Re: [PATCH RFC V2 1/4] perf/core: Add aux_pause, aux_resume, aux_start_paused

From: Arnaldo Carvalho de Melo
Date: Tue Dec 19 2023 - 08:42:36 EST

Next message: Christian Brauner: "Re: [RFC PATCH v3 3/3] devguard: added device guard for mknod in non-initial userns"
Previous message: Greg Kroah-Hartman: "[PATCH] dma-debug: make dma_debug_add_bus take a const pointer"
Next in thread: James Clark: "Re: [PATCH RFC V2 1/4] perf/core: Add aux_pause, aux_resume, aux_start_paused"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Em Fri, Dec 08, 2023 at 07:24:46PM +0200, Adrian Hunter escreveu:
> Hardware traces, such as instruction traces, can produce a vast amount of
> trace data, so being able to reduce tracing to more specific circumstances
> can be useful.

> The ability to pause or resume tracing when another event happens, can do
> that.

> Add ability for an event to "pause" or "resume" AUX area tracing.

We need this as well for the usual ring buffer, 'perf report' has:

--switch-off <event>
Stop considering events after the occurrence of this event
--switch-on <event>
Consider events after the occurrence of this event

And 'perf record' has:

--switch-output-event
Events that will cause the switch of the perf.data file, auto-selecting --switch-output=signal, the results are similar as internally the side band thread will also send a
SIGUSR2 to the main one.

But those are all in userspace, what you're doing is in the kernel, and
for the example you used synchronous, i.e. you're only interested in
what happens after you enter the syscall and then stop when the syscall
exits (but here you'll catch more stuff in the AUX trace, i.e. a "race"
from intel_pt inserting events in the AUX trace and then the syscall
exit switching it off).

Also being able to group the { resume, what-to-enable, pause } is
powerful, as we could have multiple such groups to record those
"slices", not just the --switch-off/--switch-on global ones.

> Add aux_pause bit to perf_event_attr to indicate that, if the event
> happens, the associated AUX area tracing should be paused. Ditto
> aux_resume. Do not allow aux_pause and aux_resume to be set together.
>
> Add aux_start_paused bit to perf_event_attr to indicate to an AUX area
> event that it should start in a "paused" state.
>
> Add aux_paused to struct perf_event for AUX area events to keep track of
> the "paused" state. aux_paused is initialized to aux_start_paused.
>
> Add PERF_EF_PAUSE and PERF_EF_RESUME modes for ->stop() and ->start()
> callbacks. Call as needed, during __perf_event_output(). Add
> aux_in_pause_resume to struct perf_buffer to prevent races with the NMI
> handler. Pause/resume in NMI context will miss out if it coincides with
> another pause/resume.
>
> To use aux_pause or aux_resume, an event must be in a group with the AUX
> area event as the group leader.
>
> Example (requires Intel PT and tools patches also):
>
> $ perf record --kcore -e '{intel_pt/aux-start-paused/k,syscalls:sys_enter_newuname/aux-resume/,syscalls:sys_exit_newuname/aux-pause/}' uname

User interface looks nice, asks for intel_pt to be armed but start
paused, collect just inside the kernel, then sets up another event to
enable the collection of whatever is using the aux area, intel_pt in
this case, and then one other event, sys_exit_newuname to pause it
again.

So this implicitely selects the CPU where the aux-resume took place and
in this specific case we ended up being lucky and that process wasn't
migrated to another CPU in the middle of the syscall...

Scratch that, you're not tracing system wide, but just the 'uname'
process being started from perf, perfect.

- Arnaldo

> Linux
> [ perf record: Woken up 1 times to write data ]
> [ perf record: Captured and wrote 0.041 MB perf.data ]
> $ perf script --call-trace
> uname 5712 [007] 83.855580930: name: 0x7ffd9dcebec0
> uname 5712 [007] 83.855582518: psb offs: 0
> uname 5712 [007] 83.855582518: cbr: 42 freq: 4205 MHz (150%)
> uname 5712 [007] 83.855582723: ([kernel.kallsyms]) debug_smp_processor_id
> uname 5712 [007] 83.855582723: ([kernel.kallsyms]) __x64_sys_newuname
> uname 5712 [007] 83.855582723: ([kernel.kallsyms]) down_read
> uname 5712 [007] 83.855582723: ([kernel.kallsyms]) __cond_resched
> uname 5712 [007] 83.855582932: ([kernel.kallsyms]) preempt_count_add
> uname 5712 [007] 83.855582932: ([kernel.kallsyms]) in_lock_functions
> uname 5712 [007] 83.855582932: ([kernel.kallsyms]) preempt_count_sub
> uname 5712 [007] 83.855582932: ([kernel.kallsyms]) up_read
> uname 5712 [007] 83.855582932: ([kernel.kallsyms]) preempt_count_add
> uname 5712 [007] 83.855583348: ([kernel.kallsyms]) in_lock_functions
> uname 5712 [007] 83.855583348: ([kernel.kallsyms]) preempt_count_sub
> uname 5712 [007] 83.855583348: ([kernel.kallsyms]) _copy_to_user
> uname 5712 [007] 83.855583557: ([kernel.kallsyms]) syscall_exit_to_user_mode
> uname 5712 [007] 83.855583557: ([kernel.kallsyms]) syscall_exit_work
> uname 5712 [007] 83.855583557: ([kernel.kallsyms]) perf_syscall_exit
> uname 5712 [007] 83.855583557: ([kernel.kallsyms]) debug_smp_processor_id
> uname 5712 [007] 83.855583557: ([kernel.kallsyms]) perf_trace_buf_alloc
> uname 5712 [007] 83.855583557: ([kernel.kallsyms]) perf_swevent_get_recursion_context
> uname 5712 [007] 83.855583557: ([kernel.kallsyms]) debug_smp_processor_id
> uname 5712 [007] 83.855583557: ([kernel.kallsyms]) debug_smp_processor_id
> uname 5712 [007] 83.855583557: ([kernel.kallsyms]) perf_tp_event
> uname 5712 [007] 83.855583557: ([kernel.kallsyms]) perf_trace_buf_update
> uname 5712 [007] 83.855583557: ([kernel.kallsyms]) tracing_gen_ctx_irq_test
> uname 5712 [007] 83.855583557: ([kernel.kallsyms]) perf_swevent_event
> uname 5712 [007] 83.855583765: ([kernel.kallsyms]) __perf_event_account_interrupt
> uname 5712 [007] 83.855583765: ([kernel.kallsyms]) __this_cpu_preempt_check
> uname 5712 [007] 83.855583765: ([kernel.kallsyms]) perf_event_output_forward
> uname 5712 [007] 83.855583765: ([kernel.kallsyms]) perf_event_aux_pause
> uname 5712 [007] 83.855583765: ([kernel.kallsyms]) ring_buffer_get
> uname 5712 [007] 83.855583765: ([kernel.kallsyms]) __rcu_read_lock
> uname 5712 [007] 83.855583765: ([kernel.kallsyms]) __rcu_read_unlock
> uname 5712 [007] 83.855583765: ([kernel.kallsyms]) pt_event_stop
> uname 5712 [007] 83.855583765: ([kernel.kallsyms]) debug_smp_processor_id
> uname 5712 [007] 83.855583765: ([kernel.kallsyms]) debug_smp_processor_id
> uname 5712 [007] 83.855583973: ([kernel.kallsyms]) native_write_msr
> uname 5712 [007] 83.855583973: ([kernel.kallsyms]) native_write_msr
> uname 5712 [007] 83.855584175: 0x0
>
> Signed-off-by: Adrian Hunter <adrian.hunter@xxxxxxxxx>
> ---
> include/linux/perf_event.h | 15 +++++++
> include/uapi/linux/perf_event.h | 11 ++++-
> kernel/events/core.c | 72 +++++++++++++++++++++++++++++++--
> kernel/events/internal.h | 1 +
> 4 files changed, 95 insertions(+), 4 deletions(-)
>
> diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
> index e85cd1c0eaf3..252c4aac3b79 100644
> --- a/include/linux/perf_event.h
> +++ b/include/linux/perf_event.h
> @@ -291,6 +291,7 @@ struct perf_event_pmu_context;
> #define PERF_PMU_CAP_NO_EXCLUDE 0x0040
> #define PERF_PMU_CAP_AUX_OUTPUT 0x0080
> #define PERF_PMU_CAP_EXTENDED_HW_TYPE 0x0100
> +#define PERF_PMU_CAP_AUX_PAUSE 0x0200
>
> struct perf_output_handle;
>
> @@ -363,6 +364,8 @@ struct pmu {
> #define PERF_EF_START 0x01 /* start the counter when adding */
> #define PERF_EF_RELOAD 0x02 /* reload the counter when starting */
> #define PERF_EF_UPDATE 0x04 /* update the counter when stopping */
> +#define PERF_EF_PAUSE 0x08 /* AUX area event, pause tracing */
> +#define PERF_EF_RESUME 0x10 /* AUX area event, resume tracing */
>
> /*
> * Adds/Removes a counter to/from the PMU, can be done inside a
> @@ -402,6 +405,15 @@ struct pmu {
> *
> * ->start() with PERF_EF_RELOAD will reprogram the counter
> * value, must be preceded by a ->stop() with PERF_EF_UPDATE.
> + *
> + * ->stop() with PERF_EF_PAUSE will stop as simply as possible. Will not
> + * overlap another ->stop() with PERF_EF_PAUSE nor ->start() with
> + * PERF_EF_RESUME.
> + *
> + * ->start() with PERF_EF_RESUME will start as simply as possible but
> + * only if the counter is not otherwise stopped. Will not overlap
> + * another ->start() with PERF_EF_RESUME nor ->stop() with
> + * PERF_EF_PAUSE.
> */
> void (*start) (struct perf_event *event, int flags);
> void (*stop) (struct perf_event *event, int flags);
> @@ -797,6 +809,9 @@ struct perf_event {
> /* for aux_output events */
> struct perf_event *aux_event;
>
> + /* for AUX area events */
> + unsigned int aux_paused;
> +
> void (*destroy)(struct perf_event *);
> struct rcu_head rcu_head;
>
> diff --git a/include/uapi/linux/perf_event.h b/include/uapi/linux/perf_event.h
> index 39c6a250dd1b..437bc2a8d50c 100644
> --- a/include/uapi/linux/perf_event.h
> +++ b/include/uapi/linux/perf_event.h
> @@ -507,7 +507,16 @@ struct perf_event_attr {
> __u16 sample_max_stack;
> __u16 __reserved_2;
> __u32 aux_sample_size;
> - __u32 __reserved_3;
> +
> + union {
> + __u32 aux_output_cfg;
> + struct {
> + __u64 aux_pause : 1, /* on overflow, pause AUX area tracing */
> + aux_resume : 1, /* on overflow, resume AUX area tracing */
> + aux_start_paused : 1, /* start AUX area tracing paused */
> + __reserved_3 : 29;
> + };
> + };
>
> /*
> * User provided data if sigtrap=1, passed back to user via
> diff --git a/kernel/events/core.c b/kernel/events/core.c
> index 4c72a41f11af..c1e11884d06e 100644
> --- a/kernel/events/core.c
> +++ b/kernel/events/core.c
> @@ -2060,7 +2060,8 @@ static void perf_put_aux_event(struct perf_event *event)
>
> static bool perf_need_aux_event(struct perf_event *event)
> {
> - return !!event->attr.aux_output || !!event->attr.aux_sample_size;
> + return event->attr.aux_output || event->attr.aux_sample_size ||
> + event->attr.aux_pause || event->attr.aux_resume;
> }
>
> static int perf_get_aux_event(struct perf_event *event,
> @@ -2085,6 +2086,10 @@ static int perf_get_aux_event(struct perf_event *event,
> !perf_aux_output_match(event, group_leader))
> return 0;
>
> + if ((event->attr.aux_pause || event->attr.aux_resume) &&
> + !(group_leader->pmu->capabilities & PERF_PMU_CAP_AUX_PAUSE))
> + return 0;
> +
> if (event->attr.aux_sample_size && !group_leader->pmu->snapshot_aux)
> return 0;
>
> @@ -7773,6 +7778,47 @@ void perf_prepare_header(struct perf_event_header *header,
> WARN_ON_ONCE(header->size & 7);
> }
>
> +static void __perf_event_aux_pause(struct perf_event *event, bool pause)
> +{
> + if (pause) {
> + if (!READ_ONCE(event->aux_paused)) {
> + WRITE_ONCE(event->aux_paused, 1);
> + event->pmu->stop(event, PERF_EF_PAUSE);
> + }
> + } else {
> + if (READ_ONCE(event->aux_paused)) {
> + WRITE_ONCE(event->aux_paused, 0);
> + event->pmu->start(event, PERF_EF_RESUME);
> + }
> + }
> +}
> +
> +static void perf_event_aux_pause(struct perf_event *event, bool pause)
> +{
> + struct perf_buffer *rb;
> + unsigned long flags;
> +
> + if (WARN_ON_ONCE(!event))
> + return;
> +
> + rb = ring_buffer_get(event);
> + if (!rb)
> + return;
> +
> + local_irq_save(flags);
> + /* Guard against NMI, NMI loses here */
> + if (READ_ONCE(rb->aux_in_pause_resume))
> + goto out_restore;
> + WRITE_ONCE(rb->aux_in_pause_resume, 1);
> + barrier();
> + __perf_event_aux_pause(event, pause);
> + barrier();
> + WRITE_ONCE(rb->aux_in_pause_resume, 0);
> +out_restore:
> + local_irq_restore(flags);
> + ring_buffer_put(rb);
> +}
> +
> static __always_inline int
> __perf_event_output(struct perf_event *event,
> struct perf_sample_data *data,
> @@ -7786,6 +7832,9 @@ __perf_event_output(struct perf_event *event,
> struct perf_event_header header;
> int err;
>
> + if (event->attr.aux_pause)
> + perf_event_aux_pause(event->aux_event, true);
> +
> /* protect the callchain buffers */
> rcu_read_lock();
>
> @@ -7802,6 +7851,10 @@ __perf_event_output(struct perf_event *event,
>
> exit:
> rcu_read_unlock();
> +
> + if (event->attr.aux_resume)
> + perf_event_aux_pause(event->aux_event, false);
> +
> return err;
> }
>
> @@ -11941,10 +11994,23 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu,
> }
>
> if (event->attr.aux_output &&
> - !(pmu->capabilities & PERF_PMU_CAP_AUX_OUTPUT)) {
> + (!(pmu->capabilities & PERF_PMU_CAP_AUX_OUTPUT) ||
> + event->attr.aux_pause || event->attr.aux_resume)) {
> + err = -EOPNOTSUPP;
> + goto err_pmu;
> + }
> +
> + if (event->attr.aux_pause && event->attr.aux_resume) {
> + err = -EINVAL;
> + goto err_pmu;
> + }
> +
> + if (event->attr.aux_start_paused &&
> + !(pmu->capabilities & PERF_PMU_CAP_AUX_PAUSE)) {
> err = -EOPNOTSUPP;
> goto err_pmu;
> }
> + event->aux_paused = event->attr.aux_start_paused;
>
> if (cgroup_fd != -1) {
> err = perf_cgroup_connect(cgroup_fd, event, attr, group_leader);
> @@ -12741,7 +12807,7 @@ perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu,
> * Grouping is not supported for kernel events, neither is 'AUX',
> * make sure the caller's intentions are adjusted.
> */
> - if (attr->aux_output)
> + if (attr->aux_output || attr->aux_output_cfg)
> return ERR_PTR(-EINVAL);
>
> event = perf_event_alloc(attr, cpu, task, NULL, NULL,
> diff --git a/kernel/events/internal.h b/kernel/events/internal.h
> index 5150d5f84c03..3320f78117dc 100644
> --- a/kernel/events/internal.h
> +++ b/kernel/events/internal.h
> @@ -51,6 +51,7 @@ struct perf_buffer {
> void (*free_aux)(void *);
> refcount_t aux_refcount;
> int aux_in_sampling;
> + int aux_in_pause_resume;
> void **aux_pages;
> void *aux_priv;
>
> --
> 2.34.1
>

--

- Arnaldo

Next message: Christian Brauner: "Re: [RFC PATCH v3 3/3] devguard: added device guard for mknod in non-initial userns"
Previous message: Greg Kroah-Hartman: "[PATCH] dma-debug: make dma_debug_add_bus take a const pointer"
Next in thread: James Clark: "Re: [PATCH RFC V2 1/4] perf/core: Add aux_pause, aux_resume, aux_start_paused"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]