diff options
Diffstat (limited to 'linux-2.6-xen-sparse/arch/i386/kernel')
21 files changed, 1143 insertions, 429 deletions
diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/Makefile b/linux-2.6-xen-sparse/arch/i386/kernel/Makefile index d0ccf7b375..ddaa19d896 100644 --- a/linux-2.6-xen-sparse/arch/i386/kernel/Makefile +++ b/linux-2.6-xen-sparse/arch/i386/kernel/Makefile @@ -6,8 +6,8 @@ extra-y := head.o init_task.o vmlinux.lds obj-y := process.o semaphore.o signal.o entry.o traps.o irq.o \ ptrace.o time.o ioport.o ldt.o setup.o i8259.o sys_i386.o \ - pci-dma.o i386_ksyms.o i387.o dmi_scan.o bootflag.o \ - quirks.o i8237.o topology.o + pci-dma.o i386_ksyms.o i387.o bootflag.o \ + quirks.o i8237.o topology.o alternative.o obj-y += cpu/ obj-y += timers/ @@ -37,7 +37,6 @@ obj-$(CONFIG_EFI) += efi.o efi_stub.o obj-$(CONFIG_DOUBLEFAULT) += doublefault.o obj-$(CONFIG_VM86) += vm86.o obj-$(CONFIG_EARLY_PRINTK) += early_printk.o -obj-$(CONFIG_SMP_ALTERNATIVES) += smpalts.o EXTRA_AFLAGS := -traditional diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/acpi/boot-xen.c b/linux-2.6-xen-sparse/arch/i386/kernel/acpi/boot-xen.c index 96e452577d..08e8352ea4 100644 --- a/linux-2.6-xen-sparse/arch/i386/kernel/acpi/boot-xen.c +++ b/linux-2.6-xen-sparse/arch/i386/kernel/acpi/boot-xen.c @@ -671,10 +671,10 @@ unsigned long __init acpi_find_rsdp(void) unsigned long rsdp_phys = 0; if (efi_enabled) { - if (efi.acpi20) - return __pa(efi.acpi20); - else if (efi.acpi) - return __pa(efi.acpi); + if (efi.acpi20 != EFI_INVALID_TABLE_ADDR) + return efi.acpi20; + else if (efi.acpi != EFI_INVALID_TABLE_ADDR) + return efi.acpi; } /* * Scan memory looking for the RSDP signature. First search EBDA (low @@ -696,6 +696,9 @@ static int __init acpi_parse_madt_lapic_entries(void) { int count; + if (!cpu_has_apic) + return -ENODEV; + /* * Note that the LAPIC address is obtained from the MADT (32-bit value) * and (optionally) overriden by a LAPIC_ADDR_OVR entry (64-bit value). @@ -754,6 +757,9 @@ static int __init acpi_parse_madt_ioapic_entries(void) return -ENODEV; } + if (!cpu_has_apic) + return -ENODEV; + /* * if "noapic" boot option, don't look for IO-APICs */ diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/apic-xen.c b/linux-2.6-xen-sparse/arch/i386/kernel/apic-xen.c index 07a0994307..1f8297bc65 100644 --- a/linux-2.6-xen-sparse/arch/i386/kernel/apic-xen.c +++ b/linux-2.6-xen-sparse/arch/i386/kernel/apic-xen.c @@ -38,6 +38,7 @@ #include <asm/i8253.h> #include <mach_apic.h> +#include <mach_apicdef.h> #include <mach_ipi.h> #include "io_ports.h" @@ -60,6 +61,18 @@ int enable_local_apic __initdata = 0; /* -1=force-disable, +1=force-enable */ */ int apic_verbosity; +int modern_apic(void) +{ + unsigned int lvr, version; + /* AMD systems use old APIC versions, so check the CPU */ + if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD && + boot_cpu_data.x86 >= 0xf) + return 1; + lvr = apic_read(APIC_LVR); + version = GET_APIC_VERSION(lvr); + return version >= 0x14; +} + /* * 'what should we do if we get a hw irq event on an illegal vector'. * each architecture has to answer this themselves. diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/cpu/common-xen.c b/linux-2.6-xen-sparse/arch/i386/kernel/cpu/common-xen.c index ffcee15c86..4d1a486dee 100644 --- a/linux-2.6-xen-sparse/arch/i386/kernel/cpu/common-xen.c +++ b/linux-2.6-xen-sparse/arch/i386/kernel/cpu/common-xen.c @@ -15,6 +15,10 @@ #include <asm/mpspec.h> #include <asm/apic.h> #include <mach_apic.h> +#else +#ifdef CONFIG_XEN +#define phys_pkg_id(a,b) a +#endif #endif #include <asm/hypervisor.h> @@ -28,9 +32,10 @@ DEFINE_PER_CPU(unsigned char, cpu_16bit_stack[CPU_16BIT_STACK_SIZE]); EXPORT_PER_CPU_SYMBOL(cpu_16bit_stack); #endif -static int cachesize_override __devinitdata = -1; -static int disable_x86_fxsr __devinitdata = 0; -static int disable_x86_serial_nr __devinitdata = 1; +static int cachesize_override __cpuinitdata = -1; +static int disable_x86_fxsr __cpuinitdata; +static int disable_x86_serial_nr __cpuinitdata = 1; +static int disable_x86_sep __cpuinitdata; struct cpu_dev * cpu_devs[X86_VENDOR_NUM] = {}; @@ -62,7 +67,7 @@ static int __init cachesize_setup(char *str) } __setup("cachesize=", cachesize_setup); -int __devinit get_model_name(struct cpuinfo_x86 *c) +int __cpuinit get_model_name(struct cpuinfo_x86 *c) { unsigned int *v; char *p, *q; @@ -92,7 +97,7 @@ int __devinit get_model_name(struct cpuinfo_x86 *c) } -void __devinit display_cacheinfo(struct cpuinfo_x86 *c) +void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c) { unsigned int n, dummy, ecx, edx, l2size; @@ -133,7 +138,7 @@ void __devinit display_cacheinfo(struct cpuinfo_x86 *c) /* in particular, if CPUID levels 0x80000002..4 are supported, this isn't used */ /* Look up CPU names by table lookup. */ -static char __devinit *table_lookup_model(struct cpuinfo_x86 *c) +static char __cpuinit *table_lookup_model(struct cpuinfo_x86 *c) { struct cpu_model_info *info; @@ -154,7 +159,7 @@ static char __devinit *table_lookup_model(struct cpuinfo_x86 *c) } -static void __devinit get_cpu_vendor(struct cpuinfo_x86 *c, int early) +static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c, int early) { char *v = c->x86_vendor_id; int i; @@ -190,6 +195,14 @@ static int __init x86_fxsr_setup(char * s) __setup("nofxsr", x86_fxsr_setup); +static int __init x86_sep_setup(char * s) +{ + disable_x86_sep = 1; + return 1; +} +__setup("nosep", x86_sep_setup); + + /* Standard macro to see if a specific flag is changeable */ static inline int flag_is_changeable_p(u32 flag) { @@ -213,7 +226,7 @@ static inline int flag_is_changeable_p(u32 flag) /* Probe for the CPUID instruction */ -static int __devinit have_cpuid_p(void) +static int __cpuinit have_cpuid_p(void) { return flag_is_changeable_p(X86_EFLAGS_ID); } @@ -257,10 +270,10 @@ static void __init early_cpu_detect(void) } } -void __devinit generic_identify(struct cpuinfo_x86 * c) +void __cpuinit generic_identify(struct cpuinfo_x86 * c) { u32 tfms, xlvl; - int junk; + int ebx; if (have_cpuid_p()) { /* Get vendor name */ @@ -276,7 +289,7 @@ void __devinit generic_identify(struct cpuinfo_x86 * c) /* Intel-defined flags: level 0x00000001 */ if ( c->cpuid_level >= 0x00000001 ) { u32 capability, excap; - cpuid(0x00000001, &tfms, &junk, &excap, &capability); + cpuid(0x00000001, &tfms, &ebx, &excap, &capability); c->x86_capability[0] = capability; c->x86_capability[4] = excap; c->x86 = (tfms >> 8) & 15; @@ -286,6 +299,11 @@ void __devinit generic_identify(struct cpuinfo_x86 * c) if (c->x86 >= 0x6) c->x86_model += ((tfms >> 16) & 0xF) << 4; c->x86_mask = tfms & 15; +#ifdef CONFIG_SMP + c->apicid = phys_pkg_id((ebx >> 24) & 0xFF, 0); +#else + c->apicid = (ebx >> 24) & 0xFF; +#endif } else { /* Have CPUID level 0 only - unheard of */ c->x86 = 4; @@ -310,7 +328,7 @@ void __devinit generic_identify(struct cpuinfo_x86 * c) #endif } -static void __devinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c) +static void __cpuinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c) { if (cpu_has(c, X86_FEATURE_PN) && disable_x86_serial_nr ) { /* Disable processor serial number */ @@ -338,7 +356,7 @@ __setup("serialnumber", x86_serial_nr_setup); /* * This does the hard work of actually picking apart the CPU stuff... */ -void __devinit identify_cpu(struct cpuinfo_x86 *c) +void __cpuinit identify_cpu(struct cpuinfo_x86 *c) { int i; @@ -408,6 +426,10 @@ void __devinit identify_cpu(struct cpuinfo_x86 *c) clear_bit(X86_FEATURE_XMM, c->x86_capability); } + /* SEP disabled? */ + if (disable_x86_sep) + clear_bit(X86_FEATURE_SEP, c->x86_capability); + if (disable_pse) clear_bit(X86_FEATURE_PSE, c->x86_capability); @@ -420,7 +442,7 @@ void __devinit identify_cpu(struct cpuinfo_x86 *c) else /* Last resort... */ sprintf(c->x86_model_id, "%02x/%02x", - c->x86_vendor, c->x86_model); + c->x86, c->x86_model); } /* Now the feature flags better reflect actual CPU features! */ @@ -456,7 +478,7 @@ void __devinit identify_cpu(struct cpuinfo_x86 *c) } #ifdef CONFIG_X86_HT -void __devinit detect_ht(struct cpuinfo_x86 *c) +void __cpuinit detect_ht(struct cpuinfo_x86 *c) { u32 eax, ebx, ecx, edx; int index_msb, core_bits; @@ -464,7 +486,6 @@ void __devinit detect_ht(struct cpuinfo_x86 *c) cpuid(1, &eax, &ebx, &ecx, &edx); - c->apicid = phys_pkg_id((ebx >> 24) & 0xFF, 0); if (!cpu_has(c, X86_FEATURE_HT) || cpu_has(c, X86_FEATURE_CMP_LEGACY)) return; @@ -503,7 +524,7 @@ void __devinit detect_ht(struct cpuinfo_x86 *c) } #endif -void __devinit print_cpu_info(struct cpuinfo_x86 *c) +void __cpuinit print_cpu_info(struct cpuinfo_x86 *c) { char *vendor = NULL; @@ -703,7 +724,7 @@ void __cpuinit cpu_init(void) } #ifdef CONFIG_HOTPLUG_CPU -void __devinit cpu_uninit(void) +void __cpuinit cpu_uninit(void) { int cpu = raw_smp_processor_id(); cpu_clear(cpu, cpu_initialized); diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/cpu/intel_cacheinfo-xen.c b/linux-2.6-xen-sparse/arch/i386/kernel/cpu/intel_cacheinfo-xen.c new file mode 100644 index 0000000000..f549dd2af6 --- /dev/null +++ b/linux-2.6-xen-sparse/arch/i386/kernel/cpu/intel_cacheinfo-xen.c @@ -0,0 +1,693 @@ +/* + * Routines to indentify caches on Intel CPU. + * + * Changes: + * Venkatesh Pallipadi : Adding cache identification through cpuid(4) + * Ashok Raj <ashok.raj@intel.com>: Work with CPU hotplug infrastructure. + */ + +#include <linux/init.h> +#include <linux/slab.h> +#include <linux/device.h> +#include <linux/compiler.h> +#include <linux/cpu.h> +#include <linux/sched.h> + +#include <asm/processor.h> +#include <asm/smp.h> + +#define LVL_1_INST 1 +#define LVL_1_DATA 2 +#define LVL_2 3 +#define LVL_3 4 +#define LVL_TRACE 5 + +struct _cache_table +{ + unsigned char descriptor; + char cache_type; + short size; +}; + +/* all the cache descriptor types we care about (no TLB or trace cache entries) */ +static struct _cache_table cache_table[] __cpuinitdata = +{ + { 0x06, LVL_1_INST, 8 }, /* 4-way set assoc, 32 byte line size */ + { 0x08, LVL_1_INST, 16 }, /* 4-way set assoc, 32 byte line size */ + { 0x0a, LVL_1_DATA, 8 }, /* 2 way set assoc, 32 byte line size */ + { 0x0c, LVL_1_DATA, 16 }, /* 4-way set assoc, 32 byte line size */ + { 0x22, LVL_3, 512 }, /* 4-way set assoc, sectored cache, 64 byte line size */ + { 0x23, LVL_3, 1024 }, /* 8-way set assoc, sectored cache, 64 byte line size */ + { 0x25, LVL_3, 2048 }, /* 8-way set assoc, sectored cache, 64 byte line size */ + { 0x29, LVL_3, 4096 }, /* 8-way set assoc, sectored cache, 64 byte line size */ + { 0x2c, LVL_1_DATA, 32 }, /* 8-way set assoc, 64 byte line size */ + { 0x30, LVL_1_INST, 32 }, /* 8-way set assoc, 64 byte line size */ + { 0x39, LVL_2, 128 }, /* 4-way set assoc, sectored cache, 64 byte line size */ + { 0x3a, LVL_2, 192 }, /* 6-way set assoc, sectored cache, 64 byte line size */ + { 0x3b, LVL_2, 128 }, /* 2-way set assoc, sectored cache, 64 byte line size */ + { 0x3c, LVL_2, 256 }, /* 4-way set assoc, sectored cache, 64 byte line size */ + { 0x3d, LVL_2, 384 }, /* 6-way set assoc, sectored cache, 64 byte line size */ + { 0x3e, LVL_2, 512 }, /* 4-way set assoc, sectored cache, 64 byte line size */ + { 0x41, LVL_2, 128 }, /* 4-way set assoc, 32 byte line size */ + { 0x42, LVL_2, 256 }, /* 4-way set assoc, 32 byte line size */ + { 0x43, LVL_2, 512 }, /* 4-way set assoc, 32 byte line size */ + { 0x44, LVL_2, 1024 }, /* 4-way set assoc, 32 byte line size */ + { 0x45, LVL_2, 2048 }, /* 4-way set assoc, 32 byte line size */ + { 0x46, LVL_3, 4096 }, /* 4-way set assoc, 64 byte line size */ + { 0x47, LVL_3, 8192 }, /* 8-way set assoc, 64 byte line size */ + { 0x49, LVL_3, 4096 }, /* 16-way set assoc, 64 byte line size */ + { 0x4a, LVL_3, 6144 }, /* 12-way set assoc, 64 byte line size */ + { 0x4b, LVL_3, 8192 }, /* 16-way set assoc, 64 byte line size */ + { 0x4c, LVL_3, 12288 }, /* 12-way set assoc, 64 byte line size */ + { 0x4d, LVL_3, 16384 }, /* 16-way set assoc, 64 byte line size */ + { 0x60, LVL_1_DATA, 16 }, /* 8-way set assoc, sectored cache, 64 byte line size */ + { 0x66, LVL_1_DATA, 8 }, /* 4-way set assoc, sectored cache, 64 byte line size */ + { 0x67, LVL_1_DATA, 16 }, /* 4-way set assoc, sectored cache, 64 byte line size */ + { 0x68, LVL_1_DATA, 32 }, /* 4-way set assoc, sectored cache, 64 byte line size */ + { 0x70, LVL_TRACE, 12 }, /* 8-way set assoc */ + { 0x71, LVL_TRACE, 16 }, /* 8-way set assoc */ + { 0x72, LVL_TRACE, 32 }, /* 8-way set assoc */ + { 0x73, LVL_TRACE, 64 }, /* 8-way set assoc */ + { 0x78, LVL_2, 1024 }, /* 4-way set assoc, 64 byte line size */ + { 0x79, LVL_2, 128 }, /* 8-way set assoc, sectored cache, 64 byte line size */ + { 0x7a, LVL_2, 256 }, /* 8-way set assoc, sectored cache, 64 byte line size */ + { 0x7b, LVL_2, 512 }, /* 8-way set assoc, sectored cache, 64 byte line size */ + { 0x7c, LVL_2, 1024 }, /* 8-way set assoc, sectored cache, 64 byte line size */ + { 0x7d, LVL_2, 2048 }, /* 8-way set assoc, 64 byte line size */ + { 0x7f, LVL_2, 512 }, /* 2-way set assoc, 64 byte line size */ + { 0x82, LVL_2, 256 }, /* 8-way set assoc, 32 byte line size */ + { 0x83, LVL_2, 512 }, /* 8-way set assoc, 32 byte line size */ + { 0x84, LVL_2, 1024 }, /* 8-way set assoc, 32 byte line size */ + { 0x85, LVL_2, 2048 }, /* 8-way set assoc, 32 byte line size */ + { 0x86, LVL_2, 512 }, /* 4-way set assoc, 64 byte line size */ + { 0x87, LVL_2, 1024 }, /* 8-way set assoc, 64 byte line size */ + { 0x00, 0, 0} +}; + + +enum _cache_type +{ + CACHE_TYPE_NULL = 0, + CACHE_TYPE_DATA = 1, + CACHE_TYPE_INST = 2, + CACHE_TYPE_UNIFIED = 3 +}; + +union _cpuid4_leaf_eax { + struct { + enum _cache_type type:5; + unsigned int level:3; + unsigned int is_self_initializing:1; + unsigned int is_fully_associative:1; + unsigned int reserved:4; + unsigned int num_threads_sharing:12; + unsigned int num_cores_on_die:6; + } split; + u32 full; +}; + +union _cpuid4_leaf_ebx { + struct { + unsigned int coherency_line_size:12; + unsigned int physical_line_partition:10; + unsigned int ways_of_associativity:10; + } split; + u32 full; +}; + +union _cpuid4_leaf_ecx { + struct { + unsigned int number_of_sets:32; + } split; + u32 full; +}; + +struct _cpuid4_info { + union _cpuid4_leaf_eax eax; + union _cpuid4_leaf_ebx ebx; + union _cpuid4_leaf_ecx ecx; + unsigned long size; + cpumask_t shared_cpu_map; +}; + +static unsigned short num_cache_leaves; + +static int __cpuinit cpuid4_cache_lookup(int index, struct _cpuid4_info *this_leaf) +{ + unsigned int eax, ebx, ecx, edx; + union _cpuid4_leaf_eax cache_eax; + + cpuid_count(4, index, &eax, &ebx, &ecx, &edx); + cache_eax.full = eax; + if (cache_eax.split.type == CACHE_TYPE_NULL) + return -EIO; /* better error ? */ + + this_leaf->eax.full = eax; + this_leaf->ebx.full = ebx; + this_leaf->ecx.full = ecx; + this_leaf->size = (this_leaf->ecx.split.number_of_sets + 1) * + (this_leaf->ebx.split.coherency_line_size + 1) * + (this_leaf->ebx.split.physical_line_partition + 1) * + (this_leaf->ebx.split.ways_of_associativity + 1); + return 0; +} + +/* will only be called once; __init is safe here */ +static int __init find_num_cache_leaves(void) +{ + unsigned int eax, ebx, ecx, edx; + union _cpuid4_leaf_eax cache_eax; + int i = -1; + + do { + ++i; + /* Do cpuid(4) loop to find out num_cache_leaves */ + cpuid_count(4, i, &eax, &ebx, &ecx, &edx); + cache_eax.full = eax; + } while (cache_eax.split.type != CACHE_TYPE_NULL); + return i; +} + +unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c) +{ + unsigned int trace = 0, l1i = 0, l1d = 0, l2 = 0, l3 = 0; /* Cache sizes */ + unsigned int new_l1d = 0, new_l1i = 0; /* Cache sizes from cpuid(4) */ + unsigned int new_l2 = 0, new_l3 = 0, i; /* Cache sizes from cpuid(4) */ + unsigned int l2_id = 0, l3_id = 0, num_threads_sharing, index_msb; +#ifdef CONFIG_SMP +#ifndef CONFIG_XEN + unsigned int cpu = (c == &boot_cpu_data) ? 0 : (c - cpu_data); +#endif +#endif + + if (c->cpuid_level > 3) { + static int is_initialized; + + if (is_initialized == 0) { + /* Init num_cache_leaves from boot CPU */ + num_cache_leaves = find_num_cache_leaves(); + is_initialized++; + } + + /* + * Whenever possible use cpuid(4), deterministic cache + * parameters cpuid leaf to find the cache details + */ + for (i = 0; i < num_cache_leaves; i++) { + struct _cpuid4_info this_leaf; + + int retval; + + retval = cpuid4_cache_lookup(i, &this_leaf); + if (retval >= 0) { + switch(this_leaf.eax.split.level) { + case 1: + if (this_leaf.eax.split.type == + CACHE_TYPE_DATA) + new_l1d = this_leaf.size/1024; + else if (this_leaf.eax.split.type == + CACHE_TYPE_INST) + new_l1i = this_leaf.size/1024; + break; + case 2: + new_l2 = this_leaf.size/1024; + num_threads_sharing = 1 + this_leaf.eax.split.num_threads_sharing; + index_msb = get_count_order(num_threads_sharing); + l2_id = c->apicid >> index_msb; + break; + case 3: + new_l3 = this_leaf.size/1024; + num_threads_sharing = 1 + this_leaf.eax.split.num_threads_sharing; + index_msb = get_count_order(num_threads_sharing); + l3_id = c->apicid >> index_msb; + break; + default: + break; + } + } + } + } + /* + * Don't use cpuid2 if cpuid4 is supported. For P4, we use cpuid2 for + * trace cache + */ + if ((num_cache_leaves == 0 || c->x86 == 15) && c->cpuid_level > 1) { + /* supports eax=2 call */ + int i, j, n; + int regs[4]; + unsigned char *dp = (unsigned char *)regs; + int only_trace = 0; + + if (num_cache_leaves != 0 && c->x86 == 15) + only_trace = 1; + + /* Number of times to iterate */ + n = cpuid_eax(2) & 0xFF; + + for ( i = 0 ; i < n ; i++ ) { + cpuid(2, ®s[0], ®s[1], ®s[2], ®s[3]); + + /* If bit 31 is set, this is an unknown format */ + for ( j = 0 ; j < 3 ; j++ ) { + if ( regs[j] < 0 ) regs[j] = 0; + } + + /* Byte 0 is level count, not a descriptor */ + for ( j = 1 ; j < 16 ; j++ ) { + unsigned char des = dp[j]; + unsigned char k = 0; + + /* look up this descriptor in the table */ + while (cache_table[k].descriptor != 0) + { + if (cache_table[k].descriptor == des) { + if (only_trace && cache_table[k].cache_type != LVL_TRACE) + break; + switch (cache_table[k].cache_type) { + case LVL_1_INST: + l1i += cache_table[k].size; + break; + case LVL_1_DATA: + l1d += cache_table[k].size; + break; + case LVL_2: + l2 += cache_table[k].size; + break; + case LVL_3: + l3 += cache_table[k].size; + break; + case LVL_TRACE: + trace += cache_table[k].size; + break; + } + + break; + } + + k++; + } + } + } + } + + if (new_l1d) + l1d = new_l1d; + + if (new_l1i) + l1i = new_l1i; + + if (new_l2) { + l2 = new_l2; +#ifdef CONFIG_SMP +#ifndef CONFIG_XEN + cpu_llc_id[cpu] = l2_id; +#endif +#endif + } + + if (new_l3) { + l3 = new_l3; +#ifdef CONFIG_SMP +#ifndef CONFIG_XEN + cpu_llc_id[cpu] = l3_id; +#endif +#endif + } + + if (trace) + printk (KERN_INFO "CPU: Trace cache: %dK uops", trace); + else if ( l1i ) + printk (KERN_INFO "CPU: L1 I cache: %dK", l1i); + + if (l1d) + printk(", L1 D cache: %dK\n", l1d); + else + printk("\n"); + + if (l2) + printk(KERN_INFO "CPU: L2 cache: %dK\n", l2); + + if (l3) + printk(KERN_INFO "CPU: L3 cache: %dK\n", l3); + + c->x86_cache_size = l3 ? l3 : (l2 ? l2 : (l1i+l1d)); + + return l2; +} + +/* pointer to _cpuid4_info array (for each cache leaf) */ +static struct _cpuid4_info *cpuid4_info[NR_CPUS]; +#define CPUID4_INFO_IDX(x,y) (&((cpuid4_info[x])[y])) + +#ifdef CONFIG_SMP +static void __cpuinit cache_shared_cpu_map_setup(unsigned int cpu, int index) +{ + struct _cpuid4_info *this_leaf, *sibling_leaf; + unsigned long num_threads_sharing; + int index_msb, i; + struct cpuinfo_x86 *c = cpu_data; + + this_leaf = CPUID4_INFO_IDX(cpu, index); + num_threads_sharing = 1 + this_leaf->eax.split.num_threads_sharing; + + if (num_threads_sharing == 1) + cpu_set(cpu, this_leaf->shared_cpu_map); + else { + index_msb = get_count_order(num_threads_sharing); + + for_each_online_cpu(i) { + if (c[i].apicid >> index_msb == + c[cpu].apicid >> index_msb) { + cpu_set(i, this_leaf->shared_cpu_map); + if (i != cpu && cpuid4_info[i]) { + sibling_leaf = CPUID4_INFO_IDX(i, index); + cpu_set(cpu, sibling_leaf->shared_cpu_map); + } + } + } + } +} +static void __cpuinit cache_remove_shared_cpu_map(unsigned int cpu, int index) +{ + struct _cpuid4_info *this_leaf, *sibling_leaf; + int sibling; + + this_leaf = CPUID4_INFO_IDX(cpu, index); + for_each_cpu_mask(sibling, this_leaf->shared_cpu_map) { + sibling_leaf = CPUID4_INFO_IDX(sibling, index); + cpu_clear(cpu, sibling_leaf->shared_cpu_map); + } +} +#else +static void __init cache_shared_cpu_map_setup(unsigned int cpu, int index) {} +static void __init cache_remove_shared_cpu_map(unsigned int cpu, int index) {} +#endif + +static void free_cache_attributes(unsigned int cpu) +{ + kfree(cpuid4_info[cpu]); + cpuid4_info[cpu] = NULL; +} + +static int __cpuinit detect_cache_attributes(unsigned int cpu) +{ + struct _cpuid4_info *this_leaf; + unsigned long j; + int retval; + cpumask_t oldmask; + + if (num_cache_leaves == 0) + return -ENOENT; + + cpuid4_info[cpu] = kmalloc( + sizeof(struct _cpuid4_info) * num_cache_leaves, GFP_KERNEL); + if (unlikely(cpuid4_info[cpu] == NULL)) + return -ENOMEM; + memset(cpuid4_info[cpu], 0, + sizeof(struct _cpuid4_info) * num_cache_leaves); + + oldmask = current->cpus_allowed; + retval = set_cpus_allowed(current, cpumask_of_cpu(cpu)); + if (retval) + goto out; + + /* Do cpuid and store the results */ + retval = 0; + for (j = 0; j < num_cache_leaves; j++) { + this_leaf = CPUID4_INFO_IDX(cpu, j); + retval = cpuid4_cache_lookup(j, this_leaf); + if (unlikely(retval < 0)) + break; + cache_shared_cpu_map_setup(cpu, j); + } + set_cpus_allowed(current, oldmask); + +out: + if (retval) + free_cache_attributes(cpu); + return retval; +} + +#ifdef CONFIG_SYSFS + +#include <linux/kobject.h> +#include <linux/sysfs.h> + +extern struct sysdev_class cpu_sysdev_class; /* from drivers/base/cpu.c */ + +/* pointer to kobject for cpuX/cache */ +static struct kobject * cache_kobject[NR_CPUS]; + +struct _index_kobject { + struct kobject kobj; + unsigned int cpu; + unsigned short index; +}; + +/* pointer to array of kobjects for cpuX/cache/indexY */ +static struct _index_kobject *index_kobject[NR_CPUS]; +#define INDEX_KOBJECT_PTR(x,y) (&((index_kobject[x])[y])) + +#define show_one_plus(file_name, object, val) \ +static ssize_t show_##file_name \ + (struct _cpuid4_info *this_leaf, char *buf) \ +{ \ + return sprintf (buf, "%lu\n", (unsigned long)this_leaf->object + val); \ +} + +show_one_plus(level, eax.split.level, 0); +show_one_plus(coherency_line_size, ebx.split.coherency_line_size, 1); +show_one_plus(physical_line_partition, ebx.split.physical_line_partition, 1); +show_one_plus(ways_of_associativity, ebx.split.ways_of_associativity, 1); +show_one_plus(number_of_sets, ecx.split.number_of_sets, 1); + +static ssize_t show_size(struct _cpuid4_info *this_leaf, char *buf) +{ + return sprintf (buf, "%luK\n", this_leaf->size / 1024); +} + +static ssize_t show_shared_cpu_map(struct _cpuid4_info *this_leaf, char *buf) +{ + char mask_str[NR_CPUS]; + cpumask_scnprintf(mask_str, NR_CPUS, this_leaf->shared_cpu_map); + return sprintf(buf, "%s\n", mask_str); +} + +static ssize_t show_type(struct _cpuid4_info *this_leaf, char *buf) { + switch(this_leaf->eax.split.type) { + case CACHE_TYPE_DATA: + return sprintf(buf, "Data\n"); + break; + case CACHE_TYPE_INST: + return sprintf(buf, "Instruction\n"); + break; + case CACHE_TYPE_UNIFIED: + return sprintf(buf, "Unified\n"); + break; + default: + return sprintf(buf, "Unknown\n"); + break; + } +} + +struct _cache_attr { + struct attribute attr; + ssize_t (*show)(struct _cpuid4_info *, char *); + ssize_t (*store)(struct _cpuid4_info *, const char *, size_t count); +}; + +#define define_one_ro(_name) \ +static struct _cache_attr _name = \ + __ATTR(_name, 0444, show_##_name, NULL) + +define_one_ro(level); +define_one_ro(type); +define_one_ro(coherency_line_size); +define_one_ro(physical_line_partition); +define_one_ro(ways_of_associativity); +define_one_ro(number_of_sets); +define_one_ro(size); +define_one_ro(shared_cpu_map); + +static struct attribute * default_attrs[] = { + &type.attr, + &level.attr, + &coherency_line_size.attr, + &physical_line_partition.attr, + &ways_of_associativity.attr, + &number_of_sets.attr, + &size.attr, + &shared_cpu_map.attr, + NULL +}; + +#define to_object(k) container_of(k, struct _index_kobject, kobj) +#define to_attr(a) container_of(a, struct _cache_attr, attr) + +static ssize_t show(struct kobject * kobj, struct attribute * attr, char * buf) +{ + struct _cache_attr *fattr = to_attr(attr); + struct _index_kobject *this_leaf = to_object(kobj); + ssize_t ret; + + ret = fattr->show ? + fattr->show(CPUID4_INFO_IDX(this_leaf->cpu, this_leaf->index), + buf) : + 0; + return ret; +} + +static ssize_t store(struct kobject * kobj, struct attribute * attr, + const char * buf, size_t count) +{ + return 0; +} + +static struct sysfs_ops sysfs_ops = { + .show = show, + .store = store, +}; + +static struct kobj_type ktype_cache = { + .sysfs_ops = &sysfs_ops, + .default_attrs = default_attrs, +}; + +static struct kobj_type ktype_percpu_entry = { + .sysfs_ops = &sysfs_ops, +}; + +static void cpuid4_cache_sysfs_exit(unsigned int cpu) +{ + kfree(cache_kobject[cpu]); + kfree(index_kobject[cpu]); + cache_kobject[cpu] = NULL; + index_kobject[cpu] = NULL; + free_cache_attributes(cpu); +} + +static int __cpuinit cpuid4_cache_sysfs_init(unsigned int cpu) +{ + + if (num_cache_leaves == 0) + return -ENOENT; + + detect_cache_attributes(cpu); + if (cpuid4_info[cpu] == NULL) + return -ENOENT; + + /* Allocate all required memory */ + cache_kobject[cpu] = kmalloc(sizeof(struct kobject), GFP_KERNEL); + if (unlikely(cache_kobject[cpu] == NULL)) + goto err_out; + memset(cache_kobject[cpu], 0, sizeof(struct kobject)); + + index_kobject[cpu] = kmalloc( + sizeof(struct _index_kobject ) * num_cache_leaves, GFP_KERNEL); + if (unlikely(index_kobject[cpu] == NULL)) + goto err_out; + memset(index_kobject[cpu], 0, + sizeof(struct _index_kobject) * num_cache_leaves); + + return 0; + +err_out: + cpuid4_cache_sysfs_exit(cpu); + return -ENOMEM; +} + +/* Add/Remove cache interface for CPU device */ +static int __cpuinit cache_add_dev(struct sys_device * sys_dev) +{ + unsigned int cpu = sys_dev->id; + unsigned long i, j; + struct _index_kobject *this_object; + int retval = 0; + + retval = cpuid4_cache_sysfs_init(cpu); + if (unlikely(retval < 0)) + return retval; + + cache_kobject[cpu]->parent = &sys_dev->kobj; + kobject_set_name(cache_kobject[cpu], "%s", "cache"); + cache_kobject[cpu]->ktype = &ktype_percpu_entry; + retval = kobject_register(cache_kobject[cpu]); + + for (i = 0; i < num_cache_leaves; i++) { + this_object = INDEX_KOBJECT_PTR(cpu,i); + this_object->cpu = cpu; + this_object->index = i; + this_object->kobj.parent = cache_kobject[cpu]; + kobject_set_name(&(this_object->kobj), "index%1lu", i); + this_object->kobj.ktype = &ktype_cache; + retval = kobject_register(&(this_object->kobj)); + if (unlikely(retval)) { + for (j = 0; j < i; j++) { + kobject_unregister( + &(INDEX_KOBJECT_PTR(cpu,j)->kobj)); + } + kobject_unregister(cache_kobject[cpu]); + cpuid4_cache_sysfs_exit(cpu); + break; + } + } + return retval; +} + +static void __cpuexit cache_remove_dev(struct sys_device * sys_dev) +{ + unsigned int cpu = sys_dev->id; + unsigned long i; + + for (i = 0; i < num_cache_leaves; i++) { + cache_remove_shared_cpu_map(cpu, i); + kobject_unregister(&(INDEX_KOBJECT_PTR(cpu,i)->kobj)); + } + kobject_unregister(cache_kobject[cpu]); + cpuid4_cache_sysfs_exit(cpu); + return; +} + +static int cacheinfo_cpu_callback(struct notifier_block *nfb, + unsigned long action, void *hcpu) +{ + unsigned int cpu = (unsigned long)hcpu; + struct sys_device *sys_dev; + + sys_dev = get_cpu_sysdev(cpu); + switch (action) { + case CPU_ONLINE: + cache_add_dev(sys_dev); + break; + case CPU_DEAD: + cache_remove_dev(sys_dev); + break; + } + return NOTIFY_OK; +} + +static struct notifier_block cacheinfo_cpu_notifier = +{ + .notifier_call = cacheinfo_cpu_callback, +}; + +static int __cpuinit cache_sysfs_init(void) +{ + int i; + + if (num_cache_leaves == 0) + return 0; + + register_cpu_notifier(&cacheinfo_cpu_notifier); + + for_each_online_cpu(i) { + cacheinfo_cpu_callback(&cacheinfo_cpu_notifier, CPU_ONLINE, + (void *)(long)i); + } + + return 0; +} + +device_initcall(cache_sysfs_init); + +#endif diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/cpu/mtrr/main-xen.c b/linux-2.6-xen-sparse/arch/i386/kernel/cpu/mtrr/main-xen.c index 06720038ff..380e20ec5b 100644 --- a/linux-2.6-xen-sparse/arch/i386/kernel/cpu/mtrr/main-xen.c +++ b/linux-2.6-xen-sparse/arch/i386/kernel/cpu/mtrr/main-xen.c @@ -4,11 +4,12 @@ #include <linux/module.h> #include <linux/seq_file.h> #include <asm/uaccess.h> +#include <linux/mutex.h> #include <asm/mtrr.h> #include "mtrr.h" -static DECLARE_MUTEX(mtrr_sem); +static DEFINE_MUTEX(mtrr_mutex); void generic_get_mtrr(unsigned int reg, unsigned long *base, unsigned int *size, mtrr_type * type) @@ -65,7 +66,7 @@ int mtrr_add_page(unsigned long base, unsigned long size, int error; dom0_op_t op; - down(&mtrr_sem); + mutex_lock(&mtrr_mutex); op.cmd = DOM0_ADD_MEMTYPE; op.u.add_memtype.mfn = base; @@ -73,7 +74,7 @@ int mtrr_add_page(unsigned long base, unsigned long size, op.u.add_memtype.type = type; error = HYPERVISOR_dom0_op(&op); if (error) { - up(&mtrr_sem); + mutex_unlock(&mtrr_mutex); BUG_ON(error > 0); return error; } @@ -81,7 +82,7 @@ int mtrr_add_page(unsigned long base, unsigned long size, if (increment) ++usage_table[op.u.add_memtype.reg]; - up(&mtrr_sem); + mutex_unlock(&mtrr_mutex); return op.u.add_memtype.reg; } @@ -118,7 +119,7 @@ int mtrr_del_page(int reg, unsigned long base, unsigned long size) int error = -EINVAL; dom0_op_t op; - down(&mtrr_sem); + mutex_lock(&mtrr_mutex); if (reg < 0) { /* Search for existing MTRR */ @@ -151,7 +152,7 @@ int mtrr_del_page(int reg, unsigned long base, unsigned long size) } error = reg; out: - up(&mtrr_sem); + mutex_unlock(&mtrr_mutex); return error; } diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/crash.c b/linux-2.6-xen-sparse/arch/i386/kernel/crash.c index 75aab8ef35..4b6698c7aa 100644 --- a/linux-2.6-xen-sparse/arch/i386/kernel/crash.c +++ b/linux-2.6-xen-sparse/arch/i386/kernel/crash.c @@ -69,7 +69,7 @@ static void crash_save_this_cpu(struct pt_regs *regs, int cpu) * for the data I pass, and I need tags * on the data to indicate what information I have * squirrelled away. ELF notes happen to provide - * all of that that no need to invent something new. + * all of that, so there is no need to invent something new. */ buf = (u32*)per_cpu_ptr(crash_notes, cpu); if (!buf) @@ -106,7 +106,7 @@ static int crash_nmi_callback(struct pt_regs *regs, int cpu) return 1; local_irq_disable(); - if (!user_mode(regs)) { + if (!user_mode_vm(regs)) { crash_fixup_ss_esp(&fixed_regs, regs); regs = &fixed_regs; } diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/entry-xen.S b/linux-2.6-xen-sparse/arch/i386/kernel/entry-xen.S index 8c5ace45ef..75654a1c84 100644 --- a/linux-2.6-xen-sparse/arch/i386/kernel/entry-xen.S +++ b/linux-2.6-xen-sparse/arch/i386/kernel/entry-xen.S @@ -271,6 +271,10 @@ ENTRY(system_call) pushl %eax # save orig_eax SAVE_ALL GET_THREAD_INFO(%ebp) + testl $TF_MASK,EFLAGS(%esp) + jz no_singlestep + orl $_TIF_SINGLESTEP,TI_flags(%ebp) +no_singlestep: # system call tracing in operation / emulation /* Note, _TIF_SECCOMP is bit number 8, and so it needs testw and not testb */ testw $(_TIF_SYSCALL_EMU|_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT),TI_flags(%ebp) diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/head-xen.S b/linux-2.6-xen-sparse/arch/i386/kernel/head-xen.S index 1d0278c69b..edf86d6fb1 100644 --- a/linux-2.6-xen-sparse/arch/i386/kernel/head-xen.S +++ b/linux-2.6-xen-sparse/arch/i386/kernel/head-xen.S @@ -7,6 +7,7 @@ #include <linux/linkage.h> #include <asm/segment.h> #include <asm/page.h> +#include <asm/cache.h> #include <asm/thread_info.h> #include <asm/asm-offsets.h> #include <xen/interface/xen.h> @@ -92,6 +93,7 @@ ENTRY(empty_zero_page) /* * The Global Descriptor Table contains 28 quadwords, per-CPU. */ + .align L1_CACHE_BYTES ENTRY(cpu_gdt_table) .quad 0x0000000000000000 /* NULL descriptor */ .quad 0x0000000000000000 /* 0x0b reserved */ diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/io_apic-xen.c b/linux-2.6-xen-sparse/arch/i386/kernel/io_apic-xen.c index 844d6e9ad9..4b26c5bf8a 100644 --- a/linux-2.6-xen-sparse/arch/i386/kernel/io_apic-xen.c +++ b/linux-2.6-xen-sparse/arch/i386/kernel/io_apic-xen.c @@ -394,8 +394,8 @@ static inline void rotate_irqs_among_cpus(unsigned long useful_load_threshold) { int i, j; Dprintk("Rotating IRQs among CPUs.\n"); - for (i = 0; i < NR_CPUS; i++) { - for (j = 0; cpu_online(i) && (j < NR_IRQS); j++) { + for_each_online_cpu(i) { + for (j = 0; j < NR_IRQS; j++) { if (!irq_desc[j].action) continue; /* Is it a significant load ? */ @@ -424,7 +424,7 @@ static void do_irq_balance(void) unsigned long imbalance = 0; cpumask_t allowed_mask, target_cpu_mask, tmp; - for (i = 0; i < NR_CPUS; i++) { + for_each_possible_cpu(i) { int package_index; CPU_IRQ(i) = 0; if (!cpu_online(i)) @@ -465,9 +465,7 @@ static void do_irq_balance(void) } } /* Find the least loaded processor package */ - for (i = 0; i < NR_CPUS; i++) { - if (!cpu_online(i)) - continue; + for_each_online_cpu(i) { if (i != CPU_TO_PACKAGEINDEX(i)) continue; if (min_cpu_irq > CPU_IRQ(i)) { @@ -484,9 +482,7 @@ tryanothercpu: */ tmp_cpu_irq = 0; tmp_loaded = -1; - for (i = 0; i < NR_CPUS; i++) { - if (!cpu_online(i)) - continue; + for_each_online_cpu(i) { if (i != CPU_TO_PACKAGEINDEX(i)) continue; if (max_cpu_irq <= CPU_IRQ(i)) @@ -662,9 +658,7 @@ static int __init balanced_irq_init(void) if (smp_num_siblings > 1 && !cpus_empty(tmp)) physical_balance = 1; - for (i = 0; i < NR_CPUS; i++) { - if (!cpu_online(i)) - continue; + for_each_online_cpu(i) { irq_cpu_data[i].irq_delta = kmalloc(sizeof(unsigned long) * NR_IRQS, GFP_KERNEL); irq_cpu_data[i].last_irq = kmalloc(sizeof(unsigned long) * NR_IRQS, GFP_KERNEL); if (irq_cpu_data[i].irq_delta == NULL || irq_cpu_data[i].last_irq == NULL) { @@ -681,9 +675,11 @@ static int __init balanced_irq_init(void) else printk(KERN_ERR "balanced_irq_init: failed to spawn balanced_irq"); failed: - for (i = 0; i < NR_CPUS; i++) { + for_each_possible_cpu(i) { kfree(irq_cpu_data[i].irq_delta); + irq_cpu_data[i].irq_delta = NULL; kfree(irq_cpu_data[i].last_irq); + irq_cpu_data[i].last_irq = NULL; } return 0; } @@ -691,7 +687,7 @@ failed: int __init irqbalance_disable(char *str) { irqbalance_disabled = 1; - return 0; + return 1; } __setup("noirqbalance", irqbalance_disable); @@ -1813,7 +1809,8 @@ static void __init setup_ioapic_ids_from_mpc(void) * Don't check I/O APIC IDs for xAPIC systems. They have * no meaning without the serial APIC bus. */ - if (!(boot_cpu_data.x86_vendor == X86_VENDOR_INTEL && boot_cpu_data.x86 < 15)) + if (!(boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) + || APIC_XAPIC(apic_version[boot_cpu_physical_apicid])) return; /* * This is broken; anything with a real cpu count has to @@ -2298,6 +2295,8 @@ static inline void unlock_ExtINT_logic(void) spin_unlock_irqrestore(&ioapic_lock, flags); } +int timer_uses_ioapic_pin_0; + /* * This code may look a bit paranoid, but it's supposed to cooperate with * a wide range of boards and BIOS bugs. Fortunately only the timer IRQ @@ -2334,6 +2333,9 @@ static inline void check_timer(void) pin2 = ioapic_i8259.pin; apic2 = ioapic_i8259.apic; + if (pin1 == 0) + timer_uses_ioapic_pin_0 = 1; + printk(KERN_INFO "..TIMER: vector=0x%02X apic1=%d pin1=%d apic2=%d pin2=%d\n", vector, apic1, pin1, apic2, pin2); @@ -2419,6 +2421,7 @@ static inline void check_timer(void) "report. Then try booting with the 'noapic' option"); } #else +int timer_uses_ioapic_pin_0 = 0; #define check_timer() ((void)0) #endif diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/microcode-xen.c b/linux-2.6-xen-sparse/arch/i386/kernel/microcode-xen.c index 926ba175c3..c77a40b8fd 100644 --- a/linux-2.6-xen-sparse/arch/i386/kernel/microcode-xen.c +++ b/linux-2.6-xen-sparse/arch/i386/kernel/microcode-xen.c @@ -32,6 +32,7 @@ #include <linux/miscdevice.h> #include <linux/spinlock.h> #include <linux/mm.h> +#include <linux/mutex.h> #include <linux/syscalls.h> #include <asm/msr.h> @@ -49,7 +50,7 @@ MODULE_LICENSE("GPL"); #define DEFAULT_UCODE_TOTALSIZE (DEFAULT_UCODE_DATASIZE + MC_HEADER_SIZE) /* 2048 bytes */ /* no concurrent ->write()s are allowed on /dev/cpu/microcode */ -static DECLARE_MUTEX(microcode_sem); +static DEFINE_MUTEX(microcode_mutex); static int microcode_open (struct inode *unused1, struct file *unused2) { @@ -90,37 +91,20 @@ static ssize_t microcode_write (struct file *file, const char __user *buf, size_ return -EINVAL; } - down(µcode_sem); + mutex_lock(µcode_mutex); ret = do_microcode_update(buf, len); if (!ret) ret = (ssize_t)len; - up(µcode_sem); + mutex_unlock(µcode_mutex); return ret; } -static int microcode_ioctl (struct inode *inode, struct file *file, - unsigned int cmd, unsigned long arg) -{ - switch (cmd) { - /* - * XXX: will be removed after microcode_ctl - * is updated to ignore failure of this ioctl() - */ - case MICROCODE_IOCFREE: - return 0; - default: - return -EINVAL; - } - return -EINVAL; -} - static struct file_operations microcode_fops = { .owner = THIS_MODULE, .write = microcode_write, - .ioctl = microcode_ioctl, .open = microcode_open, }; @@ -151,7 +135,6 @@ static int __init microcode_init (void) static void __exit microcode_exit (void) { misc_deregister(µcode_dev); - printk(KERN_INFO "IA-32 Microcode Update Driver v" MICROCODE_VERSION " unregistered\n"); } module_init(microcode_init) diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/mpparse-xen.c b/linux-2.6-xen-sparse/arch/i386/kernel/mpparse-xen.c index f5daedb2df..48e0bc0f8a 100644 --- a/linux-2.6-xen-sparse/arch/i386/kernel/mpparse-xen.c +++ b/linux-2.6-xen-sparse/arch/i386/kernel/mpparse-xen.c @@ -38,12 +38,6 @@ int smp_found_config; unsigned int __initdata maxcpus = NR_CPUS; -#ifdef CONFIG_HOTPLUG_CPU -#define CPU_HOTPLUG_ENABLED (1) -#else -#define CPU_HOTPLUG_ENABLED (0) -#endif - /* * Various Linux-internal data structures created from the * MP-table. @@ -110,21 +104,6 @@ static int __init mpf_checksum(unsigned char *mp, int len) static int mpc_record; static struct mpc_config_translation *translation_table[MAX_MPC_ENTRY] __initdata; -#ifdef CONFIG_X86_NUMAQ -static int MP_valid_apicid(int apicid, int version) -{ - return hweight_long(apicid & 0xf) == 1 && (apicid >> 4) != 0xf; -} -#elif !defined(CONFIG_XEN) -static int MP_valid_apicid(int apicid, int version) -{ - if (version >= 0x14) - return apicid < 0xff; - else - return apicid < 0xf; -} -#endif - #ifndef CONFIG_XEN static void __devinit MP_processor_info (struct mpc_config_processor *m) { @@ -191,12 +170,6 @@ static void __devinit MP_processor_info (struct mpc_config_processor *m) ver = m->mpc_apicver; - if (!MP_valid_apicid(apicid, ver)) { - printk(KERN_WARNING "Processor #%d INVALID. (Max ID: %d).\n", - m->mpc_apicid, MAX_APICS); - return; - } - /* * Validate version */ @@ -226,7 +199,14 @@ static void __devinit MP_processor_info (struct mpc_config_processor *m) cpu_set(num_processors, cpu_possible_map); num_processors++; - if (CPU_HOTPLUG_ENABLED || (num_processors > 8)) { + /* + * Would be preferable to switch to bigsmp when CONFIG_HOTPLUG_CPU=y + * but we need to work other dependencies like SMP_SUSPEND etc + * before this can be done without some confusion. + * if (CPU_HOTPLUG_ENABLED || num_processors > 8) + * - Ashok Raj <ashok.raj@intel.com> + */ + if (num_processors > 8) { switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_INTEL: if (!APIC_XAPIC(ver)) { @@ -256,6 +236,13 @@ static void __init MP_bus_info (struct mpc_config_bus *m) mpc_oem_bus_info(m, str, translation_table[mpc_record]); + if (m->mpc_busid >= MAX_MP_BUSSES) { + printk(KERN_WARNING "MP table busid value (%d) for bustype %s " + " is too large, max. supported is %d\n", + m->mpc_busid, str, MAX_MP_BUSSES - 1); + return; + } + if (strncmp(str, BUSTYPE_ISA, sizeof(BUSTYPE_ISA)-1) == 0) { mp_bus_id_to_type[m->mpc_busid] = MP_BUS_ISA; } else if (strncmp(str, BUSTYPE_EISA, sizeof(BUSTYPE_EISA)-1) == 0) { @@ -844,6 +831,8 @@ void __init find_smp_config (void) #endif } +int es7000_plat; + /* -------------------------------------------------------------------------- ACPI-based MP Configuration -------------------------------------------------------------------------- */ @@ -957,7 +946,8 @@ void __init mp_register_ioapic ( #ifndef CONFIG_XEN set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address); #endif - if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) && (boot_cpu_data.x86 < 15)) + if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) + && !APIC_XAPIC(apic_version[boot_cpu_physical_apicid])) tmpid = io_apic_get_unique_id(idx, id); else tmpid = id; @@ -1033,8 +1023,6 @@ void __init mp_override_legacy_irq ( return; } -int es7000_plat; - void __init mp_config_acpi_legacy_irqs (void) { struct mpc_config_intsrc intsrc; @@ -1164,7 +1152,17 @@ int mp_register_gsi (u32 gsi, int triggering, int polarity) */ int irq = gsi; if (gsi < MAX_GSI_NUM) { - if (gsi > 15) + /* + * Retain the VIA chipset work-around (gsi > 15), but + * avoid a problem where the 8254 timer (IRQ0) is setup + * via an override (so it's not on pin 0 of the ioapic), + * and at the same time, the pin 0 interrupt is a PCI + * type. The gsi > 15 test could cause these two pins + * to be shared as IRQ0, and they are not shareable. + * So test for this condition, and if necessary, avoid + * the pin collision. + */ + if (gsi > 15 || (gsi == 0 && !timer_uses_ioapic_pin_0)) gsi = pci_irq++; /* * Don't assign IRQ used by ACPI SCI diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/process-xen.c b/linux-2.6-xen-sparse/arch/i386/kernel/process-xen.c index 0dbc6a412e..b64dec23ec 100644 --- a/linux-2.6-xen-sparse/arch/i386/kernel/process-xen.c +++ b/linux-2.6-xen-sparse/arch/i386/kernel/process-xen.c @@ -38,7 +38,6 @@ #include <linux/kallsyms.h> #include <linux/ptrace.h> #include <linux/random.h> -#include <linux/kprobes.h> #include <asm/uaccess.h> #include <asm/pgtable.h> @@ -245,7 +244,7 @@ void show_regs(struct pt_regs * regs) printk("EIP: %04x:[<%08lx>] CPU: %d\n",0xffff & regs->xcs,regs->eip, smp_processor_id()); print_symbol("EIP is at %s\n", regs->eip); - if (user_mode(regs)) + if (user_mode_vm(regs)) printk(" ESP: %04x:%08lx",0xffff & regs->xss,regs->esp); printk(" EFLAGS: %08lx %s (%s %.*s)\n", regs->eflags, print_tainted(), system_utsname.release, @@ -314,13 +313,6 @@ void exit_thread(void) struct task_struct *tsk = current; struct thread_struct *t = &tsk->thread; - /* - * Remove function-return probe instances associated with this task - * and put them back on the free list. Do not insert an exit probe for - * this function, it will be disabled by kprobe_flush_task if you do. - */ - kprobe_flush_task(tsk); - /* The process may have allocated an io port bitmap... nuke it. */ if (unlikely(NULL != t->io_bitmap_ptr)) { struct physdev_set_iobitmap set_iobitmap = { 0 }; @@ -731,7 +723,6 @@ unsigned long get_wchan(struct task_struct *p) } while (count++ < 16); return 0; } -EXPORT_SYMBOL(get_wchan); /* * sys_alloc_thread_area: get a yet unused TLS descriptor index. diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/setup-xen.c b/linux-2.6-xen-sparse/arch/i386/kernel/setup-xen.c index 819f36f11e..63d993b58f 100644 --- a/linux-2.6-xen-sparse/arch/i386/kernel/setup-xen.c +++ b/linux-2.6-xen-sparse/arch/i386/kernel/setup-xen.c @@ -34,6 +34,7 @@ #include <linux/initrd.h> #include <linux/bootmem.h> #include <linux/seq_file.h> +#include <linux/platform_device.h> #include <linux/console.h> #include <linux/mca.h> #include <linux/root_dev.h> @@ -49,6 +50,7 @@ #include <linux/kexec.h> #include <linux/crash_dump.h> #include <linux/dmi.h> +#include <linux/pfn.h> #include <video/edid.h> @@ -1027,6 +1029,38 @@ efi_memory_present_wrapper(unsigned long start, unsigned long end, void *arg) return 0; } + /* + * This function checks if the entire range <start,end> is mapped with type. + * + * Note: this function only works correct if the e820 table is sorted and + * not-overlapping, which is the case + */ +int __init +e820_all_mapped(unsigned long s, unsigned long e, unsigned type) +{ + u64 start = s; + u64 end = e; + int i; + for (i = 0; i < e820.nr_map; i++) { + struct e820entry *ei = &e820.map[i]; + if (type && ei->type != type) + continue; + /* is the region (part) in overlap with the current region ?*/ + if (ei->addr >= end || ei->addr + ei->size <= start) + continue; + /* if the region is at the beginning of <start,end> we move + * start to the end of the region since it's ok until there + */ + if (ei->addr <= start) + start = ei->addr + ei->size; + /* if start is now at or beyond end, we're done, full + * coverage */ + if (start >= end) + return 1; /* we're done */ + } + return 0; +} + /* * Find the highest page frame number we have available */ @@ -1124,10 +1158,10 @@ static int __init free_available_memory(unsigned long start, unsigned long end, void *arg) { /* check max_low_pfn */ - if (start >= ((max_low_pfn + 1) << PAGE_SHIFT)) + if (start >= (max_low_pfn << PAGE_SHIFT)) return 0; - if (end >= ((max_low_pfn + 1) << PAGE_SHIFT)) - end = (max_low_pfn + 1) << PAGE_SHIFT; + if (end >= (max_low_pfn << PAGE_SHIFT)) + end = max_low_pfn << PAGE_SHIFT; if (start < end) free_bootmem(start, end - start); @@ -1380,7 +1414,7 @@ legacy_init_iomem_resources(struct e820entry *e820, int nr_map, struct resource *res; if (e820[i].addr + e820[i].size > 0x100000000ULL) continue; - res = alloc_bootmem_low(sizeof(struct resource)); + res = kzalloc(sizeof(struct resource), GFP_ATOMIC); switch (e820[i].type) { case E820_RAM: res->name = "System RAM"; break; case E820_ACPI: res->name = "ACPI Tables"; break; @@ -1467,8 +1501,11 @@ e820_setup_gap(struct e820entry *e820, int nr_map) /* * Request address space for all standard resources + * + * This is called just before pcibios_init(), which is also a + * subsys_initcall, but is linked in later (in arch/i386/pci/common.c). */ -static void __init register_memory(void) +static int __init request_standard_resources(void) { #ifdef CONFIG_XEN struct xen_memory_map memmap; @@ -1477,8 +1514,9 @@ static void __init register_memory(void) /* Nothing to do if not running in dom0. */ if (!is_initial_xendomain()) - return; + return 0; + printk("Setting up standard PCI resources\n"); #ifdef CONFIG_XEN memmap.nr_entries = E820MAX; set_xen_guest_handle(memmap.buffer, machine_e820.map); @@ -1503,6 +1541,13 @@ static void __init register_memory(void) /* request I/O space for devices used on all i[345]86 PCs */ for (i = 0; i < STANDARD_IO_RESOURCES; i++) request_resource(&ioport_resource, &standard_io_resources[i]); + return 0; +} + +subsys_initcall(request_standard_resources); + +static void __init register_memory(void) +{ #ifdef CONFIG_XEN e820_setup_gap(machine_e820.map, machine_e820.nr_map); @@ -1511,101 +1556,6 @@ static void __init register_memory(void) #endif } -/* Use inline assembly to define this because the nops are defined - as inline assembly strings in the include files and we cannot - get them easily into strings. */ -asm("\t.data\nintelnops: " - GENERIC_NOP1 GENERIC_NOP2 GENERIC_NOP3 GENERIC_NOP4 GENERIC_NOP5 GENERIC_NOP6 - GENERIC_NOP7 GENERIC_NOP8); -asm("\t.data\nk8nops: " - K8_NOP1 K8_NOP2 K8_NOP3 K8_NOP4 K8_NOP5 K8_NOP6 - K8_NOP7 K8_NOP8); -asm("\t.data\nk7nops: " - K7_NOP1 K7_NOP2 K7_NOP3 K7_NOP4 K7_NOP5 K7_NOP6 - K7_NOP7 K7_NOP8); - -extern unsigned char intelnops[], k8nops[], k7nops[]; -static unsigned char *intel_nops[ASM_NOP_MAX+1] = { - NULL, - intelnops, - intelnops + 1, - intelnops + 1 + 2, - intelnops + 1 + 2 + 3, - intelnops + 1 + 2 + 3 + 4, - intelnops + 1 + 2 + 3 + 4 + 5, - intelnops + 1 + 2 + 3 + 4 + 5 + 6, - intelnops + 1 + 2 + 3 + 4 + 5 + 6 + 7, -}; -static unsigned char *k8_nops[ASM_NOP_MAX+1] = { - NULL, - k8nops, - k8nops + 1, - k8nops + 1 + 2, - k8nops + 1 + 2 + 3, - k8nops + 1 + 2 + 3 + 4, - k8nops + 1 + 2 + 3 + 4 + 5, - k8nops + 1 + 2 + 3 + 4 + 5 + 6, - k8nops + 1 + 2 + 3 + 4 + 5 + 6 + 7, -}; -static unsigned char *k7_nops[ASM_NOP_MAX+1] = { - NULL, - k7nops, - k7nops + 1, - k7nops + 1 + 2, - k7nops + 1 + 2 + 3, - k7nops + 1 + 2 + 3 + 4, - k7nops + 1 + 2 + 3 + 4 + 5, - k7nops + 1 + 2 + 3 + 4 + 5 + 6, - k7nops + 1 + 2 + 3 + 4 + 5 + 6 + 7, -}; -static struct nop { - int cpuid; - unsigned char **noptable; -} noptypes[] = { - { X86_FEATURE_K8, k8_nops }, - { X86_FEATURE_K7, k7_nops }, - { -1, NULL } -}; - -/* Replace instructions with better alternatives for this CPU type. - - This runs before SMP is initialized to avoid SMP problems with - self modifying code. This implies that assymetric systems where - APs have less capabilities than the boot processor are not handled. - Tough. Make sure you disable such features by hand. */ -void apply_alternatives(void *start, void *end) -{ - struct alt_instr *a; - int diff, i, k; - unsigned char **noptable = intel_nops; - for (i = 0; noptypes[i].cpuid >= 0; i++) { - if (boot_cpu_has(noptypes[i].cpuid)) { - noptable = noptypes[i].noptable; - break; - } - } - for (a = start; (void *)a < end; a++) { - if (!boot_cpu_has(a->cpuid)) - continue; - BUG_ON(a->replacementlen > a->instrlen); - memcpy(a->instr, a->replacement, a->replacementlen); - diff = a->instrlen - a->replacementlen; - /* Pad the rest with nops */ - for (i = a->replacementlen; diff > 0; diff -= k, i += k) { - k = diff; - if (k > ASM_NOP_MAX) - k = ASM_NOP_MAX; - memcpy(a->instr + i, noptable[k], k); - } - } -} - -void __init alternative_instructions(void) -{ - extern struct alt_instr __alt_instructions[], __alt_instructions_end[]; - apply_alternatives(__alt_instructions, __alt_instructions_end); -} - static char * __init machine_specific_memory_setup(void); #ifdef CONFIG_MCA @@ -1636,7 +1586,7 @@ void __init setup_arch(char **cmdline_p) panic_timeout = 1; /* Register a call for panic conditions. */ - notifier_chain_register(&panic_notifier_list, &xen_panic_block); + atomic_notifier_chain_register(&panic_notifier_list, &xen_panic_block); HYPERVISOR_vm_assist(VMASST_CMD_enable, VMASST_TYPE_4gb_segments); HYPERVISOR_vm_assist(VMASST_CMD_enable, @@ -1644,6 +1594,9 @@ void __init setup_arch(char **cmdline_p) memcpy(&boot_cpu_data, &new_cpu_data, sizeof(new_cpu_data)); early_cpu_init(); +#ifdef CONFIG_SMP + prefill_possible_map(); +#endif /* * FIXME: This isn't an official loader_type right @@ -1731,6 +1684,16 @@ void __init setup_arch(char **cmdline_p) parse_cmdline_early(cmdline_p); +#ifdef CONFIG_EARLY_PRINTK + { + char *s = strstr(*cmdline_p, "earlyprintk="); + if (s) { + setup_early_printk(strchr(s, '=') + 1); + printk("early console enabled\n"); + } + } +#endif + max_low_pfn = setup_memory(); /* @@ -1801,18 +1764,6 @@ void __init setup_arch(char **cmdline_p) * NOTE: at this point the bootmem allocator is fully available. */ -#ifdef CONFIG_EARLY_PRINTK - { - char *s = strstr(*cmdline_p, "earlyprintk="); - if (s) { - extern void setup_early_printk(char *); - - setup_early_printk(strchr(s, '=') + 1); - printk("early console enabled\n"); - } - } -#endif - if (is_initial_xendomain()) dmi_scan_machine(); @@ -1825,10 +1776,6 @@ void __init setup_arch(char **cmdline_p) set_iopl.iopl = 1; HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl); -#ifdef CONFIG_X86_IO_APIC - check_acpi_pci(); /* Checks more than just ACPI actually */ -#endif - #ifdef CONFIG_ACPI if (!is_initial_xendomain()) { printk(KERN_INFO "ACPI in unprivileged domain disabled\n"); @@ -1840,6 +1787,13 @@ void __init setup_arch(char **cmdline_p) * Parse the ACPI tables for possible boot-time SMP configuration. */ acpi_boot_table_init(); +#endif + +#ifdef CONFIG_X86_IO_APIC + check_acpi_pci(); /* Checks more than just ACPI actually */ +#endif + +#ifdef CONFIG_ACPI acpi_boot_init(); #if defined(CONFIG_SMP) && defined(CONFIG_X86_PC) @@ -1882,6 +1836,23 @@ xen_panic_event(struct notifier_block *this, unsigned long event, void *ptr) return NOTIFY_DONE; } +static __init int add_pcspkr(void) +{ + struct platform_device *pd; + int ret; + + pd = platform_device_alloc("pcspkr", -1); + if (!pd) + return -ENOMEM; + + ret = platform_device_add(pd); + if (ret) + platform_device_put(pd); + + return ret; +} +device_initcall(add_pcspkr); + #include "setup_arch_post.h" /* * Local Variables: diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/smp-xen.c b/linux-2.6-xen-sparse/arch/i386/kernel/smp-xen.c index c6aaa8b822..ea58808eac 100644 --- a/linux-2.6-xen-sparse/arch/i386/kernel/smp-xen.c +++ b/linux-2.6-xen-sparse/arch/i386/kernel/smp-xen.c @@ -483,27 +483,23 @@ void unlock_ipi_call_lock(void) spin_unlock_irq(&call_lock); } -static struct call_data_struct * call_data; - -/* - * this function sends a 'generic call function' IPI to all other CPUs - * in the system. - */ - -int smp_call_function (void (*func) (void *info), void *info, int nonatomic, - int wait) -/* - * [SUMMARY] Run a function on all other CPUs. - * <func> The function to run. This must be fast and non-blocking. - * <info> An arbitrary pointer to pass to the function. - * <nonatomic> currently unused. - * <wait> If true, wait (atomically) until function has completed on other CPUs. - * [RETURNS] 0 on success, else a negative status code. Does not return until +static struct call_data_struct *call_data; + +/** + * smp_call_function(): Run a function on all other CPUs. + * @func: The function to run. This must be fast and non-blocking. + * @info: An arbitrary pointer to pass to the function. + * @nonatomic: currently unused. + * @wait: If true, wait (atomically) until function has completed on other CPUs. + * + * Returns 0 on success, else a negative status code. Does not return until * remote CPUs are nearly ready to execute <<func>> or are or have executed. * * You must not call this function with disabled interrupts or from a * hardware interrupt handler or from a bottom half handler. */ +int smp_call_function (void (*func) (void *info), void *info, int nonatomic, + int wait) { struct call_data_struct data; int cpus; diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/smpalts.c b/linux-2.6-xen-sparse/arch/i386/kernel/smpalts.c deleted file mode 100644 index 5a32e547e4..0000000000 --- a/linux-2.6-xen-sparse/arch/i386/kernel/smpalts.c +++ /dev/null @@ -1,85 +0,0 @@ -#include <linux/kernel.h> -#include <asm/system.h> -#include <asm/smp_alt.h> -#include <asm/processor.h> -#include <asm/string.h> - -struct smp_replacement_record { - unsigned char targ_size; - unsigned char smp1_size; - unsigned char smp2_size; - unsigned char up_size; - unsigned char feature; - unsigned char data[0]; -}; - -struct smp_alternative_record { - void *targ_start; - struct smp_replacement_record *repl; -}; - -extern struct smp_alternative_record __start_smp_alternatives_table, - __stop_smp_alternatives_table; -extern unsigned long __init_begin, __init_end; - -void prepare_for_smp(void) -{ - struct smp_alternative_record *r; - printk(KERN_INFO "Enabling SMP...\n"); - for (r = &__start_smp_alternatives_table; - r != &__stop_smp_alternatives_table; - r++) { - BUG_ON(r->repl->targ_size < r->repl->smp1_size); - BUG_ON(r->repl->targ_size < r->repl->smp2_size); - BUG_ON(r->repl->targ_size < r->repl->up_size); - if (system_state == SYSTEM_RUNNING && - r->targ_start >= (void *)&__init_begin && - r->targ_start < (void *)&__init_end) - continue; - if (r->repl->feature != (unsigned char)-1 && - boot_cpu_has(r->repl->feature)) { - memcpy(r->targ_start, - r->repl->data + r->repl->smp1_size, - r->repl->smp2_size); - memset(r->targ_start + r->repl->smp2_size, - 0x90, - r->repl->targ_size - r->repl->smp2_size); - } else { - memcpy(r->targ_start, - r->repl->data, - r->repl->smp1_size); - memset(r->targ_start + r->repl->smp1_size, - 0x90, - r->repl->targ_size - r->repl->smp1_size); - } - } - /* Paranoia */ - asm volatile ("jmp 1f\n1:"); - mb(); -} - -void unprepare_for_smp(void) -{ - struct smp_alternative_record *r; - printk(KERN_INFO "Disabling SMP...\n"); - for (r = &__start_smp_alternatives_table; - r != &__stop_smp_alternatives_table; - r++) { - BUG_ON(r->repl->targ_size < r->repl->smp1_size); - BUG_ON(r->repl->targ_size < r->repl->smp2_size); - BUG_ON(r->repl->targ_size < r->repl->up_size); - if (system_state == SYSTEM_RUNNING && - r->targ_start >= (void *)&__init_begin && - r->targ_start < (void *)&__init_end) - continue; - memcpy(r->targ_start, - r->repl->data + r->repl->smp1_size + r->repl->smp2_size, - r->repl->up_size); - memset(r->targ_start + r->repl->up_size, - 0x90, - r->repl->targ_size - r->repl->up_size); - } - /* Paranoia */ - asm volatile ("jmp 1f\n1:"); - mb(); -} diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/smpboot.c b/linux-2.6-xen-sparse/arch/i386/kernel/smpboot.c index 1008255c62..825b2b4ca7 100644 --- a/linux-2.6-xen-sparse/arch/i386/kernel/smpboot.c +++ b/linux-2.6-xen-sparse/arch/i386/kernel/smpboot.c @@ -72,6 +72,9 @@ int phys_proc_id[NR_CPUS] __read_mostly = {[0 ... NR_CPUS-1] = BAD_APICID}; /* Core ID of each logical CPU */ int cpu_core_id[NR_CPUS] __read_mostly = {[0 ... NR_CPUS-1] = BAD_APICID}; +/* Last level cache ID of each logical CPU */ +int cpu_llc_id[NR_CPUS] __cpuinitdata = {[0 ... NR_CPUS-1] = BAD_APICID}; + /* representing HT siblings of each logical CPU */ cpumask_t cpu_sibling_map[NR_CPUS] __read_mostly; EXPORT_SYMBOL(cpu_sibling_map); @@ -310,7 +313,9 @@ static void __init synchronize_tsc_bp (void) if (tsc_values[i] < avg) realdelta = -realdelta; - printk(KERN_INFO "CPU#%d had %ld usecs TSC skew, fixed it up.\n", i, realdelta); + if (realdelta > 0) + printk(KERN_INFO "CPU#%d had %ld usecs TSC " + "skew, fixed it up.\n", i, realdelta); } sum += delta; @@ -440,6 +445,18 @@ static void __devinit smp_callin(void) static int cpucount; +/* maps the cpu to the sched domain representing multi-core */ +cpumask_t cpu_coregroup_map(int cpu) +{ + struct cpuinfo_x86 *c = cpu_data + cpu; + /* + * For perf, we return last level cache shared map. + * TBD: when power saving sched policy is added, we will return + * cpu_core_map when power saving policy is enabled + */ + return c->llc_shared_map; +} + /* representing cpus for which sibling maps can be computed */ static cpumask_t cpu_sibling_setup_map; @@ -459,12 +476,16 @@ set_cpu_sibling_map(int cpu) cpu_set(cpu, cpu_sibling_map[i]); cpu_set(i, cpu_core_map[cpu]); cpu_set(cpu, cpu_core_map[i]); + cpu_set(i, c[cpu].llc_shared_map); + cpu_set(cpu, c[i].llc_shared_map); } } } else { cpu_set(cpu, cpu_sibling_map[cpu]); } + cpu_set(cpu, c[cpu].llc_shared_map); + if (current_cpu_data.x86_max_cores == 1) { cpu_core_map[cpu] = cpu_sibling_map[cpu]; c[cpu].booted_cores = 1; @@ -472,6 +493,11 @@ set_cpu_sibling_map(int cpu) } for_each_cpu_mask(i, cpu_sibling_setup_map) { + if (cpu_llc_id[cpu] != BAD_APICID && + cpu_llc_id[cpu] == cpu_llc_id[i]) { + cpu_set(i, c[cpu].llc_shared_map); + cpu_set(cpu, c[i].llc_shared_map); + } if (phys_proc_id[cpu] == phys_proc_id[i]) { cpu_set(i, cpu_core_map[cpu]); cpu_set(cpu, cpu_core_map[i]); @@ -899,6 +925,7 @@ static int __devinit do_boot_cpu(int apicid, int cpu) unsigned short nmi_high = 0, nmi_low = 0; ++cpucount; + alternatives_smp_switch(1); /* * We can't use kernel_thread since we must avoid to @@ -1002,7 +1029,6 @@ void cpu_exit_clear(void) cpu_clear(cpu, cpu_callout_map); cpu_clear(cpu, cpu_callin_map); - cpu_clear(cpu, cpu_present_map); cpu_clear(cpu, smp_commenced_mask); unmap_cpu_to_logical_apicid(cpu); @@ -1014,31 +1040,20 @@ struct warm_boot_cpu_info { int cpu; }; -static void __devinit do_warm_boot_cpu(void *p) +static void __cpuinit do_warm_boot_cpu(void *p) { struct warm_boot_cpu_info *info = p; do_boot_cpu(info->apicid, info->cpu); complete(info->complete); } -int __devinit smp_prepare_cpu(int cpu) +static int __cpuinit __smp_prepare_cpu(int cpu) { DECLARE_COMPLETION(done); struct warm_boot_cpu_info info; struct work_struct task; int apicid, ret; - lock_cpu_hotplug(); - - /* - * On x86, CPU0 is never offlined. Trying to bring up an - * already-booted CPU will hang. So check for that case. - */ - if (cpu_online(cpu)) { - ret = -EINVAL; - goto exit; - } - apicid = x86_cpu_to_apicid[cpu]; if (apicid == BAD_APICID) { ret = -ENODEV; @@ -1063,7 +1078,6 @@ int __devinit smp_prepare_cpu(int cpu) zap_low_mappings(); ret = 0; exit: - unlock_cpu_hotplug(); return ret; } #endif @@ -1218,11 +1232,6 @@ static void __init smp_boot_cpus(unsigned int max_cpus) if (max_cpus <= cpucount+1) continue; -#ifdef CONFIG_SMP_ALTERNATIVES - if (kicked == 1) - prepare_for_smp(); -#endif - if (((cpu = alloc_cpu_id()) <= 0) || do_boot_cpu(apicid, cpu)) printk("CPU #%d not responding - cannot use it.\n", apicid); @@ -1373,6 +1382,8 @@ void __cpu_die(unsigned int cpu) /* They ack this in play_dead by setting CPU_DEAD */ if (per_cpu(cpu_state, cpu) == CPU_DEAD) { printk ("CPU %d is now offline\n", cpu); + if (1 == num_online_cpus()) + alternatives_smp_switch(0); return; } msleep(100); @@ -1394,6 +1405,22 @@ void __cpu_die(unsigned int cpu) int __devinit __cpu_up(unsigned int cpu) { +#ifdef CONFIG_HOTPLUG_CPU + int ret=0; + + /* + * We do warm boot only on cpus that had booted earlier + * Otherwise cold boot is all handled from smp_boot_cpus(). + * cpu_callin_map is set during AP kickstart process. Its reset + * when a cpu is taken offline from cpu_exit_clear(). + */ + if (!cpu_isset(cpu, cpu_callin_map)) + ret = __smp_prepare_cpu(cpu); + + if (ret) + return -EIO; +#endif + /* In case one didn't come up */ if (!cpu_isset(cpu, cpu_callin_map)) { printk(KERN_DEBUG "skipping cpu%d, didn't come online\n", cpu); @@ -1401,11 +1428,6 @@ int __devinit __cpu_up(unsigned int cpu) return -EIO; } -#ifdef CONFIG_SMP_ALTERNATIVES - if (num_online_cpus() == 1) - prepare_for_smp(); -#endif - local_irq_enable(); per_cpu(cpu_state, cpu) = CPU_UP_PREPARE; /* Unleash the CPU! */ diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/traps-xen.c b/linux-2.6-xen-sparse/arch/i386/kernel/traps-xen.c index 1c4a42709d..be8bc18c5e 100644 --- a/linux-2.6-xen-sparse/arch/i386/kernel/traps-xen.c +++ b/linux-2.6-xen-sparse/arch/i386/kernel/traps-xen.c @@ -98,32 +98,50 @@ asmlinkage void fixup_4gb_segment(void); asmlinkage void machine_check(void); static int kstack_depth_to_print = 24; -struct notifier_block *i386die_chain; -static DEFINE_SPINLOCK(die_notifier_lock); +ATOMIC_NOTIFIER_HEAD(i386die_chain); int register_die_notifier(struct notifier_block *nb) { - int err = 0; - unsigned long flags; - spin_lock_irqsave(&die_notifier_lock, flags); - err = notifier_chain_register(&i386die_chain, nb); - spin_unlock_irqrestore(&die_notifier_lock, flags); - return err; + vmalloc_sync_all(); + return atomic_notifier_chain_register(&i386die_chain, nb); } EXPORT_SYMBOL(register_die_notifier); +int unregister_die_notifier(struct notifier_block *nb) +{ + return atomic_notifier_chain_unregister(&i386die_chain, nb); +} +EXPORT_SYMBOL(unregister_die_notifier); + static inline int valid_stack_ptr(struct thread_info *tinfo, void *p) { return p > (void *)tinfo && p < (void *)tinfo + THREAD_SIZE - 3; } -static void print_addr_and_symbol(unsigned long addr, char *log_lvl) +/* + * Print CONFIG_STACK_BACKTRACE_COLS address/symbol entries per line. + */ +static inline int print_addr_and_symbol(unsigned long addr, char *log_lvl, + int printed) { - printk(log_lvl); + if (!printed) + printk(log_lvl); + +#if CONFIG_STACK_BACKTRACE_COLS == 1 printk(" [<%08lx>] ", addr); +#else + printk(" <%08lx> ", addr); +#endif print_symbol("%s", addr); - printk("\n"); + + printed = (printed + 1) % CONFIG_STACK_BACKTRACE_COLS; + if (printed) + printk(" "); + else + printk("\n"); + + return printed; } static inline unsigned long print_context_stack(struct thread_info *tinfo, @@ -131,20 +149,24 @@ static inline unsigned long print_context_stack(struct thread_info *tinfo, char *log_lvl) { unsigned long addr; + int printed = 0; /* nr of entries already printed on current line */ #ifdef CONFIG_FRAME_POINTER while (valid_stack_ptr(tinfo, (void *)ebp)) { addr = *(unsigned long *)(ebp + 4); - print_addr_and_symbol(addr, log_lvl); + printed = print_addr_and_symbol(addr, log_lvl, printed); ebp = *(unsigned long *)ebp; } #else while (valid_stack_ptr(tinfo, stack)) { addr = *stack++; if (__kernel_text_address(addr)) - print_addr_and_symbol(addr, log_lvl); + printed = print_addr_and_symbol(addr, log_lvl, printed); } #endif + if (printed) + printk("\n"); + return ebp; } @@ -172,8 +194,7 @@ static void show_trace_log_lvl(struct task_struct *task, stack = (unsigned long*)context->previous_esp; if (!stack) break; - printk(log_lvl); - printk(" =======================\n"); + printk("%s =======================\n", log_lvl); } } @@ -196,25 +217,20 @@ static void show_stack_log_lvl(struct task_struct *task, unsigned long *esp, } stack = esp; - printk(log_lvl); for(i = 0; i < kstack_depth_to_print; i++) { if (kstack_end(stack)) break; - if (i && ((i % 8) == 0)) { - printk("\n"); - printk(log_lvl); - printk(" "); - } + if (i && ((i % 8) == 0)) + printk("\n%s ", log_lvl); printk("%08lx ", *stack++); } - printk("\n"); - printk(log_lvl); - printk("Call Trace:\n"); + printk("\n%sCall Trace:\n", log_lvl); show_trace_log_lvl(task, esp, log_lvl); } void show_stack(struct task_struct *task, unsigned long *esp) { + printk(" "); show_stack_log_lvl(task, esp, ""); } @@ -239,7 +255,7 @@ void show_registers(struct pt_regs *regs) esp = (unsigned long) (®s->esp); savesegment(ss, ss); - if (user_mode(regs)) { + if (user_mode_vm(regs)) { in_kernel = 0; esp = regs->esp; ss = regs->xss & 0xffff; @@ -339,6 +355,8 @@ void die(const char * str, struct pt_regs * regs, long err) static int die_counter; unsigned long flags; + oops_enter(); + if (die.lock_owner != raw_smp_processor_id()) { console_verbose(); spin_lock_irqsave(&die.lock, flags); @@ -351,6 +369,9 @@ void die(const char * str, struct pt_regs * regs, long err) if (++die.lock_owner_depth < 3) { int nl = 0; + unsigned long esp; + unsigned short ss; + handle_BUG(regs); printk(KERN_EMERG "%s: %04lx [#%d]\n", str, err & 0xffff, ++die_counter); #ifdef CONFIG_PREEMPT @@ -371,8 +392,23 @@ void die(const char * str, struct pt_regs * regs, long err) #endif if (nl) printk("\n"); - notify_die(DIE_OOPS, (char *)str, regs, err, 255, SIGSEGV); - show_registers(regs); + if (notify_die(DIE_OOPS, str, regs, err, + current->thread.trap_no, SIGSEGV) != + NOTIFY_STOP) { + show_registers(regs); + /* Executive summary in case the oops scrolled away */ + esp = (unsigned long) (®s->esp); + savesegment(ss, ss); + if (user_mode(regs)) { + esp = regs->esp; + ss = regs->xss & 0xffff; + } + printk(KERN_EMERG "EIP: [<%08lx>] ", regs->eip); + print_symbol("%s", regs->eip); + printk(" SS:ESP %04x:%08lx\n", ss, esp); + } + else + regs = NULL; } else printk(KERN_EMERG "Recursive die() failure, output suppressed\n"); @@ -380,6 +416,9 @@ void die(const char * str, struct pt_regs * regs, long err) die.lock_owner = -1; spin_unlock_irqrestore(&die.lock, flags); + if (!regs) + return; + if (kexec_should_crash(current)) crash_kexec(regs); @@ -391,6 +430,7 @@ void die(const char * str, struct pt_regs * regs, long err) ssleep(5); panic("Fatal exception"); } + oops_exit(); do_exit(SIGSEGV); } @@ -569,7 +609,7 @@ static DEFINE_SPINLOCK(nmi_print_lock); void die_nmi (struct pt_regs *regs, const char *msg) { - if (notify_die(DIE_NMIWATCHDOG, msg, regs, 0, 0, SIGINT) == + if (notify_die(DIE_NMIWATCHDOG, msg, regs, 0, 2, SIGINT) == NOTIFY_STOP) return; @@ -591,7 +631,7 @@ void die_nmi (struct pt_regs *regs, const char *msg) /* If we are in kernel we are probably nested up pretty bad * and might aswell get out now while we still can. */ - if (!user_mode(regs)) { + if (!user_mode_vm(regs)) { current->thread.trap_no = 2; crash_kexec(regs); } @@ -608,7 +648,7 @@ static void default_do_nmi(struct pt_regs * regs) reason = get_nmi_reason(); if (!(reason & 0xc0)) { - if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, reason, 0, SIGINT) + if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, reason, 2, SIGINT) == NOTIFY_STOP) return; #ifdef CONFIG_X86_LOCAL_APIC @@ -624,7 +664,7 @@ static void default_do_nmi(struct pt_regs * regs) unknown_nmi_error(reason, regs); return; } - if (notify_die(DIE_NMI, "nmi", regs, reason, 0, SIGINT) == NOTIFY_STOP) + if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP) return; if (reason & 0x80) mem_parity_error(reason, regs); @@ -662,6 +702,7 @@ fastcall void do_nmi(struct pt_regs * regs, long error_code) void set_nmi_callback(nmi_callback_t callback) { + vmalloc_sync_all(); rcu_assign_pointer(nmi_callback, callback); } EXPORT_SYMBOL_GPL(set_nmi_callback); @@ -1089,6 +1130,6 @@ void smp_trap_init(trap_info_t *trap_ctxt) static int __init kstack_setup(char *s) { kstack_depth_to_print = simple_strtoul(s, NULL, 0); - return 0; + return 1; } __setup("kstack=", kstack_setup); diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/traps.c b/linux-2.6-xen-sparse/arch/i386/kernel/traps.c index 85d4645c1a..3078fd86da 100644 --- a/linux-2.6-xen-sparse/arch/i386/kernel/traps.c +++ b/linux-2.6-xen-sparse/arch/i386/kernel/traps.c @@ -92,32 +92,50 @@ asmlinkage void spurious_interrupt_bug(void); asmlinkage void machine_check(void); static int kstack_depth_to_print = 24; -struct notifier_block *i386die_chain; -static DEFINE_SPINLOCK(die_notifier_lock); +ATOMIC_NOTIFIER_HEAD(i386die_chain); int register_die_notifier(struct notifier_block *nb) { - int err = 0; - unsigned long flags; - spin_lock_irqsave(&die_notifier_lock, flags); - err = notifier_chain_register(&i386die_chain, nb); - spin_unlock_irqrestore(&die_notifier_lock, flags); - return err; + vmalloc_sync_all(); + return atomic_notifier_chain_register(&i386die_chain, nb); } EXPORT_SYMBOL(register_die_notifier); +int unregister_die_notifier(struct notifier_block *nb) +{ + return atomic_notifier_chain_unregister(&i386die_chain, nb); +} +EXPORT_SYMBOL(unregister_die_notifier); + static inline int valid_stack_ptr(struct thread_info *tinfo, void *p) { return p > (void *)tinfo && p < (void *)tinfo + THREAD_SIZE - 3; } -static void print_addr_and_symbol(unsigned long addr, char *log_lvl) +/* + * Print CONFIG_STACK_BACKTRACE_COLS address/symbol entries per line. + */ +static inline int print_addr_and_symbol(unsigned long addr, char *log_lvl, + int printed) { - printk(log_lvl); + if (!printed) + printk(log_lvl); + +#if CONFIG_STACK_BACKTRACE_COLS == 1 printk(" [<%08lx>] ", addr); +#else + printk(" <%08lx> ", addr); +#endif print_symbol("%s", addr); - printk("\n"); + + printed = (printed + 1) % CONFIG_STACK_BACKTRACE_COLS; + if (printed) + printk(" "); + else + printk("\n"); + + return printed; } static inline unsigned long print_context_stack(struct thread_info *tinfo, @@ -125,20 +143,24 @@ static inline unsigned long print_context_stack(struct thread_info *tinfo, char *log_lvl) { unsigned long addr; + int printed = 0; /* nr of entries already printed on current line */ #ifdef CONFIG_FRAME_POINTER while (valid_stack_ptr(tinfo, (void *)ebp)) { addr = *(unsigned long *)(ebp + 4); - print_addr_and_symbol(addr, log_lvl); + printed = print_addr_and_symbol(addr, log_lvl, printed); ebp = *(unsigned long *)ebp; } #else while (valid_stack_ptr(tinfo, stack)) { addr = *stack++; if (__kernel_text_address(addr)) - print_addr_and_symbol(addr, log_lvl); + printed = print_addr_and_symbol(addr, log_lvl, printed); } #endif + if (printed) + printk("\n"); + return ebp; } @@ -166,8 +188,7 @@ static void show_trace_log_lvl(struct task_struct *task, stack = (unsigned long*)context->previous_esp; if (!stack) break; - printk(log_lvl); - printk(" =======================\n"); + printk("%s =======================\n", log_lvl); } } @@ -190,25 +211,20 @@ static void show_stack_log_lvl(struct task_struct *task, unsigned long *esp, } stack = esp; - printk(log_lvl); for(i = 0; i < kstack_depth_to_print; i++) { if (kstack_end(stack)) break; - if (i && ((i % 8) == 0)) { - printk("\n"); - printk(log_lvl); - printk(" "); - } + if (i && ((i % 8) == 0)) + printk("\n%s ", log_lvl); printk("%08lx ", *stack++); } - printk("\n"); - printk(log_lvl); - printk("Call Trace:\n"); + printk("\n%sCall Trace:\n", log_lvl); show_trace_log_lvl(task, esp, log_lvl); } void show_stack(struct task_struct *task, unsigned long *esp) { + printk(" "); show_stack_log_lvl(task, esp, ""); } @@ -233,7 +249,7 @@ void show_registers(struct pt_regs *regs) esp = (unsigned long) (®s->esp); savesegment(ss, ss); - if (user_mode(regs)) { + if (user_mode_vm(regs)) { in_kernel = 0; esp = regs->esp; ss = regs->xss & 0xffff; @@ -333,6 +349,8 @@ void die(const char * str, struct pt_regs * regs, long err) static int die_counter; unsigned long flags; + oops_enter(); + if (die.lock_owner != raw_smp_processor_id()) { console_verbose(); spin_lock_irqsave(&die.lock, flags); @@ -345,6 +363,9 @@ void die(const char * str, struct pt_regs * regs, long err) if (++die.lock_owner_depth < 3) { int nl = 0; + unsigned long esp; + unsigned short ss; + handle_BUG(regs); printk(KERN_EMERG "%s: %04lx [#%d]\n", str, err & 0xffff, ++die_counter); #ifdef CONFIG_PREEMPT @@ -365,8 +386,23 @@ void die(const char * str, struct pt_regs * regs, long err) #endif if (nl) printk("\n"); - notify_die(DIE_OOPS, (char *)str, regs, err, 255, SIGSEGV); - show_registers(regs); + if (notify_die(DIE_OOPS, str, regs, err, + current->thread.trap_no, SIGSEGV) != + NOTIFY_STOP) { + show_registers(regs); + /* Executive summary in case the oops scrolled away */ + esp = (unsigned long) (®s->esp); + savesegment(ss, ss); + if (user_mode(regs)) { + esp = regs->esp; + ss = regs->xss & 0xffff; + } + printk(KERN_EMERG "EIP: [<%08lx>] ", regs->eip); + print_symbol("%s", regs->eip); + printk(" SS:ESP %04x:%08lx\n", ss, esp); + } + else + regs = NULL; } else printk(KERN_EMERG "Recursive die() failure, output suppressed\n"); @@ -374,6 +410,9 @@ void die(const char * str, struct pt_regs * regs, long err) die.lock_owner = -1; spin_unlock_irqrestore(&die.lock, flags); + if (!regs) + return; + if (kexec_should_crash(current)) crash_kexec(regs); @@ -385,6 +424,7 @@ void die(const char * str, struct pt_regs * regs, long err) ssleep(5); panic("Fatal exception"); } + oops_exit(); do_exit(SIGSEGV); } @@ -594,7 +634,7 @@ static DEFINE_SPINLOCK(nmi_print_lock); void die_nmi (struct pt_regs *regs, const char *msg) { - if (notify_die(DIE_NMIWATCHDOG, msg, regs, 0, 0, SIGINT) == + if (notify_die(DIE_NMIWATCHDOG, msg, regs, 0, 2, SIGINT) == NOTIFY_STOP) return; @@ -616,7 +656,7 @@ void die_nmi (struct pt_regs *regs, const char *msg) /* If we are in kernel we are probably nested up pretty bad * and might aswell get out now while we still can. */ - if (!user_mode(regs)) { + if (!user_mode_vm(regs)) { current->thread.trap_no = 2; crash_kexec(regs); } @@ -633,7 +673,7 @@ static void default_do_nmi(struct pt_regs * regs) reason = get_nmi_reason(); if (!(reason & 0xc0)) { - if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, reason, 0, SIGINT) + if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, reason, 2, SIGINT) == NOTIFY_STOP) return; #ifdef CONFIG_X86_LOCAL_APIC @@ -649,7 +689,7 @@ static void default_do_nmi(struct pt_regs * regs) unknown_nmi_error(reason, regs); return; } - if (notify_die(DIE_NMI, "nmi", regs, reason, 0, SIGINT) == NOTIFY_STOP) + if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP) return; if (reason & 0x80) mem_parity_error(reason, regs); @@ -687,6 +727,7 @@ fastcall void do_nmi(struct pt_regs * regs, long error_code) void set_nmi_callback(nmi_callback_t callback) { + vmalloc_sync_all(); rcu_assign_pointer(nmi_callback, callback); } EXPORT_SYMBOL_GPL(set_nmi_callback); @@ -1157,6 +1198,6 @@ void __init trap_init(void) static int __init kstack_setup(char *s) { kstack_depth_to_print = simple_strtoul(s, NULL, 0); - return 0; + return 1; } __setup("kstack=", kstack_setup); diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/vm86.c b/linux-2.6-xen-sparse/arch/i386/kernel/vm86.c index 0e3a0d46c8..c0adee5da5 100644 --- a/linux-2.6-xen-sparse/arch/i386/kernel/vm86.c +++ b/linux-2.6-xen-sparse/arch/i386/kernel/vm86.c @@ -324,7 +324,7 @@ static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk /*call audit_syscall_exit since we do not exit via the normal paths */ if (unlikely(current->audit_context)) - audit_syscall_exit(current, AUDITSC_RESULT(eax), eax); + audit_syscall_exit(AUDITSC_RESULT(eax), eax); __asm__ __volatile__( "movl %0,%%esp\n\t" diff --git a/linux-2.6-xen-sparse/arch/i386/kernel/vmlinux.lds.S b/linux-2.6-xen-sparse/arch/i386/kernel/vmlinux.lds.S index 83b386d6ce..1e3f8734e5 100644 --- a/linux-2.6-xen-sparse/arch/i386/kernel/vmlinux.lds.S +++ b/linux-2.6-xen-sparse/arch/i386/kernel/vmlinux.lds.S @@ -7,6 +7,7 @@ #include <asm-generic/vmlinux.lds.h> #include <asm/thread_info.h> #include <asm/page.h> +#include <asm/cache.h> OUTPUT_FORMAT("elf32-i386", "elf32-i386", "elf32-i386") OUTPUT_ARCH(i386) @@ -40,13 +41,6 @@ SECTIONS __ex_table : AT(ADDR(__ex_table) - LOAD_OFFSET) { *(__ex_table) } __stop___ex_table = .; - . = ALIGN(16); - __start_smp_alternatives_table = .; - __smp_alternatives : AT(ADDR(__smp_alternatives) - LOAD_OFFSET) { *(__smp_alternatives) } - __stop_smp_alternatives_table = .; - - __smp_replacements : AT(ADDR(__smp_replacements) - LOAD_OFFSET) { *(__smp_replacements) } - RODATA /* writeable */ @@ -81,6 +75,26 @@ SECTIONS *(.data.init_task) } + /* might get freed after init */ + . = ALIGN(4096); + __smp_alt_begin = .; + __smp_alt_instructions = .; + .smp_altinstructions : AT(ADDR(.smp_altinstructions) - LOAD_OFFSET) { + *(.smp_altinstructions) + } + __smp_alt_instructions_end = .; + . = ALIGN(4); + __smp_locks = .; + .smp_locks : AT(ADDR(.smp_locks) - LOAD_OFFSET) { + *(.smp_locks) + } + __smp_locks_end = .; + .smp_altinstr_replacement : AT(ADDR(.smp_altinstr_replacement) - LOAD_OFFSET) { + *(.smp_altinstr_replacement) + } + . = ALIGN(4096); + __smp_alt_end = .; + /* will be freed after init */ . = ALIGN(4096); /* Init code and data */ __init_begin = .; @@ -128,7 +142,7 @@ SECTIONS __initramfs_start = .; .init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET) { *(.init.ramfs) } __initramfs_end = .; - . = ALIGN(32); + . = ALIGN(L1_CACHE_BYTES); __per_cpu_start = .; .data.percpu : AT(ADDR(.data.percpu) - LOAD_OFFSET) { *(.data.percpu) } __per_cpu_end = .; |