[V3][PATCH 6/7] perf, x86: Implement 64 bit counter support for IBS

From: Robert Richter
Date: Wed Sep 21 2011 - 06:59:10 EST


This patch implements 64 bit counter support for IBS. The sampling
period is no longer limited to the hw counter width.

The functions perf_event_set_period() and perf_event_try_update() can
be used as generic functions. They can replace similar code that is
duplicate across architectures.

V2: Added caps check for IBS_OP_CUR_CNT emulation.

Signed-off-by: Robert Richter <robert.richter@xxxxxxx>
---
arch/x86/include/asm/perf_event.h | 2 +
arch/x86/kernel/cpu/perf_event_amd_ibs.c | 204 +++++++++++++++++++++++++++---
2 files changed, 185 insertions(+), 21 deletions(-)

diff --git a/arch/x86/include/asm/perf_event.h b/arch/x86/include/asm/perf_event.h
index e7d2f15..cc44a1a 100644
--- a/arch/x86/include/asm/perf_event.h
+++ b/arch/x86/include/asm/perf_event.h
@@ -150,6 +150,8 @@ union cpuid10_edx {
#define IBS_FETCH_MAX_CNT 0x0000FFFFULL

/* IbsOpCtl bits */
+/* lower 4 bits of the current count are ignored: */
+#define IBS_OP_CUR_CNT (0xFFFF0ULL<<32)
#define IBS_OP_CNT_CTL (1ULL<<19)
#define IBS_OP_VAL (1ULL<<18)
#define IBS_OP_ENABLE (1ULL<<17)
diff --git a/arch/x86/kernel/cpu/perf_event_amd_ibs.c b/arch/x86/kernel/cpu/perf_event_amd_ibs.c
index 9713d31..6deb4d9 100644
--- a/arch/x86/kernel/cpu/perf_event_amd_ibs.c
+++ b/arch/x86/kernel/cpu/perf_event_amd_ibs.c
@@ -42,9 +42,11 @@ struct perf_ibs {
u64 cnt_mask;
u64 enable_mask;
u64 valid_mask;
+ u64 max_period;
unsigned long offset_mask[1];
int offset_max;
struct cpu_perf_ibs __percpu *pcpu;
+ u64 (*get_count)(u64 config);
};

struct perf_ibs_data {
@@ -56,6 +58,78 @@ struct perf_ibs_data {
u64 regs[MSR_AMD64_IBS_REG_COUNT_MAX];
};

+static int
+perf_event_set_period(struct hw_perf_event *hwc, u64 min, u64 max, u64 *count)
+{
+ s64 left = local64_read(&hwc->period_left);
+ s64 period = hwc->sample_period;
+ int overflow = 0;
+
+ /*
+ * If we are way outside a reasonable range then just skip forward:
+ */
+ if (unlikely(left <= -period)) {
+ left = period;
+ local64_set(&hwc->period_left, left);
+ hwc->last_period = period;
+ overflow = 1;
+ }
+
+ if (unlikely(left <= 0)) {
+ left += period;
+ local64_set(&hwc->period_left, left);
+ hwc->last_period = period;
+ overflow = 1;
+ }
+
+ if (unlikely(left < min))
+ left = min;
+
+ if (left > max)
+ left = max;
+
+ *count = (u64)left;
+
+ return overflow;
+}
+
+static int
+perf_event_try_update(struct perf_event *event, u64 new_raw_count, int width)
+{
+ struct hw_perf_event *hwc = &event->hw;
+ int shift = 64 - width;
+ u64 prev_raw_count;
+ u64 delta;
+
+ /*
+ * Careful: an NMI might modify the previous event value.
+ *
+ * Our tactic to handle this is to first atomically read and
+ * exchange a new raw count - then add that new-prev delta
+ * count to the generic event atomically:
+ */
+ prev_raw_count = local64_read(&hwc->prev_count);
+ if (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
+ new_raw_count) != prev_raw_count)
+ return 0;
+
+ /*
+ * Now we have the new raw value and have updated the prev
+ * timestamp already. We can now calculate the elapsed delta
+ * (event-)time and add that to the generic event.
+ *
+ * Careful, not all hw sign-extends above the physical width
+ * of the count.
+ */
+ delta = (new_raw_count << shift) - (prev_raw_count << shift);
+ delta >>= shift;
+
+ local64_add(delta, &event->count);
+ local64_sub(delta, &hwc->period_left);
+
+ return 1;
+}
+
static struct perf_ibs perf_ibs_fetch;
static struct perf_ibs perf_ibs_op;

@@ -89,18 +163,14 @@ static int perf_ibs_init(struct perf_event *event)
if (hwc->sample_period & 0x0f)
/* lower 4 bits can not be set in ibs max cnt */
return -EINVAL;
- max_cnt = hwc->sample_period >> 4;
- if (max_cnt & ~perf_ibs->cnt_mask)
- /* out of range */
- return -EINVAL;
- config |= max_cnt;
} else {
max_cnt = config & perf_ibs->cnt_mask;
+ config &= ~perf_ibs->cnt_mask;
event->attr.sample_period = max_cnt << 4;
hwc->sample_period = event->attr.sample_period;
}

- if (!max_cnt)
+ if (!hwc->sample_period)
return -EINVAL;

hwc->config_base = perf_ibs->msr;
@@ -109,16 +179,71 @@ static int perf_ibs_init(struct perf_event *event)
return 0;
}

+static int perf_ibs_set_period(struct perf_ibs *perf_ibs,
+ struct hw_perf_event *hwc, u64 *period)
+{
+ int ret;
+
+ /* ignore lower 4 bits in min count: */
+ ret = perf_event_set_period(hwc, 1<<4, perf_ibs->max_period, period);
+ local64_set(&hwc->prev_count, 0);
+
+ return ret;
+}
+
+static u64 get_ibs_fetch_count(u64 config)
+{
+ return (config & IBS_FETCH_CNT) >> 12;
+}
+
+static u64 get_ibs_op_count(u64 config)
+{
+ return (config & IBS_OP_CUR_CNT) >> 32;
+}
+
+static void
+perf_ibs_event_update(struct perf_ibs *perf_ibs, struct perf_event *event,
+ u64 config)
+{
+ u64 count = perf_ibs->get_count(config);
+
+ while (!perf_event_try_update(event, count, 20)) {
+ rdmsrl(event->hw.config_base, config);
+ count = perf_ibs->get_count(config);
+ }
+}
+
+/* Note: The enable mask must be encoded in the config argument. */
+static inline void perf_ibs_enable_event(struct hw_perf_event *hwc, u64 config)
+{
+ wrmsrl(hwc->config_base, hwc->config | config);
+}
+
+/*
+ * We cannot restore the ibs pmu state, so we always needs to update
+ * the event while stopping it and then reset the state when starting
+ * again. Thus, ignoring PERF_EF_RELOAD and PERF_EF_UPDATE flags in
+ * perf_ibs_start()/perf_ibs_stop() and instead always do it.
+ */
static void perf_ibs_start(struct perf_event *event, int flags)
{
struct hw_perf_event *hwc = &event->hw;
struct perf_ibs *perf_ibs = container_of(event->pmu, struct perf_ibs, pmu);
struct cpu_perf_ibs *pcpu = this_cpu_ptr(perf_ibs->pcpu);
+ u64 config;

- if (test_and_set_bit(IBS_STARTED, pcpu->state))
+ if (WARN_ON_ONCE(!(hwc->state & PERF_HES_STOPPED)))
return;

- wrmsrl(hwc->config_base, hwc->config | perf_ibs->enable_mask);
+ WARN_ON_ONCE(!(hwc->state & PERF_HES_UPTODATE));
+ hwc->state = 0;
+
+ perf_ibs_set_period(perf_ibs, hwc, &config);
+ config = (config >> 4) | perf_ibs->enable_mask;
+ set_bit(IBS_STARTED, pcpu->state);
+ perf_ibs_enable_event(hwc, config);
+
+ perf_event_update_userpage(event);
}

static void perf_ibs_stop(struct perf_event *event, int flags)
@@ -127,15 +252,28 @@ static void perf_ibs_stop(struct perf_event *event, int flags)
struct perf_ibs *perf_ibs = container_of(event->pmu, struct perf_ibs, pmu);
struct cpu_perf_ibs *pcpu = this_cpu_ptr(perf_ibs->pcpu);
u64 val;
+ int stopping;

- if (!test_and_clear_bit(IBS_STARTED, pcpu->state))
- return;
+ stopping = test_and_clear_bit(IBS_STARTED, pcpu->state);

- set_bit(IBS_STOPPING, pcpu->state);
+ if (!stopping && (hwc->state & PERF_HES_UPTODATE))
+ return;

rdmsrl(hwc->config_base, val);
- val &= ~perf_ibs->enable_mask;
- wrmsrl(hwc->config_base, val);
+
+ if (stopping) {
+ set_bit(IBS_STOPPING, pcpu->state);
+ val &= ~perf_ibs->enable_mask;
+ wrmsrl(hwc->config_base, val);
+ WARN_ON_ONCE(hwc->state & PERF_HES_STOPPED);
+ hwc->state |= PERF_HES_STOPPED;
+ }
+
+ if (hwc->state & PERF_HES_UPTODATE)
+ return;
+
+ perf_ibs_event_update(perf_ibs, event, val);
+ hwc->state |= PERF_HES_UPTODATE;
}

static int perf_ibs_add(struct perf_event *event, int flags)
@@ -146,6 +284,8 @@ static int perf_ibs_add(struct perf_event *event, int flags)
if (test_and_set_bit(IBS_ENABLED, pcpu->state))
return -ENOSPC;

+ event->hw.state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
+
pcpu->event = event;

if (flags & PERF_EF_START)
@@ -162,9 +302,11 @@ static void perf_ibs_del(struct perf_event *event, int flags)
if (!test_and_clear_bit(IBS_ENABLED, pcpu->state))
return;

- perf_ibs_stop(event, 0);
+ perf_ibs_stop(event, PERF_EF_UPDATE);

pcpu->event = NULL;
+
+ perf_event_update_userpage(event);
}

static void perf_ibs_read(struct perf_event *event) { }
@@ -185,8 +327,11 @@ static struct perf_ibs perf_ibs_fetch = {
.cnt_mask = IBS_FETCH_MAX_CNT,
.enable_mask = IBS_FETCH_ENABLE,
.valid_mask = IBS_FETCH_VAL,
+ .max_period = IBS_FETCH_MAX_CNT << 4,
.offset_mask = { MSR_AMD64_IBSFETCH_REG_MASK },
.offset_max = MSR_AMD64_IBSFETCH_REG_COUNT,
+
+ .get_count = get_ibs_fetch_count,
};

static struct perf_ibs perf_ibs_op = {
@@ -205,8 +350,11 @@ static struct perf_ibs perf_ibs_op = {
.cnt_mask = IBS_OP_MAX_CNT,
.enable_mask = IBS_OP_ENABLE,
.valid_mask = IBS_OP_VAL,
+ .max_period = IBS_OP_MAX_CNT << 4,
.offset_mask = { MSR_AMD64_IBSOP_REG_MASK },
.offset_max = MSR_AMD64_IBSOP_REG_COUNT,
+
+ .get_count = get_ibs_op_count,
};

static int perf_ibs_handle_irq(struct perf_ibs *perf_ibs, struct pt_regs *iregs)
@@ -218,9 +366,9 @@ static int perf_ibs_handle_irq(struct perf_ibs *perf_ibs, struct pt_regs *iregs)
struct perf_raw_record raw;
struct pt_regs regs;
struct perf_ibs_data ibs_data;
- int offset, size;
+ int offset, size, overflow, reenable;
unsigned int msr;
- u64 *buf;
+ u64 *buf, config;

if (!test_bit(IBS_STARTED, pcpu->state)) {
/* Catch spurious interrupts after stopping IBS: */
@@ -255,11 +403,25 @@ static int perf_ibs_handle_irq(struct perf_ibs *perf_ibs, struct pt_regs *iregs)

regs = *iregs; /* XXX: update ip from ibs sample */

- if (perf_event_overflow(event, &data, &regs))
- ; /* stop */
- else
- /* reenable */
- wrmsrl(hwc->config_base, hwc->config | perf_ibs->enable_mask);
+ /*
+ * Emulate IbsOpCurCnt in MSRC001_1033 (IbsOpCtl), not
+ * supported in all cpus. As this triggered an interrupt, we
+ * set the current count to the max count.
+ */
+ config = ibs_data.regs[0];
+ if (perf_ibs == &perf_ibs_op && !(ibs_caps & IBS_CAPS_RDWROPCNT)) {
+ config &= ~IBS_OP_CUR_CNT;
+ config |= (config & IBS_OP_MAX_CNT) << 36;
+ }
+
+ perf_ibs_event_update(perf_ibs, event, config);
+
+ overflow = perf_ibs_set_period(perf_ibs, hwc, &config);
+ reenable = !(overflow && perf_event_overflow(event, &data, &regs));
+ config = (config >> 4) | (reenable ? perf_ibs->enable_mask : 0);
+ perf_ibs_enable_event(hwc, config);
+
+ perf_event_update_userpage(event);

return 1;
}
--
1.7.6.1


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/