diff options
| author | Jiri Kosina <jkosina@suse.cz> | 2011-06-10 14:46:48 +0200 |
|---|---|---|
| committer | Jiri Kosina <jkosina@suse.cz> | 2011-06-10 14:46:57 +0200 |
| commit | 5be5758c114b18260c6fd4c8373bf89e39b0fe82 (patch) | |
| tree | 54390f904df6ff11e570f764c444356cf2709fda /arch/x86 | |
| parent | 71f66a6580c4e42df377bebbcca5c72661a40700 (diff) | |
| parent | 7f45e5cd1718ed769295033ca214032848a0097d (diff) | |
Merge branch 'master' into for-next
Sync with Linus' tree to be able to apply patches against new
code I have in queue.
Diffstat (limited to 'arch/x86')
73 files changed, 2773 insertions, 1850 deletions
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 880fcb6c86f..da349723d41 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -17,8 +17,6 @@ config X86_64 config X86 def_bool y select HAVE_AOUT if X86_32 - select HAVE_READQ - select HAVE_WRITEQ select HAVE_UNSTABLE_SCHED_CLOCK select HAVE_IDE select HAVE_OPROFILE @@ -66,7 +64,6 @@ config X86 select HAVE_GENERIC_HARDIRQS select HAVE_SPARSE_IRQ select GENERIC_FIND_FIRST_BIT - select GENERIC_FIND_NEXT_BIT select GENERIC_IRQ_PROBE select GENERIC_PENDING_IRQ if SMP select GENERIC_IRQ_SHOW @@ -917,6 +914,7 @@ config TOSHIBA config I8K tristate "Dell laptop support" + select HWMON ---help--- This adds a driver to safely access the System Management Mode of the CPU on the Dell Inspiron 8000. The System Management Mode diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug index 615e18810f4..c0f8a5c8891 100644 --- a/arch/x86/Kconfig.debug +++ b/arch/x86/Kconfig.debug @@ -66,26 +66,6 @@ config DEBUG_STACKOVERFLOW This option will cause messages to be printed if free stack space drops below a certain limit. -config DEBUG_STACK_USAGE - bool "Stack utilization instrumentation" - depends on DEBUG_KERNEL - ---help--- - Enables the display of the minimum amount of free stack which each - task has ever had available in the sysrq-T and sysrq-P debug output. - - This option will slow down process creation somewhat. - -config DEBUG_PER_CPU_MAPS - bool "Debug access to per_cpu maps" - depends on DEBUG_KERNEL - depends on SMP - ---help--- - Say Y to verify that the per_cpu map being accessed has - been setup. Adds a fair amount of code to kernel memory - and decreases performance. - - Say N if unsure. - config X86_PTDUMP bool "Export kernel pagetable layout to userspace via debugfs" depends on DEBUG_KERNEL diff --git a/arch/x86/configs/i386_defconfig b/arch/x86/configs/i386_defconfig index 6f9872658dd..2bf18059fbe 100644 --- a/arch/x86/configs/i386_defconfig +++ b/arch/x86/configs/i386_defconfig @@ -10,7 +10,6 @@ CONFIG_TASK_IO_ACCOUNTING=y CONFIG_AUDIT=y CONFIG_LOG_BUF_SHIFT=18 CONFIG_CGROUPS=y -CONFIG_CGROUP_NS=y CONFIG_CGROUP_FREEZER=y CONFIG_CPUSETS=y CONFIG_CGROUP_CPUACCT=y diff --git a/arch/x86/configs/x86_64_defconfig b/arch/x86/configs/x86_64_defconfig index ee01a9d5d4f..22a0dc8e51d 100644 --- a/arch/x86/configs/x86_64_defconfig +++ b/arch/x86/configs/x86_64_defconfig @@ -11,7 +11,6 @@ CONFIG_TASK_IO_ACCOUNTING=y CONFIG_AUDIT=y CONFIG_LOG_BUF_SHIFT=18 CONFIG_CGROUPS=y -CONFIG_CGROUP_NS=y CONFIG_CGROUP_FREEZER=y CONFIG_CPUSETS=y CONFIG_CGROUP_CPUACCT=y diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S index 95f5826be45..c1870dddd32 100644 --- a/arch/x86/ia32/ia32entry.S +++ b/arch/x86/ia32/ia32entry.S @@ -849,4 +849,5 @@ ia32_sys_call_table: .quad compat_sys_clock_adjtime .quad sys_syncfs .quad compat_sys_sendmmsg /* 345 */ + .quad sys_setns ia32_syscall_end: diff --git a/arch/x86/include/asm/acpi.h b/arch/x86/include/asm/acpi.h index 416d865eae3..610001d385d 100644 --- a/arch/x86/include/asm/acpi.h +++ b/arch/x86/include/asm/acpi.h @@ -139,7 +139,7 @@ static inline unsigned int acpi_processor_cstate_check(unsigned int max_cstate) boot_cpu_data.x86_model <= 0x05 && boot_cpu_data.x86_mask < 0x0A) return 1; - else if (c1e_detected) + else if (amd_e400_c1e_detected) return 1; else return max_cstate; diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h index 5dc6acc98db..71cc3800712 100644 --- a/arch/x86/include/asm/cpufeature.h +++ b/arch/x86/include/asm/cpufeature.h @@ -125,7 +125,7 @@ #define X86_FEATURE_OSXSAVE (4*32+27) /* "" XSAVE enabled in the OS */ #define X86_FEATURE_AVX (4*32+28) /* Advanced Vector Extensions */ #define X86_FEATURE_F16C (4*32+29) /* 16-bit fp conversions */ -#define X86_FEATURE_RDRND (4*32+30) /* The RDRAND instruction */ +#define X86_FEATURE_RDRAND (4*32+30) /* The RDRAND instruction */ #define X86_FEATURE_HYPERVISOR (4*32+31) /* Running on a hypervisor */ /* VIA/Cyrix/Centaur-defined CPU features, CPUID level 0xC0000001, word 5 */ diff --git a/arch/x86/include/asm/desc.h b/arch/x86/include/asm/desc.h index 617bd56b307..7b439d9aea2 100644 --- a/arch/x86/include/asm/desc.h +++ b/arch/x86/include/asm/desc.h @@ -4,30 +4,33 @@ #include <asm/desc_defs.h> #include <asm/ldt.h> #include <asm/mmu.h> + #include <linux/smp.h> -static inline void fill_ldt(struct desc_struct *desc, - const struct user_desc *info) -{ - desc->limit0 = info->limit & 0x0ffff; - desc->base0 = info->base_addr & 0x0000ffff; - - desc->base1 = (info->base_addr & 0x00ff0000) >> 16; - desc->type = (info->read_exec_only ^ 1) << 1; - desc->type |= info->contents << 2; - desc->s = 1; - desc->dpl = 0x3; - desc->p = info->seg_not_present ^ 1; - desc->limit = (info->limit & 0xf0000) >> 16; - desc->avl = info->useable; - desc->d = info->seg_32bit; - desc->g = info->limit_in_pages; - desc->base2 = (info->base_addr & 0xff000000) >> 24; +static inline void fill_ldt(struct desc_struct *desc, const struct user_desc *info) +{ + desc->limit0 = info->limit & 0x0ffff; + + desc->base0 = (info->base_addr & 0x0000ffff); + desc->base1 = (info->base_addr & 0x00ff0000) >> 16; + + desc->type = (info->read_exec_only ^ 1) << 1; + desc->type |= info->contents << 2; + + desc->s = 1; + desc->dpl = 0x3; + desc->p = info->seg_not_present ^ 1; + desc->limit = (info->limit & 0xf0000) >> 16; + desc->avl = info->useable; + desc->d = info->seg_32bit; + desc->g = info->limit_in_pages; + + desc->base2 = (info->base_addr & 0xff000000) >> 24; /* * Don't allow setting of the lm bit. It is useless anyway * because 64bit system calls require __USER_CS: */ - desc->l = 0; + desc->l = 0; } extern struct desc_ptr idt_descr; @@ -36,6 +39,7 @@ extern gate_desc idt_table[]; struct gdt_page { struct desc_struct gdt[GDT_ENTRIES]; } __attribute__((aligned(PAGE_SIZE))); + DECLARE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page); static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu) @@ -48,16 +52,16 @@ static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu) static inline void pack_gate(gate_desc *gate, unsigned type, unsigned long func, unsigned dpl, unsigned ist, unsigned seg) { - gate->offset_low = PTR_LOW(func); - gate->segment = __KERNEL_CS; - gate->ist = ist; - gate->p = 1; - gate->dpl = dpl; - gate->zero0 = 0; - gate->zero1 = 0; - gate->type = type; - gate->offset_middle = PTR_MIDDLE(func); - gate->offset_high = PTR_HIGH(func); + gate->offset_low = PTR_LOW(func); + gate->segment = __KERNEL_CS; + gate->ist = ist; + gate->p = 1; + gate->dpl = dpl; + gate->zero0 = 0; + gate->zero1 = 0; + gate->type = type; + gate->offset_middle = PTR_MIDDLE(func); + gate->offset_high = PTR_HIGH(func); } #else @@ -66,8 +70,7 @@ static inline void pack_gate(gate_desc *gate, unsigned char type, unsigned short seg) { gate->a = (seg << 16) | (base & 0xffff); - gate->b = (base & 0xffff0000) | - (((0x80 | type | (dpl << 5)) & 0xff) << 8); + gate->b = (base & 0xffff0000) | (((0x80 | type | (dpl << 5)) & 0xff) << 8); } #endif @@ -75,31 +78,29 @@ static inline void pack_gate(gate_desc *gate, unsigned char type, static inline int desc_empty(const void *ptr) { const u32 *desc = ptr; + return !(desc[0] | desc[1]); } #ifdef CONFIG_PARAVIRT #include <asm/paravirt.h> #else -#define load_TR_desc() native_load_tr_desc() -#define load_gdt(dtr) native_load_gdt(dtr) -#define load_idt(dtr) native_load_idt(dtr) -#define load_tr(tr) asm volatile("ltr %0"::"m" (tr)) -#define load_ldt(ldt) asm volatile("lldt %0"::"m" (ldt)) - -#define store_gdt(dtr) native_store_gdt(dtr) -#define store_idt(dtr) native_store_idt(dtr) -#define store_tr(tr) (tr = native_store_tr()) - -#define load_TLS(t, cpu) native_load_tls(t, cpu) -#define set_ldt native_set_ldt - -#define write_ldt_entry(dt, entry, desc) \ - native_write_ldt_entry(dt, entry, desc) -#define write_gdt_entry(dt, entry, desc, type) \ - native_write_gdt_entry(dt, entry, desc, type) -#define write_idt_entry(dt, entry, g) \ - native_write_idt_entry(dt, entry, g) +#define load_TR_desc() native_load_tr_desc() +#define load_gdt(dtr) native_load_gdt(dtr) +#define load_idt(dtr) native_load_idt(dtr) +#define load_tr(tr) asm volatile("ltr %0"::"m" (tr)) +#define load_ldt(ldt) asm volatile("lldt %0"::"m" (ldt)) + +#define store_gdt(dtr) native_store_gdt(dtr) +#define store_idt(dtr) native_store_idt(dtr) +#define store_tr(tr) (tr = native_store_tr()) + +#define load_TLS(t, cpu) native_load_tls(t, cpu) +#define set_ldt native_set_ldt + +#define write_ldt_entry(dt, entry, desc) native_write_ldt_entry(dt, entry, desc) +#define write_gdt_entry(dt, entry, desc, type) native_write_gdt_entry(dt, entry, desc, type) +#define write_idt_entry(dt, entry, g) native_write_idt_entry(dt, entry, g) static inline void paravirt_alloc_ldt(struct desc_struct *ldt, unsigned entries) { @@ -112,33 +113,27 @@ static inline void paravirt_free_ldt(struct desc_struct *ldt, unsigned entries) #define store_ldt(ldt) asm("sldt %0" : "=m"(ldt)) -static inline void native_write_idt_entry(gate_desc *idt, int entry, - const gate_desc *gate) +static inline void native_write_idt_entry(gate_desc *idt, int entry, const gate_desc *gate) { memcpy(&idt[entry], gate, sizeof(*gate)); } -static inline void native_write_ldt_entry(struct desc_struct *ldt, int entry, - const void *desc) +static inline void native_write_ldt_entry(struct desc_struct *ldt, int entry, const void *desc) { memcpy(&ldt[entry], desc, 8); } -static inline void native_write_gdt_entry(struct desc_struct *gdt, int entry, - const void *desc, int type) +static inline void +native_write_gdt_entry(struct desc_struct *gdt, int entry, const void *desc, int type) { unsigned int size; + switch (type) { - case DESC_TSS: - size = sizeof(tss_desc); - break; - case DESC_LDT: - size = sizeof(ldt_desc); - break; - default: - size = sizeof(struct desc_struct); - break; + case DESC_TSS: size = sizeof(tss_desc); break; + case DESC_LDT: size = sizeof(ldt_desc); break; + default: size = sizeof(*gdt); break; } + memcpy(&gdt[entry], desc, size); } @@ -154,20 +149,21 @@ static inline void pack_descriptor(struct desc_struct *desc, unsigned long base, } -static inline void set_tssldt_descriptor(void *d, unsigned long addr, - unsigned type, unsigned size) +static inline void set_tssldt_descriptor(void *d, unsigned long addr, unsigned type, unsigned size) { #ifdef CONFIG_X86_64 struct ldttss_desc64 *desc = d; + memset(desc, 0, sizeof(*desc)); - desc->limit0 = size & 0xFFFF; - desc->base0 = PTR_LOW(addr); - desc->base1 = PTR_MIDDLE(addr) & 0xFF; - desc->type = type; - desc->p = 1; - desc->limit1 = (size >> 16) & 0xF; - desc->base2 = (PTR_MIDDLE(addr) >> 8) & 0xFF; - desc->base3 = PTR_HIGH(addr); + + desc->limit0 = size & 0xFFFF; + desc->base0 = PTR_LOW(addr); + desc->base1 = PTR_MIDDLE(addr) & 0xFF; + desc->type = type; + desc->p = 1; + desc->limit1 = (size >> 16) & 0xF; + desc->base2 = (PTR_MIDDLE(addr) >> 8) & 0xFF; + desc->base3 = PTR_HIGH(addr); #else pack_descriptor((struct desc_struct *)d, addr, size, 0x80 | type, 0); #endif @@ -237,14 +233,16 @@ static inline void native_store_idt(struct desc_ptr *dtr) static inline unsigned long native_store_tr(void) { unsigned long tr; + asm volatile("str %0":"=r" (tr)); + return tr; } static inline void native_load_tls(struct thread_struct *t, unsigned int cpu) { - unsigned int i; struct desc_struct *gdt = get_cpu_gdt_table(cpu); + unsigned int i; for (i = 0; i < GDT_ENTRY_TLS_ENTRIES; i++) gdt[GDT_ENTRY_TLS_MIN + i] = t->tls_array[i]; @@ -313,6 +311,7 @@ static inline void _set_gate(int gate, unsigned type, void *addr, unsigned dpl, unsigned ist, unsigned seg) { gate_desc s; + pack_gate(&s, type, (unsigned long)addr, dpl, ist, seg); /* * does not need to be atomic because it is only done once at @@ -343,8 +342,9 @@ static inline void alloc_system_vector(int vector) set_bit(vector, used_vectors); if (first_system_vector > vector) first_system_vector = vector; - } else + } else { BUG(); + } } static inline void alloc_intr_gate(unsigned int n, void *addr) diff --git a/arch/x86/include/asm/idle.h b/arch/x86/include/asm/idle.h index 38d87379e27..f49253d7571 100644 --- a/arch/x86/include/asm/idle.h +++ b/arch/x86/include/asm/idle.h @@ -16,6 +16,6 @@ static inline void enter_idle(void) { } static inline void exit_idle(void) { } #endif /* CONFIG_X86_64 */ -void c1e_remove_cpu(int cpu); +void amd_e400_remove_cpu(int cpu); #endif /* _ASM_X86_IDLE_H */ diff --git a/arch/x86/include/asm/io.h b/arch/x86/include/asm/io.h index 07227308252..d02804d650c 100644 --- a/arch/x86/include/asm/io.h +++ b/arch/x86/include/asm/io.h @@ -38,7 +38,6 @@ #include <linux/string.h> #include <linux/compiler.h> -#include <asm-generic/int-ll64.h> #include <asm/page.h> #include <xen/xen.h> @@ -87,27 +86,6 @@ build_mmio_write(__writel, "l", unsigned int, "r", ) build_mmio_read(readq, "q", unsigned long, "=r", :"memory") build_mmio_write(writeq, "q", unsigned long, "r", :"memory") -#else - -static inline __u64 readq(const volatile void __iomem *addr) -{ - const volatile u32 __iomem *p = addr; - u32 low, high; - - low = readl(p); - high = readl(p + 1); - - return low + ((u64)high << 32); -} - -static inline void writeq(__u64 val, volatile void __iomem *addr) -{ - writel(val, addr); - writel(val >> 32, addr+4); -} - -#endif - #define readq_relaxed(a) readq(a) #define __raw_readq(a) readq(a) @@ -117,6 +95,8 @@ static inline void writeq(__u64 val, volatile void __iomem *addr) #define readq readq #define writeq writeq +#endif + /** * virt_to_phys - map virtual addresses to physical * @address: address to remap diff --git a/arch/x86/include/asm/kgdb.h b/arch/x86/include/asm/kgdb.h index 396f5b5fc4d..77e95f54570 100644 --- a/arch/x86/include/asm/kgdb.h +++ b/arch/x86/include/asm/kgdb.h @@ -77,6 +77,7 @@ static inline void arch_kgdb_breakpoint(void) } #define BREAK_INSTR_SIZE 1 #define CACHE_FLUSH_IS_SAFE 1 +#define GDB_ADJUSTS_BREAK_OFFSET extern int kgdb_ll_trap(int cmd, const char *str, struct pt_regs *regs, long err, int trap, int sig); diff --git a/arch/x86/include/asm/linkage.h b/arch/x86/include/asm/linkage.h index 12d55e773eb..48142971b25 100644 --- a/arch/x86/include/asm/linkage.h +++ b/arch/x86/include/asm/linkage.h @@ -8,11 +8,6 @@ #ifdef CONFIG_X86_32 #define asmlinkage CPP_ASMLINKAGE __attribute__((regparm(0))) -/* - * For 32-bit UML - mark functions implemented in assembly that use - * regparm input parameters: - */ -#define asmregparm __attribute__((regparm(3))) /* * Make sure the compiler doesn't do anything stupid with the diff --git a/arch/x86/include/asm/mmu.h b/arch/x86/include/asm/mmu.h index aeff3e89b22..5f55e696276 100644 --- a/arch/x86/include/asm/mmu.h +++ b/arch/x86/include/asm/mmu.h @@ -11,14 +11,14 @@ typedef struct { void *ldt; int size; - struct mutex lock; - void *vdso; #ifdef CONFIG_X86_64 /* True if mm supports a task running in 32 bit compatibility mode. */ unsigned short ia32_compat; #endif + struct mutex lock; + void *vdso; } mm_context_t; #ifdef CONFIG_SMP diff --git a/arch/x86/include/asm/percpu.h b/arch/x86/include/asm/percpu.h index 53278b0dfdf..a0a9779084d 100644 --- a/arch/x86/include/asm/percpu.h +++ b/arch/x86/include/asm/percpu.h @@ -509,6 +509,11 @@ do { \ * it in software. The address used in the cmpxchg16 instruction must be * aligned to a 16 byte boundary. */ +#ifdef CONFIG_SMP +#define CMPXCHG16B_EMU_CALL "call this_cpu_cmpxchg16b_emu\n\t" ASM_NOP3 +#else +#define CMPXCHG16B_EMU_CALL "call this_cpu_cmpxchg16b_emu\n\t" ASM_NOP2 +#endif #define percpu_cmpxchg16b_double(pcp1, o1, o2, n1, n2) \ ({ \ char __ret; \ @@ -517,7 +522,7 @@ do { \ typeof(o2) __o2 = o2; \ typeof(o2) __n2 = n2; \ typeof(o2) __dummy; \ - alternative_io("call this_cpu_cmpxchg16b_emu\n\t" ASM_NOP4, \ + alternative_io(CMPXCHG16B_EMU_CALL, \ "cmpxchg16b " __percpu_prefix "(%%rsi)\n\tsetz %0\n\t", \ X86_FEATURE_CX16, \ ASM_OUTPUT2("=a"(__ret), "=d"(__dummy)), \ diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h index 4c25ab48257..219371546af 100644 --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -754,10 +754,10 @@ static inline void __sti_mwait(unsigned long eax, unsigned long ecx) extern void mwait_idle_with_hints(unsigned long eax, unsigned long ecx); extern void select_idle_routine(const struct cpuinfo_x86 *c); -extern void init_c1e_mask(void); +extern void init_amd_e400_c1e_mask(void); extern unsigned long boot_option_idle_override; -extern bool c1e_detected; +extern bool amd_e400_c1e_detected; enum idle_boot_override {IDLE_NO_OVERRIDE=0, IDLE_HALT, IDLE_NOMWAIT, IDLE_POLL, IDLE_FORCE_MWAIT}; diff --git a/arch/x86/include/asm/ptrace.h b/arch/x86/include/asm/ptrace.h index 1babf8adecd..94e7618fcac 100644 --- a/arch/x86/include/asm/ptrace.h +++ b/arch/x86/include/asm/ptrace.h @@ -136,6 +136,7 @@ struct cpuinfo_x86; struct task_struct; extern unsigned long profile_pc(struct pt_regs *regs); +#define profile_pc profile_pc extern unsigned long convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs); @@ -202,20 +203,11 @@ static inline unsigned long kernel_stack_pointer(struct pt_regs *regs) #endif } -static inline unsigned long instruction_pointer(struct pt_regs *regs) -{ - return regs->ip; -} - -static inline unsigned long frame_pointer(struct pt_regs *regs) -{ - return regs->bp; -} +#define GET_IP(regs) ((regs)->ip) +#define GET_FP(regs) ((regs)->bp) +#define GET_USP(regs) ((regs)->sp) -static inline unsigned long user_stack_pointer(struct pt_regs *regs) -{ - return regs->sp; -} +#include <asm-generic/ptrace.h> /* Query offset/name of register from its name/offset */ extern int regs_query_register_offset(const char *name); diff --git a/arch/x86/include/asm/suspend_32.h b/arch/x86/include/asm/suspend_32.h index fd921c3a684..487055c8c1a 100644 --- a/arch/x86/include/asm/suspend_32.h +++ b/arch/x86/include/asm/suspend_32.h @@ -9,8 +9,6 @@ #include <asm/desc.h> #include <asm/i387.h> -static inline int arch_prepare_suspend(void) { return 0; } - /* image of the saved processor state */ struct saved_context { u16 es, fs, gs, ss; diff --git a/arch/x86/include/asm/suspend_64.h b/arch/x86/include/asm/suspend_64.h index 8d942afae68..09b0bf10415 100644 --- a/arch/x86/include/asm/suspend_64.h +++ b/arch/x86/include/asm/suspend_64.h @@ -9,11 +9,6 @@ #include <asm/desc.h> #include <asm/i387.h> -static inline int arch_prepare_suspend(void) -{ - return 0; -} - /* * Image of the saved processor state, used by the low level ACPI suspend to * RAM code and by the low level hibernation code. diff --git a/arch/x86/include/asm/tsc.h b/arch/x86/include/asm/tsc.h index 83e2efd181e..9db5583b6d3 100644 --- a/arch/x86/include/asm/tsc.h +++ b/arch/x86/include/asm/tsc.h @@ -51,6 +51,10 @@ extern int unsynchronized_tsc(void); extern int check_tsc_unstable(void); extern unsigned long native_calibrate_tsc(void); +#ifdef CONFIG_X86_64 +extern cycles_t vread_tsc(void); +#endif + /* * Boot-time check whether the TSCs are synchronized across * all CPUs/cores: diff --git a/arch/x86/include/asm/unistd_32.h b/arch/x86/include/asm/unistd_32.h index fb6a625c99b..593485b38ab 100644 --- a/arch/x86/include/asm/unistd_32.h +++ b/arch/x86/include/asm/unistd_32.h @@ -351,10 +351,11 @@ #define __NR_clock_adjtime 343 #define __NR_syncfs 344 #define __NR_sendmmsg 345 +#define __NR_setns 346 #ifdef __KERNEL__ -#define NR_syscalls 346 +#define NR_syscalls 347 #define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_OLD_READDIR diff --git a/arch/x86/include/asm/unistd_64.h b/arch/x86/include/asm/unistd_64.h index 79f90eb15aa..705bf139288 100644 --- a/arch/x86/include/asm/unistd_64.h +++ b/arch/x86/include/asm/unistd_64.h @@ -679,6 +679,8 @@ __SYSCALL(__NR_clock_adjtime, sys_clock_adjtime) __SYSCALL(__NR_syncfs, sys_syncfs) #define __NR_sendmmsg 307 __SYSCALL(__NR_sendmmsg, sys_sendmmsg) +#define __NR_setns 308 +__SYSCALL(__NR_setns, sys_setns) #ifndef __NO_STUBS #define __ARCH_WANT_OLD_READDIR diff --git a/arch/x86/include/asm/uv/uv_bau.h b/arch/x86/include/asm/uv/uv_bau.h index 130f1eeee5f..a291c40efd4 100644 --- a/arch/x86/include/asm/uv/uv_bau.h +++ b/arch/x86/include/asm/uv/uv_bau.h @@ -5,7 +5,7 @@ * * SGI UV Broadcast Assist Unit definitions * - * Copyright (C) 2008 Silicon Graphics, Inc. All rights reserved. + * Copyright (C) 2008-2011 Silicon Graphics, Inc. All rights reserved. */ #ifndef _ASM_X86_UV_UV_BAU_H @@ -35,17 +35,20 @@ #define MAX_CPUS_PER_UVHUB 64 #define MAX_CPUS_PER_SOCKET 32 -#define UV_ADP_SIZE 64 /* hardware-provided max. */ -#define UV_CPUS_PER_ACT_STATUS 32 /* hardware-provided max. */ -#define UV_ITEMS_PER_DESCRIPTOR 8 +#define ADP_SZ 64 /* hardware-provided max. */ +#define UV_CPUS_PER_AS 32 /* hardware-provided max. */ +#define ITEMS_PER_DESC 8 /* the 'throttle' to prevent the hardware stay-busy bug */ #define MAX_BAU_CONCURRENT 3 #define UV_ACT_STATUS_MASK 0x3 #define UV_ACT_STATUS_SIZE 2 #define UV_DISTRIBUTION_SIZE 256 #define UV_SW_ACK_NPENDING 8 -#define UV_NET_ENDPOINT_INTD 0x38 -#define UV_DESC_BASE_PNODE_SHIFT 49 +#define UV1_NET_ENDPOINT_INTD 0x38 +#define UV2_NET_ENDPOINT_INTD 0x28 +#define UV_NET_ENDPOINT_INTD (is_uv1_hub() ? \ + UV1_NET_ENDPOINT_INTD : UV2_NET_ENDPOINT_INTD) +#define UV_DESC_PSHIFT 49 #define UV_PAYLOADQ_PNODE_SHIFT 49 #define UV_PTC_BASENAME "sgi_uv/ptc_statistics" #define UV_BAU_BASENAME "sgi_uv/bau_tunables" @@ -53,29 +56,64 @@ #define UV_BAU_TUNABLES_FILE "bau_tunables" #define WHITESPACE " \t\n" #define uv_physnodeaddr(x) ((__pa((unsigned long)(x)) & uv_mmask)) -#define UV_ENABLE_INTD_SOFT_ACK_MODE_SHIFT 15 -#define UV_INTD_SOFT_ACK_TIMEOUT_PERIOD_SHIFT 16 -#define UV_INTD_SOFT_ACK_TIMEOUT_PERIOD 0x0000000009UL +#define cpubit_isset(cpu, bau_local_cpumask) \ + test_bit((cpu), (bau_local_cpumask).bits) + /* [19:16] SOFT_ACK timeout period 19: 1 is urgency 7 17:16 1 is multiplier */ -#define BAU_MISC_CONTROL_MULT_MASK 3 +/* + * UV2: Bit 19 selects between + * (0): 10 microsecond timebase and + * (1): 80 microseconds + * we're using 655us, similar to UV1: 65 units of 10us + */ +#define UV1_INTD_SOFT_ACK_TIMEOUT_PERIOD (9UL) +#define UV2_INTD_SOFT_ACK_TIMEOUT_PERIOD (65*10UL) + +#define UV_INTD_SOFT_ACK_TIMEOUT_PERIOD (is_uv1_hub() ? \ + UV1_INTD_SOFT_ACK_TIMEOUT_PERIOD : \ + UV2_INTD_SOFT_ACK_TIMEOUT_PERIOD) -#define UVH_AGING_PRESCALE_SEL 0x000000b000UL +#define BAU_MISC_CONTROL_MULT_MASK 3 + +#define UVH_AGING_PRESCALE_SEL 0x000000b000UL /* [30:28] URGENCY_7 an index into a table of times */ -#define BAU_URGENCY_7_SHIFT 28 -#define BAU_URGENCY_7_MASK 7 +#define BAU_URGENCY_7_SHIFT 28 +#define BAU_URGENCY_7_MASK 7 -#define UVH_TRANSACTION_TIMEOUT 0x000000b200UL +#define UVH_TRANSACTION_TIMEOUT 0x000000b200UL /* [45:40] BAU - BAU transaction timeout select - a multiplier */ -#define BAU_TRANS_SHIFT 40 -#define BAU_TRANS_MASK 0x3f +#define BAU_TRANS_SHIFT 40 +#define BAU_TRANS_MASK 0x3f + +/* + * shorten some awkward names + */ +#define AS_PUSH_SHIFT UVH_LB_BAU_SB_ACTIVATION_CONTROL_PUSH_SHFT +#define SOFTACK_MSHIFT UVH_LB_BAU_MISC_CONTROL_ENABLE_INTD_SOFT_ACK_MODE_SHFT +#define SOFTACK_PSHIFT UVH_LB_BAU_MISC_CONTROL_INTD_SOFT_ACK_TIMEOUT_PERIOD_SHFT +#define SOFTACK_TIMEOUT_PERIOD UV_INTD_SOFT_ACK_TIMEOUT_PERIOD +#define write_gmmr uv_write_global_mmr64 +#define write_lmmr uv_write_local_mmr +#define read_lmmr uv_read_local_mmr +#define read_gmmr uv_read_global_mmr64 /* * bits in UVH_LB_BAU_SB_ACTIVATION_STATUS_0/1 */ -#define DESC_STATUS_IDLE 0 -#define DESC_STATUS_ACTIVE 1 -#define DESC_STATUS_DESTINATION_TIMEOUT 2 -#define DESC_STATUS_SOURCE_TIMEOUT 3 +#define DS_IDLE 0 +#define DS_ACTIVE 1 +#define DS_DESTINATION_TIMEOUT 2 +#define DS_SOURCE_TIMEOUT 3 +/* + * bits put together from HRP_LB_BAU_SB_ACTIVATION_STATUS_0/1/2 + * values 1 and 5 will not occur + */ +#define UV2H_DESC_IDLE 0 +#define UV2H_DESC_DEST_TIMEOUT 2 +#define UV2H_DESC_DEST_STRONG_NACK 3 +#define UV2H_DESC_BUSY 4 +#define UV2H_DESC_SOURCE_TIMEOUT 6 +#define UV2H_DESC_DEST_PUT_ERR 7 /* * delay for 'plugged' timeout retries, in microseconds @@ -86,15 +124,24 @@ * threshholds at which to use IPI to free resources */ /* after this # consecutive 'plugged' timeouts, use IPI to release resources */ -#define PLUGSB4RESET 100 +#define PLUGSB4RESET 100 /* after this many consecutive timeouts, use IPI to release resources */ -#define TIMEOUTSB4RESET 1 +#define TIMEOUTSB4RESET 1 /* at this number uses of IPI to release resources, giveup the request */ -#define IPI_RESET_LIMIT 1 +#define IPI_RESET_LIMIT 1 /* after this # consecutive successes, bump up the throttle if it was lowered */ -#define COMPLETE_THRESHOLD 5 +#define COMPLETE_THRESHOLD 5 + +#define UV_LB_SUBNODEID 0x10 -#define UV_LB_SUBNODEID 0x10 +/* these two are the same for UV1 and UV2: */ +#define UV_SA_SHFT UVH_LB_BAU_MISC_CONTROL_INTD_SOFT_ACK_TIMEOUT_PERIOD_SHFT +#define UV_SA_MASK UVH_LB_BAU_MISC_CONTROL_INTD_SOFT_ACK_TIMEOUT_PERIOD_MASK +/* 4 bits of software ack period */ +#define UV2_ACK_MASK 0x7UL +#define UV2_ACK_UNITS_SHFT 3 +#define UV2_LEG_SHFT UV2H_LB_BAU_MISC_CONTROL_USE_LEGACY_DESCRIPTOR_FORMATS_SHFT +#define UV2_EXT_SHFT UV2H_LB_BAU_MISC_CONTROL_ENABLE_EXTENDED_SB_STATUS_SHFT /* * number of entries in the destination side payload queue @@ -115,9 +162,16 @@ /* * tuning the action when the numalink network is extremely delayed */ -#define CONGESTED_RESPONSE_US 1000 /* 'long' response time, in microseconds */ -#define CONGESTED_REPS 10 /* long delays averaged over this many broadcasts */ -#define CONGESTED_PERIOD 30 /* time for the bau to be disabled, in seconds */ +#define CONGESTED_RESPONSE_US 1000 /* 'long' response time, in + microseconds */ +#define CONGESTED_REPS 10 /* long delays averaged over + this many broadcasts */ +#define CONGESTED_PERIOD 30 /* time for the bau to be + disabled, in seconds */ +/* see msg_type: */ +#define MSG_NOOP 0 +#define MSG_REGULAR 1 +#define MSG_RETRY 2 /* * Distribution: 32 bytes (256 bits) (bytes 0-0x1f of descriptor) @@ -129,8 +183,8 @@ * 'base_dest_nasid' field of the header corresponds to the * destination nodeID associated with that specified bit. */ -struct bau_target_uvhubmask { - unsigned long bits[BITS_TO_LONGS(UV_DISTRIBUTION_SIZE)]; +struct bau_targ_hubmask { + unsigned long bits[BITS_TO_LONGS(UV_DISTRIBUTION_SIZE)]; }; /* @@ -139,7 +193,7 @@ struct bau_target_uvhubmask { * enough bits for max. cpu's per uvhub) */ struct bau_local_cpumask { - unsigned long bits; + unsigned long bits; }; /* @@ -160,14 +214,14 @@ struct bau_local_cpumask { * The payload is software-defined for INTD transactions */ struct bau_msg_payload { - unsigned long address; /* signifies a page or all TLB's - of the cpu */ + unsigned long address; /* signifies a page or all + TLB's of the cpu */ /* 64 bits */ - unsigned short sending_cpu; /* filled in by sender */ + unsigned short sending_cpu; /* filled in by sender */ /* 16 bits */ - unsigned short acknowledge_count;/* filled in by destination */ + unsigned short acknowledge_count; /* filled in by destination */ /* 16 bits */ - unsigned int reserved1:32; /* not usable */ + unsigned int reserved1:32; /* not usable */ }; @@ -176,93 +230,96 @@ struct bau_msg_payload { * see table 4.2.3.0.1 in broacast_assist spec. */ struct bau_msg_header { - unsigned int dest_subnodeid:6; /* must be 0x10, for the LB */ + unsigned int dest_subnodeid:6; /* must be 0x10, for the LB */ /* bits 5:0 */ - unsigned int base_dest_nasid:15; /* nasid of the */ - /* bits 20:6 */ /* first bit in uvhub map */ - unsigned int command:8; /* message type */ + unsigned int base_dest_nasid:15; /* nasid of the first bit */ + /* bits 20:6 */ /* in uvhub map */ + unsigned int command:8; /* message type */ /* bits 28:21 */ - /* 0x38: SN3net EndPoint Message */ - unsigned int rsvd_1:3; /* must be zero */ + /* 0x38: SN3net EndPoint Message */ + unsigned int rsvd_1:3; /* must be zero */ /* bits 31:29 */ - /* int will align on 32 bits */ - unsigned int rsvd_2:9; /* must be zero */ + /* int will align on 32 bits */ + unsigned int rsvd_2:9; /* must be zero */ /* bits 40:32 */ - /* Suppl_A is 56-41 */ - unsigned int sequence:16;/* message sequence number */ - /* bits 56:41 */ /* becomes bytes 16-17 of msg */ - /* Address field (96:57) is never used as an - address (these are address bits 42:3) */ - - unsigned int rsvd_3:1; /* must be zero */ + /* Suppl_A is 56-41 */ + unsigned int sequence:16; /* message sequence number */ + /* bits 56:41 */ /* becomes bytes 16-17 of msg */ + /* Address field (96:57) is + never used as an address + (these are address bits + 42:3) */ + + unsigned int rsvd_3:1; /* must be zero */ /* bit 57 */ - /* address bits 27:4 are payload */ + /* address bits 27:4 are payload */ /* these next 24 (58-81) bits become bytes 12-14 of msg */ - /* bits 65:58 land in byte 12 */ - unsigned int replied_to:1;/* sent as 0 by the source to byte 12 */ + unsigned int replied_to:1; /* sent as 0 by the source to + byte 12 */ /* bit 58 */ - unsigned int msg_type:3; /* software type of the message*/ + unsigned int msg_type:3; /* software type of the + message */ /* bits 61:59 */ - unsigned int canceled:1; /* message canceled, resource to be freed*/ + unsigned int canceled:1; /* message canceled, resource + is to be freed*/ /* bit 62 */ - unsigned int payload_1a:1;/* not currently used */ + unsigned int payload_1a:1; /* not currently used */ /* bit 63 */ - unsigned int payload_1b:2;/* not currently used */ + unsigned int payload_1b:2; /* not currently used */ /* bits 65:64 */ /* bits 73:66 land in byte 13 */ - unsigned int payload_1ca:6;/* not currently used */ + unsigned int payload_1ca:6; /* not currently used */ /* bits 71:66 */ - unsigned int payload_1c:2;/* not currently used */ + unsigned int payload_1c:2; /* not currently used */ /* bits 73:72 */ /* bits 81:74 land in byte 14 */ - unsigned int payload_1d:6;/* not currently used */ + unsigned int payload_1d:6; /* not currently used */ /* bits 79:74 */ - unsigned int payload_1e:2;/* not currently used */ + unsigned int payload_1e:2; /* not currently used */ /* bits 81:80 */ - unsigned int rsvd_4:7; /* must be zero */ + unsigned int rsvd_4:7; /* must be zero */ /* bits 88:82 */ - unsigned int sw_ack_flag:1;/* software acknowledge flag */ + unsigned int swack_flag:1; /* software acknowledge flag */ /* bit 89 */ - /* INTD trasactions at destination are to - wait for software acknowledge */ - unsigned int rsvd_5:6; /* must be zero */ + /* INTD trasactions at + destination are to wait for + software acknowledge */ + unsigned int rsvd_5:6; /* must be zero */ /* bits 95:90 */ - unsigned int rsvd_6:5; /* must be zero */ + unsigned int rsvd_6:5; /* must be zero */ /* bits 100:96 */ - unsigned int int_both:1;/* if 1, interrupt both sockets on the uvhub */ + unsigned int int_both:1; /* if 1, interrupt both sockets + on the uvhub */ /* bit 101*/ - unsigned int fairness:3;/* usually zero */ + unsigned int fairness:3; /* usually zero */ /* bits 104:102 */ - unsigned int multilevel:1; /* multi-level multicast format */ + unsigned int multilevel:1; /* multi-level multicast + format */ /* bit 105 */ - /* 0 for TLB: endpoint multi-unicast messages */ - unsigned int chaining:1;/* next descriptor is part of this activation*/ + /* 0 for TLB: endpoint multi-unicast messages */ + unsigned int chaining:1; /* next descriptor is part of + this activation*/ /* bit 106 */ - unsigned int rsvd_7:21; /* must be zero */ + unsigned int rsvd_7:21; /* must be zero */ /* bits 127:107 */ }; -/* see msg_type: */ -#define MSG_NOOP 0 -#define MSG_REGULAR 1 -#define MSG_RETRY 2 - /* * The activation descriptor: * The format of the message to send, plus all accompanying control * Should be 64 bytes */ struct bau_desc { - struct bau_target_uvhubmask distribution; + struct bau_targ_hubmask distribution; /* * message template, consisting of header and payload: */ - struct bau_msg_header header; - struct bau_msg_payload payload; + struct bau_msg_header header; + struct bau_msg_payload payload; }; /* * -payload-- ---------header------ @@ -281,59 +338,51 @@ struct bau_desc { * are 32 bytes (2 micropackets) (256 bits) in length, but contain only 17 * bytes of usable data, including the sw ack vector in byte 15 (bits 127:120) * (12 bytes come from bau_msg_payload, 3 from payload_1, 2 from - * sw_ack_vector and payload_2) + * swack_vec and payload_2) * "Enabling Software Acknowledgment mode (see Section 4.3.3 Software * Acknowledge Processing) also selects 32 byte (17 bytes usable) payload * operation." */ -struct bau_payload_queue_entry { - unsigned long address; /* signifies a page or all TLB's - of the cpu */ +struct bau_pq_entry { + unsigned long address; /* signifies a page or all TLB's + of the cpu */ /* 64 bits, bytes 0-7 */ - - unsigned short sending_cpu; /* cpu that sent the message */ + unsigned short sending_cpu; /* cpu that sent the message */ /* 16 bits, bytes 8-9 */ - - unsigned short acknowledge_count; /* filled in by destination */ + unsigned short acknowledge_count; /* filled in by destination */ /* 16 bits, bytes 10-11 */ - /* these next 3 bytes come from bits 58-81 of the message header */ - unsigned short replied_to:1; /* sent as 0 by the source */ - unsigned short msg_type:3; /* software message type */ - unsigned short canceled:1; /* sent as 0 by the source */ - unsigned short unused1:3; /* not currently using */ + unsigned short replied_to:1; /* sent as 0 by the source */ + unsigned short msg_type:3; /* software message type */ + unsigned short canceled:1; /* sent as 0 by the source */ + unsigned short unused1:3; /* not currently using */ /* byte 12 */ - - unsigned char unused2a; /* not currently using */ + unsigned char unused2a; /* not currently using */ /* byte 13 */ - unsigned char unused2; /* not currently using */ + unsigned char unused2; /* not currently using */ /* byte 14 */ - - unsigned char sw_ack_vector; /* filled in by the hardware */ + unsigned char swack_vec; /* filled in by the hardware */ /* byte 15 (bits 127:120) */ - - unsigned short sequence; /* message sequence number */ + unsigned short sequence; /* message sequence number */ /* bytes 16-17 */ - unsigned char unused4[2]; /* not currently using bytes 18-19 */ + unsigned char unused4[2]; /* not currently using bytes 18-19 */ /* bytes 18-19 */ - - int number_of_cpus; /* filled in at destination */ + int number_of_cpus; /* filled in at destination */ /* 32 bits, bytes 20-23 (aligned) */ - - unsigned char unused5[8]; /* not using */ + unsigned char unused5[8]; /* not using */ /* bytes 24-31 */ }; struct msg_desc { - struct bau_payload_queue_entry *msg; - int msg_slot; - int sw_ack_slot; - struct bau_payload_queue_entry *va_queue_first; - struct bau_payload_queue_entry *va_queue_last; + struct bau_pq_entry *msg; + int msg_slot; + int swack_slot; + struct bau_pq_entry *queue_first; + struct bau_pq_entry *queue_last; }; struct reset_args { - int sender; + int sender; }; /* @@ -341,112 +390,226 @@ struct reset_args { */ struct ptc_stats { /* sender statistics */ - unsigned long s_giveup; /* number of fall backs to IPI-style flushes */ - unsigned long s_requestor; /* number of shootdown requests */ - unsigned long s_stimeout; /* source side timeouts */ - unsigned long s_dtimeout; /* destination side timeouts */ - unsigned long s_time; /* time spent in sending side */ - unsigned long s_retriesok; /* successful retries */ - unsigned long s_ntargcpu; /* total number of cpu's targeted */ - unsigned long s_ntargself; /* times the sending cpu was targeted */ - unsigned long s_ntarglocals; /* targets of cpus on the local blade */ - unsigned long s_ntargremotes; /* targets of cpus on remote blades */ - unsigned long s_ntarglocaluvhub; /* targets of the local hub */ - unsigned long s_ntargremoteuvhub; /* remotes hubs targeted */ - unsigned long s_ntarguvhub; /* total number of uvhubs targeted */ - unsigned long s_ntarguvhub16; /* number of times target hubs >= 16*/ - unsigned long s_ntarguvhub8; /* number of times target hubs >= 8 */ - unsigned long s_ntarguvhub4; /* number of times target hubs >= 4 */ - unsigned long s_ntarguvhub2; /* number of times target hubs >= 2 */ - unsigned long s_ntarguvhub1; /* number of times target hubs == 1 */ - unsigned long s_resets_plug; /* ipi-style resets from plug state */ - unsigned long s_resets_timeout; /* ipi-style resets from timeouts */ - unsigned long s_busy; /* status stayed busy past s/w timer */ - unsigned long s_throttles; /* waits in throttle */ - unsigned long s_retry_messages; /* retry broadcasts */ - unsigned long s_bau_reenabled; /* for bau enable/disable */ - unsigned long s_bau_disabled; /* for bau enable/disable */ + unsigned long s_giveup; /* number of fall backs to + IPI-style flushes */ + unsigned long s_requestor; /* number of shootdown + requests */ + unsigned long s_stimeout; /* source side timeouts */ + unsigned long s_dtimeout; /* destination side timeouts */ + unsigned long s_time; /* time spent in sending side */ + unsigned long s_retriesok; /* successful retries */ + unsigned long s_ntargcpu; /* total number of cpu's + targeted */ + unsigned long s_ntargself; /* times the sending cpu was + targeted */ + unsigned long s_ntarglocals; /* targets of cpus on the local + blade */ + unsigned long s_ntargremotes; /* targets of cpus on remote + blades */ + unsigned long s_ntarglocaluvhub; /* targets of the local hub */ + unsigned long s_ntargremoteuvhub; /* remotes hubs targeted */ + unsigned long s_ntarguvhub; /* total number of uvhubs + targeted */ + unsigned long s_ntarguvhub16; /* number of times target + hubs >= 16*/ + unsigned long s_ntarguvhub8; /* number of times target + hubs >= 8 */ + unsigned long s_ntarguvhub4; /* number of times target + hubs >= 4 */ + unsigned long s_ntarguvhub2; /* number of times target + hubs >= 2 */ + unsigned long s_ntarguvhub1; /* number of times target + hubs == 1 */ + unsigned long s_resets_plug; /* ipi-style resets from plug + state */ + unsigned long s_resets_timeout; /* ipi-style resets from + timeouts */ + unsigned long s_busy; /* status stayed busy past + s/w timer */ + unsigned long s_throttles; /* waits in throttle */ + unsigned long s_retry_messages; /* retry broadcasts */ + unsigned long s_bau_reenabled; /* for bau enable/disable */ + unsigned long s_bau_disabled; /* for bau enable/disable */ /* destination statistics */ - unsigned long d_alltlb; /* times all tlb's on this cpu were flushed */ - unsigned long d_onetlb; /* times just one tlb on this cpu was flushed */ - unsigned long d_multmsg; /* interrupts with multiple messages */ - unsigned long d_nomsg; /* interrupts with no message */ - unsigned long d_time; /* time spent on destination side */ - unsigned long d_requestee; /* number of messages processed */ - unsigned long d_retries; /* number of retry messages processed */ - unsigned long d_canceled; /* number of messages canceled by retries */ - unsigned long d_nocanceled; /* retries that found nothing to cancel */ - unsigned long d_resets; /* number of ipi-style requests processed */ - unsigned long d_rcanceled; /* number of messages canceled by resets */ + unsigned long d_alltlb; /* times all tlb's on this + cpu were flushed */ + unsigned long d_onetlb; /* times just one tlb on this + cpu was flushed */ + unsigned long d_multmsg; /* interrupts with multiple + messages */ + unsigned long d_nomsg; /* interrupts with no message */ + unsigned long d_time; /* time spent on destination + side */ + unsigned long d_requestee; /* number of messages + processed */ + unsigned long d_retries; /* number of retry messages + processed */ + unsigned long d_canceled; /* number of messages canceled + by retries */ + unsigned long d_nocanceled; /* retries that found nothing + to cancel */ + unsigned long d_resets; /* number of ipi-style requests + processed */ + unsigned long d_rcanceled; /* number of messages canceled + by resets */ +}; + +struct tunables { + int *tunp; + int deflt; }; struct hub_and_pnode { - short uvhub; - short pnode; + short uvhub; + short pnode; }; + +struct socket_desc { + short num_cpus; + short cpu_number[MAX_CPUS_PER_SOCKET]; +}; + +struct uvhub_desc { + unsigned short socket_mask; + short num_cpus; + short uvhub; + short pnode; + struct socket_desc socket[2]; +}; + /* * one per-cpu; to locate the software tables */ struct bau_control { - struct bau_desc *descriptor_base; - struct bau_payload_queue_entry *va_queue_first; - struct bau_payload_queue_entry *va_queue_last; - struct bau_payload_queue_entry *bau_msg_head; - struct bau_control *uvhub_master; - struct bau_control *socket_master; - struct ptc_stats *statp; - unsigned long timeout_interval; - unsigned long set_bau_on_time; - atomic_t active_descriptor_count; - int plugged_tries; - int timeout_tries; - int ipi_attempts; - int conseccompletes; - int baudisabled; - int set_bau_off; - short cpu; - short osnode; - short uvhub_cpu; - short uvhub; - short cpus_in_socket; - short cpus_in_uvhub; - short partition_base_pnode; - unsigned short message_number; - unsigned short uvhub_quiesce; - short socket_acknowledge_count[DEST_Q_SIZE]; - cycles_t send_message; - spinlock_t uvhub_lock; - spinlock_t queue_lock; + struct bau_desc *descriptor_base; + struct bau_pq_entry *queue_first; + struct bau_pq_entry *queue_last; + struct bau_pq_entry *bau_msg_head; + struct bau_control *uvhub_master; + struct bau_control *socket_master; + struct ptc_stats *statp; + unsigned long timeout_interval; + unsigned long set_bau_on_time; + atomic_t active_descriptor_count; + int plugged_tries; + int timeout_tries; + int ipi_attempts; + int conseccompletes; + int baudisabled; + int set_bau_off; + short cpu; + short osnode; + short uvhub_cpu; + short uvhub; + short cpus_in_socket; + short cpus_in_uvhub; + short partition_base_pnode; + unsigned short message_number; + unsigned short uvhub_quiesce; + short socket_acknowledge_count[DEST_Q_SIZE]; + cycles_t send_message; + spinlock_t uvhub_lock; + spinlock_t queue_lock; /* tunables */ - int max_bau_concurrent; - int max_bau_concurrent_constant; - int plugged_delay; - int plugsb4reset; - int timeoutsb4reset; - int ipi_reset_limit; - int complete_threshold; - int congested_response_us; - int congested_reps; - int congested_period; - cycles_t period_time; - long period_requests; - struct hub_and_pnode *target_hub_and_pnode; + int max_concurr; + int max_concurr_const; + int plugged_delay; + int plugsb4reset; + int timeoutsb4reset; + int ipi_reset_limit; + int complete_threshold; + int cong_response_us; + int cong_reps; + int cong_period; + cycles_t period_time; + long period_requests; + struct hub_and_pnode *thp; }; -static inline int bau_uvhub_isset(int uvhub, struct bau_target_uvhubmask *dstp) +static unsigned long read_mmr_uv2_status(void) +{ + return read_lmmr(UV2H_LB_BAU_SB_ACTIVATION_STATUS_2); +} + +static void write_mmr_data_broadcast(int pnode, unsigned long mmr_image) +{ + write_gmmr(pnode, UVH_BAU_DATA_BROADCAST, mmr_image); +} + +static void write_mmr_descriptor_base(int pnode, unsigned long mmr_image) +{ + write_gmmr(pnode, UVH_LB_BAU_SB_DESCRIPTOR_BASE, mmr_image); +} + +static void write_mmr_activation(unsigned long index) +{ + write_lmmr(UVH_LB_BAU_SB_ACTIVATION_CONTROL, index); +} + +static void write_gmmr_activation(int pnode, unsigned long mmr_image) +{ + write_gmmr(pnode, UVH_LB_BAU_SB_ACTIVATION_CONTROL, mmr_image); +} + +static void write_mmr_payload_first(int pnode, unsigned long mmr_image) +{ + write_gmmr(pnode, UVH_LB_BAU_INTD_PAYLOAD_QUEUE_FIRST, mmr_image); +} + +static void write_mmr_payload_tail(int pnode, unsigned long mmr_image) +{ + write_gmmr(pnode, UVH_LB_BAU_INTD_PAYLOAD_QUEUE_TAIL, mmr_image); +} + +static void write_mmr_payload_last(int pnode, unsigned long mmr_image) +{ + write_gmmr(pnode, UVH_LB_BAU_INTD_PAYLOAD_QUEUE_LAST, mmr_image); +} + +static void write_mmr_misc_control(int pnode, unsigned long mmr_image) +{ + write_gmmr(pnode, UVH_LB_BAU_MISC_CONTROL, mmr_image); +} + +static unsigned long read_mmr_misc_control(int pnode) +{ + return read_gmmr(pnode, UVH_LB_BAU_MISC_CONTROL); +} + +static void write_mmr_sw_ack(unsigned long mr) +{ + uv_write_local_mmr(UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_ALIAS, mr); +} + +static unsigned long read_mmr_sw_ack(void) +{ + return read_lmmr(UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE); +} + +static unsigned long read_gmmr_sw_ack(int pnode) +{ + return read_gmmr(pnode, UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE); +} + +static void write_mmr_data_config(int pnode, unsigned long mr) +{ + uv_write_global_mmr64(pnode, UVH_BAU_DATA_CONFIG, mr); +} + +static inline int bau_uvhub_isset(int uvhub, struct bau_targ_hubmask *dstp) { return constant_test_bit(uvhub, &dstp->bits[0]); } -static inline void bau_uvhub_set(int pnode, struct bau_target_uvhubmask *dstp) +static inline void bau_uvhub_set(int pnode, struct bau_targ_hubmask *dstp) { __set_bit(pnode, &dstp->bits[0]); } -static inline void bau_uvhubs_clear(struct bau_target_uvhubmask *dstp, +static inline void bau_uvhubs_clear(struct bau_targ_hubmask *dstp, int nbits) { bitmap_zero(&dstp->bits[0], nbits); } -static inline int bau_uvhub_weight(struct bau_target_uvhubmask *dstp) +static inline int bau_uvhub_weight(struct bau_targ_hubmask *dstp) { return bitmap_weight((unsigned long *)&dstp->bits[0], UV_DISTRIBUTION_SIZE); @@ -457,9 +620,6 @@ static inline void bau_cpubits_clear(struct bau_local_cpumask *dstp, int nbits) bitmap_zero(&dstp->bits, nbits); } -#define cpubit_isset(cpu, bau_local_cpumask) \ - test_bit((cpu), (bau_local_cpumask).bits) - extern void uv_bau_message_intr1(void); extern void uv_bau_timeout_intr1(void); @@ -467,7 +627,7 @@ struct atomic_short { short counter; }; -/** +/* * atomic_read_short - read a short atomic variable * @v: pointer of type atomic_short * @@ -478,14 +638,14 @@ static inline int atomic_read_short(const struct atomic_short *v) return v->counter; } -/** - * atomic_add_short_return - add and return a short int +/* + * atom_asr - add and return a short int * @i: short value to add * @v: pointer of type atomic_short * * Atomically adds @i to @v and returns @i + @v */ -static inline int atomic_add_short_return(short i, struct atomic_short *v) +static inline int atom_asr(short i, struct atomic_short *v) { short __i = i; asm volatile(LOCK_PREFIX "xaddw %0, %1" @@ -494,4 +654,26 @@ static inline int atomic_add_short_return(short i, struct atomic_short *v) return i + __i; } +/* + * conditionally add 1 to *v, unless *v is >= u + * return 0 if we cannot add 1 to *v because it is >= u + * return 1 if we can add 1 to *v because it is < u + * the add is atomic + * + * This is close to atomic_add_unless(), but this allows the 'u' value + * to be lowered below the current 'v'. atomic_add_unless can only stop + * on equal. + */ +static inline int atomic_inc_unless_ge(spinlock_t *lock, atomic_t *v, int u) +{ + spin_lock(lock); + if (atomic_read(v) >= u) { + spin_unlock(lock); + return 0; + } + atomic_inc(v); + spin_unlock(lock); + return 1; +} + #endif /* _ASM_X86_UV_UV_BAU_H */ diff --git a/arch/x86/include/asm/uv/uv_hub.h b/arch/x86/include/asm/uv/uv_hub.h index 4298002d0c8..f26544a1521 100644 --- a/arch/x86/include/asm/uv/uv_hub.h +++ b/arch/x86/include/asm/uv/uv_hub.h @@ -77,8 +77,9 @@ * * 1111110000000000 * 5432109876543210 - * pppppppppplc0cch Nehalem-EX - * ppppppppplcc0cch Westmere-EX + * pppppppppplc0cch Nehalem-EX (12 bits in hdw reg) + * ppppppppplcc0cch Westmere-EX (12 bits in hdw reg) + * pppppppppppcccch SandyBridge (15 bits in hdw reg) * sssssssssss * * p = pnode bits @@ -87,7 +88,7 @@ * h = hyperthread * s = bits that are in the SOCKET_ID CSR * - * Note: Processor only supports 12 bits in the APICID register. The ACPI + * Note: Processor may support fewer bits in the APICID register. The ACPI * tables hold all 16 bits. Software needs to be aware of this. * * Unless otherwise specified, all references to APICID refer to @@ -138,6 +139,8 @@ struct uv_hub_info_s { unsigned long global_mmr_base; unsigned long gpa_mask; unsigned int gnode_extra; + unsigned char hub_revision; + unsigned char apic_pnode_shift; unsigned long gnode_upper; unsigned long lowmem_remap_top; unsigned long lowmem_remap_base; @@ -149,13 +152,31 @@ struct uv_hub_info_s { unsigned char m_val; unsigned char n_val; struct uv_scir_s scir; - unsigned char apic_pnode_shift; }; DECLARE_PER_CPU(struct uv_hub_info_s, __uv_hub_info); #define uv_hub_info (&__get_cpu_var(__uv_hub_info)) #define uv_cpu_hub_info(cpu) (&per_cpu(__uv_hub_info, cpu)) +/* + * Hub revisions less than UV2_HUB_REVISION_BASE are UV1 hubs. All UV2 + * hubs have revision numbers greater than or equal to UV2_HUB_REVISION_BASE. + * This is a software convention - NOT the hardware revision numbers in + * the hub chip. + */ +#define UV1_HUB_REVISION_BASE 1 +#define UV2_HUB_REVISION_BASE 3 + +static inline int is_uv1_hub(void) +{ + return uv_hub_info->hub_revision < UV2_HUB_REVISION_BASE; +} + +static inline int is_uv2_hub(void) +{ + return uv_hub_info->hub_revision >= UV2_HUB_REVISION_BASE; +} + union uvh_apicid { unsigned long v; struct uvh_apicid_s { @@ -180,11 +201,25 @@ union uvh_apicid { #define UV_PNODE_TO_GNODE(p) ((p) |uv_hub_info->gnode_extra) #define UV_PNODE_TO_NASID(p) (UV_PNODE_TO_GNODE(p) << 1) -#define UV_LOCAL_MMR_BASE 0xf4000000UL -#define UV_GLOBAL_MMR32_BASE 0xf8000000UL +#define UV1_LOCAL_MMR_BASE 0xf4000000UL +#define UV1_GLOBAL_MMR32_BASE 0xf8000000UL +#define UV1_LOCAL_MMR_SIZE (64UL * 1024 * 1024) +#define UV1_GLOBAL_MMR32_SIZE (64UL * 1024 * 1024) + +#define UV2_LOCAL_MMR_BASE 0xfa000000UL +#define UV2_GLOBAL_MMR32_BASE 0xfc000000UL +#define UV2_LOCAL_MMR_SIZE (32UL * 1024 * 1024) +#define UV2_GLOBAL_MMR32_SIZE (32UL * 1024 * 1024) + +#define UV_LOCAL_MMR_BASE (is_uv1_hub() ? UV1_LOCAL_MMR_BASE \ + : UV2_LOCAL_MMR_BASE) +#define UV_GLOBAL_MMR32_BASE (is_uv1_hub() ? UV1_GLOBAL_MMR32_BASE \ + : UV2_GLOBAL_MMR32_BASE) +#define UV_LOCAL_MMR_SIZE (is_uv1_hub() ? UV1_LOCAL_MMR_SIZE : \ + UV2_LOCAL_MMR_SIZE) +#define UV_GLOBAL_MMR32_SIZE (is_uv1_hub() ? UV1_GLOBAL_MMR32_SIZE :\ + UV2_GLOBAL_MMR32_SIZE) #define UV_GLOBAL_MMR64_BASE (uv_hub_info->global_mmr_base) -#define UV_LOCAL_MMR_SIZE (64UL * 1024 * 1024) -#define UV_GLOBAL_MMR32_SIZE (64UL * 1024 * 1024) #define UV_GLOBAL_GRU_MMR_BASE 0x4000000 @@ -301,6 +336,17 @@ static inline int uv_apicid_to_pnode(int apicid) } /* + * Convert an apicid to the socket number on the blade + */ +static inline int uv_apicid_to_socket(int apicid) +{ + if (is_uv1_hub()) + return (apicid >> (uv_hub_info->apic_pnode_shift - 1)) & 1; + else + return 0; +} + +/* * Access global MMRs using the low memory MMR32 space. This region supports * faster MMR access but not all MMRs are accessible in this space. */ @@ -519,14 +565,13 @@ static inline void uv_hub_send_ipi(int pnode, int apicid, int vector) /* * Get the minimum revision number of the hub chips within the partition. - * 1 - initial rev 1.0 silicon - * 2 - rev 2.0 production silicon + * 1 - UV1 rev 1.0 initial silicon + * 2 - UV1 rev 2.0 production silicon + * 3 - UV2 rev 1.0 initial silicon */ static inline int uv_get_min_hub_revision_id(void) { - extern int uv_min_hub_revision_id; - - return uv_min_hub_revision_id; + return uv_hub_info->hub_revision; } #endif /* CONFIG_X86_64 */ diff --git a/arch/x86/include/asm/uv/uv_mmrs.h b/arch/x86/include/asm/uv/uv_mmrs.h index f5bb64a823d..4be52c86344 100644 --- a/arch/x86/include/asm/uv/uv_mmrs.h +++ b/arch/x86/include/asm/uv/uv_mmrs.h @@ -11,13 +11,64 @@ #ifndef _ASM_X86_UV_UV_MMRS_H #define _ASM_X86_UV_UV_MMRS_H +/* + * This file contains MMR definitions for both UV1 & UV2 hubs. + * + * In general, MMR addresses and structures are identical on both hubs. + * These MMRs are identified as: + * #define UVH_xxx <address> + * union uvh_xxx { + * unsigned long v; + * struct uvh_int_cmpd_s { + * } s; + * }; + * + * If the MMR exists on both hub type but has different addresses or + * contents, the MMR definition is similar to: + * #define UV1H_xxx <uv1 address> + * #define UV2H_xxx <uv2address> + * #define UVH_xxx (is_uv1_hub() ? UV1H_xxx : UV2H_xxx) + * union uvh_xxx { + * unsigned long v; + * struct uv1h_int_cmpd_s { (Common fields only) + * } s; + * struct uv1h_int_cmpd_s { (Full UV1 definition) + * } s1; + * struct uv2h_int_cmpd_s { (Full UV2 definition) + * } s2; + * }; + * + * Only essential difference are enumerated. For example, if the address is + * the same for both UV1 & UV2, only a single #define is generated. Likewise, + * if the contents is the same for both hubs, only the "s" structure is + * generated. + * + * If the MMR exists on ONLY 1 type of hub, no generic definition is + * generated: + * #define UVnH_xxx <uvn address> + * union uvnh_xxx { + * unsigned long v; + * struct uvh_int_cmpd_s { + * } sn; + * }; + */ + #define UV_MMR_ENABLE (1UL << 63) +#define UV1_HUB_PART_NUMBER 0x88a5 +#define UV2_HUB_PART_NUMBER 0x8eb8 + +/* Compat: if this #define is present, UV headers support UV2 */ +#define UV2_HUB_IS_SUPPORTED 1 + +/* KABI compat: if this #define is present, KABI hacks are present */ +#define UV2_HUB_KABI_HACKS 1 + /* ========================================================================= */ /* UVH_BAU_DATA_BROADCAST */ /* ========================================================================= */ #define UVH_BAU_DATA_BROADCAST 0x61688UL -#define UVH_BAU_DATA_BROADCAST_32 0x0440 +#define UVH_BAU_DATA_BROADCAST_32 0x440 #define UVH_BAU_DATA_BROADCAST_ENABLE_SHFT 0 #define UVH_BAU_DATA_BROADCAST_ENABLE_MASK 0x0000000000000001UL @@ -34,7 +85,7 @@ union uvh_bau_data_broadcast_u { /* UVH_BAU_DATA_CONFIG */ /* ========================================================================= */ #define UVH_BAU_DATA_CONFIG 0x61680UL -#define UVH_BAU_DATA_CONFIG_32 0x0438 +#define UVH_BAU_DATA_CONFIG_32 0x438 #define UVH_BAU_DATA_CONFIG_VECTOR_SHFT 0 #define UVH_BAU_DATA_CONFIG_VECTOR_MASK 0x00000000000000ffUL @@ -73,125 +124,245 @@ union uvh_bau_data_config_u { /* UVH_EVENT_OCCURRED0 */ /* ========================================================================= */ #define UVH_EVENT_OCCURRED0 0x70000UL -#define UVH_EVENT_OCCURRED0_32 0x005e8 - -#define UVH_EVENT_OCCURRED0_LB_HCERR_SHFT 0 -#define UVH_EVENT_OCCURRED0_LB_HCERR_MASK 0x0000000000000001UL -#define UVH_EVENT_OCCURRED0_GR0_HCERR_SHFT 1 -#define UVH_EVENT_OCCURRED0_GR0_HCERR_MASK 0x0000000000000002UL -#define UVH_EVENT_OCCURRED0_GR1_HCERR_SHFT 2 -#define UVH_EVENT_OCCURRED0_GR1_HCERR_MASK 0x0000000000000004UL -#define UVH_EVENT_OCCURRED0_LH_HCERR_SHFT 3 -#define UVH_EVENT_OCCURRED0_LH_HCERR_MASK 0x0000000000000008UL -#define UVH_EVENT_OCCURRED0_RH_HCERR_SHFT 4 -#define UVH_EVENT_OCCURRED0_RH_HCERR_MASK 0x0000000000000010UL -#define UVH_EVENT_OCCURRED0_XN_HCERR_SHFT 5 -#define UVH_EVENT_OCCURRED0_XN_HCERR_MASK 0x0000000000000020UL -#define UVH_EVENT_OCCURRED0_SI_HCERR_SHFT 6 -#define UVH_EVENT_OCCURRED0_SI_HCERR_MASK 0x0000000000000040UL -#define UVH_EVENT_OCCURRED0_LB_AOERR0_SHFT 7 -#define UVH_EVENT_OCCURRED0_LB_AOERR0_MASK 0x0000000000000080UL -#define UVH_EVENT_OCCURRED0_GR0_AOERR0_SHFT 8 -#define UVH_EVENT_OCCURRED0_GR0_AOERR0_MASK 0x0000000000000100UL -#define UVH_EVENT_OCCURRED0_GR1_AOERR0_SHFT 9 -#define UVH_EVENT_OCCURRED0_GR1_AOERR0_MASK 0x0000000000000200UL -#define UVH_EVENT_OCCURRED0_LH_AOERR0_SHFT 10 -#define UVH_EVENT_OCCURRED0_LH_AOERR0_MASK 0x0000000000000400UL -#define UVH_EVENT_OCCURRED0_RH_AOERR0_SHFT 11 -#define UVH_EVENT_OCCURRED0_RH_AOERR0_MASK 0x0000000000000800UL -#define UVH_EVENT_OCCURRED0_XN_AOERR0_SHFT 12 -#define UVH_EVENT_OCCURRED0_XN_AOERR0_MASK 0x0000000000001000UL -#define UVH_EVENT_OCCURRED0_SI_AOERR0_SHFT 13 -#define UVH_EVENT_OCCURRED0_SI_AOERR0_MASK 0x0000000000002000UL -#define UVH_EVENT_OCCURRED0_LB_AOERR1_SHFT 14 -#define UVH_EVENT_OCCURRED0_LB_AOERR1_MASK 0x0000000000004000UL -#define UVH_EVENT_OCCURRED0_GR0_AOERR1_SHFT 15 -#define UVH_EVENT_OCCURRED0_GR0_AOERR1_MASK 0x0000000000008000UL -#define UVH_EVENT_OCCURRED0_GR1_AOERR1_SHFT 16 -#define UVH_EVENT_OCCURRED0_GR1_AOERR1_MASK 0x0000000000010000UL -#define UVH_EVENT_OCCURRED0_LH_AOERR1_SHFT 17 -#define UVH_EVENT_OCCURRED0_LH_AOERR1_MASK 0x0000000000020000UL -#define UVH_EVENT_OCCURRED0_RH_AOERR1_SHFT 18 -#define UVH_EVENT_OCCURRED0_RH_AOERR1_MASK 0x0000000000040000UL -#define UVH_EVENT_OCCURRED0_XN_AOERR1_SHFT 19 -#define UVH_EVENT_OCCURRED0_XN_AOERR1_MASK 0x0000000000080000UL -#define UVH_EVENT_OCCURRED0_SI_AOERR1_SHFT 20 -#define UVH_EVENT_OCCURRED0_SI_AOERR1_MASK 0x0000000000100000UL -#define UVH_EVENT_OCCURRED0_RH_VPI_INT_SHFT 21 -#define UVH_EVENT_OCCURRED0_RH_VPI_INT_MASK 0x0000000000200000UL -#define UVH_EVENT_OCCURRED0_SYSTEM_SHUTDOWN_INT_SHFT 22 -#define UVH_EVENT_OCCURRED0_SYSTEM_SHUTDOWN_INT_MASK 0x0000000000400000UL -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_0_SHFT 23 -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_0_MASK 0x0000000000800000UL -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_1_SHFT 24 -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_1_MASK 0x0000000001000000UL -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_2_SHFT 25 -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_2_MASK 0x0000000002000000UL -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_3_SHFT 26 -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_3_MASK 0x0000000004000000UL -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_4_SHFT 27 -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_4_MASK 0x0000000008000000UL -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_5_SHFT 28 -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_5_MASK 0x0000000010000000UL -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_6_SHFT 29 -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_6_MASK 0x0000000020000000UL -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_7_SHFT 30 -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_7_MASK 0x0000000040000000UL -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_8_SHFT 31 -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_8_MASK 0x0000000080000000UL -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_9_SHFT 32 -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_9_MASK 0x0000000100000000UL -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_10_SHFT 33 -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_10_MASK 0x0000000200000000UL -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_11_SHFT 34 -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_11_MASK 0x0000000400000000UL -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_12_SHFT 35 -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_12_MASK 0x0000000800000000UL -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_13_SHFT 36 -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_13_MASK 0x0000001000000000UL -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_14_SHFT 37 -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_14_MASK 0x0000002000000000UL -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_15_SHFT 38 -#define UVH_EVENT_OCCURRED0_LB_IRQ_INT_15_MASK 0x0000004000000000UL -#define UVH_EVENT_OCCURRED0_L1_NMI_INT_SHFT 39 -#define UVH_EVENT_OCCURRED0_L1_NMI_INT_MASK 0x0000008000000000UL -#define UVH_EVENT_OCCURRED0_STOP_CLOCK_SHFT 40 -#define UVH_EVENT_OCCURRED0_STOP_CLOCK_MASK 0x0000010000000000UL -#define UVH_EVENT_OCCURRED0_ASIC_TO_L1_SHFT 41 -#define UVH_EVENT_OCCURRED0_ASIC_TO_L1_MASK 0x0000020000000000UL -#define UVH_EVENT_OCCURRED0_L1_TO_ASIC_SHFT 42 -#define UVH_EVENT_OCCURRED0_L1_TO_ASIC_MASK 0x0000040000000000UL -#define UVH_EVENT_OCCURRED0_LTC_INT_SHFT 43 -#define UVH_EVENT_OCCURRED0_LTC_INT_MASK 0x0000080000000000UL -#define UVH_EVENT_OCCURRED0_LA_SEQ_TRIGGER_SHFT 44 -#define UVH_EVENT_OCCURRED0_LA_SEQ_TRIGGER_MASK 0x0000100000000000UL -#define UVH_EVENT_OCCURRED0_IPI_INT_SHFT 45 -#define UVH_EVENT_OCCURRED0_IPI_INT_MASK 0x0000200000000000UL -#define UVH_EVENT_OCCURRED0_EXTIO_INT0_SHFT 46 -#define UVH_EVENT_OCCURRED0_EXTIO_INT0_MASK 0x0000400000000000UL -#define UVH_EVENT_OCCURRED0_EXTIO_INT1_SHFT 47 -#define UVH_EVENT_OCCURRED0_EXTIO_INT1_MASK 0x0000800000000000UL -#define UVH_EVENT_OCCURRED0_EXTIO_INT2_SHFT 48 -#define UVH_EVENT_OCCURRED0_EXTIO_INT2_MASK 0x0001000000000000UL -#define UVH_EVENT_OCCURRED0_EXTIO_INT3_SHFT 49 -#define UVH_EVENT_OCCURRED0_EXTIO_INT3_MASK 0x0002000000000000UL -#define UVH_EVENT_OCCURRED0_PROFILE_INT_SHFT 50 -#define UVH_EVENT_OCCURRED0_PROFILE_INT_MASK 0x0004000000000000UL -#define UVH_EVENT_OCCURRED0_RTC0_SHFT 51 -#define UVH_EVENT_OCCURRED0_RTC0_MASK 0x0008000000000000UL -#define UVH_EVENT_OCCURRED0_RTC1_SHFT 52 -#define UVH_EVENT_OCCURRED0_RTC1_MASK 0x0010000000000000UL -#define UVH_EVENT_OCCURRED0_RTC2_SHFT 53 -#define UVH_EVENT_OCCURRED0_RTC2_MASK 0x0020000000000000UL -#define UVH_EVENT_OCCURRED0_RTC3_SHFT 54 -#define UVH_EVENT_OCCURRED0_RTC3_MASK 0x0040000000000000UL -#define UVH_EVENT_OCCURRED0_BAU_DATA_SHFT 55 -#define UVH_EVENT_OCCURRED0_BAU_DATA_MASK 0x0080000000000000UL -#define UVH_EVENT_OCCURRED0_POWER_MANAGEMENT_REQ_SHFT 56 -#define UVH_EVENT_OCCURRED0_POWER_MANAGEMENT_REQ_MASK 0x0100000000000000UL +#define UVH_EVENT_OCCURRED0_32 0x5e8 + +#define UV1H_EVENT_OCCURRED0_LB_HCERR_SHFT 0 +#define UV1H_EVENT_OCCURRED0_LB_HCERR_MASK 0x0000000000000001UL +#define UV1H_EVENT_OCCURRED0_GR0_HCERR_SHFT 1 +#define UV1H_EVENT_OCCURRED0_GR0_HCERR_MASK 0x0000000000000002UL +#define UV1H_EVENT_OCCURRED0_GR1_HCERR_SHFT 2 +#define UV1H_EVENT_OCCURRED0_GR1_HCERR_MASK 0x0000000000000004UL +#define UV1H_EVENT_OCCURRED0_LH_HCERR_SHFT 3 +#define UV1H_EVENT_OCCURRED0_LH_HCERR_MASK 0x0000000000000008UL +#define UV1H_EVENT_OCCURRED0_RH_HCERR_SHFT 4 +#define UV1H_EVENT_OCCURRED0_RH_HCERR_MASK 0x0000000000000010UL +#define UV1H_EVENT_OCCURRED0_XN_HCERR_SHFT 5 +#define UV1H_EVENT_OCCURRED0_XN_HCERR_MASK 0x0000000000000020UL +#define UV1H_EVENT_OCCURRED0_SI_HCERR_SHFT 6 +#define UV1H_EVENT_OCCURRED0_SI_HCERR_MASK 0x0000000000000040UL +#define UV1H_EVENT_OCCURRED0_LB_AOERR0_SHFT 7 +#define UV1H_EVENT_OCCURRED0_LB_AOERR0_MASK 0x0000000000000080UL +#define UV1H_EVENT_OCCURRED0_GR0_AOERR0_SHFT 8 +#define UV1H_EVENT_OCCURRED0_GR0_AOERR0_MASK 0x0000000000000100UL +#define UV1H_EVENT_OCCURRED0_GR1_AOERR0_SHFT 9 +#define UV1H_EVENT_OCCURRED0_GR1_AOERR0_MASK 0x0000000000000200UL +#define UV1H_EVENT_OCCURRED0_LH_AOERR0_SHFT 10 +#define UV1H_EVENT_OCCURRED0_LH_AOERR0_MASK 0x0000000000000400UL +#define UV1H_EVENT_OCCURRED0_RH_AOERR0_SHFT 11 +#define UV1H_EVENT_OCCURRED0_RH_AOERR0_MASK 0x0000000000000800UL +#define UV1H_EVENT_OCCURRED0_XN_AOERR0_SHFT 12 +#define UV1H_EVENT_OCCURRED0_XN_AOERR0_MASK 0x0000000000001000UL +#define UV1H_EVENT_OCCURRED0_SI_AOERR0_SHFT 13 +#define UV1H_EVENT_OCCURRED0_SI_AOERR0_MASK 0x0000000000002000UL +#define UV1H_EVENT_OCCURRED0_LB_AOERR1_SHFT 14 +#define UV1H_EVENT_OCCURRED0_LB_AOERR1_MASK 0x0000000000004000UL +#define UV1H_EVENT_OCCURRED0_GR0_AOERR1_SHFT 15 +#define UV1H_EVENT_OCCURRED0_GR0_AOERR1_MASK 0x0000000000008000UL +#define UV1H_EVENT_OCCURRED0_GR1_AOERR1_SHFT 16 +#define UV1H_EVENT_OCCURRED0_GR1_AOERR1_MASK 0x0000000000010000UL +#define UV1H_EVENT_OCCURRED0_LH_AOERR1_SHFT 17 +#define UV1H_EVENT_OCCURRED0_LH_AOERR1_MASK 0x0000000000020000UL +#define UV1H_EVENT_OCCURRED0_RH_AOERR1_SHFT 18 +#define UV1H_EVENT_OCCURRED0_RH_AOERR1_MASK 0x0000000000040000UL +#define UV1H_EVENT_OCCURRED0_XN_AOERR1_SHFT 19 +#define UV1H_EVENT_OCCURRED0_XN_AOERR1_MASK 0x0000000000080000UL +#define UV1H_EVENT_OCCURRED0_SI_AOERR1_SHFT 20 +#define UV1H_EVENT_OCCURRED0_SI_AOERR1_MASK 0x0000000000100000UL +#define UV1H_EVENT_OCCURRED0_RH_VPI_INT_SHFT 21 +#define UV1H_EVENT_OCCURRED0_RH_VPI_INT_MASK 0x0000000000200000UL +#define UV1H_EVENT_OCCURRED0_SYSTEM_SHUTDOWN_INT_SHFT 22 +#define UV1H_EVENT_OCCURRED0_SYSTEM_SHUTDOWN_INT_MASK 0x0000000000400000UL +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_0_SHFT 23 +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_0_MASK 0x0000000000800000UL +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_1_SHFT 24 +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_1_MASK 0x0000000001000000UL +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_2_SHFT 25 +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_2_MASK 0x0000000002000000UL +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_3_SHFT 26 +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_3_MASK 0x0000000004000000UL +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_4_SHFT 27 +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_4_MASK 0x0000000008000000UL +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_5_SHFT 28 +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_5_MASK 0x0000000010000000UL +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_6_SHFT 29 +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_6_MASK 0x0000000020000000UL +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_7_SHFT 30 +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_7_MASK 0x0000000040000000UL +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_8_SHFT 31 +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_8_MASK 0x0000000080000000UL +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_9_SHFT 32 +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_9_MASK 0x0000000100000000UL +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_10_SHFT 33 +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_10_MASK 0x0000000200000000UL +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_11_SHFT 34 +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_11_MASK 0x0000000400000000UL +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_12_SHFT 35 +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_12_MASK 0x0000000800000000UL +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_13_SHFT 36 +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_13_MASK 0x0000001000000000UL +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_14_SHFT 37 +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_14_MASK 0x0000002000000000UL +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_15_SHFT 38 +#define UV1H_EVENT_OCCURRED0_LB_IRQ_INT_15_MASK 0x0000004000000000UL +#define UV1H_EVENT_OCCURRED0_L1_NMI_INT_SHFT 39 +#define UV1H_EVENT_OCCURRED0_L1_NMI_INT_MASK 0x0000008000000000UL +#define UV1H_EVENT_OCCURRED0_STOP_CLOCK_SHFT 40 +#define UV1H_EVENT_OCCURRED0_STOP_CLOCK_MASK 0x0000010000000000UL +#define UV1H_EVENT_OCCURRED0_ASIC_TO_L1_SHFT 41 +#define UV1H_EVENT_OCCURRED0_ASIC_TO_L1_MASK 0x0000020000000000UL +#define UV1H_EVENT_OCCURRED0_L1_TO_ASIC_SHFT 42 +#define UV1H_EVENT_OCCURRED0_L1_TO_ASIC_MASK 0x0000040000000000UL +#define UV1H_EVENT_OCCURRED0_LTC_INT_SHFT 43 +#define UV1H_EVENT_OCCURRED0_LTC_INT_MASK 0x0000080000000000UL +#define UV1H_EVENT_OCCURRED0_LA_SEQ_TRIGGER_SHFT 44 +#define UV1H_EVENT_OCCURRED0_LA_SEQ_TRIGGER_MASK 0x0000100000000000UL +#define UV1H_EVENT_OCCURRED0_IPI_INT_SHFT 45 +#define UV1H_EVENT_OCCURRED0_IPI_INT_MASK 0x0000200000000000UL +#define UV1H_EVENT_OCCURRED0_EXTIO_INT0_SHFT 46 +#define UV1H_EVENT_OCCURRED0_EXTIO_INT0_MASK 0x0000400000000000UL +#define UV1H_EVENT_OCCURRED0_EXTIO_INT1_SHFT 47 +#define UV1H_EVENT_OCCURRED0_EXTIO_INT1_MASK 0x0000800000000000UL +#define UV1H_EVENT_OCCURRED0_EXTIO_INT2_SHFT 48 +#define UV1H_EVENT_OCCURRED0_EXTIO_INT2_MASK 0x0001000000000000UL +#define UV1H_EVENT_OCCURRED0_EXTIO_INT3_SHFT 49 +#define UV1H_EVENT_OCCURRED0_EXTIO_INT3_MASK 0x0002000000000000UL +#define UV1H_EVENT_OCCURRED0_PROFILE_INT_SHFT 50 +#define UV1H_EVENT_OCCURRED0_PROFILE_INT_MASK 0x0004000000000000UL +#define UV1H_EVENT_OCCURRED0_RTC0_SHFT 51 +#define UV1H_EVENT_OCCURRED0_RTC0_MASK 0x0008000000000000UL +#define UV1H_EVENT_OCCURRED0_RTC1_SHFT 52 +#define UV1H_EVENT_OCCURRED0_RTC1_MASK 0x0010000000000000UL +#define UV1H_EVENT_OCCURRED0_RTC2_SHFT 53 +#define UV1H_EVENT_OCCURRED0_RTC2_MASK 0x0020000000000000UL +#define UV1H_EVENT_OCCURRED0_RTC3_SHFT 54 +#define UV1H_EVENT_OCCURRED0_RTC3_MASK 0x0040000000000000UL +#define UV1H_EVENT_OCCURRED0_BAU_DATA_SHFT 55 +#define UV1H_EVENT_OCCURRED0_BAU_DATA_MASK 0x0080000000000000UL +#define UV1H_EVENT_OCCURRED0_POWER_MANAGEMENT_REQ_SHFT 56 +#define UV1H_EVENT_OCCURRED0_POWER_MANAGEMENT_REQ_MASK 0x0100000000000000UL + +#define UV2H_EVENT_OCCURRED0_LB_HCERR_SHFT 0 +#define UV2H_EVENT_OCCURRED0_LB_HCERR_MASK 0x0000000000000001UL +#define UV2H_EVENT_OCCURRED0_QP_HCERR_SHFT 1 +#define UV2H_EVENT_OCCURRED0_QP_HCERR_MASK 0x0000000000000002UL +#define UV2H_EVENT_OCCURRED0_RH_HCERR_SHFT 2 +#define UV2H_EVENT_OCCURRED0_RH_HCERR_MASK 0x0000000000000004UL +#define UV2H_EVENT_OCCURRED0_LH0_HCERR_SHFT 3 +#define UV2H_EVENT_OCCURRED0_LH0_HCERR_MASK 0x0000000000000008UL +#define UV2H_EVENT_OCCURRED0_LH1_HCERR_SHFT 4 +#define UV2H_EVENT_OCCURRED0_LH1_HCERR_MASK 0x0000000000000010UL +#define UV2H_EVENT_OCCURRED0_GR0_HCERR_SHFT 5 +#define UV2H_EVENT_OCCURRED0_GR0_HCERR_MASK 0x0000000000000020UL +#define UV2H_EVENT_OCCURRED0_GR1_HCERR_SHFT 6 +#define UV2H_EVENT_OCCURRED0_GR1_HCERR_MASK 0x0000000000000040UL +#define UV2H_EVENT_OCCURRED0_NI0_HCERR_SHFT 7 +#define UV2H_EVENT_OCCURRED0_NI0_HCERR_MASK 0x0000000000000080UL +#define UV2H_EVENT_OCCURRED0_NI1_HCERR_SHFT 8 +#define UV2H_EVENT_OCCURRED0_NI1_HCERR_MASK 0x0000000000000100UL +#define UV2H_EVENT_OCCURRED0_LB_AOERR0_SHFT 9 +#define UV2H_EVENT_OCCURRED0_LB_AOERR0_MASK 0x0000000000000200UL +#define UV2H_EVENT_OCCURRED0_QP_AOERR0_SHFT 10 +#define UV2H_EVENT_OCCURRED0_QP_AOERR0_MASK 0x0000000000000400UL +#define UV2H_EVENT_OCCURRED0_RH_AOERR0_SHFT 11 +#define UV2H_EVENT_OCCURRED0_RH_AOERR0_MASK 0x0000000000000800UL +#define UV2H_EVENT_OCCURRED0_LH0_AOERR0_SHFT 12 +#define UV2H_EVENT_OCCURRED0_LH0_AOERR0_MASK 0x0000000000001000UL +#define UV2H_EVENT_OCCURRED0_LH1_AOERR0_SHFT 13 +#define UV2H_EVENT_OCCURRED0_LH1_AOERR0_MASK 0x0000000000002000UL +#define UV2H_EVENT_OCCURRED0_GR0_AOERR0_SHFT 14 +#define UV2H_EVENT_OCCURRED0_GR0_AOERR0_MASK 0x0000000000004000UL +#define UV2H_EVENT_OCCURRED0_GR1_AOERR0_SHFT 15 +#define UV2H_EVENT_OCCURRED0_GR1_AOERR0_MASK 0x0000000000008000UL +#define UV2H_EVENT_OCCURRED0_XB_AOERR0_SHFT 16 +#define UV2H_EVENT_OCCURRED0_XB_AOERR0_MASK 0x0000000000010000UL +#define UV2H_EVENT_OCCURRED0_RT_AOERR0_SHFT 17 +#define UV2H_EVENT_OCCURRED0_RT_AOERR0_MASK 0x0000000000020000UL +#define UV2H_EVENT_OCCURRED0_NI0_AOERR0_SHFT 18 +#define UV2H_EVENT_OCCURRED0_NI0_AOERR0_MASK 0x0000000000040000UL +#define UV2H_EVENT_OCCURRED0_NI1_AOERR0_SHFT 19 +#define UV2H_EVENT_OCCURRED0_NI1_AOERR0_MASK 0x0000000000080000UL +#define UV2H_EVENT_OCCURRED0_LB_AOERR1_SHFT 20 +#define UV2H_EVENT_OCCURRED0_LB_AOERR1_MASK 0x0000000000100000UL +#define UV2H_EVENT_OCCURRED0_QP_AOERR1_SHFT 21 +#define UV2H_EVENT_OCCURRED0_QP_AOERR1_MASK 0x0000000000200000UL +#define UV2H_EVENT_OCCURRED0_RH_AOERR1_SHFT 22 +#define UV2H_EVENT_OCCURRED0_RH_AOERR1_MASK 0x0000000000400000UL +#define UV2H_EVENT_OCCURRED0_LH0_AOERR1_SHFT 23 +#define UV2H_EVENT_OCCURRED0_LH0_AOERR1_MASK 0x0000000000800000UL +#define UV2H_EVENT_OCCURRED0_LH1_AOERR1_SHFT 24 +#define UV2H_EVENT_OCCURRED0_LH1_AOERR1_MASK 0x0000000001000000UL +#define UV2H_EVENT_OCCURRED0_GR0_AOERR1_SHFT 25 +#define UV2H_EVENT_OCCURRED0_GR0_AOERR1_MASK 0x0000000002000000UL +#define UV2H_EVENT_OCCURRED0_GR1_AOERR1_SHFT 26 +#define UV2H_EVENT_OCCURRED0_GR1_AOERR1_MASK 0x0000000004000000UL +#define UV2H_EVENT_OCCURRED0_XB_AOERR1_SHFT 27 +#define UV2H_EVENT_OCCURRED0_XB_AOERR1_MASK 0x0000000008000000UL +#define UV2H_EVENT_OCCURRED0_RT_AOERR1_SHFT 28 +#define UV2H_EVENT_OCCURRED0_RT_AOERR1_MASK 0x0000000010000000UL +#define UV2H_EVENT_OCCURRED0_NI0_AOERR1_SHFT 29 +#define UV2H_EVENT_OCCURRED0_NI0_AOERR1_MASK 0x0000000020000000UL +#define UV2H_EVENT_OCCURRED0_NI1_AOERR1_SHFT 30 +#define UV2H_EVENT_OCCURRED0_NI1_AOERR1_MASK 0x0000000040000000UL +#define UV2H_EVENT_OCCURRED0_SYSTEM_SHUTDOWN_INT_SHFT 31 +#define UV2H_EVENT_OCCURRED0_SYSTEM_SHUTDOWN_INT_MASK 0x0000000080000000UL +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_0_SHFT 32 +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_0_MASK 0x0000000100000000UL +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_1_SHFT 33 +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_1_MASK 0x0000000200000000UL +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_2_SHFT 34 +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_2_MASK 0x0000000400000000UL +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_3_SHFT 35 +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_3_MASK 0x0000000800000000UL +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_4_SHFT 36 +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_4_MASK 0x0000001000000000UL +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_5_SHFT 37 +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_5_MASK 0x0000002000000000UL +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_6_SHFT 38 +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_6_MASK 0x0000004000000000UL +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_7_SHFT 39 +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_7_MASK 0x0000008000000000UL +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_8_SHFT 40 +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_8_MASK 0x0000010000000000UL +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_9_SHFT 41 +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_9_MASK 0x0000020000000000UL +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_10_SHFT 42 +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_10_MASK 0x0000040000000000UL +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_11_SHFT 43 +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_11_MASK 0x0000080000000000UL +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_12_SHFT 44 +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_12_MASK 0x0000100000000000UL +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_13_SHFT 45 +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_13_MASK 0x0000200000000000UL +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_14_SHFT 46 +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_14_MASK 0x0000400000000000UL +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_15_SHFT 47 +#define UV2H_EVENT_OCCURRED0_LB_IRQ_INT_15_MASK 0x0000800000000000UL +#define UV2H_EVENT_OCCURRED0_L1_NMI_INT_SHFT 48 +#define UV2H_EVENT_OCCURRED0_L1_NMI_INT_MASK 0x0001000000000000UL +#define UV2H_EVENT_OCCURRED0_STOP_CLOCK_SHFT 49 +#define UV2H_EVENT_OCCURRED0_STOP_CLOCK_MASK 0x0002000000000000UL +#define UV2H_EVENT_OCCURRED0_ASIC_TO_L1_SHFT 50 +#define UV2H_EVENT_OCCURRED0_ASIC_TO_L1_MASK 0x0004000000000000UL +#define UV2H_EVENT_OCCURRED0_L1_TO_ASIC_SHFT 51 +#define UV2H_EVENT_OCCURRED0_L1_TO_ASIC_MASK 0x0008000000000000UL +#define UV2H_EVENT_OCCURRED0_LA_SEQ_TRIGGER_SHFT 52 +#define UV2H_EVENT_OCCURRED0_LA_SEQ_TRIGGER_MASK 0x0010000000000000UL +#define UV2H_EVENT_OCCURRED0_IPI_INT_SHFT 53 +#define UV2H_EVENT_OCCURRED0_IPI_INT_MASK 0x0020000000000000UL +#define UV2H_EVENT_OCCURRED0_EXTIO_INT0_SHFT 54 +#define UV2H_EVENT_OCCURRED0_EXTIO_INT0_MASK 0x0040000000000000UL +#define UV2H_EVENT_OCCURRED0_EXTIO_INT1_SHFT 55 +#define UV2H_EVENT_OCCURRED0_EXTIO_INT1_MASK 0x0080000000000000UL +#define UV2H_EVENT_OCCURRED0_EXTIO_INT2_SHFT 56 +#define UV2H_EVENT_OCCURRED0_EXTIO_INT2_MASK 0x0100000000000000UL +#define UV2H_EVENT_OCCURRED0_EXTIO_INT3_SHFT 57 +#define UV2H_EVENT_OCCURRED0_EXTIO_INT3_MASK 0x0200000000000000UL +#define UV2H_EVENT_OCCURRED0_PROFILE_INT_SHFT 58 +#define UV2H_EVENT_OCCURRED0_PROFILE_INT_MASK 0x0400000000000000UL + union uvh_event_occurred0_u { unsigned long v; - struct uvh_event_occurred0_s { + struct uv1h_event_occurred0_s { unsigned long lb_hcerr : 1; /* RW, W1C */ unsigned long gr0_hcerr : 1; /* RW, W1C */ unsigned long gr1_hcerr : 1; /* RW, W1C */ @@ -250,14 +421,76 @@ union uvh_event_occurred0_u { unsigned long bau_data : 1; /* RW, W1C */ unsigned long power_management_req : 1; /* RW, W1C */ unsigned long rsvd_57_63 : 7; /* */ - } s; + } s1; + struct uv2h_event_occurred0_s { + unsigned long lb_hcerr : 1; /* RW */ + unsigned long qp_hcerr : 1; /* RW */ + unsigned long rh_hcerr : 1; /* RW */ + unsigned long lh0_hcerr : 1; /* RW */ + unsigned long lh1_hcerr : 1; /* RW */ + unsigned long gr0_hcerr : 1; /* RW */ + unsigned long gr1_hcerr : 1; /* RW */ + unsigned long ni0_hcerr : 1; /* RW */ + unsigned long ni1_hcerr : 1; /* RW */ + unsigned long lb_aoerr0 : 1; /* RW */ + unsigned long qp_aoerr0 : 1; /* RW */ + unsigned long rh_aoerr0 : 1; /* RW */ + unsigned long lh0_aoerr0 : 1; /* RW */ + unsigned long lh1_aoerr0 : 1; /* RW */ + unsigned long gr0_aoerr0 : 1; /* RW */ + unsigned long gr1_aoerr0 : 1; /* RW */ + unsigned long xb_aoerr0 : 1; /* RW */ + unsigned long rt_aoerr0 : 1; /* RW */ + unsigned long ni0_aoerr0 : 1; /* RW */ + unsigned long ni1_aoerr0 : 1; /* RW */ + unsigned long lb_aoerr1 : 1; /* RW */ + unsigned long qp_aoerr1 : 1; /* RW */ + unsigned long rh_aoerr1 : 1; /* RW */ + unsigned long lh0_aoerr1 : 1; /* RW */ + unsigned long lh1_aoerr1 : 1; /* RW */ + unsigned long gr0_aoerr1 : 1; /* RW */ + unsigned long gr1_aoerr1 : 1; /* RW */ + unsigned long xb_aoerr1 : 1; /* RW */ + unsigned long rt_aoerr1 : 1; /* RW */ + unsigned long ni0_aoerr1 : 1; /* RW */ + unsigned long ni1_aoerr1 : 1; /* RW */ + unsigned long system_shutdown_int : 1; /* RW */ + unsigned long lb_irq_int_0 : 1; /* RW */ + unsigned long lb_irq_int_1 : 1; /* RW */ + unsigned long lb_irq_int_2 : 1; /* RW */ + unsigned long lb_irq_int_3 : 1; /* RW */ + unsigned long lb_irq_int_4 : 1; /* RW */ + unsigned long lb_irq_int_5 : 1; /* RW */ + unsigned long lb_irq_int_6 : 1; /* RW */ + unsigned long lb_irq_int_7 : 1; /* RW */ + unsigned long lb_irq_int_8 : 1; /* RW */ + unsigned long lb_irq_int_9 : 1; /* RW */ + unsigned long lb_irq_int_10 : 1; /* RW */ + unsigned long lb_irq_int_11 : 1; /* RW */ + unsigned long lb_irq_int_12 : 1; /* RW */ + unsigned long lb_irq_int_13 : 1; /* RW */ + unsigned long lb_irq_int_14 : 1; /* RW */ + unsigned long lb_irq_int_15 : 1; /* RW */ + unsigned long l1_nmi_int : 1; /* RW */ + unsigned long stop_clock : 1; /* RW */ + unsigned long asic_to_l1 : 1; /* RW */ + unsigned long l1_to_asic : 1; /* RW */ + unsigned long la_seq_trigger : 1; /* RW */ + unsigned long ipi_int : 1; /* RW */ + unsigned long extio_int0 : 1; /* RW */ + unsigned long extio_int1 : 1; /* RW */ + unsigned long extio_int2 : 1; /* RW */ + unsigned long extio_int3 : 1; /* RW */ + unsigned long profile_int : 1; /* RW */ + unsigned long rsvd_59_63 : 5; /* */ + } s2; }; /* ========================================================================= */ /* UVH_EVENT_OCCURRED0_ALIAS */ /* ========================================================================= */ #define UVH_EVENT_OCCURRED0_ALIAS 0x0000000000070008UL -#define UVH_EVENT_OCCURRED0_ALIAS_32 0x005f0 +#define UVH_EVENT_OCCURRED0_ALIAS_32 0x5f0 /* ========================================================================= */ /* UVH_GR0_TLB_INT0_CONFIG */ @@ -432,8 +665,16 @@ union uvh_int_cmpb_u { /* ========================================================================= */ #define UVH_INT_CMPC 0x22100UL -#define UVH_INT_CMPC_REAL_TIME_CMPC_SHFT 0 -#define UVH_INT_CMPC_REAL_TIME_CMPC_MASK 0x00ffffffffffffffUL +#define UV1H_INT_CMPC_REAL_TIME_CMPC_SHFT 0 +#define UV2H_INT_CMPC_REAL_TIME_CMPC_SHFT 0 +#define UVH_INT_CMPC_REAL_TIME_CMPC_SHFT (is_uv1_hub() ? \ + UV1H_INT_CMPC_REAL_TIME_CMPC_SHFT : \ + UV2H_INT_CMPC_REAL_TIME_CMPC_SHFT) +#define UV1H_INT_CMPC_REAL_TIME_CMPC_MASK 0xffffffffffffffUL +#define UV2H_INT_CMPC_REAL_TIME_CMPC_MASK 0xffffffffffffffUL +#define UVH_INT_CMPC_REAL_TIME_CMPC_MASK (is_uv1_hub() ? \ + UV1H_INT_CMPC_REAL_TIME_CMPC_MASK : \ + UV2H_INT_CMPC_REAL_TIME_CMPC_MASK) union uvh_int_cmpc_u { unsigned long v; @@ -448,8 +689,16 @@ union uvh_int_cmpc_u { /* ========================================================================= */ #define UVH_INT_CMPD 0x22180UL -#define UVH_INT_CMPD_REAL_TIME_CMPD_SHFT 0 -#define UVH_INT_CMPD_REAL_TIME_CMPD_MASK 0x00ffffffffffffffUL +#define UV1H_INT_CMPD_REAL_TIME_CMPD_SHFT 0 +#define UV2H_INT_CMPD_REAL_TIME_CMPD_SHFT 0 +#define UVH_INT_CMPD_REAL_TIME_CMPD_SHFT (is_uv1_hub() ? \ + UV1H_INT_CMPD_REAL_TIME_CMPD_SHFT : \ + UV2H_INT_CMPD_REAL_TIME_CMPD_SHFT) +#define UV1H_INT_CMPD_REAL_TIME_CMPD_MASK 0xffffffffffffffUL +#define UV2H_INT_CMPD_REAL_TIME_CMPD_MASK 0xffffffffffffffUL +#define UVH_INT_CMPD_REAL_TIME_CMPD_MASK (is_uv1_hub() ? \ + UV1H_INT_CMPD_REAL_TIME_CMPD_MASK : \ + UV2H_INT_CMPD_REAL_TIME_CMPD_MASK) union uvh_int_cmpd_u { unsigned long v; @@ -463,7 +712,7 @@ union uvh_int_cmpd_u { /* UVH_IPI_INT */ /* ========================================================================= */ #define UVH_IPI_INT 0x60500UL -#define UVH_IPI_INT_32 0x0348 +#define UVH_IPI_INT_32 0x348 #define UVH_IPI_INT_VECTOR_SHFT 0 #define UVH_IPI_INT_VECTOR_MASK 0x00000000000000ffUL @@ -493,7 +742,7 @@ union uvh_ipi_int_u { /* UVH_LB_BAU_INTD_PAYLOAD_QUEUE_FIRST */ /* ========================================================================= */ #define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_FIRST 0x320050UL -#define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_FIRST_32 0x009c0 +#define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_FIRST_32 0x9c0 #define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_FIRST_ADDRESS_SHFT 4 #define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_FIRST_ADDRESS_MASK 0x000007fffffffff0UL @@ -515,7 +764,7 @@ union uvh_lb_bau_intd_payload_queue_first_u { /* UVH_LB_BAU_INTD_PAYLOAD_QUEUE_LAST */ /* ========================================================================= */ #define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_LAST 0x320060UL -#define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_LAST_32 0x009c8 +#define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_LAST_32 0x9c8 #define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_LAST_ADDRESS_SHFT 4 #define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_LAST_ADDRESS_MASK 0x000007fffffffff0UL @@ -533,7 +782,7 @@ union uvh_lb_bau_intd_payload_queue_last_u { /* UVH_LB_BAU_INTD_PAYLOAD_QUEUE_TAIL */ /* ========================================================================= */ #define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_TAIL 0x320070UL -#define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_TAIL_32 0x009d0 +#define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_TAIL_32 0x9d0 #define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_TAIL_ADDRESS_SHFT 4 #define UVH_LB_BAU_INTD_PAYLOAD_QUEUE_TAIL_ADDRESS_MASK 0x000007fffffffff0UL @@ -551,7 +800,7 @@ union uvh_lb_bau_intd_payload_queue_tail_u { /* UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE */ /* ========================================================================= */ #define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE 0x320080UL -#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_32 0x0a68 +#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_32 0xa68 #define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_PENDING_0_SHFT 0 #define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_PENDING_0_MASK 0x0000000000000001UL @@ -585,6 +834,7 @@ union uvh_lb_bau_intd_payload_queue_tail_u { #define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_TIMEOUT_6_MASK 0x0000000000004000UL #define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_TIMEOUT_7_SHFT 15 #define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_TIMEOUT_7_MASK 0x0000000000008000UL + union uvh_lb_bau_intd_software_acknowledge_u { unsigned long v; struct uvh_lb_bau_intd_software_acknowledge_s { @@ -612,13 +862,13 @@ union uvh_lb_bau_intd_software_acknowledge_u { /* UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_ALIAS */ /* ========================================================================= */ #define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_ALIAS 0x0000000000320088UL -#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_ALIAS_32 0x0a70 +#define UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_ALIAS_32 0xa70 /* ========================================================================= */ /* UVH_LB_BAU_MISC_CONTROL */ /* ========================================================================= */ #define UVH_LB_BAU_MISC_CONTROL 0x320170UL -#define UVH_LB_BAU_MISC_CONTROL_32 0x00a10 +#define UVH_LB_BAU_MISC_CONTROL_32 0xa10 #define UVH_LB_BAU_MISC_CONTROL_REJECTION_DELAY_SHFT 0 #define UVH_LB_BAU_MISC_CONTROL_REJECTION_DELAY_MASK 0x00000000000000ffUL @@ -628,8 +878,8 @@ union uvh_lb_bau_intd_software_acknowledge_u { #define UVH_LB_BAU_MISC_CONTROL_FORCE_BROADCAST_MASK 0x0000000000000200UL #define UVH_LB_BAU_MISC_CONTROL_FORCE_LOCK_NOP_SHFT 10 #define UVH_LB_BAU_MISC_CONTROL_FORCE_LOCK_NOP_MASK 0x0000000000000400UL -#define UVH_LB_BAU_MISC_CONTROL_CSI_AGENT_PRESENCE_VECTOR_SHFT 11 -#define UVH_LB_BAU_MISC_CONTROL_CSI_AGENT_PRESENCE_VECTOR_MASK 0x0000000000003800UL +#define UVH_LB_BAU_MISC_CONTROL_QPI_AGENT_PRESENCE_VECTOR_SHFT 11 +#define UVH_LB_BAU_MISC_CONTROL_QPI_AGENT_PRESENCE_VECTOR_MASK 0x0000000000003800UL #define UVH_LB_BAU_MISC_CONTROL_DESCRIPTOR_FETCH_MODE_SHFT 14 #define UVH_LB_BAU_MISC_CONTROL_DESCRIPTOR_FETCH_MODE_MASK 0x0000000000004000UL #define UVH_LB_BAU_MISC_CONTROL_ENABLE_INTD_SOFT_ACK_MODE_SHFT 15 @@ -650,8 +900,86 @@ union uvh_lb_bau_intd_software_acknowledge_u { #define UVH_LB_BAU_MISC_CONTROL_USE_INCOMING_PRIORITY_MASK 0x0000000008000000UL #define UVH_LB_BAU_MISC_CONTROL_ENABLE_PROGRAMMED_INITIAL_PRIORITY_SHFT 28 #define UVH_LB_BAU_MISC_CONTROL_ENABLE_PROGRAMMED_INITIAL_PRIORITY_MASK 0x0000000010000000UL -#define UVH_LB_BAU_MISC_CONTROL_FUN_SHFT 48 -#define UVH_LB_BAU_MISC_CONTROL_FUN_MASK 0xffff000000000000UL + +#define UV1H_LB_BAU_MISC_CONTROL_REJECTION_DELAY_SHFT 0 +#define UV1H_LB_BAU_MISC_CONTROL_REJECTION_DELAY_MASK 0x00000000000000ffUL +#define UV1H_LB_BAU_MISC_CONTROL_APIC_MODE_SHFT 8 +#define UV1H_LB_BAU_MISC_CONTROL_APIC_MODE_MASK 0x0000000000000100UL +#define UV1H_LB_BAU_MISC_CONTROL_FORCE_BROADCAST_SHFT 9 +#define UV1H_LB_BAU_MISC_CONTROL_FORCE_BROADCAST_MASK 0x0000000000000200UL +#define UV1H_LB_BAU_MISC_CONTROL_FORCE_LOCK_NOP_SHFT 10 +#define UV1H_LB_BAU_MISC_CONTROL_FORCE_LOCK_NOP_MASK 0x0000000000000400UL +#define UV1H_LB_BAU_MISC_CONTROL_QPI_AGENT_PRESENCE_VECTOR_SHFT 11 +#define UV1H_LB_BAU_MISC_CONTROL_QPI_AGENT_PRESENCE_VECTOR_MASK 0x0000000000003800UL +#define UV1H_LB_BAU_MISC_CONTROL_DESCRIPTOR_FETCH_MODE_SHFT 14 +#define UV1H_LB_BAU_MISC_CONTROL_DESCRIPTOR_FETCH_MODE_MASK 0x0000000000004000UL +#define UV1H_LB_BAU_MISC_CONTROL_ENABLE_INTD_SOFT_ACK_MODE_SHFT 15 +#define UV1H_LB_BAU_MISC_CONTROL_ENABLE_INTD_SOFT_ACK_MODE_MASK 0x0000000000008000UL +#define UV1H_LB_BAU_MISC_CONTROL_INTD_SOFT_ACK_TIMEOUT_PERIOD_SHFT 16 +#define UV1H_LB_BAU_MISC_CONTROL_INTD_SOFT_ACK_TIMEOUT_PERIOD_MASK 0x00000000000f0000UL +#define UV1H_LB_BAU_MISC_CONTROL_ENABLE_DUAL_MAPPING_MODE_SHFT 20 +#define UV1H_LB_BAU_MISC_CONTROL_ENABLE_DUAL_MAPPING_MODE_MASK 0x0000000000100000UL +#define UV1H_LB_BAU_MISC_CONTROL_VGA_IO_PORT_DECODE_ENABLE_SHFT 21 +#define UV1H_LB_BAU_MISC_CONTROL_VGA_IO_PORT_DECODE_ENABLE_MASK 0x0000000000200000UL +#define UV1H_LB_BAU_MISC_CONTROL_VGA_IO_PORT_16_BIT_DECODE_SHFT 22 +#define UV1H_LB_BAU_MISC_CONTROL_VGA_IO_PORT_16_BIT_DECODE_MASK 0x0000000000400000UL +#define UV1H_LB_BAU_MISC_CONTROL_SUPPRESS_DEST_REGISTRATION_SHFT 23 +#define UV1H_LB_BAU_MISC_CONTROL_SUPPRESS_DEST_REGISTRATION_MASK 0x0000000000800000UL +#define UV1H_LB_BAU_MISC_CONTROL_PROGRAMMED_INITIAL_PRIORITY_SHFT 24 +#define UV1H_LB_BAU_MISC_CONTROL_PROGRAMMED_INITIAL_PRIORITY_MASK 0x0000000007000000UL +#define UV1H_LB_BAU_MISC_CONTROL_USE_INCOMING_PRIORITY_SHFT 27 +#define UV1H_LB_BAU_MISC_CONTROL_USE_INCOMING_PRIORITY_MASK 0x0000000008000000UL +#define UV1H_LB_BAU_MISC_CONTROL_ENABLE_PROGRAMMED_INITIAL_PRIORITY_SHFT 28 +#define UV1H_LB_BAU_MISC_CONTROL_ENABLE_PROGRAMMED_INITIAL_PRIORITY_MASK 0x0000000010000000UL +#define UV1H_LB_BAU_MISC_CONTROL_FUN_SHFT 48 +#define UV1H_LB_BAU_MISC_CONTROL_FUN_MASK 0xffff000000000000UL + +#define UV2H_LB_BAU_MISC_CONTROL_REJECTION_DELAY_SHFT 0 +#define UV2H_LB_BAU_MISC_CONTROL_REJECTION_DELAY_MASK 0x00000000000000ffUL +#define UV2H_LB_BAU_MISC_CONTROL_APIC_MODE_SHFT 8 +#define UV2H_LB_BAU_MISC_CONTROL_APIC_MODE_MASK 0x0000000000000100UL +#define UV2H_LB_BAU_MISC_CONTROL_FORCE_BROADCAST_SHFT 9 +#define UV2H_LB_BAU_MISC_CONTROL_FORCE_BROADCAST_MASK 0x0000000000000200UL +#define UV2H_LB_BAU_MISC_CONTROL_FORCE_LOCK_NOP_SHFT 10 +#define UV2H_LB_BAU_MISC_CONTROL_FORCE_LOCK_NOP_MASK 0x0000000000000400UL +#define UV2H_LB_BAU_MISC_CONTROL_QPI_AGENT_PRESENCE_VECTOR_SHFT 11 +#define UV2H_LB_BAU_MISC_CONTROL_QPI_AGENT_PRESENCE_VECTOR_MASK 0x0000000000003800UL +#define UV2H_LB_BAU_MISC_CONTROL_DESCRIPTOR_FETCH_MODE_SHFT 14 +#define UV2H_LB_BAU_MISC_CONTROL_DESCRIPTOR_FETCH_MODE_MASK 0x0000000000004000UL +#define UV2H_LB_BAU_MISC_CONTROL_ENABLE_INTD_SOFT_ACK_MODE_SHFT 15 +#define UV2H_LB_BAU_MISC_CONTROL_ENABLE_INTD_SOFT_ACK_MODE_MASK 0x0000000000008000UL +#define UV2H_LB_BAU_MISC_CONTROL_INTD_SOFT_ACK_TIMEOUT_PERIOD_SHFT 16 +#define UV2H_LB_BAU_MISC_CONTROL_INTD_SOFT_ACK_TIMEOUT_PERIOD_MASK 0x00000000000f0000UL +#define UV2H_LB_BAU_MISC_CONTROL_ENABLE_DUAL_MAPPING_MODE_SHFT 20 +#define UV2H_LB_BAU_MISC_CONTROL_ENABLE_DUAL_MAPPING_MODE_MASK 0x0000000000100000UL +#define UV2H_LB_BAU_MISC_CONTROL_VGA_IO_PORT_DECODE_ENABLE_SHFT 21 +#define UV2H_LB_BAU_MISC_CONTROL_VGA_IO_PORT_DECODE_ENABLE_MASK 0x0000000000200000UL +#define UV2H_LB_BAU_MISC_CONTROL_VGA_IO_PORT_16_BIT_DECODE_SHFT 22 +#define UV2H_LB_BAU_MISC_CONTROL_VGA_IO_PORT_16_BIT_DECODE_MASK 0x0000000000400000UL +#define UV2H_LB_BAU_MISC_CONTROL_SUPPRESS_DEST_REGISTRATION_SHFT 23 +#define UV2H_LB_BAU_MISC_CONTROL_SUPPRESS_DEST_REGISTRATION_MASK 0x0000000000800000UL +#define UV2H_LB_BAU_MISC_CONTROL_PROGRAMMED_INITIAL_PRIORITY_SHFT 24 +#define UV2H_LB_BAU_MISC_CONTROL_PROGRAMMED_INITIAL_PRIORITY_MASK 0x0000000007000000UL +#define UV2H_LB_BAU_MISC_CONTROL_USE_INCOMING_PRIORITY_SHFT 27 +#define UV2H_LB_BAU_MISC_CONTROL_USE_INCOMING_PRIORITY_MASK 0x0000000008000000UL +#define UV2H_LB_BAU_MISC_CONTROL_ENABLE_PROGRAMMED_INITIAL_PRIORITY_SHFT 28 +#define UV2H_LB_BAU_MISC_CONTROL_ENABLE_PROGRAMMED_INITIAL_PRIORITY_MASK 0x0000000010000000UL +#define UV2H_LB_BAU_MISC_CONTROL_ENABLE_AUTOMATIC_APIC_MODE_SELECTION_SHFT 29 +#define UV2H_LB_BAU_MISC_CONTROL_ENABLE_AUTOMATIC_APIC_MODE_SELECTION_MASK 0x0000000020000000UL +#define UV2H_LB_BAU_MISC_CONTROL_APIC_MODE_STATUS_SHFT 30 +#define UV2H_LB_BAU_MISC_CONTROL_APIC_MODE_STATUS_MASK 0x0000000040000000UL +#define UV2H_LB_BAU_MISC_CONTROL_SUPPRESS_INTERRUPTS_TO_SELF_SHFT 31 +#define UV2H_LB_BAU_MISC_CONTROL_SUPPRESS_INTERRUPTS_TO_SELF_MASK 0x0000000080000000UL +#define UV2H_LB_BAU_MISC_CONTROL_ENABLE_LOCK_BASED_SYSTEM_FLUSH_SHFT 32 +#define UV2H_LB_BAU_MISC_CONTROL_ENABLE_LOCK_BASED_SYSTEM_FLUSH_MASK 0x0000000100000000UL +#define UV2H_LB_BAU_MISC_CONTROL_ENABLE_EXTENDED_SB_STATUS_SHFT 33 +#define UV2H_LB_BAU_MISC_CONTROL_ENABLE_EXTENDED_SB_STATUS_MASK 0x0000000200000000UL +#define UV2H_LB_BAU_MISC_CONTROL_SUPPRESS_INT_PRIO_UDT_TO_SELF_SHFT 34 +#define UV2H_LB_BAU_MISC_CONTROL_SUPPRESS_INT_PRIO_UDT_TO_SELF_MASK 0x0000000400000000UL +#define UV2H_LB_BAU_MISC_CONTROL_USE_LEGACY_DESCRIPTOR_FORMATS_SHFT 35 +#define UV2H_LB_BAU_MISC_CONTROL_USE_LEGACY_DESCRIPTOR_FORMATS_MASK 0x0000000800000000UL +#define UV2H_LB_BAU_MISC_CONTROL_FUN_SHFT 48 +#define UV2H_LB_BAU_MISC_CONTROL_FUN_MASK 0xffff000000000000UL union uvh_lb_bau_misc_control_u { unsigned long v; @@ -660,7 +988,25 @@ union uvh_lb_bau_misc_control_u { unsigned long apic_mode : 1; /* RW */ unsigned long force_broadcast : 1; /* RW */ unsigned long force_lock_nop : 1; /* RW */ - unsigned long csi_agent_presence_vector : 3; /* RW */ + unsigned long qpi_agent_presence_vector : 3; /* RW */ + unsigned long descriptor_fetch_mode : 1; /* RW */ + unsigned long enable_intd_soft_ack_mode : 1; /* RW */ + unsigned long intd_soft_ack_timeout_period : 4; /* RW */ + unsigned long enable_dual_mapping_mode : 1; /* RW */ + unsigned long vga_io_port_decode_enable : 1; /* RW */ + unsigned long vga_io_port_16_bit_decode : 1; /* RW */ + unsigned long suppress_dest_registration : 1; /* RW */ + unsigned long programmed_initial_priority : 3; /* RW */ + unsigned long use_incoming_priority : 1; /* RW */ + unsigned long enable_programmed_initial_priority : 1; /* RW */ + unsigned long rsvd_29_63 : 35; + } s; + struct uv1h_lb_bau_misc_control_s { + unsigned long rejection_delay : 8; /* RW */ + unsigned long apic_mode : 1; /* RW */ + unsigned long force_broadcast : 1; /* RW */ + unsigned long force_lock_nop : 1; /* RW */ + unsigned long qpi_agent_presence_vector : 3; /* RW */ unsigned long descriptor_fetch_mode : 1; /* RW */ unsigned long enable_intd_soft_ack_mode : 1; /* RW */ unsigned long intd_soft_ack_timeout_period : 4; /* RW */ @@ -673,14 +1019,40 @@ union uvh_lb_bau_misc_control_u { unsigned long enable_programmed_initial_priority : 1; /* RW */ unsigned long rsvd_29_47 : 19; /* */ unsigned long fun : 16; /* RW */ - } s; + } s1; + struct uv2h_lb_bau_misc_control_s { + unsigned long rejection_delay : 8; /* RW */ + unsigned long apic_mode : 1; /* RW */ + unsigned long force_broadcast : 1; /* RW */ + unsigned long force_lock_nop : 1; /* RW */ + unsigned long qpi_agent_presence_vector : 3; /* RW */ + unsigned long descriptor_fetch_mode : 1; /* RW */ + unsigned long enable_intd_soft_ack_mode : 1; /* RW */ + unsigned long intd_soft_ack_timeout_period : 4; /* RW */ + unsigned long enable_dual_mapping_mode : 1; /* RW */ + unsigned long vga_io_port_decode_enable : 1; /* RW */ + unsigned long vga_io_port_16_bit_decode : 1; /* RW */ + unsigned long suppress_dest_registration : 1; /* RW */ + unsigned long programmed_initial_priority : 3; /* RW */ + unsigned long use_incoming_priority : 1; /* RW */ + unsigned long enable_programmed_initial_priority : 1; /* RW */ + unsigned long enable_automatic_apic_mode_selection : 1; /* RW */ + unsigned long apic_mode_status : 1; /* RO */ + unsigned long suppress_interrupts_to_self : 1; /* RW */ + unsigned long enable_lock_based_system_flush : 1; /* RW */ + unsigned long enable_extended_sb_status : 1; /* RW */ + unsigned long suppress_int_prio_udt_to_self : 1; /* RW */ + unsigned long use_legacy_descriptor_formats : 1; /* RW */ + unsigned long rsvd_36_47 : 12; /* */ + unsigned long fun : 16; /* RW */ + } s2; }; /* ========================================================================= */ /* UVH_LB_BAU_SB_ACTIVATION_CONTROL */ /* ========================================================================= */ #define UVH_LB_BAU_SB_ACTIVATION_CONTROL 0x320020UL -#define UVH_LB_BAU_SB_ACTIVATION_CONTROL_32 0x009a8 +#define UVH_LB_BAU_SB_ACTIVATION_CONTROL_32 0x9a8 #define UVH_LB_BAU_SB_ACTIVATION_CONTROL_INDEX_SHFT 0 #define UVH_LB_BAU_SB_ACTIVATION_CONTROL_INDEX_MASK 0x000000000000003fUL @@ -703,7 +1075,7 @@ union uvh_lb_bau_sb_activation_control_u { /* UVH_LB_BAU_SB_ACTIVATION_STATUS_0 */ /* ========================================================================= */ #define UVH_LB_BAU_SB_ACTIVATION_STATUS_0 0x320030UL -#define UVH_LB_BAU_SB_ACTIVATION_STATUS_0_32 0x009b0 +#define UVH_LB_BAU_SB_ACTIVATION_STATUS_0_32 0x9b0 #define UVH_LB_BAU_SB_ACTIVATION_STATUS_0_STATUS_SHFT 0 #define UVH_LB_BAU_SB_ACTIVATION_STATUS_0_STATUS_MASK 0xffffffffffffffffUL @@ -719,7 +1091,7 @@ union uvh_lb_bau_sb_activation_status_0_u { /* UVH_LB_BAU_SB_ACTIVATION_STATUS_1 */ /* ========================================================================= */ #define UVH_LB_BAU_SB_ACTIVATION_STATUS_1 0x320040UL -#define UVH_LB_BAU_SB_ACTIVATION_STATUS_1_32 0x009b8 +#define UVH_LB_BAU_SB_ACTIVATION_STATUS_1_32 0x9b8 #define UVH_LB_BAU_SB_ACTIVATION_STATUS_1_STATUS_SHFT 0 #define UVH_LB_BAU_SB_ACTIVATION_STATUS_1_STATUS_MASK 0xffffffffffffffffUL @@ -735,7 +1107,7 @@ union uvh_lb_bau_sb_activation_status_1_u { /* UVH_LB_BAU_SB_DESCRIPTOR_BASE */ /* ========================================================================= */ #define UVH_LB_BAU_SB_DESCRIPTOR_BASE 0x320010UL -#define UVH_LB_BAU_SB_DESCRIPTOR_BASE_32 0x009a0 +#define UVH_LB_BAU_SB_DESCRIPTOR_BASE_32 0x9a0 #define UVH_LB_BAU_SB_DESCRIPTOR_BASE_PAGE_ADDRESS_SHFT 12 #define UVH_LB_BAU_SB_DESCRIPTOR_BASE_PAGE_ADDRESS_MASK 0x000007fffffff000UL @@ -754,23 +1126,6 @@ union uvh_lb_bau_sb_descriptor_base_u { }; /* ========================================================================= */ -/* UVH_LB_TARGET_PHYSICAL_APIC_ID_MASK */ -/* ========================================================================= */ -#define UVH_LB_TARGET_PHYSICAL_APIC_ID_MASK 0x320130UL -#define UVH_LB_TARGET_PHYSICAL_APIC_ID_MASK_32 0x009f0 - -#define UVH_LB_TARGET_PHYSICAL_APIC_ID_MASK_BIT_ENABLES_SHFT 0 -#define UVH_LB_TARGET_PHYSICAL_APIC_ID_MASK_BIT_ENABLES_MASK 0x00000000ffffffffUL - -union uvh_lb_target_physical_apic_id_mask_u { - unsigned long v; - struct uvh_lb_target_physical_apic_id_mask_s { - unsigned long bit_enables : 32; /* RW */ - unsigned long rsvd_32_63 : 32; /* */ - } s; -}; - -/* ========================================================================= */ /* UVH_NODE_ID */ /* ========================================================================= */ #define UVH_NODE_ID 0x0UL @@ -785,10 +1140,36 @@ union uvh_lb_target_physical_apic_id_mask_u { #define UVH_NODE_ID_REVISION_MASK 0x00000000f0000000UL #define UVH_NODE_ID_NODE_ID_SHFT 32 #define UVH_NODE_ID_NODE_ID_MASK 0x00007fff00000000UL -#define UVH_NODE_ID_NODES_PER_BIT_SHFT 48 -#define UVH_NODE_ID_NODES_PER_BIT_MASK 0x007f000000000000UL -#define UVH_NODE_ID_NI_PORT_SHFT 56 -#define UVH_NODE_ID_NI_PORT_MASK 0x0f00000000000000UL + +#define UV1H_NODE_ID_FORCE1_SHFT 0 +#define UV1H_NODE_ID_FORCE1_MASK 0x0000000000000001UL +#define UV1H_NODE_ID_MANUFACTURER_SHFT 1 +#define UV1H_NODE_ID_MANUFACTURER_MASK 0x0000000000000ffeUL +#define UV1H_NODE_ID_PART_NUMBER_SHFT 12 +#define UV1H_NODE_ID_PART_NUMBER_MASK 0x000000000ffff000UL +#define UV1H_NODE_ID_REVISION_SHFT 28 +#define UV1H_NODE_ID_REVISION_MASK 0x00000000f0000000UL +#define UV1H_NODE_ID_NODE_ID_SHFT 32 +#define UV1H_NODE_ID_NODE_ID_MASK 0x00007fff00000000UL +#define UV1H_NODE_ID_NODES_PER_BIT_SHFT 48 +#define UV1H_NODE_ID_NODES_PER_BIT_MASK 0x007f000000000000UL +#define UV1H_NODE_ID_NI_PORT_SHFT 56 +#define UV1H_NODE_ID_NI_PORT_MASK 0x0f00000000000000UL + +#define UV2H_NODE_ID_FORCE1_SHFT 0 +#define UV2H_NODE_ID_FORCE1_MASK 0x0000000000000001UL +#define UV2H_NODE_ID_MANUFACTURER_SHFT 1 +#define UV2H_NODE_ID_MANUFACTURER_MASK 0x0000000000000ffeUL +#define UV2H_NODE_ID_PART_NUMBER_SHFT 12 +#define UV2H_NODE_ID_PART_NUMBER_MASK 0x000000000ffff000UL +#define UV2H_NODE_ID_REVISION_SHFT 28 +#define UV2H_NODE_ID_REVISION_MASK 0x00000000f0000000UL +#define UV2H_NODE_ID_NODE_ID_SHFT 32 +#define UV2H_NODE_ID_NODE_ID_MASK 0x00007fff00000000UL +#define UV2H_NODE_ID_NODES_PER_BIT_SHFT 50 +#define UV2H_NODE_ID_NODES_PER_BIT_MASK 0x01fc000000000000UL +#define UV2H_NODE_ID_NI_PORT_SHFT 57 +#define UV2H_NODE_ID_NI_PORT_MASK 0x3e00000000000000UL union uvh_node_id_u { unsigned long v; @@ -798,12 +1179,31 @@ union uvh_node_id_u { unsigned long part_number : 16; /* RO */ unsigned long revision : 4; /* RO */ unsigned long node_id : 15; /* RW */ + unsigned long rsvd_47_63 : 17; + } s; + struct uv1h_node_id_s { + unsigned long force1 : 1; /* RO */ + unsigned long manufacturer : 11; /* RO */ + unsigned long part_number : 16; /* RO */ + unsigned long revision : 4; /* RO */ + unsigned long node_id : 15; /* RW */ unsigned long rsvd_47 : 1; /* */ unsigned long nodes_per_bit : 7; /* RW */ unsigned long rsvd_55 : 1; /* */ unsigned long ni_port : 4; /* RO */ unsigned long rsvd_60_63 : 4; /* */ - } s; + } s1; + struct uv2h_node_id_s { + unsigned long force1 : 1; /* RO */ + unsigned long manufacturer : 11; /* RO */ + unsigned long part_number : 16; /* RO */ + unsigned long revision : 4; /* RO */ + unsigned long node_id : 15; /* RW */ + unsigned long rsvd_47_49 : 3; /* */ + unsigned long nodes_per_bit : 7; /* RO */ + unsigned long ni_port : 5; /* RO */ + unsigned long rsvd_62_63 : 2; /* */ + } s2; }; /* ========================================================================= */ @@ -954,18 +1354,38 @@ union uvh_rh_gam_alias210_redirect_config_2_mmr_u { #define UVH_RH_GAM_CONFIG_MMR_M_SKT_MASK 0x000000000000003fUL #define UVH_RH_GAM_CONFIG_MMR_N_SKT_SHFT 6 #define UVH_RH_GAM_CONFIG_MMR_N_SKT_MASK 0x00000000000003c0UL -#define UVH_RH_GAM_CONFIG_MMR_MMIOL_CFG_SHFT 12 -#define UVH_RH_GAM_CONFIG_MMR_MMIOL_CFG_MASK 0x0000000000001000UL + +#define UV1H_RH_GAM_CONFIG_MMR_M_SKT_SHFT 0 +#define UV1H_RH_GAM_CONFIG_MMR_M_SKT_MASK 0x000000000000003fUL +#define UV1H_RH_GAM_CONFIG_MMR_N_SKT_SHFT 6 +#define UV1H_RH_GAM_CONFIG_MMR_N_SKT_MASK 0x00000000000003c0UL +#define UV1H_RH_GAM_CONFIG_MMR_MMIOL_CFG_SHFT 12 +#define UV1H_RH_GAM_CONFIG_MMR_MMIOL_CFG_MASK 0x0000000000001000UL + +#define UV2H_RH_GAM_CONFIG_MMR_M_SKT_SHFT 0 +#define UV2H_RH_GAM_CONFIG_MMR_M_SKT_MASK 0x000000000000003fUL +#define UV2H_RH_GAM_CONFIG_MMR_N_SKT_SHFT 6 +#define UV2H_RH_GAM_CONFIG_MMR_N_SKT_MASK 0x00000000000003c0UL union uvh_rh_gam_config_mmr_u { unsigned long v; struct uvh_rh_gam_config_mmr_s { unsigned long m_skt : 6; /* RW */ unsigned long n_skt : 4; /* RW */ + unsigned long rsvd_10_63 : 54; + } s; + struct uv1h_rh_gam_config_mmr_s { + unsigned long m_skt : 6; /* RW */ + unsigned long n_skt : 4; /* RW */ unsigned long rsvd_10_11: 2; /* */ unsigned long mmiol_cfg : 1; /* RW */ unsigned long rsvd_13_63: 51; /* */ - } s; + } s1; + struct uv2h_rh_gam_config_mmr_s { + unsigned long m_skt : 6; /* RW */ + unsigned long n_skt : 4; /* RW */ + unsigned long rsvd_10_63: 54; /* */ + } s2; }; /* ========================================================================= */ @@ -975,25 +1395,49 @@ union uvh_rh_gam_config_mmr_u { #define UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR_BASE_SHFT 28 #define UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR_BASE_MASK 0x00003ffff0000000UL -#define UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR_GR4_SHFT 48 -#define UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR_GR4_MASK 0x0001000000000000UL -#define UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR_N_GRU_SHFT 52 -#define UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR_N_GRU_MASK 0x00f0000000000000UL -#define UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR_ENABLE_SHFT 63 -#define UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR_ENABLE_MASK 0x8000000000000000UL + +#define UV1H_RH_GAM_GRU_OVERLAY_CONFIG_MMR_BASE_SHFT 28 +#define UV1H_RH_GAM_GRU_OVERLAY_CONFIG_MMR_BASE_MASK 0x00003ffff0000000UL +#define UV1H_RH_GAM_GRU_OVERLAY_CONFIG_MMR_GR4_SHFT 48 +#define UV1H_RH_GAM_GRU_OVERLAY_CONFIG_MMR_GR4_MASK 0x0001000000000000UL +#define UV1H_RH_GAM_GRU_OVERLAY_CONFIG_MMR_N_GRU_SHFT 52 +#define UV1H_RH_GAM_GRU_OVERLAY_CONFIG_MMR_N_GRU_MASK 0x00f0000000000000UL +#define UV1H_RH_GAM_GRU_OVERLAY_CONFIG_MMR_ENABLE_SHFT 63 +#define UV1H_RH_GAM_GRU_OVERLAY_CONFIG_MMR_ENABLE_MASK 0x8000000000000000UL + +#define UV2H_RH_GAM_GRU_OVERLAY_CONFIG_MMR_BASE_SHFT 28 +#define UV2H_RH_GAM_GRU_OVERLAY_CONFIG_MMR_BASE_MASK 0x00003ffff0000000UL +#define UV2H_RH_GAM_GRU_OVERLAY_CONFIG_MMR_N_GRU_SHFT 52 +#define UV2H_RH_GAM_GRU_OVERLAY_CONFIG_MMR_N_GRU_MASK 0x00f0000000000000UL +#define UV2H_RH_GAM_GRU_OVERLAY_CONFIG_MMR_ENABLE_SHFT 63 +#define UV2H_RH_GAM_GRU_OVERLAY_CONFIG_MMR_ENABLE_MASK 0x8000000000000000UL union uvh_rh_gam_gru_overlay_config_mmr_u { unsigned long v; struct uvh_rh_gam_gru_overlay_config_mmr_s { unsigned long rsvd_0_27: 28; /* */ unsigned long base : 18; /* RW */ + unsigned long rsvd_46_62 : 17; + unsigned long enable : 1; /* RW */ + } s; + struct uv1h_rh_gam_gru_overlay_config_mmr_s { + unsigned long rsvd_0_27: 28; /* */ + unsigned long base : 18; /* RW */ unsigned long rsvd_46_47: 2; /* */ unsigned long gr4 : 1; /* RW */ unsigned long rsvd_49_51: 3; /* */ unsigned long n_gru : 4; /* RW */ unsigned long rsvd_56_62: 7; /* */ unsigned long enable : 1; /* RW */ - } s; + } s1; + struct uv2h_rh_gam_gru_overlay_config_mmr_s { + unsigned long rsvd_0_27: 28; /* */ + unsigned long base : 18; /* RW */ + unsigned long rsvd_46_51: 6; /* */ + unsigned long n_gru : 4; /* RW */ + unsigned long rsvd_56_62: 7; /* */ + unsigned long enable : 1; /* RW */ + } s2; }; /* ========================================================================= */ @@ -1001,25 +1445,42 @@ union uvh_rh_gam_gru_overlay_config_mmr_u { /* ========================================================================= */ #define UVH_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR 0x1600030UL -#define UVH_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_BASE_SHFT 30 -#define UVH_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_BASE_MASK 0x00003fffc0000000UL -#define UVH_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_M_IO_SHFT 46 -#define UVH_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_M_IO_MASK 0x000fc00000000000UL -#define UVH_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_N_IO_SHFT 52 -#define UVH_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_N_IO_MASK 0x00f0000000000000UL -#define UVH_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_ENABLE_SHFT 63 -#define UVH_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_ENABLE_MASK 0x8000000000000000UL +#define UV1H_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_BASE_SHFT 30 +#define UV1H_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_BASE_MASK 0x00003fffc0000000UL +#define UV1H_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_M_IO_SHFT 46 +#define UV1H_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_M_IO_MASK 0x000fc00000000000UL +#define UV1H_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_N_IO_SHFT 52 +#define UV1H_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_N_IO_MASK 0x00f0000000000000UL +#define UV1H_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_ENABLE_SHFT 63 +#define UV1H_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_ENABLE_MASK 0x8000000000000000UL + +#define UV2H_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_BASE_SHFT 27 +#define UV2H_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_BASE_MASK 0x00003ffff8000000UL +#define UV2H_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_M_IO_SHFT 46 +#define UV2H_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_M_IO_MASK 0x000fc00000000000UL +#define UV2H_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_N_IO_SHFT 52 +#define UV2H_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_N_IO_MASK 0x00f0000000000000UL +#define UV2H_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_ENABLE_SHFT 63 +#define UV2H_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_ENABLE_MASK 0x8000000000000000UL union uvh_rh_gam_mmioh_overlay_config_mmr_u { unsigned long v; - struct uvh_rh_gam_mmioh_overlay_config_mmr_s { + struct uv1h_rh_gam_mmioh_overlay_config_mmr_s { unsigned long rsvd_0_29: 30; /* */ unsigned long base : 16; /* RW */ unsigned long m_io : 6; /* RW */ unsigned long n_io : 4; /* RW */ unsigned long rsvd_56_62: 7; /* */ unsigned long enable : 1; /* RW */ - } s; + } s1; + struct uv2h_rh_gam_mmioh_overlay_config_mmr_s { + unsigned long rsvd_0_26: 27; /* */ + unsigned long base : 19; /* RW */ + unsigned long m_io : 6; /* RW */ + unsigned long n_io : 4; /* RW */ + unsigned long rsvd_56_62: 7; /* */ + unsigned long enable : 1; /* RW */ + } s2; }; /* ========================================================================= */ @@ -1029,20 +1490,40 @@ union uvh_rh_gam_mmioh_overlay_config_mmr_u { #define UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR_BASE_SHFT 26 #define UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR_BASE_MASK 0x00003ffffc000000UL -#define UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR_DUAL_HUB_SHFT 46 -#define UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR_DUAL_HUB_MASK 0x0000400000000000UL -#define UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR_ENABLE_SHFT 63 -#define UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR_ENABLE_MASK 0x8000000000000000UL + +#define UV1H_RH_GAM_MMR_OVERLAY_CONFIG_MMR_BASE_SHFT 26 +#define UV1H_RH_GAM_MMR_OVERLAY_CONFIG_MMR_BASE_MASK 0x00003ffffc000000UL +#define UV1H_RH_GAM_MMR_OVERLAY_CONFIG_MMR_DUAL_HUB_SHFT 46 +#define UV1H_RH_GAM_MMR_OVERLAY_CONFIG_MMR_DUAL_HUB_MASK 0x0000400000000000UL +#define UV1H_RH_GAM_MMR_OVERLAY_CONFIG_MMR_ENABLE_SHFT 63 +#define UV1H_RH_GAM_MMR_OVERLAY_CONFIG_MMR_ENABLE_MASK 0x8000000000000000UL + +#define UV2H_RH_GAM_MMR_OVERLAY_CONFIG_MMR_BASE_SHFT 26 +#define UV2H_RH_GAM_MMR_OVERLAY_CONFIG_MMR_BASE_MASK 0x00003ffffc000000UL +#define UV2H_RH_GAM_MMR_OVERLAY_CONFIG_MMR_ENABLE_SHFT 63 +#define UV2H_RH_GAM_MMR_OVERLAY_CONFIG_MMR_ENABLE_MASK 0x8000000000000000UL union uvh_rh_gam_mmr_overlay_config_mmr_u { unsigned long v; struct uvh_rh_gam_mmr_overlay_config_mmr_s { unsigned long rsvd_0_25: 26; /* */ unsigned long base : 20; /* RW */ + unsigned long rsvd_46_62 : 17; + unsigned long enable : 1; /* RW */ + } s; + struct uv1h_rh_gam_mmr_overlay_config_mmr_s { + unsigned long rsvd_0_25: 26; /* */ + unsigned long base : 20; /* RW */ unsigned long dual_hub : 1; /* RW */ unsigned long rsvd_47_62: 16; /* */ unsigned long enable : 1; /* RW */ - } s; + } s1; + struct uv2h_rh_gam_mmr_overlay_config_mmr_s { + unsigned long rsvd_0_25: 26; /* */ + unsigned long base : 20; /* RW */ + unsigned long rsvd_46_62: 17; /* */ + unsigned long enable : 1; /* RW */ + } s2; }; /* ========================================================================= */ @@ -1103,10 +1584,11 @@ union uvh_rtc1_int_config_u { /* UVH_SCRATCH5 */ /* ========================================================================= */ #define UVH_SCRATCH5 0x2d0200UL -#define UVH_SCRATCH5_32 0x00778 +#define UVH_SCRATCH5_32 0x778 #define UVH_SCRATCH5_SCRATCH5_SHFT 0 #define UVH_SCRATCH5_SCRATCH5_MASK 0xffffffffffffffffUL + union uvh_scratch5_u { unsigned long v; struct uvh_scratch5_s { @@ -1114,4 +1596,154 @@ union uvh_scratch5_u { } s; }; +/* ========================================================================= */ +/* UV2H_EVENT_OCCURRED2 */ +/* ========================================================================= */ +#define UV2H_EVENT_OCCURRED2 0x70100UL +#define UV2H_EVENT_OCCURRED2_32 0xb68 + +#define UV2H_EVENT_OCCURRED2_RTC_0_SHFT 0 +#define UV2H_EVENT_OCCURRED2_RTC_0_MASK 0x0000000000000001UL +#define UV2H_EVENT_OCCURRED2_RTC_1_SHFT 1 +#define UV2H_EVENT_OCCURRED2_RTC_1_MASK 0x0000000000000002UL +#define UV2H_EVENT_OCCURRED2_RTC_2_SHFT 2 +#define UV2H_EVENT_OCCURRED2_RTC_2_MASK 0x0000000000000004UL +#define UV2H_EVENT_OCCURRED2_RTC_3_SHFT 3 +#define UV2H_EVENT_OCCURRED2_RTC_3_MASK 0x0000000000000008UL +#define UV2H_EVENT_OCCURRED2_RTC_4_SHFT 4 +#define UV2H_EVENT_OCCURRED2_RTC_4_MASK 0x0000000000000010UL +#define UV2H_EVENT_OCCURRED2_RTC_5_SHFT 5 +#define UV2H_EVENT_OCCURRED2_RTC_5_MASK 0x0000000000000020UL +#define UV2H_EVENT_OCCURRED2_RTC_6_SHFT 6 +#define UV2H_EVENT_OCCURRED2_RTC_6_MASK 0x0000000000000040UL +#define UV2H_EVENT_OCCURRED2_RTC_7_SHFT 7 +#define UV2H_EVENT_OCCURRED2_RTC_7_MASK 0x0000000000000080UL +#define UV2H_EVENT_OCCURRED2_RTC_8_SHFT 8 +#define UV2H_EVENT_OCCURRED2_RTC_8_MASK 0x0000000000000100UL +#define UV2H_EVENT_OCCURRED2_RTC_9_SHFT 9 +#define UV2H_EVENT_OCCURRED2_RTC_9_MASK 0x0000000000000200UL +#define UV2H_EVENT_OCCURRED2_RTC_10_SHFT 10 +#define UV2H_EVENT_OCCURRED2_RTC_10_MASK 0x0000000000000400UL +#define UV2H_EVENT_OCCURRED2_RTC_11_SHFT 11 +#define UV2H_EVENT_OCCURRED2_RTC_11_MASK 0x0000000000000800UL +#define UV2H_EVENT_OCCURRED2_RTC_12_SHFT 12 +#define UV2H_EVENT_OCCURRED2_RTC_12_MASK 0x0000000000001000UL +#define UV2H_EVENT_OCCURRED2_RTC_13_SHFT 13 +#define UV2H_EVENT_OCCURRED2_RTC_13_MASK 0x0000000000002000UL +#define UV2H_EVENT_OCCURRED2_RTC_14_SHFT 14 +#define UV2H_EVENT_OCCURRED2_RTC_14_MASK 0x0000000000004000UL +#define UV2H_EVENT_OCCURRED2_RTC_15_SHFT 15 +#define UV2H_EVENT_OCCURRED2_RTC_15_MASK 0x0000000000008000UL +#define UV2H_EVENT_OCCURRED2_RTC_16_SHFT 16 +#define UV2H_EVENT_OCCURRED2_RTC_16_MASK 0x0000000000010000UL +#define UV2H_EVENT_OCCURRED2_RTC_17_SHFT 17 +#define UV2H_EVENT_OCCURRED2_RTC_17_MASK 0x0000000000020000UL +#define UV2H_EVENT_OCCURRED2_RTC_18_SHFT 18 +#define UV2H_EVENT_OCCURRED2_RTC_18_MASK 0x0000000000040000UL +#define UV2H_EVENT_OCCURRED2_RTC_19_SHFT 19 +#define UV2H_EVENT_OCCURRED2_RTC_19_MASK 0x0000000000080000UL +#define UV2H_EVENT_OCCURRED2_RTC_20_SHFT 20 +#define UV2H_EVENT_OCCURRED2_RTC_20_MASK 0x0000000000100000UL +#define UV2H_EVENT_OCCURRED2_RTC_21_SHFT 21 +#define UV2H_EVENT_OCCURRED2_RTC_21_MASK 0x0000000000200000UL +#define UV2H_EVENT_OCCURRED2_RTC_22_SHFT 22 +#define UV2H_EVENT_OCCURRED2_RTC_22_MASK 0x0000000000400000UL +#define UV2H_EVENT_OCCURRED2_RTC_23_SHFT 23 +#define UV2H_EVENT_OCCURRED2_RTC_23_MASK 0x0000000000800000UL +#define UV2H_EVENT_OCCURRED2_RTC_24_SHFT 24 +#define UV2H_EVENT_OCCURRED2_RTC_24_MASK 0x0000000001000000UL +#define UV2H_EVENT_OCCURRED2_RTC_25_SHFT 25 +#define UV2H_EVENT_OCCURRED2_RTC_25_MASK 0x0000000002000000UL +#define UV2H_EVENT_OCCURRED2_RTC_26_SHFT 26 +#define UV2H_EVENT_OCCURRED2_RTC_26_MASK 0x0000000004000000UL +#define UV2H_EVENT_OCCURRED2_RTC_27_SHFT 27 +#define UV2H_EVENT_OCCURRED2_RTC_27_MASK 0x0000000008000000UL +#define UV2H_EVENT_OCCURRED2_RTC_28_SHFT 28 +#define UV2H_EVENT_OCCURRED2_RTC_28_MASK 0x0000000010000000UL +#define UV2H_EVENT_OCCURRED2_RTC_29_SHFT 29 +#define UV2H_EVENT_OCCURRED2_RTC_29_MASK 0x0000000020000000UL +#define UV2H_EVENT_OCCURRED2_RTC_30_SHFT 30 +#define UV2H_EVENT_OCCURRED2_RTC_30_MASK 0x0000000040000000UL +#define UV2H_EVENT_OCCURRED2_RTC_31_SHFT 31 +#define UV2H_EVENT_OCCURRED2_RTC_31_MASK 0x0000000080000000UL + +union uv2h_event_occurred2_u { + unsigned long v; + struct uv2h_event_occurred2_s { + unsigned long rtc_0 : 1; /* RW */ + unsigned long rtc_1 : 1; /* RW */ + unsigned long rtc_2 : 1; /* RW */ + unsigned long rtc_3 : 1; /* RW */ + unsigned long rtc_4 : 1; /* RW */ + unsigned long rtc_5 : 1; /* RW */ + unsigned long rtc_6 : 1; /* RW */ + unsigned long rtc_7 : 1; /* RW */ + unsigned long rtc_8 : 1; /* RW */ + unsigned long rtc_9 : 1; /* RW */ + unsigned long rtc_10 : 1; /* RW */ + unsigned long rtc_11 : 1; /* RW */ + unsigned long rtc_12 : 1; /* RW */ + unsigned long rtc_13 : 1; /* RW */ + unsigned long rtc_14 : 1; /* RW */ + unsigned long rtc_15 : 1; /* RW */ + unsigned long rtc_16 : 1; /* RW */ + unsigned long rtc_17 : 1; /* RW */ + unsigned long rtc_18 : 1; /* RW */ + unsigned long rtc_19 : 1; /* RW */ + unsigned long rtc_20 : 1; /* RW */ + unsigned long rtc_21 : 1; /* RW */ + unsigned long rtc_22 : 1; /* RW */ + unsigned long rtc_23 : 1; /* RW */ + unsigned long rtc_24 : 1; /* RW */ + unsigned long rtc_25 : 1; /* RW */ + unsigned long rtc_26 : 1; /* RW */ + unsigned long rtc_27 : 1; /* RW */ + unsigned long rtc_28 : 1; /* RW */ + unsigned long rtc_29 : 1; /* RW */ + unsigned long rtc_30 : 1; /* RW */ + unsigned long rtc_31 : 1; /* RW */ + unsigned long rsvd_32_63: 32; /* */ + } s1; +}; + +/* ========================================================================= */ +/* UV2H_EVENT_OCCURRED2_ALIAS */ +/* ========================================================================= */ +#define UV2H_EVENT_OCCURRED2_ALIAS 0x70108UL +#define UV2H_EVENT_OCCURRED2_ALIAS_32 0xb70 + +/* ========================================================================= */ +/* UV2H_LB_BAU_SB_ACTIVATION_STATUS_2 */ +/* ========================================================================= */ +#define UV2H_LB_BAU_SB_ACTIVATION_STATUS_2 0x320130UL +#define UV2H_LB_BAU_SB_ACTIVATION_STATUS_2_32 0x9f0 + +#define UV2H_LB_BAU_SB_ACTIVATION_STATUS_2_AUX_ERROR_SHFT 0 +#define UV2H_LB_BAU_SB_ACTIVATION_STATUS_2_AUX_ERROR_MASK 0xffffffffffffffffUL + +union uv2h_lb_bau_sb_activation_status_2_u { + unsigned long v; + struct uv2h_lb_bau_sb_activation_status_2_s { + unsigned long aux_error : 64; /* RW */ + } s1; +}; + +/* ========================================================================= */ +/* UV1H_LB_TARGET_PHYSICAL_APIC_ID_MASK */ +/* ========================================================================= */ +#define UV1H_LB_TARGET_PHYSICAL_APIC_ID_MASK 0x320130UL +#define UV1H_LB_TARGET_PHYSICAL_APIC_ID_MASK_32 0x9f0 + +#define UV1H_LB_TARGET_PHYSICAL_APIC_ID_MASK_BIT_ENABLES_SHFT 0 +#define UV1H_LB_TARGET_PHYSICAL_APIC_ID_MASK_BIT_ENABLES_MASK 0x00000000ffffffffUL + +union uv1h_lb_target_physical_apic_id_mask_u { + unsigned long v; + struct uv1h_lb_target_physical_apic_id_mask_s { + unsigned long bit_enables : 32; /* RW */ + unsigned long rsvd_32_63 : 32; /* */ + } s1; +}; + + #endif /* __ASM_UV_MMRS_X86_H__ */ diff --git a/arch/x86/include/asm/vdso.h b/arch/x86/include/asm/vdso.h index 9064052b73d..bb0522850b7 100644 --- a/arch/x86/include/asm/vdso.h +++ b/arch/x86/include/asm/vdso.h @@ -1,20 +1,6 @@ #ifndef _ASM_X86_VDSO_H #define _ASM_X86_VDSO_H -#ifdef CONFIG_X86_64 -extern const char VDSO64_PRELINK[]; - -/* - * Given a pointer to the vDSO image, find the pointer to VDSO64_name - * as that symbol is defined in the vDSO sources or linker script. - */ -#define VDSO64_SYMBOL(base, name) \ -({ \ - extern const char VDSO64_##name[]; \ - (void *)(VDSO64_##name - VDSO64_PRELINK + (unsigned long)(base)); \ -}) -#endif - #if defined CONFIG_X86_32 || defined CONFIG_COMPAT extern const char VDSO32_PRELINK[]; diff --git a/arch/x86/include/asm/vgtod.h b/arch/x86/include/asm/vgtod.h index 3d61e204826..646b4c1ca69 100644 --- a/arch/x86/include/asm/vgtod.h +++ b/arch/x86/include/asm/vgtod.h @@ -23,8 +23,6 @@ struct vsyscall_gtod_data { struct timespec wall_to_monotonic; struct timespec wall_time_coarse; }; -extern struct vsyscall_gtod_data __vsyscall_gtod_data -__section_vsyscall_gtod_data; extern struct vsyscall_gtod_data vsyscall_gtod_data; #endif /* _ASM_X86_VGTOD_H */ diff --git a/arch/x86/include/asm/vsyscall.h b/arch/x86/include/asm/vsyscall.h index d0983d255fb..d55597351f6 100644 --- a/arch/x86/include/asm/vsyscall.h +++ b/arch/x86/include/asm/vsyscall.h @@ -16,27 +16,19 @@ enum vsyscall_num { #ifdef __KERNEL__ #include <linux/seqlock.h> -#define __section_vgetcpu_mode __attribute__ ((unused, __section__ (".vgetcpu_mode"), aligned(16))) -#define __section_jiffies __attribute__ ((unused, __section__ (".jiffies"), aligned(16))) - /* Definitions for CONFIG_GENERIC_TIME definitions */ -#define __section_vsyscall_gtod_data __attribute__ \ - ((unused, __section__ (".vsyscall_gtod_data"),aligned(16))) -#define __section_vsyscall_clock __attribute__ \ - ((unused, __section__ (".vsyscall_clock"),aligned(16))) #define __vsyscall_fn \ __attribute__ ((unused, __section__(".vsyscall_fn"))) notrace #define VGETCPU_RDTSCP 1 #define VGETCPU_LSL 2 -extern int __vgetcpu_mode; -extern volatile unsigned long __jiffies; - /* kernel space (writeable) */ extern int vgetcpu_mode; extern struct timezone sys_tz; +#include <asm/vvar.h> + extern void map_vsyscall(void); #endif /* __KERNEL__ */ diff --git a/arch/x86/include/asm/vvar.h b/arch/x86/include/asm/vvar.h new file mode 100644 index 00000000000..341b3559452 --- /dev/null +++ b/arch/x86/include/asm/vvar.h @@ -0,0 +1,52 @@ +/* + * vvar.h: Shared vDSO/kernel variable declarations + * Copyright (c) 2011 Andy Lutomirski + * Subject to the GNU General Public License, version 2 + * + * A handful of variables are accessible (read-only) from userspace + * code in the vsyscall page and the vdso. They are declared here. + * Some other file must define them with DEFINE_VVAR. + * + * In normal kernel code, they are used like any other variable. + * In user code, they are accessed through the VVAR macro. + * + * Each of these variables lives in the vsyscall page, and each + * one needs a unique offset within the little piece of the page + * reserved for vvars. Specify that offset in DECLARE_VVAR. + * (There are 896 bytes available. If you mess up, the linker will + * catch it.) + */ + +/* Offset of vars within vsyscall page */ +#define VSYSCALL_VARS_OFFSET (3072 + 128) + +#if defined(__VVAR_KERNEL_LDS) + +/* The kernel linker script defines its own magic to put vvars in the + * right place. + */ +#define DECLARE_VVAR(offset, type, name) \ + EMIT_VVAR(name, VSYSCALL_VARS_OFFSET + offset) + +#else + +#define DECLARE_VVAR(offset, type, name) \ + static type const * const vvaraddr_ ## name = \ + (void *)(VSYSCALL_START + VSYSCALL_VARS_OFFSET + (offset)); + +#define DEFINE_VVAR(type, name) \ + type __vvar_ ## name \ + __attribute__((section(".vsyscall_var_" #name), aligned(16))) + +#define VVAR(name) (*vvaraddr_ ## name) + +#endif + +/* DECLARE_VVAR(offset, type, name) */ + +DECLARE_VVAR(0, volatile unsigned long, jiffies) +DECLARE_VVAR(8, int, vgetcpu_mode) +DECLARE_VVAR(128, struct vsyscall_gtod_data, vsyscall_gtod_data) + +#undef DECLARE_VVAR +#undef VSYSCALL_VARS_OFFSET diff --git a/arch/x86/include/asm/xen/hypercall.h b/arch/x86/include/asm/xen/hypercall.h index 8508bfe5229..d240ea95051 100644 --- a/arch/x86/include/asm/xen/hypercall.h +++ b/arch/x86/include/asm/xen/hypercall.h @@ -447,6 +447,13 @@ HYPERVISOR_hvm_op(int op, void *arg) return _hypercall2(unsigned long, hvm_op, op, arg); } +static inline int +HYPERVISOR_tmem_op( + struct tmem_op *op) +{ + return _hypercall1(int, tmem_op, op); +} + static inline void MULTI_fpu_taskswitch(struct multicall_entry *mcl, int set) { diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 250806472a7..90b06d4daee 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -24,13 +24,17 @@ endif nostackp := $(call cc-option, -fno-stack-protector) CFLAGS_vsyscall_64.o := $(PROFILING) -g0 $(nostackp) CFLAGS_hpet.o := $(nostackp) -CFLAGS_tsc.o := $(nostackp) +CFLAGS_vread_tsc_64.o := $(nostackp) CFLAGS_paravirt.o := $(nostackp) GCOV_PROFILE_vsyscall_64.o := n GCOV_PROFILE_hpet.o := n GCOV_PROFILE_tsc.o := n +GCOV_PROFILE_vread_tsc_64.o := n GCOV_PROFILE_paravirt.o := n +# vread_tsc_64 is hot and should be fully optimized: +CFLAGS_REMOVE_vread_tsc_64.o = -pg -fno-optimize-sibling-calls + obj-y := process_$(BITS).o signal.o entry_$(BITS).o obj-y += traps.o irq.o irq_$(BITS).o dumpstack_$(BITS).o obj-y += time.o ioport.o ldt.o dumpstack.o @@ -39,7 +43,7 @@ obj-$(CONFIG_IRQ_WORK) += irq_work.o obj-y += probe_roms.o obj-$(CONFIG_X86_32) += sys_i386_32.o i386_ksyms_32.o obj-$(CONFIG_X86_64) += sys_x86_64.o x8664_ksyms_64.o -obj-$(CONFIG_X86_64) += syscall_64.o vsyscall_64.o +obj-$(CONFIG_X86_64) += syscall_64.o vsyscall_64.o vread_tsc_64.o obj-y += bootflag.o e820.o obj-y += pci-dma.o quirks.o topology.o kdebugfs.o obj-y += alternative.o i8253.o pci-nommu.o hw_breakpoint.o diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index cd8cbeb5fa3..7c3a95e54ec 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -30,6 +30,7 @@ #include <asm/proto.h> #include <asm/iommu.h> #include <asm/gart.h> +#include <asm/dma.h> #include <asm/amd_iommu_proto.h> #include <asm/amd_iommu_types.h> #include <asm/amd_iommu.h> @@ -154,6 +155,10 @@ static int iommu_init_device(struct device *dev) pdev = pci_get_bus_and_slot(PCI_BUS(alias), alias & 0xff); if (pdev) dev_data->alias = &pdev->dev; + else { + kfree(dev_data); + return -ENOTSUPP; + } atomic_set(&dev_data->bind, 0); @@ -163,6 +168,20 @@ static int iommu_init_device(struct device *dev) return 0; } +static void iommu_ignore_device(struct device *dev) +{ + u16 devid, alias; + + devid = get_device_id(dev); + alias = amd_iommu_alias_table[devid]; + + memset(&amd_iommu_dev_table[devid], 0, sizeof(struct dev_table_entry)); + memset(&amd_iommu_dev_table[alias], 0, sizeof(struct dev_table_entry)); + + amd_iommu_rlookup_table[devid] = NULL; + amd_iommu_rlookup_table[alias] = NULL; +} + static void iommu_uninit_device(struct device *dev) { kfree(dev->archdata.iommu); @@ -192,7 +211,9 @@ int __init amd_iommu_init_devices(void) continue; ret = iommu_init_device(&pdev->dev); - if (ret) + if (ret == -ENOTSUPP) + iommu_ignore_device(&pdev->dev); + else if (ret) goto out_free; } @@ -2383,6 +2404,23 @@ static struct dma_map_ops amd_iommu_dma_ops = { .dma_supported = amd_iommu_dma_supported, }; +static unsigned device_dma_ops_init(void) +{ + struct pci_dev *pdev = NULL; + unsigned unhandled = 0; + + for_each_pci_dev(pdev) { + if (!check_device(&pdev->dev)) { + unhandled += 1; + continue; + } + + pdev->dev.archdata.dma_ops = &amd_iommu_dma_ops; + } + + return unhandled; +} + /* * The function which clues the AMD IOMMU driver into dma_ops. */ @@ -2395,7 +2433,7 @@ void __init amd_iommu_init_api(void) int __init amd_iommu_init_dma_ops(void) { struct amd_iommu *iommu; - int ret; + int ret, unhandled; /* * first allocate a default protection domain for every IOMMU we @@ -2421,7 +2459,11 @@ int __init amd_iommu_init_dma_ops(void) swiotlb = 0; /* Make the driver finally visible to the drivers */ - dma_ops = &amd_iommu_dma_ops; + unhandled = device_dma_ops_init(); + if (unhandled && max_pfn > MAX_DMA32_PFN) { + /* There are unhandled devices - initialize swiotlb for them */ + swiotlb = 1; + } amd_iommu_stats_init(); diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index 9179c21120a..bfc8453bd98 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -731,8 +731,8 @@ static void __init init_iommu_from_acpi(struct amd_iommu *iommu, { u8 *p = (u8 *)h; u8 *end = p, flags = 0; - u16 dev_i, devid = 0, devid_start = 0, devid_to = 0; - u32 ext_flags = 0; + u16 devid = 0, devid_start = 0, devid_to = 0; + u32 dev_i, ext_flags = 0; bool alias = false; struct ivhd_entry *e; @@ -887,7 +887,7 @@ static void __init init_iommu_from_acpi(struct amd_iommu *iommu, /* Initializes the device->iommu mapping for the driver */ static int __init init_iommu_devices(struct amd_iommu *iommu) { - u16 i; + u32 i; for (i = iommu->first_device; i <= iommu->last_device; ++i) set_iommu_for_device(iommu, i); @@ -1177,7 +1177,7 @@ static int __init init_memory_definitions(struct acpi_table_header *table) */ static void init_device_table(void) { - u16 devid; + u32 devid; for (devid = 0; devid <= amd_iommu_last_bdf; ++devid) { set_dev_entry_bit(devid, DEV_ENTRY_VALID); diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 9488dcff7ae..e5293394b54 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -676,7 +676,7 @@ void mask_ioapic_entries(void) int apic, pin; for (apic = 0; apic < nr_ioapics; apic++) { - if (ioapics[apic].saved_registers) + if (!ioapics[apic].saved_registers) continue; for (pin = 0; pin < ioapics[apic].nr_registers; pin++) { @@ -699,7 +699,7 @@ int restore_ioapic_entries(void) int apic, pin; for (apic = 0; apic < nr_ioapics; apic++) { - if (ioapics[apic].saved_registers) + if (!ioapics[apic].saved_registers) continue; for (pin = 0; pin < ioapics[apic].nr_registers; pin++) diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c index f450b683dfc..b511a011b7d 100644 --- a/arch/x86/kernel/apic/x2apic_uv_x.c +++ b/arch/x86/kernel/apic/x2apic_uv_x.c @@ -91,6 +91,10 @@ static int __init early_get_pnodeid(void) m_n_config.v = uv_early_read_mmr(UVH_RH_GAM_CONFIG_MMR); uv_min_hub_revision_id = node_id.s.revision; + if (node_id.s.part_number == UV2_HUB_PART_NUMBER) + uv_min_hub_revision_id += UV2_HUB_REVISION_BASE - 1; + + uv_hub_info->hub_revision = uv_min_hub_revision_id; pnode = (node_id.s.node_id >> 1) & ((1 << m_n_config.s.n_skt) - 1); return pnode; } @@ -112,17 +116,25 @@ static void __init early_get_apic_pnode_shift(void) */ static void __init uv_set_apicid_hibit(void) { - union uvh_lb_target_physical_apic_id_mask_u apicid_mask; + union uv1h_lb_target_physical_apic_id_mask_u apicid_mask; - apicid_mask.v = uv_early_read_mmr(UVH_LB_TARGET_PHYSICAL_APIC_ID_MASK); - uv_apicid_hibits = apicid_mask.s.bit_enables & UV_APICID_HIBIT_MASK; + if (is_uv1_hub()) { + apicid_mask.v = + uv_early_read_mmr(UV1H_LB_TARGET_PHYSICAL_APIC_ID_MASK); + uv_apicid_hibits = + apicid_mask.s1.bit_enables & UV_APICID_HIBIT_MASK; + } } static int __init uv_acpi_madt_oem_check(char *oem_id, char *oem_table_id) { - int pnodeid; + int pnodeid, is_uv1, is_uv2; - if (!strcmp(oem_id, "SGI")) { + is_uv1 = !strcmp(oem_id, "SGI"); + is_uv2 = !strcmp(oem_id, "SGI2"); + if (is_uv1 || is_uv2) { + uv_hub_info->hub_revision = + is_uv1 ? UV1_HUB_REVISION_BASE : UV2_HUB_REVISION_BASE; pnodeid = early_get_pnodeid(); early_get_apic_pnode_shift(); x86_platform.is_untracked_pat_range = uv_is_untracked_pat_range; @@ -484,12 +496,19 @@ static __init void map_mmr_high(int max_pnode) static __init void map_mmioh_high(int max_pnode) { union uvh_rh_gam_mmioh_overlay_config_mmr_u mmioh; - int shift = UVH_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_BASE_SHFT; + int shift; mmioh.v = uv_read_local_mmr(UVH_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR); - if (mmioh.s.enable) - map_high("MMIOH", mmioh.s.base, shift, mmioh.s.m_io, + if (is_uv1_hub() && mmioh.s1.enable) { + shift = UV1H_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_BASE_SHFT; + map_high("MMIOH", mmioh.s1.base, shift, mmioh.s1.m_io, + max_pnode, map_uc); + } + if (is_uv2_hub() && mmioh.s2.enable) { + shift = UV2H_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_BASE_SHFT; + map_high("MMIOH", mmioh.s2.base, shift, mmioh.s2.m_io, max_pnode, map_uc); + } } static __init void map_low_mmrs(void) @@ -736,13 +755,14 @@ void __init uv_system_init(void) unsigned long mmr_base, present, paddr; unsigned short pnode_mask, pnode_io_mask; + printk(KERN_INFO "UV: Found %s hub\n", is_uv1_hub() ? "UV1" : "UV2"); map_low_mmrs(); m_n_config.v = uv_read_local_mmr(UVH_RH_GAM_CONFIG_MMR ); m_val = m_n_config.s.m_skt; n_val = m_n_config.s.n_skt; mmioh.v = uv_read_local_mmr(UVH_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR); - n_io = mmioh.s.n_io; + n_io = is_uv1_hub() ? mmioh.s1.n_io : mmioh.s2.n_io; mmr_base = uv_read_local_mmr(UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR) & ~UV_MMR_ENABLE; @@ -811,6 +831,8 @@ void __init uv_system_init(void) */ uv_cpu_hub_info(cpu)->pnode_mask = pnode_mask; uv_cpu_hub_info(cpu)->apic_pnode_shift = uvh_apicid.s.pnode_shift; + uv_cpu_hub_info(cpu)->hub_revision = uv_hub_info->hub_revision; + pnode = uv_apicid_to_pnode(apicid); blade = boot_pnode_to_blade(pnode); lcpu = uv_blade_info[blade].nr_possible_cpus; diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c index 3bfa0223596..965a7666c28 100644 --- a/arch/x86/kernel/apm_32.c +++ b/arch/x86/kernel/apm_32.c @@ -361,6 +361,7 @@ struct apm_user { * idle percentage above which bios idle calls are done */ #ifdef CONFIG_APM_CPU_IDLE +#warning deprecated CONFIG_APM_CPU_IDLE will be deleted in 2012 #define DEFAULT_IDLE_THRESHOLD 95 #else #define DEFAULT_IDLE_THRESHOLD 100 @@ -904,6 +905,7 @@ static void apm_cpu_idle(void) unsigned int jiffies_since_last_check = jiffies - last_jiffies; unsigned int bucket; + WARN_ONCE(1, "deprecated apm_cpu_idle will be deleted in 2012"); recalc: if (jiffies_since_last_check > IDLE_CALC_LIMIT) { use_apm_idle = 0; diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 8f5cabb3c5b..b13ed393dfc 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -612,8 +612,11 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) } #endif - /* As a rule processors have APIC timer running in deep C states */ - if (c->x86 > 0xf && !cpu_has_amd_erratum(amd_erratum_400)) + /* + * Family 0x12 and above processors have APIC timer + * running in deep C states. + */ + if (c->x86 > 0x11) set_cpu_cap(c, X86_FEATURE_ARAT); /* diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index c39576cb301..525514cf33c 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -19,6 +19,7 @@ static int __init no_halt(char *s) { + WARN_ONCE(1, "\"no-hlt\" is deprecated, please use \"idle=poll\"\n"); boot_cpu_data.hlt_works_ok = 0; return 1; } diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index c8b41623377..22a073d7fbf 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -477,13 +477,6 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c) if (smp_num_siblings <= 1) goto out; - if (smp_num_siblings > nr_cpu_ids) { - pr_warning("CPU: Unsupported number of siblings %d", - smp_num_siblings); - smp_num_siblings = 1; - return; - } - index_msb = get_count_order(smp_num_siblings); c->phys_proc_id = apic->phys_pkg_id(c->initial_apicid, index_msb); @@ -909,7 +902,7 @@ static void vgetcpu_set_mode(void) void __init identify_boot_cpu(void) { identify_cpu(&boot_cpu_data); - init_c1e_mask(); + init_amd_e400_c1e_mask(); #ifdef CONFIG_X86_32 sysenter_setup(); enable_sep_cpu(); diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c index 0ba15a6cc57..c9a281f272f 100644 --- a/arch/x86/kernel/ftrace.c +++ b/arch/x86/kernel/ftrace.c @@ -123,7 +123,7 @@ static unsigned char *ftrace_call_replace(unsigned long ip, unsigned long addr) static atomic_t nmi_running = ATOMIC_INIT(0); static int mod_code_status; /* holds return value of text write */ static void *mod_code_ip; /* holds the IP to write to */ -static void *mod_code_newcode; /* holds the text to write to the IP */ +static const void *mod_code_newcode; /* holds the text to write to the IP */ static unsigned nmi_wait_count; static atomic_t nmi_update_count = ATOMIC_INIT(0); @@ -225,7 +225,7 @@ within(unsigned long addr, unsigned long start, unsigned long end) } static int -do_ftrace_mod_code(unsigned long ip, void *new_code) +do_ftrace_mod_code(unsigned long ip, const void *new_code) { /* * On x86_64, kernel text mappings are mapped read-only with @@ -266,8 +266,8 @@ static const unsigned char *ftrace_nop_replace(void) } static int -ftrace_modify_code(unsigned long ip, unsigned char *old_code, - unsigned char *new_code) +ftrace_modify_code(unsigned long ip, unsigned const char *old_code, + unsigned const char *new_code) { unsigned char replaced[MCOUNT_INSN_SIZE]; @@ -301,7 +301,7 @@ ftrace_modify_code(unsigned long ip, unsigned char *old_code, int ftrace_make_nop(struct module *mod, struct dyn_ftrace *rec, unsigned long addr) { - unsigned char *new, *old; + unsigned const char *new, *old; unsigned long ip = rec->ip; old = ftrace_call_replace(ip, addr); @@ -312,7 +312,7 @@ int ftrace_make_nop(struct module *mod, int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr) { - unsigned char *new, *old; + unsigned const char *new, *old; unsigned long ip = rec->ip; old = ftrace_nop_replace(); diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 88a90a977f8..2e4928d45a2 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -337,7 +337,9 @@ EXPORT_SYMBOL(boot_option_idle_override); * Powermanagement idle function, if any.. */ void (*pm_idle)(void); +#if defined(CONFIG_APM_MODULE) && defined(CONFIG_APM_CPU_IDLE) EXPORT_SYMBOL(pm_idle); +#endif #ifdef CONFIG_X86_32 /* @@ -397,7 +399,7 @@ void default_idle(void) cpu_relax(); } } -#ifdef CONFIG_APM_MODULE +#if defined(CONFIG_APM_MODULE) && defined(CONFIG_APM_CPU_IDLE) EXPORT_SYMBOL(default_idle); #endif @@ -535,45 +537,45 @@ int mwait_usable(const struct cpuinfo_x86 *c) return (edx & MWAIT_EDX_C1); } -bool c1e_detected; -EXPORT_SYMBOL(c1e_detected); +bool amd_e400_c1e_detected; +EXPORT_SYMBOL(amd_e400_c1e_detected); -static cpumask_var_t c1e_mask; +static cpumask_var_t amd_e400_c1e_mask; -void c1e_remove_cpu(int cpu) +void amd_e400_remove_cpu(int cpu) { - if (c1e_mask != NULL) - cpumask_clear_cpu(cpu, c1e_mask); + if (amd_e400_c1e_mask != NULL) + cpumask_clear_cpu(cpu, amd_e400_c1e_mask); } /* - * C1E aware idle routine. We check for C1E active in the interrupt + * AMD Erratum 400 aware idle routine. We check for C1E active in the interrupt * pending message MSR. If we detect C1E, then we handle it the same * way as C3 power states (local apic timer and TSC stop) */ -static void c1e_idle(void) +static void amd_e400_idle(void) { if (need_resched()) return; - if (!c1e_detected) { + if (!amd_e400_c1e_detected) { u32 lo, hi; rdmsr(MSR_K8_INT_PENDING_MSG, lo, hi); if (lo & K8_INTP_C1E_ACTIVE_MASK) { - c1e_detected = true; + amd_e400_c1e_detected = true; if (!boot_cpu_has(X86_FEATURE_NONSTOP_TSC)) mark_tsc_unstable("TSC halt in AMD C1E"); printk(KERN_INFO "System has AMD C1E enabled\n"); } } - if (c1e_detected) { + if (amd_e400_c1e_detected) { int cpu = smp_processor_id(); - if (!cpumask_test_cpu(cpu, c1e_mask)) { - cpumask_set_cpu(cpu, c1e_mask); + if (!cpumask_test_cpu(cpu, amd_e400_c1e_mask)) { + cpumask_set_cpu(cpu, amd_e400_c1e_mask); /* * Force broadcast so ACPI can not interfere. */ @@ -616,17 +618,17 @@ void __cpuinit select_idle_routine(const struct cpuinfo_x86 *c) pm_idle = mwait_idle; } else if (cpu_has_amd_erratum(amd_erratum_400)) { /* E400: APIC timer interrupt does not wake up CPU from C1e */ - printk(KERN_INFO "using C1E aware idle routine\n"); - pm_idle = c1e_idle; + printk(KERN_INFO "using AMD E400 aware idle routine\n"); + pm_idle = amd_e400_idle; } else pm_idle = default_idle; } -void __init init_c1e_mask(void) +void __init init_amd_e400_c1e_mask(void) { - /* If we're using c1e_idle, we need to allocate c1e_mask. */ - if (pm_idle == c1e_idle) - zalloc_cpumask_var(&c1e_mask, GFP_KERNEL); + /* If we're using amd_e400_idle, we need to allocate amd_e400_c1e_mask. */ + if (pm_idle == amd_e400_idle) + zalloc_cpumask_var(&amd_e400_c1e_mask, GFP_KERNEL); } static int __init idle_setup(char *str) @@ -640,6 +642,7 @@ static int __init idle_setup(char *str) boot_option_idle_override = IDLE_POLL; } else if (!strcmp(str, "mwait")) { boot_option_idle_override = IDLE_FORCE_MWAIT; + WARN_ONCE(1, "\"idle=mwait\" will be removed in 2012\n"); } else if (!strcmp(str, "halt")) { /* * When the boot option of idle=halt is added, halt is diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index 8d128783af4..a3d0dc59067 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -245,7 +245,6 @@ start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp) { set_user_gs(regs, 0); regs->fs = 0; - set_fs(USER_DS); regs->ds = __USER_DS; regs->es = __USER_DS; regs->ss = __USER_DS; diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 6c9dd922ac0..ca6f7ab8df3 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -338,7 +338,6 @@ start_thread_common(struct pt_regs *regs, unsigned long new_ip, regs->cs = _cs; regs->ss = _ss; regs->flags = X86_EFLAGS_IF; - set_fs(USER_DS); /* * Free the old FP and other extended state */ diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index f65e5b521db..807c2a2b80f 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -1363,7 +1363,7 @@ void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, * We must return the syscall number to actually look up in the table. * This can be -1L to skip running any syscall at all. */ -asmregparm long syscall_trace_enter(struct pt_regs *regs) +long syscall_trace_enter(struct pt_regs *regs) { long ret = 0; @@ -1408,7 +1408,7 @@ asmregparm long syscall_trace_enter(struct pt_regs *regs) return ret ?: regs->orig_ax; } -asmregparm void syscall_trace_leave(struct pt_regs *regs) +void syscall_trace_leave(struct pt_regs *regs) { bool step; diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 605e5ae19c7..afaf38447ef 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -910,6 +910,13 @@ void __init setup_arch(char **cmdline_p) memblock.current_limit = get_max_mapped(); memblock_x86_fill(); + /* + * The EFI specification says that boot service code won't be called + * after ExitBootServices(). This is, in fact, a lie. + */ + if (efi_enabled) + efi_reserve_boot_services(); + /* preallocate 4k for mptable mpc */ early_reserve_e820_mpc_new(); @@ -946,6 +953,8 @@ void __init setup_arch(char **cmdline_p) if (init_ohci1394_dma_early) init_ohci1394_dma_on_all_controllers(); #endif + /* Allocate bigger log buffer */ + setup_log_buf(1); reserve_initrd(); diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index a3c430bdfb6..33a0c11797d 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -1307,7 +1307,7 @@ void play_dead_common(void) { idle_task_exit(); reset_lazy_tlbstate(); - c1e_remove_cpu(raw_smp_processor_id()); + amd_e400_remove_cpu(raw_smp_processor_id()); mb(); /* Ack it */ @@ -1332,7 +1332,7 @@ static inline void mwait_play_dead(void) void *mwait_ptr; struct cpuinfo_x86 *c = __this_cpu_ptr(&cpu_info); - if (!this_cpu_has(X86_FEATURE_MWAIT) && mwait_usable(c)) + if (!(this_cpu_has(X86_FEATURE_MWAIT) && mwait_usable(c))) return; if (!this_cpu_has(X86_FEATURE_CLFLSH)) return; diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S index 32cbffb0c49..fbb0a045a1a 100644 --- a/arch/x86/kernel/syscall_table_32.S +++ b/arch/x86/kernel/syscall_table_32.S @@ -345,3 +345,4 @@ ENTRY(sys_call_table) .long sys_clock_adjtime .long sys_syncfs .long sys_sendmmsg /* 345 */ + .long sys_setns diff --git a/arch/x86/kernel/tboot.c b/arch/x86/kernel/tboot.c index 998e972f3b1..30ac65df7d4 100644 --- a/arch/x86/kernel/tboot.c +++ b/arch/x86/kernel/tboot.c @@ -110,7 +110,6 @@ static struct mm_struct tboot_mm = { .mmap_sem = __RWSEM_INITIALIZER(init_mm.mmap_sem), .page_table_lock = __SPIN_LOCK_UNLOCKED(init_mm.page_table_lock), .mmlist = LIST_HEAD_INIT(init_mm.mmlist), - .cpu_vm_mask = CPU_MASK_ALL, }; static inline void switch_to_tboot_pt(void) diff --git a/arch/x86/kernel/time.c b/arch/x86/kernel/time.c index 25a28a24593..00cbb272627 100644 --- a/arch/x86/kernel/time.c +++ b/arch/x86/kernel/time.c @@ -23,7 +23,7 @@ #include <asm/time.h> #ifdef CONFIG_X86_64 -volatile unsigned long __jiffies __section_jiffies = INITIAL_JIFFIES; +DEFINE_VVAR(volatile unsigned long, jiffies) = INITIAL_JIFFIES; #endif unsigned long profile_pc(struct pt_regs *regs) diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index 9335bf7dd2e..6cc6922262a 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -763,25 +763,6 @@ static cycle_t read_tsc(struct clocksource *cs) ret : clocksource_tsc.cycle_last; } -#ifdef CONFIG_X86_64 -static cycle_t __vsyscall_fn vread_tsc(void) -{ - cycle_t ret; - - /* - * Surround the RDTSC by barriers, to make sure it's not - * speculated to outside the seqlock critical section and - * does not cause time warps: - */ - rdtsc_barrier(); - ret = (cycle_t)vget_cycles(); - rdtsc_barrier(); - - return ret >= __vsyscall_gtod_data.clock.cycle_last ? - ret : __vsyscall_gtod_data.clock.cycle_last; -} -#endif - static void resume_tsc(struct clocksource *cs) { clocksource_tsc.cycle_last = 0; diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S index 49927a863cc..89aed99aafc 100644 --- a/arch/x86/kernel/vmlinux.lds.S +++ b/arch/x86/kernel/vmlinux.lds.S @@ -161,6 +161,12 @@ SECTIONS #define VVIRT_OFFSET (VSYSCALL_ADDR - __vsyscall_0) #define VVIRT(x) (ADDR(x) - VVIRT_OFFSET) +#define EMIT_VVAR(x, offset) .vsyscall_var_ ## x \ + ADDR(.vsyscall_0) + offset \ + : AT(VLOAD(.vsyscall_var_ ## x)) { \ + *(.vsyscall_var_ ## x) \ + } \ + x = VVIRT(.vsyscall_var_ ## x); . = ALIGN(4096); __vsyscall_0 = .; @@ -175,18 +181,6 @@ SECTIONS *(.vsyscall_fn) } - . = ALIGN(L1_CACHE_BYTES); - .vsyscall_gtod_data : AT(VLOAD(.vsyscall_gtod_data)) { - *(.vsyscall_gtod_data) - } - - vsyscall_gtod_data = VVIRT(.vsyscall_gtod_data); - .vsyscall_clock : AT(VLOAD(.vsyscall_clock)) { - *(.vsyscall_clock) - } - vsyscall_clock = VVIRT(.vsyscall_clock); - - .vsyscall_1 ADDR(.vsyscall_0) + 1024: AT(VLOAD(.vsyscall_1)) { *(.vsyscall_1) } @@ -194,21 +188,14 @@ SECTIONS *(.vsyscall_2) } - .vgetcpu_mode : AT(VLOAD(.vgetcpu_mode)) { - *(.vgetcpu_mode) - } - vgetcpu_mode = VVIRT(.vgetcpu_mode); - - . = ALIGN(L1_CACHE_BYTES); - .jiffies : AT(VLOAD(.jiffies)) { - *(.jiffies) - } - jiffies = VVIRT(.jiffies); - .vsyscall_3 ADDR(.vsyscall_0) + 3072: AT(VLOAD(.vsyscall_3)) { *(.vsyscall_3) } +#define __VVAR_KERNEL_LDS +#include <asm/vvar.h> +#undef __VVAR_KERNEL_LDS + . = __vsyscall_0 + PAGE_SIZE; #undef VSYSCALL_ADDR @@ -216,6 +203,7 @@ SECTIONS #undef VLOAD #undef VVIRT_OFFSET #undef VVIRT +#undef EMIT_VVAR #endif /* CONFIG_X86_64 */ @@ -326,7 +314,7 @@ SECTIONS } #if !defined(CONFIG_X86_64) || !defined(CONFIG_SMP) - PERCPU(INTERNODE_CACHE_BYTES, PAGE_SIZE) + PERCPU_SECTION(INTERNODE_CACHE_BYTES) #endif . = ALIGN(PAGE_SIZE); diff --git a/arch/x86/kernel/vread_tsc_64.c b/arch/x86/kernel/vread_tsc_64.c new file mode 100644 index 00000000000..a81aa9e9894 --- /dev/null +++ b/arch/x86/kernel/vread_tsc_64.c @@ -0,0 +1,36 @@ +/* This code runs in userspace. */ + +#define DISABLE_BRANCH_PROFILING +#include <asm/vgtod.h> + +notrace cycle_t __vsyscall_fn vread_tsc(void) +{ + cycle_t ret; + u64 last; + + /* + * Empirically, a fence (of type that depends on the CPU) + * before rdtsc is enough to ensure that rdtsc is ordered + * with respect to loads. The various CPU manuals are unclear + * as to whether rdtsc can be reordered with later loads, + * but no one has ever seen it happen. + */ + rdtsc_barrier(); + ret = (cycle_t)vget_cycles(); + + last = VVAR(vsyscall_gtod_data).clock.cycle_last; + + if (likely(ret >= last)) + return ret; + + /* + * GCC likes to generate cmov here, but this branch is extremely + * predictable (it's just a funciton of time and the likely is + * very likely) and there's a data dependence, so force GCC + * to generate a branch instead. I don't barrier() because + * we don't actually need a barrier, and if this function + * ever gets inlined it will generate worse code. + */ + asm volatile (""); + return last; +} diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c index dcbb28c4b69..3e682184d76 100644 --- a/arch/x86/kernel/vsyscall_64.c +++ b/arch/x86/kernel/vsyscall_64.c @@ -49,17 +49,10 @@ __attribute__ ((unused, __section__(".vsyscall_" #nr))) notrace #define __syscall_clobber "r11","cx","memory" -/* - * vsyscall_gtod_data contains data that is : - * - readonly from vsyscalls - * - written by timer interrupt or systcl (/proc/sys/kernel/vsyscall64) - * Try to keep this structure as small as possible to avoid cache line ping pongs - */ -int __vgetcpu_mode __section_vgetcpu_mode; - -struct vsyscall_gtod_data __vsyscall_gtod_data __section_vsyscall_gtod_data = +DEFINE_VVAR(int, vgetcpu_mode); +DEFINE_VVAR(struct vsyscall_gtod_data, vsyscall_gtod_data) = { - .lock = SEQLOCK_UNLOCKED, + .lock = __SEQLOCK_UNLOCKED(__vsyscall_gtod_data.lock), .sysctl_enabled = 1, }; @@ -97,7 +90,7 @@ void update_vsyscall(struct timespec *wall_time, struct timespec *wtm, */ static __always_inline void do_get_tz(struct timezone * tz) { - *tz = __vsyscall_gtod_data.sys_tz; + *tz = VVAR(vsyscall_gtod_data).sys_tz; } static __always_inline int gettimeofday(struct timeval *tv, struct timezone *tz) @@ -126,23 +119,24 @@ static __always_inline void do_vgettimeofday(struct timeval * tv) unsigned long mult, shift, nsec; cycle_t (*vread)(void); do { - seq = read_seqbegin(&__vsyscall_gtod_data.lock); + seq = read_seqbegin(&VVAR(vsyscall_gtod_data).lock); - vread = __vsyscall_gtod_data.clock.vread; - if (unlikely(!__vsyscall_gtod_data.sysctl_enabled || !vread)) { + vread = VVAR(vsyscall_gtod_data).clock.vread; + if (unlikely(!VVAR(vsyscall_gtod_data).sysctl_enabled || + !vread)) { gettimeofday(tv,NULL); return; } now = vread(); - base = __vsyscall_gtod_data.clock.cycle_last; - mask = __vsyscall_gtod_data.clock.mask; - mult = __vsyscall_gtod_data.clock.mult; - shift = __vsyscall_gtod_data.clock.shift; + base = VVAR(vsyscall_gtod_data).clock.cycle_last; + mask = VVAR(vsyscall_gtod_data).clock.mask; + mult = VVAR(vsyscall_gtod_data).clock.mult; + shift = VVAR(vsyscall_gtod_data).clock.shift; - tv->tv_sec = __vsyscall_gtod_data.wall_time_sec; - nsec = __vsyscall_gtod_data.wall_time_nsec; - } while (read_seqretry(&__vsyscall_gtod_data.lock, seq)); + tv->tv_sec = VVAR(vsyscall_gtod_data).wall_time_sec; + nsec = VVAR(vsyscall_gtod_data).wall_time_nsec; + } while (read_seqretry(&VVAR(vsyscall_gtod_data).lock, seq)); /* calculate interval: */ cycle_delta = (now - base) & mask; @@ -171,15 +165,15 @@ time_t __vsyscall(1) vtime(time_t *t) { unsigned seq; time_t result; - if (unlikely(!__vsyscall_gtod_data.sysctl_enabled)) + if (unlikely(!VVAR(vsyscall_gtod_data).sysctl_enabled)) return time_syscall(t); do { - seq = read_seqbegin(&__vsyscall_gtod_data.lock); + seq = read_seqbegin(&VVAR(vsyscall_gtod_data).lock); - result = __vsyscall_gtod_data.wall_time_sec; + result = VVAR(vsyscall_gtod_data).wall_time_sec; - } while (read_seqretry(&__vsyscall_gtod_data.lock, seq)); + } while (read_seqretry(&VVAR(vsyscall_gtod_data).lock, seq)); if (t) *t = result; @@ -208,9 +202,9 @@ vgetcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *tcache) We do this here because otherwise user space would do it on its own in a likely inferior way (no access to jiffies). If you don't like it pass NULL. */ - if (tcache && tcache->blob[0] == (j = __jiffies)) { + if (tcache && tcache->blob[0] == (j = VVAR(jiffies))) { p = tcache->blob[1]; - } else if (__vgetcpu_mode == VGETCPU_RDTSCP) { + } else if (VVAR(vgetcpu_mode) == VGETCPU_RDTSCP) { /* Load per CPU data from RDTSCP */ native_read_tscp(&p); } else { diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index d6e2477feb1..6df88c7885c 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -47,38 +47,40 @@ #define DstDI (5<<1) /* Destination is in ES:(E)DI */ #define DstMem64 (6<<1) /* 64bit memory operand */ #define DstImmUByte (7<<1) /* 8-bit unsigned immediate operand */ -#define DstMask (7<<1) +#define DstDX (8<<1) /* Destination is in DX register */ +#define DstMask (0xf<<1) /* Source operand type. */ -#define SrcNone (0<<4) /* No source operand. */ -#define SrcReg (1<<4) /* Register operand. */ -#define SrcMem (2<<4) /* Memory operand. */ -#define SrcMem16 (3<<4) /* Memory operand (16-bit). */ -#define SrcMem32 (4<<4) /* Memory operand (32-bit). */ -#define SrcImm (5<<4) /* Immediate operand. */ -#define SrcImmByte (6<<4) /* 8-bit sign-extended immediate operand. */ -#define SrcOne (7<<4) /* Implied '1' */ -#define SrcImmUByte (8<<4) /* 8-bit unsigned immediate operand. */ -#define SrcImmU (9<<4) /* Immediate operand, unsigned */ -#define SrcSI (0xa<<4) /* Source is in the DS:RSI */ -#define SrcImmFAddr (0xb<<4) /* Source is immediate far address */ -#define SrcMemFAddr (0xc<<4) /* Source is far address in memory */ -#define SrcAcc (0xd<<4) /* Source Accumulator */ -#define SrcImmU16 (0xe<<4) /* Immediate operand, unsigned, 16 bits */ -#define SrcMask (0xf<<4) +#define SrcNone (0<<5) /* No source operand. */ +#define SrcReg (1<<5) /* Register operand. */ +#define SrcMem (2<<5) /* Memory operand. */ +#define SrcMem16 (3<<5) /* Memory operand (16-bit). */ +#define SrcMem32 (4<<5) /* Memory operand (32-bit). */ +#define SrcImm (5<<5) /* Immediate operand. */ +#define SrcImmByte (6<<5) /* 8-bit sign-extended immediate operand. */ +#define SrcOne (7<<5) /* Implied '1' */ +#define SrcImmUByte (8<<5) /* 8-bit unsigned immediate operand. */ +#define SrcImmU (9<<5) /* Immediate operand, unsigned */ +#define SrcSI (0xa<<5) /* Source is in the DS:RSI */ +#define SrcImmFAddr (0xb<<5) /* Source is immediate far address */ +#define SrcMemFAddr (0xc<<5) /* Source is far address in memory */ +#define SrcAcc (0xd<<5) /* Source Accumulator */ +#define SrcImmU16 (0xe<<5) /* Immediate operand, unsigned, 16 bits */ +#define SrcDX (0xf<<5) /* Source is in DX register */ +#define SrcMask (0xf<<5) /* Generic ModRM decode. */ -#define ModRM (1<<8) +#define ModRM (1<<9) /* Destination is only written; never read. */ -#define Mov (1<<9) -#define BitOp (1<<10) -#define MemAbs (1<<11) /* Memory operand is absolute displacement */ -#define String (1<<12) /* String instruction (rep capable) */ -#define Stack (1<<13) /* Stack instruction (push/pop) */ -#define GroupMask (7<<14) /* Opcode uses one of the group mechanisms */ -#define Group (1<<14) /* Bits 3:5 of modrm byte extend opcode */ -#define GroupDual (2<<14) /* Alternate decoding of mod == 3 */ -#define Prefix (3<<14) /* Instruction varies with 66/f2/f3 prefix */ -#define RMExt (4<<14) /* Opcode extension in ModRM r/m if mod == 3 */ -#define Sse (1<<17) /* SSE Vector instruction */ +#define Mov (1<<10) +#define BitOp (1<<11) +#define MemAbs (1<<12) /* Memory operand is absolute displacement */ +#define String (1<<13) /* String instruction (rep capable) */ +#define Stack (1<<14) /* Stack instruction (push/pop) */ +#define GroupMask (7<<15) /* Opcode uses one of the group mechanisms */ +#define Group (1<<15) /* Bits 3:5 of modrm byte extend opcode */ +#define GroupDual (2<<15) /* Alternate decoding of mod == 3 */ +#define Prefix (3<<15) /* Instruction varies with 66/f2/f3 prefix */ +#define RMExt (4<<15) /* Opcode extension in ModRM r/m if mod == 3 */ +#define Sse (1<<18) /* SSE Vector instruction */ /* Misc flags */ #define Prot (1<<21) /* instruction generates #UD if not in prot-mode */ #define VendorSpecific (1<<22) /* Vendor specific instruction */ @@ -3154,8 +3156,8 @@ static struct opcode opcode_table[256] = { I(DstReg | SrcMem | ModRM | Src2Imm, em_imul_3op), I(SrcImmByte | Mov | Stack, em_push), I(DstReg | SrcMem | ModRM | Src2ImmByte, em_imul_3op), - D2bvIP(DstDI | Mov | String, ins, check_perm_in), /* insb, insw/insd */ - D2bvIP(SrcSI | ImplicitOps | String, outs, check_perm_out), /* outsb, outsw/outsd */ + D2bvIP(DstDI | SrcDX | Mov | String, ins, check_perm_in), /* insb, insw/insd */ + D2bvIP(SrcSI | DstDX | String, outs, check_perm_out), /* outsb, outsw/outsd */ /* 0x70 - 0x7F */ X16(D(SrcImmByte)), /* 0x80 - 0x87 */ @@ -3212,8 +3214,8 @@ static struct opcode opcode_table[256] = { /* 0xE8 - 0xEF */ D(SrcImm | Stack), D(SrcImm | ImplicitOps), D(SrcImmFAddr | No64), D(SrcImmByte | ImplicitOps), - D2bvIP(SrcNone | DstAcc, in, check_perm_in), - D2bvIP(SrcAcc | ImplicitOps, out, check_perm_out), + D2bvIP(SrcDX | DstAcc, in, check_perm_in), + D2bvIP(SrcAcc | DstDX, out, check_perm_out), /* 0xF0 - 0xF7 */ N, DI(ImplicitOps, icebp), N, N, DI(ImplicitOps | Priv, hlt), D(ImplicitOps), @@ -3613,6 +3615,12 @@ done_prefixes: memop.bytes = c->op_bytes + 2; goto srcmem_common; break; + case SrcDX: + c->src.type = OP_REG; + c->src.bytes = 2; + c->src.addr.reg = &c->regs[VCPU_REGS_RDX]; + fetch_register_operand(&c->src); + break; } if (rc != X86EMUL_CONTINUE) @@ -3682,6 +3690,12 @@ done_prefixes: c->dst.addr.mem.seg = VCPU_SREG_ES; c->dst.val = 0; break; + case DstDX: + c->dst.type = OP_REG; + c->dst.bytes = 2; + c->dst.addr.reg = &c->regs[VCPU_REGS_RDX]; + fetch_register_operand(&c->dst); + break; case ImplicitOps: /* Special instructions do their own operand decoding. */ default: @@ -4027,7 +4041,6 @@ special_insn: break; case 0xec: /* in al,dx */ case 0xed: /* in (e/r)ax,dx */ - c->src.val = c->regs[VCPU_REGS_RDX]; do_io_in: if (!pio_in_emulated(ctxt, ops, c->dst.bytes, c->src.val, &c->dst.val)) @@ -4035,7 +4048,6 @@ special_insn: break; case 0xee: /* out dx,al */ case 0xef: /* out dx,(e/r)ax */ - c->dst.val = c->regs[VCPU_REGS_RDX]; do_io_out: ops->pio_out_emulated(ctxt, c->src.bytes, c->dst.val, &c->src.val, 1); diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 28418054b88..bd14bb4c859 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -3545,10 +3545,11 @@ static int kvm_mmu_remove_some_alloc_mmu_pages(struct kvm *kvm, return kvm_mmu_prepare_zap_page(kvm, page, invalid_list); } -static int mmu_shrink(struct shrinker *shrink, int nr_to_scan, gfp_t gfp_mask) +static int mmu_shrink(struct shrinker *shrink, struct shrink_control *sc) { struct kvm *kvm; struct kvm *kvm_freed = NULL; + int nr_to_scan = sc->nr_to_scan; if (nr_to_scan == 0) goto out; diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c index e191c096ab9..db832fd65ec 100644 --- a/arch/x86/lguest/boot.c +++ b/arch/x86/lguest/boot.c @@ -993,6 +993,7 @@ static void lguest_time_irq(unsigned int irq, struct irq_desc *desc) static void lguest_time_init(void) { /* Set up the timer interrupt (0) to go to our simple timer routine */ + lguest_setup_irq(0); irq_set_handler(0, lguest_time_irq); clocksource_register_hz(&lguest_clock, NSEC_PER_SEC); diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index bcb394dfbb3..2dbf6bf4c7e 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -823,16 +823,30 @@ do_sigbus(struct pt_regs *regs, unsigned long error_code, unsigned long address, force_sig_info_fault(SIGBUS, code, address, tsk, fault); } -static noinline void +static noinline int mm_fault_error(struct pt_regs *regs, unsigned long error_code, unsigned long address, unsigned int fault) { + /* + * Pagefault was interrupted by SIGKILL. We have no reason to + * continue pagefault. + */ + if (fatal_signal_pending(current)) { + if (!(fault & VM_FAULT_RETRY)) + up_read(¤t->mm->mmap_sem); + if (!(error_code & PF_USER)) + no_context(regs, error_code, address); + return 1; + } + if (!(fault & VM_FAULT_ERROR)) + return 0; + if (fault & VM_FAULT_OOM) { /* Kernel mode? Handle exceptions or die: */ if (!(error_code & PF_USER)) { up_read(¤t->mm->mmap_sem); no_context(regs, error_code, address); - return; + return 1; } out_of_memory(regs, error_code, address); @@ -843,6 +857,7 @@ mm_fault_error(struct pt_regs *regs, unsigned long error_code, else BUG(); } + return 1; } static int spurious_fault_check(unsigned long error_code, pte_t *pte) @@ -965,7 +980,7 @@ do_page_fault(struct pt_regs *regs, unsigned long error_code) struct mm_struct *mm; int fault; int write = error_code & PF_WRITE; - unsigned int flags = FAULT_FLAG_ALLOW_RETRY | + unsigned int flags = FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_KILLABLE | (write ? FAULT_FLAG_WRITE : 0); tsk = current; @@ -1133,9 +1148,9 @@ good_area: */ fault = handle_mm_fault(mm, vma, address, flags); - if (unlikely(fault & VM_FAULT_ERROR)) { - mm_fault_error(regs, error_code, address, fault); - return; + if (unlikely(fault & (VM_FAULT_RETRY|VM_FAULT_ERROR))) { + if (mm_fault_error(regs, error_code, address, fault)) + return; } /* diff --git a/arch/x86/mm/hugetlbpage.c b/arch/x86/mm/hugetlbpage.c index d4203988504..f581a18c0d4 100644 --- a/arch/x86/mm/hugetlbpage.c +++ b/arch/x86/mm/hugetlbpage.c @@ -72,7 +72,7 @@ static void huge_pmd_share(struct mm_struct *mm, unsigned long addr, pud_t *pud) if (!vma_shareable(vma, addr)) return; - spin_lock(&mapping->i_mmap_lock); + mutex_lock(&mapping->i_mmap_mutex); vma_prio_tree_foreach(svma, &iter, &mapping->i_mmap, idx, idx) { if (svma == vma) continue; @@ -97,7 +97,7 @@ static void huge_pmd_share(struct mm_struct *mm, unsigned long addr, pud_t *pud) put_page(virt_to_page(spte)); spin_unlock(&mm->page_table_lock); out: - spin_unlock(&mapping->i_mmap_lock); + mutex_unlock(&mapping->i_mmap_mutex); } /* diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c index 37b8b0fe832..30326443ab8 100644 --- a/arch/x86/mm/init.c +++ b/arch/x86/mm/init.c @@ -16,8 +16,6 @@ #include <asm/tlb.h> #include <asm/proto.h> -DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); - unsigned long __initdata pgt_buf_start; unsigned long __meminitdata pgt_buf_end; unsigned long __meminitdata pgt_buf_top; diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c index c3b8e24f2b1..9fd8a567fe1 100644 --- a/arch/x86/oprofile/op_model_amd.c +++ b/arch/x86/oprofile/op_model_amd.c @@ -316,16 +316,23 @@ static void op_amd_stop_ibs(void) wrmsrl(MSR_AMD64_IBSOPCTL, 0); } -static inline int eilvt_is_available(int offset) +static inline int get_eilvt(int offset) { - /* check if we may assign a vector */ return !setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_NMI, 1); } +static inline int put_eilvt(int offset) +{ + return !setup_APIC_eilvt(offset, 0, 0, 1); +} + static inline int ibs_eilvt_valid(void) { int offset; u64 val; + int valid = 0; + + preempt_disable(); rdmsrl(MSR_AMD64_IBSCTL, val); offset = val & IBSCTL_LVT_OFFSET_MASK; @@ -333,16 +340,20 @@ static inline int ibs_eilvt_valid(void) if (!(val & IBSCTL_LVT_OFFSET_VALID)) { pr_err(FW_BUG "cpu %d, invalid IBS interrupt offset %d (MSR%08X=0x%016llx)\n", smp_processor_id(), offset, MSR_AMD64_IBSCTL, val); - return 0; + goto out; } - if (!eilvt_is_available(offset)) { + if (!get_eilvt(offset)) { pr_err(FW_BUG "cpu %d, IBS interrupt offset %d not available (MSR%08X=0x%016llx)\n", smp_processor_id(), offset, MSR_AMD64_IBSCTL, val); - return 0; + goto out; } - return 1; + valid = 1; +out: + preempt_enable(); + + return valid; } static inline int get_ibs_offset(void) @@ -600,67 +611,69 @@ static int setup_ibs_ctl(int ibs_eilvt_off) static int force_ibs_eilvt_setup(void) { - int i; + int offset; int ret; - /* find the next free available EILVT entry */ - for (i = 1; i < 4; i++) { - if (!eilvt_is_available(i)) - continue; - ret = setup_ibs_ctl(i); - if (ret) - return ret; - pr_err(FW_BUG "using offset %d for IBS interrupts\n", i); - return 0; + /* + * find the next free available EILVT entry, skip offset 0, + * pin search to this cpu + */ + preempt_disable(); + for (offset = 1; offset < APIC_EILVT_NR_MAX; offset++) { + if (get_eilvt(offset)) + break; } + preempt_enable(); - printk(KERN_DEBUG "No EILVT entry available\n"); - - return -EBUSY; -} - -static int __init_ibs_nmi(void) -{ - int ret; - - if (ibs_eilvt_valid()) - return 0; + if (offset == APIC_EILVT_NR_MAX) { + printk(KERN_DEBUG "No EILVT entry available\n"); + return -EBUSY; + } - ret = force_ibs_eilvt_setup(); + ret = setup_ibs_ctl(offset); if (ret) - return ret; + goto out; - if (!ibs_eilvt_valid()) - return -EFAULT; + if (!ibs_eilvt_valid()) { + ret = -EFAULT; + goto out; + } + pr_err(FW_BUG "using offset %d for IBS interrupts\n", offset); pr_err(FW_BUG "workaround enabled for IBS LVT offset\n"); return 0; +out: + preempt_disable(); + put_eilvt(offset); + preempt_enable(); + return ret; } /* * check and reserve APIC extended interrupt LVT offset for IBS if * available - * - * init_ibs() preforms implicitly cpu-local operations, so pin this - * thread to its current CPU */ static void init_ibs(void) { - preempt_disable(); - ibs_caps = get_ibs_caps(); + if (!ibs_caps) + return; + + if (ibs_eilvt_valid()) goto out; - if (__init_ibs_nmi() < 0) - ibs_caps = 0; - else - printk(KERN_INFO "oprofile: AMD IBS detected (0x%08x)\n", ibs_caps); + if (!force_ibs_eilvt_setup()) + goto out; + + /* Failed to setup ibs */ + ibs_caps = 0; + return; out: - preempt_enable(); + printk(KERN_INFO "oprofile: AMD IBS detected (0x%08x)\n", ibs_caps); } static int (*create_arch_files)(struct super_block *sb, struct dentry *root); diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c index b30aa26a8df..0d3a4fa3456 100644 --- a/arch/x86/platform/efi/efi.c +++ b/arch/x86/platform/efi/efi.c @@ -304,6 +304,40 @@ static void __init print_efi_memmap(void) } #endif /* EFI_DEBUG */ +void __init efi_reserve_boot_services(void) +{ + void *p; + + for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) { + efi_memory_desc_t *md = p; + unsigned long long start = md->phys_addr; + unsigned long long size = md->num_pages << EFI_PAGE_SHIFT; + + if (md->type != EFI_BOOT_SERVICES_CODE && + md->type != EFI_BOOT_SERVICES_DATA) + continue; + + memblock_x86_reserve_range(start, start + size, "EFI Boot"); + } +} + +static void __init efi_free_boot_services(void) +{ + void *p; + + for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) { + efi_memory_desc_t *md = p; + unsigned long long start = md->phys_addr; + unsigned long long size = md->num_pages << EFI_PAGE_SHIFT; + + if (md->type != EFI_BOOT_SERVICES_CODE && + md->type != EFI_BOOT_SERVICES_DATA) + continue; + + free_bootmem_late(start, size); + } +} + void __init efi_init(void) { efi_config_table_t *config_tables; @@ -536,7 +570,9 @@ void __init efi_enter_virtual_mode(void) for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) { md = p; - if (!(md->attribute & EFI_MEMORY_RUNTIME)) + if (!(md->attribute & EFI_MEMORY_RUNTIME) && + md->type != EFI_BOOT_SERVICES_CODE && + md->type != EFI_BOOT_SERVICES_DATA) continue; size = md->num_pages << EFI_PAGE_SHIFT; @@ -593,6 +629,13 @@ void __init efi_enter_virtual_mode(void) } /* + * Thankfully, it does seem that no runtime services other than + * SetVirtualAddressMap() will touch boot services code, so we can + * get rid of it all at this point + */ + efi_free_boot_services(); + + /* * Now that EFI is in virtual mode, update the function * pointers in the runtime service table to the new virtual addresses. * diff --git a/arch/x86/platform/efi/efi_64.c b/arch/x86/platform/efi/efi_64.c index 2649426a790..ac3aa54e265 100644 --- a/arch/x86/platform/efi/efi_64.c +++ b/arch/x86/platform/efi/efi_64.c @@ -49,10 +49,11 @@ static void __init early_code_mapping_set_exec(int executable) if (!(__supported_pte_mask & _PAGE_NX)) return; - /* Make EFI runtime service code area executable */ + /* Make EFI service code area executable */ for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) { md = p; - if (md->type == EFI_RUNTIME_SERVICES_CODE) + if (md->type == EFI_RUNTIME_SERVICES_CODE || + md->type == EFI_BOOT_SERVICES_CODE) efi_set_executable(md, executable); } } diff --git a/arch/x86/platform/uv/tlb_uv.c b/arch/x86/platform/uv/tlb_uv.c index c58e0ea39ef..68e467f69fe 100644 --- a/arch/x86/platform/uv/tlb_uv.c +++ b/arch/x86/platform/uv/tlb_uv.c @@ -1,7 +1,7 @@ /* * SGI UltraViolet TLB flush routines. * - * (c) 2008-2010 Cliff Wickman <cpw@sgi.com>, SGI. + * (c) 2008-2011 Cliff Wickman <cpw@sgi.com>, SGI. * * This code is released under the GNU General Public License version 2 or * later. @@ -35,6 +35,7 @@ static int timeout_base_ns[] = { 5242880, 167772160 }; + static int timeout_us; static int nobau; static int baudisabled; @@ -42,20 +43,70 @@ static spinlock_t disable_lock; static cycles_t congested_cycles; /* tunables: */ -static int max_bau_concurrent = MAX_BAU_CONCURRENT; -static int max_bau_concurrent_constant = MAX_BAU_CONCURRENT; -static int plugged_delay = PLUGGED_DELAY; -static int plugsb4reset = PLUGSB4RESET; -static int timeoutsb4reset = TIMEOUTSB4RESET; -static int ipi_reset_limit = IPI_RESET_LIMIT; -static int complete_threshold = COMPLETE_THRESHOLD; -static int congested_response_us = CONGESTED_RESPONSE_US; -static int congested_reps = CONGESTED_REPS; -static int congested_period = CONGESTED_PERIOD; +static int max_concurr = MAX_BAU_CONCURRENT; +static int max_concurr_const = MAX_BAU_CONCURRENT; +static int plugged_delay = PLUGGED_DELAY; +static int plugsb4reset = PLUGSB4RESET; +static int timeoutsb4reset = TIMEOUTSB4RESET; +static int ipi_reset_limit = IPI_RESET_LIMIT; +static int complete_threshold = COMPLETE_THRESHOLD; +static int congested_respns_us = CONGESTED_RESPONSE_US; +static int congested_reps = CONGESTED_REPS; +static int congested_period = CONGESTED_PERIOD; + +static struct tunables tunables[] = { + {&max_concurr, MAX_BAU_CONCURRENT}, /* must be [0] */ + {&plugged_delay, PLUGGED_DELAY}, + {&plugsb4reset, PLUGSB4RESET}, + {&timeoutsb4reset, TIMEOUTSB4RESET}, + {&ipi_reset_limit, IPI_RESET_LIMIT}, + {&complete_threshold, COMPLETE_THRESHOLD}, + {&congested_respns_us, CONGESTED_RESPONSE_US}, + {&congested_reps, CONGESTED_REPS}, + {&congested_period, CONGESTED_PERIOD} +}; + static struct dentry *tunables_dir; static struct dentry *tunables_file; -static int __init setup_nobau(char *arg) +/* these correspond to the statistics printed by ptc_seq_show() */ +static char *stat_description[] = { + "sent: number of shootdown messages sent", + "stime: time spent sending messages", + "numuvhubs: number of hubs targeted with shootdown", + "numuvhubs16: number times 16 or more hubs targeted", + "numuvhubs8: number times 8 or more hubs targeted", + "numuvhubs4: number times 4 or more hubs targeted", + "numuvhubs2: number times 2 or more hubs targeted", + "numuvhubs1: number times 1 hub targeted", + "numcpus: number of cpus targeted with shootdown", + "dto: number of destination timeouts", + "retries: destination timeout retries sent", + "rok: : destination timeouts successfully retried", + "resetp: ipi-style resource resets for plugs", + "resett: ipi-style resource resets for timeouts", + "giveup: fall-backs to ipi-style shootdowns", + "sto: number of source timeouts", + "bz: number of stay-busy's", + "throt: number times spun in throttle", + "swack: image of UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE", + "recv: shootdown messages received", + "rtime: time spent processing messages", + "all: shootdown all-tlb messages", + "one: shootdown one-tlb messages", + "mult: interrupts that found multiple messages", + "none: interrupts that found no messages", + "retry: number of retry messages processed", + "canc: number messages canceled by retries", + "nocan: number retries that found nothing to cancel", + "reset: number of ipi-style reset requests processed", + "rcan: number messages canceled by reset requests", + "disable: number times use of the BAU was disabled", + "enable: number times use of the BAU was re-enabled" +}; + +static int __init +setup_nobau(char *arg) { nobau = 1; return 0; @@ -63,7 +114,7 @@ static int __init setup_nobau(char *arg) early_param("nobau", setup_nobau); /* base pnode in this partition */ -static int uv_partition_base_pnode __read_mostly; +static int uv_base_pnode __read_mostly; /* position of pnode (which is nasid>>1): */ static int uv_nshift __read_mostly; static unsigned long uv_mmask __read_mostly; @@ -109,60 +160,52 @@ static int __init uvhub_to_first_apicid(int uvhub) * clear of the Timeout bit (as well) will free the resource. No reply will * be sent (the hardware will only do one reply per message). */ -static inline void uv_reply_to_message(struct msg_desc *mdp, - struct bau_control *bcp) +static void reply_to_message(struct msg_desc *mdp, struct bau_control *bcp) { unsigned long dw; - struct bau_payload_queue_entry *msg; + struct bau_pq_entry *msg; msg = mdp->msg; if (!msg->canceled) { - dw = (msg->sw_ack_vector << UV_SW_ACK_NPENDING) | - msg->sw_ack_vector; - uv_write_local_mmr( - UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_ALIAS, dw); + dw = (msg->swack_vec << UV_SW_ACK_NPENDING) | msg->swack_vec; + write_mmr_sw_ack(dw); } msg->replied_to = 1; - msg->sw_ack_vector = 0; + msg->swack_vec = 0; } /* * Process the receipt of a RETRY message */ -static inline void uv_bau_process_retry_msg(struct msg_desc *mdp, - struct bau_control *bcp) +static void bau_process_retry_msg(struct msg_desc *mdp, + struct bau_control *bcp) { int i; int cancel_count = 0; - int slot2; unsigned long msg_res; unsigned long mmr = 0; - struct bau_payload_queue_entry *msg; - struct bau_payload_queue_entry *msg2; - struct ptc_stats *stat; + struct bau_pq_entry *msg = mdp->msg; + struct bau_pq_entry *msg2; + struct ptc_stats *stat = bcp->statp; - msg = mdp->msg; - stat = bcp->statp; stat->d_retries++; /* * cancel any message from msg+1 to the retry itself */ for (msg2 = msg+1, i = 0; i < DEST_Q_SIZE; msg2++, i++) { - if (msg2 > mdp->va_queue_last) - msg2 = mdp->va_queue_first; + if (msg2 > mdp->queue_last) + msg2 = mdp->queue_first; if (msg2 == msg) break; - /* same conditions for cancellation as uv_do_reset */ + /* same conditions for cancellation as do_reset */ if ((msg2->replied_to == 0) && (msg2->canceled == 0) && - (msg2->sw_ack_vector) && ((msg2->sw_ack_vector & - msg->sw_ack_vector) == 0) && + (msg2->swack_vec) && ((msg2->swack_vec & + msg->swack_vec) == 0) && (msg2->sending_cpu == msg->sending_cpu) && (msg2->msg_type != MSG_NOOP)) { - slot2 = msg2 - mdp->va_queue_first; - mmr = uv_read_local_mmr - (UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE); - msg_res = msg2->sw_ack_vector; + mmr = read_mmr_sw_ack(); + msg_res = msg2->swack_vec; /* * This is a message retry; clear the resources held * by the previous message only if they timed out. @@ -170,6 +213,7 @@ static inline void uv_bau_process_retry_msg(struct msg_desc *mdp, * situation to report. */ if (mmr & (msg_res << UV_SW_ACK_NPENDING)) { + unsigned long mr; /* * is the resource timed out? * make everyone ignore the cancelled message. @@ -177,10 +221,8 @@ static inline void uv_bau_process_retry_msg(struct msg_desc *mdp, msg2->canceled = 1; stat->d_canceled++; cancel_count++; - uv_write_local_mmr( - UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_ALIAS, - (msg_res << UV_SW_ACK_NPENDING) | - msg_res); + mr = (msg_res << UV_SW_ACK_NPENDING) | msg_res; + write_mmr_sw_ack(mr); } } } @@ -192,20 +234,19 @@ static inline void uv_bau_process_retry_msg(struct msg_desc *mdp, * Do all the things a cpu should do for a TLB shootdown message. * Other cpu's may come here at the same time for this message. */ -static void uv_bau_process_message(struct msg_desc *mdp, - struct bau_control *bcp) +static void bau_process_message(struct msg_desc *mdp, + struct bau_control *bcp) { - int msg_ack_count; short socket_ack_count = 0; - struct ptc_stats *stat; - struct bau_payload_queue_entry *msg; + short *sp; + struct atomic_short *asp; + struct ptc_stats *stat = bcp->statp; + struct bau_pq_entry *msg = mdp->msg; struct bau_control *smaster = bcp->socket_master; /* * This must be a normal message, or retry of a normal message */ - msg = mdp->msg; - stat = bcp->statp; if (msg->address == TLB_FLUSH_ALL) { local_flush_tlb(); stat->d_alltlb++; @@ -222,30 +263,32 @@ static void uv_bau_process_message(struct msg_desc *mdp, * cpu number. */ if (msg->msg_type == MSG_RETRY && bcp == bcp->uvhub_master) - uv_bau_process_retry_msg(mdp, bcp); + bau_process_retry_msg(mdp, bcp); /* - * This is a sw_ack message, so we have to reply to it. + * This is a swack message, so we have to reply to it. * Count each responding cpu on the socket. This avoids * pinging the count's cache line back and forth between * the sockets. */ - socket_ack_count = atomic_add_short_return(1, (struct atomic_short *) - &smaster->socket_acknowledge_count[mdp->msg_slot]); + sp = &smaster->socket_acknowledge_count[mdp->msg_slot]; + asp = (struct atomic_short *)sp; + socket_ack_count = atom_asr(1, asp); if (socket_ack_count == bcp->cpus_in_socket) { + int msg_ack_count; /* * Both sockets dump their completed count total into * the message's count. */ smaster->socket_acknowledge_count[mdp->msg_slot] = 0; - msg_ack_count = atomic_add_short_return(socket_ack_count, - (struct atomic_short *)&msg->acknowledge_count); + asp = (struct atomic_short *)&msg->acknowledge_count; + msg_ack_count = atom_asr(socket_ack_count, asp); if (msg_ack_count == bcp->cpus_in_uvhub) { /* * All cpus in uvhub saw it; reply */ - uv_reply_to_message(mdp, bcp); + reply_to_message(mdp, bcp); } } @@ -268,62 +311,51 @@ static int uvhub_to_first_cpu(int uvhub) * Last resort when we get a large number of destination timeouts is * to clear resources held by a given cpu. * Do this with IPI so that all messages in the BAU message queue - * can be identified by their nonzero sw_ack_vector field. + * can be identified by their nonzero swack_vec field. * * This is entered for a single cpu on the uvhub. * The sender want's this uvhub to free a specific message's - * sw_ack resources. + * swack resources. */ -static void -uv_do_reset(void *ptr) +static void do_reset(void *ptr) { int i; - int slot; - int count = 0; - unsigned long mmr; - unsigned long msg_res; - struct bau_control *bcp; - struct reset_args *rap; - struct bau_payload_queue_entry *msg; - struct ptc_stats *stat; + struct bau_control *bcp = &per_cpu(bau_control, smp_processor_id()); + struct reset_args *rap = (struct reset_args *)ptr; + struct bau_pq_entry *msg; + struct ptc_stats *stat = bcp->statp; - bcp = &per_cpu(bau_control, smp_processor_id()); - rap = (struct reset_args *)ptr; - stat = bcp->statp; stat->d_resets++; - /* * We're looking for the given sender, and - * will free its sw_ack resource. + * will free its swack resource. * If all cpu's finally responded after the timeout, its * message 'replied_to' was set. */ - for (msg = bcp->va_queue_first, i = 0; i < DEST_Q_SIZE; msg++, i++) { - /* uv_do_reset: same conditions for cancellation as - uv_bau_process_retry_msg() */ + for (msg = bcp->queue_first, i = 0; i < DEST_Q_SIZE; msg++, i++) { + unsigned long msg_res; + /* do_reset: same conditions for cancellation as + bau_process_retry_msg() */ if ((msg->replied_to == 0) && (msg->canceled == 0) && (msg->sending_cpu == rap->sender) && - (msg->sw_ack_vector) && + (msg->swack_vec) && (msg->msg_type != MSG_NOOP)) { + unsigned long mmr; + unsigned long mr; /* * make everyone else ignore this message */ msg->canceled = 1; - slot = msg - bcp->va_queue_first; - count++; /* * only reset the resource if it is still pending */ - mmr = uv_read_local_mmr - (UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE); - msg_res = msg->sw_ack_vector; + mmr = read_mmr_sw_ack(); + msg_res = msg->swack_vec; + mr = (msg_res << UV_SW_ACK_NPENDING) | msg_res; if (mmr & msg_res) { stat->d_rcanceled++; - uv_write_local_mmr( - UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE_ALIAS, - (msg_res << UV_SW_ACK_NPENDING) | - msg_res); + write_mmr_sw_ack(mr); } } } @@ -334,39 +366,38 @@ uv_do_reset(void *ptr) * Use IPI to get all target uvhubs to release resources held by * a given sending cpu number. */ -static void uv_reset_with_ipi(struct bau_target_uvhubmask *distribution, - int sender) +static void reset_with_ipi(struct bau_targ_hubmask *distribution, int sender) { int uvhub; - int cpu; + int maskbits; cpumask_t mask; struct reset_args reset_args; reset_args.sender = sender; - cpus_clear(mask); /* find a single cpu for each uvhub in this distribution mask */ - for (uvhub = 0; - uvhub < sizeof(struct bau_target_uvhubmask) * BITSPERBYTE; - uvhub++) { + maskbits = sizeof(struct bau_targ_hubmask) * BITSPERBYTE; + for (uvhub = 0; uvhub < maskbits; uvhub++) { + int cpu; if (!bau_uvhub_isset(uvhub, distribution)) continue; /* find a cpu for this uvhub */ cpu = uvhub_to_first_cpu(uvhub); cpu_set(cpu, mask); } - /* IPI all cpus; Preemption is already disabled */ - smp_call_function_many(&mask, uv_do_reset, (void *)&reset_args, 1); + + /* IPI all cpus; preemption is already disabled */ + smp_call_function_many(&mask, do_reset, (void *)&reset_args, 1); return; } -static inline unsigned long -cycles_2_us(unsigned long long cyc) +static inline unsigned long cycles_2_us(unsigned long long cyc) { unsigned long long ns; unsigned long us; - ns = (cyc * per_cpu(cyc2ns, smp_processor_id())) - >> CYC2NS_SCALE_FACTOR; + int cpu = smp_processor_id(); + + ns = (cyc * per_cpu(cyc2ns, cpu)) >> CYC2NS_SCALE_FACTOR; us = ns / 1000; return us; } @@ -376,56 +407,56 @@ cycles_2_us(unsigned long long cyc) * leaves uvhub_quiesce set so that no new broadcasts are started by * bau_flush_send_and_wait() */ -static inline void -quiesce_local_uvhub(struct bau_control *hmaster) +static inline void quiesce_local_uvhub(struct bau_control *hmaster) { - atomic_add_short_return(1, (struct atomic_short *) - &hmaster->uvhub_quiesce); + atom_asr(1, (struct atomic_short *)&hmaster->uvhub_quiesce); } /* * mark this quiet-requestor as done */ -static inline void -end_uvhub_quiesce(struct bau_control *hmaster) +static inline void end_uvhub_quiesce(struct bau_control *hmaster) { - atomic_add_short_return(-1, (struct atomic_short *) - &hmaster->uvhub_quiesce); + atom_asr(-1, (struct atomic_short *)&hmaster->uvhub_quiesce); +} + +static unsigned long uv1_read_status(unsigned long mmr_offset, int right_shift) +{ + unsigned long descriptor_status; + + descriptor_status = uv_read_local_mmr(mmr_offset); + descriptor_status >>= right_shift; + descriptor_status &= UV_ACT_STATUS_MASK; + return descriptor_status; } /* * Wait for completion of a broadcast software ack message * return COMPLETE, RETRY(PLUGGED or TIMEOUT) or GIVEUP */ -static int uv_wait_completion(struct bau_desc *bau_desc, - unsigned long mmr_offset, int right_shift, int this_cpu, - struct bau_control *bcp, struct bau_control *smaster, long try) +static int uv1_wait_completion(struct bau_desc *bau_desc, + unsigned long mmr_offset, int right_shift, + struct bau_control *bcp, long try) { unsigned long descriptor_status; - cycles_t ttime; + cycles_t ttm; struct ptc_stats *stat = bcp->statp; - struct bau_control *hmaster; - - hmaster = bcp->uvhub_master; + descriptor_status = uv1_read_status(mmr_offset, right_shift); /* spin on the status MMR, waiting for it to go idle */ - while ((descriptor_status = (((unsigned long) - uv_read_local_mmr(mmr_offset) >> - right_shift) & UV_ACT_STATUS_MASK)) != - DESC_STATUS_IDLE) { + while ((descriptor_status != DS_IDLE)) { /* - * Our software ack messages may be blocked because there are - * no swack resources available. As long as none of them - * has timed out hardware will NACK our message and its - * state will stay IDLE. + * Our software ack messages may be blocked because + * there are no swack resources available. As long + * as none of them has timed out hardware will NACK + * our message and its state will stay IDLE. */ - if (descriptor_status == DESC_STATUS_SOURCE_TIMEOUT) { + if (descriptor_status == DS_SOURCE_TIMEOUT) { stat->s_stimeout++; return FLUSH_GIVEUP; - } else if (descriptor_status == - DESC_STATUS_DESTINATION_TIMEOUT) { + } else if (descriptor_status == DS_DESTINATION_TIMEOUT) { stat->s_dtimeout++; - ttime = get_cycles(); + ttm = get_cycles(); /* * Our retries may be blocked by all destination @@ -433,8 +464,7 @@ static int uv_wait_completion(struct bau_desc *bau_desc, * pending. In that case hardware returns the * ERROR that looks like a destination timeout. */ - if (cycles_2_us(ttime - bcp->send_message) < - timeout_us) { + if (cycles_2_us(ttm - bcp->send_message) < timeout_us) { bcp->conseccompletes = 0; return FLUSH_RETRY_PLUGGED; } @@ -447,80 +477,160 @@ static int uv_wait_completion(struct bau_desc *bau_desc, */ cpu_relax(); } + descriptor_status = uv1_read_status(mmr_offset, right_shift); } bcp->conseccompletes++; return FLUSH_COMPLETE; } -static inline cycles_t -sec_2_cycles(unsigned long sec) +/* + * UV2 has an extra bit of status in the ACTIVATION_STATUS_2 register. + */ +static unsigned long uv2_read_status(unsigned long offset, int rshft, int cpu) { - unsigned long ns; - cycles_t cyc; + unsigned long descriptor_status; + unsigned long descriptor_status2; - ns = sec * 1000000000; - cyc = (ns << CYC2NS_SCALE_FACTOR)/(per_cpu(cyc2ns, smp_processor_id())); - return cyc; + descriptor_status = ((read_lmmr(offset) >> rshft) & UV_ACT_STATUS_MASK); + descriptor_status2 = (read_mmr_uv2_status() >> cpu) & 0x1UL; + descriptor_status = (descriptor_status << 1) | descriptor_status2; + return descriptor_status; +} + +static int uv2_wait_completion(struct bau_desc *bau_desc, + unsigned long mmr_offset, int right_shift, + struct bau_control *bcp, long try) +{ + unsigned long descriptor_stat; + cycles_t ttm; + int cpu = bcp->uvhub_cpu; + struct ptc_stats *stat = bcp->statp; + + descriptor_stat = uv2_read_status(mmr_offset, right_shift, cpu); + + /* spin on the status MMR, waiting for it to go idle */ + while (descriptor_stat != UV2H_DESC_IDLE) { + /* + * Our software ack messages may be blocked because + * there are no swack resources available. As long + * as none of them has timed out hardware will NACK + * our message and its state will stay IDLE. + */ + if ((descriptor_stat == UV2H_DESC_SOURCE_TIMEOUT) || + (descriptor_stat == UV2H_DESC_DEST_STRONG_NACK) || + (descriptor_stat == UV2H_DESC_DEST_PUT_ERR)) { + stat->s_stimeout++; + return FLUSH_GIVEUP; + } else if (descriptor_stat == UV2H_DESC_DEST_TIMEOUT) { + stat->s_dtimeout++; + ttm = get_cycles(); + /* + * Our retries may be blocked by all destination + * swack resources being consumed, and a timeout + * pending. In that case hardware returns the + * ERROR that looks like a destination timeout. + */ + if (cycles_2_us(ttm - bcp->send_message) < timeout_us) { + bcp->conseccompletes = 0; + return FLUSH_RETRY_PLUGGED; + } + bcp->conseccompletes = 0; + return FLUSH_RETRY_TIMEOUT; + } else { + /* + * descriptor_stat is still BUSY + */ + cpu_relax(); + } + descriptor_stat = uv2_read_status(mmr_offset, right_shift, cpu); + } + bcp->conseccompletes++; + return FLUSH_COMPLETE; } /* - * conditionally add 1 to *v, unless *v is >= u - * return 0 if we cannot add 1 to *v because it is >= u - * return 1 if we can add 1 to *v because it is < u - * the add is atomic - * - * This is close to atomic_add_unless(), but this allows the 'u' value - * to be lowered below the current 'v'. atomic_add_unless can only stop - * on equal. + * There are 2 status registers; each and array[32] of 2 bits. Set up for + * which register to read and position in that register based on cpu in + * current hub. */ -static inline int atomic_inc_unless_ge(spinlock_t *lock, atomic_t *v, int u) +static int wait_completion(struct bau_desc *bau_desc, + struct bau_control *bcp, long try) { - spin_lock(lock); - if (atomic_read(v) >= u) { - spin_unlock(lock); - return 0; + int right_shift; + unsigned long mmr_offset; + int cpu = bcp->uvhub_cpu; + + if (cpu < UV_CPUS_PER_AS) { + mmr_offset = UVH_LB_BAU_SB_ACTIVATION_STATUS_0; + right_shift = cpu * UV_ACT_STATUS_SIZE; + } else { + mmr_offset = UVH_LB_BAU_SB_ACTIVATION_STATUS_1; + right_shift = ((cpu - UV_CPUS_PER_AS) * UV_ACT_STATUS_SIZE); } - atomic_inc(v); - spin_unlock(lock); - return 1; + + if (is_uv1_hub()) + return uv1_wait_completion(bau_desc, mmr_offset, right_shift, + bcp, try); + else + return uv2_wait_completion(bau_desc, mmr_offset, right_shift, + bcp, try); +} + +static inline cycles_t sec_2_cycles(unsigned long sec) +{ + unsigned long ns; + cycles_t cyc; + + ns = sec * 1000000000; + cyc = (ns << CYC2NS_SCALE_FACTOR)/(per_cpu(cyc2ns, smp_processor_id())); + return cyc; } /* - * Our retries are blocked by all destination swack resources being + * Our retries are blocked by all destination sw ack resources being * in use, and a timeout is pending. In that case hardware immediately * returns the ERROR that looks like a destination timeout. */ -static void -destination_plugged(struct bau_desc *bau_desc, struct bau_control *bcp, +static void destination_plugged(struct bau_desc *bau_desc, + struct bau_control *bcp, struct bau_control *hmaster, struct ptc_stats *stat) { udelay(bcp->plugged_delay); bcp->plugged_tries++; + if (bcp->plugged_tries >= bcp->plugsb4reset) { bcp->plugged_tries = 0; + quiesce_local_uvhub(hmaster); + spin_lock(&hmaster->queue_lock); - uv_reset_with_ipi(&bau_desc->distribution, bcp->cpu); + reset_with_ipi(&bau_desc->distribution, bcp->cpu); spin_unlock(&hmaster->queue_lock); + end_uvhub_quiesce(hmaster); + bcp->ipi_attempts++; stat->s_resets_plug++; } } -static void -destination_timeout(struct bau_desc *bau_desc, struct bau_control *bcp, - struct bau_control *hmaster, struct ptc_stats *stat) +static void destination_timeout(struct bau_desc *bau_desc, + struct bau_control *bcp, struct bau_control *hmaster, + struct ptc_stats *stat) { - hmaster->max_bau_concurrent = 1; + hmaster->max_concurr = 1; bcp->timeout_tries++; if (bcp->timeout_tries >= bcp->timeoutsb4reset) { bcp->timeout_tries = 0; + quiesce_local_uvhub(hmaster); + spin_lock(&hmaster->queue_lock); - uv_reset_with_ipi(&bau_desc->distribution, bcp->cpu); + reset_with_ipi(&bau_desc->distribution, bcp->cpu); spin_unlock(&hmaster->queue_lock); + end_uvhub_quiesce(hmaster); + bcp->ipi_attempts++; stat->s_resets_timeout++; } @@ -530,34 +640,104 @@ destination_timeout(struct bau_desc *bau_desc, struct bau_control *bcp, * Completions are taking a very long time due to a congested numalink * network. */ -static void -disable_for_congestion(struct bau_control *bcp, struct ptc_stats *stat) +static void disable_for_congestion(struct bau_control *bcp, + struct ptc_stats *stat) { - int tcpu; - struct bau_control *tbcp; - /* let only one cpu do this disabling */ spin_lock(&disable_lock); + if (!baudisabled && bcp->period_requests && ((bcp->period_time / bcp->period_requests) > congested_cycles)) { + int tcpu; + struct bau_control *tbcp; /* it becomes this cpu's job to turn on the use of the BAU again */ baudisabled = 1; bcp->set_bau_off = 1; - bcp->set_bau_on_time = get_cycles() + - sec_2_cycles(bcp->congested_period); + bcp->set_bau_on_time = get_cycles(); + bcp->set_bau_on_time += sec_2_cycles(bcp->cong_period); stat->s_bau_disabled++; for_each_present_cpu(tcpu) { tbcp = &per_cpu(bau_control, tcpu); - tbcp->baudisabled = 1; + tbcp->baudisabled = 1; } } + spin_unlock(&disable_lock); } -/** - * uv_flush_send_and_wait - * +static void count_max_concurr(int stat, struct bau_control *bcp, + struct bau_control *hmaster) +{ + bcp->plugged_tries = 0; + bcp->timeout_tries = 0; + if (stat != FLUSH_COMPLETE) + return; + if (bcp->conseccompletes <= bcp->complete_threshold) + return; + if (hmaster->max_concurr >= hmaster->max_concurr_const) + return; + hmaster->max_concurr++; +} + +static void record_send_stats(cycles_t time1, cycles_t time2, + struct bau_control *bcp, struct ptc_stats *stat, + int completion_status, int try) +{ + cycles_t elapsed; + + if (time2 > time1) { + elapsed = time2 - time1; + stat->s_time += elapsed; + + if ((completion_status == FLUSH_COMPLETE) && (try == 1)) { + bcp->period_requests++; + bcp->period_time += elapsed; + if ((elapsed > congested_cycles) && + (bcp->period_requests > bcp->cong_reps)) + disable_for_congestion(bcp, stat); + } + } else + stat->s_requestor--; + + if (completion_status == FLUSH_COMPLETE && try > 1) + stat->s_retriesok++; + else if (completion_status == FLUSH_GIVEUP) + stat->s_giveup++; +} + +/* + * Because of a uv1 hardware bug only a limited number of concurrent + * requests can be made. + */ +static void uv1_throttle(struct bau_control *hmaster, struct ptc_stats *stat) +{ + spinlock_t *lock = &hmaster->uvhub_lock; + atomic_t *v; + + v = &hmaster->active_descriptor_count; + if (!atomic_inc_unless_ge(lock, v, hmaster->max_concurr)) { + stat->s_throttles++; + do { + cpu_relax(); + } while (!atomic_inc_unless_ge(lock, v, hmaster->max_concurr)); + } +} + +/* + * Handle the completion status of a message send. + */ +static void handle_cmplt(int completion_status, struct bau_desc *bau_desc, + struct bau_control *bcp, struct bau_control *hmaster, + struct ptc_stats *stat) +{ + if (completion_status == FLUSH_RETRY_PLUGGED) + destination_plugged(bau_desc, bcp, hmaster, stat); + else if (completion_status == FLUSH_RETRY_TIMEOUT) + destination_timeout(bau_desc, bcp, hmaster, stat); +} + +/* * Send a broadcast and wait for it to complete. * * The flush_mask contains the cpus the broadcast is to be sent to including @@ -568,44 +748,23 @@ disable_for_congestion(struct bau_control *bcp, struct ptc_stats *stat) * returned to the kernel. */ int uv_flush_send_and_wait(struct bau_desc *bau_desc, - struct cpumask *flush_mask, struct bau_control *bcp) + struct cpumask *flush_mask, struct bau_control *bcp) { - int right_shift; - int completion_status = 0; int seq_number = 0; + int completion_stat = 0; long try = 0; - int cpu = bcp->uvhub_cpu; - int this_cpu = bcp->cpu; - unsigned long mmr_offset; unsigned long index; cycles_t time1; cycles_t time2; - cycles_t elapsed; struct ptc_stats *stat = bcp->statp; - struct bau_control *smaster = bcp->socket_master; struct bau_control *hmaster = bcp->uvhub_master; - if (!atomic_inc_unless_ge(&hmaster->uvhub_lock, - &hmaster->active_descriptor_count, - hmaster->max_bau_concurrent)) { - stat->s_throttles++; - do { - cpu_relax(); - } while (!atomic_inc_unless_ge(&hmaster->uvhub_lock, - &hmaster->active_descriptor_count, - hmaster->max_bau_concurrent)); - } + if (is_uv1_hub()) + uv1_throttle(hmaster, stat); + while (hmaster->uvhub_quiesce) cpu_relax(); - if (cpu < UV_CPUS_PER_ACT_STATUS) { - mmr_offset = UVH_LB_BAU_SB_ACTIVATION_STATUS_0; - right_shift = cpu * UV_ACT_STATUS_SIZE; - } else { - mmr_offset = UVH_LB_BAU_SB_ACTIVATION_STATUS_1; - right_shift = - ((cpu - UV_CPUS_PER_ACT_STATUS) * UV_ACT_STATUS_SIZE); - } time1 = get_cycles(); do { if (try == 0) { @@ -615,64 +774,134 @@ int uv_flush_send_and_wait(struct bau_desc *bau_desc, bau_desc->header.msg_type = MSG_RETRY; stat->s_retry_messages++; } + bau_desc->header.sequence = seq_number; - index = (1UL << UVH_LB_BAU_SB_ACTIVATION_CONTROL_PUSH_SHFT) | - bcp->uvhub_cpu; + index = (1UL << AS_PUSH_SHIFT) | bcp->uvhub_cpu; bcp->send_message = get_cycles(); - uv_write_local_mmr(UVH_LB_BAU_SB_ACTIVATION_CONTROL, index); + + write_mmr_activation(index); + try++; - completion_status = uv_wait_completion(bau_desc, mmr_offset, - right_shift, this_cpu, bcp, smaster, try); + completion_stat = wait_completion(bau_desc, bcp, try); + + handle_cmplt(completion_stat, bau_desc, bcp, hmaster, stat); - if (completion_status == FLUSH_RETRY_PLUGGED) { - destination_plugged(bau_desc, bcp, hmaster, stat); - } else if (completion_status == FLUSH_RETRY_TIMEOUT) { - destination_timeout(bau_desc, bcp, hmaster, stat); - } if (bcp->ipi_attempts >= bcp->ipi_reset_limit) { bcp->ipi_attempts = 0; - completion_status = FLUSH_GIVEUP; + completion_stat = FLUSH_GIVEUP; break; } cpu_relax(); - } while ((completion_status == FLUSH_RETRY_PLUGGED) || - (completion_status == FLUSH_RETRY_TIMEOUT)); + } while ((completion_stat == FLUSH_RETRY_PLUGGED) || + (completion_stat == FLUSH_RETRY_TIMEOUT)); + time2 = get_cycles(); - bcp->plugged_tries = 0; - bcp->timeout_tries = 0; - if ((completion_status == FLUSH_COMPLETE) && - (bcp->conseccompletes > bcp->complete_threshold) && - (hmaster->max_bau_concurrent < - hmaster->max_bau_concurrent_constant)) - hmaster->max_bau_concurrent++; + + count_max_concurr(completion_stat, bcp, hmaster); + while (hmaster->uvhub_quiesce) cpu_relax(); + atomic_dec(&hmaster->active_descriptor_count); - if (time2 > time1) { - elapsed = time2 - time1; - stat->s_time += elapsed; - if ((completion_status == FLUSH_COMPLETE) && (try == 1)) { - bcp->period_requests++; - bcp->period_time += elapsed; - if ((elapsed > congested_cycles) && - (bcp->period_requests > bcp->congested_reps)) { - disable_for_congestion(bcp, stat); + + record_send_stats(time1, time2, bcp, stat, completion_stat, try); + + if (completion_stat == FLUSH_GIVEUP) + return 1; + return 0; +} + +/* + * The BAU is disabled. When the disabled time period has expired, the cpu + * that disabled it must re-enable it. + * Return 0 if it is re-enabled for all cpus. + */ +static int check_enable(struct bau_control *bcp, struct ptc_stats *stat) +{ + int tcpu; + struct bau_control *tbcp; + + if (bcp->set_bau_off) { + if (get_cycles() >= bcp->set_bau_on_time) { + stat->s_bau_reenabled++; + baudisabled = 0; + for_each_present_cpu(tcpu) { + tbcp = &per_cpu(bau_control, tcpu); + tbcp->baudisabled = 0; + tbcp->period_requests = 0; + tbcp->period_time = 0; } + return 0; } + } + return -1; +} + +static void record_send_statistics(struct ptc_stats *stat, int locals, int hubs, + int remotes, struct bau_desc *bau_desc) +{ + stat->s_requestor++; + stat->s_ntargcpu += remotes + locals; + stat->s_ntargremotes += remotes; + stat->s_ntarglocals += locals; + + /* uvhub statistics */ + hubs = bau_uvhub_weight(&bau_desc->distribution); + if (locals) { + stat->s_ntarglocaluvhub++; + stat->s_ntargremoteuvhub += (hubs - 1); } else - stat->s_requestor--; - if (completion_status == FLUSH_COMPLETE && try > 1) - stat->s_retriesok++; - else if (completion_status == FLUSH_GIVEUP) { - stat->s_giveup++; - return 1; + stat->s_ntargremoteuvhub += hubs; + + stat->s_ntarguvhub += hubs; + + if (hubs >= 16) + stat->s_ntarguvhub16++; + else if (hubs >= 8) + stat->s_ntarguvhub8++; + else if (hubs >= 4) + stat->s_ntarguvhub4++; + else if (hubs >= 2) + stat->s_ntarguvhub2++; + else + stat->s_ntarguvhub1++; +} + +/* + * Translate a cpu mask to the uvhub distribution mask in the BAU + * activation descriptor. + */ +static int set_distrib_bits(struct cpumask *flush_mask, struct bau_control *bcp, + struct bau_desc *bau_desc, int *localsp, int *remotesp) +{ + int cpu; + int pnode; + int cnt = 0; + struct hub_and_pnode *hpp; + + for_each_cpu(cpu, flush_mask) { + /* + * The distribution vector is a bit map of pnodes, relative + * to the partition base pnode (and the partition base nasid + * in the header). + * Translate cpu to pnode and hub using a local memory array. + */ + hpp = &bcp->socket_master->thp[cpu]; + pnode = hpp->pnode - bcp->partition_base_pnode; + bau_uvhub_set(pnode, &bau_desc->distribution); + cnt++; + if (hpp->uvhub == bcp->uvhub) + (*localsp)++; + else + (*remotesp)++; } + if (!cnt) + return 1; return 0; } -/** - * uv_flush_tlb_others - globally purge translation cache of a virtual - * address or all TLB's +/* + * globally purge translation cache of a virtual address or all TLB's * @cpumask: mask of all cpu's in which the address is to be removed * @mm: mm_struct containing virtual address range * @va: virtual address to be removed (or TLB_FLUSH_ALL for all TLB's on cpu) @@ -696,20 +925,16 @@ int uv_flush_send_and_wait(struct bau_desc *bau_desc, * done. The returned pointer is valid till preemption is re-enabled. */ const struct cpumask *uv_flush_tlb_others(const struct cpumask *cpumask, - struct mm_struct *mm, - unsigned long va, unsigned int cpu) + struct mm_struct *mm, unsigned long va, + unsigned int cpu) { int locals = 0; int remotes = 0; int hubs = 0; - int tcpu; - int tpnode; struct bau_desc *bau_desc; struct cpumask *flush_mask; struct ptc_stats *stat; struct bau_control *bcp; - struct bau_control *tbcp; - struct hub_and_pnode *hpp; /* kernel was booted 'nobau' */ if (nobau) @@ -720,20 +945,8 @@ const struct cpumask *uv_flush_tlb_others(const struct cpumask *cpumask, /* bau was disabled due to slow response */ if (bcp->baudisabled) { - /* the cpu that disabled it must re-enable it */ - if (bcp->set_bau_off) { - if (get_cycles() >= bcp->set_bau_on_time) { - stat->s_bau_reenabled++; - baudisabled = 0; - for_each_present_cpu(tcpu) { - tbcp = &per_cpu(bau_control, tcpu); - tbcp->baudisabled = 0; - tbcp->period_requests = 0; - tbcp->period_time = 0; - } - } - } - return cpumask; + if (check_enable(bcp, stat)) + return cpumask; } /* @@ -744,59 +957,20 @@ const struct cpumask *uv_flush_tlb_others(const struct cpumask *cpumask, flush_mask = (struct cpumask *)per_cpu(uv_flush_tlb_mask, cpu); /* don't actually do a shootdown of the local cpu */ cpumask_andnot(flush_mask, cpumask, cpumask_of(cpu)); + if (cpu_isset(cpu, *cpumask)) stat->s_ntargself++; bau_desc = bcp->descriptor_base; - bau_desc += UV_ITEMS_PER_DESCRIPTOR * bcp->uvhub_cpu; + bau_desc += ITEMS_PER_DESC * bcp->uvhub_cpu; bau_uvhubs_clear(&bau_desc->distribution, UV_DISTRIBUTION_SIZE); - - for_each_cpu(tcpu, flush_mask) { - /* - * The distribution vector is a bit map of pnodes, relative - * to the partition base pnode (and the partition base nasid - * in the header). - * Translate cpu to pnode and hub using an array stored - * in local memory. - */ - hpp = &bcp->socket_master->target_hub_and_pnode[tcpu]; - tpnode = hpp->pnode - bcp->partition_base_pnode; - bau_uvhub_set(tpnode, &bau_desc->distribution); - if (hpp->uvhub == bcp->uvhub) - locals++; - else - remotes++; - } - if ((locals + remotes) == 0) + if (set_distrib_bits(flush_mask, bcp, bau_desc, &locals, &remotes)) return NULL; - stat->s_requestor++; - stat->s_ntargcpu += remotes + locals; - stat->s_ntargremotes += remotes; - stat->s_ntarglocals += locals; - remotes = bau_uvhub_weight(&bau_desc->distribution); - /* uvhub statistics */ - hubs = bau_uvhub_weight(&bau_desc->distribution); - if (locals) { - stat->s_ntarglocaluvhub++; - stat->s_ntargremoteuvhub += (hubs - 1); - } else - stat->s_ntargremoteuvhub += hubs; - stat->s_ntarguvhub += hubs; - if (hubs >= 16) - stat->s_ntarguvhub16++; - else if (hubs >= 8) - stat->s_ntarguvhub8++; - else if (hubs >= 4) - stat->s_ntarguvhub4++; - else if (hubs >= 2) - stat->s_ntarguvhub2++; - else - stat->s_ntarguvhub1++; + record_send_statistics(stat, locals, hubs, remotes, bau_desc); bau_desc->payload.address = va; bau_desc->payload.sending_cpu = cpu; - /* * uv_flush_send_and_wait returns 0 if all cpu's were messaged, * or 1 if it gave up and the original cpumask should be returned. @@ -825,26 +999,31 @@ void uv_bau_message_interrupt(struct pt_regs *regs) { int count = 0; cycles_t time_start; - struct bau_payload_queue_entry *msg; + struct bau_pq_entry *msg; struct bau_control *bcp; struct ptc_stats *stat; struct msg_desc msgdesc; time_start = get_cycles(); + bcp = &per_cpu(bau_control, smp_processor_id()); stat = bcp->statp; - msgdesc.va_queue_first = bcp->va_queue_first; - msgdesc.va_queue_last = bcp->va_queue_last; + + msgdesc.queue_first = bcp->queue_first; + msgdesc.queue_last = bcp->queue_last; + msg = bcp->bau_msg_head; - while (msg->sw_ack_vector) { + while (msg->swack_vec) { count++; - msgdesc.msg_slot = msg - msgdesc.va_queue_first; - msgdesc.sw_ack_slot = ffs(msg->sw_ack_vector) - 1; + + msgdesc.msg_slot = msg - msgdesc.queue_first; + msgdesc.swack_slot = ffs(msg->swack_vec) - 1; msgdesc.msg = msg; - uv_bau_process_message(&msgdesc, bcp); + bau_process_message(&msgdesc, bcp); + msg++; - if (msg > msgdesc.va_queue_last) - msg = msgdesc.va_queue_first; + if (msg > msgdesc.queue_last) + msg = msgdesc.queue_first; bcp->bau_msg_head = msg; } stat->d_time += (get_cycles() - time_start); @@ -852,18 +1031,17 @@ void uv_bau_message_interrupt(struct pt_regs *regs) stat->d_nomsg++; else if (count > 1) stat->d_multmsg++; + ack_APIC_irq(); } /* - * uv_enable_timeouts - * - * Each target uvhub (i.e. a uvhub that has no cpu's) needs to have + * Each target uvhub (i.e. a uvhub that has cpu's) needs to have * shootdown message timeouts enabled. The timeout does not cause * an interrupt, but causes an error message to be returned to * the sender. */ -static void __init uv_enable_timeouts(void) +static void __init enable_timeouts(void) { int uvhub; int nuvhubs; @@ -877,47 +1055,44 @@ static void __init uv_enable_timeouts(void) continue; pnode = uv_blade_to_pnode(uvhub); - mmr_image = - uv_read_global_mmr64(pnode, UVH_LB_BAU_MISC_CONTROL); + mmr_image = read_mmr_misc_control(pnode); /* * Set the timeout period and then lock it in, in three * steps; captures and locks in the period. * * To program the period, the SOFT_ACK_MODE must be off. */ - mmr_image &= ~((unsigned long)1 << - UVH_LB_BAU_MISC_CONTROL_ENABLE_INTD_SOFT_ACK_MODE_SHFT); - uv_write_global_mmr64 - (pnode, UVH_LB_BAU_MISC_CONTROL, mmr_image); + mmr_image &= ~(1L << SOFTACK_MSHIFT); + write_mmr_misc_control(pnode, mmr_image); /* * Set the 4-bit period. */ - mmr_image &= ~((unsigned long)0xf << - UVH_LB_BAU_MISC_CONTROL_INTD_SOFT_ACK_TIMEOUT_PERIOD_SHFT); - mmr_image |= (UV_INTD_SOFT_ACK_TIMEOUT_PERIOD << - UVH_LB_BAU_MISC_CONTROL_INTD_SOFT_ACK_TIMEOUT_PERIOD_SHFT); - uv_write_global_mmr64 - (pnode, UVH_LB_BAU_MISC_CONTROL, mmr_image); + mmr_image &= ~((unsigned long)0xf << SOFTACK_PSHIFT); + mmr_image |= (SOFTACK_TIMEOUT_PERIOD << SOFTACK_PSHIFT); + write_mmr_misc_control(pnode, mmr_image); /* + * UV1: * Subsequent reversals of the timebase bit (3) cause an * immediate timeout of one or all INTD resources as * indicated in bits 2:0 (7 causes all of them to timeout). */ - mmr_image |= ((unsigned long)1 << - UVH_LB_BAU_MISC_CONTROL_ENABLE_INTD_SOFT_ACK_MODE_SHFT); - uv_write_global_mmr64 - (pnode, UVH_LB_BAU_MISC_CONTROL, mmr_image); + mmr_image |= (1L << SOFTACK_MSHIFT); + if (is_uv2_hub()) { + mmr_image |= (1L << UV2_LEG_SHFT); + mmr_image |= (1L << UV2_EXT_SHFT); + } + write_mmr_misc_control(pnode, mmr_image); } } -static void *uv_ptc_seq_start(struct seq_file *file, loff_t *offset) +static void *ptc_seq_start(struct seq_file *file, loff_t *offset) { if (*offset < num_possible_cpus()) return offset; return NULL; } -static void *uv_ptc_seq_next(struct seq_file *file, void *data, loff_t *offset) +static void *ptc_seq_next(struct seq_file *file, void *data, loff_t *offset) { (*offset)++; if (*offset < num_possible_cpus()) @@ -925,12 +1100,11 @@ static void *uv_ptc_seq_next(struct seq_file *file, void *data, loff_t *offset) return NULL; } -static void uv_ptc_seq_stop(struct seq_file *file, void *data) +static void ptc_seq_stop(struct seq_file *file, void *data) { } -static inline unsigned long long -microsec_2_cycles(unsigned long microsec) +static inline unsigned long long usec_2_cycles(unsigned long microsec) { unsigned long ns; unsigned long long cyc; @@ -941,29 +1115,27 @@ microsec_2_cycles(unsigned long microsec) } /* - * Display the statistics thru /proc. + * Display the statistics thru /proc/sgi_uv/ptc_statistics * 'data' points to the cpu number + * Note: see the descriptions in stat_description[]. */ -static int uv_ptc_seq_show(struct seq_file *file, void *data) +static int ptc_seq_show(struct seq_file *file, void *data) { struct ptc_stats *stat; int cpu; cpu = *(loff_t *)data; - if (!cpu) { seq_printf(file, "# cpu sent stime self locals remotes ncpus localhub "); seq_printf(file, "remotehub numuvhubs numuvhubs16 numuvhubs8 "); seq_printf(file, - "numuvhubs4 numuvhubs2 numuvhubs1 dto "); - seq_printf(file, - "retries rok resetp resett giveup sto bz throt "); + "numuvhubs4 numuvhubs2 numuvhubs1 dto retries rok "); seq_printf(file, - "sw_ack recv rtime all "); + "resetp resett giveup sto bz throt swack recv rtime "); seq_printf(file, - "one mult none retry canc nocan reset rcan "); + "all one mult none retry canc nocan reset rcan "); seq_printf(file, "disable enable\n"); } @@ -990,8 +1162,7 @@ static int uv_ptc_seq_show(struct seq_file *file, void *data) /* destination side statistics */ seq_printf(file, "%lx %ld %ld %ld %ld %ld %ld %ld %ld %ld %ld %ld ", - uv_read_global_mmr64(uv_cpu_to_pnode(cpu), - UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE), + read_gmmr_sw_ack(uv_cpu_to_pnode(cpu)), stat->d_requestee, cycles_2_us(stat->d_time), stat->d_alltlb, stat->d_onetlb, stat->d_multmsg, stat->d_nomsg, stat->d_retries, stat->d_canceled, @@ -1000,7 +1171,6 @@ static int uv_ptc_seq_show(struct seq_file *file, void *data) seq_printf(file, "%ld %ld\n", stat->s_bau_disabled, stat->s_bau_reenabled); } - return 0; } @@ -1008,18 +1178,18 @@ static int uv_ptc_seq_show(struct seq_file *file, void *data) * Display the tunables thru debugfs */ static ssize_t tunables_read(struct file *file, char __user *userbuf, - size_t count, loff_t *ppos) + size_t count, loff_t *ppos) { char *buf; int ret; buf = kasprintf(GFP_KERNEL, "%s %s %s\n%d %d %d %d %d %d %d %d %d\n", - "max_bau_concurrent plugged_delay plugsb4reset", + "max_concur plugged_delay plugsb4reset", "timeoutsb4reset ipi_reset_limit complete_threshold", "congested_response_us congested_reps congested_period", - max_bau_concurrent, plugged_delay, plugsb4reset, + max_concurr, plugged_delay, plugsb4reset, timeoutsb4reset, ipi_reset_limit, complete_threshold, - congested_response_us, congested_reps, congested_period); + congested_respns_us, congested_reps, congested_period); if (!buf) return -ENOMEM; @@ -1030,13 +1200,16 @@ static ssize_t tunables_read(struct file *file, char __user *userbuf, } /* - * -1: resetf the statistics + * handle a write to /proc/sgi_uv/ptc_statistics + * -1: reset the statistics * 0: display meaning of the statistics */ -static ssize_t uv_ptc_proc_write(struct file *file, const char __user *user, - size_t count, loff_t *data) +static ssize_t ptc_proc_write(struct file *file, const char __user *user, + size_t count, loff_t *data) { int cpu; + int i; + int elements; long input_arg; char optstr[64]; struct ptc_stats *stat; @@ -1046,79 +1219,18 @@ static ssize_t uv_ptc_proc_write(struct file *file, const char __user *user, if (copy_from_user(optstr, user, count)) return -EFAULT; optstr[count - 1] = '\0'; + if (strict_strtol(optstr, 10, &input_arg) < 0) { printk(KERN_DEBUG "%s is invalid\n", optstr); return -EINVAL; } if (input_arg == 0) { + elements = sizeof(stat_description)/sizeof(*stat_description); printk(KERN_DEBUG "# cpu: cpu number\n"); printk(KERN_DEBUG "Sender statistics:\n"); - printk(KERN_DEBUG - "sent: number of shootdown messages sent\n"); - printk(KERN_DEBUG - "stime: time spent sending messages\n"); - printk(KERN_DEBUG - "numuvhubs: number of hubs targeted with shootdown\n"); - printk(KERN_DEBUG - "numuvhubs16: number times 16 or more hubs targeted\n"); - printk(KERN_DEBUG - "numuvhubs8: number times 8 or more hubs targeted\n"); - printk(KERN_DEBUG - "numuvhubs4: number times 4 or more hubs targeted\n"); - printk(KERN_DEBUG - "numuvhubs2: number times 2 or more hubs targeted\n"); - printk(KERN_DEBUG - "numuvhubs1: number times 1 hub targeted\n"); - printk(KERN_DEBUG - "numcpus: number of cpus targeted with shootdown\n"); - printk(KERN_DEBUG - "dto: number of destination timeouts\n"); - printk(KERN_DEBUG - "retries: destination timeout retries sent\n"); - printk(KERN_DEBUG - "rok: : destination timeouts successfully retried\n"); - printk(KERN_DEBUG - "resetp: ipi-style resource resets for plugs\n"); - printk(KERN_DEBUG - "resett: ipi-style resource resets for timeouts\n"); - printk(KERN_DEBUG - "giveup: fall-backs to ipi-style shootdowns\n"); - printk(KERN_DEBUG - "sto: number of source timeouts\n"); - printk(KERN_DEBUG - "bz: number of stay-busy's\n"); - printk(KERN_DEBUG - "throt: number times spun in throttle\n"); - printk(KERN_DEBUG "Destination side statistics:\n"); - printk(KERN_DEBUG - "sw_ack: image of UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE\n"); - printk(KERN_DEBUG - "recv: shootdown messages received\n"); - printk(KERN_DEBUG - "rtime: time spent processing messages\n"); - printk(KERN_DEBUG - "all: shootdown all-tlb messages\n"); - printk(KERN_DEBUG - "one: shootdown one-tlb messages\n"); - printk(KERN_DEBUG - "mult: interrupts that found multiple messages\n"); - printk(KERN_DEBUG - "none: interrupts that found no messages\n"); - printk(KERN_DEBUG - "retry: number of retry messages processed\n"); - printk(KERN_DEBUG - "canc: number messages canceled by retries\n"); - printk(KERN_DEBUG - "nocan: number retries that found nothing to cancel\n"); - printk(KERN_DEBUG - "reset: number of ipi-style reset requests processed\n"); - printk(KERN_DEBUG - "rcan: number messages canceled by reset requests\n"); - printk(KERN_DEBUG - "disable: number times use of the BAU was disabled\n"); - printk(KERN_DEBUG - "enable: number times use of the BAU was re-enabled\n"); + for (i = 0; i < elements; i++) + printk(KERN_DEBUG "%s\n", stat_description[i]); } else if (input_arg == -1) { for_each_present_cpu(cpu) { stat = &per_cpu(ptcstats, cpu); @@ -1145,27 +1257,18 @@ static int local_atoi(const char *name) } /* - * set the tunables - * 0 values reset them to defaults + * Parse the values written to /sys/kernel/debug/sgi_uv/bau_tunables. + * Zero values reset them to defaults. */ -static ssize_t tunables_write(struct file *file, const char __user *user, - size_t count, loff_t *data) +static int parse_tunables_write(struct bau_control *bcp, char *instr, + int count) { - int cpu; - int cnt = 0; - int val; char *p; char *q; - char instr[64]; - struct bau_control *bcp; - - if (count == 0 || count > sizeof(instr)-1) - return -EINVAL; - if (copy_from_user(instr, user, count)) - return -EFAULT; + int cnt = 0; + int val; + int e = sizeof(tunables) / sizeof(*tunables); - instr[count] = '\0'; - /* count the fields */ p = instr + strspn(instr, WHITESPACE); q = p; for (; *p; p = q + strspn(q, WHITESPACE)) { @@ -1174,8 +1277,8 @@ static ssize_t tunables_write(struct file *file, const char __user *user, if (q == p) break; } - if (cnt != 9) { - printk(KERN_INFO "bau tunable error: should be 9 numbers\n"); + if (cnt != e) { + printk(KERN_INFO "bau tunable error: should be %d values\n", e); return -EINVAL; } @@ -1187,97 +1290,80 @@ static ssize_t tunables_write(struct file *file, const char __user *user, switch (cnt) { case 0: if (val == 0) { - max_bau_concurrent = MAX_BAU_CONCURRENT; - max_bau_concurrent_constant = - MAX_BAU_CONCURRENT; + max_concurr = MAX_BAU_CONCURRENT; + max_concurr_const = MAX_BAU_CONCURRENT; continue; } - bcp = &per_cpu(bau_control, smp_processor_id()); if (val < 1 || val > bcp->cpus_in_uvhub) { printk(KERN_DEBUG "Error: BAU max concurrent %d is invalid\n", val); return -EINVAL; } - max_bau_concurrent = val; - max_bau_concurrent_constant = val; - continue; - case 1: - if (val == 0) - plugged_delay = PLUGGED_DELAY; - else - plugged_delay = val; - continue; - case 2: - if (val == 0) - plugsb4reset = PLUGSB4RESET; - else - plugsb4reset = val; - continue; - case 3: - if (val == 0) - timeoutsb4reset = TIMEOUTSB4RESET; - else - timeoutsb4reset = val; - continue; - case 4: - if (val == 0) - ipi_reset_limit = IPI_RESET_LIMIT; - else - ipi_reset_limit = val; - continue; - case 5: - if (val == 0) - complete_threshold = COMPLETE_THRESHOLD; - else - complete_threshold = val; - continue; - case 6: - if (val == 0) - congested_response_us = CONGESTED_RESPONSE_US; - else - congested_response_us = val; - continue; - case 7: - if (val == 0) - congested_reps = CONGESTED_REPS; - else - congested_reps = val; + max_concurr = val; + max_concurr_const = val; continue; - case 8: + default: if (val == 0) - congested_period = CONGESTED_PERIOD; + *tunables[cnt].tunp = tunables[cnt].deflt; else - congested_period = val; + *tunables[cnt].tunp = val; continue; } if (q == p) break; } + return 0; +} + +/* + * Handle a write to debugfs. (/sys/kernel/debug/sgi_uv/bau_tunables) + */ +static ssize_t tunables_write(struct file *file, const char __user *user, + size_t count, loff_t *data) +{ + int cpu; + int ret; + char instr[100]; + struct bau_control *bcp; + + if (count == 0 || count > sizeof(instr)-1) + return -EINVAL; + if (copy_from_user(instr, user, count)) + return -EFAULT; + + instr[count] = '\0'; + + bcp = &per_cpu(bau_control, smp_processor_id()); + + ret = parse_tunables_write(bcp, instr, count); + if (ret) + return ret; + for_each_present_cpu(cpu) { bcp = &per_cpu(bau_control, cpu); - bcp->max_bau_concurrent = max_bau_concurrent; - bcp->max_bau_concurrent_constant = max_bau_concurrent; - bcp->plugged_delay = plugged_delay; - bcp->plugsb4reset = plugsb4reset; - bcp->timeoutsb4reset = timeoutsb4reset; - bcp->ipi_reset_limit = ipi_reset_limit; - bcp->complete_threshold = complete_threshold; - bcp->congested_response_us = congested_response_us; - bcp->congested_reps = congested_reps; - bcp->congested_period = congested_period; + bcp->max_concurr = max_concurr; + bcp->max_concurr_const = max_concurr; + bcp->plugged_delay = plugged_delay; + bcp->plugsb4reset = plugsb4reset; + bcp->timeoutsb4reset = timeoutsb4reset; + bcp->ipi_reset_limit = ipi_reset_limit; + bcp->complete_threshold = complete_threshold; + bcp->cong_response_us = congested_respns_us; + bcp->cong_reps = congested_reps; + bcp->cong_period = congested_period; } return count; } static const struct seq_operations uv_ptc_seq_ops = { - .start = uv_ptc_seq_start, - .next = uv_ptc_seq_next, - .stop = uv_ptc_seq_stop, - .show = uv_ptc_seq_show + .start = ptc_seq_start, + .next = ptc_seq_next, + .stop = ptc_seq_stop, + .show = ptc_seq_show }; -static int uv_ptc_proc_open(struct inode *inode, struct file *file) +static int ptc_proc_open(struct inode *inode, struct file *file) { return seq_open(file, &uv_ptc_seq_ops); } @@ -1288,9 +1374,9 @@ static int tunables_open(struct inode *inode, struct file *file) } static const struct file_operations proc_uv_ptc_operations = { - .open = uv_ptc_proc_open, + .open = ptc_proc_open, .read = seq_read, - .write = uv_ptc_proc_write, + .write = ptc_proc_write, .llseek = seq_lseek, .release = seq_release, }; @@ -1324,7 +1410,7 @@ static int __init uv_ptc_init(void) return -EINVAL; } tunables_file = debugfs_create_file(UV_BAU_TUNABLES_FILE, 0600, - tunables_dir, NULL, &tunables_fops); + tunables_dir, NULL, &tunables_fops); if (!tunables_file) { printk(KERN_ERR "unable to create debugfs file %s\n", UV_BAU_TUNABLES_FILE); @@ -1336,24 +1422,24 @@ static int __init uv_ptc_init(void) /* * Initialize the sending side's sending buffers. */ -static void -uv_activation_descriptor_init(int node, int pnode, int base_pnode) +static void activation_descriptor_init(int node, int pnode, int base_pnode) { int i; int cpu; unsigned long pa; unsigned long m; unsigned long n; + size_t dsize; struct bau_desc *bau_desc; struct bau_desc *bd2; struct bau_control *bcp; /* - * each bau_desc is 64 bytes; there are 8 (UV_ITEMS_PER_DESCRIPTOR) - * per cpu; and one per cpu on the uvhub (UV_ADP_SIZE) + * each bau_desc is 64 bytes; there are 8 (ITEMS_PER_DESC) + * per cpu; and one per cpu on the uvhub (ADP_SZ) */ - bau_desc = kmalloc_node(sizeof(struct bau_desc) * UV_ADP_SIZE - * UV_ITEMS_PER_DESCRIPTOR, GFP_KERNEL, node); + dsize = sizeof(struct bau_desc) * ADP_SZ * ITEMS_PER_DESC; + bau_desc = kmalloc_node(dsize, GFP_KERNEL, node); BUG_ON(!bau_desc); pa = uv_gpa(bau_desc); /* need the real nasid*/ @@ -1361,27 +1447,25 @@ uv_activation_descriptor_init(int node, int pnode, int base_pnode) m = pa & uv_mmask; /* the 14-bit pnode */ - uv_write_global_mmr64(pnode, UVH_LB_BAU_SB_DESCRIPTOR_BASE, - (n << UV_DESC_BASE_PNODE_SHIFT | m)); + write_mmr_descriptor_base(pnode, (n << UV_DESC_PSHIFT | m)); /* - * Initializing all 8 (UV_ITEMS_PER_DESCRIPTOR) descriptors for each + * Initializing all 8 (ITEMS_PER_DESC) descriptors for each * cpu even though we only use the first one; one descriptor can * describe a broadcast to 256 uv hubs. */ - for (i = 0, bd2 = bau_desc; i < (UV_ADP_SIZE*UV_ITEMS_PER_DESCRIPTOR); - i++, bd2++) { + for (i = 0, bd2 = bau_desc; i < (ADP_SZ * ITEMS_PER_DESC); i++, bd2++) { memset(bd2, 0, sizeof(struct bau_desc)); - bd2->header.sw_ack_flag = 1; + bd2->header.swack_flag = 1; /* * The base_dest_nasid set in the message header is the nasid * of the first uvhub in the partition. The bit map will * indicate destination pnode numbers relative to that base. * They may not be consecutive if nasid striding is being used. */ - bd2->header.base_dest_nasid = UV_PNODE_TO_NASID(base_pnode); - bd2->header.dest_subnodeid = UV_LB_SUBNODEID; - bd2->header.command = UV_NET_ENDPOINT_INTD; - bd2->header.int_both = 1; + bd2->header.base_dest_nasid = UV_PNODE_TO_NASID(base_pnode); + bd2->header.dest_subnodeid = UV_LB_SUBNODEID; + bd2->header.command = UV_NET_ENDPOINT_INTD; + bd2->header.int_both = 1; /* * all others need to be set to zero: * fairness chaining multilevel count replied_to @@ -1401,57 +1485,55 @@ uv_activation_descriptor_init(int node, int pnode, int base_pnode) * - node is first node (kernel memory notion) on the uvhub * - pnode is the uvhub's physical identifier */ -static void -uv_payload_queue_init(int node, int pnode) +static void pq_init(int node, int pnode) { - int pn; int cpu; + size_t plsize; char *cp; - unsigned long pa; - struct bau_payload_queue_entry *pqp; - struct bau_payload_queue_entry *pqp_malloc; + void *vp; + unsigned long pn; + unsigned long first; + unsigned long pn_first; + unsigned long last; + struct bau_pq_entry *pqp; struct bau_control *bcp; - pqp = kmalloc_node((DEST_Q_SIZE + 1) - * sizeof(struct bau_payload_queue_entry), - GFP_KERNEL, node); + plsize = (DEST_Q_SIZE + 1) * sizeof(struct bau_pq_entry); + vp = kmalloc_node(plsize, GFP_KERNEL, node); + pqp = (struct bau_pq_entry *)vp; BUG_ON(!pqp); - pqp_malloc = pqp; cp = (char *)pqp + 31; - pqp = (struct bau_payload_queue_entry *)(((unsigned long)cp >> 5) << 5); + pqp = (struct bau_pq_entry *)(((unsigned long)cp >> 5) << 5); for_each_present_cpu(cpu) { if (pnode != uv_cpu_to_pnode(cpu)) continue; /* for every cpu on this pnode: */ bcp = &per_cpu(bau_control, cpu); - bcp->va_queue_first = pqp; - bcp->bau_msg_head = pqp; - bcp->va_queue_last = pqp + (DEST_Q_SIZE - 1); + bcp->queue_first = pqp; + bcp->bau_msg_head = pqp; + bcp->queue_last = pqp + (DEST_Q_SIZE - 1); } /* * need the pnode of where the memory was really allocated */ - pa = uv_gpa(pqp); - pn = pa >> uv_nshift; - uv_write_global_mmr64(pnode, - UVH_LB_BAU_INTD_PAYLOAD_QUEUE_FIRST, - ((unsigned long)pn << UV_PAYLOADQ_PNODE_SHIFT) | - uv_physnodeaddr(pqp)); - uv_write_global_mmr64(pnode, UVH_LB_BAU_INTD_PAYLOAD_QUEUE_TAIL, - uv_physnodeaddr(pqp)); - uv_write_global_mmr64(pnode, UVH_LB_BAU_INTD_PAYLOAD_QUEUE_LAST, - (unsigned long) - uv_physnodeaddr(pqp + (DEST_Q_SIZE - 1))); + pn = uv_gpa(pqp) >> uv_nshift; + first = uv_physnodeaddr(pqp); + pn_first = ((unsigned long)pn << UV_PAYLOADQ_PNODE_SHIFT) | first; + last = uv_physnodeaddr(pqp + (DEST_Q_SIZE - 1)); + write_mmr_payload_first(pnode, pn_first); + write_mmr_payload_tail(pnode, first); + write_mmr_payload_last(pnode, last); + /* in effect, all msg_type's are set to MSG_NOOP */ - memset(pqp, 0, sizeof(struct bau_payload_queue_entry) * DEST_Q_SIZE); + memset(pqp, 0, sizeof(struct bau_pq_entry) * DEST_Q_SIZE); } /* * Initialization of each UV hub's structures */ -static void __init uv_init_uvhub(int uvhub, int vector, int base_pnode) +static void __init init_uvhub(int uvhub, int vector, int base_pnode) { int node; int pnode; @@ -1459,24 +1541,24 @@ static void __init uv_init_uvhub(int uvhub, int vector, int base_pnode) node = uvhub_to_first_node(uvhub); pnode = uv_blade_to_pnode(uvhub); - uv_activation_descriptor_init(node, pnode, base_pnode); - uv_payload_queue_init(node, pnode); + + activation_descriptor_init(node, pnode, base_pnode); + + pq_init(node, pnode); /* * The below initialization can't be in firmware because the * messaging IRQ will be determined by the OS. */ apicid = uvhub_to_first_apicid(uvhub) | uv_apicid_hibits; - uv_write_global_mmr64(pnode, UVH_BAU_DATA_CONFIG, - ((apicid << 32) | vector)); + write_mmr_data_config(pnode, ((apicid << 32) | vector)); } /* * We will set BAU_MISC_CONTROL with a timeout period. * But the BIOS has set UVH_AGING_PRESCALE_SEL and UVH_TRANSACTION_TIMEOUT. - * So the destination timeout period has be be calculated from them. + * So the destination timeout period has to be calculated from them. */ -static int -calculate_destination_timeout(void) +static int calculate_destination_timeout(void) { unsigned long mmr_image; int mult1; @@ -1486,73 +1568,92 @@ calculate_destination_timeout(void) int ret; unsigned long ts_ns; - mult1 = UV_INTD_SOFT_ACK_TIMEOUT_PERIOD & BAU_MISC_CONTROL_MULT_MASK; - mmr_image = uv_read_local_mmr(UVH_AGING_PRESCALE_SEL); - index = (mmr_image >> BAU_URGENCY_7_SHIFT) & BAU_URGENCY_7_MASK; - mmr_image = uv_read_local_mmr(UVH_TRANSACTION_TIMEOUT); - mult2 = (mmr_image >> BAU_TRANS_SHIFT) & BAU_TRANS_MASK; - base = timeout_base_ns[index]; - ts_ns = base * mult1 * mult2; - ret = ts_ns / 1000; + if (is_uv1_hub()) { + mult1 = SOFTACK_TIMEOUT_PERIOD & BAU_MISC_CONTROL_MULT_MASK; + mmr_image = uv_read_local_mmr(UVH_AGING_PRESCALE_SEL); + index = (mmr_image >> BAU_URGENCY_7_SHIFT) & BAU_URGENCY_7_MASK; + mmr_image = uv_read_local_mmr(UVH_TRANSACTION_TIMEOUT); + mult2 = (mmr_image >> BAU_TRANS_SHIFT) & BAU_TRANS_MASK; + base = timeout_base_ns[index]; + ts_ns = base * mult1 * mult2; + ret = ts_ns / 1000; + } else { + /* 4 bits 0/1 for 10/80us, 3 bits of multiplier */ + mmr_image = uv_read_local_mmr(UVH_AGING_PRESCALE_SEL); + mmr_image = (mmr_image & UV_SA_MASK) >> UV_SA_SHFT; + if (mmr_image & (1L << UV2_ACK_UNITS_SHFT)) + mult1 = 80; + else + mult1 = 10; + base = mmr_image & UV2_ACK_MASK; + ret = mult1 * base; + } return ret; } +static void __init init_per_cpu_tunables(void) +{ + int cpu; + struct bau_control *bcp; + + for_each_present_cpu(cpu) { + bcp = &per_cpu(bau_control, cpu); + bcp->baudisabled = 0; + bcp->statp = &per_cpu(ptcstats, cpu); + /* time interval to catch a hardware stay-busy bug */ + bcp->timeout_interval = usec_2_cycles(2*timeout_us); + bcp->max_concurr = max_concurr; + bcp->max_concurr_const = max_concurr; + bcp->plugged_delay = plugged_delay; + bcp->plugsb4reset = plugsb4reset; + bcp->timeoutsb4reset = timeoutsb4reset; + bcp->ipi_reset_limit = ipi_reset_limit; + bcp->complete_threshold = complete_threshold; + bcp->cong_response_us = congested_respns_us; + bcp->cong_reps = congested_reps; + bcp->cong_period = congested_period; + } +} + /* - * initialize the bau_control structure for each cpu + * Scan all cpus to collect blade and socket summaries. */ -static int __init uv_init_per_cpu(int nuvhubs, int base_part_pnode) +static int __init get_cpu_topology(int base_pnode, + struct uvhub_desc *uvhub_descs, + unsigned char *uvhub_mask) { - int i; int cpu; - int tcpu; int pnode; int uvhub; - int have_hmaster; - short socket = 0; - unsigned short socket_mask; - unsigned char *uvhub_mask; + int socket; struct bau_control *bcp; struct uvhub_desc *bdp; struct socket_desc *sdp; - struct bau_control *hmaster = NULL; - struct bau_control *smaster = NULL; - struct socket_desc { - short num_cpus; - short cpu_number[MAX_CPUS_PER_SOCKET]; - }; - struct uvhub_desc { - unsigned short socket_mask; - short num_cpus; - short uvhub; - short pnode; - struct socket_desc socket[2]; - }; - struct uvhub_desc *uvhub_descs; - - timeout_us = calculate_destination_timeout(); - uvhub_descs = kmalloc(nuvhubs * sizeof(struct uvhub_desc), GFP_KERNEL); - memset(uvhub_descs, 0, nuvhubs * sizeof(struct uvhub_desc)); - uvhub_mask = kzalloc((nuvhubs+7)/8, GFP_KERNEL); for_each_present_cpu(cpu) { bcp = &per_cpu(bau_control, cpu); + memset(bcp, 0, sizeof(struct bau_control)); + pnode = uv_cpu_hub_info(cpu)->pnode; - if ((pnode - base_part_pnode) >= UV_DISTRIBUTION_SIZE) { + if ((pnode - base_pnode) >= UV_DISTRIBUTION_SIZE) { printk(KERN_EMERG "cpu %d pnode %d-%d beyond %d; BAU disabled\n", - cpu, pnode, base_part_pnode, - UV_DISTRIBUTION_SIZE); + cpu, pnode, base_pnode, UV_DISTRIBUTION_SIZE); return 1; } + bcp->osnode = cpu_to_node(cpu); - bcp->partition_base_pnode = uv_partition_base_pnode; + bcp->partition_base_pnode = base_pnode; + uvhub = uv_cpu_hub_info(cpu)->numa_blade_id; *(uvhub_mask + (uvhub/8)) |= (1 << (uvhub%8)); bdp = &uvhub_descs[uvhub]; + bdp->num_cpus++; bdp->uvhub = uvhub; bdp->pnode = pnode; + /* kludge: 'assuming' one node per socket, and assuming that disabling a socket just leaves a gap in node numbers */ socket = bcp->osnode & 1; @@ -1561,84 +1662,129 @@ static int __init uv_init_per_cpu(int nuvhubs, int base_part_pnode) sdp->cpu_number[sdp->num_cpus] = cpu; sdp->num_cpus++; if (sdp->num_cpus > MAX_CPUS_PER_SOCKET) { - printk(KERN_EMERG "%d cpus per socket invalid\n", sdp->num_cpus); + printk(KERN_EMERG "%d cpus per socket invalid\n", + sdp->num_cpus); return 1; } } + return 0; +} + +/* + * Each socket is to get a local array of pnodes/hubs. + */ +static void make_per_cpu_thp(struct bau_control *smaster) +{ + int cpu; + size_t hpsz = sizeof(struct hub_and_pnode) * num_possible_cpus(); + + smaster->thp = kmalloc_node(hpsz, GFP_KERNEL, smaster->osnode); + memset(smaster->thp, 0, hpsz); + for_each_present_cpu(cpu) { + smaster->thp[cpu].pnode = uv_cpu_hub_info(cpu)->pnode; + smaster->thp[cpu].uvhub = uv_cpu_hub_info(cpu)->numa_blade_id; + } +} + +/* + * Initialize all the per_cpu information for the cpu's on a given socket, + * given what has been gathered into the socket_desc struct. + * And reports the chosen hub and socket masters back to the caller. + */ +static int scan_sock(struct socket_desc *sdp, struct uvhub_desc *bdp, + struct bau_control **smasterp, + struct bau_control **hmasterp) +{ + int i; + int cpu; + struct bau_control *bcp; + + for (i = 0; i < sdp->num_cpus; i++) { + cpu = sdp->cpu_number[i]; + bcp = &per_cpu(bau_control, cpu); + bcp->cpu = cpu; + if (i == 0) { + *smasterp = bcp; + if (!(*hmasterp)) + *hmasterp = bcp; + } + bcp->cpus_in_uvhub = bdp->num_cpus; + bcp->cpus_in_socket = sdp->num_cpus; + bcp->socket_master = *smasterp; + bcp->uvhub = bdp->uvhub; + bcp->uvhub_master = *hmasterp; + bcp->uvhub_cpu = uv_cpu_hub_info(cpu)->blade_processor_id; + if (bcp->uvhub_cpu >= MAX_CPUS_PER_UVHUB) { + printk(KERN_EMERG "%d cpus per uvhub invalid\n", + bcp->uvhub_cpu); + return 1; + } + } + return 0; +} + +/* + * Summarize the blade and socket topology into the per_cpu structures. + */ +static int __init summarize_uvhub_sockets(int nuvhubs, + struct uvhub_desc *uvhub_descs, + unsigned char *uvhub_mask) +{ + int socket; + int uvhub; + unsigned short socket_mask; + for (uvhub = 0; uvhub < nuvhubs; uvhub++) { + struct uvhub_desc *bdp; + struct bau_control *smaster = NULL; + struct bau_control *hmaster = NULL; + if (!(*(uvhub_mask + (uvhub/8)) & (1 << (uvhub%8)))) continue; - have_hmaster = 0; + bdp = &uvhub_descs[uvhub]; socket_mask = bdp->socket_mask; socket = 0; while (socket_mask) { - if (!(socket_mask & 1)) - goto nextsocket; - sdp = &bdp->socket[socket]; - for (i = 0; i < sdp->num_cpus; i++) { - cpu = sdp->cpu_number[i]; - bcp = &per_cpu(bau_control, cpu); - bcp->cpu = cpu; - if (i == 0) { - smaster = bcp; - if (!have_hmaster) { - have_hmaster++; - hmaster = bcp; - } - } - bcp->cpus_in_uvhub = bdp->num_cpus; - bcp->cpus_in_socket = sdp->num_cpus; - bcp->socket_master = smaster; - bcp->uvhub = bdp->uvhub; - bcp->uvhub_master = hmaster; - bcp->uvhub_cpu = uv_cpu_hub_info(cpu)-> - blade_processor_id; - if (bcp->uvhub_cpu >= MAX_CPUS_PER_UVHUB) { - printk(KERN_EMERG - "%d cpus per uvhub invalid\n", - bcp->uvhub_cpu); + struct socket_desc *sdp; + if ((socket_mask & 1)) { + sdp = &bdp->socket[socket]; + if (scan_sock(sdp, bdp, &smaster, &hmaster)) return 1; - } } -nextsocket: socket++; socket_mask = (socket_mask >> 1); - /* each socket gets a local array of pnodes/hubs */ - bcp = smaster; - bcp->target_hub_and_pnode = kmalloc_node( - sizeof(struct hub_and_pnode) * - num_possible_cpus(), GFP_KERNEL, bcp->osnode); - memset(bcp->target_hub_and_pnode, 0, - sizeof(struct hub_and_pnode) * - num_possible_cpus()); - for_each_present_cpu(tcpu) { - bcp->target_hub_and_pnode[tcpu].pnode = - uv_cpu_hub_info(tcpu)->pnode; - bcp->target_hub_and_pnode[tcpu].uvhub = - uv_cpu_hub_info(tcpu)->numa_blade_id; - } + make_per_cpu_thp(smaster); } } + return 0; +} + +/* + * initialize the bau_control structure for each cpu + */ +static int __init init_per_cpu(int nuvhubs, int base_part_pnode) +{ + unsigned char *uvhub_mask; + void *vp; + struct uvhub_desc *uvhub_descs; + + timeout_us = calculate_destination_timeout(); + + vp = kmalloc(nuvhubs * sizeof(struct uvhub_desc), GFP_KERNEL); + uvhub_descs = (struct uvhub_desc *)vp; + memset(uvhub_descs, 0, nuvhubs * sizeof(struct uvhub_desc)); + uvhub_mask = kzalloc((nuvhubs+7)/8, GFP_KERNEL); + + if (get_cpu_topology(base_part_pnode, uvhub_descs, uvhub_mask)) + return 1; + + if (summarize_uvhub_sockets(nuvhubs, uvhub_descs, uvhub_mask)) + return 1; + kfree(uvhub_descs); kfree(uvhub_mask); - for_each_present_cpu(cpu) { - bcp = &per_cpu(bau_control, cpu); - bcp->baudisabled = 0; - bcp->statp = &per_cpu(ptcstats, cpu); - /* time interval to catch a hardware stay-busy bug */ - bcp->timeout_interval = microsec_2_cycles(2*timeout_us); - bcp->max_bau_concurrent = max_bau_concurrent; - bcp->max_bau_concurrent_constant = max_bau_concurrent; - bcp->plugged_delay = plugged_delay; - bcp->plugsb4reset = plugsb4reset; - bcp->timeoutsb4reset = timeoutsb4reset; - bcp->ipi_reset_limit = ipi_reset_limit; - bcp->complete_threshold = complete_threshold; - bcp->congested_response_us = congested_response_us; - bcp->congested_reps = congested_reps; - bcp->congested_period = congested_period; - } + init_per_cpu_tunables(); return 0; } @@ -1651,8 +1797,9 @@ static int __init uv_bau_init(void) int pnode; int nuvhubs; int cur_cpu; + int cpus; int vector; - unsigned long mmr; + cpumask_var_t *mask; if (!is_uv_system()) return 0; @@ -1660,24 +1807,25 @@ static int __init uv_bau_init(void) if (nobau) return 0; - for_each_possible_cpu(cur_cpu) - zalloc_cpumask_var_node(&per_cpu(uv_flush_tlb_mask, cur_cpu), - GFP_KERNEL, cpu_to_node(cur_cpu)); + for_each_possible_cpu(cur_cpu) { + mask = &per_cpu(uv_flush_tlb_mask, cur_cpu); + zalloc_cpumask_var_node(mask, GFP_KERNEL, cpu_to_node(cur_cpu)); + } uv_nshift = uv_hub_info->m_val; uv_mmask = (1UL << uv_hub_info->m_val) - 1; nuvhubs = uv_num_possible_blades(); spin_lock_init(&disable_lock); - congested_cycles = microsec_2_cycles(congested_response_us); + congested_cycles = usec_2_cycles(congested_respns_us); - uv_partition_base_pnode = 0x7fffffff; + uv_base_pnode = 0x7fffffff; for (uvhub = 0; uvhub < nuvhubs; uvhub++) { - if (uv_blade_nr_possible_cpus(uvhub) && - (uv_blade_to_pnode(uvhub) < uv_partition_base_pnode)) - uv_partition_base_pnode = uv_blade_to_pnode(uvhub); + cpus = uv_blade_nr_possible_cpus(uvhub); + if (cpus && (uv_blade_to_pnode(uvhub) < uv_base_pnode)) + uv_base_pnode = uv_blade_to_pnode(uvhub); } - if (uv_init_per_cpu(nuvhubs, uv_partition_base_pnode)) { + if (init_per_cpu(nuvhubs, uv_base_pnode)) { nobau = 1; return 0; } @@ -1685,21 +1833,21 @@ static int __init uv_bau_init(void) vector = UV_BAU_MESSAGE; for_each_possible_blade(uvhub) if (uv_blade_nr_possible_cpus(uvhub)) - uv_init_uvhub(uvhub, vector, uv_partition_base_pnode); + init_uvhub(uvhub, vector, uv_base_pnode); - uv_enable_timeouts(); + enable_timeouts(); alloc_intr_gate(vector, uv_bau_message_intr1); for_each_possible_blade(uvhub) { if (uv_blade_nr_possible_cpus(uvhub)) { + unsigned long val; + unsigned long mmr; pnode = uv_blade_to_pnode(uvhub); /* INIT the bau */ - uv_write_global_mmr64(pnode, - UVH_LB_BAU_SB_ACTIVATION_CONTROL, - ((unsigned long)1 << 63)); + val = 1L << 63; + write_gmmr_activation(pnode, val); mmr = 1; /* should be 1 to broadcast to both sockets */ - uv_write_global_mmr64(pnode, UVH_BAU_DATA_BROADCAST, - mmr); + write_mmr_data_broadcast(pnode, mmr); } } diff --git a/arch/x86/platform/uv/uv_time.c b/arch/x86/platform/uv/uv_time.c index 0eb90184515..9f29a01ee1b 100644 --- a/arch/x86/platform/uv/uv_time.c +++ b/arch/x86/platform/uv/uv_time.c @@ -99,8 +99,12 @@ static void uv_rtc_send_IPI(int cpu) /* Check for an RTC interrupt pending */ static int uv_intr_pending(int pnode) { - return uv_read_global_mmr64(pnode, UVH_EVENT_OCCURRED0) & - UVH_EVENT_OCCURRED0_RTC1_MASK; + if (is_uv1_hub()) + return uv_read_global_mmr64(pnode, UVH_EVENT_OCCURRED0) & + UV1H_EVENT_OCCURRED0_RTC1_MASK; + else + return uv_read_global_mmr64(pnode, UV2H_EVENT_OCCURRED2) & + UV2H_EVENT_OCCURRED2_RTC_1_MASK; } /* Setup interrupt and return non-zero if early expiration occurred. */ @@ -114,8 +118,12 @@ static int uv_setup_intr(int cpu, u64 expires) UVH_RTC1_INT_CONFIG_M_MASK); uv_write_global_mmr64(pnode, UVH_INT_CMPB, -1L); - uv_write_global_mmr64(pnode, UVH_EVENT_OCCURRED0_ALIAS, - UVH_EVENT_OCCURRED0_RTC1_MASK); + if (is_uv1_hub()) + uv_write_global_mmr64(pnode, UVH_EVENT_OCCURRED0_ALIAS, + UV1H_EVENT_OCCURRED0_RTC1_MASK); + else + uv_write_global_mmr64(pnode, UV2H_EVENT_OCCURRED2_ALIAS, + UV2H_EVENT_OCCURRED2_RTC_1_MASK); val = (X86_PLATFORM_IPI_VECTOR << UVH_RTC1_INT_CONFIG_VECTOR_SHFT) | ((u64)apicid << UVH_RTC1_INT_CONFIG_APIC_ID_SHFT); diff --git a/arch/x86/vdso/Makefile b/arch/x86/vdso/Makefile index b6552b189bc..bef0bc96240 100644 --- a/arch/x86/vdso/Makefile +++ b/arch/x86/vdso/Makefile @@ -11,7 +11,7 @@ vdso-install-$(VDSO32-y) += $(vdso32-images) # files to link into the vdso -vobjs-y := vdso-note.o vclock_gettime.o vgetcpu.o vvar.o +vobjs-y := vdso-note.o vclock_gettime.o vgetcpu.o # files to link into kernel obj-$(VDSO64-y) += vma.o vdso.o @@ -37,11 +37,24 @@ $(obj)/%.so: OBJCOPYFLAGS := -S $(obj)/%.so: $(obj)/%.so.dbg FORCE $(call if_changed,objcopy) +# +# Don't omit frame pointers for ease of userspace debugging, but do +# optimize sibling calls. +# CFL := $(PROFILING) -mcmodel=small -fPIC -O2 -fasynchronous-unwind-tables -m64 \ - $(filter -g%,$(KBUILD_CFLAGS)) $(call cc-option, -fno-stack-protector) + $(filter -g%,$(KBUILD_CFLAGS)) $(call cc-option, -fno-stack-protector) \ + -fno-omit-frame-pointer -foptimize-sibling-calls $(vobjs): KBUILD_CFLAGS += $(CFL) +# +# vDSO code runs in userspace and -pg doesn't help with profiling anyway. +# +CFLAGS_REMOVE_vdso-note.o = -pg +CFLAGS_REMOVE_vclock_gettime.o = -pg +CFLAGS_REMOVE_vgetcpu.o = -pg +CFLAGS_REMOVE_vvar.o = -pg + targets += vdso-syms.lds obj-$(VDSO64-y) += vdso-syms.lds diff --git a/arch/x86/vdso/vclock_gettime.c b/arch/x86/vdso/vclock_gettime.c index ee55754cc3c..a724905fdae 100644 --- a/arch/x86/vdso/vclock_gettime.c +++ b/arch/x86/vdso/vclock_gettime.c @@ -2,7 +2,7 @@ * Copyright 2006 Andi Kleen, SUSE Labs. * Subject to the GNU Public License, v.2 * - * Fast user context implementation of clock_gettime and gettimeofday. + * Fast user context implementation of clock_gettime, gettimeofday, and time. * * The code should have no internal unresolved relocations. * Check with readelf after changing. @@ -22,9 +22,8 @@ #include <asm/hpet.h> #include <asm/unistd.h> #include <asm/io.h> -#include "vextern.h" -#define gtod vdso_vsyscall_gtod_data +#define gtod (&VVAR(vsyscall_gtod_data)) notrace static long vdso_fallback_gettime(long clock, struct timespec *ts) { @@ -56,22 +55,6 @@ notrace static noinline int do_realtime(struct timespec *ts) return 0; } -/* Copy of the version in kernel/time.c which we cannot directly access */ -notrace static void -vset_normalized_timespec(struct timespec *ts, long sec, long nsec) -{ - while (nsec >= NSEC_PER_SEC) { - nsec -= NSEC_PER_SEC; - ++sec; - } - while (nsec < 0) { - nsec += NSEC_PER_SEC; - --sec; - } - ts->tv_sec = sec; - ts->tv_nsec = nsec; -} - notrace static noinline int do_monotonic(struct timespec *ts) { unsigned long seq, ns, secs; @@ -82,7 +65,17 @@ notrace static noinline int do_monotonic(struct timespec *ts) secs += gtod->wall_to_monotonic.tv_sec; ns += gtod->wall_to_monotonic.tv_nsec; } while (unlikely(read_seqretry(>od->lock, seq))); - vset_normalized_timespec(ts, secs, ns); + + /* wall_time_nsec, vgetns(), and wall_to_monotonic.tv_nsec + * are all guaranteed to be nonnegative. + */ + while (ns >= NSEC_PER_SEC) { + ns -= NSEC_PER_SEC; + ++secs; + } + ts->tv_sec = secs; + ts->tv_nsec = ns; + return 0; } @@ -107,7 +100,17 @@ notrace static noinline int do_monotonic_coarse(struct timespec *ts) secs += gtod->wall_to_monotonic.tv_sec; ns += gtod->wall_to_monotonic.tv_nsec; } while (unlikely(read_seqretry(>od->lock, seq))); - vset_normalized_timespec(ts, secs, ns); + + /* wall_time_nsec and wall_to_monotonic.tv_nsec are + * guaranteed to be between 0 and NSEC_PER_SEC. + */ + if (ns >= NSEC_PER_SEC) { + ns -= NSEC_PER_SEC; + ++secs; + } + ts->tv_sec = secs; + ts->tv_nsec = ns; + return 0; } @@ -157,3 +160,32 @@ notrace int __vdso_gettimeofday(struct timeval *tv, struct timezone *tz) } int gettimeofday(struct timeval *, struct timezone *) __attribute__((weak, alias("__vdso_gettimeofday"))); + +/* This will break when the xtime seconds get inaccurate, but that is + * unlikely */ + +static __always_inline long time_syscall(long *t) +{ + long secs; + asm volatile("syscall" + : "=a" (secs) + : "0" (__NR_time), "D" (t) : "cc", "r11", "cx", "memory"); + return secs; +} + +notrace time_t __vdso_time(time_t *t) +{ + time_t result; + + if (unlikely(!VVAR(vsyscall_gtod_data).sysctl_enabled)) + return time_syscall(t); + + /* This is atomic on x86_64 so we don't need any locks. */ + result = ACCESS_ONCE(VVAR(vsyscall_gtod_data).wall_time_sec); + + if (t) + *t = result; + return result; +} +int time(time_t *t) + __attribute__((weak, alias("__vdso_time"))); diff --git a/arch/x86/vdso/vdso.lds.S b/arch/x86/vdso/vdso.lds.S index 4e5dd3b4de7..b96b2677cad 100644 --- a/arch/x86/vdso/vdso.lds.S +++ b/arch/x86/vdso/vdso.lds.S @@ -23,15 +23,10 @@ VERSION { __vdso_gettimeofday; getcpu; __vdso_getcpu; + time; + __vdso_time; local: *; }; } VDSO64_PRELINK = VDSO_PRELINK; - -/* - * Define VDSO64_x for each VEXTERN(x), for use via VDSO64_SYMBOL. - */ -#define VEXTERN(x) VDSO64_ ## x = vdso_ ## x; -#include "vextern.h" -#undef VEXTERN diff --git a/arch/x86/vdso/vextern.h b/arch/x86/vdso/vextern.h deleted file mode 100644 index 1683ba2ae3e..00000000000 --- a/arch/x86/vdso/vextern.h +++ /dev/null @@ -1,16 +0,0 @@ -#ifndef VEXTERN -#include <asm/vsyscall.h> -#define VEXTERN(x) \ - extern typeof(x) *vdso_ ## x __attribute__((visibility("hidden"))); -#endif - -#define VMAGIC 0xfeedbabeabcdefabUL - -/* Any kernel variables used in the vDSO must be exported in the main - kernel's vmlinux.lds.S/vsyscall.h/proper __section and - put into vextern.h and be referenced as a pointer with vdso prefix. - The main kernel later fills in the values. */ - -VEXTERN(jiffies) -VEXTERN(vgetcpu_mode) -VEXTERN(vsyscall_gtod_data) diff --git a/arch/x86/vdso/vgetcpu.c b/arch/x86/vdso/vgetcpu.c index 9fbc6b20026..5463ad55857 100644 --- a/arch/x86/vdso/vgetcpu.c +++ b/arch/x86/vdso/vgetcpu.c @@ -11,14 +11,13 @@ #include <linux/time.h> #include <asm/vsyscall.h> #include <asm/vgtod.h> -#include "vextern.h" notrace long __vdso_getcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *unused) { unsigned int p; - if (*vdso_vgetcpu_mode == VGETCPU_RDTSCP) { + if (VVAR(vgetcpu_mode) == VGETCPU_RDTSCP) { /* Load per CPU data from RDTSCP */ native_read_tscp(&p); } else { diff --git a/arch/x86/vdso/vma.c b/arch/x86/vdso/vma.c index 4b5d26f108b..7abd2be0f9b 100644 --- a/arch/x86/vdso/vma.c +++ b/arch/x86/vdso/vma.c @@ -15,9 +15,6 @@ #include <asm/proto.h> #include <asm/vdso.h> -#include "vextern.h" /* Just for VMAGIC. */ -#undef VEXTERN - unsigned int __read_mostly vdso_enabled = 1; extern char vdso_start[], vdso_end[]; @@ -26,20 +23,10 @@ extern unsigned short vdso_sync_cpuid; static struct page **vdso_pages; static unsigned vdso_size; -static inline void *var_ref(void *p, char *name) -{ - if (*(void **)p != (void *)VMAGIC) { - printk("VDSO: variable %s broken\n", name); - vdso_enabled = 0; - } - return p; -} - static int __init init_vdso_vars(void) { int npages = (vdso_end - vdso_start + PAGE_SIZE - 1) / PAGE_SIZE; int i; - char *vbase; vdso_size = npages << PAGE_SHIFT; vdso_pages = kmalloc(sizeof(struct page *) * npages, GFP_KERNEL); @@ -54,20 +41,6 @@ static int __init init_vdso_vars(void) copy_page(page_address(p), vdso_start + i*PAGE_SIZE); } - vbase = vmap(vdso_pages, npages, 0, PAGE_KERNEL); - if (!vbase) - goto oom; - - if (memcmp(vbase, "\177ELF", 4)) { - printk("VDSO: I'm broken; not ELF\n"); - vdso_enabled = 0; - } - -#define VEXTERN(x) \ - *(typeof(__ ## x) **) var_ref(VDSO64_SYMBOL(vbase, x), #x) = &__ ## x; -#include "vextern.h" -#undef VEXTERN - vunmap(vbase); return 0; oom: diff --git a/arch/x86/vdso/vvar.c b/arch/x86/vdso/vvar.c deleted file mode 100644 index 1b7e703684f..00000000000 --- a/arch/x86/vdso/vvar.c +++ /dev/null @@ -1,12 +0,0 @@ -/* Define pointer to external vDSO variables. - These are part of the vDSO. The kernel fills in the real addresses - at boot time. This is done because when the vdso is linked the - kernel isn't yet and we don't know the final addresses. */ -#include <linux/kernel.h> -#include <linux/time.h> -#include <asm/vsyscall.h> -#include <asm/timex.h> -#include <asm/vgtod.h> - -#define VEXTERN(x) typeof (__ ## x) *const vdso_ ## x = (void *)VMAGIC; -#include "vextern.h" diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index 02d75246037..dc708dcc62f 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -75,67 +75,12 @@ #include "mmu.h" #include "debugfs.h" -#define MMU_UPDATE_HISTO 30 - /* * Protects atomic reservation decrease/increase against concurrent increases. * Also protects non-atomic updates of current_pages and balloon lists. */ DEFINE_SPINLOCK(xen_reservation_lock); -#ifdef CONFIG_XEN_DEBUG_FS - -static struct { - u32 pgd_update; - u32 pgd_update_pinned; - u32 pgd_update_batched; - - u32 pud_update; - u32 pud_update_pinned; - u32 pud_update_batched; - - u32 pmd_update; - u32 pmd_update_pinned; - u32 pmd_update_batched; - - u32 pte_update; - u32 pte_update_pinned; - u32 pte_update_batched; - - u32 mmu_update; - u32 mmu_update_extended; - u32 mmu_update_histo[MMU_UPDATE_HISTO]; - - u32 prot_commit; - u32 prot_commit_batched; - - u32 set_pte_at; - u32 set_pte_at_batched; - u32 set_pte_at_pinned; - u32 set_pte_at_current; - u32 set_pte_at_kernel; -} mmu_stats; - -static u8 zero_stats; - -static inline void check_zero(void) -{ - if (unlikely(zero_stats)) { - memset(&mmu_stats, 0, sizeof(mmu_stats)); - zero_stats = 0; - } -} - -#define ADD_STATS(elem, val) \ - do { check_zero(); mmu_stats.elem += (val); } while(0) - -#else /* !CONFIG_XEN_DEBUG_FS */ - -#define ADD_STATS(elem, val) do { (void)(val); } while(0) - -#endif /* CONFIG_XEN_DEBUG_FS */ - - /* * Identity map, in addition to plain kernel map. This needs to be * large enough to allocate page table pages to allocate the rest. @@ -243,11 +188,6 @@ static bool xen_page_pinned(void *ptr) return PagePinned(page); } -static bool xen_iomap_pte(pte_t pte) -{ - return pte_flags(pte) & _PAGE_IOMAP; -} - void xen_set_domain_pte(pte_t *ptep, pte_t pteval, unsigned domid) { struct multicall_space mcs; @@ -257,7 +197,7 @@ void xen_set_domain_pte(pte_t *ptep, pte_t pteval, unsigned domid) u = mcs.args; /* ptep might be kmapped when using 32-bit HIGHPTE */ - u->ptr = arbitrary_virt_to_machine(ptep).maddr; + u->ptr = virt_to_machine(ptep).maddr; u->val = pte_val_ma(pteval); MULTI_mmu_update(mcs.mc, mcs.args, 1, NULL, domid); @@ -266,11 +206,6 @@ void xen_set_domain_pte(pte_t *ptep, pte_t pteval, unsigned domid) } EXPORT_SYMBOL_GPL(xen_set_domain_pte); -static void xen_set_iomap_pte(pte_t *ptep, pte_t pteval) -{ - xen_set_domain_pte(ptep, pteval, DOMID_IO); -} - static void xen_extend_mmu_update(const struct mmu_update *update) { struct multicall_space mcs; @@ -279,27 +214,17 @@ static void xen_extend_mmu_update(const struct mmu_update *update) mcs = xen_mc_extend_args(__HYPERVISOR_mmu_update, sizeof(*u)); if (mcs.mc != NULL) { - ADD_STATS(mmu_update_extended, 1); - ADD_STATS(mmu_update_histo[mcs.mc->args[1]], -1); - mcs.mc->args[1]++; - - if (mcs.mc->args[1] < MMU_UPDATE_HISTO) - ADD_STATS(mmu_update_histo[mcs.mc->args[1]], 1); - else - ADD_STATS(mmu_update_histo[0], 1); } else { - ADD_STATS(mmu_update, 1); mcs = __xen_mc_entry(sizeof(*u)); MULTI_mmu_update(mcs.mc, mcs.args, 1, NULL, DOMID_SELF); - ADD_STATS(mmu_update_histo[1], 1); } u = mcs.args; *u = *update; } -void xen_set_pmd_hyper(pmd_t *ptr, pmd_t val) +static void xen_set_pmd_hyper(pmd_t *ptr, pmd_t val) { struct mmu_update u; @@ -312,17 +237,13 @@ void xen_set_pmd_hyper(pmd_t *ptr, pmd_t val) u.val = pmd_val_ma(val); xen_extend_mmu_update(&u); - ADD_STATS(pmd_update_batched, paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU); - xen_mc_issue(PARAVIRT_LAZY_MMU); preempt_enable(); } -void xen_set_pmd(pmd_t *ptr, pmd_t val) +static void xen_set_pmd(pmd_t *ptr, pmd_t val) { - ADD_STATS(pmd_update, 1); - /* If page is not pinned, we can just update the entry directly */ if (!xen_page_pinned(ptr)) { @@ -330,8 +251,6 @@ void xen_set_pmd(pmd_t *ptr, pmd_t val) return; } - ADD_STATS(pmd_update_pinned, 1); - xen_set_pmd_hyper(ptr, val); } @@ -344,35 +263,34 @@ void set_pte_mfn(unsigned long vaddr, unsigned long mfn, pgprot_t flags) set_pte_vaddr(vaddr, mfn_pte(mfn, flags)); } -void xen_set_pte_at(struct mm_struct *mm, unsigned long addr, - pte_t *ptep, pte_t pteval) +static bool xen_batched_set_pte(pte_t *ptep, pte_t pteval) { - if (xen_iomap_pte(pteval)) { - xen_set_iomap_pte(ptep, pteval); - goto out; - } + struct mmu_update u; - ADD_STATS(set_pte_at, 1); -// ADD_STATS(set_pte_at_pinned, xen_page_pinned(ptep)); - ADD_STATS(set_pte_at_current, mm == current->mm); - ADD_STATS(set_pte_at_kernel, mm == &init_mm); + if (paravirt_get_lazy_mode() != PARAVIRT_LAZY_MMU) + return false; - if (mm == current->mm || mm == &init_mm) { - if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU) { - struct multicall_space mcs; - mcs = xen_mc_entry(0); + xen_mc_batch(); - MULTI_update_va_mapping(mcs.mc, addr, pteval, 0); - ADD_STATS(set_pte_at_batched, 1); - xen_mc_issue(PARAVIRT_LAZY_MMU); - goto out; - } else - if (HYPERVISOR_update_va_mapping(addr, pteval, 0) == 0) - goto out; - } - xen_set_pte(ptep, pteval); + u.ptr = virt_to_machine(ptep).maddr | MMU_NORMAL_PT_UPDATE; + u.val = pte_val_ma(pteval); + xen_extend_mmu_update(&u); + + xen_mc_issue(PARAVIRT_LAZY_MMU); -out: return; + return true; +} + +static void xen_set_pte(pte_t *ptep, pte_t pteval) +{ + if (!xen_batched_set_pte(ptep, pteval)) + native_set_pte(ptep, pteval); +} + +static void xen_set_pte_at(struct mm_struct *mm, unsigned long addr, + pte_t *ptep, pte_t pteval) +{ + xen_set_pte(ptep, pteval); } pte_t xen_ptep_modify_prot_start(struct mm_struct *mm, @@ -389,13 +307,10 @@ void xen_ptep_modify_prot_commit(struct mm_struct *mm, unsigned long addr, xen_mc_batch(); - u.ptr = arbitrary_virt_to_machine(ptep).maddr | MMU_PT_UPDATE_PRESERVE_AD; + u.ptr = virt_to_machine(ptep).maddr | MMU_PT_UPDATE_PRESERVE_AD; u.val = pte_val_ma(pte); xen_extend_mmu_update(&u); - ADD_STATS(prot_commit, 1); - ADD_STATS(prot_commit_batched, paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU); - xen_mc_issue(PARAVIRT_LAZY_MMU); } @@ -463,7 +378,7 @@ static pteval_t iomap_pte(pteval_t val) return val; } -pteval_t xen_pte_val(pte_t pte) +static pteval_t xen_pte_val(pte_t pte) { pteval_t pteval = pte.pte; @@ -480,7 +395,7 @@ pteval_t xen_pte_val(pte_t pte) } PV_CALLEE_SAVE_REGS_THUNK(xen_pte_val); -pgdval_t xen_pgd_val(pgd_t pgd) +static pgdval_t xen_pgd_val(pgd_t pgd) { return pte_mfn_to_pfn(pgd.pgd); } @@ -511,7 +426,7 @@ void xen_set_pat(u64 pat) WARN_ON(pat != 0x0007010600070106ull); } -pte_t xen_make_pte(pteval_t pte) +static pte_t xen_make_pte(pteval_t pte) { phys_addr_t addr = (pte & PTE_PFN_MASK); @@ -581,20 +496,20 @@ pte_t xen_make_pte_debug(pteval_t pte) PV_CALLEE_SAVE_REGS_THUNK(xen_make_pte_debug); #endif -pgd_t xen_make_pgd(pgdval_t pgd) +static pgd_t xen_make_pgd(pgdval_t pgd) { pgd = pte_pfn_to_mfn(pgd); return native_make_pgd(pgd); } PV_CALLEE_SAVE_REGS_THUNK(xen_make_pgd); -pmdval_t xen_pmd_val(pmd_t pmd) +static pmdval_t xen_pmd_val(pmd_t pmd) { return pte_mfn_to_pfn(pmd.pmd); } PV_CALLEE_SAVE_REGS_THUNK(xen_pmd_val); -void xen_set_pud_hyper(pud_t *ptr, pud_t val) +static void xen_set_pud_hyper(pud_t *ptr, pud_t val) { struct mmu_update u; @@ -607,17 +522,13 @@ void xen_set_pud_hyper(pud_t *ptr, pud_t val) u.val = pud_val_ma(val); xen_extend_mmu_update(&u); - ADD_STATS(pud_update_batched, paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU); - xen_mc_issue(PARAVIRT_LAZY_MMU); preempt_enable(); } -void xen_set_pud(pud_t *ptr, pud_t val) +static void xen_set_pud(pud_t *ptr, pud_t val) { - ADD_STATS(pud_update, 1); - /* If page is not pinned, we can just update the entry directly */ if (!xen_page_pinned(ptr)) { @@ -625,56 +536,28 @@ void xen_set_pud(pud_t *ptr, pud_t val) return; } - ADD_STATS(pud_update_pinned, 1); - xen_set_pud_hyper(ptr, val); } -void xen_set_pte(pte_t *ptep, pte_t pte) -{ - if (xen_iomap_pte(pte)) { - xen_set_iomap_pte(ptep, pte); - return; - } - - ADD_STATS(pte_update, 1); -// ADD_STATS(pte_update_pinned, xen_page_pinned(ptep)); - ADD_STATS(pte_update_batched, paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU); - #ifdef CONFIG_X86_PAE - ptep->pte_high = pte.pte_high; - smp_wmb(); - ptep->pte_low = pte.pte_low; -#else - *ptep = pte; -#endif -} - -#ifdef CONFIG_X86_PAE -void xen_set_pte_atomic(pte_t *ptep, pte_t pte) +static void xen_set_pte_atomic(pte_t *ptep, pte_t pte) { - if (xen_iomap_pte(pte)) { - xen_set_iomap_pte(ptep, pte); - return; - } - set_64bit((u64 *)ptep, native_pte_val(pte)); } -void xen_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep) +static void xen_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep) { - ptep->pte_low = 0; - smp_wmb(); /* make sure low gets written first */ - ptep->pte_high = 0; + if (!xen_batched_set_pte(ptep, native_make_pte(0))) + native_pte_clear(mm, addr, ptep); } -void xen_pmd_clear(pmd_t *pmdp) +static void xen_pmd_clear(pmd_t *pmdp) { set_pmd(pmdp, __pmd(0)); } #endif /* CONFIG_X86_PAE */ -pmd_t xen_make_pmd(pmdval_t pmd) +static pmd_t xen_make_pmd(pmdval_t pmd) { pmd = pte_pfn_to_mfn(pmd); return native_make_pmd(pmd); @@ -682,13 +565,13 @@ pmd_t xen_make_pmd(pmdval_t pmd) PV_CALLEE_SAVE_REGS_THUNK(xen_make_pmd); #if PAGETABLE_LEVELS == 4 -pudval_t xen_pud_val(pud_t pud) +static pudval_t xen_pud_val(pud_t pud) { return pte_mfn_to_pfn(pud.pud); } PV_CALLEE_SAVE_REGS_THUNK(xen_pud_val); -pud_t xen_make_pud(pudval_t pud) +static pud_t xen_make_pud(pudval_t pud) { pud = pte_pfn_to_mfn(pud); @@ -696,7 +579,7 @@ pud_t xen_make_pud(pudval_t pud) } PV_CALLEE_SAVE_REGS_THUNK(xen_make_pud); -pgd_t *xen_get_user_pgd(pgd_t *pgd) +static pgd_t *xen_get_user_pgd(pgd_t *pgd) { pgd_t *pgd_page = (pgd_t *)(((unsigned long)pgd) & PAGE_MASK); unsigned offset = pgd - pgd_page; @@ -728,7 +611,7 @@ static void __xen_set_pgd_hyper(pgd_t *ptr, pgd_t val) * 2. It is always pinned * 3. It has no user pagetable attached to it */ -void __init xen_set_pgd_hyper(pgd_t *ptr, pgd_t val) +static void __init xen_set_pgd_hyper(pgd_t *ptr, pgd_t val) { preempt_disable(); @@ -741,12 +624,10 @@ void __init xen_set_pgd_hyper(pgd_t *ptr, pgd_t val) preempt_enable(); } -void xen_set_pgd(pgd_t *ptr, pgd_t val) +static void xen_set_pgd(pgd_t *ptr, pgd_t val) { pgd_t *user_ptr = xen_get_user_pgd(ptr); - ADD_STATS(pgd_update, 1); - /* If page is not pinned, we can just update the entry directly */ if (!xen_page_pinned(ptr)) { @@ -758,9 +639,6 @@ void xen_set_pgd(pgd_t *ptr, pgd_t val) return; } - ADD_STATS(pgd_update_pinned, 1); - ADD_STATS(pgd_update_batched, paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU); - /* If it's pinned, then we can at least batch the kernel and user updates together. */ xen_mc_batch(); @@ -1162,14 +1040,14 @@ void xen_mm_unpin_all(void) spin_unlock(&pgd_lock); } -void xen_activate_mm(struct mm_struct *prev, struct mm_struct *next) +static void xen_activate_mm(struct mm_struct *prev, struct mm_struct *next) { spin_lock(&next->page_table_lock); xen_pgd_pin(next); spin_unlock(&next->page_table_lock); } -void xen_dup_mmap(struct mm_struct *oldmm, struct mm_struct *mm) +static void xen_dup_mmap(struct mm_struct *oldmm, struct mm_struct *mm) { spin_lock(&mm->page_table_lock); xen_pgd_pin(mm); @@ -1256,7 +1134,7 @@ static void xen_drop_mm_ref(struct mm_struct *mm) * pagetable because of lazy tlb flushing. This means we need need to * switch all CPUs off this pagetable before we can unpin it. */ -void xen_exit_mmap(struct mm_struct *mm) +static void xen_exit_mmap(struct mm_struct *mm) { get_cpu(); /* make sure we don't move around */ xen_drop_mm_ref(mm); @@ -2371,7 +2249,7 @@ static int remap_area_mfn_pte_fn(pte_t *ptep, pgtable_t token, struct remap_data *rmd = data; pte_t pte = pte_mkspecial(pfn_pte(rmd->mfn++, rmd->prot)); - rmd->mmu_update->ptr = arbitrary_virt_to_machine(ptep).maddr; + rmd->mmu_update->ptr = virt_to_machine(ptep).maddr; rmd->mmu_update->val = pte_val_ma(pte); rmd->mmu_update++; @@ -2425,7 +2303,6 @@ out: EXPORT_SYMBOL_GPL(xen_remap_domain_mfn_range); #ifdef CONFIG_XEN_DEBUG_FS - static int p2m_dump_open(struct inode *inode, struct file *filp) { return single_open(filp, p2m_dump_show, NULL); @@ -2437,65 +2314,4 @@ static const struct file_operations p2m_dump_fops = { .llseek = seq_lseek, .release = single_release, }; - -static struct dentry *d_mmu_debug; - -static int __init xen_mmu_debugfs(void) -{ - struct dentry *d_xen = xen_init_debugfs(); - - if (d_xen == NULL) - return -ENOMEM; - - d_mmu_debug = debugfs_create_dir("mmu", d_xen); - - debugfs_create_u8("zero_stats", 0644, d_mmu_debug, &zero_stats); - - debugfs_create_u32("pgd_update", 0444, d_mmu_debug, &mmu_stats.pgd_update); - debugfs_create_u32("pgd_update_pinned", 0444, d_mmu_debug, - &mmu_stats.pgd_update_pinned); - debugfs_create_u32("pgd_update_batched", 0444, d_mmu_debug, - &mmu_stats.pgd_update_pinned); - - debugfs_create_u32("pud_update", 0444, d_mmu_debug, &mmu_stats.pud_update); - debugfs_create_u32("pud_update_pinned", 0444, d_mmu_debug, - &mmu_stats.pud_update_pinned); - debugfs_create_u32("pud_update_batched", 0444, d_mmu_debug, - &mmu_stats.pud_update_pinned); - - debugfs_create_u32("pmd_update", 0444, d_mmu_debug, &mmu_stats.pmd_update); - debugfs_create_u32("pmd_update_pinned", 0444, d_mmu_debug, - &mmu_stats.pmd_update_pinned); - debugfs_create_u32("pmd_update_batched", 0444, d_mmu_debug, - &mmu_stats.pmd_update_pinned); - - debugfs_create_u32("pte_update", 0444, d_mmu_debug, &mmu_stats.pte_update); -// debugfs_create_u32("pte_update_pinned", 0444, d_mmu_debug, -// &mmu_stats.pte_update_pinned); - debugfs_create_u32("pte_update_batched", 0444, d_mmu_debug, - &mmu_stats.pte_update_pinned); - - debugfs_create_u32("mmu_update", 0444, d_mmu_debug, &mmu_stats.mmu_update); - debugfs_create_u32("mmu_update_extended", 0444, d_mmu_debug, - &mmu_stats.mmu_update_extended); - xen_debugfs_create_u32_array("mmu_update_histo", 0444, d_mmu_debug, - mmu_stats.mmu_update_histo, 20); - - debugfs_create_u32("set_pte_at", 0444, d_mmu_debug, &mmu_stats.set_pte_at); - debugfs_create_u32("set_pte_at_batched", 0444, d_mmu_debug, - &mmu_stats.set_pte_at_batched); - debugfs_create_u32("set_pte_at_current", 0444, d_mmu_debug, - &mmu_stats.set_pte_at_current); - debugfs_create_u32("set_pte_at_kernel", 0444, d_mmu_debug, - &mmu_stats.set_pte_at_kernel); - - debugfs_create_u32("prot_commit", 0444, d_mmu_debug, &mmu_stats.prot_commit); - debugfs_create_u32("prot_commit_batched", 0444, d_mmu_debug, - &mmu_stats.prot_commit_batched); - - debugfs_create_file("p2m", 0600, d_mmu_debug, NULL, &p2m_dump_fops); - return 0; -} -fs_initcall(xen_mmu_debugfs); - -#endif /* CONFIG_XEN_DEBUG_FS */ +#endif /* CONFIG_XEN_DEBUG_FS */ diff --git a/arch/x86/xen/mmu.h b/arch/x86/xen/mmu.h index 537bb9aab77..73809bb951b 100644 --- a/arch/x86/xen/mmu.h +++ b/arch/x86/xen/mmu.h @@ -15,43 +15,6 @@ bool __set_phys_to_machine(unsigned long pfn, unsigned long mfn); void set_pte_mfn(unsigned long vaddr, unsigned long pfn, pgprot_t flags); - -void xen_activate_mm(struct mm_struct *prev, struct mm_struct *next); -void xen_dup_mmap(struct mm_struct *oldmm, struct mm_struct *mm); -void xen_exit_mmap(struct mm_struct *mm); - -pteval_t xen_pte_val(pte_t); -pmdval_t xen_pmd_val(pmd_t); -pgdval_t xen_pgd_val(pgd_t); - -pte_t xen_make_pte(pteval_t); -pmd_t xen_make_pmd(pmdval_t); -pgd_t xen_make_pgd(pgdval_t); - -void xen_set_pte(pte_t *ptep, pte_t pteval); -void xen_set_pte_at(struct mm_struct *mm, unsigned long addr, - pte_t *ptep, pte_t pteval); - -#ifdef CONFIG_X86_PAE -void xen_set_pte_atomic(pte_t *ptep, pte_t pte); -void xen_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep); -void xen_pmd_clear(pmd_t *pmdp); -#endif /* CONFIG_X86_PAE */ - -void xen_set_pmd(pmd_t *pmdp, pmd_t pmdval); -void xen_set_pud(pud_t *ptr, pud_t val); -void xen_set_pmd_hyper(pmd_t *pmdp, pmd_t pmdval); -void xen_set_pud_hyper(pud_t *ptr, pud_t val); - -#if PAGETABLE_LEVELS == 4 -pudval_t xen_pud_val(pud_t pud); -pud_t xen_make_pud(pudval_t pudval); -void xen_set_pgd(pgd_t *pgdp, pgd_t pgd); -void xen_set_pgd_hyper(pgd_t *pgdp, pgd_t pgd); -#endif - -pgd_t *xen_get_user_pgd(pgd_t *pgd); - pte_t xen_ptep_modify_prot_start(struct mm_struct *mm, unsigned long addr, pte_t *ptep); void xen_ptep_modify_prot_commit(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte); diff --git a/arch/x86/xen/multicalls.c b/arch/x86/xen/multicalls.c index 8bff7e7c290..1b2b73ff0a6 100644 --- a/arch/x86/xen/multicalls.c +++ b/arch/x86/xen/multicalls.c @@ -189,10 +189,10 @@ struct multicall_space __xen_mc_entry(size_t args) unsigned argidx = roundup(b->argidx, sizeof(u64)); BUG_ON(preemptible()); - BUG_ON(b->argidx > MC_ARGS); + BUG_ON(b->argidx >= MC_ARGS); if (b->mcidx == MC_BATCH || - (argidx + args) > MC_ARGS) { + (argidx + args) >= MC_ARGS) { mc_stats_flush(b->mcidx == MC_BATCH ? FL_SLOTS : FL_ARGS); xen_mc_flush(); argidx = roundup(b->argidx, sizeof(u64)); @@ -206,7 +206,7 @@ struct multicall_space __xen_mc_entry(size_t args) ret.args = &b->args[argidx]; b->argidx = argidx + args; - BUG_ON(b->argidx > MC_ARGS); + BUG_ON(b->argidx >= MC_ARGS); return ret; } @@ -216,7 +216,7 @@ struct multicall_space xen_mc_extend_args(unsigned long op, size_t size) struct multicall_space ret = { NULL, NULL }; BUG_ON(preemptible()); - BUG_ON(b->argidx > MC_ARGS); + BUG_ON(b->argidx >= MC_ARGS); if (b->mcidx == 0) return ret; @@ -224,14 +224,14 @@ struct multicall_space xen_mc_extend_args(unsigned long op, size_t size) if (b->entries[b->mcidx - 1].op != op) return ret; - if ((b->argidx + size) > MC_ARGS) + if ((b->argidx + size) >= MC_ARGS) return ret; ret.mc = &b->entries[b->mcidx - 1]; ret.args = &b->args[b->argidx]; b->argidx += size; - BUG_ON(b->argidx > MC_ARGS); + BUG_ON(b->argidx >= MC_ARGS); return ret; } |