Merge tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-nextgrafted

Pull networking updates from Jakub Kicinski:
 "Core:

   - Add dedicated kmem_cache for typical/small skb->head, avoid having
     to access struct page at kfree time, and improve memory use.

   - Introduce sysctl to set default RPS configuration for new netdevs.

   - Define Netlink protocol specification format which can be used to
     describe messages used by each family and auto-generate parsers.
     Add tools for generating kernel data structures and uAPI headers.

   - Expose all net/core sysctls inside netns.

   - Remove 4s sleep in netpoll if carrier is instantly detected on
     boot.

   - Add configurable limit of MDB entries per port, and port-vlan.

   - Continue populating drop reasons throughout the stack.

   - Retire a handful of legacy Qdiscs and classifiers.

  Protocols:

   - Support IPv4 big TCP (TSO frames larger than 64kB).

   - Add IP_LOCAL_PORT_RANGE socket option, to control local port range
     on socket by socket basis.

   - Track and report in procfs number of MPTCP sockets used.

   - Support mixing IPv4 and IPv6 flows in the in-kernel MPTCP path
     manager.

   - IPv6: don't check net.ipv6.route.max_size and rely on garbage
     collection to free memory (similarly to IPv4).

   - Support Penultimate Segment Pop (PSP) flavor in SRv6 (RFC8986).

   - ICMP: add per-rate limit counters.

   - Add support for user scanning requests in ieee802154.

   - Remove static WEP support.

   - Support minimal Wi-Fi 7 Extremely High Throughput (EHT) rate
     reporting.

   - WiFi 7 EHT channel puncturing support (client & AP).

  BPF:

   - Add a rbtree data structure following the "next-gen data structure"
     precedent set by recently added linked list, that is, by using
     kfunc + kptr instead of adding a new BPF map type.

   - Expose XDP hints via kfuncs with initial support for RX hash and
     timestamp metadata.

   - Add BPF_F_NO_TUNNEL_KEY extension to bpf_skb_set_tunnel_key to
     better support decap on GRE tunnel devices not operating in collect
     metadata.

   - Improve x86 JIT's codegen for PROBE_MEM runtime error checks.

   - Remove the need for trace_printk_lock for bpf_trace_printk and
     bpf_trace_vprintk helpers.

   - Extend libbpf's bpf_tracing.h support for tracing arguments of
     kprobes/uprobes and syscall as a special case.

   - Significantly reduce the search time for module symbols by
     livepatch and BPF.

   - Enable cpumasks to be used as kptrs, which is useful for tracing
     programs tracking which tasks end up running on which CPUs in
     different time intervals.

   - Add support for BPF trampoline on s390x and riscv64.

   - Add capability to export the XDP features supported by the NIC.

   - Add __bpf_kfunc tag for marking kernel functions as kfuncs.

   - Add cgroup.memory=nobpf kernel parameter option to disable BPF
     memory accounting for container environments.

  Netfilter:

   - Remove the CLUSTERIP target. It has been marked as obsolete for
     years, and we still have WARN splats wrt races of the out-of-band
     /proc interface installed by this target.

   - Add 'destroy' commands to nf_tables. They are identical to the
     existing 'delete' commands, but do not return an error if the
     referenced object (set, chain, rule...) did not exist.

  Driver API:

   - Improve cpumask_local_spread() locality to help NICs set the right
     IRQ affinity on AMD platforms.

   - Separate C22 and C45 MDIO bus transactions more clearly.

   - Introduce new DCB table to control DSCP rewrite on egress.

   - Support configuration of Physical Layer Collision Avoidance (PLCA)
     Reconciliation Sublayer (RS) (802.3cg-2019). Modern version of
     shared medium Ethernet.

   - Support for MAC Merge layer (IEEE 802.3-2018 clause 99). Allowing
     preemption of low priority frames by high priority frames.

   - Add support for controlling MACSec offload using netlink SET.

   - Rework devlink instance refcounts to allow registration and
     de-registration under the instance lock. Split the code into
     multiple files, drop some of the unnecessarily granular locks and
     factor out common parts of netlink operation handling.

   - Add TX frame aggregation parameters (for USB drivers).

   - Add a new attr TCA_EXT_WARN_MSG to report TC (offload) warning
     messages with notifications for debug.

   - Allow offloading of UDP NEW connections via act_ct.

   - Add support for per action HW stats in TC.

   - Support hardware miss to TC action (continue processing in SW from
     a specific point in the action chain).

   - Warn if old Wireless Extension user space interface is used with
     modern cfg80211/mac80211 drivers. Do not support Wireless
     Extensions for Wi-Fi 7 devices at all. Everyone should switch to
     using nl80211 interface instead.

   - Improve the CAN bit timing configuration. Use extack to return
     error messages directly to user space, update the SJW handling,
     including the definition of a new default value that will benefit
     CAN-FD controllers, by increasing their oscillator tolerance.

  New hardware / drivers:

   - Ethernet:
      - nVidia BlueField-3 support (control traffic driver)
      - Ethernet support for imx93 SoCs
      - Motorcomm yt8531 gigabit Ethernet PHY
      - onsemi NCN26000 10BASE-T1S PHY (with support for PLCA)
      - Microchip LAN8841 PHY (incl. cable diagnostics and PTP)
      - Amlogic gxl MDIO mux

   - WiFi:
      - RealTek RTL8188EU (rtl8xxxu)
      - Qualcomm Wi-Fi 7 devices (ath12k)

   - CAN:
      - Renesas R-Car V4H

  Drivers:

   - Bluetooth:
      - Set Per Platform Antenna Gain (PPAG) for Intel controllers.

   - Ethernet NICs:
      - Intel (1G, igc):
         - support TSN / Qbv / packet scheduling features of i226 model
      - Intel (100G, ice):
         - use GNSS subsystem instead of TTY
         - multi-buffer XDP support
         - extend support for GPIO pins to E823 devices
      - nVidia/Mellanox:
         - update the shared buffer configuration on PFC commands
         - implement PTP adjphase function for HW offset control
         - TC support for Geneve and GRE with VF tunnel offload
         - more efficient crypto key management method
         - multi-port eswitch support
      - Netronome/Corigine:
         - add DCB IEEE support
         - support IPsec offloading for NFP3800
      - Freescale/NXP (enetc):
         - support XDP_REDIRECT for XDP non-linear buffers
         - improve reconfig, avoid link flap and waiting for idle
         - support MAC Merge layer
      - Other NICs:
         - sfc/ef100: add basic devlink support for ef100
         - ionic: rx_push mode operation (writing descriptors via MMIO)
         - bnxt: use the auxiliary bus abstraction for RDMA
         - r8169: disable ASPM and reset bus in case of tx timeout
         - cpsw: support QSGMII mode for J721e CPSW9G
         - cpts: support pulse-per-second output
         - ngbe: add an mdio bus driver
         - usbnet: optimize usbnet_bh() by avoiding unnecessary queuing
         - r8152: handle devices with FW with NCM support
         - amd-xgbe: support 10Mbps, 2.5GbE speeds and rx-adaptation
         - virtio-net: support multi buffer XDP
         - virtio/vsock: replace virtio_vsock_pkt with sk_buff
         - tsnep: XDP support

   - Ethernet high-speed switches:
      - nVidia/Mellanox (mlxsw):
         - add support for latency TLV (in FW control messages)
      - Microchip (sparx5):
         - separate explicit and implicit traffic forwarding rules, make
           the implicit rules always active
         - add support for egress DSCP rewrite
         - IS0 VCAP support (Ingress Classification)
         - IS2 VCAP filters (protos, L3 addrs, L4 ports, flags, ToS
           etc.)
         - ES2 VCAP support (Egress Access Control)
         - support for Per-Stream Filtering and Policing (802.1Q,
           8.6.5.1)

   - Ethernet embedded switches:
      - Marvell (mv88e6xxx):
         - add MAB (port auth) offload support
         - enable PTP receive for mv88e6390
      - NXP (ocelot):
         - support MAC Merge layer
         - support for the the vsc7512 internal copper phys
      - Microchip:
         - lan9303: convert to PHYLINK
         - lan966x: support TC flower filter statistics
         - lan937x: PTP support for KSZ9563/KSZ8563 and LAN937x
         - lan937x: support Credit Based Shaper configuration
         - ksz9477: support Energy Efficient Ethernet
      - other:
         - qca8k: convert to regmap read/write API, use bulk operations
         - rswitch: Improve TX timestamp accuracy

   - Intel WiFi (iwlwifi):
      - EHT (Wi-Fi 7) rate reporting
      - STEP equalizer support: transfer some STEP (connection to radio
        on platforms with integrated wifi) related parameters from the
        BIOS to the firmware.

   - Qualcomm 802.11ax WiFi (ath11k):
      - IPQ5018 support
      - Fine Timing Measurement (FTM) responder role support
      - channel 177 support

   - MediaTek WiFi (mt76):
      - per-PHY LED support
      - mt7996: EHT (Wi-Fi 7) support
      - Wireless Ethernet Dispatch (WED) reset support
      - switch to using page pool allocator

   - RealTek WiFi (rtw89):
      - support new version of Bluetooth co-existance

   - Mobile:
      - rmnet: support TX aggregation"

* tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1872 commits)
  page_pool: add a comment explaining the fragment counter usage
  net: ethtool: fix __ethtool_dev_mm_supported() implementation
  ethtool: pse-pd: Fix double word in comments
  xsk: add linux/vmalloc.h to xsk.c
  sefltests: netdevsim: wait for devlink instance after netns removal
  selftest: fib_tests: Always cleanup before exit
  net/mlx5e: Align IPsec ASO result memory to be as required by hardware
  net/mlx5e: TC, Set CT miss to the specific ct action instance
  net/mlx5e: Rename CHAIN_TO_REG to MAPPED_OBJ_TO_REG
  net/mlx5: Refactor tc miss handling to a single function
  net/mlx5: Kconfig: Make tc offload depend on tc skb extension
  net/sched: flower: Support hardware miss to tc action
  net/sched: flower: Move filter handle initialization earlier
  net/sched: cls_api: Support hardware miss to tc action
  net/sched: Rename user cookie and act cookie
  sfc: fix builds without CONFIG_RTC_LIB
  sfc: clean up some inconsistent indentings
  net/mlx4_en: Introduce flexible array to silence overflow warning
  net: lan966x: Fix possible deadlock inside PTP
  net/ulp: Remove redundant ->clone() test in inet_clone_ulp().
  ...

1 files changed, 441 insertions, 0 deletions

diff --git a/Documentation/scheduler/sched-capacity.rst b/Documentation/scheduler/sched-capacity.rst
new file mode 100644
index 000000000..805f85f33
--- /dev/null
+++ b/Documentation/scheduler/sched-capacity.rst
@@ -0,0 +1,441 @@
+=========================
+Capacity Aware Scheduling
+=========================
+
+1. CPU Capacity
+===============
+
+1.1 Introduction
+----------------
+
+Conventional, homogeneous SMP platforms are composed of purely identical
+CPUs. Heterogeneous platforms on the other hand are composed of CPUs with
+different performance characteristics - on such platforms, not all CPUs can be
+considered equal.
+
+CPU capacity is a measure of the performance a CPU can reach, normalized against
+the most performant CPU in the system. Heterogeneous systems are also called
+asymmetric CPU capacity systems, as they contain CPUs of different capacities.
+
+Disparity in maximum attainable performance (IOW in maximum CPU capacity) stems
+from two factors:
+
+- not all CPUs may have the same microarchitecture (µarch).
+- with Dynamic Voltage and Frequency Scaling (DVFS), not all CPUs may be
+  physically able to attain the higher Operating Performance Points (OPP).
+
+Arm big.LITTLE systems are an example of both. The big CPUs are more
+performance-oriented than the LITTLE ones (more pipeline stages, bigger caches,
+smarter predictors, etc), and can usually reach higher OPPs than the LITTLE ones
+can.
+
+CPU performance is usually expressed in Millions of Instructions Per Second
+(MIPS), which can also be expressed as a given amount of instructions attainable
+per Hz, leading to::
+
+  capacity(cpu) = work_per_hz(cpu) * max_freq(cpu)
+
+1.2 Scheduler terms
+-------------------
+
+Two different capacity values are used within the scheduler. A CPU's
+``capacity_orig`` is its maximum attainable capacity, i.e. its maximum
+attainable performance level. A CPU's ``capacity`` is its ``capacity_orig`` to
+which some loss of available performance (e.g. time spent handling IRQs) is
+subtracted.
+
+Note that a CPU's ``capacity`` is solely intended to be used by the CFS class,
+while ``capacity_orig`` is class-agnostic. The rest of this document will use
+the term ``capacity`` interchangeably with ``capacity_orig`` for the sake of
+brevity.
+
+1.3 Platform examples
+---------------------
+
+1.3.1 Identical OPPs
+~~~~~~~~~~~~~~~~~~~~
+
+Consider an hypothetical dual-core asymmetric CPU capacity system where
+
+- work_per_hz(CPU0) = W
+- work_per_hz(CPU1) = W/2
+- all CPUs are running at the same fixed frequency
+
+By the above definition of capacity:
+
+- capacity(CPU0) = C
+- capacity(CPU1) = C/2
+
+To draw the parallel with Arm big.LITTLE, CPU0 would be a big while CPU1 would
+be a LITTLE.
+
+With a workload that periodically does a fixed amount of work, you will get an
+execution trace like so::
+
+ CPU0 work ^
+           |     ____                ____                ____
+           |    |    |              |    |              |    |
+           +----+----+----+----+----+----+----+----+----+----+-> time
+
+ CPU1 work ^
+           |     _________           _________           ____
+           |    |         |         |         |         |
+           +----+----+----+----+----+----+----+----+----+----+-> time
+
+CPU0 has the highest capacity in the system (C), and completes a fixed amount of
+work W in T units of time. On the other hand, CPU1 has half the capacity of
+CPU0, and thus only completes W/2 in T.
+
+1.3.2 Different max OPPs
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+Usually, CPUs of different capacity values also have different maximum
+OPPs. Consider the same CPUs as above (i.e. same work_per_hz()) with:
+
+- max_freq(CPU0) = F
+- max_freq(CPU1) = 2/3 * F
+
+This yields:
+
+- capacity(CPU0) = C
+- capacity(CPU1) = C/3
+
+Executing the same workload as described in 1.3.1, which each CPU running at its
+maximum frequency results in::
+
+ CPU0 work ^
+           |     ____                ____                ____
+           |    |    |              |    |              |    |
+           +----+----+----+----+----+----+----+----+----+----+-> time
+
+                            workload on CPU1
+ CPU1 work ^
+           |     ______________      ______________      ____
+           |    |              |    |              |    |
+           +----+----+----+----+----+----+----+----+----+----+-> time
+
+1.4 Representation caveat
+-------------------------
+
+It should be noted that having a *single* value to represent differences in CPU
+performance is somewhat of a contentious point. The relative performance
+difference between two different µarchs could be X% on integer operations, Y% on
+floating point operations, Z% on branches, and so on. Still, results using this
+simple approach have been satisfactory for now.
+
+2. Task utilization
+===================
+
+2.1 Introduction
+----------------
+
+Capacity aware scheduling requires an expression of a task's requirements with
+regards to CPU capacity. Each scheduler class can express this differently, and
+while task utilization is specific to CFS, it is convenient to describe it here
+in order to introduce more generic concepts.
+
+Task utilization is a percentage meant to represent the throughput requirements
+of a task. A simple approximation of it is the task's duty cycle, i.e.::
+
+  task_util(p) = duty_cycle(p)
+
+On an SMP system with fixed frequencies, 100% utilization suggests the task is a
+busy loop. Conversely, 10% utilization hints it is a small periodic task that
+spends more time sleeping than executing. Variable CPU frequencies and
+asymmetric CPU capacities complexify this somewhat; the following sections will
+expand on these.
+
+2.2 Frequency invariance
+------------------------
+
+One issue that needs to be taken into account is that a workload's duty cycle is
+directly impacted by the current OPP the CPU is running at. Consider running a
+periodic workload at a given frequency F::
+
+  CPU work ^
+           |     ____                ____                ____
+           |    |    |              |    |              |    |
+           +----+----+----+----+----+----+----+----+----+----+-> time
+
+This yields duty_cycle(p) == 25%.
+
+Now, consider running the *same* workload at frequency F/2::
+
+  CPU work ^
+           |     _________           _________           ____
+           |    |         |         |         |         |
+           +----+----+----+----+----+----+----+----+----+----+-> time
+
+This yields duty_cycle(p) == 50%, despite the task having the exact same
+behaviour (i.e. executing the same amount of work) in both executions.
+
+The task utilization signal can be made frequency invariant using the following
+formula::
+
+  task_util_freq_inv(p) = duty_cycle(p) * (curr_frequency(cpu) / max_frequency(cpu))
+
+Applying this formula to the two examples above yields a frequency invariant
+task utilization of 25%.
+
+2.3 CPU invariance
+------------------
+
+CPU capacity has a similar effect on task utilization in that running an
+identical workload on CPUs of different capacity values will yield different
+duty cycles.
+
+Consider the system described in 1.3.2., i.e.::
+
+- capacity(CPU0) = C
+- capacity(CPU1) = C/3
+
+Executing a given periodic workload on each CPU at their maximum frequency would
+result in::
+
+ CPU0 work ^
+           |     ____                ____                ____
+           |    |    |              |    |              |    |
+           +----+----+----+----+----+----+----+----+----+----+-> time
+
+ CPU1 work ^
+           |     ______________      ______________      ____
+           |    |              |    |              |    |
+           +----+----+----+----+----+----+----+----+----+----+-> time
+
+IOW,
+
+- duty_cycle(p) == 25% if p runs on CPU0 at its maximum frequency
+- duty_cycle(p) == 75% if p runs on CPU1 at its maximum frequency
+
+The task utilization signal can be made CPU invariant using the following
+formula::
+
+  task_util_cpu_inv(p) = duty_cycle(p) * (capacity(cpu) / max_capacity)
+
+with ``max_capacity`` being the highest CPU capacity value in the
+system. Applying this formula to the above example above yields a CPU
+invariant task utilization of 25%.
+
+2.4 Invariant task utilization
+------------------------------
+
+Both frequency and CPU invariance need to be applied to task utilization in
+order to obtain a truly invariant signal. The pseudo-formula for a task
+utilization that is both CPU and frequency invariant is thus, for a given
+task p::
+
+                                     curr_frequency(cpu)   capacity(cpu)
+  task_util_inv(p) = duty_cycle(p) * ------------------- * -------------
+                                     max_frequency(cpu)    max_capacity
+
+In other words, invariant task utilization describes the behaviour of a task as
+if it were running on the highest-capacity CPU in the system, running at its
+maximum frequency.
+
+Any mention of task utilization in the following sections will imply its
+invariant form.
+
+2.5 Utilization estimation
+--------------------------
+
+Without a crystal ball, task behaviour (and thus task utilization) cannot
+accurately be predicted the moment a task first becomes runnable. The CFS class
+maintains a handful of CPU and task signals based on the Per-Entity Load
+Tracking (PELT) mechanism, one of those yielding an *average* utilization (as
+opposed to instantaneous).
+
+This means that while the capacity aware scheduling criteria will be written
+considering a "true" task utilization (using a crystal ball), the implementation
+will only ever be able to use an estimator thereof.
+
+3. Capacity aware scheduling requirements
+=========================================
+
+3.1 CPU capacity
+----------------
+
+Linux cannot currently figure out CPU capacity on its own, this information thus
+needs to be handed to it. Architectures must define arch_scale_cpu_capacity()
+for that purpose.
+
+The arm and arm64 architectures directly map this to the arch_topology driver
+CPU scaling data, which is derived from the capacity-dmips-mhz CPU binding; see
+Documentation/devicetree/bindings/arm/cpu-capacity.txt.
+
+3.2 Frequency invariance
+------------------------
+
+As stated in 2.2, capacity-aware scheduling requires a frequency-invariant task
+utilization. Architectures must define arch_scale_freq_capacity(cpu) for that
+purpose.
+
+Implementing this function requires figuring out at which frequency each CPU
+have been running at. One way to implement this is to leverage hardware counters
+whose increment rate scale with a CPU's current frequency (APERF/MPERF on x86,
+AMU on arm64). Another is to directly hook into cpufreq frequency transitions,
+when the kernel is aware of the switched-to frequency (also employed by
+arm/arm64).
+
+4. Scheduler topology
+=====================
+
+During the construction of the sched domains, the scheduler will figure out
+whether the system exhibits asymmetric CPU capacities. Should that be the
+case:
+
+- The sched_asym_cpucapacity static key will be enabled.
+- The SD_ASYM_CPUCAPACITY_FULL flag will be set at the lowest sched_domain
+  level that spans all unique CPU capacity values.
+- The SD_ASYM_CPUCAPACITY flag will be set for any sched_domain that spans
+  CPUs with any range of asymmetry.
+
+The sched_asym_cpucapacity static key is intended to guard sections of code that
+cater to asymmetric CPU capacity systems. Do note however that said key is
+*system-wide*. Imagine the following setup using cpusets::
+
+  capacity    C/2          C
+            ________    ________
+           /        \  /        \
+  CPUs     0  1  2  3  4  5  6  7
+           \__/  \______________/
+  cpusets   cs0         cs1
+
+Which could be created via:
+
+.. code-block:: sh
+
+  mkdir /sys/fs/cgroup/cpuset/cs0
+  echo 0-1 > /sys/fs/cgroup/cpuset/cs0/cpuset.cpus
+  echo 0 > /sys/fs/cgroup/cpuset/cs0/cpuset.mems
+
+  mkdir /sys/fs/cgroup/cpuset/cs1
+  echo 2-7 > /sys/fs/cgroup/cpuset/cs1/cpuset.cpus
+  echo 0 > /sys/fs/cgroup/cpuset/cs1/cpuset.mems
+
+  echo 0 > /sys/fs/cgroup/cpuset/cpuset.sched_load_balance
+
+Since there *is* CPU capacity asymmetry in the system, the
+sched_asym_cpucapacity static key will be enabled. However, the sched_domain
+hierarchy of CPUs 0-1 spans a single capacity value: SD_ASYM_CPUCAPACITY isn't
+set in that hierarchy, it describes an SMP island and should be treated as such.
+
+Therefore, the 'canonical' pattern for protecting codepaths that cater to
+asymmetric CPU capacities is to:
+
+- Check the sched_asym_cpucapacity static key
+- If it is enabled, then also check for the presence of SD_ASYM_CPUCAPACITY in
+  the sched_domain hierarchy (if relevant, i.e. the codepath targets a specific
+  CPU or group thereof)
+
+5. Capacity aware scheduling implementation
+===========================================
+
+5.1 CFS
+-------
+
+5.1.1 Capacity fitness
+~~~~~~~~~~~~~~~~~~~~~~
+
+The main capacity scheduling criterion of CFS is::
+
+  task_util(p) < capacity(task_cpu(p))
+
+This is commonly called the capacity fitness criterion, i.e. CFS must ensure a
+task "fits" on its CPU. If it is violated, the task will need to achieve more
+work than what its CPU can provide: it will be CPU-bound.
+
+Furthermore, uclamp lets userspace specify a minimum and a maximum utilization
+value for a task, either via sched_setattr() or via the cgroup interface (see
+Documentation/admin-guide/cgroup-v2.rst). As its name imply, this can be used to
+clamp task_util() in the previous criterion.
+
+5.1.2 Wakeup CPU selection
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+CFS task wakeup CPU selection follows the capacity fitness criterion described
+above. On top of that, uclamp is used to clamp the task utilization values,
+which lets userspace have more leverage over the CPU selection of CFS
+tasks. IOW, CFS wakeup CPU selection searches for a CPU that satisfies::
+
+  clamp(task_util(p), task_uclamp_min(p), task_uclamp_max(p)) < capacity(cpu)
+
+By using uclamp, userspace can e.g. allow a busy loop (100% utilization) to run
+on any CPU by giving it a low uclamp.max value. Conversely, it can force a small
+periodic task (e.g. 10% utilization) to run on the highest-performance CPUs by
+giving it a high uclamp.min value.
+
+.. note::
+
+  Wakeup CPU selection in CFS can be eclipsed by Energy Aware Scheduling
+  (EAS), which is described in Documentation/scheduler/sched-energy.rst.
+
+5.1.3 Load balancing
+~~~~~~~~~~~~~~~~~~~~
+
+A pathological case in the wakeup CPU selection occurs when a task rarely
+sleeps, if at all - it thus rarely wakes up, if at all. Consider::
+
+  w == wakeup event
+
+  capacity(CPU0) = C
+  capacity(CPU1) = C / 3
+
+                           workload on CPU0
+  CPU work ^
+           |     _________           _________           ____
+           |    |         |         |         |         |
+           +----+----+----+----+----+----+----+----+----+----+-> time
+                w                   w                   w
+
+                           workload on CPU1
+  CPU work ^
+           |     ____________________________________________
+           |    |
+           +----+----+----+----+----+----+----+----+----+----+->
+                w
+
+This workload should run on CPU0, but if the task either:
+
+- was improperly scheduled from the start (inaccurate initial
+  utilization estimation)
+- was properly scheduled from the start, but suddenly needs more
+  processing power
+
+then it might become CPU-bound, IOW ``task_util(p) > capacity(task_cpu(p))``;
+the CPU capacity scheduling criterion is violated, and there may not be any more
+wakeup event to fix this up via wakeup CPU selection.
+
+Tasks that are in this situation are dubbed "misfit" tasks, and the mechanism
+put in place to handle this shares the same name. Misfit task migration
+leverages the CFS load balancer, more specifically the active load balance part
+(which caters to migrating currently running tasks). When load balance happens,
+a misfit active load balance will be triggered if a misfit task can be migrated
+to a CPU with more capacity than its current one.
+
+5.2 RT
+------
+
+5.2.1 Wakeup CPU selection
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+RT task wakeup CPU selection searches for a CPU that satisfies::
+
+  task_uclamp_min(p) <= capacity(task_cpu(cpu))
+
+while still following the usual priority constraints. If none of the candidate
+CPUs can satisfy this capacity criterion, then strict priority based scheduling
+is followed and CPU capacities are ignored.
+
+5.3 DL
+------
+
+5.3.1 Wakeup CPU selection
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+DL task wakeup CPU selection searches for a CPU that satisfies::
+
+  task_bandwidth(p) < capacity(task_cpu(p))
+
+while still respecting the usual bandwidth and deadline constraints. If
+none of the candidate CPUs can satisfy this capacity criterion, then the
+task will remain on its current CPU.