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, 812 insertions, 0 deletions

diff --git a/drivers/macintosh/windfarm_pm81.c b/drivers/macintosh/windfarm_pm81.c
new file mode 100644
index 000000000..257fb2c69
--- /dev/null
+++ b/drivers/macintosh/windfarm_pm81.c
@@ -0,0 +1,812 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Windfarm PowerMac thermal control. iMac G5
+ *
+ * (c) Copyright 2005 Benjamin Herrenschmidt, IBM Corp.
+ *                    <benh@kernel.crashing.org>
+ *
+ * The algorithm used is the PID control algorithm, used the same
+ * way the published Darwin code does, using the same values that
+ * are present in the Darwin 8.2 snapshot property lists (note however
+ * that none of the code has been re-used, it's a complete re-implementation
+ *
+ * The various control loops found in Darwin config file are:
+ *
+ * PowerMac8,1 and PowerMac8,2
+ * ===========================
+ *
+ * System Fans control loop. Different based on models. In addition to the
+ * usual PID algorithm, the control loop gets 2 additional pairs of linear
+ * scaling factors (scale/offsets) expressed as 4.12 fixed point values
+ * signed offset, unsigned scale)
+ *
+ * The targets are modified such as:
+ *  - the linked control (second control) gets the target value as-is
+ *    (typically the drive fan)
+ *  - the main control (first control) gets the target value scaled with
+ *    the first pair of factors, and is then modified as below
+ *  - the value of the target of the CPU Fan control loop is retrieved,
+ *    scaled with the second pair of factors, and the max of that and
+ *    the scaled target is applied to the main control.
+ *
+ * # model_id: 2
+ *   controls       : system-fan, drive-bay-fan
+ *   sensors        : hd-temp
+ *   PID params     : G_d = 0x15400000
+ *                    G_p = 0x00200000
+ *                    G_r = 0x000002fd
+ *                    History = 2 entries
+ *                    Input target = 0x3a0000
+ *                    Interval = 5s
+ *   linear-factors : offset = 0xff38 scale  = 0x0ccd
+ *                    offset = 0x0208 scale  = 0x07ae
+ *
+ * # model_id: 3
+ *   controls       : system-fan, drive-bay-fan
+ *   sensors        : hd-temp
+ *   PID params     : G_d = 0x08e00000
+ *                    G_p = 0x00566666
+ *                    G_r = 0x0000072b
+ *                    History = 2 entries
+ *                    Input target = 0x350000
+ *                    Interval = 5s
+ *   linear-factors : offset = 0xff38 scale  = 0x0ccd
+ *                    offset = 0x0000 scale  = 0x0000
+ *
+ * # model_id: 5
+ *   controls       : system-fan
+ *   sensors        : hd-temp
+ *   PID params     : G_d = 0x15400000
+ *                    G_p = 0x00233333
+ *                    G_r = 0x000002fd
+ *                    History = 2 entries
+ *                    Input target = 0x3a0000
+ *                    Interval = 5s
+ *   linear-factors : offset = 0x0000 scale  = 0x1000
+ *                    offset = 0x0091 scale  = 0x0bae
+ *
+ * CPU Fan control loop. The loop is identical for all models. it
+ * has an additional pair of scaling factor. This is used to scale the
+ * systems fan control loop target result (the one before it gets scaled
+ * by the System Fans control loop itself). Then, the max value of the
+ * calculated target value and system fan value is sent to the fans
+ *
+ *   controls       : cpu-fan
+ *   sensors        : cpu-temp cpu-power
+ *   PID params     : From SMU sdb partition
+ *   linear-factors : offset = 0xfb50 scale  = 0x1000
+ *
+ * CPU Slew control loop. Not implemented. The cpufreq driver in linux is
+ * completely separate for now, though we could find a way to link it, either
+ * as a client reacting to overtemp notifications, or directling monitoring
+ * the CPU temperature
+ *
+ * WARNING ! The CPU control loop requires the CPU tmax for the current
+ * operating point. However, we currently are completely separated from
+ * the cpufreq driver and thus do not know what the current operating
+ * point is. Fortunately, we also do not have any hardware supporting anything
+ * but operating point 0 at the moment, thus we just peek that value directly
+ * from the SDB partition. If we ever end up with actually slewing the system
+ * clock and thus changing operating points, we'll have to find a way to
+ * communicate with the CPU freq driver;
+ */
+
+#include <linux/types.h>
+#include <linux/errno.h>
+#include <linux/kernel.h>
+#include <linux/delay.h>
+#include <linux/slab.h>
+#include <linux/init.h>
+#include <linux/spinlock.h>
+#include <linux/wait.h>
+#include <linux/kmod.h>
+#include <linux/device.h>
+#include <linux/platform_device.h>
+#include <linux/of.h>
+
+#include <asm/machdep.h>
+#include <asm/io.h>
+#include <asm/sections.h>
+#include <asm/smu.h>
+
+#include "windfarm.h"
+#include "windfarm_pid.h"
+
+#define VERSION "0.4"
+
+#undef DEBUG
+
+#ifdef DEBUG
+#define DBG(args...)	printk(args)
+#else
+#define DBG(args...)	do { } while(0)
+#endif
+
+/* define this to force CPU overtemp to 74 degree, useful for testing
+ * the overtemp code
+ */
+#undef HACKED_OVERTEMP
+
+static int wf_smu_mach_model;	/* machine model id */
+
+/* Controls & sensors */
+static struct wf_sensor	*sensor_cpu_power;
+static struct wf_sensor	*sensor_cpu_temp;
+static struct wf_sensor	*sensor_hd_temp;
+static struct wf_control *fan_cpu_main;
+static struct wf_control *fan_hd;
+static struct wf_control *fan_system;
+static struct wf_control *cpufreq_clamp;
+
+/* Set to kick the control loop into life */
+static int wf_smu_all_controls_ok, wf_smu_all_sensors_ok;
+static bool wf_smu_started;
+
+/* Failure handling.. could be nicer */
+#define FAILURE_FAN		0x01
+#define FAILURE_SENSOR		0x02
+#define FAILURE_OVERTEMP	0x04
+
+static unsigned int wf_smu_failure_state;
+static int wf_smu_readjust, wf_smu_skipping;
+static bool wf_smu_overtemp;
+
+/*
+ * ****** System Fans Control Loop ******
+ *
+ */
+
+/* Parameters for the System Fans control loop. Parameters
+ * not in this table such as interval, history size, ...
+ * are common to all versions and thus hard coded for now.
+ */
+struct wf_smu_sys_fans_param {
+	int	model_id;
+	s32	itarget;
+	s32	gd, gp, gr;
+
+	s16	offset0;
+	u16	scale0;
+	s16	offset1;
+	u16	scale1;
+};
+
+#define WF_SMU_SYS_FANS_INTERVAL	5
+#define WF_SMU_SYS_FANS_HISTORY_SIZE	2
+
+/* State data used by the system fans control loop
+ */
+struct wf_smu_sys_fans_state {
+	int			ticks;
+	s32			sys_setpoint;
+	s32			hd_setpoint;
+	s16			offset0;
+	u16			scale0;
+	s16			offset1;
+	u16			scale1;
+	struct wf_pid_state	pid;
+};
+
+/*
+ * Configs for SMU System Fan control loop
+ */
+static struct wf_smu_sys_fans_param wf_smu_sys_all_params[] = {
+	/* Model ID 2 */
+	{
+		.model_id	= 2,
+		.itarget	= 0x3a0000,
+		.gd		= 0x15400000,
+		.gp		= 0x00200000,
+		.gr		= 0x000002fd,
+		.offset0	= 0xff38,
+		.scale0		= 0x0ccd,
+		.offset1	= 0x0208,
+		.scale1		= 0x07ae,
+	},
+	/* Model ID 3 */
+	{
+		.model_id	= 3,
+		.itarget	= 0x350000,
+		.gd		= 0x08e00000,
+		.gp		= 0x00566666,
+		.gr		= 0x0000072b,
+		.offset0	= 0xff38,
+		.scale0		= 0x0ccd,
+		.offset1	= 0x0000,
+		.scale1		= 0x0000,
+	},
+	/* Model ID 5 */
+	{
+		.model_id	= 5,
+		.itarget	= 0x3a0000,
+		.gd		= 0x15400000,
+		.gp		= 0x00233333,
+		.gr		= 0x000002fd,
+		.offset0	= 0x0000,
+		.scale0		= 0x1000,
+		.offset1	= 0x0091,
+		.scale1		= 0x0bae,
+	},
+};
+#define WF_SMU_SYS_FANS_NUM_CONFIGS ARRAY_SIZE(wf_smu_sys_all_params)
+
+static struct wf_smu_sys_fans_state *wf_smu_sys_fans;
+
+/*
+ * ****** CPU Fans Control Loop ******
+ *
+ */
+
+
+#define WF_SMU_CPU_FANS_INTERVAL	1
+#define WF_SMU_CPU_FANS_MAX_HISTORY	16
+#define WF_SMU_CPU_FANS_SIBLING_SCALE	0x00001000
+#define WF_SMU_CPU_FANS_SIBLING_OFFSET	0xfffffb50
+
+/* State data used by the cpu fans control loop
+ */
+struct wf_smu_cpu_fans_state {
+	int			ticks;
+	s32			cpu_setpoint;
+	s32			scale;
+	s32			offset;
+	struct wf_cpu_pid_state	pid;
+};
+
+static struct wf_smu_cpu_fans_state *wf_smu_cpu_fans;
+
+
+
+/*
+ * ***** Implementation *****
+ *
+ */
+
+static void wf_smu_create_sys_fans(void)
+{
+	struct wf_smu_sys_fans_param *param = NULL;
+	struct wf_pid_param pid_param;
+	int i;
+
+	/* First, locate the params for this model */
+	for (i = 0; i < WF_SMU_SYS_FANS_NUM_CONFIGS; i++)
+		if (wf_smu_sys_all_params[i].model_id == wf_smu_mach_model) {
+			param = &wf_smu_sys_all_params[i];
+			break;
+		}
+
+	/* No params found, put fans to max */
+	if (param == NULL) {
+		printk(KERN_WARNING "windfarm: System fan config not found "
+		       "for this machine model, max fan speed\n");
+		goto fail;
+	}
+
+	/* Alloc & initialize state */
+	wf_smu_sys_fans = kmalloc(sizeof(struct wf_smu_sys_fans_state),
+				  GFP_KERNEL);
+	if (wf_smu_sys_fans == NULL) {
+		printk(KERN_WARNING "windfarm: Memory allocation error"
+		       " max fan speed\n");
+		goto fail;
+	}
+	wf_smu_sys_fans->ticks = 1;
+	wf_smu_sys_fans->scale0 = param->scale0;
+	wf_smu_sys_fans->offset0 = param->offset0;
+	wf_smu_sys_fans->scale1 = param->scale1;
+	wf_smu_sys_fans->offset1 = param->offset1;
+
+	/* Fill PID params */
+	pid_param.gd = param->gd;
+	pid_param.gp = param->gp;
+	pid_param.gr = param->gr;
+	pid_param.interval = WF_SMU_SYS_FANS_INTERVAL;
+	pid_param.history_len = WF_SMU_SYS_FANS_HISTORY_SIZE;
+	pid_param.itarget = param->itarget;
+	pid_param.min = wf_control_get_min(fan_system);
+	pid_param.max = wf_control_get_max(fan_system);
+	if (fan_hd) {
+		pid_param.min =
+			max(pid_param.min, wf_control_get_min(fan_hd));
+		pid_param.max =
+			min(pid_param.max, wf_control_get_max(fan_hd));
+	}
+	wf_pid_init(&wf_smu_sys_fans->pid, &pid_param);
+
+	DBG("wf: System Fan control initialized.\n");
+	DBG("    itarged=%d.%03d, min=%d RPM, max=%d RPM\n",
+	    FIX32TOPRINT(pid_param.itarget), pid_param.min, pid_param.max);
+	return;
+
+ fail:
+
+	if (fan_system)
+		wf_control_set_max(fan_system);
+	if (fan_hd)
+		wf_control_set_max(fan_hd);
+}
+
+static void wf_smu_sys_fans_tick(struct wf_smu_sys_fans_state *st)
+{
+	s32 new_setpoint, temp, scaled, cputarget;
+	int rc;
+
+	if (--st->ticks != 0) {
+		if (wf_smu_readjust)
+			goto readjust;
+		return;
+	}
+	st->ticks = WF_SMU_SYS_FANS_INTERVAL;
+
+	rc = wf_sensor_get(sensor_hd_temp, &temp);
+	if (rc) {
+		printk(KERN_WARNING "windfarm: HD temp sensor error %d\n",
+		       rc);
+		wf_smu_failure_state |= FAILURE_SENSOR;
+		return;
+	}
+
+	DBG("wf_smu: System Fans tick ! HD temp: %d.%03d\n",
+	    FIX32TOPRINT(temp));
+
+	if (temp > (st->pid.param.itarget + 0x50000))
+		wf_smu_failure_state |= FAILURE_OVERTEMP;
+
+	new_setpoint = wf_pid_run(&st->pid, temp);
+
+	DBG("wf_smu: new_setpoint: %d RPM\n", (int)new_setpoint);
+
+	scaled = ((((s64)new_setpoint) * (s64)st->scale0) >> 12) + st->offset0;
+
+	DBG("wf_smu: scaled setpoint: %d RPM\n", (int)scaled);
+
+	cputarget = wf_smu_cpu_fans ? wf_smu_cpu_fans->pid.target : 0;
+	cputarget = ((((s64)cputarget) * (s64)st->scale1) >> 12) + st->offset1;
+	scaled = max(scaled, cputarget);
+	scaled = max(scaled, st->pid.param.min);
+	scaled = min(scaled, st->pid.param.max);
+
+	DBG("wf_smu: adjusted setpoint: %d RPM\n", (int)scaled);
+
+	if (st->sys_setpoint == scaled && new_setpoint == st->hd_setpoint)
+		return;
+	st->sys_setpoint = scaled;
+	st->hd_setpoint = new_setpoint;
+ readjust:
+	if (fan_system && wf_smu_failure_state == 0) {
+		rc = wf_control_set(fan_system, st->sys_setpoint);
+		if (rc) {
+			printk(KERN_WARNING "windfarm: Sys fan error %d\n",
+			       rc);
+			wf_smu_failure_state |= FAILURE_FAN;
+		}
+	}
+	if (fan_hd && wf_smu_failure_state == 0) {
+		rc = wf_control_set(fan_hd, st->hd_setpoint);
+		if (rc) {
+			printk(KERN_WARNING "windfarm: HD fan error %d\n",
+			       rc);
+			wf_smu_failure_state |= FAILURE_FAN;
+		}
+	}
+}
+
+static void wf_smu_create_cpu_fans(void)
+{
+	struct wf_cpu_pid_param pid_param;
+	const struct smu_sdbp_header *hdr;
+	struct smu_sdbp_cpupiddata *piddata;
+	struct smu_sdbp_fvt *fvt;
+	s32 tmax, tdelta, maxpow, powadj;
+
+	/* First, locate the PID params in SMU SBD */
+	hdr = smu_get_sdb_partition(SMU_SDB_CPUPIDDATA_ID, NULL);
+	if (!hdr) {
+		printk(KERN_WARNING "windfarm: CPU PID fan config not found "
+		       "max fan speed\n");
+		goto fail;
+	}
+	piddata = (struct smu_sdbp_cpupiddata *)&hdr[1];
+
+	/* Get the FVT params for operating point 0 (the only supported one
+	 * for now) in order to get tmax
+	 */
+	hdr = smu_get_sdb_partition(SMU_SDB_FVT_ID, NULL);
+	if (hdr) {
+		fvt = (struct smu_sdbp_fvt *)&hdr[1];
+		tmax = ((s32)fvt->maxtemp) << 16;
+	} else
+		tmax = 0x5e0000; /* 94 degree default */
+
+	/* Alloc & initialize state */
+	wf_smu_cpu_fans = kmalloc(sizeof(struct wf_smu_cpu_fans_state),
+				  GFP_KERNEL);
+	if (wf_smu_cpu_fans == NULL)
+		goto fail;
+       	wf_smu_cpu_fans->ticks = 1;
+
+	wf_smu_cpu_fans->scale = WF_SMU_CPU_FANS_SIBLING_SCALE;
+	wf_smu_cpu_fans->offset = WF_SMU_CPU_FANS_SIBLING_OFFSET;
+
+	/* Fill PID params */
+	pid_param.interval = WF_SMU_CPU_FANS_INTERVAL;
+	pid_param.history_len = piddata->history_len;
+	if (pid_param.history_len > WF_CPU_PID_MAX_HISTORY) {
+		printk(KERN_WARNING "windfarm: History size overflow on "
+		       "CPU control loop (%d)\n", piddata->history_len);
+		pid_param.history_len = WF_CPU_PID_MAX_HISTORY;
+	}
+	pid_param.gd = piddata->gd;
+	pid_param.gp = piddata->gp;
+	pid_param.gr = piddata->gr / pid_param.history_len;
+
+	tdelta = ((s32)piddata->target_temp_delta) << 16;
+	maxpow = ((s32)piddata->max_power) << 16;
+	powadj = ((s32)piddata->power_adj) << 16;
+
+	pid_param.tmax = tmax;
+	pid_param.ttarget = tmax - tdelta;
+	pid_param.pmaxadj = maxpow - powadj;
+
+	pid_param.min = wf_control_get_min(fan_cpu_main);
+	pid_param.max = wf_control_get_max(fan_cpu_main);
+
+	wf_cpu_pid_init(&wf_smu_cpu_fans->pid, &pid_param);
+
+	DBG("wf: CPU Fan control initialized.\n");
+	DBG("    ttarget=%d.%03d, tmax=%d.%03d, min=%d RPM, max=%d RPM\n",
+	    FIX32TOPRINT(pid_param.ttarget), FIX32TOPRINT(pid_param.tmax),
+	    pid_param.min, pid_param.max);
+
+	return;
+
+ fail:
+	printk(KERN_WARNING "windfarm: CPU fan config not found\n"
+	       "for this machine model, max fan speed\n");
+
+	if (cpufreq_clamp)
+		wf_control_set_max(cpufreq_clamp);
+	if (fan_cpu_main)
+		wf_control_set_max(fan_cpu_main);
+}
+
+static void wf_smu_cpu_fans_tick(struct wf_smu_cpu_fans_state *st)
+{
+	s32 new_setpoint, temp, power, systarget;
+	int rc;
+
+	if (--st->ticks != 0) {
+		if (wf_smu_readjust)
+			goto readjust;
+		return;
+	}
+	st->ticks = WF_SMU_CPU_FANS_INTERVAL;
+
+	rc = wf_sensor_get(sensor_cpu_temp, &temp);
+	if (rc) {
+		printk(KERN_WARNING "windfarm: CPU temp sensor error %d\n",
+		       rc);
+		wf_smu_failure_state |= FAILURE_SENSOR;
+		return;
+	}
+
+	rc = wf_sensor_get(sensor_cpu_power, &power);
+	if (rc) {
+		printk(KERN_WARNING "windfarm: CPU power sensor error %d\n",
+		       rc);
+		wf_smu_failure_state |= FAILURE_SENSOR;
+		return;
+	}
+
+	DBG("wf_smu: CPU Fans tick ! CPU temp: %d.%03d, power: %d.%03d\n",
+	    FIX32TOPRINT(temp), FIX32TOPRINT(power));
+
+#ifdef HACKED_OVERTEMP
+	if (temp > 0x4a0000)
+		wf_smu_failure_state |= FAILURE_OVERTEMP;
+#else
+	if (temp > st->pid.param.tmax)
+		wf_smu_failure_state |= FAILURE_OVERTEMP;
+#endif
+	new_setpoint = wf_cpu_pid_run(&st->pid, power, temp);
+
+	DBG("wf_smu: new_setpoint: %d RPM\n", (int)new_setpoint);
+
+	systarget = wf_smu_sys_fans ? wf_smu_sys_fans->pid.target : 0;
+	systarget = ((((s64)systarget) * (s64)st->scale) >> 12)
+		+ st->offset;
+	new_setpoint = max(new_setpoint, systarget);
+	new_setpoint = max(new_setpoint, st->pid.param.min);
+	new_setpoint = min(new_setpoint, st->pid.param.max);
+
+	DBG("wf_smu: adjusted setpoint: %d RPM\n", (int)new_setpoint);
+
+	if (st->cpu_setpoint == new_setpoint)
+		return;
+	st->cpu_setpoint = new_setpoint;
+ readjust:
+	if (fan_cpu_main && wf_smu_failure_state == 0) {
+		rc = wf_control_set(fan_cpu_main, st->cpu_setpoint);
+		if (rc) {
+			printk(KERN_WARNING "windfarm: CPU main fan"
+			       " error %d\n", rc);
+			wf_smu_failure_state |= FAILURE_FAN;
+		}
+	}
+}
+
+/*
+ * ****** Setup / Init / Misc ... ******
+ *
+ */
+
+static void wf_smu_tick(void)
+{
+	unsigned int last_failure = wf_smu_failure_state;
+	unsigned int new_failure;
+
+	if (!wf_smu_started) {
+		DBG("wf: creating control loops !\n");
+		wf_smu_create_sys_fans();
+		wf_smu_create_cpu_fans();
+		wf_smu_started = true;
+	}
+
+	/* Skipping ticks */
+	if (wf_smu_skipping && --wf_smu_skipping)
+		return;
+
+	wf_smu_failure_state = 0;
+	if (wf_smu_sys_fans)
+		wf_smu_sys_fans_tick(wf_smu_sys_fans);
+	if (wf_smu_cpu_fans)
+		wf_smu_cpu_fans_tick(wf_smu_cpu_fans);
+
+	wf_smu_readjust = 0;
+	new_failure = wf_smu_failure_state & ~last_failure;
+
+	/* If entering failure mode, clamp cpufreq and ramp all
+	 * fans to full speed.
+	 */
+	if (wf_smu_failure_state && !last_failure) {
+		if (cpufreq_clamp)
+			wf_control_set_max(cpufreq_clamp);
+		if (fan_system)
+			wf_control_set_max(fan_system);
+		if (fan_cpu_main)
+			wf_control_set_max(fan_cpu_main);
+		if (fan_hd)
+			wf_control_set_max(fan_hd);
+	}
+
+	/* If leaving failure mode, unclamp cpufreq and readjust
+	 * all fans on next iteration
+	 */
+	if (!wf_smu_failure_state && last_failure) {
+		if (cpufreq_clamp)
+			wf_control_set_min(cpufreq_clamp);
+		wf_smu_readjust = 1;
+	}
+
+	/* Overtemp condition detected, notify and start skipping a couple
+	 * ticks to let the temperature go down
+	 */
+	if (new_failure & FAILURE_OVERTEMP) {
+		wf_set_overtemp();
+		wf_smu_skipping = 2;
+		wf_smu_overtemp = true;
+	}
+
+	/* We only clear the overtemp condition if overtemp is cleared
+	 * _and_ no other failure is present. Since a sensor error will
+	 * clear the overtemp condition (can't measure temperature) at
+	 * the control loop levels, but we don't want to keep it clear
+	 * here in this case
+	 */
+	if (!wf_smu_failure_state && wf_smu_overtemp) {
+		wf_clear_overtemp();
+		wf_smu_overtemp = false;
+	}
+}
+
+static void wf_smu_new_control(struct wf_control *ct)
+{
+	if (wf_smu_all_controls_ok)
+		return;
+
+	if (fan_cpu_main == NULL && !strcmp(ct->name, "cpu-fan")) {
+		if (wf_get_control(ct) == 0)
+			fan_cpu_main = ct;
+	}
+
+	if (fan_system == NULL && !strcmp(ct->name, "system-fan")) {
+		if (wf_get_control(ct) == 0)
+			fan_system = ct;
+	}
+
+	if (cpufreq_clamp == NULL && !strcmp(ct->name, "cpufreq-clamp")) {
+		if (wf_get_control(ct) == 0)
+			cpufreq_clamp = ct;
+	}
+
+	/* Darwin property list says the HD fan is only for model ID
+	 * 0, 1, 2 and 3
+	 */
+
+	if (wf_smu_mach_model > 3) {
+		if (fan_system && fan_cpu_main && cpufreq_clamp)
+			wf_smu_all_controls_ok = 1;
+		return;
+	}
+
+	if (fan_hd == NULL && !strcmp(ct->name, "drive-bay-fan")) {
+		if (wf_get_control(ct) == 0)
+			fan_hd = ct;
+	}
+
+	if (fan_system && fan_hd && fan_cpu_main && cpufreq_clamp)
+		wf_smu_all_controls_ok = 1;
+}
+
+static void wf_smu_new_sensor(struct wf_sensor *sr)
+{
+	if (wf_smu_all_sensors_ok)
+		return;
+
+	if (sensor_cpu_power == NULL && !strcmp(sr->name, "cpu-power")) {
+		if (wf_get_sensor(sr) == 0)
+			sensor_cpu_power = sr;
+	}
+
+	if (sensor_cpu_temp == NULL && !strcmp(sr->name, "cpu-temp")) {
+		if (wf_get_sensor(sr) == 0)
+			sensor_cpu_temp = sr;
+	}
+
+	if (sensor_hd_temp == NULL && !strcmp(sr->name, "hd-temp")) {
+		if (wf_get_sensor(sr) == 0)
+			sensor_hd_temp = sr;
+	}
+
+	if (sensor_cpu_power && sensor_cpu_temp && sensor_hd_temp)
+		wf_smu_all_sensors_ok = 1;
+}
+
+
+static int wf_smu_notify(struct notifier_block *self,
+			       unsigned long event, void *data)
+{
+	switch(event) {
+	case WF_EVENT_NEW_CONTROL:
+		DBG("wf: new control %s detected\n",
+		    ((struct wf_control *)data)->name);
+		wf_smu_new_control(data);
+		wf_smu_readjust = 1;
+		break;
+	case WF_EVENT_NEW_SENSOR:
+		DBG("wf: new sensor %s detected\n",
+		    ((struct wf_sensor *)data)->name);
+		wf_smu_new_sensor(data);
+		break;
+	case WF_EVENT_TICK:
+		if (wf_smu_all_controls_ok && wf_smu_all_sensors_ok)
+			wf_smu_tick();
+	}
+
+	return 0;
+}
+
+static struct notifier_block wf_smu_events = {
+	.notifier_call	= wf_smu_notify,
+};
+
+static int wf_init_pm(void)
+{
+	const struct smu_sdbp_header *hdr;
+
+	hdr = smu_get_sdb_partition(SMU_SDB_SENSORTREE_ID, NULL);
+	if (hdr) {
+		struct smu_sdbp_sensortree *st =
+			(struct smu_sdbp_sensortree *)&hdr[1];
+		wf_smu_mach_model = st->model_id;
+	}
+
+	printk(KERN_INFO "windfarm: Initializing for iMacG5 model ID %d\n",
+	       wf_smu_mach_model);
+
+	return 0;
+}
+
+static int wf_smu_probe(struct platform_device *ddev)
+{
+	wf_register_client(&wf_smu_events);
+
+	return 0;
+}
+
+static int wf_smu_remove(struct platform_device *ddev)
+{
+	wf_unregister_client(&wf_smu_events);
+
+	/* XXX We don't have yet a guarantee that our callback isn't
+	 * in progress when returning from wf_unregister_client, so
+	 * we add an arbitrary delay. I'll have to fix that in the core
+	 */
+	msleep(1000);
+
+	/* Release all sensors */
+	/* One more crappy race: I don't think we have any guarantee here
+	 * that the attribute callback won't race with the sensor beeing
+	 * disposed of, and I'm not 100% certain what best way to deal
+	 * with that except by adding locks all over... I'll do that
+	 * eventually but heh, who ever rmmod this module anyway ?
+	 */
+	if (sensor_cpu_power)
+		wf_put_sensor(sensor_cpu_power);
+	if (sensor_cpu_temp)
+		wf_put_sensor(sensor_cpu_temp);
+	if (sensor_hd_temp)
+		wf_put_sensor(sensor_hd_temp);
+
+	/* Release all controls */
+	if (fan_cpu_main)
+		wf_put_control(fan_cpu_main);
+	if (fan_hd)
+		wf_put_control(fan_hd);
+	if (fan_system)
+		wf_put_control(fan_system);
+	if (cpufreq_clamp)
+		wf_put_control(cpufreq_clamp);
+
+	/* Destroy control loops state structures */
+	kfree(wf_smu_sys_fans);
+	kfree(wf_smu_cpu_fans);
+
+	return 0;
+}
+
+static struct platform_driver wf_smu_driver = {
+        .probe = wf_smu_probe,
+        .remove = wf_smu_remove,
+	.driver = {
+		.name = "windfarm",
+	},
+};
+
+
+static int __init wf_smu_init(void)
+{
+	int rc = -ENODEV;
+
+	if (of_machine_is_compatible("PowerMac8,1") ||
+	    of_machine_is_compatible("PowerMac8,2"))
+		rc = wf_init_pm();
+
+	if (rc == 0) {
+#ifdef MODULE
+		request_module("windfarm_smu_controls");
+		request_module("windfarm_smu_sensors");
+		request_module("windfarm_lm75_sensor");
+		request_module("windfarm_cpufreq_clamp");
+
+#endif /* MODULE */
+		platform_driver_register(&wf_smu_driver);
+	}
+
+	return rc;
+}
+
+static void __exit wf_smu_exit(void)
+{
+
+	platform_driver_unregister(&wf_smu_driver);
+}
+
+
+module_init(wf_smu_init);
+module_exit(wf_smu_exit);
+
+MODULE_AUTHOR("Benjamin Herrenschmidt <benh@kernel.crashing.org>");
+MODULE_DESCRIPTION("Thermal control logic for iMac G5");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:windfarm");