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, 841 insertions, 0 deletions

diff --git a/arch/sparc/kernel/pcic.c b/arch/sparc/kernel/pcic.c
new file mode 100644
index 000000000..ee4c9a9a1
--- /dev/null
+++ b/arch/sparc/kernel/pcic.c
@@ -0,0 +1,841 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * pcic.c: MicroSPARC-IIep PCI controller support
+ *
+ * Copyright (C) 1998 V. Roganov and G. Raiko
+ *
+ * Code is derived from Ultra/PCI PSYCHO controller support, see that
+ * for author info.
+ *
+ * Support for diverse IIep based platforms by Pete Zaitcev.
+ * CP-1200 by Eric Brower.
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/mm.h>
+#include <linux/slab.h>
+#include <linux/jiffies.h>
+
+#include <asm/swift.h> /* for cache flushing. */
+#include <asm/io.h>
+
+#include <linux/ctype.h>
+#include <linux/pci.h>
+#include <linux/time.h>
+#include <linux/timex.h>
+#include <linux/interrupt.h>
+#include <linux/export.h>
+
+#include <asm/irq.h>
+#include <asm/oplib.h>
+#include <asm/prom.h>
+#include <asm/pcic.h>
+#include <asm/timex.h>
+#include <asm/timer.h>
+#include <linux/uaccess.h>
+#include <asm/irq_regs.h>
+
+#include "kernel.h"
+#include "irq.h"
+
+/*
+ * I studied different documents and many live PROMs both from 2.30
+ * family and 3.xx versions. I came to the amazing conclusion: there is
+ * absolutely no way to route interrupts in IIep systems relying on
+ * information which PROM presents. We must hardcode interrupt routing
+ * schematics. And this actually sucks.   -- zaitcev 1999/05/12
+ *
+ * To find irq for a device we determine which routing map
+ * is in effect or, in other words, on which machine we are running.
+ * We use PROM name for this although other techniques may be used
+ * in special cases (Gleb reports a PROMless IIep based system).
+ * Once we know the map we take device configuration address and
+ * find PCIC pin number where INT line goes. Then we may either program
+ * preferred irq into the PCIC or supply the preexisting irq to the device.
+ */
+struct pcic_ca2irq {
+	unsigned char busno;		/* PCI bus number */
+	unsigned char devfn;		/* Configuration address */
+	unsigned char pin;		/* PCIC external interrupt pin */
+	unsigned char irq;		/* Preferred IRQ (mappable in PCIC) */
+	unsigned int force;		/* Enforce preferred IRQ */
+};
+
+struct pcic_sn2list {
+	char *sysname;
+	struct pcic_ca2irq *intmap;
+	int mapdim;
+};
+
+/*
+ * JavaEngine-1 apparently has different versions.
+ *
+ * According to communications with Sun folks, for P2 build 501-4628-03:
+ * pin 0 - parallel, audio;
+ * pin 1 - Ethernet;
+ * pin 2 - su;
+ * pin 3 - PS/2 kbd and mouse.
+ *
+ * OEM manual (805-1486):
+ * pin 0: Ethernet
+ * pin 1: All EBus
+ * pin 2: IGA (unused)
+ * pin 3: Not connected
+ * OEM manual says that 501-4628 & 501-4811 are the same thing,
+ * only the latter has NAND flash in place.
+ *
+ * So far unofficial Sun wins over the OEM manual. Poor OEMs...
+ */
+static struct pcic_ca2irq pcic_i_je1a[] = {	/* 501-4811-03 */
+	{ 0, 0x00, 2, 12, 0 },		/* EBus: hogs all */
+	{ 0, 0x01, 1,  6, 1 },		/* Happy Meal */
+	{ 0, 0x80, 0,  7, 0 },		/* IGA (unused) */
+};
+
+/* XXX JS-E entry is incomplete - PCI Slot 2 address (pin 7)? */
+static struct pcic_ca2irq pcic_i_jse[] = {
+	{ 0, 0x00, 0, 13, 0 },		/* Ebus - serial and keyboard */
+	{ 0, 0x01, 1,  6, 0 },		/* hme */
+	{ 0, 0x08, 2,  9, 0 },		/* VGA - we hope not used :) */
+	{ 0, 0x10, 6,  8, 0 },		/* PCI INTA# in Slot 1 */
+	{ 0, 0x18, 7, 12, 0 },		/* PCI INTA# in Slot 2, shared w. RTC */
+	{ 0, 0x38, 4,  9, 0 },		/* All ISA devices. Read 8259. */
+	{ 0, 0x80, 5, 11, 0 },		/* EIDE */
+	/* {0,0x88, 0,0,0} - unknown device... PMU? Probably no interrupt. */
+	{ 0, 0xA0, 4,  9, 0 },		/* USB */
+	/*
+	 * Some pins belong to non-PCI devices, we hardcode them in drivers.
+	 * sun4m timers - irq 10, 14
+	 * PC style RTC - pin 7, irq 4 ?
+	 * Smart card, Parallel - pin 4 shared with USB, ISA
+	 * audio - pin 3, irq 5 ?
+	 */
+};
+
+/* SPARCengine-6 was the original release name of CP1200.
+ * The documentation differs between the two versions
+ */
+static struct pcic_ca2irq pcic_i_se6[] = {
+	{ 0, 0x08, 0,  2, 0 },		/* SCSI	*/
+	{ 0, 0x01, 1,  6, 0 },		/* HME	*/
+	{ 0, 0x00, 3, 13, 0 },		/* EBus	*/
+};
+
+/*
+ * Krups (courtesy of Varol Kaptan)
+ * No documentation available, but it was easy to guess
+ * because it was very similar to Espresso.
+ *  
+ * pin 0 - kbd, mouse, serial;
+ * pin 1 - Ethernet;
+ * pin 2 - igs (we do not use it);
+ * pin 3 - audio;
+ * pin 4,5,6 - unused;
+ * pin 7 - RTC (from P2 onwards as David B. says).
+ */
+static struct pcic_ca2irq pcic_i_jk[] = {
+	{ 0, 0x00, 0, 13, 0 },		/* Ebus - serial and keyboard */
+	{ 0, 0x01, 1,  6, 0 },		/* hme */
+};
+
+/*
+ * Several entries in this list may point to the same routing map
+ * as several PROMs may be installed on the same physical board.
+ */
+#define SN2L_INIT(name, map)	\
+  { name, map, ARRAY_SIZE(map) }
+
+static struct pcic_sn2list pcic_known_sysnames[] = {
+	SN2L_INIT("SUNW,JavaEngine1", pcic_i_je1a),	/* JE1, PROM 2.32 */
+	SN2L_INIT("SUNW,JS-E", pcic_i_jse),	/* PROLL JavaStation-E */
+	SN2L_INIT("SUNW,SPARCengine-6", pcic_i_se6), /* SPARCengine-6/CP-1200 */
+	SN2L_INIT("SUNW,JS-NC", pcic_i_jk),	/* PROLL JavaStation-NC */
+	SN2L_INIT("SUNW,JSIIep", pcic_i_jk),	/* OBP JavaStation-NC */
+	{ NULL, NULL, 0 }
+};
+
+/*
+ * Only one PCIC per IIep,
+ * and since we have no SMP IIep, only one per system.
+ */
+static int pcic0_up;
+static struct linux_pcic pcic0;
+
+void __iomem *pcic_regs;
+static volatile int pcic_speculative;
+static volatile int pcic_trapped;
+
+/* forward */
+unsigned int pcic_build_device_irq(struct platform_device *op,
+                                   unsigned int real_irq);
+
+#define CONFIG_CMD(bus, device_fn, where) (0x80000000 | (((unsigned int)bus) << 16) | (((unsigned int)device_fn) << 8) | (where & ~3))
+
+static int pcic_read_config_dword(unsigned int busno, unsigned int devfn,
+    int where, u32 *value)
+{
+	struct linux_pcic *pcic;
+	unsigned long flags;
+
+	pcic = &pcic0;
+
+	local_irq_save(flags);
+#if 0 /* does not fail here */
+	pcic_speculative = 1;
+	pcic_trapped = 0;
+#endif
+	writel(CONFIG_CMD(busno, devfn, where), pcic->pcic_config_space_addr);
+#if 0 /* does not fail here */
+	nop();
+	if (pcic_trapped) {
+		local_irq_restore(flags);
+		*value = ~0;
+		return 0;
+	}
+#endif
+	pcic_speculative = 2;
+	pcic_trapped = 0;
+	*value = readl(pcic->pcic_config_space_data + (where&4));
+	nop();
+	if (pcic_trapped) {
+		pcic_speculative = 0;
+		local_irq_restore(flags);
+		*value = ~0;
+		return 0;
+	}
+	pcic_speculative = 0;
+	local_irq_restore(flags);
+	return 0;
+}
+
+static int pcic_read_config(struct pci_bus *bus, unsigned int devfn,
+   int where, int size, u32 *val)
+{
+	unsigned int v;
+
+	if (bus->number != 0) return -EINVAL;
+	switch (size) {
+	case 1:
+		pcic_read_config_dword(bus->number, devfn, where&~3, &v);
+		*val = 0xff & (v >> (8*(where & 3)));
+		return 0;
+	case 2:
+		if (where&1) return -EINVAL;
+		pcic_read_config_dword(bus->number, devfn, where&~3, &v);
+		*val = 0xffff & (v >> (8*(where & 3)));
+		return 0;
+	case 4:
+		if (where&3) return -EINVAL;
+		pcic_read_config_dword(bus->number, devfn, where&~3, val);
+		return 0;
+	}
+	return -EINVAL;
+}
+
+static int pcic_write_config_dword(unsigned int busno, unsigned int devfn,
+    int where, u32 value)
+{
+	struct linux_pcic *pcic;
+	unsigned long flags;
+
+	pcic = &pcic0;
+
+	local_irq_save(flags);
+	writel(CONFIG_CMD(busno, devfn, where), pcic->pcic_config_space_addr);
+	writel(value, pcic->pcic_config_space_data + (where&4));
+	local_irq_restore(flags);
+	return 0;
+}
+
+static int pcic_write_config(struct pci_bus *bus, unsigned int devfn,
+   int where, int size, u32 val)
+{
+	unsigned int v;
+
+	if (bus->number != 0) return -EINVAL;
+	switch (size) {
+	case 1:
+		pcic_read_config_dword(bus->number, devfn, where&~3, &v);
+		v = (v & ~(0xff << (8*(where&3)))) |
+		    ((0xff&val) << (8*(where&3)));
+		return pcic_write_config_dword(bus->number, devfn, where&~3, v);
+	case 2:
+		if (where&1) return -EINVAL;
+		pcic_read_config_dword(bus->number, devfn, where&~3, &v);
+		v = (v & ~(0xffff << (8*(where&3)))) |
+		    ((0xffff&val) << (8*(where&3)));
+		return pcic_write_config_dword(bus->number, devfn, where&~3, v);
+	case 4:
+		if (where&3) return -EINVAL;
+		return pcic_write_config_dword(bus->number, devfn, where, val);
+	}
+	return -EINVAL;
+}
+
+static struct pci_ops pcic_ops = {
+	.read =		pcic_read_config,
+	.write =	pcic_write_config,
+};
+
+/*
+ * On sparc64 pcibios_init() calls pci_controller_probe().
+ * We want PCIC probed little ahead so that interrupt controller
+ * would be operational.
+ */
+int __init pcic_probe(void)
+{
+	struct linux_pcic *pcic;
+	struct linux_prom_registers regs[PROMREG_MAX];
+	struct linux_pbm_info* pbm;
+	char namebuf[64];
+	phandle node;
+	int err;
+
+	if (pcic0_up) {
+		prom_printf("PCIC: called twice!\n");
+		prom_halt();
+	}
+	pcic = &pcic0;
+
+	node = prom_getchild (prom_root_node);
+	node = prom_searchsiblings (node, "pci");
+	if (node == 0)
+		return -ENODEV;
+	/*
+	 * Map in PCIC register set, config space, and IO base
+	 */
+	err = prom_getproperty(node, "reg", (char*)regs, sizeof(regs));
+	if (err == 0 || err == -1) {
+		prom_printf("PCIC: Error, cannot get PCIC registers "
+			    "from PROM.\n");
+		prom_halt();
+	}
+
+	pcic0_up = 1;
+
+	pcic->pcic_res_regs.name = "pcic_registers";
+	pcic->pcic_regs = ioremap(regs[0].phys_addr, regs[0].reg_size);
+	if (!pcic->pcic_regs) {
+		prom_printf("PCIC: Error, cannot map PCIC registers.\n");
+		prom_halt();
+	}
+
+	pcic->pcic_res_io.name = "pcic_io";
+	if ((pcic->pcic_io = (unsigned long)
+	    ioremap(regs[1].phys_addr, 0x10000)) == 0) {
+		prom_printf("PCIC: Error, cannot map PCIC IO Base.\n");
+		prom_halt();
+	}
+
+	pcic->pcic_res_cfg_addr.name = "pcic_cfg_addr";
+	if ((pcic->pcic_config_space_addr =
+	    ioremap(regs[2].phys_addr, regs[2].reg_size * 2)) == NULL) {
+		prom_printf("PCIC: Error, cannot map "
+			    "PCI Configuration Space Address.\n");
+		prom_halt();
+	}
+
+	/*
+	 * Docs say three least significant bits in address and data
+	 * must be the same. Thus, we need adjust size of data.
+	 */
+	pcic->pcic_res_cfg_data.name = "pcic_cfg_data";
+	if ((pcic->pcic_config_space_data =
+	    ioremap(regs[3].phys_addr, regs[3].reg_size * 2)) == NULL) {
+		prom_printf("PCIC: Error, cannot map "
+			    "PCI Configuration Space Data.\n");
+		prom_halt();
+	}
+
+	pbm = &pcic->pbm;
+	pbm->prom_node = node;
+	prom_getstring(node, "name", namebuf, 63);  namebuf[63] = 0;
+	strcpy(pbm->prom_name, namebuf);
+
+	{
+		extern int pcic_nmi_trap_patch[4];
+
+		t_nmi[0] = pcic_nmi_trap_patch[0];
+		t_nmi[1] = pcic_nmi_trap_patch[1];
+		t_nmi[2] = pcic_nmi_trap_patch[2];
+		t_nmi[3] = pcic_nmi_trap_patch[3];
+		swift_flush_dcache();
+		pcic_regs = pcic->pcic_regs;
+	}
+
+	prom_getstring(prom_root_node, "name", namebuf, 63);  namebuf[63] = 0;
+	{
+		struct pcic_sn2list *p;
+
+		for (p = pcic_known_sysnames; p->sysname != NULL; p++) {
+			if (strcmp(namebuf, p->sysname) == 0)
+				break;
+		}
+		pcic->pcic_imap = p->intmap;
+		pcic->pcic_imdim = p->mapdim;
+	}
+	if (pcic->pcic_imap == NULL) {
+		/*
+		 * We do not panic here for the sake of embedded systems.
+		 */
+		printk("PCIC: System %s is unknown, cannot route interrupts\n",
+		    namebuf);
+	}
+
+	return 0;
+}
+
+static void __init pcic_pbm_scan_bus(struct linux_pcic *pcic)
+{
+	struct linux_pbm_info *pbm = &pcic->pbm;
+
+	pbm->pci_bus = pci_scan_bus(pbm->pci_first_busno, &pcic_ops, pbm);
+	if (!pbm->pci_bus)
+		return;
+
+#if 0 /* deadwood transplanted from sparc64 */
+	pci_fill_in_pbm_cookies(pbm->pci_bus, pbm, pbm->prom_node);
+	pci_record_assignments(pbm, pbm->pci_bus);
+	pci_assign_unassigned(pbm, pbm->pci_bus);
+	pci_fixup_irq(pbm, pbm->pci_bus);
+#endif
+	pci_bus_add_devices(pbm->pci_bus);
+}
+
+/*
+ * Main entry point from the PCI subsystem.
+ */
+static int __init pcic_init(void)
+{
+	struct linux_pcic *pcic;
+
+	/*
+	 * PCIC should be initialized at start of the timer.
+	 * So, here we report the presence of PCIC and do some magic passes.
+	 */
+	if(!pcic0_up)
+		return 0;
+	pcic = &pcic0;
+
+	/*
+	 *      Switch off IOTLB translation.
+	 */
+	writeb(PCI_DVMA_CONTROL_IOTLB_DISABLE, 
+	       pcic->pcic_regs+PCI_DVMA_CONTROL);
+
+	/*
+	 *      Increase mapped size for PCI memory space (DMA access).
+	 *      Should be done in that order (size first, address second).
+	 *      Why we couldn't set up 4GB and forget about it? XXX
+	 */
+	writel(0xF0000000UL, pcic->pcic_regs+PCI_SIZE_0);
+	writel(0+PCI_BASE_ADDRESS_SPACE_MEMORY, 
+	       pcic->pcic_regs+PCI_BASE_ADDRESS_0);
+
+	pcic_pbm_scan_bus(pcic);
+
+	return 0;
+}
+
+int pcic_present(void)
+{
+	return pcic0_up;
+}
+
+static int pdev_to_pnode(struct linux_pbm_info *pbm, struct pci_dev *pdev)
+{
+	struct linux_prom_pci_registers regs[PROMREG_MAX];
+	int err;
+	phandle node = prom_getchild(pbm->prom_node);
+
+	while(node) {
+		err = prom_getproperty(node, "reg", 
+				       (char *)&regs[0], sizeof(regs));
+		if(err != 0 && err != -1) {
+			unsigned long devfn = (regs[0].which_io >> 8) & 0xff;
+			if(devfn == pdev->devfn)
+				return node;
+		}
+		node = prom_getsibling(node);
+	}
+	return 0;
+}
+
+static inline struct pcidev_cookie *pci_devcookie_alloc(void)
+{
+	return kmalloc(sizeof(struct pcidev_cookie), GFP_ATOMIC);
+}
+
+static void pcic_map_pci_device(struct linux_pcic *pcic,
+    struct pci_dev *dev, int node)
+{
+	char namebuf[64];
+	unsigned long address;
+	unsigned long flags;
+	int j;
+
+	if (node == 0 || node == -1) {
+		strcpy(namebuf, "???");
+	} else {
+		prom_getstring(node, "name", namebuf, 63); namebuf[63] = 0;
+	}
+
+	for (j = 0; j < 6; j++) {
+		address = dev->resource[j].start;
+		if (address == 0) break;	/* are sequential */
+		flags = dev->resource[j].flags;
+		if ((flags & IORESOURCE_IO) != 0) {
+			if (address < 0x10000) {
+				/*
+				 * A device responds to I/O cycles on PCI.
+				 * We generate these cycles with memory
+				 * access into the fixed map (phys 0x30000000).
+				 *
+				 * Since a device driver does not want to
+				 * do ioremap() before accessing PC-style I/O,
+				 * we supply virtual, ready to access address.
+				 *
+				 * Note that request_region()
+				 * works for these devices.
+				 *
+				 * XXX Neat trick, but it's a *bad* idea
+				 * to shit into regions like that.
+				 * What if we want to allocate one more
+				 * PCI base address...
+				 */
+				dev->resource[j].start =
+				    pcic->pcic_io + address;
+				dev->resource[j].end = 1;  /* XXX */
+				dev->resource[j].flags =
+				    (flags & ~IORESOURCE_IO) | IORESOURCE_MEM;
+			} else {
+				/*
+				 * OOPS... PCI Spec allows this. Sun does
+				 * not have any devices getting above 64K
+				 * so it must be user with a weird I/O
+				 * board in a PCI slot. We must remap it
+				 * under 64K but it is not done yet. XXX
+				 */
+				pci_info(dev, "PCIC: Skipping I/O space at "
+					 "0x%lx, this will Oops if a driver "
+					 "attaches device '%s'\n", address,
+					 namebuf);
+			}
+		}
+	}
+}
+
+static void
+pcic_fill_irq(struct linux_pcic *pcic, struct pci_dev *dev, int node)
+{
+	struct pcic_ca2irq *p;
+	unsigned int real_irq;
+	int i, ivec;
+	char namebuf[64];
+
+	if (node == 0 || node == -1) {
+		strcpy(namebuf, "???");
+	} else {
+		prom_getstring(node, "name", namebuf, sizeof(namebuf));
+	}
+
+	if ((p = pcic->pcic_imap) == NULL) {
+		dev->irq = 0;
+		return;
+	}
+	for (i = 0; i < pcic->pcic_imdim; i++) {
+		if (p->busno == dev->bus->number && p->devfn == dev->devfn)
+			break;
+		p++;
+	}
+	if (i >= pcic->pcic_imdim) {
+		pci_info(dev, "PCIC: device %s not found in %d\n", namebuf,
+			 pcic->pcic_imdim);
+		dev->irq = 0;
+		return;
+	}
+
+	i = p->pin;
+	if (i >= 0 && i < 4) {
+		ivec = readw(pcic->pcic_regs+PCI_INT_SELECT_LO);
+		real_irq = ivec >> (i << 2) & 0xF;
+	} else if (i >= 4 && i < 8) {
+		ivec = readw(pcic->pcic_regs+PCI_INT_SELECT_HI);
+		real_irq = ivec >> ((i-4) << 2) & 0xF;
+	} else {					/* Corrupted map */
+		pci_info(dev, "PCIC: BAD PIN %d\n", i); for (;;) {}
+	}
+/* P3 */ /* printk("PCIC: device %s pin %d ivec 0x%x irq %x\n", namebuf, i, ivec, dev->irq); */
+
+	/* real_irq means PROM did not bother to program the upper
+	 * half of PCIC. This happens on JS-E with PROM 3.11, for instance.
+	 */
+	if (real_irq == 0 || p->force) {
+		if (p->irq == 0 || p->irq >= 15) {	/* Corrupted map */
+			pci_info(dev, "PCIC: BAD IRQ %d\n", p->irq); for (;;) {}
+		}
+		pci_info(dev, "PCIC: setting irq %d at pin %d\n", p->irq,
+			 p->pin);
+		real_irq = p->irq;
+
+		i = p->pin;
+		if (i >= 4) {
+			ivec = readw(pcic->pcic_regs+PCI_INT_SELECT_HI);
+			ivec &= ~(0xF << ((i - 4) << 2));
+			ivec |= p->irq << ((i - 4) << 2);
+			writew(ivec, pcic->pcic_regs+PCI_INT_SELECT_HI);
+		} else {
+			ivec = readw(pcic->pcic_regs+PCI_INT_SELECT_LO);
+			ivec &= ~(0xF << (i << 2));
+			ivec |= p->irq << (i << 2);
+			writew(ivec, pcic->pcic_regs+PCI_INT_SELECT_LO);
+		}
+	}
+	dev->irq = pcic_build_device_irq(NULL, real_irq);
+}
+
+/*
+ * Normally called from {do_}pci_scan_bus...
+ */
+void pcibios_fixup_bus(struct pci_bus *bus)
+{
+	struct pci_dev *dev;
+	struct linux_pcic *pcic;
+	/* struct linux_pbm_info* pbm = &pcic->pbm; */
+	int node;
+	struct pcidev_cookie *pcp;
+
+	if (!pcic0_up) {
+		pci_info(bus, "pcibios_fixup_bus: no PCIC\n");
+		return;
+	}
+	pcic = &pcic0;
+
+	/*
+	 * Next crud is an equivalent of pbm = pcic_bus_to_pbm(bus);
+	 */
+	if (bus->number != 0) {
+		pci_info(bus, "pcibios_fixup_bus: nonzero bus 0x%x\n",
+			 bus->number);
+		return;
+	}
+
+	list_for_each_entry(dev, &bus->devices, bus_list) {
+		node = pdev_to_pnode(&pcic->pbm, dev);
+		if(node == 0)
+			node = -1;
+
+		/* cookies */
+		pcp = pci_devcookie_alloc();
+		pcp->pbm = &pcic->pbm;
+		pcp->prom_node = of_find_node_by_phandle(node);
+		dev->sysdata = pcp;
+
+		/* fixing I/O to look like memory */
+		if ((dev->class>>16) != PCI_BASE_CLASS_BRIDGE)
+			pcic_map_pci_device(pcic, dev, node);
+
+		pcic_fill_irq(pcic, dev, node);
+	}
+}
+
+int pcibios_enable_device(struct pci_dev *dev, int mask)
+{
+	u16 cmd, oldcmd;
+	int i;
+
+	pci_read_config_word(dev, PCI_COMMAND, &cmd);
+	oldcmd = cmd;
+
+	for (i = 0; i < PCI_NUM_RESOURCES; i++) {
+		struct resource *res = &dev->resource[i];
+
+		/* Only set up the requested stuff */
+		if (!(mask & (1<<i)))
+			continue;
+
+		if (res->flags & IORESOURCE_IO)
+			cmd |= PCI_COMMAND_IO;
+		if (res->flags & IORESOURCE_MEM)
+			cmd |= PCI_COMMAND_MEMORY;
+	}
+
+	if (cmd != oldcmd) {
+		pci_info(dev, "enabling device (%04x -> %04x)\n", oldcmd, cmd);
+		pci_write_config_word(dev, PCI_COMMAND, cmd);
+	}
+	return 0;
+}
+
+/* Makes compiler happy */
+static volatile int pcic_timer_dummy;
+
+static void pcic_clear_clock_irq(void)
+{
+	pcic_timer_dummy = readl(pcic0.pcic_regs+PCI_SYS_LIMIT);
+}
+
+/* CPU frequency is 100 MHz, timer increments every 4 CPU clocks */
+#define USECS_PER_JIFFY  (1000000 / HZ)
+#define TICK_TIMER_LIMIT ((100 * 1000000 / 4) / HZ)
+
+static unsigned int pcic_cycles_offset(void)
+{
+	u32 value, count;
+
+	value = readl(pcic0.pcic_regs + PCI_SYS_COUNTER);
+	count = value & ~PCI_SYS_COUNTER_OVERFLOW;
+
+	if (value & PCI_SYS_COUNTER_OVERFLOW)
+		count += TICK_TIMER_LIMIT;
+	/*
+	 * We divide all by HZ
+	 * to have microsecond resolution and to avoid overflow
+	 */
+	count = ((count / HZ) * USECS_PER_JIFFY) / (TICK_TIMER_LIMIT / HZ);
+
+	/* Coordinate with the sparc_config.clock_rate setting */
+	return count * 2;
+}
+
+void __init pci_time_init(void)
+{
+	struct linux_pcic *pcic = &pcic0;
+	unsigned long v;
+	int timer_irq, irq;
+	int err;
+
+#ifndef CONFIG_SMP
+	/*
+	 * The clock_rate is in SBUS dimension.
+	 * We take into account this in pcic_cycles_offset()
+	 */
+	sparc_config.clock_rate = SBUS_CLOCK_RATE / HZ;
+	sparc_config.features |= FEAT_L10_CLOCKEVENT;
+#endif
+	sparc_config.features |= FEAT_L10_CLOCKSOURCE;
+	sparc_config.get_cycles_offset = pcic_cycles_offset;
+
+	writel (TICK_TIMER_LIMIT, pcic->pcic_regs+PCI_SYS_LIMIT);
+	/* PROM should set appropriate irq */
+	v = readb(pcic->pcic_regs+PCI_COUNTER_IRQ);
+	timer_irq = PCI_COUNTER_IRQ_SYS(v);
+	writel (PCI_COUNTER_IRQ_SET(timer_irq, 0),
+		pcic->pcic_regs+PCI_COUNTER_IRQ);
+	irq = pcic_build_device_irq(NULL, timer_irq);
+	err = request_irq(irq, timer_interrupt,
+			  IRQF_TIMER, "timer", NULL);
+	if (err) {
+		prom_printf("time_init: unable to attach IRQ%d\n", timer_irq);
+		prom_halt();
+	}
+	local_irq_enable();
+}
+
+
+#if 0
+static void watchdog_reset() {
+	writeb(0, pcic->pcic_regs+PCI_SYS_STATUS);
+}
+#endif
+
+/*
+ * NMI
+ */
+void pcic_nmi(unsigned int pend, struct pt_regs *regs)
+{
+	pend = swab32(pend);
+
+	if (!pcic_speculative || (pend & PCI_SYS_INT_PENDING_PIO) == 0) {
+		/*
+		 * XXX On CP-1200 PCI #SERR may happen, we do not know
+		 * what to do about it yet.
+		 */
+		printk("Aiee, NMI pend 0x%x pc 0x%x spec %d, hanging\n",
+		    pend, (int)regs->pc, pcic_speculative);
+		for (;;) { }
+	}
+	pcic_speculative = 0;
+	pcic_trapped = 1;
+	regs->pc = regs->npc;
+	regs->npc += 4;
+}
+
+static inline unsigned long get_irqmask(int irq_nr)
+{
+	return 1 << irq_nr;
+}
+
+static void pcic_mask_irq(struct irq_data *data)
+{
+	unsigned long mask, flags;
+
+	mask = (unsigned long)data->chip_data;
+	local_irq_save(flags);
+	writel(mask, pcic0.pcic_regs+PCI_SYS_INT_TARGET_MASK_SET);
+	local_irq_restore(flags);
+}
+
+static void pcic_unmask_irq(struct irq_data *data)
+{
+	unsigned long mask, flags;
+
+	mask = (unsigned long)data->chip_data;
+	local_irq_save(flags);
+	writel(mask, pcic0.pcic_regs+PCI_SYS_INT_TARGET_MASK_CLEAR);
+	local_irq_restore(flags);
+}
+
+static unsigned int pcic_startup_irq(struct irq_data *data)
+{
+	irq_link(data->irq);
+	pcic_unmask_irq(data);
+	return 0;
+}
+
+static struct irq_chip pcic_irq = {
+	.name		= "pcic",
+	.irq_startup	= pcic_startup_irq,
+	.irq_mask	= pcic_mask_irq,
+	.irq_unmask	= pcic_unmask_irq,
+};
+
+unsigned int pcic_build_device_irq(struct platform_device *op,
+                                   unsigned int real_irq)
+{
+	unsigned int irq;
+	unsigned long mask;
+
+	irq = 0;
+	mask = get_irqmask(real_irq);
+	if (mask == 0)
+		goto out;
+
+	irq = irq_alloc(real_irq, real_irq);
+	if (irq == 0)
+		goto out;
+
+	irq_set_chip_and_handler_name(irq, &pcic_irq,
+	                              handle_level_irq, "PCIC");
+	irq_set_chip_data(irq, (void *)mask);
+
+out:
+	return irq;
+}
+
+
+static void pcic_load_profile_irq(int cpu, unsigned int limit)
+{
+	printk("PCIC: unimplemented code: FILE=%s LINE=%d", __FILE__, __LINE__);
+}
+
+void __init sun4m_pci_init_IRQ(void)
+{
+	sparc_config.build_device_irq = pcic_build_device_irq;
+	sparc_config.clear_clock_irq  = pcic_clear_clock_irq;
+	sparc_config.load_profile_irq = pcic_load_profile_irq;
+}
+
+subsys_initcall(pcic_init);