[PATCH V12 06/14] perf/x86/intel/pt: Add support for pause / resume

From: Adrian Hunter
Date: Thu Oct 10 2024 - 11:08:57 EST


Prevent tracing to start if aux_paused.

Implement support for PERF_EF_PAUSE / PERF_EF_RESUME. When aux_paused, stop
tracing. When not aux_paused, only start tracing if it isn't currently
meant to be stopped.

Signed-off-by: Adrian Hunter <adrian.hunter@xxxxxxxxx>
Reviewed-by: Andi Kleen <ak@xxxxxxxxxxxxxxx>
---


Changes in V12:
Rebase on current tip plus patch set "KVM: x86: Fix Intel PT Host/Guest
mode when host tracing"

Changes in V9:
Add more comments and barriers for resume_allowed and
pause_allowed
Always use WRITE_ONCE with resume_allowed


arch/x86/events/intel/pt.c | 69 ++++++++++++++++++++++++++++++++++++--
arch/x86/events/intel/pt.h | 4 +++
2 files changed, 70 insertions(+), 3 deletions(-)

diff --git a/arch/x86/events/intel/pt.c b/arch/x86/events/intel/pt.c
index d9469d2d6aa6..b6cfca251c07 100644
--- a/arch/x86/events/intel/pt.c
+++ b/arch/x86/events/intel/pt.c
@@ -418,6 +418,9 @@ static void pt_config_start(struct perf_event *event)
struct pt *pt = this_cpu_ptr(&pt_ctx);
u64 ctl = event->hw.aux_config;

+ if (READ_ONCE(event->hw.aux_paused))
+ return;
+
ctl |= RTIT_CTL_TRACEEN;
if (READ_ONCE(pt->vmx_on))
perf_aux_output_flag(&pt->handle, PERF_AUX_FLAG_PARTIAL);
@@ -539,11 +542,23 @@ static void pt_config(struct perf_event *event)

event->hw.aux_config = reg;

+ /*
+ * Allow resume before starting so as not to overwrite a value set by a
+ * PMI.
+ */
+ barrier();
+ WRITE_ONCE(pt->resume_allowed, 1);
/* Configuration is complete, it is now OK to handle an NMI */
barrier();
WRITE_ONCE(pt->handle_nmi, 1);
-
+ barrier();
pt_config_start(event);
+ barrier();
+ /*
+ * Allow pause after starting so its pt_config_stop() doesn't race with
+ * pt_config_start().
+ */
+ WRITE_ONCE(pt->pause_allowed, 1);
}

static void pt_config_stop(struct perf_event *event)
@@ -1526,6 +1541,7 @@ void intel_pt_interrupt(void)
buf = perf_aux_output_begin(&pt->handle, event);
if (!buf) {
event->hw.state = PERF_HES_STOPPED;
+ WRITE_ONCE(pt->resume_allowed, 0);
return;
}

@@ -1534,6 +1550,7 @@ void intel_pt_interrupt(void)
ret = pt_buffer_reset_markers(buf, &pt->handle);
if (ret) {
perf_aux_output_end(&pt->handle, 0);
+ WRITE_ONCE(pt->resume_allowed, 0);
return;
}

@@ -1588,6 +1605,26 @@ static void pt_event_start(struct perf_event *event, int mode)
struct pt *pt = this_cpu_ptr(&pt_ctx);
struct pt_buffer *buf;

+ if (mode & PERF_EF_RESUME) {
+ if (READ_ONCE(pt->resume_allowed)) {
+ u64 status;
+
+ /*
+ * Only if the trace is not active and the error and
+ * stopped bits are clear, is it safe to start, but a
+ * PMI might have just cleared these, so resume_allowed
+ * must be checked again also.
+ */
+ rdmsrl(MSR_IA32_RTIT_STATUS, status);
+ if (!(status & (RTIT_STATUS_TRIGGEREN |
+ RTIT_STATUS_ERROR |
+ RTIT_STATUS_STOPPED)) &&
+ READ_ONCE(pt->resume_allowed))
+ pt_config_start(event);
+ }
+ return;
+ }
+
buf = perf_aux_output_begin(&pt->handle, event);
if (!buf)
goto fail_stop;
@@ -1615,6 +1652,12 @@ static void pt_event_stop(struct perf_event *event, int mode)
{
struct pt *pt = this_cpu_ptr(&pt_ctx);

+ if (mode & PERF_EF_PAUSE) {
+ if (READ_ONCE(pt->pause_allowed))
+ pt_config_stop(event);
+ return;
+ }
+
/*
* Protect against the PMI racing with disabling wrmsr,
* see comment in intel_pt_interrupt().
@@ -1622,6 +1665,15 @@ static void pt_event_stop(struct perf_event *event, int mode)
WRITE_ONCE(pt->handle_nmi, 0);
barrier();

+ /*
+ * Prevent a resume from attempting to restart tracing, or a pause
+ * during a subsequent start. Do this after clearing handle_nmi so that
+ * pt_event_snapshot_aux() will not re-allow them.
+ */
+ WRITE_ONCE(pt->pause_allowed, 0);
+ WRITE_ONCE(pt->resume_allowed, 0);
+ barrier();
+
pt_config_stop(event);

if (event->hw.state == PERF_HES_STOPPED)
@@ -1787,6 +1839,10 @@ static long pt_event_snapshot_aux(struct perf_event *event,
if (WARN_ON_ONCE(!buf->snapshot))
return 0;

+ /* Prevent pause/resume from attempting to start/stop tracing */
+ WRITE_ONCE(pt->pause_allowed, 0);
+ WRITE_ONCE(pt->resume_allowed, 0);
+ barrier();
/*
* There is no PT interrupt in this mode, so stop the trace and it will
* remain stopped while the buffer is copied.
@@ -1806,8 +1862,13 @@ static long pt_event_snapshot_aux(struct perf_event *event,
* Here, handle_nmi tells us if the tracing was on.
* If the tracing was on, restart it.
*/
- if (READ_ONCE(pt->handle_nmi))
+ if (READ_ONCE(pt->handle_nmi)) {
+ WRITE_ONCE(pt->resume_allowed, 1);
+ barrier();
pt_config_start(event);
+ barrier();
+ WRITE_ONCE(pt->pause_allowed, 1);
+ }

return ret;
}
@@ -1923,7 +1984,9 @@ static __init int pt_init(void)
if (!intel_pt_validate_hw_cap(PT_CAP_topa_multiple_entries))
pt_pmu.pmu.capabilities = PERF_PMU_CAP_AUX_NO_SG;

- pt_pmu.pmu.capabilities |= PERF_PMU_CAP_EXCLUSIVE | PERF_PMU_CAP_ITRACE;
+ pt_pmu.pmu.capabilities |= PERF_PMU_CAP_EXCLUSIVE |
+ PERF_PMU_CAP_ITRACE |
+ PERF_PMU_CAP_AUX_PAUSE;
pt_pmu.pmu.attr_groups = pt_attr_groups;
pt_pmu.pmu.task_ctx_nr = perf_sw_context;
pt_pmu.pmu.event_init = pt_event_init;
diff --git a/arch/x86/events/intel/pt.h b/arch/x86/events/intel/pt.h
index 0428019b92f4..480a5a311148 100644
--- a/arch/x86/events/intel/pt.h
+++ b/arch/x86/events/intel/pt.h
@@ -119,6 +119,8 @@ struct pt_filters {
* @filters: last configured filters
* @handle_nmi: do handle PT PMI on this cpu, there's an active event
* @vmx_on: 1 if VMX is ON on this cpu
+ * @pause_allowed: PERF_EF_PAUSE is allowed to stop tracing
+ * @resume_allowed: PERF_EF_RESUME is allowed to start tracing
* @output_base: cached RTIT_OUTPUT_BASE MSR value
* @output_mask: cached RTIT_OUTPUT_MASK MSR value
* @status: cached RTIT_STATUS MSR value
@@ -132,6 +134,8 @@ struct pt {
struct pt_filters filters;
int handle_nmi;
int vmx_on;
+ int pause_allowed;
+ int resume_allowed;
u64 output_base;
u64 output_mask;
u64 status;
--
2.43.0