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'.

When an event is capable of userspace access and has been mmapped, userspace
access is enabled when the event is scheduled on a CPU's PMU. There's some
additional overhead clearing counters when disabled in order to prevent
leaking disabled counter data from other tasks.

In order to enable 64-bit counters for userspace when available, a new
config1 bit is added for userspace to indicate it wants userspace counter
access. This bit allows the kernel to decide if chaining should be
disabled and chaining and userspace access are incompatible.
The modes for config1 are as follows:

config1 = 0 or 2 : user access enabled and always 32-bit
config1 = 1 : user access disabled and always 64-bit (using chaining if needed)
config1 = 3 : user access enabled and counter size matches underlying counter.

User access is enabled with config1 == 0 so that we match x86 behavior
and don't need Arm specific code (outside the counter read).

Based on work by Raphael Gault <raphael.gault@arm.com>, but has been
completely re-written.

Signed-off-by: Rob Herring <robh@kernel.org>
---
Peter Z says (event->oncpu == smp_processor_id()) in the user page
update is always true, but my testing says otherwise[1].

v8:
 - Rework user access tracking and enabling to be done on task
   context changes. This avoids the need for any IPIs, mm_switch
   hooks or undef instr handler.
 - Clear counter when disabling the counter.

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

[1] https://lore.kernel.org/lkml/CAL_JsqK+eKef5NaVnBfARCjRE3MYhfBfe54F9YHKbsTnWqLmLw@mail.gmail.com/

1 files changed, 111 insertions, 7 deletions

diff --git a/arch/arm64/kernel/perf_event.c b/arch/arm64/kernel/perf_event.c
index f594957e29bd1..1264baec7ab88 100644
--- a/arch/arm64/kernel/perf_event.c
+++ b/arch/arm64/kernel/perf_event.c
@@ -288,15 +288,22 @@ 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 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,
 };
 
@@ -332,7 +339,7 @@ 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
@@ -344,6 +351,15 @@ static bool armv8pmu_has_long_event(struct arm_pmu *cpu_pmu)
 	return (cpu_pmu->pmuver >= ID_AA64DFR0_PMUVER_8_5);
 }
 
+static inline bool armv8pmu_event_can_chain(struct perf_event *event)
+{
+	struct arm_pmu *cpu_pmu = to_arm_pmu(event->pmu);
+
+	return !armv8pmu_event_want_user_access(event) &&
+	       armv8pmu_event_is_64bit(event) &&
+	       !armv8pmu_has_long_event(cpu_pmu);
+}
+
 /*
  * We must chain two programmable counters for 64 bit events,
  * except when we have allocated the 64bit cycle counter (for CPU
@@ -353,11 +369,9 @@ static bool armv8pmu_has_long_event(struct arm_pmu *cpu_pmu)
 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) &&
-	       armv8pmu_event_is_64bit(event) &&
-	       !armv8pmu_has_long_event(cpu_pmu) &&
+	       armv8pmu_event_can_chain(event) &&
 	       (idx != ARMV8_IDX_CYCLE_COUNTER);
 }
 
@@ -688,6 +702,16 @@ static inline u32 armv8pmu_getreset_flags(void)
 	return value;
 }
 
+static void armv8pmu_disable_user_access(void)
+{
+	write_sysreg(0, pmuserenr_el0);
+}
+
+static void armv8pmu_enable_user_access(void)
+{
+	write_sysreg(ARMV8_PMU_USERENR_ER | ARMV8_PMU_USERENR_CR, pmuserenr_el0);
+}
+
 static void armv8pmu_enable_event(struct perf_event *event)
 {
 	/*
@@ -848,13 +872,16 @@ 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_can_chain(event))
 		return	armv8pmu_get_chain_idx(cpuc, cpu_pmu);
 	else
 		return armv8pmu_get_single_idx(cpuc, cpu_pmu);
@@ -865,11 +892,61 @@ static void armv8pmu_clear_event_idx(struct pmu_hw_events *cpuc,
 {
 	int idx = event->hw.idx;
 
+	armv8pmu_write_counter(event, 0);
+
 	clear_bit(idx, cpuc->used_mask);
 	if (armv8pmu_event_is_chained(event))
 		clear_bit(idx - 1, cpuc->used_mask);
 }
 
+static int armv8pmu_access_event_idx(struct perf_event *event)
+{
+	if (armv8pmu_event_can_chain(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;
+}
+
+void armv8pmu_sched_task(struct perf_event_context *ctx, bool sched_in)
+{
+	if (sched_in && to_arm_pmu(ctx->pmu)->attr_rdpmc && atomic_read(&ctx->nr_user))
+		armv8pmu_enable_user_access();
+	else
+		armv8pmu_disable_user_access();
+}
+
+static void armv8pmu_event_mapped(struct perf_event *event, struct mm_struct *mm)
+{
+	if (armv8pmu_event_can_chain(event) || (atomic_read(&event->mmap_count) != 1))
+		return;
+
+	if (atomic_inc_return(&event->ctx->nr_user) == 1) {
+		atomic_inc(&event->pmu->sched_cb_usage);
+		armv8pmu_enable_user_access();
+	}
+	perf_event_update_userpage(event);
+}
+
+static void armv8pmu_event_unmapped(struct perf_event *event, struct mm_struct *mm)
+{
+	if (armv8pmu_event_can_chain(event) || (atomic_read(&event->mmap_count) != 1))
+		return;
+
+	if (atomic_dec_and_test(&event->ctx->nr_user)) {
+		atomic_dec(&event->pmu->sched_cb_usage);
+		armv8pmu_disable_user_access();
+	}
+	perf_event_update_userpage(event);
+}
+
 /*
  * Add an event filter to a given event.
  */
@@ -963,7 +1040,16 @@ 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;
 
 	/* Only expose micro/arch events supported by this PMU */
@@ -1097,6 +1183,11 @@ 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->pmu.event_mapped	= armv8pmu_event_mapped;
+	cpu_pmu->pmu.event_unmapped	= armv8pmu_event_unmapped;
+	cpu_pmu->pmu.sched_task		= armv8pmu_sched_task;
+
 	cpu_pmu->name			= name;
 	cpu_pmu->map_event		= map_event;
 	cpu_pmu->attr_groups[ARMPMU_ATTR_GROUP_EVENTS] = events ?
@@ -1271,6 +1362,19 @@ 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 = !!userpg->index &&
+		(event->oncpu == smp_processor_id());
+
+	if (userpg->cap_user_rdpmc) {
+		struct arm_pmu *cpu_pmu = to_arm_pmu(event->pmu);
+
+		if (armv8pmu_event_is_64bit(event) &&
+		    (armv8pmu_has_long_event(cpu_pmu) ||
+		     (userpg->index == ARMV8_IDX_CYCLE_COUNTER_USER)))
+			userpg->pmc_width = 64;
+		else
+			userpg->pmc_width = 32;
+	}
 
 	do {
 		rd = sched_clock_read_begin(&seq);