Re: [PATCH v10 4/5] arm64: perf: Enable PMU counter userspace access for perf event

[Mark Rutland <mark.rutland@arm.com>]

Hi Rob,

This looks pretty good!

I have one largish query below, and otherwise only trivialities that I'm
happy to fix up.

On Tue, Sep 14, 2021 at 03:47:59PM -0500, Rob Herring wrote:
> Arm PMUs can support direct userspace access of counters which allows for
> low overhead (i.e. no syscall) self-monitoring of tasks. The same feature
> exists on x86 called 'rdpmc'. Unlike x86, userspace access will only be
> enabled for thread bound events. This could be extended if needed, but
> simplifies the implementation and reduces the chances for any
> information leaks (which the x86 implementation suffers from).
> 
> PMU EL0 access will be enabled when an event with userspace access is
> part of the thread's context. This includes when the event is not
> scheduled on the PMU. There's some additional overhead clearing
> dirty counters when access is enabled in order to prevent leaking
> disabled counter data from other tasks.
> 
> Unlike x86, enabling of userspace access must be requested with a new
> attr bit: config1:1. If the user requests userspace access and 64-bit
> counters, then chaining will be disabled and the user will get the
> maximum size counter the underlying h/w can support. The modes for
> config1 are as follows:
> 
> config1 = 0 : user access disabled and always 32-bit
> config1 = 1 : user access disabled and always 64-bit (using chaining if needed)
> config1 = 2 : user access enabled and always 32-bit
> config1 = 3 : user access enabled and counter size matches underlying counter.

We probably need to note somewhere (i.e. in the next patch) that we mean
*logically* 32-bit, and this could be a biased 64-bit counter, so
userspace needs to treat the upper 32-bits of counters as UNKNOWN.

For the `config1 = 3` case (potentially) overriding the usual long
semantic, I'm struggling to understand why we need that rather than
forcing the use of a 64-bit counter, because in that case:

* For a CPU_CYCLES event:
  __armv8_pmuv3_map_event() will always pick 64-bits
  get_event_idx() may fail to allocate a 64-bit counter.

* For other events:
  __armv8_pmuv3_map_event() will pick 32/64 based on long counter
  support
  get_event_idx() will only fail if there are no counters free.

Whereas if __armv8_pmuv3_map_event() returned an error for the latter
when long counter support is not implemented, we'd have consistent
`long` semantics, and the CPU_CYCLES behaviour would be identical.

What's the rationale for `3` leaving the choice to the kernel?

If the problem is discoverability, I'd be happy to add something to
sysfs to describe whether the PMU has long event support.

> Based on work by Raphael Gault <raphael.gault@arm.com>, but has been
> completely re-written.
> 
> Cc: Will Deacon <will@kernel.org>
> Cc: Mark Rutland <mark.rutland@arm.com>
> Cc: Peter Zijlstra <peterz@infradead.org>
> Cc: Ingo Molnar <mingo@redhat.com>
> Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
> Cc: Jiri Olsa <jolsa@redhat.com>
> Cc: Namhyung Kim <namhyung@kernel.org>
> Cc: Catalin Marinas <catalin.marinas@arm.com>
> Cc: linux-arm-kernel@lists.infradead.org
> Cc: linux-perf-users@vger.kernel.org
> Signed-off-by: Rob Herring <robh@kernel.org>
> 
> ---
> v10:
>  - Don't control enabling user access based on mmap(). Changing the
>    event_(un)mapped to run on the event's cpu doesn't work for x86.
>    Triggering on mmap() doesn't limit access in any way and complicates
>    the implementation.
>  - Drop dirty counter tracking and just clear all unused counters.
>  - Make the sysctl immediately disable access via IPI.
>  - Merge armv8pmu_event_is_chained() and armv8pmu_event_can_chain()
> 
> v9:
>  - Enabling/disabling of user access is now controlled in .start() and
>    mmap hooks which are now called on CPUs that the event is on.
>    Depends on rework of perf core and x86 RDPMC code posted here:
>    https://lore.kernel.org/lkml/20210728230230.1911468-1-robh@kernel.org/
> 
> v8:
>  - Rework user access tracking and enabling to be done on task
>    context changes using sched_task() hook. This avoids the need for any
>    IPIs, mm_switch hooks or undef instr handler.
>  - Only support user access when explicitly requested on open and
>    only for a thread bound events. This avoids some of the information
>    leaks x86 has and simplifies the implementation.
> 
> v7:
>  - Clear disabled counters when user access is enabled for a task to
>    avoid leaking other tasks counter data.
>  - Rework context switch handling utilizing sched_task callback
>  - Add armv8pmu_event_can_chain() helper
>  - Rework config1 flags handling structure
>  - Use ARMV8_IDX_CYCLE_COUNTER_USER define for remapped user cycle
>    counter index
> 
> v6:
>  - Add new attr.config1 rdpmc bit for userspace to hint it wants
>    userspace access when also requesting 64-bit counters.
> 
> v5:
>  - Only set cap_user_rdpmc if event is on current cpu
>  - Limit enabling/disabling access to CPUs associated with the PMU
>    (supported_cpus) and with the mm_struct matching current->active_mm.
> 
> v2:
>  - Move mapped/unmapped into arm64 code. Fixes arm32.
>  - Rebase on cap_user_time_short changes
> 
> Changes from Raphael's v4:
>   - Drop homogeneous check
>   - Disable access for chained counters
>   - Set pmc_width in user page
> ---
>  arch/arm64/kernel/perf_event.c | 99 +++++++++++++++++++++++++++++++---
>  1 file changed, 92 insertions(+), 7 deletions(-)
> 
> diff --git a/arch/arm64/kernel/perf_event.c b/arch/arm64/kernel/perf_event.c
> index a8f8dd741aeb..91af3d1c254b 100644
> --- a/arch/arm64/kernel/perf_event.c
> +++ b/arch/arm64/kernel/perf_event.c
> @@ -285,6 +285,7 @@ static const struct attribute_group armv8_pmuv3_events_attr_group = {
>  
>  PMU_FORMAT_ATTR(event, "config:0-15");
>  PMU_FORMAT_ATTR(long, "config1:0");
> +PMU_FORMAT_ATTR(rdpmc, "config1:1");
>  
>  static int sysctl_perf_user_access __read_mostly;
>  
> @@ -293,9 +294,15 @@ static inline bool armv8pmu_event_is_64bit(struct perf_event *event)
>  	return event->attr.config1 & 0x1;
>  }
>  
> +static inline bool armv8pmu_event_want_user_access(struct perf_event *event)
> +{
> +	return event->attr.config1 & 0x2;
> +}
> +
>  static struct attribute *armv8_pmuv3_format_attrs[] = {
>  	&format_attr_event.attr,
>  	&format_attr_long.attr,
> +	&format_attr_rdpmc.attr,
>  	NULL,
>  };
>  
> @@ -364,7 +371,7 @@ static const struct attribute_group armv8_pmuv3_caps_attr_group = {
>   */
>  #define	ARMV8_IDX_CYCLE_COUNTER	0
>  #define	ARMV8_IDX_COUNTER0	1
> -
> +#define	ARMV8_IDX_CYCLE_COUNTER_USER	32
>  
>  /*
>   * We unconditionally enable ARMv8.5-PMU long event counter support
> @@ -379,15 +386,14 @@ static bool armv8pmu_has_long_event(struct arm_pmu *cpu_pmu)
>  /*
>   * We must chain two programmable counters for 64 bit events,
>   * except when we have allocated the 64bit cycle counter (for CPU
> - * cycles event). This must be called only when the event has
> - * a counter allocated.
> + * cycles event) or when user space counter access is enabled.
>   */
>  static inline bool armv8pmu_event_is_chained(struct perf_event *event)
>  {
>  	int idx = event->hw.idx;
>  	struct arm_pmu *cpu_pmu = to_arm_pmu(event->pmu);
>  
> -	return !WARN_ON(idx < 0) &&
> +	return !(event->hw.flags & PERF_EVENT_FLAG_USER_READ_CNT) &&
>  	       armv8pmu_event_is_64bit(event) &&
>  	       !armv8pmu_has_long_event(cpu_pmu) &&
>  	       (idx != ARMV8_IDX_CYCLE_COUNTER);
> @@ -720,6 +726,27 @@ static inline u32 armv8pmu_getreset_flags(void)
>  	return value;

Above this, could we please add:

| static inline bool armv8pmu_event_has_user_read(struct perf_event *event)
| {
| 	return event->hw.flags & PERF_EVENT_FLAG_USER_READ_CNT;
| }

... and use that where we look at PERF_EVENT_FLAG_USER_READ_CNT?

>  
> +static void armv8pmu_disable_user_access(void)
> +{
> +	write_sysreg(0, pmuserenr_el0);
> +}
> +
> +static void armv8pmu_enable_user_access(struct arm_pmu *cpu_pmu)
> +{
> +	int i;
> +	struct pmu_hw_events *cpuc = this_cpu_ptr(cpu_pmu->hw_events);
> +
> +	/* Clear any unused counters to avoid leaking their contents */
> +	for_each_clear_bit(i, cpuc->used_mask, cpu_pmu->num_events) {
> +		if (i == ARMV8_IDX_CYCLE_COUNTER)
> +			write_sysreg(0, pmccntr_el0);
> +		else
> +			armv8pmu_write_evcntr(i, 0);
> +	}
> +
> +	write_sysreg(ARMV8_PMU_USERENR_ER | ARMV8_PMU_USERENR_CR, pmuserenr_el0);
> +}
> +
>  static void armv8pmu_enable_event(struct perf_event *event)
>  {
>  	/*
> @@ -763,6 +790,14 @@ static void armv8pmu_disable_event(struct perf_event *event)
>  
>  static void armv8pmu_start(struct arm_pmu *cpu_pmu)
>  {
> +	struct perf_event_context *task_ctx =
> +		this_cpu_ptr(cpu_pmu->pmu.pmu_cpu_context)->task_ctx;
> +
> +	if (sysctl_perf_user_access && task_ctx && task_ctx->nr_user)
> +		armv8pmu_enable_user_access(cpu_pmu);
> +	else
> +		armv8pmu_disable_user_access();
> +
>  	/* Enable all counters */
>  	armv8pmu_pmcr_write(armv8pmu_pmcr_read() | ARMV8_PMU_PMCR_E);
>  }
> @@ -880,13 +915,16 @@ static int armv8pmu_get_event_idx(struct pmu_hw_events *cpuc,
>  	if (evtype == ARMV8_PMUV3_PERFCTR_CPU_CYCLES) {
>  		if (!test_and_set_bit(ARMV8_IDX_CYCLE_COUNTER, cpuc->used_mask))
>  			return ARMV8_IDX_CYCLE_COUNTER;
> +		else if (armv8pmu_event_is_64bit(event) &&
> +			   armv8pmu_event_want_user_access(event) &&
> +			   !armv8pmu_has_long_event(cpu_pmu))
> +				return -EAGAIN;
>  	}
>  
>  	/*
>  	 * Otherwise use events counters
>  	 */
> -	if (armv8pmu_event_is_64bit(event) &&
> -	    !armv8pmu_has_long_event(cpu_pmu))
> +	if (armv8pmu_event_is_chained(event))
>  		return	armv8pmu_get_chain_idx(cpuc, cpu_pmu);
>  	else
>  		return armv8pmu_get_single_idx(cpuc, cpu_pmu);
> @@ -902,6 +940,23 @@ static void armv8pmu_clear_event_idx(struct pmu_hw_events *cpuc,
>  		clear_bit(idx - 1, cpuc->used_mask);
>  }
>  
> +static int armv8pmu_access_event_idx(struct perf_event *event)

Can we please s/access/user/ here?

> +{
> +	if (!sysctl_perf_user_access ||
> +	    !(event->hw.flags & PERF_EVENT_FLAG_USER_READ_CNT))
> +		return 0;
> +
> +	/*
> +	 * We remap the cycle counter index to 32 to
> +	 * match the offset applied to the rest of
> +	 * the counter indices.
> +	 */
> +	if (event->hw.idx == ARMV8_IDX_CYCLE_COUNTER)
> +		return ARMV8_IDX_CYCLE_COUNTER_USER;
> +
> +	return event->hw.idx;
> +}
> +
>  /*
>   * Add an event filter to a given event.
>   */
> @@ -995,9 +1050,23 @@ static int __armv8_pmuv3_map_event(struct perf_event *event,
>  				       &armv8_pmuv3_perf_cache_map,
>  				       ARMV8_PMU_EVTYPE_EVENT);
>  
> -	if (armv8pmu_event_is_64bit(event))
> +	/*
> +	 * At this point, the counter is not assigned. If a 64-bit counter is
> +	 * requested, we must make sure the h/w has 64-bit counters if we set
> +	 * the event size to 64-bit because chaining is not supported with
> +	 * userspace access. This may still fail later on if the CPU cycle
> +	 * counter is in use.
> +	 */
> +	if (armv8pmu_event_is_64bit(event) &&
> +	    (!armv8pmu_event_want_user_access(event) ||
> +	     armv8pmu_has_long_event(armpmu) || (hw_event_id == ARMV8_PMUV3_PERFCTR_CPU_CYCLES)))
>  		event->hw.flags |= ARMPMU_EVT_64BIT;

If we can follow my suggestion in reply to the cover text, we can make
this:

	if (armv8pmu_event_is_64bit(event))
		event->hw.flags |= ARMPMU_EVT_64BIT;

	/*
	 * User events must be allocated into a single counter, and so
	 * must not be chained.
	 *
	 * Most 64-bit events require long counter support, but 64-bit
	 * CPU_CYCLES events can be placed into the dedicated cycle
	 * counter when this is free.
	 *
	if (armv8pmu_event_want_user_access()) {
		if (armv8pmu_event_is_64bit(event) &&
		    (hw_event_id != ARMV8_PMUV3_PERFCTR_CPU_CYCLES) &&
		    !armv8pmu_has_long_event(armpmu))
			return -EINVAL;
	}
	
> +	/* Userspace counter access only enabled if requested and a per task event */
> +	if (sysctl_perf_user_access && armv8pmu_event_want_user_access(event) &&
> +	    (event->attach_state & PERF_ATTACH_TASK))
> +		event->hw.flags |= PERF_EVENT_FLAG_USER_READ_CNT;

Can we please explicitly reject !PERF_ATTACH_TASK case?

If the user requested something we don't intend to support, I'd rather
return -EINVAL here, rather than continue on.

Thanks,
Mark.

> +
>  	/* Only expose micro/arch events supported by this PMU */
>  	if ((hw_event_id > 0) && (hw_event_id < ARMV8_PMUV3_MAX_COMMON_EVENTS)
>  	    && test_bit(hw_event_id, armpmu->pmceid_bitmap)) {
> @@ -1106,6 +1175,11 @@ static int armv8pmu_probe_pmu(struct arm_pmu *cpu_pmu)
>  	return probe.present ? 0 : -ENODEV;
>  }
>  
> +static void armv8pmu_disable_user_access_ipi(void *unused)
> +{
> +	armv8pmu_disable_user_access();
> +}
> +
>  int armv8pmu_proc_user_access_handler(struct ctl_table *table, int write,
>                  void *buffer, size_t *lenp, loff_t *ppos)
>  {
> @@ -1113,6 +1187,7 @@ int armv8pmu_proc_user_access_handler(struct ctl_table *table, int write,
>  	if (ret || !write || sysctl_perf_user_access)
>  		return ret;
>  
> +	on_each_cpu(armv8pmu_disable_user_access_ipi, NULL, 1);
>  	return 0;
>  }
>  
> @@ -1152,6 +1227,8 @@ static int armv8_pmu_init(struct arm_pmu *cpu_pmu, char *name,
>  	cpu_pmu->set_event_filter	= armv8pmu_set_event_filter;
>  	cpu_pmu->filter_match		= armv8pmu_filter_match;
>  
> +	cpu_pmu->pmu.event_idx		= armv8pmu_access_event_idx;
> +
>  	cpu_pmu->name			= name;
>  	cpu_pmu->map_event		= map_event;
>  	cpu_pmu->attr_groups[ARMPMU_ATTR_GROUP_EVENTS] = events ?
> @@ -1328,6 +1405,14 @@ void arch_perf_update_userpage(struct perf_event *event,
>  	userpg->cap_user_time = 0;
>  	userpg->cap_user_time_zero = 0;
>  	userpg->cap_user_time_short = 0;
> +	userpg->cap_user_rdpmc = !!(event->hw.flags & PERF_EVENT_FLAG_USER_READ_CNT);
> +
> +	if (userpg->cap_user_rdpmc) {
> +		if (event->hw.flags & ARMPMU_EVT_64BIT)
> +			userpg->pmc_width = 64;
> +		else
> +			userpg->pmc_width = 32;
> +	}
>  
>  	do {
>  		rd = sched_clock_read_begin(&seq);
> -- 
> 2.30.2
>