diff options
Diffstat (limited to 'arch')
30 files changed, 279 insertions, 204 deletions
diff --git a/arch/arm/xen/enlighten.c b/arch/arm/xen/enlighten.c index 7d59765aef22..c392e18f1e43 100644 --- a/arch/arm/xen/enlighten.c +++ b/arch/arm/xen/enlighten.c @@ -207,7 +207,7 @@ static void xen_power_off(void) static irqreturn_t xen_arm_callback(int irq, void *arg) { - xen_hvm_evtchn_do_upcall(); + xen_evtchn_do_upcall(); return IRQ_HANDLED; } diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 54b9387c3691..3aaadfd2c8eb 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -255,7 +255,7 @@ config PPC select HAVE_KPROBES select HAVE_KPROBES_ON_FTRACE select HAVE_KRETPROBES - select HAVE_LD_DEAD_CODE_DATA_ELIMINATION if HAVE_OBJTOOL_MCOUNT + select HAVE_LD_DEAD_CODE_DATA_ELIMINATION if HAVE_OBJTOOL_MCOUNT && (!ARCH_USING_PATCHABLE_FUNCTION_ENTRY || (!CC_IS_GCC || GCC_VERSION >= 110100)) select HAVE_LIVEPATCH if HAVE_DYNAMIC_FTRACE_WITH_REGS select HAVE_MOD_ARCH_SPECIFIC select HAVE_NMI if PERF_EVENTS || (PPC64 && PPC_BOOK3S) diff --git a/arch/powerpc/kernel/hw_breakpoint.c b/arch/powerpc/kernel/hw_breakpoint.c index b8513dc3e53a..a1318ce18d0e 100644 --- a/arch/powerpc/kernel/hw_breakpoint.c +++ b/arch/powerpc/kernel/hw_breakpoint.c @@ -230,13 +230,15 @@ void thread_change_pc(struct task_struct *tsk, struct pt_regs *regs) struct arch_hw_breakpoint *info; int i; + preempt_disable(); + for (i = 0; i < nr_wp_slots(); i++) { struct perf_event *bp = __this_cpu_read(bp_per_reg[i]); if (unlikely(bp && counter_arch_bp(bp)->perf_single_step)) goto reset; } - return; + goto out; reset: regs_set_return_msr(regs, regs->msr & ~MSR_SE); @@ -245,6 +247,9 @@ reset: __set_breakpoint(i, info); info->perf_single_step = false; } + +out: + preempt_enable(); } static bool is_larx_stcx_instr(int type) @@ -363,6 +368,11 @@ static void handle_p10dd1_spurious_exception(struct perf_event **bp, } } +/* + * Handle a DABR or DAWR exception. + * + * Called in atomic context. + */ int hw_breakpoint_handler(struct die_args *args) { bool err = false; @@ -490,6 +500,8 @@ NOKPROBE_SYMBOL(hw_breakpoint_handler); /* * Handle single-step exceptions following a DABR hit. + * + * Called in atomic context. */ static int single_step_dabr_instruction(struct die_args *args) { @@ -541,6 +553,8 @@ NOKPROBE_SYMBOL(single_step_dabr_instruction); /* * Handle debug exception notifications. + * + * Called in atomic context. */ int hw_breakpoint_exceptions_notify( struct notifier_block *unused, unsigned long val, void *data) diff --git a/arch/powerpc/kernel/hw_breakpoint_constraints.c b/arch/powerpc/kernel/hw_breakpoint_constraints.c index a74623025f3a..9e51801c4915 100644 --- a/arch/powerpc/kernel/hw_breakpoint_constraints.c +++ b/arch/powerpc/kernel/hw_breakpoint_constraints.c @@ -131,8 +131,13 @@ void wp_get_instr_detail(struct pt_regs *regs, ppc_inst_t *instr, int *type, int *size, unsigned long *ea) { struct instruction_op op; + int err; - if (__get_user_instr(*instr, (void __user *)regs->nip)) + pagefault_disable(); + err = __get_user_instr(*instr, (void __user *)regs->nip); + pagefault_enable(); + + if (err) return; analyse_instr(&op, regs, *instr); diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c index eeff136b83d9..64ff37721fd0 100644 --- a/arch/powerpc/kernel/traps.c +++ b/arch/powerpc/kernel/traps.c @@ -1512,23 +1512,11 @@ static void do_program_check(struct pt_regs *regs) return; } - if (cpu_has_feature(CPU_FTR_DEXCR_NPHIE) && user_mode(regs)) { - ppc_inst_t insn; - - if (get_user_instr(insn, (void __user *)regs->nip)) { - _exception(SIGSEGV, regs, SEGV_MAPERR, regs->nip); - return; - } - - if (ppc_inst_primary_opcode(insn) == 31 && - get_xop(ppc_inst_val(insn)) == OP_31_XOP_HASHCHK) { - _exception(SIGILL, regs, ILL_ILLOPN, regs->nip); - return; - } + /* User mode considers other cases after enabling IRQs */ + if (!user_mode(regs)) { + _exception(SIGTRAP, regs, TRAP_BRKPT, regs->nip); + return; } - - _exception(SIGTRAP, regs, TRAP_BRKPT, regs->nip); - return; } #ifdef CONFIG_PPC_TRANSACTIONAL_MEM if (reason & REASON_TM) { @@ -1561,16 +1549,44 @@ static void do_program_check(struct pt_regs *regs) /* * If we took the program check in the kernel skip down to sending a - * SIGILL. The subsequent cases all relate to emulating instructions - * which we should only do for userspace. We also do not want to enable - * interrupts for kernel faults because that might lead to further - * faults, and loose the context of the original exception. + * SIGILL. The subsequent cases all relate to user space, such as + * emulating instructions which we should only do for user space. We + * also do not want to enable interrupts for kernel faults because that + * might lead to further faults, and loose the context of the original + * exception. */ if (!user_mode(regs)) goto sigill; interrupt_cond_local_irq_enable(regs); + /* + * (reason & REASON_TRAP) is mostly handled before enabling IRQs, + * except get_user_instr() can sleep so we cannot reliably inspect the + * current instruction in that context. Now that we know we are + * handling a user space trap and can sleep, we can check if the trap + * was a hashchk failure. + */ + if (reason & REASON_TRAP) { + if (cpu_has_feature(CPU_FTR_DEXCR_NPHIE)) { + ppc_inst_t insn; + + if (get_user_instr(insn, (void __user *)regs->nip)) { + _exception(SIGSEGV, regs, SEGV_MAPERR, regs->nip); + return; + } + + if (ppc_inst_primary_opcode(insn) == 31 && + get_xop(ppc_inst_val(insn)) == OP_31_XOP_HASHCHK) { + _exception(SIGILL, regs, ILL_ILLOPN, regs->nip); + return; + } + } + + _exception(SIGTRAP, regs, TRAP_BRKPT, regs->nip); + return; + } + /* (reason & REASON_ILLEGAL) would be the obvious thing here, * but there seems to be a hardware bug on the 405GP (RevD) * that means ESR is sometimes set incorrectly - either to diff --git a/arch/powerpc/perf/hv-24x7.c b/arch/powerpc/perf/hv-24x7.c index 317175791d23..3449be7c0d51 100644 --- a/arch/powerpc/perf/hv-24x7.c +++ b/arch/powerpc/perf/hv-24x7.c @@ -1418,7 +1418,7 @@ static int h_24x7_event_init(struct perf_event *event) } domain = event_get_domain(event); - if (domain >= HV_PERF_DOMAIN_MAX) { + if (domain == 0 || domain >= HV_PERF_DOMAIN_MAX) { pr_devel("invalid domain %d\n", domain); return -EINVAL; } diff --git a/arch/powerpc/platforms/82xx/Kconfig b/arch/powerpc/platforms/82xx/Kconfig index d9f1a2a83158..1824536cf6f2 100644 --- a/arch/powerpc/platforms/82xx/Kconfig +++ b/arch/powerpc/platforms/82xx/Kconfig @@ -2,6 +2,7 @@ menuconfig PPC_82xx bool "82xx-based boards (PQ II)" depends on PPC_BOOK3S_32 + select FSL_SOC if PPC_82xx @@ -9,7 +10,6 @@ config EP8248E bool "Embedded Planet EP8248E (a.k.a. CWH-PPC-8248N-VE)" select CPM2 select PPC_INDIRECT_PCI if PCI - select FSL_SOC select PHYLIB if NETDEVICES select MDIO_BITBANG if PHYLIB help @@ -22,7 +22,6 @@ config MGCOGE bool "Keymile MGCOGE" select CPM2 select PPC_INDIRECT_PCI if PCI - select FSL_SOC help This enables support for the Keymile MGCOGE board. diff --git a/arch/riscv/include/asm/errata_list.h b/arch/riscv/include/asm/errata_list.h index e2ecd01bfac7..b55b434f0059 100644 --- a/arch/riscv/include/asm/errata_list.h +++ b/arch/riscv/include/asm/errata_list.h @@ -105,7 +105,7 @@ asm volatile(ALTERNATIVE( \ * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 | * 0000001 01001 rs1 000 00000 0001011 * dcache.cva rs1 (clean, virtual address) - * 0000001 00100 rs1 000 00000 0001011 + * 0000001 00101 rs1 000 00000 0001011 * * dcache.cipa rs1 (clean then invalidate, physical address) * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 | @@ -118,7 +118,7 @@ asm volatile(ALTERNATIVE( \ * 0000000 11001 00000 000 00000 0001011 */ #define THEAD_inval_A0 ".long 0x0265000b" -#define THEAD_clean_A0 ".long 0x0245000b" +#define THEAD_clean_A0 ".long 0x0255000b" #define THEAD_flush_A0 ".long 0x0275000b" #define THEAD_SYNC_S ".long 0x0190000b" diff --git a/arch/riscv/kernel/elf_kexec.c b/arch/riscv/kernel/elf_kexec.c index f4099059ed8f..e60fbd8660c4 100644 --- a/arch/riscv/kernel/elf_kexec.c +++ b/arch/riscv/kernel/elf_kexec.c @@ -98,7 +98,13 @@ static int elf_find_pbase(struct kimage *image, unsigned long kernel_len, kbuf.image = image; kbuf.buf_min = lowest_paddr; kbuf.buf_max = ULONG_MAX; - kbuf.buf_align = PAGE_SIZE; + + /* + * Current riscv boot protocol requires 2MB alignment for + * RV64 and 4MB alignment for RV32 + * + */ + kbuf.buf_align = PMD_SIZE; kbuf.mem = KEXEC_BUF_MEM_UNKNOWN; kbuf.memsz = ALIGN(kernel_len, PAGE_SIZE); kbuf.top_down = false; diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 982b777eadc7..66bfabae8814 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -1945,6 +1945,7 @@ config EFI select UCS2_STRING select EFI_RUNTIME_WRAPPERS select ARCH_USE_MEMREMAP_PROT + select EFI_RUNTIME_MAP if KEXEC_CORE help This enables the kernel to use EFI runtime services that are available (such as the EFI variable services). @@ -2020,7 +2021,6 @@ config EFI_MAX_FAKE_MEM config EFI_RUNTIME_MAP bool "Export EFI runtime maps to sysfs" if EXPERT depends on EFI - default KEXEC_CORE help Export EFI runtime memory regions to /sys/firmware/efi/runtime-map. That memory map is required by the 2nd kernel to set up EFI virtual diff --git a/arch/x86/boot/compressed/ident_map_64.c b/arch/x86/boot/compressed/ident_map_64.c index bcc956c17872..08f93b0401bb 100644 --- a/arch/x86/boot/compressed/ident_map_64.c +++ b/arch/x86/boot/compressed/ident_map_64.c @@ -59,6 +59,14 @@ static void *alloc_pgt_page(void *context) return NULL; } + /* Consumed more tables than expected? */ + if (pages->pgt_buf_offset == BOOT_PGT_SIZE_WARN) { + debug_putstr("pgt_buf running low in " __FILE__ "\n"); + debug_putstr("Need to raise BOOT_PGT_SIZE?\n"); + debug_putaddr(pages->pgt_buf_offset); + debug_putaddr(pages->pgt_buf_size); + } + entry = pages->pgt_buf + pages->pgt_buf_offset; pages->pgt_buf_offset += PAGE_SIZE; diff --git a/arch/x86/entry/common.c b/arch/x86/entry/common.c index 6c2826417b33..93c60c0c9d4a 100644 --- a/arch/x86/entry/common.c +++ b/arch/x86/entry/common.c @@ -294,7 +294,7 @@ static void __xen_pv_evtchn_do_upcall(struct pt_regs *regs) inc_irq_stat(irq_hv_callback_count); - xen_hvm_evtchn_do_upcall(); + xen_evtchn_do_upcall(); set_irq_regs(old_regs); } diff --git a/arch/x86/include/asm/boot.h b/arch/x86/include/asm/boot.h index 4ae14339cb8c..b3a7cfb0d99e 100644 --- a/arch/x86/include/asm/boot.h +++ b/arch/x86/include/asm/boot.h @@ -40,23 +40,40 @@ #ifdef CONFIG_X86_64 # define BOOT_STACK_SIZE 0x4000 +/* + * Used by decompressor's startup_32() to allocate page tables for identity + * mapping of the 4G of RAM in 4-level paging mode: + * - 1 level4 table; + * - 1 level3 table; + * - 4 level2 table that maps everything with 2M pages; + * + * The additional level5 table needed for 5-level paging is allocated from + * trampoline_32bit memory. + */ # define BOOT_INIT_PGT_SIZE (6*4096) -# ifdef CONFIG_RANDOMIZE_BASE + /* - * Assuming all cross the 512GB boundary: - * 1 page for level4 - * (2+2)*4 pages for kernel, param, cmd_line, and randomized kernel - * 2 pages for first 2M (video RAM: CONFIG_X86_VERBOSE_BOOTUP). - * Total is 19 pages. + * Total number of page tables kernel_add_identity_map() can allocate, + * including page tables consumed by startup_32(). + * + * Worst-case scenario: + * - 5-level paging needs 1 level5 table; + * - KASLR needs to map kernel, boot_params, cmdline and randomized kernel, + * assuming all of them cross 256T boundary: + * + 4*2 level4 table; + * + 4*2 level3 table; + * + 4*2 level2 table; + * - X86_VERBOSE_BOOTUP needs to map the first 2M (video RAM): + * + 1 level4 table; + * + 1 level3 table; + * + 1 level2 table; + * Total: 28 tables + * + * Add 4 spare table in case decompressor touches anything beyond what is + * accounted above. Warn if it happens. */ -# ifdef CONFIG_X86_VERBOSE_BOOTUP -# define BOOT_PGT_SIZE (19*4096) -# else /* !CONFIG_X86_VERBOSE_BOOTUP */ -# define BOOT_PGT_SIZE (17*4096) -# endif -# else /* !CONFIG_RANDOMIZE_BASE */ -# define BOOT_PGT_SIZE BOOT_INIT_PGT_SIZE -# endif +# define BOOT_PGT_SIZE_WARN (28*4096) +# define BOOT_PGT_SIZE (32*4096) #else /* !CONFIG_X86_64 */ # define BOOT_STACK_SIZE 0x1000 diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h index b0994ae3bc23..c4555b269a1b 100644 --- a/arch/x86/include/asm/efi.h +++ b/arch/x86/include/asm/efi.h @@ -91,19 +91,6 @@ static inline void efi_fpu_end(void) #ifdef CONFIG_X86_32 #define EFI_X86_KERNEL_ALLOC_LIMIT (SZ_512M - 1) - -#define arch_efi_call_virt_setup() \ -({ \ - efi_fpu_begin(); \ - firmware_restrict_branch_speculation_start(); \ -}) - -#define arch_efi_call_virt_teardown() \ -({ \ - firmware_restrict_branch_speculation_end(); \ - efi_fpu_end(); \ -}) - #else /* !CONFIG_X86_32 */ #define EFI_X86_KERNEL_ALLOC_LIMIT EFI_ALLOC_LIMIT @@ -116,14 +103,6 @@ extern bool efi_disable_ibt_for_runtime; __efi_call(__VA_ARGS__); \ }) -#define arch_efi_call_virt_setup() \ -({ \ - efi_sync_low_kernel_mappings(); \ - efi_fpu_begin(); \ - firmware_restrict_branch_speculation_start(); \ - efi_enter_mm(); \ -}) - #undef arch_efi_call_virt #define arch_efi_call_virt(p, f, args...) ({ \ u64 ret, ibt = ibt_save(efi_disable_ibt_for_runtime); \ @@ -132,13 +111,6 @@ extern bool efi_disable_ibt_for_runtime; ret; \ }) -#define arch_efi_call_virt_teardown() \ -({ \ - efi_leave_mm(); \ - firmware_restrict_branch_speculation_end(); \ - efi_fpu_end(); \ -}) - #ifdef CONFIG_KASAN /* * CONFIG_KASAN may redefine memset to __memset. __memset function is present @@ -168,8 +140,8 @@ extern void efi_delete_dummy_variable(void); extern void efi_crash_gracefully_on_page_fault(unsigned long phys_addr); extern void efi_free_boot_services(void); -void efi_enter_mm(void); -void efi_leave_mm(void); +void arch_efi_call_virt_setup(void); +void arch_efi_call_virt_teardown(void); /* kexec external ABI */ struct efi_setup_data { diff --git a/arch/x86/include/asm/linkage.h b/arch/x86/include/asm/linkage.h index 97a3de7892d3..5ff49fd67732 100644 --- a/arch/x86/include/asm/linkage.h +++ b/arch/x86/include/asm/linkage.h @@ -8,6 +8,14 @@ #undef notrace #define notrace __attribute__((no_instrument_function)) +#ifdef CONFIG_64BIT +/* + * The generic version tends to create spurious ENDBR instructions under + * certain conditions. + */ +#define _THIS_IP_ ({ unsigned long __here; asm ("lea 0(%%rip), %0" : "=r" (__here)); __here; }) +#endif + #ifdef CONFIG_X86_32 #define asmlinkage CPP_ASMLINKAGE __attribute__((regparm(0))) #endif /* CONFIG_X86_32 */ diff --git a/arch/x86/include/asm/paravirt_types.h b/arch/x86/include/asm/paravirt_types.h index 4acbcddddc29..772d03487520 100644 --- a/arch/x86/include/asm/paravirt_types.h +++ b/arch/x86/include/asm/paravirt_types.h @@ -9,13 +9,6 @@ struct paravirt_patch_site { u8 type; /* type of this instruction */ u8 len; /* length of original instruction */ }; - -/* Lazy mode for batching updates / context switch */ -enum paravirt_lazy_mode { - PARAVIRT_LAZY_NONE, - PARAVIRT_LAZY_MMU, - PARAVIRT_LAZY_CPU, -}; #endif #ifdef CONFIG_PARAVIRT @@ -549,14 +542,6 @@ int paravirt_disable_iospace(void); __PVOP_VCALL(op, PVOP_CALL_ARG1(arg1), PVOP_CALL_ARG2(arg2), \ PVOP_CALL_ARG3(arg3), PVOP_CALL_ARG4(arg4)) -enum paravirt_lazy_mode paravirt_get_lazy_mode(void); -void paravirt_start_context_switch(struct task_struct *prev); -void paravirt_end_context_switch(struct task_struct *next); - -void paravirt_enter_lazy_mmu(void); -void paravirt_leave_lazy_mmu(void); -void paravirt_flush_lazy_mmu(void); - void _paravirt_nop(void); void paravirt_BUG(void); unsigned long paravirt_ret0(void); diff --git a/arch/x86/include/asm/xen/hypervisor.h b/arch/x86/include/asm/xen/hypervisor.h index 5fc35f889cd1..7048dfacc04b 100644 --- a/arch/x86/include/asm/xen/hypervisor.h +++ b/arch/x86/include/asm/xen/hypervisor.h @@ -36,6 +36,7 @@ extern struct shared_info *HYPERVISOR_shared_info; extern struct start_info *xen_start_info; +#include <asm/bug.h> #include <asm/processor.h> #define XEN_SIGNATURE "XenVMMXenVMM" @@ -63,4 +64,40 @@ void __init xen_pvh_init(struct boot_params *boot_params); void __init mem_map_via_hcall(struct boot_params *boot_params_p); #endif +/* Lazy mode for batching updates / context switch */ +enum xen_lazy_mode { + XEN_LAZY_NONE, + XEN_LAZY_MMU, + XEN_LAZY_CPU, +}; + +DECLARE_PER_CPU(enum xen_lazy_mode, xen_lazy_mode); +DECLARE_PER_CPU(unsigned int, xen_lazy_nesting); + +static inline void enter_lazy(enum xen_lazy_mode mode) +{ + enum xen_lazy_mode old_mode = this_cpu_read(xen_lazy_mode); + + if (mode == old_mode) { + this_cpu_inc(xen_lazy_nesting); + return; + } + + BUG_ON(old_mode != XEN_LAZY_NONE); + + this_cpu_write(xen_lazy_mode, mode); +} + +static inline void leave_lazy(enum xen_lazy_mode mode) +{ + BUG_ON(this_cpu_read(xen_lazy_mode) != mode); + + if (this_cpu_read(xen_lazy_nesting) == 0) + this_cpu_write(xen_lazy_mode, XEN_LAZY_NONE); + else + this_cpu_dec(xen_lazy_nesting); +} + +enum xen_lazy_mode xen_get_lazy_mode(void); + #endif /* _ASM_X86_XEN_HYPERVISOR_H */ diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c index d9f5d7492f83..205cee567629 100644 --- a/arch/x86/kernel/apic/x2apic_uv_x.c +++ b/arch/x86/kernel/apic/x2apic_uv_x.c @@ -1533,7 +1533,7 @@ static void __init build_socket_tables(void) { struct uv_gam_range_entry *gre = uv_gre_table; int nums, numn, nump; - int cpu, i, lnid; + int i, lnid, apicid; int minsock = _min_socket; int maxsock = _max_socket; int minpnode = _min_pnode; @@ -1584,15 +1584,14 @@ static void __init build_socket_tables(void) /* Set socket -> node values: */ lnid = NUMA_NO_NODE; - for_each_possible_cpu(cpu) { - int nid = cpu_to_node(cpu); - int apicid, sockid; + for (apicid = 0; apicid < ARRAY_SIZE(__apicid_to_node); apicid++) { + int nid = __apicid_to_node[apicid]; + int sockid; - if (lnid == nid) + if ((nid == NUMA_NO_NODE) || (lnid == nid)) continue; lnid = nid; - apicid = per_cpu(x86_cpu_to_apicid, cpu); sockid = apicid >> uv_cpuid.socketid_shift; if (_socket_to_node[sockid - minsock] == SOCK_EMPTY) diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index 975f98d5eee5..97f1436c1a20 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -143,66 +143,7 @@ int paravirt_disable_iospace(void) return request_resource(&ioport_resource, &reserve_ioports); } -static DEFINE_PER_CPU(enum paravirt_lazy_mode, paravirt_lazy_mode) = PARAVIRT_LAZY_NONE; - -static inline void enter_lazy(enum paravirt_lazy_mode mode) -{ - BUG_ON(this_cpu_read(paravirt_lazy_mode) != PARAVIRT_LAZY_NONE); - - this_cpu_write(paravirt_lazy_mode, mode); -} - -static void leave_lazy(enum paravirt_lazy_mode mode) -{ - BUG_ON(this_cpu_read(paravirt_lazy_mode) != mode); - - this_cpu_write(paravirt_lazy_mode, PARAVIRT_LAZY_NONE); -} - -void paravirt_enter_lazy_mmu(void) -{ - enter_lazy(PARAVIRT_LAZY_MMU); -} - -void paravirt_leave_lazy_mmu(void) -{ - leave_lazy(PARAVIRT_LAZY_MMU); -} - -void paravirt_flush_lazy_mmu(void) -{ - preempt_disable(); - - if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU) { - arch_leave_lazy_mmu_mode(); - arch_enter_lazy_mmu_mode(); - } - - preempt_enable(); -} - #ifdef CONFIG_PARAVIRT_XXL -void paravirt_start_context_switch(struct task_struct *prev) -{ - BUG_ON(preemptible()); - - if (this_cpu_read(paravirt_lazy_mode) == PARAVIRT_LAZY_MMU) { - arch_leave_lazy_mmu_mode(); - set_ti_thread_flag(task_thread_info(prev), TIF_LAZY_MMU_UPDATES); - } - enter_lazy(PARAVIRT_LAZY_CPU); -} - -void paravirt_end_context_switch(struct task_struct *next) -{ - BUG_ON(preemptible()); - - leave_lazy(PARAVIRT_LAZY_CPU); - - if (test_and_clear_ti_thread_flag(task_thread_info(next), TIF_LAZY_MMU_UPDATES)) - arch_enter_lazy_mmu_mode(); -} - static noinstr void pv_native_write_cr2(unsigned long val) { native_write_cr2(val); @@ -229,14 +170,6 @@ static noinstr void pv_native_safe_halt(void) } #endif -enum paravirt_lazy_mode paravirt_get_lazy_mode(void) -{ - if (in_interrupt()) - return PARAVIRT_LAZY_NONE; - - return this_cpu_read(paravirt_lazy_mode); -} - struct pv_info pv_info = { .name = "bare hardware", #ifdef CONFIG_PARAVIRT_XXL diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 4e45ff44aa07..48e040618731 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -579,7 +579,6 @@ static bool match_llc(struct cpuinfo_x86 *c, struct cpuinfo_x86 *o) } -#if defined(CONFIG_SCHED_SMT) || defined(CONFIG_SCHED_CLUSTER) || defined(CONFIG_SCHED_MC) static inline int x86_sched_itmt_flags(void) { return sysctl_sched_itmt_enabled ? SD_ASYM_PACKING : 0; @@ -603,7 +602,14 @@ static int x86_cluster_flags(void) return cpu_cluster_flags() | x86_sched_itmt_flags(); } #endif -#endif + +static int x86_die_flags(void) +{ + if (cpu_feature_enabled(X86_FEATURE_HYBRID_CPU)) + return x86_sched_itmt_flags(); + + return 0; +} /* * Set if a package/die has multiple NUMA nodes inside. @@ -640,7 +646,7 @@ static void __init build_sched_topology(void) */ if (!x86_has_numa_in_package) { x86_topology[i++] = (struct sched_domain_topology_level){ - cpu_cpu_mask, SD_INIT_NAME(DIE) + cpu_cpu_mask, x86_die_flags, SD_INIT_NAME(DIE) }; } diff --git a/arch/x86/lib/putuser.S b/arch/x86/lib/putuser.S index 1451e0c4ae22..235bbda6fc82 100644 --- a/arch/x86/lib/putuser.S +++ b/arch/x86/lib/putuser.S @@ -56,7 +56,6 @@ SYM_FUNC_END(__put_user_1) EXPORT_SYMBOL(__put_user_1) SYM_FUNC_START(__put_user_nocheck_1) - ENDBR ASM_STAC 2: movb %al,(%_ASM_CX) xor %ecx,%ecx @@ -76,7 +75,6 @@ SYM_FUNC_END(__put_user_2) EXPORT_SYMBOL(__put_user_2) SYM_FUNC_START(__put_user_nocheck_2) - ENDBR ASM_STAC 4: movw %ax,(%_ASM_CX) xor %ecx,%ecx @@ -96,7 +94,6 @@ SYM_FUNC_END(__put_user_4) EXPORT_SYMBOL(__put_user_4) SYM_FUNC_START(__put_user_nocheck_4) - ENDBR ASM_STAC 6: movl %eax,(%_ASM_CX) xor %ecx,%ecx @@ -119,7 +116,6 @@ SYM_FUNC_END(__put_user_8) EXPORT_SYMBOL(__put_user_8) SYM_FUNC_START(__put_user_nocheck_8) - ENDBR ASM_STAC 9: mov %_ASM_AX,(%_ASM_CX) #ifdef CONFIG_X86_32 diff --git a/arch/x86/platform/efi/efi_32.c b/arch/x86/platform/efi/efi_32.c index e06a199423c0..b2cc7b4552a1 100644 --- a/arch/x86/platform/efi/efi_32.c +++ b/arch/x86/platform/efi/efi_32.c @@ -140,3 +140,15 @@ void __init efi_runtime_update_mappings(void) } } } + +void arch_efi_call_virt_setup(void) +{ + efi_fpu_begin(); + firmware_restrict_branch_speculation_start(); +} + +void arch_efi_call_virt_teardown(void) +{ + firmware_restrict_branch_speculation_end(); + efi_fpu_end(); +} diff --git a/arch/x86/platform/efi/efi_64.c b/arch/x86/platform/efi/efi_64.c index 77f7ac3668cb..91d31ac422d6 100644 --- a/arch/x86/platform/efi/efi_64.c +++ b/arch/x86/platform/efi/efi_64.c @@ -474,19 +474,34 @@ void __init efi_dump_pagetable(void) * can not change under us. * It should be ensured that there are no concurrent calls to this function. */ -void efi_enter_mm(void) +static void efi_enter_mm(void) { efi_prev_mm = current->active_mm; current->active_mm = &efi_mm; switch_mm(efi_prev_mm, &efi_mm, NULL); } -void efi_leave_mm(void) +static void efi_leave_mm(void) { current->active_mm = efi_prev_mm; switch_mm(&efi_mm, efi_prev_mm, NULL); } +void arch_efi_call_virt_setup(void) +{ + efi_sync_low_kernel_mappings(); + efi_fpu_begin(); + firmware_restrict_branch_speculation_start(); + efi_enter_mm(); +} + +void arch_efi_call_virt_teardown(void) +{ + efi_leave_mm(); + firmware_restrict_branch_speculation_end(); + efi_fpu_end(); +} + static DEFINE_SPINLOCK(efi_runtime_lock); /* diff --git a/arch/x86/purgatory/Makefile b/arch/x86/purgatory/Makefile index c2a29be35c01..08aa0f25f12a 100644 --- a/arch/x86/purgatory/Makefile +++ b/arch/x86/purgatory/Makefile @@ -19,6 +19,10 @@ CFLAGS_sha256.o := -D__DISABLE_EXPORTS -D__NO_FORTIFY # optimization flags. KBUILD_CFLAGS := $(filter-out -fprofile-sample-use=% -fprofile-use=%,$(KBUILD_CFLAGS)) +# When LTO is enabled, llvm emits many text sections, which is not supported +# by kexec. Remove -flto=* flags. +KBUILD_CFLAGS := $(filter-out $(CC_FLAGS_LTO),$(KBUILD_CFLAGS)) + # When linking purgatory.ro with -r unresolved symbols are not checked, # also link a purgatory.chk binary without -r to check for unresolved symbols. PURGATORY_LDFLAGS := -e purgatory_start -z nodefaultlib diff --git a/arch/x86/xen/efi.c b/arch/x86/xen/efi.c index 863d0d6b3edc..7250d0e0e1a9 100644 --- a/arch/x86/xen/efi.c +++ b/arch/x86/xen/efi.c @@ -138,7 +138,7 @@ void __init xen_efi_init(struct boot_params *boot_params) if (efi_systab_xen == NULL) return; - strncpy((char *)&boot_params->efi_info.efi_loader_signature, "Xen", + strscpy((char *)&boot_params->efi_info.efi_loader_signature, "Xen", sizeof(boot_params->efi_info.efi_loader_signature)); boot_params->efi_info.efi_systab = (__u32)__pa(efi_systab_xen); boot_params->efi_info.efi_systab_hi = (__u32)(__pa(efi_systab_xen) >> 32); diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index b8db2148c07d..0337392a3121 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -32,7 +32,7 @@ EXPORT_SYMBOL_GPL(hypercall_page); * &HYPERVISOR_shared_info->vcpu_info[cpu]. See xen_hvm_init_shared_info * and xen_vcpu_setup for details. By default it points to share_info->vcpu_info * but during boot it is switched to point to xen_vcpu_info. - * The pointer is used in __xen_evtchn_do_upcall to acknowledge pending events. + * The pointer is used in xen_evtchn_do_upcall to acknowledge pending events. */ DEFINE_PER_CPU(struct vcpu_info *, xen_vcpu); DEFINE_PER_CPU(struct vcpu_info, xen_vcpu_info); diff --git a/arch/x86/xen/enlighten_hvm.c b/arch/x86/xen/enlighten_hvm.c index 9a192f51f1b0..3f8c34707c50 100644 --- a/arch/x86/xen/enlighten_hvm.c +++ b/arch/x86/xen/enlighten_hvm.c @@ -136,7 +136,7 @@ DEFINE_IDTENTRY_SYSVEC(sysvec_xen_hvm_callback) inc_irq_stat(irq_hv_callback_count); - xen_hvm_evtchn_do_upcall(); + xen_evtchn_do_upcall(); set_irq_regs(old_regs); } diff --git a/arch/x86/xen/enlighten_pv.c b/arch/x86/xen/enlighten_pv.c index 49352fad7d1d..bbbfdd495ebd 100644 --- a/arch/x86/xen/enlighten_pv.c +++ b/arch/x86/xen/enlighten_pv.c @@ -101,6 +101,17 @@ struct tls_descs { struct desc_struct desc[3]; }; +DEFINE_PER_CPU(enum xen_lazy_mode, xen_lazy_mode) = XEN_LAZY_NONE; +DEFINE_PER_CPU(unsigned int, xen_lazy_nesting); + +enum xen_lazy_mode xen_get_lazy_mode(void) +{ + if (in_interrupt()) + return XEN_LAZY_NONE; + + return this_cpu_read(xen_lazy_mode); +} + /* * Updating the 3 TLS descriptors in the GDT on every task switch is * surprisingly expensive so we avoid updating them if they haven't @@ -362,10 +373,25 @@ static noinstr unsigned long xen_get_debugreg(int reg) return HYPERVISOR_get_debugreg(reg); } +static void xen_start_context_switch(struct task_struct *prev) +{ + BUG_ON(preemptible()); + + if (this_cpu_read(xen_lazy_mode) == XEN_LAZY_MMU) { + arch_leave_lazy_mmu_mode(); + set_ti_thread_flag(task_thread_info(prev), TIF_LAZY_MMU_UPDATES); + } + enter_lazy(XEN_LAZY_CPU); +} + static void xen_end_context_switch(struct task_struct *next) { + BUG_ON(preemptible()); + xen_mc_flush(); - paravirt_end_context_switch(next); + leave_lazy(XEN_LAZY_CPU); + if (test_and_clear_ti_thread_flag(task_thread_info(next), TIF_LAZY_MMU_UPDATES)) + arch_enter_lazy_mmu_mode(); } static unsigned long xen_store_tr(void) @@ -472,7 +498,7 @@ static void xen_set_ldt(const void *addr, unsigned entries) MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF); - xen_mc_issue(PARAVIRT_LAZY_CPU); + xen_mc_issue(XEN_LAZY_CPU); } static void xen_load_gdt(const struct desc_ptr *dtr) @@ -568,7 +594,7 @@ static void xen_load_tls(struct thread_struct *t, unsigned int cpu) * exception between the new %fs descriptor being loaded and * %fs being effectively cleared at __switch_to(). */ - if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_CPU) + if (xen_get_lazy_mode() == XEN_LAZY_CPU) loadsegment(fs, 0); xen_mc_batch(); @@ -577,7 +603,7 @@ static void xen_load_tls(struct thread_struct *t, unsigned int cpu) load_TLS_descriptor(t, cpu, 1); load_TLS_descriptor(t, cpu, 2); - xen_mc_issue(PARAVIRT_LAZY_CPU); + xen_mc_issue(XEN_LAZY_CPU); } static void xen_load_gs_index(unsigned int idx) @@ -909,7 +935,7 @@ static void xen_load_sp0(unsigned long sp0) mcs = xen_mc_entry(0); MULTI_stack_switch(mcs.mc, __KERNEL_DS, sp0); - xen_mc_issue(PARAVIRT_LAZY_CPU); + xen_mc_issue(XEN_LAZY_CPU); this_cpu_write(cpu_tss_rw.x86_tss.sp0, sp0); } @@ -973,7 +999,7 @@ static void xen_write_cr0(unsigned long cr0) MULTI_fpu_taskswitch(mcs.mc, (cr0 & X86_CR0_TS) != 0); - xen_mc_issue(PARAVIRT_LAZY_CPU); + xen_mc_issue(XEN_LAZY_CPU); } static void xen_write_cr4(unsigned long cr4) @@ -1156,7 +1182,7 @@ static const typeof(pv_ops) xen_cpu_ops __initconst = { #endif .io_delay = xen_io_delay, - .start_context_switch = paravirt_start_context_switch, + .start_context_switch = xen_start_context_switch, .end_context_switch = xen_end_context_switch, }, }; diff --git a/arch/x86/xen/mmu_pv.c b/arch/x86/xen/mmu_pv.c index 1652c39e3dfb..b6830554ff69 100644 --- a/arch/x86/xen/mmu_pv.c +++ b/arch/x86/xen/mmu_pv.c @@ -236,7 +236,7 @@ static void xen_set_pmd_hyper(pmd_t *ptr, pmd_t val) u.val = pmd_val_ma(val); xen_extend_mmu_update(&u); - xen_mc_issue(PARAVIRT_LAZY_MMU); + xen_mc_issue(XEN_LAZY_MMU); preempt_enable(); } @@ -270,7 +270,7 @@ static bool xen_batched_set_pte(pte_t *ptep, pte_t pteval) { struct mmu_update u; - if (paravirt_get_lazy_mode() != PARAVIRT_LAZY_MMU) + if (xen_get_lazy_mode() != XEN_LAZY_MMU) return false; xen_mc_batch(); @@ -279,7 +279,7 @@ static bool xen_batched_set_pte(pte_t *ptep, pte_t pteval) u.val = pte_val_ma(pteval); xen_extend_mmu_update(&u); - xen_mc_issue(PARAVIRT_LAZY_MMU); + xen_mc_issue(XEN_LAZY_MMU); return true; } @@ -325,7 +325,7 @@ void xen_ptep_modify_prot_commit(struct vm_area_struct *vma, unsigned long addr, u.val = pte_val_ma(pte); xen_extend_mmu_update(&u); - xen_mc_issue(PARAVIRT_LAZY_MMU); + xen_mc_issue(XEN_LAZY_MMU); } /* Assume pteval_t is equivalent to all the other *val_t types. */ @@ -419,7 +419,7 @@ static void xen_set_pud_hyper(pud_t *ptr, pud_t val) u.val = pud_val_ma(val); xen_extend_mmu_update(&u); - xen_mc_issue(PARAVIRT_LAZY_MMU); + xen_mc_issue(XEN_LAZY_MMU); preempt_enable(); } @@ -499,7 +499,7 @@ static void __init xen_set_p4d_hyper(p4d_t *ptr, p4d_t val) __xen_set_p4d_hyper(ptr, val); - xen_mc_issue(PARAVIRT_LAZY_MMU); + xen_mc_issue(XEN_LAZY_MMU); preempt_enable(); } @@ -531,7 +531,7 @@ static void xen_set_p4d(p4d_t *ptr, p4d_t val) if (user_ptr) __xen_set_p4d_hyper((p4d_t *)user_ptr, val); - xen_mc_issue(PARAVIRT_LAZY_MMU); + xen_mc_issue(XEN_LAZY_MMU); } #if CONFIG_PGTABLE_LEVELS >= 5 @@ -1245,7 +1245,7 @@ static noinline void xen_flush_tlb(void) op->cmd = MMUEXT_TLB_FLUSH_LOCAL; MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF); - xen_mc_issue(PARAVIRT_LAZY_MMU); + xen_mc_issue(XEN_LAZY_MMU); preempt_enable(); } @@ -1265,7 +1265,7 @@ static void xen_flush_tlb_one_user(unsigned long addr) op->arg1.linear_addr = addr & PAGE_MASK; MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF); - xen_mc_issue(PARAVIRT_LAZY_MMU); + xen_mc_issue(XEN_LAZY_MMU); preempt_enable(); } @@ -1302,7 +1302,7 @@ static void xen_flush_tlb_multi(const struct cpumask *cpus, MULTI_mmuext_op(mcs.mc, &args->op, 1, NULL, DOMID_SELF); - xen_mc_issue(PARAVIRT_LAZY_MMU); + xen_mc_issue(XEN_LAZY_MMU); } static unsigned long xen_read_cr3(void) @@ -1361,7 +1361,7 @@ static void xen_write_cr3(unsigned long cr3) else __xen_write_cr3(false, 0); - xen_mc_issue(PARAVIRT_LAZY_CPU); /* interrupts restored */ + xen_mc_issue(XEN_LAZY_CPU); /* interrupts restored */ } /* @@ -1396,7 +1396,7 @@ static void __init xen_write_cr3_init(unsigned long cr3) __xen_write_cr3(true, cr3); - xen_mc_issue(PARAVIRT_LAZY_CPU); /* interrupts restored */ + xen_mc_issue(XEN_LAZY_CPU); /* interrupts restored */ } static int xen_pgd_alloc(struct mm_struct *mm) @@ -1557,7 +1557,7 @@ static inline void xen_alloc_ptpage(struct mm_struct *mm, unsigned long pfn, if (level == PT_PTE && USE_SPLIT_PTE_PTLOCKS && !pinned) __pin_pagetable_pfn(MMUEXT_PIN_L1_TABLE, pfn); - xen_mc_issue(PARAVIRT_LAZY_MMU); + xen_mc_issue(XEN_LAZY_MMU); } } @@ -1587,7 +1587,7 @@ static inline void xen_release_ptpage(unsigned long pfn, unsigned level) __set_pfn_prot(pfn, PAGE_KERNEL); - xen_mc_issue(PARAVIRT_LAZY_MMU); + xen_mc_issue(XEN_LAZY_MMU); ClearPagePinned(page); } @@ -1804,7 +1804,7 @@ void __init xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn) */ xen_mc_batch(); __xen_write_cr3(true, __pa(init_top_pgt)); - xen_mc_issue(PARAVIRT_LAZY_CPU); + xen_mc_issue(XEN_LAZY_CPU); /* We can't that easily rip out L3 and L2, as the Xen pagetables are * set out this way: [L4], [L1], [L2], [L3], [L1], [L1] ... for @@ -2083,6 +2083,23 @@ static void xen_set_fixmap(unsigned idx, phys_addr_t phys, pgprot_t prot) #endif } +static void xen_enter_lazy_mmu(void) +{ + enter_lazy(XEN_LAZY_MMU); +} + +static void xen_flush_lazy_mmu(void) +{ + preempt_disable(); + + if (xen_get_lazy_mode() == XEN_LAZY_MMU) { + arch_leave_lazy_mmu_mode(); + arch_enter_lazy_mmu_mode(); + } + + preempt_enable(); +} + static void __init xen_post_allocator_init(void) { pv_ops.mmu.set_pte = xen_set_pte; @@ -2107,7 +2124,7 @@ static void xen_leave_lazy_mmu(void) { preempt_disable(); xen_mc_flush(); - paravirt_leave_lazy_mmu(); + leave_lazy(XEN_LAZY_MMU); preempt_enable(); } @@ -2166,9 +2183,9 @@ static const typeof(pv_ops) xen_mmu_ops __initconst = { .exit_mmap = xen_exit_mmap, .lazy_mode = { - .enter = paravirt_enter_lazy_mmu, + .enter = xen_enter_lazy_mmu, .leave = xen_leave_lazy_mmu, - .flush = paravirt_flush_lazy_mmu, + .flush = xen_flush_lazy_mmu, }, .set_fixmap = xen_set_fixmap, @@ -2385,7 +2402,7 @@ static noinline void xen_flush_tlb_all(void) op->cmd = MMUEXT_TLB_FLUSH_ALL; MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF); - xen_mc_issue(PARAVIRT_LAZY_MMU); + xen_mc_issue(XEN_LAZY_MMU); preempt_enable(); } diff --git a/arch/x86/xen/multicalls.h b/arch/x86/xen/multicalls.h index 1c51b2c87f30..c3867b585e0d 100644 --- a/arch/x86/xen/multicalls.h +++ b/arch/x86/xen/multicalls.h @@ -26,7 +26,7 @@ static inline void xen_mc_batch(void) /* need to disable interrupts until this entry is complete */ local_irq_save(flags); - trace_xen_mc_batch(paravirt_get_lazy_mode()); + trace_xen_mc_batch(xen_get_lazy_mode()); __this_cpu_write(xen_mc_irq_flags, flags); } @@ -44,7 +44,7 @@ static inline void xen_mc_issue(unsigned mode) { trace_xen_mc_issue(mode); - if ((paravirt_get_lazy_mode() & mode) == 0) + if ((xen_get_lazy_mode() & mode) == 0) xen_mc_flush(); /* restore flags saved in xen_mc_batch */ |