summaryrefslogtreecommitdiffstats
path: root/arch/x86/kernel
diff options
context:
space:
mode:
Diffstat (limited to 'arch/x86/kernel')
-rw-r--r--arch/x86/kernel/Makefile2
-rw-r--r--arch/x86/kernel/amd_gart_64.c2
-rw-r--r--arch/x86/kernel/apic/apic.c33
-rw-r--r--arch/x86/kernel/apic/bigsmp_32.c20
-rw-r--r--arch/x86/kernel/apic/hw_nmi.c27
-rw-r--r--arch/x86/kernel/apic/io_apic.c678
-rw-r--r--arch/x86/kernel/apic/probe_32.c10
-rw-r--r--arch/x86/kernel/apic/x2apic_uv_x.c27
-rw-r--r--arch/x86/kernel/apm_32.c2
-rw-r--r--arch/x86/kernel/cpu/Makefile8
-rw-r--r--arch/x86/kernel/cpu/amd.c49
-rw-r--r--arch/x86/kernel/cpu/common.c5
-rw-r--r--arch/x86/kernel/cpu/cpu.h1
-rw-r--r--arch/x86/kernel/cpu/intel.c24
-rw-r--r--arch/x86/kernel/cpu/intel_cacheinfo.c127
-rw-r--r--arch/x86/kernel/cpu/mcheck/mce-inject.c20
-rw-r--r--arch/x86/kernel/cpu/mcheck/mce.c50
-rw-r--r--arch/x86/kernel/cpu/mcheck/mce_intel.c10
-rw-r--r--arch/x86/kernel/cpu/mshyperv.c23
-rw-r--r--arch/x86/kernel/cpu/perf_event.c445
-rw-r--r--arch/x86/kernel/cpu/perf_event.h505
-rw-r--r--arch/x86/kernel/cpu/perf_event_amd.c38
-rw-r--r--arch/x86/kernel/cpu/perf_event_amd_ibs.c294
-rw-r--r--arch/x86/kernel/cpu/perf_event_intel.c146
-rw-r--r--arch/x86/kernel/cpu/perf_event_intel_ds.c79
-rw-r--r--arch/x86/kernel/cpu/perf_event_intel_lbr.c28
-rw-r--r--arch/x86/kernel/cpu/perf_event_p4.c10
-rw-r--r--arch/x86/kernel/cpu/perf_event_p6.c9
-rw-r--r--arch/x86/kernel/cpu/proc.c2
-rw-r--r--arch/x86/kernel/cpu/rdrand.c73
-rw-r--r--arch/x86/kernel/crash.c5
-rw-r--r--arch/x86/kernel/entry_64.S14
-rw-r--r--arch/x86/kernel/jump_label.c2
-rw-r--r--arch/x86/kernel/kgdb.c60
-rw-r--r--arch/x86/kernel/kprobes.c5
-rw-r--r--arch/x86/kernel/microcode_amd.c5
-rw-r--r--arch/x86/kernel/microcode_core.c8
-rw-r--r--arch/x86/kernel/microcode_intel.c14
-rw-r--r--arch/x86/kernel/nmi.c433
-rw-r--r--arch/x86/kernel/pci-dma.c4
-rw-r--r--arch/x86/kernel/process.c2
-rw-r--r--arch/x86/kernel/process_32.c4
-rw-r--r--arch/x86/kernel/process_64.c2
-rw-r--r--arch/x86/kernel/reboot.c23
-rw-r--r--arch/x86/kernel/rtc.c23
-rw-r--r--arch/x86/kernel/sys_x86_64.c78
-rw-r--r--arch/x86/kernel/syscall_table_32.S2
-rw-r--r--arch/x86/kernel/traps.c155
-rw-r--r--arch/x86/kernel/vsyscall_64.c2
49 files changed, 2265 insertions, 1323 deletions
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index 82f2912155a..8baca3c4871 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -19,7 +19,7 @@ endif
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
+obj-y += time.o ioport.o ldt.o dumpstack.o nmi.o
obj-y += setup.o x86_init.o i8259.o irqinit.o jump_label.o
obj-$(CONFIG_IRQ_WORK) += irq_work.o
obj-y += probe_roms.o
diff --git a/arch/x86/kernel/amd_gart_64.c b/arch/x86/kernel/amd_gart_64.c
index 8a439d364b9..b1e7c7f7a0a 100644
--- a/arch/x86/kernel/amd_gart_64.c
+++ b/arch/x86/kernel/amd_gart_64.c
@@ -5,7 +5,7 @@
* This allows to use PCI devices that only support 32bit addresses on systems
* with more than 4GB.
*
- * See Documentation/PCI/PCI-DMA-mapping.txt for the interface specification.
+ * See Documentation/DMA-API-HOWTO.txt for the interface specification.
*
* Copyright 2002 Andi Kleen, SuSE Labs.
* Subject to the GNU General Public License v2 only.
diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
index 52fa56399a5..a2fd72e0ab3 100644
--- a/arch/x86/kernel/apic/apic.c
+++ b/arch/x86/kernel/apic/apic.c
@@ -1437,27 +1437,21 @@ void enable_x2apic(void)
int __init enable_IR(void)
{
-#ifdef CONFIG_INTR_REMAP
+#ifdef CONFIG_IRQ_REMAP
if (!intr_remapping_supported()) {
pr_debug("intr-remapping not supported\n");
- return 0;
+ return -1;
}
if (!x2apic_preenabled && skip_ioapic_setup) {
pr_info("Skipped enabling intr-remap because of skipping "
"io-apic setup\n");
- return 0;
+ return -1;
}
- if (enable_intr_remapping(x2apic_supported()))
- return 0;
-
- pr_info("Enabled Interrupt-remapping\n");
-
- return 1;
-
+ return enable_intr_remapping();
#endif
- return 0;
+ return -1;
}
void __init enable_IR_x2apic(void)
@@ -1481,11 +1475,11 @@ void __init enable_IR_x2apic(void)
mask_ioapic_entries();
if (dmar_table_init_ret)
- ret = 0;
+ ret = -1;
else
ret = enable_IR();
- if (!ret) {
+ if (ret < 0) {
/* IR is required if there is APIC ID > 255 even when running
* under KVM
*/
@@ -1499,6 +1493,9 @@ void __init enable_IR_x2apic(void)
x2apic_force_phys();
}
+ if (ret == IRQ_REMAP_XAPIC_MODE)
+ goto nox2apic;
+
x2apic_enabled = 1;
if (x2apic_supported() && !x2apic_mode) {
@@ -1508,19 +1505,21 @@ void __init enable_IR_x2apic(void)
}
nox2apic:
- if (!ret) /* IR enabling failed */
+ if (ret < 0) /* IR enabling failed */
restore_ioapic_entries();
legacy_pic->restore_mask();
local_irq_restore(flags);
out:
- if (x2apic_enabled)
+ if (x2apic_enabled || !x2apic_supported())
return;
if (x2apic_preenabled)
panic("x2apic: enabled by BIOS but kernel init failed.");
- else if (cpu_has_x2apic)
- pr_info("Not enabling x2apic, Intr-remapping init failed.\n");
+ else if (ret == IRQ_REMAP_XAPIC_MODE)
+ pr_info("x2apic not enabled, IRQ remapping is in xapic mode\n");
+ else if (ret < 0)
+ pr_info("x2apic not enabled, IRQ remapping init failed\n");
}
#ifdef CONFIG_X86_64
diff --git a/arch/x86/kernel/apic/bigsmp_32.c b/arch/x86/kernel/apic/bigsmp_32.c
index efd737e827f..521bead0113 100644
--- a/arch/x86/kernel/apic/bigsmp_32.c
+++ b/arch/x86/kernel/apic/bigsmp_32.c
@@ -255,12 +255,24 @@ static struct apic apic_bigsmp = {
.x86_32_early_logical_apicid = bigsmp_early_logical_apicid,
};
-struct apic * __init generic_bigsmp_probe(void)
+void __init generic_bigsmp_probe(void)
{
- if (probe_bigsmp())
- return &apic_bigsmp;
+ unsigned int cpu;
- return NULL;
+ if (!probe_bigsmp())
+ return;
+
+ apic = &apic_bigsmp;
+
+ for_each_possible_cpu(cpu) {
+ if (early_per_cpu(x86_cpu_to_logical_apicid,
+ cpu) == BAD_APICID)
+ continue;
+ early_per_cpu(x86_cpu_to_logical_apicid, cpu) =
+ bigsmp_early_logical_apicid(cpu);
+ }
+
+ pr_info("Overriding APIC driver with %s\n", apic_bigsmp.name);
}
apic_driver(apic_bigsmp);
diff --git a/arch/x86/kernel/apic/hw_nmi.c b/arch/x86/kernel/apic/hw_nmi.c
index d5e57db0f7b..31cb9ae992b 100644
--- a/arch/x86/kernel/apic/hw_nmi.c
+++ b/arch/x86/kernel/apic/hw_nmi.c
@@ -60,22 +60,10 @@ void arch_trigger_all_cpu_backtrace(void)
}
static int __kprobes
-arch_trigger_all_cpu_backtrace_handler(struct notifier_block *self,
- unsigned long cmd, void *__args)
+arch_trigger_all_cpu_backtrace_handler(unsigned int cmd, struct pt_regs *regs)
{
- struct die_args *args = __args;
- struct pt_regs *regs;
int cpu;
- switch (cmd) {
- case DIE_NMI:
- break;
-
- default:
- return NOTIFY_DONE;
- }
-
- regs = args->regs;
cpu = smp_processor_id();
if (cpumask_test_cpu(cpu, to_cpumask(backtrace_mask))) {
@@ -86,21 +74,16 @@ arch_trigger_all_cpu_backtrace_handler(struct notifier_block *self,
show_regs(regs);
arch_spin_unlock(&lock);
cpumask_clear_cpu(cpu, to_cpumask(backtrace_mask));
- return NOTIFY_STOP;
+ return NMI_HANDLED;
}
- return NOTIFY_DONE;
+ return NMI_DONE;
}
-static __read_mostly struct notifier_block backtrace_notifier = {
- .notifier_call = arch_trigger_all_cpu_backtrace_handler,
- .next = NULL,
- .priority = NMI_LOCAL_LOW_PRIOR,
-};
-
static int __init register_trigger_all_cpu_backtrace(void)
{
- register_die_notifier(&backtrace_notifier);
+ register_nmi_handler(NMI_LOCAL, arch_trigger_all_cpu_backtrace_handler,
+ 0, "arch_bt");
return 0;
}
early_initcall(register_trigger_all_cpu_backtrace);
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 8eb863e27ea..3c31fa98af6 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -92,21 +92,21 @@ static struct ioapic {
DECLARE_BITMAP(pin_programmed, MP_MAX_IOAPIC_PIN + 1);
} ioapics[MAX_IO_APICS];
-#define mpc_ioapic_ver(id) ioapics[id].mp_config.apicver
+#define mpc_ioapic_ver(ioapic_idx) ioapics[ioapic_idx].mp_config.apicver
-int mpc_ioapic_id(int id)
+int mpc_ioapic_id(int ioapic_idx)
{
- return ioapics[id].mp_config.apicid;
+ return ioapics[ioapic_idx].mp_config.apicid;
}
-unsigned int mpc_ioapic_addr(int id)
+unsigned int mpc_ioapic_addr(int ioapic_idx)
{
- return ioapics[id].mp_config.apicaddr;
+ return ioapics[ioapic_idx].mp_config.apicaddr;
}
-struct mp_ioapic_gsi *mp_ioapic_gsi_routing(int id)
+struct mp_ioapic_gsi *mp_ioapic_gsi_routing(int ioapic_idx)
{
- return &ioapics[id].gsi_config;
+ return &ioapics[ioapic_idx].gsi_config;
}
int nr_ioapics;
@@ -186,11 +186,7 @@ static struct irq_pin_list *alloc_irq_pin_list(int node)
/* irq_cfg is indexed by the sum of all RTEs in all I/O APICs. */
-#ifdef CONFIG_SPARSE_IRQ
static struct irq_cfg irq_cfgx[NR_IRQS_LEGACY];
-#else
-static struct irq_cfg irq_cfgx[NR_IRQS];
-#endif
int __init arch_early_irq_init(void)
{
@@ -234,7 +230,6 @@ int __init arch_early_irq_init(void)
return 0;
}
-#ifdef CONFIG_SPARSE_IRQ
static struct irq_cfg *irq_cfg(unsigned int irq)
{
return irq_get_chip_data(irq);
@@ -269,22 +264,6 @@ static void free_irq_cfg(unsigned int at, struct irq_cfg *cfg)
kfree(cfg);
}
-#else
-
-struct irq_cfg *irq_cfg(unsigned int irq)
-{
- return irq < nr_irqs ? irq_cfgx + irq : NULL;
-}
-
-static struct irq_cfg *alloc_irq_cfg(unsigned int irq, int node)
-{
- return irq_cfgx + irq;
-}
-
-static inline void free_irq_cfg(unsigned int at, struct irq_cfg *cfg) { }
-
-#endif
-
static struct irq_cfg *alloc_irq_and_cfg_at(unsigned int at, int node)
{
int res = irq_alloc_desc_at(at, node);
@@ -394,13 +373,21 @@ union entry_union {
struct IO_APIC_route_entry entry;
};
+static struct IO_APIC_route_entry __ioapic_read_entry(int apic, int pin)
+{
+ union entry_union eu;
+
+ eu.w1 = io_apic_read(apic, 0x10 + 2 * pin);
+ eu.w2 = io_apic_read(apic, 0x11 + 2 * pin);
+ return eu.entry;
+}
+
static struct IO_APIC_route_entry ioapic_read_entry(int apic, int pin)
{
union entry_union eu;
unsigned long flags;
raw_spin_lock_irqsave(&ioapic_lock, flags);
- eu.w1 = io_apic_read(apic, 0x10 + 2 * pin);
- eu.w2 = io_apic_read(apic, 0x11 + 2 * pin);
+ eu.entry = __ioapic_read_entry(apic, pin);
raw_spin_unlock_irqrestore(&ioapic_lock, flags);
return eu.entry;
}
@@ -529,18 +516,6 @@ static void io_apic_modify_irq(struct irq_cfg *cfg,
__io_apic_modify_irq(entry, mask_and, mask_or, final);
}
-static void __mask_and_edge_IO_APIC_irq(struct irq_pin_list *entry)
-{
- __io_apic_modify_irq(entry, ~IO_APIC_REDIR_LEVEL_TRIGGER,
- IO_APIC_REDIR_MASKED, NULL);
-}
-
-static void __unmask_and_level_IO_APIC_irq(struct irq_pin_list *entry)
-{
- __io_apic_modify_irq(entry, ~IO_APIC_REDIR_MASKED,
- IO_APIC_REDIR_LEVEL_TRIGGER, NULL);
-}
-
static void io_apic_sync(struct irq_pin_list *entry)
{
/*
@@ -585,6 +560,66 @@ static void unmask_ioapic_irq(struct irq_data *data)
unmask_ioapic(data->chip_data);
}
+/*
+ * IO-APIC versions below 0x20 don't support EOI register.
+ * For the record, here is the information about various versions:
+ * 0Xh 82489DX
+ * 1Xh I/OAPIC or I/O(x)APIC which are not PCI 2.2 Compliant
+ * 2Xh I/O(x)APIC which is PCI 2.2 Compliant
+ * 30h-FFh Reserved
+ *
+ * Some of the Intel ICH Specs (ICH2 to ICH5) documents the io-apic
+ * version as 0x2. This is an error with documentation and these ICH chips
+ * use io-apic's of version 0x20.
+ *
+ * For IO-APIC's with EOI register, we use that to do an explicit EOI.
+ * Otherwise, we simulate the EOI message manually by changing the trigger
+ * mode to edge and then back to level, with RTE being masked during this.
+ */
+static void __eoi_ioapic_pin(int apic, int pin, int vector, struct irq_cfg *cfg)
+{
+ if (mpc_ioapic_ver(apic) >= 0x20) {
+ /*
+ * Intr-remapping uses pin number as the virtual vector
+ * in the RTE. Actual vector is programmed in
+ * intr-remapping table entry. Hence for the io-apic
+ * EOI we use the pin number.
+ */
+ if (cfg && irq_remapped(cfg))
+ io_apic_eoi(apic, pin);
+ else
+ io_apic_eoi(apic, vector);
+ } else {
+ struct IO_APIC_route_entry entry, entry1;
+
+ entry = entry1 = __ioapic_read_entry(apic, pin);
+
+ /*
+ * Mask the entry and change the trigger mode to edge.
+ */
+ entry1.mask = 1;
+ entry1.trigger = IOAPIC_EDGE;
+
+ __ioapic_write_entry(apic, pin, entry1);
+
+ /*
+ * Restore the previous level triggered entry.
+ */
+ __ioapic_write_entry(apic, pin, entry);
+ }
+}
+
+static void eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg)
+{
+ struct irq_pin_list *entry;
+ unsigned long flags;
+
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ for_each_irq_pin(entry, cfg->irq_2_pin)
+ __eoi_ioapic_pin(entry->apic, entry->pin, cfg->vector, cfg);
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+}
+
static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin)
{
struct IO_APIC_route_entry entry;
@@ -593,10 +628,44 @@ static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin)
entry = ioapic_read_entry(apic, pin);
if (entry.delivery_mode == dest_SMI)
return;
+
+ /*
+ * Make sure the entry is masked and re-read the contents to check
+ * if it is a level triggered pin and if the remote-IRR is set.
+ */
+ if (!entry.mask) {
+ entry.mask = 1;
+ ioapic_write_entry(apic, pin, entry);
+ entry = ioapic_read_entry(apic, pin);
+ }
+
+ if (entry.irr) {
+ unsigned long flags;
+
+ /*
+ * Make sure the trigger mode is set to level. Explicit EOI
+ * doesn't clear the remote-IRR if the trigger mode is not
+ * set to level.
+ */
+ if (!entry.trigger) {
+ entry.trigger = IOAPIC_LEVEL;
+ ioapic_write_entry(apic, pin, entry);
+ }
+
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ __eoi_ioapic_pin(apic, pin, entry.vector, NULL);
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+ }
+
/*
- * Disable it in the IO-APIC irq-routing table:
+ * Clear the rest of the bits in the IO-APIC RTE except for the mask
+ * bit.
*/
ioapic_mask_entry(apic, pin);
+ entry = ioapic_read_entry(apic, pin);
+ if (entry.irr)
+ printk(KERN_ERR "Unable to reset IRR for apic: %d, pin :%d\n",
+ mpc_ioapic_id(apic), pin);
}
static void clear_IO_APIC (void)
@@ -712,13 +781,13 @@ int restore_ioapic_entries(void)
/*
* Find the IRQ entry number of a certain pin.
*/
-static int find_irq_entry(int apic, int pin, int type)
+static int find_irq_entry(int ioapic_idx, int pin, int type)
{
int i;
for (i = 0; i < mp_irq_entries; i++)
if (mp_irqs[i].irqtype == type &&
- (mp_irqs[i].dstapic == mpc_ioapic_id(apic) ||
+ (mp_irqs[i].dstapic == mpc_ioapic_id(ioapic_idx) ||
mp_irqs[i].dstapic == MP_APIC_ALL) &&
mp_irqs[i].dstirq == pin)
return i;
@@ -757,12 +826,13 @@ static int __init find_isa_irq_apic(int irq, int type)
(mp_irqs[i].srcbusirq == irq))
break;
}
+
if (i < mp_irq_entries) {
- int apic;
- for(apic = 0; apic < nr_ioapics; apic++) {
- if (mpc_ioapic_id(apic) == mp_irqs[i].dstapic)
- return apic;
- }
+ int ioapic_idx;
+
+ for (ioapic_idx = 0; ioapic_idx < nr_ioapics; ioapic_idx++)
+ if (mpc_ioapic_id(ioapic_idx) == mp_irqs[i].dstapic)
+ return ioapic_idx;
}
return -1;
@@ -977,7 +1047,7 @@ static int pin_2_irq(int idx, int apic, int pin)
int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin,
struct io_apic_irq_attr *irq_attr)
{
- int apic, i, best_guess = -1;
+ int ioapic_idx, i, best_guess = -1;
apic_printk(APIC_DEBUG,
"querying PCI -> IRQ mapping bus:%d, slot:%d, pin:%d.\n",
@@ -990,8 +1060,8 @@ int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin,
for (i = 0; i < mp_irq_entries; i++) {
int lbus = mp_irqs[i].srcbus;
- for (apic = 0; apic < nr_ioapics; apic++)
- if (mpc_ioapic_id(apic) == mp_irqs[i].dstapic ||
+ for (ioapic_idx = 0; ioapic_idx < nr_ioapics; ioapic_idx++)
+ if (mpc_ioapic_id(ioapic_idx) == mp_irqs[i].dstapic ||
mp_irqs[i].dstapic == MP_APIC_ALL)
break;
@@ -999,13 +1069,13 @@ int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin,
!mp_irqs[i].irqtype &&
(bus == lbus) &&
(slot == ((mp_irqs[i].srcbusirq >> 2) & 0x1f))) {
- int irq = pin_2_irq(i, apic, mp_irqs[i].dstirq);
+ int irq = pin_2_irq(i, ioapic_idx, mp_irqs[i].dstirq);
- if (!(apic || IO_APIC_IRQ(irq)))
+ if (!(ioapic_idx || IO_APIC_IRQ(irq)))
continue;
if (pin == (mp_irqs[i].srcbusirq & 3)) {
- set_io_apic_irq_attr(irq_attr, apic,
+ set_io_apic_irq_attr(irq_attr, ioapic_idx,
mp_irqs[i].dstirq,
irq_trigger(i),
irq_polarity(i));
@@ -1016,7 +1086,7 @@ int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin,
* best-guess fuzzy result for broken mptables.
*/
if (best_guess < 0) {
- set_io_apic_irq_attr(irq_attr, apic,
+ set_io_apic_irq_attr(irq_attr, ioapic_idx,
mp_irqs[i].dstirq,
irq_trigger(i),
irq_polarity(i));
@@ -1202,7 +1272,6 @@ void __setup_vector_irq(int cpu)
}
static struct irq_chip ioapic_chip;
-static struct irq_chip ir_ioapic_chip;
#ifdef CONFIG_X86_32
static inline int IO_APIC_irq_trigger(int irq)
@@ -1246,7 +1315,7 @@ static void ioapic_register_intr(unsigned int irq, struct irq_cfg *cfg,
if (irq_remapped(cfg)) {
irq_set_status_flags(irq, IRQ_MOVE_PCNTXT);
- chip = &ir_ioapic_chip;
+ irq_remap_modify_chip_defaults(chip);
fasteoi = trigger != 0;
}
@@ -1255,77 +1324,100 @@ static void ioapic_register_intr(unsigned int irq, struct irq_cfg *cfg,
fasteoi ? "fasteoi" : "edge");
}
-static int setup_ioapic_entry(int apic_id, int irq,
- struct IO_APIC_route_entry *entry,
- unsigned int destination, int trigger,
- int polarity, int vector, int pin)
+
+static int setup_ir_ioapic_entry(int irq,
+ struct IR_IO_APIC_route_entry *entry,
+ unsigned int destination, int vector,
+ struct io_apic_irq_attr *attr)
{
- /*
- * add it to the IO-APIC irq-routing table:
- */
- memset(entry,0,sizeof(*entry));
+ int index;
+ struct irte irte;
+ int ioapic_id = mpc_ioapic_id(attr->ioapic);
+ struct intel_iommu *iommu = map_ioapic_to_ir(ioapic_id);
- if (intr_remapping_enabled) {
- struct intel_iommu *iommu = map_ioapic_to_ir(apic_id);
- struct irte irte;
- struct IR_IO_APIC_route_entry *ir_entry =
- (struct IR_IO_APIC_route_entry *) entry;
- int index;
+ if (!iommu) {
+ pr_warn("No mapping iommu for ioapic %d\n", ioapic_id);
+ return -ENODEV;
+ }
- if (!iommu)
- panic("No mapping iommu for ioapic %d\n", apic_id);
+ index = alloc_irte(iommu, irq, 1);
+ if (index < 0) {
+ pr_warn("Failed to allocate IRTE for ioapic %d\n", ioapic_id);
+ return -ENOMEM;
+ }
- index = alloc_irte(iommu, irq, 1);
- if (index < 0)
- panic("Failed to allocate IRTE for ioapic %d\n", apic_id);
+ prepare_irte(&irte, vector, destination);
- prepare_irte(&irte, vector, destination);
+ /* Set source-id of interrupt request */
+ set_ioapic_sid(&irte, ioapic_id);
- /* Set source-id of interrupt request */
- set_ioapic_sid(&irte, apic_id);
+ modify_irte(irq, &irte);
- modify_irte(irq, &irte);
+ apic_printk(APIC_VERBOSE, KERN_DEBUG "IOAPIC[%d]: "
+ "Set IRTE entry (P:%d FPD:%d Dst_Mode:%d "
+ "Redir_hint:%d Trig_Mode:%d Dlvry_Mode:%X "
+ "Avail:%X Vector:%02X Dest:%08X "
+ "SID:%04X SQ:%X SVT:%X)\n",
+ attr->ioapic, irte.present, irte.fpd, irte.dst_mode,
+ irte.redir_hint, irte.trigger_mode, irte.dlvry_mode,
+ irte.avail, irte.vector, irte.dest_id,
+ irte.sid, irte.sq, irte.svt);
+
+ memset(entry, 0, sizeof(*entry));
+
+ entry->index2 = (index >> 15) & 0x1;
+ entry->zero = 0;
+ entry->format = 1;
+ entry->index = (index & 0x7fff);
+ /*
+ * IO-APIC RTE will be configured with virtual vector.
+ * irq handler will do the explicit EOI to the io-apic.
+ */
+ entry->vector = attr->ioapic_pin;
+ entry->mask = 0; /* enable IRQ */
+ entry->trigger = attr->trigger;
+ entry->polarity = attr->polarity;
- ir_entry->index2 = (index >> 15) & 0x1;
- ir_entry->zero = 0;
- ir_entry->format = 1;
- ir_entry->index = (index & 0x7fff);
- /*
- * IO-APIC RTE will be configured with virtual vector.
- * irq handler will do the explicit EOI to the io-apic.
- */
- ir_entry->vector = pin;
-
- apic_printk(APIC_VERBOSE, KERN_DEBUG "IOAPIC[%d]: "
- "Set IRTE entry (P:%d FPD:%d Dst_Mode:%d "
- "Redir_hint:%d Trig_Mode:%d Dlvry_Mode:%X "
- "Avail:%X Vector:%02X Dest:%08X "
- "SID:%04X SQ:%X SVT:%X)\n",
- apic_id, irte.present, irte.fpd, irte.dst_mode,
- irte.redir_hint, irte.trigger_mode, irte.dlvry_mode,
- irte.avail, irte.vector, irte.dest_id,
- irte.sid, irte.sq, irte.svt);
- } else {
- entry->delivery_mode = apic->irq_delivery_mode;
- entry->dest_mode = apic->irq_dest_mode;
- entry->dest = destination;
- entry->vector = vector;
- }
+ /* Mask level triggered irqs.
+ * Use IRQ_DELAYED_DISABLE for edge triggered irqs.
+ */
+ if (attr->trigger)
+ entry->mask = 1;
- entry->mask = 0; /* enable IRQ */
- entry->trigger = trigger;
- entry->polarity = polarity;
+ return 0;
+}
- /* Mask level triggered irqs.
+static int setup_ioapic_entry(int irq, struct IO_APIC_route_entry *entry,
+ unsigned int destination, int vector,
+ struct io_apic_irq_attr *attr)
+{
+ if (intr_remapping_enabled)
+ return setup_ir_ioapic_entry(irq,
+ (struct IR_IO_APIC_route_entry *)entry,
+ destination, vector, attr);
+
+ memset(entry, 0, sizeof(*entry));
+
+ entry->delivery_mode = apic->irq_delivery_mode;
+ entry->dest_mode = apic->irq_dest_mode;
+ entry->dest = destination;
+ entry->vector = vector;
+ entry->mask = 0; /* enable IRQ */
+ entry->trigger = attr->trigger;
+ entry->polarity = attr->polarity;
+
+ /*
+ * Mask level triggered irqs.
* Use IRQ_DELAYED_DISABLE for edge triggered irqs.
*/
- if (trigger)
+ if (attr->trigger)
entry->mask = 1;
+
return 0;
}
-static void setup_ioapic_irq(int apic_id, int pin, unsigned int irq,
- struct irq_cfg *cfg, int trigger, int polarity)
+static void setup_ioapic_irq(unsigned int irq, struct irq_cfg *cfg,
+ struct io_apic_irq_attr *attr)
{
struct IO_APIC_route_entry entry;
unsigned int dest;
@@ -1348,49 +1440,48 @@ static void setup_ioapic_irq(int apic_id, int pin, unsigned int irq,
apic_printk(APIC_VERBOSE,KERN_DEBUG
"IOAPIC[%d]: Set routing entry (%d-%d -> 0x%x -> "
"IRQ %d Mode:%i Active:%i Dest:%d)\n",
- apic_id, mpc_ioapic_id(apic_id), pin, cfg->vector,
- irq, trigger, polarity, dest);
-
+ attr->ioapic, mpc_ioapic_id(attr->ioapic), attr->ioapic_pin,
+ cfg->vector, irq, attr->trigger, attr->polarity, dest);
- if (setup_ioapic_entry(mpc_ioapic_id(apic_id), irq, &entry,
- dest, trigger, polarity, cfg->vector, pin)) {
- printk("Failed to setup ioapic entry for ioapic %d, pin %d\n",
- mpc_ioapic_id(apic_id), pin);
+ if (setup_ioapic_entry(irq, &entry, dest, cfg->vector, attr)) {
+ pr_warn("Failed to setup ioapic entry for ioapic %d, pin %d\n",
+ mpc_ioapic_id(attr->ioapic), attr->ioapic_pin);
__clear_irq_vector(irq, cfg);
+
return;
}
- ioapic_register_intr(irq, cfg, trigger);
+ ioapic_register_intr(irq, cfg, attr->trigger);
if (irq < legacy_pic->nr_legacy_irqs)
legacy_pic->mask(irq);
- ioapic_write_entry(apic_id, pin, entry);
+ ioapic_write_entry(attr->ioapic, attr->ioapic_pin, entry);
}
-static bool __init io_apic_pin_not_connected(int idx, int apic_id, int pin)
+static bool __init io_apic_pin_not_connected(int idx, int ioapic_idx, int pin)
{
if (idx != -1)
return false;
apic_printk(APIC_VERBOSE, KERN_DEBUG " apic %d pin %d not connected\n",
- mpc_ioapic_id(apic_id), pin);
+ mpc_ioapic_id(ioapic_idx), pin);
return true;
}
-static void __init __io_apic_setup_irqs(unsigned int apic_id)
+static void __init __io_apic_setup_irqs(unsigned int ioapic_idx)
{
int idx, node = cpu_to_node(0);
struct io_apic_irq_attr attr;
unsigned int pin, irq;
- for (pin = 0; pin < ioapics[apic_id].nr_registers; pin++) {
- idx = find_irq_entry(apic_id, pin, mp_INT);
- if (io_apic_pin_not_connected(idx, apic_id, pin))
+ for (pin = 0; pin < ioapics[ioapic_idx].nr_registers; pin++) {
+ idx = find_irq_entry(ioapic_idx, pin, mp_INT);
+ if (io_apic_pin_not_connected(idx, ioapic_idx, pin))
continue;
- irq = pin_2_irq(idx, apic_id, pin);
+ irq = pin_2_irq(idx, ioapic_idx, pin);
- if ((apic_id > 0) && (irq > 16))
+ if ((ioapic_idx > 0) && (irq > 16))
continue;
/*
@@ -1398,10 +1489,10 @@ static void __init __io_apic_setup_irqs(unsigned int apic_id)
* installed and if it returns 1:
*/
if (apic->multi_timer_check &&
- apic->multi_timer_check(apic_id, irq))
+ apic->multi_timer_check(ioapic_idx, irq))
continue;
- set_io_apic_irq_attr(&attr, apic_id, pin, irq_trigger(idx),
+ set_io_apic_irq_attr(&attr, ioapic_idx, pin, irq_trigger(idx),
irq_polarity(idx));
io_apic_setup_irq_pin(irq, node, &attr);
@@ -1410,12 +1501,12 @@ static void __init __io_apic_setup_irqs(unsigned int apic_id)
static void __init setup_IO_APIC_irqs(void)
{
- unsigned int apic_id;
+ unsigned int ioapic_idx;
apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n");
- for (apic_id = 0; apic_id < nr_ioapics; apic_id++)
- __io_apic_setup_irqs(apic_id);
+ for (ioapic_idx = 0; ioapic_idx < nr_ioapics; ioapic_idx++)
+ __io_apic_setup_irqs(ioapic_idx);
}
/*
@@ -1425,28 +1516,28 @@ static void __init setup_IO_APIC_irqs(void)
*/
void setup_IO_APIC_irq_extra(u32 gsi)
{
- int apic_id = 0, pin, idx, irq, node = cpu_to_node(0);
+ int ioapic_idx = 0, pin, idx, irq, node = cpu_to_node(0);
struct io_apic_irq_attr attr;
/*
* Convert 'gsi' to 'ioapic.pin'.
*/
- apic_id = mp_find_ioapic(gsi);
- if (apic_id < 0)
+ ioapic_idx = mp_find_ioapic(gsi);
+ if (ioapic_idx < 0)
return;
- pin = mp_find_ioapic_pin(apic_id, gsi);
- idx = find_irq_entry(apic_id, pin, mp_INT);
+ pin = mp_find_ioapic_pin(ioapic_idx, gsi);
+ idx = find_irq_entry(ioapic_idx, pin, mp_INT);
if (idx == -1)
return;
- irq = pin_2_irq(idx, apic_id, pin);
+ irq = pin_2_irq(idx, ioapic_idx, pin);
/* Only handle the non legacy irqs on secondary ioapics */
- if (apic_id == 0 || irq < NR_IRQS_LEGACY)
+ if (ioapic_idx == 0 || irq < NR_IRQS_LEGACY)
return;
- set_io_apic_irq_attr(&attr, apic_id, pin, irq_trigger(idx),
+ set_io_apic_irq_attr(&attr, ioapic_idx, pin, irq_trigger(idx),
irq_polarity(idx));
io_apic_setup_irq_pin_once(irq, node, &attr);
@@ -1455,8 +1546,8 @@ void setup_IO_APIC_irq_extra(u32 gsi)
/*
* Set up the timer pin, possibly with the 8259A-master behind.
*/
-static void __init setup_timer_IRQ0_pin(unsigned int apic_id, unsigned int pin,
- int vector)
+static void __init setup_timer_IRQ0_pin(unsigned int ioapic_idx,
+ unsigned int pin, int vector)
{
struct IO_APIC_route_entry entry;
@@ -1487,45 +1578,29 @@ static void __init setup_timer_IRQ0_pin(unsigned int apic_id, unsigned int pin,
/*
* Add it to the IO-APIC irq-routing table:
*/
- ioapic_write_entry(apic_id, pin, entry);
+ ioapic_write_entry(ioapic_idx, pin, entry);
}
-
-__apicdebuginit(void) print_IO_APIC(void)
+__apicdebuginit(void) print_IO_APIC(int ioapic_idx)
{
- int apic, i;
+ int i;
union IO_APIC_reg_00 reg_00;
union IO_APIC_reg_01 reg_01;
union IO_APIC_reg_02 reg_02;
union IO_APIC_reg_03 reg_03;
unsigned long flags;
- struct irq_cfg *cfg;
- unsigned int irq;
-
- printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries);
- for (i = 0; i < nr_ioapics; i++)
- printk(KERN_DEBUG "number of IO-APIC #%d registers: %d.\n",
- mpc_ioapic_id(i), ioapics[i].nr_registers);
-
- /*
- * We are a bit conservative about what we expect. We have to
- * know about every hardware change ASAP.
- */
- printk(KERN_INFO "testing the IO APIC.......................\n");
-
- for (apic = 0; apic < nr_ioapics; apic++) {
raw_spin_lock_irqsave(&ioapic_lock, flags);
- reg_00.raw = io_apic_read(apic, 0);
- reg_01.raw = io_apic_read(apic, 1);
+ reg_00.raw = io_apic_read(ioapic_idx, 0);
+ reg_01.raw = io_apic_read(ioapic_idx, 1);
if (reg_01.bits.version >= 0x10)
- reg_02.raw = io_apic_read(apic, 2);
+ reg_02.raw = io_apic_read(ioapic_idx, 2);
if (reg_01.bits.version >= 0x20)
- reg_03.raw = io_apic_read(apic, 3);
+ reg_03.raw = io_apic_read(ioapic_idx, 3);
raw_spin_unlock_irqrestore(&ioapic_lock, flags);
printk("\n");
- printk(KERN_DEBUG "IO APIC #%d......\n", mpc_ioapic_id(apic));
+ printk(KERN_DEBUG "IO APIC #%d......\n", mpc_ioapic_id(ioapic_idx));
printk(KERN_DEBUG ".... register #00: %08X\n", reg_00.raw);
printk(KERN_DEBUG "....... : physical APIC id: %02X\n", reg_00.bits.ID);
printk(KERN_DEBUG "....... : Delivery Type: %X\n", reg_00.bits.delivery_type);
@@ -1575,7 +1650,7 @@ __apicdebuginit(void) print_IO_APIC(void)
struct IO_APIC_route_entry entry;
struct IR_IO_APIC_route_entry *ir_entry;
- entry = ioapic_read_entry(apic, i);
+ entry = ioapic_read_entry(ioapic_idx, i);
ir_entry = (struct IR_IO_APIC_route_entry *) &entry;
printk(KERN_DEBUG " %02x %04X ",
i,
@@ -1596,7 +1671,7 @@ __apicdebuginit(void) print_IO_APIC(void)
} else {
struct IO_APIC_route_entry entry;
- entry = ioapic_read_entry(apic, i);
+ entry = ioapic_read_entry(ioapic_idx, i);
printk(KERN_DEBUG " %02x %02X ",
i,
entry.dest
@@ -1614,7 +1689,28 @@ __apicdebuginit(void) print_IO_APIC(void)
);
}
}
- }
+}
+
+__apicdebuginit(void) print_IO_APICs(void)
+{
+ int ioapic_idx;
+ struct irq_cfg *cfg;
+ unsigned int irq;
+
+ printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries);
+ for (ioapic_idx = 0; ioapic_idx < nr_ioapics; ioapic_idx++)
+ printk(KERN_DEBUG "number of IO-APIC #%d registers: %d.\n",
+ mpc_ioapic_id(ioapic_idx),
+ ioapics[ioapic_idx].nr_registers);
+
+ /*
+ * We are a bit conservative about what we expect. We have to
+ * know about every hardware change ASAP.
+ */
+ printk(KERN_INFO "testing the IO APIC.......................\n");
+
+ for (ioapic_idx = 0; ioapic_idx < nr_ioapics; ioapic_idx++)
+ print_IO_APIC(ioapic_idx);
printk(KERN_DEBUG "IRQ to pin mappings:\n");
for_each_active_irq(irq) {
@@ -1633,8 +1729,6 @@ __apicdebuginit(void) print_IO_APIC(void)
}
printk(KERN_INFO ".................................... done.\n");
-
- return;
}
__apicdebuginit(void) print_APIC_field(int base)
@@ -1828,7 +1922,7 @@ __apicdebuginit(int) print_ICs(void)
return 0;
print_local_APICs(show_lapic);
- print_IO_APIC();
+ print_IO_APICs();
return 0;
}
@@ -1953,7 +2047,7 @@ void __init setup_ioapic_ids_from_mpc_nocheck(void)
{
union IO_APIC_reg_00 reg_00;
physid_mask_t phys_id_present_map;
- int apic_id;
+ int ioapic_idx;
int i;
unsigned char old_id;
unsigned long flags;
@@ -1967,21 +2061,20 @@ void __init setup_ioapic_ids_from_mpc_nocheck(void)
/*
* Set the IOAPIC ID to the value stored in the MPC table.
*/
- for (apic_id = 0; apic_id < nr_ioapics; apic_id++) {
-
+ for (ioapic_idx = 0; ioapic_idx < nr_ioapics; ioapic_idx++) {
/* Read the register 0 value */
raw_spin_lock_irqsave(&ioapic_lock, flags);
- reg_00.raw = io_apic_read(apic_id, 0);
+ reg_00.raw = io_apic_read(ioapic_idx, 0);
raw_spin_unlock_irqrestore(&ioapic_lock, flags);
- old_id = mpc_ioapic_id(apic_id);
+ old_id = mpc_ioapic_id(ioapic_idx);
- if (mpc_ioapic_id(apic_id) >= get_physical_broadcast()) {
+ if (mpc_ioapic_id(ioapic_idx) >= get_physical_broadcast()) {
printk(KERN_ERR "BIOS bug, IO-APIC#%d ID is %d in the MPC table!...\n",
- apic_id, mpc_ioapic_id(apic_id));
+ ioapic_idx, mpc_ioapic_id(ioapic_idx));
printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n",
reg_00.bits.ID);
- ioapics[apic_id].mp_config.apicid = reg_00.bits.ID;
+ ioapics[ioapic_idx].mp_config.apicid = reg_00.bits.ID;
}
/*
@@ -1990,9 +2083,9 @@ void __init setup_ioapic_ids_from_mpc_nocheck(void)
* 'stuck on smp_invalidate_needed IPI wait' messages.
*/
if (apic->check_apicid_used(&phys_id_present_map,
- mpc_ioapic_id(apic_id))) {
+ mpc_ioapic_id(ioapic_idx))) {
printk(KERN_ERR "BIOS bug, IO-APIC#%d ID %d is already used!...\n",
- apic_id, mpc_ioapic_id(apic_id));
+ ioapic_idx, mpc_ioapic_id(ioapic_idx));
for (i = 0; i < get_physical_broadcast(); i++)
if (!physid_isset(i, phys_id_present_map))
break;
@@ -2001,14 +2094,14 @@ void __init setup_ioapic_ids_from_mpc_nocheck(void)
printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n",
i);
physid_set(i, phys_id_present_map);
- ioapics[apic_id].mp_config.apicid = i;
+ ioapics[ioapic_idx].mp_config.apicid = i;
} else {
physid_mask_t tmp;
- apic->apicid_to_cpu_present(mpc_ioapic_id(apic_id),
+ apic->apicid_to_cpu_present(mpc_ioapic_id(ioapic_idx),
&tmp);
apic_printk(APIC_VERBOSE, "Setting %d in the "
"phys_id_present_map\n",
- mpc_ioapic_id(apic_id));
+ mpc_ioapic_id(ioapic_idx));
physids_or(phys_id_present_map, phys_id_present_map, tmp);
}
@@ -2016,35 +2109,35 @@ void __init setup_ioapic_ids_from_mpc_nocheck(void)
* We need to adjust the IRQ routing table
* if the ID changed.
*/
- if (old_id != mpc_ioapic_id(apic_id))
+ if (old_id != mpc_ioapic_id(ioapic_idx))
for (i = 0; i < mp_irq_entries; i++)
if (mp_irqs[i].dstapic == old_id)
mp_irqs[i].dstapic
- = mpc_ioapic_id(apic_id);
+ = mpc_ioapic_id(ioapic_idx);
/*
* Update the ID register according to the right value
* from the MPC table if they are different.
*/
- if (mpc_ioapic_id(apic_id) == reg_00.bits.ID)
+ if (mpc_ioapic_id(ioapic_idx) == reg_00.bits.ID)
continue;
apic_printk(APIC_VERBOSE, KERN_INFO
"...changing IO-APIC physical APIC ID to %d ...",
- mpc_ioapic_id(apic_id));
+ mpc_ioapic_id(ioapic_idx));
- reg_00.bits.ID = mpc_ioapic_id(apic_id);
+ reg_00.bits.ID = mpc_ioapic_id(ioapic_idx);
raw_spin_lock_irqsave(&ioapic_lock, flags);
- io_apic_write(apic_id, 0, reg_00.raw);
+ io_apic_write(ioapic_idx, 0, reg_00.raw);
raw_spin_unlock_irqrestore(&ioapic_lock, flags);
/*
* Sanity check
*/
raw_spin_lock_irqsave(&ioapic_lock, flags);
- reg_00.raw = io_apic_read(apic_id, 0);
+ reg_00.raw = io_apic_read(ioapic_idx, 0);
raw_spin_unlock_irqrestore(&ioapic_lock, flags);
- if (reg_00.bits.ID != mpc_ioapic_id(apic_id))
+ if (reg_00.bits.ID != mpc_ioapic_id(ioapic_idx))
printk("could not set ID!\n");
else
apic_printk(APIC_VERBOSE, " ok.\n");
@@ -2255,7 +2348,7 @@ ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask,
return ret;
}
-#ifdef CONFIG_INTR_REMAP
+#ifdef CONFIG_IRQ_REMAP
/*
* Migrate the IO-APIC irq in the presence of intr-remapping.
@@ -2267,6 +2360,9 @@ ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask,
* updated vector information), by using a virtual vector (io-apic pin number).
* Real vector that is used for interrupting cpu will be coming from
* the interrupt-remapping table entry.
+ *
+ * As the migration is a simple atomic update of IRTE, the same mechanism
+ * is used to migrate MSI irq's in the presence of interrupt-remapping.
*/
static int
ir_ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask,
@@ -2291,10 +2387,16 @@ ir_ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask,
irte.dest_id = IRTE_DEST(dest);
/*
- * Modified the IRTE and flushes the Interrupt entry cache.
+ * Atomically updates the IRTE with the new destination, vector
+ * and flushes the interrupt entry cache.
*/
modify_irte(irq, &irte);
+ /*
+ * After this point, all the interrupts will start arriving
+ * at the new destination. So, time to cleanup the previous
+ * vector allocation.
+ */
if (cfg->move_in_progress)
send_cleanup_vector(cfg);
@@ -2407,48 +2509,6 @@ static void ack_apic_edge(struct irq_data *data)
atomic_t irq_mis_count;
-/*
- * IO-APIC versions below 0x20 don't support EOI register.
- * For the record, here is the information about various versions:
- * 0Xh 82489DX
- * 1Xh I/OAPIC or I/O(x)APIC which are not PCI 2.2 Compliant
- * 2Xh I/O(x)APIC which is PCI 2.2 Compliant
- * 30h-FFh Reserved
- *
- * Some of the Intel ICH Specs (ICH2 to ICH5) documents the io-apic
- * version as 0x2. This is an error with documentation and these ICH chips
- * use io-apic's of version 0x20.
- *
- * For IO-APIC's with EOI register, we use that to do an explicit EOI.
- * Otherwise, we simulate the EOI message manually by changing the trigger
- * mode to edge and then back to level, with RTE being masked during this.
-*/
-static void eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg)
-{
- struct irq_pin_list *entry;
- unsigned long flags;
-
- raw_spin_lock_irqsave(&ioapic_lock, flags);
- for_each_irq_pin(entry, cfg->irq_2_pin) {
- if (mpc_ioapic_ver(entry->apic) >= 0x20) {
- /*
- * Intr-remapping uses pin number as the virtual vector
- * in the RTE. Actual vector is programmed in
- * intr-remapping table entry. Hence for the io-apic
- * EOI we use the pin number.
- */
- if (irq_remapped(cfg))
- io_apic_eoi(entry->apic, entry->pin);
- else
- io_apic_eoi(entry->apic, cfg->vector);
- } else {
- __mask_and_edge_IO_APIC_irq(entry);
- __unmask_and_level_IO_APIC_irq(entry);
- }
- }
- raw_spin_unlock_irqrestore(&ioapic_lock, flags);
-}
-
static void ack_apic_level(struct irq_data *data)
{
struct irq_cfg *cfg = data->chip_data;
@@ -2552,7 +2612,7 @@ static void ack_apic_level(struct irq_data *data)
}
}
-#ifdef CONFIG_INTR_REMAP
+#ifdef CONFIG_IRQ_REMAP
static void ir_ack_apic_edge(struct irq_data *data)
{
ack_APIC_irq();
@@ -2563,7 +2623,23 @@ static void ir_ack_apic_level(struct irq_data *data)
ack_APIC_irq();
eoi_ioapic_irq(data->irq, data->chip_data);
}
-#endif /* CONFIG_INTR_REMAP */
+
+static void ir_print_prefix(struct irq_data *data, struct seq_file *p)
+{
+ seq_printf(p, " IR-%s", data->chip->name);
+}
+
+static void irq_remap_modify_chip_defaults(struct irq_chip *chip)
+{
+ chip->irq_print_chip = ir_print_prefix;
+ chip->irq_ack = ir_ack_apic_edge;
+ chip->irq_eoi = ir_ack_apic_level;
+
+#ifdef CONFIG_SMP
+ chip->irq_set_affinity = ir_ioapic_set_affinity;
+#endif
+}
+#endif /* CONFIG_IRQ_REMAP */
static struct irq_chip ioapic_chip __read_mostly = {
.name = "IO-APIC",
@@ -2578,21 +2654,6 @@ static struct irq_chip ioapic_chip __read_mostly = {
.irq_retrigger = ioapic_retrigger_irq,
};
-static struct irq_chip ir_ioapic_chip __read_mostly = {
- .name = "IR-IO-APIC",
- .irq_startup = startup_ioapic_irq,
- .irq_mask = mask_ioapic_irq,
- .irq_unmask = unmask_ioapic_irq,
-#ifdef CONFIG_INTR_REMAP
- .irq_ack = ir_ack_apic_edge,
- .irq_eoi = ir_ack_apic_level,
-#ifdef CONFIG_SMP
- .irq_set_affinity = ir_ioapic_set_affinity,
-#endif
-#endif
- .irq_retrigger = ioapic_retrigger_irq,
-};
-
static inline void init_IO_APIC_traps(void)
{
struct irq_cfg *cfg;
@@ -2944,27 +3005,26 @@ static int __init io_apic_bug_finalize(void)
late_initcall(io_apic_bug_finalize);
-static void resume_ioapic_id(int ioapic_id)
+static void resume_ioapic_id(int ioapic_idx)
{
unsigned long flags;
union IO_APIC_reg_00 reg_00;
-
raw_spin_lock_irqsave(&ioapic_lock, flags);
- reg_00.raw = io_apic_read(ioapic_id, 0);
- if (reg_00.bits.ID != mpc_ioapic_id(ioapic_id)) {
- reg_00.bits.ID = mpc_ioapic_id(ioapic_id);
- io_apic_write(ioapic_id, 0, reg_00.raw);
+ reg_00.raw = io_apic_read(ioapic_idx, 0);
+ if (reg_00.bits.ID != mpc_ioapic_id(ioapic_idx)) {
+ reg_00.bits.ID = mpc_ioapic_id(ioapic_idx);
+ io_apic_write(ioapic_idx, 0, reg_00.raw);
}
raw_spin_unlock_irqrestore(&ioapic_lock, flags);
}
static void ioapic_resume(void)
{
- int ioapic_id;
+ int ioapic_idx;
- for (ioapic_id = nr_ioapics - 1; ioapic_id >= 0; ioapic_id--)
- resume_ioapic_id(ioapic_id);
+ for (ioapic_idx = nr_ioapics - 1; ioapic_idx >= 0; ioapic_idx--)
+ resume_ioapic_id(ioapic_idx);
restore_ioapic_entries();
}
@@ -3144,45 +3204,6 @@ msi_set_affinity(struct irq_data *data, const struct cpumask *mask, bool force)
return 0;
}
-#ifdef CONFIG_INTR_REMAP
-/*
- * Migrate the MSI irq to another cpumask. This migration is
- * done in the process context using interrupt-remapping hardware.
- */
-static int
-ir_msi_set_affinity(struct irq_data *data, const struct cpumask *mask,
- bool force)
-{
- struct irq_cfg *cfg = data->chip_data;
- unsigned int dest, irq = data->irq;
- struct irte irte;
-
- if (get_irte(irq, &irte))
- return -1;
-
- if (__ioapic_set_affinity(data, mask, &dest))
- return -1;
-
- irte.vector = cfg->vector;
- irte.dest_id = IRTE_DEST(dest);
-
- /*
- * atomically update the IRTE with the new destination and vector.
- */
- modify_irte(irq, &irte);
-
- /*
- * After this point, all the interrupts will start arriving
- * at the new destination. So, time to cleanup the previous
- * vector allocation.
- */
- if (cfg->move_in_progress)
- send_cleanup_vector(cfg);
-
- return 0;
-}
-
-#endif
#endif /* CONFIG_SMP */
/*
@@ -3200,19 +3221,6 @@ static struct irq_chip msi_chip = {
.irq_retrigger = ioapic_retrigger_irq,
};
-static struct irq_chip msi_ir_chip = {
- .name = "IR-PCI-MSI",
- .irq_unmask = unmask_msi_irq,
- .irq_mask = mask_msi_irq,
-#ifdef CONFIG_INTR_REMAP
- .irq_ack = ir_ack_apic_edge,
-#ifdef CONFIG_SMP
- .irq_set_affinity = ir_msi_set_affinity,
-#endif
-#endif
- .irq_retrigger = ioapic_retrigger_irq,
-};
-
/*
* Map the PCI dev to the corresponding remapping hardware unit
* and allocate 'nvec' consecutive interrupt-remapping table entries
@@ -3255,7 +3263,7 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq)
if (irq_remapped(irq_get_chip_data(irq))) {
irq_set_status_flags(irq, IRQ_MOVE_PCNTXT);
- chip = &msi_ir_chip;
+ irq_remap_modify_chip_defaults(chip);
}
irq_set_chip_and_handler_name(irq, chip, handle_edge_irq, "edge");
@@ -3328,7 +3336,7 @@ void native_teardown_msi_irq(unsigned int irq)
destroy_irq(irq);
}
-#if defined (CONFIG_DMAR) || defined (CONFIG_INTR_REMAP)
+#ifdef CONFIG_DMAR_TABLE
#ifdef CONFIG_SMP
static int
dmar_msi_set_affinity(struct irq_data *data, const struct cpumask *mask,
@@ -3409,19 +3417,6 @@ static int hpet_msi_set_affinity(struct irq_data *data,
#endif /* CONFIG_SMP */
-static struct irq_chip ir_hpet_msi_type = {
- .name = "IR-HPET_MSI",
- .irq_unmask = hpet_msi_unmask,
- .irq_mask = hpet_msi_mask,
-#ifdef CONFIG_INTR_REMAP
- .irq_ack = ir_ack_apic_edge,
-#ifdef CONFIG_SMP
- .irq_set_affinity = ir_msi_set_affinity,
-#endif
-#endif
- .irq_retrigger = ioapic_retrigger_irq,
-};
-
static struct irq_chip hpet_msi_type = {
.name = "HPET_MSI",
.irq_unmask = hpet_msi_unmask,
@@ -3458,7 +3453,7 @@ int arch_setup_hpet_msi(unsigned int irq, unsigned int id)
hpet_msi_write(irq_get_handler_data(irq), &msg);
irq_set_status_flags(irq, IRQ_MOVE_PCNTXT);
if (irq_remapped(irq_get_chip_data(irq)))
- chip = &ir_hpet_msi_type;
+ irq_remap_modify_chip_defaults(chip);
irq_set_chip_and_handler_name(irq, chip, handle_edge_irq, "edge");
return 0;
@@ -3566,26 +3561,25 @@ io_apic_setup_irq_pin(unsigned int irq, int node, struct io_apic_irq_attr *attr)
return -EINVAL;
ret = __add_pin_to_irq_node(cfg, node, attr->ioapic, attr->ioapic_pin);
if (!ret)
- setup_ioapic_irq(attr->ioapic, attr->ioapic_pin, irq, cfg,
- attr->trigger, attr->polarity);
+ setup_ioapic_irq(irq, cfg, attr);
return ret;
}
int io_apic_setup_irq_pin_once(unsigned int irq, int node,
struct io_apic_irq_attr *attr)
{
- unsigned int id = attr->ioapic, pin = attr->ioapic_pin;
+ unsigned int ioapic_idx = attr->ioapic, pin = attr->ioapic_pin;
int ret;
/* Avoid redundant programming */
- if (test_bit(pin, ioapics[id].pin_programmed)) {
+ if (test_bit(pin, ioapics[ioapic_idx].pin_programmed)) {
pr_debug("Pin %d-%d already programmed\n",
- mpc_ioapic_id(id), pin);
+ mpc_ioapic_id(ioapic_idx), pin);
return 0;
}
ret = io_apic_setup_irq_pin(irq, node, attr);
if (!ret)
- set_bit(pin, ioapics[id].pin_programmed);
+ set_bit(pin, ioapics[ioapic_idx].pin_programmed);
return ret;
}
@@ -3621,7 +3615,6 @@ int get_nr_irqs_gsi(void)
return nr_irqs_gsi;
}
-#ifdef CONFIG_SPARSE_IRQ
int __init arch_probe_nr_irqs(void)
{
int nr;
@@ -3641,7 +3634,6 @@ int __init arch_probe_nr_irqs(void)
return NR_IRQS_LEGACY;
}
-#endif
int io_apic_set_pci_routing(struct device *dev, int irq,
struct io_apic_irq_attr *irq_attr)
diff --git a/arch/x86/kernel/apic/probe_32.c b/arch/x86/kernel/apic/probe_32.c
index b5254ad044a..0787bb3412f 100644
--- a/arch/x86/kernel/apic/probe_32.c
+++ b/arch/x86/kernel/apic/probe_32.c
@@ -200,14 +200,8 @@ void __init default_setup_apic_routing(void)
* - we find more than 8 CPUs in acpi LAPIC listing with xAPIC support
*/
- if (!cmdline_apic && apic == &apic_default) {
- struct apic *bigsmp = generic_bigsmp_probe();
- if (bigsmp) {
- apic = bigsmp;
- printk(KERN_INFO "Overriding APIC driver with %s\n",
- apic->name);
- }
- }
+ if (!cmdline_apic && apic == &apic_default)
+ generic_bigsmp_probe();
#endif
if (apic->setup_apic_routing)
diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c
index 34b18594e72..62ae3001ae0 100644
--- a/arch/x86/kernel/apic/x2apic_uv_x.c
+++ b/arch/x86/kernel/apic/x2apic_uv_x.c
@@ -672,18 +672,11 @@ void __cpuinit uv_cpu_init(void)
/*
* When NMI is received, print a stack trace.
*/
-int uv_handle_nmi(struct notifier_block *self, unsigned long reason, void *data)
+int uv_handle_nmi(unsigned int reason, struct pt_regs *regs)
{
unsigned long real_uv_nmi;
int bid;
- if (reason != DIE_NMIUNKNOWN)
- return NOTIFY_OK;
-
- if (in_crash_kexec)
- /* do nothing if entering the crash kernel */
- return NOTIFY_OK;
-
/*
* Each blade has an MMR that indicates when an NMI has been sent
* to cpus on the blade. If an NMI is detected, atomically
@@ -704,7 +697,7 @@ int uv_handle_nmi(struct notifier_block *self, unsigned long reason, void *data)
}
if (likely(__get_cpu_var(cpu_last_nmi_count) == uv_blade_info[bid].nmi_count))
- return NOTIFY_DONE;
+ return NMI_DONE;
__get_cpu_var(cpu_last_nmi_count) = uv_blade_info[bid].nmi_count;
@@ -717,17 +710,12 @@ int uv_handle_nmi(struct notifier_block *self, unsigned long reason, void *data)
dump_stack();
spin_unlock(&uv_nmi_lock);
- return NOTIFY_STOP;
+ return NMI_HANDLED;
}
-static struct notifier_block uv_dump_stack_nmi_nb = {
- .notifier_call = uv_handle_nmi,
- .priority = NMI_LOCAL_LOW_PRIOR - 1,
-};
-
void uv_register_nmi_notifier(void)
{
- if (register_die_notifier(&uv_dump_stack_nmi_nb))
+ if (register_nmi_handler(NMI_UNKNOWN, uv_handle_nmi, 0, "uv"))
printk(KERN_WARNING "UV NMI handler failed to register\n");
}
@@ -832,6 +820,10 @@ void __init uv_system_init(void)
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;
+ uv_cpu_hub_info(cpu)->m_shift = 64 - m_val;
+ uv_cpu_hub_info(cpu)->n_lshift = is_uv2_1_hub() ?
+ (m_val == 40 ? 40 : 39) : m_val;
+
pnode = uv_apicid_to_pnode(apicid);
blade = boot_pnode_to_blade(pnode);
lcpu = uv_blade_info[blade].nr_possible_cpus;
@@ -862,8 +854,7 @@ void __init uv_system_init(void)
if (uv_node_to_blade[nid] >= 0)
continue;
paddr = node_start_pfn(nid) << PAGE_SHIFT;
- paddr = uv_soc_phys_ram_to_gpa(paddr);
- pnode = (paddr >> m_val) & pnode_mask;
+ pnode = uv_gpa_to_pnode(uv_soc_phys_ram_to_gpa(paddr));
blade = boot_pnode_to_blade(pnode);
uv_node_to_blade[nid] = blade;
}
diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c
index 0371c484bb8..a46bd383953 100644
--- a/arch/x86/kernel/apm_32.c
+++ b/arch/x86/kernel/apm_32.c
@@ -249,8 +249,6 @@ extern int (*console_blank_hook)(int);
#define APM_MINOR_DEV 134
/*
- * See Documentation/Config.help for the configuration options.
- *
* Various options can be changed at boot time as follows:
* (We allow underscores for compatibility with the modules code)
* apm=on/off enable/disable APM
diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile
index 6042981d030..25f24dccdcf 100644
--- a/arch/x86/kernel/cpu/Makefile
+++ b/arch/x86/kernel/cpu/Makefile
@@ -15,6 +15,7 @@ CFLAGS_common.o := $(nostackp)
obj-y := intel_cacheinfo.o scattered.o topology.o
obj-y += proc.o capflags.o powerflags.o common.o
obj-y += vmware.o hypervisor.o sched.o mshyperv.o
+obj-y += rdrand.o
obj-$(CONFIG_X86_32) += bugs.o
obj-$(CONFIG_X86_64) += bugs_64.o
@@ -28,10 +29,15 @@ obj-$(CONFIG_CPU_SUP_UMC_32) += umc.o
obj-$(CONFIG_PERF_EVENTS) += perf_event.o
+ifdef CONFIG_PERF_EVENTS
+obj-$(CONFIG_CPU_SUP_AMD) += perf_event_amd.o
+obj-$(CONFIG_CPU_SUP_INTEL) += perf_event_p6.o perf_event_p4.o perf_event_intel_lbr.o perf_event_intel_ds.o perf_event_intel.o
+endif
+
obj-$(CONFIG_X86_MCE) += mcheck/
obj-$(CONFIG_MTRR) += mtrr/
-obj-$(CONFIG_X86_LOCAL_APIC) += perfctr-watchdog.o
+obj-$(CONFIG_X86_LOCAL_APIC) += perfctr-watchdog.o perf_event_amd_ibs.o
quiet_cmd_mkcapflags = MKCAP $@
cmd_mkcapflags = $(PERL) $(srctree)/$(src)/mkcapflags.pl $< $@
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index b13ed393dfc..46ae4f65fc7 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -1,5 +1,6 @@
#include <linux/init.h>
#include <linux/bitops.h>
+#include <linux/elf.h>
#include <linux/mm.h>
#include <linux/io.h>
@@ -410,8 +411,38 @@ static void __cpuinit early_init_amd_mc(struct cpuinfo_x86 *c)
#endif
}
+static void __cpuinit bsp_init_amd(struct cpuinfo_x86 *c)
+{
+ if (cpu_has(c, X86_FEATURE_CONSTANT_TSC)) {
+
+ if (c->x86 > 0x10 ||
+ (c->x86 == 0x10 && c->x86_model >= 0x2)) {
+ u64 val;
+
+ rdmsrl(MSR_K7_HWCR, val);
+ if (!(val & BIT(24)))
+ printk(KERN_WARNING FW_BUG "TSC doesn't count "
+ "with P0 frequency!\n");
+ }
+ }
+
+ if (c->x86 == 0x15) {
+ unsigned long upperbit;
+ u32 cpuid, assoc;
+
+ cpuid = cpuid_edx(0x80000005);
+ assoc = cpuid >> 16 & 0xff;
+ upperbit = ((cpuid >> 24) << 10) / assoc;
+
+ va_align.mask = (upperbit - 1) & PAGE_MASK;
+ va_align.flags = ALIGN_VA_32 | ALIGN_VA_64;
+ }
+}
+
static void __cpuinit early_init_amd(struct cpuinfo_x86 *c)
{
+ u32 dummy;
+
early_init_amd_mc(c);
/*
@@ -442,22 +473,7 @@ static void __cpuinit early_init_amd(struct cpuinfo_x86 *c)
}
#endif
- /* We need to do the following only once */
- if (c != &boot_cpu_data)
- return;
-
- if (cpu_has(c, X86_FEATURE_CONSTANT_TSC)) {
-
- if (c->x86 > 0x10 ||
- (c->x86 == 0x10 && c->x86_model >= 0x2)) {
- u64 val;
-
- rdmsrl(MSR_K7_HWCR, val);
- if (!(val & BIT(24)))
- printk(KERN_WARNING FW_BUG "TSC doesn't count "
- "with P0 frequency!\n");
- }
- }
+ rdmsr_safe(MSR_AMD64_PATCH_LEVEL, &c->microcode, &dummy);
}
static void __cpuinit init_amd(struct cpuinfo_x86 *c)
@@ -679,6 +695,7 @@ static const struct cpu_dev __cpuinitconst amd_cpu_dev = {
.c_size_cache = amd_size_cache,
#endif
.c_early_init = early_init_amd,
+ .c_bsp_init = bsp_init_amd,
.c_init = init_amd,
.c_x86_vendor = X86_VENDOR_AMD,
};
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 62184390a60..aa003b13a83 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -15,6 +15,7 @@
#include <asm/stackprotector.h>
#include <asm/perf_event.h>
#include <asm/mmu_context.h>
+#include <asm/archrandom.h>
#include <asm/hypervisor.h>
#include <asm/processor.h>
#include <asm/sections.h>
@@ -681,6 +682,9 @@ static void __init early_identify_cpu(struct cpuinfo_x86 *c)
filter_cpuid_features(c, false);
setup_smep(c);
+
+ if (this_cpu->c_bsp_init)
+ this_cpu->c_bsp_init(c);
}
void __init early_cpu_init(void)
@@ -857,6 +861,7 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
#endif
init_hypervisor(c);
+ x86_init_rdrand(c);
/*
* Clear/Set all flags overriden by options, need do it
diff --git a/arch/x86/kernel/cpu/cpu.h b/arch/x86/kernel/cpu/cpu.h
index e765633f210..1b22dcc51af 100644
--- a/arch/x86/kernel/cpu/cpu.h
+++ b/arch/x86/kernel/cpu/cpu.h
@@ -18,6 +18,7 @@ struct cpu_dev {
struct cpu_model_info c_models[4];
void (*c_early_init)(struct cpuinfo_x86 *);
+ void (*c_bsp_init)(struct cpuinfo_x86 *);
void (*c_init)(struct cpuinfo_x86 *);
void (*c_identify)(struct cpuinfo_x86 *);
unsigned int (*c_size_cache)(struct cpuinfo_x86 *, unsigned int);
diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c
index ed6086eedf1..523131213f0 100644
--- a/arch/x86/kernel/cpu/intel.c
+++ b/arch/x86/kernel/cpu/intel.c
@@ -47,6 +47,15 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c)
(c->x86 == 0x6 && c->x86_model >= 0x0e))
set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
+ if (c->x86 >= 6 && !cpu_has(c, X86_FEATURE_IA64)) {
+ unsigned lower_word;
+
+ wrmsr(MSR_IA32_UCODE_REV, 0, 0);
+ /* Required by the SDM */
+ sync_core();
+ rdmsr(MSR_IA32_UCODE_REV, lower_word, c->microcode);
+ }
+
/*
* Atom erratum AAE44/AAF40/AAG38/AAH41:
*
@@ -55,17 +64,10 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c)
* need the microcode to have already been loaded... so if it is
* not, recommend a BIOS update and disable large pages.
*/
- if (c->x86 == 6 && c->x86_model == 0x1c && c->x86_mask <= 2) {
- u32 ucode, junk;
-
- wrmsr(MSR_IA32_UCODE_REV, 0, 0);
- sync_core();
- rdmsr(MSR_IA32_UCODE_REV, junk, ucode);
-
- if (ucode < 0x20e) {
- printk(KERN_WARNING "Atom PSE erratum detected, BIOS microcode update recommended\n");
- clear_cpu_cap(c, X86_FEATURE_PSE);
- }
+ if (c->x86 == 6 && c->x86_model == 0x1c && c->x86_mask <= 2 &&
+ c->microcode < 0x20e) {
+ printk(KERN_WARNING "Atom PSE erratum detected, BIOS microcode update recommended\n");
+ clear_cpu_cap(c, X86_FEATURE_PSE);
}
#ifdef CONFIG_X86_64
diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
index c105c533ed9..a3b0811693c 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -151,28 +151,17 @@ union _cpuid4_leaf_ecx {
u32 full;
};
-struct amd_l3_cache {
- struct amd_northbridge *nb;
- unsigned indices;
- u8 subcaches[4];
-};
-
-struct _cpuid4_info {
+struct _cpuid4_info_regs {
union _cpuid4_leaf_eax eax;
union _cpuid4_leaf_ebx ebx;
union _cpuid4_leaf_ecx ecx;
unsigned long size;
- struct amd_l3_cache *l3;
- DECLARE_BITMAP(shared_cpu_map, NR_CPUS);
+ struct amd_northbridge *nb;
};
-/* subset of above _cpuid4_info w/o shared_cpu_map */
-struct _cpuid4_info_regs {
- union _cpuid4_leaf_eax eax;
- union _cpuid4_leaf_ebx ebx;
- union _cpuid4_leaf_ecx ecx;
- unsigned long size;
- struct amd_l3_cache *l3;
+struct _cpuid4_info {
+ struct _cpuid4_info_regs base;
+ DECLARE_BITMAP(shared_cpu_map, NR_CPUS);
};
unsigned short num_cache_leaves;
@@ -314,16 +303,23 @@ struct _cache_attr {
/*
* L3 cache descriptors
*/
-static void __cpuinit amd_calc_l3_indices(struct amd_l3_cache *l3)
+static void __cpuinit amd_calc_l3_indices(struct amd_northbridge *nb)
{
+ struct amd_l3_cache *l3 = &nb->l3_cache;
unsigned int sc0, sc1, sc2, sc3;
u32 val = 0;
- pci_read_config_dword(l3->nb->misc, 0x1C4, &val);
+ pci_read_config_dword(nb->misc, 0x1C4, &val);
/* calculate subcache sizes */
l3->subcaches[0] = sc0 = !(val & BIT(0));
l3->subcaches[1] = sc1 = !(val & BIT(4));
+
+ if (boot_cpu_data.x86 == 0x15) {
+ l3->subcaches[0] = sc0 += !(val & BIT(1));
+ l3->subcaches[1] = sc1 += !(val & BIT(5));
+ }
+
l3->subcaches[2] = sc2 = !(val & BIT(8)) + !(val & BIT(9));
l3->subcaches[3] = sc3 = !(val & BIT(12)) + !(val & BIT(13));
@@ -333,33 +329,16 @@ static void __cpuinit amd_calc_l3_indices(struct amd_l3_cache *l3)
static void __cpuinit amd_init_l3_cache(struct _cpuid4_info_regs *this_leaf,
int index)
{
- static struct amd_l3_cache *__cpuinitdata l3_caches;
int node;
/* only for L3, and not in virtualized environments */
- if (index < 3 || amd_nb_num() == 0)
+ if (index < 3)
return;
- /*
- * Strictly speaking, the amount in @size below is leaked since it is
- * never freed but this is done only on shutdown so it doesn't matter.
- */
- if (!l3_caches) {
- int size = amd_nb_num() * sizeof(struct amd_l3_cache);
-
- l3_caches = kzalloc(size, GFP_ATOMIC);
- if (!l3_caches)
- return;
- }
-
node = amd_get_nb_id(smp_processor_id());
-
- if (!l3_caches[node].nb) {
- l3_caches[node].nb = node_to_amd_nb(node);
- amd_calc_l3_indices(&l3_caches[node]);
- }
-
- this_leaf->l3 = &l3_caches[node];
+ this_leaf->nb = node_to_amd_nb(node);
+ if (this_leaf->nb && !this_leaf->nb->l3_cache.indices)
+ amd_calc_l3_indices(this_leaf->nb);
}
/*
@@ -369,11 +348,11 @@ static void __cpuinit amd_init_l3_cache(struct _cpuid4_info_regs *this_leaf,
*
* @returns: the disabled index if used or negative value if slot free.
*/
-int amd_get_l3_disable_slot(struct amd_l3_cache *l3, unsigned slot)
+int amd_get_l3_disable_slot(struct amd_northbridge *nb, unsigned slot)
{
unsigned int reg = 0;
- pci_read_config_dword(l3->nb->misc, 0x1BC + slot * 4, &reg);
+ pci_read_config_dword(nb->misc, 0x1BC + slot * 4, &reg);
/* check whether this slot is activated already */
if (reg & (3UL << 30))
@@ -387,11 +366,10 @@ static ssize_t show_cache_disable(struct _cpuid4_info *this_leaf, char *buf,
{
int index;
- if (!this_leaf->l3 ||
- !amd_nb_has_feature(AMD_NB_L3_INDEX_DISABLE))
+ if (!this_leaf->base.nb || !amd_nb_has_feature(AMD_NB_L3_INDEX_DISABLE))
return -EINVAL;
- index = amd_get_l3_disable_slot(this_leaf->l3, slot);
+ index = amd_get_l3_disable_slot(this_leaf->base.nb, slot);
if (index >= 0)
return sprintf(buf, "%d\n", index);
@@ -408,7 +386,7 @@ show_cache_disable_##slot(struct _cpuid4_info *this_leaf, char *buf, \
SHOW_CACHE_DISABLE(0)
SHOW_CACHE_DISABLE(1)
-static void amd_l3_disable_index(struct amd_l3_cache *l3, int cpu,
+static void amd_l3_disable_index(struct amd_northbridge *nb, int cpu,
unsigned slot, unsigned long idx)
{
int i;
@@ -421,10 +399,10 @@ static void amd_l3_disable_index(struct amd_l3_cache *l3, int cpu,
for (i = 0; i < 4; i++) {
u32 reg = idx | (i << 20);
- if (!l3->subcaches[i])
+ if (!nb->l3_cache.subcaches[i])
continue;
- pci_write_config_dword(l3->nb->misc, 0x1BC + slot * 4, reg);
+ pci_write_config_dword(nb->misc, 0x1BC + slot * 4, reg);
/*
* We need to WBINVD on a core on the node containing the L3
@@ -434,7 +412,7 @@ static void amd_l3_disable_index(struct amd_l3_cache *l3, int cpu,
wbinvd_on_cpu(cpu);
reg |= BIT(31);
- pci_write_config_dword(l3->nb->misc, 0x1BC + slot * 4, reg);
+ pci_write_config_dword(nb->misc, 0x1BC + slot * 4, reg);
}
}
@@ -448,24 +426,24 @@ static void amd_l3_disable_index(struct amd_l3_cache *l3, int cpu,
*
* @return: 0 on success, error status on failure
*/
-int amd_set_l3_disable_slot(struct amd_l3_cache *l3, int cpu, unsigned slot,
+int amd_set_l3_disable_slot(struct amd_northbridge *nb, int cpu, unsigned slot,
unsigned long index)
{
int ret = 0;
/* check if @slot is already used or the index is already disabled */
- ret = amd_get_l3_disable_slot(l3, slot);
+ ret = amd_get_l3_disable_slot(nb, slot);
if (ret >= 0)
return -EINVAL;
- if (index > l3->indices)
+ if (index > nb->l3_cache.indices)
return -EINVAL;
/* check whether the other slot has disabled the same index already */
- if (index == amd_get_l3_disable_slot(l3, !slot))
+ if (index == amd_get_l3_disable_slot(nb, !slot))
return -EINVAL;
- amd_l3_disable_index(l3, cpu, slot, index);
+ amd_l3_disable_index(nb, cpu, slot, index);
return 0;
}
@@ -480,8 +458,7 @@ static ssize_t store_cache_disable(struct _cpuid4_info *this_leaf,
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
- if (!this_leaf->l3 ||
- !amd_nb_has_feature(AMD_NB_L3_INDEX_DISABLE))
+ if (!this_leaf->base.nb || !amd_nb_has_feature(AMD_NB_L3_INDEX_DISABLE))
return -EINVAL;
cpu = cpumask_first(to_cpumask(this_leaf->shared_cpu_map));
@@ -489,7 +466,7 @@ static ssize_t store_cache_disable(struct _cpuid4_info *this_leaf,
if (strict_strtoul(buf, 10, &val) < 0)
return -EINVAL;
- err = amd_set_l3_disable_slot(this_leaf->l3, cpu, slot, val);
+ err = amd_set_l3_disable_slot(this_leaf->base.nb, cpu, slot, val);
if (err) {
if (err == -EEXIST)
printk(KERN_WARNING "L3 disable slot %d in use!\n",
@@ -518,7 +495,7 @@ static struct _cache_attr cache_disable_1 = __ATTR(cache_disable_1, 0644,
static ssize_t
show_subcaches(struct _cpuid4_info *this_leaf, char *buf, unsigned int cpu)
{
- if (!this_leaf->l3 || !amd_nb_has_feature(AMD_NB_L3_PARTITIONING))
+ if (!this_leaf->base.nb || !amd_nb_has_feature(AMD_NB_L3_PARTITIONING))
return -EINVAL;
return sprintf(buf, "%x\n", amd_get_subcaches(cpu));
@@ -533,7 +510,7 @@ store_subcaches(struct _cpuid4_info *this_leaf, const char *buf, size_t count,
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
- if (!this_leaf->l3 || !amd_nb_has_feature(AMD_NB_L3_PARTITIONING))
+ if (!this_leaf->base.nb || !amd_nb_has_feature(AMD_NB_L3_PARTITIONING))
return -EINVAL;
if (strict_strtoul(buf, 16, &val) < 0)
@@ -769,7 +746,7 @@ static void __cpuinit cache_shared_cpu_map_setup(unsigned int cpu, int index)
return;
}
this_leaf = CPUID4_INFO_IDX(cpu, index);
- num_threads_sharing = 1 + this_leaf->eax.split.num_threads_sharing;
+ num_threads_sharing = 1 + this_leaf->base.eax.split.num_threads_sharing;
if (num_threads_sharing == 1)
cpumask_set_cpu(cpu, to_cpumask(this_leaf->shared_cpu_map));
@@ -820,29 +797,19 @@ static void __cpuinit free_cache_attributes(unsigned int cpu)
for (i = 0; i < num_cache_leaves; i++)
cache_remove_shared_cpu_map(cpu, i);
- kfree(per_cpu(ici_cpuid4_info, cpu)->l3);
kfree(per_cpu(ici_cpuid4_info, cpu));
per_cpu(ici_cpuid4_info, cpu) = NULL;
}
-static int
-__cpuinit cpuid4_cache_lookup(int index, struct _cpuid4_info *this_leaf)
-{
- struct _cpuid4_info_regs *leaf_regs =
- (struct _cpuid4_info_regs *)this_leaf;
-
- return cpuid4_cache_lookup_regs(index, leaf_regs);
-}
-
static void __cpuinit get_cpu_leaves(void *_retval)
{
int j, *retval = _retval, cpu = smp_processor_id();
/* Do cpuid and store the results */
for (j = 0; j < num_cache_leaves; j++) {
- struct _cpuid4_info *this_leaf;
- this_leaf = CPUID4_INFO_IDX(cpu, j);
- *retval = cpuid4_cache_lookup(j, this_leaf);
+ struct _cpuid4_info *this_leaf = CPUID4_INFO_IDX(cpu, j);
+
+ *retval = cpuid4_cache_lookup_regs(j, &this_leaf->base);
if (unlikely(*retval < 0)) {
int i;
@@ -900,16 +867,16 @@ 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);
+show_one_plus(level, base.eax.split.level, 0);
+show_one_plus(coherency_line_size, base.ebx.split.coherency_line_size, 1);
+show_one_plus(physical_line_partition, base.ebx.split.physical_line_partition, 1);
+show_one_plus(ways_of_associativity, base.ebx.split.ways_of_associativity, 1);
+show_one_plus(number_of_sets, base.ecx.split.number_of_sets, 1);
static ssize_t show_size(struct _cpuid4_info *this_leaf, char *buf,
unsigned int cpu)
{
- return sprintf(buf, "%luK\n", this_leaf->size / 1024);
+ return sprintf(buf, "%luK\n", this_leaf->base.size / 1024);
}
static ssize_t show_shared_cpu_map_func(struct _cpuid4_info *this_leaf,
@@ -946,7 +913,7 @@ static inline ssize_t show_shared_cpu_list(struct _cpuid4_info *leaf, char *buf,
static ssize_t show_type(struct _cpuid4_info *this_leaf, char *buf,
unsigned int cpu)
{
- switch (this_leaf->eax.split.type) {
+ switch (this_leaf->base.eax.split.type) {
case CACHE_TYPE_DATA:
return sprintf(buf, "Data\n");
case CACHE_TYPE_INST:
@@ -1135,7 +1102,7 @@ static int __cpuinit cache_add_dev(struct sys_device * sys_dev)
ktype_cache.default_attrs = default_attrs;
#ifdef CONFIG_AMD_NB
- if (this_leaf->l3)
+ if (this_leaf->base.nb)
ktype_cache.default_attrs = amd_l3_attrs();
#endif
retval = kobject_init_and_add(&(this_object->kobj),
diff --git a/arch/x86/kernel/cpu/mcheck/mce-inject.c b/arch/x86/kernel/cpu/mcheck/mce-inject.c
index 0ed633c5048..6199232161c 100644
--- a/arch/x86/kernel/cpu/mcheck/mce-inject.c
+++ b/arch/x86/kernel/cpu/mcheck/mce-inject.c
@@ -78,27 +78,20 @@ static void raise_exception(struct mce *m, struct pt_regs *pregs)
static cpumask_var_t mce_inject_cpumask;
-static int mce_raise_notify(struct notifier_block *self,
- unsigned long val, void *data)
+static int mce_raise_notify(unsigned int cmd, struct pt_regs *regs)
{
- struct die_args *args = (struct die_args *)data;
int cpu = smp_processor_id();
struct mce *m = &__get_cpu_var(injectm);
- if (val != DIE_NMI || !cpumask_test_cpu(cpu, mce_inject_cpumask))
- return NOTIFY_DONE;
+ if (!cpumask_test_cpu(cpu, mce_inject_cpumask))
+ return NMI_DONE;
cpumask_clear_cpu(cpu, mce_inject_cpumask);
if (m->inject_flags & MCJ_EXCEPTION)
- raise_exception(m, args->regs);
+ raise_exception(m, regs);
else if (m->status)
raise_poll(m);
- return NOTIFY_STOP;
+ return NMI_HANDLED;
}
-static struct notifier_block mce_raise_nb = {
- .notifier_call = mce_raise_notify,
- .priority = NMI_LOCAL_NORMAL_PRIOR,
-};
-
/* Inject mce on current CPU */
static int raise_local(void)
{
@@ -216,7 +209,8 @@ static int inject_init(void)
return -ENOMEM;
printk(KERN_INFO "Machine check injector initialized\n");
mce_chrdev_ops.write = mce_write;
- register_die_notifier(&mce_raise_nb);
+ register_nmi_handler(NMI_LOCAL, mce_raise_notify, 0,
+ "mce_notify");
return 0;
}
diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c
index 08363b04212..864830e1dd6 100644
--- a/arch/x86/kernel/cpu/mcheck/mce.c
+++ b/arch/x86/kernel/cpu/mcheck/mce.c
@@ -36,7 +36,6 @@
#include <linux/fs.h>
#include <linux/mm.h>
#include <linux/debugfs.h>
-#include <linux/edac_mce.h>
#include <linux/irq_work.h>
#include <asm/processor.h>
@@ -144,23 +143,20 @@ static struct mce_log mcelog = {
void mce_log(struct mce *mce)
{
unsigned next, entry;
+ int ret = 0;
/* Emit the trace record: */
trace_mce_record(mce);
+ ret = atomic_notifier_call_chain(&x86_mce_decoder_chain, 0, mce);
+ if (ret == NOTIFY_STOP)
+ return;
+
mce->finished = 0;
wmb();
for (;;) {
entry = rcu_dereference_check_mce(mcelog.next);
for (;;) {
- /*
- * If edac_mce is enabled, it will check the error type
- * and will process it, if it is a known error.
- * Otherwise, the error will be sent through mcelog
- * interface
- */
- if (edac_mce_parse(mce))
- return;
/*
* When the buffer fills up discard new entries.
@@ -217,8 +213,13 @@ static void print_mce(struct mce *m)
pr_cont("MISC %llx ", m->misc);
pr_cont("\n");
- pr_emerg(HW_ERR "PROCESSOR %u:%x TIME %llu SOCKET %u APIC %x\n",
- m->cpuvendor, m->cpuid, m->time, m->socketid, m->apicid);
+ /*
+ * Note this output is parsed by external tools and old fields
+ * should not be changed.
+ */
+ pr_emerg(HW_ERR "PROCESSOR %u:%x TIME %llu SOCKET %u APIC %x microcode %x\n",
+ m->cpuvendor, m->cpuid, m->time, m->socketid, m->apicid,
+ cpu_data(m->extcpu).microcode);
/*
* Print out human-readable details about the MCE error,
@@ -551,10 +552,8 @@ void machine_check_poll(enum mcp_flags flags, mce_banks_t *b)
* Don't get the IP here because it's unlikely to
* have anything to do with the actual error location.
*/
- if (!(flags & MCP_DONTLOG) && !mce_dont_log_ce) {
+ if (!(flags & MCP_DONTLOG) && !mce_dont_log_ce)
mce_log(&m);
- atomic_notifier_call_chain(&x86_mce_decoder_chain, 0, &m);
- }
/*
* Clear state for this bank.
@@ -908,9 +907,6 @@ void do_machine_check(struct pt_regs *regs, long error_code)
percpu_inc(mce_exception_count);
- if (notify_die(DIE_NMI, "machine check", regs, error_code,
- 18, SIGKILL) == NOTIFY_STOP)
- goto out;
if (!banks)
goto out;
@@ -1140,6 +1136,15 @@ static void mce_start_timer(unsigned long data)
add_timer_on(t, smp_processor_id());
}
+/* Must not be called in IRQ context where del_timer_sync() can deadlock */
+static void mce_timer_delete_all(void)
+{
+ int cpu;
+
+ for_each_online_cpu(cpu)
+ del_timer_sync(&per_cpu(mce_timer, cpu));
+}
+
static void mce_do_trigger(struct work_struct *work)
{
call_usermodehelper(mce_helper, mce_helper_argv, NULL, UMH_NO_WAIT);
@@ -1750,7 +1755,6 @@ static struct syscore_ops mce_syscore_ops = {
static void mce_cpu_restart(void *data)
{
- del_timer_sync(&__get_cpu_var(mce_timer));
if (!mce_available(__this_cpu_ptr(&cpu_info)))
return;
__mcheck_cpu_init_generic();
@@ -1760,16 +1764,15 @@ static void mce_cpu_restart(void *data)
/* Reinit MCEs after user configuration changes */
static void mce_restart(void)
{
+ mce_timer_delete_all();
on_each_cpu(mce_cpu_restart, NULL, 1);
}
/* Toggle features for corrected errors */
-static void mce_disable_ce(void *all)
+static void mce_disable_cmci(void *data)
{
if (!mce_available(__this_cpu_ptr(&cpu_info)))
return;
- if (all)
- del_timer_sync(&__get_cpu_var(mce_timer));
cmci_clear();
}
@@ -1852,7 +1855,8 @@ static ssize_t set_ignore_ce(struct sys_device *s,
if (mce_ignore_ce ^ !!new) {
if (new) {
/* disable ce features */
- on_each_cpu(mce_disable_ce, (void *)1, 1);
+ mce_timer_delete_all();
+ on_each_cpu(mce_disable_cmci, NULL, 1);
mce_ignore_ce = 1;
} else {
/* enable ce features */
@@ -1875,7 +1879,7 @@ static ssize_t set_cmci_disabled(struct sys_device *s,
if (mce_cmci_disabled ^ !!new) {
if (new) {
/* disable cmci */
- on_each_cpu(mce_disable_ce, NULL, 1);
+ on_each_cpu(mce_disable_cmci, NULL, 1);
mce_cmci_disabled = 1;
} else {
/* enable cmci */
diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel.c b/arch/x86/kernel/cpu/mcheck/mce_intel.c
index 8694ef56459..38e49bc95ff 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_intel.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_intel.c
@@ -28,7 +28,7 @@ static DEFINE_PER_CPU(mce_banks_t, mce_banks_owned);
* cmci_discover_lock protects against parallel discovery attempts
* which could race against each other.
*/
-static DEFINE_SPINLOCK(cmci_discover_lock);
+static DEFINE_RAW_SPINLOCK(cmci_discover_lock);
#define CMCI_THRESHOLD 1
@@ -85,7 +85,7 @@ static void cmci_discover(int banks, int boot)
int hdr = 0;
int i;
- spin_lock_irqsave(&cmci_discover_lock, flags);
+ raw_spin_lock_irqsave(&cmci_discover_lock, flags);
for (i = 0; i < banks; i++) {
u64 val;
@@ -116,7 +116,7 @@ static void cmci_discover(int banks, int boot)
WARN_ON(!test_bit(i, __get_cpu_var(mce_poll_banks)));
}
}
- spin_unlock_irqrestore(&cmci_discover_lock, flags);
+ raw_spin_unlock_irqrestore(&cmci_discover_lock, flags);
if (hdr)
printk(KERN_CONT "\n");
}
@@ -150,7 +150,7 @@ void cmci_clear(void)
if (!cmci_supported(&banks))
return;
- spin_lock_irqsave(&cmci_discover_lock, flags);
+ raw_spin_lock_irqsave(&cmci_discover_lock, flags);
for (i = 0; i < banks; i++) {
if (!test_bit(i, __get_cpu_var(mce_banks_owned)))
continue;
@@ -160,7 +160,7 @@ void cmci_clear(void)
wrmsrl(MSR_IA32_MCx_CTL2(i), val);
__clear_bit(i, __get_cpu_var(mce_banks_owned));
}
- spin_unlock_irqrestore(&cmci_discover_lock, flags);
+ raw_spin_unlock_irqrestore(&cmci_discover_lock, flags);
}
/*
diff --git a/arch/x86/kernel/cpu/mshyperv.c b/arch/x86/kernel/cpu/mshyperv.c
index d944bf6c50e..0a630dd4b62 100644
--- a/arch/x86/kernel/cpu/mshyperv.c
+++ b/arch/x86/kernel/cpu/mshyperv.c
@@ -11,6 +11,8 @@
*/
#include <linux/types.h>
+#include <linux/time.h>
+#include <linux/clocksource.h>
#include <linux/module.h>
#include <asm/processor.h>
#include <asm/hypervisor.h>
@@ -36,6 +38,25 @@ static bool __init ms_hyperv_platform(void)
!memcmp("Microsoft Hv", hyp_signature, 12);
}
+static cycle_t read_hv_clock(struct clocksource *arg)
+{
+ cycle_t current_tick;
+ /*
+ * Read the partition counter to get the current tick count. This count
+ * is set to 0 when the partition is created and is incremented in
+ * 100 nanosecond units.
+ */
+ rdmsrl(HV_X64_MSR_TIME_REF_COUNT, current_tick);
+ return current_tick;
+}
+
+static struct clocksource hyperv_cs = {
+ .name = "hyperv_clocksource",
+ .rating = 400, /* use this when running on Hyperv*/
+ .read = read_hv_clock,
+ .mask = CLOCKSOURCE_MASK(64),
+};
+
static void __init ms_hyperv_init_platform(void)
{
/*
@@ -46,6 +67,8 @@ static void __init ms_hyperv_init_platform(void)
printk(KERN_INFO "HyperV: features 0x%x, hints 0x%x\n",
ms_hyperv.features, ms_hyperv.hints);
+
+ clocksource_register_hz(&hyperv_cs, NSEC_PER_SEC/100);
}
const __refconst struct hypervisor_x86 x86_hyper_ms_hyperv = {
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 4ee3abf20ed..640891014b2 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -32,6 +32,8 @@
#include <asm/smp.h>
#include <asm/alternative.h>
+#include "perf_event.h"
+
#if 0
#undef wrmsrl
#define wrmsrl(msr, val) \
@@ -43,283 +45,17 @@ do { \
} while (0)
#endif
-/*
- * | NHM/WSM | SNB |
- * register -------------------------------
- * | HT | no HT | HT | no HT |
- *-----------------------------------------
- * offcore | core | core | cpu | core |
- * lbr_sel | core | core | cpu | core |
- * ld_lat | cpu | core | cpu | core |
- *-----------------------------------------
- *
- * Given that there is a small number of shared regs,
- * we can pre-allocate their slot in the per-cpu
- * per-core reg tables.
- */
-enum extra_reg_type {
- EXTRA_REG_NONE = -1, /* not used */
-
- EXTRA_REG_RSP_0 = 0, /* offcore_response_0 */
- EXTRA_REG_RSP_1 = 1, /* offcore_response_1 */
-
- EXTRA_REG_MAX /* number of entries needed */
-};
-
-struct event_constraint {
- union {
- unsigned long idxmsk[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
- u64 idxmsk64;
- };
- u64 code;
- u64 cmask;
- int weight;
-};
-
-struct amd_nb {
- int nb_id; /* NorthBridge id */
- int refcnt; /* reference count */
- struct perf_event *owners[X86_PMC_IDX_MAX];
- struct event_constraint event_constraints[X86_PMC_IDX_MAX];
-};
-
-struct intel_percore;
-
-#define MAX_LBR_ENTRIES 16
-
-struct cpu_hw_events {
- /*
- * Generic x86 PMC bits
- */
- struct perf_event *events[X86_PMC_IDX_MAX]; /* in counter order */
- unsigned long active_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
- unsigned long running[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
- int enabled;
-
- int n_events;
- int n_added;
- int n_txn;
- int assign[X86_PMC_IDX_MAX]; /* event to counter assignment */
- u64 tags[X86_PMC_IDX_MAX];
- struct perf_event *event_list[X86_PMC_IDX_MAX]; /* in enabled order */
-
- unsigned int group_flag;
-
- /*
- * Intel DebugStore bits
- */
- struct debug_store *ds;
- u64 pebs_enabled;
-
- /*
- * Intel LBR bits
- */
- int lbr_users;
- void *lbr_context;
- struct perf_branch_stack lbr_stack;
- struct perf_branch_entry lbr_entries[MAX_LBR_ENTRIES];
-
- /*
- * manage shared (per-core, per-cpu) registers
- * used on Intel NHM/WSM/SNB
- */
- struct intel_shared_regs *shared_regs;
-
- /*
- * AMD specific bits
- */
- struct amd_nb *amd_nb;
-};
-
-#define __EVENT_CONSTRAINT(c, n, m, w) {\
- { .idxmsk64 = (n) }, \
- .code = (c), \
- .cmask = (m), \
- .weight = (w), \
-}
-
-#define EVENT_CONSTRAINT(c, n, m) \
- __EVENT_CONSTRAINT(c, n, m, HWEIGHT(n))
-
-/*
- * Constraint on the Event code.
- */
-#define INTEL_EVENT_CONSTRAINT(c, n) \
- EVENT_CONSTRAINT(c, n, ARCH_PERFMON_EVENTSEL_EVENT)
-
-/*
- * Constraint on the Event code + UMask + fixed-mask
- *
- * filter mask to validate fixed counter events.
- * the following filters disqualify for fixed counters:
- * - inv
- * - edge
- * - cnt-mask
- * The other filters are supported by fixed counters.
- * The any-thread option is supported starting with v3.
- */
-#define FIXED_EVENT_CONSTRAINT(c, n) \
- EVENT_CONSTRAINT(c, (1ULL << (32+n)), X86_RAW_EVENT_MASK)
-
-/*
- * Constraint on the Event code + UMask
- */
-#define INTEL_UEVENT_CONSTRAINT(c, n) \
- EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVENT_MASK)
+struct x86_pmu x86_pmu __read_mostly;
-#define EVENT_CONSTRAINT_END \
- EVENT_CONSTRAINT(0, 0, 0)
-
-#define for_each_event_constraint(e, c) \
- for ((e) = (c); (e)->weight; (e)++)
-
-/*
- * Per register state.
- */
-struct er_account {
- raw_spinlock_t lock; /* per-core: protect structure */
- u64 config; /* extra MSR config */
- u64 reg; /* extra MSR number */
- atomic_t ref; /* reference count */
-};
-
-/*
- * Extra registers for specific events.
- *
- * Some events need large masks and require external MSRs.
- * Those extra MSRs end up being shared for all events on
- * a PMU and sometimes between PMU of sibling HT threads.
- * In either case, the kernel needs to handle conflicting
- * accesses to those extra, shared, regs. The data structure
- * to manage those registers is stored in cpu_hw_event.
- */
-struct extra_reg {
- unsigned int event;
- unsigned int msr;
- u64 config_mask;
- u64 valid_mask;
- int idx; /* per_xxx->regs[] reg index */
-};
-
-#define EVENT_EXTRA_REG(e, ms, m, vm, i) { \
- .event = (e), \
- .msr = (ms), \
- .config_mask = (m), \
- .valid_mask = (vm), \
- .idx = EXTRA_REG_##i \
- }
-
-#define INTEL_EVENT_EXTRA_REG(event, msr, vm, idx) \
- EVENT_EXTRA_REG(event, msr, ARCH_PERFMON_EVENTSEL_EVENT, vm, idx)
-
-#define EVENT_EXTRA_END EVENT_EXTRA_REG(0, 0, 0, 0, RSP_0)
-
-union perf_capabilities {
- struct {
- u64 lbr_format : 6;
- u64 pebs_trap : 1;
- u64 pebs_arch_reg : 1;
- u64 pebs_format : 4;
- u64 smm_freeze : 1;
- };
- u64 capabilities;
-};
-
-/*
- * struct x86_pmu - generic x86 pmu
- */
-struct x86_pmu {
- /*
- * Generic x86 PMC bits
- */
- const char *name;
- int version;
- int (*handle_irq)(struct pt_regs *);
- void (*disable_all)(void);
- void (*enable_all)(int added);
- void (*enable)(struct perf_event *);
- void (*disable)(struct perf_event *);
- int (*hw_config)(struct perf_event *event);
- int (*schedule_events)(struct cpu_hw_events *cpuc, int n, int *assign);
- unsigned eventsel;
- unsigned perfctr;
- u64 (*event_map)(int);
- int max_events;
- int num_counters;
- int num_counters_fixed;
- int cntval_bits;
- u64 cntval_mask;
- int apic;
- u64 max_period;
- struct event_constraint *
- (*get_event_constraints)(struct cpu_hw_events *cpuc,
- struct perf_event *event);
-
- void (*put_event_constraints)(struct cpu_hw_events *cpuc,
- struct perf_event *event);
- struct event_constraint *event_constraints;
- void (*quirks)(void);
- int perfctr_second_write;
-
- int (*cpu_prepare)(int cpu);
- void (*cpu_starting)(int cpu);
- void (*cpu_dying)(int cpu);
- void (*cpu_dead)(int cpu);
-
- /*
- * Intel Arch Perfmon v2+
- */
- u64 intel_ctrl;
- union perf_capabilities intel_cap;
-
- /*
- * Intel DebugStore bits
- */
- int bts, pebs;
- int bts_active, pebs_active;
- int pebs_record_size;
- void (*drain_pebs)(struct pt_regs *regs);
- struct event_constraint *pebs_constraints;
-
- /*
- * Intel LBR
- */
- unsigned long lbr_tos, lbr_from, lbr_to; /* MSR base regs */
- int lbr_nr; /* hardware stack size */
-
- /*
- * Extra registers for events
- */
- struct extra_reg *extra_regs;
- unsigned int er_flags;
-};
-
-#define ERF_NO_HT_SHARING 1
-#define ERF_HAS_RSP_1 2
-
-static struct x86_pmu x86_pmu __read_mostly;
-
-static DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = {
+DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = {
.enabled = 1,
};
-static int x86_perf_event_set_period(struct perf_event *event);
-
-/*
- * Generalized hw caching related hw_event table, filled
- * in on a per model basis. A value of 0 means
- * 'not supported', -1 means 'hw_event makes no sense on
- * this CPU', any other value means the raw hw_event
- * ID.
- */
-
-#define C(x) PERF_COUNT_HW_CACHE_##x
-
-static u64 __read_mostly hw_cache_event_ids
+u64 __read_mostly hw_cache_event_ids
[PERF_COUNT_HW_CACHE_MAX]
[PERF_COUNT_HW_CACHE_OP_MAX]
[PERF_COUNT_HW_CACHE_RESULT_MAX];
-static u64 __read_mostly hw_cache_extra_regs
+u64 __read_mostly hw_cache_extra_regs
[PERF_COUNT_HW_CACHE_MAX]
[PERF_COUNT_HW_CACHE_OP_MAX]
[PERF_COUNT_HW_CACHE_RESULT_MAX];
@@ -329,8 +65,7 @@ static u64 __read_mostly hw_cache_extra_regs
* Can only be executed on the CPU where the event is active.
* Returns the delta events processed.
*/
-static u64
-x86_perf_event_update(struct perf_event *event)
+u64 x86_perf_event_update(struct perf_event *event)
{
struct hw_perf_event *hwc = &event->hw;
int shift = 64 - x86_pmu.cntval_bits;
@@ -373,30 +108,6 @@ again:
return new_raw_count;
}
-static inline int x86_pmu_addr_offset(int index)
-{
- int offset;
-
- /* offset = X86_FEATURE_PERFCTR_CORE ? index << 1 : index */
- alternative_io(ASM_NOP2,
- "shll $1, %%eax",
- X86_FEATURE_PERFCTR_CORE,
- "=a" (offset),
- "a" (index));
-
- return offset;
-}
-
-static inline unsigned int x86_pmu_config_addr(int index)
-{
- return x86_pmu.eventsel + x86_pmu_addr_offset(index);
-}
-
-static inline unsigned int x86_pmu_event_addr(int index)
-{
- return x86_pmu.perfctr + x86_pmu_addr_offset(index);
-}
-
/*
* Find and validate any extra registers to set up.
*/
@@ -532,9 +243,6 @@ msr_fail:
return false;
}
-static void reserve_ds_buffers(void);
-static void release_ds_buffers(void);
-
static void hw_perf_event_destroy(struct perf_event *event)
{
if (atomic_dec_and_mutex_lock(&active_events, &pmc_reserve_mutex)) {
@@ -583,7 +291,7 @@ set_ext_hw_attr(struct hw_perf_event *hwc, struct perf_event *event)
return x86_pmu_extra_regs(val, event);
}
-static int x86_setup_perfctr(struct perf_event *event)
+int x86_setup_perfctr(struct perf_event *event)
{
struct perf_event_attr *attr = &event->attr;
struct hw_perf_event *hwc = &event->hw;
@@ -647,7 +355,7 @@ static int x86_setup_perfctr(struct perf_event *event)
return 0;
}
-static int x86_pmu_hw_config(struct perf_event *event)
+int x86_pmu_hw_config(struct perf_event *event)
{
if (event->attr.precise_ip) {
int precise = 0;
@@ -723,7 +431,7 @@ static int __x86_pmu_event_init(struct perf_event *event)
return x86_pmu.hw_config(event);
}
-static void x86_pmu_disable_all(void)
+void x86_pmu_disable_all(void)
{
struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
int idx;
@@ -758,15 +466,7 @@ static void x86_pmu_disable(struct pmu *pmu)
x86_pmu.disable_all();
}
-static inline void __x86_pmu_enable_event(struct hw_perf_event *hwc,
- u64 enable_mask)
-{
- if (hwc->extra_reg.reg)
- wrmsrl(hwc->extra_reg.reg, hwc->extra_reg.config);
- wrmsrl(hwc->config_base, hwc->config | enable_mask);
-}
-
-static void x86_pmu_enable_all(int added)
+void x86_pmu_enable_all(int added)
{
struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
int idx;
@@ -788,7 +488,7 @@ static inline int is_x86_event(struct perf_event *event)
return event->pmu == &pmu;
}
-static int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign)
+int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign)
{
struct event_constraint *c, *constraints[X86_PMC_IDX_MAX];
unsigned long used_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
@@ -959,7 +659,6 @@ static inline int match_prev_assignment(struct hw_perf_event *hwc,
}
static void x86_pmu_start(struct perf_event *event, int flags);
-static void x86_pmu_stop(struct perf_event *event, int flags);
static void x86_pmu_enable(struct pmu *pmu)
{
@@ -1031,21 +730,13 @@ static void x86_pmu_enable(struct pmu *pmu)
x86_pmu.enable_all(added);
}
-static inline void x86_pmu_disable_event(struct perf_event *event)
-{
- struct hw_perf_event *hwc = &event->hw;
-
- wrmsrl(hwc->config_base, hwc->config);
-}
-
static DEFINE_PER_CPU(u64 [X86_PMC_IDX_MAX], pmc_prev_left);
/*
* Set the next IRQ period, based on the hwc->period_left value.
* To be called with the event disabled in hw:
*/
-static int
-x86_perf_event_set_period(struct perf_event *event)
+int x86_perf_event_set_period(struct perf_event *event)
{
struct hw_perf_event *hwc = &event->hw;
s64 left = local64_read(&hwc->period_left);
@@ -1105,7 +796,7 @@ x86_perf_event_set_period(struct perf_event *event)
return ret;
}
-static void x86_pmu_enable_event(struct perf_event *event)
+void x86_pmu_enable_event(struct perf_event *event)
{
if (__this_cpu_read(cpu_hw_events.enabled))
__x86_pmu_enable_event(&event->hw,
@@ -1244,7 +935,7 @@ void perf_event_print_debug(void)
local_irq_restore(flags);
}
-static void x86_pmu_stop(struct perf_event *event, int flags)
+void x86_pmu_stop(struct perf_event *event, int flags)
{
struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
struct hw_perf_event *hwc = &event->hw;
@@ -1297,7 +988,7 @@ static void x86_pmu_del(struct perf_event *event, int flags)
perf_event_update_userpage(event);
}
-static int x86_pmu_handle_irq(struct pt_regs *regs)
+int x86_pmu_handle_irq(struct pt_regs *regs)
{
struct perf_sample_data data;
struct cpu_hw_events *cpuc;
@@ -1367,109 +1058,28 @@ void perf_events_lapic_init(void)
apic_write(APIC_LVTPC, APIC_DM_NMI);
}
-struct pmu_nmi_state {
- unsigned int marked;
- int handled;
-};
-
-static DEFINE_PER_CPU(struct pmu_nmi_state, pmu_nmi);
-
static int __kprobes
-perf_event_nmi_handler(struct notifier_block *self,
- unsigned long cmd, void *__args)
+perf_event_nmi_handler(unsigned int cmd, struct pt_regs *regs)
{
- struct die_args *args = __args;
- unsigned int this_nmi;
- int handled;
-
if (!atomic_read(&active_events))
- return NOTIFY_DONE;
-
- switch (cmd) {
- case DIE_NMI:
- break;
- case DIE_NMIUNKNOWN:
- this_nmi = percpu_read(irq_stat.__nmi_count);
- if (this_nmi != __this_cpu_read(pmu_nmi.marked))
- /* let the kernel handle the unknown nmi */
- return NOTIFY_DONE;
- /*
- * This one is a PMU back-to-back nmi. Two events
- * trigger 'simultaneously' raising two back-to-back
- * NMIs. If the first NMI handles both, the latter
- * will be empty and daze the CPU. So, we drop it to
- * avoid false-positive 'unknown nmi' messages.
- */
- return NOTIFY_STOP;
- default:
- return NOTIFY_DONE;
- }
-
- handled = x86_pmu.handle_irq(args->regs);
- if (!handled)
- return NOTIFY_DONE;
-
- this_nmi = percpu_read(irq_stat.__nmi_count);
- if ((handled > 1) ||
- /* the next nmi could be a back-to-back nmi */
- ((__this_cpu_read(pmu_nmi.marked) == this_nmi) &&
- (__this_cpu_read(pmu_nmi.handled) > 1))) {
- /*
- * We could have two subsequent back-to-back nmis: The
- * first handles more than one counter, the 2nd
- * handles only one counter and the 3rd handles no
- * counter.
- *
- * This is the 2nd nmi because the previous was
- * handling more than one counter. We will mark the
- * next (3rd) and then drop it if unhandled.
- */
- __this_cpu_write(pmu_nmi.marked, this_nmi + 1);
- __this_cpu_write(pmu_nmi.handled, handled);
- }
+ return NMI_DONE;
- return NOTIFY_STOP;
+ return x86_pmu.handle_irq(regs);
}
-static __read_mostly struct notifier_block perf_event_nmi_notifier = {
- .notifier_call = perf_event_nmi_handler,
- .next = NULL,
- .priority = NMI_LOCAL_LOW_PRIOR,
-};
-
-static struct event_constraint unconstrained;
-static struct event_constraint emptyconstraint;
-
-static struct event_constraint *
-x86_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event)
-{
- struct event_constraint *c;
-
- if (x86_pmu.event_constraints) {
- for_each_event_constraint(c, x86_pmu.event_constraints) {
- if ((event->hw.config & c->cmask) == c->code)
- return c;
- }
- }
-
- return &unconstrained;
-}
-
-#include "perf_event_amd.c"
-#include "perf_event_p6.c"
-#include "perf_event_p4.c"
-#include "perf_event_intel_lbr.c"
-#include "perf_event_intel_ds.c"
-#include "perf_event_intel.c"
+struct event_constraint emptyconstraint;
+struct event_constraint unconstrained;
static int __cpuinit
x86_pmu_notifier(struct notifier_block *self, unsigned long action, void *hcpu)
{
unsigned int cpu = (long)hcpu;
+ struct cpu_hw_events *cpuc = &per_cpu(cpu_hw_events, cpu);
int ret = NOTIFY_OK;
switch (action & ~CPU_TASKS_FROZEN) {
case CPU_UP_PREPARE:
+ cpuc->kfree_on_online = NULL;
if (x86_pmu.cpu_prepare)
ret = x86_pmu.cpu_prepare(cpu);
break;
@@ -1479,6 +1089,10 @@ x86_pmu_notifier(struct notifier_block *self, unsigned long action, void *hcpu)
x86_pmu.cpu_starting(cpu);
break;
+ case CPU_ONLINE:
+ kfree(cpuc->kfree_on_online);
+ break;
+
case CPU_DYING:
if (x86_pmu.cpu_dying)
x86_pmu.cpu_dying(cpu);
@@ -1557,7 +1171,7 @@ static int __init init_hw_perf_events(void)
((1LL << x86_pmu.num_counters_fixed)-1) << X86_PMC_IDX_FIXED;
perf_events_lapic_init();
- register_die_notifier(&perf_event_nmi_notifier);
+ register_nmi_handler(NMI_LOCAL, perf_event_nmi_handler, 0, "PMI");
unconstrained = (struct event_constraint)
__EVENT_CONSTRAINT(0, (1ULL << x86_pmu.num_counters) - 1,
@@ -1900,6 +1514,9 @@ perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
perf_callchain_store(entry, regs->ip);
+ if (!current->mm)
+ return;
+
if (perf_callchain_user32(regs, entry))
return;
diff --git a/arch/x86/kernel/cpu/perf_event.h b/arch/x86/kernel/cpu/perf_event.h
new file mode 100644
index 00000000000..b9698d40ac4
--- /dev/null
+++ b/arch/x86/kernel/cpu/perf_event.h
@@ -0,0 +1,505 @@
+/*
+ * Performance events x86 architecture header
+ *
+ * Copyright (C) 2008 Thomas Gleixner <tglx@linutronix.de>
+ * Copyright (C) 2008-2009 Red Hat, Inc., Ingo Molnar
+ * Copyright (C) 2009 Jaswinder Singh Rajput
+ * Copyright (C) 2009 Advanced Micro Devices, Inc., Robert Richter
+ * Copyright (C) 2008-2009 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com>
+ * Copyright (C) 2009 Intel Corporation, <markus.t.metzger@intel.com>
+ * Copyright (C) 2009 Google, Inc., Stephane Eranian
+ *
+ * For licencing details see kernel-base/COPYING
+ */
+
+#include <linux/perf_event.h>
+
+/*
+ * | NHM/WSM | SNB |
+ * register -------------------------------
+ * | HT | no HT | HT | no HT |
+ *-----------------------------------------
+ * offcore | core | core | cpu | core |
+ * lbr_sel | core | core | cpu | core |
+ * ld_lat | cpu | core | cpu | core |
+ *-----------------------------------------
+ *
+ * Given that there is a small number of shared regs,
+ * we can pre-allocate their slot in the per-cpu
+ * per-core reg tables.
+ */
+enum extra_reg_type {
+ EXTRA_REG_NONE = -1, /* not used */
+
+ EXTRA_REG_RSP_0 = 0, /* offcore_response_0 */
+ EXTRA_REG_RSP_1 = 1, /* offcore_response_1 */
+
+ EXTRA_REG_MAX /* number of entries needed */
+};
+
+struct event_constraint {
+ union {
+ unsigned long idxmsk[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
+ u64 idxmsk64;
+ };
+ u64 code;
+ u64 cmask;
+ int weight;
+};
+
+struct amd_nb {
+ int nb_id; /* NorthBridge id */
+ int refcnt; /* reference count */
+ struct perf_event *owners[X86_PMC_IDX_MAX];
+ struct event_constraint event_constraints[X86_PMC_IDX_MAX];
+};
+
+/* The maximal number of PEBS events: */
+#define MAX_PEBS_EVENTS 4
+
+/*
+ * A debug store configuration.
+ *
+ * We only support architectures that use 64bit fields.
+ */
+struct debug_store {
+ u64 bts_buffer_base;
+ u64 bts_index;
+ u64 bts_absolute_maximum;
+ u64 bts_interrupt_threshold;
+ u64 pebs_buffer_base;
+ u64 pebs_index;
+ u64 pebs_absolute_maximum;
+ u64 pebs_interrupt_threshold;
+ u64 pebs_event_reset[MAX_PEBS_EVENTS];
+};
+
+/*
+ * Per register state.
+ */
+struct er_account {
+ raw_spinlock_t lock; /* per-core: protect structure */
+ u64 config; /* extra MSR config */
+ u64 reg; /* extra MSR number */
+ atomic_t ref; /* reference count */
+};
+
+/*
+ * Per core/cpu state
+ *
+ * Used to coordinate shared registers between HT threads or
+ * among events on a single PMU.
+ */
+struct intel_shared_regs {
+ struct er_account regs[EXTRA_REG_MAX];
+ int refcnt; /* per-core: #HT threads */
+ unsigned core_id; /* per-core: core id */
+};
+
+#define MAX_LBR_ENTRIES 16
+
+struct cpu_hw_events {
+ /*
+ * Generic x86 PMC bits
+ */
+ struct perf_event *events[X86_PMC_IDX_MAX]; /* in counter order */
+ unsigned long active_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
+ unsigned long running[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
+ int enabled;
+
+ int n_events;
+ int n_added;
+ int n_txn;
+ int assign[X86_PMC_IDX_MAX]; /* event to counter assignment */
+ u64 tags[X86_PMC_IDX_MAX];
+ struct perf_event *event_list[X86_PMC_IDX_MAX]; /* in enabled order */
+
+ unsigned int group_flag;
+
+ /*
+ * Intel DebugStore bits
+ */
+ struct debug_store *ds;
+ u64 pebs_enabled;
+
+ /*
+ * Intel LBR bits
+ */
+ int lbr_users;
+ void *lbr_context;
+ struct perf_branch_stack lbr_stack;
+ struct perf_branch_entry lbr_entries[MAX_LBR_ENTRIES];
+
+ /*
+ * Intel host/guest exclude bits
+ */
+ u64 intel_ctrl_guest_mask;
+ u64 intel_ctrl_host_mask;
+ struct perf_guest_switch_msr guest_switch_msrs[X86_PMC_IDX_MAX];
+
+ /*
+ * manage shared (per-core, per-cpu) registers
+ * used on Intel NHM/WSM/SNB
+ */
+ struct intel_shared_regs *shared_regs;
+
+ /*
+ * AMD specific bits
+ */
+ struct amd_nb *amd_nb;
+
+ void *kfree_on_online;
+};
+
+#define __EVENT_CONSTRAINT(c, n, m, w) {\
+ { .idxmsk64 = (n) }, \
+ .code = (c), \
+ .cmask = (m), \
+ .weight = (w), \
+}
+
+#define EVENT_CONSTRAINT(c, n, m) \
+ __EVENT_CONSTRAINT(c, n, m, HWEIGHT(n))
+
+/*
+ * Constraint on the Event code.
+ */
+#define INTEL_EVENT_CONSTRAINT(c, n) \
+ EVENT_CONSTRAINT(c, n, ARCH_PERFMON_EVENTSEL_EVENT)
+
+/*
+ * Constraint on the Event code + UMask + fixed-mask
+ *
+ * filter mask to validate fixed counter events.
+ * the following filters disqualify for fixed counters:
+ * - inv
+ * - edge
+ * - cnt-mask
+ * The other filters are supported by fixed counters.
+ * The any-thread option is supported starting with v3.
+ */
+#define FIXED_EVENT_CONSTRAINT(c, n) \
+ EVENT_CONSTRAINT(c, (1ULL << (32+n)), X86_RAW_EVENT_MASK)
+
+/*
+ * Constraint on the Event code + UMask
+ */
+#define INTEL_UEVENT_CONSTRAINT(c, n) \
+ EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVENT_MASK)
+
+#define EVENT_CONSTRAINT_END \
+ EVENT_CONSTRAINT(0, 0, 0)
+
+#define for_each_event_constraint(e, c) \
+ for ((e) = (c); (e)->weight; (e)++)
+
+/*
+ * Extra registers for specific events.
+ *
+ * Some events need large masks and require external MSRs.
+ * Those extra MSRs end up being shared for all events on
+ * a PMU and sometimes between PMU of sibling HT threads.
+ * In either case, the kernel needs to handle conflicting
+ * accesses to those extra, shared, regs. The data structure
+ * to manage those registers is stored in cpu_hw_event.
+ */
+struct extra_reg {
+ unsigned int event;
+ unsigned int msr;
+ u64 config_mask;
+ u64 valid_mask;
+ int idx; /* per_xxx->regs[] reg index */
+};
+
+#define EVENT_EXTRA_REG(e, ms, m, vm, i) { \
+ .event = (e), \
+ .msr = (ms), \
+ .config_mask = (m), \
+ .valid_mask = (vm), \
+ .idx = EXTRA_REG_##i \
+ }
+
+#define INTEL_EVENT_EXTRA_REG(event, msr, vm, idx) \
+ EVENT_EXTRA_REG(event, msr, ARCH_PERFMON_EVENTSEL_EVENT, vm, idx)
+
+#define EVENT_EXTRA_END EVENT_EXTRA_REG(0, 0, 0, 0, RSP_0)
+
+union perf_capabilities {
+ struct {
+ u64 lbr_format:6;
+ u64 pebs_trap:1;
+ u64 pebs_arch_reg:1;
+ u64 pebs_format:4;
+ u64 smm_freeze:1;
+ };
+ u64 capabilities;
+};
+
+/*
+ * struct x86_pmu - generic x86 pmu
+ */
+struct x86_pmu {
+ /*
+ * Generic x86 PMC bits
+ */
+ const char *name;
+ int version;
+ int (*handle_irq)(struct pt_regs *);
+ void (*disable_all)(void);
+ void (*enable_all)(int added);
+ void (*enable)(struct perf_event *);
+ void (*disable)(struct perf_event *);
+ int (*hw_config)(struct perf_event *event);
+ int (*schedule_events)(struct cpu_hw_events *cpuc, int n, int *assign);
+ unsigned eventsel;
+ unsigned perfctr;
+ u64 (*event_map)(int);
+ int max_events;
+ int num_counters;
+ int num_counters_fixed;
+ int cntval_bits;
+ u64 cntval_mask;
+ int apic;
+ u64 max_period;
+ struct event_constraint *
+ (*get_event_constraints)(struct cpu_hw_events *cpuc,
+ struct perf_event *event);
+
+ void (*put_event_constraints)(struct cpu_hw_events *cpuc,
+ struct perf_event *event);
+ struct event_constraint *event_constraints;
+ void (*quirks)(void);
+ int perfctr_second_write;
+
+ int (*cpu_prepare)(int cpu);
+ void (*cpu_starting)(int cpu);
+ void (*cpu_dying)(int cpu);
+ void (*cpu_dead)(int cpu);
+
+ /*
+ * Intel Arch Perfmon v2+
+ */
+ u64 intel_ctrl;
+ union perf_capabilities intel_cap;
+
+ /*
+ * Intel DebugStore bits
+ */
+ int bts, pebs;
+ int bts_active, pebs_active;
+ int pebs_record_size;
+ void (*drain_pebs)(struct pt_regs *regs);
+ struct event_constraint *pebs_constraints;
+
+ /*
+ * Intel LBR
+ */
+ unsigned long lbr_tos, lbr_from, lbr_to; /* MSR base regs */
+ int lbr_nr; /* hardware stack size */
+
+ /*
+ * Extra registers for events
+ */
+ struct extra_reg *extra_regs;
+ unsigned int er_flags;
+
+ /*
+ * Intel host/guest support (KVM)
+ */
+ struct perf_guest_switch_msr *(*guest_get_msrs)(int *nr);
+};
+
+#define ERF_NO_HT_SHARING 1
+#define ERF_HAS_RSP_1 2
+
+extern struct x86_pmu x86_pmu __read_mostly;
+
+DECLARE_PER_CPU(struct cpu_hw_events, cpu_hw_events);
+
+int x86_perf_event_set_period(struct perf_event *event);
+
+/*
+ * Generalized hw caching related hw_event table, filled
+ * in on a per model basis. A value of 0 means
+ * 'not supported', -1 means 'hw_event makes no sense on
+ * this CPU', any other value means the raw hw_event
+ * ID.
+ */
+
+#define C(x) PERF_COUNT_HW_CACHE_##x
+
+extern u64 __read_mostly hw_cache_event_ids
+ [PERF_COUNT_HW_CACHE_MAX]
+ [PERF_COUNT_HW_CACHE_OP_MAX]
+ [PERF_COUNT_HW_CACHE_RESULT_MAX];
+extern u64 __read_mostly hw_cache_extra_regs
+ [PERF_COUNT_HW_CACHE_MAX]
+ [PERF_COUNT_HW_CACHE_OP_MAX]
+ [PERF_COUNT_HW_CACHE_RESULT_MAX];
+
+u64 x86_perf_event_update(struct perf_event *event);
+
+static inline int x86_pmu_addr_offset(int index)
+{
+ int offset;
+
+ /* offset = X86_FEATURE_PERFCTR_CORE ? index << 1 : index */
+ alternative_io(ASM_NOP2,
+ "shll $1, %%eax",
+ X86_FEATURE_PERFCTR_CORE,
+ "=a" (offset),
+ "a" (index));
+
+ return offset;
+}
+
+static inline unsigned int x86_pmu_config_addr(int index)
+{
+ return x86_pmu.eventsel + x86_pmu_addr_offset(index);
+}
+
+static inline unsigned int x86_pmu_event_addr(int index)
+{
+ return x86_pmu.perfctr + x86_pmu_addr_offset(index);
+}
+
+int x86_setup_perfctr(struct perf_event *event);
+
+int x86_pmu_hw_config(struct perf_event *event);
+
+void x86_pmu_disable_all(void);
+
+static inline void __x86_pmu_enable_event(struct hw_perf_event *hwc,
+ u64 enable_mask)
+{
+ if (hwc->extra_reg.reg)
+ wrmsrl(hwc->extra_reg.reg, hwc->extra_reg.config);
+ wrmsrl(hwc->config_base, hwc->config | enable_mask);
+}
+
+void x86_pmu_enable_all(int added);
+
+int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign);
+
+void x86_pmu_stop(struct perf_event *event, int flags);
+
+static inline void x86_pmu_disable_event(struct perf_event *event)
+{
+ struct hw_perf_event *hwc = &event->hw;
+
+ wrmsrl(hwc->config_base, hwc->config);
+}
+
+void x86_pmu_enable_event(struct perf_event *event);
+
+int x86_pmu_handle_irq(struct pt_regs *regs);
+
+extern struct event_constraint emptyconstraint;
+
+extern struct event_constraint unconstrained;
+
+#ifdef CONFIG_CPU_SUP_AMD
+
+int amd_pmu_init(void);
+
+#else /* CONFIG_CPU_SUP_AMD */
+
+static inline int amd_pmu_init(void)
+{
+ return 0;
+}
+
+#endif /* CONFIG_CPU_SUP_AMD */
+
+#ifdef CONFIG_CPU_SUP_INTEL
+
+int intel_pmu_save_and_restart(struct perf_event *event);
+
+struct event_constraint *
+x86_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event);
+
+struct intel_shared_regs *allocate_shared_regs(int cpu);
+
+int intel_pmu_init(void);
+
+void init_debug_store_on_cpu(int cpu);
+
+void fini_debug_store_on_cpu(int cpu);
+
+void release_ds_buffers(void);
+
+void reserve_ds_buffers(void);
+
+extern struct event_constraint bts_constraint;
+
+void intel_pmu_enable_bts(u64 config);
+
+void intel_pmu_disable_bts(void);
+
+int intel_pmu_drain_bts_buffer(void);
+
+extern struct event_constraint intel_core2_pebs_event_constraints[];
+
+extern struct event_constraint intel_atom_pebs_event_constraints[];
+
+extern struct event_constraint intel_nehalem_pebs_event_constraints[];
+
+extern struct event_constraint intel_westmere_pebs_event_constraints[];
+
+extern struct event_constraint intel_snb_pebs_event_constraints[];
+
+struct event_constraint *intel_pebs_constraints(struct perf_event *event);
+
+void intel_pmu_pebs_enable(struct perf_event *event);
+
+void intel_pmu_pebs_disable(struct perf_event *event);
+
+void intel_pmu_pebs_enable_all(void);
+
+void intel_pmu_pebs_disable_all(void);
+
+void intel_ds_init(void);
+
+void intel_pmu_lbr_reset(void);
+
+void intel_pmu_lbr_enable(struct perf_event *event);
+
+void intel_pmu_lbr_disable(struct perf_event *event);
+
+void intel_pmu_lbr_enable_all(void);
+
+void intel_pmu_lbr_disable_all(void);
+
+void intel_pmu_lbr_read(void);
+
+void intel_pmu_lbr_init_core(void);
+
+void intel_pmu_lbr_init_nhm(void);
+
+void intel_pmu_lbr_init_atom(void);
+
+int p4_pmu_init(void);
+
+int p6_pmu_init(void);
+
+#else /* CONFIG_CPU_SUP_INTEL */
+
+static inline void reserve_ds_buffers(void)
+{
+}
+
+static inline void release_ds_buffers(void)
+{
+}
+
+static inline int intel_pmu_init(void)
+{
+ return 0;
+}
+
+static inline struct intel_shared_regs *allocate_shared_regs(int cpu)
+{
+ return NULL;
+}
+
+#endif /* CONFIG_CPU_SUP_INTEL */
diff --git a/arch/x86/kernel/cpu/perf_event_amd.c b/arch/x86/kernel/cpu/perf_event_amd.c
index 941caa2e449..aeefd45697a 100644
--- a/arch/x86/kernel/cpu/perf_event_amd.c
+++ b/arch/x86/kernel/cpu/perf_event_amd.c
@@ -1,4 +1,10 @@
-#ifdef CONFIG_CPU_SUP_AMD
+#include <linux/perf_event.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <asm/apicdef.h>
+
+#include "perf_event.h"
static __initconst const u64 amd_hw_cache_event_ids
[PERF_COUNT_HW_CACHE_MAX]
@@ -132,6 +138,19 @@ static int amd_pmu_hw_config(struct perf_event *event)
if (ret)
return ret;
+ if (event->attr.exclude_host && event->attr.exclude_guest)
+ /*
+ * When HO == GO == 1 the hardware treats that as GO == HO == 0
+ * and will count in both modes. We don't want to count in that
+ * case so we emulate no-counting by setting US = OS = 0.
+ */
+ event->hw.config &= ~(ARCH_PERFMON_EVENTSEL_USR |
+ ARCH_PERFMON_EVENTSEL_OS);
+ else if (event->attr.exclude_host)
+ event->hw.config |= AMD_PERFMON_EVENTSEL_GUESTONLY;
+ else if (event->attr.exclude_guest)
+ event->hw.config |= AMD_PERFMON_EVENTSEL_HOSTONLY;
+
if (event->attr.type != PERF_TYPE_RAW)
return 0;
@@ -350,7 +369,7 @@ static void amd_pmu_cpu_starting(int cpu)
continue;
if (nb->nb_id == nb_id) {
- kfree(cpuc->amd_nb);
+ cpuc->kfree_on_online = cpuc->amd_nb;
cpuc->amd_nb = nb;
break;
}
@@ -392,7 +411,7 @@ static __initconst const struct x86_pmu amd_pmu = {
.perfctr = MSR_K7_PERFCTR0,
.event_map = amd_pmu_event_map,
.max_events = ARRAY_SIZE(amd_perfmon_event_map),
- .num_counters = 4,
+ .num_counters = AMD64_NUM_COUNTERS,
.cntval_bits = 48,
.cntval_mask = (1ULL << 48) - 1,
.apic = 1,
@@ -556,7 +575,7 @@ static __initconst const struct x86_pmu amd_pmu_f15h = {
.perfctr = MSR_F15H_PERF_CTR,
.event_map = amd_pmu_event_map,
.max_events = ARRAY_SIZE(amd_perfmon_event_map),
- .num_counters = 6,
+ .num_counters = AMD64_NUM_COUNTERS_F15H,
.cntval_bits = 48,
.cntval_mask = (1ULL << 48) - 1,
.apic = 1,
@@ -573,7 +592,7 @@ static __initconst const struct x86_pmu amd_pmu_f15h = {
#endif
};
-static __init int amd_pmu_init(void)
+__init int amd_pmu_init(void)
{
/* Performance-monitoring supported from K7 and later: */
if (boot_cpu_data.x86 < 6)
@@ -602,12 +621,3 @@ static __init int amd_pmu_init(void)
return 0;
}
-
-#else /* CONFIG_CPU_SUP_AMD */
-
-static int amd_pmu_init(void)
-{
- return 0;
-}
-
-#endif
diff --git a/arch/x86/kernel/cpu/perf_event_amd_ibs.c b/arch/x86/kernel/cpu/perf_event_amd_ibs.c
new file mode 100644
index 00000000000..ab6343d2182
--- /dev/null
+++ b/arch/x86/kernel/cpu/perf_event_amd_ibs.c
@@ -0,0 +1,294 @@
+/*
+ * Performance events - AMD IBS
+ *
+ * Copyright (C) 2011 Advanced Micro Devices, Inc., Robert Richter
+ *
+ * For licencing details see kernel-base/COPYING
+ */
+
+#include <linux/perf_event.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+
+#include <asm/apic.h>
+
+static u32 ibs_caps;
+
+#if defined(CONFIG_PERF_EVENTS) && defined(CONFIG_CPU_SUP_AMD)
+
+static struct pmu perf_ibs;
+
+static int perf_ibs_init(struct perf_event *event)
+{
+ if (perf_ibs.type != event->attr.type)
+ return -ENOENT;
+ return 0;
+}
+
+static int perf_ibs_add(struct perf_event *event, int flags)
+{
+ return 0;
+}
+
+static void perf_ibs_del(struct perf_event *event, int flags)
+{
+}
+
+static struct pmu perf_ibs = {
+ .event_init= perf_ibs_init,
+ .add= perf_ibs_add,
+ .del= perf_ibs_del,
+};
+
+static __init int perf_event_ibs_init(void)
+{
+ if (!ibs_caps)
+ return -ENODEV; /* ibs not supported by the cpu */
+
+ perf_pmu_register(&perf_ibs, "ibs", -1);
+ printk(KERN_INFO "perf: AMD IBS detected (0x%08x)\n", ibs_caps);
+
+ return 0;
+}
+
+#else /* defined(CONFIG_PERF_EVENTS) && defined(CONFIG_CPU_SUP_AMD) */
+
+static __init int perf_event_ibs_init(void) { return 0; }
+
+#endif
+
+/* IBS - apic initialization, for perf and oprofile */
+
+static __init u32 __get_ibs_caps(void)
+{
+ u32 caps;
+ unsigned int max_level;
+
+ if (!boot_cpu_has(X86_FEATURE_IBS))
+ return 0;
+
+ /* check IBS cpuid feature flags */
+ max_level = cpuid_eax(0x80000000);
+ if (max_level < IBS_CPUID_FEATURES)
+ return IBS_CAPS_DEFAULT;
+
+ caps = cpuid_eax(IBS_CPUID_FEATURES);
+ if (!(caps & IBS_CAPS_AVAIL))
+ /* cpuid flags not valid */
+ return IBS_CAPS_DEFAULT;
+
+ return caps;
+}
+
+u32 get_ibs_caps(void)
+{
+ return ibs_caps;
+}
+
+EXPORT_SYMBOL(get_ibs_caps);
+
+static inline int get_eilvt(int offset)
+{
+ 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);
+}
+
+/*
+ * Check and reserve APIC extended interrupt LVT offset for IBS if available.
+ */
+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;
+
+ 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);
+ goto out;
+ }
+
+ 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);
+ goto out;
+ }
+
+ valid = 1;
+out:
+ preempt_enable();
+
+ return valid;
+}
+
+static int setup_ibs_ctl(int ibs_eilvt_off)
+{
+ struct pci_dev *cpu_cfg;
+ int nodes;
+ u32 value = 0;
+
+ nodes = 0;
+ cpu_cfg = NULL;
+ do {
+ cpu_cfg = pci_get_device(PCI_VENDOR_ID_AMD,
+ PCI_DEVICE_ID_AMD_10H_NB_MISC,
+ cpu_cfg);
+ if (!cpu_cfg)
+ break;
+ ++nodes;
+ pci_write_config_dword(cpu_cfg, IBSCTL, ibs_eilvt_off
+ | IBSCTL_LVT_OFFSET_VALID);
+ pci_read_config_dword(cpu_cfg, IBSCTL, &value);
+ if (value != (ibs_eilvt_off | IBSCTL_LVT_OFFSET_VALID)) {
+ pci_dev_put(cpu_cfg);
+ printk(KERN_DEBUG "Failed to setup IBS LVT offset, "
+ "IBSCTL = 0x%08x\n", value);
+ return -EINVAL;
+ }
+ } while (1);
+
+ if (!nodes) {
+ printk(KERN_DEBUG "No CPU node configured for IBS\n");
+ return -ENODEV;
+ }
+
+ return 0;
+}
+
+/*
+ * This runs only on the current cpu. We try to find an LVT offset and
+ * setup the local APIC. For this we must disable preemption. On
+ * success we initialize all nodes with this offset. This updates then
+ * the offset in the IBS_CTL per-node msr. The per-core APIC setup of
+ * the IBS interrupt vector is handled by perf_ibs_cpu_notifier that
+ * is using the new offset.
+ */
+static int force_ibs_eilvt_setup(void)
+{
+ int offset;
+ int ret;
+
+ preempt_disable();
+ /* find the next free available EILVT entry, skip offset 0 */
+ for (offset = 1; offset < APIC_EILVT_NR_MAX; offset++) {
+ if (get_eilvt(offset))
+ break;
+ }
+ preempt_enable();
+
+ if (offset == APIC_EILVT_NR_MAX) {
+ printk(KERN_DEBUG "No EILVT entry available\n");
+ return -EBUSY;
+ }
+
+ ret = setup_ibs_ctl(offset);
+ if (ret)
+ goto out;
+
+ 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;
+}
+
+static inline int get_ibs_lvt_offset(void)
+{
+ u64 val;
+
+ rdmsrl(MSR_AMD64_IBSCTL, val);
+ if (!(val & IBSCTL_LVT_OFFSET_VALID))
+ return -EINVAL;
+
+ return val & IBSCTL_LVT_OFFSET_MASK;
+}
+
+static void setup_APIC_ibs(void *dummy)
+{
+ int offset;
+
+ offset = get_ibs_lvt_offset();
+ if (offset < 0)
+ goto failed;
+
+ if (!setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_NMI, 0))
+ return;
+failed:
+ pr_warn("perf: IBS APIC setup failed on cpu #%d\n",
+ smp_processor_id());
+}
+
+static void clear_APIC_ibs(void *dummy)
+{
+ int offset;
+
+ offset = get_ibs_lvt_offset();
+ if (offset >= 0)
+ setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_FIX, 1);
+}
+
+static int __cpuinit
+perf_ibs_cpu_notifier(struct notifier_block *self, unsigned long action, void *hcpu)
+{
+ switch (action & ~CPU_TASKS_FROZEN) {
+ case CPU_STARTING:
+ setup_APIC_ibs(NULL);
+ break;
+ case CPU_DYING:
+ clear_APIC_ibs(NULL);
+ break;
+ default:
+ break;
+ }
+
+ return NOTIFY_OK;
+}
+
+static __init int amd_ibs_init(void)
+{
+ u32 caps;
+ int ret;
+
+ caps = __get_ibs_caps();
+ if (!caps)
+ return -ENODEV; /* ibs not supported by the cpu */
+
+ if (!ibs_eilvt_valid()) {
+ ret = force_ibs_eilvt_setup();
+ if (ret) {
+ pr_err("Failed to setup IBS, %d\n", ret);
+ return ret;
+ }
+ }
+
+ get_online_cpus();
+ ibs_caps = caps;
+ /* make ibs_caps visible to other cpus: */
+ smp_mb();
+ perf_cpu_notifier(perf_ibs_cpu_notifier);
+ smp_call_function(setup_APIC_ibs, NULL, 1);
+ put_online_cpus();
+
+ return perf_event_ibs_init();
+}
+
+/* Since we need the pci subsystem to init ibs we can't do this earlier: */
+device_initcall(amd_ibs_init);
diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c
index f88af2c2a56..e09ca20e86e 100644
--- a/arch/x86/kernel/cpu/perf_event_intel.c
+++ b/arch/x86/kernel/cpu/perf_event_intel.c
@@ -1,16 +1,19 @@
-#ifdef CONFIG_CPU_SUP_INTEL
-
/*
* Per core/cpu state
*
* Used to coordinate shared registers between HT threads or
* among events on a single PMU.
*/
-struct intel_shared_regs {
- struct er_account regs[EXTRA_REG_MAX];
- int refcnt; /* per-core: #HT threads */
- unsigned core_id; /* per-core: core id */
-};
+
+#include <linux/stddef.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+
+#include <asm/hardirq.h>
+#include <asm/apic.h>
+
+#include "perf_event.h"
/*
* Intel PerfMon, used on Core and later.
@@ -746,7 +749,8 @@ static void intel_pmu_enable_all(int added)
intel_pmu_pebs_enable_all();
intel_pmu_lbr_enable_all();
- wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, x86_pmu.intel_ctrl);
+ wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL,
+ x86_pmu.intel_ctrl & ~cpuc->intel_ctrl_guest_mask);
if (test_bit(X86_PMC_IDX_FIXED_BTS, cpuc->active_mask)) {
struct perf_event *event =
@@ -869,6 +873,7 @@ static void intel_pmu_disable_fixed(struct hw_perf_event *hwc)
static void intel_pmu_disable_event(struct perf_event *event)
{
struct hw_perf_event *hwc = &event->hw;
+ struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
if (unlikely(hwc->idx == X86_PMC_IDX_FIXED_BTS)) {
intel_pmu_disable_bts();
@@ -876,6 +881,9 @@ static void intel_pmu_disable_event(struct perf_event *event)
return;
}
+ cpuc->intel_ctrl_guest_mask &= ~(1ull << hwc->idx);
+ cpuc->intel_ctrl_host_mask &= ~(1ull << hwc->idx);
+
if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) {
intel_pmu_disable_fixed(hwc);
return;
@@ -921,6 +929,7 @@ static void intel_pmu_enable_fixed(struct hw_perf_event *hwc)
static void intel_pmu_enable_event(struct perf_event *event)
{
struct hw_perf_event *hwc = &event->hw;
+ struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
if (unlikely(hwc->idx == X86_PMC_IDX_FIXED_BTS)) {
if (!__this_cpu_read(cpu_hw_events.enabled))
@@ -930,6 +939,11 @@ static void intel_pmu_enable_event(struct perf_event *event)
return;
}
+ if (event->attr.exclude_host)
+ cpuc->intel_ctrl_guest_mask |= (1ull << hwc->idx);
+ if (event->attr.exclude_guest)
+ cpuc->intel_ctrl_host_mask |= (1ull << hwc->idx);
+
if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) {
intel_pmu_enable_fixed(hwc);
return;
@@ -945,7 +959,7 @@ static void intel_pmu_enable_event(struct perf_event *event)
* Save and restart an expired event. Called by NMI contexts,
* so it has to be careful about preempting normal event ops:
*/
-static int intel_pmu_save_and_restart(struct perf_event *event)
+int intel_pmu_save_and_restart(struct perf_event *event)
{
x86_perf_event_update(event);
return x86_perf_event_set_period(event);
@@ -1197,6 +1211,21 @@ intel_shared_regs_constraints(struct cpu_hw_events *cpuc,
return c;
}
+struct event_constraint *
+x86_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event)
+{
+ struct event_constraint *c;
+
+ if (x86_pmu.event_constraints) {
+ for_each_event_constraint(c, x86_pmu.event_constraints) {
+ if ((event->hw.config & c->cmask) == c->code)
+ return c;
+ }
+ }
+
+ return &unconstrained;
+}
+
static struct event_constraint *
intel_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event)
{
@@ -1284,12 +1313,84 @@ static int intel_pmu_hw_config(struct perf_event *event)
return 0;
}
+struct perf_guest_switch_msr *perf_guest_get_msrs(int *nr)
+{
+ if (x86_pmu.guest_get_msrs)
+ return x86_pmu.guest_get_msrs(nr);
+ *nr = 0;
+ return NULL;
+}
+EXPORT_SYMBOL_GPL(perf_guest_get_msrs);
+
+static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr)
+{
+ struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+ struct perf_guest_switch_msr *arr = cpuc->guest_switch_msrs;
+
+ arr[0].msr = MSR_CORE_PERF_GLOBAL_CTRL;
+ arr[0].host = x86_pmu.intel_ctrl & ~cpuc->intel_ctrl_guest_mask;
+ arr[0].guest = x86_pmu.intel_ctrl & ~cpuc->intel_ctrl_host_mask;
+
+ *nr = 1;
+ return arr;
+}
+
+static struct perf_guest_switch_msr *core_guest_get_msrs(int *nr)
+{
+ struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+ struct perf_guest_switch_msr *arr = cpuc->guest_switch_msrs;
+ int idx;
+
+ for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+ struct perf_event *event = cpuc->events[idx];
+
+ arr[idx].msr = x86_pmu_config_addr(idx);
+ arr[idx].host = arr[idx].guest = 0;
+
+ if (!test_bit(idx, cpuc->active_mask))
+ continue;
+
+ arr[idx].host = arr[idx].guest =
+ event->hw.config | ARCH_PERFMON_EVENTSEL_ENABLE;
+
+ if (event->attr.exclude_host)
+ arr[idx].host &= ~ARCH_PERFMON_EVENTSEL_ENABLE;
+ else if (event->attr.exclude_guest)
+ arr[idx].guest &= ~ARCH_PERFMON_EVENTSEL_ENABLE;
+ }
+
+ *nr = x86_pmu.num_counters;
+ return arr;
+}
+
+static void core_pmu_enable_event(struct perf_event *event)
+{
+ if (!event->attr.exclude_host)
+ x86_pmu_enable_event(event);
+}
+
+static void core_pmu_enable_all(int added)
+{
+ struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+ int idx;
+
+ for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+ struct hw_perf_event *hwc = &cpuc->events[idx]->hw;
+
+ if (!test_bit(idx, cpuc->active_mask) ||
+ cpuc->events[idx]->attr.exclude_host)
+ continue;
+
+ __x86_pmu_enable_event(hwc, ARCH_PERFMON_EVENTSEL_ENABLE);
+ }
+}
+
static __initconst const struct x86_pmu core_pmu = {
.name = "core",
.handle_irq = x86_pmu_handle_irq,
.disable_all = x86_pmu_disable_all,
- .enable_all = x86_pmu_enable_all,
- .enable = x86_pmu_enable_event,
+ .enable_all = core_pmu_enable_all,
+ .enable = core_pmu_enable_event,
.disable = x86_pmu_disable_event,
.hw_config = x86_pmu_hw_config,
.schedule_events = x86_schedule_events,
@@ -1307,9 +1408,10 @@ static __initconst const struct x86_pmu core_pmu = {
.get_event_constraints = intel_get_event_constraints,
.put_event_constraints = intel_put_event_constraints,
.event_constraints = intel_core_event_constraints,
+ .guest_get_msrs = core_guest_get_msrs,
};
-static struct intel_shared_regs *allocate_shared_regs(int cpu)
+struct intel_shared_regs *allocate_shared_regs(int cpu)
{
struct intel_shared_regs *regs;
int i;
@@ -1362,7 +1464,7 @@ static void intel_pmu_cpu_starting(int cpu)
pc = per_cpu(cpu_hw_events, i).shared_regs;
if (pc && pc->core_id == core_id) {
- kfree(cpuc->shared_regs);
+ cpuc->kfree_on_online = cpuc->shared_regs;
cpuc->shared_regs = pc;
break;
}
@@ -1413,6 +1515,7 @@ static __initconst const struct x86_pmu intel_pmu = {
.cpu_prepare = intel_pmu_cpu_prepare,
.cpu_starting = intel_pmu_cpu_starting,
.cpu_dying = intel_pmu_cpu_dying,
+ .guest_get_msrs = intel_guest_get_msrs,
};
static void intel_clovertown_quirks(void)
@@ -1441,7 +1544,7 @@ static void intel_clovertown_quirks(void)
x86_pmu.pebs_constraints = NULL;
}
-static __init int intel_pmu_init(void)
+__init int intel_pmu_init(void)
{
union cpuid10_edx edx;
union cpuid10_eax eax;
@@ -1597,7 +1700,7 @@ static __init int intel_pmu_init(void)
intel_pmu_lbr_init_nhm();
x86_pmu.event_constraints = intel_snb_event_constraints;
- x86_pmu.pebs_constraints = intel_snb_pebs_events;
+ x86_pmu.pebs_constraints = intel_snb_pebs_event_constraints;
x86_pmu.extra_regs = intel_snb_extra_regs;
/* all extra regs are per-cpu when HT is on */
x86_pmu.er_flags |= ERF_HAS_RSP_1;
@@ -1628,16 +1731,3 @@ static __init int intel_pmu_init(void)
}
return 0;
}
-
-#else /* CONFIG_CPU_SUP_INTEL */
-
-static int intel_pmu_init(void)
-{
- return 0;
-}
-
-static struct intel_shared_regs *allocate_shared_regs(int cpu)
-{
- return NULL;
-}
-#endif /* CONFIG_CPU_SUP_INTEL */
diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c
index 1b1ef3addcf..c0d238f49db 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_ds.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c
@@ -1,7 +1,10 @@
-#ifdef CONFIG_CPU_SUP_INTEL
+#include <linux/bitops.h>
+#include <linux/types.h>
+#include <linux/slab.h>
-/* The maximal number of PEBS events: */
-#define MAX_PEBS_EVENTS 4
+#include <asm/perf_event.h>
+
+#include "perf_event.h"
/* The size of a BTS record in bytes: */
#define BTS_RECORD_SIZE 24
@@ -37,24 +40,7 @@ struct pebs_record_nhm {
u64 status, dla, dse, lat;
};
-/*
- * A debug store configuration.
- *
- * We only support architectures that use 64bit fields.
- */
-struct debug_store {
- u64 bts_buffer_base;
- u64 bts_index;
- u64 bts_absolute_maximum;
- u64 bts_interrupt_threshold;
- u64 pebs_buffer_base;
- u64 pebs_index;
- u64 pebs_absolute_maximum;
- u64 pebs_interrupt_threshold;
- u64 pebs_event_reset[MAX_PEBS_EVENTS];
-};
-
-static void init_debug_store_on_cpu(int cpu)
+void init_debug_store_on_cpu(int cpu)
{
struct debug_store *ds = per_cpu(cpu_hw_events, cpu).ds;
@@ -66,7 +52,7 @@ static void init_debug_store_on_cpu(int cpu)
(u32)((u64)(unsigned long)ds >> 32));
}
-static void fini_debug_store_on_cpu(int cpu)
+void fini_debug_store_on_cpu(int cpu)
{
if (!per_cpu(cpu_hw_events, cpu).ds)
return;
@@ -175,7 +161,7 @@ static void release_ds_buffer(int cpu)
kfree(ds);
}
-static void release_ds_buffers(void)
+void release_ds_buffers(void)
{
int cpu;
@@ -194,7 +180,7 @@ static void release_ds_buffers(void)
put_online_cpus();
}
-static void reserve_ds_buffers(void)
+void reserve_ds_buffers(void)
{
int bts_err = 0, pebs_err = 0;
int cpu;
@@ -260,10 +246,10 @@ static void reserve_ds_buffers(void)
* BTS
*/
-static struct event_constraint bts_constraint =
+struct event_constraint bts_constraint =
EVENT_CONSTRAINT(0, 1ULL << X86_PMC_IDX_FIXED_BTS, 0);
-static void intel_pmu_enable_bts(u64 config)
+void intel_pmu_enable_bts(u64 config)
{
unsigned long debugctlmsr;
@@ -282,7 +268,7 @@ static void intel_pmu_enable_bts(u64 config)
update_debugctlmsr(debugctlmsr);
}
-static void intel_pmu_disable_bts(void)
+void intel_pmu_disable_bts(void)
{
struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
unsigned long debugctlmsr;
@@ -299,7 +285,7 @@ static void intel_pmu_disable_bts(void)
update_debugctlmsr(debugctlmsr);
}
-static int intel_pmu_drain_bts_buffer(void)
+int intel_pmu_drain_bts_buffer(void)
{
struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
struct debug_store *ds = cpuc->ds;
@@ -361,7 +347,7 @@ static int intel_pmu_drain_bts_buffer(void)
/*
* PEBS
*/
-static struct event_constraint intel_core2_pebs_event_constraints[] = {
+struct event_constraint intel_core2_pebs_event_constraints[] = {
INTEL_UEVENT_CONSTRAINT(0x00c0, 0x1), /* INST_RETIRED.ANY */
INTEL_UEVENT_CONSTRAINT(0xfec1, 0x1), /* X87_OPS_RETIRED.ANY */
INTEL_UEVENT_CONSTRAINT(0x00c5, 0x1), /* BR_INST_RETIRED.MISPRED */
@@ -370,14 +356,14 @@ static struct event_constraint intel_core2_pebs_event_constraints[] = {
EVENT_CONSTRAINT_END
};
-static struct event_constraint intel_atom_pebs_event_constraints[] = {
+struct event_constraint intel_atom_pebs_event_constraints[] = {
INTEL_UEVENT_CONSTRAINT(0x00c0, 0x1), /* INST_RETIRED.ANY */
INTEL_UEVENT_CONSTRAINT(0x00c5, 0x1), /* MISPREDICTED_BRANCH_RETIRED */
INTEL_EVENT_CONSTRAINT(0xcb, 0x1), /* MEM_LOAD_RETIRED.* */
EVENT_CONSTRAINT_END
};
-static struct event_constraint intel_nehalem_pebs_event_constraints[] = {
+struct event_constraint intel_nehalem_pebs_event_constraints[] = {
INTEL_EVENT_CONSTRAINT(0x0b, 0xf), /* MEM_INST_RETIRED.* */
INTEL_EVENT_CONSTRAINT(0x0f, 0xf), /* MEM_UNCORE_RETIRED.* */
INTEL_UEVENT_CONSTRAINT(0x010c, 0xf), /* MEM_STORE_RETIRED.DTLB_MISS */
@@ -392,7 +378,7 @@ static struct event_constraint intel_nehalem_pebs_event_constraints[] = {
EVENT_CONSTRAINT_END
};
-static struct event_constraint intel_westmere_pebs_event_constraints[] = {
+struct event_constraint intel_westmere_pebs_event_constraints[] = {
INTEL_EVENT_CONSTRAINT(0x0b, 0xf), /* MEM_INST_RETIRED.* */
INTEL_EVENT_CONSTRAINT(0x0f, 0xf), /* MEM_UNCORE_RETIRED.* */
INTEL_UEVENT_CONSTRAINT(0x010c, 0xf), /* MEM_STORE_RETIRED.DTLB_MISS */
@@ -407,7 +393,7 @@ static struct event_constraint intel_westmere_pebs_event_constraints[] = {
EVENT_CONSTRAINT_END
};
-static struct event_constraint intel_snb_pebs_events[] = {
+struct event_constraint intel_snb_pebs_event_constraints[] = {
INTEL_UEVENT_CONSTRAINT(0x01c0, 0x2), /* INST_RETIRED.PRECDIST */
INTEL_UEVENT_CONSTRAINT(0x01c2, 0xf), /* UOPS_RETIRED.ALL */
INTEL_UEVENT_CONSTRAINT(0x02c2, 0xf), /* UOPS_RETIRED.RETIRE_SLOTS */
@@ -428,8 +414,7 @@ static struct event_constraint intel_snb_pebs_events[] = {
EVENT_CONSTRAINT_END
};
-static struct event_constraint *
-intel_pebs_constraints(struct perf_event *event)
+struct event_constraint *intel_pebs_constraints(struct perf_event *event)
{
struct event_constraint *c;
@@ -446,7 +431,7 @@ intel_pebs_constraints(struct perf_event *event)
return &emptyconstraint;
}
-static void intel_pmu_pebs_enable(struct perf_event *event)
+void intel_pmu_pebs_enable(struct perf_event *event)
{
struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
struct hw_perf_event *hwc = &event->hw;
@@ -460,7 +445,7 @@ static void intel_pmu_pebs_enable(struct perf_event *event)
intel_pmu_lbr_enable(event);
}
-static void intel_pmu_pebs_disable(struct perf_event *event)
+void intel_pmu_pebs_disable(struct perf_event *event)
{
struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
struct hw_perf_event *hwc = &event->hw;
@@ -475,7 +460,7 @@ static void intel_pmu_pebs_disable(struct perf_event *event)
intel_pmu_lbr_disable(event);
}
-static void intel_pmu_pebs_enable_all(void)
+void intel_pmu_pebs_enable_all(void)
{
struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
@@ -483,7 +468,7 @@ static void intel_pmu_pebs_enable_all(void)
wrmsrl(MSR_IA32_PEBS_ENABLE, cpuc->pebs_enabled);
}
-static void intel_pmu_pebs_disable_all(void)
+void intel_pmu_pebs_disable_all(void)
{
struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
@@ -576,8 +561,6 @@ static int intel_pmu_pebs_fixup_ip(struct pt_regs *regs)
return 0;
}
-static int intel_pmu_save_and_restart(struct perf_event *event);
-
static void __intel_pmu_pebs_event(struct perf_event *event,
struct pt_regs *iregs, void *__pebs)
{
@@ -716,7 +699,7 @@ static void intel_pmu_drain_pebs_nhm(struct pt_regs *iregs)
* BTS, PEBS probe and setup
*/
-static void intel_ds_init(void)
+void intel_ds_init(void)
{
/*
* No support for 32bit formats
@@ -749,15 +732,3 @@ static void intel_ds_init(void)
}
}
}
-
-#else /* CONFIG_CPU_SUP_INTEL */
-
-static void reserve_ds_buffers(void)
-{
-}
-
-static void release_ds_buffers(void)
-{
-}
-
-#endif /* CONFIG_CPU_SUP_INTEL */
diff --git a/arch/x86/kernel/cpu/perf_event_intel_lbr.c b/arch/x86/kernel/cpu/perf_event_intel_lbr.c
index d202c1bece1..3fab3de3ce9 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_lbr.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_lbr.c
@@ -1,4 +1,10 @@
-#ifdef CONFIG_CPU_SUP_INTEL
+#include <linux/perf_event.h>
+#include <linux/types.h>
+
+#include <asm/perf_event.h>
+#include <asm/msr.h>
+
+#include "perf_event.h"
enum {
LBR_FORMAT_32 = 0x00,
@@ -48,7 +54,7 @@ static void intel_pmu_lbr_reset_64(void)
}
}
-static void intel_pmu_lbr_reset(void)
+void intel_pmu_lbr_reset(void)
{
if (!x86_pmu.lbr_nr)
return;
@@ -59,7 +65,7 @@ static void intel_pmu_lbr_reset(void)
intel_pmu_lbr_reset_64();
}
-static void intel_pmu_lbr_enable(struct perf_event *event)
+void intel_pmu_lbr_enable(struct perf_event *event)
{
struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
@@ -81,7 +87,7 @@ static void intel_pmu_lbr_enable(struct perf_event *event)
cpuc->lbr_users++;
}
-static void intel_pmu_lbr_disable(struct perf_event *event)
+void intel_pmu_lbr_disable(struct perf_event *event)
{
struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
@@ -95,7 +101,7 @@ static void intel_pmu_lbr_disable(struct perf_event *event)
__intel_pmu_lbr_disable();
}
-static void intel_pmu_lbr_enable_all(void)
+void intel_pmu_lbr_enable_all(void)
{
struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
@@ -103,7 +109,7 @@ static void intel_pmu_lbr_enable_all(void)
__intel_pmu_lbr_enable();
}
-static void intel_pmu_lbr_disable_all(void)
+void intel_pmu_lbr_disable_all(void)
{
struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
@@ -178,7 +184,7 @@ static void intel_pmu_lbr_read_64(struct cpu_hw_events *cpuc)
cpuc->lbr_stack.nr = i;
}
-static void intel_pmu_lbr_read(void)
+void intel_pmu_lbr_read(void)
{
struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
@@ -191,7 +197,7 @@ static void intel_pmu_lbr_read(void)
intel_pmu_lbr_read_64(cpuc);
}
-static void intel_pmu_lbr_init_core(void)
+void intel_pmu_lbr_init_core(void)
{
x86_pmu.lbr_nr = 4;
x86_pmu.lbr_tos = 0x01c9;
@@ -199,7 +205,7 @@ static void intel_pmu_lbr_init_core(void)
x86_pmu.lbr_to = 0x60;
}
-static void intel_pmu_lbr_init_nhm(void)
+void intel_pmu_lbr_init_nhm(void)
{
x86_pmu.lbr_nr = 16;
x86_pmu.lbr_tos = 0x01c9;
@@ -207,12 +213,10 @@ static void intel_pmu_lbr_init_nhm(void)
x86_pmu.lbr_to = 0x6c0;
}
-static void intel_pmu_lbr_init_atom(void)
+void intel_pmu_lbr_init_atom(void)
{
x86_pmu.lbr_nr = 8;
x86_pmu.lbr_tos = 0x01c9;
x86_pmu.lbr_from = 0x40;
x86_pmu.lbr_to = 0x60;
}
-
-#endif /* CONFIG_CPU_SUP_INTEL */
diff --git a/arch/x86/kernel/cpu/perf_event_p4.c b/arch/x86/kernel/cpu/perf_event_p4.c
index 7809d2bcb20..492bf1358a7 100644
--- a/arch/x86/kernel/cpu/perf_event_p4.c
+++ b/arch/x86/kernel/cpu/perf_event_p4.c
@@ -7,9 +7,13 @@
* For licencing details see kernel-base/COPYING
*/
-#ifdef CONFIG_CPU_SUP_INTEL
+#include <linux/perf_event.h>
#include <asm/perf_event_p4.h>
+#include <asm/hardirq.h>
+#include <asm/apic.h>
+
+#include "perf_event.h"
#define P4_CNTR_LIMIT 3
/*
@@ -1303,7 +1307,7 @@ static __initconst const struct x86_pmu p4_pmu = {
.perfctr_second_write = 1,
};
-static __init int p4_pmu_init(void)
+__init int p4_pmu_init(void)
{
unsigned int low, high;
@@ -1326,5 +1330,3 @@ static __init int p4_pmu_init(void)
return 0;
}
-
-#endif /* CONFIG_CPU_SUP_INTEL */
diff --git a/arch/x86/kernel/cpu/perf_event_p6.c b/arch/x86/kernel/cpu/perf_event_p6.c
index 20c097e3386..c7181befecd 100644
--- a/arch/x86/kernel/cpu/perf_event_p6.c
+++ b/arch/x86/kernel/cpu/perf_event_p6.c
@@ -1,4 +1,7 @@
-#ifdef CONFIG_CPU_SUP_INTEL
+#include <linux/perf_event.h>
+#include <linux/types.h>
+
+#include "perf_event.h"
/*
* Not sure about some of these
@@ -114,7 +117,7 @@ static __initconst const struct x86_pmu p6_pmu = {
.event_constraints = p6_event_constraints,
};
-static __init int p6_pmu_init(void)
+__init int p6_pmu_init(void)
{
switch (boot_cpu_data.x86_model) {
case 1:
@@ -138,5 +141,3 @@ static __init int p6_pmu_init(void)
return 0;
}
-
-#endif /* CONFIG_CPU_SUP_INTEL */
diff --git a/arch/x86/kernel/cpu/proc.c b/arch/x86/kernel/cpu/proc.c
index 62ac8cb6ba2..14b23140e81 100644
--- a/arch/x86/kernel/cpu/proc.c
+++ b/arch/x86/kernel/cpu/proc.c
@@ -85,6 +85,8 @@ static int show_cpuinfo(struct seq_file *m, void *v)
seq_printf(m, "stepping\t: %d\n", c->x86_mask);
else
seq_printf(m, "stepping\t: unknown\n");
+ if (c->microcode)
+ seq_printf(m, "microcode\t: 0x%x\n", c->microcode);
if (cpu_has(c, X86_FEATURE_TSC)) {
unsigned int freq = cpufreq_quick_get(cpu);
diff --git a/arch/x86/kernel/cpu/rdrand.c b/arch/x86/kernel/cpu/rdrand.c
new file mode 100644
index 00000000000..feca286c2bb
--- /dev/null
+++ b/arch/x86/kernel/cpu/rdrand.c
@@ -0,0 +1,73 @@
+/*
+ * This file is part of the Linux kernel.
+ *
+ * Copyright (c) 2011, Intel Corporation
+ * Authors: Fenghua Yu <fenghua.yu@intel.com>,
+ * H. Peter Anvin <hpa@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc.,
+ * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ */
+
+#include <asm/processor.h>
+#include <asm/archrandom.h>
+#include <asm/sections.h>
+
+static int __init x86_rdrand_setup(char *s)
+{
+ setup_clear_cpu_cap(X86_FEATURE_RDRAND);
+ return 1;
+}
+__setup("nordrand", x86_rdrand_setup);
+
+/* We can't use arch_get_random_long() here since alternatives haven't run */
+static inline int rdrand_long(unsigned long *v)
+{
+ int ok;
+ asm volatile("1: " RDRAND_LONG "\n\t"
+ "jc 2f\n\t"
+ "decl %0\n\t"
+ "jnz 1b\n\t"
+ "2:"
+ : "=r" (ok), "=a" (*v)
+ : "0" (RDRAND_RETRY_LOOPS));
+ return ok;
+}
+
+/*
+ * Force a reseed cycle; we are architecturally guaranteed a reseed
+ * after no more than 512 128-bit chunks of random data. This also
+ * acts as a test of the CPU capability.
+ */
+#define RESEED_LOOP ((512*128)/sizeof(unsigned long))
+
+void __cpuinit x86_init_rdrand(struct cpuinfo_x86 *c)
+{
+#ifdef CONFIG_ARCH_RANDOM
+ unsigned long tmp;
+ int i, count, ok;
+
+ if (!cpu_has(c, X86_FEATURE_RDRAND))
+ return; /* Nothing to do */
+
+ for (count = i = 0; i < RESEED_LOOP; i++) {
+ ok = rdrand_long(&tmp);
+ if (ok)
+ count++;
+ }
+
+ if (count != RESEED_LOOP)
+ clear_cpu_cap(c, X86_FEATURE_RDRAND);
+#endif
+}
diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c
index 764c7c2b181..13ad89971d4 100644
--- a/arch/x86/kernel/crash.c
+++ b/arch/x86/kernel/crash.c
@@ -32,15 +32,12 @@ int in_crash_kexec;
#if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC)
-static void kdump_nmi_callback(int cpu, struct die_args *args)
+static void kdump_nmi_callback(int cpu, struct pt_regs *regs)
{
- struct pt_regs *regs;
#ifdef CONFIG_X86_32
struct pt_regs fixed_regs;
#endif
- regs = args->regs;
-
#ifdef CONFIG_X86_32
if (!user_mode_vm(regs)) {
crash_fixup_ss_esp(&fixed_regs, regs);
diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
index 6419bb05ecd..faf8d5e74b0 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -331,10 +331,15 @@ ENDPROC(native_usergs_sysret64)
1: incl PER_CPU_VAR(irq_count)
jne 2f
mov PER_CPU_VAR(irq_stack_ptr),%rsp
- EMPTY_FRAME 0
+ CFI_DEF_CFA_REGISTER rsi
2: /* Store previous stack value */
pushq %rsi
+ CFI_ESCAPE 0x0f /* DW_CFA_def_cfa_expression */, 6, \
+ 0x77 /* DW_OP_breg7 */, 0, \
+ 0x06 /* DW_OP_deref */, \
+ 0x08 /* DW_OP_const1u */, SS+8-RBP, \
+ 0x22 /* DW_OP_plus */
/* We entered an interrupt context - irqs are off: */
TRACE_IRQS_OFF
.endm
@@ -788,7 +793,6 @@ END(interrupt)
subq $ORIG_RAX-RBP, %rsp
CFI_ADJUST_CFA_OFFSET ORIG_RAX-RBP
SAVE_ARGS_IRQ
- PARTIAL_FRAME 0
call \func
.endm
@@ -813,10 +817,10 @@ ret_from_intr:
/* Restore saved previous stack */
popq %rsi
- leaq 16(%rsi), %rsp
-
+ CFI_DEF_CFA_REGISTER rsi
+ leaq ARGOFFSET-RBP(%rsi), %rsp
CFI_DEF_CFA_REGISTER rsp
- CFI_ADJUST_CFA_OFFSET -16
+ CFI_ADJUST_CFA_OFFSET RBP-ARGOFFSET
exit_intr:
GET_THREAD_INFO(%rcx)
diff --git a/arch/x86/kernel/jump_label.c b/arch/x86/kernel/jump_label.c
index 3fee346ef54..cacdd46d184 100644
--- a/arch/x86/kernel/jump_label.c
+++ b/arch/x86/kernel/jump_label.c
@@ -42,7 +42,7 @@ void arch_jump_label_transform(struct jump_entry *entry,
put_online_cpus();
}
-void arch_jump_label_text_poke_early(jump_label_t addr)
+void __init_or_module arch_jump_label_text_poke_early(jump_label_t addr)
{
text_poke_early((void *)addr, ideal_nops[NOP_ATOMIC5],
JUMP_LABEL_NOP_SIZE);
diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c
index 00354d4919a..faba5771aca 100644
--- a/arch/x86/kernel/kgdb.c
+++ b/arch/x86/kernel/kgdb.c
@@ -511,28 +511,37 @@ single_step_cont(struct pt_regs *regs, struct die_args *args)
static int was_in_debug_nmi[NR_CPUS];
-static int __kgdb_notify(struct die_args *args, unsigned long cmd)
+static int kgdb_nmi_handler(unsigned int cmd, struct pt_regs *regs)
{
- struct pt_regs *regs = args->regs;
-
switch (cmd) {
- case DIE_NMI:
+ case NMI_LOCAL:
if (atomic_read(&kgdb_active) != -1) {
/* KGDB CPU roundup */
kgdb_nmicallback(raw_smp_processor_id(), regs);
was_in_debug_nmi[raw_smp_processor_id()] = 1;
touch_nmi_watchdog();
- return NOTIFY_STOP;
+ return NMI_HANDLED;
}
- return NOTIFY_DONE;
+ break;
- case DIE_NMIUNKNOWN:
+ case NMI_UNKNOWN:
if (was_in_debug_nmi[raw_smp_processor_id()]) {
was_in_debug_nmi[raw_smp_processor_id()] = 0;
- return NOTIFY_STOP;
+ return NMI_HANDLED;
}
- return NOTIFY_DONE;
+ break;
+ default:
+ /* do nothing */
+ break;
+ }
+ return NMI_DONE;
+}
+
+static int __kgdb_notify(struct die_args *args, unsigned long cmd)
+{
+ struct pt_regs *regs = args->regs;
+ switch (cmd) {
case DIE_DEBUG:
if (atomic_read(&kgdb_cpu_doing_single_step) != -1) {
if (user_mode(regs))
@@ -590,11 +599,6 @@ kgdb_notify(struct notifier_block *self, unsigned long cmd, void *ptr)
static struct notifier_block kgdb_notifier = {
.notifier_call = kgdb_notify,
-
- /*
- * Lowest-prio notifier priority, we want to be notified last:
- */
- .priority = NMI_LOCAL_LOW_PRIOR,
};
/**
@@ -605,7 +609,31 @@ static struct notifier_block kgdb_notifier = {
*/
int kgdb_arch_init(void)
{
- return register_die_notifier(&kgdb_notifier);
+ int retval;
+
+ retval = register_die_notifier(&kgdb_notifier);
+ if (retval)
+ goto out;
+
+ retval = register_nmi_handler(NMI_LOCAL, kgdb_nmi_handler,
+ 0, "kgdb");
+ if (retval)
+ goto out1;
+
+ retval = register_nmi_handler(NMI_UNKNOWN, kgdb_nmi_handler,
+ 0, "kgdb");
+
+ if (retval)
+ goto out2;
+
+ return retval;
+
+out2:
+ unregister_nmi_handler(NMI_LOCAL, "kgdb");
+out1:
+ unregister_die_notifier(&kgdb_notifier);
+out:
+ return retval;
}
static void kgdb_hw_overflow_handler(struct perf_event *event,
@@ -673,6 +701,8 @@ void kgdb_arch_exit(void)
breakinfo[i].pev = NULL;
}
}
+ unregister_nmi_handler(NMI_UNKNOWN, "kgdb");
+ unregister_nmi_handler(NMI_LOCAL, "kgdb");
unregister_die_notifier(&kgdb_notifier);
}
diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
index f1a6244d7d9..7da647d8b64 100644
--- a/arch/x86/kernel/kprobes.c
+++ b/arch/x86/kernel/kprobes.c
@@ -75,8 +75,11 @@ DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk);
/*
* Undefined/reserved opcodes, conditional jump, Opcode Extension
* Groups, and some special opcodes can not boost.
+ * This is non-const and volatile to keep gcc from statically
+ * optimizing it out, as variable_test_bit makes gcc think only
+ * *(unsigned long*) is used.
*/
-static const u32 twobyte_is_boostable[256 / 32] = {
+static volatile u32 twobyte_is_boostable[256 / 32] = {
/* 0 1 2 3 4 5 6 7 8 9 a b c d e f */
/* ---------------------------------------------- */
W(0x00, 0, 0, 1, 1, 0, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 0) | /* 00 */
diff --git a/arch/x86/kernel/microcode_amd.c b/arch/x86/kernel/microcode_amd.c
index 591be0ee193..d494799aafc 100644
--- a/arch/x86/kernel/microcode_amd.c
+++ b/arch/x86/kernel/microcode_amd.c
@@ -74,14 +74,13 @@ static struct equiv_cpu_entry *equiv_cpu_table;
static int collect_cpu_info_amd(int cpu, struct cpu_signature *csig)
{
struct cpuinfo_x86 *c = &cpu_data(cpu);
- u32 dummy;
if (c->x86_vendor != X86_VENDOR_AMD || c->x86 < 0x10) {
pr_warning("CPU%d: family %d not supported\n", cpu, c->x86);
return -1;
}
- rdmsr(MSR_AMD64_PATCH_LEVEL, csig->rev, dummy);
+ csig->rev = c->microcode;
pr_info("CPU%d: patch_level=0x%08x\n", cpu, csig->rev);
return 0;
@@ -130,6 +129,7 @@ static int apply_microcode_amd(int cpu)
int cpu_num = raw_smp_processor_id();
struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num;
struct microcode_amd *mc_amd = uci->mc;
+ struct cpuinfo_x86 *c = &cpu_data(cpu);
/* We should bind the task to the CPU */
BUG_ON(cpu_num != cpu);
@@ -150,6 +150,7 @@ static int apply_microcode_amd(int cpu)
pr_info("CPU%d: new patch_level=0x%08x\n", cpu, rev);
uci->cpu_sig.rev = rev;
+ c->microcode = rev;
return 0;
}
diff --git a/arch/x86/kernel/microcode_core.c b/arch/x86/kernel/microcode_core.c
index f9242800bc8..f2d2a664e79 100644
--- a/arch/x86/kernel/microcode_core.c
+++ b/arch/x86/kernel/microcode_core.c
@@ -483,7 +483,13 @@ mc_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu)
sysfs_remove_group(&sys_dev->kobj, &mc_attr_group);
pr_debug("CPU%d removed\n", cpu);
break;
- case CPU_DEAD:
+
+ /*
+ * When a CPU goes offline, don't free up or invalidate the copy of
+ * the microcode in kernel memory, so that we can reuse it when the
+ * CPU comes back online without unnecessarily requesting the userspace
+ * for it again.
+ */
case CPU_UP_CANCELED_FROZEN:
/* The CPU refused to come up during a system resume */
microcode_fini_cpu(cpu);
diff --git a/arch/x86/kernel/microcode_intel.c b/arch/x86/kernel/microcode_intel.c
index 1a1b606d3e9..3ca42d0e43a 100644
--- a/arch/x86/kernel/microcode_intel.c
+++ b/arch/x86/kernel/microcode_intel.c
@@ -161,12 +161,7 @@ static int collect_cpu_info(int cpu_num, struct cpu_signature *csig)
csig->pf = 1 << ((val[1] >> 18) & 7);
}
- wrmsr(MSR_IA32_UCODE_REV, 0, 0);
- /* see notes above for revision 1.07. Apparent chip bug */
- sync_core();
- /* get the current revision from MSR 0x8B */
- rdmsr(MSR_IA32_UCODE_REV, val[0], csig->rev);
-
+ csig->rev = c->microcode;
pr_info("CPU%d sig=0x%x, pf=0x%x, revision=0x%x\n",
cpu_num, csig->sig, csig->pf, csig->rev);
@@ -299,9 +294,9 @@ static int apply_microcode(int cpu)
struct microcode_intel *mc_intel;
struct ucode_cpu_info *uci;
unsigned int val[2];
- int cpu_num;
+ int cpu_num = raw_smp_processor_id();
+ struct cpuinfo_x86 *c = &cpu_data(cpu_num);
- cpu_num = raw_smp_processor_id();
uci = ucode_cpu_info + cpu;
mc_intel = uci->mc;
@@ -317,7 +312,7 @@ static int apply_microcode(int cpu)
(unsigned long) mc_intel->bits >> 16 >> 16);
wrmsr(MSR_IA32_UCODE_REV, 0, 0);
- /* see notes above for revision 1.07. Apparent chip bug */
+ /* As documented in the SDM: Do a CPUID 1 here */
sync_core();
/* get the current revision from MSR 0x8B */
@@ -335,6 +330,7 @@ static int apply_microcode(int cpu)
(mc_intel->hdr.date >> 16) & 0xff);
uci->cpu_sig.rev = val[1];
+ c->microcode = val[1];
return 0;
}
diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c
new file mode 100644
index 00000000000..7ec5bd140b8
--- /dev/null
+++ b/arch/x86/kernel/nmi.c
@@ -0,0 +1,433 @@
+/*
+ * Copyright (C) 1991, 1992 Linus Torvalds
+ * Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs
+ * Copyright (C) 2011 Don Zickus Red Hat, Inc.
+ *
+ * Pentium III FXSR, SSE support
+ * Gareth Hughes <gareth@valinux.com>, May 2000
+ */
+
+/*
+ * Handle hardware traps and faults.
+ */
+#include <linux/spinlock.h>
+#include <linux/kprobes.h>
+#include <linux/kdebug.h>
+#include <linux/nmi.h>
+#include <linux/delay.h>
+#include <linux/hardirq.h>
+#include <linux/slab.h>
+
+#include <linux/mca.h>
+
+#if defined(CONFIG_EDAC)
+#include <linux/edac.h>
+#endif
+
+#include <linux/atomic.h>
+#include <asm/traps.h>
+#include <asm/mach_traps.h>
+#include <asm/nmi.h>
+
+#define NMI_MAX_NAMELEN 16
+struct nmiaction {
+ struct list_head list;
+ nmi_handler_t handler;
+ unsigned int flags;
+ char *name;
+};
+
+struct nmi_desc {
+ spinlock_t lock;
+ struct list_head head;
+};
+
+static struct nmi_desc nmi_desc[NMI_MAX] =
+{
+ {
+ .lock = __SPIN_LOCK_UNLOCKED(&nmi_desc[0].lock),
+ .head = LIST_HEAD_INIT(nmi_desc[0].head),
+ },
+ {
+ .lock = __SPIN_LOCK_UNLOCKED(&nmi_desc[1].lock),
+ .head = LIST_HEAD_INIT(nmi_desc[1].head),
+ },
+
+};
+
+struct nmi_stats {
+ unsigned int normal;
+ unsigned int unknown;
+ unsigned int external;
+ unsigned int swallow;
+};
+
+static DEFINE_PER_CPU(struct nmi_stats, nmi_stats);
+
+static int ignore_nmis;
+
+int unknown_nmi_panic;
+/*
+ * Prevent NMI reason port (0x61) being accessed simultaneously, can
+ * only be used in NMI handler.
+ */
+static DEFINE_RAW_SPINLOCK(nmi_reason_lock);
+
+static int __init setup_unknown_nmi_panic(char *str)
+{
+ unknown_nmi_panic = 1;
+ return 1;
+}
+__setup("unknown_nmi_panic", setup_unknown_nmi_panic);
+
+#define nmi_to_desc(type) (&nmi_desc[type])
+
+static int notrace __kprobes nmi_handle(unsigned int type, struct pt_regs *regs, bool b2b)
+{
+ struct nmi_desc *desc = nmi_to_desc(type);
+ struct nmiaction *a;
+ int handled=0;
+
+ rcu_read_lock();
+
+ /*
+ * NMIs are edge-triggered, which means if you have enough
+ * of them concurrently, you can lose some because only one
+ * can be latched at any given time. Walk the whole list
+ * to handle those situations.
+ */
+ list_for_each_entry_rcu(a, &desc->head, list)
+ handled += a->handler(type, regs);
+
+ rcu_read_unlock();
+
+ /* return total number of NMI events handled */
+ return handled;
+}
+
+static int __setup_nmi(unsigned int type, struct nmiaction *action)
+{
+ struct nmi_desc *desc = nmi_to_desc(type);
+ unsigned long flags;
+
+ spin_lock_irqsave(&desc->lock, flags);
+
+ /*
+ * most handlers of type NMI_UNKNOWN never return because
+ * they just assume the NMI is theirs. Just a sanity check
+ * to manage expectations
+ */
+ WARN_ON_ONCE(type == NMI_UNKNOWN && !list_empty(&desc->head));
+
+ /*
+ * some handlers need to be executed first otherwise a fake
+ * event confuses some handlers (kdump uses this flag)
+ */
+ if (action->flags & NMI_FLAG_FIRST)
+ list_add_rcu(&action->list, &desc->head);
+ else
+ list_add_tail_rcu(&action->list, &desc->head);
+
+ spin_unlock_irqrestore(&desc->lock, flags);
+ return 0;
+}
+
+static struct nmiaction *__free_nmi(unsigned int type, const char *name)
+{
+ struct nmi_desc *desc = nmi_to_desc(type);
+ struct nmiaction *n;
+ unsigned long flags;
+
+ spin_lock_irqsave(&desc->lock, flags);
+
+ list_for_each_entry_rcu(n, &desc->head, list) {
+ /*
+ * the name passed in to describe the nmi handler
+ * is used as the lookup key
+ */
+ if (!strcmp(n->name, name)) {
+ WARN(in_nmi(),
+ "Trying to free NMI (%s) from NMI context!\n", n->name);
+ list_del_rcu(&n->list);
+ break;
+ }
+ }
+
+ spin_unlock_irqrestore(&desc->lock, flags);
+ synchronize_rcu();
+ return (n);
+}
+
+int register_nmi_handler(unsigned int type, nmi_handler_t handler,
+ unsigned long nmiflags, const char *devname)
+{
+ struct nmiaction *action;
+ int retval = -ENOMEM;
+
+ if (!handler)
+ return -EINVAL;
+
+ action = kzalloc(sizeof(struct nmiaction), GFP_KERNEL);
+ if (!action)
+ goto fail_action;
+
+ action->handler = handler;
+ action->flags = nmiflags;
+ action->name = kstrndup(devname, NMI_MAX_NAMELEN, GFP_KERNEL);
+ if (!action->name)
+ goto fail_action_name;
+
+ retval = __setup_nmi(type, action);
+
+ if (retval)
+ goto fail_setup_nmi;
+
+ return retval;
+
+fail_setup_nmi:
+ kfree(action->name);
+fail_action_name:
+ kfree(action);
+fail_action:
+
+ return retval;
+}
+EXPORT_SYMBOL_GPL(register_nmi_handler);
+
+void unregister_nmi_handler(unsigned int type, const char *name)
+{
+ struct nmiaction *a;
+
+ a = __free_nmi(type, name);
+ if (a) {
+ kfree(a->name);
+ kfree(a);
+ }
+}
+
+EXPORT_SYMBOL_GPL(unregister_nmi_handler);
+
+static notrace __kprobes void
+pci_serr_error(unsigned char reason, struct pt_regs *regs)
+{
+ pr_emerg("NMI: PCI system error (SERR) for reason %02x on CPU %d.\n",
+ reason, smp_processor_id());
+
+ /*
+ * On some machines, PCI SERR line is used to report memory
+ * errors. EDAC makes use of it.
+ */
+#if defined(CONFIG_EDAC)
+ if (edac_handler_set()) {
+ edac_atomic_assert_error();
+ return;
+ }
+#endif
+
+ if (panic_on_unrecovered_nmi)
+ panic("NMI: Not continuing");
+
+ pr_emerg("Dazed and confused, but trying to continue\n");
+
+ /* Clear and disable the PCI SERR error line. */
+ reason = (reason & NMI_REASON_CLEAR_MASK) | NMI_REASON_CLEAR_SERR;
+ outb(reason, NMI_REASON_PORT);
+}
+
+static notrace __kprobes void
+io_check_error(unsigned char reason, struct pt_regs *regs)
+{
+ unsigned long i;
+
+ pr_emerg(
+ "NMI: IOCK error (debug interrupt?) for reason %02x on CPU %d.\n",
+ reason, smp_processor_id());
+ show_registers(regs);
+
+ if (panic_on_io_nmi)
+ panic("NMI IOCK error: Not continuing");
+
+ /* Re-enable the IOCK line, wait for a few seconds */
+ reason = (reason & NMI_REASON_CLEAR_MASK) | NMI_REASON_CLEAR_IOCHK;
+ outb(reason, NMI_REASON_PORT);
+
+ i = 20000;
+ while (--i) {
+ touch_nmi_watchdog();
+ udelay(100);
+ }
+
+ reason &= ~NMI_REASON_CLEAR_IOCHK;
+ outb(reason, NMI_REASON_PORT);
+}
+
+static notrace __kprobes void
+unknown_nmi_error(unsigned char reason, struct pt_regs *regs)
+{
+ int handled;
+
+ /*
+ * Use 'false' as back-to-back NMIs are dealt with one level up.
+ * Of course this makes having multiple 'unknown' handlers useless
+ * as only the first one is ever run (unless it can actually determine
+ * if it caused the NMI)
+ */
+ handled = nmi_handle(NMI_UNKNOWN, regs, false);
+ if (handled) {
+ __this_cpu_add(nmi_stats.unknown, handled);
+ return;
+ }
+
+ __this_cpu_add(nmi_stats.unknown, 1);
+
+#ifdef CONFIG_MCA
+ /*
+ * Might actually be able to figure out what the guilty party
+ * is:
+ */
+ if (MCA_bus) {
+ mca_handle_nmi();
+ return;
+ }
+#endif
+ pr_emerg("Uhhuh. NMI received for unknown reason %02x on CPU %d.\n",
+ reason, smp_processor_id());
+
+ pr_emerg("Do you have a strange power saving mode enabled?\n");
+ if (unknown_nmi_panic || panic_on_unrecovered_nmi)
+ panic("NMI: Not continuing");
+
+ pr_emerg("Dazed and confused, but trying to continue\n");
+}
+
+static DEFINE_PER_CPU(bool, swallow_nmi);
+static DEFINE_PER_CPU(unsigned long, last_nmi_rip);
+
+static notrace __kprobes void default_do_nmi(struct pt_regs *regs)
+{
+ unsigned char reason = 0;
+ int handled;
+ bool b2b = false;
+
+ /*
+ * CPU-specific NMI must be processed before non-CPU-specific
+ * NMI, otherwise we may lose it, because the CPU-specific
+ * NMI can not be detected/processed on other CPUs.
+ */
+
+ /*
+ * Back-to-back NMIs are interesting because they can either
+ * be two NMI or more than two NMIs (any thing over two is dropped
+ * due to NMI being edge-triggered). If this is the second half
+ * of the back-to-back NMI, assume we dropped things and process
+ * more handlers. Otherwise reset the 'swallow' NMI behaviour
+ */
+ if (regs->ip == __this_cpu_read(last_nmi_rip))
+ b2b = true;
+ else
+ __this_cpu_write(swallow_nmi, false);
+
+ __this_cpu_write(last_nmi_rip, regs->ip);
+
+ handled = nmi_handle(NMI_LOCAL, regs, b2b);
+ __this_cpu_add(nmi_stats.normal, handled);
+ if (handled) {
+ /*
+ * There are cases when a NMI handler handles multiple
+ * events in the current NMI. One of these events may
+ * be queued for in the next NMI. Because the event is
+ * already handled, the next NMI will result in an unknown
+ * NMI. Instead lets flag this for a potential NMI to
+ * swallow.
+ */
+ if (handled > 1)
+ __this_cpu_write(swallow_nmi, true);
+ return;
+ }
+
+ /* Non-CPU-specific NMI: NMI sources can be processed on any CPU */
+ raw_spin_lock(&nmi_reason_lock);
+ reason = get_nmi_reason();
+
+ if (reason & NMI_REASON_MASK) {
+ if (reason & NMI_REASON_SERR)
+ pci_serr_error(reason, regs);
+ else if (reason & NMI_REASON_IOCHK)
+ io_check_error(reason, regs);
+#ifdef CONFIG_X86_32
+ /*
+ * Reassert NMI in case it became active
+ * meanwhile as it's edge-triggered:
+ */
+ reassert_nmi();
+#endif
+ __this_cpu_add(nmi_stats.external, 1);
+ raw_spin_unlock(&nmi_reason_lock);
+ return;
+ }
+ raw_spin_unlock(&nmi_reason_lock);
+
+ /*
+ * Only one NMI can be latched at a time. To handle
+ * this we may process multiple nmi handlers at once to
+ * cover the case where an NMI is dropped. The downside
+ * to this approach is we may process an NMI prematurely,
+ * while its real NMI is sitting latched. This will cause
+ * an unknown NMI on the next run of the NMI processing.
+ *
+ * We tried to flag that condition above, by setting the
+ * swallow_nmi flag when we process more than one event.
+ * This condition is also only present on the second half
+ * of a back-to-back NMI, so we flag that condition too.
+ *
+ * If both are true, we assume we already processed this
+ * NMI previously and we swallow it. Otherwise we reset
+ * the logic.
+ *
+ * There are scenarios where we may accidentally swallow
+ * a 'real' unknown NMI. For example, while processing
+ * a perf NMI another perf NMI comes in along with a
+ * 'real' unknown NMI. These two NMIs get combined into
+ * one (as descibed above). When the next NMI gets
+ * processed, it will be flagged by perf as handled, but
+ * noone will know that there was a 'real' unknown NMI sent
+ * also. As a result it gets swallowed. Or if the first
+ * perf NMI returns two events handled then the second
+ * NMI will get eaten by the logic below, again losing a
+ * 'real' unknown NMI. But this is the best we can do
+ * for now.
+ */
+ if (b2b && __this_cpu_read(swallow_nmi))
+ __this_cpu_add(nmi_stats.swallow, 1);
+ else
+ unknown_nmi_error(reason, regs);
+}
+
+dotraplinkage notrace __kprobes void
+do_nmi(struct pt_regs *regs, long error_code)
+{
+ nmi_enter();
+
+ inc_irq_stat(__nmi_count);
+
+ if (!ignore_nmis)
+ default_do_nmi(regs);
+
+ nmi_exit();
+}
+
+void stop_nmi(void)
+{
+ ignore_nmis++;
+}
+
+void restart_nmi(void)
+{
+ ignore_nmis--;
+}
+
+/* reset the back-to-back NMI logic */
+void local_touch_nmi(void)
+{
+ __this_cpu_write(last_nmi_rip, 0);
+}
diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c
index b49d00da2ae..622872054fb 100644
--- a/arch/x86/kernel/pci-dma.c
+++ b/arch/x86/kernel/pci-dma.c
@@ -117,8 +117,8 @@ again:
}
/*
- * See <Documentation/x86_64/boot-options.txt> for the iommu kernel parameter
- * documentation.
+ * See <Documentation/x86/x86_64/boot-options.txt> for the iommu kernel
+ * parameter documentation.
*/
static __init int iommu_setup(char *p)
{
diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index e7e3b019c43..b9b3b1a5164 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -49,7 +49,7 @@ void free_thread_xstate(struct task_struct *tsk)
void free_thread_info(struct thread_info *ti)
{
free_thread_xstate(ti->task);
- free_pages((unsigned long)ti, get_order(THREAD_SIZE));
+ free_pages((unsigned long)ti, THREAD_ORDER);
}
void arch_task_cache_init(void)
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 7a3b65107a2..795b79f984c 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -57,6 +57,7 @@
#include <asm/idle.h>
#include <asm/syscalls.h>
#include <asm/debugreg.h>
+#include <asm/nmi.h>
asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
@@ -107,6 +108,7 @@ void cpu_idle(void)
if (cpu_is_offline(cpu))
play_dead();
+ local_touch_nmi();
local_irq_disable();
/* Don't trace irqs off for idle */
stop_critical_timings();
@@ -262,7 +264,7 @@ EXPORT_SYMBOL_GPL(start_thread);
/*
- * switch_to(x,yn) should switch tasks from x to y.
+ * switch_to(x,y) should switch tasks from x to y.
*
* We fsave/fwait so that an exception goes off at the right time
* (as a call from the fsave or fwait in effect) rather than to
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index f693e44e1bf..3bd7e6eebf3 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -51,6 +51,7 @@
#include <asm/idle.h>
#include <asm/syscalls.h>
#include <asm/debugreg.h>
+#include <asm/nmi.h>
asmlinkage extern void ret_from_fork(void);
@@ -133,6 +134,7 @@ void cpu_idle(void)
* from here on, until they go to idle.
* Otherwise, idle callbacks can misfire.
*/
+ local_touch_nmi();
local_irq_disable();
enter_idle();
/* Don't trace irqs off for idle */
diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c
index 9242436e993..e334be1182b 100644
--- a/arch/x86/kernel/reboot.c
+++ b/arch/x86/kernel/reboot.c
@@ -464,7 +464,7 @@ static inline void kb_wait(void)
}
}
-static void vmxoff_nmi(int cpu, struct die_args *args)
+static void vmxoff_nmi(int cpu, struct pt_regs *regs)
{
cpu_emergency_vmxoff();
}
@@ -736,14 +736,10 @@ static nmi_shootdown_cb shootdown_callback;
static atomic_t waiting_for_crash_ipi;
-static int crash_nmi_callback(struct notifier_block *self,
- unsigned long val, void *data)
+static int crash_nmi_callback(unsigned int val, struct pt_regs *regs)
{
int cpu;
- if (val != DIE_NMI)
- return NOTIFY_OK;
-
cpu = raw_smp_processor_id();
/* Don't do anything if this handler is invoked on crashing cpu.
@@ -751,10 +747,10 @@ static int crash_nmi_callback(struct notifier_block *self,
* an NMI if system was initially booted with nmi_watchdog parameter.
*/
if (cpu == crashing_cpu)
- return NOTIFY_STOP;
+ return NMI_HANDLED;
local_irq_disable();
- shootdown_callback(cpu, (struct die_args *)data);
+ shootdown_callback(cpu, regs);
atomic_dec(&waiting_for_crash_ipi);
/* Assume hlt works */
@@ -762,7 +758,7 @@ static int crash_nmi_callback(struct notifier_block *self,
for (;;)
cpu_relax();
- return 1;
+ return NMI_HANDLED;
}
static void smp_send_nmi_allbutself(void)
@@ -770,12 +766,6 @@ static void smp_send_nmi_allbutself(void)
apic->send_IPI_allbutself(NMI_VECTOR);
}
-static struct notifier_block crash_nmi_nb = {
- .notifier_call = crash_nmi_callback,
- /* we want to be the first one called */
- .priority = NMI_LOCAL_HIGH_PRIOR+1,
-};
-
/* Halt all other CPUs, calling the specified function on each of them
*
* This function can be used to halt all other CPUs on crash
@@ -794,7 +784,8 @@ void nmi_shootdown_cpus(nmi_shootdown_cb callback)
atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1);
/* Would it be better to replace the trap vector here? */
- if (register_die_notifier(&crash_nmi_nb))
+ if (register_nmi_handler(NMI_LOCAL, crash_nmi_callback,
+ NMI_FLAG_FIRST, "crash"))
return; /* return what? */
/* Ensure the new callback function is set before sending
* out the NMI
diff --git a/arch/x86/kernel/rtc.c b/arch/x86/kernel/rtc.c
index 3f2ad2640d8..ccdbc16b894 100644
--- a/arch/x86/kernel/rtc.c
+++ b/arch/x86/kernel/rtc.c
@@ -42,8 +42,11 @@ int mach_set_rtc_mmss(unsigned long nowtime)
{
int real_seconds, real_minutes, cmos_minutes;
unsigned char save_control, save_freq_select;
+ unsigned long flags;
int retval = 0;
+ spin_lock_irqsave(&rtc_lock, flags);
+
/* tell the clock it's being set */
save_control = CMOS_READ(RTC_CONTROL);
CMOS_WRITE((save_control|RTC_SET), RTC_CONTROL);
@@ -93,12 +96,17 @@ int mach_set_rtc_mmss(unsigned long nowtime)
CMOS_WRITE(save_control, RTC_CONTROL);
CMOS_WRITE(save_freq_select, RTC_FREQ_SELECT);
+ spin_unlock_irqrestore(&rtc_lock, flags);
+
return retval;
}
unsigned long mach_get_cmos_time(void)
{
unsigned int status, year, mon, day, hour, min, sec, century = 0;
+ unsigned long flags;
+
+ spin_lock_irqsave(&rtc_lock, flags);
/*
* If UIP is clear, then we have >= 244 microseconds before
@@ -125,6 +133,8 @@ unsigned long mach_get_cmos_time(void)
status = CMOS_READ(RTC_CONTROL);
WARN_ON_ONCE(RTC_ALWAYS_BCD && (status & RTC_DM_BINARY));
+ spin_unlock_irqrestore(&rtc_lock, flags);
+
if (RTC_ALWAYS_BCD || !(status & RTC_DM_BINARY)) {
sec = bcd2bin(sec);
min = bcd2bin(min);
@@ -169,24 +179,15 @@ EXPORT_SYMBOL(rtc_cmos_write);
int update_persistent_clock(struct timespec now)
{
- unsigned long flags;
- int retval;
-
- spin_lock_irqsave(&rtc_lock, flags);
- retval = x86_platform.set_wallclock(now.tv_sec);
- spin_unlock_irqrestore(&rtc_lock, flags);
-
- return retval;
+ return x86_platform.set_wallclock(now.tv_sec);
}
/* not static: needed by APM */
void read_persistent_clock(struct timespec *ts)
{
- unsigned long retval, flags;
+ unsigned long retval;
- spin_lock_irqsave(&rtc_lock, flags);
retval = x86_platform.get_wallclock();
- spin_unlock_irqrestore(&rtc_lock, flags);
ts->tv_sec = retval;
ts->tv_nsec = 0;
diff --git a/arch/x86/kernel/sys_x86_64.c b/arch/x86/kernel/sys_x86_64.c
index ff14a5044ce..051489082d5 100644
--- a/arch/x86/kernel/sys_x86_64.c
+++ b/arch/x86/kernel/sys_x86_64.c
@@ -14,10 +14,73 @@
#include <linux/personality.h>
#include <linux/random.h>
#include <linux/uaccess.h>
+#include <linux/elf.h>
#include <asm/ia32.h>
#include <asm/syscalls.h>
+/*
+ * Align a virtual address to avoid aliasing in the I$ on AMD F15h.
+ *
+ * @flags denotes the allocation direction - bottomup or topdown -
+ * or vDSO; see call sites below.
+ */
+unsigned long align_addr(unsigned long addr, struct file *filp,
+ enum align_flags flags)
+{
+ unsigned long tmp_addr;
+
+ /* handle 32- and 64-bit case with a single conditional */
+ if (va_align.flags < 0 || !(va_align.flags & (2 - mmap_is_ia32())))
+ return addr;
+
+ if (!(current->flags & PF_RANDOMIZE))
+ return addr;
+
+ if (!((flags & ALIGN_VDSO) || filp))
+ return addr;
+
+ tmp_addr = addr;
+
+ /*
+ * We need an address which is <= than the original
+ * one only when in topdown direction.
+ */
+ if (!(flags & ALIGN_TOPDOWN))
+ tmp_addr += va_align.mask;
+
+ tmp_addr &= ~va_align.mask;
+
+ return tmp_addr;
+}
+
+static int __init control_va_addr_alignment(char *str)
+{
+ /* guard against enabling this on other CPU families */
+ if (va_align.flags < 0)
+ return 1;
+
+ if (*str == 0)
+ return 1;
+
+ if (*str == '=')
+ str++;
+
+ if (!strcmp(str, "32"))
+ va_align.flags = ALIGN_VA_32;
+ else if (!strcmp(str, "64"))
+ va_align.flags = ALIGN_VA_64;
+ else if (!strcmp(str, "off"))
+ va_align.flags = 0;
+ else if (!strcmp(str, "on"))
+ va_align.flags = ALIGN_VA_32 | ALIGN_VA_64;
+ else
+ return 0;
+
+ return 1;
+}
+__setup("align_va_addr", control_va_addr_alignment);
+
SYSCALL_DEFINE6(mmap, unsigned long, addr, unsigned long, len,
unsigned long, prot, unsigned long, flags,
unsigned long, fd, unsigned long, off)
@@ -92,6 +155,9 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr,
start_addr = addr;
full_search:
+
+ addr = align_addr(addr, filp, 0);
+
for (vma = find_vma(mm, addr); ; vma = vma->vm_next) {
/* At this point: (!vma || addr < vma->vm_end). */
if (end - len < addr) {
@@ -117,6 +183,7 @@ full_search:
mm->cached_hole_size = vma->vm_start - addr;
addr = vma->vm_end;
+ addr = align_addr(addr, filp, 0);
}
}
@@ -161,10 +228,13 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
/* make sure it can fit in the remaining address space */
if (addr > len) {
- vma = find_vma(mm, addr-len);
- if (!vma || addr <= vma->vm_start)
+ unsigned long tmp_addr = align_addr(addr - len, filp,
+ ALIGN_TOPDOWN);
+
+ vma = find_vma(mm, tmp_addr);
+ if (!vma || tmp_addr + len <= vma->vm_start)
/* remember the address as a hint for next time */
- return mm->free_area_cache = addr-len;
+ return mm->free_area_cache = tmp_addr;
}
if (mm->mmap_base < len)
@@ -173,6 +243,8 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
addr = mm->mmap_base-len;
do {
+ addr = align_addr(addr, filp, ALIGN_TOPDOWN);
+
/*
* Lookup failure means no vma is above this address,
* else if new region fits below vma->vm_start,
diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S
index bc19be332bc..9a0e3129392 100644
--- a/arch/x86/kernel/syscall_table_32.S
+++ b/arch/x86/kernel/syscall_table_32.S
@@ -346,3 +346,5 @@ ENTRY(sys_call_table)
.long sys_syncfs
.long sys_sendmmsg /* 345 */
.long sys_setns
+ .long sys_process_vm_readv
+ .long sys_process_vm_writev
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index 6913369c234..a8e3eb83466 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -81,15 +81,6 @@ gate_desc idt_table[NR_VECTORS] __page_aligned_data = { { { { 0, 0 } } }, };
DECLARE_BITMAP(used_vectors, NR_VECTORS);
EXPORT_SYMBOL_GPL(used_vectors);
-static int ignore_nmis;
-
-int unknown_nmi_panic;
-/*
- * Prevent NMI reason port (0x61) being accessed simultaneously, can
- * only be used in NMI handler.
- */
-static DEFINE_RAW_SPINLOCK(nmi_reason_lock);
-
static inline void conditional_sti(struct pt_regs *regs)
{
if (regs->flags & X86_EFLAGS_IF)
@@ -307,152 +298,6 @@ gp_in_kernel:
die("general protection fault", regs, error_code);
}
-static int __init setup_unknown_nmi_panic(char *str)
-{
- unknown_nmi_panic = 1;
- return 1;
-}
-__setup("unknown_nmi_panic", setup_unknown_nmi_panic);
-
-static notrace __kprobes void
-pci_serr_error(unsigned char reason, struct pt_regs *regs)
-{
- pr_emerg("NMI: PCI system error (SERR) for reason %02x on CPU %d.\n",
- reason, smp_processor_id());
-
- /*
- * On some machines, PCI SERR line is used to report memory
- * errors. EDAC makes use of it.
- */
-#if defined(CONFIG_EDAC)
- if (edac_handler_set()) {
- edac_atomic_assert_error();
- return;
- }
-#endif
-
- if (panic_on_unrecovered_nmi)
- panic("NMI: Not continuing");
-
- pr_emerg("Dazed and confused, but trying to continue\n");
-
- /* Clear and disable the PCI SERR error line. */
- reason = (reason & NMI_REASON_CLEAR_MASK) | NMI_REASON_CLEAR_SERR;
- outb(reason, NMI_REASON_PORT);
-}
-
-static notrace __kprobes void
-io_check_error(unsigned char reason, struct pt_regs *regs)
-{
- unsigned long i;
-
- pr_emerg(
- "NMI: IOCK error (debug interrupt?) for reason %02x on CPU %d.\n",
- reason, smp_processor_id());
- show_registers(regs);
-
- if (panic_on_io_nmi)
- panic("NMI IOCK error: Not continuing");
-
- /* Re-enable the IOCK line, wait for a few seconds */
- reason = (reason & NMI_REASON_CLEAR_MASK) | NMI_REASON_CLEAR_IOCHK;
- outb(reason, NMI_REASON_PORT);
-
- i = 20000;
- while (--i) {
- touch_nmi_watchdog();
- udelay(100);
- }
-
- reason &= ~NMI_REASON_CLEAR_IOCHK;
- outb(reason, NMI_REASON_PORT);
-}
-
-static notrace __kprobes void
-unknown_nmi_error(unsigned char reason, struct pt_regs *regs)
-{
- if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) ==
- NOTIFY_STOP)
- return;
-#ifdef CONFIG_MCA
- /*
- * Might actually be able to figure out what the guilty party
- * is:
- */
- if (MCA_bus) {
- mca_handle_nmi();
- return;
- }
-#endif
- pr_emerg("Uhhuh. NMI received for unknown reason %02x on CPU %d.\n",
- reason, smp_processor_id());
-
- pr_emerg("Do you have a strange power saving mode enabled?\n");
- if (unknown_nmi_panic || panic_on_unrecovered_nmi)
- panic("NMI: Not continuing");
-
- pr_emerg("Dazed and confused, but trying to continue\n");
-}
-
-static notrace __kprobes void default_do_nmi(struct pt_regs *regs)
-{
- unsigned char reason = 0;
-
- /*
- * CPU-specific NMI must be processed before non-CPU-specific
- * NMI, otherwise we may lose it, because the CPU-specific
- * NMI can not be detected/processed on other CPUs.
- */
- if (notify_die(DIE_NMI, "nmi", regs, 0, 2, SIGINT) == NOTIFY_STOP)
- return;
-
- /* Non-CPU-specific NMI: NMI sources can be processed on any CPU */
- raw_spin_lock(&nmi_reason_lock);
- reason = get_nmi_reason();
-
- if (reason & NMI_REASON_MASK) {
- if (reason & NMI_REASON_SERR)
- pci_serr_error(reason, regs);
- else if (reason & NMI_REASON_IOCHK)
- io_check_error(reason, regs);
-#ifdef CONFIG_X86_32
- /*
- * Reassert NMI in case it became active
- * meanwhile as it's edge-triggered:
- */
- reassert_nmi();
-#endif
- raw_spin_unlock(&nmi_reason_lock);
- return;
- }
- raw_spin_unlock(&nmi_reason_lock);
-
- unknown_nmi_error(reason, regs);
-}
-
-dotraplinkage notrace __kprobes void
-do_nmi(struct pt_regs *regs, long error_code)
-{
- nmi_enter();
-
- inc_irq_stat(__nmi_count);
-
- if (!ignore_nmis)
- default_do_nmi(regs);
-
- nmi_exit();
-}
-
-void stop_nmi(void)
-{
- ignore_nmis++;
-}
-
-void restart_nmi(void)
-{
- ignore_nmis--;
-}
-
/* May run on IST stack. */
dotraplinkage void __kprobes do_int3(struct pt_regs *regs, long error_code)
{
diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c
index 18ae83dd1cd..b56c65de384 100644
--- a/arch/x86/kernel/vsyscall_64.c
+++ b/arch/x86/kernel/vsyscall_64.c
@@ -56,7 +56,7 @@ DEFINE_VVAR(struct vsyscall_gtod_data, vsyscall_gtod_data) =
.lock = __SEQLOCK_UNLOCKED(__vsyscall_gtod_data.lock),
};
-static enum { EMULATE, NATIVE, NONE } vsyscall_mode = EMULATE;
+static enum { EMULATE, NATIVE, NONE } vsyscall_mode = NATIVE;
static int __init vsyscall_setup(char *str)
{