arm64: perf: Enable PMU counter userspace access for perf event

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 with 64-bit
counters, then the event open will fail if the h/w doesn't support
64-bit counters. Chaining is not supported with userspace access. 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 always 64-bit

Based on work by Raphael Gault <[e-mail]>, but has been
completely re-written.

Cc: Will Deacon <[e-mail]>
Cc: Mark Rutland <[e-mail]>
Cc: Peter Zijlstra <[e-mail]>
Cc: Ingo Molnar <[e-mail]>
Cc: Arnaldo Carvalho de Melo <[e-mail]>
Cc: Alexander Shishkin <[e-mail]>
Cc: Jiri Olsa <[e-mail]>
Cc: Namhyung Kim <[e-mail]>
Cc: Catalin Marinas <[e-mail]>
Cc: [e-mail]
Cc: [e-mail]
Signed-off-by: Rob Herring <[e-mail]>
Link: https://lore.kernel.org/r/[e-mail]
[will: Made armv8pmu_proc_user_access_handler() static]
Signed-off-by: Will Deacon <[e-mail]>

arch/arm64/kernel/perf_event.c

···
 
 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;
 
···
 	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,
 };
 
···
  */
 #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
···
 	return (cpu_pmu->pmuver >= ID_AA64DFR0_PMUVER_8_5);
 }
 
+static inline bool armv8pmu_event_has_user_read(struct perf_event *event)
+{
+	return event->hw.flags & PERF_EVENT_FLAG_USER_READ_CNT;
+}
+
 /*
  * 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 !armv8pmu_event_has_user_read(event) &&
 	       armv8pmu_event_is_64bit(event) &&
 	       !armv8pmu_has_long_event(cpu_pmu) &&
 	       (idx != ARMV8_IDX_CYCLE_COUNTER);
···
 	return value;
 }
 
+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(0, pmuserenr_el0);
+	write_sysreg(ARMV8_PMU_USERENR_ER | ARMV8_PMU_USERENR_CR, pmuserenr_el0);
+}
+
 static void armv8pmu_enable_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);
 }
···
 	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);
···
 	clear_bit(idx, cpuc->used_mask);
 	if (armv8pmu_event_is_chained(event))
 		clear_bit(idx - 1, cpuc->used_mask);
+}
+
+static int armv8pmu_user_event_idx(struct perf_event *event)
+{
+	if (!sysctl_perf_user_access || !armv8pmu_event_has_user_read(event))
+		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;
 }
 
 /*
···
 
 	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(event)) {
+		if (!(event->attach_state & PERF_ATTACH_TASK))
+			return -EINVAL;
+		if (armv8pmu_event_is_64bit(event) &&
+		    (hw_event_id != ARMV8_PMUV3_PERFCTR_CPU_CYCLES) &&
+		    !armv8pmu_has_long_event(armpmu))
+			return -EOPNOTSUPP;
+
+		event->hw.flags |= PERF_EVENT_FLAG_USER_READ_CNT;
+	}
 
 	/* Only expose micro/arch events supported by this PMU */
 	if ((hw_event_id > 0) && (hw_event_id < ARMV8_PMUV3_MAX_COMMON_EVENTS)
···
 	return probe.present ? 0 : -ENODEV;
 }
 
+static void armv8pmu_disable_user_access_ipi(void *unused)
+{
+	armv8pmu_disable_user_access();
+}
+
+static int armv8pmu_proc_user_access_handler(struct ctl_table *table, int write,
+		void *buffer, size_t *lenp, loff_t *ppos)
+{
+	int ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
+	if (ret || !write || sysctl_perf_user_access)
+		return ret;
+
+	on_each_cpu(armv8pmu_disable_user_access_ipi, NULL, 1);
+	return 0;
+}
+
 static struct ctl_table armv8_pmu_sysctl_table[] = {
 	{
 		.procname       = "perf_user_access",
 		.data		= &sysctl_perf_user_access,
 		.maxlen		= sizeof(unsigned int),
 		.mode           = 0644,
-		.proc_handler	= proc_dointvec_minmax,
+		.proc_handler	= armv8pmu_proc_user_access_handler,
 		.extra1		= SYSCTL_ZERO,
 		.extra2		= SYSCTL_ONE,
 	},
···
 	cpu_pmu->reset			= armv8pmu_reset;
 	cpu_pmu->set_event_filter	= armv8pmu_set_event_filter;
 	cpu_pmu->filter_match		= armv8pmu_filter_match;
+
+	cpu_pmu->pmu.event_idx		= armv8pmu_user_event_idx;
 
 	cpu_pmu->name			= name;
 	cpu_pmu->map_event		= map_event;
···
 	userpg->cap_user_time = 0;
 	userpg->cap_user_time_zero = 0;
 	userpg->cap_user_time_short = 0;
+	userpg->cap_user_rdpmc = armv8pmu_event_has_user_read(event);
+
+	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);