summaryrefslogtreecommitdiff
path: root/arch
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2009-09-11 13:22:30 -0700
committerLinus Torvalds <torvalds@linux-foundation.org>2009-09-11 13:22:30 -0700
commitb9356c53ba2f593081e5aa45eb67adcce243d1c0 (patch)
treedebf2b72d5241dbaa5b4f2088feb0a2584388792 /arch
parentd90a7e86401ffea2163a4337f3a47f3909c4e255 (diff)
parent4680e64a88c4ce2c4e736dade99233e3def13fa7 (diff)
Merge branch 'oprofile-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip
* 'oprofile-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (55 commits) arch/x86/oprofile/op_model_amd.c: fix op_amd_handle_ibs() return type Revert "x86: oprofile/op_model_amd.c set return values for op_amd_handle_ibs()" x86/oprofile: Small coding style fixes x86/oprofile: Add counter reservation check for virtual counters x86/oprofile: Implement op_x86_virt_to_phys() oprofile: Adding switch counter to oprofile statistic variables x86/oprofile: Implement mux_clone() x86/oprofile: Enable multiplexing only if the model supports it x86/oprofile: Add function has_mux() to check multiplexing support x86/oprofile: Modify initialization of num_virt_counters x86/oprofile: Remove unused num_virt_controls from struct op_x86_model_spec x86/oprofile: Remove const qualifier from struct op_x86_model_spec x86/oprofile: Moving nmi_cpu_switch() in nmi_int.c x86/oprofile: Moving nmi_cpu_save/restore_mpx_registers() in nmi_int.c x86/oprofile: Moving nmi_setup_cpu_mux() in nmi_int.c x86/oprofile: Implement multiplexing setup/shutdown functions oprofile: Grouping multiplexing code in op_model_amd.c oprofile: Introduce op_x86_phys_to_virt() oprofile: Grouping multiplexing code in oprof.c oprofile: Remove oprofile_multiplexing_init() ...
Diffstat (limited to 'arch')
-rw-r--r--arch/Kconfig12
-rw-r--r--arch/x86/oprofile/nmi_int.c404
-rw-r--r--arch/x86/oprofile/op_counter.h2
-rw-r--r--arch/x86/oprofile/op_model_amd.c372
-rw-r--r--arch/x86/oprofile/op_model_p4.c72
-rw-r--r--arch/x86/oprofile/op_model_ppro.c101
-rw-r--r--arch/x86/oprofile/op_x86_model.h59
7 files changed, 622 insertions, 400 deletions
diff --git a/arch/Kconfig b/arch/Kconfig
index 99193b160232..beea3ccebb5e 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -30,6 +30,18 @@ config OPROFILE_IBS
If unsure, say N.
+config OPROFILE_EVENT_MULTIPLEX
+ bool "OProfile multiplexing support (EXPERIMENTAL)"
+ default n
+ depends on OPROFILE && X86
+ help
+ The number of hardware counters is limited. The multiplexing
+ feature enables OProfile to gather more events than counters
+ are provided by the hardware. This is realized by switching
+ between events at an user specified time interval.
+
+ If unsure, say N.
+
config HAVE_OPROFILE
bool
diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 89b9a5cd63da..cb88b1a0bd5f 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -1,11 +1,14 @@
/**
* @file nmi_int.c
*
- * @remark Copyright 2002-2008 OProfile authors
+ * @remark Copyright 2002-2009 OProfile authors
* @remark Read the file COPYING
*
* @author John Levon <levon@movementarian.org>
* @author Robert Richter <robert.richter@amd.com>
+ * @author Barry Kasindorf <barry.kasindorf@amd.com>
+ * @author Jason Yeh <jason.yeh@amd.com>
+ * @author Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
*/
#include <linux/init.h>
@@ -24,13 +27,35 @@
#include "op_counter.h"
#include "op_x86_model.h"
-static struct op_x86_model_spec const *model;
+static struct op_x86_model_spec *model;
static DEFINE_PER_CPU(struct op_msrs, cpu_msrs);
static DEFINE_PER_CPU(unsigned long, saved_lvtpc);
/* 0 == registered but off, 1 == registered and on */
static int nmi_enabled = 0;
+struct op_counter_config counter_config[OP_MAX_COUNTER];
+
+/* common functions */
+
+u64 op_x86_get_ctrl(struct op_x86_model_spec const *model,
+ struct op_counter_config *counter_config)
+{
+ u64 val = 0;
+ u16 event = (u16)counter_config->event;
+
+ val |= ARCH_PERFMON_EVENTSEL_INT;
+ val |= counter_config->user ? ARCH_PERFMON_EVENTSEL_USR : 0;
+ val |= counter_config->kernel ? ARCH_PERFMON_EVENTSEL_OS : 0;
+ val |= (counter_config->unit_mask & 0xFF) << 8;
+ event &= model->event_mask ? model->event_mask : 0xFF;
+ val |= event & 0xFF;
+ val |= (event & 0x0F00) << 24;
+
+ return val;
+}
+
+
static int profile_exceptions_notify(struct notifier_block *self,
unsigned long val, void *data)
{
@@ -52,36 +77,214 @@ static int profile_exceptions_notify(struct notifier_block *self,
static void nmi_cpu_save_registers(struct op_msrs *msrs)
{
- unsigned int const nr_ctrs = model->num_counters;
- unsigned int const nr_ctrls = model->num_controls;
struct op_msr *counters = msrs->counters;
struct op_msr *controls = msrs->controls;
unsigned int i;
- for (i = 0; i < nr_ctrs; ++i) {
- if (counters[i].addr) {
- rdmsr(counters[i].addr,
- counters[i].saved.low,
- counters[i].saved.high);
- }
+ for (i = 0; i < model->num_counters; ++i) {
+ if (counters[i].addr)
+ rdmsrl(counters[i].addr, counters[i].saved);
+ }
+
+ for (i = 0; i < model->num_controls; ++i) {
+ if (controls[i].addr)
+ rdmsrl(controls[i].addr, controls[i].saved);
+ }
+}
+
+static void nmi_cpu_start(void *dummy)
+{
+ struct op_msrs const *msrs = &__get_cpu_var(cpu_msrs);
+ model->start(msrs);
+}
+
+static int nmi_start(void)
+{
+ on_each_cpu(nmi_cpu_start, NULL, 1);
+ return 0;
+}
+
+static void nmi_cpu_stop(void *dummy)
+{
+ struct op_msrs const *msrs = &__get_cpu_var(cpu_msrs);
+ model->stop(msrs);
+}
+
+static void nmi_stop(void)
+{
+ on_each_cpu(nmi_cpu_stop, NULL, 1);
+}
+
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+
+static DEFINE_PER_CPU(int, switch_index);
+
+static inline int has_mux(void)
+{
+ return !!model->switch_ctrl;
+}
+
+inline int op_x86_phys_to_virt(int phys)
+{
+ return __get_cpu_var(switch_index) + phys;
+}
+
+inline int op_x86_virt_to_phys(int virt)
+{
+ return virt % model->num_counters;
+}
+
+static void nmi_shutdown_mux(void)
+{
+ int i;
+
+ if (!has_mux())
+ return;
+
+ for_each_possible_cpu(i) {
+ kfree(per_cpu(cpu_msrs, i).multiplex);
+ per_cpu(cpu_msrs, i).multiplex = NULL;
+ per_cpu(switch_index, i) = 0;
}
+}
+
+static int nmi_setup_mux(void)
+{
+ size_t multiplex_size =
+ sizeof(struct op_msr) * model->num_virt_counters;
+ int i;
+
+ if (!has_mux())
+ return 1;
+
+ for_each_possible_cpu(i) {
+ per_cpu(cpu_msrs, i).multiplex =
+ kmalloc(multiplex_size, GFP_KERNEL);
+ if (!per_cpu(cpu_msrs, i).multiplex)
+ return 0;
+ }
+
+ return 1;
+}
+
+static void nmi_cpu_setup_mux(int cpu, struct op_msrs const * const msrs)
+{
+ int i;
+ struct op_msr *multiplex = msrs->multiplex;
+
+ if (!has_mux())
+ return;
- for (i = 0; i < nr_ctrls; ++i) {
- if (controls[i].addr) {
- rdmsr(controls[i].addr,
- controls[i].saved.low,
- controls[i].saved.high);
+ for (i = 0; i < model->num_virt_counters; ++i) {
+ if (counter_config[i].enabled) {
+ multiplex[i].saved = -(u64)counter_config[i].count;
+ } else {
+ multiplex[i].addr = 0;
+ multiplex[i].saved = 0;
}
}
+
+ per_cpu(switch_index, cpu) = 0;
+}
+
+static void nmi_cpu_save_mpx_registers(struct op_msrs *msrs)
+{
+ struct op_msr *multiplex = msrs->multiplex;
+ int i;
+
+ for (i = 0; i < model->num_counters; ++i) {
+ int virt = op_x86_phys_to_virt(i);
+ if (multiplex[virt].addr)
+ rdmsrl(multiplex[virt].addr, multiplex[virt].saved);
+ }
+}
+
+static void nmi_cpu_restore_mpx_registers(struct op_msrs *msrs)
+{
+ struct op_msr *multiplex = msrs->multiplex;
+ int i;
+
+ for (i = 0; i < model->num_counters; ++i) {
+ int virt = op_x86_phys_to_virt(i);
+ if (multiplex[virt].addr)
+ wrmsrl(multiplex[virt].addr, multiplex[virt].saved);
+ }
}
-static void nmi_save_registers(void *dummy)
+static void nmi_cpu_switch(void *dummy)
{
int cpu = smp_processor_id();
+ int si = per_cpu(switch_index, cpu);
struct op_msrs *msrs = &per_cpu(cpu_msrs, cpu);
- nmi_cpu_save_registers(msrs);
+
+ nmi_cpu_stop(NULL);
+ nmi_cpu_save_mpx_registers(msrs);
+
+ /* move to next set */
+ si += model->num_counters;
+ if ((si > model->num_virt_counters) || (counter_config[si].count == 0))
+ per_cpu(switch_index, cpu) = 0;
+ else
+ per_cpu(switch_index, cpu) = si;
+
+ model->switch_ctrl(model, msrs);
+ nmi_cpu_restore_mpx_registers(msrs);
+
+ nmi_cpu_start(NULL);
+}
+
+
+/*
+ * Quick check to see if multiplexing is necessary.
+ * The check should be sufficient since counters are used
+ * in ordre.
+ */
+static int nmi_multiplex_on(void)
+{
+ return counter_config[model->num_counters].count ? 0 : -EINVAL;
+}
+
+static int nmi_switch_event(void)
+{
+ if (!has_mux())
+ return -ENOSYS; /* not implemented */
+ if (nmi_multiplex_on() < 0)
+ return -EINVAL; /* not necessary */
+
+ on_each_cpu(nmi_cpu_switch, NULL, 1);
+
+ return 0;
+}
+
+static inline void mux_init(struct oprofile_operations *ops)
+{
+ if (has_mux())
+ ops->switch_events = nmi_switch_event;
+}
+
+static void mux_clone(int cpu)
+{
+ if (!has_mux())
+ return;
+
+ memcpy(per_cpu(cpu_msrs, cpu).multiplex,
+ per_cpu(cpu_msrs, 0).multiplex,
+ sizeof(struct op_msr) * model->num_virt_counters);
}
+#else
+
+inline int op_x86_phys_to_virt(int phys) { return phys; }
+inline int op_x86_virt_to_phys(int virt) { return virt; }
+static inline void nmi_shutdown_mux(void) { }
+static inline int nmi_setup_mux(void) { return 1; }
+static inline void
+nmi_cpu_setup_mux(int cpu, struct op_msrs const * const msrs) { }
+static inline void mux_init(struct oprofile_operations *ops) { }
+static void mux_clone(int cpu) { }
+
+#endif
+
static void free_msrs(void)
{
int i;
@@ -95,38 +298,32 @@ static void free_msrs(void)
static int allocate_msrs(void)
{
- int success = 1;
size_t controls_size = sizeof(struct op_msr) * model->num_controls;
size_t counters_size = sizeof(struct op_msr) * model->num_counters;
int i;
for_each_possible_cpu(i) {
per_cpu(cpu_msrs, i).counters = kmalloc(counters_size,
- GFP_KERNEL);
- if (!per_cpu(cpu_msrs, i).counters) {
- success = 0;
- break;
- }
+ GFP_KERNEL);
+ if (!per_cpu(cpu_msrs, i).counters)
+ return 0;
per_cpu(cpu_msrs, i).controls = kmalloc(controls_size,
- GFP_KERNEL);
- if (!per_cpu(cpu_msrs, i).controls) {
- success = 0;
- break;
- }
+ GFP_KERNEL);
+ if (!per_cpu(cpu_msrs, i).controls)
+ return 0;
}
- if (!success)
- free_msrs();
-
- return success;
+ return 1;
}
static void nmi_cpu_setup(void *dummy)
{
int cpu = smp_processor_id();
struct op_msrs *msrs = &per_cpu(cpu_msrs, cpu);
+ nmi_cpu_save_registers(msrs);
spin_lock(&oprofilefs_lock);
- model->setup_ctrs(msrs);
+ model->setup_ctrs(model, msrs);
+ nmi_cpu_setup_mux(cpu, msrs);
spin_unlock(&oprofilefs_lock);
per_cpu(saved_lvtpc, cpu) = apic_read(APIC_LVTPC);
apic_write(APIC_LVTPC, APIC_DM_NMI);
@@ -144,11 +341,15 @@ static int nmi_setup(void)
int cpu;
if (!allocate_msrs())
- return -ENOMEM;
+ err = -ENOMEM;
+ else if (!nmi_setup_mux())
+ err = -ENOMEM;
+ else
+ err = register_die_notifier(&profile_exceptions_nb);
- err = register_die_notifier(&profile_exceptions_nb);
if (err) {
free_msrs();
+ nmi_shutdown_mux();
return err;
}
@@ -159,45 +360,38 @@ static int nmi_setup(void)
/* Assume saved/restored counters are the same on all CPUs */
model->fill_in_addresses(&per_cpu(cpu_msrs, 0));
for_each_possible_cpu(cpu) {
- if (cpu != 0) {
- memcpy(per_cpu(cpu_msrs, cpu).counters,
- per_cpu(cpu_msrs, 0).counters,
- sizeof(struct op_msr) * model->num_counters);
-
- memcpy(per_cpu(cpu_msrs, cpu).controls,
- per_cpu(cpu_msrs, 0).controls,
- sizeof(struct op_msr) * model->num_controls);
- }
+ if (!cpu)
+ continue;
+
+ memcpy(per_cpu(cpu_msrs, cpu).counters,
+ per_cpu(cpu_msrs, 0).counters,
+ sizeof(struct op_msr) * model->num_counters);
+
+ memcpy(per_cpu(cpu_msrs, cpu).controls,
+ per_cpu(cpu_msrs, 0).controls,
+ sizeof(struct op_msr) * model->num_controls);
+ mux_clone(cpu);
}
- on_each_cpu(nmi_save_registers, NULL, 1);
on_each_cpu(nmi_cpu_setup, NULL, 1);
nmi_enabled = 1;
return 0;
}
-static void nmi_restore_registers(struct op_msrs *msrs)
+static void nmi_cpu_restore_registers(struct op_msrs *msrs)
{
- unsigned int const nr_ctrs = model->num_counters;
- unsigned int const nr_ctrls = model->num_controls;
struct op_msr *counters = msrs->counters;
struct op_msr *controls = msrs->controls;
unsigned int i;
- for (i = 0; i < nr_ctrls; ++i) {
- if (controls[i].addr) {
- wrmsr(controls[i].addr,
- controls[i].saved.low,
- controls[i].saved.high);
- }
+ for (i = 0; i < model->num_controls; ++i) {
+ if (controls[i].addr)
+ wrmsrl(controls[i].addr, controls[i].saved);
}
- for (i = 0; i < nr_ctrs; ++i) {
- if (counters[i].addr) {
- wrmsr(counters[i].addr,
- counters[i].saved.low,
- counters[i].saved.high);
- }
+ for (i = 0; i < model->num_counters; ++i) {
+ if (counters[i].addr)
+ wrmsrl(counters[i].addr, counters[i].saved);
}
}
@@ -205,7 +399,7 @@ static void nmi_cpu_shutdown(void *dummy)
{
unsigned int v;
int cpu = smp_processor_id();
- struct op_msrs *msrs = &__get_cpu_var(cpu_msrs);
+ struct op_msrs *msrs = &per_cpu(cpu_msrs, cpu);
/* restoring APIC_LVTPC can trigger an apic error because the delivery
* mode and vector nr combination can be illegal. That's by design: on
@@ -216,7 +410,7 @@ static void nmi_cpu_shutdown(void *dummy)
apic_write(APIC_LVTERR, v | APIC_LVT_MASKED);
apic_write(APIC_LVTPC, per_cpu(saved_lvtpc, cpu));
apic_write(APIC_LVTERR, v);
- nmi_restore_registers(msrs);
+ nmi_cpu_restore_registers(msrs);
}
static void nmi_shutdown(void)
@@ -226,42 +420,18 @@ static void nmi_shutdown(void)
nmi_enabled = 0;
on_each_cpu(nmi_cpu_shutdown, NULL, 1);
unregister_die_notifier(&profile_exceptions_nb);
+ nmi_shutdown_mux();
msrs = &get_cpu_var(cpu_msrs);
model->shutdown(msrs);
free_msrs();
put_cpu_var(cpu_msrs);
}
-static void nmi_cpu_start(void *dummy)
-{
- struct op_msrs const *msrs = &__get_cpu_var(cpu_msrs);
- model->start(msrs);
-}
-
-static int nmi_start(void)
-{
- on_each_cpu(nmi_cpu_start, NULL, 1);
- return 0;
-}
-
-static void nmi_cpu_stop(void *dummy)
-{
- struct op_msrs const *msrs = &__get_cpu_var(cpu_msrs);
- model->stop(msrs);
-}
-
-static void nmi_stop(void)
-{
- on_each_cpu(nmi_cpu_stop, NULL, 1);
-}
-
-struct op_counter_config counter_config[OP_MAX_COUNTER];
-
static int nmi_create_files(struct super_block *sb, struct dentry *root)
{
unsigned int i;
- for (i = 0; i < model->num_counters; ++i) {
+ for (i = 0; i < model->num_virt_counters; ++i) {
struct dentry *dir;
char buf[4];
@@ -270,7 +440,7 @@ static int nmi_create_files(struct super_block *sb, struct dentry *root)
* NOTE: assumes 1:1 mapping here (that counters are organized
* sequentially in their struct assignment).
*/
- if (unlikely(!avail_to_resrv_perfctr_nmi_bit(i)))
+ if (!avail_to_resrv_perfctr_nmi_bit(op_x86_virt_to_phys(i)))
continue;
snprintf(buf, sizeof(buf), "%d", i);
@@ -402,6 +572,7 @@ module_param_call(cpu_type, force_cpu_type, NULL, NULL, 0);
static int __init ppro_init(char **cpu_type)
{
__u8 cpu_model = boot_cpu_data.x86_model;
+ struct op_x86_model_spec *spec = &op_ppro_spec; /* default */
if (force_arch_perfmon && cpu_has_arch_perfmon)
return 0;
@@ -428,7 +599,7 @@ static int __init ppro_init(char **cpu_type)
*cpu_type = "i386/core_2";
break;
case 26:
- arch_perfmon_setup_counters();
+ spec = &op_arch_perfmon_spec;
*cpu_type = "i386/core_i7";
break;
case 28:
@@ -439,17 +610,7 @@ static int __init ppro_init(char **cpu_type)
return 0;
}
- model = &op_ppro_spec;
- return 1;
-}
-
-static int __init arch_perfmon_init(char **cpu_type)
-{
- if (!cpu_has_arch_perfmon)
- return 0;
- *cpu_type = "i386/arch_perfmon";
- model = &op_arch_perfmon_spec;
- arch_perfmon_setup_counters();
+ model = spec;
return 1;
}
@@ -471,27 +632,26 @@ int __init op_nmi_init(struct oprofile_operations *ops)
/* Needs to be at least an Athlon (or hammer in 32bit mode) */
switch (family) {
- default:
- return -ENODEV;
case 6:
- model = &op_amd_spec;
cpu_type = "i386/athlon";
break;
case 0xf:
- model = &op_amd_spec;
- /* Actually it could be i386/hammer too, but give
- user space an consistent name. */
+ /*
+ * Actually it could be i386/hammer too, but
+ * give user space an consistent name.
+ */
cpu_type = "x86-64/hammer";
break;
case 0x10:
- model = &op_amd_spec;
cpu_type = "x86-64/family10";
break;
case 0x11:
- model = &op_amd_spec;
cpu_type = "x86-64/family11h";
break;
+ default:
+ return -ENODEV;
}
+ model = &op_amd_spec;
break;
case X86_VENDOR_INTEL:
@@ -510,8 +670,15 @@ int __init op_nmi_init(struct oprofile_operations *ops)
break;
}
- if (!cpu_type && !arch_perfmon_init(&cpu_type))
+ if (cpu_type)
+ break;
+
+ if (!cpu_has_arch_perfmon)
return -ENODEV;
+
+ /* use arch perfmon as fallback */
+ cpu_type = "i386/arch_perfmon";
+ model = &op_arch_perfmon_spec;
break;
default:
@@ -522,18 +689,23 @@ int __init op_nmi_init(struct oprofile_operations *ops)
register_cpu_notifier(&oprofile_cpu_nb);
#endif
/* default values, can be overwritten by model */
- ops->create_files = nmi_create_files;
- ops->setup = nmi_setup;
- ops->shutdown = nmi_shutdown;
- ops->start = nmi_start;
- ops->stop = nmi_stop;
- ops->cpu_type = cpu_type;
+ ops->create_files = nmi_create_files;
+ ops->setup = nmi_setup;
+ ops->shutdown = nmi_shutdown;
+ ops->start = nmi_start;
+ ops->stop = nmi_stop;
+ ops->cpu_type = cpu_type;
if (model->init)
ret = model->init(ops);
if (ret)
return ret;
+ if (!model->num_virt_counters)
+ model->num_virt_counters = model->num_counters;
+
+ mux_init(ops);
+
init_sysfs();
using_nmi = 1;
printk(KERN_INFO "oprofile: using NMI interrupt.\n");
diff --git a/arch/x86/oprofile/op_counter.h b/arch/x86/oprofile/op_counter.h
index 91b6a116165e..e28398df0df2 100644
--- a/arch/x86/oprofile/op_counter.h
+++ b/arch/x86/oprofile/op_counter.h
@@ -10,7 +10,7 @@
#ifndef OP_COUNTER_H
#define OP_COUNTER_H
-#define OP_MAX_COUNTER 8
+#define OP_MAX_COUNTER 32
/* Per-perfctr configuration as set via
* oprofilefs.
diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index 8fdf06e4edf9..39686c29f03a 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -9,12 +9,15 @@
* @author Philippe Elie
* @author Graydon Hoare
* @author Robert Richter <robert.richter@amd.com>
- * @author Barry Kasindorf
+ * @author Barry Kasindorf <barry.kasindorf@amd.com>
+ * @author Jason Yeh <jason.yeh@amd.com>
+ * @author Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
*/
#include <linux/oprofile.h>
#include <linux/device.h>
#include <linux/pci.h>
+#include <linux/percpu.h>
#include <asm/ptrace.h>
#include <asm/msr.h>
@@ -25,43 +28,36 @@
#define NUM_COUNTERS 4
#define NUM_CONTROLS 4
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+#define NUM_VIRT_COUNTERS 32
+#define NUM_VIRT_CONTROLS 32
+#else
+#define NUM_VIRT_COUNTERS NUM_COUNTERS
+#define NUM_VIRT_CONTROLS NUM_CONTROLS
+#endif
+
+#define OP_EVENT_MASK 0x0FFF
+#define OP_CTR_OVERFLOW (1ULL<<31)
-#define CTR_IS_RESERVED(msrs, c) (msrs->counters[(c)].addr ? 1 : 0)
-#define CTR_READ(l, h, msrs, c) do {rdmsr(msrs->counters[(c)].addr, (l), (h)); } while (0)
-#define CTR_WRITE(l, msrs, c) do {wrmsr(msrs->counters[(c)].addr, -(unsigned int)(l), -1); } while (0)
-#define CTR_OVERFLOWED(n) (!((n) & (1U<<31)))
-
-#define CTRL_IS_RESERVED(msrs, c) (msrs->controls[(c)].addr ? 1 : 0)
-#define CTRL_READ(l, h, msrs, c) do {rdmsr(msrs->controls[(c)].addr, (l), (h)); } while (0)
-#define CTRL_WRITE(l, h, msrs, c) do {wrmsr(msrs->controls[(c)].addr, (l), (h)); } while (0)
-#define CTRL_SET_ACTIVE(n) (n |= (1<<22))
-#define CTRL_SET_INACTIVE(n) (n &= ~(1<<22))
-#define CTRL_CLEAR_LO(x) (x &= (1<<21))
-#define CTRL_CLEAR_HI(x) (x &= 0xfffffcf0)
-#define CTRL_SET_ENABLE(val) (val |= 1<<20)
-#define CTRL_SET_USR(val, u) (val |= ((u & 1) << 16))
-#define CTRL_SET_KERN(val, k) (val |= ((k & 1) << 17))
-#define CTRL_SET_UM(val, m) (val |= (m << 8))
-#define CTRL_SET_EVENT_LOW(val, e) (val |= (e & 0xff))
-#define CTRL_SET_EVENT_HIGH(val, e) (val |= ((e >> 8) & 0xf))
-#define CTRL_SET_HOST_ONLY(val, h) (val |= ((h & 1) << 9))
-#define CTRL_SET_GUEST_ONLY(val, h) (val |= ((h & 1) << 8))
-
-static unsigned long reset_value[NUM_COUNTERS];
+#define MSR_AMD_EVENTSEL_RESERVED ((0xFFFFFCF0ULL<<32)|(1ULL<<21))
+
+static unsigned long reset_value[NUM_VIRT_COUNTERS];
#ifdef CONFIG_OPROFILE_IBS
/* IbsFetchCtl bits/masks */
-#define IBS_FETCH_HIGH_VALID_BIT (1UL << 17) /* bit 49 */
-#define IBS_FETCH_HIGH_ENABLE (1UL << 16) /* bit 48 */
-#define IBS_FETCH_LOW_MAX_CNT_MASK 0x0000FFFFUL /* MaxCnt mask */
+#define IBS_FETCH_RAND_EN (1ULL<<57)
+#define IBS_FETCH_VAL (1ULL<<49)
+#define IBS_FETCH_ENABLE (1ULL<<48)
+#define IBS_FETCH_CNT_MASK 0xFFFF0000ULL
/*IbsOpCtl bits */
-#define IBS_OP_LOW_VALID_BIT (1ULL<<18) /* bit 18 */
-#define IBS_OP_LOW_ENABLE (1ULL<<17) /* bit 17 */
+#define IBS_OP_CNT_CTL (1ULL<<19)
+#define IBS_OP_VAL (1ULL<<18)
+#define IBS_OP_ENABLE (1ULL<<17)
-#define IBS_FETCH_SIZE 6
-#define IBS_OP_SIZE 12
+#define IBS_FETCH_SIZE 6
+#define IBS_OP_SIZE 12
static int has_ibs; /* AMD Family10h and later */
@@ -78,6 +74,45 @@ static struct op_ibs_config ibs_config;
#endif
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+
+static void op_mux_fill_in_addresses(struct op_msrs * const msrs)
+{
+ int i;
+
+ for (i = 0; i < NUM_VIRT_COUNTERS; i++) {
+ int hw_counter = op_x86_virt_to_phys(i);
+ if (reserve_perfctr_nmi(MSR_K7_PERFCTR0 + i))
+ msrs->multiplex[i].addr = MSR_K7_PERFCTR0 + hw_counter;
+ else
+ msrs->multiplex[i].addr = 0;
+ }
+}
+
+static void op_mux_switch_ctrl(struct op_x86_model_spec const *model,
+ struct op_msrs const * const msrs)
+{
+ u64 val;
+ int i;
+
+ /* enable active counters */
+ for (i = 0; i < NUM_COUNTERS; ++i) {
+ int virt = op_x86_phys_to_virt(i);
+ if (!counter_config[virt].enabled)
+ continue;
+ rdmsrl(msrs->controls[i].addr, val);
+ val &= model->reserved;
+ val |= op_x86_get_ctrl(model, &counter_config[virt]);
+ wrmsrl(msrs->controls[i].addr, val);
+ }
+}
+
+#else
+
+static inline void op_mux_fill_in_addresses(struct op_msrs * const msrs) { }
+
+#endif
+
/* functions for op_amd_spec */
static void op_amd_fill_in_addresses(struct op_msrs * const msrs)
@@ -97,150 +132,174 @@ static void op_amd_fill_in_addresses(struct op_msrs * const msrs)
else
msrs->controls[i].addr = 0;
}
-}
+ op_mux_fill_in_addresses(msrs);
+}
-static void op_amd_setup_ctrs(struct op_msrs const * const msrs)
+static void op_amd_setup_ctrs(struct op_x86_model_spec const *model,
+ struct op_msrs const * const msrs)
{
- unsigned int low, high;
+ u64 val;
int i;
+ /* setup reset_value */
+ for (i = 0; i < NUM_VIRT_COUNTERS; ++i) {
+ if (counter_config[i].enabled)
+ reset_value[i] = counter_config[i].count;
+ else
+ reset_value[i] = 0;
+ }
+
/* clear all counters */
- for (i = 0 ; i < NUM_CONTROLS; ++i) {
- if (unlikely(!CTRL_IS_RESERVED(msrs, i)))
+ for (i = 0; i < NUM_CONTROLS; ++i) {
+ if (unlikely(!msrs->controls[i].addr))
continue;
- CTRL_READ(low, high, msrs, i);
- CTRL_CLEAR_LO(low);
- CTRL_CLEAR_HI(high);
- CTRL_WRITE(low, high, msrs, i);
+ rdmsrl(msrs->controls[i].addr, val);
+ val &= model->reserved;
+ wrmsrl(msrs->controls[i].addr, val);
}
/* avoid a false detection of ctr overflows in NMI handler */
for (i = 0; i < NUM_COUNTERS; ++i) {
- if (unlikely(!CTR_IS_RESERVED(msrs, i)))
+ if (unlikely(!msrs->counters[i].addr))
continue;
- CTR_WRITE(1, msrs, i);
+ wrmsrl(msrs->counters[i].addr, -1LL);
}
/* enable active counters */
for (i = 0; i < NUM_COUNTERS; ++i) {
- if ((counter_config[i].enabled) && (CTR_IS_RESERVED(msrs, i))) {
- reset_value[i] = counter_config[i].count;
+ int virt = op_x86_phys_to_virt(i);
+ if (!counter_config[virt].enabled)
+ continue;
+ if (!msrs->counters[i].addr)
+ continue;
- CTR_WRITE(counter_config[i].count, msrs, i);
-
- CTRL_READ(low, high, msrs, i);
- CTRL_CLEAR_LO(low);
- CTRL_CLEAR_HI(high);
- CTRL_SET_ENABLE(low);
- CTRL_SET_USR(low, counter_config[i].user);
- CTRL_SET_KERN(low, counter_config[i].kernel);
- CTRL_SET_UM(low, counter_config[i].unit_mask);
- CTRL_SET_EVENT_LOW(low, counter_config[i].event);
- CTRL_SET_EVENT_HIGH(high, counter_config[i].event);
- CTRL_SET_HOST_ONLY(high, 0);
- CTRL_SET_GUEST_ONLY(high, 0);
-
- CTRL_WRITE(low, high, msrs, i);
- } else {
- reset_value[i] = 0;
- }
+ /* setup counter registers */
+ wrmsrl(msrs->counters[i].addr, -(u64)reset_value[virt]);
+
+ /* setup control registers */
+ rdmsrl(msrs->controls[i].addr, val);
+ val &= model->reserved;
+ val |= op_x86_get_ctrl(model, &counter_config[virt]);
+ wrmsrl(msrs->controls[i].addr, val);
}
}
#ifdef CONFIG_OPROFILE_IBS
-static inline int
+static inline void
op_amd_handle_ibs(struct pt_regs * const regs,
struct op_msrs const * const msrs)
{
- u32 low, high;
- u64 msr;
+ u64 val, ctl;
struct op_entry entry;
if (!has_ibs)
- return 1;
+ return;
if (ibs_config.fetch_enabled) {
- rdmsr(MSR_AMD64_IBSFETCHCTL, low, high);
- if (high & IBS_FETCH_HIGH_VALID_BIT) {
- rdmsrl(MSR_AMD64_IBSFETCHLINAD, msr);
- oprofile_write_reserve(&entry, regs, msr,
+ rdmsrl(MSR_AMD64_IBSFETCHCTL, ctl);
+ if (ctl & IBS_FETCH_VAL) {
+ rdmsrl(MSR_AMD64_IBSFETCHLINAD, val);
+ oprofile_write_reserve(&entry, regs, val,
IBS_FETCH_CODE, IBS_FETCH_SIZE);
- oprofile_add_data(&entry, (u32)msr);
- oprofile_add_data(&entry, (u32)(msr >> 32));
- oprofile_add_data(&entry, low);
- oprofile_add_data(&entry, high);
- rdmsrl(MSR_AMD64_IBSFETCHPHYSAD, msr);
- oprofile_add_data(&entry, (u32)msr);
- oprofile_add_data(&entry, (u32)(msr >> 32));
+ oprofile_add_data64(&entry, val);
+ oprofile_add_data64(&entry, ctl);
+ rdmsrl(MSR_AMD64_IBSFETCHPHYSAD, val);
+ oprofile_add_data64(&entry, val);
oprofile_write_commit(&entry);
/* reenable the IRQ */
- high &= ~IBS_FETCH_HIGH_VALID_BIT;
- high |= IBS_FETCH_HIGH_ENABLE;
- low &= IBS_FETCH_LOW_MAX_CNT_MASK;
- wrmsr(MSR_AMD64_IBSFETCHCTL, low, high);
+ ctl &= ~(IBS_FETCH_VAL | IBS_FETCH_CNT_MASK);
+ ctl |= IBS_FETCH_ENABLE;
+ wrmsrl(MSR_AMD64_IBSFETCHCTL, ctl);
}
}
if (ibs_config.op_enabled) {
- rdmsr(MSR_AMD64_IBSOPCTL, low, high);
- if (low & IBS_OP_LOW_VALID_BIT) {
- rdmsrl(MSR_AMD64_IBSOPRIP, msr);
- oprofile_write_reserve(&entry, regs, msr,
+ rdmsrl(MSR_AMD64_IBSOPCTL, ctl);
+ if (ctl & IBS_OP_VAL) {
+ rdmsrl(MSR_AMD64_IBSOPRIP, val);
+ oprofile_write_reserve(&entry, regs, val,
IBS_OP_CODE, IBS_OP_SIZE);
- oprofile_add_data(&entry, (u32)msr);
- oprofile_add_data(&entry, (u32)(msr >> 32));
- rdmsrl(MSR_AMD64_IBSOPDATA, msr);
- oprofile_add_data(&entry, (u32)msr);
- oprofile_add_data(&entry, (u32)(msr >> 32));
- rdmsrl(MSR_AMD64_IBSOPDATA2, msr);
- oprofile_add_data(&entry, (u32)msr);
- oprofile_add_data(&entry, (u32)(msr >> 32));
- rdmsrl(MSR_AMD64_IBSOPDATA3, msr);
- oprofile_add_data(&entry, (u32)msr);
- oprofile_add_data(&entry, (u32)(msr >> 32));
- rdmsrl(MSR_AMD64_IBSDCLINAD, msr);
- oprofile_add_data(&entry, (u32)msr);
- oprofile_add_data(&entry, (u32)(msr >> 32));
- rdmsrl(MSR_AMD64_IBSDCPHYSAD, msr);
- oprofile_add_data(&entry, (u32)msr);
- oprofile_add_data(&entry, (u32)(msr >> 32));
+ oprofile_add_data64(&entry, val);
+ rdmsrl(MSR_AMD64_IBSOPDATA, val);
+ oprofile_add_data64(&entry, val);
+ rdmsrl(MSR_AMD64_IBSOPDATA2, val);
+ oprofile_add_data64(&entry, val);
+ rdmsrl(MSR_AMD64_IBSOPDATA3, val);
+ oprofile_add_data64(&entry, val);
+ rdmsrl(MSR_AMD64_IBSDCLINAD, val);
+ oprofile_add_data64(&entry, val);
+ rdmsrl(MSR_AMD64_IBSDCPHYSAD, val);
+ oprofile_add_data64(&entry, val);
oprofile_write_commit(&entry);
/* reenable the IRQ */
- high = 0;
- low &= ~IBS_OP_LOW_VALID_BIT;
- low |= IBS_OP_LOW_ENABLE;
- wrmsr(MSR_AMD64_IBSOPCTL, low, high);
+ ctl &= ~IBS_OP_VAL & 0xFFFFFFFF;
+ ctl |= IBS_OP_ENABLE;
+ wrmsrl(MSR_AMD64_IBSOPCTL, ctl);
}
}
+}
- return 1;
+static inline void op_amd_start_ibs(void)
+{
+ u64 val;
+ if (has_ibs && ibs_config.fetch_enabled) {
+ val = (ibs_config.max_cnt_fetch >> 4) & 0xFFFF;
+ val |= ibs_config.rand_en ? IBS_FETCH_RAND_EN : 0;
+ val |= IBS_FETCH_ENABLE;
+ wrmsrl(MSR_AMD64_IBSFETCHCTL, val);
+ }
+
+ if (has_ibs && ibs_config.op_enabled) {
+ val = (ibs_config.max_cnt_op >> 4) & 0xFFFF;
+ val |= ibs_config.dispatched_ops ? IBS_OP_CNT_CTL : 0;
+ val |= IBS_OP_ENABLE;
+ wrmsrl(MSR_AMD64_IBSOPCTL, val);
+ }
+}
+
+static void op_amd_stop_ibs(void)
+{
+ if (has_ibs && ibs_config.fetch_enabled)
+ /* clear max count and enable */
+ wrmsrl(MSR_AMD64_IBSFETCHCTL, 0);
+
+ if (has_ibs && ibs_config.op_enabled)
+ /* clear max count and enable */
+ wrmsrl(MSR_AMD64_IBSOPCTL, 0);
}
+#else
+
+static inline void op_amd_handle_ibs(struct pt_regs * const regs,
+ struct op_msrs const * const msrs) { }
+static inline void op_amd_start_ibs(void) { }
+static inline void op_amd_stop_ibs(void) { }
+
#endif
static int op_amd_check_ctrs(struct pt_regs * const regs,
struct op_msrs const * const msrs)
{
- unsigned int low, high;
+ u64 val;
int i;
- for (i = 0 ; i < NUM_COUNTERS; ++i) {
- if (!reset_value[i])
+ for (i = 0; i < NUM_COUNTERS; ++i) {
+ int virt = op_x86_phys_to_virt(i);
+ if (!reset_value[virt])
continue;
- CTR_READ(low, high, msrs, i);
- if (CTR_OVERFLOWED(low)) {
- oprofile_add_sample(regs, i);
- CTR_WRITE(reset_value[i], msrs, i);
- }
+ rdmsrl(msrs->counters[i].addr, val);
+ /* bit is clear if overflowed: */
+ if (val & OP_CTR_OVERFLOW)
+ continue;
+ oprofile_add_sample(regs, virt);
+ wrmsrl(msrs->counters[i].addr, -(u64)reset_value[virt]);
}
-#ifdef CONFIG_OPROFILE_IBS
op_amd_handle_ibs(regs, msrs);
-#endif
/* See op_model_ppro.c */
return 1;
@@ -248,79 +307,50 @@ static int op_amd_check_ctrs(struct pt_regs * const regs,
static void op_amd_start(struct op_msrs const * const msrs)
{
- unsigned int low, high;
+ u64 val;
int i;
- for (i = 0 ; i < NUM_COUNTERS ; ++i) {
- if (reset_value[i]) {
- CTRL_READ(low, high, msrs, i);
- CTRL_SET_ACTIVE(low);
- CTRL_WRITE(low, high, msrs, i);
- }
- }
-#ifdef CONFIG_OPROFILE_IBS
- if (has_ibs && ibs_config.fetch_enabled) {
- low = (ibs_config.max_cnt_fetch >> 4) & 0xFFFF;
- high = ((ibs_config.rand_en & 0x1) << 25) /* bit 57 */
- + IBS_FETCH_HIGH_ENABLE;
- wrmsr(MSR_AMD64_IBSFETCHCTL, low, high);
+ for (i = 0; i < NUM_COUNTERS; ++i) {
+ if (!reset_value[op_x86_phys_to_virt(i)])
+ continue;
+ rdmsrl(msrs->controls[i].addr, val);
+ val |= ARCH_PERFMON_EVENTSEL0_ENABLE;
+ wrmsrl(msrs->controls[i].addr, val);
}
- if (has_ibs && ibs_config.op_enabled) {
- low = ((ibs_config.max_cnt_op >> 4) & 0xFFFF)
- + ((ibs_config.dispatched_ops & 0x1) << 19) /* bit 19 */
- + IBS_OP_LOW_ENABLE;
- high = 0;
- wrmsr(MSR_AMD64_IBSOPCTL, low, high);
- }
-#endif
+ op_amd_start_ibs();
}
-
static void op_amd_stop(struct op_msrs const * const msrs)
{
- unsigned int low, high;
+ u64 val;
int i;
/*
* Subtle: stop on all counters to avoid race with setting our
* pm callback
*/
- for (i = 0 ; i < NUM_COUNTERS ; ++i) {
- if (!reset_value[i])
+ for (i = 0; i < NUM_COUNTERS; ++i) {
+ if (!reset_value[op_x86_phys_to_virt(i)])
continue;
- CTRL_READ(low, high, msrs, i);
- CTRL_SET_INACTIVE(low);
- CTRL_WRITE(low, high, msrs, i);
- }
-
-#ifdef CONFIG_OPROFILE_IBS
- if (has_ibs && ibs_config.fetch_enabled) {
- /* clear max count and enable */
- low = 0;
- high = 0;
- wrmsr(MSR_AMD64_IBSFETCHCTL, low, high);
+ rdmsrl(msrs->controls[i].addr, val);
+ val &= ~ARCH_PERFMON_EVENTSEL0_ENABLE;
+ wrmsrl(msrs->controls[i].addr, val);
}
- if (has_ibs && ibs_config.op_enabled) {
- /* clear max count and enable */
- low = 0;
- high = 0;
- wrmsr(MSR_AMD64_IBSOPCTL, low, high);
- }
-#endif
+ op_amd_stop_ibs();
}
static void op_amd_shutdown(struct op_msrs const * const msrs)
{
int i;
- for (i = 0 ; i < NUM_COUNTERS ; ++i) {
- if (CTR_IS_RESERVED(msrs, i))
+ for (i = 0; i < NUM_COUNTERS; ++i) {
+ if (msrs->counters[i].addr)
release_perfctr_nmi(MSR_K7_PERFCTR0 + i);
}
- for (i = 0 ; i < NUM_CONTROLS ; ++i) {
- if (CTRL_IS_RESERVED(msrs, i))
+ for (i = 0; i < NUM_CONTROLS; ++i) {
+ if (msrs->controls[i].addr)
release_evntsel_nmi(MSR_K7_EVNTSEL0 + i);
}
}
@@ -490,15 +520,21 @@ static void op_amd_exit(void) {}
#endif /* CONFIG_OPROFILE_IBS */
-struct op_x86_model_spec const op_amd_spec = {
- .init = op_amd_init,
- .exit = op_amd_exit,
+struct op_x86_model_spec op_amd_spec = {
.num_counters = NUM_COUNTERS,
.num_controls = NUM_CONTROLS,
+ .num_virt_counters = NUM_VIRT_COUNTERS,
+ .reserved = MSR_AMD_EVENTSEL_RESERVED,
+ .event_mask = OP_EVENT_MASK,
+ .init = op_amd_init,
+ .exit = op_amd_exit,
.fill_in_addresses = &op_amd_fill_in_addresses,
.setup_ctrs = &op_amd_setup_ctrs,
.check_ctrs = &op_amd_check_ctrs,
.start = &op_amd_start,
.stop = &op_amd_stop,
- .shutdown = &op_amd_shutdown
+ .shutdown = &op_amd_shutdown,
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+ .switch_ctrl = &op_mux_switch_ctrl,
+#endif
};
diff --git a/arch/x86/oprofile/op_model_p4.c b/arch/x86/oprofile/op_model_p4.c
index 819b131fd752..ac6b354becdf 100644
--- a/arch/x86/oprofile/op_model_p4.c
+++ b/arch/x86/oprofile/op_model_p4.c
@@ -32,6 +32,8 @@
#define NUM_CCCRS_HT2 9
#define NUM_CONTROLS_HT2 (NUM_ESCRS_HT2 + NUM_CCCRS_HT2)
+#define OP_CTR_OVERFLOW (1ULL<<31)
+
static unsigned int num_counters = NUM_COUNTERS_NON_HT;
static unsigned int num_controls = NUM_CONTROLS_NON_HT;
@@ -350,8 +352,6 @@ static struct p4_event_binding p4_events[NUM_EVENTS] = {
#define ESCR_SET_OS_1(escr, os) ((escr) |= (((os) & 1) << 1))
#define ESCR_SET_EVENT_SELECT(escr, sel) ((escr) |= (((sel) & 0x3f) << 25))
#define ESCR_SET_EVENT_MASK(escr, mask) ((escr) |= (((mask) & 0xffff) << 9))
-#define ESCR_READ(escr, high, ev, i) do {rdmsr(ev->bindings[(i)].escr_address, (escr), (high)); } while (0)
-#define ESCR_WRITE(escr, high, ev, i) do {wrmsr(ev->bindings[(i)].escr_address, (escr), (high)); } while (0)
#define CCCR_RESERVED_BITS 0x38030FFF
#define CCCR_CLEAR(cccr) ((cccr) &= CCCR_RESERVED_BITS)
@@ -361,17 +361,9 @@ static struct p4_event_binding p4_events[NUM_EVENTS] = {
#define CCCR_SET_PMI_OVF_1(cccr) ((cccr) |= (1<<27))
#define CCCR_SET_ENABLE(cccr) ((cccr) |= (1<<12))
#define CCCR_SET_DISABLE(cccr) ((cccr) &= ~(1<<12))
-#define CCCR_READ(low, high, i) do {rdmsr(p4_counters[(i)].cccr_address, (low), (high)); } while (0)
-#define CCCR_WRITE(low, high, i) do {wrmsr(p4_counters[(i)].cccr_address, (low), (high)); } while (0)
#define CCCR_OVF_P(cccr) ((cccr) & (1U<<31))
#define CCCR_CLEAR_OVF(cccr) ((cccr) &= (~(1U<<31)))
-#define CTRL_IS_RESERVED(msrs, c) (msrs->controls[(c)].addr ? 1 : 0)
-#define CTR_IS_RESERVED(msrs, c) (msrs->counters[(c)].addr ? 1 : 0)
-#define CTR_READ(l, h, i) do {rdmsr(p4_counters[(i)].counter_address, (l), (h)); } while (0)
-#define CTR_WRITE(l, i) do {wrmsr(p4_counters[(i)].counter_address, -(u32)(l), -1); } while (0)
-#define CTR_OVERFLOW_P(ctr) (!((ctr) & 0x80000000))
-
/* this assigns a "stagger" to the current CPU, which is used throughout
the code in this module as an extra array offset, to select the "even"
@@ -515,7 +507,7 @@ static void pmc_setup_one_p4_counter(unsigned int ctr)
if (ev->bindings[i].virt_counter & counter_bit) {
/* modify ESCR */
- ESCR_READ(escr, high, ev, i);
+ rdmsr(ev->bindings[i].escr_address, escr, high);
ESCR_CLEAR(escr);
if (stag == 0) {
ESCR_SET_USR_0(escr, counter_config[ctr].user);
@@ -526,10 +518,11 @@ static void pmc_setup_one_p4_counter(unsigned int ctr)
}
ESCR_SET_EVENT_SELECT(escr, ev->event_select);
ESCR_SET_EVENT_MASK(escr, counter_config[ctr].unit_mask);
- ESCR_WRITE(escr, high, ev, i);
+ wrmsr(ev->bindings[i].escr_address, escr, high);
/* modify CCCR */
- CCCR_READ(cccr, high, VIRT_CTR(stag, ctr));
+ rdmsr(p4_counters[VIRT_CTR(stag, ctr)].cccr_address,
+ cccr, high);
CCCR_CLEAR(cccr);
CCCR_SET_REQUIRED_BITS(cccr);
CCCR_SET_ESCR_SELECT(cccr, ev->escr_select);
@@ -537,7 +530,8 @@ static void pmc_setup_one_p4_counter(unsigned int ctr)
CCCR_SET_PMI_OVF_0(cccr);
else
CCCR_SET_PMI_OVF_1(cccr);
- CCCR_WRITE(cccr, high, VIRT_CTR(stag, ctr));
+ wrmsr(p4_counters[VIRT_CTR(stag, ctr)].cccr_address,
+ cccr, high);
return;
}
}
@@ -548,7 +542,8 @@ static void pmc_setup_one_p4_counter(unsigned int ctr)
}
-static void p4_setup_ctrs(struct op_msrs const * const msrs)
+static void p4_setup_ctrs(struct op_x86_model_spec const *model,
+ struct op_msrs const * const msrs)
{
unsigned int i;
unsigned int low, high;
@@ -563,8 +558,8 @@ static void p4_setup_ctrs(struct op_msrs const * const msrs)
}
/* clear the cccrs we will use */
- for (i = 0 ; i < num_counters ; i++) {
- if (unlikely(!CTRL_IS_RESERVED(msrs, i)))
+ for (i = 0; i < num_counters; i++) {
+ if (unlikely(!msrs->controls[i].addr))
continue;
rdmsr(p4_counters[VIRT_CTR(stag, i)].cccr_address, low, high);
CCCR_CLEAR(low);
@@ -574,17 +569,18 @@ static void p4_setup_ctrs(struct op_msrs const * const msrs)
/* clear all escrs (including those outside our concern) */
for (i = num_counters; i < num_controls; i++) {
- if (unlikely(!CTRL_IS_RESERVED(msrs, i)))
+ if (unlikely(!msrs->controls[i].addr))
continue;
wrmsr(msrs->controls[i].addr, 0, 0);
}
/* setup all counters */
- for (i = 0 ; i < num_counters ; ++i) {
- if ((counter_config[i].enabled) && (CTRL_IS_RESERVED(msrs, i))) {
+ for (i = 0; i < num_counters; ++i) {
+ if (counter_config[i].enabled && msrs->controls[i].addr) {
reset_value[i] = counter_config[i].count;
pmc_setup_one_p4_counter(i);
- CTR_WRITE(counter_config[i].count, VIRT_CTR(stag, i));
+ wrmsrl(p4_counters[VIRT_CTR(stag, i)].counter_address,
+ -(u64)counter_config[i].count);
} else {
reset_value[i] = 0;
}
@@ -624,14 +620,16 @@ static int p4_check_ctrs(struct pt_regs * const regs,
real = VIRT_CTR(stag, i);
- CCCR_READ(low, high, real);
- CTR_READ(ctr, high, real);
- if (CCCR_OVF_P(low) || CTR_OVERFLOW_P(ctr)) {
+ rdmsr(p4_counters[real].cccr_address, low, high);
+ rdmsr(p4_counters[real].counter_address, ctr, high);
+ if (CCCR_OVF_P(low) || !(ctr & OP_CTR_OVERFLOW)) {
oprofile_add_sample(regs, i);
- CTR_WRITE(reset_value[i], real);
+ wrmsrl(p4_counters[real].counter_address,
+ -(u64)reset_value[i]);
CCCR_CLEAR_OVF(low);
- CCCR_WRITE(low, high, real);
- CTR_WRITE(reset_value[i], real);
+ wrmsr(p4_counters[real].cccr_address, low, high);
+ wrmsrl(p4_counters[real].counter_address,
+ -(u64)reset_value[i]);
}
}
@@ -653,9 +651,9 @@ static void p4_start(struct op_msrs const * const msrs)
for (i = 0; i < num_counters; ++i) {
if (!reset_value[i])
continue;
- CCCR_READ(low, high, VIRT_CTR(stag, i));
+ rdmsr(p4_counters[VIRT_CTR(stag, i)].cccr_address, low, high);
CCCR_SET_ENABLE(low);
- CCCR_WRITE(low, high, VIRT_CTR(stag, i));
+ wrmsr(p4_counters[VIRT_CTR(stag, i)].cccr_address, low, high);
}
}
@@ -670,9 +668,9 @@ static void p4_stop(struct op_msrs const * const msrs)
for (i = 0; i < num_counters; ++i) {
if (!reset_value[i])
continue;
- CCCR_READ(low, high, VIRT_CTR(stag, i));
+ rdmsr(p4_counters[VIRT_CTR(stag, i)].cccr_address, low, high);
CCCR_SET_DISABLE(low);
- CCCR_WRITE(low, high, VIRT_CTR(stag, i));
+ wrmsr(p4_counters[VIRT_CTR(stag, i)].cccr_address, low, high);
}
}
@@ -680,8 +678,8 @@ static void p4_shutdown(struct op_msrs const * const msrs)
{
int i;
- for (i = 0 ; i < num_counters ; ++i) {
- if (CTR_IS_RESERVED(msrs, i))
+ for (i = 0; i < num_counters; ++i) {
+ if (msrs->counters[i].addr)
release_perfctr_nmi(msrs->counters[i].addr);
}
/*
@@ -689,15 +687,15 @@ static void p4_shutdown(struct op_msrs const * const msrs)
* conjunction with the counter registers (hence the starting offset).
* This saves a few bits.
*/
- for (i = num_counters ; i < num_controls ; ++i) {
- if (CTRL_IS_RESERVED(msrs, i))
+ for (i = num_counters; i < num_controls; ++i) {
+ if (msrs->controls[i].addr)
release_evntsel_nmi(msrs->controls[i].addr);
}
}
#ifdef CONFIG_SMP
-struct op_x86_model_spec const op_p4_ht2_spec = {
+struct op_x86_model_spec op_p4_ht2_spec = {
.num_counters = NUM_COUNTERS_HT2,
.num_controls = NUM_CONTROLS_HT2,
.fill_in_addresses = &p4_fill_in_addresses,
@@ -709,7 +707,7 @@ struct op_x86_model_spec const op_p4_ht2_spec = {
};
#endif
-struct op_x86_model_spec const op_p4_spec = {
+struct op_x86_model_spec op_p4_spec = {
.num_counters = NUM_COUNTERS_NON_HT,
.num_controls = NUM_CONTROLS_NON_HT,
.fill_in_addresses = &p4_fill_in_addresses,
diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c
index 4da7230b3d17..4899215999de 100644
--- a/arch/x86/oprofile/op_model_ppro.c
+++ b/arch/x86/oprofile/op_model_ppro.c
@@ -10,6 +10,7 @@
* @author Philippe Elie
* @author Graydon Hoare
* @author Andi Kleen
+ * @author Robert Richter <robert.richter@amd.com>
*/
#include <linux/oprofile.h>
@@ -18,7 +19,6 @@
#include <asm/msr.h>
#include <asm/apic.h>
#include <asm/nmi.h>
-#include <asm/perf_counter.h>
#include "op_x86_model.h"
#include "op_counter.h"
@@ -26,20 +26,7 @@
static int num_counters = 2;
static int counter_width = 32;
-#define CTR_IS_RESERVED(msrs, c) (msrs->counters[(c)].addr ? 1 : 0)
-#define CTR_OVERFLOWED(n) (!((n) & (1ULL<<(counter_width-1))))
-
-#define CTRL_IS_RESERVED(msrs, c) (msrs->controls[(c)].addr ? 1 : 0)
-#define CTRL_READ(l, h, msrs, c) do {rdmsr((msrs->controls[(c)].addr), (l), (h)); } while (0)
-#define CTRL_WRITE(l, h, msrs, c) do {wrmsr((msrs->controls[(c)].addr), (l), (h)); } while (0)
-#define CTRL_SET_ACTIVE(n) (n |= (1<<22))
-#define CTRL_SET_INACTIVE(n) (n &= ~(1<<22))
-#define CTRL_CLEAR(x) (x &= (1<<21))
-#define CTRL_SET_ENABLE(val) (val |= 1<<20)
-#define CTRL_SET_USR(val, u) (val |= ((u & 1) << 16))
-#define CTRL_SET_KERN(val, k) (val |= ((k & 1) << 17))
-#define CTRL_SET_UM(val, m) (val |= (m << 8))
-#define CTRL_SET_EVENT(val, e) (val |= e)
+#define MSR_PPRO_EVENTSEL_RESERVED ((0xFFFFFFFFULL<<32)|(1ULL<<21))
static u64 *reset_value;
@@ -63,9 +50,10 @@ static void ppro_fill_in_addresses(struct op_msrs * const msrs)
}
-static void ppro_setup_ctrs(struct op_msrs const * const msrs)
+static void ppro_setup_ctrs(struct op_x86_model_spec const *model,
+ struct op_msrs const * const msrs)
{
- unsigned int low, high;
+ u64 val;
int i;
if (!reset_value) {
@@ -93,36 +81,30 @@ static void ppro_setup_ctrs(struct op_msrs const * const msrs)
}
/* clear all counters */
- for (i = 0 ; i < num_counters; ++i) {
- if (unlikely(!CTRL_IS_RESERVED(msrs, i)))
+ for (i = 0; i < num_counters; ++i) {
+ if (unlikely(!msrs->controls[i].addr))
continue;
- CTRL_READ(low, high, msrs, i);
- CTRL_CLEAR(low);
- CTRL_WRITE(low, high, msrs, i);
+ rdmsrl(msrs->controls[i].addr, val);
+ val &= model->reserved;
+ wrmsrl(msrs->controls[i].addr, val);
}
/* avoid a false detection of ctr overflows in NMI handler */
for (i = 0; i < num_counters; ++i) {
- if (unlikely(!CTR_IS_RESERVED(msrs, i)))
+ if (unlikely(!msrs->counters[i].addr))
continue;
wrmsrl(msrs->counters[i].addr, -1LL);
}
/* enable active counters */
for (i = 0; i < num_counters; ++i) {
- if ((counter_config[i].enabled) && (CTR_IS_RESERVED(msrs, i))) {
+ if (counter_config[i].enabled && msrs->counters[i].addr) {
reset_value[i] = counter_config[i].count;
-
wrmsrl(msrs->counters[i].addr, -reset_value[i]);
-
- CTRL_READ(low, high, msrs, i);
- CTRL_CLEAR(low);
- CTRL_SET_ENABLE(low);
- CTRL_SET_USR(low, counter_config[i].user);
- CTRL_SET_KERN(low, counter_config[i].kernel);
- CTRL_SET_UM(low, counter_config[i].unit_mask);
- CTRL_SET_EVENT(low, counter_config[i].event);
- CTRL_WRITE(low, high, msrs, i);
+ rdmsrl(msrs->controls[i].addr, val);
+ val &= model->reserved;
+ val |= op_x86_get_ctrl(model, &counter_config[i]);
+ wrmsrl(msrs->controls[i].addr, val);
} else {
reset_value[i] = 0;
}
@@ -143,14 +125,14 @@ static int ppro_check_ctrs(struct pt_regs * const regs,
if (unlikely(!reset_value))
goto out;
- for (i = 0 ; i < num_counters; ++i) {
+ for (i = 0; i < num_counters; ++i) {
if (!reset_value[i])
continue;
rdmsrl(msrs->counters[i].addr, val);
- if (CTR_OVERFLOWED(val)) {
- oprofile_add_sample(regs, i);
- wrmsrl(msrs->counters[i].addr, -reset_value[i]);
- }
+ if (val & (1ULL << (counter_width - 1)))
+ continue;
+ oprofile_add_sample(regs, i);
+ wrmsrl(msrs->counters[i].addr, -reset_value[i]);
}
out:
@@ -171,16 +153,16 @@ out:
static void ppro_start(struct op_msrs const * const msrs)
{
- unsigned int low, high;
+ u64 val;
int i;
if (!reset_value)
return;
for (i = 0; i < num_counters; ++i) {
if (reset_value[i]) {
- CTRL_READ(low, high, msrs, i);
- CTRL_SET_ACTIVE(low);
- CTRL_WRITE(low, high, msrs, i);
+ rdmsrl(msrs->controls[i].addr, val);
+ val |= ARCH_PERFMON_EVENTSEL0_ENABLE;
+ wrmsrl(msrs->controls[i].addr, val);
}
}
}
@@ -188,7 +170,7 @@ static void ppro_start(struct op_msrs const * const msrs)
static void ppro_stop(struct op_msrs const * const msrs)
{
- unsigned int low, high;
+ u64 val;
int i;
if (!reset_value)
@@ -196,9 +178,9 @@ static void ppro_stop(struct op_msrs const * const msrs)
for (i = 0; i < num_counters; ++i) {
if (!reset_value[i])
continue;
- CTRL_READ(low, high, msrs, i);
- CTRL_SET_INACTIVE(low);
- CTRL_WRITE(low, high, msrs, i);
+ rdmsrl(msrs->controls[i].addr, val);
+ val &= ~ARCH_PERFMON_EVENTSEL0_ENABLE;
+ wrmsrl(msrs->controls[i].addr, val);
}
}
@@ -206,12 +188,12 @@ static void ppro_shutdown(struct op_msrs const * const msrs)
{
int i;
- for (i = 0 ; i < num_counters ; ++i) {
- if (CTR_IS_RESERVED(msrs, i))
+ for (i = 0; i < num_counters; ++i) {
+ if (msrs->counters[i].addr)
release_perfctr_nmi(MSR_P6_PERFCTR0 + i);
}
- for (i = 0 ; i < num_counters ; ++i) {
- if (CTRL_IS_RESERVED(msrs, i))
+ for (i = 0; i < num_counters; ++i) {
+ if (msrs->controls[i].addr)
release_evntsel_nmi(MSR_P6_EVNTSEL0 + i);
}
if (reset_value) {
@@ -222,8 +204,9 @@ static void ppro_shutdown(struct op_msrs const * const msrs)
struct op_x86_model_spec op_ppro_spec = {
- .num_counters = 2, /* can be overriden */
- .num_controls = 2, /* dito */
+ .num_counters = 2,
+ .num_controls = 2,
+ .reserved = MSR_PPRO_EVENTSEL_RESERVED,
.fill_in_addresses = &ppro_fill_in_addresses,
.setup_ctrs = &ppro_setup_ctrs,
.check_ctrs = &ppro_check_ctrs,
@@ -241,7 +224,7 @@ struct op_x86_model_spec op_ppro_spec = {
* the specific CPU.
*/
-void arch_perfmon_setup_counters(void)
+static void arch_perfmon_setup_counters(void)
{
union cpuid10_eax eax;
@@ -259,11 +242,17 @@ void arch_perfmon_setup_counters(void)
op_arch_perfmon_spec.num_counters = num_counters;
op_arch_perfmon_spec.num_controls = num_counters;
- op_ppro_spec.num_counters = num_counters;
- op_ppro_spec.num_controls = num_counters;
+}
+
+static int arch_perfmon_init(struct oprofile_operations *ignore)
+{
+ arch_perfmon_setup_counters();
+ return 0;
}
struct op_x86_model_spec op_arch_perfmon_spec = {
+ .reserved = MSR_PPRO_EVENTSEL_RESERVED,
+ .init = &arch_perfmon_init,
/* num_counters/num_controls filled in at runtime */
.fill_in_addresses = &ppro_fill_in_addresses,
/* user space does the cpuid check for available events */
diff --git a/arch/x86/oprofile/op_x86_model.h b/arch/x86/oprofile/op_x86_model.h
index 825e79064d64..b83776180c7f 100644
--- a/arch/x86/oprofile/op_x86_model.h
+++ b/arch/x86/oprofile/op_x86_model.h
@@ -6,51 +6,66 @@
* @remark Read the file COPYING
*
* @author Graydon Hoare
+ * @author Robert Richter <robert.richter@amd.com>
*/
#ifndef OP_X86_MODEL_H
#define OP_X86_MODEL_H
-struct op_saved_msr {
- unsigned int high;
- unsigned int low;
-};
+#include <asm/types.h>
+#include <asm/perf_counter.h>
struct op_msr {
- unsigned long addr;
- struct op_saved_msr saved;
+ unsigned long addr;
+ u64 saved;
};
struct op_msrs {
struct op_msr *counters;
struct op_msr *controls;
+ struct op_msr *multiplex;
};
struct pt_regs;
+struct oprofile_operations;
+
/* The model vtable abstracts the differences between
* various x86 CPU models' perfctr support.
*/
struct op_x86_model_spec {
- int (*init)(struct oprofile_operations *ops);
- void (*exit)(void);
- unsigned int num_counters;
- unsigned int num_controls;
- void (*fill_in_addresses)(struct op_msrs * const msrs);
- void (*setup_ctrs)(struct op_msrs const * const msrs);
- int (*check_ctrs)(struct pt_regs * const regs,
- struct op_msrs const * const msrs);
- void (*start)(struct op_msrs const * const msrs);
- void (*stop)(struct op_msrs const * const msrs);
- void (*shutdown)(struct op_msrs const * const msrs);
+ unsigned int num_counters;
+ unsigned int num_controls;
+ unsigned int num_virt_counters;
+ u64 reserved;
+ u16 event_mask;
+ int (*init)(struct oprofile_operations *ops);
+ void (*exit)(void);
+ void (*fill_in_addresses)(struct op_msrs * const msrs);
+ void (*setup_ctrs)(struct op_x86_model_spec const *model,
+ struct op_msrs const * const msrs);
+ int (*check_ctrs)(struct pt_regs * const regs,
+ struct op_msrs const * const msrs);
+ void (*start)(struct op_msrs const * const msrs);
+ void (*stop)(struct op_msrs const * const msrs);
+ void (*shutdown)(struct op_msrs const * const msrs);
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+ void (*switch_ctrl)(struct op_x86_model_spec const *model,
+ struct op_msrs const * const msrs);
+#endif
};
+struct op_counter_config;
+
+extern u64 op_x86_get_ctrl(struct op_x86_model_spec const *model,
+ struct op_counter_config *counter_config);
+extern int op_x86_phys_to_virt(int phys);
+extern int op_x86_virt_to_phys(int virt);
+
extern struct op_x86_model_spec op_ppro_spec;
-extern struct op_x86_model_spec const op_p4_spec;
-extern struct op_x86_model_spec const op_p4_ht2_spec;
-extern struct op_x86_model_spec const op_amd_spec;
+extern struct op_x86_model_spec op_p4_spec;
+extern struct op_x86_model_spec op_p4_ht2_spec;
+extern struct op_x86_model_spec op_amd_spec;
extern struct op_x86_model_spec op_arch_perfmon_spec;
-extern void arch_perfmon_setup_counters(void);
-
#endif /* OP_X86_MODEL_H */