diff options
| author |   José Pekkarinen <jose.pekkarinen@nokia.com> | 2016-04-11 10:41:07 +0300 |
|---|---|---|
| committer | José Pekkarinen <jose.pekkarinen@nokia.com> | 2016-04-13 08:17:18 +0300 |
| commit | e09b41010ba33a20a87472ee821fa407a5b8da36 (patch) | |
| tree | d10dc367189862e7ca5c592f033dc3726e1df4e3 /kernel/arch/x86/kernel/cpu | |
| parent | f93b97fd65072de626c074dbe099a1fff05ce060 (diff) | |
These changes are the raw update to linux-4.4.6-rt14. Kernel sources
are taken from kernel.org, and rt patch from the rt wiki download page.
During the rebasing, the following patch collided:
Force tick interrupt and get rid of softirq magic(I70131fb85).
Collisions have been removed because its logic was found on the
source already.
Change-Id: I7f57a4081d9deaa0d9ccfc41a6c8daccdee3b769
Signed-off-by: José Pekkarinen <jose.pekkarinen@nokia.com>
Diffstat (limited to 'kernel/arch/x86/kernel/cpu')
48 files changed, 4756 insertions, 2504 deletions
diff --git a/kernel/arch/x86/kernel/cpu/Makefile b/kernel/arch/x86/kernel/cpu/Makefile index 9bff68798..58031303e 100644 --- a/kernel/arch/x86/kernel/cpu/Makefile +++ b/kernel/arch/x86/kernel/cpu/Makefile @@ -41,11 +41,14 @@ obj-$(CONFIG_CPU_SUP_INTEL) += perf_event_p6.o perf_event_knc.o perf_event_p4.o obj-$(CONFIG_CPU_SUP_INTEL) += perf_event_intel_lbr.o perf_event_intel_ds.o perf_event_intel.o obj-$(CONFIG_CPU_SUP_INTEL) += perf_event_intel_rapl.o perf_event_intel_cqm.o obj-$(CONFIG_CPU_SUP_INTEL) += perf_event_intel_pt.o perf_event_intel_bts.o +obj-$(CONFIG_CPU_SUP_INTEL) += perf_event_intel_cstate.o obj-$(CONFIG_PERF_EVENTS_INTEL_UNCORE) += perf_event_intel_uncore.o \ perf_event_intel_uncore_snb.o \ perf_event_intel_uncore_snbep.o \ perf_event_intel_uncore_nhmex.o +obj-$(CONFIG_CPU_SUP_INTEL) += perf_event_msr.o +obj-$(CONFIG_CPU_SUP_AMD) += perf_event_msr.o endif diff --git a/kernel/arch/x86/kernel/cpu/amd.c b/kernel/arch/x86/kernel/cpu/amd.c index e4cf63301..a8816b325 100644 --- a/kernel/arch/x86/kernel/cpu/amd.c +++ b/kernel/arch/x86/kernel/cpu/amd.c @@ -11,6 +11,7 @@ #include <asm/cpu.h> #include <asm/smp.h> #include <asm/pci-direct.h> +#include <asm/delay.h> #ifdef CONFIG_X86_64 # include <asm/mmconfig.h> @@ -19,6 +20,13 @@ #include "cpu.h" +/* + * nodes_per_socket: Stores the number of nodes per socket. + * Refer to Fam15h Models 00-0fh BKDG - CPUID Fn8000_001E_ECX + * Node Identifiers[10:8] + */ +static u32 nodes_per_socket = 1; + static inline int rdmsrl_amd_safe(unsigned msr, unsigned long long *p) { u32 gprs[8] = { 0 }; @@ -107,7 +115,7 @@ static void init_amd_k6(struct cpuinfo_x86 *c) const int K6_BUG_LOOP = 1000000; int n; void (*f_vide)(void); - unsigned long d, d2; + u64 d, d2; printk(KERN_INFO "AMD K6 stepping B detected - "); @@ -118,10 +126,10 @@ static void init_amd_k6(struct cpuinfo_x86 *c) n = K6_BUG_LOOP; f_vide = vide; - rdtscl(d); + d = rdtsc(); while (n--) f_vide(); - rdtscl(d2); + d2 = rdtsc(); d = d2-d; if (d > 20*K6_BUG_LOOP) @@ -288,10 +296,10 @@ static int nearby_node(int apicid) * Assumption: Number of cores in each internal node is the same. * (2) AMD processors supporting compute units */ -#ifdef CONFIG_X86_HT +#ifdef CONFIG_SMP static void amd_get_topology(struct cpuinfo_x86 *c) { - u32 nodes, cores_per_cu = 1; + u32 cores_per_cu = 1; u8 node_id; int cpu = smp_processor_id(); @@ -300,7 +308,7 @@ static void amd_get_topology(struct cpuinfo_x86 *c) u32 eax, ebx, ecx, edx; cpuid(0x8000001e, &eax, &ebx, &ecx, &edx); - nodes = ((ecx >> 8) & 7) + 1; + nodes_per_socket = ((ecx >> 8) & 7) + 1; node_id = ecx & 7; /* get compute unit information */ @@ -311,18 +319,18 @@ static void amd_get_topology(struct cpuinfo_x86 *c) u64 value; rdmsrl(MSR_FAM10H_NODE_ID, value); - nodes = ((value >> 3) & 7) + 1; + nodes_per_socket = ((value >> 3) & 7) + 1; node_id = value & 7; } else return; /* fixup multi-node processor information */ - if (nodes > 1) { + if (nodes_per_socket > 1) { u32 cores_per_node; u32 cus_per_node; set_cpu_cap(c, X86_FEATURE_AMD_DCM); - cores_per_node = c->x86_max_cores / nodes; + cores_per_node = c->x86_max_cores / nodes_per_socket; cus_per_node = cores_per_node / cores_per_cu; /* store NodeID, use llc_shared_map to store sibling info */ @@ -341,9 +349,10 @@ static void amd_get_topology(struct cpuinfo_x86 *c) */ static void amd_detect_cmp(struct cpuinfo_x86 *c) { -#ifdef CONFIG_X86_HT +#ifdef CONFIG_SMP unsigned bits; int cpu = smp_processor_id(); + unsigned int socket_id, core_complex_id; bits = c->x86_coreid_bits; /* Low order bits define the core id (index of core in socket) */ @@ -353,6 +362,18 @@ static void amd_detect_cmp(struct cpuinfo_x86 *c) /* use socket ID also for last level cache */ per_cpu(cpu_llc_id, cpu) = c->phys_proc_id; amd_get_topology(c); + + /* + * Fix percpu cpu_llc_id here as LLC topology is different + * for Fam17h systems. + */ + if (c->x86 != 0x17 || !cpuid_edx(0x80000006)) + return; + + socket_id = (c->apicid >> bits) - 1; + core_complex_id = (c->apicid & ((1 << bits) - 1)) >> 3; + + per_cpu(cpu_llc_id, cpu) = (socket_id << 3) | core_complex_id; #endif } @@ -366,6 +387,12 @@ u16 amd_get_nb_id(int cpu) } EXPORT_SYMBOL_GPL(amd_get_nb_id); +u32 amd_get_nodes_per_socket(void) +{ + return nodes_per_socket; +} +EXPORT_SYMBOL_GPL(amd_get_nodes_per_socket); + static void srat_detect_node(struct cpuinfo_x86 *c) { #ifdef CONFIG_NUMA @@ -420,7 +447,7 @@ static void srat_detect_node(struct cpuinfo_x86 *c) static void early_init_amd_mc(struct cpuinfo_x86 *c) { -#ifdef CONFIG_X86_HT +#ifdef CONFIG_SMP unsigned bits, ecx; /* Multi core CPU? */ @@ -493,6 +520,9 @@ static void bsp_init_amd(struct cpuinfo_x86 *c) /* A random value per boot for bit slice [12:upper_bit) */ va_align.bits = get_random_int() & va_align.mask; } + + if (cpu_has(c, X86_FEATURE_MWAITX)) + use_mwaitx_delay(); } static void early_init_amd(struct cpuinfo_x86 *c) @@ -520,8 +550,16 @@ static void early_init_amd(struct cpuinfo_x86 *c) set_cpu_cap(c, X86_FEATURE_K6_MTRR); #endif #if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_PCI) - /* check CPU config space for extended APIC ID */ - if (cpu_has_apic && c->x86 >= 0xf) { + /* + * ApicID can always be treated as an 8-bit value for AMD APIC versions + * >= 0x10, but even old K8s came out of reset with version 0x10. So, we + * can safely set X86_FEATURE_EXTD_APICID unconditionally for families + * after 16h. + */ + if (cpu_has_apic && c->x86 > 0x16) { + set_cpu_cap(c, X86_FEATURE_EXTD_APICID); + } else if (cpu_has_apic && c->x86 >= 0xf) { + /* check CPU config space for extended APIC ID */ unsigned int val; val = read_pci_config(0, 24, 0, 0x68); if ((val & ((1 << 17) | (1 << 18))) == ((1 << 17) | (1 << 18))) diff --git a/kernel/arch/x86/kernel/cpu/bugs.c b/kernel/arch/x86/kernel/cpu/bugs.c index 03445346e..bd17db15a 100644 --- a/kernel/arch/x86/kernel/cpu/bugs.c +++ b/kernel/arch/x86/kernel/cpu/bugs.c @@ -12,57 +12,11 @@ #include <asm/bugs.h> #include <asm/processor.h> #include <asm/processor-flags.h> -#include <asm/i387.h> +#include <asm/fpu/internal.h> #include <asm/msr.h> #include <asm/paravirt.h> #include <asm/alternative.h> -static double __initdata x = 4195835.0; -static double __initdata y = 3145727.0; - -/* - * This used to check for exceptions.. - * However, it turns out that to support that, - * the XMM trap handlers basically had to - * be buggy. So let's have a correct XMM trap - * handler, and forget about printing out - * some status at boot. - * - * We should really only care about bugs here - * anyway. Not features. - */ -static void __init check_fpu(void) -{ - s32 fdiv_bug; - - kernel_fpu_begin(); - - /* - * trap_init() enabled FXSR and company _before_ testing for FP - * problems here. - * - * Test for the divl bug: http://en.wikipedia.org/wiki/Fdiv_bug - */ - __asm__("fninit\n\t" - "fldl %1\n\t" - "fdivl %2\n\t" - "fmull %2\n\t" - "fldl %1\n\t" - "fsubp %%st,%%st(1)\n\t" - "fistpl %0\n\t" - "fwait\n\t" - "fninit" - : "=m" (*&fdiv_bug) - : "m" (*&x), "m" (*&y)); - - kernel_fpu_end(); - - if (fdiv_bug) { - set_cpu_bug(&boot_cpu_data, X86_BUG_FDIV); - pr_warn("Hmm, FPU with FDIV bug\n"); - } -} - void __init check_bugs(void) { identify_boot_cpu(); @@ -85,10 +39,5 @@ void __init check_bugs(void) '0' + (boot_cpu_data.x86 > 6 ? 6 : boot_cpu_data.x86); alternative_instructions(); - /* - * kernel_fpu_begin/end() in check_fpu() relies on the patched - * alternative instructions. - */ - if (cpu_has_fpu) - check_fpu(); + fpu__init_check_bugs(); } diff --git a/kernel/arch/x86/kernel/cpu/common.c b/kernel/arch/x86/kernel/cpu/common.c index 205e0f3df..c2b7522cb 100644 --- a/kernel/arch/x86/kernel/cpu/common.c +++ b/kernel/arch/x86/kernel/cpu/common.c @@ -5,6 +5,7 @@ #include <linux/module.h> #include <linux/percpu.h> #include <linux/string.h> +#include <linux/ctype.h> #include <linux/delay.h> #include <linux/sched.h> #include <linux/init.h> @@ -12,6 +13,7 @@ #include <linux/kgdb.h> #include <linux/smp.h> #include <linux/io.h> +#include <linux/syscore_ops.h> #include <asm/stackprotector.h> #include <asm/perf_event.h> @@ -31,8 +33,7 @@ #include <asm/setup.h> #include <asm/apic.h> #include <asm/desc.h> -#include <asm/i387.h> -#include <asm/fpu-internal.h> +#include <asm/fpu/internal.h> #include <asm/mtrr.h> #include <linux/numa.h> #include <asm/asm.h> @@ -145,32 +146,21 @@ DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = { } }; EXPORT_PER_CPU_SYMBOL_GPL(gdt_page); -static int __init x86_xsave_setup(char *s) +static int __init x86_mpx_setup(char *s) { + /* require an exact match without trailing characters */ if (strlen(s)) return 0; - setup_clear_cpu_cap(X86_FEATURE_XSAVE); - setup_clear_cpu_cap(X86_FEATURE_XSAVEOPT); - setup_clear_cpu_cap(X86_FEATURE_XSAVES); - setup_clear_cpu_cap(X86_FEATURE_AVX); - setup_clear_cpu_cap(X86_FEATURE_AVX2); - return 1; -} -__setup("noxsave", x86_xsave_setup); -static int __init x86_xsaveopt_setup(char *s) -{ - setup_clear_cpu_cap(X86_FEATURE_XSAVEOPT); - return 1; -} -__setup("noxsaveopt", x86_xsaveopt_setup); + /* do not emit a message if the feature is not present */ + if (!boot_cpu_has(X86_FEATURE_MPX)) + return 1; -static int __init x86_xsaves_setup(char *s) -{ - setup_clear_cpu_cap(X86_FEATURE_XSAVES); + setup_clear_cpu_cap(X86_FEATURE_MPX); + pr_info("nompx: Intel Memory Protection Extensions (MPX) disabled\n"); return 1; } -__setup("noxsaves", x86_xsaves_setup); +__setup("nompx", x86_mpx_setup); #ifdef CONFIG_X86_32 static int cachesize_override = -1; @@ -183,14 +173,6 @@ static int __init cachesize_setup(char *str) } __setup("cachesize=", cachesize_setup); -static int __init x86_fxsr_setup(char *s) -{ - setup_clear_cpu_cap(X86_FEATURE_FXSR); - setup_clear_cpu_cap(X86_FEATURE_XMM); - return 1; -} -__setup("nofxsr", x86_fxsr_setup); - static int __init x86_sep_setup(char *s) { setup_clear_cpu_cap(X86_FEATURE_SEP); @@ -291,10 +273,9 @@ __setup("nosmap", setup_disable_smap); static __always_inline void setup_smap(struct cpuinfo_x86 *c) { - unsigned long eflags; + unsigned long eflags = native_save_fl(); /* This should have been cleared long ago */ - raw_local_save_flags(eflags); BUG_ON(eflags & X86_EFLAGS_AC); if (cpu_has(c, X86_FEATURE_SMAP)) { @@ -419,7 +400,7 @@ static const struct cpu_dev *cpu_devs[X86_VENDOR_NUM] = {}; static void get_model_name(struct cpuinfo_x86 *c) { unsigned int *v; - char *p, *q; + char *p, *q, *s; if (c->extended_cpuid_level < 0x80000004) return; @@ -430,19 +411,21 @@ static void get_model_name(struct cpuinfo_x86 *c) cpuid(0x80000004, &v[8], &v[9], &v[10], &v[11]); c->x86_model_id[48] = 0; - /* - * Intel chips right-justify this string for some dumb reason; - * undo that brain damage: - */ - p = q = &c->x86_model_id[0]; + /* Trim whitespace */ + p = q = s = &c->x86_model_id[0]; + while (*p == ' ') p++; - if (p != q) { - while (*p) - *q++ = *p++; - while (q <= &c->x86_model_id[48]) - *q++ = '\0'; /* Zero-pad the rest */ + + while (*p) { + /* Note the last non-whitespace index */ + if (!isspace(*p)) + s = q; + + *q++ = *p++; } + + *(s + 1) = '\0'; } void cpu_detect_cache_sizes(struct cpuinfo_x86 *c) @@ -508,7 +491,7 @@ static void cpu_detect_tlb(struct cpuinfo_x86 *c) void detect_ht(struct cpuinfo_x86 *c) { -#ifdef CONFIG_X86_HT +#ifdef CONFIG_SMP u32 eax, ebx, ecx, edx; int index_msb, core_bits; static bool printed; @@ -686,6 +669,7 @@ void get_cpu_cap(struct cpuinfo_x86 *c) c->x86_virt_bits = (eax >> 8) & 0xff; c->x86_phys_bits = eax & 0xff; + c->x86_capability[13] = cpuid_ebx(0x80000008); } #ifdef CONFIG_X86_32 else if (cpu_has(c, X86_FEATURE_PAE) || cpu_has(c, X86_FEATURE_PSE36)) @@ -759,7 +743,6 @@ static void __init early_identify_cpu(struct cpuinfo_x86 *c) cpu_detect(c); get_cpu_vendor(c); get_cpu_cap(c); - fpu_detect(c); if (this_cpu->c_early_init) this_cpu->c_early_init(c); @@ -771,6 +754,7 @@ static void __init early_identify_cpu(struct cpuinfo_x86 *c) this_cpu->c_bsp_init(c); setup_force_cpu_cap(X86_FEATURE_ALWAYS); + fpu__init_system(c); } void __init early_cpu_init(void) @@ -844,7 +828,7 @@ static void generic_identify(struct cpuinfo_x86 *c) if (c->cpuid_level >= 0x00000001) { c->initial_apicid = (cpuid_ebx(1) >> 24) & 0xFF; #ifdef CONFIG_X86_32 -# ifdef CONFIG_X86_HT +# ifdef CONFIG_SMP c->apicid = apic->phys_pkg_id(c->initial_apicid, 0); # else c->apicid = c->initial_apicid; @@ -1026,7 +1010,7 @@ void enable_sep_cpu(void) (unsigned long)tss + offsetofend(struct tss_struct, SYSENTER_stack), 0); - wrmsr(MSR_IA32_SYSENTER_EIP, (unsigned long)ia32_sysenter_target, 0); + wrmsr(MSR_IA32_SYSENTER_EIP, (unsigned long)entry_SYSENTER_32, 0); out: put_cpu(); @@ -1122,14 +1106,14 @@ void print_cpu_info(struct cpuinfo_x86 *c) printk(KERN_CONT "%s ", vendor); if (c->x86_model_id[0]) - printk(KERN_CONT "%s", strim(c->x86_model_id)); + printk(KERN_CONT "%s", c->x86_model_id); else printk(KERN_CONT "%d86", c->x86); - printk(KERN_CONT " (fam: %02x, model: %02x", c->x86, c->x86_model); + printk(KERN_CONT " (family: 0x%x, model: 0x%x", c->x86, c->x86_model); if (c->x86_mask || c->cpuid_level >= 0) - printk(KERN_CONT ", stepping: %02x)\n", c->x86_mask); + printk(KERN_CONT ", stepping: 0x%x)\n", c->x86_mask); else printk(KERN_CONT ")\n"); @@ -1155,10 +1139,6 @@ static __init int setup_disablecpuid(char *arg) } __setup("clearcpuid=", setup_disablecpuid); -DEFINE_PER_CPU(unsigned long, kernel_stack) = - (unsigned long)&init_thread_union + THREAD_SIZE; -EXPORT_PER_CPU_SYMBOL(kernel_stack); - #ifdef CONFIG_X86_64 struct desc_ptr idt_descr = { NR_VECTORS * 16 - 1, (unsigned long) idt_table }; struct desc_ptr debug_idt_descr = { NR_VECTORS * 16 - 1, @@ -1183,8 +1163,6 @@ DEFINE_PER_CPU(unsigned int, irq_count) __visible = -1; DEFINE_PER_CPU(int, __preempt_count) = INIT_PREEMPT_COUNT; EXPORT_PER_CPU_SYMBOL(__preempt_count); -DEFINE_PER_CPU(struct task_struct *, fpu_owner_task); - /* * Special IST stacks which the CPU switches to when it calls * an IST-marked descriptor entry. Up to 7 stacks (hardware @@ -1208,10 +1186,10 @@ void syscall_init(void) * set CS/DS but only a 32bit target. LSTAR sets the 64bit rip. */ wrmsrl(MSR_STAR, ((u64)__USER32_CS)<<48 | ((u64)__KERNEL_CS)<<32); - wrmsrl(MSR_LSTAR, system_call); + wrmsrl(MSR_LSTAR, (unsigned long)entry_SYSCALL_64); #ifdef CONFIG_IA32_EMULATION - wrmsrl(MSR_CSTAR, ia32_cstar_target); + wrmsrl(MSR_CSTAR, (unsigned long)entry_SYSCALL_compat); /* * This only works on Intel CPUs. * On AMD CPUs these MSRs are 32-bit, CPU truncates MSR_IA32_SYSENTER_EIP. @@ -1220,9 +1198,9 @@ void syscall_init(void) */ wrmsrl_safe(MSR_IA32_SYSENTER_CS, (u64)__KERNEL_CS); wrmsrl_safe(MSR_IA32_SYSENTER_ESP, 0ULL); - wrmsrl_safe(MSR_IA32_SYSENTER_EIP, (u64)ia32_sysenter_target); + wrmsrl_safe(MSR_IA32_SYSENTER_EIP, (u64)entry_SYSENTER_compat); #else - wrmsrl(MSR_CSTAR, ignore_sysret); + wrmsrl(MSR_CSTAR, (unsigned long)ignore_sysret); wrmsrl_safe(MSR_IA32_SYSENTER_CS, (u64)GDT_ENTRY_INVALID_SEG); wrmsrl_safe(MSR_IA32_SYSENTER_ESP, 0ULL); wrmsrl_safe(MSR_IA32_SYSENTER_EIP, 0ULL); @@ -1275,7 +1253,6 @@ DEFINE_PER_CPU(struct task_struct *, current_task) = &init_task; EXPORT_PER_CPU_SYMBOL(current_task); DEFINE_PER_CPU(int, __preempt_count) = INIT_PREEMPT_COUNT; EXPORT_PER_CPU_SYMBOL(__preempt_count); -DEFINE_PER_CPU(struct task_struct *, fpu_owner_task); /* * On x86_32, vm86 modifies tss.sp0, so sp0 isn't a reliable way to find @@ -1439,7 +1416,7 @@ void cpu_init(void) clear_all_debug_regs(); dbg_restore_debug_regs(); - fpu_init(); + fpu__init_cpu(); if (is_uv_system()) uv_cpu_init(); @@ -1495,7 +1472,7 @@ void cpu_init(void) clear_all_debug_regs(); dbg_restore_debug_regs(); - fpu_init(); + fpu__init_cpu(); } #endif @@ -1512,3 +1489,20 @@ inline bool __static_cpu_has_safe(u16 bit) return boot_cpu_has(bit); } EXPORT_SYMBOL_GPL(__static_cpu_has_safe); + +static void bsp_resume(void) +{ + if (this_cpu->c_bsp_resume) + this_cpu->c_bsp_resume(&boot_cpu_data); +} + +static struct syscore_ops cpu_syscore_ops = { + .resume = bsp_resume, +}; + +static int __init init_cpu_syscore(void) +{ + register_syscore_ops(&cpu_syscore_ops); + return 0; +} +core_initcall(init_cpu_syscore); diff --git a/kernel/arch/x86/kernel/cpu/cpu.h b/kernel/arch/x86/kernel/cpu/cpu.h index c37dc37e8..2584265d4 100644 --- a/kernel/arch/x86/kernel/cpu/cpu.h +++ b/kernel/arch/x86/kernel/cpu/cpu.h @@ -13,6 +13,7 @@ struct cpu_dev { void (*c_init)(struct cpuinfo_x86 *); void (*c_identify)(struct cpuinfo_x86 *); void (*c_detect_tlb)(struct cpuinfo_x86 *); + void (*c_bsp_resume)(struct cpuinfo_x86 *); int c_x86_vendor; #ifdef CONFIG_X86_32 /* Optional vendor specific routine to obtain the cache size. */ diff --git a/kernel/arch/x86/kernel/cpu/intel.c b/kernel/arch/x86/kernel/cpu/intel.c index 50163fa90..209ac1e7d 100644 --- a/kernel/arch/x86/kernel/cpu/intel.c +++ b/kernel/arch/x86/kernel/cpu/intel.c @@ -97,6 +97,7 @@ static void early_init_intel(struct cpuinfo_x86 *c) switch (c->x86_model) { case 0x27: /* Penwell */ case 0x35: /* Cloverview */ + case 0x4a: /* Merrifield */ set_cpu_cap(c, X86_FEATURE_NONSTOP_TSC_S3); break; default: @@ -371,6 +372,36 @@ static void detect_vmx_virtcap(struct cpuinfo_x86 *c) } } +static void init_intel_energy_perf(struct cpuinfo_x86 *c) +{ + u64 epb; + + /* + * Initialize MSR_IA32_ENERGY_PERF_BIAS if not already initialized. + * (x86_energy_perf_policy(8) is available to change it at run-time.) + */ + if (!cpu_has(c, X86_FEATURE_EPB)) + return; + + rdmsrl(MSR_IA32_ENERGY_PERF_BIAS, epb); + if ((epb & 0xF) != ENERGY_PERF_BIAS_PERFORMANCE) + return; + + pr_warn_once("ENERGY_PERF_BIAS: Set to 'normal', was 'performance'\n"); + pr_warn_once("ENERGY_PERF_BIAS: View and update with x86_energy_perf_policy(8)\n"); + epb = (epb & ~0xF) | ENERGY_PERF_BIAS_NORMAL; + wrmsrl(MSR_IA32_ENERGY_PERF_BIAS, epb); +} + +static void intel_bsp_resume(struct cpuinfo_x86 *c) +{ + /* + * MSR_IA32_ENERGY_PERF_BIAS is lost across suspend/resume, + * so reinitialize it properly like during bootup: + */ + init_intel_energy_perf(c); +} + static void init_intel(struct cpuinfo_x86 *c) { unsigned int l2 = 0; @@ -478,21 +509,7 @@ static void init_intel(struct cpuinfo_x86 *c) if (cpu_has(c, X86_FEATURE_VMX)) detect_vmx_virtcap(c); - /* - * Initialize MSR_IA32_ENERGY_PERF_BIAS if BIOS did not. - * x86_energy_perf_policy(8) is available to change it at run-time - */ - if (cpu_has(c, X86_FEATURE_EPB)) { - u64 epb; - - rdmsrl(MSR_IA32_ENERGY_PERF_BIAS, epb); - if ((epb & 0xF) == ENERGY_PERF_BIAS_PERFORMANCE) { - pr_warn_once("ENERGY_PERF_BIAS: Set to 'normal', was 'performance'\n"); - pr_warn_once("ENERGY_PERF_BIAS: View and update with x86_energy_perf_policy(8)\n"); - epb = (epb & ~0xF) | ENERGY_PERF_BIAS_NORMAL; - wrmsrl(MSR_IA32_ENERGY_PERF_BIAS, epb); - } - } + init_intel_energy_perf(c); } #ifdef CONFIG_X86_32 @@ -747,6 +764,7 @@ static const struct cpu_dev intel_cpu_dev = { .c_detect_tlb = intel_detect_tlb, .c_early_init = early_init_intel, .c_init = init_intel, + .c_bsp_resume = intel_bsp_resume, .c_x86_vendor = X86_VENDOR_INTEL, }; diff --git a/kernel/arch/x86/kernel/cpu/intel_cacheinfo.c b/kernel/arch/x86/kernel/cpu/intel_cacheinfo.c index edcb0e28c..e38d338a6 100644 --- a/kernel/arch/x86/kernel/cpu/intel_cacheinfo.c +++ b/kernel/arch/x86/kernel/cpu/intel_cacheinfo.c @@ -157,7 +157,7 @@ struct _cpuid4_info_regs { struct amd_northbridge *nb; }; -unsigned short num_cache_leaves; +static unsigned short num_cache_leaves; /* AMD doesn't have CPUID4. Emulate it here to report the same information to the user. This makes some assumptions about the machine: @@ -326,7 +326,7 @@ static void amd_calc_l3_indices(struct amd_northbridge *nb) * * @returns: the disabled index if used or negative value if slot free. */ -int amd_get_l3_disable_slot(struct amd_northbridge *nb, unsigned slot) +static int amd_get_l3_disable_slot(struct amd_northbridge *nb, unsigned slot) { unsigned int reg = 0; @@ -403,8 +403,8 @@ static void amd_l3_disable_index(struct amd_northbridge *nb, int cpu, * * @return: 0 on success, error status on failure */ -int amd_set_l3_disable_slot(struct amd_northbridge *nb, int cpu, unsigned slot, - unsigned long index) +static int amd_set_l3_disable_slot(struct amd_northbridge *nb, int cpu, + unsigned slot, unsigned long index) { int ret = 0; @@ -654,7 +654,7 @@ unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c) unsigned int new_l1d = 0, new_l1i = 0; /* Cache sizes from cpuid(4) */ unsigned int new_l2 = 0, new_l3 = 0, i; /* Cache sizes from cpuid(4) */ unsigned int l2_id = 0, l3_id = 0, num_threads_sharing, index_msb; -#ifdef CONFIG_X86_HT +#ifdef CONFIG_SMP unsigned int cpu = c->cpu_index; #endif @@ -773,19 +773,19 @@ unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c) if (new_l2) { l2 = new_l2; -#ifdef CONFIG_X86_HT +#ifdef CONFIG_SMP per_cpu(cpu_llc_id, cpu) = l2_id; #endif } if (new_l3) { l3 = new_l3; -#ifdef CONFIG_X86_HT +#ifdef CONFIG_SMP per_cpu(cpu_llc_id, cpu) = l3_id; #endif } -#ifdef CONFIG_X86_HT +#ifdef CONFIG_SMP /* * If cpu_llc_id is not yet set, this means cpuid_level < 4 which in * turns means that the only possibility is SMT (as indicated in diff --git a/kernel/arch/x86/kernel/cpu/intel_pt.h b/kernel/arch/x86/kernel/cpu/intel_pt.h index 1c338b0eb..336878a5d 100644 --- a/kernel/arch/x86/kernel/cpu/intel_pt.h +++ b/kernel/arch/x86/kernel/cpu/intel_pt.h @@ -25,32 +25,11 @@ */ #define TOPA_PMI_MARGIN 512 -/* - * Table of Physical Addresses bits - */ -enum topa_sz { - TOPA_4K = 0, - TOPA_8K, - TOPA_16K, - TOPA_32K, - TOPA_64K, - TOPA_128K, - TOPA_256K, - TOPA_512K, - TOPA_1MB, - TOPA_2MB, - TOPA_4MB, - TOPA_8MB, - TOPA_16MB, - TOPA_32MB, - TOPA_64MB, - TOPA_128MB, - TOPA_SZ_END, -}; +#define TOPA_SHIFT 12 -static inline unsigned int sizes(enum topa_sz tsz) +static inline unsigned int sizes(unsigned int tsz) { - return 1 << (tsz + 12); + return 1 << (tsz + TOPA_SHIFT); }; struct topa_entry { @@ -66,20 +45,26 @@ struct topa_entry { u64 rsvd4 : 16; }; -#define TOPA_SHIFT 12 -#define PT_CPUID_LEAVES 2 +#define PT_CPUID_LEAVES 2 +#define PT_CPUID_REGS_NUM 4 /* number of regsters (eax, ebx, ecx, edx) */ enum pt_capabilities { PT_CAP_max_subleaf = 0, PT_CAP_cr3_filtering, + PT_CAP_psb_cyc, + PT_CAP_mtc, PT_CAP_topa_output, PT_CAP_topa_multiple_entries, + PT_CAP_single_range_output, PT_CAP_payloads_lip, + PT_CAP_mtc_periods, + PT_CAP_cycle_thresholds, + PT_CAP_psb_periods, }; struct pt_pmu { struct pmu pmu; - u32 caps[4 * PT_CPUID_LEAVES]; + u32 caps[PT_CPUID_REGS_NUM * PT_CPUID_LEAVES]; }; /** diff --git a/kernel/arch/x86/kernel/cpu/mcheck/Makefile b/kernel/arch/x86/kernel/cpu/mcheck/Makefile index bb34b03af..a3311c886 100644 --- a/kernel/arch/x86/kernel/cpu/mcheck/Makefile +++ b/kernel/arch/x86/kernel/cpu/mcheck/Makefile @@ -1,4 +1,4 @@ -obj-y = mce.o mce-severity.o +obj-y = mce.o mce-severity.o mce-genpool.o obj-$(CONFIG_X86_ANCIENT_MCE) += winchip.o p5.o obj-$(CONFIG_X86_MCE_INTEL) += mce_intel.o diff --git a/kernel/arch/x86/kernel/cpu/mcheck/mce-apei.c b/kernel/arch/x86/kernel/cpu/mcheck/mce-apei.c index a1aef9533..34c89a3e8 100644 --- a/kernel/arch/x86/kernel/cpu/mcheck/mce-apei.c +++ b/kernel/arch/x86/kernel/cpu/mcheck/mce-apei.c @@ -57,7 +57,6 @@ void apei_mce_report_mem_error(int severity, struct cper_sec_mem_err *mem_err) m.addr = mem_err->physical_addr; mce_log(&m); - mce_notify_irq(); } EXPORT_SYMBOL_GPL(apei_mce_report_mem_error); diff --git a/kernel/arch/x86/kernel/cpu/mcheck/mce-genpool.c b/kernel/arch/x86/kernel/cpu/mcheck/mce-genpool.c new file mode 100644 index 000000000..0a850100c --- /dev/null +++ b/kernel/arch/x86/kernel/cpu/mcheck/mce-genpool.c @@ -0,0 +1,99 @@ +/* + * MCE event pool management in MCE context + * + * Copyright (C) 2015 Intel Corp. + * Author: Chen, Gong <gong.chen@linux.intel.com> + * + * This file is licensed under GPLv2. + */ +#include <linux/smp.h> +#include <linux/mm.h> +#include <linux/genalloc.h> +#include <linux/llist.h> +#include "mce-internal.h" + +/* + * printk() is not safe in MCE context. This is a lock-less memory allocator + * used to save error information organized in a lock-less list. + * + * This memory pool is only to be used to save MCE records in MCE context. + * MCE events are rare, so a fixed size memory pool should be enough. Use + * 2 pages to save MCE events for now (~80 MCE records at most). + */ +#define MCE_POOLSZ (2 * PAGE_SIZE) + +static struct gen_pool *mce_evt_pool; +static LLIST_HEAD(mce_event_llist); +static char gen_pool_buf[MCE_POOLSZ]; + +void mce_gen_pool_process(void) +{ + struct llist_node *head; + struct mce_evt_llist *node; + struct mce *mce; + + head = llist_del_all(&mce_event_llist); + if (!head) + return; + + head = llist_reverse_order(head); + llist_for_each_entry(node, head, llnode) { + mce = &node->mce; + atomic_notifier_call_chain(&x86_mce_decoder_chain, 0, mce); + gen_pool_free(mce_evt_pool, (unsigned long)node, sizeof(*node)); + } +} + +bool mce_gen_pool_empty(void) +{ + return llist_empty(&mce_event_llist); +} + +int mce_gen_pool_add(struct mce *mce) +{ + struct mce_evt_llist *node; + + if (!mce_evt_pool) + return -EINVAL; + + node = (void *)gen_pool_alloc(mce_evt_pool, sizeof(*node)); + if (!node) { + pr_warn_ratelimited("MCE records pool full!\n"); + return -ENOMEM; + } + + memcpy(&node->mce, mce, sizeof(*mce)); + llist_add(&node->llnode, &mce_event_llist); + + return 0; +} + +static int mce_gen_pool_create(void) +{ + struct gen_pool *tmpp; + int ret = -ENOMEM; + + tmpp = gen_pool_create(ilog2(sizeof(struct mce_evt_llist)), -1); + if (!tmpp) + goto out; + + ret = gen_pool_add(tmpp, (unsigned long)gen_pool_buf, MCE_POOLSZ, -1); + if (ret) { + gen_pool_destroy(tmpp); + goto out; + } + + mce_evt_pool = tmpp; + +out: + return ret; +} + +int mce_gen_pool_init(void) +{ + /* Just init mce_gen_pool once. */ + if (mce_evt_pool) + return 0; + + return mce_gen_pool_create(); +} diff --git a/kernel/arch/x86/kernel/cpu/mcheck/mce-internal.h b/kernel/arch/x86/kernel/cpu/mcheck/mce-internal.h index fe32074b8..547720efd 100644 --- a/kernel/arch/x86/kernel/cpu/mcheck/mce-internal.h +++ b/kernel/arch/x86/kernel/cpu/mcheck/mce-internal.h @@ -13,6 +13,8 @@ enum severity_level { MCE_PANIC_SEVERITY, }; +extern struct atomic_notifier_head x86_mce_decoder_chain; + #define ATTR_LEN 16 #define INITIAL_CHECK_INTERVAL 5 * 60 /* 5 minutes */ @@ -24,6 +26,16 @@ struct mce_bank { char attrname[ATTR_LEN]; /* attribute name */ }; +struct mce_evt_llist { + struct llist_node llnode; + struct mce mce; +}; + +void mce_gen_pool_process(void); +bool mce_gen_pool_empty(void); +int mce_gen_pool_add(struct mce *mce); +int mce_gen_pool_init(void); + extern int (*mce_severity)(struct mce *a, int tolerant, char **msg, bool is_excp); struct dentry *mce_get_debugfs_dir(void); @@ -67,3 +79,5 @@ static inline int apei_clear_mce(u64 record_id) return -EINVAL; } #endif + +void mce_inject_log(struct mce *m); diff --git a/kernel/arch/x86/kernel/cpu/mcheck/mce.c b/kernel/arch/x86/kernel/cpu/mcheck/mce.c index 9d46f9a13..a080b4939 100644 --- a/kernel/arch/x86/kernel/cpu/mcheck/mce.c +++ b/kernel/arch/x86/kernel/cpu/mcheck/mce.c @@ -54,10 +54,13 @@ static DEFINE_MUTEX(mce_chrdev_read_mutex); -#define rcu_dereference_check_mce(p) \ - rcu_dereference_index_check((p), \ - rcu_read_lock_sched_held() || \ - lockdep_is_held(&mce_chrdev_read_mutex)) +#define mce_log_get_idx_check(p) \ +({ \ + RCU_LOCKDEP_WARN(!rcu_read_lock_sched_held() && \ + !lockdep_is_held(&mce_chrdev_read_mutex), \ + "suspicious mce_log_get_idx_check() usage"); \ + smp_load_acquire(&(p)); \ +}) #define CREATE_TRACE_POINTS #include <trace/events/mce.h> @@ -109,22 +112,24 @@ DEFINE_PER_CPU(mce_banks_t, mce_poll_banks) = { */ mce_banks_t mce_banks_ce_disabled; -static DEFINE_PER_CPU(struct work_struct, mce_work); +static struct work_struct mce_work; +static struct irq_work mce_irq_work; static void (*quirk_no_way_out)(int bank, struct mce *m, struct pt_regs *regs); +static int mce_usable_address(struct mce *m); /* * CPU/chipset specific EDAC code can register a notifier call here to print * MCE errors in a human-readable form. */ -static ATOMIC_NOTIFIER_HEAD(x86_mce_decoder_chain); +ATOMIC_NOTIFIER_HEAD(x86_mce_decoder_chain); /* Do initial initialization of a struct mce */ void mce_setup(struct mce *m) { memset(m, 0, sizeof(struct mce)); m->cpu = m->extcpu = smp_processor_id(); - rdtscll(m->tsc); + m->tsc = rdtsc(); /* We hope get_seconds stays lockless */ m->time = get_seconds(); m->cpuvendor = boot_cpu_data.x86_vendor; @@ -156,12 +161,13 @@ void mce_log(struct mce *mce) /* Emit the trace record: */ trace_mce_record(mce); - atomic_notifier_call_chain(&x86_mce_decoder_chain, 0, mce); + if (!mce_gen_pool_add(mce)) + irq_work_queue(&mce_irq_work); mce->finished = 0; wmb(); for (;;) { - entry = rcu_dereference_check_mce(mcelog.next); + entry = mce_log_get_idx_check(mcelog.next); for (;;) { /* @@ -195,48 +201,23 @@ void mce_log(struct mce *mce) set_bit(0, &mce_need_notify); } -static void drain_mcelog_buffer(void) +void mce_inject_log(struct mce *m) { - unsigned int next, i, prev = 0; - - next = ACCESS_ONCE(mcelog.next); - - do { - struct mce *m; - - /* drain what was logged during boot */ - for (i = prev; i < next; i++) { - unsigned long start = jiffies; - unsigned retries = 1; - - m = &mcelog.entry[i]; - - while (!m->finished) { - if (time_after_eq(jiffies, start + 2*retries)) - retries++; - - cpu_relax(); - - if (!m->finished && retries >= 4) { - pr_err("skipping error being logged currently!\n"); - break; - } - } - smp_rmb(); - atomic_notifier_call_chain(&x86_mce_decoder_chain, 0, m); - } - - memset(mcelog.entry + prev, 0, (next - prev) * sizeof(*m)); - prev = next; - next = cmpxchg(&mcelog.next, prev, 0); - } while (next != prev); + mutex_lock(&mce_chrdev_read_mutex); + mce_log(m); + mutex_unlock(&mce_chrdev_read_mutex); } +EXPORT_SYMBOL_GPL(mce_inject_log); +static struct notifier_block mce_srao_nb; void mce_register_decode_chain(struct notifier_block *nb) { + /* Ensure SRAO notifier has the highest priority in the decode chain. */ + if (nb != &mce_srao_nb && nb->priority == INT_MAX) + nb->priority -= 1; + atomic_notifier_chain_register(&x86_mce_decoder_chain, nb); - drain_mcelog_buffer(); } EXPORT_SYMBOL_GPL(mce_register_decode_chain); @@ -460,61 +441,6 @@ static inline void mce_gather_info(struct mce *m, struct pt_regs *regs) } } -/* - * Simple lockless ring to communicate PFNs from the exception handler with the - * process context work function. This is vastly simplified because there's - * only a single reader and a single writer. - */ -#define MCE_RING_SIZE 16 /* we use one entry less */ - -struct mce_ring { - unsigned short start; - unsigned short end; - unsigned long ring[MCE_RING_SIZE]; -}; -static DEFINE_PER_CPU(struct mce_ring, mce_ring); - -/* Runs with CPU affinity in workqueue */ -static int mce_ring_empty(void) -{ - struct mce_ring *r = this_cpu_ptr(&mce_ring); - - return r->start == r->end; -} - -static int mce_ring_get(unsigned long *pfn) -{ - struct mce_ring *r; - int ret = 0; - - *pfn = 0; - get_cpu(); - r = this_cpu_ptr(&mce_ring); - if (r->start == r->end) - goto out; - *pfn = r->ring[r->start]; - r->start = (r->start + 1) % MCE_RING_SIZE; - ret = 1; -out: - put_cpu(); - return ret; -} - -/* Always runs in MCE context with preempt off */ -static int mce_ring_add(unsigned long pfn) -{ - struct mce_ring *r = this_cpu_ptr(&mce_ring); - unsigned next; - - next = (r->end + 1) % MCE_RING_SIZE; - if (next == r->start) - return -1; - r->ring[r->end] = pfn; - wmb(); - r->end = next; - return 0; -} - int mce_available(struct cpuinfo_x86 *c) { if (mca_cfg.disabled) @@ -524,12 +450,10 @@ int mce_available(struct cpuinfo_x86 *c) static void mce_schedule_work(void) { - if (!mce_ring_empty()) - schedule_work(this_cpu_ptr(&mce_work)); + if (!mce_gen_pool_empty() && keventd_up()) + schedule_work(&mce_work); } -static DEFINE_PER_CPU(struct irq_work, mce_irq_work); - static void mce_irq_work_cb(struct irq_work *entry) { mce_notify_irq(); @@ -550,9 +474,30 @@ static void mce_report_event(struct pt_regs *regs) return; } - irq_work_queue(this_cpu_ptr(&mce_irq_work)); + irq_work_queue(&mce_irq_work); } +static int srao_decode_notifier(struct notifier_block *nb, unsigned long val, + void *data) +{ + struct mce *mce = (struct mce *)data; + unsigned long pfn; + + if (!mce) + return NOTIFY_DONE; + + if (mce->usable_addr && (mce->severity == MCE_AO_SEVERITY)) { + pfn = mce->addr >> PAGE_SHIFT; + memory_failure(pfn, MCE_VECTOR, 0); + } + + return NOTIFY_OK; +} +static struct notifier_block mce_srao_nb = { + .notifier_call = srao_decode_notifier, + .priority = INT_MAX, +}; + /* * Read ADDR and MISC registers. */ @@ -671,8 +616,11 @@ bool machine_check_poll(enum mcp_flags flags, mce_banks_t *b) */ if (severity == MCE_DEFERRED_SEVERITY && memory_error(&m)) { if (m.status & MCI_STATUS_ADDRV) { - mce_ring_add(m.addr >> PAGE_SHIFT); - mce_schedule_work(); + m.severity = severity; + m.usable_addr = mce_usable_address(&m); + + if (!mce_gen_pool_add(&m)) + mce_schedule_work(); } } @@ -1028,7 +976,6 @@ void do_machine_check(struct pt_regs *regs, long error_code) { struct mca_config *cfg = &mca_cfg; struct mce m, *final; - enum ctx_state prev_state; int i; int worst = 0; int severity; @@ -1052,8 +999,20 @@ void do_machine_check(struct pt_regs *regs, long error_code) char *msg = "Unknown"; u64 recover_paddr = ~0ull; int flags = MF_ACTION_REQUIRED; + int lmce = 0; - prev_state = ist_enter(regs); + /* If this CPU is offline, just bail out. */ + if (cpu_is_offline(smp_processor_id())) { + u64 mcgstatus; + + mcgstatus = mce_rdmsrl(MSR_IA32_MCG_STATUS); + if (mcgstatus & MCG_STATUS_RIPV) { + mce_wrmsrl(MSR_IA32_MCG_STATUS, 0); + return; + } + } + + ist_enter(regs); this_cpu_inc(mce_exception_count); @@ -1079,11 +1038,20 @@ void do_machine_check(struct pt_regs *regs, long error_code) kill_it = 1; /* - * Go through all the banks in exclusion of the other CPUs. - * This way we don't report duplicated events on shared banks - * because the first one to see it will clear it. + * Check if this MCE is signaled to only this logical processor */ - order = mce_start(&no_way_out); + if (m.mcgstatus & MCG_STATUS_LMCES) + lmce = 1; + else { + /* + * Go through all the banks in exclusion of the other CPUs. + * This way we don't report duplicated events on shared banks + * because the first one to see it will clear it. + * If this is a Local MCE, then no need to perform rendezvous. + */ + order = mce_start(&no_way_out); + } + for (i = 0; i < cfg->banks; i++) { __clear_bit(i, toclear); if (!test_bit(i, valid_banks)) @@ -1132,15 +1100,9 @@ void do_machine_check(struct pt_regs *regs, long error_code) mce_read_aux(&m, i); - /* - * Action optional error. Queue address for later processing. - * When the ring overflows we just ignore the AO error. - * RED-PEN add some logging mechanism when - * usable_address or mce_add_ring fails. - * RED-PEN don't ignore overflow for mca_cfg.tolerant == 0 - */ - if (severity == MCE_AO_SEVERITY && mce_usable_address(&m)) - mce_ring_add(m.addr >> PAGE_SHIFT); + /* assuming valid severity level != 0 */ + m.severity = severity; + m.usable_addr = mce_usable_address(&m); mce_log(&m); @@ -1160,8 +1122,18 @@ void do_machine_check(struct pt_regs *regs, long error_code) * Do most of the synchronization with other CPUs. * When there's any problem use only local no_way_out state. */ - if (mce_end(order) < 0) - no_way_out = worst >= MCE_PANIC_SEVERITY; + if (!lmce) { + if (mce_end(order) < 0) + no_way_out = worst >= MCE_PANIC_SEVERITY; + } else { + /* + * Local MCE skipped calling mce_reign() + * If we found a fatal error, we need to panic here. + */ + if (worst >= MCE_PANIC_SEVERITY && mca_cfg.tolerant < 3) + mce_panic("Machine check from unknown source", + NULL, NULL); + } /* * At insane "tolerant" levels we take no action. Otherwise @@ -1206,7 +1178,7 @@ out: local_irq_disable(); ist_end_non_atomic(); done: - ist_exit(regs, prev_state); + ist_exit(regs); } EXPORT_SYMBOL_GPL(do_machine_check); @@ -1226,14 +1198,11 @@ int memory_failure(unsigned long pfn, int vector, int flags) /* * Action optional processing happens here (picking up * from the list of faulting pages that do_machine_check() - * placed into the "ring"). + * placed into the genpool). */ static void mce_process_work(struct work_struct *dummy) { - unsigned long pfn; - - while (mce_ring_get(&pfn)) - memory_failure(pfn, MCE_VECTOR, 0); + mce_gen_pool_process(); } #ifdef CONFIG_X86_MCE_INTEL @@ -1655,6 +1624,8 @@ static int __mcheck_cpu_ancient_init(struct cpuinfo_x86 *c) winchip_mcheck_init(c); return 1; break; + default: + return 0; } return 0; @@ -1667,9 +1638,28 @@ static void __mcheck_cpu_init_vendor(struct cpuinfo_x86 *c) mce_intel_feature_init(c); mce_adjust_timer = cmci_intel_adjust_timer; break; - case X86_VENDOR_AMD: + + case X86_VENDOR_AMD: { + u32 ebx = cpuid_ebx(0x80000007); + mce_amd_feature_init(c); - mce_flags.overflow_recov = cpuid_ebx(0x80000007) & 0x1; + mce_flags.overflow_recov = !!(ebx & BIT(0)); + mce_flags.succor = !!(ebx & BIT(1)); + mce_flags.smca = !!(ebx & BIT(3)); + + break; + } + + default: + break; + } +} + +static void __mcheck_cpu_clear_vendor(struct cpuinfo_x86 *c) +{ + switch (c->x86_vendor) { + case X86_VENDOR_INTEL: + mce_intel_feature_clear(c); break; default: break; @@ -1730,13 +1720,36 @@ void mcheck_cpu_init(struct cpuinfo_x86 *c) return; } + if (mce_gen_pool_init()) { + mca_cfg.disabled = true; + pr_emerg("Couldn't allocate MCE records pool!\n"); + return; + } + machine_check_vector = do_machine_check; __mcheck_cpu_init_generic(); __mcheck_cpu_init_vendor(c); __mcheck_cpu_init_timer(); - INIT_WORK(this_cpu_ptr(&mce_work), mce_process_work); - init_irq_work(this_cpu_ptr(&mce_irq_work), &mce_irq_work_cb); +} + +/* + * Called for each booted CPU to clear some machine checks opt-ins + */ +void mcheck_cpu_clear(struct cpuinfo_x86 *c) +{ + if (mca_cfg.disabled) + return; + + if (!mce_available(c)) + return; + + /* + * Possibly to clear general settings generic to x86 + * __mcheck_cpu_clear_generic(c); + */ + __mcheck_cpu_clear_vendor(c); + } /* @@ -1783,7 +1796,7 @@ static void collect_tscs(void *data) { unsigned long *cpu_tsc = (unsigned long *)data; - rdtscll(cpu_tsc[smp_processor_id()]); + cpu_tsc[smp_processor_id()] = rdtsc(); } static int mce_apei_read_done; @@ -1849,7 +1862,7 @@ static ssize_t mce_chrdev_read(struct file *filp, char __user *ubuf, goto out; } - next = rcu_dereference_check_mce(mcelog.next); + next = mce_log_get_idx_check(mcelog.next); /* Only supports full reads right now */ err = -EINVAL; @@ -1915,7 +1928,7 @@ out: static unsigned int mce_chrdev_poll(struct file *file, poll_table *wait) { poll_wait(file, &mce_chrdev_wait, wait); - if (rcu_access_index(mcelog.next)) + if (READ_ONCE(mcelog.next)) return POLLIN | POLLRDNORM; if (!mce_apei_read_done && apei_check_mce()) return POLLIN | POLLRDNORM; @@ -1960,8 +1973,8 @@ void register_mce_write_callback(ssize_t (*fn)(struct file *filp, } EXPORT_SYMBOL_GPL(register_mce_write_callback); -ssize_t mce_chrdev_write(struct file *filp, const char __user *ubuf, - size_t usize, loff_t *off) +static ssize_t mce_chrdev_write(struct file *filp, const char __user *ubuf, + size_t usize, loff_t *off) { if (mce_write) return mce_write(filp, ubuf, usize, off); @@ -2007,6 +2020,7 @@ void mce_disable_bank(int bank) /* * mce=off Disables machine check * mce=no_cmci Disables CMCI + * mce=no_lmce Disables LMCE * mce=dont_log_ce Clears corrected events silently, no log created for CEs. * mce=ignore_ce Disables polling and CMCI, corrected events are not cleared. * mce=TOLERANCELEVEL[,monarchtimeout] (number, see above) @@ -2030,6 +2044,8 @@ static int __init mcheck_enable(char *str) cfg->disabled = true; else if (!strcmp(str, "no_cmci")) cfg->cmci_disabled = true; + else if (!strcmp(str, "no_lmce")) + cfg->lmce_disabled = true; else if (!strcmp(str, "dont_log_ce")) cfg->dont_log_ce = true; else if (!strcmp(str, "ignore_ce")) @@ -2039,11 +2055,8 @@ static int __init mcheck_enable(char *str) else if (!strcmp(str, "bios_cmci_threshold")) cfg->bios_cmci_threshold = true; else if (isdigit(str[0])) { - get_option(&str, &(cfg->tolerant)); - if (*str == ',') { - ++str; + if (get_option(&str, &cfg->tolerant) == 2) get_option(&str, &(cfg->monarch_timeout)); - } } else { pr_info("mce argument %s ignored. Please use /sys\n", str); return 0; @@ -2055,8 +2068,12 @@ __setup("mce", mcheck_enable); int __init mcheck_init(void) { mcheck_intel_therm_init(); + mce_register_decode_chain(&mce_srao_nb); mcheck_vendor_init_severity(); + INIT_WORK(&mce_work, mce_process_work); + init_irq_work(&mce_irq_work, mce_irq_work_cb); + return 0; } @@ -2068,7 +2085,7 @@ int __init mcheck_init(void) * Disable machine checks on suspend and shutdown. We can't really handle * them later. */ -static int mce_disable_error_reporting(void) +static void mce_disable_error_reporting(void) { int i; @@ -2078,17 +2095,32 @@ static int mce_disable_error_reporting(void) if (b->init) wrmsrl(MSR_IA32_MCx_CTL(i), 0); } - return 0; + return; +} + +static void vendor_disable_error_reporting(void) +{ + /* + * Don't clear on Intel CPUs. Some of these MSRs are socket-wide. + * Disabling them for just a single offlined CPU is bad, since it will + * inhibit reporting for all shared resources on the socket like the + * last level cache (LLC), the integrated memory controller (iMC), etc. + */ + if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) + return; + + mce_disable_error_reporting(); } static int mce_syscore_suspend(void) { - return mce_disable_error_reporting(); + vendor_disable_error_reporting(); + return 0; } static void mce_syscore_shutdown(void) { - mce_disable_error_reporting(); + vendor_disable_error_reporting(); } /* @@ -2368,7 +2400,6 @@ static void mce_device_remove(unsigned int cpu) static void mce_disable_cpu(void *h) { unsigned long action = *(unsigned long *)h; - int i; if (!mce_available(raw_cpu_ptr(&cpu_info))) return; @@ -2377,12 +2408,8 @@ static void mce_disable_cpu(void *h) if (!(action & CPU_TASKS_FROZEN)) cmci_clear(); - for (i = 0; i < mca_cfg.banks; i++) { - struct mce_bank *b = &mce_banks[i]; - if (b->init) - wrmsrl(MSR_IA32_MCx_CTL(i), 0); - } + vendor_disable_error_reporting(); } static void mce_reenable_cpu(void *h) @@ -2594,5 +2621,20 @@ static int __init mcheck_debugfs_init(void) return 0; } -late_initcall(mcheck_debugfs_init); +#else +static int __init mcheck_debugfs_init(void) { return -EINVAL; } #endif + +static int __init mcheck_late_init(void) +{ + mcheck_debugfs_init(); + + /* + * Flush out everything that has been logged during early boot, now that + * everything has been initialized (workqueues, decoders, ...). + */ + mce_schedule_work(); + + return 0; +} +late_initcall(mcheck_late_init); diff --git a/kernel/arch/x86/kernel/cpu/mcheck/mce_amd.c b/kernel/arch/x86/kernel/cpu/mcheck/mce_amd.c index 55ad9b37c..e99b15077 100644 --- a/kernel/arch/x86/kernel/cpu/mcheck/mce_amd.c +++ b/kernel/arch/x86/kernel/cpu/mcheck/mce_amd.c @@ -1,19 +1,13 @@ /* - * (c) 2005-2012 Advanced Micro Devices, Inc. + * (c) 2005-2015 Advanced Micro Devices, Inc. * Your use of this code is subject to the terms and conditions of the * GNU general public license version 2. See "COPYING" or * http://www.gnu.org/licenses/gpl.html * * Written by Jacob Shin - AMD, Inc. - * * Maintained by: Borislav Petkov <bp@alien8.de> * - * April 2006 - * - added support for AMD Family 0x10 processors - * May 2012 - * - major scrubbing - * - * All MC4_MISCi registers are shared between multi-cores + * All MC4_MISCi registers are shared between cores on a node. */ #include <linux/interrupt.h> #include <linux/notifier.h> @@ -32,6 +26,7 @@ #include <asm/idle.h> #include <asm/mce.h> #include <asm/msr.h> +#include <asm/trace/irq_vectors.h> #define NR_BLOCKS 9 #define THRESHOLD_MAX 0xFFF @@ -47,6 +42,13 @@ #define MASK_BLKPTR_LO 0xFF000000 #define MCG_XBLK_ADDR 0xC0000400 +/* Deferred error settings */ +#define MSR_CU_DEF_ERR 0xC0000410 +#define MASK_DEF_LVTOFF 0x000000F0 +#define MASK_DEF_INT_TYPE 0x00000006 +#define DEF_LVT_OFF 0x2 +#define DEF_INT_TYPE_APIC 0x2 + static const char * const th_names[] = { "load_store", "insn_fetch", @@ -60,6 +62,13 @@ static DEFINE_PER_CPU(struct threshold_bank **, threshold_banks); static DEFINE_PER_CPU(unsigned char, bank_map); /* see which banks are on */ static void amd_threshold_interrupt(void); +static void amd_deferred_error_interrupt(void); + +static void default_deferred_error_interrupt(void) +{ + pr_err("Unexpected deferred interrupt at vector %x\n", DEFERRED_ERROR_VECTOR); +} +void (*deferred_error_int_vector)(void) = default_deferred_error_interrupt; /* * CPU Initialization @@ -196,7 +205,7 @@ static void mce_threshold_block_init(struct threshold_block *b, int offset) threshold_restart_bank(&tr); }; -static int setup_APIC_mce(int reserved, int new) +static int setup_APIC_mce_threshold(int reserved, int new) { if (reserved < 0 && !setup_APIC_eilvt(new, THRESHOLD_APIC_VECTOR, APIC_EILVT_MSG_FIX, 0)) @@ -205,6 +214,39 @@ static int setup_APIC_mce(int reserved, int new) return reserved; } +static int setup_APIC_deferred_error(int reserved, int new) +{ + if (reserved < 0 && !setup_APIC_eilvt(new, DEFERRED_ERROR_VECTOR, + APIC_EILVT_MSG_FIX, 0)) + return new; + + return reserved; +} + +static void deferred_error_interrupt_enable(struct cpuinfo_x86 *c) +{ + u32 low = 0, high = 0; + int def_offset = -1, def_new; + + if (rdmsr_safe(MSR_CU_DEF_ERR, &low, &high)) + return; + + def_new = (low & MASK_DEF_LVTOFF) >> 4; + if (!(low & MASK_DEF_LVTOFF)) { + pr_err(FW_BUG "Your BIOS is not setting up LVT offset 0x2 for deferred error IRQs correctly.\n"); + def_new = DEF_LVT_OFF; + low = (low & ~MASK_DEF_LVTOFF) | (DEF_LVT_OFF << 4); + } + + def_offset = setup_APIC_deferred_error(def_offset, def_new); + if ((def_offset == def_new) && + (deferred_error_int_vector != amd_deferred_error_interrupt)) + deferred_error_int_vector = amd_deferred_error_interrupt; + + low = (low & ~MASK_DEF_INT_TYPE) | DEF_INT_TYPE_APIC; + wrmsr(MSR_CU_DEF_ERR, low, high); +} + /* cpu init entry point, called from mce.c with preempt off */ void mce_amd_feature_init(struct cpuinfo_x86 *c) { @@ -252,7 +294,7 @@ void mce_amd_feature_init(struct cpuinfo_x86 *c) b.interrupt_enable = 1; new = (high & MASK_LVTOFF_HI) >> 20; - offset = setup_APIC_mce(offset, new); + offset = setup_APIC_mce_threshold(offset, new); if ((offset == new) && (mce_threshold_vector != amd_threshold_interrupt)) @@ -262,6 +304,73 @@ init: mce_threshold_block_init(&b, offset); } } + + if (mce_flags.succor) + deferred_error_interrupt_enable(c); +} + +static void __log_error(unsigned int bank, bool threshold_err, u64 misc) +{ + struct mce m; + u64 status; + + rdmsrl(MSR_IA32_MCx_STATUS(bank), status); + if (!(status & MCI_STATUS_VAL)) + return; + + mce_setup(&m); + + m.status = status; + m.bank = bank; + + if (threshold_err) + m.misc = misc; + + if (m.status & MCI_STATUS_ADDRV) + rdmsrl(MSR_IA32_MCx_ADDR(bank), m.addr); + + mce_log(&m); + wrmsrl(MSR_IA32_MCx_STATUS(bank), 0); +} + +static inline void __smp_deferred_error_interrupt(void) +{ + inc_irq_stat(irq_deferred_error_count); + deferred_error_int_vector(); +} + +asmlinkage __visible void smp_deferred_error_interrupt(void) +{ + entering_irq(); + __smp_deferred_error_interrupt(); + exiting_ack_irq(); +} + +asmlinkage __visible void smp_trace_deferred_error_interrupt(void) +{ + entering_irq(); + trace_deferred_error_apic_entry(DEFERRED_ERROR_VECTOR); + __smp_deferred_error_interrupt(); + trace_deferred_error_apic_exit(DEFERRED_ERROR_VECTOR); + exiting_ack_irq(); +} + +/* APIC interrupt handler for deferred errors */ +static void amd_deferred_error_interrupt(void) +{ + u64 status; + unsigned int bank; + + for (bank = 0; bank < mca_cfg.banks; ++bank) { + rdmsrl(MSR_IA32_MCx_STATUS(bank), status); + + if (!(status & MCI_STATUS_VAL) || + !(status & MCI_STATUS_DEFERRED)) + continue; + + __log_error(bank, false, 0); + break; + } } /* @@ -273,12 +382,12 @@ init: * the interrupt goes off when error_count reaches threshold_limit. * the handler will simply log mcelog w/ software defined bank number. */ + static void amd_threshold_interrupt(void) { u32 low = 0, high = 0, address = 0; int cpu = smp_processor_id(); unsigned int bank, block; - struct mce m; /* assume first bank caused it */ for (bank = 0; bank < mca_cfg.banks; ++bank) { @@ -321,15 +430,7 @@ static void amd_threshold_interrupt(void) return; log: - mce_setup(&m); - rdmsrl(MSR_IA32_MCx_STATUS(bank), m.status); - if (!(m.status & MCI_STATUS_VAL)) - return; - m.misc = ((u64)high << 32) | low; - m.bank = bank; - mce_log(&m); - - wrmsrl(MSR_IA32_MCx_STATUS(bank), 0); + __log_error(bank, true, ((u64)high << 32) | low); } /* diff --git a/kernel/arch/x86/kernel/cpu/mcheck/mce_intel.c b/kernel/arch/x86/kernel/cpu/mcheck/mce_intel.c index e166d833c..1e8bb6c94 100644 --- a/kernel/arch/x86/kernel/cpu/mcheck/mce_intel.c +++ b/kernel/arch/x86/kernel/cpu/mcheck/mce_intel.c @@ -91,6 +91,36 @@ static int cmci_supported(int *banks) return !!(cap & MCG_CMCI_P); } +static bool lmce_supported(void) +{ + u64 tmp; + + if (mca_cfg.lmce_disabled) + return false; + + rdmsrl(MSR_IA32_MCG_CAP, tmp); + + /* + * LMCE depends on recovery support in the processor. Hence both + * MCG_SER_P and MCG_LMCE_P should be present in MCG_CAP. + */ + if ((tmp & (MCG_SER_P | MCG_LMCE_P)) != + (MCG_SER_P | MCG_LMCE_P)) + return false; + + /* + * BIOS should indicate support for LMCE by setting bit 20 in + * IA32_FEATURE_CONTROL without which touching MCG_EXT_CTL will + * generate a #GP fault. + */ + rdmsrl(MSR_IA32_FEATURE_CONTROL, tmp); + if ((tmp & (FEATURE_CONTROL_LOCKED | FEATURE_CONTROL_LMCE)) == + (FEATURE_CONTROL_LOCKED | FEATURE_CONTROL_LMCE)) + return true; + + return false; +} + bool mce_intel_cmci_poll(void) { if (__this_cpu_read(cmci_storm_state) == CMCI_STORM_NONE) @@ -221,7 +251,6 @@ static void intel_threshold_interrupt(void) return; machine_check_poll(MCP_TIMESTAMP, this_cpu_ptr(&mce_banks_owned)); - mce_notify_irq(); } /* @@ -410,8 +439,39 @@ static void intel_init_cmci(void) cmci_recheck(); } +static void intel_init_lmce(void) +{ + u64 val; + + if (!lmce_supported()) + return; + + rdmsrl(MSR_IA32_MCG_EXT_CTL, val); + + if (!(val & MCG_EXT_CTL_LMCE_EN)) + wrmsrl(MSR_IA32_MCG_EXT_CTL, val | MCG_EXT_CTL_LMCE_EN); +} + +static void intel_clear_lmce(void) +{ + u64 val; + + if (!lmce_supported()) + return; + + rdmsrl(MSR_IA32_MCG_EXT_CTL, val); + val &= ~MCG_EXT_CTL_LMCE_EN; + wrmsrl(MSR_IA32_MCG_EXT_CTL, val); +} + void mce_intel_feature_init(struct cpuinfo_x86 *c) { intel_init_thermal(c); intel_init_cmci(); + intel_init_lmce(); +} + +void mce_intel_feature_clear(struct cpuinfo_x86 *c) +{ + intel_clear_lmce(); } diff --git a/kernel/arch/x86/kernel/cpu/mcheck/p5.c b/kernel/arch/x86/kernel/cpu/mcheck/p5.c index 737b0ad4e..12402e10a 100644 --- a/kernel/arch/x86/kernel/cpu/mcheck/p5.c +++ b/kernel/arch/x86/kernel/cpu/mcheck/p5.c @@ -19,10 +19,9 @@ int mce_p5_enabled __read_mostly; /* Machine check handler for Pentium class Intel CPUs: */ static void pentium_machine_check(struct pt_regs *regs, long error_code) { - enum ctx_state prev_state; u32 loaddr, hi, lotype; - prev_state = ist_enter(regs); + ist_enter(regs); rdmsr(MSR_IA32_P5_MC_ADDR, loaddr, hi); rdmsr(MSR_IA32_P5_MC_TYPE, lotype, hi); @@ -39,7 +38,7 @@ static void pentium_machine_check(struct pt_regs *regs, long error_code) add_taint(TAINT_MACHINE_CHECK, LOCKDEP_NOW_UNRELIABLE); - ist_exit(regs, prev_state); + ist_exit(regs); } /* Set up machine check reporting for processors with Intel style MCE: */ diff --git a/kernel/arch/x86/kernel/cpu/mcheck/therm_throt.c b/kernel/arch/x86/kernel/cpu/mcheck/therm_throt.c index 1af51b158..2c5aaf8c2 100644 --- a/kernel/arch/x86/kernel/cpu/mcheck/therm_throt.c +++ b/kernel/arch/x86/kernel/cpu/mcheck/therm_throt.c @@ -503,14 +503,6 @@ void intel_init_thermal(struct cpuinfo_x86 *c) return; } - /* Check whether a vector already exists */ - if (h & APIC_VECTOR_MASK) { - printk(KERN_DEBUG - "CPU%d: Thermal LVT vector (%#x) already installed\n", - cpu, (h & APIC_VECTOR_MASK)); - return; - } - /* early Pentium M models use different method for enabling TM2 */ if (cpu_has(c, X86_FEATURE_TM2)) { if (c->x86 == 6 && (c->x86_model == 9 || c->x86_model == 13)) { diff --git a/kernel/arch/x86/kernel/cpu/mcheck/winchip.c b/kernel/arch/x86/kernel/cpu/mcheck/winchip.c index 44f138296..01dd87028 100644 --- a/kernel/arch/x86/kernel/cpu/mcheck/winchip.c +++ b/kernel/arch/x86/kernel/cpu/mcheck/winchip.c @@ -15,12 +15,12 @@ /* Machine check handler for WinChip C6: */ static void winchip_machine_check(struct pt_regs *regs, long error_code) { - enum ctx_state prev_state = ist_enter(regs); + ist_enter(regs); printk(KERN_EMERG "CPU0: Machine Check Exception.\n"); add_taint(TAINT_MACHINE_CHECK, LOCKDEP_NOW_UNRELIABLE); - ist_exit(regs, prev_state); + ist_exit(regs); } /* Set up machine check reporting on the Winchip C6 series */ diff --git a/kernel/arch/x86/kernel/cpu/microcode/Makefile b/kernel/arch/x86/kernel/cpu/microcode/Makefile index 285c85427..220b1a508 100644 --- a/kernel/arch/x86/kernel/cpu/microcode/Makefile +++ b/kernel/arch/x86/kernel/cpu/microcode/Makefile @@ -2,6 +2,3 @@ microcode-y := core.o obj-$(CONFIG_MICROCODE) += microcode.o microcode-$(CONFIG_MICROCODE_INTEL) += intel.o intel_lib.o microcode-$(CONFIG_MICROCODE_AMD) += amd.o -obj-$(CONFIG_MICROCODE_EARLY) += core_early.o -obj-$(CONFIG_MICROCODE_INTEL_EARLY) += intel_early.o -obj-$(CONFIG_MICROCODE_AMD_EARLY) += amd_early.o diff --git a/kernel/arch/x86/kernel/cpu/microcode/amd.c b/kernel/arch/x86/kernel/cpu/microcode/amd.c index 12829c3ce..2233f8a76 100644 --- a/kernel/arch/x86/kernel/cpu/microcode/amd.c +++ b/kernel/arch/x86/kernel/cpu/microcode/amd.c @@ -1,5 +1,9 @@ /* * AMD CPU Microcode Update Driver for Linux + * + * This driver allows to upgrade microcode on F10h AMD + * CPUs and later. + * * Copyright (C) 2008-2011 Advanced Micro Devices Inc. * * Author: Peter Oruba <peter.oruba@amd.com> @@ -7,34 +11,31 @@ * Based on work by: * Tigran Aivazian <tigran@aivazian.fsnet.co.uk> * - * Maintainers: - * Andreas Herrmann <herrmann.der.user@googlemail.com> - * Borislav Petkov <bp@alien8.de> + * early loader: + * Copyright (C) 2013 Advanced Micro Devices, Inc. * - * This driver allows to upgrade microcode on F10h AMD - * CPUs and later. + * Author: Jacob Shin <jacob.shin@amd.com> + * Fixes: Borislav Petkov <bp@suse.de> * * Licensed under the terms of the GNU General Public * License version 2. See file COPYING for details. */ +#define pr_fmt(fmt) "microcode: " fmt -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - +#include <linux/earlycpio.h> #include <linux/firmware.h> #include <linux/uaccess.h> #include <linux/vmalloc.h> +#include <linux/initrd.h> #include <linux/kernel.h> -#include <linux/module.h> #include <linux/pci.h> +#include <asm/microcode_amd.h> #include <asm/microcode.h> #include <asm/processor.h> +#include <asm/setup.h> +#include <asm/cpu.h> #include <asm/msr.h> -#include <asm/microcode_amd.h> - -MODULE_DESCRIPTION("AMD Microcode Update Driver"); -MODULE_AUTHOR("Peter Oruba"); -MODULE_LICENSE("GPL v2"); static struct equiv_cpu_entry *equiv_cpu_table; @@ -47,6 +48,432 @@ struct ucode_patch { static LIST_HEAD(pcache); +/* + * This points to the current valid container of microcode patches which we will + * save from the initrd before jettisoning its contents. + */ +static u8 *container; +static size_t container_size; + +static u32 ucode_new_rev; +u8 amd_ucode_patch[PATCH_MAX_SIZE]; +static u16 this_equiv_id; + +static struct cpio_data ucode_cpio; + +/* + * Microcode patch container file is prepended to the initrd in cpio format. + * See Documentation/x86/early-microcode.txt + */ +static __initdata char ucode_path[] = "kernel/x86/microcode/AuthenticAMD.bin"; + +static struct cpio_data __init find_ucode_in_initrd(void) +{ + long offset = 0; + char *path; + void *start; + size_t size; + +#ifdef CONFIG_X86_32 + struct boot_params *p; + + /* + * On 32-bit, early load occurs before paging is turned on so we need + * to use physical addresses. + */ + p = (struct boot_params *)__pa_nodebug(&boot_params); + path = (char *)__pa_nodebug(ucode_path); + start = (void *)p->hdr.ramdisk_image; + size = p->hdr.ramdisk_size; +#else + path = ucode_path; + start = (void *)(boot_params.hdr.ramdisk_image + PAGE_OFFSET); + size = boot_params.hdr.ramdisk_size; +#endif + + return find_cpio_data(path, start, size, &offset); +} + +static size_t compute_container_size(u8 *data, u32 total_size) +{ + size_t size = 0; + u32 *header = (u32 *)data; + + if (header[0] != UCODE_MAGIC || + header[1] != UCODE_EQUIV_CPU_TABLE_TYPE || /* type */ + header[2] == 0) /* size */ + return size; + + size = header[2] + CONTAINER_HDR_SZ; + total_size -= size; + data += size; + + while (total_size) { + u16 patch_size; + + header = (u32 *)data; + + if (header[0] != UCODE_UCODE_TYPE) + break; + + /* + * Sanity-check patch size. + */ + patch_size = header[1]; + if (patch_size > PATCH_MAX_SIZE) + break; + + size += patch_size + SECTION_HDR_SIZE; + data += patch_size + SECTION_HDR_SIZE; + total_size -= patch_size + SECTION_HDR_SIZE; + } + + return size; +} + +/* + * Early load occurs before we can vmalloc(). So we look for the microcode + * patch container file in initrd, traverse equivalent cpu table, look for a + * matching microcode patch, and update, all in initrd memory in place. + * When vmalloc() is available for use later -- on 64-bit during first AP load, + * and on 32-bit during save_microcode_in_initrd_amd() -- we can call + * load_microcode_amd() to save equivalent cpu table and microcode patches in + * kernel heap memory. + */ +static void apply_ucode_in_initrd(void *ucode, size_t size, bool save_patch) +{ + struct equiv_cpu_entry *eq; + size_t *cont_sz; + u32 *header; + u8 *data, **cont; + u8 (*patch)[PATCH_MAX_SIZE]; + u16 eq_id = 0; + int offset, left; + u32 rev, eax, ebx, ecx, edx; + u32 *new_rev; + +#ifdef CONFIG_X86_32 + new_rev = (u32 *)__pa_nodebug(&ucode_new_rev); + cont_sz = (size_t *)__pa_nodebug(&container_size); + cont = (u8 **)__pa_nodebug(&container); + patch = (u8 (*)[PATCH_MAX_SIZE])__pa_nodebug(&amd_ucode_patch); +#else + new_rev = &ucode_new_rev; + cont_sz = &container_size; + cont = &container; + patch = &amd_ucode_patch; +#endif + + data = ucode; + left = size; + header = (u32 *)data; + + /* find equiv cpu table */ + if (header[0] != UCODE_MAGIC || + header[1] != UCODE_EQUIV_CPU_TABLE_TYPE || /* type */ + header[2] == 0) /* size */ + return; + + eax = 0x00000001; + ecx = 0; + native_cpuid(&eax, &ebx, &ecx, &edx); + + while (left > 0) { + eq = (struct equiv_cpu_entry *)(data + CONTAINER_HDR_SZ); + + *cont = data; + + /* Advance past the container header */ + offset = header[2] + CONTAINER_HDR_SZ; + data += offset; + left -= offset; + + eq_id = find_equiv_id(eq, eax); + if (eq_id) { + this_equiv_id = eq_id; + *cont_sz = compute_container_size(*cont, left + offset); + + /* + * truncate how much we need to iterate over in the + * ucode update loop below + */ + left = *cont_sz - offset; + break; + } + + /* + * support multiple container files appended together. if this + * one does not have a matching equivalent cpu entry, we fast + * forward to the next container file. + */ + while (left > 0) { + header = (u32 *)data; + if (header[0] == UCODE_MAGIC && + header[1] == UCODE_EQUIV_CPU_TABLE_TYPE) + break; + + offset = header[1] + SECTION_HDR_SIZE; + data += offset; + left -= offset; + } + + /* mark where the next microcode container file starts */ + offset = data - (u8 *)ucode; + ucode = data; + } + + if (!eq_id) { + *cont = NULL; + *cont_sz = 0; + return; + } + + if (check_current_patch_level(&rev, true)) + return; + + while (left > 0) { + struct microcode_amd *mc; + + header = (u32 *)data; + if (header[0] != UCODE_UCODE_TYPE || /* type */ + header[1] == 0) /* size */ + break; + + mc = (struct microcode_amd *)(data + SECTION_HDR_SIZE); + + if (eq_id == mc->hdr.processor_rev_id && rev < mc->hdr.patch_id) { + + if (!__apply_microcode_amd(mc)) { + rev = mc->hdr.patch_id; + *new_rev = rev; + + if (save_patch) + memcpy(patch, mc, + min_t(u32, header[1], PATCH_MAX_SIZE)); + } + } + + offset = header[1] + SECTION_HDR_SIZE; + data += offset; + left -= offset; + } +} + +static bool __init load_builtin_amd_microcode(struct cpio_data *cp, + unsigned int family) +{ +#ifdef CONFIG_X86_64 + char fw_name[36] = "amd-ucode/microcode_amd.bin"; + + if (family >= 0x15) + snprintf(fw_name, sizeof(fw_name), + "amd-ucode/microcode_amd_fam%.2xh.bin", family); + + return get_builtin_firmware(cp, fw_name); +#else + return false; +#endif +} + +void __init load_ucode_amd_bsp(unsigned int family) +{ + struct cpio_data cp; + void **data; + size_t *size; + +#ifdef CONFIG_X86_32 + data = (void **)__pa_nodebug(&ucode_cpio.data); + size = (size_t *)__pa_nodebug(&ucode_cpio.size); +#else + data = &ucode_cpio.data; + size = &ucode_cpio.size; +#endif + + cp = find_ucode_in_initrd(); + if (!cp.data) { + if (!load_builtin_amd_microcode(&cp, family)) + return; + } + + *data = cp.data; + *size = cp.size; + + apply_ucode_in_initrd(cp.data, cp.size, true); +} + +#ifdef CONFIG_X86_32 +/* + * On 32-bit, since AP's early load occurs before paging is turned on, we + * cannot traverse cpu_equiv_table and pcache in kernel heap memory. So during + * cold boot, AP will apply_ucode_in_initrd() just like the BSP. During + * save_microcode_in_initrd_amd() BSP's patch is copied to amd_ucode_patch, + * which is used upon resume from suspend. + */ +void load_ucode_amd_ap(void) +{ + struct microcode_amd *mc; + size_t *usize; + void **ucode; + + mc = (struct microcode_amd *)__pa_nodebug(amd_ucode_patch); + if (mc->hdr.patch_id && mc->hdr.processor_rev_id) { + __apply_microcode_amd(mc); + return; + } + + ucode = (void *)__pa_nodebug(&container); + usize = (size_t *)__pa_nodebug(&container_size); + + if (!*ucode || !*usize) + return; + + apply_ucode_in_initrd(*ucode, *usize, false); +} + +static void __init collect_cpu_sig_on_bsp(void *arg) +{ + unsigned int cpu = smp_processor_id(); + struct ucode_cpu_info *uci = ucode_cpu_info + cpu; + + uci->cpu_sig.sig = cpuid_eax(0x00000001); +} + +static void __init get_bsp_sig(void) +{ + unsigned int bsp = boot_cpu_data.cpu_index; + struct ucode_cpu_info *uci = ucode_cpu_info + bsp; + + if (!uci->cpu_sig.sig) + smp_call_function_single(bsp, collect_cpu_sig_on_bsp, NULL, 1); +} +#else +void load_ucode_amd_ap(void) +{ + unsigned int cpu = smp_processor_id(); + struct equiv_cpu_entry *eq; + struct microcode_amd *mc; + u32 rev, eax; + u16 eq_id; + + /* Exit if called on the BSP. */ + if (!cpu) + return; + + if (!container) + return; + + /* + * 64-bit runs with paging enabled, thus early==false. + */ + if (check_current_patch_level(&rev, false)) + return; + + eax = cpuid_eax(0x00000001); + eq = (struct equiv_cpu_entry *)(container + CONTAINER_HDR_SZ); + + eq_id = find_equiv_id(eq, eax); + if (!eq_id) + return; + + if (eq_id == this_equiv_id) { + mc = (struct microcode_amd *)amd_ucode_patch; + + if (mc && rev < mc->hdr.patch_id) { + if (!__apply_microcode_amd(mc)) + ucode_new_rev = mc->hdr.patch_id; + } + + } else { + if (!ucode_cpio.data) + return; + + /* + * AP has a different equivalence ID than BSP, looks like + * mixed-steppings silicon so go through the ucode blob anew. + */ + apply_ucode_in_initrd(ucode_cpio.data, ucode_cpio.size, false); + } +} +#endif + +int __init save_microcode_in_initrd_amd(void) +{ + unsigned long cont; + int retval = 0; + enum ucode_state ret; + u8 *cont_va; + u32 eax; + + if (!container) + return -EINVAL; + +#ifdef CONFIG_X86_32 + get_bsp_sig(); + cont = (unsigned long)container; + cont_va = __va(container); +#else + /* + * We need the physical address of the container for both bitness since + * boot_params.hdr.ramdisk_image is a physical address. + */ + cont = __pa(container); + cont_va = container; +#endif + + /* + * Take into account the fact that the ramdisk might get relocated and + * therefore we need to recompute the container's position in virtual + * memory space. + */ + if (relocated_ramdisk) + container = (u8 *)(__va(relocated_ramdisk) + + (cont - boot_params.hdr.ramdisk_image)); + else + container = cont_va; + + if (ucode_new_rev) + pr_info("microcode: updated early to new patch_level=0x%08x\n", + ucode_new_rev); + + eax = cpuid_eax(0x00000001); + eax = ((eax >> 8) & 0xf) + ((eax >> 20) & 0xff); + + ret = load_microcode_amd(smp_processor_id(), eax, container, container_size); + if (ret != UCODE_OK) + retval = -EINVAL; + + /* + * This will be freed any msec now, stash patches for the current + * family and switch to patch cache for cpu hotplug, etc later. + */ + container = NULL; + container_size = 0; + + return retval; +} + +void reload_ucode_amd(void) +{ + struct microcode_amd *mc; + u32 rev; + + /* + * early==false because this is a syscore ->resume path and by + * that time paging is long enabled. + */ + if (check_current_patch_level(&rev, false)) + return; + + mc = (struct microcode_amd *)amd_ucode_patch; + + if (mc && rev < mc->hdr.patch_id) { + if (!__apply_microcode_amd(mc)) { + ucode_new_rev = mc->hdr.patch_id; + pr_info("microcode: reload patch_level=0x%08x\n", + ucode_new_rev); + } + } +} static u16 __find_equiv_id(unsigned int cpu) { struct ucode_cpu_info *uci = ucode_cpu_info + cpu; @@ -177,6 +604,53 @@ static unsigned int verify_patch_size(u8 family, u32 patch_size, return patch_size; } +/* + * Those patch levels cannot be updated to newer ones and thus should be final. + */ +static u32 final_levels[] = { + 0x01000098, + 0x0100009f, + 0x010000af, + 0, /* T-101 terminator */ +}; + +/* + * Check the current patch level on this CPU. + * + * @rev: Use it to return the patch level. It is set to 0 in the case of + * error. + * + * Returns: + * - true: if update should stop + * - false: otherwise + */ +bool check_current_patch_level(u32 *rev, bool early) +{ + u32 lvl, dummy, i; + bool ret = false; + u32 *levels; + + native_rdmsr(MSR_AMD64_PATCH_LEVEL, lvl, dummy); + + if (IS_ENABLED(CONFIG_X86_32) && early) + levels = (u32 *)__pa_nodebug(&final_levels); + else + levels = final_levels; + + for (i = 0; levels[i]; i++) { + if (lvl == levels[i]) { + lvl = 0; + ret = true; + break; + } + } + + if (rev) + *rev = lvl; + + return ret; +} + int __apply_microcode_amd(struct microcode_amd *mc_amd) { u32 rev, dummy; @@ -197,7 +671,7 @@ int apply_microcode_amd(int cpu) struct microcode_amd *mc_amd; struct ucode_cpu_info *uci; struct ucode_patch *p; - u32 rev, dummy; + u32 rev; BUG_ON(raw_smp_processor_id() != cpu); @@ -210,7 +684,8 @@ int apply_microcode_amd(int cpu) mc_amd = p->data; uci->mc = p->data; - rdmsr(MSR_AMD64_PATCH_LEVEL, rev, dummy); + if (check_current_patch_level(&rev, false)) + return -1; /* need to apply patch? */ if (rev >= mc_amd->hdr.patch_id) { @@ -387,7 +862,7 @@ enum ucode_state load_microcode_amd(int cpu, u8 family, const u8 *data, size_t s if (ret != UCODE_OK) cleanup(); -#if defined(CONFIG_MICROCODE_AMD_EARLY) && defined(CONFIG_X86_32) +#ifdef CONFIG_X86_32 /* save BSP's matching patch for early load */ if (cpu_data(cpu).cpu_index == boot_cpu_data.cpu_index) { struct ucode_patch *p = find_patch(cpu); @@ -475,7 +950,7 @@ static struct microcode_ops microcode_amd_ops = { struct microcode_ops * __init init_amd_microcode(void) { - struct cpuinfo_x86 *c = &cpu_data(0); + struct cpuinfo_x86 *c = &boot_cpu_data; if (c->x86_vendor != X86_VENDOR_AMD || c->x86 < 0x10) { pr_warning("AMD CPU family 0x%x not supported\n", c->x86); diff --git a/kernel/arch/x86/kernel/cpu/microcode/amd_early.c b/kernel/arch/x86/kernel/cpu/microcode/amd_early.c deleted file mode 100644 index 737737edb..000000000 --- a/kernel/arch/x86/kernel/cpu/microcode/amd_early.c +++ /dev/null @@ -1,422 +0,0 @@ -/* - * Copyright (C) 2013 Advanced Micro Devices, Inc. - * - * Author: Jacob Shin <jacob.shin@amd.com> - * Fixes: Borislav Petkov <bp@suse.de> - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - */ - -#include <linux/earlycpio.h> -#include <linux/initrd.h> - -#include <asm/cpu.h> -#include <asm/setup.h> -#include <asm/microcode_amd.h> - -/* - * This points to the current valid container of microcode patches which we will - * save from the initrd before jettisoning its contents. - */ -static u8 *container; -static size_t container_size; - -static u32 ucode_new_rev; -u8 amd_ucode_patch[PATCH_MAX_SIZE]; -static u16 this_equiv_id; - -static struct cpio_data ucode_cpio; - -/* - * Microcode patch container file is prepended to the initrd in cpio format. - * See Documentation/x86/early-microcode.txt - */ -static __initdata char ucode_path[] = "kernel/x86/microcode/AuthenticAMD.bin"; - -static struct cpio_data __init find_ucode_in_initrd(void) -{ - long offset = 0; - char *path; - void *start; - size_t size; - -#ifdef CONFIG_X86_32 - struct boot_params *p; - - /* - * On 32-bit, early load occurs before paging is turned on so we need - * to use physical addresses. - */ - p = (struct boot_params *)__pa_nodebug(&boot_params); - path = (char *)__pa_nodebug(ucode_path); - start = (void *)p->hdr.ramdisk_image; - size = p->hdr.ramdisk_size; -#else - path = ucode_path; - start = (void *)(boot_params.hdr.ramdisk_image + PAGE_OFFSET); - size = boot_params.hdr.ramdisk_size; -#endif - - return find_cpio_data(path, start, size, &offset); -} - -static size_t compute_container_size(u8 *data, u32 total_size) -{ - size_t size = 0; - u32 *header = (u32 *)data; - - if (header[0] != UCODE_MAGIC || - header[1] != UCODE_EQUIV_CPU_TABLE_TYPE || /* type */ - header[2] == 0) /* size */ - return size; - - size = header[2] + CONTAINER_HDR_SZ; - total_size -= size; - data += size; - - while (total_size) { - u16 patch_size; - - header = (u32 *)data; - - if (header[0] != UCODE_UCODE_TYPE) - break; - - /* - * Sanity-check patch size. - */ - patch_size = header[1]; - if (patch_size > PATCH_MAX_SIZE) - break; - - size += patch_size + SECTION_HDR_SIZE; - data += patch_size + SECTION_HDR_SIZE; - total_size -= patch_size + SECTION_HDR_SIZE; - } - - return size; -} - -/* - * Early load occurs before we can vmalloc(). So we look for the microcode - * patch container file in initrd, traverse equivalent cpu table, look for a - * matching microcode patch, and update, all in initrd memory in place. - * When vmalloc() is available for use later -- on 64-bit during first AP load, - * and on 32-bit during save_microcode_in_initrd_amd() -- we can call - * load_microcode_amd() to save equivalent cpu table and microcode patches in - * kernel heap memory. - */ -static void apply_ucode_in_initrd(void *ucode, size_t size, bool save_patch) -{ - struct equiv_cpu_entry *eq; - size_t *cont_sz; - u32 *header; - u8 *data, **cont; - u8 (*patch)[PATCH_MAX_SIZE]; - u16 eq_id = 0; - int offset, left; - u32 rev, eax, ebx, ecx, edx; - u32 *new_rev; - -#ifdef CONFIG_X86_32 - new_rev = (u32 *)__pa_nodebug(&ucode_new_rev); - cont_sz = (size_t *)__pa_nodebug(&container_size); - cont = (u8 **)__pa_nodebug(&container); - patch = (u8 (*)[PATCH_MAX_SIZE])__pa_nodebug(&amd_ucode_patch); -#else - new_rev = &ucode_new_rev; - cont_sz = &container_size; - cont = &container; - patch = &amd_ucode_patch; -#endif - - data = ucode; - left = size; - header = (u32 *)data; - - /* find equiv cpu table */ - if (header[0] != UCODE_MAGIC || - header[1] != UCODE_EQUIV_CPU_TABLE_TYPE || /* type */ - header[2] == 0) /* size */ - return; - - eax = 0x00000001; - ecx = 0; - native_cpuid(&eax, &ebx, &ecx, &edx); - - while (left > 0) { - eq = (struct equiv_cpu_entry *)(data + CONTAINER_HDR_SZ); - - *cont = data; - - /* Advance past the container header */ - offset = header[2] + CONTAINER_HDR_SZ; - data += offset; - left -= offset; - - eq_id = find_equiv_id(eq, eax); - if (eq_id) { - this_equiv_id = eq_id; - *cont_sz = compute_container_size(*cont, left + offset); - - /* - * truncate how much we need to iterate over in the - * ucode update loop below - */ - left = *cont_sz - offset; - break; - } - - /* - * support multiple container files appended together. if this - * one does not have a matching equivalent cpu entry, we fast - * forward to the next container file. - */ - while (left > 0) { - header = (u32 *)data; - if (header[0] == UCODE_MAGIC && - header[1] == UCODE_EQUIV_CPU_TABLE_TYPE) - break; - - offset = header[1] + SECTION_HDR_SIZE; - data += offset; - left -= offset; - } - - /* mark where the next microcode container file starts */ - offset = data - (u8 *)ucode; - ucode = data; - } - - if (!eq_id) { - *cont = NULL; - *cont_sz = 0; - return; - } - - /* find ucode and update if needed */ - - native_rdmsr(MSR_AMD64_PATCH_LEVEL, rev, eax); - - while (left > 0) { - struct microcode_amd *mc; - - header = (u32 *)data; - if (header[0] != UCODE_UCODE_TYPE || /* type */ - header[1] == 0) /* size */ - break; - - mc = (struct microcode_amd *)(data + SECTION_HDR_SIZE); - - if (eq_id == mc->hdr.processor_rev_id && rev < mc->hdr.patch_id) { - - if (!__apply_microcode_amd(mc)) { - rev = mc->hdr.patch_id; - *new_rev = rev; - - if (save_patch) - memcpy(patch, mc, - min_t(u32, header[1], PATCH_MAX_SIZE)); - } - } - - offset = header[1] + SECTION_HDR_SIZE; - data += offset; - left -= offset; - } -} - -void __init load_ucode_amd_bsp(void) -{ - struct cpio_data cp; - void **data; - size_t *size; - -#ifdef CONFIG_X86_32 - data = (void **)__pa_nodebug(&ucode_cpio.data); - size = (size_t *)__pa_nodebug(&ucode_cpio.size); -#else - data = &ucode_cpio.data; - size = &ucode_cpio.size; -#endif - - cp = find_ucode_in_initrd(); - if (!cp.data) - return; - - *data = cp.data; - *size = cp.size; - - apply_ucode_in_initrd(cp.data, cp.size, true); -} - -#ifdef CONFIG_X86_32 -/* - * On 32-bit, since AP's early load occurs before paging is turned on, we - * cannot traverse cpu_equiv_table and pcache in kernel heap memory. So during - * cold boot, AP will apply_ucode_in_initrd() just like the BSP. During - * save_microcode_in_initrd_amd() BSP's patch is copied to amd_ucode_patch, - * which is used upon resume from suspend. - */ -void load_ucode_amd_ap(void) -{ - struct microcode_amd *mc; - size_t *usize; - void **ucode; - - mc = (struct microcode_amd *)__pa_nodebug(amd_ucode_patch); - if (mc->hdr.patch_id && mc->hdr.processor_rev_id) { - __apply_microcode_amd(mc); - return; - } - - ucode = (void *)__pa_nodebug(&container); - usize = (size_t *)__pa_nodebug(&container_size); - - if (!*ucode || !*usize) - return; - - apply_ucode_in_initrd(*ucode, *usize, false); -} - -static void __init collect_cpu_sig_on_bsp(void *arg) -{ - unsigned int cpu = smp_processor_id(); - struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - - uci->cpu_sig.sig = cpuid_eax(0x00000001); -} - -static void __init get_bsp_sig(void) -{ - unsigned int bsp = boot_cpu_data.cpu_index; - struct ucode_cpu_info *uci = ucode_cpu_info + bsp; - - if (!uci->cpu_sig.sig) - smp_call_function_single(bsp, collect_cpu_sig_on_bsp, NULL, 1); -} -#else -void load_ucode_amd_ap(void) -{ - unsigned int cpu = smp_processor_id(); - struct ucode_cpu_info *uci = ucode_cpu_info + cpu; - struct equiv_cpu_entry *eq; - struct microcode_amd *mc; - u32 rev, eax; - u16 eq_id; - - /* Exit if called on the BSP. */ - if (!cpu) - return; - - if (!container) - return; - - rdmsr(MSR_AMD64_PATCH_LEVEL, rev, eax); - - uci->cpu_sig.rev = rev; - uci->cpu_sig.sig = eax; - - eax = cpuid_eax(0x00000001); - eq = (struct equiv_cpu_entry *)(container + CONTAINER_HDR_SZ); - - eq_id = find_equiv_id(eq, eax); - if (!eq_id) - return; - - if (eq_id == this_equiv_id) { - mc = (struct microcode_amd *)amd_ucode_patch; - - if (mc && rev < mc->hdr.patch_id) { - if (!__apply_microcode_amd(mc)) - ucode_new_rev = mc->hdr.patch_id; - } - - } else { - if (!ucode_cpio.data) - return; - - /* - * AP has a different equivalence ID than BSP, looks like - * mixed-steppings silicon so go through the ucode blob anew. - */ - apply_ucode_in_initrd(ucode_cpio.data, ucode_cpio.size, false); - } -} -#endif - -int __init save_microcode_in_initrd_amd(void) -{ - unsigned long cont; - int retval = 0; - enum ucode_state ret; - u8 *cont_va; - u32 eax; - - if (!container) - return -EINVAL; - -#ifdef CONFIG_X86_32 - get_bsp_sig(); - cont = (unsigned long)container; - cont_va = __va(container); -#else - /* - * We need the physical address of the container for both bitness since - * boot_params.hdr.ramdisk_image is a physical address. - */ - cont = __pa(container); - cont_va = container; -#endif - - /* - * Take into account the fact that the ramdisk might get relocated and - * therefore we need to recompute the container's position in virtual - * memory space. - */ - if (relocated_ramdisk) - container = (u8 *)(__va(relocated_ramdisk) + - (cont - boot_params.hdr.ramdisk_image)); - else - container = cont_va; - - if (ucode_new_rev) - pr_info("microcode: updated early to new patch_level=0x%08x\n", - ucode_new_rev); - - eax = cpuid_eax(0x00000001); - eax = ((eax >> 8) & 0xf) + ((eax >> 20) & 0xff); - - ret = load_microcode_amd(smp_processor_id(), eax, container, container_size); - if (ret != UCODE_OK) - retval = -EINVAL; - - /* - * This will be freed any msec now, stash patches for the current - * family and switch to patch cache for cpu hotplug, etc later. - */ - container = NULL; - container_size = 0; - - return retval; -} - -void reload_ucode_amd(void) -{ - struct microcode_amd *mc; - u32 rev, eax; - - rdmsr(MSR_AMD64_PATCH_LEVEL, rev, eax); - - mc = (struct microcode_amd *)amd_ucode_patch; - - if (mc && rev < mc->hdr.patch_id) { - if (!__apply_microcode_amd(mc)) { - ucode_new_rev = mc->hdr.patch_id; - pr_info("microcode: reload patch_level=0x%08x\n", - ucode_new_rev); - } - } -} diff --git a/kernel/arch/x86/kernel/cpu/microcode/core.c b/kernel/arch/x86/kernel/cpu/microcode/core.c index 36a83617e..b3e94ef46 100644 --- a/kernel/arch/x86/kernel/cpu/microcode/core.c +++ b/kernel/arch/x86/kernel/cpu/microcode/core.c @@ -1,104 +1,57 @@ /* - * Intel CPU Microcode Update Driver for Linux + * CPU Microcode Update Driver for Linux * - * Copyright (C) 2000-2006 Tigran Aivazian <tigran@aivazian.fsnet.co.uk> - * 2006 Shaohua Li <shaohua.li@intel.com> + * Copyright (C) 2000-2006 Tigran Aivazian <tigran@aivazian.fsnet.co.uk> + * 2006 Shaohua Li <shaohua.li@intel.com> + * 2013-2015 Borislav Petkov <bp@alien8.de> * - * This driver allows to upgrade microcode on Intel processors - * belonging to IA-32 family - PentiumPro, Pentium II, - * Pentium III, Xeon, Pentium 4, etc. + * X86 CPU microcode early update for Linux: * - * Reference: Section 8.11 of Volume 3a, IA-32 Intel? Architecture - * Software Developer's Manual - * Order Number 253668 or free download from: + * Copyright (C) 2012 Fenghua Yu <fenghua.yu@intel.com> + * H Peter Anvin" <hpa@zytor.com> + * (C) 2015 Borislav Petkov <bp@alien8.de> * - * http://developer.intel.com/Assets/PDF/manual/253668.pdf + * This driver allows to upgrade microcode on x86 processors. * - * For more information, go to http://www.urbanmyth.org/microcode - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - * - * 1.0 16 Feb 2000, Tigran Aivazian <tigran@sco.com> - * Initial release. - * 1.01 18 Feb 2000, Tigran Aivazian <tigran@sco.com> - * Added read() support + cleanups. - * 1.02 21 Feb 2000, Tigran Aivazian <tigran@sco.com> - * Added 'device trimming' support. open(O_WRONLY) zeroes - * and frees the saved copy of applied microcode. - * 1.03 29 Feb 2000, Tigran Aivazian <tigran@sco.com> - * Made to use devfs (/dev/cpu/microcode) + cleanups. - * 1.04 06 Jun 2000, Simon Trimmer <simon@veritas.com> - * Added misc device support (now uses both devfs and misc). - * Added MICROCODE_IOCFREE ioctl to clear memory. - * 1.05 09 Jun 2000, Simon Trimmer <simon@veritas.com> - * Messages for error cases (non Intel & no suitable microcode). - * 1.06 03 Aug 2000, Tigran Aivazian <tigran@veritas.com> - * Removed ->release(). Removed exclusive open and status bitmap. - * Added microcode_rwsem to serialize read()/write()/ioctl(). - * Removed global kernel lock usage. - * 1.07 07 Sep 2000, Tigran Aivazian <tigran@veritas.com> - * Write 0 to 0x8B msr and then cpuid before reading revision, - * so that it works even if there were no update done by the - * BIOS. Otherwise, reading from 0x8B gives junk (which happened - * to be 0 on my machine which is why it worked even when I - * disabled update by the BIOS) - * Thanks to Eric W. Biederman <ebiederman@lnxi.com> for the fix. - * 1.08 11 Dec 2000, Richard Schaal <richard.schaal@intel.com> and - * Tigran Aivazian <tigran@veritas.com> - * Intel Pentium 4 processor support and bugfixes. - * 1.09 30 Oct 2001, Tigran Aivazian <tigran@veritas.com> - * Bugfix for HT (Hyper-Threading) enabled processors - * whereby processor resources are shared by all logical processors - * in a single CPU package. - * 1.10 28 Feb 2002 Asit K Mallick <asit.k.mallick@intel.com> and - * Tigran Aivazian <tigran@veritas.com>, - * Serialize updates as required on HT processors due to - * speculative nature of implementation. - * 1.11 22 Mar 2002 Tigran Aivazian <tigran@veritas.com> - * Fix the panic when writing zero-length microcode chunk. - * 1.12 29 Sep 2003 Nitin Kamble <nitin.a.kamble@intel.com>, - * Jun Nakajima <jun.nakajima@intel.com> - * Support for the microcode updates in the new format. - * 1.13 10 Oct 2003 Tigran Aivazian <tigran@veritas.com> - * Removed ->read() method and obsoleted MICROCODE_IOCFREE ioctl - * because we no longer hold a copy of applied microcode - * in kernel memory. - * 1.14 25 Jun 2004 Tigran Aivazian <tigran@veritas.com> - * Fix sigmatch() macro to handle old CPUs with pf == 0. - * Thanks to Stuart Swales for pointing out this bug. + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. */ -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt +#define pr_fmt(fmt) "microcode: " fmt #include <linux/platform_device.h> +#include <linux/syscore_ops.h> #include <linux/miscdevice.h> #include <linux/capability.h> +#include <linux/firmware.h> #include <linux/kernel.h> -#include <linux/module.h> #include <linux/mutex.h> #include <linux/cpu.h> #include <linux/fs.h> #include <linux/mm.h> -#include <linux/syscore_ops.h> -#include <asm/microcode.h> -#include <asm/processor.h> +#include <asm/microcode_intel.h> #include <asm/cpu_device_id.h> +#include <asm/microcode_amd.h> #include <asm/perf_event.h> +#include <asm/microcode.h> +#include <asm/processor.h> +#include <asm/cmdline.h> -MODULE_DESCRIPTION("Microcode Update Driver"); -MODULE_AUTHOR("Tigran Aivazian <tigran@aivazian.fsnet.co.uk>"); -MODULE_LICENSE("GPL"); - -#define MICROCODE_VERSION "2.00" +#define MICROCODE_VERSION "2.01" static struct microcode_ops *microcode_ops; -bool dis_ucode_ldr; -module_param(dis_ucode_ldr, bool, 0); +static bool dis_ucode_ldr; + +static int __init disable_loader(char *str) +{ + dis_ucode_ldr = true; + return 1; +} +__setup("dis_ucode_ldr", disable_loader); /* * Synchronization. @@ -126,6 +79,150 @@ struct cpu_info_ctx { int err; }; +static bool __init check_loader_disabled_bsp(void) +{ +#ifdef CONFIG_X86_32 + const char *cmdline = (const char *)__pa_nodebug(boot_command_line); + const char *opt = "dis_ucode_ldr"; + const char *option = (const char *)__pa_nodebug(opt); + bool *res = (bool *)__pa_nodebug(&dis_ucode_ldr); + +#else /* CONFIG_X86_64 */ + const char *cmdline = boot_command_line; + const char *option = "dis_ucode_ldr"; + bool *res = &dis_ucode_ldr; +#endif + + if (cmdline_find_option_bool(cmdline, option)) + *res = true; + + return *res; +} + +extern struct builtin_fw __start_builtin_fw[]; +extern struct builtin_fw __end_builtin_fw[]; + +bool get_builtin_firmware(struct cpio_data *cd, const char *name) +{ +#ifdef CONFIG_FW_LOADER + struct builtin_fw *b_fw; + + for (b_fw = __start_builtin_fw; b_fw != __end_builtin_fw; b_fw++) { + if (!strcmp(name, b_fw->name)) { + cd->size = b_fw->size; + cd->data = b_fw->data; + return true; + } + } +#endif + return false; +} + +void __init load_ucode_bsp(void) +{ + int vendor; + unsigned int family; + + if (check_loader_disabled_bsp()) + return; + + if (!have_cpuid_p()) + return; + + vendor = x86_vendor(); + family = x86_family(); + + switch (vendor) { + case X86_VENDOR_INTEL: + if (family >= 6) + load_ucode_intel_bsp(); + break; + case X86_VENDOR_AMD: + if (family >= 0x10) + load_ucode_amd_bsp(family); + break; + default: + break; + } +} + +static bool check_loader_disabled_ap(void) +{ +#ifdef CONFIG_X86_32 + return *((bool *)__pa_nodebug(&dis_ucode_ldr)); +#else + return dis_ucode_ldr; +#endif +} + +void load_ucode_ap(void) +{ + int vendor, family; + + if (check_loader_disabled_ap()) + return; + + if (!have_cpuid_p()) + return; + + vendor = x86_vendor(); + family = x86_family(); + + switch (vendor) { + case X86_VENDOR_INTEL: + if (family >= 6) + load_ucode_intel_ap(); + break; + case X86_VENDOR_AMD: + if (family >= 0x10) + load_ucode_amd_ap(); + break; + default: + break; + } +} + +int __init save_microcode_in_initrd(void) +{ + struct cpuinfo_x86 *c = &boot_cpu_data; + + switch (c->x86_vendor) { + case X86_VENDOR_INTEL: + if (c->x86 >= 6) + save_microcode_in_initrd_intel(); + break; + case X86_VENDOR_AMD: + if (c->x86 >= 0x10) + save_microcode_in_initrd_amd(); + break; + default: + break; + } + + return 0; +} + +void reload_early_microcode(void) +{ + int vendor, family; + + vendor = x86_vendor(); + family = x86_family(); + + switch (vendor) { + case X86_VENDOR_INTEL: + if (family >= 6) + reload_ucode_intel(); + break; + case X86_VENDOR_AMD: + if (family >= 0x10) + reload_ucode_amd(); + break; + default: + break; + } +} + static void collect_cpu_info_local(void *arg) { struct cpu_info_ctx *ctx = arg; @@ -268,9 +365,6 @@ static void __exit microcode_dev_exit(void) { misc_deregister(µcode_dev); } - -MODULE_ALIAS_MISCDEV(MICROCODE_MINOR); -MODULE_ALIAS("devname:cpu/microcode"); #else #define microcode_dev_init() 0 #define microcode_dev_exit() do { } while (0) @@ -435,17 +529,16 @@ static int mc_device_add(struct device *dev, struct subsys_interface *sif) return err; } -static int mc_device_remove(struct device *dev, struct subsys_interface *sif) +static void mc_device_remove(struct device *dev, struct subsys_interface *sif) { int cpu = dev->id; if (!cpu_online(cpu)) - return 0; + return; pr_debug("CPU%d removed\n", cpu); microcode_fini_cpu(cpu); sysfs_remove_group(&dev->kobj, &mc_attr_group); - return 0; } static struct subsys_interface mc_cpu_interface = { @@ -518,24 +611,10 @@ mc_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu) return NOTIFY_OK; } -static struct notifier_block __refdata mc_cpu_notifier = { +static struct notifier_block mc_cpu_notifier = { .notifier_call = mc_cpu_callback, }; -#ifdef MODULE -/* Autoload on Intel and AMD systems */ -static const struct x86_cpu_id __initconst microcode_id[] = { -#ifdef CONFIG_MICROCODE_INTEL - { X86_VENDOR_INTEL, X86_FAMILY_ANY, X86_MODEL_ANY, }, -#endif -#ifdef CONFIG_MICROCODE_AMD - { X86_VENDOR_AMD, X86_FAMILY_ANY, X86_MODEL_ANY, }, -#endif - {} -}; -MODULE_DEVICE_TABLE(x86cpu, microcode_id); -#endif - static struct attribute *cpu_root_microcode_attrs[] = { &dev_attr_reload.attr, NULL @@ -546,9 +625,9 @@ static struct attribute_group cpu_root_microcode_group = { .attrs = cpu_root_microcode_attrs, }; -static int __init microcode_init(void) +int __init microcode_init(void) { - struct cpuinfo_x86 *c = &cpu_data(0); + struct cpuinfo_x86 *c = &boot_cpu_data; int error; if (paravirt_enabled() || dis_ucode_ldr) @@ -619,35 +698,4 @@ static int __init microcode_init(void) return error; } -module_init(microcode_init); - -static void __exit microcode_exit(void) -{ - struct cpuinfo_x86 *c = &cpu_data(0); - - microcode_dev_exit(); - - unregister_hotcpu_notifier(&mc_cpu_notifier); - unregister_syscore_ops(&mc_syscore_ops); - - sysfs_remove_group(&cpu_subsys.dev_root->kobj, - &cpu_root_microcode_group); - - get_online_cpus(); - mutex_lock(µcode_mutex); - - subsys_interface_unregister(&mc_cpu_interface); - - mutex_unlock(µcode_mutex); - put_online_cpus(); - - platform_device_unregister(microcode_pdev); - - microcode_ops = NULL; - - if (c->x86_vendor == X86_VENDOR_AMD) - exit_amd_microcode(); - - pr_info("Microcode Update Driver: v" MICROCODE_VERSION " removed.\n"); -} -module_exit(microcode_exit); +late_initcall(microcode_init); diff --git a/kernel/arch/x86/kernel/cpu/microcode/core_early.c b/kernel/arch/x86/kernel/cpu/microcode/core_early.c deleted file mode 100644 index a413a69cb..000000000 --- a/kernel/arch/x86/kernel/cpu/microcode/core_early.c +++ /dev/null @@ -1,148 +0,0 @@ -/* - * X86 CPU microcode early update for Linux - * - * Copyright (C) 2012 Fenghua Yu <fenghua.yu@intel.com> - * H Peter Anvin" <hpa@zytor.com> - * - * This driver allows to early upgrade microcode on Intel processors - * belonging to IA-32 family - PentiumPro, Pentium II, - * Pentium III, Xeon, Pentium 4, etc. - * - * Reference: Section 9.11 of Volume 3, IA-32 Intel Architecture - * Software Developer's Manual. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ -#include <linux/module.h> -#include <asm/microcode.h> -#include <asm/microcode_intel.h> -#include <asm/microcode_amd.h> -#include <asm/processor.h> -#include <asm/cmdline.h> - -static bool __init check_loader_disabled_bsp(void) -{ -#ifdef CONFIG_X86_32 - const char *cmdline = (const char *)__pa_nodebug(boot_command_line); - const char *opt = "dis_ucode_ldr"; - const char *option = (const char *)__pa_nodebug(opt); - bool *res = (bool *)__pa_nodebug(&dis_ucode_ldr); - -#else /* CONFIG_X86_64 */ - const char *cmdline = boot_command_line; - const char *option = "dis_ucode_ldr"; - bool *res = &dis_ucode_ldr; -#endif - - if (cmdline_find_option_bool(cmdline, option)) - *res = true; - - return *res; -} - -void __init load_ucode_bsp(void) -{ - int vendor, family; - - if (check_loader_disabled_bsp()) - return; - - if (!have_cpuid_p()) - return; - - vendor = x86_vendor(); - family = x86_family(); - - switch (vendor) { - case X86_VENDOR_INTEL: - if (family >= 6) - load_ucode_intel_bsp(); - break; - case X86_VENDOR_AMD: - if (family >= 0x10) - load_ucode_amd_bsp(); - break; - default: - break; - } -} - -static bool check_loader_disabled_ap(void) -{ -#ifdef CONFIG_X86_32 - return *((bool *)__pa_nodebug(&dis_ucode_ldr)); -#else - return dis_ucode_ldr; -#endif -} - -void load_ucode_ap(void) -{ - int vendor, family; - - if (check_loader_disabled_ap()) - return; - - if (!have_cpuid_p()) - return; - - vendor = x86_vendor(); - family = x86_family(); - - switch (vendor) { - case X86_VENDOR_INTEL: - if (family >= 6) - load_ucode_intel_ap(); - break; - case X86_VENDOR_AMD: - if (family >= 0x10) - load_ucode_amd_ap(); - break; - default: - break; - } -} - -int __init save_microcode_in_initrd(void) -{ - struct cpuinfo_x86 *c = &boot_cpu_data; - - switch (c->x86_vendor) { - case X86_VENDOR_INTEL: - if (c->x86 >= 6) - save_microcode_in_initrd_intel(); - break; - case X86_VENDOR_AMD: - if (c->x86 >= 0x10) - save_microcode_in_initrd_amd(); - break; - default: - break; - } - - return 0; -} - -void reload_early_microcode(void) -{ - int vendor, family; - - vendor = x86_vendor(); - family = x86_family(); - - switch (vendor) { - case X86_VENDOR_INTEL: - if (family >= 6) - reload_ucode_intel(); - break; - case X86_VENDOR_AMD: - if (family >= 0x10) - reload_ucode_amd(); - break; - default: - break; - } -} diff --git a/kernel/arch/x86/kernel/cpu/microcode/intel.c b/kernel/arch/x86/kernel/cpu/microcode/intel.c index a41beadb3..ce47402eb 100644 --- a/kernel/arch/x86/kernel/cpu/microcode/intel.c +++ b/kernel/arch/x86/kernel/cpu/microcode/intel.c @@ -1,91 +1,807 @@ /* - * Intel CPU Microcode Update Driver for Linux + * Intel CPU Microcode Update Driver for Linux * - * Copyright (C) 2000-2006 Tigran Aivazian <tigran@aivazian.fsnet.co.uk> - * 2006 Shaohua Li <shaohua.li@intel.com> + * Copyright (C) 2000-2006 Tigran Aivazian <tigran@aivazian.fsnet.co.uk> + * 2006 Shaohua Li <shaohua.li@intel.com> * - * This driver allows to upgrade microcode on Intel processors - * belonging to IA-32 family - PentiumPro, Pentium II, - * Pentium III, Xeon, Pentium 4, etc. + * Intel CPU microcode early update for Linux * - * Reference: Section 8.11 of Volume 3a, IA-32 Intel? Architecture - * Software Developer's Manual - * Order Number 253668 or free download from: + * Copyright (C) 2012 Fenghua Yu <fenghua.yu@intel.com> + * H Peter Anvin" <hpa@zytor.com> * - * http://developer.intel.com/Assets/PDF/manual/253668.pdf - * - * For more information, go to http://www.urbanmyth.org/microcode - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - * - * 1.0 16 Feb 2000, Tigran Aivazian <tigran@sco.com> - * Initial release. - * 1.01 18 Feb 2000, Tigran Aivazian <tigran@sco.com> - * Added read() support + cleanups. - * 1.02 21 Feb 2000, Tigran Aivazian <tigran@sco.com> - * Added 'device trimming' support. open(O_WRONLY) zeroes - * and frees the saved copy of applied microcode. - * 1.03 29 Feb 2000, Tigran Aivazian <tigran@sco.com> - * Made to use devfs (/dev/cpu/microcode) + cleanups. - * 1.04 06 Jun 2000, Simon Trimmer <simon@veritas.com> - * Added misc device support (now uses both devfs and misc). - * Added MICROCODE_IOCFREE ioctl to clear memory. - * 1.05 09 Jun 2000, Simon Trimmer <simon@veritas.com> - * Messages for error cases (non Intel & no suitable microcode). - * 1.06 03 Aug 2000, Tigran Aivazian <tigran@veritas.com> - * Removed ->release(). Removed exclusive open and status bitmap. - * Added microcode_rwsem to serialize read()/write()/ioctl(). - * Removed global kernel lock usage. - * 1.07 07 Sep 2000, Tigran Aivazian <tigran@veritas.com> - * Write 0 to 0x8B msr and then cpuid before reading revision, - * so that it works even if there were no update done by the - * BIOS. Otherwise, reading from 0x8B gives junk (which happened - * to be 0 on my machine which is why it worked even when I - * disabled update by the BIOS) - * Thanks to Eric W. Biederman <ebiederman@lnxi.com> for the fix. - * 1.08 11 Dec 2000, Richard Schaal <richard.schaal@intel.com> and - * Tigran Aivazian <tigran@veritas.com> - * Intel Pentium 4 processor support and bugfixes. - * 1.09 30 Oct 2001, Tigran Aivazian <tigran@veritas.com> - * Bugfix for HT (Hyper-Threading) enabled processors - * whereby processor resources are shared by all logical processors - * in a single CPU package. - * 1.10 28 Feb 2002 Asit K Mallick <asit.k.mallick@intel.com> and - * Tigran Aivazian <tigran@veritas.com>, - * Serialize updates as required on HT processors due to - * speculative nature of implementation. - * 1.11 22 Mar 2002 Tigran Aivazian <tigran@veritas.com> - * Fix the panic when writing zero-length microcode chunk. - * 1.12 29 Sep 2003 Nitin Kamble <nitin.a.kamble@intel.com>, - * Jun Nakajima <jun.nakajima@intel.com> - * Support for the microcode updates in the new format. - * 1.13 10 Oct 2003 Tigran Aivazian <tigran@veritas.com> - * Removed ->read() method and obsoleted MICROCODE_IOCFREE ioctl - * because we no longer hold a copy of applied microcode - * in kernel memory. - * 1.14 25 Jun 2004 Tigran Aivazian <tigran@veritas.com> - * Fix sigmatch() macro to handle old CPUs with pf == 0. - * Thanks to Stuart Swales for pointing out this bug. + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. */ -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt +/* + * This needs to be before all headers so that pr_debug in printk.h doesn't turn + * printk calls into no_printk(). + * + *#define DEBUG + */ +#define pr_fmt(fmt) "microcode: " fmt +#include <linux/earlycpio.h> #include <linux/firmware.h> #include <linux/uaccess.h> -#include <linux/kernel.h> -#include <linux/module.h> #include <linux/vmalloc.h> +#include <linux/initrd.h> +#include <linux/kernel.h> +#include <linux/slab.h> +#include <linux/cpu.h> +#include <linux/mm.h> #include <asm/microcode_intel.h> #include <asm/processor.h> +#include <asm/tlbflush.h> +#include <asm/setup.h> #include <asm/msr.h> -MODULE_DESCRIPTION("Microcode Update Driver"); -MODULE_AUTHOR("Tigran Aivazian <tigran@aivazian.fsnet.co.uk>"); -MODULE_LICENSE("GPL"); +static unsigned long mc_saved_in_initrd[MAX_UCODE_COUNT]; +static struct mc_saved_data { + unsigned int mc_saved_count; + struct microcode_intel **mc_saved; +} mc_saved_data; + +static enum ucode_state +load_microcode_early(struct microcode_intel **saved, + unsigned int num_saved, struct ucode_cpu_info *uci) +{ + struct microcode_intel *ucode_ptr, *new_mc = NULL; + struct microcode_header_intel *mc_hdr; + int new_rev, ret, i; + + new_rev = uci->cpu_sig.rev; + + for (i = 0; i < num_saved; i++) { + ucode_ptr = saved[i]; + mc_hdr = (struct microcode_header_intel *)ucode_ptr; + + ret = has_newer_microcode(ucode_ptr, + uci->cpu_sig.sig, + uci->cpu_sig.pf, + new_rev); + if (!ret) + continue; + + new_rev = mc_hdr->rev; + new_mc = ucode_ptr; + } + + if (!new_mc) + return UCODE_NFOUND; + + uci->mc = (struct microcode_intel *)new_mc; + return UCODE_OK; +} + +static inline void +copy_initrd_ptrs(struct microcode_intel **mc_saved, unsigned long *initrd, + unsigned long off, int num_saved) +{ + int i; + + for (i = 0; i < num_saved; i++) + mc_saved[i] = (struct microcode_intel *)(initrd[i] + off); +} + +#ifdef CONFIG_X86_32 +static void +microcode_phys(struct microcode_intel **mc_saved_tmp, + struct mc_saved_data *mc_saved_data) +{ + int i; + struct microcode_intel ***mc_saved; + + mc_saved = (struct microcode_intel ***) + __pa_nodebug(&mc_saved_data->mc_saved); + for (i = 0; i < mc_saved_data->mc_saved_count; i++) { + struct microcode_intel *p; + + p = *(struct microcode_intel **) + __pa_nodebug(mc_saved_data->mc_saved + i); + mc_saved_tmp[i] = (struct microcode_intel *)__pa_nodebug(p); + } +} +#endif + +static enum ucode_state +load_microcode(struct mc_saved_data *mc_saved_data, unsigned long *initrd, + unsigned long initrd_start, struct ucode_cpu_info *uci) +{ + struct microcode_intel *mc_saved_tmp[MAX_UCODE_COUNT]; + unsigned int count = mc_saved_data->mc_saved_count; + + if (!mc_saved_data->mc_saved) { + copy_initrd_ptrs(mc_saved_tmp, initrd, initrd_start, count); + + return load_microcode_early(mc_saved_tmp, count, uci); + } else { +#ifdef CONFIG_X86_32 + microcode_phys(mc_saved_tmp, mc_saved_data); + return load_microcode_early(mc_saved_tmp, count, uci); +#else + return load_microcode_early(mc_saved_data->mc_saved, + count, uci); +#endif + } +} + +/* + * Given CPU signature and a microcode patch, this function finds if the + * microcode patch has matching family and model with the CPU. + */ +static enum ucode_state +matching_model_microcode(struct microcode_header_intel *mc_header, + unsigned long sig) +{ + unsigned int fam, model; + unsigned int fam_ucode, model_ucode; + struct extended_sigtable *ext_header; + unsigned long total_size = get_totalsize(mc_header); + unsigned long data_size = get_datasize(mc_header); + int ext_sigcount, i; + struct extended_signature *ext_sig; + + fam = __x86_family(sig); + model = x86_model(sig); + + fam_ucode = __x86_family(mc_header->sig); + model_ucode = x86_model(mc_header->sig); + + if (fam == fam_ucode && model == model_ucode) + return UCODE_OK; + + /* Look for ext. headers: */ + if (total_size <= data_size + MC_HEADER_SIZE) + return UCODE_NFOUND; + + ext_header = (void *) mc_header + data_size + MC_HEADER_SIZE; + ext_sig = (void *)ext_header + EXT_HEADER_SIZE; + ext_sigcount = ext_header->count; + + for (i = 0; i < ext_sigcount; i++) { + fam_ucode = __x86_family(ext_sig->sig); + model_ucode = x86_model(ext_sig->sig); + + if (fam == fam_ucode && model == model_ucode) + return UCODE_OK; + + ext_sig++; + } + return UCODE_NFOUND; +} + +static int +save_microcode(struct mc_saved_data *mc_saved_data, + struct microcode_intel **mc_saved_src, + unsigned int mc_saved_count) +{ + int i, j; + struct microcode_intel **saved_ptr; + int ret; + + if (!mc_saved_count) + return -EINVAL; + + /* + * Copy new microcode data. + */ + saved_ptr = kcalloc(mc_saved_count, sizeof(struct microcode_intel *), GFP_KERNEL); + if (!saved_ptr) + return -ENOMEM; + + for (i = 0; i < mc_saved_count; i++) { + struct microcode_header_intel *mc_hdr; + struct microcode_intel *mc; + unsigned long size; + + if (!mc_saved_src[i]) { + ret = -EINVAL; + goto err; + } + + mc = mc_saved_src[i]; + mc_hdr = &mc->hdr; + size = get_totalsize(mc_hdr); + + saved_ptr[i] = kmalloc(size, GFP_KERNEL); + if (!saved_ptr[i]) { + ret = -ENOMEM; + goto err; + } + + memcpy(saved_ptr[i], mc, size); + } + + /* + * Point to newly saved microcode. + */ + mc_saved_data->mc_saved = saved_ptr; + mc_saved_data->mc_saved_count = mc_saved_count; + + return 0; + +err: + for (j = 0; j <= i; j++) + kfree(saved_ptr[j]); + kfree(saved_ptr); + + return ret; +} + +/* + * A microcode patch in ucode_ptr is saved into mc_saved + * - if it has matching signature and newer revision compared to an existing + * patch mc_saved. + * - or if it is a newly discovered microcode patch. + * + * The microcode patch should have matching model with CPU. + * + * Returns: The updated number @num_saved of saved microcode patches. + */ +static unsigned int _save_mc(struct microcode_intel **mc_saved, + u8 *ucode_ptr, unsigned int num_saved) +{ + struct microcode_header_intel *mc_hdr, *mc_saved_hdr; + unsigned int sig, pf; + int found = 0, i; + + mc_hdr = (struct microcode_header_intel *)ucode_ptr; + + for (i = 0; i < num_saved; i++) { + mc_saved_hdr = (struct microcode_header_intel *)mc_saved[i]; + sig = mc_saved_hdr->sig; + pf = mc_saved_hdr->pf; + + if (!find_matching_signature(ucode_ptr, sig, pf)) + continue; + + found = 1; + + if (mc_hdr->rev <= mc_saved_hdr->rev) + continue; + + /* + * Found an older ucode saved earlier. Replace it with + * this newer one. + */ + mc_saved[i] = (struct microcode_intel *)ucode_ptr; + break; + } + + /* Newly detected microcode, save it to memory. */ + if (i >= num_saved && !found) + mc_saved[num_saved++] = (struct microcode_intel *)ucode_ptr; + + return num_saved; +} + +/* + * Get microcode matching with BSP's model. Only CPUs with the same model as + * BSP can stay in the platform. + */ +static enum ucode_state __init +get_matching_model_microcode(int cpu, unsigned long start, + void *data, size_t size, + struct mc_saved_data *mc_saved_data, + unsigned long *mc_saved_in_initrd, + struct ucode_cpu_info *uci) +{ + u8 *ucode_ptr = data; + unsigned int leftover = size; + enum ucode_state state = UCODE_OK; + unsigned int mc_size; + struct microcode_header_intel *mc_header; + struct microcode_intel *mc_saved_tmp[MAX_UCODE_COUNT]; + unsigned int mc_saved_count = mc_saved_data->mc_saved_count; + int i; + + while (leftover && mc_saved_count < ARRAY_SIZE(mc_saved_tmp)) { + + if (leftover < sizeof(mc_header)) + break; + + mc_header = (struct microcode_header_intel *)ucode_ptr; + + mc_size = get_totalsize(mc_header); + if (!mc_size || mc_size > leftover || + microcode_sanity_check(ucode_ptr, 0) < 0) + break; + + leftover -= mc_size; + + /* + * Since APs with same family and model as the BSP may boot in + * the platform, we need to find and save microcode patches + * with the same family and model as the BSP. + */ + if (matching_model_microcode(mc_header, uci->cpu_sig.sig) != + UCODE_OK) { + ucode_ptr += mc_size; + continue; + } + + mc_saved_count = _save_mc(mc_saved_tmp, ucode_ptr, mc_saved_count); + + ucode_ptr += mc_size; + } + + if (leftover) { + state = UCODE_ERROR; + goto out; + } + + if (mc_saved_count == 0) { + state = UCODE_NFOUND; + goto out; + } + + for (i = 0; i < mc_saved_count; i++) + mc_saved_in_initrd[i] = (unsigned long)mc_saved_tmp[i] - start; + + mc_saved_data->mc_saved_count = mc_saved_count; +out: + return state; +} + +static int collect_cpu_info_early(struct ucode_cpu_info *uci) +{ + unsigned int val[2]; + unsigned int family, model; + struct cpu_signature csig; + unsigned int eax, ebx, ecx, edx; + + csig.sig = 0; + csig.pf = 0; + csig.rev = 0; + + memset(uci, 0, sizeof(*uci)); + + eax = 0x00000001; + ecx = 0; + native_cpuid(&eax, &ebx, &ecx, &edx); + csig.sig = eax; + + family = __x86_family(csig.sig); + model = x86_model(csig.sig); + + if ((model >= 5) || (family > 6)) { + /* get processor flags from MSR 0x17 */ + native_rdmsr(MSR_IA32_PLATFORM_ID, val[0], val[1]); + csig.pf = 1 << ((val[1] >> 18) & 7); + } + native_wrmsr(MSR_IA32_UCODE_REV, 0, 0); + + /* As documented in the SDM: Do a CPUID 1 here */ + sync_core(); + + /* get the current revision from MSR 0x8B */ + native_rdmsr(MSR_IA32_UCODE_REV, val[0], val[1]); + + csig.rev = val[1]; + + uci->cpu_sig = csig; + uci->valid = 1; + + return 0; +} + +static void show_saved_mc(void) +{ +#ifdef DEBUG + int i, j; + unsigned int sig, pf, rev, total_size, data_size, date; + struct ucode_cpu_info uci; + + if (mc_saved_data.mc_saved_count == 0) { + pr_debug("no microcode data saved.\n"); + return; + } + pr_debug("Total microcode saved: %d\n", mc_saved_data.mc_saved_count); + + collect_cpu_info_early(&uci); + + sig = uci.cpu_sig.sig; + pf = uci.cpu_sig.pf; + rev = uci.cpu_sig.rev; + pr_debug("CPU: sig=0x%x, pf=0x%x, rev=0x%x\n", sig, pf, rev); + + for (i = 0; i < mc_saved_data.mc_saved_count; i++) { + struct microcode_header_intel *mc_saved_header; + struct extended_sigtable *ext_header; + int ext_sigcount; + struct extended_signature *ext_sig; + + mc_saved_header = (struct microcode_header_intel *) + mc_saved_data.mc_saved[i]; + sig = mc_saved_header->sig; + pf = mc_saved_header->pf; + rev = mc_saved_header->rev; + total_size = get_totalsize(mc_saved_header); + data_size = get_datasize(mc_saved_header); + date = mc_saved_header->date; + + pr_debug("mc_saved[%d]: sig=0x%x, pf=0x%x, rev=0x%x, toal size=0x%x, date = %04x-%02x-%02x\n", + i, sig, pf, rev, total_size, + date & 0xffff, + date >> 24, + (date >> 16) & 0xff); + + /* Look for ext. headers: */ + if (total_size <= data_size + MC_HEADER_SIZE) + continue; + + ext_header = (void *) mc_saved_header + data_size + MC_HEADER_SIZE; + ext_sigcount = ext_header->count; + ext_sig = (void *)ext_header + EXT_HEADER_SIZE; + + for (j = 0; j < ext_sigcount; j++) { + sig = ext_sig->sig; + pf = ext_sig->pf; + + pr_debug("\tExtended[%d]: sig=0x%x, pf=0x%x\n", + j, sig, pf); + + ext_sig++; + } + + } +#endif +} + +#ifdef CONFIG_HOTPLUG_CPU +static DEFINE_MUTEX(x86_cpu_microcode_mutex); +/* + * Save this mc into mc_saved_data. So it will be loaded early when a CPU is + * hot added or resumes. + * + * Please make sure this mc should be a valid microcode patch before calling + * this function. + */ +int save_mc_for_early(u8 *mc) +{ + struct microcode_intel *mc_saved_tmp[MAX_UCODE_COUNT]; + unsigned int mc_saved_count_init; + unsigned int mc_saved_count; + struct microcode_intel **mc_saved; + int ret = 0; + int i; + + /* + * Hold hotplug lock so mc_saved_data is not accessed by a CPU in + * hotplug. + */ + mutex_lock(&x86_cpu_microcode_mutex); + + mc_saved_count_init = mc_saved_data.mc_saved_count; + mc_saved_count = mc_saved_data.mc_saved_count; + mc_saved = mc_saved_data.mc_saved; + + if (mc_saved && mc_saved_count) + memcpy(mc_saved_tmp, mc_saved, + mc_saved_count * sizeof(struct microcode_intel *)); + /* + * Save the microcode patch mc in mc_save_tmp structure if it's a newer + * version. + */ + mc_saved_count = _save_mc(mc_saved_tmp, mc, mc_saved_count); + + /* + * Save the mc_save_tmp in global mc_saved_data. + */ + ret = save_microcode(&mc_saved_data, mc_saved_tmp, mc_saved_count); + if (ret) { + pr_err("Cannot save microcode patch.\n"); + goto out; + } + + show_saved_mc(); + + /* + * Free old saved microcode data. + */ + if (mc_saved) { + for (i = 0; i < mc_saved_count_init; i++) + kfree(mc_saved[i]); + kfree(mc_saved); + } + +out: + mutex_unlock(&x86_cpu_microcode_mutex); + + return ret; +} +EXPORT_SYMBOL_GPL(save_mc_for_early); +#endif + +static bool __init load_builtin_intel_microcode(struct cpio_data *cp) +{ +#ifdef CONFIG_X86_64 + unsigned int eax = 0x00000001, ebx, ecx = 0, edx; + unsigned int family, model, stepping; + char name[30]; + + native_cpuid(&eax, &ebx, &ecx, &edx); + + family = __x86_family(eax); + model = x86_model(eax); + stepping = eax & 0xf; + + sprintf(name, "intel-ucode/%02x-%02x-%02x", family, model, stepping); + + return get_builtin_firmware(cp, name); +#else + return false; +#endif +} + +static __initdata char ucode_name[] = "kernel/x86/microcode/GenuineIntel.bin"; +static __init enum ucode_state +scan_microcode(struct mc_saved_data *mc_saved_data, unsigned long *initrd, + unsigned long start, unsigned long size, + struct ucode_cpu_info *uci) +{ + struct cpio_data cd; + long offset = 0; +#ifdef CONFIG_X86_32 + char *p = (char *)__pa_nodebug(ucode_name); +#else + char *p = ucode_name; +#endif + + cd.data = NULL; + cd.size = 0; + + cd = find_cpio_data(p, (void *)start, size, &offset); + if (!cd.data) { + if (!load_builtin_intel_microcode(&cd)) + return UCODE_ERROR; + } + + return get_matching_model_microcode(0, start, cd.data, cd.size, + mc_saved_data, initrd, uci); +} + +/* + * Print ucode update info. + */ +static void +print_ucode_info(struct ucode_cpu_info *uci, unsigned int date) +{ + int cpu = smp_processor_id(); + + pr_info("CPU%d microcode updated early to revision 0x%x, date = %04x-%02x-%02x\n", + cpu, + uci->cpu_sig.rev, + date & 0xffff, + date >> 24, + (date >> 16) & 0xff); +} + +#ifdef CONFIG_X86_32 + +static int delay_ucode_info; +static int current_mc_date; + +/* + * Print early updated ucode info after printk works. This is delayed info dump. + */ +void show_ucode_info_early(void) +{ + struct ucode_cpu_info uci; + + if (delay_ucode_info) { + collect_cpu_info_early(&uci); + print_ucode_info(&uci, current_mc_date); + delay_ucode_info = 0; + } +} + +/* + * At this point, we can not call printk() yet. Keep microcode patch number in + * mc_saved_data.mc_saved and delay printing microcode info in + * show_ucode_info_early() until printk() works. + */ +static void print_ucode(struct ucode_cpu_info *uci) +{ + struct microcode_intel *mc_intel; + int *delay_ucode_info_p; + int *current_mc_date_p; + + mc_intel = uci->mc; + if (mc_intel == NULL) + return; + + delay_ucode_info_p = (int *)__pa_nodebug(&delay_ucode_info); + current_mc_date_p = (int *)__pa_nodebug(¤t_mc_date); + + *delay_ucode_info_p = 1; + *current_mc_date_p = mc_intel->hdr.date; +} +#else + +/* + * Flush global tlb. We only do this in x86_64 where paging has been enabled + * already and PGE should be enabled as well. + */ +static inline void flush_tlb_early(void) +{ + __native_flush_tlb_global_irq_disabled(); +} + +static inline void print_ucode(struct ucode_cpu_info *uci) +{ + struct microcode_intel *mc_intel; + + mc_intel = uci->mc; + if (mc_intel == NULL) + return; + + print_ucode_info(uci, mc_intel->hdr.date); +} +#endif + +static int apply_microcode_early(struct ucode_cpu_info *uci, bool early) +{ + struct microcode_intel *mc_intel; + unsigned int val[2]; + + mc_intel = uci->mc; + if (mc_intel == NULL) + return 0; + + /* write microcode via MSR 0x79 */ + native_wrmsr(MSR_IA32_UCODE_WRITE, + (unsigned long) mc_intel->bits, + (unsigned long) mc_intel->bits >> 16 >> 16); + native_wrmsr(MSR_IA32_UCODE_REV, 0, 0); + + /* As documented in the SDM: Do a CPUID 1 here */ + sync_core(); + + /* get the current revision from MSR 0x8B */ + native_rdmsr(MSR_IA32_UCODE_REV, val[0], val[1]); + if (val[1] != mc_intel->hdr.rev) + return -1; + +#ifdef CONFIG_X86_64 + /* Flush global tlb. This is precaution. */ + flush_tlb_early(); +#endif + uci->cpu_sig.rev = val[1]; + + if (early) + print_ucode(uci); + else + print_ucode_info(uci, mc_intel->hdr.date); + + return 0; +} + +/* + * This function converts microcode patch offsets previously stored in + * mc_saved_in_initrd to pointers and stores the pointers in mc_saved_data. + */ +int __init save_microcode_in_initrd_intel(void) +{ + unsigned int count = mc_saved_data.mc_saved_count; + struct microcode_intel *mc_saved[MAX_UCODE_COUNT]; + int ret = 0; + + if (count == 0) + return ret; + + copy_initrd_ptrs(mc_saved, mc_saved_in_initrd, initrd_start, count); + ret = save_microcode(&mc_saved_data, mc_saved, count); + if (ret) + pr_err("Cannot save microcode patches from initrd.\n"); + + show_saved_mc(); + + return ret; +} + +static void __init +_load_ucode_intel_bsp(struct mc_saved_data *mc_saved_data, + unsigned long *initrd, + unsigned long start, unsigned long size) +{ + struct ucode_cpu_info uci; + enum ucode_state ret; + + collect_cpu_info_early(&uci); + + ret = scan_microcode(mc_saved_data, initrd, start, size, &uci); + if (ret != UCODE_OK) + return; + + ret = load_microcode(mc_saved_data, initrd, start, &uci); + if (ret != UCODE_OK) + return; + + apply_microcode_early(&uci, true); +} + +void __init load_ucode_intel_bsp(void) +{ + u64 start, size; +#ifdef CONFIG_X86_32 + struct boot_params *p; + + p = (struct boot_params *)__pa_nodebug(&boot_params); + start = p->hdr.ramdisk_image; + size = p->hdr.ramdisk_size; + + _load_ucode_intel_bsp( + (struct mc_saved_data *)__pa_nodebug(&mc_saved_data), + (unsigned long *)__pa_nodebug(&mc_saved_in_initrd), + start, size); +#else + start = boot_params.hdr.ramdisk_image + PAGE_OFFSET; + size = boot_params.hdr.ramdisk_size; + + _load_ucode_intel_bsp(&mc_saved_data, mc_saved_in_initrd, start, size); +#endif +} + +void load_ucode_intel_ap(void) +{ + struct mc_saved_data *mc_saved_data_p; + struct ucode_cpu_info uci; + unsigned long *mc_saved_in_initrd_p; + unsigned long initrd_start_addr; + enum ucode_state ret; +#ifdef CONFIG_X86_32 + unsigned long *initrd_start_p; + + mc_saved_in_initrd_p = + (unsigned long *)__pa_nodebug(mc_saved_in_initrd); + mc_saved_data_p = (struct mc_saved_data *)__pa_nodebug(&mc_saved_data); + initrd_start_p = (unsigned long *)__pa_nodebug(&initrd_start); + initrd_start_addr = (unsigned long)__pa_nodebug(*initrd_start_p); +#else + mc_saved_data_p = &mc_saved_data; + mc_saved_in_initrd_p = mc_saved_in_initrd; + initrd_start_addr = initrd_start; +#endif + + /* + * If there is no valid ucode previously saved in memory, no need to + * update ucode on this AP. + */ + if (mc_saved_data_p->mc_saved_count == 0) + return; + + collect_cpu_info_early(&uci); + ret = load_microcode(mc_saved_data_p, mc_saved_in_initrd_p, + initrd_start_addr, &uci); + + if (ret != UCODE_OK) + return; + + apply_microcode_early(&uci, true); +} + +void reload_ucode_intel(void) +{ + struct ucode_cpu_info uci; + enum ucode_state ret; + + if (!mc_saved_data.mc_saved_count) + return; + + collect_cpu_info_early(&uci); + + ret = load_microcode_early(mc_saved_data.mc_saved, + mc_saved_data.mc_saved_count, &uci); + if (ret != UCODE_OK) + return; + + apply_microcode_early(&uci, false); +} static int collect_cpu_info(int cpu_num, struct cpu_signature *csig) { @@ -124,7 +840,7 @@ static int get_matching_mc(struct microcode_intel *mc_intel, int cpu) cpf = cpu_sig.pf; crev = cpu_sig.rev; - return get_matching_microcode(csig, cpf, crev, mc_intel); + return has_newer_microcode(mc_intel, csig, cpf, crev); } static int apply_microcode_intel(int cpu) @@ -226,7 +942,7 @@ static enum ucode_state generic_load_microcode(int cpu, void *data, size_t size, csig = uci->cpu_sig.sig; cpf = uci->cpu_sig.pf; - if (get_matching_microcode(csig, cpf, new_rev, mc)) { + if (has_newer_microcode(mc, csig, cpf, new_rev)) { vfree(new_mc); new_rev = mc_header.rev; new_mc = mc; @@ -325,7 +1041,7 @@ static struct microcode_ops microcode_intel_ops = { struct microcode_ops * __init init_intel_microcode(void) { - struct cpuinfo_x86 *c = &cpu_data(0); + struct cpuinfo_x86 *c = &boot_cpu_data; if (c->x86_vendor != X86_VENDOR_INTEL || c->x86 < 6 || cpu_has(c, X86_FEATURE_IA64)) { diff --git a/kernel/arch/x86/kernel/cpu/microcode/intel_early.c b/kernel/arch/x86/kernel/cpu/microcode/intel_early.c deleted file mode 100644 index 2f49ab4ac..000000000 --- a/kernel/arch/x86/kernel/cpu/microcode/intel_early.c +++ /dev/null @@ -1,786 +0,0 @@ -/* - * Intel CPU microcode early update for Linux - * - * Copyright (C) 2012 Fenghua Yu <fenghua.yu@intel.com> - * H Peter Anvin" <hpa@zytor.com> - * - * This allows to early upgrade microcode on Intel processors - * belonging to IA-32 family - PentiumPro, Pentium II, - * Pentium III, Xeon, Pentium 4, etc. - * - * Reference: Section 9.11 of Volume 3, IA-32 Intel Architecture - * Software Developer's Manual. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -/* - * This needs to be before all headers so that pr_debug in printk.h doesn't turn - * printk calls into no_printk(). - * - *#define DEBUG - */ - -#include <linux/module.h> -#include <linux/mm.h> -#include <linux/slab.h> -#include <linux/earlycpio.h> -#include <linux/initrd.h> -#include <linux/cpu.h> -#include <asm/msr.h> -#include <asm/microcode_intel.h> -#include <asm/processor.h> -#include <asm/tlbflush.h> -#include <asm/setup.h> - -#undef pr_fmt -#define pr_fmt(fmt) "microcode: " fmt - -static unsigned long mc_saved_in_initrd[MAX_UCODE_COUNT]; -static struct mc_saved_data { - unsigned int mc_saved_count; - struct microcode_intel **mc_saved; -} mc_saved_data; - -static enum ucode_state -load_microcode_early(struct microcode_intel **saved, - unsigned int num_saved, struct ucode_cpu_info *uci) -{ - struct microcode_intel *ucode_ptr, *new_mc = NULL; - struct microcode_header_intel *mc_hdr; - int new_rev, ret, i; - - new_rev = uci->cpu_sig.rev; - - for (i = 0; i < num_saved; i++) { - ucode_ptr = saved[i]; - mc_hdr = (struct microcode_header_intel *)ucode_ptr; - - ret = get_matching_microcode(uci->cpu_sig.sig, - uci->cpu_sig.pf, - new_rev, - ucode_ptr); - if (!ret) - continue; - - new_rev = mc_hdr->rev; - new_mc = ucode_ptr; - } - - if (!new_mc) - return UCODE_NFOUND; - - uci->mc = (struct microcode_intel *)new_mc; - return UCODE_OK; -} - -static inline void -copy_initrd_ptrs(struct microcode_intel **mc_saved, unsigned long *initrd, - unsigned long off, int num_saved) -{ - int i; - - for (i = 0; i < num_saved; i++) - mc_saved[i] = (struct microcode_intel *)(initrd[i] + off); -} - -#ifdef CONFIG_X86_32 -static void -microcode_phys(struct microcode_intel **mc_saved_tmp, - struct mc_saved_data *mc_saved_data) -{ - int i; - struct microcode_intel ***mc_saved; - - mc_saved = (struct microcode_intel ***) - __pa_nodebug(&mc_saved_data->mc_saved); - for (i = 0; i < mc_saved_data->mc_saved_count; i++) { - struct microcode_intel *p; - - p = *(struct microcode_intel **) - __pa_nodebug(mc_saved_data->mc_saved + i); - mc_saved_tmp[i] = (struct microcode_intel *)__pa_nodebug(p); - } -} -#endif - -static enum ucode_state -load_microcode(struct mc_saved_data *mc_saved_data, unsigned long *initrd, - unsigned long initrd_start, struct ucode_cpu_info *uci) -{ - struct microcode_intel *mc_saved_tmp[MAX_UCODE_COUNT]; - unsigned int count = mc_saved_data->mc_saved_count; - - if (!mc_saved_data->mc_saved) { - copy_initrd_ptrs(mc_saved_tmp, initrd, initrd_start, count); - - return load_microcode_early(mc_saved_tmp, count, uci); - } else { -#ifdef CONFIG_X86_32 - microcode_phys(mc_saved_tmp, mc_saved_data); - return load_microcode_early(mc_saved_tmp, count, uci); -#else - return load_microcode_early(mc_saved_data->mc_saved, - count, uci); -#endif - } -} - -/* - * Given CPU signature and a microcode patch, this function finds if the - * microcode patch has matching family and model with the CPU. - */ -static enum ucode_state -matching_model_microcode(struct microcode_header_intel *mc_header, - unsigned long sig) -{ - unsigned int fam, model; - unsigned int fam_ucode, model_ucode; - struct extended_sigtable *ext_header; - unsigned long total_size = get_totalsize(mc_header); - unsigned long data_size = get_datasize(mc_header); - int ext_sigcount, i; - struct extended_signature *ext_sig; - - fam = __x86_family(sig); - model = x86_model(sig); - - fam_ucode = __x86_family(mc_header->sig); - model_ucode = x86_model(mc_header->sig); - - if (fam == fam_ucode && model == model_ucode) - return UCODE_OK; - - /* Look for ext. headers: */ - if (total_size <= data_size + MC_HEADER_SIZE) - return UCODE_NFOUND; - - ext_header = (void *) mc_header + data_size + MC_HEADER_SIZE; - ext_sig = (void *)ext_header + EXT_HEADER_SIZE; - ext_sigcount = ext_header->count; - - for (i = 0; i < ext_sigcount; i++) { - fam_ucode = __x86_family(ext_sig->sig); - model_ucode = x86_model(ext_sig->sig); - - if (fam == fam_ucode && model == model_ucode) - return UCODE_OK; - - ext_sig++; - } - return UCODE_NFOUND; -} - -static int -save_microcode(struct mc_saved_data *mc_saved_data, - struct microcode_intel **mc_saved_src, - unsigned int mc_saved_count) -{ - int i, j; - struct microcode_intel **saved_ptr; - int ret; - - if (!mc_saved_count) - return -EINVAL; - - /* - * Copy new microcode data. - */ - saved_ptr = kcalloc(mc_saved_count, sizeof(struct microcode_intel *), GFP_KERNEL); - if (!saved_ptr) - return -ENOMEM; - - for (i = 0; i < mc_saved_count; i++) { - struct microcode_header_intel *mc_hdr; - struct microcode_intel *mc; - unsigned long size; - - if (!mc_saved_src[i]) { - ret = -EINVAL; - goto err; - } - - mc = mc_saved_src[i]; - mc_hdr = &mc->hdr; - size = get_totalsize(mc_hdr); - - saved_ptr[i] = kmalloc(size, GFP_KERNEL); - if (!saved_ptr[i]) { - ret = -ENOMEM; - goto err; - } - - memcpy(saved_ptr[i], mc, size); - } - - /* - * Point to newly saved microcode. - */ - mc_saved_data->mc_saved = saved_ptr; - mc_saved_data->mc_saved_count = mc_saved_count; - - return 0; - -err: - for (j = 0; j <= i; j++) - kfree(saved_ptr[j]); - kfree(saved_ptr); - - return ret; -} - -/* - * A microcode patch in ucode_ptr is saved into mc_saved - * - if it has matching signature and newer revision compared to an existing - * patch mc_saved. - * - or if it is a newly discovered microcode patch. - * - * The microcode patch should have matching model with CPU. - * - * Returns: The updated number @num_saved of saved microcode patches. - */ -static unsigned int _save_mc(struct microcode_intel **mc_saved, - u8 *ucode_ptr, unsigned int num_saved) -{ - struct microcode_header_intel *mc_hdr, *mc_saved_hdr; - unsigned int sig, pf, new_rev; - int found = 0, i; - - mc_hdr = (struct microcode_header_intel *)ucode_ptr; - - for (i = 0; i < num_saved; i++) { - mc_saved_hdr = (struct microcode_header_intel *)mc_saved[i]; - sig = mc_saved_hdr->sig; - pf = mc_saved_hdr->pf; - new_rev = mc_hdr->rev; - - if (!get_matching_sig(sig, pf, new_rev, ucode_ptr)) - continue; - - found = 1; - - if (!revision_is_newer(mc_hdr, new_rev)) - continue; - - /* - * Found an older ucode saved earlier. Replace it with - * this newer one. - */ - mc_saved[i] = (struct microcode_intel *)ucode_ptr; - break; - } - - /* Newly detected microcode, save it to memory. */ - if (i >= num_saved && !found) - mc_saved[num_saved++] = (struct microcode_intel *)ucode_ptr; - - return num_saved; -} - -/* - * Get microcode matching with BSP's model. Only CPUs with the same model as - * BSP can stay in the platform. - */ -static enum ucode_state __init -get_matching_model_microcode(int cpu, unsigned long start, - void *data, size_t size, - struct mc_saved_data *mc_saved_data, - unsigned long *mc_saved_in_initrd, - struct ucode_cpu_info *uci) -{ - u8 *ucode_ptr = data; - unsigned int leftover = size; - enum ucode_state state = UCODE_OK; - unsigned int mc_size; - struct microcode_header_intel *mc_header; - struct microcode_intel *mc_saved_tmp[MAX_UCODE_COUNT]; - unsigned int mc_saved_count = mc_saved_data->mc_saved_count; - int i; - - while (leftover && mc_saved_count < ARRAY_SIZE(mc_saved_tmp)) { - - if (leftover < sizeof(mc_header)) - break; - - mc_header = (struct microcode_header_intel *)ucode_ptr; - - mc_size = get_totalsize(mc_header); - if (!mc_size || mc_size > leftover || - microcode_sanity_check(ucode_ptr, 0) < 0) - break; - - leftover -= mc_size; - - /* - * Since APs with same family and model as the BSP may boot in - * the platform, we need to find and save microcode patches - * with the same family and model as the BSP. - */ - if (matching_model_microcode(mc_header, uci->cpu_sig.sig) != - UCODE_OK) { - ucode_ptr += mc_size; - continue; - } - - mc_saved_count = _save_mc(mc_saved_tmp, ucode_ptr, mc_saved_count); - - ucode_ptr += mc_size; - } - - if (leftover) { - state = UCODE_ERROR; - goto out; - } - - if (mc_saved_count == 0) { - state = UCODE_NFOUND; - goto out; - } - - for (i = 0; i < mc_saved_count; i++) - mc_saved_in_initrd[i] = (unsigned long)mc_saved_tmp[i] - start; - - mc_saved_data->mc_saved_count = mc_saved_count; -out: - return state; -} - -static int collect_cpu_info_early(struct ucode_cpu_info *uci) -{ - unsigned int val[2]; - unsigned int family, model; - struct cpu_signature csig; - unsigned int eax, ebx, ecx, edx; - - csig.sig = 0; - csig.pf = 0; - csig.rev = 0; - - memset(uci, 0, sizeof(*uci)); - - eax = 0x00000001; - ecx = 0; - native_cpuid(&eax, &ebx, &ecx, &edx); - csig.sig = eax; - - family = __x86_family(csig.sig); - model = x86_model(csig.sig); - - if ((model >= 5) || (family > 6)) { - /* get processor flags from MSR 0x17 */ - native_rdmsr(MSR_IA32_PLATFORM_ID, val[0], val[1]); - csig.pf = 1 << ((val[1] >> 18) & 7); - } - native_wrmsr(MSR_IA32_UCODE_REV, 0, 0); - - /* As documented in the SDM: Do a CPUID 1 here */ - sync_core(); - - /* get the current revision from MSR 0x8B */ - native_rdmsr(MSR_IA32_UCODE_REV, val[0], val[1]); - - csig.rev = val[1]; - - uci->cpu_sig = csig; - uci->valid = 1; - - return 0; -} - -#ifdef DEBUG -static void __ref show_saved_mc(void) -{ - int i, j; - unsigned int sig, pf, rev, total_size, data_size, date; - struct ucode_cpu_info uci; - - if (mc_saved_data.mc_saved_count == 0) { - pr_debug("no microcode data saved.\n"); - return; - } - pr_debug("Total microcode saved: %d\n", mc_saved_data.mc_saved_count); - - collect_cpu_info_early(&uci); - - sig = uci.cpu_sig.sig; - pf = uci.cpu_sig.pf; - rev = uci.cpu_sig.rev; - pr_debug("CPU: sig=0x%x, pf=0x%x, rev=0x%x\n", sig, pf, rev); - - for (i = 0; i < mc_saved_data.mc_saved_count; i++) { - struct microcode_header_intel *mc_saved_header; - struct extended_sigtable *ext_header; - int ext_sigcount; - struct extended_signature *ext_sig; - - mc_saved_header = (struct microcode_header_intel *) - mc_saved_data.mc_saved[i]; - sig = mc_saved_header->sig; - pf = mc_saved_header->pf; - rev = mc_saved_header->rev; - total_size = get_totalsize(mc_saved_header); - data_size = get_datasize(mc_saved_header); - date = mc_saved_header->date; - - pr_debug("mc_saved[%d]: sig=0x%x, pf=0x%x, rev=0x%x, toal size=0x%x, date = %04x-%02x-%02x\n", - i, sig, pf, rev, total_size, - date & 0xffff, - date >> 24, - (date >> 16) & 0xff); - - /* Look for ext. headers: */ - if (total_size <= data_size + MC_HEADER_SIZE) - continue; - - ext_header = (void *) mc_saved_header + data_size + MC_HEADER_SIZE; - ext_sigcount = ext_header->count; - ext_sig = (void *)ext_header + EXT_HEADER_SIZE; - - for (j = 0; j < ext_sigcount; j++) { - sig = ext_sig->sig; - pf = ext_sig->pf; - - pr_debug("\tExtended[%d]: sig=0x%x, pf=0x%x\n", - j, sig, pf); - - ext_sig++; - } - - } -} -#else -static inline void show_saved_mc(void) -{ -} -#endif - -#if defined(CONFIG_MICROCODE_INTEL_EARLY) && defined(CONFIG_HOTPLUG_CPU) -static DEFINE_MUTEX(x86_cpu_microcode_mutex); -/* - * Save this mc into mc_saved_data. So it will be loaded early when a CPU is - * hot added or resumes. - * - * Please make sure this mc should be a valid microcode patch before calling - * this function. - */ -int save_mc_for_early(u8 *mc) -{ - struct microcode_intel *mc_saved_tmp[MAX_UCODE_COUNT]; - unsigned int mc_saved_count_init; - unsigned int mc_saved_count; - struct microcode_intel **mc_saved; - int ret = 0; - int i; - - /* - * Hold hotplug lock so mc_saved_data is not accessed by a CPU in - * hotplug. - */ - mutex_lock(&x86_cpu_microcode_mutex); - - mc_saved_count_init = mc_saved_data.mc_saved_count; - mc_saved_count = mc_saved_data.mc_saved_count; - mc_saved = mc_saved_data.mc_saved; - - if (mc_saved && mc_saved_count) - memcpy(mc_saved_tmp, mc_saved, - mc_saved_count * sizeof(struct microcode_intel *)); - /* - * Save the microcode patch mc in mc_save_tmp structure if it's a newer - * version. - */ - mc_saved_count = _save_mc(mc_saved_tmp, mc, mc_saved_count); - - /* - * Save the mc_save_tmp in global mc_saved_data. - */ - ret = save_microcode(&mc_saved_data, mc_saved_tmp, mc_saved_count); - if (ret) { - pr_err("Cannot save microcode patch.\n"); - goto out; - } - - show_saved_mc(); - - /* - * Free old saved microcode data. - */ - if (mc_saved) { - for (i = 0; i < mc_saved_count_init; i++) - kfree(mc_saved[i]); - kfree(mc_saved); - } - -out: - mutex_unlock(&x86_cpu_microcode_mutex); - - return ret; -} -EXPORT_SYMBOL_GPL(save_mc_for_early); -#endif - -static __initdata char ucode_name[] = "kernel/x86/microcode/GenuineIntel.bin"; -static __init enum ucode_state -scan_microcode(struct mc_saved_data *mc_saved_data, unsigned long *initrd, - unsigned long start, unsigned long size, - struct ucode_cpu_info *uci) -{ - struct cpio_data cd; - long offset = 0; -#ifdef CONFIG_X86_32 - char *p = (char *)__pa_nodebug(ucode_name); -#else - char *p = ucode_name; -#endif - - cd.data = NULL; - cd.size = 0; - - cd = find_cpio_data(p, (void *)start, size, &offset); - if (!cd.data) - return UCODE_ERROR; - - return get_matching_model_microcode(0, start, cd.data, cd.size, - mc_saved_data, initrd, uci); -} - -/* - * Print ucode update info. - */ -static void -print_ucode_info(struct ucode_cpu_info *uci, unsigned int date) -{ - int cpu = smp_processor_id(); - - pr_info("CPU%d microcode updated early to revision 0x%x, date = %04x-%02x-%02x\n", - cpu, - uci->cpu_sig.rev, - date & 0xffff, - date >> 24, - (date >> 16) & 0xff); -} - -#ifdef CONFIG_X86_32 - -static int delay_ucode_info; -static int current_mc_date; - -/* - * Print early updated ucode info after printk works. This is delayed info dump. - */ -void show_ucode_info_early(void) -{ - struct ucode_cpu_info uci; - - if (delay_ucode_info) { - collect_cpu_info_early(&uci); - print_ucode_info(&uci, current_mc_date); - delay_ucode_info = 0; - } -} - -/* - * At this point, we can not call printk() yet. Keep microcode patch number in - * mc_saved_data.mc_saved and delay printing microcode info in - * show_ucode_info_early() until printk() works. - */ -static void print_ucode(struct ucode_cpu_info *uci) -{ - struct microcode_intel *mc_intel; - int *delay_ucode_info_p; - int *current_mc_date_p; - - mc_intel = uci->mc; - if (mc_intel == NULL) - return; - - delay_ucode_info_p = (int *)__pa_nodebug(&delay_ucode_info); - current_mc_date_p = (int *)__pa_nodebug(¤t_mc_date); - - *delay_ucode_info_p = 1; - *current_mc_date_p = mc_intel->hdr.date; -} -#else - -/* - * Flush global tlb. We only do this in x86_64 where paging has been enabled - * already and PGE should be enabled as well. - */ -static inline void flush_tlb_early(void) -{ - __native_flush_tlb_global_irq_disabled(); -} - -static inline void print_ucode(struct ucode_cpu_info *uci) -{ - struct microcode_intel *mc_intel; - - mc_intel = uci->mc; - if (mc_intel == NULL) - return; - - print_ucode_info(uci, mc_intel->hdr.date); -} -#endif - -static int apply_microcode_early(struct ucode_cpu_info *uci, bool early) -{ - struct microcode_intel *mc_intel; - unsigned int val[2]; - - mc_intel = uci->mc; - if (mc_intel == NULL) - return 0; - - /* write microcode via MSR 0x79 */ - native_wrmsr(MSR_IA32_UCODE_WRITE, - (unsigned long) mc_intel->bits, - (unsigned long) mc_intel->bits >> 16 >> 16); - native_wrmsr(MSR_IA32_UCODE_REV, 0, 0); - - /* As documented in the SDM: Do a CPUID 1 here */ - sync_core(); - - /* get the current revision from MSR 0x8B */ - native_rdmsr(MSR_IA32_UCODE_REV, val[0], val[1]); - if (val[1] != mc_intel->hdr.rev) - return -1; - -#ifdef CONFIG_X86_64 - /* Flush global tlb. This is precaution. */ - flush_tlb_early(); -#endif - uci->cpu_sig.rev = val[1]; - - if (early) - print_ucode(uci); - else - print_ucode_info(uci, mc_intel->hdr.date); - - return 0; -} - -/* - * This function converts microcode patch offsets previously stored in - * mc_saved_in_initrd to pointers and stores the pointers in mc_saved_data. - */ -int __init save_microcode_in_initrd_intel(void) -{ - unsigned int count = mc_saved_data.mc_saved_count; - struct microcode_intel *mc_saved[MAX_UCODE_COUNT]; - int ret = 0; - - if (count == 0) - return ret; - - copy_initrd_ptrs(mc_saved, mc_saved_in_initrd, initrd_start, count); - ret = save_microcode(&mc_saved_data, mc_saved, count); - if (ret) - pr_err("Cannot save microcode patches from initrd.\n"); - - show_saved_mc(); - - return ret; -} - -static void __init -_load_ucode_intel_bsp(struct mc_saved_data *mc_saved_data, - unsigned long *initrd, - unsigned long start, unsigned long size) -{ - struct ucode_cpu_info uci; - enum ucode_state ret; - - collect_cpu_info_early(&uci); - - ret = scan_microcode(mc_saved_data, initrd, start, size, &uci); - if (ret != UCODE_OK) - return; - - ret = load_microcode(mc_saved_data, initrd, start, &uci); - if (ret != UCODE_OK) - return; - - apply_microcode_early(&uci, true); -} - -void __init load_ucode_intel_bsp(void) -{ - u64 start, size; -#ifdef CONFIG_X86_32 - struct boot_params *p; - - p = (struct boot_params *)__pa_nodebug(&boot_params); - start = p->hdr.ramdisk_image; - size = p->hdr.ramdisk_size; - - _load_ucode_intel_bsp( - (struct mc_saved_data *)__pa_nodebug(&mc_saved_data), - (unsigned long *)__pa_nodebug(&mc_saved_in_initrd), - start, size); -#else - start = boot_params.hdr.ramdisk_image + PAGE_OFFSET; - size = boot_params.hdr.ramdisk_size; - - _load_ucode_intel_bsp(&mc_saved_data, mc_saved_in_initrd, start, size); -#endif -} - -void load_ucode_intel_ap(void) -{ - struct mc_saved_data *mc_saved_data_p; - struct ucode_cpu_info uci; - unsigned long *mc_saved_in_initrd_p; - unsigned long initrd_start_addr; - enum ucode_state ret; -#ifdef CONFIG_X86_32 - unsigned long *initrd_start_p; - - mc_saved_in_initrd_p = - (unsigned long *)__pa_nodebug(mc_saved_in_initrd); - mc_saved_data_p = (struct mc_saved_data *)__pa_nodebug(&mc_saved_data); - initrd_start_p = (unsigned long *)__pa_nodebug(&initrd_start); - initrd_start_addr = (unsigned long)__pa_nodebug(*initrd_start_p); -#else - mc_saved_data_p = &mc_saved_data; - mc_saved_in_initrd_p = mc_saved_in_initrd; - initrd_start_addr = initrd_start; -#endif - - /* - * If there is no valid ucode previously saved in memory, no need to - * update ucode on this AP. - */ - if (mc_saved_data_p->mc_saved_count == 0) - return; - - collect_cpu_info_early(&uci); - ret = load_microcode(mc_saved_data_p, mc_saved_in_initrd_p, - initrd_start_addr, &uci); - - if (ret != UCODE_OK) - return; - - apply_microcode_early(&uci, true); -} - -void reload_ucode_intel(void) -{ - struct ucode_cpu_info uci; - enum ucode_state ret; - - if (!mc_saved_data.mc_saved_count) - return; - - collect_cpu_info_early(&uci); - - ret = load_microcode_early(mc_saved_data.mc_saved, - mc_saved_data.mc_saved_count, &uci); - if (ret != UCODE_OK) - return; - - apply_microcode_early(&uci, false); -} diff --git a/kernel/arch/x86/kernel/cpu/microcode/intel_lib.c b/kernel/arch/x86/kernel/cpu/microcode/intel_lib.c index cd47a510a..b96896bcb 100644 --- a/kernel/arch/x86/kernel/cpu/microcode/intel_lib.c +++ b/kernel/arch/x86/kernel/cpu/microcode/intel_lib.c @@ -25,17 +25,23 @@ #include <linux/firmware.h> #include <linux/uaccess.h> #include <linux/kernel.h> -#include <linux/module.h> #include <asm/microcode_intel.h> #include <asm/processor.h> #include <asm/msr.h> -static inline int -update_match_cpu(unsigned int csig, unsigned int cpf, - unsigned int sig, unsigned int pf) +static inline bool cpu_signatures_match(unsigned int s1, unsigned int p1, + unsigned int s2, unsigned int p2) { - return (!sigmatch(sig, csig, pf, cpf)) ? 0 : 1; + if (s1 != s2) + return false; + + /* Processor flags are either both 0 ... */ + if (!p1 && !p2) + return true; + + /* ... or they intersect. */ + return p1 & p2; } int microcode_sanity_check(void *mc, int print_err) @@ -124,27 +130,25 @@ EXPORT_SYMBOL_GPL(microcode_sanity_check); /* * Returns 1 if update has been found, 0 otherwise. */ -int get_matching_sig(unsigned int csig, int cpf, int rev, void *mc) +int find_matching_signature(void *mc, unsigned int csig, int cpf) { - struct microcode_header_intel *mc_header = mc; - struct extended_sigtable *ext_header; - unsigned long total_size = get_totalsize(mc_header); - int ext_sigcount, i; + struct microcode_header_intel *mc_hdr = mc; + struct extended_sigtable *ext_hdr; struct extended_signature *ext_sig; + int i; - if (update_match_cpu(csig, cpf, mc_header->sig, mc_header->pf)) + if (cpu_signatures_match(csig, cpf, mc_hdr->sig, mc_hdr->pf)) return 1; /* Look for ext. headers: */ - if (total_size <= get_datasize(mc_header) + MC_HEADER_SIZE) + if (get_totalsize(mc_hdr) <= get_datasize(mc_hdr) + MC_HEADER_SIZE) return 0; - ext_header = mc + get_datasize(mc_header) + MC_HEADER_SIZE; - ext_sigcount = ext_header->count; - ext_sig = (void *)ext_header + EXT_HEADER_SIZE; + ext_hdr = mc + get_datasize(mc_hdr) + MC_HEADER_SIZE; + ext_sig = (void *)ext_hdr + EXT_HEADER_SIZE; - for (i = 0; i < ext_sigcount; i++) { - if (update_match_cpu(csig, cpf, ext_sig->sig, ext_sig->pf)) + for (i = 0; i < ext_hdr->count; i++) { + if (cpu_signatures_match(csig, cpf, ext_sig->sig, ext_sig->pf)) return 1; ext_sig++; } @@ -154,13 +158,13 @@ int get_matching_sig(unsigned int csig, int cpf, int rev, void *mc) /* * Returns 1 if update has been found, 0 otherwise. */ -int get_matching_microcode(unsigned int csig, int cpf, int rev, void *mc) +int has_newer_microcode(void *mc, unsigned int csig, int cpf, int new_rev) { struct microcode_header_intel *mc_hdr = mc; - if (!revision_is_newer(mc_hdr, rev)) + if (mc_hdr->rev <= new_rev) return 0; - return get_matching_sig(csig, cpf, rev, mc); + return find_matching_signature(mc, csig, cpf); } -EXPORT_SYMBOL_GPL(get_matching_microcode); +EXPORT_SYMBOL_GPL(has_newer_microcode); diff --git a/kernel/arch/x86/kernel/cpu/mshyperv.c b/kernel/arch/x86/kernel/cpu/mshyperv.c index 939155ffd..20e242ea1 100644 --- a/kernel/arch/x86/kernel/cpu/mshyperv.c +++ b/kernel/arch/x86/kernel/cpu/mshyperv.c @@ -18,6 +18,7 @@ #include <linux/efi.h> #include <linux/interrupt.h> #include <linux/irq.h> +#include <linux/kexec.h> #include <asm/processor.h> #include <asm/hypervisor.h> #include <asm/hyperv.h> @@ -28,25 +29,26 @@ #include <asm/i8259.h> #include <asm/apic.h> #include <asm/timer.h> +#include <asm/reboot.h> struct ms_hyperv_info ms_hyperv; EXPORT_SYMBOL_GPL(ms_hyperv); #if IS_ENABLED(CONFIG_HYPERV) static void (*vmbus_handler)(void); +static void (*hv_kexec_handler)(void); +static void (*hv_crash_handler)(struct pt_regs *regs); void hyperv_vector_handler(struct pt_regs *regs) { struct pt_regs *old_regs = set_irq_regs(regs); - irq_enter(); - exit_idle(); - + entering_irq(); inc_irq_stat(irq_hv_callback_count); if (vmbus_handler) vmbus_handler(); - irq_exit(); + exiting_irq(); set_irq_regs(old_regs); } @@ -69,7 +71,47 @@ void hv_remove_vmbus_irq(void) } EXPORT_SYMBOL_GPL(hv_setup_vmbus_irq); EXPORT_SYMBOL_GPL(hv_remove_vmbus_irq); -#endif + +void hv_setup_kexec_handler(void (*handler)(void)) +{ + hv_kexec_handler = handler; +} +EXPORT_SYMBOL_GPL(hv_setup_kexec_handler); + +void hv_remove_kexec_handler(void) +{ + hv_kexec_handler = NULL; +} +EXPORT_SYMBOL_GPL(hv_remove_kexec_handler); + +void hv_setup_crash_handler(void (*handler)(struct pt_regs *regs)) +{ + hv_crash_handler = handler; +} +EXPORT_SYMBOL_GPL(hv_setup_crash_handler); + +void hv_remove_crash_handler(void) +{ + hv_crash_handler = NULL; +} +EXPORT_SYMBOL_GPL(hv_remove_crash_handler); + +#ifdef CONFIG_KEXEC_CORE +static void hv_machine_shutdown(void) +{ + if (kexec_in_progress && hv_kexec_handler) + hv_kexec_handler(); + native_machine_shutdown(); +} + +static void hv_machine_crash_shutdown(struct pt_regs *regs) +{ + if (hv_crash_handler) + hv_crash_handler(regs); + native_machine_crash_shutdown(regs); +} +#endif /* CONFIG_KEXEC_CORE */ +#endif /* CONFIG_HYPERV */ static uint32_t __init ms_hyperv_platform(void) { @@ -116,6 +158,7 @@ static void __init ms_hyperv_init_platform(void) * Extract the features and hints */ ms_hyperv.features = cpuid_eax(HYPERV_CPUID_FEATURES); + ms_hyperv.misc_features = cpuid_edx(HYPERV_CPUID_FEATURES); ms_hyperv.hints = cpuid_eax(HYPERV_CPUID_ENLIGHTMENT_INFO); printk(KERN_INFO "HyperV: features 0x%x, hints 0x%x\n", @@ -143,6 +186,11 @@ static void __init ms_hyperv_init_platform(void) no_timer_check = 1; #endif +#if IS_ENABLED(CONFIG_HYPERV) && defined(CONFIG_KEXEC_CORE) + machine_ops.shutdown = hv_machine_shutdown; + machine_ops.crash_shutdown = hv_machine_crash_shutdown; +#endif + mark_tsc_unstable("running on Hyper-V"); } const __refconst struct hypervisor_x86 x86_hyper_ms_hyperv = { diff --git a/kernel/arch/x86/kernel/cpu/mtrr/cleanup.c b/kernel/arch/x86/kernel/cpu/mtrr/cleanup.c index 5f90b85ff..70d7c93f4 100644 --- a/kernel/arch/x86/kernel/cpu/mtrr/cleanup.c +++ b/kernel/arch/x86/kernel/cpu/mtrr/cleanup.c @@ -98,7 +98,8 @@ x86_get_mtrr_mem_range(struct range *range, int nr_range, continue; base = range_state[i].base_pfn; if (base < (1<<(20-PAGE_SHIFT)) && mtrr_state.have_fixed && - (mtrr_state.enabled & 1)) { + (mtrr_state.enabled & MTRR_STATE_MTRR_ENABLED) && + (mtrr_state.enabled & MTRR_STATE_MTRR_FIXED_ENABLED)) { /* Var MTRR contains UC entry below 1M? Skip it: */ printk(BIOS_BUG_MSG, i); if (base + size <= (1<<(20-PAGE_SHIFT))) diff --git a/kernel/arch/x86/kernel/cpu/mtrr/generic.c b/kernel/arch/x86/kernel/cpu/mtrr/generic.c index 7d74f7b3c..3b533cf37 100644 --- a/kernel/arch/x86/kernel/cpu/mtrr/generic.c +++ b/kernel/arch/x86/kernel/cpu/mtrr/generic.c @@ -102,59 +102,76 @@ static int check_type_overlap(u8 *prev, u8 *curr) return 0; } -/* - * Error/Semi-error returns: - * 0xFF - when MTRR is not enabled - * *repeat == 1 implies [start:end] spanned across MTRR range and type returned - * corresponds only to [start:*partial_end]. - * Caller has to lookup again for [*partial_end:end]. +/** + * mtrr_type_lookup_fixed - look up memory type in MTRR fixed entries + * + * Return the MTRR fixed memory type of 'start'. + * + * MTRR fixed entries are divided into the following ways: + * 0x00000 - 0x7FFFF : This range is divided into eight 64KB sub-ranges + * 0x80000 - 0xBFFFF : This range is divided into sixteen 16KB sub-ranges + * 0xC0000 - 0xFFFFF : This range is divided into sixty-four 4KB sub-ranges + * + * Return Values: + * MTRR_TYPE_(type) - Matched memory type + * MTRR_TYPE_INVALID - Unmatched + */ +static u8 mtrr_type_lookup_fixed(u64 start, u64 end) +{ + int idx; + + if (start >= 0x100000) + return MTRR_TYPE_INVALID; + + /* 0x0 - 0x7FFFF */ + if (start < 0x80000) { + idx = 0; + idx += (start >> 16); + return mtrr_state.fixed_ranges[idx]; + /* 0x80000 - 0xBFFFF */ + } else if (start < 0xC0000) { + idx = 1 * 8; + idx += ((start - 0x80000) >> 14); + return mtrr_state.fixed_ranges[idx]; + } + + /* 0xC0000 - 0xFFFFF */ + idx = 3 * 8; + idx += ((start - 0xC0000) >> 12); + return mtrr_state.fixed_ranges[idx]; +} + +/** + * mtrr_type_lookup_variable - look up memory type in MTRR variable entries + * + * Return Value: + * MTRR_TYPE_(type) - Matched memory type or default memory type (unmatched) + * + * Output Arguments: + * repeat - Set to 1 when [start:end] spanned across MTRR range and type + * returned corresponds only to [start:*partial_end]. Caller has + * to lookup again for [*partial_end:end]. + * + * uniform - Set to 1 when an MTRR covers the region uniformly, i.e. the + * region is fully covered by a single MTRR entry or the default + * type. */ -static u8 __mtrr_type_lookup(u64 start, u64 end, u64 *partial_end, int *repeat) +static u8 mtrr_type_lookup_variable(u64 start, u64 end, u64 *partial_end, + int *repeat, u8 *uniform) { int i; u64 base, mask; u8 prev_match, curr_match; *repeat = 0; - if (!mtrr_state_set) - return 0xFF; - - if (!mtrr_state.enabled) - return 0xFF; + *uniform = 1; - /* Make end inclusive end, instead of exclusive */ + /* Make end inclusive instead of exclusive */ end--; - /* Look in fixed ranges. Just return the type as per start */ - if (mtrr_state.have_fixed && (start < 0x100000)) { - int idx; - - if (start < 0x80000) { - idx = 0; - idx += (start >> 16); - return mtrr_state.fixed_ranges[idx]; - } else if (start < 0xC0000) { - idx = 1 * 8; - idx += ((start - 0x80000) >> 14); - return mtrr_state.fixed_ranges[idx]; - } else if (start < 0x1000000) { - idx = 3 * 8; - idx += ((start - 0xC0000) >> 12); - return mtrr_state.fixed_ranges[idx]; - } - } - - /* - * Look in variable ranges - * Look of multiple ranges matching this address and pick type - * as per MTRR precedence - */ - if (!(mtrr_state.enabled & 2)) - return mtrr_state.def_type; - - prev_match = 0xFF; + prev_match = MTRR_TYPE_INVALID; for (i = 0; i < num_var_ranges; ++i) { - unsigned short start_state, end_state; + unsigned short start_state, end_state, inclusive; if (!(mtrr_state.var_ranges[i].mask_lo & (1 << 11))) continue; @@ -166,20 +183,29 @@ static u8 __mtrr_type_lookup(u64 start, u64 end, u64 *partial_end, int *repeat) start_state = ((start & mask) == (base & mask)); end_state = ((end & mask) == (base & mask)); + inclusive = ((start < base) && (end > base)); - if (start_state != end_state) { + if ((start_state != end_state) || inclusive) { /* * We have start:end spanning across an MTRR. - * We split the region into - * either - * (start:mtrr_end) (mtrr_end:end) - * or - * (start:mtrr_start) (mtrr_start:end) + * We split the region into either + * + * - start_state:1 + * (start:mtrr_end)(mtrr_end:end) + * - end_state:1 + * (start:mtrr_start)(mtrr_start:end) + * - inclusive:1 + * (start:mtrr_start)(mtrr_start:mtrr_end)(mtrr_end:end) + * * depending on kind of overlap. - * Return the type for first region and a pointer to - * the start of second region so that caller will - * lookup again on the second region. - * Note: This way we handle multiple overlaps as well. + * + * Return the type of the first region and a pointer + * to the start of next region so that caller will be + * advised to lookup again after having adjusted start + * and end. + * + * Note: This way we handle overlaps with multiple + * entries and the default type properly. */ if (start_state) *partial_end = base + get_mtrr_size(mask); @@ -193,59 +219,94 @@ static u8 __mtrr_type_lookup(u64 start, u64 end, u64 *partial_end, int *repeat) end = *partial_end - 1; /* end is inclusive */ *repeat = 1; + *uniform = 0; } if ((start & mask) != (base & mask)) continue; curr_match = mtrr_state.var_ranges[i].base_lo & 0xff; - if (prev_match == 0xFF) { + if (prev_match == MTRR_TYPE_INVALID) { prev_match = curr_match; continue; } + *uniform = 0; if (check_type_overlap(&prev_match, &curr_match)) return curr_match; } - if (mtrr_tom2) { - if (start >= (1ULL<<32) && (end < mtrr_tom2)) - return MTRR_TYPE_WRBACK; - } - - if (prev_match != 0xFF) + if (prev_match != MTRR_TYPE_INVALID) return prev_match; return mtrr_state.def_type; } -/* - * Returns the effective MTRR type for the region - * Error return: - * 0xFF - when MTRR is not enabled +/** + * mtrr_type_lookup - look up memory type in MTRR + * + * Return Values: + * MTRR_TYPE_(type) - The effective MTRR type for the region + * MTRR_TYPE_INVALID - MTRR is disabled + * + * Output Argument: + * uniform - Set to 1 when an MTRR covers the region uniformly, i.e. the + * region is fully covered by a single MTRR entry or the default + * type. */ -u8 mtrr_type_lookup(u64 start, u64 end) +u8 mtrr_type_lookup(u64 start, u64 end, u8 *uniform) { - u8 type, prev_type; + u8 type, prev_type, is_uniform = 1, dummy; int repeat; u64 partial_end; - type = __mtrr_type_lookup(start, end, &partial_end, &repeat); + if (!mtrr_state_set) + return MTRR_TYPE_INVALID; + + if (!(mtrr_state.enabled & MTRR_STATE_MTRR_ENABLED)) + return MTRR_TYPE_INVALID; + + /* + * Look up the fixed ranges first, which take priority over + * the variable ranges. + */ + if ((start < 0x100000) && + (mtrr_state.have_fixed) && + (mtrr_state.enabled & MTRR_STATE_MTRR_FIXED_ENABLED)) { + is_uniform = 0; + type = mtrr_type_lookup_fixed(start, end); + goto out; + } + + /* + * Look up the variable ranges. Look of multiple ranges matching + * this address and pick type as per MTRR precedence. + */ + type = mtrr_type_lookup_variable(start, end, &partial_end, + &repeat, &is_uniform); /* * Common path is with repeat = 0. * However, we can have cases where [start:end] spans across some - * MTRR range. Do repeated lookups for that case here. + * MTRR ranges and/or the default type. Do repeated lookups for + * that case here. */ while (repeat) { prev_type = type; start = partial_end; - type = __mtrr_type_lookup(start, end, &partial_end, &repeat); + is_uniform = 0; + type = mtrr_type_lookup_variable(start, end, &partial_end, + &repeat, &dummy); if (check_type_overlap(&prev_type, &type)) - return type; + goto out; } + if (mtrr_tom2 && (start >= (1ULL<<32)) && (end < mtrr_tom2)) + type = MTRR_TYPE_WRBACK; + +out: + *uniform = is_uniform; return type; } @@ -347,7 +408,9 @@ static void __init print_mtrr_state(void) mtrr_attrib_to_str(mtrr_state.def_type)); if (mtrr_state.have_fixed) { pr_debug("MTRR fixed ranges %sabled:\n", - mtrr_state.enabled & 1 ? "en" : "dis"); + ((mtrr_state.enabled & MTRR_STATE_MTRR_ENABLED) && + (mtrr_state.enabled & MTRR_STATE_MTRR_FIXED_ENABLED)) ? + "en" : "dis"); print_fixed(0x00000, 0x10000, mtrr_state.fixed_ranges + 0); for (i = 0; i < 2; ++i) print_fixed(0x80000 + i * 0x20000, 0x04000, @@ -360,7 +423,7 @@ static void __init print_mtrr_state(void) print_fixed_last(); } pr_debug("MTRR variable ranges %sabled:\n", - mtrr_state.enabled & 2 ? "en" : "dis"); + mtrr_state.enabled & MTRR_STATE_MTRR_ENABLED ? "en" : "dis"); high_width = (__ffs64(size_or_mask) - (32 - PAGE_SHIFT) + 3) / 4; for (i = 0; i < num_var_ranges; ++i) { @@ -382,7 +445,7 @@ static void __init print_mtrr_state(void) } /* Grab all of the MTRR state for this CPU into *state */ -void __init get_mtrr_state(void) +bool __init get_mtrr_state(void) { struct mtrr_var_range *vrs; unsigned long flags; @@ -426,6 +489,8 @@ void __init get_mtrr_state(void) post_set(); local_irq_restore(flags); + + return !!(mtrr_state.enabled & MTRR_STATE_MTRR_ENABLED); } /* Some BIOS's are messed up and don't set all MTRRs the same! */ diff --git a/kernel/arch/x86/kernel/cpu/mtrr/main.c b/kernel/arch/x86/kernel/cpu/mtrr/main.c index ea5f363a1..f891b4750 100644 --- a/kernel/arch/x86/kernel/cpu/mtrr/main.c +++ b/kernel/arch/x86/kernel/cpu/mtrr/main.c @@ -59,6 +59,12 @@ #define MTRR_TO_PHYS_WC_OFFSET 1000 u32 num_var_ranges; +static bool __mtrr_enabled; + +static bool mtrr_enabled(void) +{ + return __mtrr_enabled; +} unsigned int mtrr_usage_table[MTRR_MAX_VAR_RANGES]; static DEFINE_MUTEX(mtrr_mutex); @@ -286,7 +292,7 @@ int mtrr_add_page(unsigned long base, unsigned long size, int i, replace, error; mtrr_type ltype; - if (!mtrr_if) + if (!mtrr_enabled()) return -ENXIO; error = mtrr_if->validate_add_page(base, size, type); @@ -435,12 +441,13 @@ static int mtrr_check(unsigned long base, unsigned long size) int mtrr_add(unsigned long base, unsigned long size, unsigned int type, bool increment) { + if (!mtrr_enabled()) + return -ENODEV; if (mtrr_check(base, size)) return -EINVAL; return mtrr_add_page(base >> PAGE_SHIFT, size >> PAGE_SHIFT, type, increment); } -EXPORT_SYMBOL(mtrr_add); /** * mtrr_del_page - delete a memory type region @@ -463,8 +470,8 @@ int mtrr_del_page(int reg, unsigned long base, unsigned long size) unsigned long lbase, lsize; int error = -EINVAL; - if (!mtrr_if) - return -ENXIO; + if (!mtrr_enabled()) + return -ENODEV; max = num_var_ranges; /* No CPU hotplug when we change MTRR entries */ @@ -523,11 +530,12 @@ int mtrr_del_page(int reg, unsigned long base, unsigned long size) */ int mtrr_del(int reg, unsigned long base, unsigned long size) { + if (!mtrr_enabled()) + return -ENODEV; if (mtrr_check(base, size)) return -EINVAL; return mtrr_del_page(reg, base >> PAGE_SHIFT, size >> PAGE_SHIFT); } -EXPORT_SYMBOL(mtrr_del); /** * arch_phys_wc_add - add a WC MTRR and handle errors if PAT is unavailable @@ -538,6 +546,9 @@ EXPORT_SYMBOL(mtrr_del); * attempts to add a WC MTRR covering size bytes starting at base and * logs an error if this fails. * + * The called should provide a power of two size on an equivalent + * power of two boundary. + * * Drivers must store the return value to pass to mtrr_del_wc_if_needed, * but drivers should not try to interpret that return value. */ @@ -545,7 +556,7 @@ int arch_phys_wc_add(unsigned long base, unsigned long size) { int ret; - if (pat_enabled) + if (pat_enabled() || !mtrr_enabled()) return 0; /* Success! (We don't need to do anything.) */ ret = mtrr_add(base, size, MTRR_TYPE_WRCOMB, true); @@ -577,7 +588,7 @@ void arch_phys_wc_del(int handle) EXPORT_SYMBOL(arch_phys_wc_del); /* - * phys_wc_to_mtrr_index - translates arch_phys_wc_add's return value + * arch_phys_wc_index - translates arch_phys_wc_add's return value * @handle: Return value from arch_phys_wc_add * * This will turn the return value from arch_phys_wc_add into an mtrr @@ -587,14 +598,14 @@ EXPORT_SYMBOL(arch_phys_wc_del); * in printk line. Alas there is an illegitimate use in some ancient * drm ioctls. */ -int phys_wc_to_mtrr_index(int handle) +int arch_phys_wc_index(int handle) { if (handle < MTRR_TO_PHYS_WC_OFFSET) return -1; else return handle - MTRR_TO_PHYS_WC_OFFSET; } -EXPORT_SYMBOL_GPL(phys_wc_to_mtrr_index); +EXPORT_SYMBOL_GPL(arch_phys_wc_index); /* * HACK ALERT! @@ -734,10 +745,12 @@ void __init mtrr_bp_init(void) } if (mtrr_if) { + __mtrr_enabled = true; set_num_var_ranges(); init_table(); if (use_intel()) { - get_mtrr_state(); + /* BIOS may override */ + __mtrr_enabled = get_mtrr_state(); if (mtrr_cleanup(phys_addr)) { changed_by_mtrr_cleanup = 1; @@ -745,10 +758,16 @@ void __init mtrr_bp_init(void) } } } + + if (!mtrr_enabled()) + pr_info("MTRR: Disabled\n"); } void mtrr_ap_init(void) { + if (!mtrr_enabled()) + return; + if (!use_intel() || mtrr_aps_delayed_init) return; /* @@ -774,6 +793,9 @@ void mtrr_save_state(void) { int first_cpu; + if (!mtrr_enabled()) + return; + get_online_cpus(); first_cpu = cpumask_first(cpu_online_mask); smp_call_function_single(first_cpu, mtrr_save_fixed_ranges, NULL, 1); @@ -782,6 +804,8 @@ void mtrr_save_state(void) void set_mtrr_aps_delayed_init(void) { + if (!mtrr_enabled()) + return; if (!use_intel()) return; @@ -793,7 +817,7 @@ void set_mtrr_aps_delayed_init(void) */ void mtrr_aps_init(void) { - if (!use_intel()) + if (!use_intel() || !mtrr_enabled()) return; /* @@ -810,7 +834,7 @@ void mtrr_aps_init(void) void mtrr_bp_restore(void) { - if (!use_intel()) + if (!use_intel() || !mtrr_enabled()) return; mtrr_if->set_all(); @@ -818,7 +842,7 @@ void mtrr_bp_restore(void) static int __init mtrr_init_finialize(void) { - if (!mtrr_if) + if (!mtrr_enabled()) return 0; if (use_intel()) { diff --git a/kernel/arch/x86/kernel/cpu/mtrr/mtrr.h b/kernel/arch/x86/kernel/cpu/mtrr/mtrr.h index df5e41f31..951884dcc 100644 --- a/kernel/arch/x86/kernel/cpu/mtrr/mtrr.h +++ b/kernel/arch/x86/kernel/cpu/mtrr/mtrr.h @@ -51,7 +51,7 @@ void set_mtrr_prepare_save(struct set_mtrr_context *ctxt); void fill_mtrr_var_range(unsigned int index, u32 base_lo, u32 base_hi, u32 mask_lo, u32 mask_hi); -void get_mtrr_state(void); +bool get_mtrr_state(void); extern void set_mtrr_ops(const struct mtrr_ops *ops); diff --git a/kernel/arch/x86/kernel/cpu/perf_event.c b/kernel/arch/x86/kernel/cpu/perf_event.c index 4cc98a4e8..2bf79d7c9 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event.c +++ b/kernel/arch/x86/kernel/cpu/perf_event.c @@ -5,7 +5,7 @@ * Copyright (C) 2008-2009 Red Hat, Inc., Ingo Molnar * Copyright (C) 2009 Jaswinder Singh Rajput * Copyright (C) 2009 Advanced Micro Devices, Inc., Robert Richter - * Copyright (C) 2008-2009 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com> + * Copyright (C) 2008-2009 Red Hat, Inc., Peter Zijlstra * Copyright (C) 2009 Intel Corporation, <markus.t.metzger@intel.com> * Copyright (C) 2009 Google, Inc., Stephane Eranian * @@ -135,6 +135,7 @@ static int x86_pmu_extra_regs(u64 config, struct perf_event *event) } static atomic_t active_events; +static atomic_t pmc_refcount; static DEFINE_MUTEX(pmc_reserve_mutex); #ifdef CONFIG_X86_LOCAL_APIC @@ -271,6 +272,7 @@ msr_fail: static void hw_perf_event_destroy(struct perf_event *event) { x86_release_hardware(); + atomic_dec(&active_events); } void hw_perf_lbr_event_destroy(struct perf_event *event) @@ -324,16 +326,16 @@ int x86_reserve_hardware(void) { int err = 0; - if (!atomic_inc_not_zero(&active_events)) { + if (!atomic_inc_not_zero(&pmc_refcount)) { mutex_lock(&pmc_reserve_mutex); - if (atomic_read(&active_events) == 0) { + if (atomic_read(&pmc_refcount) == 0) { if (!reserve_pmc_hardware()) err = -EBUSY; else reserve_ds_buffers(); } if (!err) - atomic_inc(&active_events); + atomic_inc(&pmc_refcount); mutex_unlock(&pmc_reserve_mutex); } @@ -342,7 +344,7 @@ int x86_reserve_hardware(void) void x86_release_hardware(void) { - if (atomic_dec_and_mutex_lock(&active_events, &pmc_reserve_mutex)) { + if (atomic_dec_and_mutex_lock(&pmc_refcount, &pmc_reserve_mutex)) { release_pmc_hardware(); release_ds_buffers(); mutex_unlock(&pmc_reserve_mutex); @@ -355,28 +357,30 @@ void x86_release_hardware(void) */ int x86_add_exclusive(unsigned int what) { - int ret = -EBUSY, i; - - if (atomic_inc_not_zero(&x86_pmu.lbr_exclusive[what])) - return 0; + int i; - mutex_lock(&pmc_reserve_mutex); - for (i = 0; i < ARRAY_SIZE(x86_pmu.lbr_exclusive); i++) { - if (i != what && atomic_read(&x86_pmu.lbr_exclusive[i])) - goto out; + if (!atomic_inc_not_zero(&x86_pmu.lbr_exclusive[what])) { + mutex_lock(&pmc_reserve_mutex); + for (i = 0; i < ARRAY_SIZE(x86_pmu.lbr_exclusive); i++) { + if (i != what && atomic_read(&x86_pmu.lbr_exclusive[i])) + goto fail_unlock; + } + atomic_inc(&x86_pmu.lbr_exclusive[what]); + mutex_unlock(&pmc_reserve_mutex); } - atomic_inc(&x86_pmu.lbr_exclusive[what]); - ret = 0; + atomic_inc(&active_events); + return 0; -out: +fail_unlock: mutex_unlock(&pmc_reserve_mutex); - return ret; + return -EBUSY; } void x86_del_exclusive(unsigned int what) { atomic_dec(&x86_pmu.lbr_exclusive[what]); + atomic_dec(&active_events); } int x86_setup_perfctr(struct perf_event *event) @@ -557,6 +561,7 @@ static int __x86_pmu_event_init(struct perf_event *event) if (err) return err; + atomic_inc(&active_events); event->destroy = hw_perf_event_destroy; event->hw.idx = -1; @@ -895,10 +900,7 @@ int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign) if (x86_pmu.commit_scheduling) x86_pmu.commit_scheduling(cpuc, i, assign[i]); } - } - - if (!assign || unsched) { - + } else { for (i = 0; i < n; i++) { e = cpuc->event_list[i]; /* @@ -1111,13 +1113,16 @@ int x86_perf_event_set_period(struct perf_event *event) per_cpu(pmc_prev_left[idx], smp_processor_id()) = left; - /* - * The hw event starts counting from this event offset, - * mark it to be able to extra future deltas: - */ - local64_set(&hwc->prev_count, (u64)-left); + if (!(hwc->flags & PERF_X86_EVENT_AUTO_RELOAD) || + local64_read(&hwc->prev_count) != (u64)-left) { + /* + * The hw event starts counting from this event offset, + * mark it to be able to extra future deltas: + */ + local64_set(&hwc->prev_count, (u64)-left); - wrmsrl(hwc->event_base, (u64)(-left) & x86_pmu.cntval_mask); + wrmsrl(hwc->event_base, (u64)(-left) & x86_pmu.cntval_mask); + } /* * Due to erratum on certan cpu we need @@ -1170,7 +1175,7 @@ static int x86_pmu_add(struct perf_event *event, int flags) * skip the schedulability test here, it will be performed * at commit time (->commit_txn) as a whole. */ - if (cpuc->group_flag & PERF_EVENT_TXN) + if (cpuc->txn_flags & PERF_PMU_TXN_ADD) goto done_collect; ret = x86_pmu.schedule_events(cpuc, n, assign); @@ -1321,7 +1326,7 @@ static void x86_pmu_del(struct perf_event *event, int flags) * XXX assumes any ->del() called during a TXN will only be on * an event added during that same TXN. */ - if (cpuc->group_flag & PERF_EVENT_TXN) + if (cpuc->txn_flags & PERF_PMU_TXN_ADD) return; /* @@ -1429,6 +1434,10 @@ perf_event_nmi_handler(unsigned int cmd, struct pt_regs *regs) u64 finish_clock; int ret; + /* + * All PMUs/events that share this PMI handler should make sure to + * increment active_events for their events. + */ if (!atomic_read(&active_events)) return NMI_DONE; @@ -1542,7 +1551,7 @@ static void __init filter_events(struct attribute **attrs) } /* Merge two pointer arrays */ -static __init struct attribute **merge_attr(struct attribute **a, struct attribute **b) +__init struct attribute **merge_attr(struct attribute **a, struct attribute **b) { struct attribute **new; int j, i; @@ -1739,11 +1748,22 @@ static inline void x86_pmu_read(struct perf_event *event) * Start group events scheduling transaction * Set the flag to make pmu::enable() not perform the * schedulability test, it will be performed at commit time + * + * We only support PERF_PMU_TXN_ADD transactions. Save the + * transaction flags but otherwise ignore non-PERF_PMU_TXN_ADD + * transactions. */ -static void x86_pmu_start_txn(struct pmu *pmu) +static void x86_pmu_start_txn(struct pmu *pmu, unsigned int txn_flags) { + struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); + + WARN_ON_ONCE(cpuc->txn_flags); /* txn already in flight */ + + cpuc->txn_flags = txn_flags; + if (txn_flags & ~PERF_PMU_TXN_ADD) + return; + perf_pmu_disable(pmu); - __this_cpu_or(cpu_hw_events.group_flag, PERF_EVENT_TXN); __this_cpu_write(cpu_hw_events.n_txn, 0); } @@ -1754,7 +1774,16 @@ static void x86_pmu_start_txn(struct pmu *pmu) */ static void x86_pmu_cancel_txn(struct pmu *pmu) { - __this_cpu_and(cpu_hw_events.group_flag, ~PERF_EVENT_TXN); + unsigned int txn_flags; + struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); + + WARN_ON_ONCE(!cpuc->txn_flags); /* no txn in flight */ + + txn_flags = cpuc->txn_flags; + cpuc->txn_flags = 0; + if (txn_flags & ~PERF_PMU_TXN_ADD) + return; + /* * Truncate collected array by the number of events added in this * transaction. See x86_pmu_add() and x86_pmu_*_txn(). @@ -1777,6 +1806,13 @@ static int x86_pmu_commit_txn(struct pmu *pmu) int assign[X86_PMC_IDX_MAX]; int n, ret; + WARN_ON_ONCE(!cpuc->txn_flags); /* no txn in flight */ + + if (cpuc->txn_flags & ~PERF_PMU_TXN_ADD) { + cpuc->txn_flags = 0; + return 0; + } + n = cpuc->n_events; if (!x86_pmu_initialized()) @@ -1792,7 +1828,7 @@ static int x86_pmu_commit_txn(struct pmu *pmu) */ memcpy(cpuc->assign, assign, n*sizeof(int)); - cpuc->group_flag &= ~PERF_EVENT_TXN; + cpuc->txn_flags = 0; perf_pmu_enable(pmu); return 0; } @@ -2170,6 +2206,7 @@ static unsigned long get_segment_base(unsigned int segment) int idx = segment >> 3; if ((segment & SEGMENT_TI_MASK) == SEGMENT_LDT) { +#ifdef CONFIG_MODIFY_LDT_SYSCALL struct ldt_struct *ldt; if (idx > LDT_ENTRIES) @@ -2181,6 +2218,9 @@ static unsigned long get_segment_base(unsigned int segment) return 0; desc = &ldt->entries[idx]; +#else + return 0; +#endif } else { if (idx > GDT_ENTRIES) return 0; @@ -2191,7 +2231,7 @@ static unsigned long get_segment_base(unsigned int segment) return get_desc_base(desc); } -#ifdef CONFIG_COMPAT +#ifdef CONFIG_IA32_EMULATION #include <asm/compat.h> diff --git a/kernel/arch/x86/kernel/cpu/perf_event.h b/kernel/arch/x86/kernel/cpu/perf_event.h index f068695ea..d0e35ebb2 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event.h +++ b/kernel/arch/x86/kernel/cpu/perf_event.h @@ -5,7 +5,7 @@ * Copyright (C) 2008-2009 Red Hat, Inc., Ingo Molnar * Copyright (C) 2009 Jaswinder Singh Rajput * Copyright (C) 2009 Advanced Micro Devices, Inc., Robert Richter - * Copyright (C) 2008-2009 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com> + * Copyright (C) 2008-2009 Red Hat, Inc., Peter Zijlstra * Copyright (C) 2009 Intel Corporation, <markus.t.metzger@intel.com> * Copyright (C) 2009 Google, Inc., Stephane Eranian * @@ -47,6 +47,7 @@ enum extra_reg_type { EXTRA_REG_RSP_1 = 1, /* offcore_response_1 */ EXTRA_REG_LBR = 2, /* lbr_select */ EXTRA_REG_LDLAT = 3, /* ld_lat_threshold */ + EXTRA_REG_FE = 4, /* fe_* */ EXTRA_REG_MAX /* number of entries needed */ }; @@ -75,6 +76,8 @@ struct event_constraint { #define PERF_X86_EVENT_DYNAMIC 0x0080 /* dynamic alloc'd constraint */ #define PERF_X86_EVENT_RDPMC_ALLOWED 0x0100 /* grant rdpmc permission */ #define PERF_X86_EVENT_EXCL_ACCT 0x0200 /* accounted EXCL event */ +#define PERF_X86_EVENT_AUTO_RELOAD 0x0400 /* use PEBS auto-reload */ +#define PERF_X86_EVENT_FREERUNNING 0x0800 /* use freerunning PEBS */ struct amd_nb { @@ -88,6 +91,18 @@ struct amd_nb { #define MAX_PEBS_EVENTS 8 /* + * Flags PEBS can handle without an PMI. + * + * TID can only be handled by flushing at context switch. + * + */ +#define PEBS_FREERUNNING_FLAGS \ + (PERF_SAMPLE_IP | PERF_SAMPLE_TID | PERF_SAMPLE_ADDR | \ + PERF_SAMPLE_ID | PERF_SAMPLE_CPU | PERF_SAMPLE_STREAM_ID | \ + PERF_SAMPLE_DATA_SRC | PERF_SAMPLE_IDENTIFIER | \ + PERF_SAMPLE_TRANSACTION) + +/* * A debug store configuration. * * We only support architectures that use 64bit fields. @@ -133,7 +148,6 @@ enum intel_excl_state_type { }; struct intel_excl_states { - enum intel_excl_state_type init_state[X86_PMC_IDX_MAX]; enum intel_excl_state_type state[X86_PMC_IDX_MAX]; bool sched_started; /* true if scheduling has started */ }; @@ -152,7 +166,7 @@ struct intel_excl_cntrs { unsigned core_id; /* per-core: core id */ }; -#define MAX_LBR_ENTRIES 16 +#define MAX_LBR_ENTRIES 32 enum { X86_PERF_KFREE_SHARED = 0, @@ -182,7 +196,7 @@ struct cpu_hw_events { int n_excl; /* the number of exclusive events */ - unsigned int group_flag; + unsigned int txn_flags; int is_fake; /* @@ -373,7 +387,7 @@ struct cpu_hw_events { /* Check flags and event code/umask, and set the HSW N/A flag */ #define INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_NA(code, n) \ __EVENT_CONSTRAINT(code, n, \ - INTEL_ARCH_EVENT_MASK|INTEL_ARCH_EVENT_MASK, \ + INTEL_ARCH_EVENT_MASK|X86_ALL_EVENT_FLAGS, \ HWEIGHT(n), 0, PERF_X86_EVENT_PEBS_NA_HSW) @@ -527,10 +541,10 @@ struct x86_pmu { void (*put_event_constraints)(struct cpu_hw_events *cpuc, struct perf_event *event); - void (*commit_scheduling)(struct cpu_hw_events *cpuc, int idx, int cntr); - void (*start_scheduling)(struct cpu_hw_events *cpuc); + void (*commit_scheduling)(struct cpu_hw_events *cpuc, int idx, int cntr); + void (*stop_scheduling)(struct cpu_hw_events *cpuc); struct event_constraint *event_constraints; @@ -581,6 +595,7 @@ struct x86_pmu { struct event_constraint *pebs_constraints; void (*pebs_aliases)(struct perf_event *event); int max_pebs_events; + unsigned long free_running_flags; /* * Intel LBR @@ -611,6 +626,8 @@ struct x86_pmu { struct x86_perf_task_context { u64 lbr_from[MAX_LBR_ENTRIES]; u64 lbr_to[MAX_LBR_ENTRIES]; + u64 lbr_info[MAX_LBR_ENTRIES]; + int tos; int lbr_callstack_users; int lbr_stack_state; }; @@ -780,6 +797,8 @@ static inline void set_linear_ip(struct pt_regs *regs, unsigned long ip) ssize_t x86_event_sysfs_show(char *page, u64 config, u64 event); ssize_t intel_event_sysfs_show(char *page, u64 config); +struct attribute **merge_attr(struct attribute **a, struct attribute **b); + #ifdef CONFIG_CPU_SUP_AMD int amd_pmu_init(void); @@ -795,20 +814,6 @@ static inline int amd_pmu_init(void) #ifdef CONFIG_CPU_SUP_INTEL -static inline bool intel_pmu_needs_lbr_smpl(struct perf_event *event) -{ - /* user explicitly requested branch sampling */ - if (has_branch_stack(event)) - return true; - - /* implicit branch sampling to correct PEBS skid */ - if (x86_pmu.intel_cap.pebs_trap && event->attr.precise_ip > 1 && - x86_pmu.intel_cap.pebs_format < 2) - return true; - - return false; -} - static inline bool intel_pmu_has_bts(struct perf_event *event) { if (event->attr.config == PERF_COUNT_HW_BRANCH_INSTRUCTIONS && @@ -860,6 +865,8 @@ extern struct event_constraint intel_ivb_pebs_event_constraints[]; extern struct event_constraint intel_hsw_pebs_event_constraints[]; +extern struct event_constraint intel_skl_pebs_event_constraints[]; + struct event_constraint *intel_pebs_constraints(struct perf_event *event); void intel_pmu_pebs_enable(struct perf_event *event); @@ -870,6 +877,8 @@ void intel_pmu_pebs_enable_all(void); void intel_pmu_pebs_disable_all(void); +void intel_pmu_pebs_sched_task(struct perf_event_context *ctx, bool sched_in); + void intel_ds_init(void); void intel_pmu_lbr_sched_task(struct perf_event_context *ctx, bool sched_in); @@ -896,6 +905,8 @@ void intel_pmu_lbr_init_snb(void); void intel_pmu_lbr_init_hsw(void); +void intel_pmu_lbr_init_skl(void); + int intel_pmu_setup_lbr_filter(struct perf_event *event); void intel_pt_interrupt(void); @@ -919,6 +930,7 @@ static inline int is_ht_workaround_enabled(void) { return !!(x86_pmu.flags & PMU_FL_EXCL_ENABLED); } + #else /* CONFIG_CPU_SUP_INTEL */ static inline void reserve_ds_buffers(void) diff --git a/kernel/arch/x86/kernel/cpu/perf_event_intel.c b/kernel/arch/x86/kernel/cpu/perf_event_intel.c index 2813ea0f1..e2a430021 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event_intel.c +++ b/kernel/arch/x86/kernel/cpu/perf_event_intel.c @@ -12,7 +12,7 @@ #include <linux/init.h> #include <linux/slab.h> #include <linux/export.h> -#include <linux/watchdog.h> +#include <linux/nmi.h> #include <asm/cpufeature.h> #include <asm/hardirq.h> @@ -177,6 +177,14 @@ static struct event_constraint intel_slm_event_constraints[] __read_mostly = EVENT_CONSTRAINT_END }; +struct event_constraint intel_skl_event_constraints[] = { + FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */ + FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ + FIXED_EVENT_CONSTRAINT(0x0300, 2), /* CPU_CLK_UNHALTED.REF */ + INTEL_UEVENT_CONSTRAINT(0x1c0, 0x2), /* INST_RETIRED.PREC_DIST */ + EVENT_CONSTRAINT_END +}; + static struct extra_reg intel_snb_extra_regs[] __read_mostly = { /* must define OFFCORE_RSP_X first, see intel_fixup_er() */ INTEL_UEVENT_EXTRA_REG(0x01b7, MSR_OFFCORE_RSP_0, 0x3f807f8fffull, RSP_0), @@ -193,6 +201,18 @@ static struct extra_reg intel_snbep_extra_regs[] __read_mostly = { EVENT_EXTRA_END }; +static struct extra_reg intel_skl_extra_regs[] __read_mostly = { + INTEL_UEVENT_EXTRA_REG(0x01b7, MSR_OFFCORE_RSP_0, 0x3fffff8fffull, RSP_0), + INTEL_UEVENT_EXTRA_REG(0x01bb, MSR_OFFCORE_RSP_1, 0x3fffff8fffull, RSP_1), + INTEL_UEVENT_PEBS_LDLAT_EXTRA_REG(0x01cd), + /* + * Note the low 8 bits eventsel code is not a continuous field, containing + * some #GPing bits. These are masked out. + */ + INTEL_UEVENT_EXTRA_REG(0x01c6, MSR_PEBS_FRONTEND, 0x7fff17, FE), + EVENT_EXTRA_END +}; + EVENT_ATTR_STR(mem-loads, mem_ld_nhm, "event=0x0b,umask=0x10,ldlat=3"); EVENT_ATTR_STR(mem-loads, mem_ld_snb, "event=0xcd,umask=0x1,ldlat=3"); EVENT_ATTR_STR(mem-stores, mem_st_snb, "event=0xcd,umask=0x2"); @@ -212,7 +232,7 @@ static struct event_constraint intel_hsw_event_constraints[] = { FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */ FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ FIXED_EVENT_CONSTRAINT(0x0300, 2), /* CPU_CLK_UNHALTED.REF */ - INTEL_EVENT_CONSTRAINT(0x48, 0x4), /* L1D_PEND_MISS.* */ + INTEL_UEVENT_CONSTRAINT(0x148, 0x4), /* L1D_PEND_MISS.PENDING */ INTEL_UEVENT_CONSTRAINT(0x01c0, 0x2), /* INST_RETIRED.PREC_DIST */ INTEL_EVENT_CONSTRAINT(0xcd, 0x8), /* MEM_TRANS_RETIRED.LOAD_LATENCY */ /* CYCLE_ACTIVITY.CYCLES_L1D_PENDING */ @@ -235,7 +255,7 @@ struct event_constraint intel_bdw_event_constraints[] = { FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ FIXED_EVENT_CONSTRAINT(0x0300, 2), /* CPU_CLK_UNHALTED.REF */ INTEL_UEVENT_CONSTRAINT(0x148, 0x4), /* L1D_PEND_MISS.PENDING */ - INTEL_EVENT_CONSTRAINT(0xa3, 0x4), /* CYCLE_ACTIVITY.* */ + INTEL_UEVENT_CONSTRAINT(0x8a3, 0x4), /* CYCLE_ACTIVITY.CYCLES_L1D_MISS */ EVENT_CONSTRAINT_END }; @@ -244,6 +264,200 @@ static u64 intel_pmu_event_map(int hw_event) return intel_perfmon_event_map[hw_event]; } +/* + * Notes on the events: + * - data reads do not include code reads (comparable to earlier tables) + * - data counts include speculative execution (except L1 write, dtlb, bpu) + * - remote node access includes remote memory, remote cache, remote mmio. + * - prefetches are not included in the counts. + * - icache miss does not include decoded icache + */ + +#define SKL_DEMAND_DATA_RD BIT_ULL(0) +#define SKL_DEMAND_RFO BIT_ULL(1) +#define SKL_ANY_RESPONSE BIT_ULL(16) +#define SKL_SUPPLIER_NONE BIT_ULL(17) +#define SKL_L3_MISS_LOCAL_DRAM BIT_ULL(26) +#define SKL_L3_MISS_REMOTE_HOP0_DRAM BIT_ULL(27) +#define SKL_L3_MISS_REMOTE_HOP1_DRAM BIT_ULL(28) +#define SKL_L3_MISS_REMOTE_HOP2P_DRAM BIT_ULL(29) +#define SKL_L3_MISS (SKL_L3_MISS_LOCAL_DRAM| \ + SKL_L3_MISS_REMOTE_HOP0_DRAM| \ + SKL_L3_MISS_REMOTE_HOP1_DRAM| \ + SKL_L3_MISS_REMOTE_HOP2P_DRAM) +#define SKL_SPL_HIT BIT_ULL(30) +#define SKL_SNOOP_NONE BIT_ULL(31) +#define SKL_SNOOP_NOT_NEEDED BIT_ULL(32) +#define SKL_SNOOP_MISS BIT_ULL(33) +#define SKL_SNOOP_HIT_NO_FWD BIT_ULL(34) +#define SKL_SNOOP_HIT_WITH_FWD BIT_ULL(35) +#define SKL_SNOOP_HITM BIT_ULL(36) +#define SKL_SNOOP_NON_DRAM BIT_ULL(37) +#define SKL_ANY_SNOOP (SKL_SPL_HIT|SKL_SNOOP_NONE| \ + SKL_SNOOP_NOT_NEEDED|SKL_SNOOP_MISS| \ + SKL_SNOOP_HIT_NO_FWD|SKL_SNOOP_HIT_WITH_FWD| \ + SKL_SNOOP_HITM|SKL_SNOOP_NON_DRAM) +#define SKL_DEMAND_READ SKL_DEMAND_DATA_RD +#define SKL_SNOOP_DRAM (SKL_SNOOP_NONE| \ + SKL_SNOOP_NOT_NEEDED|SKL_SNOOP_MISS| \ + SKL_SNOOP_HIT_NO_FWD|SKL_SNOOP_HIT_WITH_FWD| \ + SKL_SNOOP_HITM|SKL_SPL_HIT) +#define SKL_DEMAND_WRITE SKL_DEMAND_RFO +#define SKL_LLC_ACCESS SKL_ANY_RESPONSE +#define SKL_L3_MISS_REMOTE (SKL_L3_MISS_REMOTE_HOP0_DRAM| \ + SKL_L3_MISS_REMOTE_HOP1_DRAM| \ + SKL_L3_MISS_REMOTE_HOP2P_DRAM) + +static __initconst const u64 skl_hw_cache_event_ids + [PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX] = +{ + [ C(L1D ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x81d0, /* MEM_INST_RETIRED.ALL_LOADS */ + [ C(RESULT_MISS) ] = 0x151, /* L1D.REPLACEMENT */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0x82d0, /* MEM_INST_RETIRED.ALL_STORES */ + [ C(RESULT_MISS) ] = 0x0, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0x0, + [ C(RESULT_MISS) ] = 0x0, + }, + }, + [ C(L1I ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x0, + [ C(RESULT_MISS) ] = 0x283, /* ICACHE_64B.MISS */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0x0, + [ C(RESULT_MISS) ] = 0x0, + }, + }, + [ C(LL ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x1b7, /* OFFCORE_RESPONSE */ + [ C(RESULT_MISS) ] = 0x1b7, /* OFFCORE_RESPONSE */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0x1b7, /* OFFCORE_RESPONSE */ + [ C(RESULT_MISS) ] = 0x1b7, /* OFFCORE_RESPONSE */ + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0x0, + [ C(RESULT_MISS) ] = 0x0, + }, + }, + [ C(DTLB) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x81d0, /* MEM_INST_RETIRED.ALL_LOADS */ + [ C(RESULT_MISS) ] = 0x608, /* DTLB_LOAD_MISSES.WALK_COMPLETED */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0x82d0, /* MEM_INST_RETIRED.ALL_STORES */ + [ C(RESULT_MISS) ] = 0x649, /* DTLB_STORE_MISSES.WALK_COMPLETED */ + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0x0, + [ C(RESULT_MISS) ] = 0x0, + }, + }, + [ C(ITLB) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x2085, /* ITLB_MISSES.STLB_HIT */ + [ C(RESULT_MISS) ] = 0xe85, /* ITLB_MISSES.WALK_COMPLETED */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + }, + [ C(BPU ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0xc4, /* BR_INST_RETIRED.ALL_BRANCHES */ + [ C(RESULT_MISS) ] = 0xc5, /* BR_MISP_RETIRED.ALL_BRANCHES */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = -1, + [ C(RESULT_MISS) ] = -1, + }, + }, + [ C(NODE) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = 0x1b7, /* OFFCORE_RESPONSE */ + [ C(RESULT_MISS) ] = 0x1b7, /* OFFCORE_RESPONSE */ + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = 0x1b7, /* OFFCORE_RESPONSE */ + [ C(RESULT_MISS) ] = 0x1b7, /* OFFCORE_RESPONSE */ + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0x0, + [ C(RESULT_MISS) ] = 0x0, + }, + }, +}; + +static __initconst const u64 skl_hw_cache_extra_regs + [PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX] = +{ + [ C(LL ) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = SKL_DEMAND_READ| + SKL_LLC_ACCESS|SKL_ANY_SNOOP, + [ C(RESULT_MISS) ] = SKL_DEMAND_READ| + SKL_L3_MISS|SKL_ANY_SNOOP| + SKL_SUPPLIER_NONE, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = SKL_DEMAND_WRITE| + SKL_LLC_ACCESS|SKL_ANY_SNOOP, + [ C(RESULT_MISS) ] = SKL_DEMAND_WRITE| + SKL_L3_MISS|SKL_ANY_SNOOP| + SKL_SUPPLIER_NONE, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0x0, + [ C(RESULT_MISS) ] = 0x0, + }, + }, + [ C(NODE) ] = { + [ C(OP_READ) ] = { + [ C(RESULT_ACCESS) ] = SKL_DEMAND_READ| + SKL_L3_MISS_LOCAL_DRAM|SKL_SNOOP_DRAM, + [ C(RESULT_MISS) ] = SKL_DEMAND_READ| + SKL_L3_MISS_REMOTE|SKL_SNOOP_DRAM, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = SKL_DEMAND_WRITE| + SKL_L3_MISS_LOCAL_DRAM|SKL_SNOOP_DRAM, + [ C(RESULT_MISS) ] = SKL_DEMAND_WRITE| + SKL_L3_MISS_REMOTE|SKL_SNOOP_DRAM, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = 0x0, + [ C(RESULT_MISS) ] = 0x0, + }, + }, +}; + #define SNB_DMND_DATA_RD (1ULL << 0) #define SNB_DMND_RFO (1ULL << 1) #define SNB_DMND_IFETCH (1ULL << 2) @@ -1114,7 +1328,7 @@ static struct extra_reg intel_slm_extra_regs[] __read_mostly = { /* must define OFFCORE_RSP_X first, see intel_fixup_er() */ INTEL_UEVENT_EXTRA_REG(0x01b7, MSR_OFFCORE_RSP_0, 0x768005ffffull, RSP_0), - INTEL_UEVENT_EXTRA_REG(0x02b7, MSR_OFFCORE_RSP_1, 0x768005ffffull, RSP_1), + INTEL_UEVENT_EXTRA_REG(0x02b7, MSR_OFFCORE_RSP_1, 0x368005ffffull, RSP_1), EVENT_EXTRA_END }; @@ -1594,6 +1808,7 @@ static int intel_pmu_handle_irq(struct pt_regs *regs) loops = 0; again: + intel_pmu_lbr_read(); intel_pmu_ack_status(status); if (++loops > 100) { static bool warned = false; @@ -1608,16 +1823,16 @@ again: inc_irq_stat(apic_perf_irqs); - intel_pmu_lbr_read(); /* - * CondChgd bit 63 doesn't mean any overflow status. Ignore - * and clear the bit. + * Ignore a range of extra bits in status that do not indicate + * overflow by themselves. */ - if (__test_and_clear_bit(63, (unsigned long *)&status)) { - if (!status) - goto done; - } + status &= ~(GLOBAL_STATUS_COND_CHG | + GLOBAL_STATUS_ASIF | + GLOBAL_STATUS_LBRS_FROZEN); + if (!status) + goto done; /* * PEBS overflow sets bit 62 in the global status register @@ -1699,18 +1914,22 @@ intel_bts_constraints(struct perf_event *event) return NULL; } -static int intel_alt_er(int idx) +static int intel_alt_er(int idx, u64 config) { + int alt_idx; if (!(x86_pmu.flags & PMU_FL_HAS_RSP_1)) return idx; if (idx == EXTRA_REG_RSP_0) - return EXTRA_REG_RSP_1; + alt_idx = EXTRA_REG_RSP_1; if (idx == EXTRA_REG_RSP_1) - return EXTRA_REG_RSP_0; + alt_idx = EXTRA_REG_RSP_0; + + if (config & ~x86_pmu.extra_regs[alt_idx].valid_mask) + return idx; - return idx; + return alt_idx; } static void intel_fixup_er(struct perf_event *event, int idx) @@ -1799,7 +2018,7 @@ again: */ c = NULL; } else { - idx = intel_alt_er(idx); + idx = intel_alt_er(idx, reg->config); if (idx != reg->idx) { raw_spin_unlock_irqrestore(&era->lock, flags); goto again; @@ -1903,9 +2122,8 @@ static void intel_start_scheduling(struct cpu_hw_events *cpuc) { struct intel_excl_cntrs *excl_cntrs = cpuc->excl_cntrs; - struct intel_excl_states *xl, *xlo; + struct intel_excl_states *xl; int tid = cpuc->excl_thread_id; - int o_tid = 1 - tid; /* sibling thread */ /* * nothing needed if in group validation mode @@ -1916,10 +2134,9 @@ intel_start_scheduling(struct cpu_hw_events *cpuc) /* * no exclusion needed */ - if (!excl_cntrs) + if (WARN_ON_ONCE(!excl_cntrs)) return; - xlo = &excl_cntrs->states[o_tid]; xl = &excl_cntrs->states[tid]; xl->sched_started = true; @@ -1928,22 +2145,41 @@ intel_start_scheduling(struct cpu_hw_events *cpuc) * in stop_event_scheduling() * makes scheduling appear as a transaction */ - WARN_ON_ONCE(!irqs_disabled()); raw_spin_lock(&excl_cntrs->lock); +} - /* - * save initial state of sibling thread - */ - memcpy(xlo->init_state, xlo->state, sizeof(xlo->init_state)); +static void intel_commit_scheduling(struct cpu_hw_events *cpuc, int idx, int cntr) +{ + struct intel_excl_cntrs *excl_cntrs = cpuc->excl_cntrs; + struct event_constraint *c = cpuc->event_constraint[idx]; + struct intel_excl_states *xl; + int tid = cpuc->excl_thread_id; + + if (cpuc->is_fake || !is_ht_workaround_enabled()) + return; + + if (WARN_ON_ONCE(!excl_cntrs)) + return; + + if (!(c->flags & PERF_X86_EVENT_DYNAMIC)) + return; + + xl = &excl_cntrs->states[tid]; + + lockdep_assert_held(&excl_cntrs->lock); + + if (c->flags & PERF_X86_EVENT_EXCL) + xl->state[cntr] = INTEL_EXCL_EXCLUSIVE; + else + xl->state[cntr] = INTEL_EXCL_SHARED; } static void intel_stop_scheduling(struct cpu_hw_events *cpuc) { struct intel_excl_cntrs *excl_cntrs = cpuc->excl_cntrs; - struct intel_excl_states *xl, *xlo; + struct intel_excl_states *xl; int tid = cpuc->excl_thread_id; - int o_tid = 1 - tid; /* sibling thread */ /* * nothing needed if in group validation mode @@ -1953,17 +2189,11 @@ intel_stop_scheduling(struct cpu_hw_events *cpuc) /* * no exclusion needed */ - if (!excl_cntrs) + if (WARN_ON_ONCE(!excl_cntrs)) return; - xlo = &excl_cntrs->states[o_tid]; xl = &excl_cntrs->states[tid]; - /* - * make new sibling thread state visible - */ - memcpy(xlo->state, xlo->init_state, sizeof(xlo->state)); - xl->sched_started = false; /* * release shared state lock (acquired in intel_start_scheduling()) @@ -1975,12 +2205,10 @@ static struct event_constraint * intel_get_excl_constraints(struct cpu_hw_events *cpuc, struct perf_event *event, int idx, struct event_constraint *c) { - struct event_constraint *cx; struct intel_excl_cntrs *excl_cntrs = cpuc->excl_cntrs; - struct intel_excl_states *xl, *xlo; - int is_excl, i; + struct intel_excl_states *xlo; int tid = cpuc->excl_thread_id; - int o_tid = 1 - tid; /* alternate */ + int is_excl, i; /* * validating a group does not require @@ -1992,27 +2220,8 @@ intel_get_excl_constraints(struct cpu_hw_events *cpuc, struct perf_event *event, /* * no exclusion needed */ - if (!excl_cntrs) + if (WARN_ON_ONCE(!excl_cntrs)) return c; - /* - * event requires exclusive counter access - * across HT threads - */ - is_excl = c->flags & PERF_X86_EVENT_EXCL; - if (is_excl && !(event->hw.flags & PERF_X86_EVENT_EXCL_ACCT)) { - event->hw.flags |= PERF_X86_EVENT_EXCL_ACCT; - if (!cpuc->n_excl++) - WRITE_ONCE(excl_cntrs->has_exclusive[tid], 1); - } - - /* - * xl = state of current HT - * xlo = state of sibling HT - */ - xl = &excl_cntrs->states[tid]; - xlo = &excl_cntrs->states[o_tid]; - - cx = c; /* * because we modify the constraint, we need @@ -2023,10 +2232,7 @@ intel_get_excl_constraints(struct cpu_hw_events *cpuc, struct perf_event *event, * been cloned (marked dynamic) */ if (!(c->flags & PERF_X86_EVENT_DYNAMIC)) { - - /* sanity check */ - if (idx < 0) - return &emptyconstraint; + struct event_constraint *cx; /* * grab pre-allocated constraint entry @@ -2037,13 +2243,14 @@ intel_get_excl_constraints(struct cpu_hw_events *cpuc, struct perf_event *event, * initialize dynamic constraint * with static constraint */ - memcpy(cx, c, sizeof(*cx)); + *cx = *c; /* * mark constraint as dynamic, so we * can free it later on */ cx->flags |= PERF_X86_EVENT_DYNAMIC; + c = cx; } /* @@ -2054,6 +2261,22 @@ intel_get_excl_constraints(struct cpu_hw_events *cpuc, struct perf_event *event, */ /* + * state of sibling HT + */ + xlo = &excl_cntrs->states[tid ^ 1]; + + /* + * event requires exclusive counter access + * across HT threads + */ + is_excl = c->flags & PERF_X86_EVENT_EXCL; + if (is_excl && !(event->hw.flags & PERF_X86_EVENT_EXCL_ACCT)) { + event->hw.flags |= PERF_X86_EVENT_EXCL_ACCT; + if (!cpuc->n_excl++) + WRITE_ONCE(excl_cntrs->has_exclusive[tid], 1); + } + + /* * Modify static constraint with current dynamic * state of thread * @@ -2061,46 +2284,49 @@ intel_get_excl_constraints(struct cpu_hw_events *cpuc, struct perf_event *event, * SHARED : sibling counter measuring non-exclusive event * UNUSED : sibling counter unused */ - for_each_set_bit(i, cx->idxmsk, X86_PMC_IDX_MAX) { + for_each_set_bit(i, c->idxmsk, X86_PMC_IDX_MAX) { /* * exclusive event in sibling counter * our corresponding counter cannot be used * regardless of our event */ - if (xl->state[i] == INTEL_EXCL_EXCLUSIVE) - __clear_bit(i, cx->idxmsk); + if (xlo->state[i] == INTEL_EXCL_EXCLUSIVE) + __clear_bit(i, c->idxmsk); /* * if measuring an exclusive event, sibling * measuring non-exclusive, then counter cannot * be used */ - if (is_excl && xl->state[i] == INTEL_EXCL_SHARED) - __clear_bit(i, cx->idxmsk); + if (is_excl && xlo->state[i] == INTEL_EXCL_SHARED) + __clear_bit(i, c->idxmsk); } /* * recompute actual bit weight for scheduling algorithm */ - cx->weight = hweight64(cx->idxmsk64); + c->weight = hweight64(c->idxmsk64); /* * if we return an empty mask, then switch * back to static empty constraint to avoid * the cost of freeing later on */ - if (cx->weight == 0) - cx = &emptyconstraint; + if (c->weight == 0) + c = &emptyconstraint; - return cx; + return c; } static struct event_constraint * intel_get_event_constraints(struct cpu_hw_events *cpuc, int idx, struct perf_event *event) { - struct event_constraint *c1 = cpuc->event_constraint[idx]; + struct event_constraint *c1 = NULL; struct event_constraint *c2; + if (idx >= 0) /* fake does < 0 */ + c1 = cpuc->event_constraint[idx]; + /* * first time only * - static constraint: no change across incremental scheduling calls @@ -2124,10 +2350,8 @@ static void intel_put_excl_constraints(struct cpu_hw_events *cpuc, { struct hw_perf_event *hwc = &event->hw; struct intel_excl_cntrs *excl_cntrs = cpuc->excl_cntrs; - struct intel_excl_states *xlo, *xl; - unsigned long flags = 0; /* keep compiler happy */ int tid = cpuc->excl_thread_id; - int o_tid = 1 - tid; + struct intel_excl_states *xl; /* * nothing needed if in group validation mode @@ -2135,13 +2359,9 @@ static void intel_put_excl_constraints(struct cpu_hw_events *cpuc, if (cpuc->is_fake) return; - WARN_ON_ONCE(!excl_cntrs); - - if (!excl_cntrs) + if (WARN_ON_ONCE(!excl_cntrs)) return; - xl = &excl_cntrs->states[tid]; - xlo = &excl_cntrs->states[o_tid]; if (hwc->flags & PERF_X86_EVENT_EXCL_ACCT) { hwc->flags &= ~PERF_X86_EVENT_EXCL_ACCT; if (!--cpuc->n_excl) @@ -2149,22 +2369,25 @@ static void intel_put_excl_constraints(struct cpu_hw_events *cpuc, } /* - * put_constraint may be called from x86_schedule_events() - * which already has the lock held so here make locking - * conditional + * If event was actually assigned, then mark the counter state as + * unused now. */ - if (!xl->sched_started) - raw_spin_lock_irqsave(&excl_cntrs->lock, flags); + if (hwc->idx >= 0) { + xl = &excl_cntrs->states[tid]; - /* - * if event was actually assigned, then mark the - * counter state as unused now - */ - if (hwc->idx >= 0) - xlo->state[hwc->idx] = INTEL_EXCL_UNUSED; + /* + * put_constraint may be called from x86_schedule_events() + * which already has the lock held so here make locking + * conditional. + */ + if (!xl->sched_started) + raw_spin_lock(&excl_cntrs->lock); + + xl->state[hwc->idx] = INTEL_EXCL_UNUSED; - if (!xl->sched_started) - raw_spin_unlock_irqrestore(&excl_cntrs->lock, flags); + if (!xl->sched_started) + raw_spin_unlock(&excl_cntrs->lock); + } } static void @@ -2196,41 +2419,6 @@ static void intel_put_event_constraints(struct cpu_hw_events *cpuc, intel_put_excl_constraints(cpuc, event); } -static void intel_commit_scheduling(struct cpu_hw_events *cpuc, int idx, int cntr) -{ - struct intel_excl_cntrs *excl_cntrs = cpuc->excl_cntrs; - struct event_constraint *c = cpuc->event_constraint[idx]; - struct intel_excl_states *xlo, *xl; - int tid = cpuc->excl_thread_id; - int o_tid = 1 - tid; - int is_excl; - - if (cpuc->is_fake || !c) - return; - - is_excl = c->flags & PERF_X86_EVENT_EXCL; - - if (!(c->flags & PERF_X86_EVENT_DYNAMIC)) - return; - - WARN_ON_ONCE(!excl_cntrs); - - if (!excl_cntrs) - return; - - xl = &excl_cntrs->states[tid]; - xlo = &excl_cntrs->states[o_tid]; - - WARN_ON_ONCE(!raw_spin_is_locked(&excl_cntrs->lock)); - - if (cntr >= 0) { - if (is_excl) - xlo->init_state[cntr] = INTEL_EXCL_EXCLUSIVE; - else - xlo->init_state[cntr] = INTEL_EXCL_SHARED; - } -} - static void intel_pebs_aliases_core2(struct perf_event *event) { if ((event->hw.config & X86_RAW_EVENT_MASK) == 0x003c) { @@ -2287,6 +2475,15 @@ static void intel_pebs_aliases_snb(struct perf_event *event) } } +static unsigned long intel_pmu_free_running_flags(struct perf_event *event) +{ + unsigned long flags = x86_pmu.free_running_flags; + + if (event->attr.use_clockid) + flags &= ~PERF_SAMPLE_TIME; + return flags; +} + static int intel_pmu_hw_config(struct perf_event *event) { int ret = x86_pmu_hw_config(event); @@ -2294,8 +2491,16 @@ static int intel_pmu_hw_config(struct perf_event *event) if (ret) return ret; - if (event->attr.precise_ip && x86_pmu.pebs_aliases) - x86_pmu.pebs_aliases(event); + if (event->attr.precise_ip) { + if (!event->attr.freq) { + event->hw.flags |= PERF_X86_EVENT_AUTO_RELOAD; + if (!(event->attr.sample_type & + ~intel_pmu_free_running_flags(event))) + event->hw.flags |= PERF_X86_EVENT_FREERUNNING; + } + if (x86_pmu.pebs_aliases) + x86_pmu.pebs_aliases(event); + } if (needs_branch_stack(event)) { ret = intel_pmu_setup_lbr_filter(event); @@ -2544,19 +2749,11 @@ struct intel_shared_regs *allocate_shared_regs(int cpu) static struct intel_excl_cntrs *allocate_excl_cntrs(int cpu) { struct intel_excl_cntrs *c; - int i; c = kzalloc_node(sizeof(struct intel_excl_cntrs), GFP_KERNEL, cpu_to_node(cpu)); if (c) { raw_spin_lock_init(&c->lock); - for (i = 0; i < X86_PMC_IDX_MAX; i++) { - c->states[0].state[i] = INTEL_EXCL_UNUSED; - c->states[0].init_state[i] = INTEL_EXCL_UNUSED; - - c->states[1].state[i] = INTEL_EXCL_UNUSED; - c->states[1].init_state[i] = INTEL_EXCL_UNUSED; - } c->core_id = -1; } return c; @@ -2569,7 +2766,7 @@ static int intel_pmu_cpu_prepare(int cpu) if (x86_pmu.extra_regs || x86_pmu.lbr_sel_map) { cpuc->shared_regs = allocate_shared_regs(cpu); if (!cpuc->shared_regs) - return NOTIFY_BAD; + goto err; } if (x86_pmu.flags & PMU_FL_EXCL_CNTRS) { @@ -2577,18 +2774,27 @@ static int intel_pmu_cpu_prepare(int cpu) cpuc->constraint_list = kzalloc(sz, GFP_KERNEL); if (!cpuc->constraint_list) - return NOTIFY_BAD; + goto err_shared_regs; cpuc->excl_cntrs = allocate_excl_cntrs(cpu); - if (!cpuc->excl_cntrs) { - kfree(cpuc->constraint_list); - kfree(cpuc->shared_regs); - return NOTIFY_BAD; - } + if (!cpuc->excl_cntrs) + goto err_constraint_list; + cpuc->excl_thread_id = 0; } return NOTIFY_OK; + +err_constraint_list: + kfree(cpuc->constraint_list); + cpuc->constraint_list = NULL; + +err_shared_regs: + kfree(cpuc->shared_regs); + cpuc->shared_regs = NULL; + +err: + return NOTIFY_BAD; } static void intel_pmu_cpu_starting(int cpu) @@ -2611,7 +2817,7 @@ static void intel_pmu_cpu_starting(int cpu) if (!(x86_pmu.flags & PMU_FL_NO_HT_SHARING)) { void **onln = &cpuc->kfree_on_online[X86_PERF_KFREE_SHARED]; - for_each_cpu(i, topology_thread_cpumask(cpu)) { + for_each_cpu(i, topology_sibling_cpumask(cpu)) { struct intel_shared_regs *pc; pc = per_cpu(cpu_hw_events, i).shared_regs; @@ -2629,7 +2835,7 @@ static void intel_pmu_cpu_starting(int cpu) cpuc->lbr_sel = &cpuc->shared_regs->regs[EXTRA_REG_LBR]; if (x86_pmu.flags & PMU_FL_EXCL_CNTRS) { - for_each_cpu(i, topology_thread_cpumask(cpu)) { + for_each_cpu(i, topology_sibling_cpumask(cpu)) { struct intel_excl_cntrs *c; c = per_cpu(cpu_hw_events, i).excl_cntrs; @@ -2677,10 +2883,21 @@ static void intel_pmu_cpu_dying(int cpu) fini_debug_store_on_cpu(cpu); } +static void intel_pmu_sched_task(struct perf_event_context *ctx, + bool sched_in) +{ + if (x86_pmu.pebs_active) + intel_pmu_pebs_sched_task(ctx, sched_in); + if (x86_pmu.lbr_nr) + intel_pmu_lbr_sched_task(ctx, sched_in); +} + PMU_FORMAT_ATTR(offcore_rsp, "config1:0-63"); PMU_FORMAT_ATTR(ldlat, "config1:0-15"); +PMU_FORMAT_ATTR(frontend, "config1:0-23"); + static struct attribute *intel_arch3_formats_attr[] = { &format_attr_event.attr, &format_attr_umask.attr, @@ -2697,6 +2914,11 @@ static struct attribute *intel_arch3_formats_attr[] = { NULL, }; +static struct attribute *skl_format_attr[] = { + &format_attr_frontend.attr, + NULL, +}; + static __initconst const struct x86_pmu core_pmu = { .name = "core", .handle_irq = x86_pmu_handle_irq, @@ -2711,6 +2933,8 @@ static __initconst const struct x86_pmu core_pmu = { .event_map = intel_pmu_event_map, .max_events = ARRAY_SIZE(intel_perfmon_event_map), .apic = 1, + .free_running_flags = PEBS_FREERUNNING_FLAGS, + /* * Intel PMCs cannot be accessed sanely above 32-bit width, * so we install an artificial 1<<31 period regardless of @@ -2749,6 +2973,7 @@ static __initconst const struct x86_pmu intel_pmu = { .event_map = intel_pmu_event_map, .max_events = ARRAY_SIZE(intel_perfmon_event_map), .apic = 1, + .free_running_flags = PEBS_FREERUNNING_FLAGS, /* * Intel PMCs cannot be accessed sanely above 32 bit width, * so we install an artificial 1<<31 period regardless of @@ -2766,7 +2991,7 @@ static __initconst const struct x86_pmu intel_pmu = { .cpu_starting = intel_pmu_cpu_starting, .cpu_dying = intel_pmu_cpu_dying, .guest_get_msrs = intel_guest_get_msrs, - .sched_task = intel_pmu_lbr_sched_task, + .sched_task = intel_pmu_sched_task, }; static __init void intel_clovertown_quirk(void) @@ -2939,8 +3164,8 @@ static __init void intel_ht_bug(void) { x86_pmu.flags |= PMU_FL_EXCL_CNTRS | PMU_FL_EXCL_ENABLED; - x86_pmu.commit_scheduling = intel_commit_scheduling; x86_pmu.start_scheduling = intel_start_scheduling; + x86_pmu.commit_scheduling = intel_commit_scheduling; x86_pmu.stop_scheduling = intel_stop_scheduling; } @@ -3286,6 +3511,30 @@ __init int intel_pmu_init(void) pr_cont("Broadwell events, "); break; + case 78: /* 14nm Skylake Mobile */ + case 94: /* 14nm Skylake Desktop */ + x86_pmu.late_ack = true; + memcpy(hw_cache_event_ids, skl_hw_cache_event_ids, sizeof(hw_cache_event_ids)); + memcpy(hw_cache_extra_regs, skl_hw_cache_extra_regs, sizeof(hw_cache_extra_regs)); + intel_pmu_lbr_init_skl(); + + x86_pmu.event_constraints = intel_skl_event_constraints; + x86_pmu.pebs_constraints = intel_skl_pebs_event_constraints; + x86_pmu.extra_regs = intel_skl_extra_regs; + x86_pmu.pebs_aliases = intel_pebs_aliases_snb; + /* all extra regs are per-cpu when HT is on */ + x86_pmu.flags |= PMU_FL_HAS_RSP_1; + x86_pmu.flags |= PMU_FL_NO_HT_SHARING; + + x86_pmu.hw_config = hsw_hw_config; + x86_pmu.get_event_constraints = hsw_get_event_constraints; + x86_pmu.format_attrs = merge_attr(intel_arch3_formats_attr, + skl_format_attr); + WARN_ON(!x86_pmu.format_attrs); + x86_pmu.cpu_events = hsw_events_attrs; + pr_cont("Skylake events, "); + break; + default: switch (x86_pmu.version) { case 1: @@ -3355,7 +3604,7 @@ __init int intel_pmu_init(void) */ if (x86_pmu.extra_regs) { for (er = x86_pmu.extra_regs; er->msr; er++) { - er->extra_msr_access = check_msr(er->msr, 0x1ffUL); + er->extra_msr_access = check_msr(er->msr, 0x11UL); /* Disable LBR select mapping */ if ((er->idx == EXTRA_REG_LBR) && !er->extra_msr_access) x86_pmu.lbr_sel_map = NULL; @@ -3388,21 +3637,24 @@ static __init int fixup_ht_bug(void) if (!(x86_pmu.flags & PMU_FL_EXCL_ENABLED)) return 0; - w = cpumask_weight(topology_thread_cpumask(cpu)); + w = cpumask_weight(topology_sibling_cpumask(cpu)); if (w > 1) { pr_info("PMU erratum BJ122, BV98, HSD29 worked around, HT is on\n"); return 0; } - watchdog_nmi_disable_all(); + if (lockup_detector_suspend() != 0) { + pr_debug("failed to disable PMU erratum BJ122, BV98, HSD29 workaround\n"); + return 0; + } x86_pmu.flags &= ~(PMU_FL_EXCL_CNTRS | PMU_FL_EXCL_ENABLED); - x86_pmu.commit_scheduling = NULL; x86_pmu.start_scheduling = NULL; + x86_pmu.commit_scheduling = NULL; x86_pmu.stop_scheduling = NULL; - watchdog_nmi_enable_all(); + lockup_detector_resume(); get_online_cpus(); diff --git a/kernel/arch/x86/kernel/cpu/perf_event_intel_bts.c b/kernel/arch/x86/kernel/cpu/perf_event_intel_bts.c index 7795f3f8b..2cad71d1b 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event_intel_bts.c +++ b/kernel/arch/x86/kernel/cpu/perf_event_intel_bts.c @@ -62,9 +62,6 @@ struct bts_buffer { struct pmu bts_pmu; -void intel_pmu_enable_bts(u64 config); -void intel_pmu_disable_bts(void); - static size_t buf_size(struct page *page) { return 1 << (PAGE_SHIFT + page_private(page)); @@ -225,6 +222,7 @@ static void __bts_event_start(struct perf_event *event) if (!buf || bts_buffer_is_full(buf, bts)) return; + event->hw.itrace_started = 1; event->hw.state = 0; if (!buf->snapshot) @@ -497,6 +495,19 @@ static int bts_event_init(struct perf_event *event) if (x86_add_exclusive(x86_lbr_exclusive_bts)) return -EBUSY; + /* + * BTS leaks kernel addresses even when CPL0 tracing is + * disabled, so disallow intel_bts driver for unprivileged + * users on paranoid systems since it provides trace data + * to the user in a zero-copy fashion. + * + * Note that the default paranoia setting permits unprivileged + * users to profile the kernel. + */ + if (event->attr.exclude_kernel && perf_paranoid_kernel() && + !capable(CAP_SYS_ADMIN)) + return -EACCES; + ret = x86_reserve_hardware(); if (ret) { x86_del_exclusive(x86_lbr_exclusive_bts); @@ -530,5 +541,4 @@ static __init int bts_init(void) return perf_pmu_register(&bts_pmu, "intel_bts", -1); } - -module_init(bts_init); +arch_initcall(bts_init); diff --git a/kernel/arch/x86/kernel/cpu/perf_event_intel_cqm.c b/kernel/arch/x86/kernel/cpu/perf_event_intel_cqm.c index cb77b11bc..a316ca96f 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event_intel_cqm.c +++ b/kernel/arch/x86/kernel/cpu/perf_event_intel_cqm.c @@ -13,16 +13,35 @@ #define MSR_IA32_QM_CTR 0x0c8e #define MSR_IA32_QM_EVTSEL 0x0c8d -static unsigned int cqm_max_rmid = -1; +static u32 cqm_max_rmid = -1; static unsigned int cqm_l3_scale; /* supposedly cacheline size */ -struct intel_cqm_state { - raw_spinlock_t lock; - int rmid; - int cnt; +/** + * struct intel_pqr_state - State cache for the PQR MSR + * @rmid: The cached Resource Monitoring ID + * @closid: The cached Class Of Service ID + * @rmid_usecnt: The usage counter for rmid + * + * The upper 32 bits of MSR_IA32_PQR_ASSOC contain closid and the + * lower 10 bits rmid. The update to MSR_IA32_PQR_ASSOC always + * contains both parts, so we need to cache them. + * + * The cache also helps to avoid pointless updates if the value does + * not change. + */ +struct intel_pqr_state { + u32 rmid; + u32 closid; + int rmid_usecnt; }; -static DEFINE_PER_CPU(struct intel_cqm_state, cqm_state); +/* + * The cached intel_pqr_state is strictly per CPU and can never be + * updated from a remote CPU. Both functions which modify the state + * (intel_cqm_event_start and intel_cqm_event_stop) are called with + * interrupts disabled, which is sufficient for the protection. + */ +static DEFINE_PER_CPU(struct intel_pqr_state, pqr_state); /* * Protects cache_cgroups and cqm_rmid_free_lru and cqm_rmid_limbo_lru. @@ -57,7 +76,7 @@ static cpumask_t cqm_cpumask; * near-zero occupancy value, i.e. no cachelines are tagged with this * RMID, once __intel_cqm_rmid_rotate() returns. */ -static unsigned int intel_cqm_rotation_rmid; +static u32 intel_cqm_rotation_rmid; #define INVALID_RMID (-1) @@ -69,7 +88,7 @@ static unsigned int intel_cqm_rotation_rmid; * Likewise, an rmid value of -1 is used to indicate "no rmid currently * assigned" and is used as part of the rotation code. */ -static inline bool __rmid_valid(unsigned int rmid) +static inline bool __rmid_valid(u32 rmid) { if (!rmid || rmid == INVALID_RMID) return false; @@ -77,7 +96,7 @@ static inline bool __rmid_valid(unsigned int rmid) return true; } -static u64 __rmid_read(unsigned int rmid) +static u64 __rmid_read(u32 rmid) { u64 val; @@ -102,7 +121,7 @@ enum rmid_recycle_state { }; struct cqm_rmid_entry { - unsigned int rmid; + u32 rmid; enum rmid_recycle_state state; struct list_head list; unsigned long queue_time; @@ -147,7 +166,7 @@ static LIST_HEAD(cqm_rmid_limbo_lru); */ static struct cqm_rmid_entry **cqm_rmid_ptrs; -static inline struct cqm_rmid_entry *__rmid_entry(int rmid) +static inline struct cqm_rmid_entry *__rmid_entry(u32 rmid) { struct cqm_rmid_entry *entry; @@ -162,7 +181,7 @@ static inline struct cqm_rmid_entry *__rmid_entry(int rmid) * * We expect to be called with cache_mutex held. */ -static int __get_rmid(void) +static u32 __get_rmid(void) { struct cqm_rmid_entry *entry; @@ -177,7 +196,7 @@ static int __get_rmid(void) return entry->rmid; } -static void __put_rmid(unsigned int rmid) +static void __put_rmid(u32 rmid) { struct cqm_rmid_entry *entry; @@ -279,7 +298,7 @@ static bool __match_event(struct perf_event *a, struct perf_event *b) static inline struct perf_cgroup *event_to_cgroup(struct perf_event *event) { if (event->attach_state & PERF_ATTACH_TASK) - return perf_cgroup_from_task(event->hw.target); + return perf_cgroup_from_task(event->hw.target, event->ctx); return event->cgrp; } @@ -372,7 +391,7 @@ static bool __conflict_event(struct perf_event *a, struct perf_event *b) } struct rmid_read { - unsigned int rmid; + u32 rmid; atomic64_t value; }; @@ -381,12 +400,11 @@ static void __intel_cqm_event_count(void *info); /* * Exchange the RMID of a group of events. */ -static unsigned int -intel_cqm_xchg_rmid(struct perf_event *group, unsigned int rmid) +static u32 intel_cqm_xchg_rmid(struct perf_event *group, u32 rmid) { struct perf_event *event; - unsigned int old_rmid = group->hw.cqm_rmid; struct list_head *head = &group->hw.cqm_group_entry; + u32 old_rmid = group->hw.cqm_rmid; lockdep_assert_held(&cache_mutex); @@ -451,7 +469,7 @@ static void intel_cqm_stable(void *arg) * If we have group events waiting for an RMID that don't conflict with * events already running, assign @rmid. */ -static bool intel_cqm_sched_in_event(unsigned int rmid) +static bool intel_cqm_sched_in_event(u32 rmid) { struct perf_event *leader, *event; @@ -598,7 +616,7 @@ static bool intel_cqm_rmid_stabilize(unsigned int *available) static void __intel_cqm_pick_and_rotate(struct perf_event *next) { struct perf_event *rotor; - unsigned int rmid; + u32 rmid; lockdep_assert_held(&cache_mutex); @@ -626,7 +644,7 @@ static void __intel_cqm_pick_and_rotate(struct perf_event *next) static void intel_cqm_sched_out_conflicting_events(struct perf_event *event) { struct perf_event *group, *g; - unsigned int rmid; + u32 rmid; lockdep_assert_held(&cache_mutex); @@ -828,8 +846,8 @@ static void intel_cqm_setup_event(struct perf_event *event, struct perf_event **group) { struct perf_event *iter; - unsigned int rmid; bool conflict = false; + u32 rmid; list_for_each_entry(iter, &cache_groups, hw.cqm_groups_entry) { rmid = iter->hw.cqm_rmid; @@ -860,7 +878,7 @@ static void intel_cqm_setup_event(struct perf_event *event, static void intel_cqm_event_read(struct perf_event *event) { unsigned long flags; - unsigned int rmid; + u32 rmid; u64 val; /* @@ -969,55 +987,48 @@ out: static void intel_cqm_event_start(struct perf_event *event, int mode) { - struct intel_cqm_state *state = this_cpu_ptr(&cqm_state); - unsigned int rmid = event->hw.cqm_rmid; - unsigned long flags; + struct intel_pqr_state *state = this_cpu_ptr(&pqr_state); + u32 rmid = event->hw.cqm_rmid; if (!(event->hw.cqm_state & PERF_HES_STOPPED)) return; event->hw.cqm_state &= ~PERF_HES_STOPPED; - raw_spin_lock_irqsave(&state->lock, flags); - - if (state->cnt++) - WARN_ON_ONCE(state->rmid != rmid); - else + if (state->rmid_usecnt++) { + if (!WARN_ON_ONCE(state->rmid != rmid)) + return; + } else { WARN_ON_ONCE(state->rmid); + } state->rmid = rmid; - wrmsrl(MSR_IA32_PQR_ASSOC, state->rmid); - - raw_spin_unlock_irqrestore(&state->lock, flags); + wrmsr(MSR_IA32_PQR_ASSOC, rmid, state->closid); } static void intel_cqm_event_stop(struct perf_event *event, int mode) { - struct intel_cqm_state *state = this_cpu_ptr(&cqm_state); - unsigned long flags; + struct intel_pqr_state *state = this_cpu_ptr(&pqr_state); if (event->hw.cqm_state & PERF_HES_STOPPED) return; event->hw.cqm_state |= PERF_HES_STOPPED; - raw_spin_lock_irqsave(&state->lock, flags); intel_cqm_event_read(event); - if (!--state->cnt) { + if (!--state->rmid_usecnt) { state->rmid = 0; - wrmsrl(MSR_IA32_PQR_ASSOC, 0); + wrmsr(MSR_IA32_PQR_ASSOC, 0, state->closid); } else { WARN_ON_ONCE(!state->rmid); } - - raw_spin_unlock_irqrestore(&state->lock, flags); } static int intel_cqm_event_add(struct perf_event *event, int mode) { unsigned long flags; - unsigned int rmid; + u32 rmid; raw_spin_lock_irqsave(&cache_lock, flags); @@ -1032,11 +1043,6 @@ static int intel_cqm_event_add(struct perf_event *event, int mode) return 0; } -static void intel_cqm_event_del(struct perf_event *event, int mode) -{ - intel_cqm_event_stop(event, mode); -} - static void intel_cqm_event_destroy(struct perf_event *event) { struct perf_event *group_other = NULL; @@ -1065,7 +1071,7 @@ static void intel_cqm_event_destroy(struct perf_event *event) list_replace(&event->hw.cqm_groups_entry, &group_other->hw.cqm_groups_entry); } else { - unsigned int rmid = event->hw.cqm_rmid; + u32 rmid = event->hw.cqm_rmid; if (__rmid_valid(rmid)) __put_rmid(rmid); @@ -1229,7 +1235,7 @@ static struct pmu intel_cqm_pmu = { .task_ctx_nr = perf_sw_context, .event_init = intel_cqm_event_init, .add = intel_cqm_event_add, - .del = intel_cqm_event_del, + .del = intel_cqm_event_stop, .start = intel_cqm_event_start, .stop = intel_cqm_event_stop, .read = intel_cqm_event_read, @@ -1249,14 +1255,14 @@ static inline void cqm_pick_event_reader(int cpu) cpumask_set_cpu(cpu, &cqm_cpumask); } -static void intel_cqm_cpu_prepare(unsigned int cpu) +static void intel_cqm_cpu_starting(unsigned int cpu) { - struct intel_cqm_state *state = &per_cpu(cqm_state, cpu); + struct intel_pqr_state *state = &per_cpu(pqr_state, cpu); struct cpuinfo_x86 *c = &cpu_data(cpu); - raw_spin_lock_init(&state->lock); state->rmid = 0; - state->cnt = 0; + state->closid = 0; + state->rmid_usecnt = 0; WARN_ON(c->x86_cache_max_rmid != cqm_max_rmid); WARN_ON(c->x86_cache_occ_scale != cqm_l3_scale); @@ -1290,13 +1296,11 @@ static int intel_cqm_cpu_notifier(struct notifier_block *nb, unsigned int cpu = (unsigned long)hcpu; switch (action & ~CPU_TASKS_FROZEN) { - case CPU_UP_PREPARE: - intel_cqm_cpu_prepare(cpu); - break; case CPU_DOWN_PREPARE: intel_cqm_cpu_exit(cpu); break; case CPU_STARTING: + intel_cqm_cpu_starting(cpu); cqm_pick_event_reader(cpu); break; } @@ -1367,7 +1371,7 @@ static int __init intel_cqm_init(void) goto out; for_each_online_cpu(i) { - intel_cqm_cpu_prepare(i); + intel_cqm_cpu_starting(i); cqm_pick_event_reader(i); } diff --git a/kernel/arch/x86/kernel/cpu/perf_event_intel_cstate.c b/kernel/arch/x86/kernel/cpu/perf_event_intel_cstate.c new file mode 100644 index 000000000..75a38b5a2 --- /dev/null +++ b/kernel/arch/x86/kernel/cpu/perf_event_intel_cstate.c @@ -0,0 +1,694 @@ +/* + * perf_event_intel_cstate.c: support cstate residency counters + * + * Copyright (C) 2015, Intel Corp. + * Author: Kan Liang (kan.liang@intel.com) + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Library General Public + * License as published by the Free Software Foundation; either + * version 2 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Library General Public License for more details. + * + */ + +/* + * This file export cstate related free running (read-only) counters + * for perf. These counters may be use simultaneously by other tools, + * such as turbostat. However, it still make sense to implement them + * in perf. Because we can conveniently collect them together with + * other events, and allow to use them from tools without special MSR + * access code. + * + * The events only support system-wide mode counting. There is no + * sampling support because it is not supported by the hardware. + * + * According to counters' scope and category, two PMUs are registered + * with the perf_event core subsystem. + * - 'cstate_core': The counter is available for each physical core. + * The counters include CORE_C*_RESIDENCY. + * - 'cstate_pkg': The counter is available for each physical package. + * The counters include PKG_C*_RESIDENCY. + * + * All of these counters are specified in the Intel® 64 and IA-32 + * Architectures Software Developer.s Manual Vol3b. + * + * Model specific counters: + * MSR_CORE_C1_RES: CORE C1 Residency Counter + * perf code: 0x00 + * Available model: SLM,AMT + * Scope: Core (each processor core has a MSR) + * MSR_CORE_C3_RESIDENCY: CORE C3 Residency Counter + * perf code: 0x01 + * Available model: NHM,WSM,SNB,IVB,HSW,BDW,SKL + * Scope: Core + * MSR_CORE_C6_RESIDENCY: CORE C6 Residency Counter + * perf code: 0x02 + * Available model: SLM,AMT,NHM,WSM,SNB,IVB,HSW,BDW,SKL + * Scope: Core + * MSR_CORE_C7_RESIDENCY: CORE C7 Residency Counter + * perf code: 0x03 + * Available model: SNB,IVB,HSW,BDW,SKL + * Scope: Core + * MSR_PKG_C2_RESIDENCY: Package C2 Residency Counter. + * perf code: 0x00 + * Available model: SNB,IVB,HSW,BDW,SKL + * Scope: Package (physical package) + * MSR_PKG_C3_RESIDENCY: Package C3 Residency Counter. + * perf code: 0x01 + * Available model: NHM,WSM,SNB,IVB,HSW,BDW,SKL + * Scope: Package (physical package) + * MSR_PKG_C6_RESIDENCY: Package C6 Residency Counter. + * perf code: 0x02 + * Available model: SLM,AMT,NHM,WSM,SNB,IVB,HSW,BDW,SKL + * Scope: Package (physical package) + * MSR_PKG_C7_RESIDENCY: Package C7 Residency Counter. + * perf code: 0x03 + * Available model: NHM,WSM,SNB,IVB,HSW,BDW,SKL + * Scope: Package (physical package) + * MSR_PKG_C8_RESIDENCY: Package C8 Residency Counter. + * perf code: 0x04 + * Available model: HSW ULT only + * Scope: Package (physical package) + * MSR_PKG_C9_RESIDENCY: Package C9 Residency Counter. + * perf code: 0x05 + * Available model: HSW ULT only + * Scope: Package (physical package) + * MSR_PKG_C10_RESIDENCY: Package C10 Residency Counter. + * perf code: 0x06 + * Available model: HSW ULT only + * Scope: Package (physical package) + * + */ + +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/perf_event.h> +#include <asm/cpu_device_id.h> +#include "perf_event.h" + +#define DEFINE_CSTATE_FORMAT_ATTR(_var, _name, _format) \ +static ssize_t __cstate_##_var##_show(struct kobject *kobj, \ + struct kobj_attribute *attr, \ + char *page) \ +{ \ + BUILD_BUG_ON(sizeof(_format) >= PAGE_SIZE); \ + return sprintf(page, _format "\n"); \ +} \ +static struct kobj_attribute format_attr_##_var = \ + __ATTR(_name, 0444, __cstate_##_var##_show, NULL) + +static ssize_t cstate_get_attr_cpumask(struct device *dev, + struct device_attribute *attr, + char *buf); + +struct perf_cstate_msr { + u64 msr; + struct perf_pmu_events_attr *attr; + bool (*test)(int idx); +}; + + +/* cstate_core PMU */ + +static struct pmu cstate_core_pmu; +static bool has_cstate_core; + +enum perf_cstate_core_id { + /* + * cstate_core events + */ + PERF_CSTATE_CORE_C1_RES = 0, + PERF_CSTATE_CORE_C3_RES, + PERF_CSTATE_CORE_C6_RES, + PERF_CSTATE_CORE_C7_RES, + + PERF_CSTATE_CORE_EVENT_MAX, +}; + +bool test_core(int idx) +{ + if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL || + boot_cpu_data.x86 != 6) + return false; + + switch (boot_cpu_data.x86_model) { + case 30: /* 45nm Nehalem */ + case 26: /* 45nm Nehalem-EP */ + case 46: /* 45nm Nehalem-EX */ + + case 37: /* 32nm Westmere */ + case 44: /* 32nm Westmere-EP */ + case 47: /* 32nm Westmere-EX */ + if (idx == PERF_CSTATE_CORE_C3_RES || + idx == PERF_CSTATE_CORE_C6_RES) + return true; + break; + case 42: /* 32nm SandyBridge */ + case 45: /* 32nm SandyBridge-E/EN/EP */ + + case 58: /* 22nm IvyBridge */ + case 62: /* 22nm IvyBridge-EP/EX */ + + case 60: /* 22nm Haswell Core */ + case 63: /* 22nm Haswell Server */ + case 69: /* 22nm Haswell ULT */ + case 70: /* 22nm Haswell + GT3e (Intel Iris Pro graphics) */ + + case 61: /* 14nm Broadwell Core-M */ + case 86: /* 14nm Broadwell Xeon D */ + case 71: /* 14nm Broadwell + GT3e (Intel Iris Pro graphics) */ + case 79: /* 14nm Broadwell Server */ + + case 78: /* 14nm Skylake Mobile */ + case 94: /* 14nm Skylake Desktop */ + if (idx == PERF_CSTATE_CORE_C3_RES || + idx == PERF_CSTATE_CORE_C6_RES || + idx == PERF_CSTATE_CORE_C7_RES) + return true; + break; + case 55: /* 22nm Atom "Silvermont" */ + case 77: /* 22nm Atom "Silvermont Avoton/Rangely" */ + case 76: /* 14nm Atom "Airmont" */ + if (idx == PERF_CSTATE_CORE_C1_RES || + idx == PERF_CSTATE_CORE_C6_RES) + return true; + break; + } + + return false; +} + +PMU_EVENT_ATTR_STRING(c1-residency, evattr_cstate_core_c1, "event=0x00"); +PMU_EVENT_ATTR_STRING(c3-residency, evattr_cstate_core_c3, "event=0x01"); +PMU_EVENT_ATTR_STRING(c6-residency, evattr_cstate_core_c6, "event=0x02"); +PMU_EVENT_ATTR_STRING(c7-residency, evattr_cstate_core_c7, "event=0x03"); + +static struct perf_cstate_msr core_msr[] = { + [PERF_CSTATE_CORE_C1_RES] = { MSR_CORE_C1_RES, &evattr_cstate_core_c1, test_core, }, + [PERF_CSTATE_CORE_C3_RES] = { MSR_CORE_C3_RESIDENCY, &evattr_cstate_core_c3, test_core, }, + [PERF_CSTATE_CORE_C6_RES] = { MSR_CORE_C6_RESIDENCY, &evattr_cstate_core_c6, test_core, }, + [PERF_CSTATE_CORE_C7_RES] = { MSR_CORE_C7_RESIDENCY, &evattr_cstate_core_c7, test_core, }, +}; + +static struct attribute *core_events_attrs[PERF_CSTATE_CORE_EVENT_MAX + 1] = { + NULL, +}; + +static struct attribute_group core_events_attr_group = { + .name = "events", + .attrs = core_events_attrs, +}; + +DEFINE_CSTATE_FORMAT_ATTR(core_event, event, "config:0-63"); +static struct attribute *core_format_attrs[] = { + &format_attr_core_event.attr, + NULL, +}; + +static struct attribute_group core_format_attr_group = { + .name = "format", + .attrs = core_format_attrs, +}; + +static cpumask_t cstate_core_cpu_mask; +static DEVICE_ATTR(cpumask, S_IRUGO, cstate_get_attr_cpumask, NULL); + +static struct attribute *cstate_cpumask_attrs[] = { + &dev_attr_cpumask.attr, + NULL, +}; + +static struct attribute_group cpumask_attr_group = { + .attrs = cstate_cpumask_attrs, +}; + +static const struct attribute_group *core_attr_groups[] = { + &core_events_attr_group, + &core_format_attr_group, + &cpumask_attr_group, + NULL, +}; + +/* cstate_core PMU end */ + + +/* cstate_pkg PMU */ + +static struct pmu cstate_pkg_pmu; +static bool has_cstate_pkg; + +enum perf_cstate_pkg_id { + /* + * cstate_pkg events + */ + PERF_CSTATE_PKG_C2_RES = 0, + PERF_CSTATE_PKG_C3_RES, + PERF_CSTATE_PKG_C6_RES, + PERF_CSTATE_PKG_C7_RES, + PERF_CSTATE_PKG_C8_RES, + PERF_CSTATE_PKG_C9_RES, + PERF_CSTATE_PKG_C10_RES, + + PERF_CSTATE_PKG_EVENT_MAX, +}; + +bool test_pkg(int idx) +{ + if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL || + boot_cpu_data.x86 != 6) + return false; + + switch (boot_cpu_data.x86_model) { + case 30: /* 45nm Nehalem */ + case 26: /* 45nm Nehalem-EP */ + case 46: /* 45nm Nehalem-EX */ + + case 37: /* 32nm Westmere */ + case 44: /* 32nm Westmere-EP */ + case 47: /* 32nm Westmere-EX */ + if (idx == PERF_CSTATE_CORE_C3_RES || + idx == PERF_CSTATE_CORE_C6_RES || + idx == PERF_CSTATE_CORE_C7_RES) + return true; + break; + case 42: /* 32nm SandyBridge */ + case 45: /* 32nm SandyBridge-E/EN/EP */ + + case 58: /* 22nm IvyBridge */ + case 62: /* 22nm IvyBridge-EP/EX */ + + case 60: /* 22nm Haswell Core */ + case 63: /* 22nm Haswell Server */ + case 70: /* 22nm Haswell + GT3e (Intel Iris Pro graphics) */ + + case 61: /* 14nm Broadwell Core-M */ + case 86: /* 14nm Broadwell Xeon D */ + case 71: /* 14nm Broadwell + GT3e (Intel Iris Pro graphics) */ + case 79: /* 14nm Broadwell Server */ + + case 78: /* 14nm Skylake Mobile */ + case 94: /* 14nm Skylake Desktop */ + if (idx == PERF_CSTATE_PKG_C2_RES || + idx == PERF_CSTATE_PKG_C3_RES || + idx == PERF_CSTATE_PKG_C6_RES || + idx == PERF_CSTATE_PKG_C7_RES) + return true; + break; + case 55: /* 22nm Atom "Silvermont" */ + case 77: /* 22nm Atom "Silvermont Avoton/Rangely" */ + case 76: /* 14nm Atom "Airmont" */ + if (idx == PERF_CSTATE_CORE_C6_RES) + return true; + break; + case 69: /* 22nm Haswell ULT */ + if (idx == PERF_CSTATE_PKG_C2_RES || + idx == PERF_CSTATE_PKG_C3_RES || + idx == PERF_CSTATE_PKG_C6_RES || + idx == PERF_CSTATE_PKG_C7_RES || + idx == PERF_CSTATE_PKG_C8_RES || + idx == PERF_CSTATE_PKG_C9_RES || + idx == PERF_CSTATE_PKG_C10_RES) + return true; + break; + } + + return false; +} + +PMU_EVENT_ATTR_STRING(c2-residency, evattr_cstate_pkg_c2, "event=0x00"); +PMU_EVENT_ATTR_STRING(c3-residency, evattr_cstate_pkg_c3, "event=0x01"); +PMU_EVENT_ATTR_STRING(c6-residency, evattr_cstate_pkg_c6, "event=0x02"); +PMU_EVENT_ATTR_STRING(c7-residency, evattr_cstate_pkg_c7, "event=0x03"); +PMU_EVENT_ATTR_STRING(c8-residency, evattr_cstate_pkg_c8, "event=0x04"); +PMU_EVENT_ATTR_STRING(c9-residency, evattr_cstate_pkg_c9, "event=0x05"); +PMU_EVENT_ATTR_STRING(c10-residency, evattr_cstate_pkg_c10, "event=0x06"); + +static struct perf_cstate_msr pkg_msr[] = { + [PERF_CSTATE_PKG_C2_RES] = { MSR_PKG_C2_RESIDENCY, &evattr_cstate_pkg_c2, test_pkg, }, + [PERF_CSTATE_PKG_C3_RES] = { MSR_PKG_C3_RESIDENCY, &evattr_cstate_pkg_c3, test_pkg, }, + [PERF_CSTATE_PKG_C6_RES] = { MSR_PKG_C6_RESIDENCY, &evattr_cstate_pkg_c6, test_pkg, }, + [PERF_CSTATE_PKG_C7_RES] = { MSR_PKG_C7_RESIDENCY, &evattr_cstate_pkg_c7, test_pkg, }, + [PERF_CSTATE_PKG_C8_RES] = { MSR_PKG_C8_RESIDENCY, &evattr_cstate_pkg_c8, test_pkg, }, + [PERF_CSTATE_PKG_C9_RES] = { MSR_PKG_C9_RESIDENCY, &evattr_cstate_pkg_c9, test_pkg, }, + [PERF_CSTATE_PKG_C10_RES] = { MSR_PKG_C10_RESIDENCY, &evattr_cstate_pkg_c10, test_pkg, }, +}; + +static struct attribute *pkg_events_attrs[PERF_CSTATE_PKG_EVENT_MAX + 1] = { + NULL, +}; + +static struct attribute_group pkg_events_attr_group = { + .name = "events", + .attrs = pkg_events_attrs, +}; + +DEFINE_CSTATE_FORMAT_ATTR(pkg_event, event, "config:0-63"); +static struct attribute *pkg_format_attrs[] = { + &format_attr_pkg_event.attr, + NULL, +}; +static struct attribute_group pkg_format_attr_group = { + .name = "format", + .attrs = pkg_format_attrs, +}; + +static cpumask_t cstate_pkg_cpu_mask; + +static const struct attribute_group *pkg_attr_groups[] = { + &pkg_events_attr_group, + &pkg_format_attr_group, + &cpumask_attr_group, + NULL, +}; + +/* cstate_pkg PMU end*/ + +static ssize_t cstate_get_attr_cpumask(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct pmu *pmu = dev_get_drvdata(dev); + + if (pmu == &cstate_core_pmu) + return cpumap_print_to_pagebuf(true, buf, &cstate_core_cpu_mask); + else if (pmu == &cstate_pkg_pmu) + return cpumap_print_to_pagebuf(true, buf, &cstate_pkg_cpu_mask); + else + return 0; +} + +static int cstate_pmu_event_init(struct perf_event *event) +{ + u64 cfg = event->attr.config; + int ret = 0; + + if (event->attr.type != event->pmu->type) + return -ENOENT; + + /* unsupported modes and filters */ + if (event->attr.exclude_user || + event->attr.exclude_kernel || + event->attr.exclude_hv || + event->attr.exclude_idle || + event->attr.exclude_host || + event->attr.exclude_guest || + event->attr.sample_period) /* no sampling */ + return -EINVAL; + + if (event->pmu == &cstate_core_pmu) { + if (cfg >= PERF_CSTATE_CORE_EVENT_MAX) + return -EINVAL; + if (!core_msr[cfg].attr) + return -EINVAL; + event->hw.event_base = core_msr[cfg].msr; + } else if (event->pmu == &cstate_pkg_pmu) { + if (cfg >= PERF_CSTATE_PKG_EVENT_MAX) + return -EINVAL; + if (!pkg_msr[cfg].attr) + return -EINVAL; + event->hw.event_base = pkg_msr[cfg].msr; + } else + return -ENOENT; + + /* must be done before validate_group */ + event->hw.config = cfg; + event->hw.idx = -1; + + return ret; +} + +static inline u64 cstate_pmu_read_counter(struct perf_event *event) +{ + u64 val; + + rdmsrl(event->hw.event_base, val); + return val; +} + +static void cstate_pmu_event_update(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + u64 prev_raw_count, new_raw_count; + +again: + prev_raw_count = local64_read(&hwc->prev_count); + new_raw_count = cstate_pmu_read_counter(event); + + if (local64_cmpxchg(&hwc->prev_count, prev_raw_count, + new_raw_count) != prev_raw_count) + goto again; + + local64_add(new_raw_count - prev_raw_count, &event->count); +} + +static void cstate_pmu_event_start(struct perf_event *event, int mode) +{ + local64_set(&event->hw.prev_count, cstate_pmu_read_counter(event)); +} + +static void cstate_pmu_event_stop(struct perf_event *event, int mode) +{ + cstate_pmu_event_update(event); +} + +static void cstate_pmu_event_del(struct perf_event *event, int mode) +{ + cstate_pmu_event_stop(event, PERF_EF_UPDATE); +} + +static int cstate_pmu_event_add(struct perf_event *event, int mode) +{ + if (mode & PERF_EF_START) + cstate_pmu_event_start(event, mode); + + return 0; +} + +static void cstate_cpu_exit(int cpu) +{ + int i, id, target; + + /* cpu exit for cstate core */ + if (has_cstate_core) { + id = topology_core_id(cpu); + target = -1; + + for_each_online_cpu(i) { + if (i == cpu) + continue; + if (id == topology_core_id(i)) { + target = i; + break; + } + } + if (cpumask_test_and_clear_cpu(cpu, &cstate_core_cpu_mask) && target >= 0) + cpumask_set_cpu(target, &cstate_core_cpu_mask); + WARN_ON(cpumask_empty(&cstate_core_cpu_mask)); + if (target >= 0) + perf_pmu_migrate_context(&cstate_core_pmu, cpu, target); + } + + /* cpu exit for cstate pkg */ + if (has_cstate_pkg) { + id = topology_physical_package_id(cpu); + target = -1; + + for_each_online_cpu(i) { + if (i == cpu) + continue; + if (id == topology_physical_package_id(i)) { + target = i; + break; + } + } + if (cpumask_test_and_clear_cpu(cpu, &cstate_pkg_cpu_mask) && target >= 0) + cpumask_set_cpu(target, &cstate_pkg_cpu_mask); + WARN_ON(cpumask_empty(&cstate_pkg_cpu_mask)); + if (target >= 0) + perf_pmu_migrate_context(&cstate_pkg_pmu, cpu, target); + } +} + +static void cstate_cpu_init(int cpu) +{ + int i, id; + + /* cpu init for cstate core */ + if (has_cstate_core) { + id = topology_core_id(cpu); + for_each_cpu(i, &cstate_core_cpu_mask) { + if (id == topology_core_id(i)) + break; + } + if (i >= nr_cpu_ids) + cpumask_set_cpu(cpu, &cstate_core_cpu_mask); + } + + /* cpu init for cstate pkg */ + if (has_cstate_pkg) { + id = topology_physical_package_id(cpu); + for_each_cpu(i, &cstate_pkg_cpu_mask) { + if (id == topology_physical_package_id(i)) + break; + } + if (i >= nr_cpu_ids) + cpumask_set_cpu(cpu, &cstate_pkg_cpu_mask); + } +} + +static int cstate_cpu_notifier(struct notifier_block *self, + unsigned long action, void *hcpu) +{ + unsigned int cpu = (long)hcpu; + + switch (action & ~CPU_TASKS_FROZEN) { + case CPU_UP_PREPARE: + break; + case CPU_STARTING: + cstate_cpu_init(cpu); + break; + case CPU_UP_CANCELED: + case CPU_DYING: + break; + case CPU_ONLINE: + case CPU_DEAD: + break; + case CPU_DOWN_PREPARE: + cstate_cpu_exit(cpu); + break; + default: + break; + } + + return NOTIFY_OK; +} + +/* + * Probe the cstate events and insert the available one into sysfs attrs + * Return false if there is no available events. + */ +static bool cstate_probe_msr(struct perf_cstate_msr *msr, + struct attribute **events_attrs, + int max_event_nr) +{ + int i, j = 0; + u64 val; + + /* Probe the cstate events. */ + for (i = 0; i < max_event_nr; i++) { + if (!msr[i].test(i) || rdmsrl_safe(msr[i].msr, &val)) + msr[i].attr = NULL; + } + + /* List remaining events in the sysfs attrs. */ + for (i = 0; i < max_event_nr; i++) { + if (msr[i].attr) + events_attrs[j++] = &msr[i].attr->attr.attr; + } + events_attrs[j] = NULL; + + return (j > 0) ? true : false; +} + +static int __init cstate_init(void) +{ + /* SLM has different MSR for PKG C6 */ + switch (boot_cpu_data.x86_model) { + case 55: + case 76: + case 77: + pkg_msr[PERF_CSTATE_PKG_C6_RES].msr = MSR_PKG_C7_RESIDENCY; + } + + if (cstate_probe_msr(core_msr, core_events_attrs, PERF_CSTATE_CORE_EVENT_MAX)) + has_cstate_core = true; + + if (cstate_probe_msr(pkg_msr, pkg_events_attrs, PERF_CSTATE_PKG_EVENT_MAX)) + has_cstate_pkg = true; + + return (has_cstate_core || has_cstate_pkg) ? 0 : -ENODEV; +} + +static void __init cstate_cpumask_init(void) +{ + int cpu; + + cpu_notifier_register_begin(); + + for_each_online_cpu(cpu) + cstate_cpu_init(cpu); + + __perf_cpu_notifier(cstate_cpu_notifier); + + cpu_notifier_register_done(); +} + +static struct pmu cstate_core_pmu = { + .attr_groups = core_attr_groups, + .name = "cstate_core", + .task_ctx_nr = perf_invalid_context, + .event_init = cstate_pmu_event_init, + .add = cstate_pmu_event_add, /* must have */ + .del = cstate_pmu_event_del, /* must have */ + .start = cstate_pmu_event_start, + .stop = cstate_pmu_event_stop, + .read = cstate_pmu_event_update, + .capabilities = PERF_PMU_CAP_NO_INTERRUPT, +}; + +static struct pmu cstate_pkg_pmu = { + .attr_groups = pkg_attr_groups, + .name = "cstate_pkg", + .task_ctx_nr = perf_invalid_context, + .event_init = cstate_pmu_event_init, + .add = cstate_pmu_event_add, /* must have */ + .del = cstate_pmu_event_del, /* must have */ + .start = cstate_pmu_event_start, + .stop = cstate_pmu_event_stop, + .read = cstate_pmu_event_update, + .capabilities = PERF_PMU_CAP_NO_INTERRUPT, +}; + +static void __init cstate_pmus_register(void) +{ + int err; + + if (has_cstate_core) { + err = perf_pmu_register(&cstate_core_pmu, cstate_core_pmu.name, -1); + if (WARN_ON(err)) + pr_info("Failed to register PMU %s error %d\n", + cstate_core_pmu.name, err); + } + + if (has_cstate_pkg) { + err = perf_pmu_register(&cstate_pkg_pmu, cstate_pkg_pmu.name, -1); + if (WARN_ON(err)) + pr_info("Failed to register PMU %s error %d\n", + cstate_pkg_pmu.name, err); + } +} + +static int __init cstate_pmu_init(void) +{ + int err; + + if (cpu_has_hypervisor) + return -ENODEV; + + err = cstate_init(); + if (err) + return err; + + cstate_cpumask_init(); + + cstate_pmus_register(); + + return 0; +} + +device_initcall(cstate_pmu_init); diff --git a/kernel/arch/x86/kernel/cpu/perf_event_intel_ds.c b/kernel/arch/x86/kernel/cpu/perf_event_intel_ds.c index 7f73b3553..5db1c7755 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event_intel_ds.c +++ b/kernel/arch/x86/kernel/cpu/perf_event_intel_ds.c @@ -11,7 +11,7 @@ #define BTS_RECORD_SIZE 24 #define BTS_BUFFER_SIZE (PAGE_SIZE << 4) -#define PEBS_BUFFER_SIZE PAGE_SIZE +#define PEBS_BUFFER_SIZE (PAGE_SIZE << 4) #define PEBS_FIXUP_SIZE PAGE_SIZE /* @@ -224,6 +224,19 @@ union hsw_tsx_tuning { #define PEBS_HSW_TSX_FLAGS 0xff00000000ULL +/* Same as HSW, plus TSC */ + +struct pebs_record_skl { + u64 flags, ip; + u64 ax, bx, cx, dx; + u64 si, di, bp, sp; + u64 r8, r9, r10, r11; + u64 r12, r13, r14, r15; + u64 status, dla, dse, lat; + u64 real_ip, tsx_tuning; + u64 tsc; +}; + void init_debug_store_on_cpu(int cpu) { struct debug_store *ds = per_cpu(cpu_hw_events, cpu).ds; @@ -250,7 +263,7 @@ static int alloc_pebs_buffer(int cpu) { struct debug_store *ds = per_cpu(cpu_hw_events, cpu).ds; int node = cpu_to_node(cpu); - int max, thresh = 1; /* always use a single PEBS record */ + int max; void *buffer, *ibuffer; if (!x86_pmu.pebs) @@ -280,9 +293,6 @@ static int alloc_pebs_buffer(int cpu) ds->pebs_absolute_maximum = ds->pebs_buffer_base + max * x86_pmu.pebs_record_size; - ds->pebs_interrupt_threshold = ds->pebs_buffer_base + - thresh * x86_pmu.pebs_record_size; - return 0; } @@ -500,10 +510,11 @@ int intel_pmu_drain_bts_buffer(void) u64 flags; }; struct perf_event *event = cpuc->events[INTEL_PMC_IDX_FIXED_BTS]; - struct bts_record *at, *top; + struct bts_record *at, *base, *top; struct perf_output_handle handle; struct perf_event_header header; struct perf_sample_data data; + unsigned long skip = 0; struct pt_regs regs; if (!event) @@ -512,10 +523,10 @@ int intel_pmu_drain_bts_buffer(void) if (!x86_pmu.bts_active) return 0; - at = (struct bts_record *)(unsigned long)ds->bts_buffer_base; - top = (struct bts_record *)(unsigned long)ds->bts_index; + base = (struct bts_record *)(unsigned long)ds->bts_buffer_base; + top = (struct bts_record *)(unsigned long)ds->bts_index; - if (top <= at) + if (top <= base) return 0; memset(®s, 0, sizeof(regs)); @@ -525,16 +536,43 @@ int intel_pmu_drain_bts_buffer(void) perf_sample_data_init(&data, 0, event->hw.last_period); /* + * BTS leaks kernel addresses in branches across the cpl boundary, + * such as traps or system calls, so unless the user is asking for + * kernel tracing (and right now it's not possible), we'd need to + * filter them out. But first we need to count how many of those we + * have in the current batch. This is an extra O(n) pass, however, + * it's much faster than the other one especially considering that + * n <= 2560 (BTS_BUFFER_SIZE / BTS_RECORD_SIZE * 15/16; see the + * alloc_bts_buffer()). + */ + for (at = base; at < top; at++) { + /* + * Note that right now *this* BTS code only works if + * attr::exclude_kernel is set, but let's keep this extra + * check here in case that changes. + */ + if (event->attr.exclude_kernel && + (kernel_ip(at->from) || kernel_ip(at->to))) + skip++; + } + + /* * Prepare a generic sample, i.e. fill in the invariant fields. * We will overwrite the from and to address before we output * the sample. */ perf_prepare_sample(&header, &data, event, ®s); - if (perf_output_begin(&handle, event, header.size * (top - at))) + if (perf_output_begin(&handle, event, header.size * + (top - base - skip))) return 1; - for (; at < top; at++) { + for (at = base; at < top; at++) { + /* Filter out any records that contain kernel addresses. */ + if (event->attr.exclude_kernel && + (kernel_ip(at->from) || kernel_ip(at->to))) + continue; + data.ip = at->from; data.addr = at->to; @@ -549,6 +587,19 @@ int intel_pmu_drain_bts_buffer(void) return 1; } +static inline void intel_pmu_drain_pebs_buffer(void) +{ + struct pt_regs regs; + + x86_pmu.drain_pebs(®s); +} + +void intel_pmu_pebs_sched_task(struct perf_event_context *ctx, bool sched_in) +{ + if (!sched_in) + intel_pmu_drain_pebs_buffer(); +} + /* * PEBS */ @@ -665,6 +716,28 @@ struct event_constraint intel_hsw_pebs_event_constraints[] = { EVENT_CONSTRAINT_END }; +struct event_constraint intel_skl_pebs_event_constraints[] = { + INTEL_FLAGS_UEVENT_CONSTRAINT(0x1c0, 0x2), /* INST_RETIRED.PREC_DIST */ + INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_NA(0x01c2, 0xf), /* UOPS_RETIRED.ALL */ + /* UOPS_RETIRED.ALL, inv=1, cmask=16 (cycles:p). */ + INTEL_FLAGS_EVENT_CONSTRAINT(0x108001c2, 0xf), + INTEL_PLD_CONSTRAINT(0x1cd, 0xf), /* MEM_TRANS_RETIRED.* */ + INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x11d0, 0xf), /* MEM_INST_RETIRED.STLB_MISS_LOADS */ + INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_ST(0x12d0, 0xf), /* MEM_INST_RETIRED.STLB_MISS_STORES */ + INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x21d0, 0xf), /* MEM_INST_RETIRED.LOCK_LOADS */ + INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_ST(0x22d0, 0xf), /* MEM_INST_RETIRED.LOCK_STORES */ + INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x41d0, 0xf), /* MEM_INST_RETIRED.SPLIT_LOADS */ + INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_ST(0x42d0, 0xf), /* MEM_INST_RETIRED.SPLIT_STORES */ + INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x81d0, 0xf), /* MEM_INST_RETIRED.ALL_LOADS */ + INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_ST(0x82d0, 0xf), /* MEM_INST_RETIRED.ALL_STORES */ + INTEL_FLAGS_EVENT_CONSTRAINT_DATALA_LD(0xd1, 0xf), /* MEM_LOAD_RETIRED.* */ + INTEL_FLAGS_EVENT_CONSTRAINT_DATALA_LD(0xd2, 0xf), /* MEM_LOAD_L3_HIT_RETIRED.* */ + INTEL_FLAGS_EVENT_CONSTRAINT_DATALA_LD(0xd3, 0xf), /* MEM_LOAD_L3_MISS_RETIRED.* */ + /* Allow all events as PEBS with no flags */ + INTEL_ALL_EVENT_CONSTRAINT(0, 0xf), + EVENT_CONSTRAINT_END +}; + struct event_constraint *intel_pebs_constraints(struct perf_event *event) { struct event_constraint *c; @@ -684,25 +757,71 @@ struct event_constraint *intel_pebs_constraints(struct perf_event *event) return &emptyconstraint; } +static inline bool pebs_is_enabled(struct cpu_hw_events *cpuc) +{ + return (cpuc->pebs_enabled & ((1ULL << MAX_PEBS_EVENTS) - 1)); +} + void intel_pmu_pebs_enable(struct perf_event *event) { struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); struct hw_perf_event *hwc = &event->hw; + struct debug_store *ds = cpuc->ds; + bool first_pebs; + u64 threshold; hwc->config &= ~ARCH_PERFMON_EVENTSEL_INT; + first_pebs = !pebs_is_enabled(cpuc); cpuc->pebs_enabled |= 1ULL << hwc->idx; if (event->hw.flags & PERF_X86_EVENT_PEBS_LDLAT) cpuc->pebs_enabled |= 1ULL << (hwc->idx + 32); else if (event->hw.flags & PERF_X86_EVENT_PEBS_ST) cpuc->pebs_enabled |= 1ULL << 63; + + /* + * When the event is constrained enough we can use a larger + * threshold and run the event with less frequent PMI. + */ + if (hwc->flags & PERF_X86_EVENT_FREERUNNING) { + threshold = ds->pebs_absolute_maximum - + x86_pmu.max_pebs_events * x86_pmu.pebs_record_size; + + if (first_pebs) + perf_sched_cb_inc(event->ctx->pmu); + } else { + threshold = ds->pebs_buffer_base + x86_pmu.pebs_record_size; + + /* + * If not all events can use larger buffer, + * roll back to threshold = 1 + */ + if (!first_pebs && + (ds->pebs_interrupt_threshold > threshold)) + perf_sched_cb_dec(event->ctx->pmu); + } + + /* Use auto-reload if possible to save a MSR write in the PMI */ + if (hwc->flags & PERF_X86_EVENT_AUTO_RELOAD) { + ds->pebs_event_reset[hwc->idx] = + (u64)(-hwc->sample_period) & x86_pmu.cntval_mask; + } + + if (first_pebs || ds->pebs_interrupt_threshold > threshold) + ds->pebs_interrupt_threshold = threshold; } void intel_pmu_pebs_disable(struct perf_event *event) { struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); struct hw_perf_event *hwc = &event->hw; + struct debug_store *ds = cpuc->ds; + bool large_pebs = ds->pebs_interrupt_threshold > + ds->pebs_buffer_base + x86_pmu.pebs_record_size; + + if (large_pebs) + intel_pmu_drain_pebs_buffer(); cpuc->pebs_enabled &= ~(1ULL << hwc->idx); @@ -711,6 +830,9 @@ void intel_pmu_pebs_disable(struct perf_event *event) else if (event->hw.flags & PERF_X86_EVENT_PEBS_ST) cpuc->pebs_enabled &= ~(1ULL << 63); + if (large_pebs && !pebs_is_enabled(cpuc)) + perf_sched_cb_dec(event->ctx->pmu); + if (cpuc->enabled) wrmsrl(MSR_IA32_PEBS_ENABLE, cpuc->pebs_enabled); @@ -827,7 +949,7 @@ static int intel_pmu_pebs_fixup_ip(struct pt_regs *regs) return 0; } -static inline u64 intel_hsw_weight(struct pebs_record_hsw *pebs) +static inline u64 intel_hsw_weight(struct pebs_record_skl *pebs) { if (pebs->tsx_tuning) { union hsw_tsx_tuning tsx = { .value = pebs->tsx_tuning }; @@ -836,7 +958,7 @@ static inline u64 intel_hsw_weight(struct pebs_record_hsw *pebs) return 0; } -static inline u64 intel_hsw_transaction(struct pebs_record_hsw *pebs) +static inline u64 intel_hsw_transaction(struct pebs_record_skl *pebs) { u64 txn = (pebs->tsx_tuning & PEBS_HSW_TSX_FLAGS) >> 32; @@ -846,8 +968,10 @@ static inline u64 intel_hsw_transaction(struct pebs_record_hsw *pebs) return txn; } -static void __intel_pmu_pebs_event(struct perf_event *event, - struct pt_regs *iregs, void *__pebs) +static void setup_pebs_sample_data(struct perf_event *event, + struct pt_regs *iregs, void *__pebs, + struct perf_sample_data *data, + struct pt_regs *regs) { #define PERF_X86_EVENT_PEBS_HSW_PREC \ (PERF_X86_EVENT_PEBS_ST_HSW | \ @@ -858,14 +982,12 @@ static void __intel_pmu_pebs_event(struct perf_event *event, * unconditionally access the 'extra' entries. */ struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); - struct pebs_record_hsw *pebs = __pebs; - struct perf_sample_data data; - struct pt_regs regs; + struct pebs_record_skl *pebs = __pebs; u64 sample_type; int fll, fst, dsrc; int fl = event->hw.flags; - if (!intel_pmu_save_and_restart(event)) + if (pebs == NULL) return; sample_type = event->attr.sample_type; @@ -874,15 +996,15 @@ static void __intel_pmu_pebs_event(struct perf_event *event, fll = fl & PERF_X86_EVENT_PEBS_LDLAT; fst = fl & (PERF_X86_EVENT_PEBS_ST | PERF_X86_EVENT_PEBS_HSW_PREC); - perf_sample_data_init(&data, 0, event->hw.last_period); + perf_sample_data_init(data, 0, event->hw.last_period); - data.period = event->hw.last_period; + data->period = event->hw.last_period; /* * Use latency for weight (only avail with PEBS-LL) */ if (fll && (sample_type & PERF_SAMPLE_WEIGHT)) - data.weight = pebs->lat; + data->weight = pebs->lat; /* * data.data_src encodes the data source @@ -895,7 +1017,7 @@ static void __intel_pmu_pebs_event(struct perf_event *event, val = precise_datala_hsw(event, pebs->dse); else if (fst) val = precise_store_data(pebs->dse); - data.data_src.val = val; + data->data_src.val = val; } /* @@ -908,61 +1030,133 @@ static void __intel_pmu_pebs_event(struct perf_event *event, * PERF_SAMPLE_IP and PERF_SAMPLE_CALLCHAIN to function properly. * A possible PERF_SAMPLE_REGS will have to transfer all regs. */ - regs = *iregs; - regs.flags = pebs->flags; - set_linear_ip(®s, pebs->ip); - regs.bp = pebs->bp; - regs.sp = pebs->sp; + *regs = *iregs; + regs->flags = pebs->flags; + set_linear_ip(regs, pebs->ip); + regs->bp = pebs->bp; + regs->sp = pebs->sp; if (sample_type & PERF_SAMPLE_REGS_INTR) { - regs.ax = pebs->ax; - regs.bx = pebs->bx; - regs.cx = pebs->cx; - regs.dx = pebs->dx; - regs.si = pebs->si; - regs.di = pebs->di; - regs.bp = pebs->bp; - regs.sp = pebs->sp; - - regs.flags = pebs->flags; + regs->ax = pebs->ax; + regs->bx = pebs->bx; + regs->cx = pebs->cx; + regs->dx = pebs->dx; + regs->si = pebs->si; + regs->di = pebs->di; + regs->bp = pebs->bp; + regs->sp = pebs->sp; + + regs->flags = pebs->flags; #ifndef CONFIG_X86_32 - regs.r8 = pebs->r8; - regs.r9 = pebs->r9; - regs.r10 = pebs->r10; - regs.r11 = pebs->r11; - regs.r12 = pebs->r12; - regs.r13 = pebs->r13; - regs.r14 = pebs->r14; - regs.r15 = pebs->r15; + regs->r8 = pebs->r8; + regs->r9 = pebs->r9; + regs->r10 = pebs->r10; + regs->r11 = pebs->r11; + regs->r12 = pebs->r12; + regs->r13 = pebs->r13; + regs->r14 = pebs->r14; + regs->r15 = pebs->r15; #endif } if (event->attr.precise_ip > 1 && x86_pmu.intel_cap.pebs_format >= 2) { - regs.ip = pebs->real_ip; - regs.flags |= PERF_EFLAGS_EXACT; - } else if (event->attr.precise_ip > 1 && intel_pmu_pebs_fixup_ip(®s)) - regs.flags |= PERF_EFLAGS_EXACT; + regs->ip = pebs->real_ip; + regs->flags |= PERF_EFLAGS_EXACT; + } else if (event->attr.precise_ip > 1 && intel_pmu_pebs_fixup_ip(regs)) + regs->flags |= PERF_EFLAGS_EXACT; else - regs.flags &= ~PERF_EFLAGS_EXACT; + regs->flags &= ~PERF_EFLAGS_EXACT; if ((sample_type & PERF_SAMPLE_ADDR) && x86_pmu.intel_cap.pebs_format >= 1) - data.addr = pebs->dla; + data->addr = pebs->dla; if (x86_pmu.intel_cap.pebs_format >= 2) { /* Only set the TSX weight when no memory weight. */ if ((sample_type & PERF_SAMPLE_WEIGHT) && !fll) - data.weight = intel_hsw_weight(pebs); + data->weight = intel_hsw_weight(pebs); if (sample_type & PERF_SAMPLE_TRANSACTION) - data.txn = intel_hsw_transaction(pebs); + data->txn = intel_hsw_transaction(pebs); } + /* + * v3 supplies an accurate time stamp, so we use that + * for the time stamp. + * + * We can only do this for the default trace clock. + */ + if (x86_pmu.intel_cap.pebs_format >= 3 && + event->attr.use_clockid == 0) + data->time = native_sched_clock_from_tsc(pebs->tsc); + if (has_branch_stack(event)) - data.br_stack = &cpuc->lbr_stack; + data->br_stack = &cpuc->lbr_stack; +} + +static inline void * +get_next_pebs_record_by_bit(void *base, void *top, int bit) +{ + struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); + void *at; + u64 pebs_status; + + if (base == NULL) + return NULL; + + for (at = base; at < top; at += x86_pmu.pebs_record_size) { + struct pebs_record_nhm *p = at; + + if (test_bit(bit, (unsigned long *)&p->status)) { + /* PEBS v3 has accurate status bits */ + if (x86_pmu.intel_cap.pebs_format >= 3) + return at; + + if (p->status == (1 << bit)) + return at; + + /* clear non-PEBS bit and re-check */ + pebs_status = p->status & cpuc->pebs_enabled; + pebs_status &= (1ULL << MAX_PEBS_EVENTS) - 1; + if (pebs_status == (1 << bit)) + return at; + } + } + return NULL; +} + +static void __intel_pmu_pebs_event(struct perf_event *event, + struct pt_regs *iregs, + void *base, void *top, + int bit, int count) +{ + struct perf_sample_data data; + struct pt_regs regs; + void *at = get_next_pebs_record_by_bit(base, top, bit); + + if (!intel_pmu_save_and_restart(event) && + !(event->hw.flags & PERF_X86_EVENT_AUTO_RELOAD)) + return; + + while (count > 1) { + setup_pebs_sample_data(event, iregs, at, &data, ®s); + perf_event_output(event, &data, ®s); + at += x86_pmu.pebs_record_size; + at = get_next_pebs_record_by_bit(at, top, bit); + count--; + } + + setup_pebs_sample_data(event, iregs, at, &data, ®s); - if (perf_event_overflow(event, &data, ®s)) + /* + * All but the last records are processed. + * The last one is left to be able to call the overflow handler. + */ + if (perf_event_overflow(event, &data, ®s)) { x86_pmu_stop(event, 0); + return; + } + } static void intel_pmu_drain_pebs_core(struct pt_regs *iregs) @@ -992,72 +1186,99 @@ static void intel_pmu_drain_pebs_core(struct pt_regs *iregs) if (!event->attr.precise_ip) return; - n = top - at; + n = (top - at) / x86_pmu.pebs_record_size; if (n <= 0) return; - /* - * Should not happen, we program the threshold at 1 and do not - * set a reset value. - */ - WARN_ONCE(n > 1, "bad leftover pebs %d\n", n); - at += n - 1; - - __intel_pmu_pebs_event(event, iregs, at); + __intel_pmu_pebs_event(event, iregs, at, top, 0, n); } static void intel_pmu_drain_pebs_nhm(struct pt_regs *iregs) { struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); struct debug_store *ds = cpuc->ds; - struct perf_event *event = NULL; - void *at, *top; - u64 status = 0; - int bit; + struct perf_event *event; + void *base, *at, *top; + short counts[MAX_PEBS_EVENTS] = {}; + short error[MAX_PEBS_EVENTS] = {}; + int bit, i; if (!x86_pmu.pebs_active) return; - at = (struct pebs_record_nhm *)(unsigned long)ds->pebs_buffer_base; + base = (struct pebs_record_nhm *)(unsigned long)ds->pebs_buffer_base; top = (struct pebs_record_nhm *)(unsigned long)ds->pebs_index; ds->pebs_index = ds->pebs_buffer_base; - if (unlikely(at > top)) + if (unlikely(base >= top)) return; - /* - * Should not happen, we program the threshold at 1 and do not - * set a reset value. - */ - WARN_ONCE(top - at > x86_pmu.max_pebs_events * x86_pmu.pebs_record_size, - "Unexpected number of pebs records %ld\n", - (long)(top - at) / x86_pmu.pebs_record_size); - - for (; at < top; at += x86_pmu.pebs_record_size) { + for (at = base; at < top; at += x86_pmu.pebs_record_size) { struct pebs_record_nhm *p = at; + u64 pebs_status; - for_each_set_bit(bit, (unsigned long *)&p->status, - x86_pmu.max_pebs_events) { - event = cpuc->events[bit]; - if (!test_bit(bit, cpuc->active_mask)) - continue; + /* PEBS v3 has accurate status bits */ + if (x86_pmu.intel_cap.pebs_format >= 3) { + for_each_set_bit(bit, (unsigned long *)&p->status, + MAX_PEBS_EVENTS) + counts[bit]++; - WARN_ON_ONCE(!event); + continue; + } - if (!event->attr.precise_ip) - continue; + pebs_status = p->status & cpuc->pebs_enabled; + pebs_status &= (1ULL << x86_pmu.max_pebs_events) - 1; - if (__test_and_set_bit(bit, (unsigned long *)&status)) - continue; + bit = find_first_bit((unsigned long *)&pebs_status, + x86_pmu.max_pebs_events); + if (WARN(bit >= x86_pmu.max_pebs_events, + "PEBS record without PEBS event! status=%Lx pebs_enabled=%Lx active_mask=%Lx", + (unsigned long long)p->status, (unsigned long long)cpuc->pebs_enabled, + *(unsigned long long *)cpuc->active_mask)) + continue; - break; + /* + * The PEBS hardware does not deal well with the situation + * when events happen near to each other and multiple bits + * are set. But it should happen rarely. + * + * If these events include one PEBS and multiple non-PEBS + * events, it doesn't impact PEBS record. The record will + * be handled normally. (slow path) + * + * If these events include two or more PEBS events, the + * records for the events can be collapsed into a single + * one, and it's not possible to reconstruct all events + * that caused the PEBS record. It's called collision. + * If collision happened, the record will be dropped. + */ + if (p->status != (1ULL << bit)) { + for_each_set_bit(i, (unsigned long *)&pebs_status, + x86_pmu.max_pebs_events) + error[i]++; + continue; } - if (!event || bit >= x86_pmu.max_pebs_events) + counts[bit]++; + } + + for (bit = 0; bit < x86_pmu.max_pebs_events; bit++) { + if ((counts[bit] == 0) && (error[bit] == 0)) continue; - __intel_pmu_pebs_event(event, iregs, at); + event = cpuc->events[bit]; + WARN_ON_ONCE(!event); + WARN_ON_ONCE(!event->attr.precise_ip); + + /* log dropped samples number */ + if (error[bit]) + perf_log_lost_samples(event, error[bit]); + + if (counts[bit]) { + __intel_pmu_pebs_event(event, iregs, base, + top, bit, counts[bit]); + } } } @@ -1098,6 +1319,14 @@ void __init intel_ds_init(void) x86_pmu.drain_pebs = intel_pmu_drain_pebs_nhm; break; + case 3: + pr_cont("PEBS fmt3%c, ", pebs_type); + x86_pmu.pebs_record_size = + sizeof(struct pebs_record_skl); + x86_pmu.drain_pebs = intel_pmu_drain_pebs_nhm; + x86_pmu.free_running_flags |= PERF_SAMPLE_TIME; + break; + default: printk(KERN_CONT "no PEBS fmt%d%c, ", format, pebs_type); x86_pmu.pebs = 0; diff --git a/kernel/arch/x86/kernel/cpu/perf_event_intel_lbr.c b/kernel/arch/x86/kernel/cpu/perf_event_intel_lbr.c index 94e5b506c..659f01e16 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event_intel_lbr.c +++ b/kernel/arch/x86/kernel/cpu/perf_event_intel_lbr.c @@ -13,7 +13,8 @@ enum { LBR_FORMAT_EIP = 0x02, LBR_FORMAT_EIP_FLAGS = 0x03, LBR_FORMAT_EIP_FLAGS2 = 0x04, - LBR_FORMAT_MAX_KNOWN = LBR_FORMAT_EIP_FLAGS2, + LBR_FORMAT_INFO = 0x05, + LBR_FORMAT_MAX_KNOWN = LBR_FORMAT_INFO, }; static enum { @@ -96,6 +97,7 @@ enum { X86_BR_NO_TX = 1 << 14,/* not in transaction */ X86_BR_ZERO_CALL = 1 << 15,/* zero length call */ X86_BR_CALL_STACK = 1 << 16,/* call stack */ + X86_BR_IND_JMP = 1 << 17,/* indirect jump */ }; #define X86_BR_PLM (X86_BR_USER | X86_BR_KERNEL) @@ -113,6 +115,7 @@ enum { X86_BR_IRQ |\ X86_BR_ABORT |\ X86_BR_IND_CALL |\ + X86_BR_IND_JMP |\ X86_BR_ZERO_CALL) #define X86_BR_ALL (X86_BR_PLM | X86_BR_ANY) @@ -138,13 +141,20 @@ static void __intel_pmu_lbr_enable(bool pmi) u64 debugctl, lbr_select = 0, orig_debugctl; /* + * No need to unfreeze manually, as v4 can do that as part + * of the GLOBAL_STATUS ack. + */ + if (pmi && x86_pmu.version >= 4) + return; + + /* * No need to reprogram LBR_SELECT in a PMI, as it * did not change. */ - if (cpuc->lbr_sel && !pmi) { + if (cpuc->lbr_sel) lbr_select = cpuc->lbr_sel->config; + if (!pmi) wrmsrl(MSR_LBR_SELECT, lbr_select); - } rdmsrl(MSR_IA32_DEBUGCTLMSR, debugctl); orig_debugctl = debugctl; @@ -184,6 +194,8 @@ static void intel_pmu_lbr_reset_64(void) for (i = 0; i < x86_pmu.lbr_nr; i++) { wrmsrl(x86_pmu.lbr_from + i, 0); wrmsrl(x86_pmu.lbr_to + i, 0); + if (x86_pmu.intel_cap.lbr_format == LBR_FORMAT_INFO) + wrmsrl(MSR_LBR_INFO_0 + i, 0); } } @@ -227,12 +239,15 @@ static void __intel_pmu_lbr_restore(struct x86_perf_task_context *task_ctx) } mask = x86_pmu.lbr_nr - 1; - tos = intel_pmu_lbr_tos(); - for (i = 0; i < x86_pmu.lbr_nr; i++) { + tos = task_ctx->tos; + for (i = 0; i < tos; i++) { lbr_idx = (tos - i) & mask; wrmsrl(x86_pmu.lbr_from + lbr_idx, task_ctx->lbr_from[i]); wrmsrl(x86_pmu.lbr_to + lbr_idx, task_ctx->lbr_to[i]); + if (x86_pmu.intel_cap.lbr_format == LBR_FORMAT_INFO) + wrmsrl(MSR_LBR_INFO_0 + lbr_idx, task_ctx->lbr_info[i]); } + wrmsrl(x86_pmu.lbr_tos, tos); task_ctx->lbr_stack_state = LBR_NONE; } @@ -249,11 +264,14 @@ static void __intel_pmu_lbr_save(struct x86_perf_task_context *task_ctx) mask = x86_pmu.lbr_nr - 1; tos = intel_pmu_lbr_tos(); - for (i = 0; i < x86_pmu.lbr_nr; i++) { + for (i = 0; i < tos; i++) { lbr_idx = (tos - i) & mask; rdmsrl(x86_pmu.lbr_from + lbr_idx, task_ctx->lbr_from[i]); rdmsrl(x86_pmu.lbr_to + lbr_idx, task_ctx->lbr_to[i]); + if (x86_pmu.intel_cap.lbr_format == LBR_FORMAT_INFO) + rdmsrl(MSR_LBR_INFO_0 + lbr_idx, task_ctx->lbr_info[i]); } + task_ctx->tos = tos; task_ctx->lbr_stack_state = LBR_VALID; } @@ -262,9 +280,6 @@ void intel_pmu_lbr_sched_task(struct perf_event_context *ctx, bool sched_in) struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); struct x86_perf_task_context *task_ctx; - if (!x86_pmu.lbr_nr) - return; - /* * If LBR callstack feature is enabled and the stack was saved when * the task was scheduled out, restore the stack. Otherwise flush @@ -412,16 +427,31 @@ static void intel_pmu_lbr_read_64(struct cpu_hw_events *cpuc) u64 tos = intel_pmu_lbr_tos(); int i; int out = 0; + int num = x86_pmu.lbr_nr; - for (i = 0; i < x86_pmu.lbr_nr; i++) { + if (cpuc->lbr_sel->config & LBR_CALL_STACK) + num = tos; + + for (i = 0; i < num; i++) { unsigned long lbr_idx = (tos - i) & mask; u64 from, to, mis = 0, pred = 0, in_tx = 0, abort = 0; int skip = 0; + u16 cycles = 0; int lbr_flags = lbr_desc[lbr_format]; rdmsrl(x86_pmu.lbr_from + lbr_idx, from); rdmsrl(x86_pmu.lbr_to + lbr_idx, to); + if (lbr_format == LBR_FORMAT_INFO) { + u64 info; + + rdmsrl(MSR_LBR_INFO_0 + lbr_idx, info); + mis = !!(info & LBR_INFO_MISPRED); + pred = !mis; + in_tx = !!(info & LBR_INFO_IN_TX); + abort = !!(info & LBR_INFO_ABORT); + cycles = (info & LBR_INFO_CYCLES); + } if (lbr_flags & LBR_EIP_FLAGS) { mis = !!(from & LBR_FROM_FLAG_MISPRED); pred = !mis; @@ -451,6 +481,7 @@ static void intel_pmu_lbr_read_64(struct cpu_hw_events *cpuc) cpuc->lbr_entries[out].predicted = pred; cpuc->lbr_entries[out].in_tx = in_tx; cpuc->lbr_entries[out].abort = abort; + cpuc->lbr_entries[out].cycles = cycles; cpuc->lbr_entries[out].reserved = 0; out++; } @@ -523,6 +554,11 @@ static int intel_pmu_setup_sw_lbr_filter(struct perf_event *event) X86_BR_CALL_STACK; } + if (br_type & PERF_SAMPLE_BRANCH_IND_JUMP) + mask |= X86_BR_IND_JMP; + + if (br_type & PERF_SAMPLE_BRANCH_CALL) + mask |= X86_BR_CALL | X86_BR_ZERO_CALL; /* * stash actual user request into reg, it may * be used by fixup code for some CPU @@ -736,7 +772,7 @@ static int branch_type(unsigned long from, unsigned long to, int abort) break; case 4: case 5: - ret = X86_BR_JMP; + ret = X86_BR_IND_JMP; break; } break; @@ -844,6 +880,7 @@ static const int nhm_lbr_sel_map[PERF_SAMPLE_BRANCH_MAX_SHIFT] = { */ [PERF_SAMPLE_BRANCH_IND_CALL_SHIFT] = LBR_IND_CALL | LBR_IND_JMP, [PERF_SAMPLE_BRANCH_COND_SHIFT] = LBR_JCC, + [PERF_SAMPLE_BRANCH_IND_JUMP_SHIFT] = LBR_IND_JMP, }; static const int snb_lbr_sel_map[PERF_SAMPLE_BRANCH_MAX_SHIFT] = { @@ -856,6 +893,8 @@ static const int snb_lbr_sel_map[PERF_SAMPLE_BRANCH_MAX_SHIFT] = { | LBR_FAR, [PERF_SAMPLE_BRANCH_IND_CALL_SHIFT] = LBR_IND_CALL, [PERF_SAMPLE_BRANCH_COND_SHIFT] = LBR_JCC, + [PERF_SAMPLE_BRANCH_IND_JUMP_SHIFT] = LBR_IND_JMP, + [PERF_SAMPLE_BRANCH_CALL_SHIFT] = LBR_REL_CALL, }; static const int hsw_lbr_sel_map[PERF_SAMPLE_BRANCH_MAX_SHIFT] = { @@ -870,6 +909,8 @@ static const int hsw_lbr_sel_map[PERF_SAMPLE_BRANCH_MAX_SHIFT] = { [PERF_SAMPLE_BRANCH_COND_SHIFT] = LBR_JCC, [PERF_SAMPLE_BRANCH_CALL_STACK_SHIFT] = LBR_REL_CALL | LBR_IND_CALL | LBR_RETURN | LBR_CALL_STACK, + [PERF_SAMPLE_BRANCH_IND_JUMP_SHIFT] = LBR_IND_JMP, + [PERF_SAMPLE_BRANCH_CALL_SHIFT] = LBR_REL_CALL, }; /* core */ @@ -942,6 +983,26 @@ void intel_pmu_lbr_init_hsw(void) pr_cont("16-deep LBR, "); } +/* skylake */ +__init void intel_pmu_lbr_init_skl(void) +{ + x86_pmu.lbr_nr = 32; + x86_pmu.lbr_tos = MSR_LBR_TOS; + x86_pmu.lbr_from = MSR_LBR_NHM_FROM; + x86_pmu.lbr_to = MSR_LBR_NHM_TO; + + x86_pmu.lbr_sel_mask = LBR_SEL_MASK; + x86_pmu.lbr_sel_map = hsw_lbr_sel_map; + + /* + * SW branch filter usage: + * - support syscall, sysret capture. + * That requires LBR_FAR but that means far + * jmp need to be filtered out + */ + pr_cont("32-deep LBR, "); +} + /* atom */ void __init intel_pmu_lbr_init_atom(void) { diff --git a/kernel/arch/x86/kernel/cpu/perf_event_intel_pt.c b/kernel/arch/x86/kernel/cpu/perf_event_intel_pt.c index 123ff1bb2..868e11943 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event_intel_pt.c +++ b/kernel/arch/x86/kernel/cpu/perf_event_intel_pt.c @@ -65,15 +65,21 @@ static struct pt_cap_desc { } pt_caps[] = { PT_CAP(max_subleaf, 0, CR_EAX, 0xffffffff), PT_CAP(cr3_filtering, 0, CR_EBX, BIT(0)), + PT_CAP(psb_cyc, 0, CR_EBX, BIT(1)), + PT_CAP(mtc, 0, CR_EBX, BIT(3)), PT_CAP(topa_output, 0, CR_ECX, BIT(0)), PT_CAP(topa_multiple_entries, 0, CR_ECX, BIT(1)), + PT_CAP(single_range_output, 0, CR_ECX, BIT(2)), PT_CAP(payloads_lip, 0, CR_ECX, BIT(31)), + PT_CAP(mtc_periods, 1, CR_EAX, 0xffff0000), + PT_CAP(cycle_thresholds, 1, CR_EBX, 0xffff), + PT_CAP(psb_periods, 1, CR_EBX, 0xffff0000), }; static u32 pt_cap_get(enum pt_capabilities cap) { struct pt_cap_desc *cd = &pt_caps[cap]; - u32 c = pt_pmu.caps[cd->leaf * 4 + cd->reg]; + u32 c = pt_pmu.caps[cd->leaf * PT_CPUID_REGS_NUM + cd->reg]; unsigned int shift = __ffs(cd->mask); return (c & cd->mask) >> shift; @@ -94,12 +100,22 @@ static struct attribute_group pt_cap_group = { .name = "caps", }; +PMU_FORMAT_ATTR(cyc, "config:1" ); +PMU_FORMAT_ATTR(mtc, "config:9" ); PMU_FORMAT_ATTR(tsc, "config:10" ); PMU_FORMAT_ATTR(noretcomp, "config:11" ); +PMU_FORMAT_ATTR(mtc_period, "config:14-17" ); +PMU_FORMAT_ATTR(cyc_thresh, "config:19-22" ); +PMU_FORMAT_ATTR(psb_period, "config:24-27" ); static struct attribute *pt_formats_attr[] = { + &format_attr_cyc.attr, + &format_attr_mtc.attr, &format_attr_tsc.attr, &format_attr_noretcomp.attr, + &format_attr_mtc_period.attr, + &format_attr_cyc_thresh.attr, + &format_attr_psb_period.attr, NULL, }; @@ -123,16 +139,13 @@ static int __init pt_pmu_hw_init(void) long i; attrs = NULL; - ret = -ENODEV; - if (!test_cpu_cap(&boot_cpu_data, X86_FEATURE_INTEL_PT)) - goto fail; for (i = 0; i < PT_CPUID_LEAVES; i++) { cpuid_count(20, i, - &pt_pmu.caps[CR_EAX + i*4], - &pt_pmu.caps[CR_EBX + i*4], - &pt_pmu.caps[CR_ECX + i*4], - &pt_pmu.caps[CR_EDX + i*4]); + &pt_pmu.caps[CR_EAX + i*PT_CPUID_REGS_NUM], + &pt_pmu.caps[CR_EBX + i*PT_CPUID_REGS_NUM], + &pt_pmu.caps[CR_ECX + i*PT_CPUID_REGS_NUM], + &pt_pmu.caps[CR_EDX + i*PT_CPUID_REGS_NUM]); } ret = -ENOMEM; @@ -170,15 +183,65 @@ fail: return ret; } -#define PT_CONFIG_MASK (RTIT_CTL_TSC_EN | RTIT_CTL_DISRETC) +#define RTIT_CTL_CYC_PSB (RTIT_CTL_CYCLEACC | \ + RTIT_CTL_CYC_THRESH | \ + RTIT_CTL_PSB_FREQ) + +#define RTIT_CTL_MTC (RTIT_CTL_MTC_EN | \ + RTIT_CTL_MTC_RANGE) + +#define PT_CONFIG_MASK (RTIT_CTL_TSC_EN | \ + RTIT_CTL_DISRETC | \ + RTIT_CTL_CYC_PSB | \ + RTIT_CTL_MTC) static bool pt_event_valid(struct perf_event *event) { u64 config = event->attr.config; + u64 allowed, requested; if ((config & PT_CONFIG_MASK) != config) return false; + if (config & RTIT_CTL_CYC_PSB) { + if (!pt_cap_get(PT_CAP_psb_cyc)) + return false; + + allowed = pt_cap_get(PT_CAP_psb_periods); + requested = (config & RTIT_CTL_PSB_FREQ) >> + RTIT_CTL_PSB_FREQ_OFFSET; + if (requested && (!(allowed & BIT(requested)))) + return false; + + allowed = pt_cap_get(PT_CAP_cycle_thresholds); + requested = (config & RTIT_CTL_CYC_THRESH) >> + RTIT_CTL_CYC_THRESH_OFFSET; + if (requested && (!(allowed & BIT(requested)))) + return false; + } + + if (config & RTIT_CTL_MTC) { + /* + * In the unlikely case that CPUID lists valid mtc periods, + * but not the mtc capability, drop out here. + * + * Spec says that setting mtc period bits while mtc bit in + * CPUID is 0 will #GP, so better safe than sorry. + */ + if (!pt_cap_get(PT_CAP_mtc)) + return false; + + allowed = pt_cap_get(PT_CAP_mtc_periods); + if (!allowed) + return false; + + requested = (config & RTIT_CTL_MTC_RANGE) >> + RTIT_CTL_MTC_RANGE_OFFSET; + + if (!(allowed & BIT(requested))) + return false; + } + return true; } @@ -187,19 +250,15 @@ static bool pt_event_valid(struct perf_event *event) * These all are cpu affine and operate on a local PT */ -static bool pt_is_running(void) -{ - u64 ctl; - - rdmsrl(MSR_IA32_RTIT_CTL, ctl); - - return !!(ctl & RTIT_CTL_TRACEEN); -} - static void pt_config(struct perf_event *event) { u64 reg; + if (!event->hw.itrace_started) { + event->hw.itrace_started = 1; + wrmsrl(MSR_IA32_RTIT_STATUS, 0); + } + reg = RTIT_CTL_TOPA | RTIT_CTL_BRANCH_EN | RTIT_CTL_TRACEEN; if (!event->attr.exclude_kernel) @@ -609,7 +668,12 @@ static unsigned int pt_topa_next_entry(struct pt_buffer *buf, unsigned int pg) * @handle: Current output handle. * * Place INT and STOP marks to prevent overwriting old data that the consumer - * hasn't yet collected. + * hasn't yet collected and waking up the consumer after a certain fraction of + * the buffer has filled up. Only needed and sensible for non-snapshot counters. + * + * This obviously relies on buf::head to figure out buffer markers, so it has + * to be called after pt_buffer_reset_offsets() and before the hardware tracing + * is enabled. */ static int pt_buffer_reset_markers(struct pt_buffer *buf, struct perf_output_handle *handle) @@ -618,9 +682,6 @@ static int pt_buffer_reset_markers(struct pt_buffer *buf, unsigned long head = local64_read(&buf->head); unsigned long idx, npages, wakeup; - if (buf->snapshot) - return 0; - /* can't stop in the middle of an output region */ if (buf->output_off + handle->size + 1 < sizes(TOPA_ENTRY(buf->cur, buf->cur_idx)->size)) @@ -674,7 +735,7 @@ static void pt_buffer_setup_topa_index(struct pt_buffer *buf) struct topa *cur = buf->first, *prev = buf->last; struct topa_entry *te_cur = TOPA_ENTRY(cur, 0), *te_prev = TOPA_ENTRY(prev, prev->last - 1); - int pg = 0, idx = 0, ntopa = 0; + int pg = 0, idx = 0; while (pg < buf->nr_pages) { int tidx; @@ -689,9 +750,9 @@ static void pt_buffer_setup_topa_index(struct pt_buffer *buf) /* advance to next topa table */ idx = 0; cur = list_entry(cur->list.next, struct topa, list); - ntopa++; - } else + } else { idx++; + } te_cur = TOPA_ENTRY(cur, idx); } @@ -703,7 +764,14 @@ static void pt_buffer_setup_topa_index(struct pt_buffer *buf) * @head: Write pointer (aux_head) from AUX buffer. * * Find the ToPA table and entry corresponding to given @head and set buffer's - * "current" pointers accordingly. + * "current" pointers accordingly. This is done after we have obtained the + * current aux_head position from a successful call to perf_aux_output_begin() + * to make sure the hardware is writing to the right place. + * + * This function modifies buf::{cur,cur_idx,output_off} that will be programmed + * into PT msrs when the tracing is enabled and buf::head and buf::data_size, + * which are used to determine INT and STOP markers' locations by a subsequent + * call to pt_buffer_reset_markers(). */ static void pt_buffer_reset_offsets(struct pt_buffer *buf, unsigned long head) { @@ -901,6 +969,7 @@ void intel_pt_interrupt(void) } pt_buffer_reset_offsets(buf, pt->handle.head); + /* snapshot counters don't use PMI, so it's safe */ ret = pt_buffer_reset_markers(buf, &pt->handle); if (ret) { perf_aux_output_end(&pt->handle, 0, true); @@ -909,7 +978,6 @@ void intel_pt_interrupt(void) pt_config_buffer(buf->cur->table, buf->cur_idx, buf->output_off); - wrmsrl(MSR_IA32_RTIT_STATUS, 0); pt_config(event); } } @@ -923,7 +991,7 @@ static void pt_event_start(struct perf_event *event, int mode) struct pt *pt = this_cpu_ptr(&pt_ctx); struct pt_buffer *buf = perf_get_aux(&pt->handle); - if (pt_is_running() || !buf || pt_buffer_is_full(buf, pt)) { + if (!buf || pt_buffer_is_full(buf, pt)) { event->hw.state = PERF_HES_STOPPED; return; } @@ -933,7 +1001,6 @@ static void pt_event_start(struct perf_event *event, int mode) pt_config_buffer(buf->cur->table, buf->cur_idx, buf->output_off); - wrmsrl(MSR_IA32_RTIT_STATUS, 0); pt_config(event); } @@ -954,7 +1021,6 @@ static void pt_event_stop(struct perf_event *event, int mode) event->hw.state = PERF_HES_STOPPED; if (mode & PERF_EF_UPDATE) { - struct pt *pt = this_cpu_ptr(&pt_ctx); struct pt_buffer *buf = perf_get_aux(&pt->handle); if (!buf) @@ -1061,6 +1127,10 @@ static __init int pt_init(void) int ret, cpu, prior_warn = 0; BUILD_BUG_ON(sizeof(struct topa) > PAGE_SIZE); + + if (!test_cpu_cap(&boot_cpu_data, X86_FEATURE_INTEL_PT)) + return -ENODEV; + get_online_cpus(); for_each_online_cpu(cpu) { u64 ctl; @@ -1106,5 +1176,4 @@ static __init int pt_init(void) return ret; } - -module_init(pt_init); +arch_initcall(pt_init); diff --git a/kernel/arch/x86/kernel/cpu/perf_event_intel_rapl.c b/kernel/arch/x86/kernel/cpu/perf_event_intel_rapl.c index 358c54ad2..ed446bdcb 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event_intel_rapl.c +++ b/kernel/arch/x86/kernel/cpu/perf_event_intel_rapl.c @@ -86,6 +86,10 @@ static const char *rapl_domain_names[NR_RAPL_DOMAINS] __initconst = { 1<<RAPL_IDX_RAM_NRG_STAT|\ 1<<RAPL_IDX_PP1_NRG_STAT) +/* Knights Landing has PKG, RAM */ +#define RAPL_IDX_KNL (1<<RAPL_IDX_PKG_NRG_STAT|\ + 1<<RAPL_IDX_RAM_NRG_STAT) + /* * event code: LSB 8 bits, passed in attr->config * any other bit is reserved @@ -103,12 +107,6 @@ static ssize_t __rapl_##_var##_show(struct kobject *kobj, \ static struct kobj_attribute format_attr_##_var = \ __ATTR(_name, 0444, __rapl_##_var##_show, NULL) -#define RAPL_EVENT_DESC(_name, _config) \ -{ \ - .attr = __ATTR(_name, 0444, rapl_event_show, NULL), \ - .config = _config, \ -} - #define RAPL_CNTR_WIDTH 32 /* 32-bit rapl counters */ #define RAPL_EVENT_ATTR_STR(_name, v, str) \ @@ -204,9 +202,8 @@ again: static void rapl_start_hrtimer(struct rapl_pmu *pmu) { - __hrtimer_start_range_ns(&pmu->hrtimer, - pmu->timer_interval, 0, - HRTIMER_MODE_REL_PINNED, 0); + hrtimer_start(&pmu->hrtimer, pmu->timer_interval, + HRTIMER_MODE_REL_PINNED); } static void rapl_stop_hrtimer(struct rapl_pmu *pmu) @@ -487,6 +484,18 @@ static struct attribute *rapl_events_hsw_attr[] = { NULL, }; +static struct attribute *rapl_events_knl_attr[] = { + EVENT_PTR(rapl_pkg), + EVENT_PTR(rapl_ram), + + EVENT_PTR(rapl_pkg_unit), + EVENT_PTR(rapl_ram_unit), + + EVENT_PTR(rapl_pkg_scale), + EVENT_PTR(rapl_ram_scale), + NULL, +}; + static struct attribute_group rapl_pmu_events_group = { .name = "events", .attrs = NULL, /* patched at runtime */ @@ -731,6 +740,10 @@ static int __init rapl_pmu_init(void) rapl_cntr_mask = RAPL_IDX_SRV; rapl_pmu_events_group.attrs = rapl_events_srv_attr; break; + case 87: /* Knights Landing */ + rapl_add_quirk(rapl_hsw_server_quirk); + rapl_cntr_mask = RAPL_IDX_KNL; + rapl_pmu_events_group.attrs = rapl_events_knl_attr; default: /* unsupported */ diff --git a/kernel/arch/x86/kernel/cpu/perf_event_intel_uncore.c b/kernel/arch/x86/kernel/cpu/perf_event_intel_uncore.c index 90b7c501c..61215a69b 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event_intel_uncore.c +++ b/kernel/arch/x86/kernel/cpu/perf_event_intel_uncore.c @@ -7,7 +7,8 @@ struct intel_uncore_type **uncore_pci_uncores = empty_uncore; static bool pcidrv_registered; struct pci_driver *uncore_pci_driver; /* pci bus to socket mapping */ -int uncore_pcibus_to_physid[256] = { [0 ... 255] = -1, }; +DEFINE_RAW_SPINLOCK(pci2phy_map_lock); +struct list_head pci2phy_map_head = LIST_HEAD_INIT(pci2phy_map_head); struct pci_dev *uncore_extra_pci_dev[UNCORE_SOCKET_MAX][UNCORE_EXTRA_PCI_DEV_MAX]; static DEFINE_RAW_SPINLOCK(uncore_box_lock); @@ -20,6 +21,59 @@ static struct event_constraint uncore_constraint_fixed = struct event_constraint uncore_constraint_empty = EVENT_CONSTRAINT(0, 0, 0); +int uncore_pcibus_to_physid(struct pci_bus *bus) +{ + struct pci2phy_map *map; + int phys_id = -1; + + raw_spin_lock(&pci2phy_map_lock); + list_for_each_entry(map, &pci2phy_map_head, list) { + if (map->segment == pci_domain_nr(bus)) { + phys_id = map->pbus_to_physid[bus->number]; + break; + } + } + raw_spin_unlock(&pci2phy_map_lock); + + return phys_id; +} + +struct pci2phy_map *__find_pci2phy_map(int segment) +{ + struct pci2phy_map *map, *alloc = NULL; + int i; + + lockdep_assert_held(&pci2phy_map_lock); + +lookup: + list_for_each_entry(map, &pci2phy_map_head, list) { + if (map->segment == segment) + goto end; + } + + if (!alloc) { + raw_spin_unlock(&pci2phy_map_lock); + alloc = kmalloc(sizeof(struct pci2phy_map), GFP_KERNEL); + raw_spin_lock(&pci2phy_map_lock); + + if (!alloc) + return NULL; + + goto lookup; + } + + map = alloc; + alloc = NULL; + map->segment = segment; + for (i = 0; i < 256; i++) + map->pbus_to_physid[i] = -1; + list_add_tail(&map->list, &pci2phy_map_head); + +end: + kfree(alloc); + return map; +} + ssize_t uncore_event_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { @@ -233,9 +287,8 @@ static enum hrtimer_restart uncore_pmu_hrtimer(struct hrtimer *hrtimer) void uncore_pmu_start_hrtimer(struct intel_uncore_box *box) { - __hrtimer_start_range_ns(&box->hrtimer, - ns_to_ktime(box->hrtimer_duration), 0, - HRTIMER_MODE_REL_PINNED, 0); + hrtimer_start(&box->hrtimer, ns_to_ktime(box->hrtimer_duration), + HRTIMER_MODE_REL_PINNED); } void uncore_pmu_cancel_hrtimer(struct intel_uncore_box *box) @@ -810,7 +863,7 @@ static int uncore_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id int phys_id; bool first_box = false; - phys_id = uncore_pcibus_to_physid[pdev->bus->number]; + phys_id = uncore_pcibus_to_physid(pdev->bus); if (phys_id < 0) return -ENODEV; @@ -857,9 +910,10 @@ static void uncore_pci_remove(struct pci_dev *pdev) { struct intel_uncore_box *box = pci_get_drvdata(pdev); struct intel_uncore_pmu *pmu; - int i, cpu, phys_id = uncore_pcibus_to_physid[pdev->bus->number]; + int i, cpu, phys_id; bool last_box = false; + phys_id = uncore_pcibus_to_physid(pdev->bus); box = pci_get_drvdata(pdev); if (!box) { for (i = 0; i < UNCORE_EXTRA_PCI_DEV_MAX; i++) { @@ -912,6 +966,9 @@ static int __init uncore_pci_init(void) case 63: /* Haswell-EP */ ret = hswep_uncore_pci_init(); break; + case 86: /* BDX-DE */ + ret = bdx_uncore_pci_init(); + break; case 42: /* Sandy Bridge */ ret = snb_uncore_pci_init(); break; @@ -922,6 +979,9 @@ static int __init uncore_pci_init(void) case 69: /* Haswell Celeron */ ret = hsw_uncore_pci_init(); break; + case 61: /* Broadwell */ + ret = bdw_uncore_pci_init(); + break; default: return 0; } @@ -1207,6 +1267,11 @@ static int __init uncore_cpu_init(void) break; case 42: /* Sandy Bridge */ case 58: /* Ivy Bridge */ + case 60: /* Haswell */ + case 69: /* Haswell */ + case 70: /* Haswell */ + case 61: /* Broadwell */ + case 71: /* Broadwell */ snb_uncore_cpu_init(); break; case 45: /* Sandy Bridge-EP */ @@ -1222,6 +1287,9 @@ static int __init uncore_cpu_init(void) case 63: /* Haswell-EP */ hswep_uncore_cpu_init(); break; + case 86: /* BDX-DE */ + bdx_uncore_cpu_init(); + break; default: return 0; } diff --git a/kernel/arch/x86/kernel/cpu/perf_event_intel_uncore.h b/kernel/arch/x86/kernel/cpu/perf_event_intel_uncore.h index ceac8f5dc..2f0a4a98e 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event_intel_uncore.h +++ b/kernel/arch/x86/kernel/cpu/perf_event_intel_uncore.h @@ -117,6 +117,15 @@ struct uncore_event_desc { const char *config; }; +struct pci2phy_map { + struct list_head list; + int segment; + int pbus_to_physid[256]; +}; + +int uncore_pcibus_to_physid(struct pci_bus *bus); +struct pci2phy_map *__find_pci2phy_map(int segment); + ssize_t uncore_event_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf); @@ -317,7 +326,8 @@ u64 uncore_shared_reg_config(struct intel_uncore_box *box, int idx); extern struct intel_uncore_type **uncore_msr_uncores; extern struct intel_uncore_type **uncore_pci_uncores; extern struct pci_driver *uncore_pci_driver; -extern int uncore_pcibus_to_physid[256]; +extern raw_spinlock_t pci2phy_map_lock; +extern struct list_head pci2phy_map_head; extern struct pci_dev *uncore_extra_pci_dev[UNCORE_SOCKET_MAX][UNCORE_EXTRA_PCI_DEV_MAX]; extern struct event_constraint uncore_constraint_empty; @@ -325,6 +335,7 @@ extern struct event_constraint uncore_constraint_empty; int snb_uncore_pci_init(void); int ivb_uncore_pci_init(void); int hsw_uncore_pci_init(void); +int bdw_uncore_pci_init(void); void snb_uncore_cpu_init(void); void nhm_uncore_cpu_init(void); @@ -335,6 +346,8 @@ int ivbep_uncore_pci_init(void); void ivbep_uncore_cpu_init(void); int hswep_uncore_pci_init(void); void hswep_uncore_cpu_init(void); +int bdx_uncore_pci_init(void); +void bdx_uncore_cpu_init(void); /* perf_event_intel_uncore_nhmex.c */ void nhmex_uncore_cpu_init(void); diff --git a/kernel/arch/x86/kernel/cpu/perf_event_intel_uncore_snb.c b/kernel/arch/x86/kernel/cpu/perf_event_intel_uncore_snb.c index 4562e9e22..845256158 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event_intel_uncore_snb.c +++ b/kernel/arch/x86/kernel/cpu/perf_event_intel_uncore_snb.c @@ -7,6 +7,7 @@ #define PCI_DEVICE_ID_INTEL_IVB_E3_IMC 0x0150 #define PCI_DEVICE_ID_INTEL_HSW_IMC 0x0c00 #define PCI_DEVICE_ID_INTEL_HSW_U_IMC 0x0a04 +#define PCI_DEVICE_ID_INTEL_BDW_IMC 0x1604 /* SNB event control */ #define SNB_UNC_CTL_EV_SEL_MASK 0x000000ff @@ -44,6 +45,11 @@ #define SNB_UNC_CBO_0_PER_CTR0 0x706 #define SNB_UNC_CBO_MSR_OFFSET 0x10 +/* SNB ARB register */ +#define SNB_UNC_ARB_PER_CTR0 0x3b0 +#define SNB_UNC_ARB_PERFEVTSEL0 0x3b2 +#define SNB_UNC_ARB_MSR_OFFSET 0x10 + /* NHM global control register */ #define NHM_UNC_PERF_GLOBAL_CTL 0x391 #define NHM_UNC_FIXED_CTR 0x394 @@ -114,7 +120,7 @@ static struct intel_uncore_ops snb_uncore_msr_ops = { .read_counter = uncore_msr_read_counter, }; -static struct event_constraint snb_uncore_cbox_constraints[] = { +static struct event_constraint snb_uncore_arb_constraints[] = { UNCORE_EVENT_CONSTRAINT(0x80, 0x1), UNCORE_EVENT_CONSTRAINT(0x83, 0x1), EVENT_CONSTRAINT_END @@ -133,14 +139,28 @@ static struct intel_uncore_type snb_uncore_cbox = { .single_fixed = 1, .event_mask = SNB_UNC_RAW_EVENT_MASK, .msr_offset = SNB_UNC_CBO_MSR_OFFSET, - .constraints = snb_uncore_cbox_constraints, .ops = &snb_uncore_msr_ops, .format_group = &snb_uncore_format_group, .event_descs = snb_uncore_events, }; +static struct intel_uncore_type snb_uncore_arb = { + .name = "arb", + .num_counters = 2, + .num_boxes = 1, + .perf_ctr_bits = 44, + .perf_ctr = SNB_UNC_ARB_PER_CTR0, + .event_ctl = SNB_UNC_ARB_PERFEVTSEL0, + .event_mask = SNB_UNC_RAW_EVENT_MASK, + .msr_offset = SNB_UNC_ARB_MSR_OFFSET, + .constraints = snb_uncore_arb_constraints, + .ops = &snb_uncore_msr_ops, + .format_group = &snb_uncore_format_group, +}; + static struct intel_uncore_type *snb_msr_uncores[] = { &snb_uncore_cbox, + &snb_uncore_arb, NULL, }; @@ -400,15 +420,25 @@ static void snb_uncore_imc_event_del(struct perf_event *event, int flags) static int snb_pci2phy_map_init(int devid) { struct pci_dev *dev = NULL; - int bus; + struct pci2phy_map *map; + int bus, segment; dev = pci_get_device(PCI_VENDOR_ID_INTEL, devid, dev); if (!dev) return -ENOTTY; bus = dev->bus->number; - - uncore_pcibus_to_physid[bus] = 0; + segment = pci_domain_nr(dev->bus); + + raw_spin_lock(&pci2phy_map_lock); + map = __find_pci2phy_map(segment); + if (!map) { + raw_spin_unlock(&pci2phy_map_lock); + pci_dev_put(dev); + return -ENOMEM; + } + map->pbus_to_physid[bus] = 0; + raw_spin_unlock(&pci2phy_map_lock); pci_dev_put(dev); @@ -486,6 +516,14 @@ static const struct pci_device_id hsw_uncore_pci_ids[] = { { /* end: all zeroes */ }, }; +static const struct pci_device_id bdw_uncore_pci_ids[] = { + { /* IMC */ + PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_BDW_IMC), + .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0), + }, + { /* end: all zeroes */ }, +}; + static struct pci_driver snb_uncore_pci_driver = { .name = "snb_uncore", .id_table = snb_uncore_pci_ids, @@ -501,6 +539,11 @@ static struct pci_driver hsw_uncore_pci_driver = { .id_table = hsw_uncore_pci_ids, }; +static struct pci_driver bdw_uncore_pci_driver = { + .name = "bdw_uncore", + .id_table = bdw_uncore_pci_ids, +}; + struct imc_uncore_pci_dev { __u32 pci_id; struct pci_driver *driver; @@ -514,6 +557,7 @@ static const struct imc_uncore_pci_dev desktop_imc_pci_ids[] = { IMC_DEV(IVB_E3_IMC, &ivb_uncore_pci_driver), /* Xeon E3-1200 v2/3rd Gen Core processor */ IMC_DEV(HSW_IMC, &hsw_uncore_pci_driver), /* 4th Gen Core Processor */ IMC_DEV(HSW_U_IMC, &hsw_uncore_pci_driver), /* 4th Gen Core ULT Mobile Processor */ + IMC_DEV(BDW_IMC, &bdw_uncore_pci_driver), /* 5th Gen Core U */ { /* end marker */ } }; @@ -561,6 +605,11 @@ int hsw_uncore_pci_init(void) return imc_uncore_pci_init(); } +int bdw_uncore_pci_init(void) +{ + return imc_uncore_pci_init(); +} + /* end of Sandy Bridge uncore support */ /* Nehalem uncore support */ diff --git a/kernel/arch/x86/kernel/cpu/perf_event_intel_uncore_snbep.c b/kernel/arch/x86/kernel/cpu/perf_event_intel_uncore_snbep.c index 6d6e85dd5..f0f4fcba2 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event_intel_uncore_snbep.c +++ b/kernel/arch/x86/kernel/cpu/perf_event_intel_uncore_snbep.c @@ -1087,7 +1087,8 @@ static struct pci_driver snbep_uncore_pci_driver = { static int snbep_pci2phy_map_init(int devid) { struct pci_dev *ubox_dev = NULL; - int i, bus, nodeid; + int i, bus, nodeid, segment; + struct pci2phy_map *map; int err = 0; u32 config = 0; @@ -1106,16 +1107,27 @@ static int snbep_pci2phy_map_init(int devid) err = pci_read_config_dword(ubox_dev, 0x54, &config); if (err) break; + + segment = pci_domain_nr(ubox_dev->bus); + raw_spin_lock(&pci2phy_map_lock); + map = __find_pci2phy_map(segment); + if (!map) { + raw_spin_unlock(&pci2phy_map_lock); + err = -ENOMEM; + break; + } + /* * every three bits in the Node ID mapping register maps * to a particular node. */ for (i = 0; i < 8; i++) { if (nodeid == ((config >> (3 * i)) & 0x7)) { - uncore_pcibus_to_physid[bus] = i; + map->pbus_to_physid[bus] = i; break; } } + raw_spin_unlock(&pci2phy_map_lock); } if (!err) { @@ -1123,13 +1135,17 @@ static int snbep_pci2phy_map_init(int devid) * For PCI bus with no UBOX device, find the next bus * that has UBOX device and use its mapping. */ - i = -1; - for (bus = 255; bus >= 0; bus--) { - if (uncore_pcibus_to_physid[bus] >= 0) - i = uncore_pcibus_to_physid[bus]; - else - uncore_pcibus_to_physid[bus] = i; + raw_spin_lock(&pci2phy_map_lock); + list_for_each_entry(map, &pci2phy_map_head, list) { + i = -1; + for (bus = 255; bus >= 0; bus--) { + if (map->pbus_to_physid[bus] >= 0) + i = map->pbus_to_physid[bus]; + else + map->pbus_to_physid[bus] = i; + } } + raw_spin_unlock(&pci2phy_map_lock); } pci_dev_put(ubox_dev); @@ -2215,7 +2231,7 @@ static struct intel_uncore_type *hswep_pci_uncores[] = { NULL, }; -static DEFINE_PCI_DEVICE_TABLE(hswep_uncore_pci_ids) = { +static const struct pci_device_id hswep_uncore_pci_ids[] = { { /* Home Agent 0 */ PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2f30), .driver_data = UNCORE_PCI_DEV_DATA(HSWEP_PCI_UNCORE_HA, 0), @@ -2321,3 +2337,167 @@ int hswep_uncore_pci_init(void) return 0; } /* end of Haswell-EP uncore support */ + +/* BDX-DE uncore support */ + +static struct intel_uncore_type bdx_uncore_ubox = { + .name = "ubox", + .num_counters = 2, + .num_boxes = 1, + .perf_ctr_bits = 48, + .fixed_ctr_bits = 48, + .perf_ctr = HSWEP_U_MSR_PMON_CTR0, + .event_ctl = HSWEP_U_MSR_PMON_CTL0, + .event_mask = SNBEP_U_MSR_PMON_RAW_EVENT_MASK, + .fixed_ctr = HSWEP_U_MSR_PMON_UCLK_FIXED_CTR, + .fixed_ctl = HSWEP_U_MSR_PMON_UCLK_FIXED_CTL, + .num_shared_regs = 1, + .ops = &ivbep_uncore_msr_ops, + .format_group = &ivbep_uncore_ubox_format_group, +}; + +static struct event_constraint bdx_uncore_cbox_constraints[] = { + UNCORE_EVENT_CONSTRAINT(0x09, 0x3), + UNCORE_EVENT_CONSTRAINT(0x11, 0x1), + UNCORE_EVENT_CONSTRAINT(0x36, 0x1), + EVENT_CONSTRAINT_END +}; + +static struct intel_uncore_type bdx_uncore_cbox = { + .name = "cbox", + .num_counters = 4, + .num_boxes = 8, + .perf_ctr_bits = 48, + .event_ctl = HSWEP_C0_MSR_PMON_CTL0, + .perf_ctr = HSWEP_C0_MSR_PMON_CTR0, + .event_mask = SNBEP_CBO_MSR_PMON_RAW_EVENT_MASK, + .box_ctl = HSWEP_C0_MSR_PMON_BOX_CTL, + .msr_offset = HSWEP_CBO_MSR_OFFSET, + .num_shared_regs = 1, + .constraints = bdx_uncore_cbox_constraints, + .ops = &hswep_uncore_cbox_ops, + .format_group = &hswep_uncore_cbox_format_group, +}; + +static struct intel_uncore_type *bdx_msr_uncores[] = { + &bdx_uncore_ubox, + &bdx_uncore_cbox, + &hswep_uncore_pcu, + NULL, +}; + +void bdx_uncore_cpu_init(void) +{ + if (bdx_uncore_cbox.num_boxes > boot_cpu_data.x86_max_cores) + bdx_uncore_cbox.num_boxes = boot_cpu_data.x86_max_cores; + uncore_msr_uncores = bdx_msr_uncores; +} + +static struct intel_uncore_type bdx_uncore_ha = { + .name = "ha", + .num_counters = 4, + .num_boxes = 1, + .perf_ctr_bits = 48, + SNBEP_UNCORE_PCI_COMMON_INIT(), +}; + +static struct intel_uncore_type bdx_uncore_imc = { + .name = "imc", + .num_counters = 5, + .num_boxes = 2, + .perf_ctr_bits = 48, + .fixed_ctr_bits = 48, + .fixed_ctr = SNBEP_MC_CHy_PCI_PMON_FIXED_CTR, + .fixed_ctl = SNBEP_MC_CHy_PCI_PMON_FIXED_CTL, + .event_descs = hswep_uncore_imc_events, + SNBEP_UNCORE_PCI_COMMON_INIT(), +}; + +static struct intel_uncore_type bdx_uncore_irp = { + .name = "irp", + .num_counters = 4, + .num_boxes = 1, + .perf_ctr_bits = 48, + .event_mask = SNBEP_PMON_RAW_EVENT_MASK, + .box_ctl = SNBEP_PCI_PMON_BOX_CTL, + .ops = &hswep_uncore_irp_ops, + .format_group = &snbep_uncore_format_group, +}; + + +static struct event_constraint bdx_uncore_r2pcie_constraints[] = { + UNCORE_EVENT_CONSTRAINT(0x10, 0x3), + UNCORE_EVENT_CONSTRAINT(0x11, 0x3), + UNCORE_EVENT_CONSTRAINT(0x13, 0x1), + UNCORE_EVENT_CONSTRAINT(0x23, 0x1), + UNCORE_EVENT_CONSTRAINT(0x25, 0x1), + UNCORE_EVENT_CONSTRAINT(0x26, 0x3), + UNCORE_EVENT_CONSTRAINT(0x2d, 0x3), + EVENT_CONSTRAINT_END +}; + +static struct intel_uncore_type bdx_uncore_r2pcie = { + .name = "r2pcie", + .num_counters = 4, + .num_boxes = 1, + .perf_ctr_bits = 48, + .constraints = bdx_uncore_r2pcie_constraints, + SNBEP_UNCORE_PCI_COMMON_INIT(), +}; + +enum { + BDX_PCI_UNCORE_HA, + BDX_PCI_UNCORE_IMC, + BDX_PCI_UNCORE_IRP, + BDX_PCI_UNCORE_R2PCIE, +}; + +static struct intel_uncore_type *bdx_pci_uncores[] = { + [BDX_PCI_UNCORE_HA] = &bdx_uncore_ha, + [BDX_PCI_UNCORE_IMC] = &bdx_uncore_imc, + [BDX_PCI_UNCORE_IRP] = &bdx_uncore_irp, + [BDX_PCI_UNCORE_R2PCIE] = &bdx_uncore_r2pcie, + NULL, +}; + +static const struct pci_device_id bdx_uncore_pci_ids[] = { + { /* Home Agent 0 */ + PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6f30), + .driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_HA, 0), + }, + { /* MC0 Channel 0 */ + PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6fb0), + .driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_IMC, 0), + }, + { /* MC0 Channel 1 */ + PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6fb1), + .driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_IMC, 1), + }, + { /* IRP */ + PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6f39), + .driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_IRP, 0), + }, + { /* R2PCIe */ + PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6f34), + .driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_R2PCIE, 0), + }, + { /* end: all zeroes */ } +}; + +static struct pci_driver bdx_uncore_pci_driver = { + .name = "bdx_uncore", + .id_table = bdx_uncore_pci_ids, +}; + +int bdx_uncore_pci_init(void) +{ + int ret = snbep_pci2phy_map_init(0x6f1e); + + if (ret) + return ret; + uncore_pci_uncores = bdx_pci_uncores; + uncore_pci_driver = &bdx_uncore_pci_driver; + return 0; +} + +/* end of BDX-DE uncore support */ diff --git a/kernel/arch/x86/kernel/cpu/perf_event_msr.c b/kernel/arch/x86/kernel/cpu/perf_event_msr.c new file mode 100644 index 000000000..ec863b9a9 --- /dev/null +++ b/kernel/arch/x86/kernel/cpu/perf_event_msr.c @@ -0,0 +1,241 @@ +#include <linux/perf_event.h> + +enum perf_msr_id { + PERF_MSR_TSC = 0, + PERF_MSR_APERF = 1, + PERF_MSR_MPERF = 2, + PERF_MSR_PPERF = 3, + PERF_MSR_SMI = 4, + + PERF_MSR_EVENT_MAX, +}; + +static bool test_aperfmperf(int idx) +{ + return boot_cpu_has(X86_FEATURE_APERFMPERF); +} + +static bool test_intel(int idx) +{ + if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL || + boot_cpu_data.x86 != 6) + return false; + + switch (boot_cpu_data.x86_model) { + case 30: /* 45nm Nehalem */ + case 26: /* 45nm Nehalem-EP */ + case 46: /* 45nm Nehalem-EX */ + + case 37: /* 32nm Westmere */ + case 44: /* 32nm Westmere-EP */ + case 47: /* 32nm Westmere-EX */ + + case 42: /* 32nm SandyBridge */ + case 45: /* 32nm SandyBridge-E/EN/EP */ + + case 58: /* 22nm IvyBridge */ + case 62: /* 22nm IvyBridge-EP/EX */ + + case 60: /* 22nm Haswell Core */ + case 63: /* 22nm Haswell Server */ + case 69: /* 22nm Haswell ULT */ + case 70: /* 22nm Haswell + GT3e (Intel Iris Pro graphics) */ + + case 61: /* 14nm Broadwell Core-M */ + case 86: /* 14nm Broadwell Xeon D */ + case 71: /* 14nm Broadwell + GT3e (Intel Iris Pro graphics) */ + case 79: /* 14nm Broadwell Server */ + + case 55: /* 22nm Atom "Silvermont" */ + case 77: /* 22nm Atom "Silvermont Avoton/Rangely" */ + case 76: /* 14nm Atom "Airmont" */ + if (idx == PERF_MSR_SMI) + return true; + break; + + case 78: /* 14nm Skylake Mobile */ + case 94: /* 14nm Skylake Desktop */ + if (idx == PERF_MSR_SMI || idx == PERF_MSR_PPERF) + return true; + break; + } + + return false; +} + +struct perf_msr { + u64 msr; + struct perf_pmu_events_attr *attr; + bool (*test)(int idx); +}; + +PMU_EVENT_ATTR_STRING(tsc, evattr_tsc, "event=0x00"); +PMU_EVENT_ATTR_STRING(aperf, evattr_aperf, "event=0x01"); +PMU_EVENT_ATTR_STRING(mperf, evattr_mperf, "event=0x02"); +PMU_EVENT_ATTR_STRING(pperf, evattr_pperf, "event=0x03"); +PMU_EVENT_ATTR_STRING(smi, evattr_smi, "event=0x04"); + +static struct perf_msr msr[] = { + [PERF_MSR_TSC] = { 0, &evattr_tsc, NULL, }, + [PERF_MSR_APERF] = { MSR_IA32_APERF, &evattr_aperf, test_aperfmperf, }, + [PERF_MSR_MPERF] = { MSR_IA32_MPERF, &evattr_mperf, test_aperfmperf, }, + [PERF_MSR_PPERF] = { MSR_PPERF, &evattr_pperf, test_intel, }, + [PERF_MSR_SMI] = { MSR_SMI_COUNT, &evattr_smi, test_intel, }, +}; + +static struct attribute *events_attrs[PERF_MSR_EVENT_MAX + 1] = { + NULL, +}; + +static struct attribute_group events_attr_group = { + .name = "events", + .attrs = events_attrs, +}; + +PMU_FORMAT_ATTR(event, "config:0-63"); +static struct attribute *format_attrs[] = { + &format_attr_event.attr, + NULL, +}; +static struct attribute_group format_attr_group = { + .name = "format", + .attrs = format_attrs, +}; + +static const struct attribute_group *attr_groups[] = { + &events_attr_group, + &format_attr_group, + NULL, +}; + +static int msr_event_init(struct perf_event *event) +{ + u64 cfg = event->attr.config; + + if (event->attr.type != event->pmu->type) + return -ENOENT; + + if (cfg >= PERF_MSR_EVENT_MAX) + return -EINVAL; + + /* unsupported modes and filters */ + if (event->attr.exclude_user || + event->attr.exclude_kernel || + event->attr.exclude_hv || + event->attr.exclude_idle || + event->attr.exclude_host || + event->attr.exclude_guest || + event->attr.sample_period) /* no sampling */ + return -EINVAL; + + if (!msr[cfg].attr) + return -EINVAL; + + event->hw.idx = -1; + event->hw.event_base = msr[cfg].msr; + event->hw.config = cfg; + + return 0; +} + +static inline u64 msr_read_counter(struct perf_event *event) +{ + u64 now; + + if (event->hw.event_base) + rdmsrl(event->hw.event_base, now); + else + rdtscll(now); + + return now; +} +static void msr_event_update(struct perf_event *event) +{ + u64 prev, now; + s64 delta; + + /* Careful, an NMI might modify the previous event value. */ +again: + prev = local64_read(&event->hw.prev_count); + now = msr_read_counter(event); + + if (local64_cmpxchg(&event->hw.prev_count, prev, now) != prev) + goto again; + + delta = now - prev; + if (unlikely(event->hw.event_base == MSR_SMI_COUNT)) + delta = sign_extend64(delta, 31); + + local64_add(now - prev, &event->count); +} + +static void msr_event_start(struct perf_event *event, int flags) +{ + u64 now; + + now = msr_read_counter(event); + local64_set(&event->hw.prev_count, now); +} + +static void msr_event_stop(struct perf_event *event, int flags) +{ + msr_event_update(event); +} + +static void msr_event_del(struct perf_event *event, int flags) +{ + msr_event_stop(event, PERF_EF_UPDATE); +} + +static int msr_event_add(struct perf_event *event, int flags) +{ + if (flags & PERF_EF_START) + msr_event_start(event, flags); + + return 0; +} + +static struct pmu pmu_msr = { + .task_ctx_nr = perf_sw_context, + .attr_groups = attr_groups, + .event_init = msr_event_init, + .add = msr_event_add, + .del = msr_event_del, + .start = msr_event_start, + .stop = msr_event_stop, + .read = msr_event_update, + .capabilities = PERF_PMU_CAP_NO_INTERRUPT, +}; + +static int __init msr_init(void) +{ + int i, j = 0; + + if (!boot_cpu_has(X86_FEATURE_TSC)) { + pr_cont("no MSR PMU driver.\n"); + return 0; + } + + /* Probe the MSRs. */ + for (i = PERF_MSR_TSC + 1; i < PERF_MSR_EVENT_MAX; i++) { + u64 val; + + /* + * Virt sucks arse; you cannot tell if a R/O MSR is present :/ + */ + if (!msr[i].test(i) || rdmsrl_safe(msr[i].msr, &val)) + msr[i].attr = NULL; + } + + /* List remaining MSRs in the sysfs attrs. */ + for (i = 0; i < PERF_MSR_EVENT_MAX; i++) { + if (msr[i].attr) + events_attrs[j++] = &msr[i].attr->attr.attr; + } + events_attrs[j] = NULL; + + perf_pmu_register(&pmu_msr, "msr", -1); + + return 0; +} +device_initcall(msr_init); diff --git a/kernel/arch/x86/kernel/cpu/proc.c b/kernel/arch/x86/kernel/cpu/proc.c index e7d8c7608..18ca99f27 100644 --- a/kernel/arch/x86/kernel/cpu/proc.c +++ b/kernel/arch/x86/kernel/cpu/proc.c @@ -12,7 +12,8 @@ static void show_cpuinfo_core(struct seq_file *m, struct cpuinfo_x86 *c, { #ifdef CONFIG_SMP seq_printf(m, "physical id\t: %d\n", c->phys_proc_id); - seq_printf(m, "siblings\t: %d\n", cpumask_weight(cpu_core_mask(cpu))); + seq_printf(m, "siblings\t: %d\n", + cpumask_weight(topology_core_cpumask(cpu))); seq_printf(m, "core id\t\t: %d\n", c->cpu_core_id); seq_printf(m, "cpu cores\t: %d\n", c->booted_cores); seq_printf(m, "apicid\t\t: %d\n", c->apicid); diff --git a/kernel/arch/x86/kernel/cpu/scattered.c b/kernel/arch/x86/kernel/cpu/scattered.c index 3d423a101..608fb26c7 100644 --- a/kernel/arch/x86/kernel/cpu/scattered.c +++ b/kernel/arch/x86/kernel/cpu/scattered.c @@ -37,7 +37,7 @@ void init_scattered_cpuid_features(struct cpuinfo_x86 *c) { X86_FEATURE_PLN, CR_EAX, 4, 0x00000006, 0 }, { X86_FEATURE_PTS, CR_EAX, 6, 0x00000006, 0 }, { X86_FEATURE_HWP, CR_EAX, 7, 0x00000006, 0 }, - { X86_FEATURE_HWP_NOITFY, CR_EAX, 8, 0x00000006, 0 }, + { X86_FEATURE_HWP_NOTIFY, CR_EAX, 8, 0x00000006, 0 }, { X86_FEATURE_HWP_ACT_WINDOW, CR_EAX, 9, 0x00000006, 0 }, { X86_FEATURE_HWP_EPP, CR_EAX,10, 0x00000006, 0 }, { X86_FEATURE_HWP_PKG_REQ, CR_EAX,11, 0x00000006, 0 }, |