From 5b7c4cabbb65f5c469464da6c5f614cbd7f730f2 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 21 Feb 2023 18:24:12 -0800 Subject: Merge tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next 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(). ... --- arch/sparc/kernel/leon_pci_grpci1.c | 722 ++++++++++++++++++++++++++++++++++++ 1 file changed, 722 insertions(+) create mode 100644 arch/sparc/kernel/leon_pci_grpci1.c (limited to 'arch/sparc/kernel/leon_pci_grpci1.c') diff --git a/arch/sparc/kernel/leon_pci_grpci1.c b/arch/sparc/kernel/leon_pci_grpci1.c new file mode 100644 index 000000000..e6935d0ac --- /dev/null +++ b/arch/sparc/kernel/leon_pci_grpci1.c @@ -0,0 +1,722 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * leon_pci_grpci1.c: GRPCI1 Host PCI driver + * + * Copyright (C) 2013 Aeroflex Gaisler AB + * + * This GRPCI1 driver does not support PCI interrupts taken from + * GPIO pins. Interrupt generation at PCI parity and system error + * detection is by default turned off since some GRPCI1 cores does + * not support detection. It can be turned on from the bootloader + * using the all_pci_errors property. + * + * Contributors: Daniel Hellstrom + */ + +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +#include "irq.h" + +/* Enable/Disable Debugging Configuration Space Access */ +#undef GRPCI1_DEBUG_CFGACCESS + +/* + * GRPCI1 APB Register MAP + */ +struct grpci1_regs { + unsigned int cfg_stat; /* 0x00 Configuration / Status */ + unsigned int bar0; /* 0x04 BAR0 (RO) */ + unsigned int page0; /* 0x08 PAGE0 (RO) */ + unsigned int bar1; /* 0x0C BAR1 (RO) */ + unsigned int page1; /* 0x10 PAGE1 */ + unsigned int iomap; /* 0x14 IO Map */ + unsigned int stat_cmd; /* 0x18 PCI Status & Command (RO) */ + unsigned int irq; /* 0x1C Interrupt register */ +}; + +#define REGLOAD(a) (be32_to_cpu(__raw_readl(&(a)))) +#define REGSTORE(a, v) (__raw_writel(cpu_to_be32(v), &(a))) + +#define PAGE0_BTEN_BIT 0 +#define PAGE0_BTEN (1 << PAGE0_BTEN_BIT) + +#define CFGSTAT_HOST_BIT 13 +#define CFGSTAT_CTO_BIT 8 +#define CFGSTAT_HOST (1 << CFGSTAT_HOST_BIT) +#define CFGSTAT_CTO (1 << CFGSTAT_CTO_BIT) + +#define IRQ_DPE (1 << 9) +#define IRQ_SSE (1 << 8) +#define IRQ_RMA (1 << 7) +#define IRQ_RTA (1 << 6) +#define IRQ_STA (1 << 5) +#define IRQ_DPED (1 << 4) +#define IRQ_INTD (1 << 3) +#define IRQ_INTC (1 << 2) +#define IRQ_INTB (1 << 1) +#define IRQ_INTA (1 << 0) +#define IRQ_DEF_ERRORS (IRQ_RMA | IRQ_RTA | IRQ_STA) +#define IRQ_ALL_ERRORS (IRQ_DPED | IRQ_DEF_ERRORS | IRQ_SSE | IRQ_DPE) +#define IRQ_INTX (IRQ_INTA | IRQ_INTB | IRQ_INTC | IRQ_INTD) +#define IRQ_MASK_BIT 16 + +#define DEF_PCI_ERRORS (PCI_STATUS_SIG_TARGET_ABORT | \ + PCI_STATUS_REC_TARGET_ABORT | \ + PCI_STATUS_REC_MASTER_ABORT) +#define ALL_PCI_ERRORS (PCI_STATUS_PARITY | PCI_STATUS_DETECTED_PARITY | \ + PCI_STATUS_SIG_SYSTEM_ERROR | DEF_PCI_ERRORS) + +#define TGT 256 + +struct grpci1_priv { + struct leon_pci_info info; /* must be on top of this structure */ + struct grpci1_regs __iomem *regs; /* GRPCI register map */ + struct device *dev; + int pci_err_mask; /* STATUS register error mask */ + int irq; /* LEON irqctrl GRPCI IRQ */ + unsigned char irq_map[4]; /* GRPCI nexus PCI INTX# IRQs */ + unsigned int irq_err; /* GRPCI nexus Virt Error IRQ */ + + /* AHB PCI Windows */ + unsigned long pci_area; /* MEMORY */ + unsigned long pci_area_end; + unsigned long pci_io; /* I/O */ + unsigned long pci_conf; /* CONFIGURATION */ + unsigned long pci_conf_end; + unsigned long pci_io_va; +}; + +static struct grpci1_priv *grpci1priv; + +static int grpci1_cfg_w32(struct grpci1_priv *priv, unsigned int bus, + unsigned int devfn, int where, u32 val); + +static int grpci1_map_irq(const struct pci_dev *dev, u8 slot, u8 pin) +{ + struct grpci1_priv *priv = dev->bus->sysdata; + int irq_group; + + /* Use default IRQ decoding on PCI BUS0 according slot numbering */ + irq_group = slot & 0x3; + pin = ((pin - 1) + irq_group) & 0x3; + + return priv->irq_map[pin]; +} + +static int grpci1_cfg_r32(struct grpci1_priv *priv, unsigned int bus, + unsigned int devfn, int where, u32 *val) +{ + u32 *pci_conf, tmp, cfg; + + if (where & 0x3) + return -EINVAL; + + if (bus == 0) { + devfn += (0x8 * 6); /* start at AD16=Device0 */ + } else if (bus == TGT) { + bus = 0; + devfn = 0; /* special case: bridge controller itself */ + } + + /* Select bus */ + cfg = REGLOAD(priv->regs->cfg_stat); + REGSTORE(priv->regs->cfg_stat, (cfg & ~(0xf << 23)) | (bus << 23)); + + /* do read access */ + pci_conf = (u32 *) (priv->pci_conf | (devfn << 8) | (where & 0xfc)); + tmp = LEON3_BYPASS_LOAD_PA(pci_conf); + + /* check if master abort was received */ + if (REGLOAD(priv->regs->cfg_stat) & CFGSTAT_CTO) { + *val = 0xffffffff; + /* Clear Master abort bit in PCI cfg space (is set) */ + tmp = REGLOAD(priv->regs->stat_cmd); + grpci1_cfg_w32(priv, TGT, 0, PCI_COMMAND, tmp); + } else { + /* Bus always little endian (unaffected by byte-swapping) */ + *val = swab32(tmp); + } + + return 0; +} + +static int grpci1_cfg_r16(struct grpci1_priv *priv, unsigned int bus, + unsigned int devfn, int where, u32 *val) +{ + u32 v; + int ret; + + if (where & 0x1) + return -EINVAL; + ret = grpci1_cfg_r32(priv, bus, devfn, where & ~0x3, &v); + *val = 0xffff & (v >> (8 * (where & 0x3))); + return ret; +} + +static int grpci1_cfg_r8(struct grpci1_priv *priv, unsigned int bus, + unsigned int devfn, int where, u32 *val) +{ + u32 v; + int ret; + + ret = grpci1_cfg_r32(priv, bus, devfn, where & ~0x3, &v); + *val = 0xff & (v >> (8 * (where & 3))); + + return ret; +} + +static int grpci1_cfg_w32(struct grpci1_priv *priv, unsigned int bus, + unsigned int devfn, int where, u32 val) +{ + unsigned int *pci_conf; + u32 cfg; + + if (where & 0x3) + return -EINVAL; + + if (bus == 0) { + devfn += (0x8 * 6); /* start at AD16=Device0 */ + } else if (bus == TGT) { + bus = 0; + devfn = 0; /* special case: bridge controller itself */ + } + + /* Select bus */ + cfg = REGLOAD(priv->regs->cfg_stat); + REGSTORE(priv->regs->cfg_stat, (cfg & ~(0xf << 23)) | (bus << 23)); + + pci_conf = (unsigned int *) (priv->pci_conf | + (devfn << 8) | (where & 0xfc)); + LEON3_BYPASS_STORE_PA(pci_conf, swab32(val)); + + return 0; +} + +static int grpci1_cfg_w16(struct grpci1_priv *priv, unsigned int bus, + unsigned int devfn, int where, u32 val) +{ + int ret; + u32 v; + + if (where & 0x1) + return -EINVAL; + ret = grpci1_cfg_r32(priv, bus, devfn, where&~3, &v); + if (ret) + return ret; + v = (v & ~(0xffff << (8 * (where & 0x3)))) | + ((0xffff & val) << (8 * (where & 0x3))); + return grpci1_cfg_w32(priv, bus, devfn, where & ~0x3, v); +} + +static int grpci1_cfg_w8(struct grpci1_priv *priv, unsigned int bus, + unsigned int devfn, int where, u32 val) +{ + int ret; + u32 v; + + ret = grpci1_cfg_r32(priv, bus, devfn, where & ~0x3, &v); + if (ret != 0) + return ret; + v = (v & ~(0xff << (8 * (where & 0x3)))) | + ((0xff & val) << (8 * (where & 0x3))); + return grpci1_cfg_w32(priv, bus, devfn, where & ~0x3, v); +} + +/* Read from Configuration Space. When entering here the PCI layer has taken + * the pci_lock spinlock and IRQ is off. + */ +static int grpci1_read_config(struct pci_bus *bus, unsigned int devfn, + int where, int size, u32 *val) +{ + struct grpci1_priv *priv = grpci1priv; + unsigned int busno = bus->number; + int ret; + + if (PCI_SLOT(devfn) > 15 || busno > 15) { + *val = ~0; + return 0; + } + + switch (size) { + case 1: + ret = grpci1_cfg_r8(priv, busno, devfn, where, val); + break; + case 2: + ret = grpci1_cfg_r16(priv, busno, devfn, where, val); + break; + case 4: + ret = grpci1_cfg_r32(priv, busno, devfn, where, val); + break; + default: + ret = -EINVAL; + break; + } + +#ifdef GRPCI1_DEBUG_CFGACCESS + printk(KERN_INFO + "grpci1_read_config: [%02x:%02x:%x] ofs=%d val=%x size=%d\n", + busno, PCI_SLOT(devfn), PCI_FUNC(devfn), where, *val, size); +#endif + + return ret; +} + +/* Write to Configuration Space. When entering here the PCI layer has taken + * the pci_lock spinlock and IRQ is off. + */ +static int grpci1_write_config(struct pci_bus *bus, unsigned int devfn, + int where, int size, u32 val) +{ + struct grpci1_priv *priv = grpci1priv; + unsigned int busno = bus->number; + + if (PCI_SLOT(devfn) > 15 || busno > 15) + return 0; + +#ifdef GRPCI1_DEBUG_CFGACCESS + printk(KERN_INFO + "grpci1_write_config: [%02x:%02x:%x] ofs=%d size=%d val=%x\n", + busno, PCI_SLOT(devfn), PCI_FUNC(devfn), where, size, val); +#endif + + switch (size) { + default: + return -EINVAL; + case 1: + return grpci1_cfg_w8(priv, busno, devfn, where, val); + case 2: + return grpci1_cfg_w16(priv, busno, devfn, where, val); + case 4: + return grpci1_cfg_w32(priv, busno, devfn, where, val); + } +} + +static struct pci_ops grpci1_ops = { + .read = grpci1_read_config, + .write = grpci1_write_config, +}; + +/* GENIRQ IRQ chip implementation for grpci1 irqmode=0..2. In configuration + * 3 where all PCI Interrupts has a separate IRQ on the system IRQ controller + * this is not needed and the standard IRQ controller can be used. + */ + +static void grpci1_mask_irq(struct irq_data *data) +{ + u32 irqidx; + struct grpci1_priv *priv = grpci1priv; + + irqidx = (u32)data->chip_data - 1; + if (irqidx > 3) /* only mask PCI interrupts here */ + return; + irqidx += IRQ_MASK_BIT; + + REGSTORE(priv->regs->irq, REGLOAD(priv->regs->irq) & ~(1 << irqidx)); +} + +static void grpci1_unmask_irq(struct irq_data *data) +{ + u32 irqidx; + struct grpci1_priv *priv = grpci1priv; + + irqidx = (u32)data->chip_data - 1; + if (irqidx > 3) /* only unmask PCI interrupts here */ + return; + irqidx += IRQ_MASK_BIT; + + REGSTORE(priv->regs->irq, REGLOAD(priv->regs->irq) | (1 << irqidx)); +} + +static unsigned int grpci1_startup_irq(struct irq_data *data) +{ + grpci1_unmask_irq(data); + return 0; +} + +static void grpci1_shutdown_irq(struct irq_data *data) +{ + grpci1_mask_irq(data); +} + +static struct irq_chip grpci1_irq = { + .name = "grpci1", + .irq_startup = grpci1_startup_irq, + .irq_shutdown = grpci1_shutdown_irq, + .irq_mask = grpci1_mask_irq, + .irq_unmask = grpci1_unmask_irq, +}; + +/* Handle one or multiple IRQs from the PCI core */ +static void grpci1_pci_flow_irq(struct irq_desc *desc) +{ + struct grpci1_priv *priv = grpci1priv; + int i, ack = 0; + unsigned int irqreg; + + irqreg = REGLOAD(priv->regs->irq); + irqreg = (irqreg >> IRQ_MASK_BIT) & irqreg; + + /* Error Interrupt? */ + if (irqreg & IRQ_ALL_ERRORS) { + generic_handle_irq(priv->irq_err); + ack = 1; + } + + /* PCI Interrupt? */ + if (irqreg & IRQ_INTX) { + /* Call respective PCI Interrupt handler */ + for (i = 0; i < 4; i++) { + if (irqreg & (1 << i)) + generic_handle_irq(priv->irq_map[i]); + } + ack = 1; + } + + /* + * Call "first level" IRQ chip end-of-irq handler. It will ACK LEON IRQ + * Controller, this must be done after IRQ sources have been handled to + * avoid double IRQ generation + */ + if (ack) + desc->irq_data.chip->irq_eoi(&desc->irq_data); +} + +/* Create a virtual IRQ */ +static unsigned int grpci1_build_device_irq(unsigned int irq) +{ + unsigned int virq = 0, pil; + + pil = 1 << 8; + virq = irq_alloc(irq, pil); + if (virq == 0) + goto out; + + irq_set_chip_and_handler_name(virq, &grpci1_irq, handle_simple_irq, + "pcilvl"); + irq_set_chip_data(virq, (void *)irq); + +out: + return virq; +} + +/* + * Initialize mappings AMBA<->PCI, clear IRQ state, setup PCI interface + * + * Target BARs: + * BAR0: unused in this implementation + * BAR1: peripheral DMA to host's memory (size at least 256MByte) + * BAR2..BAR5: not implemented in hardware + */ +static void grpci1_hw_init(struct grpci1_priv *priv) +{ + u32 ahbadr, bar_sz, data, pciadr; + struct grpci1_regs __iomem *regs = priv->regs; + + /* set 1:1 mapping between AHB -> PCI memory space */ + REGSTORE(regs->cfg_stat, priv->pci_area & 0xf0000000); + + /* map PCI accesses to target BAR1 to Linux kernel memory 1:1 */ + ahbadr = 0xf0000000 & (u32)__pa(PAGE_ALIGN((unsigned long) &_end)); + REGSTORE(regs->page1, ahbadr); + + /* translate I/O accesses to 0, I/O Space always @ PCI low 64Kbytes */ + REGSTORE(regs->iomap, REGLOAD(regs->iomap) & 0x0000ffff); + + /* disable and clear pending interrupts */ + REGSTORE(regs->irq, 0); + + /* Setup BAR0 outside access range so that it does not conflict with + * peripheral DMA. There is no need to set up the PAGE0 register. + */ + grpci1_cfg_w32(priv, TGT, 0, PCI_BASE_ADDRESS_0, 0xffffffff); + grpci1_cfg_r32(priv, TGT, 0, PCI_BASE_ADDRESS_0, &bar_sz); + bar_sz = ~bar_sz + 1; + pciadr = priv->pci_area - bar_sz; + grpci1_cfg_w32(priv, TGT, 0, PCI_BASE_ADDRESS_0, pciadr); + + /* + * Setup the Host's PCI Target BAR1 for other peripherals to access, + * and do DMA to the host's memory. + */ + grpci1_cfg_w32(priv, TGT, 0, PCI_BASE_ADDRESS_1, ahbadr); + + /* + * Setup Latency Timer and cache line size. Default cache line + * size will result in poor performance (256 word fetches), 0xff + * will set it according to the max size of the PCI FIFO. + */ + grpci1_cfg_w8(priv, TGT, 0, PCI_CACHE_LINE_SIZE, 0xff); + grpci1_cfg_w8(priv, TGT, 0, PCI_LATENCY_TIMER, 0x40); + + /* set as bus master, enable pci memory responses, clear status bits */ + grpci1_cfg_r32(priv, TGT, 0, PCI_COMMAND, &data); + data |= (PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER); + grpci1_cfg_w32(priv, TGT, 0, PCI_COMMAND, data); +} + +static irqreturn_t grpci1_jump_interrupt(int irq, void *arg) +{ + struct grpci1_priv *priv = arg; + dev_err(priv->dev, "Jump IRQ happened\n"); + return IRQ_NONE; +} + +/* Handle GRPCI1 Error Interrupt */ +static irqreturn_t grpci1_err_interrupt(int irq, void *arg) +{ + struct grpci1_priv *priv = arg; + u32 status; + + grpci1_cfg_r16(priv, TGT, 0, PCI_STATUS, &status); + status &= priv->pci_err_mask; + + if (status == 0) + return IRQ_NONE; + + if (status & PCI_STATUS_PARITY) + dev_err(priv->dev, "Data Parity Error\n"); + + if (status & PCI_STATUS_SIG_TARGET_ABORT) + dev_err(priv->dev, "Signalled Target Abort\n"); + + if (status & PCI_STATUS_REC_TARGET_ABORT) + dev_err(priv->dev, "Received Target Abort\n"); + + if (status & PCI_STATUS_REC_MASTER_ABORT) + dev_err(priv->dev, "Received Master Abort\n"); + + if (status & PCI_STATUS_SIG_SYSTEM_ERROR) + dev_err(priv->dev, "Signalled System Error\n"); + + if (status & PCI_STATUS_DETECTED_PARITY) + dev_err(priv->dev, "Parity Error\n"); + + /* Clear handled INT TYPE IRQs */ + grpci1_cfg_w16(priv, TGT, 0, PCI_STATUS, status); + + return IRQ_HANDLED; +} + +static int grpci1_of_probe(struct platform_device *ofdev) +{ + struct grpci1_regs __iomem *regs; + struct grpci1_priv *priv; + int err, len; + const int *tmp; + u32 cfg, size, err_mask; + struct resource *res; + + if (grpci1priv) { + dev_err(&ofdev->dev, "only one GRPCI1 supported\n"); + return -ENODEV; + } + + if (ofdev->num_resources < 3) { + dev_err(&ofdev->dev, "not enough APB/AHB resources\n"); + return -EIO; + } + + priv = devm_kzalloc(&ofdev->dev, sizeof(*priv), GFP_KERNEL); + if (!priv) { + dev_err(&ofdev->dev, "memory allocation failed\n"); + return -ENOMEM; + } + platform_set_drvdata(ofdev, priv); + priv->dev = &ofdev->dev; + + /* find device register base address */ + res = platform_get_resource(ofdev, IORESOURCE_MEM, 0); + regs = devm_ioremap_resource(&ofdev->dev, res); + if (IS_ERR(regs)) + return PTR_ERR(regs); + + /* + * check that we're in Host Slot and that we can act as a Host Bridge + * and not only as target/peripheral. + */ + cfg = REGLOAD(regs->cfg_stat); + if ((cfg & CFGSTAT_HOST) == 0) { + dev_err(&ofdev->dev, "not in host system slot\n"); + return -EIO; + } + + /* check that BAR1 support 256 MByte so that we can map kernel space */ + REGSTORE(regs->page1, 0xffffffff); + size = ~REGLOAD(regs->page1) + 1; + if (size < 0x10000000) { + dev_err(&ofdev->dev, "BAR1 must be at least 256MByte\n"); + return -EIO; + } + + /* hardware must support little-endian PCI (byte-twisting) */ + if ((REGLOAD(regs->page0) & PAGE0_BTEN) == 0) { + dev_err(&ofdev->dev, "byte-twisting is required\n"); + return -EIO; + } + + priv->regs = regs; + priv->irq = irq_of_parse_and_map(ofdev->dev.of_node, 0); + dev_info(&ofdev->dev, "host found at 0x%p, irq%d\n", regs, priv->irq); + + /* Find PCI Memory, I/O and Configuration Space Windows */ + priv->pci_area = ofdev->resource[1].start; + priv->pci_area_end = ofdev->resource[1].end+1; + priv->pci_io = ofdev->resource[2].start; + priv->pci_conf = ofdev->resource[2].start + 0x10000; + priv->pci_conf_end = priv->pci_conf + 0x10000; + priv->pci_io_va = (unsigned long)ioremap(priv->pci_io, 0x10000); + if (!priv->pci_io_va) { + dev_err(&ofdev->dev, "unable to map PCI I/O area\n"); + return -EIO; + } + + printk(KERN_INFO + "GRPCI1: MEMORY SPACE [0x%08lx - 0x%08lx]\n" + " I/O SPACE [0x%08lx - 0x%08lx]\n" + " CONFIG SPACE [0x%08lx - 0x%08lx]\n", + priv->pci_area, priv->pci_area_end-1, + priv->pci_io, priv->pci_conf-1, + priv->pci_conf, priv->pci_conf_end-1); + + /* + * I/O Space resources in I/O Window mapped into Virtual Adr Space + * We never use low 4KB because some devices seem have problems using + * address 0. + */ + priv->info.io_space.name = "GRPCI1 PCI I/O Space"; + priv->info.io_space.start = priv->pci_io_va + 0x1000; + priv->info.io_space.end = priv->pci_io_va + 0x10000 - 1; + priv->info.io_space.flags = IORESOURCE_IO; + + /* + * grpci1 has no prefetchable memory, map everything as + * non-prefetchable memory + */ + priv->info.mem_space.name = "GRPCI1 PCI MEM Space"; + priv->info.mem_space.start = priv->pci_area; + priv->info.mem_space.end = priv->pci_area_end - 1; + priv->info.mem_space.flags = IORESOURCE_MEM; + + if (request_resource(&iomem_resource, &priv->info.mem_space) < 0) { + dev_err(&ofdev->dev, "unable to request PCI memory area\n"); + err = -ENOMEM; + goto err1; + } + + if (request_resource(&ioport_resource, &priv->info.io_space) < 0) { + dev_err(&ofdev->dev, "unable to request PCI I/O area\n"); + err = -ENOMEM; + goto err2; + } + + /* setup maximum supported PCI buses */ + priv->info.busn.name = "GRPCI1 busn"; + priv->info.busn.start = 0; + priv->info.busn.end = 15; + + grpci1priv = priv; + + /* Initialize hardware */ + grpci1_hw_init(priv); + + /* + * Get PCI Interrupt to System IRQ mapping and setup IRQ handling + * Error IRQ. All PCI and PCI-Error interrupts are shared using the + * same system IRQ. + */ + leon_update_virq_handling(priv->irq, grpci1_pci_flow_irq, "pcilvl", 0); + + priv->irq_map[0] = grpci1_build_device_irq(1); + priv->irq_map[1] = grpci1_build_device_irq(2); + priv->irq_map[2] = grpci1_build_device_irq(3); + priv->irq_map[3] = grpci1_build_device_irq(4); + priv->irq_err = grpci1_build_device_irq(5); + + printk(KERN_INFO " PCI INTA..D#: IRQ%d, IRQ%d, IRQ%d, IRQ%d\n", + priv->irq_map[0], priv->irq_map[1], priv->irq_map[2], + priv->irq_map[3]); + + /* Enable IRQs on LEON IRQ controller */ + err = devm_request_irq(&ofdev->dev, priv->irq, grpci1_jump_interrupt, 0, + "GRPCI1_JUMP", priv); + if (err) { + dev_err(&ofdev->dev, "ERR IRQ request failed: %d\n", err); + goto err3; + } + + /* Setup IRQ handler for access errors */ + err = devm_request_irq(&ofdev->dev, priv->irq_err, + grpci1_err_interrupt, IRQF_SHARED, "GRPCI1_ERR", + priv); + if (err) { + dev_err(&ofdev->dev, "ERR VIRQ request failed: %d\n", err); + goto err3; + } + + tmp = of_get_property(ofdev->dev.of_node, "all_pci_errors", &len); + if (tmp && (len == 4)) { + priv->pci_err_mask = ALL_PCI_ERRORS; + err_mask = IRQ_ALL_ERRORS << IRQ_MASK_BIT; + } else { + priv->pci_err_mask = DEF_PCI_ERRORS; + err_mask = IRQ_DEF_ERRORS << IRQ_MASK_BIT; + } + + /* + * Enable Error Interrupts. PCI interrupts are unmasked once request_irq + * is called by the PCI Device drivers + */ + REGSTORE(regs->irq, err_mask); + + /* Init common layer and scan buses */ + priv->info.ops = &grpci1_ops; + priv->info.map_irq = grpci1_map_irq; + leon_pci_init(ofdev, &priv->info); + + return 0; + +err3: + release_resource(&priv->info.io_space); +err2: + release_resource(&priv->info.mem_space); +err1: + iounmap((void __iomem *)priv->pci_io_va); + grpci1priv = NULL; + return err; +} + +static const struct of_device_id grpci1_of_match[] __initconst = { + { + .name = "GAISLER_PCIFBRG", + }, + { + .name = "01_014", + }, + {}, +}; + +static struct platform_driver grpci1_of_driver = { + .driver = { + .name = "grpci1", + .of_match_table = grpci1_of_match, + }, + .probe = grpci1_of_probe, +}; + +static int __init grpci1_init(void) +{ + return platform_driver_register(&grpci1_of_driver); +} + +subsys_initcall(grpci1_init); -- cgit v1.2.3