + struct page *page; + unsigned long addr; + + page = alloc_pages_node(node, flags | __GFP_ZERO, order); + if (!page) + return NULL; + addr = (unsigned long)page_address(page); + if (kaiser_add_mapping(addr, size, __PAGE_KERNEL) < 0) { + __free_pages(page, order); + addr = 0; + } + return (void *)addr; +#else + return kmalloc_node(size, flags | __GFP_ZERO, node); +#endif +} + +static void dsfree(const void *buffer, size_t size) +{ +#ifdef CONFIG_PAGE_TABLE_ISOLATION + if (!buffer) + return; + kaiser_remove_mapping((unsigned long)buffer, size); + free_pages((unsigned long)buffer, get_order(size)); +#else + kfree(buffer); +#endif +} + static int alloc_pebs_buffer(int cpu) { struct debug_store *ds = per_cpu(cpu_hw_events, cpu).ds; @@ -266,7 +303,7 @@ static int alloc_pebs_buffer(int cpu) if (!x86_pmu.pebs) return 0; - buffer = kzalloc_node(x86_pmu.pebs_buffer_size, GFP_KERNEL, node); + buffer = dsalloc(x86_pmu.pebs_buffer_size, GFP_KERNEL, node); if (unlikely(!buffer)) return -ENOMEM; @@ -277,7 +314,7 @@ static int alloc_pebs_buffer(int cpu) if (x86_pmu.intel_cap.pebs_format < 2) { ibuffer = kzalloc_node(PEBS_FIXUP_SIZE, GFP_KERNEL, node); if (!ibuffer) { - kfree(buffer); + dsfree(buffer, x86_pmu.pebs_buffer_size); return -ENOMEM; } per_cpu(insn_buffer, cpu) = ibuffer; @@ -306,7 +343,7 @@ static void release_pebs_buffer(int cpu) kfree(per_cpu(insn_buffer, cpu)); per_cpu(insn_buffer, cpu) = NULL; - kfree((void *)(unsigned long)ds->pebs_buffer_base); + dsfree((void *)(unsigned long)ds->pebs_buffer_base, x86_pmu.pebs_buffer_size); ds->pebs_buffer_base = 0; } @@ -320,7 +357,7 @@ static int alloc_bts_buffer(int cpu) if (!x86_pmu.bts) return 0; - buffer = kzalloc_node(BTS_BUFFER_SIZE, GFP_KERNEL | __GFP_NOWARN, node); + buffer = dsalloc(BTS_BUFFER_SIZE, GFP_KERNEL | __GFP_NOWARN, node); if (unlikely(!buffer)) { WARN_ONCE(1, "%s: BTS buffer allocation failure\n", __func__); return -ENOMEM; @@ -346,19 +383,15 @@ static void release_bts_buffer(int cpu) if (!ds || !x86_pmu.bts) return; - kfree((void *)(unsigned long)ds->bts_buffer_base); + dsfree((void *)(unsigned long)ds->bts_buffer_base, BTS_BUFFER_SIZE); ds->bts_buffer_base = 0; } static int alloc_ds_buffer(int cpu) { - int node = cpu_to_node(cpu); - struct debug_store *ds; - - ds = kzalloc_node(sizeof(*ds), GFP_KERNEL, node); - if (unlikely(!ds)) - return -ENOMEM; + struct debug_store *ds = per_cpu_ptr(&cpu_debug_store, cpu); + memset(ds, 0, sizeof(*ds)); per_cpu(cpu_hw_events, cpu).ds = ds; return 0; @@ -372,7 +405,6 @@ static void release_ds_buffer(int cpu) return; per_cpu(cpu_hw_events, cpu).ds = NULL; - kfree(ds); } void release_ds_buffers(void) diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S index 2c123171944a..bfcc300d8b3c 100644 --- a/arch/x86/kernel/entry_32.S +++ b/arch/x86/kernel/entry_32.S @@ -821,7 +821,7 @@ ENTRY(simd_coprocessor_error) 661: pushl_cfi $do_general_protection 662: .section .altinstructions,"a" - altinstruction_entry 661b, 663f, X86_FEATURE_XMM, 662b-661b, 664f-663f + altinstruction_entry 661b, 663f, X86_FEATURE_XMM, 662b-661b, 664f-663f, 0 .previous .section .altinstr_replacement,"ax" 663: pushl $do_simd_coprocessor_error diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index 42a970d19ab3..5dc1043544ad 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -58,6 +58,7 @@ #include #include #include +#include #include /* Avoid __ASSEMBLER__'ifying just for this. */ @@ -263,6 +264,7 @@ ENDPROC(native_usergs_sysret64) testl $3, CS-RBP(%rsi) je 1f SWAPGS + SWITCH_KERNEL_CR3 /* * irq_count is used to check if a CPU is already on an interrupt stack * or not. While this is essentially redundant with preempt_count it is @@ -284,6 +286,12 @@ ENDPROC(native_usergs_sysret64) TRACE_IRQS_OFF .endm +/* + * Return: ebx=0: needs swapgs but not SWITCH_USER_CR3 in paranoid_exit + * ebx=1: needs neither swapgs nor SWITCH_USER_CR3 in paranoid_exit + * ebx=2: needs both swapgs and SWITCH_USER_CR3 in paranoid_exit + * ebx=3: needs SWITCH_USER_CR3 but not swapgs in paranoid_exit + */ ENTRY(save_paranoid) XCPT_FRAME 1 RDI+8 cld @@ -309,7 +317,26 @@ ENTRY(save_paranoid) js 1f /* negative -> in kernel */ SWAPGS xorl %ebx,%ebx -1: ret +1: +#ifdef CONFIG_PAGE_TABLE_ISOLATION + /* + * We might have come in between a swapgs and a SWITCH_KERNEL_CR3 + * on entry, or between a SWITCH_USER_CR3 and a swapgs on exit. + * Do a conditional SWITCH_KERNEL_CR3: this could safely be done + * unconditionally, but we need to find out whether the reverse + * should be done on return (conveyed to paranoid_exit in %ebx). + */ + ALTERNATIVE "jmp 2f", "movq %cr3, %rax", X86_FEATURE_KAISER + testl $KAISER_SHADOW_PGD_OFFSET, %eax + jz 2f + orl $2, %ebx + andq $(~(X86_CR3_PCID_ASID_MASK | KAISER_SHADOW_PGD_OFFSET)), %rax + /* If PCID enabled, set X86_CR3_PCID_NOFLUSH_BIT */ + ALTERNATIVE "", "bts $63, %rax", X86_FEATURE_PCID + movq %rax, %cr3 +2: +#endif + ret CFI_ENDPROC END(save_paranoid) @@ -394,6 +421,7 @@ ENTRY(system_call) CFI_REGISTER rip,rcx /*CFI_REGISTER rflags,r11*/ SWAPGS_UNSAFE_STACK + SWITCH_KERNEL_CR3_NO_STACK /* * A hypervisor implementation might want to use a label * after the swapgs, so that it can do the swapgs @@ -448,6 +476,14 @@ GLOBAL(system_call_after_swapgs) CFI_REGISTER rip,rcx RESTORE_ARGS 1,-ARG_SKIP,0 /*CFI_REGISTER rflags,r11*/ + /* + * This opens a window where we have a user CR3, but are + * running in the kernel. This makes using the CS + * register useless for telling whether or not we need to + * switch CR3 in NMIs. Normal interrupts are OK because + * they are off here. + */ + SWITCH_USER_CR3 movq PER_CPU_VAR(old_rsp), %rsp USERGS_SYSRET64 @@ -820,6 +856,14 @@ retint_swapgs: /* return to user-space */ */ DISABLE_INTERRUPTS(CLBR_ANY) TRACE_IRQS_IRETQ + /* + * This opens a window where we have a user CR3, but are + * running in the kernel. This makes using the CS + * register useless for telling whether or not we need to + * switch CR3 in NMIs. Normal interrupts are OK because + * they are off here. + */ + SWITCH_USER_CR3 SWAPGS jmp restore_args @@ -860,6 +904,7 @@ ENTRY(native_iret) pushq_cfi %rax pushq_cfi %rdi SWAPGS + SWITCH_KERNEL_CR3 movq PER_CPU_VAR(espfix_waddr),%rdi movq %rax,(0*8)(%rdi) /* RAX */ movq (2*8)(%rsp),%rax /* RIP */ @@ -875,6 +920,7 @@ ENTRY(native_iret) andl $0xffff0000,%eax popq_cfi %rdi orq PER_CPU_VAR(espfix_stack),%rax + SWITCH_USER_CR3 SWAPGS movq %rax,%rsp popq_cfi %rax @@ -1289,30 +1335,41 @@ idtentry machine_check has_error_code=0 paranoid=1 do_sym=*machine_check_vector( * is fundamentally NMI-unsafe. (we cannot change the soft and * hard flags at once, atomically) */ - - /* ebx: no swapgs flag */ +/* + * On entry: ebx=0: needs swapgs but not SWITCH_USER_CR3 + * ebx=1: needs neither swapgs nor SWITCH_USER_CR3 + * ebx=2: needs both swapgs and SWITCH_USER_CR3 + * ebx=3: needs SWITCH_USER_CR3 but not swapgs + */ ENTRY(paranoid_exit) DEFAULT_FRAME DISABLE_INTERRUPTS(CLBR_NONE) TRACE_IRQS_OFF_DEBUG - testl %ebx,%ebx /* swapgs needed? */ - jnz paranoid_restore - testl $3,CS(%rsp) - jnz paranoid_userspace -paranoid_swapgs: - TRACE_IRQS_IRETQ 0 - SWAPGS_UNSAFE_STACK - RESTORE_ALL 8 - jmp irq_return -paranoid_restore: + movq %rbx, %r12 /* paranoid_userspace uses %ebx */ + testl $3, CS(%rsp) + jnz paranoid_userspace +paranoid_kernel: + movq %r12, %rbx /* restore after paranoid_userspace */ TRACE_IRQS_IRETQ_DEBUG 0 +#ifdef CONFIG_PAGE_TABLE_ISOLATION + /* No ALTERNATIVE for X86_FEATURE_KAISER: paranoid_entry sets %ebx */ + testl $2, %ebx /* SWITCH_USER_CR3 needed? */ + jz paranoid_exit_no_switch + SWITCH_USER_CR3 +paranoid_exit_no_switch: +#endif + testl $1, %ebx /* swapgs needed? */ + jnz paranoid_exit_no_swapgs + SWAPGS_UNSAFE_STACK +paranoid_exit_no_swapgs: RESTORE_ALL 8 - jmp irq_return + jmp irq_return + paranoid_userspace: GET_THREAD_INFO(%rcx) movl TI_flags(%rcx),%ebx andl $_TIF_WORK_MASK,%ebx - jz paranoid_swapgs + jz paranoid_kernel movq %rsp,%rdi /* &pt_regs */ call sync_regs movq %rax,%rsp /* switch stack for scheduling */ @@ -1361,6 +1418,13 @@ ENTRY(error_entry) movq_cfi r13, R13+8 movq_cfi r14, R14+8 movq_cfi r15, R15+8 + /* + * error_entry() always returns with a kernel gsbase and + * CR3. We must also have a kernel CR3/gsbase before + * calling TRACE_IRQS_*. Just unconditionally switch to + * the kernel CR3 here. + */ + SWITCH_KERNEL_CR3 xorl %ebx,%ebx testl $3,CS+8(%rsp) je error_kernelspace @@ -1497,6 +1561,10 @@ ENTRY(nmi) */ SWAPGS_UNSAFE_STACK + /* + * percpu variables are mapped with user CR3, so no need + * to switch CR3 here. + */ cld movq %rsp, %rdx movq PER_CPU_VAR(kernel_stack), %rsp @@ -1531,12 +1599,34 @@ ENTRY(nmi) movq %rsp, %rdi movq $-1, %rsi +#ifdef CONFIG_PAGE_TABLE_ISOLATION + /* Unconditionally use kernel CR3 for do_nmi() */ + /* %rax is saved above, so OK to clobber here */ + ALTERNATIVE "jmp 2f", "movq %cr3, %rax", X86_FEATURE_KAISER + /* If PCID enabled, NOFLUSH now and NOFLUSH on return */ + ALTERNATIVE "", "bts $63, %rax", X86_FEATURE_PCID + pushq %rax + /* mask off "user" bit of pgd address and 12 PCID bits: */ + andq $(~(X86_CR3_PCID_ASID_MASK | KAISER_SHADOW_PGD_OFFSET)), %rax + movq %rax, %cr3 +2: +#endif call do_nmi +#ifdef CONFIG_PAGE_TABLE_ISOLATION + /* + * Unconditionally restore CR3. I know we return to + * kernel code that needs user CR3, but do we ever return + * to "user mode" where we need the kernel CR3? + */ + ALTERNATIVE "", "popq %rax; movq %rax, %cr3", X86_FEATURE_KAISER +#endif + /* * Return back to user mode. We must *not* do the normal exit - * work, because we don't want to enable interrupts. Fortunately, - * do_nmi doesn't modify pt_regs. + * work, because we don't want to enable interrupts. Do not + * switch to user CR3: we might be going back to kernel code + * that had a user CR3 set. */ SWAPGS @@ -1746,23 +1836,69 @@ ENTRY(nmi) subq $ORIG_RAX-R15, %rsp CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15 /* - * Use save_paranoid to handle SWAPGS, but no need to use paranoid_exit - * as we should not be calling schedule in NMI context. - * Even with normal interrupts enabled. An NMI should not be - * setting NEED_RESCHED or anything that normal interrupts and - * exceptions might do. + * Use the same approach as save_paranoid to handle SWAPGS, but + * without CR3 handling since we do that differently in NMIs. No + * need to use paranoid_exit as we should not be calling schedule + * in NMI context. Even with normal interrupts enabled. An NMI + * should not be setting NEED_RESCHED or anything that normal + * interrupts and exceptions might do. */ - call save_paranoid - DEFAULT_FRAME 0 + cld + movq %rdi, RDI(%rsp) + movq %rsi, RSI(%rsp) + movq_cfi rdx, RDX + movq_cfi rcx, RCX + movq_cfi rax, RAX + movq %r8, R8(%rsp) + movq %r9, R9(%rsp) + movq %r10, R10(%rsp) + movq %r11, R11(%rsp) + movq_cfi rbx, RBX + movq %rbp, RBP(%rsp) + movq %r12, R12(%rsp) + movq %r13, R13(%rsp) + movq %r14, R14(%rsp) + movq %r15, R15(%rsp) + movl $1, %ebx + movl $MSR_GS_BASE, %ecx + rdmsr + testl %edx, %edx + js 1f /* negative -> in kernel */ + SWAPGS + xorl %ebx,%ebx +1: + movq %rsp,%rdi + movq $-1,%rsi +#ifdef CONFIG_PAGE_TABLE_ISOLATION + /* Unconditionally use kernel CR3 for do_nmi() */ + /* %rax is saved above, so OK to clobber here */ + ALTERNATIVE "jmp 2f", "movq %cr3, %rax", X86_FEATURE_KAISER + /* If PCID enabled, NOFLUSH now and NOFLUSH on return */ + ALTERNATIVE "", "bts $63, %rax", X86_FEATURE_PCID + pushq %rax + /* mask off "user" bit of pgd address and 12 PCID bits: */ + andq $(~(X86_CR3_PCID_ASID_MASK | KAISER_SHADOW_PGD_OFFSET)), %rax + movq %rax, %cr3 +2: +#endif + DEFAULT_FRAME 0 /* ???? */ /* paranoidentry do_nmi, 0; without TRACE_IRQS_OFF */ - movq %rsp,%rdi - movq $-1,%rsi - call do_nmi + call do_nmi + +#ifdef CONFIG_PAGE_TABLE_ISOLATION + /* + * Unconditionally restore CR3. We might be returning to + * kernel code that needs user CR3, like just just before + * a sysret. + */ + ALTERNATIVE "", "popq %rax; movq %rax, %cr3", X86_FEATURE_KAISER +#endif testl %ebx,%ebx /* swapgs needed? */ jnz nmi_restore nmi_swapgs: + /* We fixed up CR3 above, so no need to switch it here */ SWAPGS_UNSAFE_STACK nmi_restore: diff --git a/arch/x86/kernel/espfix_64.c b/arch/x86/kernel/espfix_64.c index 94d857fb1033..a1944faa739c 100644 --- a/arch/x86/kernel/espfix_64.c +++ b/arch/x86/kernel/espfix_64.c @@ -41,6 +41,7 @@ #include #include #include +#include /* * Note: we only need 6*8 = 48 bytes for the espfix stack, but round @@ -129,6 +130,15 @@ void __init init_espfix_bsp(void) /* Install the espfix pud into the kernel page directory */ pgd_p = &init_level4_pgt[pgd_index(ESPFIX_BASE_ADDR)]; pgd_populate(&init_mm, pgd_p, (pud_t *)espfix_pud_page); + /* + * Just copy the top-level PGD that is mapping the espfix + * area to ensure it is mapped into the shadow user page + * tables. + */ + if (kaiser_enabled) { + set_pgd(native_get_shadow_pgd(pgd_p), + __pgd(_KERNPG_TABLE | __pa((pud_t *)espfix_pud_page))); + } /* Randomize the locations */ init_espfix_random(); diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S index 761fd69df6d9..8c3cdb9ba14b 100644 --- a/arch/x86/kernel/head_64.S +++ b/arch/x86/kernel/head_64.S @@ -183,8 +183,8 @@ ENTRY(secondary_startup_64) movq $(init_level4_pgt - __START_KERNEL_map), %rax 1: - /* Enable PAE mode and PGE */ - movl $(X86_CR4_PAE | X86_CR4_PGE), %ecx + /* Enable PAE and PSE, but defer PGE until kaiser_enabled is decided */ + movl $(X86_CR4_PAE | X86_CR4_PSE), %ecx movq %rcx, %cr4 /* Setup early boot stage 4 level pagetables. */ @@ -441,6 +441,27 @@ ENDPROC(early_idt_handler_common) .balign PAGE_SIZE; \ GLOBAL(name) +#ifdef CONFIG_PAGE_TABLE_ISOLATION +/* + * Each PGD needs to be 8k long and 8k aligned. We do not + * ever go out to userspace with these, so we do not + * strictly *need* the second page, but this allows us to + * have a single set_pgd() implementation that does not + * need to worry about whether it has 4k or 8k to work + * with. + * + * This ensures PGDs are 8k long: + */ +#define KAISER_USER_PGD_FILL 512 +/* This ensures they are 8k-aligned: */ +#define NEXT_PGD_PAGE(name) \ + .balign 2 * PAGE_SIZE; \ +GLOBAL(name) +#else +#define NEXT_PGD_PAGE(name) NEXT_PAGE(name) +#define KAISER_USER_PGD_FILL 0 +#endif + /* Automate the creation of 1 to 1 mapping pmd entries */ #define PMDS(START, PERM, COUNT) \ i = 0 ; \ @@ -450,9 +471,10 @@ GLOBAL(name) .endr __INITDATA -NEXT_PAGE(early_level4_pgt) +NEXT_PGD_PAGE(early_level4_pgt) .fill 511,8,0 .quad level3_kernel_pgt - __START_KERNEL_map + _PAGE_TABLE + .fill KAISER_USER_PGD_FILL,8,0 NEXT_PAGE(early_dynamic_pgts) .fill 512*EARLY_DYNAMIC_PAGE_TABLES,8,0 @@ -460,16 +482,18 @@ NEXT_PAGE(early_dynamic_pgts) .data #ifndef CONFIG_XEN -NEXT_PAGE(init_level4_pgt) +NEXT_PGD_PAGE(init_level4_pgt) .fill 512,8,0 + .fill KAISER_USER_PGD_FILL,8,0 #else -NEXT_PAGE(init_level4_pgt) +NEXT_PGD_PAGE(init_level4_pgt) .quad level3_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE .org init_level4_pgt + L4_PAGE_OFFSET*8, 0 .quad level3_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE .org init_level4_pgt + L4_START_KERNEL*8, 0 /* (2^48-(2*1024*1024*1024))/(2^39) = 511 */ .quad level3_kernel_pgt - __START_KERNEL_map + _PAGE_TABLE + .fill KAISER_USER_PGD_FILL,8,0 NEXT_PAGE(level3_ident_pgt) .quad level2_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE @@ -480,6 +504,7 @@ NEXT_PAGE(level2_ident_pgt) */ PMDS(0, __PAGE_KERNEL_IDENT_LARGE_EXEC, PTRS_PER_PMD) #endif + .fill KAISER_USER_PGD_FILL,8,0 NEXT_PAGE(level3_kernel_pgt) .fill L3_START_KERNEL,8,0 diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index 5d7acb77cc1d..99df035d7cb1 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include @@ -181,7 +182,7 @@ void fpu_init(void) if (cpu_has_xmm) cr4_mask |= X86_CR4_OSXMMEXCPT; if (cr4_mask) - set_in_cr4(cr4_mask); + cr4_set_bits(cr4_mask); cr0 = read_cr0(); cr0 &= ~(X86_CR0_TS|X86_CR0_EM); /* clear TS and EM */ diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c index 37907756fc41..eedac3e88156 100644 --- a/arch/x86/kernel/irq.c +++ b/arch/x86/kernel/irq.c @@ -96,8 +96,7 @@ int arch_show_interrupts(struct seq_file *p, int prec) seq_printf(p, " Rescheduling interrupts\n"); seq_printf(p, "%*s: ", prec, "CAL"); for_each_online_cpu(j) - seq_printf(p, "%10u ", irq_stats(j)->irq_call_count - - irq_stats(j)->irq_tlb_count); + seq_printf(p, "%10u ", irq_stats(j)->irq_call_count); seq_printf(p, " Function call interrupts\n"); seq_printf(p, "%*s: ", prec, "TLB"); for_each_online_cpu(j) diff --git a/arch/x86/kernel/irqinit.c b/arch/x86/kernel/irqinit.c index 7f50156542fb..31045946cfe9 100644 --- a/arch/x86/kernel/irqinit.c +++ b/arch/x86/kernel/irqinit.c @@ -51,7 +51,7 @@ static struct irqaction irq2 = { .flags = IRQF_NO_THREAD, }; -DEFINE_PER_CPU(vector_irq_t, vector_irq) = { +DEFINE_PER_CPU_USER_MAPPED(vector_irq_t, vector_irq) = { [0 ... NR_VECTORS - 1] = VECTOR_UNDEFINED, }; diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c index a2de9bc7ac0b..206aa5a2afe0 100644 --- a/arch/x86/kernel/kvmclock.c +++ b/arch/x86/kernel/kvmclock.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include @@ -281,6 +282,10 @@ int __init kvm_setup_vsyscall_timeinfo(void) if (!hv_clock) return 0; + /* FIXME: Need to add pvclock pages to user-space page tables */ + if (kaiser_enabled) + return 0; + size = PAGE_ALIGN(sizeof(struct pvclock_vsyscall_time_info)*NR_CPUS); preempt_disable(); diff --git a/arch/x86/kernel/ldt.c b/arch/x86/kernel/ldt.c index 2bcc0525f1c1..7bfd9ae0d228 100644 --- a/arch/x86/kernel/ldt.c +++ b/arch/x86/kernel/ldt.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include @@ -33,11 +34,21 @@ static void flush_ldt(void *current_mm) set_ldt(pc->ldt->entries, pc->ldt->size); } +static void __free_ldt_struct(struct ldt_struct *ldt) +{ + if (ldt->size * LDT_ENTRY_SIZE > PAGE_SIZE) + vfree(ldt->entries); + else + free_page((unsigned long)ldt->entries); + kfree(ldt); +} + /* The caller must call finalize_ldt_struct on the result. LDT starts zeroed. */ static struct ldt_struct *alloc_ldt_struct(int size) { struct ldt_struct *new_ldt; int alloc_size; + int ret; if (size > LDT_ENTRIES) return NULL; @@ -58,14 +69,20 @@ static struct ldt_struct *alloc_ldt_struct(int size) if (alloc_size > PAGE_SIZE) new_ldt->entries = vzalloc(alloc_size); else - new_ldt->entries = kzalloc(PAGE_SIZE, GFP_KERNEL); + new_ldt->entries = (void *)get_zeroed_page(GFP_KERNEL); if (!new_ldt->entries) { kfree(new_ldt); return NULL; } + ret = kaiser_add_mapping((unsigned long)new_ldt->entries, alloc_size, + __PAGE_KERNEL); new_ldt->size = size; + if (ret) { + __free_ldt_struct(new_ldt); + return NULL; + } return new_ldt; } @@ -91,12 +108,10 @@ static void free_ldt_struct(struct ldt_struct *ldt) if (likely(!ldt)) return; + kaiser_remove_mapping((unsigned long)ldt->entries, + ldt->size * LDT_ENTRY_SIZE); paravirt_free_ldt(ldt->entries, ldt->size); - if (ldt->size * LDT_ENTRY_SIZE > PAGE_SIZE) - vfree(ldt->entries); - else - kfree(ldt->entries); - kfree(ldt); + __free_ldt_struct(ldt); } /* diff --git a/arch/x86/kernel/paravirt_patch_64.c b/arch/x86/kernel/paravirt_patch_64.c index a1da6737ba5b..a91d9b9b4bde 100644 --- a/arch/x86/kernel/paravirt_patch_64.c +++ b/arch/x86/kernel/paravirt_patch_64.c @@ -9,7 +9,6 @@ DEF_NATIVE(pv_irq_ops, save_fl, "pushfq; popq %rax"); DEF_NATIVE(pv_mmu_ops, read_cr2, "movq %cr2, %rax"); DEF_NATIVE(pv_mmu_ops, read_cr3, "movq %cr3, %rax"); DEF_NATIVE(pv_mmu_ops, write_cr3, "movq %rdi, %cr3"); -DEF_NATIVE(pv_mmu_ops, flush_tlb_single, "invlpg (%rdi)"); DEF_NATIVE(pv_cpu_ops, clts, "clts"); DEF_NATIVE(pv_cpu_ops, wbinvd, "wbinvd"); @@ -57,7 +56,6 @@ unsigned native_patch(u8 type, u16 clobbers, void *ibuf, PATCH_SITE(pv_mmu_ops, read_cr3); PATCH_SITE(pv_mmu_ops, write_cr3); PATCH_SITE(pv_cpu_ops, clts); - PATCH_SITE(pv_mmu_ops, flush_tlb_single); PATCH_SITE(pv_cpu_ops, wbinvd); patch_site: diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index c98cff71760d..ee6014b2c43b 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -29,6 +29,7 @@ #include #include #include +#include /* * per-CPU TSS segments. Threads are completely 'soft' on Linux, @@ -37,7 +38,7 @@ * section. Since TSS's are completely CPU-local, we want them * on exact cacheline boundaries, to eliminate cacheline ping-pong. */ -__visible DEFINE_PER_CPU_SHARED_ALIGNED(struct tss_struct, init_tss) = INIT_TSS; +__visible DEFINE_PER_CPU_SHARED_ALIGNED_USER_MAPPED(struct tss_struct, init_tss) = INIT_TSS; #ifdef CONFIG_X86_64 static DEFINE_PER_CPU(unsigned char, is_idle); @@ -139,7 +140,7 @@ void flush_thread(void) static void hard_disable_TSC(void) { - write_cr4(read_cr4() | X86_CR4_TSD); + cr4_set_bits(X86_CR4_TSD); } void disable_TSC(void) @@ -156,7 +157,7 @@ void disable_TSC(void) static void hard_enable_TSC(void) { - write_cr4(read_cr4() & ~X86_CR4_TSD); + cr4_clear_bits(X86_CR4_TSD); } static void enable_TSC(void) diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index a6f51ad11d89..838d259bca74 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -53,7 +53,7 @@ asmlinkage extern void ret_from_fork(void); -__visible DEFINE_PER_CPU(unsigned long, old_rsp); +__visible DEFINE_PER_CPU_USER_MAPPED(unsigned long, old_rsp); /* Prints also some state that isn't saved in the pt_regs */ void __show_regs(struct pt_regs *regs, int all) diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 5e8c0f1c99c4..281aa6bcafd9 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -92,6 +92,10 @@ void __noreturn machine_real_restart(unsigned int type) load_cr3(initial_page_table); #else write_cr3(real_mode_header->trampoline_pgd); + + /* Exiting long mode will fail if CR4.PCIDE is set. */ + if (static_cpu_has(X86_FEATURE_PCID)) + cr4_clear_bits(X86_CR4_PCIDE); #endif /* Jump to the identity-mapped low memory code */ diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index b1077b8ee7af..28373e51b94a 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -110,6 +110,7 @@ #include #include #include +#include /* * max_low_pfn_mapped: highest direct mapped pfn under 4GB @@ -1019,6 +1020,12 @@ void __init setup_arch(char **cmdline_p) */ init_hypervisor_platform(); + /* + * This needs to happen right after XENPV is set on xen and + * kaiser_enabled is checked below in cleanup_highmap(). + */ + kaiser_check_boottime_disable(); + x86_init.resources.probe_roms(); /* after parse_early_param, so could debug it */ diff --git a/arch/x86/kernel/tracepoint.c b/arch/x86/kernel/tracepoint.c index 1c113db9ed57..2bb5ee464df3 100644 --- a/arch/x86/kernel/tracepoint.c +++ b/arch/x86/kernel/tracepoint.c @@ -9,10 +9,12 @@ #include atomic_t trace_idt_ctr = ATOMIC_INIT(0); +__aligned(PAGE_SIZE) struct desc_ptr trace_idt_descr = { NR_VECTORS * 16 - 1, (unsigned long) trace_idt_table }; /* No need to be aligned, but done to keep all IDTs defined the same way. */ +__aligned(PAGE_SIZE) gate_desc trace_idt_table[NR_VECTORS] __page_aligned_bss; static int trace_irq_vector_refcount; diff --git a/arch/x86/kernel/vm86_32.c b/arch/x86/kernel/vm86_32.c index e8edcf52e069..3dd3508ad297 100644 --- a/arch/x86/kernel/vm86_32.c +++ b/arch/x86/kernel/vm86_32.c @@ -194,7 +194,7 @@ static void mark_screen_rdonly(struct mm_struct *mm) pte_unmap_unlock(pte, ptl); out: up_write(&mm->mmap_sem); - flush_tlb(); + flush_tlb_mm_range(mm, 0xA0000, 0xA0000 + 32*PAGE_SIZE, 0UL); } diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c index e1e1e80fc6a6..5a615d801815 100644 --- a/arch/x86/kernel/vsyscall_64.c +++ b/arch/x86/kernel/vsyscall_64.c @@ -55,6 +55,7 @@ DEFINE_VVAR(int, vgetcpu_mode); static enum { EMULATE, NATIVE, NONE } vsyscall_mode = EMULATE; +unsigned long vsyscall_pgprot = __PAGE_KERNEL_VSYSCALL; static int __init vsyscall_setup(char *str) { @@ -75,6 +76,11 @@ static int __init vsyscall_setup(char *str) } early_param("vsyscall", vsyscall_setup); +bool vsyscall_enabled(void) +{ + return vsyscall_mode != NONE; +} + static void warn_bad_vsyscall(const char *level, struct pt_regs *regs, const char *message) { @@ -331,10 +337,10 @@ void __init map_vsyscall(void) extern char __vsyscall_page; unsigned long physaddr_vsyscall = __pa_symbol(&__vsyscall_page); + if (vsyscall_mode != NATIVE) + vsyscall_pgprot = __PAGE_KERNEL_VVAR; __set_fixmap(VSYSCALL_PAGE, physaddr_vsyscall, - vsyscall_mode == NATIVE - ? PAGE_KERNEL_VSYSCALL - : PAGE_KERNEL_VVAR); + __pgprot(vsyscall_pgprot)); BUILD_BUG_ON((unsigned long)__fix_to_virt(VSYSCALL_PAGE) != (unsigned long)VSYSCALL_ADDR); } diff --git a/arch/x86/kernel/xsave.c b/arch/x86/kernel/xsave.c index beddb0344d52..cfc2dbffa16c 100644 --- a/arch/x86/kernel/xsave.c +++ b/arch/x86/kernel/xsave.c @@ -11,6 +11,7 @@ #include #include #include +#include #include /* @@ -450,7 +451,7 @@ static void prepare_fx_sw_frame(void) */ static inline void xstate_enable(void) { - set_in_cr4(X86_CR4_OSXSAVE); + cr4_set_bits(X86_CR4_OSXSAVE); xsetbv(XCR_XFEATURE_ENABLED_MASK, pcntxt_mask); } diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index f9a6d8d68720..6f6f77a5145c 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -2738,7 +2738,7 @@ static int hardware_enable(void *garbage) /* enable and lock */ wrmsrl(MSR_IA32_FEATURE_CONTROL, old | test_bits); } - write_cr4(read_cr4() | X86_CR4_VMXE); /* FIXME: not cpu hotplug safe */ + cr4_set_bits(X86_CR4_VMXE); if (vmm_exclusive) { kvm_cpu_vmxon(phys_addr); @@ -2775,7 +2775,7 @@ static void hardware_disable(void *garbage) vmclear_local_loaded_vmcss(); kvm_cpu_vmxoff(); } - write_cr4(read_cr4() & ~X86_CR4_VMXE); + cr4_clear_bits(X86_CR4_VMXE); } static __init int adjust_vmx_controls(u32 ctl_min, u32 ctl_opt, diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 2d7ccb5a2bdd..d0847b33282f 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -688,7 +688,8 @@ int kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) return 1; /* PCID can not be enabled when cr3[11:0]!=000H or EFER.LMA=0 */ - if ((kvm_read_cr3(vcpu) & X86_CR3_PCID_MASK) || !is_long_mode(vcpu)) + if ((kvm_read_cr3(vcpu) & X86_CR3_PCID_ASID_MASK) || + !is_long_mode(vcpu)) return 1; } diff --git a/arch/x86/lib/clear_page_64.S b/arch/x86/lib/clear_page_64.S index f2145cfa12a6..38e57faefd71 100644 --- a/arch/x86/lib/clear_page_64.S +++ b/arch/x86/lib/clear_page_64.S @@ -67,7 +67,7 @@ ENDPROC(clear_page) .previous .section .altinstructions,"a" altinstruction_entry clear_page,1b,X86_FEATURE_REP_GOOD,\ - .Lclear_page_end-clear_page, 2b-1b + .Lclear_page_end-clear_page, 2b-1b, 0 altinstruction_entry clear_page,2b,X86_FEATURE_ERMS, \ - .Lclear_page_end-clear_page,3b-2b + .Lclear_page_end-clear_page,3b-2b, 0 .previous diff --git a/arch/x86/lib/cmdline.c b/arch/x86/lib/cmdline.c index 422db000d727..3261abb21ef4 100644 --- a/arch/x86/lib/cmdline.c +++ b/arch/x86/lib/cmdline.c @@ -21,12 +21,16 @@ static inline int myisspace(u8 c) * @option: option string to look for * * Returns the position of that @option (starts counting with 1) - * or 0 on not found. + * or 0 on not found. @option will only be found if it is found + * as an entire word in @cmdline. For instance, if @option="car" + * then a cmdline which contains "cart" will not match. */ -int cmdline_find_option_bool(const char *cmdline, const char *option) +static int +__cmdline_find_option_bool(const char *cmdline, int max_cmdline_size, + const char *option) { char c; - int len, pos = 0, wstart = 0; + int pos = 0, wstart = 0; const char *opptr = NULL; enum { st_wordstart = 0, /* Start of word/after whitespace */ @@ -37,11 +41,11 @@ int cmdline_find_option_bool(const char *cmdline, const char *option) if (!cmdline) return -1; /* No command line */ - len = min_t(int, strlen(cmdline), COMMAND_LINE_SIZE); - if (!len) - return 0; - - while (len--) { + /* + * This 'pos' check ensures we do not overrun + * a non-NULL-terminated 'cmdline' + */ + while (pos < max_cmdline_size) { c = *(char *)cmdline++; pos++; @@ -58,18 +62,35 @@ int cmdline_find_option_bool(const char *cmdline, const char *option) /* fall through */ case st_wordcmp: - if (!*opptr) + if (!*opptr) { + /* + * We matched all the way to the end of the + * option we were looking for. If the + * command-line has a space _or_ ends, then + * we matched! + */ if (!c || myisspace(c)) return wstart; - else - state = st_wordskip; - else if (!c) + /* + * We hit the end of the option, but _not_ + * the end of a word on the cmdline. Not + * a match. + */ + } else if (!c) { + /* + * Hit the NULL terminator on the end of + * cmdline. + */ return 0; - else if (c != *opptr++) - state = st_wordskip; - else if (!len) /* last word and is matching */ - return wstart; - break; + } else if (c == *opptr++) { + /* + * We are currently matching, so continue + * to the next character on the cmdline. + */ + break; + } + state = st_wordskip; + /* fall through */ case st_wordskip: if (!c) @@ -82,3 +103,113 @@ int cmdline_find_option_bool(const char *cmdline, const char *option) return 0; /* Buffer overrun */ } + +/* + * Find a non-boolean option (i.e. option=argument). In accordance with + * standard Linux practice, if this option is repeated, this returns the + * last instance on the command line. + * + * @cmdline: the cmdline string + * @max_cmdline_size: the maximum size of cmdline + * @option: option string to look for + * @buffer: memory buffer to return the option argument + * @bufsize: size of the supplied memory buffer + * + * Returns the length of the argument (regardless of if it was + * truncated to fit in the buffer), or -1 on not found. + */ +static int +__cmdline_find_option(const char *cmdline, int max_cmdline_size, + const char *option, char *buffer, int bufsize) +{ + char c; + int pos = 0, len = -1; + const char *opptr = NULL; + char *bufptr = buffer; + enum { + st_wordstart = 0, /* Start of word/after whitespace */ + st_wordcmp, /* Comparing this word */ + st_wordskip, /* Miscompare, skip */ + st_bufcpy, /* Copying this to buffer */ + } state = st_wordstart; + + if (!cmdline) + return -1; /* No command line */ + + /* + * This 'pos' check ensures we do not overrun + * a non-NULL-terminated 'cmdline' + */ + while (pos++ < max_cmdline_size) { + c = *(char *)cmdline++; + if (!c) + break; + + switch (state) { + case st_wordstart: + if (myisspace(c)) + break; + + state = st_wordcmp; + opptr = option; + /* fall through */ + + case st_wordcmp: + if ((c == '=') && !*opptr) { + /* + * We matched all the way to the end of the + * option we were looking for, prepare to + * copy the argument. + */ + len = 0; + bufptr = buffer; + state = st_bufcpy; + break; + } else if (c == *opptr++) { + /* + * We are currently matching, so continue + * to the next character on the cmdline. + */ + break; + } + state = st_wordskip; + /* fall through */ + + case st_wordskip: + if (myisspace(c)) + state = st_wordstart; + break; + + case st_bufcpy: + if (myisspace(c)) { + state = st_wordstart; + } else { + /* + * Increment len, but don't overrun the + * supplied buffer and leave room for the + * NULL terminator. + */ + if (++len < bufsize) + *bufptr++ = c; + } + break; + } + } + + if (bufsize) + *bufptr = '\0'; + + return len; +} + +int cmdline_find_option_bool(const char *cmdline, const char *option) +{ + return __cmdline_find_option_bool(cmdline, COMMAND_LINE_SIZE, option); +} + +int cmdline_find_option(const char *cmdline, const char *option, char *buffer, + int bufsize) +{ + return __cmdline_find_option(cmdline, COMMAND_LINE_SIZE, option, + buffer, bufsize); +} diff --git a/arch/x86/lib/copy_page_64.S b/arch/x86/lib/copy_page_64.S index 176cca67212b..f1ffdbb07755 100644 --- a/arch/x86/lib/copy_page_64.S +++ b/arch/x86/lib/copy_page_64.S @@ -106,5 +106,5 @@ ENDPROC(copy_page) .previous .section .altinstructions,"a" altinstruction_entry copy_page, 1b, X86_FEATURE_REP_GOOD, \ - .Lcopy_page_end-copy_page, 2b-1b + .Lcopy_page_end-copy_page, 2b-1b, 0 .previous diff --git a/arch/x86/lib/copy_user_64.S b/arch/x86/lib/copy_user_64.S index dee945d55594..fca1f3a32bf5 100644 --- a/arch/x86/lib/copy_user_64.S +++ b/arch/x86/lib/copy_user_64.S @@ -28,19 +28,18 @@ */ .macro ALTERNATIVE_JUMP feature1,feature2,orig,alt1,alt2 0: - .byte 0xe9 /* 32bit jump */ - .long \orig-1f /* by default jump to orig */ + jmp \orig 1: .section .altinstr_replacement,"ax" -2: .byte 0xe9 /* near jump with 32bit immediate */ - .long \alt1-1b /* offset */ /* or alternatively to alt1 */ -3: .byte 0xe9 /* near jump with 32bit immediate */ - .long \alt2-1b /* offset */ /* or alternatively to alt2 */ +2: + jmp \alt1 +3: + jmp \alt2 .previous .section .altinstructions,"a" - altinstruction_entry 0b,2b,\feature1,5,5 - altinstruction_entry 0b,3b,\feature2,5,5 + altinstruction_entry 0b,2b,\feature1,5,5,0 + altinstruction_entry 0b,3b,\feature2,5,5,0 .previous .endm diff --git a/arch/x86/lib/memcpy_64.S b/arch/x86/lib/memcpy_64.S index 56313a326188..f7766e8a497d 100644 --- a/arch/x86/lib/memcpy_64.S +++ b/arch/x86/lib/memcpy_64.S @@ -199,8 +199,8 @@ ENDPROC(__memcpy) * only outcome... */ .section .altinstructions, "a" - altinstruction_entry memcpy,.Lmemcpy_c,X86_FEATURE_REP_GOOD,\ - .Lmemcpy_e-.Lmemcpy_c,.Lmemcpy_e-.Lmemcpy_c - altinstruction_entry memcpy,.Lmemcpy_c_e,X86_FEATURE_ERMS, \ - .Lmemcpy_e_e-.Lmemcpy_c_e,.Lmemcpy_e_e-.Lmemcpy_c_e + altinstruction_entry __memcpy,.Lmemcpy_c,X86_FEATURE_REP_GOOD,\ + .Lmemcpy_e-.Lmemcpy_c,.Lmemcpy_e-.Lmemcpy_c,0 + altinstruction_entry __memcpy,.Lmemcpy_c_e,X86_FEATURE_ERMS, \ + .Lmemcpy_e_e-.Lmemcpy_c_e,.Lmemcpy_e_e-.Lmemcpy_c_e,0 .previous diff --git a/arch/x86/lib/memmove_64.S b/arch/x86/lib/memmove_64.S index 65268a6104f4..d290a599996c 100644 --- a/arch/x86/lib/memmove_64.S +++ b/arch/x86/lib/memmove_64.S @@ -218,6 +218,6 @@ ENTRY(memmove) altinstruction_entry .Lmemmove_begin_forward, \ .Lmemmove_begin_forward_efs,X86_FEATURE_ERMS, \ .Lmemmove_end_forward-.Lmemmove_begin_forward, \ - .Lmemmove_end_forward_efs-.Lmemmove_begin_forward_efs + .Lmemmove_end_forward_efs-.Lmemmove_begin_forward_efs,0 .previous ENDPROC(memmove) diff --git a/arch/x86/lib/memset_64.S b/arch/x86/lib/memset_64.S index 2dcb3808cbda..aba502738043 100644 --- a/arch/x86/lib/memset_64.S +++ b/arch/x86/lib/memset_64.S @@ -147,8 +147,8 @@ ENDPROC(__memset) * feature to implement the right patch order. */ .section .altinstructions,"a" - altinstruction_entry memset,.Lmemset_c,X86_FEATURE_REP_GOOD,\ - .Lfinal-memset,.Lmemset_e-.Lmemset_c - altinstruction_entry memset,.Lmemset_c_e,X86_FEATURE_ERMS, \ - .Lfinal-memset,.Lmemset_e_e-.Lmemset_c_e + altinstruction_entry __memset,.Lmemset_c,X86_FEATURE_REP_GOOD,\ + .Lfinal-__memset,.Lmemset_e-.Lmemset_c,0 + altinstruction_entry __memset,.Lmemset_c_e,X86_FEATURE_ERMS, \ + .Lfinal-__memset,.Lmemset_e_e-.Lmemset_c_e,0 .previous diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile index 6a19ad9f370d..8a8ac114358f 100644 --- a/arch/x86/mm/Makefile +++ b/arch/x86/mm/Makefile @@ -1,5 +1,5 @@ obj-y := init.o init_$(BITS).o fault.o ioremap.o extable.o pageattr.o mmap.o \ - pat.o pgtable.o physaddr.o gup.o setup_nx.o + pat.o pgtable.o physaddr.o gup.o setup_nx.o tlb.o # Make sure __phys_addr has no stackprotector nostackp := $(call cc-option, -fno-stack-protector) @@ -9,7 +9,6 @@ CFLAGS_setup_nx.o := $(nostackp) CFLAGS_fault.o := -I$(src)/../include/asm/trace obj-$(CONFIG_X86_PAT) += pat_rbtree.o -obj-$(CONFIG_SMP) += tlb.o obj-$(CONFIG_X86_32) += pgtable_32.o iomap_32.o @@ -30,3 +29,4 @@ obj-$(CONFIG_ACPI_NUMA) += srat.o obj-$(CONFIG_NUMA_EMU) += numa_emulation.o obj-$(CONFIG_MEMTEST) += memtest.o +obj-$(CONFIG_PAGE_TABLE_ISOLATION) += kaiser.o diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c index 89c43a1ce82b..471f1ff32eb1 100644 --- a/arch/x86/mm/init.c +++ b/arch/x86/mm/init.c @@ -137,11 +137,11 @@ static void __init probe_page_size_mask(void) /* Enable PSE if available */ if (cpu_has_pse) - set_in_cr4(X86_CR4_PSE); + cr4_set_bits_and_update_boot(X86_CR4_PSE); /* Enable PGE if available */ - if (cpu_has_pge) { - set_in_cr4(X86_CR4_PGE); + if (cpu_has_pge && !kaiser_enabled) { + cr4_set_bits_and_update_boot(X86_CR4_PGE); __supported_pte_mask |= _PAGE_GLOBAL; } } diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 12e5ac7885f8..10a4b72dcc1b 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -396,6 +396,16 @@ void __init cleanup_highmap(void) continue; if (vaddr < (unsigned long) _text || vaddr > end) set_pmd(pmd, __pmd(0)); + else if (kaiser_enabled) { + /* + * level2_kernel_pgt is initialized with _PAGE_GLOBAL: + * clear that now. This is not important, so long as + * CR4.PGE remains clear, but it removes an anomaly. + * Physical mapping setup below avoids _PAGE_GLOBAL + * by use of massage_pgprot() inside pfn_pte() etc. + */ + set_pmd(pmd, pmd_clear_flags(*pmd, _PAGE_GLOBAL)); + } } } diff --git a/arch/x86/mm/kaiser.c b/arch/x86/mm/kaiser.c new file mode 100644 index 000000000000..8d1019d176e2 --- /dev/null +++ b/arch/x86/mm/kaiser.c @@ -0,0 +1,469 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#undef pr_fmt +#define pr_fmt(fmt) "Kernel/User page tables isolation: " fmt + +extern struct mm_struct init_mm; + +#include +#include /* to verify its kaiser declarations */ +#include +#include +#include +#include +#include + +int kaiser_enabled __read_mostly = 1; +EXPORT_SYMBOL(kaiser_enabled); /* for inlined TLB flush functions */ + +__visible +DEFINE_PER_CPU_USER_MAPPED(unsigned long, unsafe_stack_register_backup); + +/* + * These can have bit 63 set, so we can not just use a plain "or" + * instruction to get their value or'd into CR3. It would take + * another register. So, we use a memory reference to these instead. + * + * This is also handy because systems that do not support PCIDs + * just end up or'ing a 0 into their CR3, which does no harm. + */ +DEFINE_PER_CPU(unsigned long, x86_cr3_pcid_user); + +/* + * At runtime, the only things we map are some things for CPU + * hotplug, and stacks for new processes. No two CPUs will ever + * be populating the same addresses, so we only need to ensure + * that we protect between two CPUs trying to allocate and + * populate the same page table page. + * + * Only take this lock when doing a set_p[4um]d(), but it is not + * needed for doing a set_pte(). We assume that only the *owner* + * of a given allocation will be doing this for _their_ + * allocation. + * + * This ensures that once a system has been running for a while + * and there have been stacks all over and these page tables + * are fully populated, there will be no further acquisitions of + * this lock. + */ +static DEFINE_SPINLOCK(shadow_table_allocation_lock); + +/* + * Returns -1 on error. + */ +static inline unsigned long get_pa_from_mapping(unsigned long vaddr) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + + pgd = pgd_offset_k(vaddr); + /* + * We made all the kernel PGDs present in kaiser_init(). + * We expect them to stay that way. + */ + BUG_ON(pgd_none(*pgd)); + /* + * PGDs are either 512GB or 128TB on all x86_64 + * configurations. We don't handle these. + */ + BUG_ON(pgd_large(*pgd)); + + pud = pud_offset(pgd, vaddr); + if (pud_none(*pud)) { + WARN_ON_ONCE(1); + return -1; + } + + if (pud_large(*pud)) + return (pud_pfn(*pud) << PAGE_SHIFT) | (vaddr & ~PUD_PAGE_MASK); + + pmd = pmd_offset(pud, vaddr); + if (pmd_none(*pmd)) { + WARN_ON_ONCE(1); + return -1; + } + + if (pmd_large(*pmd)) + return (pmd_pfn(*pmd) << PAGE_SHIFT) | (vaddr & ~PMD_PAGE_MASK); + + pte = pte_offset_kernel(pmd, vaddr); + if (pte_none(*pte)) { + WARN_ON_ONCE(1); + return -1; + } + + return (pte_pfn(*pte) << PAGE_SHIFT) | (vaddr & ~PAGE_MASK); +} + +/* + * This is a relatively normal page table walk, except that it + * also tries to allocate page tables pages along the way. + * + * Returns a pointer to a PTE on success, or NULL on failure. + */ +static pte_t *kaiser_pagetable_walk(unsigned long address, bool user) +{ + pmd_t *pmd; + pud_t *pud; + pgd_t *pgd = native_get_shadow_pgd(pgd_offset_k(address)); + gfp_t gfp = (GFP_KERNEL | __GFP_NOTRACK | __GFP_ZERO); + unsigned long prot = _KERNPG_TABLE; + + if (pgd_none(*pgd)) { + WARN_ONCE(1, "All shadow pgds should have been populated"); + return NULL; + } + BUILD_BUG_ON(pgd_large(*pgd) != 0); + + if (user) { + set_pgd(pgd, __pgd(pgd_val(*pgd) | _PAGE_USER)); + prot = _PAGE_TABLE; + } + + pud = pud_offset(pgd, address); + /* The shadow page tables do not use large mappings: */ + if (pud_large(*pud)) { + WARN_ON(1); + return NULL; + } + if (pud_none(*pud)) { + unsigned long new_pmd_page = __get_free_page(gfp); + if (!new_pmd_page) + return NULL; + spin_lock(&shadow_table_allocation_lock); + if (pud_none(*pud)) { + set_pud(pud, __pud(prot | __pa(new_pmd_page))); + __inc_zone_page_state(virt_to_page((void *) + new_pmd_page), NR_KAISERTABLE); + } else + free_page(new_pmd_page); + spin_unlock(&shadow_table_allocation_lock); + } + + pmd = pmd_offset(pud, address); + /* The shadow page tables do not use large mappings: */ + if (pmd_large(*pmd)) { + WARN_ON(1); + return NULL; + } + if (pmd_none(*pmd)) { + unsigned long new_pte_page = __get_free_page(gfp); + if (!new_pte_page) + return NULL; + spin_lock(&shadow_table_allocation_lock); + if (pmd_none(*pmd)) { + set_pmd(pmd, __pmd(prot | __pa(new_pte_page))); + __inc_zone_page_state(virt_to_page((void *) + new_pte_page), NR_KAISERTABLE); + } else + free_page(new_pte_page); + spin_unlock(&shadow_table_allocation_lock); + } + + return pte_offset_kernel(pmd, address); +} + +static int kaiser_add_user_map(const void *__start_addr, unsigned long size, + unsigned long flags) +{ + int ret = 0; + pte_t *pte; + unsigned long start_addr = (unsigned long )__start_addr; + unsigned long address = start_addr & PAGE_MASK; + unsigned long end_addr = PAGE_ALIGN(start_addr + size); + unsigned long target_address; + + /* + * It is convenient for callers to pass in __PAGE_KERNEL etc, + * and there is no actual harm from setting _PAGE_GLOBAL, so + * long as CR4.PGE is not set. But it is nonetheless troubling + * to see Kaiser itself setting _PAGE_GLOBAL (now that "nokaiser" + * requires that not to be #defined to 0): so mask it off here. + */ + flags &= ~_PAGE_GLOBAL; + + for (; address < end_addr; address += PAGE_SIZE) { + target_address = get_pa_from_mapping(address); + if (target_address == -1) { + ret = -EIO; + break; + } + pte = kaiser_pagetable_walk(address, flags & _PAGE_USER); + if (!pte) { + ret = -ENOMEM; + break; + } + if (pte_none(*pte)) { + set_pte(pte, __pte(flags | target_address)); + } else { + pte_t tmp; + set_pte(&tmp, __pte(flags | target_address)); + WARN_ON_ONCE(!pte_same(*pte, tmp)); + } + } + return ret; +} + +static int kaiser_add_user_map_ptrs(const void *start, const void *end, unsigned long flags) +{ + unsigned long size = end - start; + + return kaiser_add_user_map(start, size, flags); +} + +/* + * Ensure that the top level of the (shadow) page tables are + * entirely populated. This ensures that all processes that get + * forked have the same entries. This way, we do not have to + * ever go set up new entries in older processes. + * + * Note: we never free these, so there are no updates to them + * after this. + */ +static void __init kaiser_init_all_pgds(void) +{ + pgd_t *pgd; + int i = 0; + + pgd = native_get_shadow_pgd(pgd_offset_k((unsigned long )0)); + for (i = PTRS_PER_PGD / 2; i < PTRS_PER_PGD; i++) { + pgd_t new_pgd; + pud_t *pud = pud_alloc_one(&init_mm, + PAGE_OFFSET + i * PGDIR_SIZE); + if (!pud) { + WARN_ON(1); + break; + } + inc_zone_page_state(virt_to_page(pud), NR_KAISERTABLE); + new_pgd = __pgd(_KERNPG_TABLE |__pa(pud)); + /* + * Make sure not to stomp on some other pgd entry. + */ + if (!pgd_none(pgd[i])) { + WARN_ON(1); + continue; + } + set_pgd(pgd + i, new_pgd); + } +} + +#define kaiser_add_user_map_early(start, size, flags) do { \ + int __ret = kaiser_add_user_map(start, size, flags); \ + WARN_ON(__ret); \ +} while (0) + +#define kaiser_add_user_map_ptrs_early(start, end, flags) do { \ + int __ret = kaiser_add_user_map_ptrs(start, end, flags); \ + WARN_ON(__ret); \ +} while (0) + +void __init kaiser_check_boottime_disable(void) +{ + bool enable = true; + char arg[5]; + int ret; + + if (xen_pv_domain()) + goto silent_disable; + + ret = cmdline_find_option(boot_command_line, "pti", arg, sizeof(arg)); + if (ret > 0) { + if (!strncmp(arg, "on", 2)) + goto enable; + + if (!strncmp(arg, "off", 3)) + goto disable; + + if (!strncmp(arg, "auto", 4)) + goto skip; + } + + if (cmdline_find_option_bool(boot_command_line, "nopti")) + goto disable; + +skip: + if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) + goto disable; + +enable: + if (enable) + setup_force_cpu_cap(X86_FEATURE_KAISER); + + return; + +disable: + pr_info("disabled\n"); + +silent_disable: + kaiser_enabled = 0; + setup_clear_cpu_cap(X86_FEATURE_KAISER); +} + +/* + * If anything in here fails, we will likely die on one of the + * first kernel->user transitions and init will die. But, we + * will have most of the kernel up by then and should be able to + * get a clean warning out of it. If we BUG_ON() here, we run + * the risk of being before we have good console output. + */ +void __init kaiser_init(void) +{ + int cpu, idx; + + if (!kaiser_enabled) + return; + + kaiser_init_all_pgds(); + + for_each_possible_cpu(cpu) { + void *percpu_vaddr = __per_cpu_user_mapped_start + + per_cpu_offset(cpu); + unsigned long percpu_sz = __per_cpu_user_mapped_end - + __per_cpu_user_mapped_start; + kaiser_add_user_map_early(percpu_vaddr, percpu_sz, + __PAGE_KERNEL); + } + + /* + * Map the entry/exit text section, which is needed at + * switches from user to and from kernel. + */ + kaiser_add_user_map_ptrs_early(__entry_text_start, __entry_text_end, + __PAGE_KERNEL_RX); + +#if defined(CONFIG_FUNCTION_GRAPH_TRACER) || defined(CONFIG_KASAN) + kaiser_add_user_map_ptrs_early(__irqentry_text_start, + __irqentry_text_end, + __PAGE_KERNEL_RX); +#endif + kaiser_add_user_map_early((void *)idt_descr.address, + sizeof(gate_desc) * NR_VECTORS, + __PAGE_KERNEL_RO); +#ifdef CONFIG_TRACING + kaiser_add_user_map_early(&trace_idt_descr, + sizeof(trace_idt_descr), + __PAGE_KERNEL); + kaiser_add_user_map_early(&trace_idt_table, + sizeof(gate_desc) * NR_VECTORS, + __PAGE_KERNEL); +#endif + kaiser_add_user_map_early(&debug_idt_descr, sizeof(debug_idt_descr), + __PAGE_KERNEL); + kaiser_add_user_map_early(&debug_idt_table, + sizeof(gate_desc) * NR_VECTORS, + __PAGE_KERNEL); + + if (vsyscall_enabled()) + kaiser_add_user_map_early((void *)VSYSCALL_ADDR, PAGE_SIZE, + vsyscall_pgprot); + + pr_info("enabled\n"); +} + +/* Add a mapping to the shadow mapping, and synchronize the mappings */ +int kaiser_add_mapping(unsigned long addr, unsigned long size, unsigned long flags) +{ + if (!kaiser_enabled) + return 0; + return kaiser_add_user_map((const void *)addr, size, flags); +} + +void kaiser_remove_mapping(unsigned long start, unsigned long size) +{ + extern void unmap_pud_range_nofree(pgd_t *pgd, + unsigned long start, unsigned long end); + unsigned long end = start + size; + unsigned long addr, next; + pgd_t *pgd; + + if (!kaiser_enabled) + return; + pgd = native_get_shadow_pgd(pgd_offset_k(start)); + for (addr = start; addr < end; pgd++, addr = next) { + next = pgd_addr_end(addr, end); + unmap_pud_range_nofree(pgd, addr, next); + } +} + +/* + * Page table pages are page-aligned. The lower half of the top + * level is used for userspace and the top half for the kernel. + * This returns true for user pages that need to get copied into + * both the user and kernel copies of the page tables, and false + * for kernel pages that should only be in the kernel copy. + */ +static inline bool is_userspace_pgd(pgd_t *pgdp) +{ + return ((unsigned long)pgdp % PAGE_SIZE) < (PAGE_SIZE / 2); +} + +pgd_t kaiser_set_shadow_pgd(pgd_t *pgdp, pgd_t pgd) +{ + if (!kaiser_enabled) + return pgd; + /* + * Do we need to also populate the shadow pgd? Check _PAGE_USER to + * skip cases like kexec and EFI which make temporary low mappings. + */ + if (pgd.pgd & _PAGE_USER) { + if (is_userspace_pgd(pgdp)) { + native_get_shadow_pgd(pgdp)->pgd = pgd.pgd; + /* + * Even if the entry is *mapping* userspace, ensure + * that userspace can not use it. This way, if we + * get out to userspace running on the kernel CR3, + * userspace will crash instead of running. + */ + pgd.pgd |= _PAGE_NX; + } + } else if (!pgd.pgd) { + /* + * pgd_clear() cannot check _PAGE_USER, and is even used to + * clear corrupted pgd entries: so just rely on cases like + * kexec and EFI never to be using pgd_clear(). + */ + if (!WARN_ON_ONCE((unsigned long)pgdp & PAGE_SIZE) && + is_userspace_pgd(pgdp)) + native_get_shadow_pgd(pgdp)->pgd = pgd.pgd; + } + return pgd; +} + +void kaiser_setup_pcid(void) +{ + unsigned long user_cr3 = KAISER_SHADOW_PGD_OFFSET; + + if (this_cpu_has(X86_FEATURE_PCID)) + user_cr3 |= X86_CR3_PCID_USER_NOFLUSH; + /* + * These variables are used by the entry/exit + * code to change PCID and pgd and TLB flushing. + */ + this_cpu_write(x86_cr3_pcid_user, user_cr3); +} + +/* + * Make a note that this cpu will need to flush USER tlb on return to user. + * If cpu does not have PCID, then the NOFLUSH bit will never have been set. + */ +void kaiser_flush_tlb_on_return_to_user(void) +{ + if (this_cpu_has(X86_FEATURE_PCID)) + this_cpu_write(x86_cr3_pcid_user, + X86_CR3_PCID_USER_FLUSH | KAISER_SHADOW_PGD_OFFSET); +} +EXPORT_SYMBOL(kaiser_flush_tlb_on_return_to_user); diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index e5545f2105f6..aa7cdf54ad85 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -52,6 +52,7 @@ static DEFINE_SPINLOCK(cpa_lock); #define CPA_FLUSHTLB 1 #define CPA_ARRAY 2 #define CPA_PAGES_ARRAY 4 +#define CPA_FREE_PAGETABLES 8 #ifdef CONFIG_PROC_FS static unsigned long direct_pages_count[PG_LEVEL_NUM]; @@ -672,10 +673,13 @@ static int split_large_page(struct cpa_data *cpa, pte_t *kpte, return 0; } -static bool try_to_free_pte_page(pte_t *pte) +static bool try_to_free_pte_page(struct cpa_data *cpa, pte_t *pte) { int i; + if (!(cpa->flags & CPA_FREE_PAGETABLES)) + return false; + for (i = 0; i < PTRS_PER_PTE; i++) if (!pte_none(pte[i])) return false; @@ -684,10 +688,13 @@ static bool try_to_free_pte_page(pte_t *pte) return true; } -static bool try_to_free_pmd_page(pmd_t *pmd) +static bool try_to_free_pmd_page(struct cpa_data *cpa, pmd_t *pmd) { int i; + if (!(cpa->flags & CPA_FREE_PAGETABLES)) + return false; + for (i = 0; i < PTRS_PER_PMD; i++) if (!pmd_none(pmd[i])) return false; @@ -708,7 +715,9 @@ static bool try_to_free_pud_page(pud_t *pud) return true; } -static bool unmap_pte_range(pmd_t *pmd, unsigned long start, unsigned long end) +static bool unmap_pte_range(struct cpa_data *cpa, pmd_t *pmd, + unsigned long start, + unsigned long end) { pte_t *pte = pte_offset_kernel(pmd, start); @@ -719,22 +728,23 @@ static bool unmap_pte_range(pmd_t *pmd, unsigned long start, unsigned long end) pte++; } - if (try_to_free_pte_page((pte_t *)pmd_page_vaddr(*pmd))) { + if (try_to_free_pte_page(cpa, (pte_t *)pmd_page_vaddr(*pmd))) { pmd_clear(pmd); return true; } return false; } -static void __unmap_pmd_range(pud_t *pud, pmd_t *pmd, +static void __unmap_pmd_range(struct cpa_data *cpa, pud_t *pud, pmd_t *pmd, unsigned long start, unsigned long end) { - if (unmap_pte_range(pmd, start, end)) - if (try_to_free_pmd_page((pmd_t *)pud_page_vaddr(*pud))) + if (unmap_pte_range(cpa, pmd, start, end)) + if (try_to_free_pmd_page(cpa, (pmd_t *)pud_page_vaddr(*pud))) pud_clear(pud); } -static void unmap_pmd_range(pud_t *pud, unsigned long start, unsigned long end) +static void unmap_pmd_range(struct cpa_data *cpa, pud_t *pud, + unsigned long start, unsigned long end) { pmd_t *pmd = pmd_offset(pud, start); @@ -745,7 +755,7 @@ static void unmap_pmd_range(pud_t *pud, unsigned long start, unsigned long end) unsigned long next_page = (start + PMD_SIZE) & PMD_MASK; unsigned long pre_end = min_t(unsigned long, end, next_page); - __unmap_pmd_range(pud, pmd, start, pre_end); + __unmap_pmd_range(cpa, pud, pmd, start, pre_end); start = pre_end; pmd++; @@ -758,7 +768,8 @@ static void unmap_pmd_range(pud_t *pud, unsigned long start, unsigned long end) if (pmd_large(*pmd)) pmd_clear(pmd); else - __unmap_pmd_range(pud, pmd, start, start + PMD_SIZE); + __unmap_pmd_range(cpa, pud, pmd, + start, start + PMD_SIZE); start += PMD_SIZE; pmd++; @@ -768,17 +779,19 @@ static void unmap_pmd_range(pud_t *pud, unsigned long start, unsigned long end) * 4K leftovers? */ if (start < end) - return __unmap_pmd_range(pud, pmd, start, end); + return __unmap_pmd_range(cpa, pud, pmd, start, end); /* * Try again to free the PMD page if haven't succeeded above. */ if (!pud_none(*pud)) - if (try_to_free_pmd_page((pmd_t *)pud_page_vaddr(*pud))) + if (try_to_free_pmd_page(cpa, (pmd_t *)pud_page_vaddr(*pud))) pud_clear(pud); } -static void unmap_pud_range(pgd_t *pgd, unsigned long start, unsigned long end) +static void __unmap_pud_range(struct cpa_data *cpa, pgd_t *pgd, + unsigned long start, + unsigned long end) { pud_t *pud = pud_offset(pgd, start); @@ -789,7 +802,7 @@ static void unmap_pud_range(pgd_t *pgd, unsigned long start, unsigned long end) unsigned long next_page = (start + PUD_SIZE) & PUD_MASK; unsigned long pre_end = min_t(unsigned long, end, next_page); - unmap_pmd_range(pud, start, pre_end); + unmap_pmd_range(cpa, pud, start, pre_end); start = pre_end; pud++; @@ -803,7 +816,7 @@ static void unmap_pud_range(pgd_t *pgd, unsigned long start, unsigned long end) if (pud_large(*pud)) pud_clear(pud); else - unmap_pmd_range(pud, start, start + PUD_SIZE); + unmap_pmd_range(cpa, pud, start, start + PUD_SIZE); start += PUD_SIZE; pud++; @@ -813,7 +826,7 @@ static void unmap_pud_range(pgd_t *pgd, unsigned long start, unsigned long end) * 2M leftovers? */ if (start < end) - unmap_pmd_range(pud, start, end); + unmap_pmd_range(cpa, pud, start, end); /* * No need to try to free the PUD page because we'll free it in @@ -821,6 +834,24 @@ static void unmap_pud_range(pgd_t *pgd, unsigned long start, unsigned long end) */ } +static void unmap_pud_range(pgd_t *pgd, unsigned long start, unsigned long end) +{ + struct cpa_data cpa = { + .flags = CPA_FREE_PAGETABLES, + }; + + __unmap_pud_range(&cpa, pgd, start, end); +} + +void unmap_pud_range_nofree(pgd_t *pgd, unsigned long start, unsigned long end) +{ + struct cpa_data cpa = { + .flags = 0, + }; + + __unmap_pud_range(&cpa, pgd, start, end); +} + static void unmap_pgd_range(pgd_t *root, unsigned long addr, unsigned long end) { pgd_t *pgd_entry = root + pgd_index(addr); diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c index 6fb6927f9e76..d230776e9b19 100644 --- a/arch/x86/mm/pgtable.c +++ b/arch/x86/mm/pgtable.c @@ -5,7 +5,7 @@ #include #include -#define PGALLOC_GFP GFP_KERNEL | __GFP_NOTRACK | __GFP_REPEAT | __GFP_ZERO +#define PGALLOC_GFP (GFP_KERNEL | __GFP_NOTRACK | __GFP_REPEAT | __GFP_ZERO) #ifdef CONFIG_HIGHPTE #define PGALLOC_USER_GFP __GFP_HIGHMEM @@ -271,12 +271,31 @@ static void pgd_prepopulate_pmd(struct mm_struct *mm, pgd_t *pgd, pmd_t *pmds[]) } } +/* + * Instead of one pgd, Kaiser acquires two pgds. Being order-1, it is + * both 8k in size and 8k-aligned. That lets us just flip bit 12 + * in a pointer to swap between the two 4k halves. + */ +#define PGD_ALLOCATION_ORDER kaiser_enabled + +static inline pgd_t *_pgd_alloc(void) +{ + /* No __GFP_REPEAT: to avoid page allocation stalls in order-1 case */ + return (pgd_t *)__get_free_pages(PGALLOC_GFP & ~__GFP_REPEAT, + PGD_ALLOCATION_ORDER); +} + +static inline void _pgd_free(pgd_t *pgd) +{ + free_pages((unsigned long)pgd, PGD_ALLOCATION_ORDER); +} + pgd_t *pgd_alloc(struct mm_struct *mm) { pgd_t *pgd; pmd_t *pmds[PREALLOCATED_PMDS]; - pgd = (pgd_t *)__get_free_page(PGALLOC_GFP); + pgd = _pgd_alloc(); if (pgd == NULL) goto out; @@ -306,7 +325,7 @@ pgd_t *pgd_alloc(struct mm_struct *mm) out_free_pmds: free_pmds(pmds); out_free_pgd: - free_page((unsigned long)pgd); + _pgd_free(pgd); out: return NULL; } @@ -316,7 +335,7 @@ void pgd_free(struct mm_struct *mm, pgd_t *pgd) pgd_mop_up_pmds(mm, pgd); pgd_dtor(pgd); paravirt_pgd_free(mm, pgd); - free_page((unsigned long)pgd); + _pgd_free(pgd); } /* diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c index 3465445be664..efa4839c3b59 100644 --- a/arch/x86/mm/tlb.c +++ b/arch/x86/mm/tlb.c @@ -6,19 +6,20 @@ #include #include #include +#include #include #include #include #include #include -#include +#include DEFINE_PER_CPU_SHARED_ALIGNED(struct tlb_state, cpu_tlbstate) = { &init_mm, 0, }; /* - * Smarter SMP flushing macros. + * TLB flushing, formerly SMP-only * c/o Linus Torvalds. * * These mean you can really definitely utterly forget about @@ -37,6 +38,36 @@ struct flush_tlb_info { unsigned long flush_end; }; +static void load_new_mm_cr3(pgd_t *pgdir) +{ + unsigned long new_mm_cr3 = __pa(pgdir); + + if (kaiser_enabled) { + /* + * We reuse the same PCID for different tasks, so we must + * flush all the entries for the PCID out when we change tasks. + * Flush KERN below, flush USER when returning to userspace in + * kaiser's SWITCH_USER_CR3 (_SWITCH_TO_USER_CR3) macro. + * + * invpcid_flush_single_context(X86_CR3_PCID_ASID_USER) could + * do it here, but can only be used if X86_FEATURE_INVPCID is + * available - and many machines support pcid without invpcid. + * + * If X86_CR3_PCID_KERN_FLUSH actually added something, then it + * would be needed in the write_cr3() below - if PCIDs enabled. + */ + BUILD_BUG_ON(X86_CR3_PCID_KERN_FLUSH); + kaiser_flush_tlb_on_return_to_user(); + } + + /* + * Caution: many callers of this function expect + * that load_cr3() is serializing and orders TLB + * fills with respect to the mm_cpumask writes. + */ + write_cr3(new_mm_cr3); +} + /* * We cannot call mmdrop() because we are in interrupt context, * instead update mm->cpu_vm_mask. @@ -48,11 +79,94 @@ void leave_mm(int cpu) BUG(); if (cpumask_test_cpu(cpu, mm_cpumask(active_mm))) { cpumask_clear_cpu(cpu, mm_cpumask(active_mm)); - load_cr3(swapper_pg_dir); + load_new_mm_cr3(swapper_pg_dir); } } EXPORT_SYMBOL_GPL(leave_mm); +void switch_mm(struct mm_struct *prev, struct mm_struct *next, + struct task_struct *tsk) +{ + unsigned long flags; + + local_irq_save(flags); + switch_mm_irqs_off(prev, next, tsk); + local_irq_restore(flags); +} + +void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next, + struct task_struct *tsk) +{ + unsigned cpu = smp_processor_id(); + + if (likely(prev != next)) { + this_cpu_write(cpu_tlbstate.state, TLBSTATE_OK); + this_cpu_write(cpu_tlbstate.active_mm, next); + cpumask_set_cpu(cpu, mm_cpumask(next)); + + /* + * Re-load page tables. + * + * This logic has an ordering constraint: + * + * CPU 0: Write to a PTE for 'next' + * CPU 0: load bit 1 in mm_cpumask. if nonzero, send IPI. + * CPU 1: set bit 1 in next's mm_cpumask + * CPU 1: load from the PTE that CPU 0 writes (implicit) + * + * We need to prevent an outcome in which CPU 1 observes + * the new PTE value and CPU 0 observes bit 1 clear in + * mm_cpumask. (If that occurs, then the IPI will never + * be sent, and CPU 0's TLB will contain a stale entry.) + * + * The bad outcome can occur if either CPU's load is + * reordered before that CPU's store, so both CPUs must + * execute full barriers to prevent this from happening. + * + * Thus, switch_mm needs a full barrier between the + * store to mm_cpumask and any operation that could load + * from next->pgd. TLB fills are special and can happen + * due to instruction fetches or for no reason at all, + * and neither LOCK nor MFENCE orders them. + * Fortunately, load_cr3() is serializing and gives the + * ordering guarantee we need. + * + */ + load_new_mm_cr3(next->pgd); + + /* Stop flush ipis for the previous mm */ + cpumask_clear_cpu(cpu, mm_cpumask(prev)); + + /* Load the LDT, if the LDT is different: */ + if (unlikely(prev->context.ldt != next->context.ldt)) + load_mm_ldt(next); + } else { + this_cpu_write(cpu_tlbstate.state, TLBSTATE_OK); + BUG_ON(this_cpu_read(cpu_tlbstate.active_mm) != next); + + if (!cpumask_test_cpu(cpu, mm_cpumask(next))) { + /* + * On established mms, the mm_cpumask is only changed + * from irq context, from ptep_clear_flush() while in + * lazy tlb mode, and here. Irqs are blocked during + * schedule, protecting us from simultaneous changes. + */ + cpumask_set_cpu(cpu, mm_cpumask(next)); + + /* + * We were in lazy tlb mode and leave_mm disabled + * tlb flush IPI delivery. We must reload CR3 + * to make sure to use no freed page tables. + * + * As above, load_cr3() is serializing and orders TLB + * fills with respect to the mm_cpumask write. + */ + load_new_mm_cr3(next->pgd); + load_mm_ldt(next); + } + } +} + /* * The flush IPI assumes that a thread switch happens in this order: * [cpu0: the cpu that switches] @@ -143,35 +257,37 @@ void native_flush_tlb_others(const struct cpumask *cpumask, smp_call_function_many(cpumask, flush_tlb_func, &info, 1); } -void flush_tlb_current_task(void) -{ - struct mm_struct *mm = current->mm; - - preempt_disable(); - - count_vm_tlb_event(NR_TLB_LOCAL_FLUSH_ALL); - - /* This is an implicit full barrier that synchronizes with switch_mm. */ - local_flush_tlb(); - - if (cpumask_any_but(mm_cpumask(mm), smp_processor_id()) < nr_cpu_ids) - flush_tlb_others(mm_cpumask(mm), mm, 0UL, TLB_FLUSH_ALL); - preempt_enable(); -} +/* + * See Documentation/x86/tlb.txt for details. We choose 33 + * because it is large enough to cover the vast majority (at + * least 95%) of allocations, and is small enough that we are + * confident it will not cause too much overhead. Each single + * flush is about 100 ns, so this caps the maximum overhead at + * _about_ 3,000 ns. + * + * This is in units of pages. + */ +static unsigned long tlb_single_page_flush_ceiling __read_mostly = 33; void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start, unsigned long end, unsigned long vmflag) { unsigned long addr; - unsigned act_entries, tlb_entries = 0; - unsigned long nr_base_pages; + /* do a global flush by default */ + unsigned long base_pages_to_flush = TLB_FLUSH_ALL; preempt_disable(); + + if ((end != TLB_FLUSH_ALL) && !(vmflag & VM_HUGETLB)) + base_pages_to_flush = (end - start) >> PAGE_SHIFT; + if (base_pages_to_flush > tlb_single_page_flush_ceiling) + base_pages_to_flush = TLB_FLUSH_ALL; + if (current->active_mm != mm) { /* Synchronize with switch_mm. */ smp_mb(); - goto flush_all; + goto out; } if (!current->mm) { @@ -180,28 +296,14 @@ void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start, /* Synchronize with switch_mm. */ smp_mb(); - goto flush_all; - } - - if (end == TLB_FLUSH_ALL || tlb_flushall_shift == -1 - || vmflag & VM_HUGETLB) { - local_flush_tlb(); - goto flush_all; + goto out; } - /* In modern CPU, last level tlb used for both data/ins */ - if (vmflag & VM_EXEC) - tlb_entries = tlb_lli_4k[ENTRIES]; - else - tlb_entries = tlb_lld_4k[ENTRIES]; - - /* Assume all of TLB entries was occupied by this task */ - act_entries = tlb_entries >> tlb_flushall_shift; - act_entries = mm->total_vm > act_entries ? act_entries : mm->total_vm; - nr_base_pages = (end - start) >> PAGE_SHIFT; - - /* tlb_flushall_shift is on balance point, details in commit log */ - if (nr_base_pages > act_entries) { + /* + * Both branches below are implicit full barriers (MOV to CR or + * INVLPG) that synchronize with switch_mm. + */ + if (base_pages_to_flush == TLB_FLUSH_ALL) { count_vm_tlb_event(NR_TLB_LOCAL_FLUSH_ALL); local_flush_tlb(); } else { @@ -210,44 +312,14 @@ void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start, count_vm_tlb_event(NR_TLB_LOCAL_FLUSH_ONE); __flush_tlb_single(addr); } - - if (cpumask_any_but(mm_cpumask(mm), - smp_processor_id()) < nr_cpu_ids) - flush_tlb_others(mm_cpumask(mm), mm, start, end); - preempt_enable(); - return; } - -flush_all: - if (cpumask_any_but(mm_cpumask(mm), smp_processor_id()) < nr_cpu_ids) - flush_tlb_others(mm_cpumask(mm), mm, 0UL, TLB_FLUSH_ALL); - preempt_enable(); -} - -void flush_tlb_page(struct vm_area_struct *vma, unsigned long start) -{ - struct mm_struct *mm = vma->vm_mm; - - preempt_disable(); - - if (current->active_mm == mm) { - if (current->mm) { - /* - * Implicit full barrier (INVLPG) that synchronizes - * with switch_mm. - */ - __flush_tlb_one(start); - } else { - leave_mm(smp_processor_id()); - - /* Synchronize with switch_mm. */ - smp_mb(); - } +out: + if (base_pages_to_flush == TLB_FLUSH_ALL) { + start = 0UL; + end = TLB_FLUSH_ALL; } - if (cpumask_any_but(mm_cpumask(mm), smp_processor_id()) < nr_cpu_ids) - flush_tlb_others(mm_cpumask(mm), mm, start, start + PAGE_SIZE); - + flush_tlb_others(mm_cpumask(mm), mm, start, end); preempt_enable(); } @@ -277,32 +349,26 @@ static void do_kernel_range_flush(void *info) void flush_tlb_kernel_range(unsigned long start, unsigned long end) { - unsigned act_entries; - struct flush_tlb_info info; - - /* In modern CPU, last level tlb used for both data/ins */ - act_entries = tlb_lld_4k[ENTRIES]; /* Balance as user space task's flush, a bit conservative */ - if (end == TLB_FLUSH_ALL || tlb_flushall_shift == -1 || - (end - start) >> PAGE_SHIFT > act_entries >> tlb_flushall_shift) - + if (end == TLB_FLUSH_ALL || + (end - start) > tlb_single_page_flush_ceiling * PAGE_SIZE) { on_each_cpu(do_flush_tlb_all, NULL, 1); - else { + } else { + struct flush_tlb_info info; info.flush_start = start; info.flush_end = end; on_each_cpu(do_kernel_range_flush, &info, 1); } } -#ifdef CONFIG_DEBUG_TLBFLUSH static ssize_t tlbflush_read_file(struct file *file, char __user *user_buf, size_t count, loff_t *ppos) { char buf[32]; unsigned int len; - len = sprintf(buf, "%hd\n", tlb_flushall_shift); + len = sprintf(buf, "%ld\n", tlb_single_page_flush_ceiling); return simple_read_from_buffer(user_buf, count, ppos, buf, len); } @@ -311,20 +377,20 @@ static ssize_t tlbflush_write_file(struct file *file, { char buf[32]; ssize_t len; - s8 shift; + int ceiling; len = min(count, sizeof(buf) - 1); if (copy_from_user(buf, user_buf, len)) return -EFAULT; buf[len] = '\0'; - if (kstrtos8(buf, 0, &shift)) + if (kstrtoint(buf, 0, &ceiling)) return -EINVAL; - if (shift < -1 || shift >= BITS_PER_LONG) + if (ceiling < 0) return -EINVAL; - tlb_flushall_shift = shift; + tlb_single_page_flush_ceiling = ceiling; return count; } @@ -334,11 +400,10 @@ static const struct file_operations fops_tlbflush = { .llseek = default_llseek, }; -static int __init create_tlb_flushall_shift(void) +static int __init create_tlb_single_page_flush_ceiling(void) { - debugfs_create_file("tlb_flushall_shift", S_IRUSR | S_IWUSR, + debugfs_create_file("tlb_single_page_flush_ceiling", S_IRUSR | S_IWUSR, arch_debugfs_dir, NULL, &fops_tlbflush); return 0; } -late_initcall(create_tlb_flushall_shift); -#endif +late_initcall(create_tlb_single_page_flush_ceiling); diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index e1e7f4f1fb80..10130e9afd6e 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -430,6 +430,12 @@ static void __init xen_init_cpuid_mask(void) ~((1 << X86_FEATURE_MTRR) | /* disable MTRR */ (1 << X86_FEATURE_ACC)); /* thermal monitoring */ + /* + * Xen PV would need some work to support PCID: CR3 handling as well + * as xen_flush_tlb_others() would need updating. + */ + cpuid_leaf1_ecx_mask &= ~(1 << (X86_FEATURE_PCID % 32)); /* disable PCID */ + if (!xen_initial_domain()) cpuid_leaf1_edx_mask &= ~((1 << X86_FEATURE_ACPI)); /* disable ACPI */ @@ -1526,10 +1532,10 @@ static void xen_pvh_set_cr_flags(int cpu) * set them here. For all, OSFXSR OSXMMEXCPT are set in fpu_init. */ if (cpu_has_pse) - set_in_cr4(X86_CR4_PSE); + cr4_set_bits_and_update_boot(X86_CR4_PSE); if (cpu_has_pge) - set_in_cr4(X86_CR4_PGE); + cr4_set_bits_and_update_boot(X86_CR4_PGE); } /* diff --git a/drivers/lguest/x86/core.c b/drivers/lguest/x86/core.c index 922a1acbf652..6adfd7ba4c97 100644 --- a/drivers/lguest/x86/core.c +++ b/drivers/lguest/x86/core.c @@ -47,6 +47,7 @@ #include #include #include +#include #include "../lg.h" static int cpu_had_pge; @@ -452,9 +453,9 @@ void lguest_arch_handle_trap(struct lg_cpu *cpu) static void adjust_pge(void *on) { if (on) - write_cr4(read_cr4() | X86_CR4_PGE); + cr4_set_bits(X86_CR4_PGE); else - write_cr4(read_cr4() & ~X86_CR4_PGE); + cr4_clear_bits(X86_CR4_PGE); } /*H:020 diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c index 1fe2c8115be0..a9dfe6bd70eb 100644 --- a/drivers/vhost/vhost.c +++ b/drivers/vhost/vhost.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include #include diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index c1c0b0cf39b4..4191efe77e36 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -689,7 +689,14 @@ */ #define PERCPU_INPUT(cacheline) \ VMLINUX_SYMBOL(__per_cpu_start) = .; \ + VMLINUX_SYMBOL(__per_cpu_user_mapped_start) = .; \ *(.data..percpu..first) \ + . = ALIGN(cacheline); \ + *(.data..percpu..user_mapped) \ + *(.data..percpu..user_mapped..shared_aligned) \ + . = ALIGN(PAGE_SIZE); \ + *(.data..percpu..user_mapped..page_aligned) \ + VMLINUX_SYMBOL(__per_cpu_user_mapped_end) = .; \ . = ALIGN(PAGE_SIZE); \ *(.data..percpu..page_aligned) \ . = ALIGN(cacheline); \ diff --git a/include/linux/kaiser.h b/include/linux/kaiser.h new file mode 100644 index 000000000000..58c55b1589d0 --- /dev/null +++ b/include/linux/kaiser.h @@ -0,0 +1,52 @@ +#ifndef _LINUX_KAISER_H +#define _LINUX_KAISER_H + +#ifdef CONFIG_PAGE_TABLE_ISOLATION +#include + +static inline int kaiser_map_thread_stack(void *stack) +{ + /* + * Map that page of kernel stack on which we enter from user context. + */ + return kaiser_add_mapping((unsigned long)stack + + THREAD_SIZE - PAGE_SIZE, PAGE_SIZE, __PAGE_KERNEL); +} + +static inline void kaiser_unmap_thread_stack(void *stack) +{ + /* + * Note: may be called even when kaiser_map_thread_stack() failed. + */ + kaiser_remove_mapping((unsigned long)stack + + THREAD_SIZE - PAGE_SIZE, PAGE_SIZE); +} +#else + +/* + * These stubs are used whenever CONFIG_PAGE_TABLE_ISOLATION is off, which + * includes architectures that support KAISER, but have it disabled. + */ + +static inline void kaiser_init(void) +{ +} +static inline int kaiser_add_mapping(unsigned long addr, + unsigned long size, unsigned long flags) +{ + return 0; +} +static inline void kaiser_remove_mapping(unsigned long start, + unsigned long size) +{ +} +static inline int kaiser_map_thread_stack(void *stack) +{ + return 0; +} +static inline void kaiser_unmap_thread_stack(void *stack) +{ +} + +#endif /* !CONFIG_PAGE_TABLE_ISOLATION */ +#endif /* _LINUX_KAISER_H */ diff --git a/include/linux/mmu_context.h b/include/linux/mmu_context.h index 70fffeba7495..a4441784503b 100644 --- a/include/linux/mmu_context.h +++ b/include/linux/mmu_context.h @@ -1,9 +1,16 @@ #ifndef _LINUX_MMU_CONTEXT_H #define _LINUX_MMU_CONTEXT_H +#include + struct mm_struct; void use_mm(struct mm_struct *mm); void unuse_mm(struct mm_struct *mm); +/* Architectures that care about IRQ state in switch_mm can override this. */ +#ifndef switch_mm_irqs_off +# define switch_mm_irqs_off switch_mm +#endif + #endif diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index c7eb42235fbe..9fe9377821a5 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -131,8 +131,9 @@ enum zone_stat_item { NR_SLAB_RECLAIMABLE, NR_SLAB_UNRECLAIMABLE, NR_PAGETABLE, /* used for pagetables */ - NR_KERNEL_STACK, /* Second 128 byte cacheline */ + NR_KERNEL_STACK, + NR_KAISERTABLE, NR_UNSTABLE_NFS, /* NFS unstable pages */ NR_BOUNCE, NR_VMSCAN_WRITE, diff --git a/include/linux/percpu-defs.h b/include/linux/percpu-defs.h index dec01d6c3f80..c2d267b53d40 100644 --- a/include/linux/percpu-defs.h +++ b/include/linux/percpu-defs.h @@ -1,6 +1,12 @@ #ifndef _LINUX_PERCPU_DEFS_H #define _LINUX_PERCPU_DEFS_H +#ifdef CONFIG_PAGE_TABLE_ISOLATION +#define USER_MAPPED_SECTION "..user_mapped" +#else +#define USER_MAPPED_SECTION "" +#endif + /* * Base implementations of per-CPU variable declarations and definitions, where * the section in which the variable is to be placed is provided by the @@ -94,6 +100,12 @@ #define DEFINE_PER_CPU(type, name) \ DEFINE_PER_CPU_SECTION(type, name, "") +#define DECLARE_PER_CPU_USER_MAPPED(type, name) \ + DECLARE_PER_CPU_SECTION(type, name, USER_MAPPED_SECTION) + +#define DEFINE_PER_CPU_USER_MAPPED(type, name) \ + DEFINE_PER_CPU_SECTION(type, name, USER_MAPPED_SECTION) + /* * Declaration/definition used for per-CPU variables that must come first in * the set of variables. @@ -123,6 +135,14 @@ DEFINE_PER_CPU_SECTION(type, name, PER_CPU_SHARED_ALIGNED_SECTION) \ ____cacheline_aligned_in_smp +#define DECLARE_PER_CPU_SHARED_ALIGNED_USER_MAPPED(type, name) \ + DECLARE_PER_CPU_SECTION(type, name, USER_MAPPED_SECTION PER_CPU_SHARED_ALIGNED_SECTION) \ + ____cacheline_aligned_in_smp + +#define DEFINE_PER_CPU_SHARED_ALIGNED_USER_MAPPED(type, name) \ + DEFINE_PER_CPU_SECTION(type, name, USER_MAPPED_SECTION PER_CPU_SHARED_ALIGNED_SECTION) \ + ____cacheline_aligned_in_smp + #define DECLARE_PER_CPU_ALIGNED(type, name) \ DECLARE_PER_CPU_SECTION(type, name, PER_CPU_ALIGNED_SECTION) \ ____cacheline_aligned @@ -141,11 +161,21 @@ #define DEFINE_PER_CPU_PAGE_ALIGNED(type, name) \ DEFINE_PER_CPU_SECTION(type, name, "..page_aligned") \ __aligned(PAGE_SIZE) +/* + * Declaration/definition used for per-CPU variables that must be page aligned and need to be mapped in user mode. + */ +#define DECLARE_PER_CPU_PAGE_ALIGNED_USER_MAPPED(type, name) \ + DECLARE_PER_CPU_SECTION(type, name, USER_MAPPED_SECTION"..page_aligned") \ + __aligned(PAGE_SIZE) + +#define DEFINE_PER_CPU_PAGE_ALIGNED_USER_MAPPED(type, name) \ + DEFINE_PER_CPU_SECTION(type, name, USER_MAPPED_SECTION"..page_aligned") \ + __aligned(PAGE_SIZE) /* * Declaration/definition used for per-CPU variables that must be read mostly. */ -#define DECLARE_PER_CPU_READ_MOSTLY(type, name) \ +#define DECLARE_PER_CPU_READ_MOSTLY(type, name) \ DECLARE_PER_CPU_SECTION(type, name, "..read_mostly") #define DEFINE_PER_CPU_READ_MOSTLY(type, name) \ diff --git a/init/main.c b/init/main.c index 3861fe648fbd..0190bb249401 100644 --- a/init/main.c +++ b/init/main.c @@ -78,6 +78,7 @@ #include #include #include +#include #include #include @@ -497,6 +498,7 @@ static void __init mm_init(void) percpu_init_late(); pgtable_init(); vmalloc_init(); + kaiser_init(); } asmlinkage __visible void __init start_kernel(void) diff --git a/kernel/fork.c b/kernel/fork.c index 0a2f907b1302..315342c2aac4 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -58,6 +58,7 @@ #include #include #include +#include #include #include #include @@ -158,6 +159,7 @@ static struct thread_info *alloc_thread_info_node(struct task_struct *tsk, static inline void free_thread_info(struct thread_info *ti) { + kaiser_unmap_thread_stack(ti); free_kmem_pages((unsigned long)ti, THREAD_SIZE_ORDER); } # else @@ -316,6 +318,10 @@ static struct task_struct *dup_task_struct(struct task_struct *orig) tsk->stack = ti; + err = kaiser_map_thread_stack(tsk->stack); + if (err) + goto free_ti; + setup_thread_stack(tsk, orig); clear_user_return_notifier(tsk); clear_tsk_need_resched(tsk); diff --git a/kernel/sched/core.c b/kernel/sched/core.c index e6c9458f4b49..d776fb7d902a 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -32,7 +32,7 @@ #include #include #include -#include +#include #include #include #include @@ -2374,7 +2374,7 @@ context_switch(struct rq *rq, struct task_struct *prev, atomic_inc(&oldmm->mm_count); enter_lazy_tlb(oldmm, next); } else - switch_mm(oldmm, mm, next); + switch_mm_irqs_off(oldmm, mm, next); if (!prev->mm) { prev->active_mm = NULL; diff --git a/mm/mmu_context.c b/mm/mmu_context.c index f802c2d216a7..6f4d27c5bb32 100644 --- a/mm/mmu_context.c +++ b/mm/mmu_context.c @@ -4,9 +4,9 @@ */ #include +#include #include #include -#include #include diff --git a/mm/vmstat.c b/mm/vmstat.c index 8c0c76083562..6c70f6647907 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -753,6 +753,7 @@ const char * const vmstat_text[] = { "nr_slab_unreclaimable", "nr_page_table_pages", "nr_kernel_stack", + "nr_overhead", "nr_unstable", "nr_bounce", "nr_vmscan_write", diff --git a/security/Kconfig b/security/Kconfig index beb86b500adf..89e13fb728f5 100644 --- a/security/Kconfig +++ b/security/Kconfig @@ -30,6 +30,16 @@ config SECURITY If you are unsure how to answer this question, answer N. +config PAGE_TABLE_ISOLATION + bool "Remove the kernel mapping in user mode" + default y + depends on X86_64 && SMP + help + This enforces a strict kernel and user space isolation, in order + to close hardware side channels on kernel address information. + + If you are unsure how to answer this question, answer Y. + config SECURITYFS bool "Enable the securityfs filesystem" help