[PATCH v2 0/2] perf: Add Raspberry Pi AXI PMU driver

[Ian Rogers <irogers@google.com>]

This patch series adds an uncore Performance Monitoring Unit (PMU) driver
for Broadcom AXI system and VideoCore VPU performance monitors found on
Raspberry Pi SoCs (BCM2835 through BCM2712 / Raspberry Pi 1 through 5).

Motivation & Background
------------------------
Currently, Linux lacks a standard perf-API compatible driver for the
Broadcom AXI performance counter blocks on Raspberry Pi platforms. Prior
out-of-tree vendor solutions relied on custom debugfs nodes and ad-hoc
kthreads, preventing integration with standard Linux perf tooling (`perf stat`,
`perf list`, etc.).

This driver implements standard `struct pmu` hardware uncore callbacks
under `drivers/perf/`, exposing human-readable sysfs event aliases, unit
scaling (`Bytes`), and bus filtering directly to user space.

Key Architectural Improvements & Features
------------------------------------------
1. Standard Linux Perf Integration:
   - Exposes uncore AXI interconnect events via `/sys/bus/event_source/devices/rpi_axi_pmu/`.
   - Supports event sampling and hardware counter accumulation (`local64_add`),
     automatically managing 31-bit hardware counter wraparound across high-bandwidth
     interconnect transfers.

2. CPU Hotplug Support (`cpuhp`):
   - Registers dynamic CPU hotplug notifiers (`CPUHP_AP_ONLINE_DYN`).
   - Automatically migrates PMU context (`perf_pmu_migrate_context`) to an online
     CPU core when a designated CPU goes offline, avoiding stale uncore state.

3. Hybrid Memory-Mapped & Mailbox Work Queue Architecture:
   - System Monitor (MMIO): Performs fast atomic-safe memory reads (~15ns)
     directly mapped over ARM physical memory space (`MON__SYSTEM`).
   - VPU Monitor (Mailbox IPC): For Broadcom BCM2835-BCM2711 platforms (RPi 1-4),
     VideoCore VPU monitor IPC calls are offloaded to process context via a dedicated
     workqueue (`vpu_work`) and serialized under `vpu_mutex`. This avoids atomic
     sleeps or blocking in timer/interrupt context.

4. SoC Generation Support:
   - Patch 1 adds core driver support for Broadcom BCM2835-BCM2711 (RPi 1-4).
   - Patch 2 expands support for Broadcom BCM2712 (Raspberry Pi 5), adding PCIe RP1
     Southbridge links, HEVC decoder, HVS display engine, and Cortex-A76 DSU L3
     interconnect monitoring.

Hardware Validation
-------------------
The driver has been validated on real hardware across multiple SoC generations:
- Raspberry Pi 400 (BCM2711): Validated System L2, ARM CPU, and VideoCore VPU
  firmware mailbox IPC performance counters.
- Raspberry Pi 5 (BCM2712): Validated live byte throughput across HVS display
  refresh cycles, Cortex-A76 DSU L3 interconnect memory traffic, and PCIe RP1
  Southbridge transfers.

Changes in v2
-------------
- Match Data Initialization: In rpi_axi_pmu__init(), set pmu->chip via
  of_device_get_match_data(dev) to ensure BCM2712 match data is correctly parsed.
- Sysfs Event Visibility Filtering: Updated rpi_axi_pmu_events_is_visible() to
  properly filter all 7 BCM2712 master filtered event aliases (arm_, pcie0_,
  hvs_filtered, emmc0_) on BCM2835-BCM2711 systems, and hide legacy RPi 1-4
  aliases on BCM2712 systems.
- Unmapped Monitor Check: Added monitor base_address and initialization check to
  config_is_valid(). Events targeting uninitialized or unmapped monitors (e.g. missing
  memory resource or monitor=1 on BCM2712) are properly rejected with -EINVAL.
- CPU Affinity Enforcement: Restrict uncore events in event_init() to event->cpu == pmu->cpu
  to prevent orphaned uncore events when a non-PMU CPU goes offline.
- 31-bit Counter Wrap Delta: Mask counter accumulation with delta = (new_count - prev_count) & 0x7FFFFFFF
  in read(), timer_handler(), and vpu_work_handler() to prevent 2B+ event spikes on counter overflow.
- Stopped Event Guard: Return early in rpi_axi_pmu_read() if event->hw.state & PERF_HES_STOPPED.
- Lockless Race Fix: Pass bus and filter explicitly to rpi_axi_pmu_enable_bus_watcher()
  under spinlock to prevent lockless read races in vpu_work_handler().
- Slot Reuse Clean: Set enabled[idx] = false in rpi_axi_pmu_del() when refcount[idx] == 0 so
  re-allocated watcher slots properly enable hardware watchers.
- Events Array Bounds: Return -ENOSPC in rpi_axi_pmu_add() if pmu->events array is full.
- Timer Cancel Race Fix: Re-acquire pmu->lock after hrtimer_cancel() in rpi_axi_pmu_del()
  and re-verify active_events > 0 before returning.
- Hotplug Timer Migration: Cancel and re-pin pmu->hrtimer to target CPU in rpi_axi_pmu_offline_cpu().
- Registration Order: Call perf_pmu_register() before cpuhp_state_add_instance() in probe,
  and cancel hrtimer before cancel_work_sync() and unregister in exit teardown.
- Header & Style Cleanups: Removed <linux/mod_devicetable.h> and <linux/version.h>,
  updated rpi_axi_pmu_match[] terminator to { }, and fixed C variable declaration style.

Ian Rogers (2):
  perf: Add Raspberry Pi BCM2835 AXI PMU driver
  perf: Add Raspberry Pi 5 (BCM2712) AXI PMU support

 drivers/perf/Kconfig       |   10 +
 drivers/perf/Makefile      |    1 +
 drivers/perf/rpi_axi_pmu.c | 2001 ++++++++++++++++++++++++++++++++++++
 3 files changed, 2012 insertions(+)
 create mode 100644 drivers/perf/rpi_axi_pmu.c

-- 
2.55.0.679.g6767b8d81c-goog