diff options
191 files changed, 16508 insertions, 2937 deletions
diff --git a/Documentation/devicetree/bindings/net/apm-xgene-enet.txt b/Documentation/devicetree/bindings/net/apm-xgene-enet.txt index e41b2d59ca7f..f591ab782dbc 100644 --- a/Documentation/devicetree/bindings/net/apm-xgene-enet.txt +++ b/Documentation/devicetree/bindings/net/apm-xgene-enet.txt @@ -47,6 +47,9 @@ Optional properties: Valid values are between 0 to 7, that maps to 273, 589, 899, 1222, 1480, 1806, 2147, 2464 ps Default value is 2, which corresponds to 899 ps +- rxlos-gpios: Input gpio from SFP+ module to indicate availability of + incoming signal. + Example: menetclk: menetclk { diff --git a/Documentation/devicetree/bindings/net/xilinx_gmii2rgmii.txt b/Documentation/devicetree/bindings/net/xilinx_gmii2rgmii.txt new file mode 100644 index 000000000000..038dda48b8e6 --- /dev/null +++ b/Documentation/devicetree/bindings/net/xilinx_gmii2rgmii.txt @@ -0,0 +1,35 @@ +XILINX GMIITORGMII Converter Driver Device Tree Bindings +-------------------------------------------------------- + +The Gigabit Media Independent Interface (GMII) to Reduced Gigabit Media +Independent Interface (RGMII) core provides the RGMII between RGMII-compliant +Ethernet physical media devices (PHY) and the Gigabit Ethernet controller. +This core can be used in all three modes of operation(10/100/1000 Mb/s). +The Management Data Input/Output (MDIO) interface is used to configure the +Speed of operation. This core can switch dynamically between the three +Different speed modes by configuring the conveter register through mdio write. + +This converter sits between the ethernet MAC and the external phy. +MAC <==> GMII2RGMII <==> RGMII_PHY + +For more details about mdio please refer phy.txt file in the same directory. + +Required properties: +- compatible : Should be "xlnx,gmii-to-rgmii-1.0" +- reg : The ID number for the phy, usually a small integer +- phy-handle : Should point to the external phy device. + See ethernet.txt file in the same directory. + +Example: + mdio { + #address-cells = <1>; + #size-cells = <0>; + phy: ethernet-phy@0 { + ...... + }; + gmiitorgmii: gmiitorgmii@8 { + compatible = "xlnx,gmii-to-rgmii-1.0"; + reg = <8>; + phy-handle = <&phy>; + }; + }; diff --git a/Documentation/networking/00-INDEX b/Documentation/networking/00-INDEX index 415154a487d0..a7697783ac4c 100644 --- a/Documentation/networking/00-INDEX +++ b/Documentation/networking/00-INDEX @@ -74,6 +74,8 @@ dns_resolver.txt - The DNS resolver module allows kernel servies to make DNS queries. driver.txt - Softnet driver issues. +ena.txt + - info on Amazon's Elastic Network Adapter (ENA) e100.txt - info on Intel's EtherExpress PRO/100 line of 10/100 boards e1000.txt diff --git a/Documentation/networking/ena.txt b/Documentation/networking/ena.txt new file mode 100644 index 000000000000..2b4b6f57e549 --- /dev/null +++ b/Documentation/networking/ena.txt @@ -0,0 +1,305 @@ +Linux kernel driver for Elastic Network Adapter (ENA) family: +============================================================= + +Overview: +========= +ENA is a networking interface designed to make good use of modern CPU +features and system architectures. + +The ENA device exposes a lightweight management interface with a +minimal set of memory mapped registers and extendable command set +through an Admin Queue. + +The driver supports a range of ENA devices, is link-speed independent +(i.e., the same driver is used for 10GbE, 25GbE, 40GbE, etc.), and has +a negotiated and extendable feature set. + +Some ENA devices support SR-IOV. This driver is used for both the +SR-IOV Physical Function (PF) and Virtual Function (VF) devices. + +ENA devices enable high speed and low overhead network traffic +processing by providing multiple Tx/Rx queue pairs (the maximum number +is advertised by the device via the Admin Queue), a dedicated MSI-X +interrupt vector per Tx/Rx queue pair, adaptive interrupt moderation, +and CPU cacheline optimized data placement. + +The ENA driver supports industry standard TCP/IP offload features such +as checksum offload and TCP transmit segmentation offload (TSO). +Receive-side scaling (RSS) is supported for multi-core scaling. + +The ENA driver and its corresponding devices implement health +monitoring mechanisms such as watchdog, enabling the device and driver +to recover in a manner transparent to the application, as well as +debug logs. + +Some of the ENA devices support a working mode called Low-latency +Queue (LLQ), which saves several more microseconds. + +Supported PCI vendor ID/device IDs: +=================================== +1d0f:0ec2 - ENA PF +1d0f:1ec2 - ENA PF with LLQ support +1d0f:ec20 - ENA VF +1d0f:ec21 - ENA VF with LLQ support + +ENA Source Code Directory Structure: +==================================== +ena_com.[ch] - Management communication layer. This layer is + responsible for the handling all the management + (admin) communication between the device and the + driver. +ena_eth_com.[ch] - Tx/Rx data path. +ena_admin_defs.h - Definition of ENA management interface. +ena_eth_io_defs.h - Definition of ENA data path interface. +ena_common_defs.h - Common definitions for ena_com layer. +ena_regs_defs.h - Definition of ENA PCI memory-mapped (MMIO) registers. +ena_netdev.[ch] - Main Linux kernel driver. +ena_syfsfs.[ch] - Sysfs files. +ena_ethtool.c - ethtool callbacks. +ena_pci_id_tbl.h - Supported device IDs. + +Management Interface: +===================== +ENA management interface is exposed by means of: +- PCIe Configuration Space +- Device Registers +- Admin Queue (AQ) and Admin Completion Queue (ACQ) +- Asynchronous Event Notification Queue (AENQ) + +ENA device MMIO Registers are accessed only during driver +initialization and are not involved in further normal device +operation. + +AQ is used for submitting management commands, and the +results/responses are reported asynchronously through ACQ. + +ENA introduces a very small set of management commands with room for +vendor-specific extensions. Most of the management operations are +framed in a generic Get/Set feature command. + +The following admin queue commands are supported: +- Create I/O submission queue +- Create I/O completion queue +- Destroy I/O submission queue +- Destroy I/O completion queue +- Get feature +- Set feature +- Configure AENQ +- Get statistics + +Refer to ena_admin_defs.h for the list of supported Get/Set Feature +properties. + +The Asynchronous Event Notification Queue (AENQ) is a uni-directional +queue used by the ENA device to send to the driver events that cannot +be reported using ACQ. AENQ events are subdivided into groups. Each +group may have multiple syndromes, as shown below + +The events are: + Group Syndrome + Link state change - X - + Fatal error - X - + Notification Suspend traffic + Notification Resume traffic + Keep-Alive - X - + +ACQ and AENQ share the same MSI-X vector. + +Keep-Alive is a special mechanism that allows monitoring of the +device's health. The driver maintains a watchdog (WD) handler which, +if fired, logs the current state and statistics then resets and +restarts the ENA device and driver. A Keep-Alive event is delivered by +the device every second. The driver re-arms the WD upon reception of a +Keep-Alive event. A missed Keep-Alive event causes the WD handler to +fire. + +Data Path Interface: +==================== +I/O operations are based on Tx and Rx Submission Queues (Tx SQ and Rx +SQ correspondingly). Each SQ has a completion queue (CQ) associated +with it. + +The SQs and CQs are implemented as descriptor rings in contiguous +physical memory. + +The ENA driver supports two Queue Operation modes for Tx SQs: +- Regular mode + * In this mode the Tx SQs reside in the host's memory. The ENA + device fetches the ENA Tx descriptors and packet data from host + memory. +- Low Latency Queue (LLQ) mode or "push-mode". + * In this mode the driver pushes the transmit descriptors and the + first 128 bytes of the packet directly to the ENA device memory + space. The rest of the packet payload is fetched by the + device. For this operation mode, the driver uses a dedicated PCI + device memory BAR, which is mapped with write-combine capability. + +The Rx SQs support only the regular mode. + +Note: Not all ENA devices support LLQ, and this feature is negotiated + with the device upon initialization. If the ENA device does not + support LLQ mode, the driver falls back to the regular mode. + +The driver supports multi-queue for both Tx and Rx. This has various +benefits: +- Reduced CPU/thread/process contention on a given Ethernet interface. +- Cache miss rate on completion is reduced, particularly for data + cache lines that hold the sk_buff structures. +- Increased process-level parallelism when handling received packets. +- Increased data cache hit rate, by steering kernel processing of + packets to the CPU, where the application thread consuming the + packet is running. +- In hardware interrupt re-direction. + +Interrupt Modes: +================ +The driver assigns a single MSI-X vector per queue pair (for both Tx +and Rx directions). The driver assigns an additional dedicated MSI-X vector +for management (for ACQ and AENQ). + +Management interrupt registration is performed when the Linux kernel +probes the adapter, and it is de-registered when the adapter is +removed. I/O queue interrupt registration is performed when the Linux +interface of the adapter is opened, and it is de-registered when the +interface is closed. + +The management interrupt is named: + ena-mgmnt@pci:<PCI domain:bus:slot.function> +and for each queue pair, an interrupt is named: + <interface name>-Tx-Rx-<queue index> + +The ENA device operates in auto-mask and auto-clear interrupt +modes. That is, once MSI-X is delivered to the host, its Cause bit is +automatically cleared and the interrupt is masked. The interrupt is +unmasked by the driver after NAPI processing is complete. + +Interrupt Moderation: +===================== +ENA driver and device can operate in conventional or adaptive interrupt +moderation mode. + +In conventional mode the driver instructs device to postpone interrupt +posting according to static interrupt delay value. The interrupt delay +value can be configured through ethtool(8). The following ethtool +parameters are supported by the driver: tx-usecs, rx-usecs + +In adaptive interrupt moderation mode the interrupt delay value is +updated by the driver dynamically and adjusted every NAPI cycle +according to the traffic nature. + +By default ENA driver applies adaptive coalescing on Rx traffic and +conventional coalescing on Tx traffic. + +Adaptive coalescing can be switched on/off through ethtool(8) +adaptive_rx on|off parameter. + +The driver chooses interrupt delay value according to the number of +bytes and packets received between interrupt unmasking and interrupt +posting. The driver uses interrupt delay table that subdivides the +range of received bytes/packets into 5 levels and assigns interrupt +delay value to each level. + +The user can enable/disable adaptive moderation, modify the interrupt +delay table and restore its default values through sysfs. + +The rx_copybreak is initialized by default to ENA_DEFAULT_RX_COPYBREAK +and can be configured by the ETHTOOL_STUNABLE command of the +SIOCETHTOOL ioctl. + +SKB: +The driver-allocated SKB for frames received from Rx handling using +NAPI context. The allocation method depends on the size of the packet. +If the frame length is larger than rx_copybreak, napi_get_frags() +is used, otherwise netdev_alloc_skb_ip_align() is used, the buffer +content is copied (by CPU) to the SKB, and the buffer is recycled. + +Statistics: +=========== +The user can obtain ENA device and driver statistics using ethtool. +The driver can collect regular or extended statistics (including +per-queue stats) from the device. + +In addition the driver logs the stats to syslog upon device reset. + +MTU: +==== +The driver supports an arbitrarily large MTU with a maximum that is +negotiated with the device. The driver configures MTU using the +SetFeature command (ENA_ADMIN_MTU property). The user can change MTU +via ip(8) and similar legacy tools. + +Stateless Offloads: +=================== +The ENA driver supports: +- TSO over IPv4/IPv6 +- TSO with ECN +- IPv4 header checksum offload +- TCP/UDP over IPv4/IPv6 checksum offloads + +RSS: +==== +- The ENA device supports RSS that allows flexible Rx traffic + steering. +- Toeplitz and CRC32 hash functions are supported. +- Different combinations of L2/L3/L4 fields can be configured as + inputs for hash functions. +- The driver configures RSS settings using the AQ SetFeature command + (ENA_ADMIN_RSS_HASH_FUNCTION, ENA_ADMIN_RSS_HASH_INPUT and + ENA_ADMIN_RSS_REDIRECTION_TABLE_CONFIG properties). +- If the NETIF_F_RXHASH flag is set, the 32-bit result of the hash + function delivered in the Rx CQ descriptor is set in the received + SKB. +- The user can provide a hash key, hash function, and configure the + indirection table through ethtool(8). + +DATA PATH: +========== +Tx: +--- +end_start_xmit() is called by the stack. This function does the following: +- Maps data buffers (skb->data and frags). +- Populates ena_buf for the push buffer (if the driver and device are + in push mode.) +- Prepares ENA bufs for the remaining frags. +- Allocates a new request ID from the empty req_id ring. The request + ID is the index of the packet in the Tx info. This is used for + out-of-order TX completions. +- Adds the packet to the proper place in the Tx ring. +- Calls ena_com_prepare_tx(), an ENA communication layer that converts + the ena_bufs to ENA descriptors (and adds meta ENA descriptors as + needed.) + * This function also copies the ENA descriptors and the push buffer + to the Device memory space (if in push mode.) +- Writes doorbell to the ENA device. +- When the ENA device finishes sending the packet, a completion + interrupt is raised. +- The interrupt handler schedules NAPI. +- The ena_clean_tx_irq() function is called. This function handles the + completion descriptors generated by the ENA, with a single + completion descriptor per completed packet. + * req_id is retrieved from the completion descriptor. The tx_info of + the packet is retrieved via the req_id. The data buffers are + unmapped and req_id is returned to the empty req_id ring. + * The function stops when the completion descriptors are completed or + the budget is reached. + +Rx: +--- +- When a packet is received from the ENA device. +- The interrupt handler schedules NAPI. +- The ena_clean_rx_irq() function is called. This function calls + ena_rx_pkt(), an ENA communication layer function, which returns the + number of descriptors used for a new unhandled packet, and zero if + no new packet is found. +- Then it calls the ena_clean_rx_irq() function. +- ena_eth_rx_skb() checks packet length: + * If the packet is small (len < rx_copybreak), the driver allocates + a SKB for the new packet, and copies the packet payload into the + SKB data buffer. + - In this way the original data buffer is not passed to the stack + and is reused for future Rx packets. + * Otherwise the function unmaps the Rx buffer, then allocates the + new SKB structure and hooks the Rx buffer to the SKB frags. +- The new SKB is updated with the necessary information (protocol, + checksum hw verify result, etc.), and then passed to the network + stack, using the NAPI interface function napi_gro_receive(). diff --git a/MAINTAINERS b/MAINTAINERS index 429fc61bee81..50ce33dc62f6 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -636,6 +636,15 @@ F: drivers/tty/serial/altera_jtaguart.c F: include/linux/altera_uart.h F: include/linux/altera_jtaguart.h +AMAZON ETHERNET DRIVERS +M: Netanel Belgazal <netanel@annapurnalabs.com> +R: Saeed Bishara <saeed@annapurnalabs.com> +R: Zorik Machulsky <zorik@annapurnalabs.com> +L: netdev@vger.kernel.org +S: Supported +F: Documentation/networking/ena.txt +F: drivers/net/ethernet/amazon/ + AMD CRYPTOGRAPHIC COPROCESSOR (CCP) DRIVER M: Tom Lendacky <thomas.lendacky@amd.com> M: Gary Hook <gary.hook@amd.com> diff --git a/arch/arm64/boot/dts/apm/apm-mustang.dts b/arch/arm64/boot/dts/apm/apm-mustang.dts index b7fb5d9295c2..32a961c5e98a 100644 --- a/arch/arm64/boot/dts/apm/apm-mustang.dts +++ b/arch/arm64/boot/dts/apm/apm-mustang.dts @@ -74,6 +74,7 @@ &xgenet { status = "ok"; + rxlos-gpios = <&sbgpio 12 1>; }; &mmc0 { diff --git a/arch/arm64/boot/dts/apm/apm-storm.dtsi b/arch/arm64/boot/dts/apm/apm-storm.dtsi index f1c2c713f9b0..d5c3435324e8 100644 --- a/arch/arm64/boot/dts/apm/apm-storm.dtsi +++ b/arch/arm64/boot/dts/apm/apm-storm.dtsi @@ -923,7 +923,7 @@ /* mac address will be overwritten by the bootloader */ local-mac-address = [00 00 00 00 00 00]; phy-connection-type = "rgmii"; - phy-handle = <&menet0phy>,<&menetphy>; + phy-handle = <&menetphy>,<&menet0phy>; mdio { compatible = "apm,xgene-mdio"; #address-cells = <1>; diff --git a/drivers/net/can/usb/ems_usb.c b/drivers/net/can/usb/ems_usb.c index 71f0e791355b..b3d02759c226 100644 --- a/drivers/net/can/usb/ems_usb.c +++ b/drivers/net/can/usb/ems_usb.c @@ -600,7 +600,6 @@ static int ems_usb_start(struct ems_usb *dev) /* create a URB, and a buffer for it */ urb = usb_alloc_urb(0, GFP_KERNEL); if (!urb) { - netdev_err(netdev, "No memory left for URBs\n"); err = -ENOMEM; break; } @@ -752,10 +751,8 @@ static netdev_tx_t ems_usb_start_xmit(struct sk_buff *skb, struct net_device *ne /* create a URB, and a buffer for it, and copy the data to the URB */ urb = usb_alloc_urb(0, GFP_ATOMIC); - if (!urb) { - netdev_err(netdev, "No memory left for URBs\n"); + if (!urb) goto nomem; - } buf = usb_alloc_coherent(dev->udev, size, GFP_ATOMIC, &urb->transfer_dma); if (!buf) { @@ -1007,10 +1004,8 @@ static int ems_usb_probe(struct usb_interface *intf, dev->tx_contexts[i].echo_index = MAX_TX_URBS; dev->intr_urb = usb_alloc_urb(0, GFP_KERNEL); - if (!dev->intr_urb) { - dev_err(&intf->dev, "Couldn't alloc intr URB\n"); + if (!dev->intr_urb) goto cleanup_candev; - } dev->intr_in_buffer = kzalloc(INTR_IN_BUFFER_SIZE, GFP_KERNEL); if (!dev->intr_in_buffer) diff --git a/drivers/net/can/usb/esd_usb2.c b/drivers/net/can/usb/esd_usb2.c index 784a9002fbb9..be928ce62d32 100644 --- a/drivers/net/can/usb/esd_usb2.c +++ b/drivers/net/can/usb/esd_usb2.c @@ -558,8 +558,6 @@ static int esd_usb2_setup_rx_urbs(struct esd_usb2 *dev) /* create a URB, and a buffer for it */ urb = usb_alloc_urb(0, GFP_KERNEL); if (!urb) { - dev_warn(dev->udev->dev.parent, - "No memory left for URBs\n"); err = -ENOMEM; break; } @@ -730,7 +728,6 @@ static netdev_tx_t esd_usb2_start_xmit(struct sk_buff *skb, /* create a URB, and a buffer for it, and copy the data to the URB */ urb = usb_alloc_urb(0, GFP_ATOMIC); if (!urb) { - netdev_err(netdev, "No memory left for URBs\n"); stats->tx_dropped++; dev_kfree_skb(skb); goto nourbmem; diff --git a/drivers/net/can/usb/gs_usb.c b/drivers/net/can/usb/gs_usb.c index 6f0cbc38782e..77e3cc06a30c 100644 --- a/drivers/net/can/usb/gs_usb.c +++ b/drivers/net/can/usb/gs_usb.c @@ -493,10 +493,8 @@ static netdev_tx_t gs_can_start_xmit(struct sk_buff *skb, /* create a URB, and a buffer for it */ urb = usb_alloc_urb(0, GFP_ATOMIC); - if (!urb) { - netdev_err(netdev, "No memory left for URB\n"); + if (!urb) goto nomem_urb; - } hf = usb_alloc_coherent(dev->udev, sizeof(*hf), GFP_ATOMIC, &urb->transfer_dma); @@ -600,11 +598,8 @@ static int gs_can_open(struct net_device *netdev) /* alloc rx urb */ urb = usb_alloc_urb(0, GFP_KERNEL); - if (!urb) { - netdev_err(netdev, - "No memory left for URB\n"); + if (!urb) return -ENOMEM; - } /* alloc rx buffer */ buf = usb_alloc_coherent(dev->udev, diff --git a/drivers/net/can/usb/kvaser_usb.c b/drivers/net/can/usb/kvaser_usb.c index 6f1f3b675ff5..d51e0c401b48 100644 --- a/drivers/net/can/usb/kvaser_usb.c +++ b/drivers/net/can/usb/kvaser_usb.c @@ -787,10 +787,8 @@ static int kvaser_usb_simple_msg_async(struct kvaser_usb_net_priv *priv, int err; urb = usb_alloc_urb(0, GFP_ATOMIC); - if (!urb) { - netdev_err(netdev, "No memory left for URBs\n"); + if (!urb) return -ENOMEM; - } buf = kmalloc(sizeof(struct kvaser_msg), GFP_ATOMIC); if (!buf) { @@ -1393,8 +1391,6 @@ static int kvaser_usb_setup_rx_urbs(struct kvaser_usb *dev) urb = usb_alloc_urb(0, GFP_KERNEL); if (!urb) { - dev_warn(dev->udev->dev.parent, - "No memory left for URBs\n"); err = -ENOMEM; break; } @@ -1670,7 +1666,6 @@ static netdev_tx_t kvaser_usb_start_xmit(struct sk_buff *skb, urb = usb_alloc_urb(0, GFP_ATOMIC); if (!urb) { - netdev_err(netdev, "No memory left for URBs\n"); stats->tx_dropped++; dev_kfree_skb(skb); return NETDEV_TX_OK; diff --git a/drivers/net/can/usb/peak_usb/pcan_usb_core.c b/drivers/net/can/usb/peak_usb/pcan_usb_core.c index bfb91d8fa460..c06382cdfdfe 100644 --- a/drivers/net/can/usb/peak_usb/pcan_usb_core.c +++ b/drivers/net/can/usb/peak_usb/pcan_usb_core.c @@ -399,7 +399,6 @@ static int peak_usb_start(struct peak_usb_device *dev) /* create a URB, and a buffer for it, to receive usb messages */ urb = usb_alloc_urb(0, GFP_KERNEL); if (!urb) { - netdev_err(netdev, "No memory left for URBs\n"); err = -ENOMEM; break; } @@ -454,7 +453,6 @@ static int peak_usb_start(struct peak_usb_device *dev) /* create a URB and a buffer for it, to transmit usb messages */ urb = usb_alloc_urb(0, GFP_KERNEL); if (!urb) { - netdev_err(netdev, "No memory left for URBs\n"); err = -ENOMEM; break; } @@ -651,10 +649,8 @@ static int peak_usb_restart(struct peak_usb_device *dev) /* first allocate a urb to handle the asynchronous steps */ urb = usb_alloc_urb(0, GFP_ATOMIC); - if (!urb) { - netdev_err(dev->netdev, "no memory left for urb\n"); + if (!urb) return -ENOMEM; - } /* also allocate enough space for the commands to send */ buf = kmalloc(PCAN_USB_MAX_CMD_LEN, GFP_ATOMIC); diff --git a/drivers/net/can/usb/usb_8dev.c b/drivers/net/can/usb/usb_8dev.c index a731720f1d13..108a30e15097 100644 --- a/drivers/net/can/usb/usb_8dev.c +++ b/drivers/net/can/usb/usb_8dev.c @@ -623,10 +623,8 @@ static netdev_tx_t usb_8dev_start_xmit(struct sk_buff *skb, /* create a URB, and a buffer for it, and copy the data to the URB */ urb = usb_alloc_urb(0, GFP_ATOMIC); - if (!urb) { - netdev_err(netdev, "No memory left for URBs\n"); + if (!urb) goto nomem; - } buf = usb_alloc_coherent(priv->udev, size, GFP_ATOMIC, &urb->transfer_dma); @@ -748,7 +746,6 @@ static int usb_8dev_start(struct usb_8dev_priv *priv) /* create a URB, and a buffer for it */ urb = usb_alloc_urb(0, GFP_KERNEL); if (!urb) { - netdev_err(netdev, "No memory left for URBs\n"); err = -ENOMEM; break; } diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index bda37d336736..38ee10de7884 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -1681,7 +1681,8 @@ static int b53_switch_init(struct b53_device *dev) return 0; } -struct b53_device *b53_switch_alloc(struct device *base, struct b53_io_ops *ops, +struct b53_device *b53_switch_alloc(struct device *base, + const struct b53_io_ops *ops, void *priv) { struct dsa_switch *ds; diff --git a/drivers/net/dsa/b53/b53_mdio.c b/drivers/net/dsa/b53/b53_mdio.c index aa87c3fffdac..477a16b5660a 100644 --- a/drivers/net/dsa/b53/b53_mdio.c +++ b/drivers/net/dsa/b53/b53_mdio.c @@ -267,7 +267,7 @@ static int b53_mdio_phy_write16(struct b53_device *dev, int addr, int reg, return mdiobus_write_nested(bus, addr, reg, value); } -static struct b53_io_ops b53_mdio_ops = { +static const struct b53_io_ops b53_mdio_ops = { .read8 = b53_mdio_read8, .read16 = b53_mdio_read16, .read32 = b53_mdio_read32, diff --git a/drivers/net/dsa/b53/b53_mmap.c b/drivers/net/dsa/b53/b53_mmap.c index 77ffc4312808..cc9e6bd83e0e 100644 --- a/drivers/net/dsa/b53/b53_mmap.c +++ b/drivers/net/dsa/b53/b53_mmap.c @@ -208,7 +208,7 @@ static int b53_mmap_write64(struct b53_device *dev, u8 page, u8 reg, return 0; } -static struct b53_io_ops b53_mmap_ops = { +static const struct b53_io_ops b53_mmap_ops = { .read8 = b53_mmap_read8, .read16 = b53_mmap_read16, .read32 = b53_mmap_read32, diff --git a/drivers/net/dsa/b53/b53_priv.h b/drivers/net/dsa/b53/b53_priv.h index 835a744f206e..d268493a5fec 100644 --- a/drivers/net/dsa/b53/b53_priv.h +++ b/drivers/net/dsa/b53/b53_priv.h @@ -182,7 +182,8 @@ static inline int is_cpu_port(struct b53_device *dev, int port) return dev->cpu_port; } -struct b53_device *b53_switch_alloc(struct device *base, struct b53_io_ops *ops, +struct b53_device *b53_switch_alloc(struct device *base, + const struct b53_io_ops *ops, void *priv); int b53_switch_detect(struct b53_device *dev); diff --git a/drivers/net/dsa/b53/b53_spi.c b/drivers/net/dsa/b53/b53_spi.c index 2bda0b5f1578..f89f5308a99b 100644 --- a/drivers/net/dsa/b53/b53_spi.c +++ b/drivers/net/dsa/b53/b53_spi.c @@ -270,7 +270,7 @@ static int b53_spi_write64(struct b53_device *dev, u8 page, u8 reg, u64 value) return spi_write(spi, txbuf, sizeof(txbuf)); } -static struct b53_io_ops b53_spi_ops = { +static const struct b53_io_ops b53_spi_ops = { .read8 = b53_spi_read8, .read16 = b53_spi_read16, .read32 = b53_spi_read32, @@ -317,8 +317,6 @@ static int b53_spi_remove(struct spi_device *spi) static struct spi_driver b53_spi_driver = { .driver = { .name = "b53-switch", - .bus = &spi_bus_type, - .owner = THIS_MODULE, }, .probe = b53_spi_probe, .remove = b53_spi_remove, diff --git a/drivers/net/dsa/b53/b53_srab.c b/drivers/net/dsa/b53/b53_srab.c index 3e2d4a5fcd5a..8a62b6a69703 100644 --- a/drivers/net/dsa/b53/b53_srab.c +++ b/drivers/net/dsa/b53/b53_srab.c @@ -344,7 +344,7 @@ err: return ret; } -static struct b53_io_ops b53_srab_ops = { +static const struct b53_io_ops b53_srab_ops = { .read8 = b53_srab_read8, .read16 = b53_srab_read16, .read32 = b53_srab_read32, diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index d36aedde8cb9..a230fcba5b64 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -216,6 +216,118 @@ static int mv88e6xxx_write(struct mv88e6xxx_chip *chip, return 0; } +static int mv88e6xxx_phy_read(struct mv88e6xxx_chip *chip, int phy, + int reg, u16 *val) +{ + int addr = phy; /* PHY devices addresses start at 0x0 */ + + if (!chip->phy_ops) + return -EOPNOTSUPP; + + return chip->phy_ops->read(chip, addr, reg, val); +} + +static int mv88e6xxx_phy_write(struct mv88e6xxx_chip *chip, int phy, + int reg, u16 val) +{ + int addr = phy; /* PHY devices addresses start at 0x0 */ + + if (!chip->phy_ops) + return -EOPNOTSUPP; + + return chip->phy_ops->write(chip, addr, reg, val); +} + +static int mv88e6xxx_phy_page_get(struct mv88e6xxx_chip *chip, int phy, u8 page) +{ + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_PHY_PAGE)) + return -EOPNOTSUPP; + + return mv88e6xxx_phy_write(chip, phy, PHY_PAGE, page); +} + +static void mv88e6xxx_phy_page_put(struct mv88e6xxx_chip *chip, int phy) +{ + int err; + + /* Restore PHY page Copper 0x0 for access via the registered MDIO bus */ + err = mv88e6xxx_phy_write(chip, phy, PHY_PAGE, PHY_PAGE_COPPER); + if (unlikely(err)) { + dev_err(chip->dev, "failed to restore PHY %d page Copper (%d)\n", + phy, err); + } +} + +static int mv88e6xxx_phy_page_read(struct mv88e6xxx_chip *chip, int phy, + u8 page, int reg, u16 *val) +{ + int err; + + /* There is no paging for registers 22 */ + if (reg == PHY_PAGE) + return -EINVAL; + + err = mv88e6xxx_phy_page_get(chip, phy, page); + if (!err) { + err = mv88e6xxx_phy_read(chip, phy, reg, val); + mv88e6xxx_phy_page_put(chip, phy); + } + + return err; +} + +static int mv88e6xxx_phy_page_write(struct mv88e6xxx_chip *chip, int phy, + u8 page, int reg, u16 val) +{ + int err; + + /* There is no paging for registers 22 */ + if (reg == PHY_PAGE) + return -EINVAL; + + err = mv88e6xxx_phy_page_get(chip, phy, page); + if (!err) { + err = mv88e6xxx_phy_write(chip, phy, PHY_PAGE, page); + mv88e6xxx_phy_page_put(chip, phy); + } + + return err; +} + +static int mv88e6xxx_serdes_read(struct mv88e6xxx_chip *chip, int reg, u16 *val) +{ + return mv88e6xxx_phy_page_read(chip, ADDR_SERDES, SERDES_PAGE_FIBER, + reg, val); +} + +static int mv88e6xxx_serdes_write(struct mv88e6xxx_chip *chip, int reg, u16 val) +{ + return mv88e6xxx_phy_page_write(chip, ADDR_SERDES, SERDES_PAGE_FIBER, + reg, val); +} + +static int mv88e6xxx_wait(struct mv88e6xxx_chip *chip, int addr, int reg, + u16 mask) +{ + unsigned long timeout = jiffies + HZ / 10; + + while (time_before(jiffies, timeout)) { + u16 val; + int err; + + err = mv88e6xxx_read(chip, addr, reg, &val); + if (err) + return err; + + if (!(val & mask)) + return 0; + + usleep_range(1000, 2000); + } + + return -ETIMEDOUT; +} + /* Indirect write to single pointer-data register with an Update bit */ static int mv88e6xxx_update(struct mv88e6xxx_chip *chip, int addr, int reg, u16 update) @@ -260,22 +372,6 @@ static int _mv88e6xxx_reg_write(struct mv88e6xxx_chip *chip, int addr, return mv88e6xxx_write(chip, addr, reg, val); } -static int mv88e6xxx_mdio_read_direct(struct mv88e6xxx_chip *chip, - int addr, int regnum) -{ - if (addr >= 0) - return _mv88e6xxx_reg_read(chip, addr, regnum); - return 0xffff; -} - -static int mv88e6xxx_mdio_write_direct(struct mv88e6xxx_chip *chip, - int addr, int regnum, u16 val) -{ - if (addr >= 0) - return _mv88e6xxx_reg_write(chip, addr, regnum, val); - return 0; -} - static int mv88e6xxx_ppu_disable(struct mv88e6xxx_chip *chip) { int ret; @@ -400,34 +496,39 @@ static void mv88e6xxx_ppu_state_init(struct mv88e6xxx_chip *chip) chip->ppu_timer.function = mv88e6xxx_ppu_reenable_timer; } -static int mv88e6xxx_mdio_read_ppu(struct mv88e6xxx_chip *chip, int addr, - int regnum) +static int mv88e6xxx_phy_ppu_read(struct mv88e6xxx_chip *chip, int addr, + int reg, u16 *val) { - int ret; + int err; - ret = mv88e6xxx_ppu_access_get(chip); - if (ret >= 0) { - ret = _mv88e6xxx_reg_read(chip, addr, regnum); + err = mv88e6xxx_ppu_access_get(chip); + if (!err) { + err = mv88e6xxx_read(chip, addr, reg, val); mv88e6xxx_ppu_access_put(chip); } - return ret; + return err; } -static int mv88e6xxx_mdio_write_ppu(struct mv88e6xxx_chip *chip, int addr, - int regnum, u16 val) +static int mv88e6xxx_phy_ppu_write(struct mv88e6xxx_chip *chip, int addr, + int reg, u16 val) { - int ret; + int err; - ret = mv88e6xxx_ppu_access_get(chip); - if (ret >= 0) { - ret = _mv88e6xxx_reg_write(chip, addr, regnum, val); + err = mv88e6xxx_ppu_access_get(chip); + if (!err) { + err = mv88e6xxx_write(chip, addr, reg, val); mv88e6xxx_ppu_access_put(chip); } - return ret; + return err; } +static const struct mv88e6xxx_ops mv88e6xxx_phy_ppu_ops = { + .read = mv88e6xxx_phy_ppu_read, + .write = mv88e6xxx_phy_ppu_write, +}; + static bool mv88e6xxx_6065_family(struct mv88e6xxx_chip *chip) { return chip->info->family == MV88E6XXX_FAMILY_6065; @@ -819,130 +920,69 @@ static void mv88e6xxx_get_regs(struct dsa_switch *ds, int port, mutex_unlock(&chip->reg_lock); } -static int _mv88e6xxx_wait(struct mv88e6xxx_chip *chip, int reg, int offset, - u16 mask) -{ - unsigned long timeout = jiffies + HZ / 10; - - while (time_before(jiffies, timeout)) { - int ret; - - ret = _mv88e6xxx_reg_read(chip, reg, offset); - if (ret < 0) - return ret; - if (!(ret & mask)) - return 0; - - usleep_range(1000, 2000); - } - return -ETIMEDOUT; -} - -static int mv88e6xxx_mdio_wait(struct mv88e6xxx_chip *chip) -{ - return _mv88e6xxx_wait(chip, REG_GLOBAL2, GLOBAL2_SMI_OP, - GLOBAL2_SMI_OP_BUSY); -} - static int _mv88e6xxx_atu_wait(struct mv88e6xxx_chip *chip) { - return _mv88e6xxx_wait(chip, REG_GLOBAL, GLOBAL_ATU_OP, - GLOBAL_ATU_OP_BUSY); -} - -static int mv88e6xxx_mdio_read_indirect(struct mv88e6xxx_chip *chip, - int addr, int regnum) -{ - int ret; - - ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, GLOBAL2_SMI_OP, - GLOBAL2_SMI_OP_22_READ | (addr << 5) | - regnum); - if (ret < 0) - return ret; - - ret = mv88e6xxx_mdio_wait(chip); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_reg_read(chip, REG_GLOBAL2, GLOBAL2_SMI_DATA); - - return ret; -} - -static int mv88e6xxx_mdio_write_indirect(struct mv88e6xxx_chip *chip, - int addr, int regnum, u16 val) -{ - int ret; - - ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, GLOBAL2_SMI_DATA, val); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, GLOBAL2_SMI_OP, - GLOBAL2_SMI_OP_22_WRITE | (addr << 5) | - regnum); - - return mv88e6xxx_mdio_wait(chip); + return mv88e6xxx_wait(chip, REG_GLOBAL, GLOBAL_ATU_OP, + GLOBAL_ATU_OP_BUSY); } static int mv88e6xxx_get_eee(struct dsa_switch *ds, int port, struct ethtool_eee *e) { struct mv88e6xxx_chip *chip = ds_to_priv(ds); - int reg; + u16 reg; + int err; if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_EEE)) return -EOPNOTSUPP; mutex_lock(&chip->reg_lock); - reg = mv88e6xxx_mdio_read_indirect(chip, port, 16); - if (reg < 0) + err = mv88e6xxx_phy_read(chip, port, 16, ®); + if (err) goto out; e->eee_enabled = !!(reg & 0x0200); e->tx_lpi_enabled = !!(reg & 0x0100); - reg = _mv88e6xxx_reg_read(chip, REG_PORT(port), PORT_STATUS); - if (reg < 0) + err = mv88e6xxx_read(chip, REG_PORT(port), PORT_STATUS, ®); + if (err) goto out; e->eee_active = !!(reg & PORT_STATUS_EEE); - reg = 0; - out: mutex_unlock(&chip->reg_lock); - return reg; + + return err; } static int mv88e6xxx_set_eee(struct dsa_switch *ds, int port, struct phy_device *phydev, struct ethtool_eee *e) { struct mv88e6xxx_chip *chip = ds_to_priv(ds); - int reg; - int ret; + u16 reg; + int err; if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_EEE)) return -EOPNOTSUPP; mutex_lock(&chip->reg_lock); - ret = mv88e6xxx_mdio_read_indirect(chip, port, 16); - if (ret < 0) + err = mv88e6xxx_phy_read(chip, port, 16, ®); + if (err) goto out; - reg = ret & ~0x0300; + reg &= ~0x0300; if (e->eee_enabled) reg |= 0x0200; if (e->tx_lpi_enabled) reg |= 0x0100; - ret = mv88e6xxx_mdio_write_indirect(chip, port, 16, reg); + err = mv88e6xxx_phy_write(chip, port, 16, reg); out: mutex_unlock(&chip->reg_lock); - return ret; + return err; } static int _mv88e6xxx_atu_cmd(struct mv88e6xxx_chip *chip, u16 fid, u16 cmd) @@ -1227,8 +1267,8 @@ static int _mv88e6xxx_port_pvid_set(struct mv88e6xxx_chip *chip, static int _mv88e6xxx_vtu_wait(struct mv88e6xxx_chip *chip) { - return _mv88e6xxx_wait(chip, REG_GLOBAL, GLOBAL_VTU_OP, - GLOBAL_VTU_OP_BUSY); + return mv88e6xxx_wait(chip, REG_GLOBAL, GLOBAL_VTU_OP, + GLOBAL_VTU_OP_BUSY); } static int _mv88e6xxx_vtu_cmd(struct mv88e6xxx_chip *chip, u16 op) @@ -2302,38 +2342,6 @@ static void mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port) mutex_unlock(&chip->reg_lock); } -static int _mv88e6xxx_mdio_page_write(struct mv88e6xxx_chip *chip, - int port, int page, int reg, int val) -{ - int ret; - - ret = mv88e6xxx_mdio_write_indirect(chip, port, 0x16, page); - if (ret < 0) - goto restore_page_0; - - ret = mv88e6xxx_mdio_write_indirect(chip, port, reg, val); -restore_page_0: - mv88e6xxx_mdio_write_indirect(chip, port, 0x16, 0x0); - - return ret; -} - -static int _mv88e6xxx_mdio_page_read(struct mv88e6xxx_chip *chip, - int port, int page, int reg) -{ - int ret; - - ret = mv88e6xxx_mdio_write_indirect(chip, port, 0x16, page); - if (ret < 0) - goto restore_page_0; - - ret = mv88e6xxx_mdio_read_indirect(chip, port, reg); -restore_page_0: - mv88e6xxx_mdio_write_indirect(chip, port, 0x16, 0x0); - - return ret; -} - static int mv88e6xxx_switch_reset(struct mv88e6xxx_chip *chip) { bool ppu_active = mv88e6xxx_has(chip, MV88E6XXX_FLAG_PPU_ACTIVE); @@ -2396,23 +2404,22 @@ static int mv88e6xxx_switch_reset(struct mv88e6xxx_chip *chip) return ret; } -static int mv88e6xxx_power_on_serdes(struct mv88e6xxx_chip *chip) +static int mv88e6xxx_serdes_power_on(struct mv88e6xxx_chip *chip) { - int ret; + u16 val; + int err; - ret = _mv88e6xxx_mdio_page_read(chip, REG_FIBER_SERDES, - PAGE_FIBER_SERDES, MII_BMCR); - if (ret < 0) - return ret; + /* Clear Power Down bit */ + err = mv88e6xxx_serdes_read(chip, MII_BMCR, &val); + if (err) + return err; - if (ret & BMCR_PDOWN) { - ret &= ~BMCR_PDOWN; - ret = _mv88e6xxx_mdio_page_write(chip, REG_FIBER_SERDES, - PAGE_FIBER_SERDES, MII_BMCR, - ret); + if (val & BMCR_PDOWN) { + val &= ~BMCR_PDOWN; + err = mv88e6xxx_serdes_write(chip, MII_BMCR, val); } - return ret; + return err; } static int mv88e6xxx_port_read(struct mv88e6xxx_chip *chip, int port, @@ -2535,7 +2542,7 @@ static int mv88e6xxx_setup_port(struct mv88e6xxx_chip *chip, int port) /* If this port is connected to a SerDes, make sure the SerDes is not * powered down. */ - if (mv88e6xxx_6352_family(chip)) { + if (mv88e6xxx_has(chip, MV88E6XXX_FLAGS_SERDES)) { ret = _mv88e6xxx_reg_read(chip, REG_PORT(port), PORT_STATUS); if (ret < 0) return ret; @@ -2543,7 +2550,7 @@ static int mv88e6xxx_setup_port(struct mv88e6xxx_chip *chip, int port) if ((ret == PORT_STATUS_CMODE_100BASE_X) || (ret == PORT_STATUS_CMODE_1000BASE_X) || (ret == PORT_STATUS_CMODE_SGMII)) { - ret = mv88e6xxx_power_on_serdes(chip); + ret = mv88e6xxx_serdes_power_on(chip); if (ret < 0) return ret; } @@ -2949,8 +2956,8 @@ static int mv88e6xxx_g2_clear_irl(struct mv88e6xxx_chip *chip) break; /* Wait for the operation to complete */ - err = _mv88e6xxx_wait(chip, REG_GLOBAL2, GLOBAL2_IRL_CMD, - GLOBAL2_IRL_CMD_BUSY); + err = mv88e6xxx_wait(chip, REG_GLOBAL2, GLOBAL2_IRL_CMD, + GLOBAL2_IRL_CMD_BUSY); if (err) break; } @@ -3004,9 +3011,9 @@ static int mv88e6xxx_g2_clear_pot(struct mv88e6xxx_chip *chip) static int mv88e6xxx_g2_eeprom_wait(struct mv88e6xxx_chip *chip) { - return _mv88e6xxx_wait(chip, REG_GLOBAL2, GLOBAL2_EEPROM_CMD, - GLOBAL2_EEPROM_CMD_BUSY | - GLOBAL2_EEPROM_CMD_RUNNING); + return mv88e6xxx_wait(chip, REG_GLOBAL2, GLOBAL2_EEPROM_CMD, + GLOBAL2_EEPROM_CMD_BUSY | + GLOBAL2_EEPROM_CMD_RUNNING); } static int mv88e6xxx_g2_eeprom_cmd(struct mv88e6xxx_chip *chip, u16 cmd) @@ -3054,6 +3061,62 @@ static int mv88e6xxx_g2_eeprom_write16(struct mv88e6xxx_chip *chip, return mv88e6xxx_g2_eeprom_cmd(chip, cmd); } +static int mv88e6xxx_g2_smi_phy_wait(struct mv88e6xxx_chip *chip) +{ + return mv88e6xxx_wait(chip, REG_GLOBAL2, GLOBAL2_SMI_PHY_CMD, + GLOBAL2_SMI_PHY_CMD_BUSY); +} + +static int mv88e6xxx_g2_smi_phy_cmd(struct mv88e6xxx_chip *chip, u16 cmd) +{ + int err; + + err = mv88e6xxx_write(chip, REG_GLOBAL2, GLOBAL2_SMI_PHY_CMD, cmd); + if (err) + return err; + + return mv88e6xxx_g2_smi_phy_wait(chip); +} + +static int mv88e6xxx_g2_smi_phy_read(struct mv88e6xxx_chip *chip, int addr, + int reg, u16 *val) +{ + u16 cmd = GLOBAL2_SMI_PHY_CMD_OP_22_READ_DATA | (addr << 5) | reg; + int err; + + err = mv88e6xxx_g2_smi_phy_wait(chip); + if (err) + return err; + + err = mv88e6xxx_g2_smi_phy_cmd(chip, cmd); + if (err) + return err; + + return mv88e6xxx_read(chip, REG_GLOBAL2, GLOBAL2_SMI_PHY_DATA, val); +} + +static int mv88e6xxx_g2_smi_phy_write(struct mv88e6xxx_chip *chip, int addr, + int reg, u16 val) +{ + u16 cmd = GLOBAL2_SMI_PHY_CMD_OP_22_WRITE_DATA | (addr << 5) | reg; + int err; + + err = mv88e6xxx_g2_smi_phy_wait(chip); + if (err) + return err; + + err = mv88e6xxx_write(chip, REG_GLOBAL2, GLOBAL2_SMI_PHY_DATA, val); + if (err) + return err; + + return mv88e6xxx_g2_smi_phy_cmd(chip, cmd); +} + +static const struct mv88e6xxx_ops mv88e6xxx_g2_smi_phy_ops = { + .read = mv88e6xxx_g2_smi_phy_read, + .write = mv88e6xxx_g2_smi_phy_write, +}; + static int mv88e6xxx_g2_setup(struct mv88e6xxx_chip *chip) { u16 reg; @@ -3187,82 +3250,35 @@ static int mv88e6xxx_set_addr(struct dsa_switch *ds, u8 *addr) return err; } -static int mv88e6xxx_mdio_page_read(struct dsa_switch *ds, int port, int page, - int reg) -{ - struct mv88e6xxx_chip *chip = ds_to_priv(ds); - int ret; - - mutex_lock(&chip->reg_lock); - ret = _mv88e6xxx_mdio_page_read(chip, port, page, reg); - mutex_unlock(&chip->reg_lock); - - return ret; -} - -static int mv88e6xxx_mdio_page_write(struct dsa_switch *ds, int port, int page, - int reg, int val) -{ - struct mv88e6xxx_chip *chip = ds_to_priv(ds); - int ret; - - mutex_lock(&chip->reg_lock); - ret = _mv88e6xxx_mdio_page_write(chip, port, page, reg, val); - mutex_unlock(&chip->reg_lock); - - return ret; -} - -static int mv88e6xxx_port_to_mdio_addr(struct mv88e6xxx_chip *chip, int port) -{ - if (port >= 0 && port < chip->info->num_ports) - return port; - return -EINVAL; -} - -static int mv88e6xxx_mdio_read(struct mii_bus *bus, int port, int regnum) +static int mv88e6xxx_mdio_read(struct mii_bus *bus, int phy, int reg) { struct mv88e6xxx_chip *chip = bus->priv; - int addr = mv88e6xxx_port_to_mdio_addr(chip, port); - int ret; + u16 val; + int err; - if (addr < 0) + if (phy >= chip->info->num_ports) return 0xffff; mutex_lock(&chip->reg_lock); - - if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_PPU)) - ret = mv88e6xxx_mdio_read_ppu(chip, addr, regnum); - else if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_SMI_PHY)) - ret = mv88e6xxx_mdio_read_indirect(chip, addr, regnum); - else - ret = mv88e6xxx_mdio_read_direct(chip, addr, regnum); - + err = mv88e6xxx_phy_read(chip, phy, reg, &val); mutex_unlock(&chip->reg_lock); - return ret; + + return err ? err : val; } -static int mv88e6xxx_mdio_write(struct mii_bus *bus, int port, int regnum, - u16 val) +static int mv88e6xxx_mdio_write(struct mii_bus *bus, int phy, int reg, u16 val) { struct mv88e6xxx_chip *chip = bus->priv; - int addr = mv88e6xxx_port_to_mdio_addr(chip, port); - int ret; + int err; - if (addr < 0) + if (phy >= chip->info->num_ports) return 0xffff; mutex_lock(&chip->reg_lock); - - if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_PPU)) - ret = mv88e6xxx_mdio_write_ppu(chip, addr, regnum, val); - else if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_SMI_PHY)) - ret = mv88e6xxx_mdio_write_indirect(chip, addr, regnum, val); - else - ret = mv88e6xxx_mdio_write_direct(chip, addr, regnum, val); - + err = mv88e6xxx_phy_write(chip, phy, reg, val); mutex_unlock(&chip->reg_lock); - return ret; + + return err; } static int mv88e6xxx_mdio_register(struct mv88e6xxx_chip *chip, @@ -3272,9 +3288,6 @@ static int mv88e6xxx_mdio_register(struct mv88e6xxx_chip *chip, struct mii_bus *bus; int err; - if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_PPU)) - mv88e6xxx_ppu_state_init(chip); - if (np) chip->mdio_np = of_get_child_by_name(np, "mdio"); @@ -3330,44 +3343,42 @@ static void mv88e6xxx_mdio_unregister(struct mv88e6xxx_chip *chip) static int mv88e61xx_get_temp(struct dsa_switch *ds, int *temp) { struct mv88e6xxx_chip *chip = ds_to_priv(ds); + u16 val; int ret; - int val; *temp = 0; mutex_lock(&chip->reg_lock); - ret = mv88e6xxx_mdio_write_direct(chip, 0x0, 0x16, 0x6); + ret = mv88e6xxx_phy_write(chip, 0x0, 0x16, 0x6); if (ret < 0) goto error; /* Enable temperature sensor */ - ret = mv88e6xxx_mdio_read_direct(chip, 0x0, 0x1a); + ret = mv88e6xxx_phy_read(chip, 0x0, 0x1a, &val); if (ret < 0) goto error; - ret = mv88e6xxx_mdio_write_direct(chip, 0x0, 0x1a, ret | (1 << 5)); + ret = mv88e6xxx_phy_write(chip, 0x0, 0x1a, val | (1 << 5)); if (ret < 0) goto error; /* Wait for temperature to stabilize */ usleep_range(10000, 12000); - val = mv88e6xxx_mdio_read_direct(chip, 0x0, 0x1a); - if (val < 0) { - ret = val; + ret = mv88e6xxx_phy_read(chip, 0x0, 0x1a, &val); + if (ret < 0) goto error; - } /* Disable temperature sensor */ - ret = mv88e6xxx_mdio_write_direct(chip, 0x0, 0x1a, ret & ~(1 << 5)); + ret = mv88e6xxx_phy_write(chip, 0x0, 0x1a, val & ~(1 << 5)); if (ret < 0) goto error; *temp = ((val & 0x1f) - 5) * 5; error: - mv88e6xxx_mdio_write_direct(chip, 0x0, 0x16, 0x0); + mv88e6xxx_phy_write(chip, 0x0, 0x16, 0x0); mutex_unlock(&chip->reg_lock); return ret; } @@ -3376,15 +3387,18 @@ static int mv88e63xx_get_temp(struct dsa_switch *ds, int *temp) { struct mv88e6xxx_chip *chip = ds_to_priv(ds); int phy = mv88e6xxx_6320_family(chip) ? 3 : 0; + u16 val; int ret; *temp = 0; - ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 27); + mutex_lock(&chip->reg_lock); + ret = mv88e6xxx_phy_page_read(chip, phy, 6, 27, &val); + mutex_unlock(&chip->reg_lock); if (ret < 0) return ret; - *temp = (ret & 0xff) - 25; + *temp = (val & 0xff) - 25; return 0; } @@ -3406,6 +3420,7 @@ static int mv88e6xxx_get_temp_limit(struct dsa_switch *ds, int *temp) { struct mv88e6xxx_chip *chip = ds_to_priv(ds); int phy = mv88e6xxx_6320_family(chip) ? 3 : 0; + u16 val; int ret; if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_TEMP_LIMIT)) @@ -3413,11 +3428,13 @@ static int mv88e6xxx_get_temp_limit(struct dsa_switch *ds, int *temp) *temp = 0; - ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 26); + mutex_lock(&chip->reg_lock); + ret = mv88e6xxx_phy_page_read(chip, phy, 6, 26, &val); + mutex_unlock(&chip->reg_lock); if (ret < 0) return ret; - *temp = (((ret >> 8) & 0x1f) * 5) - 25; + *temp = (((val >> 8) & 0x1f) * 5) - 25; return 0; } @@ -3426,23 +3443,30 @@ static int mv88e6xxx_set_temp_limit(struct dsa_switch *ds, int temp) { struct mv88e6xxx_chip *chip = ds_to_priv(ds); int phy = mv88e6xxx_6320_family(chip) ? 3 : 0; - int ret; + u16 val; + int err; if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_TEMP_LIMIT)) return -EOPNOTSUPP; - ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 26); - if (ret < 0) - return ret; + mutex_lock(&chip->reg_lock); + err = mv88e6xxx_phy_page_read(chip, phy, 6, 26, &val); + if (err) + goto unlock; temp = clamp_val(DIV_ROUND_CLOSEST(temp, 5) + 5, 0, 0x1f); - return mv88e6xxx_mdio_page_write(ds, phy, 6, 26, - (ret & 0xe0ff) | (temp << 8)); + err = mv88e6xxx_phy_page_write(chip, phy, 6, 26, + (val & 0xe0ff) | (temp << 8)); +unlock: + mutex_unlock(&chip->reg_lock); + + return err; } static int mv88e6xxx_get_temp_alarm(struct dsa_switch *ds, bool *alarm) { struct mv88e6xxx_chip *chip = ds_to_priv(ds); int phy = mv88e6xxx_6320_family(chip) ? 3 : 0; + u16 val; int ret; if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_TEMP_LIMIT)) @@ -3450,11 +3474,13 @@ static int mv88e6xxx_get_temp_alarm(struct dsa_switch *ds, bool *alarm) *alarm = false; - ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 26); + mutex_lock(&chip->reg_lock); + ret = mv88e6xxx_phy_page_read(chip, phy, 6, 26, &val); + mutex_unlock(&chip->reg_lock); if (ret < 0) return ret; - *alarm = !!(ret & 0x40); + *alarm = !!(val & 0x40); return 0; } @@ -3871,6 +3897,23 @@ static struct mv88e6xxx_chip *mv88e6xxx_alloc_chip(struct device *dev) return chip; } +static const struct mv88e6xxx_ops mv88e6xxx_phy_ops = { + .read = mv88e6xxx_read, + .write = mv88e6xxx_write, +}; + +static void mv88e6xxx_phy_init(struct mv88e6xxx_chip *chip) +{ + if (mv88e6xxx_has(chip, MV88E6XXX_FLAGS_SMI_PHY)) { + chip->phy_ops = &mv88e6xxx_g2_smi_phy_ops; + } else if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_PPU)) { + chip->phy_ops = &mv88e6xxx_phy_ppu_ops; + mv88e6xxx_ppu_state_init(chip); + } else { + chip->phy_ops = &mv88e6xxx_phy_ops; + } +} + static int mv88e6xxx_smi_init(struct mv88e6xxx_chip *chip, struct mii_bus *bus, int sw_addr) { @@ -3880,7 +3923,7 @@ static int mv88e6xxx_smi_init(struct mv88e6xxx_chip *chip, if (sw_addr == 0) chip->smi_ops = &mv88e6xxx_smi_single_chip_ops; - else if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_MULTI_CHIP)) + else if (mv88e6xxx_has(chip, MV88E6XXX_FLAGS_MULTI_CHIP)) chip->smi_ops = &mv88e6xxx_smi_multi_chip_ops; else return -EINVAL; @@ -3918,6 +3961,8 @@ static const char *mv88e6xxx_drv_probe(struct device *dsa_dev, if (err) goto free; + mv88e6xxx_phy_init(chip); + err = mv88e6xxx_mdio_register(chip, NULL); if (err) goto free; @@ -4019,6 +4064,8 @@ static int mv88e6xxx_probe(struct mdio_device *mdiodev) if (err) return err; + mv88e6xxx_phy_init(chip); + chip->reset = devm_gpiod_get_optional(dev, "reset", GPIOD_ASIS); if (IS_ERR(chip->reset)) return PTR_ERR(chip->reset); diff --git a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h index 48d6ea77f9bd..1f9bab5891b1 100644 --- a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h +++ b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h @@ -30,9 +30,12 @@ #define SMI_CMD_OP_45_READ_DATA_INC ((3 << 10) | SMI_CMD_BUSY) #define SMI_DATA 0x01 -/* Fiber/SERDES Registers are located at SMI address F, page 1 */ -#define REG_FIBER_SERDES 0x0f -#define PAGE_FIBER_SERDES 0x01 +/* PHY Registers */ +#define PHY_PAGE 0x16 +#define PHY_PAGE_COPPER 0x00 + +#define ADDR_SERDES 0x0f +#define SERDES_PAGE_FIBER 0x01 #define REG_PORT(p) (0x10 + (p)) #define PORT_STATUS 0x00 @@ -329,17 +332,16 @@ #define GLOBAL2_EEPROM_DATA 0x15 #define GLOBAL2_PTP_AVB_OP 0x16 #define GLOBAL2_PTP_AVB_DATA 0x17 -#define GLOBAL2_SMI_OP 0x18 -#define GLOBAL2_SMI_OP_BUSY BIT(15) -#define GLOBAL2_SMI_OP_CLAUSE_22 BIT(12) -#define GLOBAL2_SMI_OP_22_WRITE ((1 << 10) | GLOBAL2_SMI_OP_BUSY | \ - GLOBAL2_SMI_OP_CLAUSE_22) -#define GLOBAL2_SMI_OP_22_READ ((2 << 10) | GLOBAL2_SMI_OP_BUSY | \ - GLOBAL2_SMI_OP_CLAUSE_22) -#define GLOBAL2_SMI_OP_45_WRITE_ADDR ((0 << 10) | GLOBAL2_SMI_OP_BUSY) -#define GLOBAL2_SMI_OP_45_WRITE_DATA ((1 << 10) | GLOBAL2_SMI_OP_BUSY) -#define GLOBAL2_SMI_OP_45_READ_DATA ((2 << 10) | GLOBAL2_SMI_OP_BUSY) -#define GLOBAL2_SMI_DATA 0x19 +#define GLOBAL2_SMI_PHY_CMD 0x18 +#define GLOBAL2_SMI_PHY_CMD_BUSY BIT(15) +#define GLOBAL2_SMI_PHY_CMD_MODE_22 BIT(12) +#define GLOBAL2_SMI_PHY_CMD_OP_22_WRITE_DATA ((0x1 << 10) | \ + GLOBAL2_SMI_PHY_CMD_MODE_22 | \ + GLOBAL2_SMI_PHY_CMD_BUSY) +#define GLOBAL2_SMI_PHY_CMD_OP_22_READ_DATA ((0x2 << 10) | \ + GLOBAL2_SMI_PHY_CMD_MODE_22 | \ + GLOBAL2_SMI_PHY_CMD_BUSY) +#define GLOBAL2_SMI_PHY_DATA 0x19 #define GLOBAL2_SCRATCH_MISC 0x1a #define GLOBAL2_SCRATCH_BUSY BIT(15) #define GLOBAL2_SCRATCH_REGISTER_SHIFT 8 @@ -388,6 +390,21 @@ enum mv88e6xxx_cap { */ MV88E6XXX_CAP_EEE, + /* Multi-chip Addressing Mode. + * Some chips respond to only 2 registers of its own SMI device address + * when it is non-zero, and use indirect access to internal registers. + */ + MV88E6XXX_CAP_SMI_CMD, /* (0x00) SMI Command */ + MV88E6XXX_CAP_SMI_DATA, /* (0x01) SMI Data */ + + /* PHY Registers. + */ + MV88E6XXX_CAP_PHY_PAGE, /* (0x16) Page Register */ + + /* Fiber/SERDES Registers (SMI address F). + */ + MV88E6XXX_CAP_SERDES, + /* Switch Global 2 Registers. * The device contains a second set of global 16-bit registers. */ @@ -402,12 +419,8 @@ enum mv88e6xxx_cap { MV88E6XXX_CAP_G2_POT, /* (0x0f) Priority Override Table */ MV88E6XXX_CAP_G2_EEPROM_CMD, /* (0x14) EEPROM Command */ MV88E6XXX_CAP_G2_EEPROM_DATA, /* (0x15) EEPROM Data */ - - /* Multi-chip Addressing Mode. - * Some chips require an indirect SMI access when their SMI device - * address is not zero. See SMI_CMD and SMI_DATA. - */ - MV88E6XXX_CAP_MULTI_CHIP, + MV88E6XXX_CAP_G2_SMI_PHY_CMD, /* (0x18) SMI PHY Command */ + MV88E6XXX_CAP_G2_SMI_PHY_DATA, /* (0x19) SMI PHY Data */ /* PHY Polling Unit. * See GLOBAL_CONTROL_PPU_ENABLE and GLOBAL_STATUS_PPU_POLLING. @@ -415,12 +428,6 @@ enum mv88e6xxx_cap { MV88E6XXX_CAP_PPU, MV88E6XXX_CAP_PPU_ACTIVE, - /* SMI PHY Command and Data registers. - * This requires an indirect access to PHY registers through - * GLOBAL2_SMI_OP, otherwise direct access to PHY registers is done. - */ - MV88E6XXX_CAP_SMI_PHY, - /* Per VLAN Spanning Tree Unit (STU). * The Port State database, if present, is accessed through VTU * operations and dedicated SID registers. See GLOBAL_VTU_SID. @@ -441,6 +448,14 @@ enum mv88e6xxx_cap { /* Bitmask of capabilities */ #define MV88E6XXX_FLAG_EEE BIT(MV88E6XXX_CAP_EEE) + +#define MV88E6XXX_FLAG_SMI_CMD BIT(MV88E6XXX_CAP_SMI_CMD) +#define MV88E6XXX_FLAG_SMI_DATA BIT(MV88E6XXX_CAP_SMI_DATA) + +#define MV88E6XXX_FLAG_PHY_PAGE BIT(MV88E6XXX_CAP_PHY_PAGE) + +#define MV88E6XXX_FLAG_SERDES BIT(MV88E6XXX_CAP_SERDES) + #define MV88E6XXX_FLAG_GLOBAL2 BIT(MV88E6XXX_CAP_GLOBAL2) #define MV88E6XXX_FLAG_G2_MGMT_EN_2X BIT(MV88E6XXX_CAP_G2_MGMT_EN_2X) #define MV88E6XXX_FLAG_G2_MGMT_EN_0X BIT(MV88E6XXX_CAP_G2_MGMT_EN_0X) @@ -452,10 +467,11 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAG_G2_POT BIT(MV88E6XXX_CAP_G2_POT) #define MV88E6XXX_FLAG_G2_EEPROM_CMD BIT(MV88E6XXX_CAP_G2_EEPROM_CMD) #define MV88E6XXX_FLAG_G2_EEPROM_DATA BIT(MV88E6XXX_CAP_G2_EEPROM_DATA) -#define MV88E6XXX_FLAG_MULTI_CHIP BIT(MV88E6XXX_CAP_MULTI_CHIP) +#define MV88E6XXX_FLAG_G2_SMI_PHY_CMD BIT(MV88E6XXX_CAP_G2_SMI_PHY_CMD) +#define MV88E6XXX_FLAG_G2_SMI_PHY_DATA BIT(MV88E6XXX_CAP_G2_SMI_PHY_DATA) + #define MV88E6XXX_FLAG_PPU BIT(MV88E6XXX_CAP_PPU) #define MV88E6XXX_FLAG_PPU_ACTIVE BIT(MV88E6XXX_CAP_PPU_ACTIVE) -#define MV88E6XXX_FLAG_SMI_PHY BIT(MV88E6XXX_CAP_SMI_PHY) #define MV88E6XXX_FLAG_STU BIT(MV88E6XXX_CAP_STU) #define MV88E6XXX_FLAG_TEMP BIT(MV88E6XXX_CAP_TEMP) #define MV88E6XXX_FLAG_TEMP_LIMIT BIT(MV88E6XXX_CAP_TEMP_LIMIT) @@ -471,28 +487,43 @@ enum mv88e6xxx_cap { (MV88E6XXX_FLAG_G2_IRL_CMD | \ MV88E6XXX_FLAG_G2_IRL_DATA) +/* Multi-chip Addressing Mode */ +#define MV88E6XXX_FLAGS_MULTI_CHIP \ + (MV88E6XXX_FLAG_SMI_CMD | \ + MV88E6XXX_FLAG_SMI_DATA) + /* Cross-chip Port VLAN Table */ #define MV88E6XXX_FLAGS_PVT \ (MV88E6XXX_FLAG_G2_PVT_ADDR | \ MV88E6XXX_FLAG_G2_PVT_DATA) +/* Fiber/SERDES Registers at SMI address F, page 1 */ +#define MV88E6XXX_FLAGS_SERDES \ + (MV88E6XXX_FLAG_PHY_PAGE | \ + MV88E6XXX_FLAG_SERDES) + +/* Indirect PHY access via Global2 SMI PHY registers */ +#define MV88E6XXX_FLAGS_SMI_PHY \ + (MV88E6XXX_FLAG_G2_SMI_PHY_CMD |\ + MV88E6XXX_FLAG_G2_SMI_PHY_DATA) + #define MV88E6XXX_FLAGS_FAMILY_6095 \ (MV88E6XXX_FLAG_GLOBAL2 | \ MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ - MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU | \ - MV88E6XXX_FLAG_VTU) + MV88E6XXX_FLAG_VTU | \ + MV88E6XXX_FLAGS_MULTI_CHIP) #define MV88E6XXX_FLAGS_FAMILY_6097 \ (MV88E6XXX_FLAG_GLOBAL2 | \ MV88E6XXX_FLAG_G2_MGMT_EN_2X | \ MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ MV88E6XXX_FLAG_G2_POT | \ - MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU | \ MV88E6XXX_FLAG_STU | \ MV88E6XXX_FLAG_VTU | \ MV88E6XXX_FLAGS_IRL | \ + MV88E6XXX_FLAGS_MULTI_CHIP | \ MV88E6XXX_FLAGS_PVT) #define MV88E6XXX_FLAGS_FAMILY_6165 \ @@ -501,17 +532,17 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ MV88E6XXX_FLAG_G2_SWITCH_MAC | \ MV88E6XXX_FLAG_G2_POT | \ - MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_STU | \ MV88E6XXX_FLAG_TEMP | \ MV88E6XXX_FLAG_VTU | \ MV88E6XXX_FLAGS_IRL | \ + MV88E6XXX_FLAGS_MULTI_CHIP | \ MV88E6XXX_FLAGS_PVT) #define MV88E6XXX_FLAGS_FAMILY_6185 \ (MV88E6XXX_FLAG_GLOBAL2 | \ MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ - MV88E6XXX_FLAG_MULTI_CHIP | \ + MV88E6XXX_FLAGS_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU | \ MV88E6XXX_FLAG_VTU) @@ -522,15 +553,15 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ MV88E6XXX_FLAG_G2_SWITCH_MAC | \ MV88E6XXX_FLAG_G2_POT | \ - MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ - MV88E6XXX_FLAG_SMI_PHY | \ MV88E6XXX_FLAG_TEMP | \ MV88E6XXX_FLAG_TEMP_LIMIT | \ MV88E6XXX_FLAG_VTU | \ MV88E6XXX_FLAGS_EEPROM16 | \ MV88E6XXX_FLAGS_IRL | \ - MV88E6XXX_FLAGS_PVT) + MV88E6XXX_FLAGS_MULTI_CHIP | \ + MV88E6XXX_FLAGS_PVT | \ + MV88E6XXX_FLAGS_SMI_PHY) #define MV88E6XXX_FLAGS_FAMILY_6351 \ (MV88E6XXX_FLAG_GLOBAL2 | \ @@ -538,14 +569,14 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ MV88E6XXX_FLAG_G2_SWITCH_MAC | \ MV88E6XXX_FLAG_G2_POT | \ - MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ - MV88E6XXX_FLAG_SMI_PHY | \ MV88E6XXX_FLAG_STU | \ MV88E6XXX_FLAG_TEMP | \ MV88E6XXX_FLAG_VTU | \ MV88E6XXX_FLAGS_IRL | \ - MV88E6XXX_FLAGS_PVT) + MV88E6XXX_FLAGS_MULTI_CHIP | \ + MV88E6XXX_FLAGS_PVT | \ + MV88E6XXX_FLAGS_SMI_PHY) #define MV88E6XXX_FLAGS_FAMILY_6352 \ (MV88E6XXX_FLAG_EEE | \ @@ -554,16 +585,17 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ MV88E6XXX_FLAG_G2_SWITCH_MAC | \ MV88E6XXX_FLAG_G2_POT | \ - MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ - MV88E6XXX_FLAG_SMI_PHY | \ MV88E6XXX_FLAG_STU | \ MV88E6XXX_FLAG_TEMP | \ MV88E6XXX_FLAG_TEMP_LIMIT | \ MV88E6XXX_FLAG_VTU | \ MV88E6XXX_FLAGS_EEPROM16 | \ MV88E6XXX_FLAGS_IRL | \ - MV88E6XXX_FLAGS_PVT) + MV88E6XXX_FLAGS_MULTI_CHIP | \ + MV88E6XXX_FLAGS_PVT | \ + MV88E6XXX_FLAGS_SERDES | \ + MV88E6XXX_FLAGS_SMI_PHY) struct mv88e6xxx_info { enum mv88e6xxx_family family; @@ -623,6 +655,7 @@ struct mv88e6xxx_chip { /* Handles automatic disabling and re-enabling of the PHY * polling unit. */ + const struct mv88e6xxx_ops *phy_ops; struct mutex ppu_mutex; int ppu_disabled; struct work_struct ppu_work; diff --git a/drivers/net/ethernet/Kconfig b/drivers/net/ethernet/Kconfig index 2ffd63463299..8cc7467b6c1f 100644 --- a/drivers/net/ethernet/Kconfig +++ b/drivers/net/ethernet/Kconfig @@ -24,6 +24,7 @@ source "drivers/net/ethernet/agere/Kconfig" source "drivers/net/ethernet/allwinner/Kconfig" source "drivers/net/ethernet/alteon/Kconfig" source "drivers/net/ethernet/altera/Kconfig" +source "drivers/net/ethernet/amazon/Kconfig" source "drivers/net/ethernet/amd/Kconfig" source "drivers/net/ethernet/apm/Kconfig" source "drivers/net/ethernet/apple/Kconfig" diff --git a/drivers/net/ethernet/Makefile b/drivers/net/ethernet/Makefile index 1d349e9aa9a6..a09423df83f2 100644 --- a/drivers/net/ethernet/Makefile +++ b/drivers/net/ethernet/Makefile @@ -10,6 +10,7 @@ obj-$(CONFIG_NET_VENDOR_AGERE) += agere/ obj-$(CONFIG_NET_VENDOR_ALLWINNER) += allwinner/ obj-$(CONFIG_NET_VENDOR_ALTEON) += alteon/ obj-$(CONFIG_ALTERA_TSE) += altera/ +obj-$(CONFIG_NET_VENDOR_AMAZON) += amazon/ obj-$(CONFIG_NET_VENDOR_AMD) += amd/ obj-$(CONFIG_NET_XGENE) += apm/ obj-$(CONFIG_NET_VENDOR_APPLE) += apple/ diff --git a/drivers/net/ethernet/adi/bfin_mac.c b/drivers/net/ethernet/adi/bfin_mac.c index 38eaea18da23..00f9ee3fc3e5 100644 --- a/drivers/net/ethernet/adi/bfin_mac.c +++ b/drivers/net/ethernet/adi/bfin_mac.c @@ -192,8 +192,8 @@ static int desc_list_init(struct net_device *dev) goto init_error; skb_reserve(new_skb, NET_IP_ALIGN); - /* Invidate the data cache of skb->data range when it is write back - * cache. It will prevent overwritting the new data from DMA + /* Invalidate the data cache of skb->data range when it is write back + * cache. It will prevent overwriting the new data from DMA */ blackfin_dcache_invalidate_range((unsigned long)new_skb->head, (unsigned long)new_skb->end); @@ -1205,7 +1205,7 @@ static void bfin_mac_rx(struct bfin_mac_local *lp) } /* reserve 2 bytes for RXDWA padding */ skb_reserve(new_skb, NET_IP_ALIGN); - /* Invidate the data cache of skb->data range when it is write back + /* Invalidate the data cache of skb->data range when it is write back * cache. It will prevent overwritting the new data from DMA */ blackfin_dcache_invalidate_range((unsigned long)new_skb->head, @@ -1599,7 +1599,7 @@ static int bfin_mac_probe(struct platform_device *pdev) *(__le16 *) (&(ndev->dev_addr[4])) = cpu_to_le16((u16) bfin_read_EMAC_ADDRHI()); /* probe mac */ - /*todo: how to proble? which is revision_register */ + /*todo: how to probe? which is revision_register */ bfin_write_EMAC_ADDRLO(0x12345678); if (bfin_read_EMAC_ADDRLO() != 0x12345678) { dev_err(&pdev->dev, "Cannot detect Blackfin on-chip ethernet MAC controller!\n"); diff --git a/drivers/net/ethernet/aeroflex/greth.c b/drivers/net/ethernet/aeroflex/greth.c index bca07c5c94bd..f8df8248035e 100644 --- a/drivers/net/ethernet/aeroflex/greth.c +++ b/drivers/net/ethernet/aeroflex/greth.c @@ -1105,27 +1105,6 @@ static void greth_set_msglevel(struct net_device *dev, u32 value) struct greth_private *greth = netdev_priv(dev); greth->msg_enable = value; } -static int greth_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct greth_private *greth = netdev_priv(dev); - struct phy_device *phy = greth->phy; - - if (!phy) - return -ENODEV; - - return phy_ethtool_gset(phy, cmd); -} - -static int greth_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct greth_private *greth = netdev_priv(dev); - struct phy_device *phy = greth->phy; - - if (!phy) - return -ENODEV; - - return phy_ethtool_sset(phy, cmd); -} static int greth_get_regs_len(struct net_device *dev) { @@ -1157,12 +1136,12 @@ static void greth_get_regs(struct net_device *dev, struct ethtool_regs *regs, vo static const struct ethtool_ops greth_ethtool_ops = { .get_msglevel = greth_get_msglevel, .set_msglevel = greth_set_msglevel, - .get_settings = greth_get_settings, - .set_settings = greth_set_settings, .get_drvinfo = greth_get_drvinfo, .get_regs_len = greth_get_regs_len, .get_regs = greth_get_regs, .get_link = ethtool_op_get_link, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static struct net_device_ops greth_netdev_ops = { @@ -1224,7 +1203,7 @@ static int greth_mdio_write(struct mii_bus *bus, int phy, int reg, u16 val) static void greth_link_change(struct net_device *dev) { struct greth_private *greth = netdev_priv(dev); - struct phy_device *phydev = greth->phy; + struct phy_device *phydev = dev->phydev; unsigned long flags; int status_change = 0; u32 ctrl; @@ -1307,7 +1286,6 @@ static int greth_mdio_probe(struct net_device *dev) greth->link = 0; greth->speed = 0; greth->duplex = -1; - greth->phy = phy; return 0; } @@ -1325,6 +1303,7 @@ static int greth_mdio_init(struct greth_private *greth) { int ret; unsigned long timeout; + struct net_device *ndev = greth->netdev; greth->mdio = mdiobus_alloc(); if (!greth->mdio) { @@ -1349,15 +1328,16 @@ static int greth_mdio_init(struct greth_private *greth) goto unreg_mdio; } - phy_start(greth->phy); + phy_start(ndev->phydev); /* If Ethernet debug link is used make autoneg happen right away */ if (greth->edcl && greth_edcl == 1) { - phy_start_aneg(greth->phy); + phy_start_aneg(ndev->phydev); timeout = jiffies + 6*HZ; - while (!phy_aneg_done(greth->phy) && time_before(jiffies, timeout)) { + while (!phy_aneg_done(ndev->phydev) && + time_before(jiffies, timeout)) { } - phy_read_status(greth->phy); + phy_read_status(ndev->phydev); greth_link_change(greth->netdev); } @@ -1569,8 +1549,8 @@ static int greth_of_remove(struct platform_device *of_dev) dma_free_coherent(&of_dev->dev, 1024, greth->tx_bd_base, greth->tx_bd_base_phys); - if (greth->phy) - phy_stop(greth->phy); + if (ndev->phydev) + phy_stop(ndev->phydev); mdiobus_unregister(greth->mdio); unregister_netdev(ndev); diff --git a/drivers/net/ethernet/aeroflex/greth.h b/drivers/net/ethernet/aeroflex/greth.h index 92dd918e4a83..9c07140a5d8d 100644 --- a/drivers/net/ethernet/aeroflex/greth.h +++ b/drivers/net/ethernet/aeroflex/greth.h @@ -123,7 +123,6 @@ struct greth_private { struct napi_struct napi; spinlock_t devlock; - struct phy_device *phy; struct mii_bus *mdio; unsigned int link; unsigned int speed; diff --git a/drivers/net/ethernet/amazon/Kconfig b/drivers/net/ethernet/amazon/Kconfig new file mode 100644 index 000000000000..99b30353541a --- /dev/null +++ b/drivers/net/ethernet/amazon/Kconfig @@ -0,0 +1,27 @@ +# +# Amazon network device configuration +# + +config NET_VENDOR_AMAZON + bool "Amazon Devices" + default y + ---help--- + If you have a network (Ethernet) device belonging to this class, say Y. + + Note that the answer to this question doesn't directly affect the + kernel: saying N will just cause the configurator to skip all + the questions about Amazon devices. If you say Y, you will be asked + for your specific device in the following questions. + +if NET_VENDOR_AMAZON + +config ENA_ETHERNET + tristate "Elastic Network Adapter (ENA) support" + depends on (PCI_MSI && X86) + ---help--- + This driver supports Elastic Network Adapter (ENA)" + + To compile this driver as a module, choose M here. + The module will be called ena. + +endif #NET_VENDOR_AMAZON diff --git a/drivers/net/ethernet/amazon/Makefile b/drivers/net/ethernet/amazon/Makefile new file mode 100644 index 000000000000..8e0b73f60d51 --- /dev/null +++ b/drivers/net/ethernet/amazon/Makefile @@ -0,0 +1,5 @@ +# +# Makefile for the Amazon network device drivers. +# + +obj-$(CONFIG_ENA_ETHERNET) += ena/ diff --git a/drivers/net/ethernet/amazon/ena/Makefile b/drivers/net/ethernet/amazon/ena/Makefile new file mode 100644 index 000000000000..eaeeae06c5d9 --- /dev/null +++ b/drivers/net/ethernet/amazon/ena/Makefile @@ -0,0 +1,7 @@ +# +# Makefile for the Elastic Network Adapter (ENA) device drivers. +# + +obj-$(CONFIG_ENA_ETHERNET) += ena.o + +ena-y := ena_netdev.o ena_com.o ena_eth_com.o ena_ethtool.o diff --git a/drivers/net/ethernet/amazon/ena/ena_admin_defs.h b/drivers/net/ethernet/amazon/ena/ena_admin_defs.h new file mode 100644 index 000000000000..a46e749bf226 --- /dev/null +++ b/drivers/net/ethernet/amazon/ena/ena_admin_defs.h @@ -0,0 +1,973 @@ +/* + * Copyright 2015 - 2016 Amazon.com, Inc. or its affiliates. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _ENA_ADMIN_H_ +#define _ENA_ADMIN_H_ + +enum ena_admin_aq_opcode { + ENA_ADMIN_CREATE_SQ = 1, + + ENA_ADMIN_DESTROY_SQ = 2, + + ENA_ADMIN_CREATE_CQ = 3, + + ENA_ADMIN_DESTROY_CQ = 4, + + ENA_ADMIN_GET_FEATURE = 8, + + ENA_ADMIN_SET_FEATURE = 9, + + ENA_ADMIN_GET_STATS = 11, +}; + +enum ena_admin_aq_completion_status { + ENA_ADMIN_SUCCESS = 0, + + ENA_ADMIN_RESOURCE_ALLOCATION_FAILURE = 1, + + ENA_ADMIN_BAD_OPCODE = 2, + + ENA_ADMIN_UNSUPPORTED_OPCODE = 3, + + ENA_ADMIN_MALFORMED_REQUEST = 4, + + /* Additional status is provided in ACQ entry extended_status */ + ENA_ADMIN_ILLEGAL_PARAMETER = 5, + + ENA_ADMIN_UNKNOWN_ERROR = 6, +}; + +enum ena_admin_aq_feature_id { + ENA_ADMIN_DEVICE_ATTRIBUTES = 1, + + ENA_ADMIN_MAX_QUEUES_NUM = 2, + + ENA_ADMIN_RSS_HASH_FUNCTION = 10, + + ENA_ADMIN_STATELESS_OFFLOAD_CONFIG = 11, + + ENA_ADMIN_RSS_REDIRECTION_TABLE_CONFIG = 12, + + ENA_ADMIN_MTU = 14, + + ENA_ADMIN_RSS_HASH_INPUT = 18, + + ENA_ADMIN_INTERRUPT_MODERATION = 20, + + ENA_ADMIN_AENQ_CONFIG = 26, + + ENA_ADMIN_LINK_CONFIG = 27, + + ENA_ADMIN_HOST_ATTR_CONFIG = 28, + + ENA_ADMIN_FEATURES_OPCODE_NUM = 32, +}; + +enum ena_admin_placement_policy_type { + /* descriptors and headers are in host memory */ + ENA_ADMIN_PLACEMENT_POLICY_HOST = 1, + + /* descriptors and headers are in device memory (a.k.a Low Latency + * Queue) + */ + ENA_ADMIN_PLACEMENT_POLICY_DEV = 3, +}; + +enum ena_admin_link_types { + ENA_ADMIN_LINK_SPEED_1G = 0x1, + + ENA_ADMIN_LINK_SPEED_2_HALF_G = 0x2, + + ENA_ADMIN_LINK_SPEED_5G = 0x4, + + ENA_ADMIN_LINK_SPEED_10G = 0x8, + + ENA_ADMIN_LINK_SPEED_25G = 0x10, + + ENA_ADMIN_LINK_SPEED_40G = 0x20, + + ENA_ADMIN_LINK_SPEED_50G = 0x40, + + ENA_ADMIN_LINK_SPEED_100G = 0x80, + + ENA_ADMIN_LINK_SPEED_200G = 0x100, + + ENA_ADMIN_LINK_SPEED_400G = 0x200, +}; + +enum ena_admin_completion_policy_type { + /* completion queue entry for each sq descriptor */ + ENA_ADMIN_COMPLETION_POLICY_DESC = 0, + + /* completion queue entry upon request in sq descriptor */ + ENA_ADMIN_COMPLETION_POLICY_DESC_ON_DEMAND = 1, + + /* current queue head pointer is updated in OS memory upon sq + * descriptor request + */ + ENA_ADMIN_COMPLETION_POLICY_HEAD_ON_DEMAND = 2, + + /* current queue head pointer is updated in OS memory for each sq + * descriptor + */ + ENA_ADMIN_COMPLETION_POLICY_HEAD = 3, +}; + +/* basic stats return ena_admin_basic_stats while extanded stats return a + * buffer (string format) with additional statistics per queue and per + * device id + */ +enum ena_admin_get_stats_type { + ENA_ADMIN_GET_STATS_TYPE_BASIC = 0, + + ENA_ADMIN_GET_STATS_TYPE_EXTENDED = 1, +}; + +enum ena_admin_get_stats_scope { + ENA_ADMIN_SPECIFIC_QUEUE = 0, + + ENA_ADMIN_ETH_TRAFFIC = 1, +}; + +struct ena_admin_aq_common_desc { + /* 11:0 : command_id + * 15:12 : reserved12 + */ + u16 command_id; + + /* as appears in ena_admin_aq_opcode */ + u8 opcode; + + /* 0 : phase + * 1 : ctrl_data - control buffer address valid + * 2 : ctrl_data_indirect - control buffer address + * points to list of pages with addresses of control + * buffers + * 7:3 : reserved3 + */ + u8 flags; +}; + +/* used in ena_admin_aq_entry. Can point directly to control data, or to a + * page list chunk. Used also at the end of indirect mode page list chunks, + * for chaining. + */ +struct ena_admin_ctrl_buff_info { + u32 length; + + struct ena_common_mem_addr address; +}; + +struct ena_admin_sq { + u16 sq_idx; + + /* 4:0 : reserved + * 7:5 : sq_direction - 0x1 - Tx; 0x2 - Rx + */ + u8 sq_identity; + + u8 reserved1; +}; + +struct ena_admin_aq_entry { + struct ena_admin_aq_common_desc aq_common_descriptor; + + union { + u32 inline_data_w1[3]; + + struct ena_admin_ctrl_buff_info control_buffer; + } u; + + u32 inline_data_w4[12]; +}; + +struct ena_admin_acq_common_desc { + /* command identifier to associate it with the aq descriptor + * 11:0 : command_id + * 15:12 : reserved12 + */ + u16 command; + + u8 status; + + /* 0 : phase + * 7:1 : reserved1 + */ + u8 flags; + + u16 extended_status; + + /* serves as a hint what AQ entries can be revoked */ + u16 sq_head_indx; +}; + +struct ena_admin_acq_entry { + struct ena_admin_acq_common_desc acq_common_descriptor; + + u32 response_specific_data[14]; +}; + +struct ena_admin_aq_create_sq_cmd { + struct ena_admin_aq_common_desc aq_common_descriptor; + + /* 4:0 : reserved0_w1 + * 7:5 : sq_direction - 0x1 - Tx, 0x2 - Rx + */ + u8 sq_identity; + + u8 reserved8_w1; + + /* 3:0 : placement_policy - Describing where the SQ + * descriptor ring and the SQ packet headers reside: + * 0x1 - descriptors and headers are in OS memory, + * 0x3 - descriptors and headers in device memory + * (a.k.a Low Latency Queue) + * 6:4 : completion_policy - Describing what policy + * to use for generation completion entry (cqe) in + * the CQ associated with this SQ: 0x0 - cqe for each + * sq descriptor, 0x1 - cqe upon request in sq + * descriptor, 0x2 - current queue head pointer is + * updated in OS memory upon sq descriptor request + * 0x3 - current queue head pointer is updated in OS + * memory for each sq descriptor + * 7 : reserved15_w1 + */ + u8 sq_caps_2; + + /* 0 : is_physically_contiguous - Described if the + * queue ring memory is allocated in physical + * contiguous pages or split. + * 7:1 : reserved17_w1 + */ + u8 sq_caps_3; + + /* associated completion queue id. This CQ must be created prior to + * SQ creation + */ + u16 cq_idx; + + /* submission queue depth in entries */ + u16 sq_depth; + + /* SQ physical base address in OS memory. This field should not be + * used for Low Latency queues. Has to be page aligned. + */ + struct ena_common_mem_addr sq_ba; + + /* specifies queue head writeback location in OS memory. Valid if + * completion_policy is set to completion_policy_head_on_demand or + * completion_policy_head. Has to be cache aligned + */ + struct ena_common_mem_addr sq_head_writeback; + + u32 reserved0_w7; + + u32 reserved0_w8; +}; + +enum ena_admin_sq_direction { + ENA_ADMIN_SQ_DIRECTION_TX = 1, + + ENA_ADMIN_SQ_DIRECTION_RX = 2, +}; + +struct ena_admin_acq_create_sq_resp_desc { + struct ena_admin_acq_common_desc acq_common_desc; + + u16 sq_idx; + + u16 reserved; + + /* queue doorbell address as an offset to PCIe MMIO REG BAR */ + u32 sq_doorbell_offset; + + /* low latency queue ring base address as an offset to PCIe MMIO + * LLQ_MEM BAR + */ + u32 llq_descriptors_offset; + + /* low latency queue headers' memory as an offset to PCIe MMIO + * LLQ_MEM BAR + */ + u32 llq_headers_offset; +}; + +struct ena_admin_aq_destroy_sq_cmd { + struct ena_admin_aq_common_desc aq_common_descriptor; + + struct ena_admin_sq sq; +}; + +struct ena_admin_acq_destroy_sq_resp_desc { + struct ena_admin_acq_common_desc acq_common_desc; +}; + +struct ena_admin_aq_create_cq_cmd { + struct ena_admin_aq_common_desc aq_common_descriptor; + + /* 4:0 : reserved5 + * 5 : interrupt_mode_enabled - if set, cq operates + * in interrupt mode, otherwise - polling + * 7:6 : reserved6 + */ + u8 cq_caps_1; + + /* 4:0 : cq_entry_size_words - size of CQ entry in + * 32-bit words, valid values: 4, 8. + * 7:5 : reserved7 + */ + u8 cq_caps_2; + + /* completion queue depth in # of entries. must be power of 2 */ + u16 cq_depth; + + /* msix vector assigned to this cq */ + u32 msix_vector; + + /* cq physical base address in OS memory. CQ must be physically + * contiguous + */ + struct ena_common_mem_addr cq_ba; +}; + +struct ena_admin_acq_create_cq_resp_desc { + struct ena_admin_acq_common_desc acq_common_desc; + + u16 cq_idx; + + /* actual cq depth in number of entries */ + u16 cq_actual_depth; + + u32 numa_node_register_offset; + + u32 cq_head_db_register_offset; + + u32 cq_interrupt_unmask_register_offset; +}; + +struct ena_admin_aq_destroy_cq_cmd { + struct ena_admin_aq_common_desc aq_common_descriptor; + + u16 cq_idx; + + u16 reserved1; +}; + +struct ena_admin_acq_destroy_cq_resp_desc { + struct ena_admin_acq_common_desc acq_common_desc; +}; + +/* ENA AQ Get Statistics command. Extended statistics are placed in control + * buffer pointed by AQ entry + */ +struct ena_admin_aq_get_stats_cmd { + struct ena_admin_aq_common_desc aq_common_descriptor; + + union { + /* command specific inline data */ + u32 inline_data_w1[3]; + + struct ena_admin_ctrl_buff_info control_buffer; + } u; + + /* stats type as defined in enum ena_admin_get_stats_type */ + u8 type; + + /* stats scope defined in enum ena_admin_get_stats_scope */ + u8 scope; + + u16 reserved3; + + /* queue id. used when scope is specific_queue */ + u16 queue_idx; + + /* device id, value 0xFFFF means mine. only privileged device can get + * stats of other device + */ + u16 device_id; +}; + +/* Basic Statistics Command. */ +struct ena_admin_basic_stats { + u32 tx_bytes_low; + + u32 tx_bytes_high; + + u32 tx_pkts_low; + + u32 tx_pkts_high; + + u32 rx_bytes_low; + + u32 rx_bytes_high; + + u32 rx_pkts_low; + + u32 rx_pkts_high; + + u32 rx_drops_low; + + u32 rx_drops_high; +}; + +struct ena_admin_acq_get_stats_resp { + struct ena_admin_acq_common_desc acq_common_desc; + + struct ena_admin_basic_stats basic_stats; +}; + +struct ena_admin_get_set_feature_common_desc { + /* 1:0 : select - 0x1 - current value; 0x3 - default + * value + * 7:3 : reserved3 + */ + u8 flags; + + /* as appears in ena_admin_aq_feature_id */ + u8 feature_id; + + u16 reserved16; +}; + +struct ena_admin_device_attr_feature_desc { + u32 impl_id; + + u32 device_version; + + /* bitmap of ena_admin_aq_feature_id */ + u32 supported_features; + + u32 reserved3; + + /* Indicates how many bits are used physical address access. */ + u32 phys_addr_width; + + /* Indicates how many bits are used virtual address access. */ + u32 virt_addr_width; + + /* unicast MAC address (in Network byte order) */ + u8 mac_addr[6]; + + u8 reserved7[2]; + + u32 max_mtu; +}; + +struct ena_admin_queue_feature_desc { + /* including LLQs */ + u32 max_sq_num; + + u32 max_sq_depth; + + u32 max_cq_num; + + u32 max_cq_depth; + + u32 max_llq_num; + + u32 max_llq_depth; + + u32 max_header_size; + + /* Maximum Descriptors number, including meta descriptor, allowed for + * a single Tx packet + */ + u16 max_packet_tx_descs; + + /* Maximum Descriptors number allowed for a single Rx packet */ + u16 max_packet_rx_descs; +}; + +struct ena_admin_set_feature_mtu_desc { + /* exclude L2 */ + u32 mtu; +}; + +struct ena_admin_set_feature_host_attr_desc { + /* host OS info base address in OS memory. host info is 4KB of + * physically contiguous + */ + struct ena_common_mem_addr os_info_ba; + + /* host debug area base address in OS memory. debug area must be + * physically contiguous + */ + struct ena_common_mem_addr debug_ba; + + /* debug area size */ + u32 debug_area_size; +}; + +struct ena_admin_feature_intr_moder_desc { + /* interrupt delay granularity in usec */ + u16 intr_delay_resolution; + + u16 reserved; +}; + +struct ena_admin_get_feature_link_desc { + /* Link speed in Mb */ + u32 speed; + + /* bit field of enum ena_admin_link types */ + u32 supported; + + /* 0 : autoneg + * 1 : duplex - Full Duplex + * 31:2 : reserved2 + */ + u32 flags; +}; + +struct ena_admin_feature_aenq_desc { + /* bitmask for AENQ groups the device can report */ + u32 supported_groups; + + /* bitmask for AENQ groups to report */ + u32 enabled_groups; +}; + +struct ena_admin_feature_offload_desc { + /* 0 : TX_L3_csum_ipv4 + * 1 : TX_L4_ipv4_csum_part - The checksum field + * should be initialized with pseudo header checksum + * 2 : TX_L4_ipv4_csum_full + * 3 : TX_L4_ipv6_csum_part - The checksum field + * should be initialized with pseudo header checksum + * 4 : TX_L4_ipv6_csum_full + * 5 : tso_ipv4 + * 6 : tso_ipv6 + * 7 : tso_ecn + */ + u32 tx; + + /* Receive side supported stateless offload + * 0 : RX_L3_csum_ipv4 - IPv4 checksum + * 1 : RX_L4_ipv4_csum - TCP/UDP/IPv4 checksum + * 2 : RX_L4_ipv6_csum - TCP/UDP/IPv6 checksum + * 3 : RX_hash - Hash calculation + */ + u32 rx_supported; + + u32 rx_enabled; +}; + +enum ena_admin_hash_functions { + ENA_ADMIN_TOEPLITZ = 1, + + ENA_ADMIN_CRC32 = 2, +}; + +struct ena_admin_feature_rss_flow_hash_control { + u32 keys_num; + + u32 reserved; + + u32 key[10]; +}; + +struct ena_admin_feature_rss_flow_hash_function { + /* 7:0 : funcs - bitmask of ena_admin_hash_functions */ + u32 supported_func; + + /* 7:0 : selected_func - bitmask of + * ena_admin_hash_functions + */ + u32 selected_func; + + /* initial value */ + u32 init_val; +}; + +/* RSS flow hash protocols */ +enum ena_admin_flow_hash_proto { + ENA_ADMIN_RSS_TCP4 = 0, + + ENA_ADMIN_RSS_UDP4 = 1, + + ENA_ADMIN_RSS_TCP6 = 2, + + ENA_ADMIN_RSS_UDP6 = 3, + + ENA_ADMIN_RSS_IP4 = 4, + + ENA_ADMIN_RSS_IP6 = 5, + + ENA_ADMIN_RSS_IP4_FRAG = 6, + + ENA_ADMIN_RSS_NOT_IP = 7, + + ENA_ADMIN_RSS_PROTO_NUM = 16, +}; + +/* RSS flow hash fields */ +enum ena_admin_flow_hash_fields { + /* Ethernet Dest Addr */ + ENA_ADMIN_RSS_L2_DA = 0, + + /* Ethernet Src Addr */ + ENA_ADMIN_RSS_L2_SA = 1, + + /* ipv4/6 Dest Addr */ + ENA_ADMIN_RSS_L3_DA = 2, + + /* ipv4/6 Src Addr */ + ENA_ADMIN_RSS_L3_SA = 5, + + /* tcp/udp Dest Port */ + ENA_ADMIN_RSS_L4_DP = 6, + + /* tcp/udp Src Port */ + ENA_ADMIN_RSS_L4_SP = 7, +}; + +struct ena_admin_proto_input { + /* flow hash fields (bitwise according to ena_admin_flow_hash_fields) */ + u16 fields; + + u16 reserved2; +}; + +struct ena_admin_feature_rss_hash_control { + struct ena_admin_proto_input supported_fields[ENA_ADMIN_RSS_PROTO_NUM]; + + struct ena_admin_proto_input selected_fields[ENA_ADMIN_RSS_PROTO_NUM]; + + struct ena_admin_proto_input reserved2[ENA_ADMIN_RSS_PROTO_NUM]; + + struct ena_admin_proto_input reserved3[ENA_ADMIN_RSS_PROTO_NUM]; +}; + +struct ena_admin_feature_rss_flow_hash_input { + /* supported hash input sorting + * 1 : L3_sort - support swap L3 addresses if DA is + * smaller than SA + * 2 : L4_sort - support swap L4 ports if DP smaller + * SP + */ + u16 supported_input_sort; + + /* enabled hash input sorting + * 1 : enable_L3_sort - enable swap L3 addresses if + * DA smaller than SA + * 2 : enable_L4_sort - enable swap L4 ports if DP + * smaller than SP + */ + u16 enabled_input_sort; +}; + +enum ena_admin_os_type { + ENA_ADMIN_OS_LINUX = 1, + + ENA_ADMIN_OS_WIN = 2, + + ENA_ADMIN_OS_DPDK = 3, + + ENA_ADMIN_OS_FREEBSD = 4, + + ENA_ADMIN_OS_IPXE = 5, +}; + +struct ena_admin_host_info { + /* defined in enum ena_admin_os_type */ + u32 os_type; + + /* os distribution string format */ + u8 os_dist_str[128]; + + /* OS distribution numeric format */ + u32 os_dist; + + /* kernel version string format */ + u8 kernel_ver_str[32]; + + /* Kernel version numeric format */ + u32 kernel_ver; + + /* 7:0 : major + * 15:8 : minor + * 23:16 : sub_minor + */ + u32 driver_version; + + /* features bitmap */ + u32 supported_network_features[4]; +}; + +struct ena_admin_rss_ind_table_entry { + u16 cq_idx; + + u16 reserved; +}; + +struct ena_admin_feature_rss_ind_table { + /* min supported table size (2^min_size) */ + u16 min_size; + + /* max supported table size (2^max_size) */ + u16 max_size; + + /* table size (2^size) */ + u16 size; + + u16 reserved; + + /* index of the inline entry. 0xFFFFFFFF means invalid */ + u32 inline_index; + + /* used for updating single entry, ignored when setting the entire + * table through the control buffer. + */ + struct ena_admin_rss_ind_table_entry inline_entry; +}; + +struct ena_admin_get_feat_cmd { + struct ena_admin_aq_common_desc aq_common_descriptor; + + struct ena_admin_ctrl_buff_info control_buffer; + + struct ena_admin_get_set_feature_common_desc feat_common; + + u32 raw[11]; +}; + +struct ena_admin_get_feat_resp { + struct ena_admin_acq_common_desc acq_common_desc; + + union { + u32 raw[14]; + + struct ena_admin_device_attr_feature_desc dev_attr; + + struct ena_admin_queue_feature_desc max_queue; + + struct ena_admin_feature_aenq_desc aenq; + + struct ena_admin_get_feature_link_desc link; + + struct ena_admin_feature_offload_desc offload; + + struct ena_admin_feature_rss_flow_hash_function flow_hash_func; + + struct ena_admin_feature_rss_flow_hash_input flow_hash_input; + + struct ena_admin_feature_rss_ind_table ind_table; + + struct ena_admin_feature_intr_moder_desc intr_moderation; + } u; +}; + +struct ena_admin_set_feat_cmd { + struct ena_admin_aq_common_desc aq_common_descriptor; + + struct ena_admin_ctrl_buff_info control_buffer; + + struct ena_admin_get_set_feature_common_desc feat_common; + + union { + u32 raw[11]; + + /* mtu size */ + struct ena_admin_set_feature_mtu_desc mtu; + + /* host attributes */ + struct ena_admin_set_feature_host_attr_desc host_attr; + + /* AENQ configuration */ + struct ena_admin_feature_aenq_desc aenq; + + /* rss flow hash function */ + struct ena_admin_feature_rss_flow_hash_function flow_hash_func; + + /* rss flow hash input */ + struct ena_admin_feature_rss_flow_hash_input flow_hash_input; + + /* rss indirection table */ + struct ena_admin_feature_rss_ind_table ind_table; + } u; +}; + +struct ena_admin_set_feat_resp { + struct ena_admin_acq_common_desc acq_common_desc; + + union { + u32 raw[14]; + } u; +}; + +struct ena_admin_aenq_common_desc { + u16 group; + + u16 syndrom; + + /* 0 : phase */ + u8 flags; + + u8 reserved1[3]; + + u32 timestamp_low; + + u32 timestamp_high; +}; + +/* asynchronous event notification groups */ +enum ena_admin_aenq_group { + ENA_ADMIN_LINK_CHANGE = 0, + + ENA_ADMIN_FATAL_ERROR = 1, + + ENA_ADMIN_WARNING = 2, + + ENA_ADMIN_NOTIFICATION = 3, + + ENA_ADMIN_KEEP_ALIVE = 4, + + ENA_ADMIN_AENQ_GROUPS_NUM = 5, +}; + +enum ena_admin_aenq_notification_syndrom { + ENA_ADMIN_SUSPEND = 0, + + ENA_ADMIN_RESUME = 1, +}; + +struct ena_admin_aenq_entry { + struct ena_admin_aenq_common_desc aenq_common_desc; + + /* command specific inline data */ + u32 inline_data_w4[12]; +}; + +struct ena_admin_aenq_link_change_desc { + struct ena_admin_aenq_common_desc aenq_common_desc; + + /* 0 : link_status */ + u32 flags; +}; + +struct ena_admin_ena_mmio_req_read_less_resp { + u16 req_id; + + u16 reg_off; + + /* value is valid when poll is cleared */ + u32 reg_val; +}; + +/* aq_common_desc */ +#define ENA_ADMIN_AQ_COMMON_DESC_COMMAND_ID_MASK GENMASK(11, 0) +#define ENA_ADMIN_AQ_COMMON_DESC_PHASE_MASK BIT(0) +#define ENA_ADMIN_AQ_COMMON_DESC_CTRL_DATA_SHIFT 1 +#define ENA_ADMIN_AQ_COMMON_DESC_CTRL_DATA_MASK BIT(1) +#define ENA_ADMIN_AQ_COMMON_DESC_CTRL_DATA_INDIRECT_SHIFT 2 +#define ENA_ADMIN_AQ_COMMON_DESC_CTRL_DATA_INDIRECT_MASK BIT(2) + +/* sq */ +#define ENA_ADMIN_SQ_SQ_DIRECTION_SHIFT 5 +#define ENA_ADMIN_SQ_SQ_DIRECTION_MASK GENMASK(7, 5) + +/* acq_common_desc */ +#define ENA_ADMIN_ACQ_COMMON_DESC_COMMAND_ID_MASK GENMASK(11, 0) +#define ENA_ADMIN_ACQ_COMMON_DESC_PHASE_MASK BIT(0) + +/* aq_create_sq_cmd */ +#define ENA_ADMIN_AQ_CREATE_SQ_CMD_SQ_DIRECTION_SHIFT 5 +#define ENA_ADMIN_AQ_CREATE_SQ_CMD_SQ_DIRECTION_MASK GENMASK(7, 5) +#define ENA_ADMIN_AQ_CREATE_SQ_CMD_PLACEMENT_POLICY_MASK GENMASK(3, 0) +#define ENA_ADMIN_AQ_CREATE_SQ_CMD_COMPLETION_POLICY_SHIFT 4 +#define ENA_ADMIN_AQ_CREATE_SQ_CMD_COMPLETION_POLICY_MASK GENMASK(6, 4) +#define ENA_ADMIN_AQ_CREATE_SQ_CMD_IS_PHYSICALLY_CONTIGUOUS_MASK BIT(0) + +/* aq_create_cq_cmd */ +#define ENA_ADMIN_AQ_CREATE_CQ_CMD_INTERRUPT_MODE_ENABLED_SHIFT 5 +#define ENA_ADMIN_AQ_CREATE_CQ_CMD_INTERRUPT_MODE_ENABLED_MASK BIT(5) +#define ENA_ADMIN_AQ_CREATE_CQ_CMD_CQ_ENTRY_SIZE_WORDS_MASK GENMASK(4, 0) + +/* get_set_feature_common_desc */ +#define ENA_ADMIN_GET_SET_FEATURE_COMMON_DESC_SELECT_MASK GENMASK(1, 0) + +/* get_feature_link_desc */ +#define ENA_ADMIN_GET_FEATURE_LINK_DESC_AUTONEG_MASK BIT(0) +#define ENA_ADMIN_GET_FEATURE_LINK_DESC_DUPLEX_SHIFT 1 +#define ENA_ADMIN_GET_FEATURE_LINK_DESC_DUPLEX_MASK BIT(1) + +/* feature_offload_desc */ +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TX_L3_CSUM_IPV4_MASK BIT(0) +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TX_L4_IPV4_CSUM_PART_SHIFT 1 +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TX_L4_IPV4_CSUM_PART_MASK BIT(1) +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TX_L4_IPV4_CSUM_FULL_SHIFT 2 +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TX_L4_IPV4_CSUM_FULL_MASK BIT(2) +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TX_L4_IPV6_CSUM_PART_SHIFT 3 +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TX_L4_IPV6_CSUM_PART_MASK BIT(3) +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TX_L4_IPV6_CSUM_FULL_SHIFT 4 +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TX_L4_IPV6_CSUM_FULL_MASK BIT(4) +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_IPV4_SHIFT 5 +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_IPV4_MASK BIT(5) +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_IPV6_SHIFT 6 +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_IPV6_MASK BIT(6) +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_ECN_SHIFT 7 +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_ECN_MASK BIT(7) +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_RX_L3_CSUM_IPV4_MASK BIT(0) +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_RX_L4_IPV4_CSUM_SHIFT 1 +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_RX_L4_IPV4_CSUM_MASK BIT(1) +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_RX_L4_IPV6_CSUM_SHIFT 2 +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_RX_L4_IPV6_CSUM_MASK BIT(2) +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_RX_HASH_SHIFT 3 +#define ENA_ADMIN_FEATURE_OFFLOAD_DESC_RX_HASH_MASK BIT(3) + +/* feature_rss_flow_hash_function */ +#define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_FUNCTION_FUNCS_MASK GENMASK(7, 0) +#define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_FUNCTION_SELECTED_FUNC_MASK GENMASK(7, 0) + +/* feature_rss_flow_hash_input */ +#define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_INPUT_L3_SORT_SHIFT 1 +#define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_INPUT_L3_SORT_MASK BIT(1) +#define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_INPUT_L4_SORT_SHIFT 2 +#define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_INPUT_L4_SORT_MASK BIT(2) +#define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_INPUT_ENABLE_L3_SORT_SHIFT 1 +#define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_INPUT_ENABLE_L3_SORT_MASK BIT(1) +#define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_INPUT_ENABLE_L4_SORT_SHIFT 2 +#define ENA_ADMIN_FEATURE_RSS_FLOW_HASH_INPUT_ENABLE_L4_SORT_MASK BIT(2) + +/* host_info */ +#define ENA_ADMIN_HOST_INFO_MAJOR_MASK GENMASK(7, 0) +#define ENA_ADMIN_HOST_INFO_MINOR_SHIFT 8 +#define ENA_ADMIN_HOST_INFO_MINOR_MASK GENMASK(15, 8) +#define ENA_ADMIN_HOST_INFO_SUB_MINOR_SHIFT 16 +#define ENA_ADMIN_HOST_INFO_SUB_MINOR_MASK GENMASK(23, 16) + +/* aenq_common_desc */ +#define ENA_ADMIN_AENQ_COMMON_DESC_PHASE_MASK BIT(0) + +/* aenq_link_change_desc */ +#define ENA_ADMIN_AENQ_LINK_CHANGE_DESC_LINK_STATUS_MASK BIT(0) + +#endif /*_ENA_ADMIN_H_ */ diff --git a/drivers/net/ethernet/amazon/ena/ena_com.c b/drivers/net/ethernet/amazon/ena/ena_com.c new file mode 100644 index 000000000000..3066d9c99984 --- /dev/null +++ b/drivers/net/ethernet/amazon/ena/ena_com.c @@ -0,0 +1,2666 @@ +/* + * Copyright 2015 Amazon.com, Inc. or its affiliates. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "ena_com.h" + +/*****************************************************************************/ +/*****************************************************************************/ + +/* Timeout in micro-sec */ +#define ADMIN_CMD_TIMEOUT_US (1000000) + +#define ENA_ASYNC_QUEUE_DEPTH 4 +#define ENA_ADMIN_QUEUE_DEPTH 32 + +#define MIN_ENA_VER (((ENA_COMMON_SPEC_VERSION_MAJOR) << \ + ENA_REGS_VERSION_MAJOR_VERSION_SHIFT) \ + | (ENA_COMMON_SPEC_VERSION_MINOR)) + +#define ENA_CTRL_MAJOR 0 +#define ENA_CTRL_MINOR 0 +#define ENA_CTRL_SUB_MINOR 1 + +#define MIN_ENA_CTRL_VER \ + (((ENA_CTRL_MAJOR) << \ + (ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_SHIFT)) | \ + ((ENA_CTRL_MINOR) << \ + (ENA_REGS_CONTROLLER_VERSION_MINOR_VERSION_SHIFT)) | \ + (ENA_CTRL_SUB_MINOR)) + +#define ENA_DMA_ADDR_TO_UINT32_LOW(x) ((u32)((u64)(x))) +#define ENA_DMA_ADDR_TO_UINT32_HIGH(x) ((u32)(((u64)(x)) >> 32)) + +#define ENA_MMIO_READ_TIMEOUT 0xFFFFFFFF + +/*****************************************************************************/ +/*****************************************************************************/ +/*****************************************************************************/ + +enum ena_cmd_status { + ENA_CMD_SUBMITTED, + ENA_CMD_COMPLETED, + /* Abort - canceled by the driver */ + ENA_CMD_ABORTED, +}; + +struct ena_comp_ctx { + struct completion wait_event; + struct ena_admin_acq_entry *user_cqe; + u32 comp_size; + enum ena_cmd_status status; + /* status from the device */ + u8 comp_status; + u8 cmd_opcode; + bool occupied; +}; + +struct ena_com_stats_ctx { + struct ena_admin_aq_get_stats_cmd get_cmd; + struct ena_admin_acq_get_stats_resp get_resp; +}; + +static inline int ena_com_mem_addr_set(struct ena_com_dev *ena_dev, + struct ena_common_mem_addr *ena_addr, + dma_addr_t addr) +{ + if ((addr & GENMASK_ULL(ena_dev->dma_addr_bits - 1, 0)) != addr) { + pr_err("dma address has more bits that the device supports\n"); + return -EINVAL; + } + + ena_addr->mem_addr_low = (u32)addr; + ena_addr->mem_addr_high = (u64)addr >> 32; + + return 0; +} + +static int ena_com_admin_init_sq(struct ena_com_admin_queue *queue) +{ + struct ena_com_admin_sq *sq = &queue->sq; + u16 size = ADMIN_SQ_SIZE(queue->q_depth); + + sq->entries = dma_zalloc_coherent(queue->q_dmadev, size, &sq->dma_addr, + GFP_KERNEL); + + if (!sq->entries) { + pr_err("memory allocation failed"); + return -ENOMEM; + } + + sq->head = 0; + sq->tail = 0; + sq->phase = 1; + + sq->db_addr = NULL; + + return 0; +} + +static int ena_com_admin_init_cq(struct ena_com_admin_queue *queue) +{ + struct ena_com_admin_cq *cq = &queue->cq; + u16 size = ADMIN_CQ_SIZE(queue->q_depth); + + cq->entries = dma_zalloc_coherent(queue->q_dmadev, size, &cq->dma_addr, + GFP_KERNEL); + + if (!cq->entries) { + pr_err("memory allocation failed"); + return -ENOMEM; + } + + cq->head = 0; + cq->phase = 1; + + return 0; +} + +static int ena_com_admin_init_aenq(struct ena_com_dev *dev, + struct ena_aenq_handlers *aenq_handlers) +{ + struct ena_com_aenq *aenq = &dev->aenq; + u32 addr_low, addr_high, aenq_caps; + u16 size; + + dev->aenq.q_depth = ENA_ASYNC_QUEUE_DEPTH; + size = ADMIN_AENQ_SIZE(ENA_ASYNC_QUEUE_DEPTH); + aenq->entries = dma_zalloc_coherent(dev->dmadev, size, &aenq->dma_addr, + GFP_KERNEL); + + if (!aenq->entries) { + pr_err("memory allocation failed"); + return -ENOMEM; + } + + aenq->head = aenq->q_depth; + aenq->phase = 1; + + addr_low = ENA_DMA_ADDR_TO_UINT32_LOW(aenq->dma_addr); + addr_high = ENA_DMA_ADDR_TO_UINT32_HIGH(aenq->dma_addr); + + writel(addr_low, dev->reg_bar + ENA_REGS_AENQ_BASE_LO_OFF); + writel(addr_high, dev->reg_bar + ENA_REGS_AENQ_BASE_HI_OFF); + + aenq_caps = 0; + aenq_caps |= dev->aenq.q_depth & ENA_REGS_AENQ_CAPS_AENQ_DEPTH_MASK; + aenq_caps |= (sizeof(struct ena_admin_aenq_entry) + << ENA_REGS_AENQ_CAPS_AENQ_ENTRY_SIZE_SHIFT) & + ENA_REGS_AENQ_CAPS_AENQ_ENTRY_SIZE_MASK; + writel(aenq_caps, dev->reg_bar + ENA_REGS_AENQ_CAPS_OFF); + + if (unlikely(!aenq_handlers)) { + pr_err("aenq handlers pointer is NULL\n"); + return -EINVAL; + } + + aenq->aenq_handlers = aenq_handlers; + + return 0; +} + +static inline void comp_ctxt_release(struct ena_com_admin_queue *queue, + struct ena_comp_ctx *comp_ctx) +{ + comp_ctx->occupied = false; + atomic_dec(&queue->outstanding_cmds); +} + +static struct ena_comp_ctx *get_comp_ctxt(struct ena_com_admin_queue *queue, + u16 command_id, bool capture) +{ + if (unlikely(command_id >= queue->q_depth)) { + pr_err("command id is larger than the queue size. cmd_id: %u queue size %d\n", + command_id, queue->q_depth); + return NULL; + } + + if (unlikely(queue->comp_ctx[command_id].occupied && capture)) { + pr_err("Completion context is occupied\n"); + return NULL; + } + + if (capture) { + atomic_inc(&queue->outstanding_cmds); + queue->comp_ctx[command_id].occupied = true; + } + + return &queue->comp_ctx[command_id]; +} + +static struct ena_comp_ctx *__ena_com_submit_admin_cmd(struct ena_com_admin_queue *admin_queue, + struct ena_admin_aq_entry *cmd, + size_t cmd_size_in_bytes, + struct ena_admin_acq_entry *comp, + size_t comp_size_in_bytes) +{ + struct ena_comp_ctx *comp_ctx; + u16 tail_masked, cmd_id; + u16 queue_size_mask; + u16 cnt; + + queue_size_mask = admin_queue->q_depth - 1; + + tail_masked = admin_queue->sq.tail & queue_size_mask; + + /* In case of queue FULL */ + cnt = admin_queue->sq.tail - admin_queue->sq.head; + if (cnt >= admin_queue->q_depth) { + pr_debug("admin queue is FULL (tail %d head %d depth: %d)\n", + admin_queue->sq.tail, admin_queue->sq.head, + admin_queue->q_depth); + admin_queue->stats.out_of_space++; + return ERR_PTR(-ENOSPC); + } + + cmd_id = admin_queue->curr_cmd_id; + + cmd->aq_common_descriptor.flags |= admin_queue->sq.phase & + ENA_ADMIN_AQ_COMMON_DESC_PHASE_MASK; + + cmd->aq_common_descriptor.command_id |= cmd_id & + ENA_ADMIN_AQ_COMMON_DESC_COMMAND_ID_MASK; + + comp_ctx = get_comp_ctxt(admin_queue, cmd_id, true); + if (unlikely(!comp_ctx)) + return ERR_PTR(-EINVAL); + + comp_ctx->status = ENA_CMD_SUBMITTED; + comp_ctx->comp_size = (u32)comp_size_in_bytes; + comp_ctx->user_cqe = comp; + comp_ctx->cmd_opcode = cmd->aq_common_descriptor.opcode; + + reinit_completion(&comp_ctx->wait_event); + + memcpy(&admin_queue->sq.entries[tail_masked], cmd, cmd_size_in_bytes); + + admin_queue->curr_cmd_id = (admin_queue->curr_cmd_id + 1) & + queue_size_mask; + + admin_queue->sq.tail++; + admin_queue->stats.submitted_cmd++; + + if (unlikely((admin_queue->sq.tail & queue_size_mask) == 0)) + admin_queue->sq.phase = !admin_queue->sq.phase; + + writel(admin_queue->sq.tail, admin_queue->sq.db_addr); + + return comp_ctx; +} + +static inline int ena_com_init_comp_ctxt(struct ena_com_admin_queue *queue) +{ + size_t size = queue->q_depth * sizeof(struct ena_comp_ctx); + struct ena_comp_ctx *comp_ctx; + u16 i; + + queue->comp_ctx = devm_kzalloc(queue->q_dmadev, size, GFP_KERNEL); + if (unlikely(!queue->comp_ctx)) { + pr_err("memory allocation failed"); + return -ENOMEM; + } + + for (i = 0; i < queue->q_depth; i++) { + comp_ctx = get_comp_ctxt(queue, i, false); + if (comp_ctx) + init_completion(&comp_ctx->wait_event); + } + + return 0; +} + +static struct ena_comp_ctx *ena_com_submit_admin_cmd(struct ena_com_admin_queue *admin_queue, + struct ena_admin_aq_entry *cmd, + size_t cmd_size_in_bytes, + struct ena_admin_acq_entry *comp, + size_t comp_size_in_bytes) +{ + unsigned long flags; + struct ena_comp_ctx *comp_ctx; + + spin_lock_irqsave(&admin_queue->q_lock, flags); + if (unlikely(!admin_queue->running_state)) { + spin_unlock_irqrestore(&admin_queue->q_lock, flags); + return ERR_PTR(-ENODEV); + } + comp_ctx = __ena_com_submit_admin_cmd(admin_queue, cmd, + cmd_size_in_bytes, + comp, + comp_size_in_bytes); + if (unlikely(IS_ERR(comp_ctx))) + admin_queue->running_state = false; + spin_unlock_irqrestore(&admin_queue->q_lock, flags); + + return comp_ctx; +} + +static int ena_com_init_io_sq(struct ena_com_dev *ena_dev, + struct ena_com_create_io_ctx *ctx, + struct ena_com_io_sq *io_sq) +{ + size_t size; + int dev_node = 0; + + memset(&io_sq->desc_addr, 0x0, sizeof(struct ena_com_io_desc_addr)); + + io_sq->desc_entry_size = + (io_sq->direction == ENA_COM_IO_QUEUE_DIRECTION_TX) ? + sizeof(struct ena_eth_io_tx_desc) : + sizeof(struct ena_eth_io_rx_desc); + + size = io_sq->desc_entry_size * io_sq->q_depth; + + if (io_sq->mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_HOST) { + dev_node = dev_to_node(ena_dev->dmadev); + set_dev_node(ena_dev->dmadev, ctx->numa_node); + io_sq->desc_addr.virt_addr = + dma_zalloc_coherent(ena_dev->dmadev, size, + &io_sq->desc_addr.phys_addr, + GFP_KERNEL); + set_dev_node(ena_dev->dmadev, dev_node); + if (!io_sq->desc_addr.virt_addr) { + io_sq->desc_addr.virt_addr = + dma_zalloc_coherent(ena_dev->dmadev, size, + &io_sq->desc_addr.phys_addr, + GFP_KERNEL); + } + } else { + dev_node = dev_to_node(ena_dev->dmadev); + set_dev_node(ena_dev->dmadev, ctx->numa_node); + io_sq->desc_addr.virt_addr = + devm_kzalloc(ena_dev->dmadev, size, GFP_KERNEL); + set_dev_node(ena_dev->dmadev, dev_node); + if (!io_sq->desc_addr.virt_addr) { + io_sq->desc_addr.virt_addr = + devm_kzalloc(ena_dev->dmadev, size, GFP_KERNEL); + } + } + + if (!io_sq->desc_addr.virt_addr) { + pr_err("memory allocation failed"); + return -ENOMEM; + } + + io_sq->tail = 0; + io_sq->next_to_comp = 0; + io_sq->phase = 1; + + return 0; +} + +static int ena_com_init_io_cq(struct ena_com_dev *ena_dev, + struct ena_com_create_io_ctx *ctx, + struct ena_com_io_cq *io_cq) +{ + size_t size; + int prev_node = 0; + + memset(&io_cq->cdesc_addr, 0x0, sizeof(struct ena_com_io_desc_addr)); + + /* Use the basic completion descriptor for Rx */ + io_cq->cdesc_entry_size_in_bytes = + (io_cq->direction == ENA_COM_IO_QUEUE_DIRECTION_TX) ? + sizeof(struct ena_eth_io_tx_cdesc) : + sizeof(struct ena_eth_io_rx_cdesc_base); + + size = io_cq->cdesc_entry_size_in_bytes * io_cq->q_depth; + + prev_node = dev_to_node(ena_dev->dmadev); + set_dev_node(ena_dev->dmadev, ctx->numa_node); + io_cq->cdesc_addr.virt_addr = + dma_zalloc_coherent(ena_dev->dmadev, size, + &io_cq->cdesc_addr.phys_addr, GFP_KERNEL); + set_dev_node(ena_dev->dmadev, prev_node); + if (!io_cq->cdesc_addr.virt_addr) { + io_cq->cdesc_addr.virt_addr = + dma_zalloc_coherent(ena_dev->dmadev, size, + &io_cq->cdesc_addr.phys_addr, + GFP_KERNEL); + } + + if (!io_cq->cdesc_addr.virt_addr) { + pr_err("memory allocation failed"); + return -ENOMEM; + } + + io_cq->phase = 1; + io_cq->head = 0; + + return 0; +} + +static void ena_com_handle_single_admin_completion(struct ena_com_admin_queue *admin_queue, + struct ena_admin_acq_entry *cqe) +{ + struct ena_comp_ctx *comp_ctx; + u16 cmd_id; + + cmd_id = cqe->acq_common_descriptor.command & + ENA_ADMIN_ACQ_COMMON_DESC_COMMAND_ID_MASK; + + comp_ctx = get_comp_ctxt(admin_queue, cmd_id, false); + if (unlikely(!comp_ctx)) { + pr_err("comp_ctx is NULL. Changing the admin queue running state\n"); + admin_queue->running_state = false; + return; + } + + comp_ctx->status = ENA_CMD_COMPLETED; + comp_ctx->comp_status = cqe->acq_common_descriptor.status; + + if (comp_ctx->user_cqe) + memcpy(comp_ctx->user_cqe, (void *)cqe, comp_ctx->comp_size); + + if (!admin_queue->polling) + complete(&comp_ctx->wait_event); +} + +static void ena_com_handle_admin_completion(struct ena_com_admin_queue *admin_queue) +{ + struct ena_admin_acq_entry *cqe = NULL; + u16 comp_num = 0; + u16 head_masked; + u8 phase; + + head_masked = admin_queue->cq.head & (admin_queue->q_depth - 1); + phase = admin_queue->cq.phase; + + cqe = &admin_queue->cq.entries[head_masked]; + + /* Go over all the completions */ + while ((cqe->acq_common_descriptor.flags & + ENA_ADMIN_ACQ_COMMON_DESC_PHASE_MASK) == phase) { + /* Do not read the rest of the completion entry before the + * phase bit was validated + */ + rmb(); + ena_com_handle_single_admin_completion(admin_queue, cqe); + + head_masked++; + comp_num++; + if (unlikely(head_masked == admin_queue->q_depth)) { + head_masked = 0; + phase = !phase; + } + + cqe = &admin_queue->cq.entries[head_masked]; + } + + admin_queue->cq.head += comp_num; + admin_queue->cq.phase = phase; + admin_queue->sq.head += comp_num; + admin_queue->stats.completed_cmd += comp_num; +} + +static int ena_com_comp_status_to_errno(u8 comp_status) +{ + if (unlikely(comp_status != 0)) + pr_err("admin command failed[%u]\n", comp_status); + + if (unlikely(comp_status > ENA_ADMIN_UNKNOWN_ERROR)) + return -EINVAL; + + switch (comp_status) { + case ENA_ADMIN_SUCCESS: + return 0; + case ENA_ADMIN_RESOURCE_ALLOCATION_FAILURE: + return -ENOMEM; + case ENA_ADMIN_UNSUPPORTED_OPCODE: + return -EPERM; + case ENA_ADMIN_BAD_OPCODE: + case ENA_ADMIN_MALFORMED_REQUEST: + case ENA_ADMIN_ILLEGAL_PARAMETER: + case ENA_ADMIN_UNKNOWN_ERROR: + return -EINVAL; + } + + return 0; +} + +static int ena_com_wait_and_process_admin_cq_polling(struct ena_comp_ctx *comp_ctx, + struct ena_com_admin_queue *admin_queue) +{ + unsigned long flags; + u32 start_time; + int ret; + + start_time = ((u32)jiffies_to_usecs(jiffies)); + + while (comp_ctx->status == ENA_CMD_SUBMITTED) { + if ((((u32)jiffies_to_usecs(jiffies)) - start_time) > + ADMIN_CMD_TIMEOUT_US) { + pr_err("Wait for completion (polling) timeout\n"); + /* ENA didn't have any completion */ + spin_lock_irqsave(&admin_queue->q_lock, flags); + admin_queue->stats.no_completion++; + admin_queue->running_state = false; + spin_unlock_irqrestore(&admin_queue->q_lock, flags); + + ret = -ETIME; + goto err; + } + + spin_lock_irqsave(&admin_queue->q_lock, flags); + ena_com_handle_admin_completion(admin_queue); + spin_unlock_irqrestore(&admin_queue->q_lock, flags); + + msleep(100); + } + + if (unlikely(comp_ctx->status == ENA_CMD_ABORTED)) { + pr_err("Command was aborted\n"); + spin_lock_irqsave(&admin_queue->q_lock, flags); + admin_queue->stats.aborted_cmd++; + spin_unlock_irqrestore(&admin_queue->q_lock, flags); + ret = -ENODEV; + goto err; + } + + WARN(comp_ctx->status != ENA_CMD_COMPLETED, "Invalid comp status %d\n", + comp_ctx->status); + + ret = ena_com_comp_status_to_errno(comp_ctx->comp_status); +err: + comp_ctxt_release(admin_queue, comp_ctx); + return ret; +} + +static int ena_com_wait_and_process_admin_cq_interrupts(struct ena_comp_ctx *comp_ctx, + struct ena_com_admin_queue *admin_queue) +{ + unsigned long flags; + int ret; + + wait_for_completion_timeout(&comp_ctx->wait_event, + usecs_to_jiffies(ADMIN_CMD_TIMEOUT_US)); + + /* In case the command wasn't completed find out the root cause. + * There might be 2 kinds of errors + * 1) No completion (timeout reached) + * 2) There is completion but the device didn't get any msi-x interrupt. + */ + if (unlikely(comp_ctx->status == ENA_CMD_SUBMITTED)) { + spin_lock_irqsave(&admin_queue->q_lock, flags); + ena_com_handle_admin_completion(admin_queue); + admin_queue->stats.no_completion++; + spin_unlock_irqrestore(&admin_queue->q_lock, flags); + + if (comp_ctx->status == ENA_CMD_COMPLETED) + pr_err("The ena device have completion but the driver didn't receive any MSI-X interrupt (cmd %d)\n", + comp_ctx->cmd_opcode); + else + pr_err("The ena device doesn't send any completion for the admin cmd %d status %d\n", + comp_ctx->cmd_opcode, comp_ctx->status); + + admin_queue->running_state = false; + ret = -ETIME; + goto err; + } + + ret = ena_com_comp_status_to_errno(comp_ctx->comp_status); +err: + comp_ctxt_release(admin_queue, comp_ctx); + return ret; +} + +/* This method read the hardware device register through posting writes + * and waiting for response + * On timeout the function will return ENA_MMIO_READ_TIMEOUT + */ +static u32 ena_com_reg_bar_read32(struct ena_com_dev *ena_dev, u16 offset) +{ + struct ena_com_mmio_read *mmio_read = &ena_dev->mmio_read; + volatile struct ena_admin_ena_mmio_req_read_less_resp *read_resp = + mmio_read->read_resp; + u32 mmio_read_reg, ret; + unsigned long flags; + int i; + + might_sleep(); + + /* If readless is disabled, perform regular read */ + if (!mmio_read->readless_supported) + return readl(ena_dev->reg_bar + offset); + + spin_lock_irqsave(&mmio_read->lock, flags); + mmio_read->seq_num++; + + read_resp->req_id = mmio_read->seq_num + 0xDEAD; + mmio_read_reg = (offset << ENA_REGS_MMIO_REG_READ_REG_OFF_SHIFT) & + ENA_REGS_MMIO_REG_READ_REG_OFF_MASK; + mmio_read_reg |= mmio_read->seq_num & + ENA_REGS_MMIO_REG_READ_REQ_ID_MASK; + + /* make sure read_resp->req_id get updated before the hw can write + * there + */ + wmb(); + + writel(mmio_read_reg, ena_dev->reg_bar + ENA_REGS_MMIO_REG_READ_OFF); + + for (i = 0; i < ENA_REG_READ_TIMEOUT; i++) { + if (read_resp->req_id == mmio_read->seq_num) + break; + + udelay(1); + } + + if (unlikely(i == ENA_REG_READ_TIMEOUT)) { + pr_err("reading reg failed for timeout. expected: req id[%hu] offset[%hu] actual: req id[%hu] offset[%hu]\n", + mmio_read->seq_num, offset, read_resp->req_id, + read_resp->reg_off); + ret = ENA_MMIO_READ_TIMEOUT; + goto err; + } + + if (read_resp->reg_off != offset) { + pr_err("Read failure: wrong offset provided"); + ret = ENA_MMIO_READ_TIMEOUT; + } else { + ret = read_resp->reg_val; + } +err: + spin_unlock_irqrestore(&mmio_read->lock, flags); + + return ret; +} + +/* There are two types to wait for completion. + * Polling mode - wait until the completion is available. + * Async mode - wait on wait queue until the completion is ready + * (or the timeout expired). + * It is expected that the IRQ called ena_com_handle_admin_completion + * to mark the completions. + */ +static int ena_com_wait_and_process_admin_cq(struct ena_comp_ctx *comp_ctx, + struct ena_com_admin_queue *admin_queue) +{ + if (admin_queue->polling) + return ena_com_wait_and_process_admin_cq_polling(comp_ctx, + admin_queue); + + return ena_com_wait_and_process_admin_cq_interrupts(comp_ctx, + admin_queue); +} + +static int ena_com_destroy_io_sq(struct ena_com_dev *ena_dev, + struct ena_com_io_sq *io_sq) +{ + struct ena_com_admin_queue *admin_queue = &ena_dev->admin_queue; + struct ena_admin_aq_destroy_sq_cmd destroy_cmd; + struct ena_admin_acq_destroy_sq_resp_desc destroy_resp; + u8 direction; + int ret; + + memset(&destroy_cmd, 0x0, sizeof(struct ena_admin_aq_destroy_sq_cmd)); + + if (io_sq->direction == ENA_COM_IO_QUEUE_DIRECTION_TX) + direction = ENA_ADMIN_SQ_DIRECTION_TX; + else + direction = ENA_ADMIN_SQ_DIRECTION_RX; + + destroy_cmd.sq.sq_identity |= (direction << + ENA_ADMIN_SQ_SQ_DIRECTION_SHIFT) & + ENA_ADMIN_SQ_SQ_DIRECTION_MASK; + + destroy_cmd.sq.sq_idx = io_sq->idx; + destroy_cmd.aq_common_descriptor.opcode = ENA_ADMIN_DESTROY_SQ; + + ret = ena_com_execute_admin_command(admin_queue, + (struct ena_admin_aq_entry *)&destroy_cmd, + sizeof(destroy_cmd), + (struct ena_admin_acq_entry *)&destroy_resp, + sizeof(destroy_resp)); + + if (unlikely(ret && (ret != -ENODEV))) + pr_err("failed to destroy io sq error: %d\n", ret); + + return ret; +} + +static void ena_com_io_queue_free(struct ena_com_dev *ena_dev, + struct ena_com_io_sq *io_sq, + struct ena_com_io_cq *io_cq) +{ + size_t size; + + if (io_cq->cdesc_addr.virt_addr) { + size = io_cq->cdesc_entry_size_in_bytes * io_cq->q_depth; + + dma_free_coherent(ena_dev->dmadev, size, + io_cq->cdesc_addr.virt_addr, + io_cq->cdesc_addr.phys_addr); + + io_cq->cdesc_addr.virt_addr = NULL; + } + + if (io_sq->desc_addr.virt_addr) { + size = io_sq->desc_entry_size * io_sq->q_depth; + + if (io_sq->mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_HOST) + dma_free_coherent(ena_dev->dmadev, size, + io_sq->desc_addr.virt_addr, + io_sq->desc_addr.phys_addr); + else + devm_kfree(ena_dev->dmadev, io_sq->desc_addr.virt_addr); + + io_sq->desc_addr.virt_addr = NULL; + } +} + +static int wait_for_reset_state(struct ena_com_dev *ena_dev, u32 timeout, + u16 exp_state) +{ + u32 val, i; + + for (i = 0; i < timeout; i++) { + val = ena_com_reg_bar_read32(ena_dev, ENA_REGS_DEV_STS_OFF); + + if (unlikely(val == ENA_MMIO_READ_TIMEOUT)) { + pr_err("Reg read timeout occurred\n"); + return -ETIME; + } + + if ((val & ENA_REGS_DEV_STS_RESET_IN_PROGRESS_MASK) == + exp_state) + return 0; + + /* The resolution of the timeout is 100ms */ + msleep(100); + } + + return -ETIME; +} + +static bool ena_com_check_supported_feature_id(struct ena_com_dev *ena_dev, + enum ena_admin_aq_feature_id feature_id) +{ + u32 feature_mask = 1 << feature_id; + + /* Device attributes is always supported */ + if ((feature_id != ENA_ADMIN_DEVICE_ATTRIBUTES) && + !(ena_dev->supported_features & feature_mask)) + return false; + + return true; +} + +static int ena_com_get_feature_ex(struct ena_com_dev *ena_dev, + struct ena_admin_get_feat_resp *get_resp, + enum ena_admin_aq_feature_id feature_id, + dma_addr_t control_buf_dma_addr, + u32 control_buff_size) +{ + struct ena_com_admin_queue *admin_queue; + struct ena_admin_get_feat_cmd get_cmd; + int ret; + + if (!ena_com_check_supported_feature_id(ena_dev, feature_id)) { + pr_info("Feature %d isn't supported\n", feature_id); + return -EPERM; + } + + memset(&get_cmd, 0x0, sizeof(get_cmd)); + admin_queue = &ena_dev->admin_queue; + + get_cmd.aq_common_descriptor.opcode = ENA_ADMIN_GET_FEATURE; + + if (control_buff_size) + get_cmd.aq_common_descriptor.flags = + ENA_ADMIN_AQ_COMMON_DESC_CTRL_DATA_INDIRECT_MASK; + else + get_cmd.aq_common_descriptor.flags = 0; + + ret = ena_com_mem_addr_set(ena_dev, + &get_cmd.control_buffer.address, + control_buf_dma_addr); + if (unlikely(ret)) { + pr_err("memory address set failed\n"); + return ret; + } + + get_cmd.control_buffer.length = control_buff_size; + + get_cmd.feat_common.feature_id = feature_id; + + ret = ena_com_execute_admin_command(admin_queue, + (struct ena_admin_aq_entry *) + &get_cmd, + sizeof(get_cmd), + (struct ena_admin_acq_entry *) + get_resp, + sizeof(*get_resp)); + + if (unlikely(ret)) + pr_err("Failed to submit get_feature command %d error: %d\n", + feature_id, ret); + + return ret; +} + +static int ena_com_get_feature(struct ena_com_dev *ena_dev, + struct ena_admin_get_feat_resp *get_resp, + enum ena_admin_aq_feature_id feature_id) +{ + return ena_com_get_feature_ex(ena_dev, + get_resp, + feature_id, + 0, + 0); +} + +static int ena_com_hash_key_allocate(struct ena_com_dev *ena_dev) +{ + struct ena_rss *rss = &ena_dev->rss; + + rss->hash_key = + dma_zalloc_coherent(ena_dev->dmadev, sizeof(*rss->hash_key), + &rss->hash_key_dma_addr, GFP_KERNEL); + + if (unlikely(!rss->hash_key)) + return -ENOMEM; + + return 0; +} + +static void ena_com_hash_key_destroy(struct ena_com_dev *ena_dev) +{ + struct ena_rss *rss = &ena_dev->rss; + + if (rss->hash_key) + dma_free_coherent(ena_dev->dmadev, sizeof(*rss->hash_key), + rss->hash_key, rss->hash_key_dma_addr); + rss->hash_key = NULL; +} + +static int ena_com_hash_ctrl_init(struct ena_com_dev *ena_dev) +{ + struct ena_rss *rss = &ena_dev->rss; + + rss->hash_ctrl = + dma_zalloc_coherent(ena_dev->dmadev, sizeof(*rss->hash_ctrl), + &rss->hash_ctrl_dma_addr, GFP_KERNEL); + + if (unlikely(!rss->hash_ctrl)) + return -ENOMEM; + + return 0; +} + +static void ena_com_hash_ctrl_destroy(struct ena_com_dev *ena_dev) +{ + struct ena_rss *rss = &ena_dev->rss; + + if (rss->hash_ctrl) + dma_free_coherent(ena_dev->dmadev, sizeof(*rss->hash_ctrl), + rss->hash_ctrl, rss->hash_ctrl_dma_addr); + rss->hash_ctrl = NULL; +} + +static int ena_com_indirect_table_allocate(struct ena_com_dev *ena_dev, + u16 log_size) +{ + struct ena_rss *rss = &ena_dev->rss; + struct ena_admin_get_feat_resp get_resp; + size_t tbl_size; + int ret; + + ret = ena_com_get_feature(ena_dev, &get_resp, + ENA_ADMIN_RSS_REDIRECTION_TABLE_CONFIG); + if (unlikely(ret)) + return ret; + + if ((get_resp.u.ind_table.min_size > log_size) || + (get_resp.u.ind_table.max_size < log_size)) { + pr_err("indirect table size doesn't fit. requested size: %d while min is:%d and max %d\n", + 1 << log_size, 1 << get_resp.u.ind_table.min_size, + 1 << get_resp.u.ind_table.max_size); + return -EINVAL; + } + + tbl_size = (1ULL << log_size) * + sizeof(struct ena_admin_rss_ind_table_entry); + + rss->rss_ind_tbl = + dma_zalloc_coherent(ena_dev->dmadev, tbl_size, + &rss->rss_ind_tbl_dma_addr, GFP_KERNEL); + if (unlikely(!rss->rss_ind_tbl)) + goto mem_err1; + + tbl_size = (1ULL << log_size) * sizeof(u16); + rss->host_rss_ind_tbl = + devm_kzalloc(ena_dev->dmadev, tbl_size, GFP_KERNEL); + if (unlikely(!rss->host_rss_ind_tbl)) + goto mem_err2; + + rss->tbl_log_size = log_size; + + return 0; + +mem_err2: + tbl_size = (1ULL << log_size) * + sizeof(struct ena_admin_rss_ind_table_entry); + + dma_free_coherent(ena_dev->dmadev, tbl_size, rss->rss_ind_tbl, + rss->rss_ind_tbl_dma_addr); + rss->rss_ind_tbl = NULL; +mem_err1: + rss->tbl_log_size = 0; + return -ENOMEM; +} + +static void ena_com_indirect_table_destroy(struct ena_com_dev *ena_dev) +{ + struct ena_rss *rss = &ena_dev->rss; + size_t tbl_size = (1ULL << rss->tbl_log_size) * + sizeof(struct ena_admin_rss_ind_table_entry); + + if (rss->rss_ind_tbl) + dma_free_coherent(ena_dev->dmadev, tbl_size, rss->rss_ind_tbl, + rss->rss_ind_tbl_dma_addr); + rss->rss_ind_tbl = NULL; + + if (rss->host_rss_ind_tbl) + devm_kfree(ena_dev->dmadev, rss->host_rss_ind_tbl); + rss->host_rss_ind_tbl = NULL; +} + +static int ena_com_create_io_sq(struct ena_com_dev *ena_dev, + struct ena_com_io_sq *io_sq, u16 cq_idx) +{ + struct ena_com_admin_queue *admin_queue = &ena_dev->admin_queue; + struct ena_admin_aq_create_sq_cmd create_cmd; + struct ena_admin_acq_create_sq_resp_desc cmd_completion; + u8 direction; + int ret; + + memset(&create_cmd, 0x0, sizeof(struct ena_admin_aq_create_sq_cmd)); + + create_cmd.aq_common_descriptor.opcode = ENA_ADMIN_CREATE_SQ; + + if (io_sq->direction == ENA_COM_IO_QUEUE_DIRECTION_TX) + direction = ENA_ADMIN_SQ_DIRECTION_TX; + else + direction = ENA_ADMIN_SQ_DIRECTION_RX; + + create_cmd.sq_identity |= (direction << + ENA_ADMIN_AQ_CREATE_SQ_CMD_SQ_DIRECTION_SHIFT) & + ENA_ADMIN_AQ_CREATE_SQ_CMD_SQ_DIRECTION_MASK; + + create_cmd.sq_caps_2 |= io_sq->mem_queue_type & + ENA_ADMIN_AQ_CREATE_SQ_CMD_PLACEMENT_POLICY_MASK; + + create_cmd.sq_caps_2 |= (ENA_ADMIN_COMPLETION_POLICY_DESC << + ENA_ADMIN_AQ_CREATE_SQ_CMD_COMPLETION_POLICY_SHIFT) & + ENA_ADMIN_AQ_CREATE_SQ_CMD_COMPLETION_POLICY_MASK; + + create_cmd.sq_caps_3 |= + ENA_ADMIN_AQ_CREATE_SQ_CMD_IS_PHYSICALLY_CONTIGUOUS_MASK; + + create_cmd.cq_idx = cq_idx; + create_cmd.sq_depth = io_sq->q_depth; + + if (io_sq->mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_HOST) { + ret = ena_com_mem_addr_set(ena_dev, + &create_cmd.sq_ba, + io_sq->desc_addr.phys_addr); + if (unlikely(ret)) { + pr_err("memory address set failed\n"); + return ret; + } + } + + ret = ena_com_execute_admin_command(admin_queue, + (struct ena_admin_aq_entry *)&create_cmd, + sizeof(create_cmd), + (struct ena_admin_acq_entry *)&cmd_completion, + sizeof(cmd_completion)); + if (unlikely(ret)) { + pr_err("Failed to create IO SQ. error: %d\n", ret); + return ret; + } + + io_sq->idx = cmd_completion.sq_idx; + + io_sq->db_addr = (u32 __iomem *)((uintptr_t)ena_dev->reg_bar + + (uintptr_t)cmd_completion.sq_doorbell_offset); + + if (io_sq->mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV) { + io_sq->header_addr = (u8 __iomem *)((uintptr_t)ena_dev->mem_bar + + cmd_completion.llq_headers_offset); + + io_sq->desc_addr.pbuf_dev_addr = + (u8 __iomem *)((uintptr_t)ena_dev->mem_bar + + cmd_completion.llq_descriptors_offset); + } + + pr_debug("created sq[%u], depth[%u]\n", io_sq->idx, io_sq->q_depth); + + return ret; +} + +static int ena_com_ind_tbl_convert_to_device(struct ena_com_dev *ena_dev) +{ + struct ena_rss *rss = &ena_dev->rss; + struct ena_com_io_sq *io_sq; + u16 qid; + int i; + + for (i = 0; i < 1 << rss->tbl_log_size; i++) { + qid = rss->host_rss_ind_tbl[i]; + if (qid >= ENA_TOTAL_NUM_QUEUES) + return -EINVAL; + + io_sq = &ena_dev->io_sq_queues[qid]; + + if (io_sq->direction != ENA_COM_IO_QUEUE_DIRECTION_RX) + return -EINVAL; + + rss->rss_ind_tbl[i].cq_idx = io_sq->idx; + } + + return 0; +} + +static int ena_com_ind_tbl_convert_from_device(struct ena_com_dev *ena_dev) +{ + u16 dev_idx_to_host_tbl[ENA_TOTAL_NUM_QUEUES] = { (u16)-1 }; + struct ena_rss *rss = &ena_dev->rss; + u8 idx; + u16 i; + + for (i = 0; i < ENA_TOTAL_NUM_QUEUES; i++) + dev_idx_to_host_tbl[ena_dev->io_sq_queues[i].idx] = i; + + for (i = 0; i < 1 << rss->tbl_log_size; i++) { + if (rss->rss_ind_tbl[i].cq_idx > ENA_TOTAL_NUM_QUEUES) + return -EINVAL; + idx = (u8)rss->rss_ind_tbl[i].cq_idx; + + if (dev_idx_to_host_tbl[idx] > ENA_TOTAL_NUM_QUEUES) + return -EINVAL; + + rss->host_rss_ind_tbl[i] = dev_idx_to_host_tbl[idx]; + } + + return 0; +} + +static int ena_com_init_interrupt_moderation_table(struct ena_com_dev *ena_dev) +{ + size_t size; + + size = sizeof(struct ena_intr_moder_entry) * ENA_INTR_MAX_NUM_OF_LEVELS; + + ena_dev->intr_moder_tbl = + devm_kzalloc(ena_dev->dmadev, size, GFP_KERNEL); + if (!ena_dev->intr_moder_tbl) + return -ENOMEM; + + ena_com_config_default_interrupt_moderation_table(ena_dev); + + return 0; +} + +static void ena_com_update_intr_delay_resolution(struct ena_com_dev *ena_dev, + u16 intr_delay_resolution) +{ + struct ena_intr_moder_entry *intr_moder_tbl = ena_dev->intr_moder_tbl; + unsigned int i; + + if (!intr_delay_resolution) { + pr_err("Illegal intr_delay_resolution provided. Going to use default 1 usec resolution\n"); + intr_delay_resolution = 1; + } + ena_dev->intr_delay_resolution = intr_delay_resolution; + + /* update Rx */ + for (i = 0; i < ENA_INTR_MAX_NUM_OF_LEVELS; i++) + intr_moder_tbl[i].intr_moder_interval /= intr_delay_resolution; + + /* update Tx */ + ena_dev->intr_moder_tx_interval /= intr_delay_resolution; +} + +/*****************************************************************************/ +/******************************* API ******************************/ +/*****************************************************************************/ + +int ena_com_execute_admin_command(struct ena_com_admin_queue *admin_queue, + struct ena_admin_aq_entry *cmd, + size_t cmd_size, + struct ena_admin_acq_entry *comp, + size_t comp_size) +{ + struct ena_comp_ctx *comp_ctx; + int ret; + + comp_ctx = ena_com_submit_admin_cmd(admin_queue, cmd, cmd_size, + comp, comp_size); + if (unlikely(IS_ERR(comp_ctx))) { + pr_err("Failed to submit command [%ld]\n", PTR_ERR(comp_ctx)); + return PTR_ERR(comp_ctx); + } + + ret = ena_com_wait_and_process_admin_cq(comp_ctx, admin_queue); + if (unlikely(ret)) { + if (admin_queue->running_state) + pr_err("Failed to process command. ret = %d\n", ret); + else + pr_debug("Failed to process command. ret = %d\n", ret); + } + return ret; +} + +int ena_com_create_io_cq(struct ena_com_dev *ena_dev, + struct ena_com_io_cq *io_cq) +{ + struct ena_com_admin_queue *admin_queue = &ena_dev->admin_queue; + struct ena_admin_aq_create_cq_cmd create_cmd; + struct ena_admin_acq_create_cq_resp_desc cmd_completion; + int ret; + + memset(&create_cmd, 0x0, sizeof(struct ena_admin_aq_create_cq_cmd)); + + create_cmd.aq_common_descriptor.opcode = ENA_ADMIN_CREATE_CQ; + + create_cmd.cq_caps_2 |= (io_cq->cdesc_entry_size_in_bytes / 4) & + ENA_ADMIN_AQ_CREATE_CQ_CMD_CQ_ENTRY_SIZE_WORDS_MASK; + create_cmd.cq_caps_1 |= + ENA_ADMIN_AQ_CREATE_CQ_CMD_INTERRUPT_MODE_ENABLED_MASK; + + create_cmd.msix_vector = io_cq->msix_vector; + create_cmd.cq_depth = io_cq->q_depth; + + ret = ena_com_mem_addr_set(ena_dev, + &create_cmd.cq_ba, + io_cq->cdesc_addr.phys_addr); + if (unlikely(ret)) { + pr_err("memory address set failed\n"); + return ret; + } + + ret = ena_com_execute_admin_command(admin_queue, + (struct ena_admin_aq_entry *)&create_cmd, + sizeof(create_cmd), + (struct ena_admin_acq_entry *)&cmd_completion, + sizeof(cmd_completion)); + if (unlikely(ret)) { + pr_err("Failed to create IO CQ. error: %d\n", ret); + return ret; + } + + io_cq->idx = cmd_completion.cq_idx; + + io_cq->unmask_reg = (u32 __iomem *)((uintptr_t)ena_dev->reg_bar + + cmd_completion.cq_interrupt_unmask_register_offset); + + if (cmd_completion.cq_head_db_register_offset) + io_cq->cq_head_db_reg = + (u32 __iomem *)((uintptr_t)ena_dev->reg_bar + + cmd_completion.cq_head_db_register_offset); + + if (cmd_completion.numa_node_register_offset) + io_cq->numa_node_cfg_reg = + (u32 __iomem *)((uintptr_t)ena_dev->reg_bar + + cmd_completion.numa_node_register_offset); + + pr_debug("created cq[%u], depth[%u]\n", io_cq->idx, io_cq->q_depth); + + return ret; +} + +int ena_com_get_io_handlers(struct ena_com_dev *ena_dev, u16 qid, + struct ena_com_io_sq **io_sq, + struct ena_com_io_cq **io_cq) +{ + if (qid >= ENA_TOTAL_NUM_QUEUES) { + pr_err("Invalid queue number %d but the max is %d\n", qid, + ENA_TOTAL_NUM_QUEUES); + return -EINVAL; + } + + *io_sq = &ena_dev->io_sq_queues[qid]; + *io_cq = &ena_dev->io_cq_queues[qid]; + + return 0; +} + +void ena_com_abort_admin_commands(struct ena_com_dev *ena_dev) +{ + struct ena_com_admin_queue *admin_queue = &ena_dev->admin_queue; + struct ena_comp_ctx *comp_ctx; + u16 i; + + if (!admin_queue->comp_ctx) + return; + + for (i = 0; i < admin_queue->q_depth; i++) { + comp_ctx = get_comp_ctxt(admin_queue, i, false); + if (unlikely(!comp_ctx)) + break; + + comp_ctx->status = ENA_CMD_ABORTED; + + complete(&comp_ctx->wait_event); + } +} + +void ena_com_wait_for_abort_completion(struct ena_com_dev *ena_dev) +{ + struct ena_com_admin_queue *admin_queue = &ena_dev->admin_queue; + unsigned long flags; + + spin_lock_irqsave(&admin_queue->q_lock, flags); + while (atomic_read(&admin_queue->outstanding_cmds) != 0) { + spin_unlock_irqrestore(&admin_queue->q_lock, flags); + msleep(20); + spin_lock_irqsave(&admin_queue->q_lock, flags); + } + spin_unlock_irqrestore(&admin_queue->q_lock, flags); +} + +int ena_com_destroy_io_cq(struct ena_com_dev *ena_dev, + struct ena_com_io_cq *io_cq) +{ + struct ena_com_admin_queue *admin_queue = &ena_dev->admin_queue; + struct ena_admin_aq_destroy_cq_cmd destroy_cmd; + struct ena_admin_acq_destroy_cq_resp_desc destroy_resp; + int ret; + + memset(&destroy_cmd, 0x0, sizeof(struct ena_admin_aq_destroy_sq_cmd)); + + destroy_cmd.cq_idx = io_cq->idx; + destroy_cmd.aq_common_descriptor.opcode = ENA_ADMIN_DESTROY_CQ; + + ret = ena_com_execute_admin_command(admin_queue, + (struct ena_admin_aq_entry *)&destroy_cmd, + sizeof(destroy_cmd), + (struct ena_admin_acq_entry *)&destroy_resp, + sizeof(destroy_resp)); + + if (unlikely(ret && (ret != -ENODEV))) + pr_err("Failed to destroy IO CQ. error: %d\n", ret); + + return ret; +} + +bool ena_com_get_admin_running_state(struct ena_com_dev *ena_dev) +{ + return ena_dev->admin_queue.running_state; +} + +void ena_com_set_admin_running_state(struct ena_com_dev *ena_dev, bool state) +{ + struct ena_com_admin_queue *admin_queue = &ena_dev->admin_queue; + unsigned long flags; + + spin_lock_irqsave(&admin_queue->q_lock, flags); + ena_dev->admin_queue.running_state = state; + spin_unlock_irqrestore(&admin_queue->q_lock, flags); +} + +void ena_com_admin_aenq_enable(struct ena_com_dev *ena_dev) +{ + u16 depth = ena_dev->aenq.q_depth; + + WARN(ena_dev->aenq.head != depth, "Invalid AENQ state\n"); + + /* Init head_db to mark that all entries in the queue + * are initially available + */ + writel(depth, ena_dev->reg_bar + ENA_REGS_AENQ_HEAD_DB_OFF); +} + +int ena_com_set_aenq_config(struct ena_com_dev *ena_dev, u32 groups_flag) +{ + struct ena_com_admin_queue *admin_queue; + struct ena_admin_set_feat_cmd cmd; + struct ena_admin_set_feat_resp resp; + struct ena_admin_get_feat_resp get_resp; + int ret; + + ret = ena_com_get_feature(ena_dev, &get_resp, ENA_ADMIN_AENQ_CONFIG); + if (ret) { + pr_info("Can't get aenq configuration\n"); + return ret; + } + + if ((get_resp.u.aenq.supported_groups & groups_flag) != groups_flag) { + pr_warn("Trying to set unsupported aenq events. supported flag: %x asked flag: %x\n", + get_resp.u.aenq.supported_groups, groups_flag); + return -EPERM; + } + + memset(&cmd, 0x0, sizeof(cmd)); + admin_queue = &ena_dev->admin_queue; + + cmd.aq_common_descriptor.opcode = ENA_ADMIN_SET_FEATURE; + cmd.aq_common_descriptor.flags = 0; + cmd.feat_common.feature_id = ENA_ADMIN_AENQ_CONFIG; + cmd.u.aenq.enabled_groups = groups_flag; + + ret = ena_com_execute_admin_command(admin_queue, + (struct ena_admin_aq_entry *)&cmd, + sizeof(cmd), + (struct ena_admin_acq_entry *)&resp, + sizeof(resp)); + + if (unlikely(ret)) + pr_err("Failed to config AENQ ret: %d\n", ret); + + return ret; +} + +int ena_com_get_dma_width(struct ena_com_dev *ena_dev) +{ + u32 caps = ena_com_reg_bar_read32(ena_dev, ENA_REGS_CAPS_OFF); + int width; + + if (unlikely(caps == ENA_MMIO_READ_TIMEOUT)) { + pr_err("Reg read timeout occurred\n"); + return -ETIME; + } + + width = (caps & ENA_REGS_CAPS_DMA_ADDR_WIDTH_MASK) >> + ENA_REGS_CAPS_DMA_ADDR_WIDTH_SHIFT; + + pr_debug("ENA dma width: %d\n", width); + + if ((width < 32) || width > ENA_MAX_PHYS_ADDR_SIZE_BITS) { + pr_err("DMA width illegal value: %d\n", width); + return -EINVAL; + } + + ena_dev->dma_addr_bits = width; + + return width; +} + +int ena_com_validate_version(struct ena_com_dev *ena_dev) +{ + u32 ver; + u32 ctrl_ver; + u32 ctrl_ver_masked; + + /* Make sure the ENA version and the controller version are at least + * as the driver expects + */ + ver = ena_com_reg_bar_read32(ena_dev, ENA_REGS_VERSION_OFF); + ctrl_ver = ena_com_reg_bar_read32(ena_dev, + ENA_REGS_CONTROLLER_VERSION_OFF); + + if (unlikely((ver == ENA_MMIO_READ_TIMEOUT) || + (ctrl_ver == ENA_MMIO_READ_TIMEOUT))) { + pr_err("Reg read timeout occurred\n"); + return -ETIME; + } + + pr_info("ena device version: %d.%d\n", + (ver & ENA_REGS_VERSION_MAJOR_VERSION_MASK) >> + ENA_REGS_VERSION_MAJOR_VERSION_SHIFT, + ver & ENA_REGS_VERSION_MINOR_VERSION_MASK); + + if (ver < MIN_ENA_VER) { + pr_err("ENA version is lower than the minimal version the driver supports\n"); + return -1; + } + + pr_info("ena controller version: %d.%d.%d implementation version %d\n", + (ctrl_ver & ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_MASK) >> + ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_SHIFT, + (ctrl_ver & ENA_REGS_CONTROLLER_VERSION_MINOR_VERSION_MASK) >> + ENA_REGS_CONTROLLER_VERSION_MINOR_VERSION_SHIFT, + (ctrl_ver & ENA_REGS_CONTROLLER_VERSION_SUBMINOR_VERSION_MASK), + (ctrl_ver & ENA_REGS_CONTROLLER_VERSION_IMPL_ID_MASK) >> + ENA_REGS_CONTROLLER_VERSION_IMPL_ID_SHIFT); + + ctrl_ver_masked = + (ctrl_ver & ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_MASK) | + (ctrl_ver & ENA_REGS_CONTROLLER_VERSION_MINOR_VERSION_MASK) | + (ctrl_ver & ENA_REGS_CONTROLLER_VERSION_SUBMINOR_VERSION_MASK); + + /* Validate the ctrl version without the implementation ID */ + if (ctrl_ver_masked < MIN_ENA_CTRL_VER) { + pr_err("ENA ctrl version is lower than the minimal ctrl version the driver supports\n"); + return -1; + } + + return 0; +} + +void ena_com_admin_destroy(struct ena_com_dev *ena_dev) +{ + struct ena_com_admin_queue *admin_queue = &ena_dev->admin_queue; + struct ena_com_admin_cq *cq = &admin_queue->cq; + struct ena_com_admin_sq *sq = &admin_queue->sq; + struct ena_com_aenq *aenq = &ena_dev->aenq; + u16 size; + + if (admin_queue->comp_ctx) + devm_kfree(ena_dev->dmadev, admin_queue->comp_ctx); + admin_queue->comp_ctx = NULL; + size = ADMIN_SQ_SIZE(admin_queue->q_depth); + if (sq->entries) + dma_free_coherent(ena_dev->dmadev, size, sq->entries, + sq->dma_addr); + sq->entries = NULL; + + size = ADMIN_CQ_SIZE(admin_queue->q_depth); + if (cq->entries) + dma_free_coherent(ena_dev->dmadev, size, cq->entries, + cq->dma_addr); + cq->entries = NULL; + + size = ADMIN_AENQ_SIZE(aenq->q_depth); + if (ena_dev->aenq.entries) + dma_free_coherent(ena_dev->dmadev, size, aenq->entries, + aenq->dma_addr); + aenq->entries = NULL; +} + +void ena_com_set_admin_polling_mode(struct ena_com_dev *ena_dev, bool polling) +{ + ena_dev->admin_queue.polling = polling; +} + +int ena_com_mmio_reg_read_request_init(struct ena_com_dev *ena_dev) +{ + struct ena_com_mmio_read *mmio_read = &ena_dev->mmio_read; + + spin_lock_init(&mmio_read->lock); + mmio_read->read_resp = + dma_zalloc_coherent(ena_dev->dmadev, + sizeof(*mmio_read->read_resp), + &mmio_read->read_resp_dma_addr, GFP_KERNEL); + if (unlikely(!mmio_read->read_resp)) + return -ENOMEM; + + ena_com_mmio_reg_read_request_write_dev_addr(ena_dev); + + mmio_read->read_resp->req_id = 0x0; + mmio_read->seq_num = 0x0; + mmio_read->readless_supported = true; + + return 0; +} + +void ena_com_set_mmio_read_mode(struct ena_com_dev *ena_dev, bool readless_supported) +{ + struct ena_com_mmio_read *mmio_read = &ena_dev->mmio_read; + + mmio_read->readless_supported = readless_supported; +} + +void ena_com_mmio_reg_read_request_destroy(struct ena_com_dev *ena_dev) +{ + struct ena_com_mmio_read *mmio_read = &ena_dev->mmio_read; + + writel(0x0, ena_dev->reg_bar + ENA_REGS_MMIO_RESP_LO_OFF); + writel(0x0, ena_dev->reg_bar + ENA_REGS_MMIO_RESP_HI_OFF); + + dma_free_coherent(ena_dev->dmadev, sizeof(*mmio_read->read_resp), + mmio_read->read_resp, mmio_read->read_resp_dma_addr); + + mmio_read->read_resp = NULL; +} + +void ena_com_mmio_reg_read_request_write_dev_addr(struct ena_com_dev *ena_dev) +{ + struct ena_com_mmio_read *mmio_read = &ena_dev->mmio_read; + u32 addr_low, addr_high; + + addr_low = ENA_DMA_ADDR_TO_UINT32_LOW(mmio_read->read_resp_dma_addr); + addr_high = ENA_DMA_ADDR_TO_UINT32_HIGH(mmio_read->read_resp_dma_addr); + + writel(addr_low, ena_dev->reg_bar + ENA_REGS_MMIO_RESP_LO_OFF); + writel(addr_high, ena_dev->reg_bar + ENA_REGS_MMIO_RESP_HI_OFF); +} + +int ena_com_admin_init(struct ena_com_dev *ena_dev, + struct ena_aenq_handlers *aenq_handlers, + bool init_spinlock) +{ + struct ena_com_admin_queue *admin_queue = &ena_dev->admin_queue; + u32 aq_caps, acq_caps, dev_sts, addr_low, addr_high; + int ret; + + dev_sts = ena_com_reg_bar_read32(ena_dev, ENA_REGS_DEV_STS_OFF); + + if (unlikely(dev_sts == ENA_MMIO_READ_TIMEOUT)) { + pr_err("Reg read timeout occurred\n"); + return -ETIME; + } + + if (!(dev_sts & ENA_REGS_DEV_STS_READY_MASK)) { + pr_err("Device isn't ready, abort com init\n"); + return -ENODEV; + } + + admin_queue->q_depth = ENA_ADMIN_QUEUE_DEPTH; + + admin_queue->q_dmadev = ena_dev->dmadev; + admin_queue->polling = false; + admin_queue->curr_cmd_id = 0; + + atomic_set(&admin_queue->outstanding_cmds, 0); + + if (init_spinlock) + spin_lock_init(&admin_queue->q_lock); + + ret = ena_com_init_comp_ctxt(admin_queue); + if (ret) + goto error; + + ret = ena_com_admin_init_sq(admin_queue); + if (ret) + goto error; + + ret = ena_com_admin_init_cq(admin_queue); + if (ret) + goto error; + + admin_queue->sq.db_addr = (u32 __iomem *)((uintptr_t)ena_dev->reg_bar + + ENA_REGS_AQ_DB_OFF); + + addr_low = ENA_DMA_ADDR_TO_UINT32_LOW(admin_queue->sq.dma_addr); + addr_high = ENA_DMA_ADDR_TO_UINT32_HIGH(admin_queue->sq.dma_addr); + + writel(addr_low, ena_dev->reg_bar + ENA_REGS_AQ_BASE_LO_OFF); + writel(addr_high, ena_dev->reg_bar + ENA_REGS_AQ_BASE_HI_OFF); + + addr_low = ENA_DMA_ADDR_TO_UINT32_LOW(admin_queue->cq.dma_addr); + addr_high = ENA_DMA_ADDR_TO_UINT32_HIGH(admin_queue->cq.dma_addr); + + writel(addr_low, ena_dev->reg_bar + ENA_REGS_ACQ_BASE_LO_OFF); + writel(addr_high, ena_dev->reg_bar + ENA_REGS_ACQ_BASE_HI_OFF); + + aq_caps = 0; + aq_caps |= admin_queue->q_depth & ENA_REGS_AQ_CAPS_AQ_DEPTH_MASK; + aq_caps |= (sizeof(struct ena_admin_aq_entry) << + ENA_REGS_AQ_CAPS_AQ_ENTRY_SIZE_SHIFT) & + ENA_REGS_AQ_CAPS_AQ_ENTRY_SIZE_MASK; + + acq_caps = 0; + acq_caps |= admin_queue->q_depth & ENA_REGS_ACQ_CAPS_ACQ_DEPTH_MASK; + acq_caps |= (sizeof(struct ena_admin_acq_entry) << + ENA_REGS_ACQ_CAPS_ACQ_ENTRY_SIZE_SHIFT) & + ENA_REGS_ACQ_CAPS_ACQ_ENTRY_SIZE_MASK; + + writel(aq_caps, ena_dev->reg_bar + ENA_REGS_AQ_CAPS_OFF); + writel(acq_caps, ena_dev->reg_bar + ENA_REGS_ACQ_CAPS_OFF); + ret = ena_com_admin_init_aenq(ena_dev, aenq_handlers); + if (ret) + goto error; + + admin_queue->running_state = true; + + return 0; +error: + ena_com_admin_destroy(ena_dev); + + return ret; +} + +int ena_com_create_io_queue(struct ena_com_dev *ena_dev, + struct ena_com_create_io_ctx *ctx) +{ + struct ena_com_io_sq *io_sq; + struct ena_com_io_cq *io_cq; + int ret; + + if (ctx->qid >= ENA_TOTAL_NUM_QUEUES) { + pr_err("Qid (%d) is bigger than max num of queues (%d)\n", + ctx->qid, ENA_TOTAL_NUM_QUEUES); + return -EINVAL; + } + + io_sq = &ena_dev->io_sq_queues[ctx->qid]; + io_cq = &ena_dev->io_cq_queues[ctx->qid]; + + memset(io_sq, 0x0, sizeof(struct ena_com_io_sq)); + memset(io_cq, 0x0, sizeof(struct ena_com_io_cq)); + + /* Init CQ */ + io_cq->q_depth = ctx->queue_size; + io_cq->direction = ctx->direction; + io_cq->qid = ctx->qid; + + io_cq->msix_vector = ctx->msix_vector; + + io_sq->q_depth = ctx->queue_size; + io_sq->direction = ctx->direction; + io_sq->qid = ctx->qid; + + io_sq->mem_queue_type = ctx->mem_queue_type; + + if (ctx->direction == ENA_COM_IO_QUEUE_DIRECTION_TX) + /* header length is limited to 8 bits */ + io_sq->tx_max_header_size = + min_t(u32, ena_dev->tx_max_header_size, SZ_256); + + ret = ena_com_init_io_sq(ena_dev, ctx, io_sq); + if (ret) + goto error; + ret = ena_com_init_io_cq(ena_dev, ctx, io_cq); + if (ret) + goto error; + + ret = ena_com_create_io_cq(ena_dev, io_cq); + if (ret) + goto error; + + ret = ena_com_create_io_sq(ena_dev, io_sq, io_cq->idx); + if (ret) + goto destroy_io_cq; + + return 0; + +destroy_io_cq: + ena_com_destroy_io_cq(ena_dev, io_cq); +error: + ena_com_io_queue_free(ena_dev, io_sq, io_cq); + return ret; +} + +void ena_com_destroy_io_queue(struct ena_com_dev *ena_dev, u16 qid) +{ + struct ena_com_io_sq *io_sq; + struct ena_com_io_cq *io_cq; + + if (qid >= ENA_TOTAL_NUM_QUEUES) { + pr_err("Qid (%d) is bigger than max num of queues (%d)\n", qid, + ENA_TOTAL_NUM_QUEUES); + return; + } + + io_sq = &ena_dev->io_sq_queues[qid]; + io_cq = &ena_dev->io_cq_queues[qid]; + + ena_com_destroy_io_sq(ena_dev, io_sq); + ena_com_destroy_io_cq(ena_dev, io_cq); + + ena_com_io_queue_free(ena_dev, io_sq, io_cq); +} + +int ena_com_get_link_params(struct ena_com_dev *ena_dev, + struct ena_admin_get_feat_resp *resp) +{ + return ena_com_get_feature(ena_dev, resp, ENA_ADMIN_LINK_CONFIG); +} + +int ena_com_get_dev_attr_feat(struct ena_com_dev *ena_dev, + struct ena_com_dev_get_features_ctx *get_feat_ctx) +{ + struct ena_admin_get_feat_resp get_resp; + int rc; + + rc = ena_com_get_feature(ena_dev, &get_resp, + ENA_ADMIN_DEVICE_ATTRIBUTES); + if (rc) + return rc; + + memcpy(&get_feat_ctx->dev_attr, &get_resp.u.dev_attr, + sizeof(get_resp.u.dev_attr)); + ena_dev->supported_features = get_resp.u.dev_attr.supported_features; + + rc = ena_com_get_feature(ena_dev, &get_resp, + ENA_ADMIN_MAX_QUEUES_NUM); + if (rc) + return rc; + + memcpy(&get_feat_ctx->max_queues, &get_resp.u.max_queue, + sizeof(get_resp.u.max_queue)); + ena_dev->tx_max_header_size = get_resp.u.max_queue.max_header_size; + + rc = ena_com_get_feature(ena_dev, &get_resp, + ENA_ADMIN_AENQ_CONFIG); + if (rc) + return rc; + + memcpy(&get_feat_ctx->aenq, &get_resp.u.aenq, + sizeof(get_resp.u.aenq)); + + rc = ena_com_get_feature(ena_dev, &get_resp, + ENA_ADMIN_STATELESS_OFFLOAD_CONFIG); + if (rc) + return rc; + + memcpy(&get_feat_ctx->offload, &get_resp.u.offload, + sizeof(get_resp.u.offload)); + + return 0; +} + +void ena_com_admin_q_comp_intr_handler(struct ena_com_dev *ena_dev) +{ + ena_com_handle_admin_completion(&ena_dev->admin_queue); +} + +/* ena_handle_specific_aenq_event: + * return the handler that is relevant to the specific event group + */ +static ena_aenq_handler ena_com_get_specific_aenq_cb(struct ena_com_dev *dev, + u16 group) +{ + struct ena_aenq_handlers *aenq_handlers = dev->aenq.aenq_handlers; + + if ((group < ENA_MAX_HANDLERS) && aenq_handlers->handlers[group]) + return aenq_handlers->handlers[group]; + + return aenq_handlers->unimplemented_handler; +} + +/* ena_aenq_intr_handler: + * handles the aenq incoming events. + * pop events from the queue and apply the specific handler + */ +void ena_com_aenq_intr_handler(struct ena_com_dev *dev, void *data) +{ + struct ena_admin_aenq_entry *aenq_e; + struct ena_admin_aenq_common_desc *aenq_common; + struct ena_com_aenq *aenq = &dev->aenq; + ena_aenq_handler handler_cb; + u16 masked_head, processed = 0; + u8 phase; + + masked_head = aenq->head & (aenq->q_depth - 1); + phase = aenq->phase; + aenq_e = &aenq->entries[masked_head]; /* Get first entry */ + aenq_common = &aenq_e->aenq_common_desc; + + /* Go over all the events */ + while ((aenq_common->flags & ENA_ADMIN_AENQ_COMMON_DESC_PHASE_MASK) == + phase) { + pr_debug("AENQ! Group[%x] Syndrom[%x] timestamp: [%llus]\n", + aenq_common->group, aenq_common->syndrom, + (u64)aenq_common->timestamp_low + + ((u64)aenq_common->timestamp_high << 32)); + + /* Handle specific event*/ + handler_cb = ena_com_get_specific_aenq_cb(dev, + aenq_common->group); + handler_cb(data, aenq_e); /* call the actual event handler*/ + + /* Get next event entry */ + masked_head++; + processed++; + + if (unlikely(masked_head == aenq->q_depth)) { + masked_head = 0; + phase = !phase; + } + aenq_e = &aenq->entries[masked_head]; + aenq_common = &aenq_e->aenq_common_desc; + } + + aenq->head += processed; + aenq->phase = phase; + + /* Don't update aenq doorbell if there weren't any processed events */ + if (!processed) + return; + + /* write the aenq doorbell after all AENQ descriptors were read */ + mb(); + writel((u32)aenq->head, dev->reg_bar + ENA_REGS_AENQ_HEAD_DB_OFF); +} + +int ena_com_dev_reset(struct ena_com_dev *ena_dev) +{ + u32 stat, timeout, cap, reset_val; + int rc; + + stat = ena_com_reg_bar_read32(ena_dev, ENA_REGS_DEV_STS_OFF); + cap = ena_com_reg_bar_read32(ena_dev, ENA_REGS_CAPS_OFF); + + if (unlikely((stat == ENA_MMIO_READ_TIMEOUT) || + (cap == ENA_MMIO_READ_TIMEOUT))) { + pr_err("Reg read32 timeout occurred\n"); + return -ETIME; + } + + if ((stat & ENA_REGS_DEV_STS_READY_MASK) == 0) { + pr_err("Device isn't ready, can't reset device\n"); + return -EINVAL; + } + + timeout = (cap & ENA_REGS_CAPS_RESET_TIMEOUT_MASK) >> + ENA_REGS_CAPS_RESET_TIMEOUT_SHIFT; + if (timeout == 0) { + pr_err("Invalid timeout value\n"); + return -EINVAL; + } + + /* start reset */ + reset_val = ENA_REGS_DEV_CTL_DEV_RESET_MASK; + writel(reset_val, ena_dev->reg_bar + ENA_REGS_DEV_CTL_OFF); + + /* Write again the MMIO read request address */ + ena_com_mmio_reg_read_request_write_dev_addr(ena_dev); + + rc = wait_for_reset_state(ena_dev, timeout, + ENA_REGS_DEV_STS_RESET_IN_PROGRESS_MASK); + if (rc != 0) { + pr_err("Reset indication didn't turn on\n"); + return rc; + } + + /* reset done */ + writel(0, ena_dev->reg_bar + ENA_REGS_DEV_CTL_OFF); + rc = wait_for_reset_state(ena_dev, timeout, 0); + if (rc != 0) { + pr_err("Reset indication didn't turn off\n"); + return rc; + } + + return 0; +} + +static int ena_get_dev_stats(struct ena_com_dev *ena_dev, + struct ena_com_stats_ctx *ctx, + enum ena_admin_get_stats_type type) +{ + struct ena_admin_aq_get_stats_cmd *get_cmd = &ctx->get_cmd; + struct ena_admin_acq_get_stats_resp *get_resp = &ctx->get_resp; + struct ena_com_admin_queue *admin_queue; + int ret; + + admin_queue = &ena_dev->admin_queue; + + get_cmd->aq_common_descriptor.opcode = ENA_ADMIN_GET_STATS; + get_cmd->aq_common_descriptor.flags = 0; + get_cmd->type = type; + + ret = ena_com_execute_admin_command(admin_queue, + (struct ena_admin_aq_entry *)get_cmd, + sizeof(*get_cmd), + (struct ena_admin_acq_entry *)get_resp, + sizeof(*get_resp)); + + if (unlikely(ret)) + pr_err("Failed to get stats. error: %d\n", ret); + + return ret; +} + +int ena_com_get_dev_basic_stats(struct ena_com_dev *ena_dev, + struct ena_admin_basic_stats *stats) +{ + struct ena_com_stats_ctx ctx; + int ret; + + memset(&ctx, 0x0, sizeof(ctx)); + ret = ena_get_dev_stats(ena_dev, &ctx, ENA_ADMIN_GET_STATS_TYPE_BASIC); + if (likely(ret == 0)) + memcpy(stats, &ctx.get_resp.basic_stats, + sizeof(ctx.get_resp.basic_stats)); + + return ret; +} + +int ena_com_set_dev_mtu(struct ena_com_dev *ena_dev, int mtu) +{ + struct ena_com_admin_queue *admin_queue; + struct ena_admin_set_feat_cmd cmd; + struct ena_admin_set_feat_resp resp; + int ret; + + if (!ena_com_check_supported_feature_id(ena_dev, ENA_ADMIN_MTU)) { + pr_info("Feature %d isn't supported\n", ENA_ADMIN_MTU); + return -EPERM; + } + + memset(&cmd, 0x0, sizeof(cmd)); + admin_queue = &ena_dev->admin_queue; + + cmd.aq_common_descriptor.opcode = ENA_ADMIN_SET_FEATURE; + cmd.aq_common_descriptor.flags = 0; + cmd.feat_common.feature_id = ENA_ADMIN_MTU; + cmd.u.mtu.mtu = mtu; + + ret = ena_com_execute_admin_command(admin_queue, + (struct ena_admin_aq_entry *)&cmd, + sizeof(cmd), + (struct ena_admin_acq_entry *)&resp, + sizeof(resp)); + + if (unlikely(ret)) + pr_err("Failed to set mtu %d. error: %d\n", mtu, ret); + + return ret; +} + +int ena_com_get_offload_settings(struct ena_com_dev *ena_dev, + struct ena_admin_feature_offload_desc *offload) +{ + int ret; + struct ena_admin_get_feat_resp resp; + + ret = ena_com_get_feature(ena_dev, &resp, + ENA_ADMIN_STATELESS_OFFLOAD_CONFIG); + if (unlikely(ret)) { + pr_err("Failed to get offload capabilities %d\n", ret); + return ret; + } + + memcpy(offload, &resp.u.offload, sizeof(resp.u.offload)); + + return 0; +} + +int ena_com_set_hash_function(struct ena_com_dev *ena_dev) +{ + struct ena_com_admin_queue *admin_queue = &ena_dev->admin_queue; + struct ena_rss *rss = &ena_dev->rss; + struct ena_admin_set_feat_cmd cmd; + struct ena_admin_set_feat_resp resp; + struct ena_admin_get_feat_resp get_resp; + int ret; + + if (!ena_com_check_supported_feature_id(ena_dev, + ENA_ADMIN_RSS_HASH_FUNCTION)) { + pr_info("Feature %d isn't supported\n", + ENA_ADMIN_RSS_HASH_FUNCTION); + return -EPERM; + } + + /* Validate hash function is supported */ + ret = ena_com_get_feature(ena_dev, &get_resp, + ENA_ADMIN_RSS_HASH_FUNCTION); + if (unlikely(ret)) + return ret; + + if (get_resp.u.flow_hash_func.supported_func & (1 << rss->hash_func)) { + pr_err("Func hash %d isn't supported by device, abort\n", + rss->hash_func); + return -EPERM; + } + + memset(&cmd, 0x0, sizeof(cmd)); + + cmd.aq_common_descriptor.opcode = ENA_ADMIN_SET_FEATURE; + cmd.aq_common_descriptor.flags = + ENA_ADMIN_AQ_COMMON_DESC_CTRL_DATA_INDIRECT_MASK; + cmd.feat_common.feature_id = ENA_ADMIN_RSS_HASH_FUNCTION; + cmd.u.flow_hash_func.init_val = rss->hash_init_val; + cmd.u.flow_hash_func.selected_func = 1 << rss->hash_func; + + ret = ena_com_mem_addr_set(ena_dev, + &cmd.control_buffer.address, + rss->hash_key_dma_addr); + if (unlikely(ret)) { + pr_err("memory address set failed\n"); + return ret; + } + + cmd.control_buffer.length = sizeof(*rss->hash_key); + + ret = ena_com_execute_admin_command(admin_queue, + (struct ena_admin_aq_entry *)&cmd, + sizeof(cmd), + (struct ena_admin_acq_entry *)&resp, + sizeof(resp)); + if (unlikely(ret)) { + pr_err("Failed to set hash function %d. error: %d\n", + rss->hash_func, ret); + return -EINVAL; + } + + return 0; +} + +int ena_com_fill_hash_function(struct ena_com_dev *ena_dev, + enum ena_admin_hash_functions func, + const u8 *key, u16 key_len, u32 init_val) +{ + struct ena_rss *rss = &ena_dev->rss; + struct ena_admin_get_feat_resp get_resp; + struct ena_admin_feature_rss_flow_hash_control *hash_key = + rss->hash_key; + int rc; + + /* Make sure size is a mult of DWs */ + if (unlikely(key_len & 0x3)) + return -EINVAL; + + rc = ena_com_get_feature_ex(ena_dev, &get_resp, + ENA_ADMIN_RSS_HASH_FUNCTION, + rss->hash_key_dma_addr, + sizeof(*rss->hash_key)); + if (unlikely(rc)) + return rc; + + if (!((1 << func) & get_resp.u.flow_hash_func.supported_func)) { + pr_err("Flow hash function %d isn't supported\n", func); + return -EPERM; + } + + switch (func) { + case ENA_ADMIN_TOEPLITZ: + if (key_len > sizeof(hash_key->key)) { + pr_err("key len (%hu) is bigger than the max supported (%zu)\n", + key_len, sizeof(hash_key->key)); + return -EINVAL; + } + + memcpy(hash_key->key, key, key_len); + rss->hash_init_val = init_val; + hash_key->keys_num = key_len >> 2; + break; + case ENA_ADMIN_CRC32: + rss->hash_init_val = init_val; + break; + default: + pr_err("Invalid hash function (%d)\n", func); + return -EINVAL; + } + + rc = ena_com_set_hash_function(ena_dev); + + /* Restore the old function */ + if (unlikely(rc)) + ena_com_get_hash_function(ena_dev, NULL, NULL); + + return rc; +} + +int ena_com_get_hash_function(struct ena_com_dev *ena_dev, + enum ena_admin_hash_functions *func, + u8 *key) +{ + struct ena_rss *rss = &ena_dev->rss; + struct ena_admin_get_feat_resp get_resp; + struct ena_admin_feature_rss_flow_hash_control *hash_key = + rss->hash_key; + int rc; + + rc = ena_com_get_feature_ex(ena_dev, &get_resp, + ENA_ADMIN_RSS_HASH_FUNCTION, + rss->hash_key_dma_addr, + sizeof(*rss->hash_key)); + if (unlikely(rc)) + return rc; + + rss->hash_func = get_resp.u.flow_hash_func.selected_func; + if (func) + *func = rss->hash_func; + + if (key) + memcpy(key, hash_key->key, (size_t)(hash_key->keys_num) << 2); + + return 0; +} + +int ena_com_get_hash_ctrl(struct ena_com_dev *ena_dev, + enum ena_admin_flow_hash_proto proto, + u16 *fields) +{ + struct ena_rss *rss = &ena_dev->rss; + struct ena_admin_get_feat_resp get_resp; + int rc; + + rc = ena_com_get_feature_ex(ena_dev, &get_resp, + ENA_ADMIN_RSS_HASH_INPUT, + rss->hash_ctrl_dma_addr, + sizeof(*rss->hash_ctrl)); + if (unlikely(rc)) + return rc; + + if (fields) + *fields = rss->hash_ctrl->selected_fields[proto].fields; + + return 0; +} + +int ena_com_set_hash_ctrl(struct ena_com_dev *ena_dev) +{ + struct ena_com_admin_queue *admin_queue = &ena_dev->admin_queue; + struct ena_rss *rss = &ena_dev->rss; + struct ena_admin_feature_rss_hash_control *hash_ctrl = rss->hash_ctrl; + struct ena_admin_set_feat_cmd cmd; + struct ena_admin_set_feat_resp resp; + int ret; + + if (!ena_com_check_supported_feature_id(ena_dev, + ENA_ADMIN_RSS_HASH_INPUT)) { + pr_info("Feature %d isn't supported\n", ENA_ADMIN_RSS_HASH_INPUT); + return -EPERM; + } + + memset(&cmd, 0x0, sizeof(cmd)); + + cmd.aq_common_descriptor.opcode = ENA_ADMIN_SET_FEATURE; + cmd.aq_common_descriptor.flags = + ENA_ADMIN_AQ_COMMON_DESC_CTRL_DATA_INDIRECT_MASK; + cmd.feat_common.feature_id = ENA_ADMIN_RSS_HASH_INPUT; + cmd.u.flow_hash_input.enabled_input_sort = + ENA_ADMIN_FEATURE_RSS_FLOW_HASH_INPUT_L3_SORT_MASK | + ENA_ADMIN_FEATURE_RSS_FLOW_HASH_INPUT_L4_SORT_MASK; + + ret = ena_com_mem_addr_set(ena_dev, + &cmd.control_buffer.address, + rss->hash_ctrl_dma_addr); + if (unlikely(ret)) { + pr_err("memory address set failed\n"); + return ret; + } + cmd.control_buffer.length = sizeof(*hash_ctrl); + + ret = ena_com_execute_admin_command(admin_queue, + (struct ena_admin_aq_entry *)&cmd, + sizeof(cmd), + (struct ena_admin_acq_entry *)&resp, + sizeof(resp)); + if (unlikely(ret)) + pr_err("Failed to set hash input. error: %d\n", ret); + + return ret; +} + +int ena_com_set_default_hash_ctrl(struct ena_com_dev *ena_dev) +{ + struct ena_rss *rss = &ena_dev->rss; + struct ena_admin_feature_rss_hash_control *hash_ctrl = + rss->hash_ctrl; + u16 available_fields = 0; + int rc, i; + + /* Get the supported hash input */ + rc = ena_com_get_hash_ctrl(ena_dev, 0, NULL); + if (unlikely(rc)) + return rc; + + hash_ctrl->selected_fields[ENA_ADMIN_RSS_TCP4].fields = + ENA_ADMIN_RSS_L3_SA | ENA_ADMIN_RSS_L3_DA | + ENA_ADMIN_RSS_L4_DP | ENA_ADMIN_RSS_L4_SP; + + hash_ctrl->selected_fields[ENA_ADMIN_RSS_UDP4].fields = + ENA_ADMIN_RSS_L3_SA | ENA_ADMIN_RSS_L3_DA | + ENA_ADMIN_RSS_L4_DP | ENA_ADMIN_RSS_L4_SP; + + hash_ctrl->selected_fields[ENA_ADMIN_RSS_TCP6].fields = + ENA_ADMIN_RSS_L3_SA | ENA_ADMIN_RSS_L3_DA | + ENA_ADMIN_RSS_L4_DP | ENA_ADMIN_RSS_L4_SP; + + hash_ctrl->selected_fields[ENA_ADMIN_RSS_UDP6].fields = + ENA_ADMIN_RSS_L3_SA | ENA_ADMIN_RSS_L3_DA | + ENA_ADMIN_RSS_L4_DP | ENA_ADMIN_RSS_L4_SP; + + hash_ctrl->selected_fields[ENA_ADMIN_RSS_IP4].fields = + ENA_ADMIN_RSS_L3_SA | ENA_ADMIN_RSS_L3_DA; + + hash_ctrl->selected_fields[ENA_ADMIN_RSS_IP6].fields = + ENA_ADMIN_RSS_L3_SA | ENA_ADMIN_RSS_L3_DA; + + hash_ctrl->selected_fields[ENA_ADMIN_RSS_IP4_FRAG].fields = + ENA_ADMIN_RSS_L3_SA | ENA_ADMIN_RSS_L3_DA; + + hash_ctrl->selected_fields[ENA_ADMIN_RSS_IP4_FRAG].fields = + ENA_ADMIN_RSS_L2_DA | ENA_ADMIN_RSS_L2_SA; + + for (i = 0; i < ENA_ADMIN_RSS_PROTO_NUM; i++) { + available_fields = hash_ctrl->selected_fields[i].fields & + hash_ctrl->supported_fields[i].fields; + if (available_fields != hash_ctrl->selected_fields[i].fields) { + pr_err("hash control doesn't support all the desire configuration. proto %x supported %x selected %x\n", + i, hash_ctrl->supported_fields[i].fields, + hash_ctrl->selected_fields[i].fields); + return -EPERM; + } + } + + rc = ena_com_set_hash_ctrl(ena_dev); + + /* In case of failure, restore the old hash ctrl */ + if (unlikely(rc)) + ena_com_get_hash_ctrl(ena_dev, 0, NULL); + + return rc; +} + +int ena_com_fill_hash_ctrl(struct ena_com_dev *ena_dev, + enum ena_admin_flow_hash_proto proto, + u16 hash_fields) +{ + struct ena_rss *rss = &ena_dev->rss; + struct ena_admin_feature_rss_hash_control *hash_ctrl = rss->hash_ctrl; + u16 supported_fields; + int rc; + + if (proto >= ENA_ADMIN_RSS_PROTO_NUM) { + pr_err("Invalid proto num (%u)\n", proto); + return -EINVAL; + } + + /* Get the ctrl table */ + rc = ena_com_get_hash_ctrl(ena_dev, proto, NULL); + if (unlikely(rc)) + return rc; + + /* Make sure all the fields are supported */ + supported_fields = hash_ctrl->supported_fields[proto].fields; + if ((hash_fields & supported_fields) != hash_fields) { + pr_err("proto %d doesn't support the required fields %x. supports only: %x\n", + proto, hash_fields, supported_fields); + } + + hash_ctrl->selected_fields[proto].fields = hash_fields; + + rc = ena_com_set_hash_ctrl(ena_dev); + + /* In case of failure, restore the old hash ctrl */ + if (unlikely(rc)) + ena_com_get_hash_ctrl(ena_dev, 0, NULL); + + return 0; +} + +int ena_com_indirect_table_fill_entry(struct ena_com_dev *ena_dev, + u16 entry_idx, u16 entry_value) +{ + struct ena_rss *rss = &ena_dev->rss; + + if (unlikely(entry_idx >= (1 << rss->tbl_log_size))) + return -EINVAL; + + if (unlikely((entry_value > ENA_TOTAL_NUM_QUEUES))) + return -EINVAL; + + rss->host_rss_ind_tbl[entry_idx] = entry_value; + + return 0; +} + +int ena_com_indirect_table_set(struct ena_com_dev *ena_dev) +{ + struct ena_com_admin_queue *admin_queue = &ena_dev->admin_queue; + struct ena_rss *rss = &ena_dev->rss; + struct ena_admin_set_feat_cmd cmd; + struct ena_admin_set_feat_resp resp; + int ret; + + if (!ena_com_check_supported_feature_id( + ena_dev, ENA_ADMIN_RSS_REDIRECTION_TABLE_CONFIG)) { + pr_info("Feature %d isn't supported\n", + ENA_ADMIN_RSS_REDIRECTION_TABLE_CONFIG); + return -EPERM; + } + + ret = ena_com_ind_tbl_convert_to_device(ena_dev); + if (ret) { + pr_err("Failed to convert host indirection table to device table\n"); + return ret; + } + + memset(&cmd, 0x0, sizeof(cmd)); + + cmd.aq_common_descriptor.opcode = ENA_ADMIN_SET_FEATURE; + cmd.aq_common_descriptor.flags = + ENA_ADMIN_AQ_COMMON_DESC_CTRL_DATA_INDIRECT_MASK; + cmd.feat_common.feature_id = ENA_ADMIN_RSS_REDIRECTION_TABLE_CONFIG; + cmd.u.ind_table.size = rss->tbl_log_size; + cmd.u.ind_table.inline_index = 0xFFFFFFFF; + + ret = ena_com_mem_addr_set(ena_dev, + &cmd.control_buffer.address, + rss->rss_ind_tbl_dma_addr); + if (unlikely(ret)) { + pr_err("memory address set failed\n"); + return ret; + } + + cmd.control_buffer.length = (1ULL << rss->tbl_log_size) * + sizeof(struct ena_admin_rss_ind_table_entry); + + ret = ena_com_execute_admin_command(admin_queue, + (struct ena_admin_aq_entry *)&cmd, + sizeof(cmd), + (struct ena_admin_acq_entry *)&resp, + sizeof(resp)); + + if (unlikely(ret)) + pr_err("Failed to set indirect table. error: %d\n", ret); + + return ret; +} + +int ena_com_indirect_table_get(struct ena_com_dev *ena_dev, u32 *ind_tbl) +{ + struct ena_rss *rss = &ena_dev->rss; + struct ena_admin_get_feat_resp get_resp; + u32 tbl_size; + int i, rc; + + tbl_size = (1ULL << rss->tbl_log_size) * + sizeof(struct ena_admin_rss_ind_table_entry); + + rc = ena_com_get_feature_ex(ena_dev, &get_resp, + ENA_ADMIN_RSS_REDIRECTION_TABLE_CONFIG, + rss->rss_ind_tbl_dma_addr, + tbl_size); + if (unlikely(rc)) + return rc; + + if (!ind_tbl) + return 0; + + rc = ena_com_ind_tbl_convert_from_device(ena_dev); + if (unlikely(rc)) + return rc; + + for (i = 0; i < (1 << rss->tbl_log_size); i++) + ind_tbl[i] = rss->host_rss_ind_tbl[i]; + + return 0; +} + +int ena_com_rss_init(struct ena_com_dev *ena_dev, u16 indr_tbl_log_size) +{ + int rc; + + memset(&ena_dev->rss, 0x0, sizeof(ena_dev->rss)); + + rc = ena_com_indirect_table_allocate(ena_dev, indr_tbl_log_size); + if (unlikely(rc)) + goto err_indr_tbl; + + rc = ena_com_hash_key_allocate(ena_dev); + if (unlikely(rc)) + goto err_hash_key; + + rc = ena_com_hash_ctrl_init(ena_dev); + if (unlikely(rc)) + goto err_hash_ctrl; + + return 0; + +err_hash_ctrl: + ena_com_hash_key_destroy(ena_dev); +err_hash_key: + ena_com_indirect_table_destroy(ena_dev); +err_indr_tbl: + + return rc; +} + +void ena_com_rss_destroy(struct ena_com_dev *ena_dev) +{ + ena_com_indirect_table_destroy(ena_dev); + ena_com_hash_key_destroy(ena_dev); + ena_com_hash_ctrl_destroy(ena_dev); + + memset(&ena_dev->rss, 0x0, sizeof(ena_dev->rss)); +} + +int ena_com_allocate_host_info(struct ena_com_dev *ena_dev) +{ + struct ena_host_attribute *host_attr = &ena_dev->host_attr; + + host_attr->host_info = + dma_zalloc_coherent(ena_dev->dmadev, SZ_4K, + &host_attr->host_info_dma_addr, GFP_KERNEL); + if (unlikely(!host_attr->host_info)) + return -ENOMEM; + + return 0; +} + +int ena_com_allocate_debug_area(struct ena_com_dev *ena_dev, + u32 debug_area_size) +{ + struct ena_host_attribute *host_attr = &ena_dev->host_attr; + + host_attr->debug_area_virt_addr = + dma_zalloc_coherent(ena_dev->dmadev, debug_area_size, + &host_attr->debug_area_dma_addr, GFP_KERNEL); + if (unlikely(!host_attr->debug_area_virt_addr)) { + host_attr->debug_area_size = 0; + return -ENOMEM; + } + + host_attr->debug_area_size = debug_area_size; + + return 0; +} + +void ena_com_delete_host_info(struct ena_com_dev *ena_dev) +{ + struct ena_host_attribute *host_attr = &ena_dev->host_attr; + + if (host_attr->host_info) { + dma_free_coherent(ena_dev->dmadev, SZ_4K, host_attr->host_info, + host_attr->host_info_dma_addr); + host_attr->host_info = NULL; + } +} + +void ena_com_delete_debug_area(struct ena_com_dev *ena_dev) +{ + struct ena_host_attribute *host_attr = &ena_dev->host_attr; + + if (host_attr->debug_area_virt_addr) { + dma_free_coherent(ena_dev->dmadev, host_attr->debug_area_size, + host_attr->debug_area_virt_addr, + host_attr->debug_area_dma_addr); + host_attr->debug_area_virt_addr = NULL; + } +} + +int ena_com_set_host_attributes(struct ena_com_dev *ena_dev) +{ + struct ena_host_attribute *host_attr = &ena_dev->host_attr; + struct ena_com_admin_queue *admin_queue; + struct ena_admin_set_feat_cmd cmd; + struct ena_admin_set_feat_resp resp; + + int ret; + + if (!ena_com_check_supported_feature_id(ena_dev, + ENA_ADMIN_HOST_ATTR_CONFIG)) { + pr_warn("Set host attribute isn't supported\n"); + return -EPERM; + } + + memset(&cmd, 0x0, sizeof(cmd)); + admin_queue = &ena_dev->admin_queue; + + cmd.aq_common_descriptor.opcode = ENA_ADMIN_SET_FEATURE; + cmd.feat_common.feature_id = ENA_ADMIN_HOST_ATTR_CONFIG; + + ret = ena_com_mem_addr_set(ena_dev, + &cmd.u.host_attr.debug_ba, + host_attr->debug_area_dma_addr); + if (unlikely(ret)) { + pr_err("memory address set failed\n"); + return ret; + } + + ret = ena_com_mem_addr_set(ena_dev, + &cmd.u.host_attr.os_info_ba, + host_attr->host_info_dma_addr); + if (unlikely(ret)) { + pr_err("memory address set failed\n"); + return ret; + } + + cmd.u.host_attr.debug_area_size = host_attr->debug_area_size; + + ret = ena_com_execute_admin_command(admin_queue, + (struct ena_admin_aq_entry *)&cmd, + sizeof(cmd), + (struct ena_admin_acq_entry *)&resp, + sizeof(resp)); + + if (unlikely(ret)) + pr_err("Failed to set host attributes: %d\n", ret); + + return ret; +} + +/* Interrupt moderation */ +bool ena_com_interrupt_moderation_supported(struct ena_com_dev *ena_dev) +{ + return ena_com_check_supported_feature_id(ena_dev, + ENA_ADMIN_INTERRUPT_MODERATION); +} + +int ena_com_update_nonadaptive_moderation_interval_tx(struct ena_com_dev *ena_dev, + u32 tx_coalesce_usecs) +{ + if (!ena_dev->intr_delay_resolution) { + pr_err("Illegal interrupt delay granularity value\n"); + return -EFAULT; + } + + ena_dev->intr_moder_tx_interval = tx_coalesce_usecs / + ena_dev->intr_delay_resolution; + + return 0; +} + +int ena_com_update_nonadaptive_moderation_interval_rx(struct ena_com_dev *ena_dev, + u32 rx_coalesce_usecs) +{ + if (!ena_dev->intr_delay_resolution) { + pr_err("Illegal interrupt delay granularity value\n"); + return -EFAULT; + } + + /* We use LOWEST entry of moderation table for storing + * nonadaptive interrupt coalescing values + */ + ena_dev->intr_moder_tbl[ENA_INTR_MODER_LOWEST].intr_moder_interval = + rx_coalesce_usecs / ena_dev->intr_delay_resolution; + + return 0; +} + +void ena_com_destroy_interrupt_moderation(struct ena_com_dev *ena_dev) +{ + if (ena_dev->intr_moder_tbl) + devm_kfree(ena_dev->dmadev, ena_dev->intr_moder_tbl); + ena_dev->intr_moder_tbl = NULL; +} + +int ena_com_init_interrupt_moderation(struct ena_com_dev *ena_dev) +{ + struct ena_admin_get_feat_resp get_resp; + u16 delay_resolution; + int rc; + + rc = ena_com_get_feature(ena_dev, &get_resp, + ENA_ADMIN_INTERRUPT_MODERATION); + + if (rc) { + if (rc == -EPERM) { + pr_info("Feature %d isn't supported\n", + ENA_ADMIN_INTERRUPT_MODERATION); + rc = 0; + } else { + pr_err("Failed to get interrupt moderation admin cmd. rc: %d\n", + rc); + } + + /* no moderation supported, disable adaptive support */ + ena_com_disable_adaptive_moderation(ena_dev); + return rc; + } + + rc = ena_com_init_interrupt_moderation_table(ena_dev); + if (rc) + goto err; + + /* if moderation is supported by device we set adaptive moderation */ + delay_resolution = get_resp.u.intr_moderation.intr_delay_resolution; + ena_com_update_intr_delay_resolution(ena_dev, delay_resolution); + ena_com_enable_adaptive_moderation(ena_dev); + + return 0; +err: + ena_com_destroy_interrupt_moderation(ena_dev); + return rc; +} + +void ena_com_config_default_interrupt_moderation_table(struct ena_com_dev *ena_dev) +{ + struct ena_intr_moder_entry *intr_moder_tbl = ena_dev->intr_moder_tbl; + + if (!intr_moder_tbl) + return; + + intr_moder_tbl[ENA_INTR_MODER_LOWEST].intr_moder_interval = + ENA_INTR_LOWEST_USECS; + intr_moder_tbl[ENA_INTR_MODER_LOWEST].pkts_per_interval = + ENA_INTR_LOWEST_PKTS; + intr_moder_tbl[ENA_INTR_MODER_LOWEST].bytes_per_interval = + ENA_INTR_LOWEST_BYTES; + + intr_moder_tbl[ENA_INTR_MODER_LOW].intr_moder_interval = + ENA_INTR_LOW_USECS; + intr_moder_tbl[ENA_INTR_MODER_LOW].pkts_per_interval = + ENA_INTR_LOW_PKTS; + intr_moder_tbl[ENA_INTR_MODER_LOW].bytes_per_interval = + ENA_INTR_LOW_BYTES; + + intr_moder_tbl[ENA_INTR_MODER_MID].intr_moder_interval = + ENA_INTR_MID_USECS; + intr_moder_tbl[ENA_INTR_MODER_MID].pkts_per_interval = + ENA_INTR_MID_PKTS; + intr_moder_tbl[ENA_INTR_MODER_MID].bytes_per_interval = + ENA_INTR_MID_BYTES; + + intr_moder_tbl[ENA_INTR_MODER_HIGH].intr_moder_interval = + ENA_INTR_HIGH_USECS; + intr_moder_tbl[ENA_INTR_MODER_HIGH].pkts_per_interval = + ENA_INTR_HIGH_PKTS; + intr_moder_tbl[ENA_INTR_MODER_HIGH].bytes_per_interval = + ENA_INTR_HIGH_BYTES; + + intr_moder_tbl[ENA_INTR_MODER_HIGHEST].intr_moder_interval = + ENA_INTR_HIGHEST_USECS; + intr_moder_tbl[ENA_INTR_MODER_HIGHEST].pkts_per_interval = + ENA_INTR_HIGHEST_PKTS; + intr_moder_tbl[ENA_INTR_MODER_HIGHEST].bytes_per_interval = + ENA_INTR_HIGHEST_BYTES; +} + +unsigned int ena_com_get_nonadaptive_moderation_interval_tx(struct ena_com_dev *ena_dev) +{ + return ena_dev->intr_moder_tx_interval; +} + +unsigned int ena_com_get_nonadaptive_moderation_interval_rx(struct ena_com_dev *ena_dev) +{ + struct ena_intr_moder_entry *intr_moder_tbl = ena_dev->intr_moder_tbl; + + if (intr_moder_tbl) + return intr_moder_tbl[ENA_INTR_MODER_LOWEST].intr_moder_interval; + + return 0; +} + +void ena_com_init_intr_moderation_entry(struct ena_com_dev *ena_dev, + enum ena_intr_moder_level level, + struct ena_intr_moder_entry *entry) +{ + struct ena_intr_moder_entry *intr_moder_tbl = ena_dev->intr_moder_tbl; + + if (level >= ENA_INTR_MAX_NUM_OF_LEVELS) + return; + + intr_moder_tbl[level].intr_moder_interval = entry->intr_moder_interval; + if (ena_dev->intr_delay_resolution) + intr_moder_tbl[level].intr_moder_interval /= + ena_dev->intr_delay_resolution; + intr_moder_tbl[level].pkts_per_interval = entry->pkts_per_interval; + + /* use hardcoded value until ethtool supports bytecount parameter */ + if (entry->bytes_per_interval != ENA_INTR_BYTE_COUNT_NOT_SUPPORTED) + intr_moder_tbl[level].bytes_per_interval = entry->bytes_per_interval; +} + +void ena_com_get_intr_moderation_entry(struct ena_com_dev *ena_dev, + enum ena_intr_moder_level level, + struct ena_intr_moder_entry *entry) +{ + struct ena_intr_moder_entry *intr_moder_tbl = ena_dev->intr_moder_tbl; + + if (level >= ENA_INTR_MAX_NUM_OF_LEVELS) + return; + + entry->intr_moder_interval = intr_moder_tbl[level].intr_moder_interval; + if (ena_dev->intr_delay_resolution) + entry->intr_moder_interval *= ena_dev->intr_delay_resolution; + entry->pkts_per_interval = + intr_moder_tbl[level].pkts_per_interval; + entry->bytes_per_interval = intr_moder_tbl[level].bytes_per_interval; +} diff --git a/drivers/net/ethernet/amazon/ena/ena_com.h b/drivers/net/ethernet/amazon/ena/ena_com.h new file mode 100644 index 000000000000..509d7b8e15ab --- /dev/null +++ b/drivers/net/ethernet/amazon/ena/ena_com.h @@ -0,0 +1,1038 @@ +/* + * Copyright 2015 Amazon.com, Inc. or its affiliates. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef ENA_COM +#define ENA_COM + +#include <linux/delay.h> +#include <linux/dma-mapping.h> +#include <linux/gfp.h> +#include <linux/sched.h> +#include <linux/sizes.h> +#include <linux/spinlock.h> +#include <linux/types.h> +#include <linux/wait.h> + +#include "ena_common_defs.h" +#include "ena_admin_defs.h" +#include "ena_eth_io_defs.h" +#include "ena_regs_defs.h" + +#undef pr_fmt +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#define ENA_MAX_NUM_IO_QUEUES 128U +/* We need to queues for each IO (on for Tx and one for Rx) */ +#define ENA_TOTAL_NUM_QUEUES (2 * (ENA_MAX_NUM_IO_QUEUES)) + +#define ENA_MAX_HANDLERS 256 + +#define ENA_MAX_PHYS_ADDR_SIZE_BITS 48 + +/* Unit in usec */ +#define ENA_REG_READ_TIMEOUT 200000 + +#define ADMIN_SQ_SIZE(depth) ((depth) * sizeof(struct ena_admin_aq_entry)) +#define ADMIN_CQ_SIZE(depth) ((depth) * sizeof(struct ena_admin_acq_entry)) +#define ADMIN_AENQ_SIZE(depth) ((depth) * sizeof(struct ena_admin_aenq_entry)) + +/*****************************************************************************/ +/*****************************************************************************/ +/* ENA adaptive interrupt moderation settings */ + +#define ENA_INTR_LOWEST_USECS (0) +#define ENA_INTR_LOWEST_PKTS (3) +#define ENA_INTR_LOWEST_BYTES (2 * 1524) + +#define ENA_INTR_LOW_USECS (32) +#define ENA_INTR_LOW_PKTS (12) +#define ENA_INTR_LOW_BYTES (16 * 1024) + +#define ENA_INTR_MID_USECS (80) +#define ENA_INTR_MID_PKTS (48) +#define ENA_INTR_MID_BYTES (64 * 1024) + +#define ENA_INTR_HIGH_USECS (128) +#define ENA_INTR_HIGH_PKTS (96) +#define ENA_INTR_HIGH_BYTES (128 * 1024) + +#define ENA_INTR_HIGHEST_USECS (192) +#define ENA_INTR_HIGHEST_PKTS (128) +#define ENA_INTR_HIGHEST_BYTES (192 * 1024) + +#define ENA_INTR_INITIAL_TX_INTERVAL_USECS 196 +#define ENA_INTR_INITIAL_RX_INTERVAL_USECS 4 +#define ENA_INTR_DELAY_OLD_VALUE_WEIGHT 6 +#define ENA_INTR_DELAY_NEW_VALUE_WEIGHT 4 +#define ENA_INTR_MODER_LEVEL_STRIDE 2 +#define ENA_INTR_BYTE_COUNT_NOT_SUPPORTED 0xFFFFFF + +enum ena_intr_moder_level { + ENA_INTR_MODER_LOWEST = 0, + ENA_INTR_MODER_LOW, + ENA_INTR_MODER_MID, + ENA_INTR_MODER_HIGH, + ENA_INTR_MODER_HIGHEST, + ENA_INTR_MAX_NUM_OF_LEVELS, +}; + +struct ena_intr_moder_entry { + unsigned int intr_moder_interval; + unsigned int pkts_per_interval; + unsigned int bytes_per_interval; +}; + +enum queue_direction { + ENA_COM_IO_QUEUE_DIRECTION_TX, + ENA_COM_IO_QUEUE_DIRECTION_RX +}; + +struct ena_com_buf { + dma_addr_t paddr; /**< Buffer physical address */ + u16 len; /**< Buffer length in bytes */ +}; + +struct ena_com_rx_buf_info { + u16 len; + u16 req_id; +}; + +struct ena_com_io_desc_addr { + u8 __iomem *pbuf_dev_addr; /* LLQ address */ + u8 *virt_addr; + dma_addr_t phys_addr; +}; + +struct ena_com_tx_meta { + u16 mss; + u16 l3_hdr_len; + u16 l3_hdr_offset; + u16 l4_hdr_len; /* In words */ +}; + +struct ena_com_io_cq { + struct ena_com_io_desc_addr cdesc_addr; + + /* Interrupt unmask register */ + u32 __iomem *unmask_reg; + + /* The completion queue head doorbell register */ + u32 __iomem *cq_head_db_reg; + + /* numa configuration register (for TPH) */ + u32 __iomem *numa_node_cfg_reg; + + /* The value to write to the above register to unmask + * the interrupt of this queue + */ + u32 msix_vector; + + enum queue_direction direction; + + /* holds the number of cdesc of the current packet */ + u16 cur_rx_pkt_cdesc_count; + /* save the firt cdesc idx of the current packet */ + u16 cur_rx_pkt_cdesc_start_idx; + + u16 q_depth; + /* Caller qid */ + u16 qid; + + /* Device queue index */ + u16 idx; + u16 head; + u16 last_head_update; + u8 phase; + u8 cdesc_entry_size_in_bytes; + +} ____cacheline_aligned; + +struct ena_com_io_sq { + struct ena_com_io_desc_addr desc_addr; + + u32 __iomem *db_addr; + u8 __iomem *header_addr; + + enum queue_direction direction; + enum ena_admin_placement_policy_type mem_queue_type; + + u32 msix_vector; + struct ena_com_tx_meta cached_tx_meta; + + u16 q_depth; + u16 qid; + + u16 idx; + u16 tail; + u16 next_to_comp; + u32 tx_max_header_size; + u8 phase; + u8 desc_entry_size; + u8 dma_addr_bits; +} ____cacheline_aligned; + +struct ena_com_admin_cq { + struct ena_admin_acq_entry *entries; + dma_addr_t dma_addr; + + u16 head; + u8 phase; +}; + +struct ena_com_admin_sq { + struct ena_admin_aq_entry *entries; + dma_addr_t dma_addr; + + u32 __iomem *db_addr; + + u16 head; + u16 tail; + u8 phase; + +}; + +struct ena_com_stats_admin { + u32 aborted_cmd; + u32 submitted_cmd; + u32 completed_cmd; + u32 out_of_space; + u32 no_completion; +}; + +struct ena_com_admin_queue { + void *q_dmadev; + spinlock_t q_lock; /* spinlock for the admin queue */ + struct ena_comp_ctx *comp_ctx; + u16 q_depth; + struct ena_com_admin_cq cq; + struct ena_com_admin_sq sq; + + /* Indicate if the admin queue should poll for completion */ + bool polling; + + u16 curr_cmd_id; + + /* Indicate that the ena was initialized and can + * process new admin commands + */ + bool running_state; + + /* Count the number of outstanding admin commands */ + atomic_t outstanding_cmds; + + struct ena_com_stats_admin stats; +}; + +struct ena_aenq_handlers; + +struct ena_com_aenq { + u16 head; + u8 phase; + struct ena_admin_aenq_entry *entries; + dma_addr_t dma_addr; + u16 q_depth; + struct ena_aenq_handlers *aenq_handlers; +}; + +struct ena_com_mmio_read { + struct ena_admin_ena_mmio_req_read_less_resp *read_resp; + dma_addr_t read_resp_dma_addr; + u16 seq_num; + bool readless_supported; + /* spin lock to ensure a single outstanding read */ + spinlock_t lock; +}; + +struct ena_rss { + /* Indirect table */ + u16 *host_rss_ind_tbl; + struct ena_admin_rss_ind_table_entry *rss_ind_tbl; + dma_addr_t rss_ind_tbl_dma_addr; + u16 tbl_log_size; + + /* Hash key */ + enum ena_admin_hash_functions hash_func; + struct ena_admin_feature_rss_flow_hash_control *hash_key; + dma_addr_t hash_key_dma_addr; + u32 hash_init_val; + + /* Flow Control */ + struct ena_admin_feature_rss_hash_control *hash_ctrl; + dma_addr_t hash_ctrl_dma_addr; + +}; + +struct ena_host_attribute { + /* Debug area */ + u8 *debug_area_virt_addr; + dma_addr_t debug_area_dma_addr; + u32 debug_area_size; + + /* Host information */ + struct ena_admin_host_info *host_info; + dma_addr_t host_info_dma_addr; +}; + +/* Each ena_dev is a PCI function. */ +struct ena_com_dev { + struct ena_com_admin_queue admin_queue; + struct ena_com_aenq aenq; + struct ena_com_io_cq io_cq_queues[ENA_TOTAL_NUM_QUEUES]; + struct ena_com_io_sq io_sq_queues[ENA_TOTAL_NUM_QUEUES]; + u8 __iomem *reg_bar; + void __iomem *mem_bar; + void *dmadev; + + enum ena_admin_placement_policy_type tx_mem_queue_type; + u32 tx_max_header_size; + u16 stats_func; /* Selected function for extended statistic dump */ + u16 stats_queue; /* Selected queue for extended statistic dump */ + + struct ena_com_mmio_read mmio_read; + + struct ena_rss rss; + u32 supported_features; + u32 dma_addr_bits; + + struct ena_host_attribute host_attr; + bool adaptive_coalescing; + u16 intr_delay_resolution; + u32 intr_moder_tx_interval; + struct ena_intr_moder_entry *intr_moder_tbl; +}; + +struct ena_com_dev_get_features_ctx { + struct ena_admin_queue_feature_desc max_queues; + struct ena_admin_device_attr_feature_desc dev_attr; + struct ena_admin_feature_aenq_desc aenq; + struct ena_admin_feature_offload_desc offload; +}; + +struct ena_com_create_io_ctx { + enum ena_admin_placement_policy_type mem_queue_type; + enum queue_direction direction; + int numa_node; + u32 msix_vector; + u16 queue_size; + u16 qid; +}; + +typedef void (*ena_aenq_handler)(void *data, + struct ena_admin_aenq_entry *aenq_e); + +/* Holds aenq handlers. Indexed by AENQ event group */ +struct ena_aenq_handlers { + ena_aenq_handler handlers[ENA_MAX_HANDLERS]; + ena_aenq_handler unimplemented_handler; +}; + +/*****************************************************************************/ +/*****************************************************************************/ + +/* ena_com_mmio_reg_read_request_init - Init the mmio reg read mechanism + * @ena_dev: ENA communication layer struct + * + * Initialize the register read mechanism. + * + * @note: This method must be the first stage in the initialization sequence. + * + * @return - 0 on success, negative value on failure. + */ +int ena_com_mmio_reg_read_request_init(struct ena_com_dev *ena_dev); + +/* ena_com_set_mmio_read_mode - Enable/disable the mmio reg read mechanism + * @ena_dev: ENA communication layer struct + * @readless_supported: readless mode (enable/disable) + */ +void ena_com_set_mmio_read_mode(struct ena_com_dev *ena_dev, + bool readless_supported); + +/* ena_com_mmio_reg_read_request_write_dev_addr - Write the mmio reg read return + * value physical address. + * @ena_dev: ENA communication layer struct + */ +void ena_com_mmio_reg_read_request_write_dev_addr(struct ena_com_dev *ena_dev); + +/* ena_com_mmio_reg_read_request_destroy - Destroy the mmio reg read mechanism + * @ena_dev: ENA communication layer struct + */ +void ena_com_mmio_reg_read_request_destroy(struct ena_com_dev *ena_dev); + +/* ena_com_admin_init - Init the admin and the async queues + * @ena_dev: ENA communication layer struct + * @aenq_handlers: Those handlers to be called upon event. + * @init_spinlock: Indicate if this method should init the admin spinlock or + * the spinlock was init before (for example, in a case of FLR). + * + * Initialize the admin submission and completion queues. + * Initialize the asynchronous events notification queues. + * + * @return - 0 on success, negative value on failure. + */ +int ena_com_admin_init(struct ena_com_dev *ena_dev, + struct ena_aenq_handlers *aenq_handlers, + bool init_spinlock); + +/* ena_com_admin_destroy - Destroy the admin and the async events queues. + * @ena_dev: ENA communication layer struct + * + * @note: Before calling this method, the caller must validate that the device + * won't send any additional admin completions/aenq. + * To achieve that, a FLR is recommended. + */ +void ena_com_admin_destroy(struct ena_com_dev *ena_dev); + +/* ena_com_dev_reset - Perform device FLR to the device. + * @ena_dev: ENA communication layer struct + * + * @return - 0 on success, negative value on failure. + */ +int ena_com_dev_reset(struct ena_com_dev *ena_dev); + +/* ena_com_create_io_queue - Create io queue. + * @ena_dev: ENA communication layer struct + * @ctx - create context structure + * + * Create the submission and the completion queues. + * + * @return - 0 on success, negative value on failure. + */ +int ena_com_create_io_queue(struct ena_com_dev *ena_dev, + struct ena_com_create_io_ctx *ctx); + +/* ena_com_destroy_io_queue - Destroy IO queue with the queue id - qid. + * @ena_dev: ENA communication layer struct + * @qid - the caller virtual queue id. + */ +void ena_com_destroy_io_queue(struct ena_com_dev *ena_dev, u16 qid); + +/* ena_com_get_io_handlers - Return the io queue handlers + * @ena_dev: ENA communication layer struct + * @qid - the caller virtual queue id. + * @io_sq - IO submission queue handler + * @io_cq - IO completion queue handler. + * + * @return - 0 on success, negative value on failure. + */ +int ena_com_get_io_handlers(struct ena_com_dev *ena_dev, u16 qid, + struct ena_com_io_sq **io_sq, + struct ena_com_io_cq **io_cq); + +/* ena_com_admin_aenq_enable - ENAble asynchronous event notifications + * @ena_dev: ENA communication layer struct + * + * After this method, aenq event can be received via AENQ. + */ +void ena_com_admin_aenq_enable(struct ena_com_dev *ena_dev); + +/* ena_com_set_admin_running_state - Set the state of the admin queue + * @ena_dev: ENA communication layer struct + * + * Change the state of the admin queue (enable/disable) + */ +void ena_com_set_admin_running_state(struct ena_com_dev *ena_dev, bool state); + +/* ena_com_get_admin_running_state - Get the admin queue state + * @ena_dev: ENA communication layer struct + * + * Retrieve the state of the admin queue (enable/disable) + * + * @return - current polling mode (enable/disable) + */ +bool ena_com_get_admin_running_state(struct ena_com_dev *ena_dev); + +/* ena_com_set_admin_polling_mode - Set the admin completion queue polling mode + * @ena_dev: ENA communication layer struct + * @polling: ENAble/Disable polling mode + * + * Set the admin completion mode. + */ +void ena_com_set_admin_polling_mode(struct ena_com_dev *ena_dev, bool polling); + +/* ena_com_set_admin_polling_mode - Get the admin completion queue polling mode + * @ena_dev: ENA communication layer struct + * + * Get the admin completion mode. + * If polling mode is on, ena_com_execute_admin_command will perform a + * polling on the admin completion queue for the commands completion, + * otherwise it will wait on wait event. + * + * @return state + */ +bool ena_com_get_ena_admin_polling_mode(struct ena_com_dev *ena_dev); + +/* ena_com_admin_q_comp_intr_handler - admin queue interrupt handler + * @ena_dev: ENA communication layer struct + * + * This method go over the admin completion queue and wake up all the pending + * threads that wait on the commands wait event. + * + * @note: Should be called after MSI-X interrupt. + */ +void ena_com_admin_q_comp_intr_handler(struct ena_com_dev *ena_dev); + +/* ena_com_aenq_intr_handler - AENQ interrupt handler + * @ena_dev: ENA communication layer struct + * + * This method go over the async event notification queue and call the proper + * aenq handler. + */ +void ena_com_aenq_intr_handler(struct ena_com_dev *dev, void *data); + +/* ena_com_abort_admin_commands - Abort all the outstanding admin commands. + * @ena_dev: ENA communication layer struct + * + * This method aborts all the outstanding admin commands. + * The caller should then call ena_com_wait_for_abort_completion to make sure + * all the commands were completed. + */ +void ena_com_abort_admin_commands(struct ena_com_dev *ena_dev); + +/* ena_com_wait_for_abort_completion - Wait for admin commands abort. + * @ena_dev: ENA communication layer struct + * + * This method wait until all the outstanding admin commands will be completed. + */ +void ena_com_wait_for_abort_completion(struct ena_com_dev *ena_dev); + +/* ena_com_validate_version - Validate the device parameters + * @ena_dev: ENA communication layer struct + * + * This method validate the device parameters are the same as the saved + * parameters in ena_dev. + * This method is useful after device reset, to validate the device mac address + * and the device offloads are the same as before the reset. + * + * @return - 0 on success negative value otherwise. + */ +int ena_com_validate_version(struct ena_com_dev *ena_dev); + +/* ena_com_get_link_params - Retrieve physical link parameters. + * @ena_dev: ENA communication layer struct + * @resp: Link parameters + * + * Retrieve the physical link parameters, + * like speed, auto-negotiation and full duplex support. + * + * @return - 0 on Success negative value otherwise. + */ +int ena_com_get_link_params(struct ena_com_dev *ena_dev, + struct ena_admin_get_feat_resp *resp); + +/* ena_com_get_dma_width - Retrieve physical dma address width the device + * supports. + * @ena_dev: ENA communication layer struct + * + * Retrieve the maximum physical address bits the device can handle. + * + * @return: > 0 on Success and negative value otherwise. + */ +int ena_com_get_dma_width(struct ena_com_dev *ena_dev); + +/* ena_com_set_aenq_config - Set aenq groups configurations + * @ena_dev: ENA communication layer struct + * @groups flag: bit fields flags of enum ena_admin_aenq_group. + * + * Configure which aenq event group the driver would like to receive. + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_set_aenq_config(struct ena_com_dev *ena_dev, u32 groups_flag); + +/* ena_com_get_dev_attr_feat - Get device features + * @ena_dev: ENA communication layer struct + * @get_feat_ctx: returned context that contain the get features. + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_get_dev_attr_feat(struct ena_com_dev *ena_dev, + struct ena_com_dev_get_features_ctx *get_feat_ctx); + +/* ena_com_get_dev_basic_stats - Get device basic statistics + * @ena_dev: ENA communication layer struct + * @stats: stats return value + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_get_dev_basic_stats(struct ena_com_dev *ena_dev, + struct ena_admin_basic_stats *stats); + +/* ena_com_set_dev_mtu - Configure the device mtu. + * @ena_dev: ENA communication layer struct + * @mtu: mtu value + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_set_dev_mtu(struct ena_com_dev *ena_dev, int mtu); + +/* ena_com_get_offload_settings - Retrieve the device offloads capabilities + * @ena_dev: ENA communication layer struct + * @offlad: offload return value + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_get_offload_settings(struct ena_com_dev *ena_dev, + struct ena_admin_feature_offload_desc *offload); + +/* ena_com_rss_init - Init RSS + * @ena_dev: ENA communication layer struct + * @log_size: indirection log size + * + * Allocate RSS/RFS resources. + * The caller then can configure rss using ena_com_set_hash_function, + * ena_com_set_hash_ctrl and ena_com_indirect_table_set. + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_rss_init(struct ena_com_dev *ena_dev, u16 log_size); + +/* ena_com_rss_destroy - Destroy rss + * @ena_dev: ENA communication layer struct + * + * Free all the RSS/RFS resources. + */ +void ena_com_rss_destroy(struct ena_com_dev *ena_dev); + +/* ena_com_fill_hash_function - Fill RSS hash function + * @ena_dev: ENA communication layer struct + * @func: The hash function (Toeplitz or crc) + * @key: Hash key (for toeplitz hash) + * @key_len: key length (max length 10 DW) + * @init_val: initial value for the hash function + * + * Fill the ena_dev resources with the desire hash function, hash key, key_len + * and key initial value (if needed by the hash function). + * To flush the key into the device the caller should call + * ena_com_set_hash_function. + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_fill_hash_function(struct ena_com_dev *ena_dev, + enum ena_admin_hash_functions func, + const u8 *key, u16 key_len, u32 init_val); + +/* ena_com_set_hash_function - Flush the hash function and it dependencies to + * the device. + * @ena_dev: ENA communication layer struct + * + * Flush the hash function and it dependencies (key, key length and + * initial value) if needed. + * + * @note: Prior to this method the caller should call ena_com_fill_hash_function + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_set_hash_function(struct ena_com_dev *ena_dev); + +/* ena_com_get_hash_function - Retrieve the hash function and the hash key + * from the device. + * @ena_dev: ENA communication layer struct + * @func: hash function + * @key: hash key + * + * Retrieve the hash function and the hash key from the device. + * + * @note: If the caller called ena_com_fill_hash_function but didn't flash + * it to the device, the new configuration will be lost. + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_get_hash_function(struct ena_com_dev *ena_dev, + enum ena_admin_hash_functions *func, + u8 *key); + +/* ena_com_fill_hash_ctrl - Fill RSS hash control + * @ena_dev: ENA communication layer struct. + * @proto: The protocol to configure. + * @hash_fields: bit mask of ena_admin_flow_hash_fields + * + * Fill the ena_dev resources with the desire hash control (the ethernet + * fields that take part of the hash) for a specific protocol. + * To flush the hash control to the device, the caller should call + * ena_com_set_hash_ctrl. + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_fill_hash_ctrl(struct ena_com_dev *ena_dev, + enum ena_admin_flow_hash_proto proto, + u16 hash_fields); + +/* ena_com_set_hash_ctrl - Flush the hash control resources to the device. + * @ena_dev: ENA communication layer struct + * + * Flush the hash control (the ethernet fields that take part of the hash) + * + * @note: Prior to this method the caller should call ena_com_fill_hash_ctrl. + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_set_hash_ctrl(struct ena_com_dev *ena_dev); + +/* ena_com_get_hash_ctrl - Retrieve the hash control from the device. + * @ena_dev: ENA communication layer struct + * @proto: The protocol to retrieve. + * @fields: bit mask of ena_admin_flow_hash_fields. + * + * Retrieve the hash control from the device. + * + * @note, If the caller called ena_com_fill_hash_ctrl but didn't flash + * it to the device, the new configuration will be lost. + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_get_hash_ctrl(struct ena_com_dev *ena_dev, + enum ena_admin_flow_hash_proto proto, + u16 *fields); + +/* ena_com_set_default_hash_ctrl - Set the hash control to a default + * configuration. + * @ena_dev: ENA communication layer struct + * + * Fill the ena_dev resources with the default hash control configuration. + * To flush the hash control to the device, the caller should call + * ena_com_set_hash_ctrl. + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_set_default_hash_ctrl(struct ena_com_dev *ena_dev); + +/* ena_com_indirect_table_fill_entry - Fill a single entry in the RSS + * indirection table + * @ena_dev: ENA communication layer struct. + * @entry_idx - indirection table entry. + * @entry_value - redirection value + * + * Fill a single entry of the RSS indirection table in the ena_dev resources. + * To flush the indirection table to the device, the called should call + * ena_com_indirect_table_set. + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_indirect_table_fill_entry(struct ena_com_dev *ena_dev, + u16 entry_idx, u16 entry_value); + +/* ena_com_indirect_table_set - Flush the indirection table to the device. + * @ena_dev: ENA communication layer struct + * + * Flush the indirection hash control to the device. + * Prior to this method the caller should call ena_com_indirect_table_fill_entry + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_indirect_table_set(struct ena_com_dev *ena_dev); + +/* ena_com_indirect_table_get - Retrieve the indirection table from the device. + * @ena_dev: ENA communication layer struct + * @ind_tbl: indirection table + * + * Retrieve the RSS indirection table from the device. + * + * @note: If the caller called ena_com_indirect_table_fill_entry but didn't flash + * it to the device, the new configuration will be lost. + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_indirect_table_get(struct ena_com_dev *ena_dev, u32 *ind_tbl); + +/* ena_com_allocate_host_info - Allocate host info resources. + * @ena_dev: ENA communication layer struct + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_allocate_host_info(struct ena_com_dev *ena_dev); + +/* ena_com_allocate_debug_area - Allocate debug area. + * @ena_dev: ENA communication layer struct + * @debug_area_size - debug area size. + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_allocate_debug_area(struct ena_com_dev *ena_dev, + u32 debug_area_size); + +/* ena_com_delete_debug_area - Free the debug area resources. + * @ena_dev: ENA communication layer struct + * + * Free the allocate debug area. + */ +void ena_com_delete_debug_area(struct ena_com_dev *ena_dev); + +/* ena_com_delete_host_info - Free the host info resources. + * @ena_dev: ENA communication layer struct + * + * Free the allocate host info. + */ +void ena_com_delete_host_info(struct ena_com_dev *ena_dev); + +/* ena_com_set_host_attributes - Update the device with the host + * attributes (debug area and host info) base address. + * @ena_dev: ENA communication layer struct + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_set_host_attributes(struct ena_com_dev *ena_dev); + +/* ena_com_create_io_cq - Create io completion queue. + * @ena_dev: ENA communication layer struct + * @io_cq - io completion queue handler + + * Create IO completion queue. + * + * @return - 0 on success, negative value on failure. + */ +int ena_com_create_io_cq(struct ena_com_dev *ena_dev, + struct ena_com_io_cq *io_cq); + +/* ena_com_destroy_io_cq - Destroy io completion queue. + * @ena_dev: ENA communication layer struct + * @io_cq - io completion queue handler + + * Destroy IO completion queue. + * + * @return - 0 on success, negative value on failure. + */ +int ena_com_destroy_io_cq(struct ena_com_dev *ena_dev, + struct ena_com_io_cq *io_cq); + +/* ena_com_execute_admin_command - Execute admin command + * @admin_queue: admin queue. + * @cmd: the admin command to execute. + * @cmd_size: the command size. + * @cmd_completion: command completion return value. + * @cmd_comp_size: command completion size. + + * Submit an admin command and then wait until the device will return a + * completion. + * The completion will be copyed into cmd_comp. + * + * @return - 0 on success, negative value on failure. + */ +int ena_com_execute_admin_command(struct ena_com_admin_queue *admin_queue, + struct ena_admin_aq_entry *cmd, + size_t cmd_size, + struct ena_admin_acq_entry *cmd_comp, + size_t cmd_comp_size); + +/* ena_com_init_interrupt_moderation - Init interrupt moderation + * @ena_dev: ENA communication layer struct + * + * @return - 0 on success, negative value on failure. + */ +int ena_com_init_interrupt_moderation(struct ena_com_dev *ena_dev); + +/* ena_com_destroy_interrupt_moderation - Destroy interrupt moderation resources + * @ena_dev: ENA communication layer struct + */ +void ena_com_destroy_interrupt_moderation(struct ena_com_dev *ena_dev); + +/* ena_com_interrupt_moderation_supported - Return if interrupt moderation + * capability is supported by the device. + * + * @return - supported or not. + */ +bool ena_com_interrupt_moderation_supported(struct ena_com_dev *ena_dev); + +/* ena_com_config_default_interrupt_moderation_table - Restore the interrupt + * moderation table back to the default parameters. + * @ena_dev: ENA communication layer struct + */ +void ena_com_config_default_interrupt_moderation_table(struct ena_com_dev *ena_dev); + +/* ena_com_update_nonadaptive_moderation_interval_tx - Update the + * non-adaptive interval in Tx direction. + * @ena_dev: ENA communication layer struct + * @tx_coalesce_usecs: Interval in usec. + * + * @return - 0 on success, negative value on failure. + */ +int ena_com_update_nonadaptive_moderation_interval_tx(struct ena_com_dev *ena_dev, + u32 tx_coalesce_usecs); + +/* ena_com_update_nonadaptive_moderation_interval_rx - Update the + * non-adaptive interval in Rx direction. + * @ena_dev: ENA communication layer struct + * @rx_coalesce_usecs: Interval in usec. + * + * @return - 0 on success, negative value on failure. + */ +int ena_com_update_nonadaptive_moderation_interval_rx(struct ena_com_dev *ena_dev, + u32 rx_coalesce_usecs); + +/* ena_com_get_nonadaptive_moderation_interval_tx - Retrieve the + * non-adaptive interval in Tx direction. + * @ena_dev: ENA communication layer struct + * + * @return - interval in usec + */ +unsigned int ena_com_get_nonadaptive_moderation_interval_tx(struct ena_com_dev *ena_dev); + +/* ena_com_get_nonadaptive_moderation_interval_rx - Retrieve the + * non-adaptive interval in Rx direction. + * @ena_dev: ENA communication layer struct + * + * @return - interval in usec + */ +unsigned int ena_com_get_nonadaptive_moderation_interval_rx(struct ena_com_dev *ena_dev); + +/* ena_com_init_intr_moderation_entry - Update a single entry in the interrupt + * moderation table. + * @ena_dev: ENA communication layer struct + * @level: Interrupt moderation table level + * @entry: Entry value + * + * Update a single entry in the interrupt moderation table. + */ +void ena_com_init_intr_moderation_entry(struct ena_com_dev *ena_dev, + enum ena_intr_moder_level level, + struct ena_intr_moder_entry *entry); + +/* ena_com_get_intr_moderation_entry - Init ena_intr_moder_entry. + * @ena_dev: ENA communication layer struct + * @level: Interrupt moderation table level + * @entry: Entry to fill. + * + * Initialize the entry according to the adaptive interrupt moderation table. + */ +void ena_com_get_intr_moderation_entry(struct ena_com_dev *ena_dev, + enum ena_intr_moder_level level, + struct ena_intr_moder_entry *entry); + +static inline bool ena_com_get_adaptive_moderation_enabled(struct ena_com_dev *ena_dev) +{ + return ena_dev->adaptive_coalescing; +} + +static inline void ena_com_enable_adaptive_moderation(struct ena_com_dev *ena_dev) +{ + ena_dev->adaptive_coalescing = true; +} + +static inline void ena_com_disable_adaptive_moderation(struct ena_com_dev *ena_dev) +{ + ena_dev->adaptive_coalescing = false; +} + +/* ena_com_calculate_interrupt_delay - Calculate new interrupt delay + * @ena_dev: ENA communication layer struct + * @pkts: Number of packets since the last update + * @bytes: Number of bytes received since the last update. + * @smoothed_interval: Returned interval + * @moder_tbl_idx: Current table level as input update new level as return + * value. + */ +static inline void ena_com_calculate_interrupt_delay(struct ena_com_dev *ena_dev, + unsigned int pkts, + unsigned int bytes, + unsigned int *smoothed_interval, + unsigned int *moder_tbl_idx) +{ + enum ena_intr_moder_level curr_moder_idx, new_moder_idx; + struct ena_intr_moder_entry *curr_moder_entry; + struct ena_intr_moder_entry *pred_moder_entry; + struct ena_intr_moder_entry *new_moder_entry; + struct ena_intr_moder_entry *intr_moder_tbl = ena_dev->intr_moder_tbl; + unsigned int interval; + + /* We apply adaptive moderation on Rx path only. + * Tx uses static interrupt moderation. + */ + if (!pkts || !bytes) + /* Tx interrupt, or spurious interrupt, + * in both cases we just use same delay values + */ + return; + + curr_moder_idx = (enum ena_intr_moder_level)(*moder_tbl_idx); + if (unlikely(curr_moder_idx >= ENA_INTR_MAX_NUM_OF_LEVELS)) { + pr_err("Wrong moderation index %u\n", curr_moder_idx); + return; + } + + curr_moder_entry = &intr_moder_tbl[curr_moder_idx]; + new_moder_idx = curr_moder_idx; + + if (curr_moder_idx == ENA_INTR_MODER_LOWEST) { + if ((pkts > curr_moder_entry->pkts_per_interval) || + (bytes > curr_moder_entry->bytes_per_interval)) + new_moder_idx = + (enum ena_intr_moder_level)(curr_moder_idx + ENA_INTR_MODER_LEVEL_STRIDE); + } else { + pred_moder_entry = &intr_moder_tbl[curr_moder_idx - ENA_INTR_MODER_LEVEL_STRIDE]; + + if ((pkts <= pred_moder_entry->pkts_per_interval) || + (bytes <= pred_moder_entry->bytes_per_interval)) + new_moder_idx = + (enum ena_intr_moder_level)(curr_moder_idx - ENA_INTR_MODER_LEVEL_STRIDE); + else if ((pkts > curr_moder_entry->pkts_per_interval) || + (bytes > curr_moder_entry->bytes_per_interval)) { + if (curr_moder_idx != ENA_INTR_MODER_HIGHEST) + new_moder_idx = + (enum ena_intr_moder_level)(curr_moder_idx + ENA_INTR_MODER_LEVEL_STRIDE); + } + } + new_moder_entry = &intr_moder_tbl[new_moder_idx]; + + interval = new_moder_entry->intr_moder_interval; + *smoothed_interval = ( + (interval * ENA_INTR_DELAY_NEW_VALUE_WEIGHT + + ENA_INTR_DELAY_OLD_VALUE_WEIGHT * (*smoothed_interval)) + 5) / + 10; + + *moder_tbl_idx = new_moder_idx; +} + +/* ena_com_update_intr_reg - Prepare interrupt register + * @intr_reg: interrupt register to update. + * @rx_delay_interval: Rx interval in usecs + * @tx_delay_interval: Tx interval in usecs + * @unmask: unask enable/disable + * + * Prepare interrupt update register with the supplied parameters. + */ +static inline void ena_com_update_intr_reg(struct ena_eth_io_intr_reg *intr_reg, + u32 rx_delay_interval, + u32 tx_delay_interval, + bool unmask) +{ + intr_reg->intr_control = 0; + intr_reg->intr_control |= rx_delay_interval & + ENA_ETH_IO_INTR_REG_RX_INTR_DELAY_MASK; + + intr_reg->intr_control |= + (tx_delay_interval << ENA_ETH_IO_INTR_REG_TX_INTR_DELAY_SHIFT) + & ENA_ETH_IO_INTR_REG_TX_INTR_DELAY_MASK; + + if (unmask) + intr_reg->intr_control |= ENA_ETH_IO_INTR_REG_INTR_UNMASK_MASK; +} + +#endif /* !(ENA_COM) */ diff --git a/drivers/net/ethernet/amazon/ena/ena_common_defs.h b/drivers/net/ethernet/amazon/ena/ena_common_defs.h new file mode 100644 index 000000000000..bb8d73676eab --- /dev/null +++ b/drivers/net/ethernet/amazon/ena/ena_common_defs.h @@ -0,0 +1,48 @@ +/* + * Copyright 2015 - 2016 Amazon.com, Inc. or its affiliates. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _ENA_COMMON_H_ +#define _ENA_COMMON_H_ + +#define ENA_COMMON_SPEC_VERSION_MAJOR 0 /* */ +#define ENA_COMMON_SPEC_VERSION_MINOR 10 /* */ + +/* ENA operates with 48-bit memory addresses. ena_mem_addr_t */ +struct ena_common_mem_addr { + u32 mem_addr_low; + + u16 mem_addr_high; + + /* MBZ */ + u16 reserved16; +}; + +#endif /*_ENA_COMMON_H_ */ diff --git a/drivers/net/ethernet/amazon/ena/ena_eth_com.c b/drivers/net/ethernet/amazon/ena/ena_eth_com.c new file mode 100644 index 000000000000..539c536464a5 --- /dev/null +++ b/drivers/net/ethernet/amazon/ena/ena_eth_com.c @@ -0,0 +1,501 @@ +/* + * Copyright 2015 Amazon.com, Inc. or its affiliates. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "ena_eth_com.h" + +static inline struct ena_eth_io_rx_cdesc_base *ena_com_get_next_rx_cdesc( + struct ena_com_io_cq *io_cq) +{ + struct ena_eth_io_rx_cdesc_base *cdesc; + u16 expected_phase, head_masked; + u16 desc_phase; + + head_masked = io_cq->head & (io_cq->q_depth - 1); + expected_phase = io_cq->phase; + + cdesc = (struct ena_eth_io_rx_cdesc_base *)(io_cq->cdesc_addr.virt_addr + + (head_masked * io_cq->cdesc_entry_size_in_bytes)); + + desc_phase = (cdesc->status & ENA_ETH_IO_RX_CDESC_BASE_PHASE_MASK) >> + ENA_ETH_IO_RX_CDESC_BASE_PHASE_SHIFT; + + if (desc_phase != expected_phase) + return NULL; + + return cdesc; +} + +static inline void ena_com_cq_inc_head(struct ena_com_io_cq *io_cq) +{ + io_cq->head++; + + /* Switch phase bit in case of wrap around */ + if (unlikely((io_cq->head & (io_cq->q_depth - 1)) == 0)) + io_cq->phase ^= 1; +} + +static inline void *get_sq_desc(struct ena_com_io_sq *io_sq) +{ + u16 tail_masked; + u32 offset; + + tail_masked = io_sq->tail & (io_sq->q_depth - 1); + + offset = tail_masked * io_sq->desc_entry_size; + + return (void *)((uintptr_t)io_sq->desc_addr.virt_addr + offset); +} + +static inline void ena_com_copy_curr_sq_desc_to_dev(struct ena_com_io_sq *io_sq) +{ + u16 tail_masked = io_sq->tail & (io_sq->q_depth - 1); + u32 offset = tail_masked * io_sq->desc_entry_size; + + /* In case this queue isn't a LLQ */ + if (io_sq->mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_HOST) + return; + + memcpy_toio(io_sq->desc_addr.pbuf_dev_addr + offset, + io_sq->desc_addr.virt_addr + offset, + io_sq->desc_entry_size); +} + +static inline void ena_com_sq_update_tail(struct ena_com_io_sq *io_sq) +{ + io_sq->tail++; + + /* Switch phase bit in case of wrap around */ + if (unlikely((io_sq->tail & (io_sq->q_depth - 1)) == 0)) + io_sq->phase ^= 1; +} + +static inline int ena_com_write_header(struct ena_com_io_sq *io_sq, + u8 *head_src, u16 header_len) +{ + u16 tail_masked = io_sq->tail & (io_sq->q_depth - 1); + u8 __iomem *dev_head_addr = + io_sq->header_addr + (tail_masked * io_sq->tx_max_header_size); + + if (io_sq->mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_HOST) + return 0; + + if (unlikely(!io_sq->header_addr)) { + pr_err("Push buffer header ptr is NULL\n"); + return -EINVAL; + } + + memcpy_toio(dev_head_addr, head_src, header_len); + + return 0; +} + +static inline struct ena_eth_io_rx_cdesc_base * + ena_com_rx_cdesc_idx_to_ptr(struct ena_com_io_cq *io_cq, u16 idx) +{ + idx &= (io_cq->q_depth - 1); + return (struct ena_eth_io_rx_cdesc_base *) + ((uintptr_t)io_cq->cdesc_addr.virt_addr + + idx * io_cq->cdesc_entry_size_in_bytes); +} + +static inline u16 ena_com_cdesc_rx_pkt_get(struct ena_com_io_cq *io_cq, + u16 *first_cdesc_idx) +{ + struct ena_eth_io_rx_cdesc_base *cdesc; + u16 count = 0, head_masked; + u32 last = 0; + + do { + cdesc = ena_com_get_next_rx_cdesc(io_cq); + if (!cdesc) + break; + + ena_com_cq_inc_head(io_cq); + count++; + last = (cdesc->status & ENA_ETH_IO_RX_CDESC_BASE_LAST_MASK) >> + ENA_ETH_IO_RX_CDESC_BASE_LAST_SHIFT; + } while (!last); + + if (last) { + *first_cdesc_idx = io_cq->cur_rx_pkt_cdesc_start_idx; + count += io_cq->cur_rx_pkt_cdesc_count; + + head_masked = io_cq->head & (io_cq->q_depth - 1); + + io_cq->cur_rx_pkt_cdesc_count = 0; + io_cq->cur_rx_pkt_cdesc_start_idx = head_masked; + + pr_debug("ena q_id: %d packets were completed. first desc idx %u descs# %d\n", + io_cq->qid, *first_cdesc_idx, count); + } else { + io_cq->cur_rx_pkt_cdesc_count += count; + count = 0; + } + + return count; +} + +static inline bool ena_com_meta_desc_changed(struct ena_com_io_sq *io_sq, + struct ena_com_tx_ctx *ena_tx_ctx) +{ + int rc; + + if (ena_tx_ctx->meta_valid) { + rc = memcmp(&io_sq->cached_tx_meta, + &ena_tx_ctx->ena_meta, + sizeof(struct ena_com_tx_meta)); + + if (unlikely(rc != 0)) + return true; + } + + return false; +} + +static inline void ena_com_create_and_store_tx_meta_desc(struct ena_com_io_sq *io_sq, + struct ena_com_tx_ctx *ena_tx_ctx) +{ + struct ena_eth_io_tx_meta_desc *meta_desc = NULL; + struct ena_com_tx_meta *ena_meta = &ena_tx_ctx->ena_meta; + + meta_desc = get_sq_desc(io_sq); + memset(meta_desc, 0x0, sizeof(struct ena_eth_io_tx_meta_desc)); + + meta_desc->len_ctrl |= ENA_ETH_IO_TX_META_DESC_META_DESC_MASK; + + meta_desc->len_ctrl |= ENA_ETH_IO_TX_META_DESC_EXT_VALID_MASK; + + /* bits 0-9 of the mss */ + meta_desc->word2 |= (ena_meta->mss << + ENA_ETH_IO_TX_META_DESC_MSS_LO_SHIFT) & + ENA_ETH_IO_TX_META_DESC_MSS_LO_MASK; + /* bits 10-13 of the mss */ + meta_desc->len_ctrl |= ((ena_meta->mss >> 10) << + ENA_ETH_IO_TX_META_DESC_MSS_HI_SHIFT) & + ENA_ETH_IO_TX_META_DESC_MSS_HI_MASK; + + /* Extended meta desc */ + meta_desc->len_ctrl |= ENA_ETH_IO_TX_META_DESC_ETH_META_TYPE_MASK; + meta_desc->len_ctrl |= ENA_ETH_IO_TX_META_DESC_META_STORE_MASK; + meta_desc->len_ctrl |= (io_sq->phase << + ENA_ETH_IO_TX_META_DESC_PHASE_SHIFT) & + ENA_ETH_IO_TX_META_DESC_PHASE_MASK; + + meta_desc->len_ctrl |= ENA_ETH_IO_TX_META_DESC_FIRST_MASK; + meta_desc->word2 |= ena_meta->l3_hdr_len & + ENA_ETH_IO_TX_META_DESC_L3_HDR_LEN_MASK; + meta_desc->word2 |= (ena_meta->l3_hdr_offset << + ENA_ETH_IO_TX_META_DESC_L3_HDR_OFF_SHIFT) & + ENA_ETH_IO_TX_META_DESC_L3_HDR_OFF_MASK; + + meta_desc->word2 |= (ena_meta->l4_hdr_len << + ENA_ETH_IO_TX_META_DESC_L4_HDR_LEN_IN_WORDS_SHIFT) & + ENA_ETH_IO_TX_META_DESC_L4_HDR_LEN_IN_WORDS_MASK; + + meta_desc->len_ctrl |= ENA_ETH_IO_TX_META_DESC_META_STORE_MASK; + + /* Cached the meta desc */ + memcpy(&io_sq->cached_tx_meta, ena_meta, + sizeof(struct ena_com_tx_meta)); + + ena_com_copy_curr_sq_desc_to_dev(io_sq); + ena_com_sq_update_tail(io_sq); +} + +static inline void ena_com_rx_set_flags(struct ena_com_rx_ctx *ena_rx_ctx, + struct ena_eth_io_rx_cdesc_base *cdesc) +{ + ena_rx_ctx->l3_proto = cdesc->status & + ENA_ETH_IO_RX_CDESC_BASE_L3_PROTO_IDX_MASK; + ena_rx_ctx->l4_proto = + (cdesc->status & ENA_ETH_IO_RX_CDESC_BASE_L4_PROTO_IDX_MASK) >> + ENA_ETH_IO_RX_CDESC_BASE_L4_PROTO_IDX_SHIFT; + ena_rx_ctx->l3_csum_err = + (cdesc->status & ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM_ERR_MASK) >> + ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM_ERR_SHIFT; + ena_rx_ctx->l4_csum_err = + (cdesc->status & ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_ERR_MASK) >> + ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_ERR_SHIFT; + ena_rx_ctx->hash = cdesc->hash; + ena_rx_ctx->frag = + (cdesc->status & ENA_ETH_IO_RX_CDESC_BASE_IPV4_FRAG_MASK) >> + ENA_ETH_IO_RX_CDESC_BASE_IPV4_FRAG_SHIFT; + + pr_debug("ena_rx_ctx->l3_proto %d ena_rx_ctx->l4_proto %d\nena_rx_ctx->l3_csum_err %d ena_rx_ctx->l4_csum_err %d\nhash frag %d frag: %d cdesc_status: %x\n", + ena_rx_ctx->l3_proto, ena_rx_ctx->l4_proto, + ena_rx_ctx->l3_csum_err, ena_rx_ctx->l4_csum_err, + ena_rx_ctx->hash, ena_rx_ctx->frag, cdesc->status); +} + +/*****************************************************************************/ +/***************************** API **********************************/ +/*****************************************************************************/ + +int ena_com_prepare_tx(struct ena_com_io_sq *io_sq, + struct ena_com_tx_ctx *ena_tx_ctx, + int *nb_hw_desc) +{ + struct ena_eth_io_tx_desc *desc = NULL; + struct ena_com_buf *ena_bufs = ena_tx_ctx->ena_bufs; + void *push_header = ena_tx_ctx->push_header; + u16 header_len = ena_tx_ctx->header_len; + u16 num_bufs = ena_tx_ctx->num_bufs; + int total_desc, i, rc; + bool have_meta; + u64 addr_hi; + + WARN(io_sq->direction != ENA_COM_IO_QUEUE_DIRECTION_TX, "wrong Q type"); + + /* num_bufs +1 for potential meta desc */ + if (ena_com_sq_empty_space(io_sq) < (num_bufs + 1)) { + pr_err("Not enough space in the tx queue\n"); + return -ENOMEM; + } + + if (unlikely(header_len > io_sq->tx_max_header_size)) { + pr_err("header size is too large %d max header: %d\n", + header_len, io_sq->tx_max_header_size); + return -EINVAL; + } + + /* start with pushing the header (if needed) */ + rc = ena_com_write_header(io_sq, push_header, header_len); + if (unlikely(rc)) + return rc; + + have_meta = ena_tx_ctx->meta_valid && ena_com_meta_desc_changed(io_sq, + ena_tx_ctx); + if (have_meta) + ena_com_create_and_store_tx_meta_desc(io_sq, ena_tx_ctx); + + /* If the caller doesn't want send packets */ + if (unlikely(!num_bufs && !header_len)) { + *nb_hw_desc = have_meta ? 0 : 1; + return 0; + } + + desc = get_sq_desc(io_sq); + memset(desc, 0x0, sizeof(struct ena_eth_io_tx_desc)); + + /* Set first desc when we don't have meta descriptor */ + if (!have_meta) + desc->len_ctrl |= ENA_ETH_IO_TX_DESC_FIRST_MASK; + + desc->buff_addr_hi_hdr_sz |= (header_len << + ENA_ETH_IO_TX_DESC_HEADER_LENGTH_SHIFT) & + ENA_ETH_IO_TX_DESC_HEADER_LENGTH_MASK; + desc->len_ctrl |= (io_sq->phase << ENA_ETH_IO_TX_DESC_PHASE_SHIFT) & + ENA_ETH_IO_TX_DESC_PHASE_MASK; + + desc->len_ctrl |= ENA_ETH_IO_TX_DESC_COMP_REQ_MASK; + + /* Bits 0-9 */ + desc->meta_ctrl |= (ena_tx_ctx->req_id << + ENA_ETH_IO_TX_DESC_REQ_ID_LO_SHIFT) & + ENA_ETH_IO_TX_DESC_REQ_ID_LO_MASK; + + desc->meta_ctrl |= (ena_tx_ctx->df << + ENA_ETH_IO_TX_DESC_DF_SHIFT) & + ENA_ETH_IO_TX_DESC_DF_MASK; + + /* Bits 10-15 */ + desc->len_ctrl |= ((ena_tx_ctx->req_id >> 10) << + ENA_ETH_IO_TX_DESC_REQ_ID_HI_SHIFT) & + ENA_ETH_IO_TX_DESC_REQ_ID_HI_MASK; + + if (ena_tx_ctx->meta_valid) { + desc->meta_ctrl |= (ena_tx_ctx->tso_enable << + ENA_ETH_IO_TX_DESC_TSO_EN_SHIFT) & + ENA_ETH_IO_TX_DESC_TSO_EN_MASK; + desc->meta_ctrl |= ena_tx_ctx->l3_proto & + ENA_ETH_IO_TX_DESC_L3_PROTO_IDX_MASK; + desc->meta_ctrl |= (ena_tx_ctx->l4_proto << + ENA_ETH_IO_TX_DESC_L4_PROTO_IDX_SHIFT) & + ENA_ETH_IO_TX_DESC_L4_PROTO_IDX_MASK; + desc->meta_ctrl |= (ena_tx_ctx->l3_csum_enable << + ENA_ETH_IO_TX_DESC_L3_CSUM_EN_SHIFT) & + ENA_ETH_IO_TX_DESC_L3_CSUM_EN_MASK; + desc->meta_ctrl |= (ena_tx_ctx->l4_csum_enable << + ENA_ETH_IO_TX_DESC_L4_CSUM_EN_SHIFT) & + ENA_ETH_IO_TX_DESC_L4_CSUM_EN_MASK; + desc->meta_ctrl |= (ena_tx_ctx->l4_csum_partial << + ENA_ETH_IO_TX_DESC_L4_CSUM_PARTIAL_SHIFT) & + ENA_ETH_IO_TX_DESC_L4_CSUM_PARTIAL_MASK; + } + + for (i = 0; i < num_bufs; i++) { + /* The first desc share the same desc as the header */ + if (likely(i != 0)) { + ena_com_copy_curr_sq_desc_to_dev(io_sq); + ena_com_sq_update_tail(io_sq); + + desc = get_sq_desc(io_sq); + memset(desc, 0x0, sizeof(struct ena_eth_io_tx_desc)); + + desc->len_ctrl |= (io_sq->phase << + ENA_ETH_IO_TX_DESC_PHASE_SHIFT) & + ENA_ETH_IO_TX_DESC_PHASE_MASK; + } + + desc->len_ctrl |= ena_bufs->len & + ENA_ETH_IO_TX_DESC_LENGTH_MASK; + + addr_hi = ((ena_bufs->paddr & + GENMASK_ULL(io_sq->dma_addr_bits - 1, 32)) >> 32); + + desc->buff_addr_lo = (u32)ena_bufs->paddr; + desc->buff_addr_hi_hdr_sz |= addr_hi & + ENA_ETH_IO_TX_DESC_ADDR_HI_MASK; + ena_bufs++; + } + + /* set the last desc indicator */ + desc->len_ctrl |= ENA_ETH_IO_TX_DESC_LAST_MASK; + + ena_com_copy_curr_sq_desc_to_dev(io_sq); + + ena_com_sq_update_tail(io_sq); + + total_desc = max_t(u16, num_bufs, 1); + total_desc += have_meta ? 1 : 0; + + *nb_hw_desc = total_desc; + return 0; +} + +int ena_com_rx_pkt(struct ena_com_io_cq *io_cq, + struct ena_com_io_sq *io_sq, + struct ena_com_rx_ctx *ena_rx_ctx) +{ + struct ena_com_rx_buf_info *ena_buf = &ena_rx_ctx->ena_bufs[0]; + struct ena_eth_io_rx_cdesc_base *cdesc = NULL; + u16 cdesc_idx = 0; + u16 nb_hw_desc; + u16 i; + + WARN(io_cq->direction != ENA_COM_IO_QUEUE_DIRECTION_RX, "wrong Q type"); + + nb_hw_desc = ena_com_cdesc_rx_pkt_get(io_cq, &cdesc_idx); + if (nb_hw_desc == 0) { + ena_rx_ctx->descs = nb_hw_desc; + return 0; + } + + pr_debug("fetch rx packet: queue %d completed desc: %d\n", io_cq->qid, + nb_hw_desc); + + if (unlikely(nb_hw_desc > ena_rx_ctx->max_bufs)) { + pr_err("Too many RX cdescs (%d) > MAX(%d)\n", nb_hw_desc, + ena_rx_ctx->max_bufs); + return -ENOSPC; + } + + for (i = 0; i < nb_hw_desc; i++) { + cdesc = ena_com_rx_cdesc_idx_to_ptr(io_cq, cdesc_idx + i); + + ena_buf->len = cdesc->length; + ena_buf->req_id = cdesc->req_id; + ena_buf++; + } + + /* Update SQ head ptr */ + io_sq->next_to_comp += nb_hw_desc; + + pr_debug("[%s][QID#%d] Updating SQ head to: %d\n", __func__, io_sq->qid, + io_sq->next_to_comp); + + /* Get rx flags from the last pkt */ + ena_com_rx_set_flags(ena_rx_ctx, cdesc); + + ena_rx_ctx->descs = nb_hw_desc; + return 0; +} + +int ena_com_add_single_rx_desc(struct ena_com_io_sq *io_sq, + struct ena_com_buf *ena_buf, + u16 req_id) +{ + struct ena_eth_io_rx_desc *desc; + + WARN(io_sq->direction != ENA_COM_IO_QUEUE_DIRECTION_RX, "wrong Q type"); + + if (unlikely(ena_com_sq_empty_space(io_sq) == 0)) + return -ENOSPC; + + desc = get_sq_desc(io_sq); + memset(desc, 0x0, sizeof(struct ena_eth_io_rx_desc)); + + desc->length = ena_buf->len; + + desc->ctrl |= ENA_ETH_IO_RX_DESC_FIRST_MASK; + desc->ctrl |= ENA_ETH_IO_RX_DESC_LAST_MASK; + desc->ctrl |= io_sq->phase & ENA_ETH_IO_RX_DESC_PHASE_MASK; + desc->ctrl |= ENA_ETH_IO_RX_DESC_COMP_REQ_MASK; + + desc->req_id = req_id; + + desc->buff_addr_lo = (u32)ena_buf->paddr; + desc->buff_addr_hi = + ((ena_buf->paddr & GENMASK_ULL(io_sq->dma_addr_bits - 1, 32)) >> 32); + + ena_com_sq_update_tail(io_sq); + + return 0; +} + +int ena_com_tx_comp_req_id_get(struct ena_com_io_cq *io_cq, u16 *req_id) +{ + u8 expected_phase, cdesc_phase; + struct ena_eth_io_tx_cdesc *cdesc; + u16 masked_head; + + masked_head = io_cq->head & (io_cq->q_depth - 1); + expected_phase = io_cq->phase; + + cdesc = (struct ena_eth_io_tx_cdesc *) + ((uintptr_t)io_cq->cdesc_addr.virt_addr + + (masked_head * io_cq->cdesc_entry_size_in_bytes)); + + /* When the current completion descriptor phase isn't the same as the + * expected, it mean that the device still didn't update + * this completion. + */ + cdesc_phase = cdesc->flags & ENA_ETH_IO_TX_CDESC_PHASE_MASK; + if (cdesc_phase != expected_phase) + return -EAGAIN; + + ena_com_cq_inc_head(io_cq); + + *req_id = cdesc->req_id; + + return 0; +} diff --git a/drivers/net/ethernet/amazon/ena/ena_eth_com.h b/drivers/net/ethernet/amazon/ena/ena_eth_com.h new file mode 100644 index 000000000000..bb53c3a4f8e9 --- /dev/null +++ b/drivers/net/ethernet/amazon/ena/ena_eth_com.h @@ -0,0 +1,160 @@ +/* + * Copyright 2015 Amazon.com, Inc. or its affiliates. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef ENA_ETH_COM_H_ +#define ENA_ETH_COM_H_ + +#include "ena_com.h" + +/* head update threshold in units of (queue size / ENA_COMP_HEAD_THRESH) */ +#define ENA_COMP_HEAD_THRESH 4 + +struct ena_com_tx_ctx { + struct ena_com_tx_meta ena_meta; + struct ena_com_buf *ena_bufs; + /* For LLQ, header buffer - pushed to the device mem space */ + void *push_header; + + enum ena_eth_io_l3_proto_index l3_proto; + enum ena_eth_io_l4_proto_index l4_proto; + u16 num_bufs; + u16 req_id; + /* For regular queue, indicate the size of the header + * For LLQ, indicate the size of the pushed buffer + */ + u16 header_len; + + u8 meta_valid; + u8 tso_enable; + u8 l3_csum_enable; + u8 l4_csum_enable; + u8 l4_csum_partial; + u8 df; /* Don't fragment */ +}; + +struct ena_com_rx_ctx { + struct ena_com_rx_buf_info *ena_bufs; + enum ena_eth_io_l3_proto_index l3_proto; + enum ena_eth_io_l4_proto_index l4_proto; + bool l3_csum_err; + bool l4_csum_err; + /* fragmented packet */ + bool frag; + u32 hash; + u16 descs; + int max_bufs; +}; + +int ena_com_prepare_tx(struct ena_com_io_sq *io_sq, + struct ena_com_tx_ctx *ena_tx_ctx, + int *nb_hw_desc); + +int ena_com_rx_pkt(struct ena_com_io_cq *io_cq, + struct ena_com_io_sq *io_sq, + struct ena_com_rx_ctx *ena_rx_ctx); + +int ena_com_add_single_rx_desc(struct ena_com_io_sq *io_sq, + struct ena_com_buf *ena_buf, + u16 req_id); + +int ena_com_tx_comp_req_id_get(struct ena_com_io_cq *io_cq, u16 *req_id); + +static inline void ena_com_unmask_intr(struct ena_com_io_cq *io_cq, + struct ena_eth_io_intr_reg *intr_reg) +{ + writel(intr_reg->intr_control, io_cq->unmask_reg); +} + +static inline int ena_com_sq_empty_space(struct ena_com_io_sq *io_sq) +{ + u16 tail, next_to_comp, cnt; + + next_to_comp = io_sq->next_to_comp; + tail = io_sq->tail; + cnt = tail - next_to_comp; + + return io_sq->q_depth - 1 - cnt; +} + +static inline int ena_com_write_sq_doorbell(struct ena_com_io_sq *io_sq) +{ + u16 tail; + + tail = io_sq->tail; + + pr_debug("write submission queue doorbell for queue: %d tail: %d\n", + io_sq->qid, tail); + + writel(tail, io_sq->db_addr); + + return 0; +} + +static inline int ena_com_update_dev_comp_head(struct ena_com_io_cq *io_cq) +{ + u16 unreported_comp, head; + bool need_update; + + head = io_cq->head; + unreported_comp = head - io_cq->last_head_update; + need_update = unreported_comp > (io_cq->q_depth / ENA_COMP_HEAD_THRESH); + + if (io_cq->cq_head_db_reg && need_update) { + pr_debug("Write completion queue doorbell for queue %d: head: %d\n", + io_cq->qid, head); + writel(head, io_cq->cq_head_db_reg); + io_cq->last_head_update = head; + } + + return 0; +} + +static inline void ena_com_update_numa_node(struct ena_com_io_cq *io_cq, + u8 numa_node) +{ + struct ena_eth_io_numa_node_cfg_reg numa_cfg; + + if (!io_cq->numa_node_cfg_reg) + return; + + numa_cfg.numa_cfg = (numa_node & ENA_ETH_IO_NUMA_NODE_CFG_REG_NUMA_MASK) + | ENA_ETH_IO_NUMA_NODE_CFG_REG_ENABLED_MASK; + + writel(numa_cfg.numa_cfg, io_cq->numa_node_cfg_reg); +} + +static inline void ena_com_comp_ack(struct ena_com_io_sq *io_sq, u16 elem) +{ + io_sq->next_to_comp += elem; +} + +#endif /* ENA_ETH_COM_H_ */ diff --git a/drivers/net/ethernet/amazon/ena/ena_eth_io_defs.h b/drivers/net/ethernet/amazon/ena/ena_eth_io_defs.h new file mode 100644 index 000000000000..f320c58793a5 --- /dev/null +++ b/drivers/net/ethernet/amazon/ena/ena_eth_io_defs.h @@ -0,0 +1,416 @@ +/* + * Copyright 2015 - 2016 Amazon.com, Inc. or its affiliates. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _ENA_ETH_IO_H_ +#define _ENA_ETH_IO_H_ + +enum ena_eth_io_l3_proto_index { + ENA_ETH_IO_L3_PROTO_UNKNOWN = 0, + + ENA_ETH_IO_L3_PROTO_IPV4 = 8, + + ENA_ETH_IO_L3_PROTO_IPV6 = 11, + + ENA_ETH_IO_L3_PROTO_FCOE = 21, + + ENA_ETH_IO_L3_PROTO_ROCE = 22, +}; + +enum ena_eth_io_l4_proto_index { + ENA_ETH_IO_L4_PROTO_UNKNOWN = 0, + + ENA_ETH_IO_L4_PROTO_TCP = 12, + + ENA_ETH_IO_L4_PROTO_UDP = 13, + + ENA_ETH_IO_L4_PROTO_ROUTEABLE_ROCE = 23, +}; + +struct ena_eth_io_tx_desc { + /* 15:0 : length - Buffer length in bytes, must + * include any packet trailers that the ENA supposed + * to update like End-to-End CRC, Authentication GMAC + * etc. This length must not include the + * 'Push_Buffer' length. This length must not include + * the 4-byte added in the end for 802.3 Ethernet FCS + * 21:16 : req_id_hi - Request ID[15:10] + * 22 : reserved22 - MBZ + * 23 : meta_desc - MBZ + * 24 : phase + * 25 : reserved1 - MBZ + * 26 : first - Indicates first descriptor in + * transaction + * 27 : last - Indicates last descriptor in + * transaction + * 28 : comp_req - Indicates whether completion + * should be posted, after packet is transmitted. + * Valid only for first descriptor + * 30:29 : reserved29 - MBZ + * 31 : reserved31 - MBZ + */ + u32 len_ctrl; + + /* 3:0 : l3_proto_idx - L3 protocol. This field + * required when l3_csum_en,l3_csum or tso_en are set. + * 4 : DF - IPv4 DF, must be 0 if packet is IPv4 and + * DF flags of the IPv4 header is 0. Otherwise must + * be set to 1 + * 6:5 : reserved5 + * 7 : tso_en - Enable TSO, For TCP only. + * 12:8 : l4_proto_idx - L4 protocol. This field need + * to be set when l4_csum_en or tso_en are set. + * 13 : l3_csum_en - enable IPv4 header checksum. + * 14 : l4_csum_en - enable TCP/UDP checksum. + * 15 : ethernet_fcs_dis - when set, the controller + * will not append the 802.3 Ethernet Frame Check + * Sequence to the packet + * 16 : reserved16 + * 17 : l4_csum_partial - L4 partial checksum. when + * set to 0, the ENA calculates the L4 checksum, + * where the Destination Address required for the + * TCP/UDP pseudo-header is taken from the actual + * packet L3 header. when set to 1, the ENA doesn't + * calculate the sum of the pseudo-header, instead, + * the checksum field of the L4 is used instead. When + * TSO enabled, the checksum of the pseudo-header + * must not include the tcp length field. L4 partial + * checksum should be used for IPv6 packet that + * contains Routing Headers. + * 20:18 : reserved18 - MBZ + * 21 : reserved21 - MBZ + * 31:22 : req_id_lo - Request ID[9:0] + */ + u32 meta_ctrl; + + u32 buff_addr_lo; + + /* address high and header size + * 15:0 : addr_hi - Buffer Pointer[47:32] + * 23:16 : reserved16_w2 + * 31:24 : header_length - Header length. For Low + * Latency Queues, this fields indicates the number + * of bytes written to the headers' memory. For + * normal queues, if packet is TCP or UDP, and longer + * than max_header_size, then this field should be + * set to the sum of L4 header offset and L4 header + * size(without options), otherwise, this field + * should be set to 0. For both modes, this field + * must not exceed the max_header_size. + * max_header_size value is reported by the Max + * Queues Feature descriptor + */ + u32 buff_addr_hi_hdr_sz; +}; + +struct ena_eth_io_tx_meta_desc { + /* 9:0 : req_id_lo - Request ID[9:0] + * 11:10 : reserved10 - MBZ + * 12 : reserved12 - MBZ + * 13 : reserved13 - MBZ + * 14 : ext_valid - if set, offset fields in Word2 + * are valid Also MSS High in Word 0 and bits [31:24] + * in Word 3 + * 15 : reserved15 + * 19:16 : mss_hi + * 20 : eth_meta_type - 0: Tx Metadata Descriptor, 1: + * Extended Metadata Descriptor + * 21 : meta_store - Store extended metadata in queue + * cache + * 22 : reserved22 - MBZ + * 23 : meta_desc - MBO + * 24 : phase + * 25 : reserved25 - MBZ + * 26 : first - Indicates first descriptor in + * transaction + * 27 : last - Indicates last descriptor in + * transaction + * 28 : comp_req - Indicates whether completion + * should be posted, after packet is transmitted. + * Valid only for first descriptor + * 30:29 : reserved29 - MBZ + * 31 : reserved31 - MBZ + */ + u32 len_ctrl; + + /* 5:0 : req_id_hi + * 31:6 : reserved6 - MBZ + */ + u32 word1; + + /* 7:0 : l3_hdr_len + * 15:8 : l3_hdr_off + * 21:16 : l4_hdr_len_in_words - counts the L4 header + * length in words. there is an explicit assumption + * that L4 header appears right after L3 header and + * L4 offset is based on l3_hdr_off+l3_hdr_len + * 31:22 : mss_lo + */ + u32 word2; + + u32 reserved; +}; + +struct ena_eth_io_tx_cdesc { + /* Request ID[15:0] */ + u16 req_id; + + u8 status; + + /* flags + * 0 : phase + * 7:1 : reserved1 + */ + u8 flags; + + u16 sub_qid; + + u16 sq_head_idx; +}; + +struct ena_eth_io_rx_desc { + /* In bytes. 0 means 64KB */ + u16 length; + + /* MBZ */ + u8 reserved2; + + /* 0 : phase + * 1 : reserved1 - MBZ + * 2 : first - Indicates first descriptor in + * transaction + * 3 : last - Indicates last descriptor in transaction + * 4 : comp_req + * 5 : reserved5 - MBO + * 7:6 : reserved6 - MBZ + */ + u8 ctrl; + + u16 req_id; + + /* MBZ */ + u16 reserved6; + + u32 buff_addr_lo; + + u16 buff_addr_hi; + + /* MBZ */ + u16 reserved16_w3; +}; + +/* 4-word format Note: all ethernet parsing information are valid only when + * last=1 + */ +struct ena_eth_io_rx_cdesc_base { + /* 4:0 : l3_proto_idx + * 6:5 : src_vlan_cnt + * 7 : reserved7 - MBZ + * 12:8 : l4_proto_idx + * 13 : l3_csum_err - when set, either the L3 + * checksum error detected, or, the controller didn't + * validate the checksum. This bit is valid only when + * l3_proto_idx indicates IPv4 packet + * 14 : l4_csum_err - when set, either the L4 + * checksum error detected, or, the controller didn't + * validate the checksum. This bit is valid only when + * l4_proto_idx indicates TCP/UDP packet, and, + * ipv4_frag is not set + * 15 : ipv4_frag - Indicates IPv4 fragmented packet + * 23:16 : reserved16 + * 24 : phase + * 25 : l3_csum2 - second checksum engine result + * 26 : first - Indicates first descriptor in + * transaction + * 27 : last - Indicates last descriptor in + * transaction + * 29:28 : reserved28 + * 30 : buffer - 0: Metadata descriptor. 1: Buffer + * Descriptor was used + * 31 : reserved31 + */ + u32 status; + + u16 length; + + u16 req_id; + + /* 32-bit hash result */ + u32 hash; + + u16 sub_qid; + + u16 reserved; +}; + +/* 8-word format */ +struct ena_eth_io_rx_cdesc_ext { + struct ena_eth_io_rx_cdesc_base base; + + u32 buff_addr_lo; + + u16 buff_addr_hi; + + u16 reserved16; + + u32 reserved_w6; + + u32 reserved_w7; +}; + +struct ena_eth_io_intr_reg { + /* 14:0 : rx_intr_delay + * 29:15 : tx_intr_delay + * 30 : intr_unmask + * 31 : reserved + */ + u32 intr_control; +}; + +struct ena_eth_io_numa_node_cfg_reg { + /* 7:0 : numa + * 30:8 : reserved + * 31 : enabled + */ + u32 numa_cfg; +}; + +/* tx_desc */ +#define ENA_ETH_IO_TX_DESC_LENGTH_MASK GENMASK(15, 0) +#define ENA_ETH_IO_TX_DESC_REQ_ID_HI_SHIFT 16 +#define ENA_ETH_IO_TX_DESC_REQ_ID_HI_MASK GENMASK(21, 16) +#define ENA_ETH_IO_TX_DESC_META_DESC_SHIFT 23 +#define ENA_ETH_IO_TX_DESC_META_DESC_MASK BIT(23) +#define ENA_ETH_IO_TX_DESC_PHASE_SHIFT 24 +#define ENA_ETH_IO_TX_DESC_PHASE_MASK BIT(24) +#define ENA_ETH_IO_TX_DESC_FIRST_SHIFT 26 +#define ENA_ETH_IO_TX_DESC_FIRST_MASK BIT(26) +#define ENA_ETH_IO_TX_DESC_LAST_SHIFT 27 +#define ENA_ETH_IO_TX_DESC_LAST_MASK BIT(27) +#define ENA_ETH_IO_TX_DESC_COMP_REQ_SHIFT 28 +#define ENA_ETH_IO_TX_DESC_COMP_REQ_MASK BIT(28) +#define ENA_ETH_IO_TX_DESC_L3_PROTO_IDX_MASK GENMASK(3, 0) +#define ENA_ETH_IO_TX_DESC_DF_SHIFT 4 +#define ENA_ETH_IO_TX_DESC_DF_MASK BIT(4) +#define ENA_ETH_IO_TX_DESC_TSO_EN_SHIFT 7 +#define ENA_ETH_IO_TX_DESC_TSO_EN_MASK BIT(7) +#define ENA_ETH_IO_TX_DESC_L4_PROTO_IDX_SHIFT 8 +#define ENA_ETH_IO_TX_DESC_L4_PROTO_IDX_MASK GENMASK(12, 8) +#define ENA_ETH_IO_TX_DESC_L3_CSUM_EN_SHIFT 13 +#define ENA_ETH_IO_TX_DESC_L3_CSUM_EN_MASK BIT(13) +#define ENA_ETH_IO_TX_DESC_L4_CSUM_EN_SHIFT 14 +#define ENA_ETH_IO_TX_DESC_L4_CSUM_EN_MASK BIT(14) +#define ENA_ETH_IO_TX_DESC_ETHERNET_FCS_DIS_SHIFT 15 +#define ENA_ETH_IO_TX_DESC_ETHERNET_FCS_DIS_MASK BIT(15) +#define ENA_ETH_IO_TX_DESC_L4_CSUM_PARTIAL_SHIFT 17 +#define ENA_ETH_IO_TX_DESC_L4_CSUM_PARTIAL_MASK BIT(17) +#define ENA_ETH_IO_TX_DESC_REQ_ID_LO_SHIFT 22 +#define ENA_ETH_IO_TX_DESC_REQ_ID_LO_MASK GENMASK(31, 22) +#define ENA_ETH_IO_TX_DESC_ADDR_HI_MASK GENMASK(15, 0) +#define ENA_ETH_IO_TX_DESC_HEADER_LENGTH_SHIFT 24 +#define ENA_ETH_IO_TX_DESC_HEADER_LENGTH_MASK GENMASK(31, 24) + +/* tx_meta_desc */ +#define ENA_ETH_IO_TX_META_DESC_REQ_ID_LO_MASK GENMASK(9, 0) +#define ENA_ETH_IO_TX_META_DESC_EXT_VALID_SHIFT 14 +#define ENA_ETH_IO_TX_META_DESC_EXT_VALID_MASK BIT(14) +#define ENA_ETH_IO_TX_META_DESC_MSS_HI_SHIFT 16 +#define ENA_ETH_IO_TX_META_DESC_MSS_HI_MASK GENMASK(19, 16) +#define ENA_ETH_IO_TX_META_DESC_ETH_META_TYPE_SHIFT 20 +#define ENA_ETH_IO_TX_META_DESC_ETH_META_TYPE_MASK BIT(20) +#define ENA_ETH_IO_TX_META_DESC_META_STORE_SHIFT 21 +#define ENA_ETH_IO_TX_META_DESC_META_STORE_MASK BIT(21) +#define ENA_ETH_IO_TX_META_DESC_META_DESC_SHIFT 23 +#define ENA_ETH_IO_TX_META_DESC_META_DESC_MASK BIT(23) +#define ENA_ETH_IO_TX_META_DESC_PHASE_SHIFT 24 +#define ENA_ETH_IO_TX_META_DESC_PHASE_MASK BIT(24) +#define ENA_ETH_IO_TX_META_DESC_FIRST_SHIFT 26 +#define ENA_ETH_IO_TX_META_DESC_FIRST_MASK BIT(26) +#define ENA_ETH_IO_TX_META_DESC_LAST_SHIFT 27 +#define ENA_ETH_IO_TX_META_DESC_LAST_MASK BIT(27) +#define ENA_ETH_IO_TX_META_DESC_COMP_REQ_SHIFT 28 +#define ENA_ETH_IO_TX_META_DESC_COMP_REQ_MASK BIT(28) +#define ENA_ETH_IO_TX_META_DESC_REQ_ID_HI_MASK GENMASK(5, 0) +#define ENA_ETH_IO_TX_META_DESC_L3_HDR_LEN_MASK GENMASK(7, 0) +#define ENA_ETH_IO_TX_META_DESC_L3_HDR_OFF_SHIFT 8 +#define ENA_ETH_IO_TX_META_DESC_L3_HDR_OFF_MASK GENMASK(15, 8) +#define ENA_ETH_IO_TX_META_DESC_L4_HDR_LEN_IN_WORDS_SHIFT 16 +#define ENA_ETH_IO_TX_META_DESC_L4_HDR_LEN_IN_WORDS_MASK GENMASK(21, 16) +#define ENA_ETH_IO_TX_META_DESC_MSS_LO_SHIFT 22 +#define ENA_ETH_IO_TX_META_DESC_MSS_LO_MASK GENMASK(31, 22) + +/* tx_cdesc */ +#define ENA_ETH_IO_TX_CDESC_PHASE_MASK BIT(0) + +/* rx_desc */ +#define ENA_ETH_IO_RX_DESC_PHASE_MASK BIT(0) +#define ENA_ETH_IO_RX_DESC_FIRST_SHIFT 2 +#define ENA_ETH_IO_RX_DESC_FIRST_MASK BIT(2) +#define ENA_ETH_IO_RX_DESC_LAST_SHIFT 3 +#define ENA_ETH_IO_RX_DESC_LAST_MASK BIT(3) +#define ENA_ETH_IO_RX_DESC_COMP_REQ_SHIFT 4 +#define ENA_ETH_IO_RX_DESC_COMP_REQ_MASK BIT(4) + +/* rx_cdesc_base */ +#define ENA_ETH_IO_RX_CDESC_BASE_L3_PROTO_IDX_MASK GENMASK(4, 0) +#define ENA_ETH_IO_RX_CDESC_BASE_SRC_VLAN_CNT_SHIFT 5 +#define ENA_ETH_IO_RX_CDESC_BASE_SRC_VLAN_CNT_MASK GENMASK(6, 5) +#define ENA_ETH_IO_RX_CDESC_BASE_L4_PROTO_IDX_SHIFT 8 +#define ENA_ETH_IO_RX_CDESC_BASE_L4_PROTO_IDX_MASK GENMASK(12, 8) +#define ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM_ERR_SHIFT 13 +#define ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM_ERR_MASK BIT(13) +#define ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_ERR_SHIFT 14 +#define ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_ERR_MASK BIT(14) +#define ENA_ETH_IO_RX_CDESC_BASE_IPV4_FRAG_SHIFT 15 +#define ENA_ETH_IO_RX_CDESC_BASE_IPV4_FRAG_MASK BIT(15) +#define ENA_ETH_IO_RX_CDESC_BASE_PHASE_SHIFT 24 +#define ENA_ETH_IO_RX_CDESC_BASE_PHASE_MASK BIT(24) +#define ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM2_SHIFT 25 +#define ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM2_MASK BIT(25) +#define ENA_ETH_IO_RX_CDESC_BASE_FIRST_SHIFT 26 +#define ENA_ETH_IO_RX_CDESC_BASE_FIRST_MASK BIT(26) +#define ENA_ETH_IO_RX_CDESC_BASE_LAST_SHIFT 27 +#define ENA_ETH_IO_RX_CDESC_BASE_LAST_MASK BIT(27) +#define ENA_ETH_IO_RX_CDESC_BASE_BUFFER_SHIFT 30 +#define ENA_ETH_IO_RX_CDESC_BASE_BUFFER_MASK BIT(30) + +/* intr_reg */ +#define ENA_ETH_IO_INTR_REG_RX_INTR_DELAY_MASK GENMASK(14, 0) +#define ENA_ETH_IO_INTR_REG_TX_INTR_DELAY_SHIFT 15 +#define ENA_ETH_IO_INTR_REG_TX_INTR_DELAY_MASK GENMASK(29, 15) +#define ENA_ETH_IO_INTR_REG_INTR_UNMASK_SHIFT 30 +#define ENA_ETH_IO_INTR_REG_INTR_UNMASK_MASK BIT(30) + +/* numa_node_cfg_reg */ +#define ENA_ETH_IO_NUMA_NODE_CFG_REG_NUMA_MASK GENMASK(7, 0) +#define ENA_ETH_IO_NUMA_NODE_CFG_REG_ENABLED_SHIFT 31 +#define ENA_ETH_IO_NUMA_NODE_CFG_REG_ENABLED_MASK BIT(31) + +#endif /*_ENA_ETH_IO_H_ */ diff --git a/drivers/net/ethernet/amazon/ena/ena_ethtool.c b/drivers/net/ethernet/amazon/ena/ena_ethtool.c new file mode 100644 index 000000000000..67b2338f8fb3 --- /dev/null +++ b/drivers/net/ethernet/amazon/ena/ena_ethtool.c @@ -0,0 +1,895 @@ +/* + * Copyright 2015 Amazon.com, Inc. or its affiliates. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include <linux/pci.h> + +#include "ena_netdev.h" + +struct ena_stats { + char name[ETH_GSTRING_LEN]; + int stat_offset; +}; + +#define ENA_STAT_ENA_COM_ENTRY(stat) { \ + .name = #stat, \ + .stat_offset = offsetof(struct ena_com_stats_admin, stat) \ +} + +#define ENA_STAT_ENTRY(stat, stat_type) { \ + .name = #stat, \ + .stat_offset = offsetof(struct ena_stats_##stat_type, stat) \ +} + +#define ENA_STAT_RX_ENTRY(stat) \ + ENA_STAT_ENTRY(stat, rx) + +#define ENA_STAT_TX_ENTRY(stat) \ + ENA_STAT_ENTRY(stat, tx) + +#define ENA_STAT_GLOBAL_ENTRY(stat) \ + ENA_STAT_ENTRY(stat, dev) + +static const struct ena_stats ena_stats_global_strings[] = { + ENA_STAT_GLOBAL_ENTRY(tx_timeout), + ENA_STAT_GLOBAL_ENTRY(io_suspend), + ENA_STAT_GLOBAL_ENTRY(io_resume), + ENA_STAT_GLOBAL_ENTRY(wd_expired), + ENA_STAT_GLOBAL_ENTRY(interface_up), + ENA_STAT_GLOBAL_ENTRY(interface_down), + ENA_STAT_GLOBAL_ENTRY(admin_q_pause), +}; + +static const struct ena_stats ena_stats_tx_strings[] = { + ENA_STAT_TX_ENTRY(cnt), + ENA_STAT_TX_ENTRY(bytes), + ENA_STAT_TX_ENTRY(queue_stop), + ENA_STAT_TX_ENTRY(queue_wakeup), + ENA_STAT_TX_ENTRY(dma_mapping_err), + ENA_STAT_TX_ENTRY(linearize), + ENA_STAT_TX_ENTRY(linearize_failed), + ENA_STAT_TX_ENTRY(napi_comp), + ENA_STAT_TX_ENTRY(tx_poll), + ENA_STAT_TX_ENTRY(doorbells), + ENA_STAT_TX_ENTRY(prepare_ctx_err), + ENA_STAT_TX_ENTRY(missing_tx_comp), + ENA_STAT_TX_ENTRY(bad_req_id), +}; + +static const struct ena_stats ena_stats_rx_strings[] = { + ENA_STAT_RX_ENTRY(cnt), + ENA_STAT_RX_ENTRY(bytes), + ENA_STAT_RX_ENTRY(refil_partial), + ENA_STAT_RX_ENTRY(bad_csum), + ENA_STAT_RX_ENTRY(page_alloc_fail), + ENA_STAT_RX_ENTRY(skb_alloc_fail), + ENA_STAT_RX_ENTRY(dma_mapping_err), + ENA_STAT_RX_ENTRY(bad_desc_num), + ENA_STAT_RX_ENTRY(rx_copybreak_pkt), +}; + +static const struct ena_stats ena_stats_ena_com_strings[] = { + ENA_STAT_ENA_COM_ENTRY(aborted_cmd), + ENA_STAT_ENA_COM_ENTRY(submitted_cmd), + ENA_STAT_ENA_COM_ENTRY(completed_cmd), + ENA_STAT_ENA_COM_ENTRY(out_of_space), + ENA_STAT_ENA_COM_ENTRY(no_completion), +}; + +#define ENA_STATS_ARRAY_GLOBAL ARRAY_SIZE(ena_stats_global_strings) +#define ENA_STATS_ARRAY_TX ARRAY_SIZE(ena_stats_tx_strings) +#define ENA_STATS_ARRAY_RX ARRAY_SIZE(ena_stats_rx_strings) +#define ENA_STATS_ARRAY_ENA_COM ARRAY_SIZE(ena_stats_ena_com_strings) + +static void ena_safe_update_stat(u64 *src, u64 *dst, + struct u64_stats_sync *syncp) +{ + unsigned int start; + + do { + start = u64_stats_fetch_begin_irq(syncp); + *(dst) = *src; + } while (u64_stats_fetch_retry_irq(syncp, start)); +} + +static void ena_queue_stats(struct ena_adapter *adapter, u64 **data) +{ + const struct ena_stats *ena_stats; + struct ena_ring *ring; + + u64 *ptr; + int i, j; + + for (i = 0; i < adapter->num_queues; i++) { + /* Tx stats */ + ring = &adapter->tx_ring[i]; + + for (j = 0; j < ENA_STATS_ARRAY_TX; j++) { + ena_stats = &ena_stats_tx_strings[j]; + + ptr = (u64 *)((uintptr_t)&ring->tx_stats + + (uintptr_t)ena_stats->stat_offset); + + ena_safe_update_stat(ptr, (*data)++, &ring->syncp); + } + + /* Rx stats */ + ring = &adapter->rx_ring[i]; + + for (j = 0; j < ENA_STATS_ARRAY_RX; j++) { + ena_stats = &ena_stats_rx_strings[j]; + + ptr = (u64 *)((uintptr_t)&ring->rx_stats + + (uintptr_t)ena_stats->stat_offset); + + ena_safe_update_stat(ptr, (*data)++, &ring->syncp); + } + } +} + +static void ena_dev_admin_queue_stats(struct ena_adapter *adapter, u64 **data) +{ + const struct ena_stats *ena_stats; + u32 *ptr; + int i; + + for (i = 0; i < ENA_STATS_ARRAY_ENA_COM; i++) { + ena_stats = &ena_stats_ena_com_strings[i]; + + ptr = (u32 *)((uintptr_t)&adapter->ena_dev->admin_queue.stats + + (uintptr_t)ena_stats->stat_offset); + + *(*data)++ = *ptr; + } +} + +static void ena_get_ethtool_stats(struct net_device *netdev, + struct ethtool_stats *stats, + u64 *data) +{ + struct ena_adapter *adapter = netdev_priv(netdev); + const struct ena_stats *ena_stats; + u64 *ptr; + int i; + + for (i = 0; i < ENA_STATS_ARRAY_GLOBAL; i++) { + ena_stats = &ena_stats_global_strings[i]; + + ptr = (u64 *)((uintptr_t)&adapter->dev_stats + + (uintptr_t)ena_stats->stat_offset); + + ena_safe_update_stat(ptr, data++, &adapter->syncp); + } + + ena_queue_stats(adapter, &data); + ena_dev_admin_queue_stats(adapter, &data); +} + +int ena_get_sset_count(struct net_device *netdev, int sset) +{ + struct ena_adapter *adapter = netdev_priv(netdev); + + if (sset != ETH_SS_STATS) + return -EOPNOTSUPP; + + return adapter->num_queues * (ENA_STATS_ARRAY_TX + ENA_STATS_ARRAY_RX) + + ENA_STATS_ARRAY_GLOBAL + ENA_STATS_ARRAY_ENA_COM; +} + +static void ena_queue_strings(struct ena_adapter *adapter, u8 **data) +{ + const struct ena_stats *ena_stats; + int i, j; + + for (i = 0; i < adapter->num_queues; i++) { + /* Tx stats */ + for (j = 0; j < ENA_STATS_ARRAY_TX; j++) { + ena_stats = &ena_stats_tx_strings[j]; + + snprintf(*data, ETH_GSTRING_LEN, + "queue_%u_tx_%s", i, ena_stats->name); + (*data) += ETH_GSTRING_LEN; + } + /* Rx stats */ + for (j = 0; j < ENA_STATS_ARRAY_RX; j++) { + ena_stats = &ena_stats_rx_strings[j]; + + snprintf(*data, ETH_GSTRING_LEN, + "queue_%u_rx_%s", i, ena_stats->name); + (*data) += ETH_GSTRING_LEN; + } + } +} + +static void ena_com_dev_strings(u8 **data) +{ + const struct ena_stats *ena_stats; + int i; + + for (i = 0; i < ENA_STATS_ARRAY_ENA_COM; i++) { + ena_stats = &ena_stats_ena_com_strings[i]; + + snprintf(*data, ETH_GSTRING_LEN, + "ena_admin_q_%s", ena_stats->name); + (*data) += ETH_GSTRING_LEN; + } +} + +static void ena_get_strings(struct net_device *netdev, u32 sset, u8 *data) +{ + struct ena_adapter *adapter = netdev_priv(netdev); + const struct ena_stats *ena_stats; + int i; + + if (sset != ETH_SS_STATS) + return; + + for (i = 0; i < ENA_STATS_ARRAY_GLOBAL; i++) { + ena_stats = &ena_stats_global_strings[i]; + + memcpy(data, ena_stats->name, ETH_GSTRING_LEN); + data += ETH_GSTRING_LEN; + } + + ena_queue_strings(adapter, &data); + ena_com_dev_strings(&data); +} + +static int ena_get_link_ksettings(struct net_device *netdev, + struct ethtool_link_ksettings *link_ksettings) +{ + struct ena_adapter *adapter = netdev_priv(netdev); + struct ena_com_dev *ena_dev = adapter->ena_dev; + struct ena_admin_get_feature_link_desc *link; + struct ena_admin_get_feat_resp feat_resp; + int rc; + + rc = ena_com_get_link_params(ena_dev, &feat_resp); + if (rc) + return rc; + + link = &feat_resp.u.link; + link_ksettings->base.speed = link->speed; + + if (link->flags & ENA_ADMIN_GET_FEATURE_LINK_DESC_AUTONEG_MASK) { + ethtool_link_ksettings_add_link_mode(link_ksettings, + supported, Autoneg); + ethtool_link_ksettings_add_link_mode(link_ksettings, + supported, Autoneg); + } + + link_ksettings->base.autoneg = + (link->flags & ENA_ADMIN_GET_FEATURE_LINK_DESC_AUTONEG_MASK) ? + AUTONEG_ENABLE : AUTONEG_DISABLE; + + link_ksettings->base.duplex = DUPLEX_FULL; + + return 0; +} + +static int ena_get_coalesce(struct net_device *net_dev, + struct ethtool_coalesce *coalesce) +{ + struct ena_adapter *adapter = netdev_priv(net_dev); + struct ena_com_dev *ena_dev = adapter->ena_dev; + struct ena_intr_moder_entry intr_moder_entry; + + if (!ena_com_interrupt_moderation_supported(ena_dev)) { + /* the devie doesn't support interrupt moderation */ + return -EOPNOTSUPP; + } + coalesce->tx_coalesce_usecs = + ena_com_get_nonadaptive_moderation_interval_tx(ena_dev) / + ena_dev->intr_delay_resolution; + if (!ena_com_get_adaptive_moderation_enabled(ena_dev)) { + coalesce->rx_coalesce_usecs = + ena_com_get_nonadaptive_moderation_interval_rx(ena_dev) + / ena_dev->intr_delay_resolution; + } else { + ena_com_get_intr_moderation_entry(adapter->ena_dev, ENA_INTR_MODER_LOWEST, &intr_moder_entry); + coalesce->rx_coalesce_usecs_low = intr_moder_entry.intr_moder_interval; + coalesce->rx_max_coalesced_frames_low = intr_moder_entry.pkts_per_interval; + + ena_com_get_intr_moderation_entry(adapter->ena_dev, ENA_INTR_MODER_MID, &intr_moder_entry); + coalesce->rx_coalesce_usecs = intr_moder_entry.intr_moder_interval; + coalesce->rx_max_coalesced_frames = intr_moder_entry.pkts_per_interval; + + ena_com_get_intr_moderation_entry(adapter->ena_dev, ENA_INTR_MODER_HIGHEST, &intr_moder_entry); + coalesce->rx_coalesce_usecs_high = intr_moder_entry.intr_moder_interval; + coalesce->rx_max_coalesced_frames_high = intr_moder_entry.pkts_per_interval; + } + coalesce->use_adaptive_rx_coalesce = + ena_com_get_adaptive_moderation_enabled(ena_dev); + + return 0; +} + +static void ena_update_tx_rings_intr_moderation(struct ena_adapter *adapter) +{ + unsigned int val; + int i; + + val = ena_com_get_nonadaptive_moderation_interval_tx(adapter->ena_dev); + + for (i = 0; i < adapter->num_queues; i++) + adapter->tx_ring[i].smoothed_interval = val; +} + +static int ena_set_coalesce(struct net_device *net_dev, + struct ethtool_coalesce *coalesce) +{ + struct ena_adapter *adapter = netdev_priv(net_dev); + struct ena_com_dev *ena_dev = adapter->ena_dev; + struct ena_intr_moder_entry intr_moder_entry; + int rc; + + if (!ena_com_interrupt_moderation_supported(ena_dev)) { + /* the devie doesn't support interrupt moderation */ + return -EOPNOTSUPP; + } + + if (coalesce->rx_coalesce_usecs_irq || + coalesce->rx_max_coalesced_frames_irq || + coalesce->tx_coalesce_usecs_irq || + coalesce->tx_max_coalesced_frames || + coalesce->tx_max_coalesced_frames_irq || + coalesce->stats_block_coalesce_usecs || + coalesce->use_adaptive_tx_coalesce || + coalesce->pkt_rate_low || + coalesce->tx_coalesce_usecs_low || + coalesce->tx_max_coalesced_frames_low || + coalesce->pkt_rate_high || + coalesce->tx_coalesce_usecs_high || + coalesce->tx_max_coalesced_frames_high || + coalesce->rate_sample_interval) + return -EINVAL; + + rc = ena_com_update_nonadaptive_moderation_interval_tx(ena_dev, + coalesce->tx_coalesce_usecs); + if (rc) + return rc; + + ena_update_tx_rings_intr_moderation(adapter); + + if (ena_com_get_adaptive_moderation_enabled(ena_dev)) { + if (!coalesce->use_adaptive_rx_coalesce) { + ena_com_disable_adaptive_moderation(ena_dev); + rc = ena_com_update_nonadaptive_moderation_interval_rx(ena_dev, + coalesce->rx_coalesce_usecs); + return rc; + } + } else { /* was in non-adaptive mode */ + if (coalesce->use_adaptive_rx_coalesce) { + ena_com_enable_adaptive_moderation(ena_dev); + } else { + rc = ena_com_update_nonadaptive_moderation_interval_rx(ena_dev, + coalesce->rx_coalesce_usecs); + return rc; + } + } + + intr_moder_entry.intr_moder_interval = coalesce->rx_coalesce_usecs_low; + intr_moder_entry.pkts_per_interval = coalesce->rx_max_coalesced_frames_low; + intr_moder_entry.bytes_per_interval = ENA_INTR_BYTE_COUNT_NOT_SUPPORTED; + ena_com_init_intr_moderation_entry(adapter->ena_dev, ENA_INTR_MODER_LOWEST, &intr_moder_entry); + + intr_moder_entry.intr_moder_interval = coalesce->rx_coalesce_usecs; + intr_moder_entry.pkts_per_interval = coalesce->rx_max_coalesced_frames; + intr_moder_entry.bytes_per_interval = ENA_INTR_BYTE_COUNT_NOT_SUPPORTED; + ena_com_init_intr_moderation_entry(adapter->ena_dev, ENA_INTR_MODER_MID, &intr_moder_entry); + + intr_moder_entry.intr_moder_interval = coalesce->rx_coalesce_usecs_high; + intr_moder_entry.pkts_per_interval = coalesce->rx_max_coalesced_frames_high; + intr_moder_entry.bytes_per_interval = ENA_INTR_BYTE_COUNT_NOT_SUPPORTED; + ena_com_init_intr_moderation_entry(adapter->ena_dev, ENA_INTR_MODER_HIGHEST, &intr_moder_entry); + + return 0; +} + +static u32 ena_get_msglevel(struct net_device *netdev) +{ + struct ena_adapter *adapter = netdev_priv(netdev); + + return adapter->msg_enable; +} + +static void ena_set_msglevel(struct net_device *netdev, u32 value) +{ + struct ena_adapter *adapter = netdev_priv(netdev); + + adapter->msg_enable = value; +} + +static void ena_get_drvinfo(struct net_device *dev, + struct ethtool_drvinfo *info) +{ + struct ena_adapter *adapter = netdev_priv(dev); + + strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(adapter->pdev), + sizeof(info->bus_info)); +} + +static void ena_get_ringparam(struct net_device *netdev, + struct ethtool_ringparam *ring) +{ + struct ena_adapter *adapter = netdev_priv(netdev); + struct ena_ring *tx_ring = &adapter->tx_ring[0]; + struct ena_ring *rx_ring = &adapter->rx_ring[0]; + + ring->rx_max_pending = rx_ring->ring_size; + ring->tx_max_pending = tx_ring->ring_size; + ring->rx_pending = rx_ring->ring_size; + ring->tx_pending = tx_ring->ring_size; +} + +static u32 ena_flow_hash_to_flow_type(u16 hash_fields) +{ + u32 data = 0; + + if (hash_fields & ENA_ADMIN_RSS_L2_DA) + data |= RXH_L2DA; + + if (hash_fields & ENA_ADMIN_RSS_L3_DA) + data |= RXH_IP_DST; + + if (hash_fields & ENA_ADMIN_RSS_L3_SA) + data |= RXH_IP_SRC; + + if (hash_fields & ENA_ADMIN_RSS_L4_DP) + data |= RXH_L4_B_2_3; + + if (hash_fields & ENA_ADMIN_RSS_L4_SP) + data |= RXH_L4_B_0_1; + + return data; +} + +static u16 ena_flow_data_to_flow_hash(u32 hash_fields) +{ + u16 data = 0; + + if (hash_fields & RXH_L2DA) + data |= ENA_ADMIN_RSS_L2_DA; + + if (hash_fields & RXH_IP_DST) + data |= ENA_ADMIN_RSS_L3_DA; + + if (hash_fields & RXH_IP_SRC) + data |= ENA_ADMIN_RSS_L3_SA; + + if (hash_fields & RXH_L4_B_2_3) + data |= ENA_ADMIN_RSS_L4_DP; + + if (hash_fields & RXH_L4_B_0_1) + data |= ENA_ADMIN_RSS_L4_SP; + + return data; +} + +static int ena_get_rss_hash(struct ena_com_dev *ena_dev, + struct ethtool_rxnfc *cmd) +{ + enum ena_admin_flow_hash_proto proto; + u16 hash_fields; + int rc; + + cmd->data = 0; + + switch (cmd->flow_type) { + case TCP_V4_FLOW: + proto = ENA_ADMIN_RSS_TCP4; + break; + case UDP_V4_FLOW: + proto = ENA_ADMIN_RSS_UDP4; + break; + case TCP_V6_FLOW: + proto = ENA_ADMIN_RSS_TCP6; + break; + case UDP_V6_FLOW: + proto = ENA_ADMIN_RSS_UDP6; + break; + case IPV4_FLOW: + proto = ENA_ADMIN_RSS_IP4; + break; + case IPV6_FLOW: + proto = ENA_ADMIN_RSS_IP6; + break; + case ETHER_FLOW: + proto = ENA_ADMIN_RSS_NOT_IP; + break; + case AH_V4_FLOW: + case ESP_V4_FLOW: + case AH_V6_FLOW: + case ESP_V6_FLOW: + case SCTP_V4_FLOW: + case AH_ESP_V4_FLOW: + return -EOPNOTSUPP; + default: + return -EINVAL; + } + + rc = ena_com_get_hash_ctrl(ena_dev, proto, &hash_fields); + if (rc) { + /* If device don't have permission, return unsupported */ + if (rc == -EPERM) + rc = -EOPNOTSUPP; + return rc; + } + + cmd->data = ena_flow_hash_to_flow_type(hash_fields); + + return 0; +} + +static int ena_set_rss_hash(struct ena_com_dev *ena_dev, + struct ethtool_rxnfc *cmd) +{ + enum ena_admin_flow_hash_proto proto; + u16 hash_fields; + + switch (cmd->flow_type) { + case TCP_V4_FLOW: + proto = ENA_ADMIN_RSS_TCP4; + break; + case UDP_V4_FLOW: + proto = ENA_ADMIN_RSS_UDP4; + break; + case TCP_V6_FLOW: + proto = ENA_ADMIN_RSS_TCP6; + break; + case UDP_V6_FLOW: + proto = ENA_ADMIN_RSS_UDP6; + break; + case IPV4_FLOW: + proto = ENA_ADMIN_RSS_IP4; + break; + case IPV6_FLOW: + proto = ENA_ADMIN_RSS_IP6; + break; + case ETHER_FLOW: + proto = ENA_ADMIN_RSS_NOT_IP; + break; + case AH_V4_FLOW: + case ESP_V4_FLOW: + case AH_V6_FLOW: + case ESP_V6_FLOW: + case SCTP_V4_FLOW: + case AH_ESP_V4_FLOW: + return -EOPNOTSUPP; + default: + return -EINVAL; + } + + hash_fields = ena_flow_data_to_flow_hash(cmd->data); + + return ena_com_fill_hash_ctrl(ena_dev, proto, hash_fields); +} + +static int ena_set_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *info) +{ + struct ena_adapter *adapter = netdev_priv(netdev); + int rc = 0; + + switch (info->cmd) { + case ETHTOOL_SRXFH: + rc = ena_set_rss_hash(adapter->ena_dev, info); + break; + case ETHTOOL_SRXCLSRLDEL: + case ETHTOOL_SRXCLSRLINS: + default: + netif_err(adapter, drv, netdev, + "Command parameter %d is not supported\n", info->cmd); + rc = -EOPNOTSUPP; + } + + return (rc == -EPERM) ? -EOPNOTSUPP : rc; +} + +static int ena_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *info, + u32 *rules) +{ + struct ena_adapter *adapter = netdev_priv(netdev); + int rc = 0; + + switch (info->cmd) { + case ETHTOOL_GRXRINGS: + info->data = adapter->num_queues; + rc = 0; + break; + case ETHTOOL_GRXFH: + rc = ena_get_rss_hash(adapter->ena_dev, info); + break; + case ETHTOOL_GRXCLSRLCNT: + case ETHTOOL_GRXCLSRULE: + case ETHTOOL_GRXCLSRLALL: + default: + netif_err(adapter, drv, netdev, + "Command parameter %d is not supported\n", info->cmd); + rc = -EOPNOTSUPP; + } + + return (rc == -EPERM) ? -EOPNOTSUPP : rc; +} + +static u32 ena_get_rxfh_indir_size(struct net_device *netdev) +{ + return ENA_RX_RSS_TABLE_SIZE; +} + +static u32 ena_get_rxfh_key_size(struct net_device *netdev) +{ + return ENA_HASH_KEY_SIZE; +} + +static int ena_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key, + u8 *hfunc) +{ + struct ena_adapter *adapter = netdev_priv(netdev); + enum ena_admin_hash_functions ena_func; + u8 func; + int rc; + + rc = ena_com_indirect_table_get(adapter->ena_dev, indir); + if (rc) + return rc; + + rc = ena_com_get_hash_function(adapter->ena_dev, &ena_func, key); + if (rc) + return rc; + + switch (ena_func) { + case ENA_ADMIN_TOEPLITZ: + func = ETH_RSS_HASH_TOP; + break; + case ENA_ADMIN_CRC32: + func = ETH_RSS_HASH_XOR; + break; + default: + netif_err(adapter, drv, netdev, + "Command parameter is not supported\n"); + return -EOPNOTSUPP; + } + + if (hfunc) + *hfunc = func; + + return rc; +} + +static int ena_set_rxfh(struct net_device *netdev, const u32 *indir, + const u8 *key, const u8 hfunc) +{ + struct ena_adapter *adapter = netdev_priv(netdev); + struct ena_com_dev *ena_dev = adapter->ena_dev; + enum ena_admin_hash_functions func; + int rc, i; + + if (indir) { + for (i = 0; i < ENA_RX_RSS_TABLE_SIZE; i++) { + rc = ena_com_indirect_table_fill_entry(ena_dev, + ENA_IO_RXQ_IDX(indir[i]), + i); + if (unlikely(rc)) { + netif_err(adapter, drv, netdev, + "Cannot fill indirect table (index is too large)\n"); + return rc; + } + } + + rc = ena_com_indirect_table_set(ena_dev); + if (rc) { + netif_err(adapter, drv, netdev, + "Cannot set indirect table\n"); + return rc == -EPERM ? -EOPNOTSUPP : rc; + } + } + + switch (hfunc) { + case ETH_RSS_HASH_TOP: + func = ENA_ADMIN_TOEPLITZ; + break; + case ETH_RSS_HASH_XOR: + func = ENA_ADMIN_CRC32; + break; + default: + netif_err(adapter, drv, netdev, "Unsupported hfunc %d\n", + hfunc); + return -EOPNOTSUPP; + } + + if (key) { + rc = ena_com_fill_hash_function(ena_dev, func, key, + ENA_HASH_KEY_SIZE, + 0xFFFFFFFF); + if (unlikely(rc)) { + netif_err(adapter, drv, netdev, "Cannot fill key\n"); + return rc == -EPERM ? -EOPNOTSUPP : rc; + } + } + + return 0; +} + +static void ena_get_channels(struct net_device *netdev, + struct ethtool_channels *channels) +{ + struct ena_adapter *adapter = netdev_priv(netdev); + + channels->max_rx = ENA_MAX_NUM_IO_QUEUES; + channels->max_tx = ENA_MAX_NUM_IO_QUEUES; + channels->max_other = 0; + channels->max_combined = 0; + channels->rx_count = adapter->num_queues; + channels->tx_count = adapter->num_queues; + channels->other_count = 0; + channels->combined_count = 0; +} + +static int ena_get_tunable(struct net_device *netdev, + const struct ethtool_tunable *tuna, void *data) +{ + struct ena_adapter *adapter = netdev_priv(netdev); + int ret = 0; + + switch (tuna->id) { + case ETHTOOL_RX_COPYBREAK: + *(u32 *)data = adapter->rx_copybreak; + break; + default: + ret = -EINVAL; + break; + } + + return ret; +} + +static int ena_set_tunable(struct net_device *netdev, + const struct ethtool_tunable *tuna, + const void *data) +{ + struct ena_adapter *adapter = netdev_priv(netdev); + int ret = 0; + u32 len; + + switch (tuna->id) { + case ETHTOOL_RX_COPYBREAK: + len = *(u32 *)data; + if (len > adapter->netdev->mtu) { + ret = -EINVAL; + break; + } + adapter->rx_copybreak = len; + break; + default: + ret = -EINVAL; + break; + } + + return ret; +} + +static const struct ethtool_ops ena_ethtool_ops = { + .get_link_ksettings = ena_get_link_ksettings, + .get_drvinfo = ena_get_drvinfo, + .get_msglevel = ena_get_msglevel, + .set_msglevel = ena_set_msglevel, + .get_link = ethtool_op_get_link, + .get_coalesce = ena_get_coalesce, + .set_coalesce = ena_set_coalesce, + .get_ringparam = ena_get_ringparam, + .get_sset_count = ena_get_sset_count, + .get_strings = ena_get_strings, + .get_ethtool_stats = ena_get_ethtool_stats, + .get_rxnfc = ena_get_rxnfc, + .set_rxnfc = ena_set_rxnfc, + .get_rxfh_indir_size = ena_get_rxfh_indir_size, + .get_rxfh_key_size = ena_get_rxfh_key_size, + .get_rxfh = ena_get_rxfh, + .set_rxfh = ena_set_rxfh, + .get_channels = ena_get_channels, + .get_tunable = ena_get_tunable, + .set_tunable = ena_set_tunable, +}; + +void ena_set_ethtool_ops(struct net_device *netdev) +{ + netdev->ethtool_ops = &ena_ethtool_ops; +} + +static void ena_dump_stats_ex(struct ena_adapter *adapter, u8 *buf) +{ + struct net_device *netdev = adapter->netdev; + u8 *strings_buf; + u64 *data_buf; + int strings_num; + int i, rc; + + strings_num = ena_get_sset_count(netdev, ETH_SS_STATS); + if (strings_num <= 0) { + netif_err(adapter, drv, netdev, "Can't get stats num\n"); + return; + } + + strings_buf = devm_kzalloc(&adapter->pdev->dev, + strings_num * ETH_GSTRING_LEN, + GFP_ATOMIC); + if (!strings_buf) { + netif_err(adapter, drv, netdev, + "failed to alloc strings_buf\n"); + return; + } + + data_buf = devm_kzalloc(&adapter->pdev->dev, + strings_num * sizeof(u64), + GFP_ATOMIC); + if (!data_buf) { + netif_err(adapter, drv, netdev, + "failed to allocate data buf\n"); + devm_kfree(&adapter->pdev->dev, strings_buf); + return; + } + + ena_get_strings(netdev, ETH_SS_STATS, strings_buf); + ena_get_ethtool_stats(netdev, NULL, data_buf); + + /* If there is a buffer, dump stats, otherwise print them to dmesg */ + if (buf) + for (i = 0; i < strings_num; i++) { + rc = snprintf(buf, ETH_GSTRING_LEN + sizeof(u64), + "%s %llu\n", + strings_buf + i * ETH_GSTRING_LEN, + data_buf[i]); + buf += rc; + } + else + for (i = 0; i < strings_num; i++) + netif_err(adapter, drv, netdev, "%s: %llu\n", + strings_buf + i * ETH_GSTRING_LEN, + data_buf[i]); + + devm_kfree(&adapter->pdev->dev, strings_buf); + devm_kfree(&adapter->pdev->dev, data_buf); +} + +void ena_dump_stats_to_buf(struct ena_adapter *adapter, u8 *buf) +{ + if (!buf) + return; + + ena_dump_stats_ex(adapter, buf); +} + +void ena_dump_stats_to_dmesg(struct ena_adapter *adapter) +{ + ena_dump_stats_ex(adapter, NULL); +} diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c new file mode 100644 index 000000000000..5c536b8ec99c --- /dev/null +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c @@ -0,0 +1,3278 @@ +/* + * Copyright 2015 Amazon.com, Inc. or its affiliates. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#ifdef CONFIG_RFS_ACCEL +#include <linux/cpu_rmap.h> +#endif /* CONFIG_RFS_ACCEL */ +#include <linux/ethtool.h> +#include <linux/if_vlan.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/moduleparam.h> +#include <linux/numa.h> +#include <linux/pci.h> +#include <linux/utsname.h> +#include <linux/version.h> +#include <linux/vmalloc.h> +#include <net/ip.h> + +#include "ena_netdev.h" +#include "ena_pci_id_tbl.h" + +static char version[] = DEVICE_NAME " v" DRV_MODULE_VERSION "\n"; + +MODULE_AUTHOR("Amazon.com, Inc. or its affiliates"); +MODULE_DESCRIPTION(DEVICE_NAME); +MODULE_LICENSE("GPL"); +MODULE_VERSION(DRV_MODULE_VERSION); + +/* Time in jiffies before concluding the transmitter is hung. */ +#define TX_TIMEOUT (5 * HZ) + +#define ENA_NAPI_BUDGET 64 + +#define DEFAULT_MSG_ENABLE (NETIF_MSG_DRV | NETIF_MSG_PROBE | NETIF_MSG_IFUP | \ + NETIF_MSG_TX_DONE | NETIF_MSG_TX_ERR | NETIF_MSG_RX_ERR) +static int debug = -1; +module_param(debug, int, 0); +MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all)"); + +static struct ena_aenq_handlers aenq_handlers; + +static struct workqueue_struct *ena_wq; + +MODULE_DEVICE_TABLE(pci, ena_pci_tbl); + +static int ena_rss_init_default(struct ena_adapter *adapter); + +static void ena_tx_timeout(struct net_device *dev) +{ + struct ena_adapter *adapter = netdev_priv(dev); + + u64_stats_update_begin(&adapter->syncp); + adapter->dev_stats.tx_timeout++; + u64_stats_update_end(&adapter->syncp); + + netif_err(adapter, tx_err, dev, "Transmit time out\n"); + + /* Change the state of the device to trigger reset */ + set_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags); +} + +static void update_rx_ring_mtu(struct ena_adapter *adapter, int mtu) +{ + int i; + + for (i = 0; i < adapter->num_queues; i++) + adapter->rx_ring[i].mtu = mtu; +} + +static int ena_change_mtu(struct net_device *dev, int new_mtu) +{ + struct ena_adapter *adapter = netdev_priv(dev); + int ret; + + if ((new_mtu > adapter->max_mtu) || (new_mtu < ENA_MIN_MTU)) { + netif_err(adapter, drv, dev, + "Invalid MTU setting. new_mtu: %d\n", new_mtu); + + return -EINVAL; + } + + ret = ena_com_set_dev_mtu(adapter->ena_dev, new_mtu); + if (!ret) { + netif_dbg(adapter, drv, dev, "set MTU to %d\n", new_mtu); + update_rx_ring_mtu(adapter, new_mtu); + dev->mtu = new_mtu; + } else { + netif_err(adapter, drv, dev, "Failed to set MTU to %d\n", + new_mtu); + } + + return ret; +} + +static int ena_init_rx_cpu_rmap(struct ena_adapter *adapter) +{ +#ifdef CONFIG_RFS_ACCEL + u32 i; + int rc; + + adapter->netdev->rx_cpu_rmap = alloc_irq_cpu_rmap(adapter->num_queues); + if (!adapter->netdev->rx_cpu_rmap) + return -ENOMEM; + for (i = 0; i < adapter->num_queues; i++) { + int irq_idx = ENA_IO_IRQ_IDX(i); + + rc = irq_cpu_rmap_add(adapter->netdev->rx_cpu_rmap, + adapter->msix_entries[irq_idx].vector); + if (rc) { + free_irq_cpu_rmap(adapter->netdev->rx_cpu_rmap); + adapter->netdev->rx_cpu_rmap = NULL; + return rc; + } + } +#endif /* CONFIG_RFS_ACCEL */ + return 0; +} + +static void ena_init_io_rings_common(struct ena_adapter *adapter, + struct ena_ring *ring, u16 qid) +{ + ring->qid = qid; + ring->pdev = adapter->pdev; + ring->dev = &adapter->pdev->dev; + ring->netdev = adapter->netdev; + ring->napi = &adapter->ena_napi[qid].napi; + ring->adapter = adapter; + ring->ena_dev = adapter->ena_dev; + ring->per_napi_packets = 0; + ring->per_napi_bytes = 0; + ring->cpu = 0; + u64_stats_init(&ring->syncp); +} + +static void ena_init_io_rings(struct ena_adapter *adapter) +{ + struct ena_com_dev *ena_dev; + struct ena_ring *txr, *rxr; + int i; + + ena_dev = adapter->ena_dev; + + for (i = 0; i < adapter->num_queues; i++) { + txr = &adapter->tx_ring[i]; + rxr = &adapter->rx_ring[i]; + + /* TX/RX common ring state */ + ena_init_io_rings_common(adapter, txr, i); + ena_init_io_rings_common(adapter, rxr, i); + + /* TX specific ring state */ + txr->ring_size = adapter->tx_ring_size; + txr->tx_max_header_size = ena_dev->tx_max_header_size; + txr->tx_mem_queue_type = ena_dev->tx_mem_queue_type; + txr->sgl_size = adapter->max_tx_sgl_size; + txr->smoothed_interval = + ena_com_get_nonadaptive_moderation_interval_tx(ena_dev); + + /* RX specific ring state */ + rxr->ring_size = adapter->rx_ring_size; + rxr->rx_copybreak = adapter->rx_copybreak; + rxr->sgl_size = adapter->max_rx_sgl_size; + rxr->smoothed_interval = + ena_com_get_nonadaptive_moderation_interval_rx(ena_dev); + } +} + +/* ena_setup_tx_resources - allocate I/O Tx resources (Descriptors) + * @adapter: network interface device structure + * @qid: queue index + * + * Return 0 on success, negative on failure + */ +static int ena_setup_tx_resources(struct ena_adapter *adapter, int qid) +{ + struct ena_ring *tx_ring = &adapter->tx_ring[qid]; + struct ena_irq *ena_irq = &adapter->irq_tbl[ENA_IO_IRQ_IDX(qid)]; + int size, i, node; + + if (tx_ring->tx_buffer_info) { + netif_err(adapter, ifup, + adapter->netdev, "tx_buffer_info info is not NULL"); + return -EEXIST; + } + + size = sizeof(struct ena_tx_buffer) * tx_ring->ring_size; + node = cpu_to_node(ena_irq->cpu); + + tx_ring->tx_buffer_info = vzalloc_node(size, node); + if (!tx_ring->tx_buffer_info) { + tx_ring->tx_buffer_info = vzalloc(size); + if (!tx_ring->tx_buffer_info) + return -ENOMEM; + } + + size = sizeof(u16) * tx_ring->ring_size; + tx_ring->free_tx_ids = vzalloc_node(size, node); + if (!tx_ring->free_tx_ids) { + tx_ring->free_tx_ids = vzalloc(size); + if (!tx_ring->free_tx_ids) { + vfree(tx_ring->tx_buffer_info); + return -ENOMEM; + } + } + + /* Req id ring for TX out of order completions */ + for (i = 0; i < tx_ring->ring_size; i++) + tx_ring->free_tx_ids[i] = i; + + /* Reset tx statistics */ + memset(&tx_ring->tx_stats, 0x0, sizeof(tx_ring->tx_stats)); + + tx_ring->next_to_use = 0; + tx_ring->next_to_clean = 0; + tx_ring->cpu = ena_irq->cpu; + return 0; +} + +/* ena_free_tx_resources - Free I/O Tx Resources per Queue + * @adapter: network interface device structure + * @qid: queue index + * + * Free all transmit software resources + */ +static void ena_free_tx_resources(struct ena_adapter *adapter, int qid) +{ + struct ena_ring *tx_ring = &adapter->tx_ring[qid]; + + vfree(tx_ring->tx_buffer_info); + tx_ring->tx_buffer_info = NULL; + + vfree(tx_ring->free_tx_ids); + tx_ring->free_tx_ids = NULL; +} + +/* ena_setup_all_tx_resources - allocate I/O Tx queues resources for All queues + * @adapter: private structure + * + * Return 0 on success, negative on failure + */ +static int ena_setup_all_tx_resources(struct ena_adapter *adapter) +{ + int i, rc = 0; + + for (i = 0; i < adapter->num_queues; i++) { + rc = ena_setup_tx_resources(adapter, i); + if (rc) + goto err_setup_tx; + } + + return 0; + +err_setup_tx: + + netif_err(adapter, ifup, adapter->netdev, + "Tx queue %d: allocation failed\n", i); + + /* rewind the index freeing the rings as we go */ + while (i--) + ena_free_tx_resources(adapter, i); + return rc; +} + +/* ena_free_all_io_tx_resources - Free I/O Tx Resources for All Queues + * @adapter: board private structure + * + * Free all transmit software resources + */ +static void ena_free_all_io_tx_resources(struct ena_adapter *adapter) +{ + int i; + + for (i = 0; i < adapter->num_queues; i++) + ena_free_tx_resources(adapter, i); +} + +/* ena_setup_rx_resources - allocate I/O Rx resources (Descriptors) + * @adapter: network interface device structure + * @qid: queue index + * + * Returns 0 on success, negative on failure + */ +static int ena_setup_rx_resources(struct ena_adapter *adapter, + u32 qid) +{ + struct ena_ring *rx_ring = &adapter->rx_ring[qid]; + struct ena_irq *ena_irq = &adapter->irq_tbl[ENA_IO_IRQ_IDX(qid)]; + int size, node; + + if (rx_ring->rx_buffer_info) { + netif_err(adapter, ifup, adapter->netdev, + "rx_buffer_info is not NULL"); + return -EEXIST; + } + + /* alloc extra element so in rx path + * we can always prefetch rx_info + 1 + */ + size = sizeof(struct ena_rx_buffer) * (rx_ring->ring_size + 1); + node = cpu_to_node(ena_irq->cpu); + + rx_ring->rx_buffer_info = vzalloc_node(size, node); + if (!rx_ring->rx_buffer_info) { + rx_ring->rx_buffer_info = vzalloc(size); + if (!rx_ring->rx_buffer_info) + return -ENOMEM; + } + + /* Reset rx statistics */ + memset(&rx_ring->rx_stats, 0x0, sizeof(rx_ring->rx_stats)); + + rx_ring->next_to_clean = 0; + rx_ring->next_to_use = 0; + rx_ring->cpu = ena_irq->cpu; + + return 0; +} + +/* ena_free_rx_resources - Free I/O Rx Resources + * @adapter: network interface device structure + * @qid: queue index + * + * Free all receive software resources + */ +static void ena_free_rx_resources(struct ena_adapter *adapter, + u32 qid) +{ + struct ena_ring *rx_ring = &adapter->rx_ring[qid]; + + vfree(rx_ring->rx_buffer_info); + rx_ring->rx_buffer_info = NULL; +} + +/* ena_setup_all_rx_resources - allocate I/O Rx queues resources for all queues + * @adapter: board private structure + * + * Return 0 on success, negative on failure + */ +static int ena_setup_all_rx_resources(struct ena_adapter *adapter) +{ + int i, rc = 0; + + for (i = 0; i < adapter->num_queues; i++) { + rc = ena_setup_rx_resources(adapter, i); + if (rc) + goto err_setup_rx; + } + + return 0; + +err_setup_rx: + + netif_err(adapter, ifup, adapter->netdev, + "Rx queue %d: allocation failed\n", i); + + /* rewind the index freeing the rings as we go */ + while (i--) + ena_free_rx_resources(adapter, i); + return rc; +} + +/* ena_free_all_io_rx_resources - Free I/O Rx Resources for All Queues + * @adapter: board private structure + * + * Free all receive software resources + */ +static void ena_free_all_io_rx_resources(struct ena_adapter *adapter) +{ + int i; + + for (i = 0; i < adapter->num_queues; i++) + ena_free_rx_resources(adapter, i); +} + +static inline int ena_alloc_rx_page(struct ena_ring *rx_ring, + struct ena_rx_buffer *rx_info, gfp_t gfp) +{ + struct ena_com_buf *ena_buf; + struct page *page; + dma_addr_t dma; + + /* if previous allocated page is not used */ + if (unlikely(rx_info->page)) + return 0; + + page = alloc_page(gfp); + if (unlikely(!page)) { + u64_stats_update_begin(&rx_ring->syncp); + rx_ring->rx_stats.page_alloc_fail++; + u64_stats_update_end(&rx_ring->syncp); + return -ENOMEM; + } + + dma = dma_map_page(rx_ring->dev, page, 0, PAGE_SIZE, + DMA_FROM_DEVICE); + if (unlikely(dma_mapping_error(rx_ring->dev, dma))) { + u64_stats_update_begin(&rx_ring->syncp); + rx_ring->rx_stats.dma_mapping_err++; + u64_stats_update_end(&rx_ring->syncp); + + __free_page(page); + return -EIO; + } + netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev, + "alloc page %p, rx_info %p\n", page, rx_info); + + rx_info->page = page; + rx_info->page_offset = 0; + ena_buf = &rx_info->ena_buf; + ena_buf->paddr = dma; + ena_buf->len = PAGE_SIZE; + + return 0; +} + +static void ena_free_rx_page(struct ena_ring *rx_ring, + struct ena_rx_buffer *rx_info) +{ + struct page *page = rx_info->page; + struct ena_com_buf *ena_buf = &rx_info->ena_buf; + + if (unlikely(!page)) { + netif_warn(rx_ring->adapter, rx_err, rx_ring->netdev, + "Trying to free unallocated buffer\n"); + return; + } + + dma_unmap_page(rx_ring->dev, ena_buf->paddr, PAGE_SIZE, + DMA_FROM_DEVICE); + + __free_page(page); + rx_info->page = NULL; +} + +static int ena_refill_rx_bufs(struct ena_ring *rx_ring, u32 num) +{ + u16 next_to_use; + u32 i; + int rc; + + next_to_use = rx_ring->next_to_use; + + for (i = 0; i < num; i++) { + struct ena_rx_buffer *rx_info = + &rx_ring->rx_buffer_info[next_to_use]; + + rc = ena_alloc_rx_page(rx_ring, rx_info, + __GFP_COLD | GFP_ATOMIC | __GFP_COMP); + if (unlikely(rc < 0)) { + netif_warn(rx_ring->adapter, rx_err, rx_ring->netdev, + "failed to alloc buffer for rx queue %d\n", + rx_ring->qid); + break; + } + rc = ena_com_add_single_rx_desc(rx_ring->ena_com_io_sq, + &rx_info->ena_buf, + next_to_use); + if (unlikely(rc)) { + netif_warn(rx_ring->adapter, rx_status, rx_ring->netdev, + "failed to add buffer for rx queue %d\n", + rx_ring->qid); + break; + } + next_to_use = ENA_RX_RING_IDX_NEXT(next_to_use, + rx_ring->ring_size); + } + + if (unlikely(i < num)) { + u64_stats_update_begin(&rx_ring->syncp); + rx_ring->rx_stats.refil_partial++; + u64_stats_update_end(&rx_ring->syncp); + netdev_warn(rx_ring->netdev, + "refilled rx qid %d with only %d buffers (from %d)\n", + rx_ring->qid, i, num); + } + + if (likely(i)) { + /* Add memory barrier to make sure the desc were written before + * issue a doorbell + */ + wmb(); + ena_com_write_sq_doorbell(rx_ring->ena_com_io_sq); + } + + rx_ring->next_to_use = next_to_use; + + return i; +} + +static void ena_free_rx_bufs(struct ena_adapter *adapter, + u32 qid) +{ + struct ena_ring *rx_ring = &adapter->rx_ring[qid]; + u32 i; + + for (i = 0; i < rx_ring->ring_size; i++) { + struct ena_rx_buffer *rx_info = &rx_ring->rx_buffer_info[i]; + + if (rx_info->page) + ena_free_rx_page(rx_ring, rx_info); + } +} + +/* ena_refill_all_rx_bufs - allocate all queues Rx buffers + * @adapter: board private structure + * + */ +static void ena_refill_all_rx_bufs(struct ena_adapter *adapter) +{ + struct ena_ring *rx_ring; + int i, rc, bufs_num; + + for (i = 0; i < adapter->num_queues; i++) { + rx_ring = &adapter->rx_ring[i]; + bufs_num = rx_ring->ring_size - 1; + rc = ena_refill_rx_bufs(rx_ring, bufs_num); + + if (unlikely(rc != bufs_num)) + netif_warn(rx_ring->adapter, rx_status, rx_ring->netdev, + "refilling Queue %d failed. allocated %d buffers from: %d\n", + i, rc, bufs_num); + } +} + +static void ena_free_all_rx_bufs(struct ena_adapter *adapter) +{ + int i; + + for (i = 0; i < adapter->num_queues; i++) + ena_free_rx_bufs(adapter, i); +} + +/* ena_free_tx_bufs - Free Tx Buffers per Queue + * @tx_ring: TX ring for which buffers be freed + */ +static void ena_free_tx_bufs(struct ena_ring *tx_ring) +{ + u32 i; + + for (i = 0; i < tx_ring->ring_size; i++) { + struct ena_tx_buffer *tx_info = &tx_ring->tx_buffer_info[i]; + struct ena_com_buf *ena_buf; + int nr_frags; + int j; + + if (!tx_info->skb) + continue; + + netdev_notice(tx_ring->netdev, + "free uncompleted tx skb qid %d idx 0x%x\n", + tx_ring->qid, i); + + ena_buf = tx_info->bufs; + dma_unmap_single(tx_ring->dev, + ena_buf->paddr, + ena_buf->len, + DMA_TO_DEVICE); + + /* unmap remaining mapped pages */ + nr_frags = tx_info->num_of_bufs - 1; + for (j = 0; j < nr_frags; j++) { + ena_buf++; + dma_unmap_page(tx_ring->dev, + ena_buf->paddr, + ena_buf->len, + DMA_TO_DEVICE); + } + + dev_kfree_skb_any(tx_info->skb); + } + netdev_tx_reset_queue(netdev_get_tx_queue(tx_ring->netdev, + tx_ring->qid)); +} + +static void ena_free_all_tx_bufs(struct ena_adapter *adapter) +{ + struct ena_ring *tx_ring; + int i; + + for (i = 0; i < adapter->num_queues; i++) { + tx_ring = &adapter->tx_ring[i]; + ena_free_tx_bufs(tx_ring); + } +} + +static void ena_destroy_all_tx_queues(struct ena_adapter *adapter) +{ + u16 ena_qid; + int i; + + for (i = 0; i < adapter->num_queues; i++) { + ena_qid = ENA_IO_TXQ_IDX(i); + ena_com_destroy_io_queue(adapter->ena_dev, ena_qid); + } +} + +static void ena_destroy_all_rx_queues(struct ena_adapter *adapter) +{ + u16 ena_qid; + int i; + + for (i = 0; i < adapter->num_queues; i++) { + ena_qid = ENA_IO_RXQ_IDX(i); + ena_com_destroy_io_queue(adapter->ena_dev, ena_qid); + } +} + +static void ena_destroy_all_io_queues(struct ena_adapter *adapter) +{ + ena_destroy_all_tx_queues(adapter); + ena_destroy_all_rx_queues(adapter); +} + +static int validate_tx_req_id(struct ena_ring *tx_ring, u16 req_id) +{ + struct ena_tx_buffer *tx_info = NULL; + + if (likely(req_id < tx_ring->ring_size)) { + tx_info = &tx_ring->tx_buffer_info[req_id]; + if (likely(tx_info->skb)) + return 0; + } + + if (tx_info) + netif_err(tx_ring->adapter, tx_done, tx_ring->netdev, + "tx_info doesn't have valid skb\n"); + else + netif_err(tx_ring->adapter, tx_done, tx_ring->netdev, + "Invalid req_id: %hu\n", req_id); + + u64_stats_update_begin(&tx_ring->syncp); + tx_ring->tx_stats.bad_req_id++; + u64_stats_update_end(&tx_ring->syncp); + + /* Trigger device reset */ + set_bit(ENA_FLAG_TRIGGER_RESET, &tx_ring->adapter->flags); + return -EFAULT; +} + +static int ena_clean_tx_irq(struct ena_ring *tx_ring, u32 budget) +{ + struct netdev_queue *txq; + bool above_thresh; + u32 tx_bytes = 0; + u32 total_done = 0; + u16 next_to_clean; + u16 req_id; + int tx_pkts = 0; + int rc; + + next_to_clean = tx_ring->next_to_clean; + txq = netdev_get_tx_queue(tx_ring->netdev, tx_ring->qid); + + while (tx_pkts < budget) { + struct ena_tx_buffer *tx_info; + struct sk_buff *skb; + struct ena_com_buf *ena_buf; + int i, nr_frags; + + rc = ena_com_tx_comp_req_id_get(tx_ring->ena_com_io_cq, + &req_id); + if (rc) + break; + + rc = validate_tx_req_id(tx_ring, req_id); + if (rc) + break; + + tx_info = &tx_ring->tx_buffer_info[req_id]; + skb = tx_info->skb; + + /* prefetch skb_end_pointer() to speedup skb_shinfo(skb) */ + prefetch(&skb->end); + + tx_info->skb = NULL; + tx_info->last_jiffies = 0; + + if (likely(tx_info->num_of_bufs != 0)) { + ena_buf = tx_info->bufs; + + dma_unmap_single(tx_ring->dev, + dma_unmap_addr(ena_buf, paddr), + dma_unmap_len(ena_buf, len), + DMA_TO_DEVICE); + + /* unmap remaining mapped pages */ + nr_frags = tx_info->num_of_bufs - 1; + for (i = 0; i < nr_frags; i++) { + ena_buf++; + dma_unmap_page(tx_ring->dev, + dma_unmap_addr(ena_buf, paddr), + dma_unmap_len(ena_buf, len), + DMA_TO_DEVICE); + } + } + + netif_dbg(tx_ring->adapter, tx_done, tx_ring->netdev, + "tx_poll: q %d skb %p completed\n", tx_ring->qid, + skb); + + tx_bytes += skb->len; + dev_kfree_skb(skb); + tx_pkts++; + total_done += tx_info->tx_descs; + + tx_ring->free_tx_ids[next_to_clean] = req_id; + next_to_clean = ENA_TX_RING_IDX_NEXT(next_to_clean, + tx_ring->ring_size); + } + + tx_ring->next_to_clean = next_to_clean; + ena_com_comp_ack(tx_ring->ena_com_io_sq, total_done); + ena_com_update_dev_comp_head(tx_ring->ena_com_io_cq); + + netdev_tx_completed_queue(txq, tx_pkts, tx_bytes); + + netif_dbg(tx_ring->adapter, tx_done, tx_ring->netdev, + "tx_poll: q %d done. total pkts: %d\n", + tx_ring->qid, tx_pkts); + + /* need to make the rings circular update visible to + * ena_start_xmit() before checking for netif_queue_stopped(). + */ + smp_mb(); + + above_thresh = ena_com_sq_empty_space(tx_ring->ena_com_io_sq) > + ENA_TX_WAKEUP_THRESH; + if (unlikely(netif_tx_queue_stopped(txq) && above_thresh)) { + __netif_tx_lock(txq, smp_processor_id()); + above_thresh = ena_com_sq_empty_space(tx_ring->ena_com_io_sq) > + ENA_TX_WAKEUP_THRESH; + if (netif_tx_queue_stopped(txq) && above_thresh) { + netif_tx_wake_queue(txq); + u64_stats_update_begin(&tx_ring->syncp); + tx_ring->tx_stats.queue_wakeup++; + u64_stats_update_end(&tx_ring->syncp); + } + __netif_tx_unlock(txq); + } + + tx_ring->per_napi_bytes += tx_bytes; + tx_ring->per_napi_packets += tx_pkts; + + return tx_pkts; +} + +static struct sk_buff *ena_rx_skb(struct ena_ring *rx_ring, + struct ena_com_rx_buf_info *ena_bufs, + u32 descs, + u16 *next_to_clean) +{ + struct sk_buff *skb; + struct ena_rx_buffer *rx_info = + &rx_ring->rx_buffer_info[*next_to_clean]; + u32 len; + u32 buf = 0; + void *va; + + len = ena_bufs[0].len; + if (unlikely(!rx_info->page)) { + netif_err(rx_ring->adapter, rx_err, rx_ring->netdev, + "Page is NULL\n"); + return NULL; + } + + netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev, + "rx_info %p page %p\n", + rx_info, rx_info->page); + + /* save virt address of first buffer */ + va = page_address(rx_info->page) + rx_info->page_offset; + prefetch(va + NET_IP_ALIGN); + + if (len <= rx_ring->rx_copybreak) { + skb = netdev_alloc_skb_ip_align(rx_ring->netdev, + rx_ring->rx_copybreak); + if (unlikely(!skb)) { + u64_stats_update_begin(&rx_ring->syncp); + rx_ring->rx_stats.skb_alloc_fail++; + u64_stats_update_end(&rx_ring->syncp); + netif_err(rx_ring->adapter, rx_err, rx_ring->netdev, + "Failed to allocate skb\n"); + return NULL; + } + + netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev, + "rx allocated small packet. len %d. data_len %d\n", + skb->len, skb->data_len); + + /* sync this buffer for CPU use */ + dma_sync_single_for_cpu(rx_ring->dev, + dma_unmap_addr(&rx_info->ena_buf, paddr), + len, + DMA_FROM_DEVICE); + skb_copy_to_linear_data(skb, va, len); + dma_sync_single_for_device(rx_ring->dev, + dma_unmap_addr(&rx_info->ena_buf, paddr), + len, + DMA_FROM_DEVICE); + + skb_put(skb, len); + skb->protocol = eth_type_trans(skb, rx_ring->netdev); + *next_to_clean = ENA_RX_RING_IDX_ADD(*next_to_clean, descs, + rx_ring->ring_size); + return skb; + } + + skb = napi_get_frags(rx_ring->napi); + if (unlikely(!skb)) { + netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev, + "Failed allocating skb\n"); + u64_stats_update_begin(&rx_ring->syncp); + rx_ring->rx_stats.skb_alloc_fail++; + u64_stats_update_end(&rx_ring->syncp); + return NULL; + } + + do { + dma_unmap_page(rx_ring->dev, + dma_unmap_addr(&rx_info->ena_buf, paddr), + PAGE_SIZE, DMA_FROM_DEVICE); + + skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_info->page, + rx_info->page_offset, len, PAGE_SIZE); + + netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev, + "rx skb updated. len %d. data_len %d\n", + skb->len, skb->data_len); + + rx_info->page = NULL; + *next_to_clean = + ENA_RX_RING_IDX_NEXT(*next_to_clean, + rx_ring->ring_size); + if (likely(--descs == 0)) + break; + rx_info = &rx_ring->rx_buffer_info[*next_to_clean]; + len = ena_bufs[++buf].len; + } while (1); + + return skb; +} + +/* ena_rx_checksum - indicate in skb if hw indicated a good cksum + * @adapter: structure containing adapter specific data + * @ena_rx_ctx: received packet context/metadata + * @skb: skb currently being received and modified + */ +static inline void ena_rx_checksum(struct ena_ring *rx_ring, + struct ena_com_rx_ctx *ena_rx_ctx, + struct sk_buff *skb) +{ + /* Rx csum disabled */ + if (unlikely(!(rx_ring->netdev->features & NETIF_F_RXCSUM))) { + skb->ip_summed = CHECKSUM_NONE; + return; + } + + /* For fragmented packets the checksum isn't valid */ + if (ena_rx_ctx->frag) { + skb->ip_summed = CHECKSUM_NONE; + return; + } + + /* if IP and error */ + if (unlikely((ena_rx_ctx->l3_proto == ENA_ETH_IO_L3_PROTO_IPV4) && + (ena_rx_ctx->l3_csum_err))) { + /* ipv4 checksum error */ + skb->ip_summed = CHECKSUM_NONE; + u64_stats_update_begin(&rx_ring->syncp); + rx_ring->rx_stats.bad_csum++; + u64_stats_update_end(&rx_ring->syncp); + netif_err(rx_ring->adapter, rx_err, rx_ring->netdev, + "RX IPv4 header checksum error\n"); + return; + } + + /* if TCP/UDP */ + if (likely((ena_rx_ctx->l4_proto == ENA_ETH_IO_L4_PROTO_TCP) || + (ena_rx_ctx->l4_proto == ENA_ETH_IO_L4_PROTO_UDP))) { + if (unlikely(ena_rx_ctx->l4_csum_err)) { + /* TCP/UDP checksum error */ + u64_stats_update_begin(&rx_ring->syncp); + rx_ring->rx_stats.bad_csum++; + u64_stats_update_end(&rx_ring->syncp); + netif_err(rx_ring->adapter, rx_err, rx_ring->netdev, + "RX L4 checksum error\n"); + skb->ip_summed = CHECKSUM_NONE; + return; + } + + skb->ip_summed = CHECKSUM_UNNECESSARY; + } +} + +static void ena_set_rx_hash(struct ena_ring *rx_ring, + struct ena_com_rx_ctx *ena_rx_ctx, + struct sk_buff *skb) +{ + enum pkt_hash_types hash_type; + + if (likely(rx_ring->netdev->features & NETIF_F_RXHASH)) { + if (likely((ena_rx_ctx->l4_proto == ENA_ETH_IO_L4_PROTO_TCP) || + (ena_rx_ctx->l4_proto == ENA_ETH_IO_L4_PROTO_UDP))) + + hash_type = PKT_HASH_TYPE_L4; + else + hash_type = PKT_HASH_TYPE_NONE; + + /* Override hash type if the packet is fragmented */ + if (ena_rx_ctx->frag) + hash_type = PKT_HASH_TYPE_NONE; + + skb_set_hash(skb, ena_rx_ctx->hash, hash_type); + } +} + +/* ena_clean_rx_irq - Cleanup RX irq + * @rx_ring: RX ring to clean + * @napi: napi handler + * @budget: how many packets driver is allowed to clean + * + * Returns the number of cleaned buffers. + */ +static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi, + u32 budget) +{ + u16 next_to_clean = rx_ring->next_to_clean; + u32 res_budget, work_done; + + struct ena_com_rx_ctx ena_rx_ctx; + struct ena_adapter *adapter; + struct sk_buff *skb; + int refill_required; + int refill_threshold; + int rc = 0; + int total_len = 0; + int rx_copybreak_pkt = 0; + + netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev, + "%s qid %d\n", __func__, rx_ring->qid); + res_budget = budget; + + do { + ena_rx_ctx.ena_bufs = rx_ring->ena_bufs; + ena_rx_ctx.max_bufs = rx_ring->sgl_size; + ena_rx_ctx.descs = 0; + rc = ena_com_rx_pkt(rx_ring->ena_com_io_cq, + rx_ring->ena_com_io_sq, + &ena_rx_ctx); + if (unlikely(rc)) + goto error; + + if (unlikely(ena_rx_ctx.descs == 0)) + break; + + netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev, + "rx_poll: q %d got packet from ena. descs #: %d l3 proto %d l4 proto %d hash: %x\n", + rx_ring->qid, ena_rx_ctx.descs, ena_rx_ctx.l3_proto, + ena_rx_ctx.l4_proto, ena_rx_ctx.hash); + + /* allocate skb and fill it */ + skb = ena_rx_skb(rx_ring, rx_ring->ena_bufs, ena_rx_ctx.descs, + &next_to_clean); + + /* exit if we failed to retrieve a buffer */ + if (unlikely(!skb)) { + next_to_clean = ENA_RX_RING_IDX_ADD(next_to_clean, + ena_rx_ctx.descs, + rx_ring->ring_size); + break; + } + + ena_rx_checksum(rx_ring, &ena_rx_ctx, skb); + + ena_set_rx_hash(rx_ring, &ena_rx_ctx, skb); + + skb_record_rx_queue(skb, rx_ring->qid); + + if (rx_ring->ena_bufs[0].len <= rx_ring->rx_copybreak) { + total_len += rx_ring->ena_bufs[0].len; + rx_copybreak_pkt++; + napi_gro_receive(napi, skb); + } else { + total_len += skb->len; + napi_gro_frags(napi); + } + + res_budget--; + } while (likely(res_budget)); + + work_done = budget - res_budget; + rx_ring->per_napi_bytes += total_len; + rx_ring->per_napi_packets += work_done; + u64_stats_update_begin(&rx_ring->syncp); + rx_ring->rx_stats.bytes += total_len; + rx_ring->rx_stats.cnt += work_done; + rx_ring->rx_stats.rx_copybreak_pkt += rx_copybreak_pkt; + u64_stats_update_end(&rx_ring->syncp); + + rx_ring->next_to_clean = next_to_clean; + + refill_required = ena_com_sq_empty_space(rx_ring->ena_com_io_sq); + refill_threshold = rx_ring->ring_size / ENA_RX_REFILL_THRESH_DIVIDER; + + /* Optimization, try to batch new rx buffers */ + if (refill_required > refill_threshold) { + ena_com_update_dev_comp_head(rx_ring->ena_com_io_cq); + ena_refill_rx_bufs(rx_ring, refill_required); + } + + return work_done; + +error: + adapter = netdev_priv(rx_ring->netdev); + + u64_stats_update_begin(&rx_ring->syncp); + rx_ring->rx_stats.bad_desc_num++; + u64_stats_update_end(&rx_ring->syncp); + + /* Too many desc from the device. Trigger reset */ + set_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags); + + return 0; +} + +inline void ena_adjust_intr_moderation(struct ena_ring *rx_ring, + struct ena_ring *tx_ring) +{ + /* We apply adaptive moderation on Rx path only. + * Tx uses static interrupt moderation. + */ + ena_com_calculate_interrupt_delay(rx_ring->ena_dev, + rx_ring->per_napi_packets, + rx_ring->per_napi_bytes, + &rx_ring->smoothed_interval, + &rx_ring->moder_tbl_idx); + + /* Reset per napi packets/bytes */ + tx_ring->per_napi_packets = 0; + tx_ring->per_napi_bytes = 0; + rx_ring->per_napi_packets = 0; + rx_ring->per_napi_bytes = 0; +} + +static inline void ena_update_ring_numa_node(struct ena_ring *tx_ring, + struct ena_ring *rx_ring) +{ + int cpu = get_cpu(); + int numa_node; + + /* Check only one ring since the 2 rings are running on the same cpu */ + if (likely(tx_ring->cpu == cpu)) + goto out; + + numa_node = cpu_to_node(cpu); + put_cpu(); + + if (numa_node != NUMA_NO_NODE) { + ena_com_update_numa_node(tx_ring->ena_com_io_cq, numa_node); + ena_com_update_numa_node(rx_ring->ena_com_io_cq, numa_node); + } + + tx_ring->cpu = cpu; + rx_ring->cpu = cpu; + + return; +out: + put_cpu(); +} + +static int ena_io_poll(struct napi_struct *napi, int budget) +{ + struct ena_napi *ena_napi = container_of(napi, struct ena_napi, napi); + struct ena_ring *tx_ring, *rx_ring; + struct ena_eth_io_intr_reg intr_reg; + + u32 tx_work_done; + u32 rx_work_done; + int tx_budget; + int napi_comp_call = 0; + int ret; + + tx_ring = ena_napi->tx_ring; + rx_ring = ena_napi->rx_ring; + + tx_budget = tx_ring->ring_size / ENA_TX_POLL_BUDGET_DIVIDER; + + if (!test_bit(ENA_FLAG_DEV_UP, &tx_ring->adapter->flags)) { + napi_complete_done(napi, 0); + return 0; + } + + tx_work_done = ena_clean_tx_irq(tx_ring, tx_budget); + rx_work_done = ena_clean_rx_irq(rx_ring, napi, budget); + + if ((budget > rx_work_done) && (tx_budget > tx_work_done)) { + napi_complete_done(napi, rx_work_done); + + napi_comp_call = 1; + /* Tx and Rx share the same interrupt vector */ + if (ena_com_get_adaptive_moderation_enabled(rx_ring->ena_dev)) + ena_adjust_intr_moderation(rx_ring, tx_ring); + + /* Update intr register: rx intr delay, tx intr delay and + * interrupt unmask + */ + ena_com_update_intr_reg(&intr_reg, + rx_ring->smoothed_interval, + tx_ring->smoothed_interval, + true); + + /* It is a shared MSI-X. Tx and Rx CQ have pointer to it. + * So we use one of them to reach the intr reg + */ + ena_com_unmask_intr(rx_ring->ena_com_io_cq, &intr_reg); + + ena_update_ring_numa_node(tx_ring, rx_ring); + + ret = rx_work_done; + } else { + ret = budget; + } + + u64_stats_update_begin(&tx_ring->syncp); + tx_ring->tx_stats.napi_comp += napi_comp_call; + tx_ring->tx_stats.tx_poll++; + u64_stats_update_end(&tx_ring->syncp); + + return ret; +} + +static irqreturn_t ena_intr_msix_mgmnt(int irq, void *data) +{ + struct ena_adapter *adapter = (struct ena_adapter *)data; + + ena_com_admin_q_comp_intr_handler(adapter->ena_dev); + + /* Don't call the aenq handler before probe is done */ + if (likely(test_bit(ENA_FLAG_DEVICE_RUNNING, &adapter->flags))) + ena_com_aenq_intr_handler(adapter->ena_dev, data); + + return IRQ_HANDLED; +} + +/* ena_intr_msix_io - MSI-X Interrupt Handler for Tx/Rx + * @irq: interrupt number + * @data: pointer to a network interface private napi device structure + */ +static irqreturn_t ena_intr_msix_io(int irq, void *data) +{ + struct ena_napi *ena_napi = data; + + napi_schedule(&ena_napi->napi); + + return IRQ_HANDLED; +} + +static int ena_enable_msix(struct ena_adapter *adapter, int num_queues) +{ + int i, msix_vecs, rc; + + if (test_bit(ENA_FLAG_MSIX_ENABLED, &adapter->flags)) { + netif_err(adapter, probe, adapter->netdev, + "Error, MSI-X is already enabled\n"); + return -EPERM; + } + + /* Reserved the max msix vectors we might need */ + msix_vecs = ENA_MAX_MSIX_VEC(num_queues); + + netif_dbg(adapter, probe, adapter->netdev, + "trying to enable MSI-X, vectors %d\n", msix_vecs); + + adapter->msix_entries = vzalloc(msix_vecs * sizeof(struct msix_entry)); + + if (!adapter->msix_entries) + return -ENOMEM; + + for (i = 0; i < msix_vecs; i++) + adapter->msix_entries[i].entry = i; + + rc = pci_enable_msix(adapter->pdev, adapter->msix_entries, msix_vecs); + if (rc != 0) { + netif_err(adapter, probe, adapter->netdev, + "Failed to enable MSI-X, vectors %d rc %d\n", + msix_vecs, rc); + return -ENOSPC; + } + + netif_dbg(adapter, probe, adapter->netdev, "enable MSI-X, vectors %d\n", + msix_vecs); + + if (msix_vecs >= 1) { + if (ena_init_rx_cpu_rmap(adapter)) + netif_warn(adapter, probe, adapter->netdev, + "Failed to map IRQs to CPUs\n"); + } + + adapter->msix_vecs = msix_vecs; + set_bit(ENA_FLAG_MSIX_ENABLED, &adapter->flags); + + return 0; +} + +static void ena_setup_mgmnt_intr(struct ena_adapter *adapter) +{ + u32 cpu; + + snprintf(adapter->irq_tbl[ENA_MGMNT_IRQ_IDX].name, + ENA_IRQNAME_SIZE, "ena-mgmnt@pci:%s", + pci_name(adapter->pdev)); + adapter->irq_tbl[ENA_MGMNT_IRQ_IDX].handler = + ena_intr_msix_mgmnt; + adapter->irq_tbl[ENA_MGMNT_IRQ_IDX].data = adapter; + adapter->irq_tbl[ENA_MGMNT_IRQ_IDX].vector = + adapter->msix_entries[ENA_MGMNT_IRQ_IDX].vector; + cpu = cpumask_first(cpu_online_mask); + adapter->irq_tbl[ENA_MGMNT_IRQ_IDX].cpu = cpu; + cpumask_set_cpu(cpu, + &adapter->irq_tbl[ENA_MGMNT_IRQ_IDX].affinity_hint_mask); +} + +static void ena_setup_io_intr(struct ena_adapter *adapter) +{ + struct net_device *netdev; + int irq_idx, i, cpu; + + netdev = adapter->netdev; + + for (i = 0; i < adapter->num_queues; i++) { + irq_idx = ENA_IO_IRQ_IDX(i); + cpu = i % num_online_cpus(); + + snprintf(adapter->irq_tbl[irq_idx].name, ENA_IRQNAME_SIZE, + "%s-Tx-Rx-%d", netdev->name, i); + adapter->irq_tbl[irq_idx].handler = ena_intr_msix_io; + adapter->irq_tbl[irq_idx].data = &adapter->ena_napi[i]; + adapter->irq_tbl[irq_idx].vector = + adapter->msix_entries[irq_idx].vector; + adapter->irq_tbl[irq_idx].cpu = cpu; + + cpumask_set_cpu(cpu, + &adapter->irq_tbl[irq_idx].affinity_hint_mask); + } +} + +static int ena_request_mgmnt_irq(struct ena_adapter *adapter) +{ + unsigned long flags = 0; + struct ena_irq *irq; + int rc; + + irq = &adapter->irq_tbl[ENA_MGMNT_IRQ_IDX]; + rc = request_irq(irq->vector, irq->handler, flags, irq->name, + irq->data); + if (rc) { + netif_err(adapter, probe, adapter->netdev, + "failed to request admin irq\n"); + return rc; + } + + netif_dbg(adapter, probe, adapter->netdev, + "set affinity hint of mgmnt irq.to 0x%lx (irq vector: %d)\n", + irq->affinity_hint_mask.bits[0], irq->vector); + + irq_set_affinity_hint(irq->vector, &irq->affinity_hint_mask); + + return rc; +} + +static int ena_request_io_irq(struct ena_adapter *adapter) +{ + unsigned long flags = 0; + struct ena_irq *irq; + int rc = 0, i, k; + + if (!test_bit(ENA_FLAG_MSIX_ENABLED, &adapter->flags)) { + netif_err(adapter, ifup, adapter->netdev, + "Failed to request I/O IRQ: MSI-X is not enabled\n"); + return -EINVAL; + } + + for (i = ENA_IO_IRQ_FIRST_IDX; i < adapter->msix_vecs; i++) { + irq = &adapter->irq_tbl[i]; + rc = request_irq(irq->vector, irq->handler, flags, irq->name, + irq->data); + if (rc) { + netif_err(adapter, ifup, adapter->netdev, + "Failed to request I/O IRQ. index %d rc %d\n", + i, rc); + goto err; + } + + netif_dbg(adapter, ifup, adapter->netdev, + "set affinity hint of irq. index %d to 0x%lx (irq vector: %d)\n", + i, irq->affinity_hint_mask.bits[0], irq->vector); + + irq_set_affinity_hint(irq->vector, &irq->affinity_hint_mask); + } + + return rc; + +err: + for (k = ENA_IO_IRQ_FIRST_IDX; k < i; k++) { + irq = &adapter->irq_tbl[k]; + free_irq(irq->vector, irq->data); + } + + return rc; +} + +static void ena_free_mgmnt_irq(struct ena_adapter *adapter) +{ + struct ena_irq *irq; + + irq = &adapter->irq_tbl[ENA_MGMNT_IRQ_IDX]; + synchronize_irq(irq->vector); + irq_set_affinity_hint(irq->vector, NULL); + free_irq(irq->vector, irq->data); +} + +static void ena_free_io_irq(struct ena_adapter *adapter) +{ + struct ena_irq *irq; + int i; + +#ifdef CONFIG_RFS_ACCEL + if (adapter->msix_vecs >= 1) { + free_irq_cpu_rmap(adapter->netdev->rx_cpu_rmap); + adapter->netdev->rx_cpu_rmap = NULL; + } +#endif /* CONFIG_RFS_ACCEL */ + + for (i = ENA_IO_IRQ_FIRST_IDX; i < adapter->msix_vecs; i++) { + irq = &adapter->irq_tbl[i]; + irq_set_affinity_hint(irq->vector, NULL); + free_irq(irq->vector, irq->data); + } +} + +static void ena_disable_msix(struct ena_adapter *adapter) +{ + if (test_and_clear_bit(ENA_FLAG_MSIX_ENABLED, &adapter->flags)) + pci_disable_msix(adapter->pdev); + + if (adapter->msix_entries) + vfree(adapter->msix_entries); + adapter->msix_entries = NULL; +} + +static void ena_disable_io_intr_sync(struct ena_adapter *adapter) +{ + int i; + + if (!netif_running(adapter->netdev)) + return; + + for (i = ENA_IO_IRQ_FIRST_IDX; i < adapter->msix_vecs; i++) + synchronize_irq(adapter->irq_tbl[i].vector); +} + +static void ena_del_napi(struct ena_adapter *adapter) +{ + int i; + + for (i = 0; i < adapter->num_queues; i++) + netif_napi_del(&adapter->ena_napi[i].napi); +} + +static void ena_init_napi(struct ena_adapter *adapter) +{ + struct ena_napi *napi; + int i; + + for (i = 0; i < adapter->num_queues; i++) { + napi = &adapter->ena_napi[i]; + + netif_napi_add(adapter->netdev, + &adapter->ena_napi[i].napi, + ena_io_poll, + ENA_NAPI_BUDGET); + napi->rx_ring = &adapter->rx_ring[i]; + napi->tx_ring = &adapter->tx_ring[i]; + napi->qid = i; + } +} + +static void ena_napi_disable_all(struct ena_adapter *adapter) +{ + int i; + + for (i = 0; i < adapter->num_queues; i++) + napi_disable(&adapter->ena_napi[i].napi); +} + +static void ena_napi_enable_all(struct ena_adapter *adapter) +{ + int i; + + for (i = 0; i < adapter->num_queues; i++) + napi_enable(&adapter->ena_napi[i].napi); +} + +static void ena_restore_ethtool_params(struct ena_adapter *adapter) +{ + adapter->tx_usecs = 0; + adapter->rx_usecs = 0; + adapter->tx_frames = 1; + adapter->rx_frames = 1; +} + +/* Configure the Rx forwarding */ +static int ena_rss_configure(struct ena_adapter *adapter) +{ + struct ena_com_dev *ena_dev = adapter->ena_dev; + int rc; + + /* In case the RSS table wasn't initialized by probe */ + if (!ena_dev->rss.tbl_log_size) { + rc = ena_rss_init_default(adapter); + if (rc && (rc != -EPERM)) { + netif_err(adapter, ifup, adapter->netdev, + "Failed to init RSS rc: %d\n", rc); + return rc; + } + } + + /* Set indirect table */ + rc = ena_com_indirect_table_set(ena_dev); + if (unlikely(rc && rc != -EPERM)) + return rc; + + /* Configure hash function (if supported) */ + rc = ena_com_set_hash_function(ena_dev); + if (unlikely(rc && (rc != -EPERM))) + return rc; + + /* Configure hash inputs (if supported) */ + rc = ena_com_set_hash_ctrl(ena_dev); + if (unlikely(rc && (rc != -EPERM))) + return rc; + + return 0; +} + +static int ena_up_complete(struct ena_adapter *adapter) +{ + int rc, i; + + rc = ena_rss_configure(adapter); + if (rc) + return rc; + + ena_init_napi(adapter); + + ena_change_mtu(adapter->netdev, adapter->netdev->mtu); + + ena_refill_all_rx_bufs(adapter); + + /* enable transmits */ + netif_tx_start_all_queues(adapter->netdev); + + ena_restore_ethtool_params(adapter); + + ena_napi_enable_all(adapter); + + /* schedule napi in case we had pending packets + * from the last time we disable napi + */ + for (i = 0; i < adapter->num_queues; i++) + napi_schedule(&adapter->ena_napi[i].napi); + + return 0; +} + +static int ena_create_io_tx_queue(struct ena_adapter *adapter, int qid) +{ + struct ena_com_create_io_ctx ctx = { 0 }; + struct ena_com_dev *ena_dev; + struct ena_ring *tx_ring; + u32 msix_vector; + u16 ena_qid; + int rc; + + ena_dev = adapter->ena_dev; + + tx_ring = &adapter->tx_ring[qid]; + msix_vector = ENA_IO_IRQ_IDX(qid); + ena_qid = ENA_IO_TXQ_IDX(qid); + + ctx.direction = ENA_COM_IO_QUEUE_DIRECTION_TX; + ctx.qid = ena_qid; + ctx.mem_queue_type = ena_dev->tx_mem_queue_type; + ctx.msix_vector = msix_vector; + ctx.queue_size = adapter->tx_ring_size; + ctx.numa_node = cpu_to_node(tx_ring->cpu); + + rc = ena_com_create_io_queue(ena_dev, &ctx); + if (rc) { + netif_err(adapter, ifup, adapter->netdev, + "Failed to create I/O TX queue num %d rc: %d\n", + qid, rc); + return rc; + } + + rc = ena_com_get_io_handlers(ena_dev, ena_qid, + &tx_ring->ena_com_io_sq, + &tx_ring->ena_com_io_cq); + if (rc) { + netif_err(adapter, ifup, adapter->netdev, + "Failed to get TX queue handlers. TX queue num %d rc: %d\n", + qid, rc); + ena_com_destroy_io_queue(ena_dev, ena_qid); + } + + ena_com_update_numa_node(tx_ring->ena_com_io_cq, ctx.numa_node); + return rc; +} + +static int ena_create_all_io_tx_queues(struct ena_adapter *adapter) +{ + struct ena_com_dev *ena_dev = adapter->ena_dev; + int rc, i; + + for (i = 0; i < adapter->num_queues; i++) { + rc = ena_create_io_tx_queue(adapter, i); + if (rc) + goto create_err; + } + + return 0; + +create_err: + while (i--) + ena_com_destroy_io_queue(ena_dev, ENA_IO_TXQ_IDX(i)); + + return rc; +} + +static int ena_create_io_rx_queue(struct ena_adapter *adapter, int qid) +{ + struct ena_com_dev *ena_dev; + struct ena_com_create_io_ctx ctx = { 0 }; + struct ena_ring *rx_ring; + u32 msix_vector; + u16 ena_qid; + int rc; + + ena_dev = adapter->ena_dev; + + rx_ring = &adapter->rx_ring[qid]; + msix_vector = ENA_IO_IRQ_IDX(qid); + ena_qid = ENA_IO_RXQ_IDX(qid); + + ctx.qid = ena_qid; + ctx.direction = ENA_COM_IO_QUEUE_DIRECTION_RX; + ctx.mem_queue_type = ENA_ADMIN_PLACEMENT_POLICY_HOST; + ctx.msix_vector = msix_vector; + ctx.queue_size = adapter->rx_ring_size; + ctx.numa_node = cpu_to_node(rx_ring->cpu); + + rc = ena_com_create_io_queue(ena_dev, &ctx); + if (rc) { + netif_err(adapter, ifup, adapter->netdev, + "Failed to create I/O RX queue num %d rc: %d\n", + qid, rc); + return rc; + } + + rc = ena_com_get_io_handlers(ena_dev, ena_qid, + &rx_ring->ena_com_io_sq, + &rx_ring->ena_com_io_cq); + if (rc) { + netif_err(adapter, ifup, adapter->netdev, + "Failed to get RX queue handlers. RX queue num %d rc: %d\n", + qid, rc); + ena_com_destroy_io_queue(ena_dev, ena_qid); + } + + ena_com_update_numa_node(rx_ring->ena_com_io_cq, ctx.numa_node); + + return rc; +} + +static int ena_create_all_io_rx_queues(struct ena_adapter *adapter) +{ + struct ena_com_dev *ena_dev = adapter->ena_dev; + int rc, i; + + for (i = 0; i < adapter->num_queues; i++) { + rc = ena_create_io_rx_queue(adapter, i); + if (rc) + goto create_err; + } + + return 0; + +create_err: + while (i--) + ena_com_destroy_io_queue(ena_dev, ENA_IO_RXQ_IDX(i)); + + return rc; +} + +static int ena_up(struct ena_adapter *adapter) +{ + int rc; + + netdev_dbg(adapter->netdev, "%s\n", __func__); + + ena_setup_io_intr(adapter); + + rc = ena_request_io_irq(adapter); + if (rc) + goto err_req_irq; + + /* allocate transmit descriptors */ + rc = ena_setup_all_tx_resources(adapter); + if (rc) + goto err_setup_tx; + + /* allocate receive descriptors */ + rc = ena_setup_all_rx_resources(adapter); + if (rc) + goto err_setup_rx; + + /* Create TX queues */ + rc = ena_create_all_io_tx_queues(adapter); + if (rc) + goto err_create_tx_queues; + + /* Create RX queues */ + rc = ena_create_all_io_rx_queues(adapter); + if (rc) + goto err_create_rx_queues; + + rc = ena_up_complete(adapter); + if (rc) + goto err_up; + + if (test_bit(ENA_FLAG_LINK_UP, &adapter->flags)) + netif_carrier_on(adapter->netdev); + + u64_stats_update_begin(&adapter->syncp); + adapter->dev_stats.interface_up++; + u64_stats_update_end(&adapter->syncp); + + set_bit(ENA_FLAG_DEV_UP, &adapter->flags); + + return rc; + +err_up: + ena_destroy_all_rx_queues(adapter); +err_create_rx_queues: + ena_destroy_all_tx_queues(adapter); +err_create_tx_queues: + ena_free_all_io_rx_resources(adapter); +err_setup_rx: + ena_free_all_io_tx_resources(adapter); +err_setup_tx: + ena_free_io_irq(adapter); +err_req_irq: + + return rc; +} + +static void ena_down(struct ena_adapter *adapter) +{ + netif_info(adapter, ifdown, adapter->netdev, "%s\n", __func__); + + clear_bit(ENA_FLAG_DEV_UP, &adapter->flags); + + u64_stats_update_begin(&adapter->syncp); + adapter->dev_stats.interface_down++; + u64_stats_update_end(&adapter->syncp); + + /* After this point the napi handler won't enable the tx queue */ + ena_napi_disable_all(adapter); + netif_carrier_off(adapter->netdev); + netif_tx_disable(adapter->netdev); + + /* After destroy the queue there won't be any new interrupts */ + ena_destroy_all_io_queues(adapter); + + ena_disable_io_intr_sync(adapter); + ena_free_io_irq(adapter); + ena_del_napi(adapter); + + ena_free_all_tx_bufs(adapter); + ena_free_all_rx_bufs(adapter); + ena_free_all_io_tx_resources(adapter); + ena_free_all_io_rx_resources(adapter); +} + +/* ena_open - Called when a network interface is made active + * @netdev: network interface device structure + * + * Returns 0 on success, negative value on failure + * + * The open entry point is called when a network interface is made + * active by the system (IFF_UP). At this point all resources needed + * for transmit and receive operations are allocated, the interrupt + * handler is registered with the OS, the watchdog timer is started, + * and the stack is notified that the interface is ready. + */ +static int ena_open(struct net_device *netdev) +{ + struct ena_adapter *adapter = netdev_priv(netdev); + int rc; + + /* Notify the stack of the actual queue counts. */ + rc = netif_set_real_num_tx_queues(netdev, adapter->num_queues); + if (rc) { + netif_err(adapter, ifup, netdev, "Can't set num tx queues\n"); + return rc; + } + + rc = netif_set_real_num_rx_queues(netdev, adapter->num_queues); + if (rc) { + netif_err(adapter, ifup, netdev, "Can't set num rx queues\n"); + return rc; + } + + rc = ena_up(adapter); + if (rc) + return rc; + + return rc; +} + +/* ena_close - Disables a network interface + * @netdev: network interface device structure + * + * Returns 0, this is not allowed to fail + * + * The close entry point is called when an interface is de-activated + * by the OS. The hardware is still under the drivers control, but + * needs to be disabled. A global MAC reset is issued to stop the + * hardware, and all transmit and receive resources are freed. + */ +static int ena_close(struct net_device *netdev) +{ + struct ena_adapter *adapter = netdev_priv(netdev); + + netif_dbg(adapter, ifdown, netdev, "%s\n", __func__); + + if (test_bit(ENA_FLAG_DEV_UP, &adapter->flags)) + ena_down(adapter); + + return 0; +} + +static void ena_tx_csum(struct ena_com_tx_ctx *ena_tx_ctx, struct sk_buff *skb) +{ + u32 mss = skb_shinfo(skb)->gso_size; + struct ena_com_tx_meta *ena_meta = &ena_tx_ctx->ena_meta; + u8 l4_protocol = 0; + + if ((skb->ip_summed == CHECKSUM_PARTIAL) || mss) { + ena_tx_ctx->l4_csum_enable = 1; + if (mss) { + ena_tx_ctx->tso_enable = 1; + ena_meta->l4_hdr_len = tcp_hdr(skb)->doff; + ena_tx_ctx->l4_csum_partial = 0; + } else { + ena_tx_ctx->tso_enable = 0; + ena_meta->l4_hdr_len = 0; + ena_tx_ctx->l4_csum_partial = 1; + } + + switch (ip_hdr(skb)->version) { + case IPVERSION: + ena_tx_ctx->l3_proto = ENA_ETH_IO_L3_PROTO_IPV4; + if (ip_hdr(skb)->frag_off & htons(IP_DF)) + ena_tx_ctx->df = 1; + if (mss) + ena_tx_ctx->l3_csum_enable = 1; + l4_protocol = ip_hdr(skb)->protocol; + break; + case 6: + ena_tx_ctx->l3_proto = ENA_ETH_IO_L3_PROTO_IPV6; + l4_protocol = ipv6_hdr(skb)->nexthdr; + break; + default: + break; + } + + if (l4_protocol == IPPROTO_TCP) + ena_tx_ctx->l4_proto = ENA_ETH_IO_L4_PROTO_TCP; + else + ena_tx_ctx->l4_proto = ENA_ETH_IO_L4_PROTO_UDP; + + ena_meta->mss = mss; + ena_meta->l3_hdr_len = skb_network_header_len(skb); + ena_meta->l3_hdr_offset = skb_network_offset(skb); + ena_tx_ctx->meta_valid = 1; + + } else { + ena_tx_ctx->meta_valid = 0; + } +} + +static int ena_check_and_linearize_skb(struct ena_ring *tx_ring, + struct sk_buff *skb) +{ + int num_frags, header_len, rc; + + num_frags = skb_shinfo(skb)->nr_frags; + header_len = skb_headlen(skb); + + if (num_frags < tx_ring->sgl_size) + return 0; + + if ((num_frags == tx_ring->sgl_size) && + (header_len < tx_ring->tx_max_header_size)) + return 0; + + u64_stats_update_begin(&tx_ring->syncp); + tx_ring->tx_stats.linearize++; + u64_stats_update_end(&tx_ring->syncp); + + rc = skb_linearize(skb); + if (unlikely(rc)) { + u64_stats_update_begin(&tx_ring->syncp); + tx_ring->tx_stats.linearize_failed++; + u64_stats_update_end(&tx_ring->syncp); + } + + return rc; +} + +/* Called with netif_tx_lock. */ +static netdev_tx_t ena_start_xmit(struct sk_buff *skb, struct net_device *dev) +{ + struct ena_adapter *adapter = netdev_priv(dev); + struct ena_tx_buffer *tx_info; + struct ena_com_tx_ctx ena_tx_ctx; + struct ena_ring *tx_ring; + struct netdev_queue *txq; + struct ena_com_buf *ena_buf; + void *push_hdr; + u32 len, last_frag; + u16 next_to_use; + u16 req_id; + u16 push_len; + u16 header_len; + dma_addr_t dma; + int qid, rc, nb_hw_desc; + int i = -1; + + netif_dbg(adapter, tx_queued, dev, "%s skb %p\n", __func__, skb); + /* Determine which tx ring we will be placed on */ + qid = skb_get_queue_mapping(skb); + tx_ring = &adapter->tx_ring[qid]; + txq = netdev_get_tx_queue(dev, qid); + + rc = ena_check_and_linearize_skb(tx_ring, skb); + if (unlikely(rc)) + goto error_drop_packet; + + skb_tx_timestamp(skb); + len = skb_headlen(skb); + + next_to_use = tx_ring->next_to_use; + req_id = tx_ring->free_tx_ids[next_to_use]; + tx_info = &tx_ring->tx_buffer_info[req_id]; + tx_info->num_of_bufs = 0; + + WARN(tx_info->skb, "SKB isn't NULL req_id %d\n", req_id); + ena_buf = tx_info->bufs; + tx_info->skb = skb; + + if (tx_ring->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV) { + /* prepared the push buffer */ + push_len = min_t(u32, len, tx_ring->tx_max_header_size); + header_len = push_len; + push_hdr = skb->data; + } else { + push_len = 0; + header_len = min_t(u32, len, tx_ring->tx_max_header_size); + push_hdr = NULL; + } + + netif_dbg(adapter, tx_queued, dev, + "skb: %p header_buf->vaddr: %p push_len: %d\n", skb, + push_hdr, push_len); + + if (len > push_len) { + dma = dma_map_single(tx_ring->dev, skb->data + push_len, + len - push_len, DMA_TO_DEVICE); + if (dma_mapping_error(tx_ring->dev, dma)) + goto error_report_dma_error; + + ena_buf->paddr = dma; + ena_buf->len = len - push_len; + + ena_buf++; + tx_info->num_of_bufs++; + } + + last_frag = skb_shinfo(skb)->nr_frags; + + for (i = 0; i < last_frag; i++) { + const skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; + + len = skb_frag_size(frag); + dma = skb_frag_dma_map(tx_ring->dev, frag, 0, len, + DMA_TO_DEVICE); + if (dma_mapping_error(tx_ring->dev, dma)) + goto error_report_dma_error; + + ena_buf->paddr = dma; + ena_buf->len = len; + ena_buf++; + } + + tx_info->num_of_bufs += last_frag; + + memset(&ena_tx_ctx, 0x0, sizeof(struct ena_com_tx_ctx)); + ena_tx_ctx.ena_bufs = tx_info->bufs; + ena_tx_ctx.push_header = push_hdr; + ena_tx_ctx.num_bufs = tx_info->num_of_bufs; + ena_tx_ctx.req_id = req_id; + ena_tx_ctx.header_len = header_len; + + /* set flags and meta data */ + ena_tx_csum(&ena_tx_ctx, skb); + + /* prepare the packet's descriptors to dma engine */ + rc = ena_com_prepare_tx(tx_ring->ena_com_io_sq, &ena_tx_ctx, + &nb_hw_desc); + + if (unlikely(rc)) { + netif_err(adapter, tx_queued, dev, + "failed to prepare tx bufs\n"); + u64_stats_update_begin(&tx_ring->syncp); + tx_ring->tx_stats.queue_stop++; + tx_ring->tx_stats.prepare_ctx_err++; + u64_stats_update_end(&tx_ring->syncp); + netif_tx_stop_queue(txq); + goto error_unmap_dma; + } + + netdev_tx_sent_queue(txq, skb->len); + + u64_stats_update_begin(&tx_ring->syncp); + tx_ring->tx_stats.cnt++; + tx_ring->tx_stats.bytes += skb->len; + u64_stats_update_end(&tx_ring->syncp); + + tx_info->tx_descs = nb_hw_desc; + tx_info->last_jiffies = jiffies; + + tx_ring->next_to_use = ENA_TX_RING_IDX_NEXT(next_to_use, + tx_ring->ring_size); + + /* This WMB is aimed to: + * 1 - perform smp barrier before reading next_to_completion + * 2 - make sure the desc were written before trigger DB + */ + wmb(); + + /* stop the queue when no more space available, the packet can have up + * to sgl_size + 2. one for the meta descriptor and one for header + * (if the header is larger than tx_max_header_size). + */ + if (unlikely(ena_com_sq_empty_space(tx_ring->ena_com_io_sq) < + (tx_ring->sgl_size + 2))) { + netif_dbg(adapter, tx_queued, dev, "%s stop queue %d\n", + __func__, qid); + + netif_tx_stop_queue(txq); + u64_stats_update_begin(&tx_ring->syncp); + tx_ring->tx_stats.queue_stop++; + u64_stats_update_end(&tx_ring->syncp); + + /* There is a rare condition where this function decide to + * stop the queue but meanwhile clean_tx_irq updates + * next_to_completion and terminates. + * The queue will remain stopped forever. + * To solve this issue this function perform rmb, check + * the wakeup condition and wake up the queue if needed. + */ + smp_rmb(); + + if (ena_com_sq_empty_space(tx_ring->ena_com_io_sq) + > ENA_TX_WAKEUP_THRESH) { + netif_tx_wake_queue(txq); + u64_stats_update_begin(&tx_ring->syncp); + tx_ring->tx_stats.queue_wakeup++; + u64_stats_update_end(&tx_ring->syncp); + } + } + + if (netif_xmit_stopped(txq) || !skb->xmit_more) { + /* trigger the dma engine */ + ena_com_write_sq_doorbell(tx_ring->ena_com_io_sq); + u64_stats_update_begin(&tx_ring->syncp); + tx_ring->tx_stats.doorbells++; + u64_stats_update_end(&tx_ring->syncp); + } + + return NETDEV_TX_OK; + +error_report_dma_error: + u64_stats_update_begin(&tx_ring->syncp); + tx_ring->tx_stats.dma_mapping_err++; + u64_stats_update_end(&tx_ring->syncp); + netdev_warn(adapter->netdev, "failed to map skb\n"); + + tx_info->skb = NULL; + +error_unmap_dma: + if (i >= 0) { + /* save value of frag that failed */ + last_frag = i; + + /* start back at beginning and unmap skb */ + tx_info->skb = NULL; + ena_buf = tx_info->bufs; + dma_unmap_single(tx_ring->dev, dma_unmap_addr(ena_buf, paddr), + dma_unmap_len(ena_buf, len), DMA_TO_DEVICE); + + /* unmap remaining mapped pages */ + for (i = 0; i < last_frag; i++) { + ena_buf++; + dma_unmap_page(tx_ring->dev, dma_unmap_addr(ena_buf, paddr), + dma_unmap_len(ena_buf, len), DMA_TO_DEVICE); + } + } + +error_drop_packet: + + dev_kfree_skb(skb); + return NETDEV_TX_OK; +} + +#ifdef CONFIG_NET_POLL_CONTROLLER +static void ena_netpoll(struct net_device *netdev) +{ + struct ena_adapter *adapter = netdev_priv(netdev); + int i; + + for (i = 0; i < adapter->num_queues; i++) + napi_schedule(&adapter->ena_napi[i].napi); +} +#endif /* CONFIG_NET_POLL_CONTROLLER */ + +static u16 ena_select_queue(struct net_device *dev, struct sk_buff *skb, + void *accel_priv, select_queue_fallback_t fallback) +{ + u16 qid; + /* we suspect that this is good for in--kernel network services that + * want to loop incoming skb rx to tx in normal user generated traffic, + * most probably we will not get to this + */ + if (skb_rx_queue_recorded(skb)) + qid = skb_get_rx_queue(skb); + else + qid = fallback(dev, skb); + + return qid; +} + +static void ena_config_host_info(struct ena_com_dev *ena_dev) +{ + struct ena_admin_host_info *host_info; + int rc; + + /* Allocate only the host info */ + rc = ena_com_allocate_host_info(ena_dev); + if (rc) { + pr_err("Cannot allocate host info\n"); + return; + } + + host_info = ena_dev->host_attr.host_info; + + host_info->os_type = ENA_ADMIN_OS_LINUX; + host_info->kernel_ver = LINUX_VERSION_CODE; + strncpy(host_info->kernel_ver_str, utsname()->version, + sizeof(host_info->kernel_ver_str) - 1); + host_info->os_dist = 0; + strncpy(host_info->os_dist_str, utsname()->release, + sizeof(host_info->os_dist_str) - 1); + host_info->driver_version = + (DRV_MODULE_VER_MAJOR) | + (DRV_MODULE_VER_MINOR << ENA_ADMIN_HOST_INFO_MINOR_SHIFT) | + (DRV_MODULE_VER_SUBMINOR << ENA_ADMIN_HOST_INFO_SUB_MINOR_SHIFT); + + rc = ena_com_set_host_attributes(ena_dev); + if (rc) { + if (rc == -EPERM) + pr_warn("Cannot set host attributes\n"); + else + pr_err("Cannot set host attributes\n"); + + goto err; + } + + return; + +err: + ena_com_delete_host_info(ena_dev); +} + +static void ena_config_debug_area(struct ena_adapter *adapter) +{ + u32 debug_area_size; + int rc, ss_count; + + ss_count = ena_get_sset_count(adapter->netdev, ETH_SS_STATS); + if (ss_count <= 0) { + netif_err(adapter, drv, adapter->netdev, + "SS count is negative\n"); + return; + } + + /* allocate 32 bytes for each string and 64bit for the value */ + debug_area_size = ss_count * ETH_GSTRING_LEN + sizeof(u64) * ss_count; + + rc = ena_com_allocate_debug_area(adapter->ena_dev, debug_area_size); + if (rc) { + pr_err("Cannot allocate debug area\n"); + return; + } + + rc = ena_com_set_host_attributes(adapter->ena_dev); + if (rc) { + if (rc == -EPERM) + netif_warn(adapter, drv, adapter->netdev, + "Cannot set host attributes\n"); + else + netif_err(adapter, drv, adapter->netdev, + "Cannot set host attributes\n"); + goto err; + } + + return; +err: + ena_com_delete_debug_area(adapter->ena_dev); +} + +static struct rtnl_link_stats64 *ena_get_stats64(struct net_device *netdev, + struct rtnl_link_stats64 *stats) +{ + struct ena_adapter *adapter = netdev_priv(netdev); + struct ena_admin_basic_stats ena_stats; + int rc; + + if (!test_bit(ENA_FLAG_DEV_UP, &adapter->flags)) + return NULL; + + rc = ena_com_get_dev_basic_stats(adapter->ena_dev, &ena_stats); + if (rc) + return NULL; + + stats->tx_bytes = ((u64)ena_stats.tx_bytes_high << 32) | + ena_stats.tx_bytes_low; + stats->rx_bytes = ((u64)ena_stats.rx_bytes_high << 32) | + ena_stats.rx_bytes_low; + + stats->rx_packets = ((u64)ena_stats.rx_pkts_high << 32) | + ena_stats.rx_pkts_low; + stats->tx_packets = ((u64)ena_stats.tx_pkts_high << 32) | + ena_stats.tx_pkts_low; + + stats->rx_dropped = ((u64)ena_stats.rx_drops_high << 32) | + ena_stats.rx_drops_low; + + stats->multicast = 0; + stats->collisions = 0; + + stats->rx_length_errors = 0; + stats->rx_crc_errors = 0; + stats->rx_frame_errors = 0; + stats->rx_fifo_errors = 0; + stats->rx_missed_errors = 0; + stats->tx_window_errors = 0; + + stats->rx_errors = 0; + stats->tx_errors = 0; + + return stats; +} + +static const struct net_device_ops ena_netdev_ops = { + .ndo_open = ena_open, + .ndo_stop = ena_close, + .ndo_start_xmit = ena_start_xmit, + .ndo_select_queue = ena_select_queue, + .ndo_get_stats64 = ena_get_stats64, + .ndo_tx_timeout = ena_tx_timeout, + .ndo_change_mtu = ena_change_mtu, + .ndo_set_mac_address = NULL, + .ndo_validate_addr = eth_validate_addr, +#ifdef CONFIG_NET_POLL_CONTROLLER + .ndo_poll_controller = ena_netpoll, +#endif /* CONFIG_NET_POLL_CONTROLLER */ +}; + +static void ena_device_io_suspend(struct work_struct *work) +{ + struct ena_adapter *adapter = + container_of(work, struct ena_adapter, suspend_io_task); + struct net_device *netdev = adapter->netdev; + + /* ena_napi_disable_all disables only the IO handling. + * We are still subject to AENQ keep alive watchdog. + */ + u64_stats_update_begin(&adapter->syncp); + adapter->dev_stats.io_suspend++; + u64_stats_update_begin(&adapter->syncp); + ena_napi_disable_all(adapter); + netif_tx_lock(netdev); + netif_device_detach(netdev); + netif_tx_unlock(netdev); +} + +static void ena_device_io_resume(struct work_struct *work) +{ + struct ena_adapter *adapter = + container_of(work, struct ena_adapter, resume_io_task); + struct net_device *netdev = adapter->netdev; + + u64_stats_update_begin(&adapter->syncp); + adapter->dev_stats.io_resume++; + u64_stats_update_end(&adapter->syncp); + + netif_device_attach(netdev); + ena_napi_enable_all(adapter); +} + +static int ena_device_validate_params(struct ena_adapter *adapter, + struct ena_com_dev_get_features_ctx *get_feat_ctx) +{ + struct net_device *netdev = adapter->netdev; + int rc; + + rc = ether_addr_equal(get_feat_ctx->dev_attr.mac_addr, + adapter->mac_addr); + if (!rc) { + netif_err(adapter, drv, netdev, + "Error, mac address are different\n"); + return -EINVAL; + } + + if ((get_feat_ctx->max_queues.max_cq_num < adapter->num_queues) || + (get_feat_ctx->max_queues.max_sq_num < adapter->num_queues)) { + netif_err(adapter, drv, netdev, + "Error, device doesn't support enough queues\n"); + return -EINVAL; + } + + if (get_feat_ctx->dev_attr.max_mtu < netdev->mtu) { + netif_err(adapter, drv, netdev, + "Error, device max mtu is smaller than netdev MTU\n"); + return -EINVAL; + } + + return 0; +} + +static int ena_device_init(struct ena_com_dev *ena_dev, struct pci_dev *pdev, + struct ena_com_dev_get_features_ctx *get_feat_ctx, + bool *wd_state) +{ + struct device *dev = &pdev->dev; + bool readless_supported; + u32 aenq_groups; + int dma_width; + int rc; + + rc = ena_com_mmio_reg_read_request_init(ena_dev); + if (rc) { + dev_err(dev, "failed to init mmio read less\n"); + return rc; + } + + /* The PCIe configuration space revision id indicate if mmio reg + * read is disabled + */ + readless_supported = !(pdev->revision & ENA_MMIO_DISABLE_REG_READ); + ena_com_set_mmio_read_mode(ena_dev, readless_supported); + + rc = ena_com_dev_reset(ena_dev); + if (rc) { + dev_err(dev, "Can not reset device\n"); + goto err_mmio_read_less; + } + + rc = ena_com_validate_version(ena_dev); + if (rc) { + dev_err(dev, "device version is too low\n"); + goto err_mmio_read_less; + } + + dma_width = ena_com_get_dma_width(ena_dev); + if (dma_width < 0) { + dev_err(dev, "Invalid dma width value %d", dma_width); + rc = dma_width; + goto err_mmio_read_less; + } + + rc = pci_set_dma_mask(pdev, DMA_BIT_MASK(dma_width)); + if (rc) { + dev_err(dev, "pci_set_dma_mask failed 0x%x\n", rc); + goto err_mmio_read_less; + } + + rc = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(dma_width)); + if (rc) { + dev_err(dev, "err_pci_set_consistent_dma_mask failed 0x%x\n", + rc); + goto err_mmio_read_less; + } + + /* ENA admin level init */ + rc = ena_com_admin_init(ena_dev, &aenq_handlers, true); + if (rc) { + dev_err(dev, + "Can not initialize ena admin queue with device\n"); + goto err_mmio_read_less; + } + + /* To enable the msix interrupts the driver needs to know the number + * of queues. So the driver uses polling mode to retrieve this + * information + */ + ena_com_set_admin_polling_mode(ena_dev, true); + + /* Get Device Attributes*/ + rc = ena_com_get_dev_attr_feat(ena_dev, get_feat_ctx); + if (rc) { + dev_err(dev, "Cannot get attribute for ena device rc=%d\n", rc); + goto err_admin_init; + } + + /* Try to turn all the available aenq groups */ + aenq_groups = BIT(ENA_ADMIN_LINK_CHANGE) | + BIT(ENA_ADMIN_FATAL_ERROR) | + BIT(ENA_ADMIN_WARNING) | + BIT(ENA_ADMIN_NOTIFICATION) | + BIT(ENA_ADMIN_KEEP_ALIVE); + + aenq_groups &= get_feat_ctx->aenq.supported_groups; + + rc = ena_com_set_aenq_config(ena_dev, aenq_groups); + if (rc) { + dev_err(dev, "Cannot configure aenq groups rc= %d\n", rc); + goto err_admin_init; + } + + *wd_state = !!(aenq_groups & BIT(ENA_ADMIN_KEEP_ALIVE)); + + ena_config_host_info(ena_dev); + + return 0; + +err_admin_init: + ena_com_admin_destroy(ena_dev); +err_mmio_read_less: + ena_com_mmio_reg_read_request_destroy(ena_dev); + + return rc; +} + +static int ena_enable_msix_and_set_admin_interrupts(struct ena_adapter *adapter, + int io_vectors) +{ + struct ena_com_dev *ena_dev = adapter->ena_dev; + struct device *dev = &adapter->pdev->dev; + int rc; + + rc = ena_enable_msix(adapter, io_vectors); + if (rc) { + dev_err(dev, "Can not reserve msix vectors\n"); + return rc; + } + + ena_setup_mgmnt_intr(adapter); + + rc = ena_request_mgmnt_irq(adapter); + if (rc) { + dev_err(dev, "Can not setup management interrupts\n"); + goto err_disable_msix; + } + + ena_com_set_admin_polling_mode(ena_dev, false); + + ena_com_admin_aenq_enable(ena_dev); + + return 0; + +err_disable_msix: + ena_disable_msix(adapter); + + return rc; +} + +static void ena_fw_reset_device(struct work_struct *work) +{ + struct ena_com_dev_get_features_ctx get_feat_ctx; + struct ena_adapter *adapter = + container_of(work, struct ena_adapter, reset_task); + struct net_device *netdev = adapter->netdev; + struct ena_com_dev *ena_dev = adapter->ena_dev; + struct pci_dev *pdev = adapter->pdev; + bool dev_up, wd_state; + int rc; + + del_timer_sync(&adapter->timer_service); + + rtnl_lock(); + + dev_up = test_bit(ENA_FLAG_DEV_UP, &adapter->flags); + ena_com_set_admin_running_state(ena_dev, false); + + /* After calling ena_close the tx queues and the napi + * are disabled so no one can interfere or touch the + * data structures + */ + ena_close(netdev); + + rc = ena_com_dev_reset(ena_dev); + if (rc) { + dev_err(&pdev->dev, "Device reset failed\n"); + goto err; + } + + ena_free_mgmnt_irq(adapter); + + ena_disable_msix(adapter); + + ena_com_abort_admin_commands(ena_dev); + + ena_com_wait_for_abort_completion(ena_dev); + + ena_com_admin_destroy(ena_dev); + + ena_com_mmio_reg_read_request_destroy(ena_dev); + + /* Finish with the destroy part. Start the init part */ + + rc = ena_device_init(ena_dev, adapter->pdev, &get_feat_ctx, &wd_state); + if (rc) { + dev_err(&pdev->dev, "Can not initialize device\n"); + goto err; + } + adapter->wd_state = wd_state; + + rc = ena_device_validate_params(adapter, &get_feat_ctx); + if (rc) { + dev_err(&pdev->dev, "Validation of device parameters failed\n"); + goto err_device_destroy; + } + + rc = ena_enable_msix_and_set_admin_interrupts(adapter, + adapter->num_queues); + if (rc) { + dev_err(&pdev->dev, "Enable MSI-X failed\n"); + goto err_device_destroy; + } + /* If the interface was up before the reset bring it up */ + if (dev_up) { + rc = ena_up(adapter); + if (rc) { + dev_err(&pdev->dev, "Failed to create I/O queues\n"); + goto err_disable_msix; + } + } + + mod_timer(&adapter->timer_service, round_jiffies(jiffies + HZ)); + + rtnl_unlock(); + + dev_err(&pdev->dev, "Device reset completed successfully\n"); + + return; +err_disable_msix: + ena_free_mgmnt_irq(adapter); + ena_disable_msix(adapter); +err_device_destroy: + ena_com_admin_destroy(ena_dev); +err: + rtnl_unlock(); + + dev_err(&pdev->dev, + "Reset attempt failed. Can not reset the device\n"); +} + +static void check_for_missing_tx_completions(struct ena_adapter *adapter) +{ + struct ena_tx_buffer *tx_buf; + unsigned long last_jiffies; + struct ena_ring *tx_ring; + int i, j, budget; + u32 missed_tx; + + /* Make sure the driver doesn't turn the device in other process */ + smp_rmb(); + + if (!test_bit(ENA_FLAG_DEV_UP, &adapter->flags)) + return; + + budget = ENA_MONITORED_TX_QUEUES; + + for (i = adapter->last_monitored_tx_qid; i < adapter->num_queues; i++) { + tx_ring = &adapter->tx_ring[i]; + + for (j = 0; j < tx_ring->ring_size; j++) { + tx_buf = &tx_ring->tx_buffer_info[j]; + last_jiffies = tx_buf->last_jiffies; + if (unlikely(last_jiffies && time_is_before_jiffies(last_jiffies + TX_TIMEOUT))) { + netif_notice(adapter, tx_err, adapter->netdev, + "Found a Tx that wasn't completed on time, qid %d, index %d.\n", + tx_ring->qid, j); + + u64_stats_update_begin(&tx_ring->syncp); + missed_tx = tx_ring->tx_stats.missing_tx_comp++; + u64_stats_update_end(&tx_ring->syncp); + + /* Clear last jiffies so the lost buffer won't + * be counted twice. + */ + tx_buf->last_jiffies = 0; + + if (unlikely(missed_tx > MAX_NUM_OF_TIMEOUTED_PACKETS)) { + netif_err(adapter, tx_err, adapter->netdev, + "The number of lost tx completion is above the threshold (%d > %d). Reset the device\n", + missed_tx, MAX_NUM_OF_TIMEOUTED_PACKETS); + set_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags); + } + } + } + + budget--; + if (!budget) + break; + } + + adapter->last_monitored_tx_qid = i % adapter->num_queues; +} + +/* Check for keep alive expiration */ +static void check_for_missing_keep_alive(struct ena_adapter *adapter) +{ + unsigned long keep_alive_expired; + + if (!adapter->wd_state) + return; + + keep_alive_expired = round_jiffies(adapter->last_keep_alive_jiffies + + ENA_DEVICE_KALIVE_TIMEOUT); + if (unlikely(time_is_before_jiffies(keep_alive_expired))) { + netif_err(adapter, drv, adapter->netdev, + "Keep alive watchdog timeout.\n"); + u64_stats_update_begin(&adapter->syncp); + adapter->dev_stats.wd_expired++; + u64_stats_update_end(&adapter->syncp); + set_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags); + } +} + +static void check_for_admin_com_state(struct ena_adapter *adapter) +{ + if (unlikely(!ena_com_get_admin_running_state(adapter->ena_dev))) { + netif_err(adapter, drv, adapter->netdev, + "ENA admin queue is not in running state!\n"); + u64_stats_update_begin(&adapter->syncp); + adapter->dev_stats.admin_q_pause++; + u64_stats_update_end(&adapter->syncp); + set_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags); + } +} + +static void ena_update_host_info(struct ena_admin_host_info *host_info, + struct net_device *netdev) +{ + host_info->supported_network_features[0] = + netdev->features & GENMASK_ULL(31, 0); + host_info->supported_network_features[1] = + (netdev->features & GENMASK_ULL(63, 32)) >> 32; +} + +static void ena_timer_service(unsigned long data) +{ + struct ena_adapter *adapter = (struct ena_adapter *)data; + u8 *debug_area = adapter->ena_dev->host_attr.debug_area_virt_addr; + struct ena_admin_host_info *host_info = + adapter->ena_dev->host_attr.host_info; + + check_for_missing_keep_alive(adapter); + + check_for_admin_com_state(adapter); + + check_for_missing_tx_completions(adapter); + + if (debug_area) + ena_dump_stats_to_buf(adapter, debug_area); + + if (host_info) + ena_update_host_info(host_info, adapter->netdev); + + if (unlikely(test_and_clear_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags))) { + netif_err(adapter, drv, adapter->netdev, + "Trigger reset is on\n"); + ena_dump_stats_to_dmesg(adapter); + queue_work(ena_wq, &adapter->reset_task); + return; + } + + /* Reset the timer */ + mod_timer(&adapter->timer_service, jiffies + HZ); +} + +static int ena_calc_io_queue_num(struct pci_dev *pdev, + struct ena_com_dev *ena_dev, + struct ena_com_dev_get_features_ctx *get_feat_ctx) +{ + int io_sq_num, io_queue_num; + + /* In case of LLQ use the llq number in the get feature cmd */ + if (ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV) { + io_sq_num = get_feat_ctx->max_queues.max_llq_num; + + if (io_sq_num == 0) { + dev_err(&pdev->dev, + "Trying to use LLQ but llq_num is 0. Fall back into regular queues\n"); + + ena_dev->tx_mem_queue_type = + ENA_ADMIN_PLACEMENT_POLICY_HOST; + io_sq_num = get_feat_ctx->max_queues.max_sq_num; + } + } else { + io_sq_num = get_feat_ctx->max_queues.max_sq_num; + } + + io_queue_num = min_t(int, num_possible_cpus(), ENA_MAX_NUM_IO_QUEUES); + io_queue_num = min_t(int, io_queue_num, io_sq_num); + io_queue_num = min_t(int, io_queue_num, + get_feat_ctx->max_queues.max_cq_num); + /* 1 IRQ for for mgmnt and 1 IRQs for each IO direction */ + io_queue_num = min_t(int, io_queue_num, pci_msix_vec_count(pdev) - 1); + if (unlikely(!io_queue_num)) { + dev_err(&pdev->dev, "The device doesn't have io queues\n"); + return -EFAULT; + } + + return io_queue_num; +} + +static int ena_set_push_mode(struct pci_dev *pdev, struct ena_com_dev *ena_dev, + struct ena_com_dev_get_features_ctx *get_feat_ctx) +{ + bool has_mem_bar; + + has_mem_bar = pci_select_bars(pdev, IORESOURCE_MEM) & BIT(ENA_MEM_BAR); + + /* Enable push mode if device supports LLQ */ + if (has_mem_bar && (get_feat_ctx->max_queues.max_llq_num > 0)) + ena_dev->tx_mem_queue_type = ENA_ADMIN_PLACEMENT_POLICY_DEV; + else + ena_dev->tx_mem_queue_type = ENA_ADMIN_PLACEMENT_POLICY_HOST; + + return 0; +} + +static void ena_set_dev_offloads(struct ena_com_dev_get_features_ctx *feat, + struct net_device *netdev) +{ + netdev_features_t dev_features = 0; + + /* Set offload features */ + if (feat->offload.tx & + ENA_ADMIN_FEATURE_OFFLOAD_DESC_TX_L4_IPV4_CSUM_PART_MASK) + dev_features |= NETIF_F_IP_CSUM; + + if (feat->offload.tx & + ENA_ADMIN_FEATURE_OFFLOAD_DESC_TX_L4_IPV6_CSUM_PART_MASK) + dev_features |= NETIF_F_IPV6_CSUM; + + if (feat->offload.tx & ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_IPV4_MASK) + dev_features |= NETIF_F_TSO; + + if (feat->offload.tx & ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_IPV6_MASK) + dev_features |= NETIF_F_TSO6; + + if (feat->offload.tx & ENA_ADMIN_FEATURE_OFFLOAD_DESC_TSO_ECN_MASK) + dev_features |= NETIF_F_TSO_ECN; + + if (feat->offload.rx_supported & + ENA_ADMIN_FEATURE_OFFLOAD_DESC_RX_L4_IPV4_CSUM_MASK) + dev_features |= NETIF_F_RXCSUM; + + if (feat->offload.rx_supported & + ENA_ADMIN_FEATURE_OFFLOAD_DESC_RX_L4_IPV6_CSUM_MASK) + dev_features |= NETIF_F_RXCSUM; + + netdev->features = + dev_features | + NETIF_F_SG | + NETIF_F_NTUPLE | + NETIF_F_RXHASH | + NETIF_F_HIGHDMA; + + netdev->hw_features |= netdev->features; + netdev->vlan_features |= netdev->features; +} + +static void ena_set_conf_feat_params(struct ena_adapter *adapter, + struct ena_com_dev_get_features_ctx *feat) +{ + struct net_device *netdev = adapter->netdev; + + /* Copy mac address */ + if (!is_valid_ether_addr(feat->dev_attr.mac_addr)) { + eth_hw_addr_random(netdev); + ether_addr_copy(adapter->mac_addr, netdev->dev_addr); + } else { + ether_addr_copy(adapter->mac_addr, feat->dev_attr.mac_addr); + ether_addr_copy(netdev->dev_addr, adapter->mac_addr); + } + + /* Set offload features */ + ena_set_dev_offloads(feat, netdev); + + adapter->max_mtu = feat->dev_attr.max_mtu; +} + +static int ena_rss_init_default(struct ena_adapter *adapter) +{ + struct ena_com_dev *ena_dev = adapter->ena_dev; + struct device *dev = &adapter->pdev->dev; + int rc, i; + u32 val; + + rc = ena_com_rss_init(ena_dev, ENA_RX_RSS_TABLE_LOG_SIZE); + if (unlikely(rc)) { + dev_err(dev, "Cannot init indirect table\n"); + goto err_rss_init; + } + + for (i = 0; i < ENA_RX_RSS_TABLE_SIZE; i++) { + val = ethtool_rxfh_indir_default(i, adapter->num_queues); + rc = ena_com_indirect_table_fill_entry(ena_dev, i, + ENA_IO_RXQ_IDX(val)); + if (unlikely(rc && (rc != -EPERM))) { + dev_err(dev, "Cannot fill indirect table\n"); + goto err_fill_indir; + } + } + + rc = ena_com_fill_hash_function(ena_dev, ENA_ADMIN_CRC32, NULL, + ENA_HASH_KEY_SIZE, 0xFFFFFFFF); + if (unlikely(rc && (rc != -EPERM))) { + dev_err(dev, "Cannot fill hash function\n"); + goto err_fill_indir; + } + + rc = ena_com_set_default_hash_ctrl(ena_dev); + if (unlikely(rc && (rc != -EPERM))) { + dev_err(dev, "Cannot fill hash control\n"); + goto err_fill_indir; + } + + return 0; + +err_fill_indir: + ena_com_rss_destroy(ena_dev); +err_rss_init: + + return rc; +} + +static void ena_release_bars(struct ena_com_dev *ena_dev, struct pci_dev *pdev) +{ + int release_bars; + + release_bars = pci_select_bars(pdev, IORESOURCE_MEM) & ENA_BAR_MASK; + pci_release_selected_regions(pdev, release_bars); +} + +static int ena_calc_queue_size(struct pci_dev *pdev, + struct ena_com_dev *ena_dev, + u16 *max_tx_sgl_size, + u16 *max_rx_sgl_size, + struct ena_com_dev_get_features_ctx *get_feat_ctx) +{ + u32 queue_size = ENA_DEFAULT_RING_SIZE; + + queue_size = min_t(u32, queue_size, + get_feat_ctx->max_queues.max_cq_depth); + queue_size = min_t(u32, queue_size, + get_feat_ctx->max_queues.max_sq_depth); + + if (ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV) + queue_size = min_t(u32, queue_size, + get_feat_ctx->max_queues.max_llq_depth); + + queue_size = rounddown_pow_of_two(queue_size); + + if (unlikely(!queue_size)) { + dev_err(&pdev->dev, "Invalid queue size\n"); + return -EFAULT; + } + + *max_tx_sgl_size = min_t(u16, ENA_PKT_MAX_BUFS, + get_feat_ctx->max_queues.max_packet_tx_descs); + *max_rx_sgl_size = min_t(u16, ENA_PKT_MAX_BUFS, + get_feat_ctx->max_queues.max_packet_rx_descs); + + return queue_size; +} + +/* ena_probe - Device Initialization Routine + * @pdev: PCI device information struct + * @ent: entry in ena_pci_tbl + * + * Returns 0 on success, negative on failure + * + * ena_probe initializes an adapter identified by a pci_dev structure. + * The OS initialization, configuring of the adapter private structure, + * and a hardware reset occur. + */ +static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent) +{ + struct ena_com_dev_get_features_ctx get_feat_ctx; + static int version_printed; + struct net_device *netdev; + struct ena_adapter *adapter; + struct ena_com_dev *ena_dev = NULL; + static int adapters_found; + int io_queue_num, bars, rc; + int queue_size; + u16 tx_sgl_size = 0; + u16 rx_sgl_size = 0; + bool wd_state; + + dev_dbg(&pdev->dev, "%s\n", __func__); + + if (version_printed++ == 0) + dev_info(&pdev->dev, "%s", version); + + rc = pci_enable_device_mem(pdev); + if (rc) { + dev_err(&pdev->dev, "pci_enable_device_mem() failed!\n"); + return rc; + } + + pci_set_master(pdev); + + ena_dev = vzalloc(sizeof(*ena_dev)); + if (!ena_dev) { + rc = -ENOMEM; + goto err_disable_device; + } + + bars = pci_select_bars(pdev, IORESOURCE_MEM) & ENA_BAR_MASK; + rc = pci_request_selected_regions(pdev, bars, DRV_MODULE_NAME); + if (rc) { + dev_err(&pdev->dev, "pci_request_selected_regions failed %d\n", + rc); + goto err_free_ena_dev; + } + + ena_dev->reg_bar = ioremap(pci_resource_start(pdev, ENA_REG_BAR), + pci_resource_len(pdev, ENA_REG_BAR)); + if (!ena_dev->reg_bar) { + dev_err(&pdev->dev, "failed to remap regs bar\n"); + rc = -EFAULT; + goto err_free_region; + } + + ena_dev->dmadev = &pdev->dev; + + rc = ena_device_init(ena_dev, pdev, &get_feat_ctx, &wd_state); + if (rc) { + dev_err(&pdev->dev, "ena device init failed\n"); + if (rc == -ETIME) + rc = -EPROBE_DEFER; + goto err_free_region; + } + + rc = ena_set_push_mode(pdev, ena_dev, &get_feat_ctx); + if (rc) { + dev_err(&pdev->dev, "Invalid module param(push_mode)\n"); + goto err_device_destroy; + } + + if (ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_DEV) { + ena_dev->mem_bar = ioremap_wc(pci_resource_start(pdev, ENA_MEM_BAR), + pci_resource_len(pdev, ENA_MEM_BAR)); + if (!ena_dev->mem_bar) { + rc = -EFAULT; + goto err_device_destroy; + } + } + + /* initial Tx interrupt delay, Assumes 1 usec granularity. + * Updated during device initialization with the real granularity + */ + ena_dev->intr_moder_tx_interval = ENA_INTR_INITIAL_TX_INTERVAL_USECS; + io_queue_num = ena_calc_io_queue_num(pdev, ena_dev, &get_feat_ctx); + queue_size = ena_calc_queue_size(pdev, ena_dev, &tx_sgl_size, + &rx_sgl_size, &get_feat_ctx); + if ((queue_size <= 0) || (io_queue_num <= 0)) { + rc = -EFAULT; + goto err_device_destroy; + } + + dev_info(&pdev->dev, "creating %d io queues. queue size: %d\n", + io_queue_num, queue_size); + + /* dev zeroed in init_etherdev */ + netdev = alloc_etherdev_mq(sizeof(struct ena_adapter), io_queue_num); + if (!netdev) { + dev_err(&pdev->dev, "alloc_etherdev_mq failed\n"); + rc = -ENOMEM; + goto err_device_destroy; + } + + SET_NETDEV_DEV(netdev, &pdev->dev); + + adapter = netdev_priv(netdev); + pci_set_drvdata(pdev, adapter); + + adapter->ena_dev = ena_dev; + adapter->netdev = netdev; + adapter->pdev = pdev; + + ena_set_conf_feat_params(adapter, &get_feat_ctx); + + adapter->msg_enable = netif_msg_init(debug, DEFAULT_MSG_ENABLE); + + adapter->tx_ring_size = queue_size; + adapter->rx_ring_size = queue_size; + + adapter->max_tx_sgl_size = tx_sgl_size; + adapter->max_rx_sgl_size = rx_sgl_size; + + adapter->num_queues = io_queue_num; + adapter->last_monitored_tx_qid = 0; + + adapter->rx_copybreak = ENA_DEFAULT_RX_COPYBREAK; + adapter->wd_state = wd_state; + + snprintf(adapter->name, ENA_NAME_MAX_LEN, "ena_%d", adapters_found); + + rc = ena_com_init_interrupt_moderation(adapter->ena_dev); + if (rc) { + dev_err(&pdev->dev, + "Failed to query interrupt moderation feature\n"); + goto err_netdev_destroy; + } + ena_init_io_rings(adapter); + + netdev->netdev_ops = &ena_netdev_ops; + netdev->watchdog_timeo = TX_TIMEOUT; + ena_set_ethtool_ops(netdev); + + netdev->priv_flags |= IFF_UNICAST_FLT; + + u64_stats_init(&adapter->syncp); + + rc = ena_enable_msix_and_set_admin_interrupts(adapter, io_queue_num); + if (rc) { + dev_err(&pdev->dev, + "Failed to enable and set the admin interrupts\n"); + goto err_worker_destroy; + } + rc = ena_rss_init_default(adapter); + if (rc && (rc != -EPERM)) { + dev_err(&pdev->dev, "Cannot init RSS rc: %d\n", rc); + goto err_free_msix; + } + + ena_config_debug_area(adapter); + + memcpy(adapter->netdev->perm_addr, adapter->mac_addr, netdev->addr_len); + + netif_carrier_off(netdev); + + rc = register_netdev(netdev); + if (rc) { + dev_err(&pdev->dev, "Cannot register net device\n"); + goto err_rss; + } + + INIT_WORK(&adapter->suspend_io_task, ena_device_io_suspend); + INIT_WORK(&adapter->resume_io_task, ena_device_io_resume); + INIT_WORK(&adapter->reset_task, ena_fw_reset_device); + + adapter->last_keep_alive_jiffies = jiffies; + + init_timer(&adapter->timer_service); + adapter->timer_service.expires = round_jiffies(jiffies + HZ); + adapter->timer_service.function = ena_timer_service; + adapter->timer_service.data = (unsigned long)adapter; + + add_timer(&adapter->timer_service); + + dev_info(&pdev->dev, "%s found at mem %lx, mac addr %pM Queues %d\n", + DEVICE_NAME, (long)pci_resource_start(pdev, 0), + netdev->dev_addr, io_queue_num); + + set_bit(ENA_FLAG_DEVICE_RUNNING, &adapter->flags); + + adapters_found++; + + return 0; + +err_rss: + ena_com_delete_debug_area(ena_dev); + ena_com_rss_destroy(ena_dev); +err_free_msix: + ena_com_dev_reset(ena_dev); + ena_free_mgmnt_irq(adapter); + ena_disable_msix(adapter); +err_worker_destroy: + ena_com_destroy_interrupt_moderation(ena_dev); + del_timer(&adapter->timer_service); + cancel_work_sync(&adapter->suspend_io_task); + cancel_work_sync(&adapter->resume_io_task); +err_netdev_destroy: + free_netdev(netdev); +err_device_destroy: + ena_com_delete_host_info(ena_dev); + ena_com_admin_destroy(ena_dev); +err_free_region: + ena_release_bars(ena_dev, pdev); +err_free_ena_dev: + vfree(ena_dev); +err_disable_device: + pci_disable_device(pdev); + return rc; +} + +/*****************************************************************************/ +static int ena_sriov_configure(struct pci_dev *dev, int numvfs) +{ + int rc; + + if (numvfs > 0) { + rc = pci_enable_sriov(dev, numvfs); + if (rc != 0) { + dev_err(&dev->dev, + "pci_enable_sriov failed to enable: %d vfs with the error: %d\n", + numvfs, rc); + return rc; + } + + return numvfs; + } + + if (numvfs == 0) { + pci_disable_sriov(dev); + return 0; + } + + return -EINVAL; +} + +/*****************************************************************************/ +/*****************************************************************************/ + +/* ena_remove - Device Removal Routine + * @pdev: PCI device information struct + * + * ena_remove is called by the PCI subsystem to alert the driver + * that it should release a PCI device. + */ +static void ena_remove(struct pci_dev *pdev) +{ + struct ena_adapter *adapter = pci_get_drvdata(pdev); + struct ena_com_dev *ena_dev; + struct net_device *netdev; + + if (!adapter) + /* This device didn't load properly and it's resources + * already released, nothing to do + */ + return; + + ena_dev = adapter->ena_dev; + netdev = adapter->netdev; + +#ifdef CONFIG_RFS_ACCEL + if ((adapter->msix_vecs >= 1) && (netdev->rx_cpu_rmap)) { + free_irq_cpu_rmap(netdev->rx_cpu_rmap); + netdev->rx_cpu_rmap = NULL; + } +#endif /* CONFIG_RFS_ACCEL */ + + unregister_netdev(netdev); + del_timer_sync(&adapter->timer_service); + + cancel_work_sync(&adapter->reset_task); + + cancel_work_sync(&adapter->suspend_io_task); + + cancel_work_sync(&adapter->resume_io_task); + + ena_com_dev_reset(ena_dev); + + ena_free_mgmnt_irq(adapter); + + ena_disable_msix(adapter); + + free_netdev(netdev); + + ena_com_mmio_reg_read_request_destroy(ena_dev); + + ena_com_abort_admin_commands(ena_dev); + + ena_com_wait_for_abort_completion(ena_dev); + + ena_com_admin_destroy(ena_dev); + + ena_com_rss_destroy(ena_dev); + + ena_com_delete_debug_area(ena_dev); + + ena_com_delete_host_info(ena_dev); + + ena_release_bars(ena_dev, pdev); + + pci_disable_device(pdev); + + ena_com_destroy_interrupt_moderation(ena_dev); + + vfree(ena_dev); +} + +static struct pci_driver ena_pci_driver = { + .name = DRV_MODULE_NAME, + .id_table = ena_pci_tbl, + .probe = ena_probe, + .remove = ena_remove, + .sriov_configure = ena_sriov_configure, +}; + +static int __init ena_init(void) +{ + pr_info("%s", version); + + ena_wq = create_singlethread_workqueue(DRV_MODULE_NAME); + if (!ena_wq) { + pr_err("Failed to create workqueue\n"); + return -ENOMEM; + } + + return pci_register_driver(&ena_pci_driver); +} + +static void __exit ena_cleanup(void) +{ + pci_unregister_driver(&ena_pci_driver); + + if (ena_wq) { + destroy_workqueue(ena_wq); + ena_wq = NULL; + } +} + +/****************************************************************************** + ******************************** AENQ Handlers ******************************* + *****************************************************************************/ +/* ena_update_on_link_change: + * Notify the network interface about the change in link status + */ +static void ena_update_on_link_change(void *adapter_data, + struct ena_admin_aenq_entry *aenq_e) +{ + struct ena_adapter *adapter = (struct ena_adapter *)adapter_data; + struct ena_admin_aenq_link_change_desc *aenq_desc = + (struct ena_admin_aenq_link_change_desc *)aenq_e; + int status = aenq_desc->flags & + ENA_ADMIN_AENQ_LINK_CHANGE_DESC_LINK_STATUS_MASK; + + if (status) { + netdev_dbg(adapter->netdev, "%s\n", __func__); + set_bit(ENA_FLAG_LINK_UP, &adapter->flags); + netif_carrier_on(adapter->netdev); + } else { + clear_bit(ENA_FLAG_LINK_UP, &adapter->flags); + netif_carrier_off(adapter->netdev); + } +} + +static void ena_keep_alive_wd(void *adapter_data, + struct ena_admin_aenq_entry *aenq_e) +{ + struct ena_adapter *adapter = (struct ena_adapter *)adapter_data; + + adapter->last_keep_alive_jiffies = jiffies; +} + +static void ena_notification(void *adapter_data, + struct ena_admin_aenq_entry *aenq_e) +{ + struct ena_adapter *adapter = (struct ena_adapter *)adapter_data; + + WARN(aenq_e->aenq_common_desc.group != ENA_ADMIN_NOTIFICATION, + "Invalid group(%x) expected %x\n", + aenq_e->aenq_common_desc.group, + ENA_ADMIN_NOTIFICATION); + + switch (aenq_e->aenq_common_desc.syndrom) { + case ENA_ADMIN_SUSPEND: + /* Suspend just the IO queues. + * We deliberately don't suspend admin so the timer and + * the keep_alive events should remain. + */ + queue_work(ena_wq, &adapter->suspend_io_task); + break; + case ENA_ADMIN_RESUME: + queue_work(ena_wq, &adapter->resume_io_task); + break; + default: + netif_err(adapter, drv, adapter->netdev, + "Invalid aenq notification link state %d\n", + aenq_e->aenq_common_desc.syndrom); + } +} + +/* This handler will called for unknown event group or unimplemented handlers*/ +static void unimplemented_aenq_handler(void *data, + struct ena_admin_aenq_entry *aenq_e) +{ + struct ena_adapter *adapter = (struct ena_adapter *)data; + + netif_err(adapter, drv, adapter->netdev, + "Unknown event was received or event with unimplemented handler\n"); +} + +static struct ena_aenq_handlers aenq_handlers = { + .handlers = { + [ENA_ADMIN_LINK_CHANGE] = ena_update_on_link_change, + [ENA_ADMIN_NOTIFICATION] = ena_notification, + [ENA_ADMIN_KEEP_ALIVE] = ena_keep_alive_wd, + }, + .unimplemented_handler = unimplemented_aenq_handler +}; + +module_init(ena_init); +module_exit(ena_cleanup); diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.h b/drivers/net/ethernet/amazon/ena/ena_netdev.h new file mode 100644 index 000000000000..69d7e9ed5bc8 --- /dev/null +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.h @@ -0,0 +1,324 @@ +/* + * Copyright 2015 Amazon.com, Inc. or its affiliates. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef ENA_H +#define ENA_H + +#include <linux/bitops.h> +#include <linux/etherdevice.h> +#include <linux/inetdevice.h> +#include <linux/interrupt.h> +#include <linux/netdevice.h> +#include <linux/skbuff.h> + +#include "ena_com.h" +#include "ena_eth_com.h" + +#define DRV_MODULE_VER_MAJOR 1 +#define DRV_MODULE_VER_MINOR 0 +#define DRV_MODULE_VER_SUBMINOR 2 + +#define DRV_MODULE_NAME "ena" +#ifndef DRV_MODULE_VERSION +#define DRV_MODULE_VERSION \ + __stringify(DRV_MODULE_VER_MAJOR) "." \ + __stringify(DRV_MODULE_VER_MINOR) "." \ + __stringify(DRV_MODULE_VER_SUBMINOR) +#endif + +#define DEVICE_NAME "Elastic Network Adapter (ENA)" + +/* 1 for AENQ + ADMIN */ +#define ENA_MAX_MSIX_VEC(io_queues) (1 + (io_queues)) + +#define ENA_REG_BAR 0 +#define ENA_MEM_BAR 2 +#define ENA_BAR_MASK (BIT(ENA_REG_BAR) | BIT(ENA_MEM_BAR)) + +#define ENA_DEFAULT_RING_SIZE (1024) + +#define ENA_TX_WAKEUP_THRESH (MAX_SKB_FRAGS + 2) +#define ENA_DEFAULT_RX_COPYBREAK (128 - NET_IP_ALIGN) + +/* limit the buffer size to 600 bytes to handle MTU changes from very + * small to very large, in which case the number of buffers per packet + * could exceed ENA_PKT_MAX_BUFS + */ +#define ENA_DEFAULT_MIN_RX_BUFF_ALLOC_SIZE 600 + +#define ENA_MIN_MTU 128 + +#define ENA_NAME_MAX_LEN 20 +#define ENA_IRQNAME_SIZE 40 + +#define ENA_PKT_MAX_BUFS 19 + +#define ENA_RX_RSS_TABLE_LOG_SIZE 7 +#define ENA_RX_RSS_TABLE_SIZE (1 << ENA_RX_RSS_TABLE_LOG_SIZE) + +#define ENA_HASH_KEY_SIZE 40 + +/* The number of tx packet completions that will be handled each NAPI poll + * cycle is ring_size / ENA_TX_POLL_BUDGET_DIVIDER. + */ +#define ENA_TX_POLL_BUDGET_DIVIDER 4 + +/* Refill Rx queue when number of available descriptors is below + * QUEUE_SIZE / ENA_RX_REFILL_THRESH_DIVIDER + */ +#define ENA_RX_REFILL_THRESH_DIVIDER 8 + +/* Number of queues to check for missing queues per timer service */ +#define ENA_MONITORED_TX_QUEUES 4 +/* Max timeout packets before device reset */ +#define MAX_NUM_OF_TIMEOUTED_PACKETS 32 + +#define ENA_TX_RING_IDX_NEXT(idx, ring_size) (((idx) + 1) & ((ring_size) - 1)) + +#define ENA_RX_RING_IDX_NEXT(idx, ring_size) (((idx) + 1) & ((ring_size) - 1)) +#define ENA_RX_RING_IDX_ADD(idx, n, ring_size) \ + (((idx) + (n)) & ((ring_size) - 1)) + +#define ENA_IO_TXQ_IDX(q) (2 * (q)) +#define ENA_IO_RXQ_IDX(q) (2 * (q) + 1) + +#define ENA_MGMNT_IRQ_IDX 0 +#define ENA_IO_IRQ_FIRST_IDX 1 +#define ENA_IO_IRQ_IDX(q) (ENA_IO_IRQ_FIRST_IDX + (q)) + +/* ENA device should send keep alive msg every 1 sec. + * We wait for 3 sec just to be on the safe side. + */ +#define ENA_DEVICE_KALIVE_TIMEOUT (3 * HZ) + +#define ENA_MMIO_DISABLE_REG_READ BIT(0) + +struct ena_irq { + irq_handler_t handler; + void *data; + int cpu; + u32 vector; + cpumask_t affinity_hint_mask; + char name[ENA_IRQNAME_SIZE]; +}; + +struct ena_napi { + struct napi_struct napi ____cacheline_aligned; + struct ena_ring *tx_ring; + struct ena_ring *rx_ring; + u32 qid; +}; + +struct ena_tx_buffer { + struct sk_buff *skb; + /* num of ena desc for this specific skb + * (includes data desc and metadata desc) + */ + u32 tx_descs; + /* num of buffers used by this skb */ + u32 num_of_bufs; + /* Save the last jiffies to detect missing tx packets */ + unsigned long last_jiffies; + struct ena_com_buf bufs[ENA_PKT_MAX_BUFS]; +} ____cacheline_aligned; + +struct ena_rx_buffer { + struct sk_buff *skb; + struct page *page; + u32 page_offset; + struct ena_com_buf ena_buf; +} ____cacheline_aligned; + +struct ena_stats_tx { + u64 cnt; + u64 bytes; + u64 queue_stop; + u64 prepare_ctx_err; + u64 queue_wakeup; + u64 dma_mapping_err; + u64 linearize; + u64 linearize_failed; + u64 napi_comp; + u64 tx_poll; + u64 doorbells; + u64 missing_tx_comp; + u64 bad_req_id; +}; + +struct ena_stats_rx { + u64 cnt; + u64 bytes; + u64 refil_partial; + u64 bad_csum; + u64 page_alloc_fail; + u64 skb_alloc_fail; + u64 dma_mapping_err; + u64 bad_desc_num; + u64 rx_copybreak_pkt; +}; + +struct ena_ring { + /* Holds the empty requests for TX out of order completions */ + u16 *free_tx_ids; + union { + struct ena_tx_buffer *tx_buffer_info; + struct ena_rx_buffer *rx_buffer_info; + }; + + /* cache ptr to avoid using the adapter */ + struct device *dev; + struct pci_dev *pdev; + struct napi_struct *napi; + struct net_device *netdev; + struct ena_com_dev *ena_dev; + struct ena_adapter *adapter; + struct ena_com_io_cq *ena_com_io_cq; + struct ena_com_io_sq *ena_com_io_sq; + + u16 next_to_use; + u16 next_to_clean; + u16 rx_copybreak; + u16 qid; + u16 mtu; + u16 sgl_size; + + /* The maximum header length the device can handle */ + u8 tx_max_header_size; + + /* cpu for TPH */ + int cpu; + /* number of tx/rx_buffer_info's entries */ + int ring_size; + + enum ena_admin_placement_policy_type tx_mem_queue_type; + + struct ena_com_rx_buf_info ena_bufs[ENA_PKT_MAX_BUFS]; + u32 smoothed_interval; + u32 per_napi_packets; + u32 per_napi_bytes; + enum ena_intr_moder_level moder_tbl_idx; + struct u64_stats_sync syncp; + union { + struct ena_stats_tx tx_stats; + struct ena_stats_rx rx_stats; + }; +} ____cacheline_aligned; + +struct ena_stats_dev { + u64 tx_timeout; + u64 io_suspend; + u64 io_resume; + u64 wd_expired; + u64 interface_up; + u64 interface_down; + u64 admin_q_pause; +}; + +enum ena_flags_t { + ENA_FLAG_DEVICE_RUNNING, + ENA_FLAG_DEV_UP, + ENA_FLAG_LINK_UP, + ENA_FLAG_MSIX_ENABLED, + ENA_FLAG_TRIGGER_RESET +}; + +/* adapter specific private data structure */ +struct ena_adapter { + struct ena_com_dev *ena_dev; + /* OS defined structs */ + struct net_device *netdev; + struct pci_dev *pdev; + + /* rx packets that shorter that this len will be copied to the skb + * header + */ + u32 rx_copybreak; + u32 max_mtu; + + int num_queues; + + struct msix_entry *msix_entries; + int msix_vecs; + + u32 tx_usecs, rx_usecs; /* interrupt moderation */ + u32 tx_frames, rx_frames; /* interrupt moderation */ + + u32 tx_ring_size; + u32 rx_ring_size; + + u32 msg_enable; + + u16 max_tx_sgl_size; + u16 max_rx_sgl_size; + + u8 mac_addr[ETH_ALEN]; + + char name[ENA_NAME_MAX_LEN]; + + unsigned long flags; + /* TX */ + struct ena_ring tx_ring[ENA_MAX_NUM_IO_QUEUES] + ____cacheline_aligned_in_smp; + + /* RX */ + struct ena_ring rx_ring[ENA_MAX_NUM_IO_QUEUES] + ____cacheline_aligned_in_smp; + + struct ena_napi ena_napi[ENA_MAX_NUM_IO_QUEUES]; + + struct ena_irq irq_tbl[ENA_MAX_MSIX_VEC(ENA_MAX_NUM_IO_QUEUES)]; + + /* timer service */ + struct work_struct reset_task; + struct work_struct suspend_io_task; + struct work_struct resume_io_task; + struct timer_list timer_service; + + bool wd_state; + unsigned long last_keep_alive_jiffies; + + struct u64_stats_sync syncp; + struct ena_stats_dev dev_stats; + + /* last queue index that was checked for uncompleted tx packets */ + u32 last_monitored_tx_qid; +}; + +void ena_set_ethtool_ops(struct net_device *netdev); + +void ena_dump_stats_to_dmesg(struct ena_adapter *adapter); + +void ena_dump_stats_to_buf(struct ena_adapter *adapter, u8 *buf); + +int ena_get_sset_count(struct net_device *netdev, int sset); + +#endif /* !(ENA_H) */ diff --git a/drivers/net/ethernet/amazon/ena/ena_pci_id_tbl.h b/drivers/net/ethernet/amazon/ena/ena_pci_id_tbl.h new file mode 100644 index 000000000000..f80d2a47fa94 --- /dev/null +++ b/drivers/net/ethernet/amazon/ena/ena_pci_id_tbl.h @@ -0,0 +1,67 @@ +/* + * Copyright 2015 Amazon.com, Inc. or its affiliates. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#ifndef ENA_PCI_ID_TBL_H_ +#define ENA_PCI_ID_TBL_H_ + +#ifndef PCI_VENDOR_ID_AMAZON +#define PCI_VENDOR_ID_AMAZON 0x1d0f +#endif + +#ifndef PCI_DEV_ID_ENA_PF +#define PCI_DEV_ID_ENA_PF 0x0ec2 +#endif + +#ifndef PCI_DEV_ID_ENA_LLQ_PF +#define PCI_DEV_ID_ENA_LLQ_PF 0x1ec2 +#endif + +#ifndef PCI_DEV_ID_ENA_VF +#define PCI_DEV_ID_ENA_VF 0xec20 +#endif + +#ifndef PCI_DEV_ID_ENA_LLQ_VF +#define PCI_DEV_ID_ENA_LLQ_VF 0xec21 +#endif + +#define ENA_PCI_ID_TABLE_ENTRY(devid) \ + {PCI_DEVICE(PCI_VENDOR_ID_AMAZON, devid)}, + +static const struct pci_device_id ena_pci_tbl[] = { + ENA_PCI_ID_TABLE_ENTRY(PCI_DEV_ID_ENA_PF) + ENA_PCI_ID_TABLE_ENTRY(PCI_DEV_ID_ENA_LLQ_PF) + ENA_PCI_ID_TABLE_ENTRY(PCI_DEV_ID_ENA_VF) + ENA_PCI_ID_TABLE_ENTRY(PCI_DEV_ID_ENA_LLQ_VF) + { } +}; + +#endif /* ENA_PCI_ID_TBL_H_ */ diff --git a/drivers/net/ethernet/amazon/ena/ena_regs_defs.h b/drivers/net/ethernet/amazon/ena/ena_regs_defs.h new file mode 100644 index 000000000000..26097a2b6030 --- /dev/null +++ b/drivers/net/ethernet/amazon/ena/ena_regs_defs.h @@ -0,0 +1,133 @@ +/* + * Copyright 2015 - 2016 Amazon.com, Inc. or its affiliates. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _ENA_REGS_H_ +#define _ENA_REGS_H_ + +/* ena_registers offsets */ +#define ENA_REGS_VERSION_OFF 0x0 +#define ENA_REGS_CONTROLLER_VERSION_OFF 0x4 +#define ENA_REGS_CAPS_OFF 0x8 +#define ENA_REGS_CAPS_EXT_OFF 0xc +#define ENA_REGS_AQ_BASE_LO_OFF 0x10 +#define ENA_REGS_AQ_BASE_HI_OFF 0x14 +#define ENA_REGS_AQ_CAPS_OFF 0x18 +#define ENA_REGS_ACQ_BASE_LO_OFF 0x20 +#define ENA_REGS_ACQ_BASE_HI_OFF 0x24 +#define ENA_REGS_ACQ_CAPS_OFF 0x28 +#define ENA_REGS_AQ_DB_OFF 0x2c +#define ENA_REGS_ACQ_TAIL_OFF 0x30 +#define ENA_REGS_AENQ_CAPS_OFF 0x34 +#define ENA_REGS_AENQ_BASE_LO_OFF 0x38 +#define ENA_REGS_AENQ_BASE_HI_OFF 0x3c +#define ENA_REGS_AENQ_HEAD_DB_OFF 0x40 +#define ENA_REGS_AENQ_TAIL_OFF 0x44 +#define ENA_REGS_INTR_MASK_OFF 0x4c +#define ENA_REGS_DEV_CTL_OFF 0x54 +#define ENA_REGS_DEV_STS_OFF 0x58 +#define ENA_REGS_MMIO_REG_READ_OFF 0x5c +#define ENA_REGS_MMIO_RESP_LO_OFF 0x60 +#define ENA_REGS_MMIO_RESP_HI_OFF 0x64 +#define ENA_REGS_RSS_IND_ENTRY_UPDATE_OFF 0x68 + +/* version register */ +#define ENA_REGS_VERSION_MINOR_VERSION_MASK 0xff +#define ENA_REGS_VERSION_MAJOR_VERSION_SHIFT 8 +#define ENA_REGS_VERSION_MAJOR_VERSION_MASK 0xff00 + +/* controller_version register */ +#define ENA_REGS_CONTROLLER_VERSION_SUBMINOR_VERSION_MASK 0xff +#define ENA_REGS_CONTROLLER_VERSION_MINOR_VERSION_SHIFT 8 +#define ENA_REGS_CONTROLLER_VERSION_MINOR_VERSION_MASK 0xff00 +#define ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_SHIFT 16 +#define ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_MASK 0xff0000 +#define ENA_REGS_CONTROLLER_VERSION_IMPL_ID_SHIFT 24 +#define ENA_REGS_CONTROLLER_VERSION_IMPL_ID_MASK 0xff000000 + +/* caps register */ +#define ENA_REGS_CAPS_CONTIGUOUS_QUEUE_REQUIRED_MASK 0x1 +#define ENA_REGS_CAPS_RESET_TIMEOUT_SHIFT 1 +#define ENA_REGS_CAPS_RESET_TIMEOUT_MASK 0x3e +#define ENA_REGS_CAPS_DMA_ADDR_WIDTH_SHIFT 8 +#define ENA_REGS_CAPS_DMA_ADDR_WIDTH_MASK 0xff00 + +/* aq_caps register */ +#define ENA_REGS_AQ_CAPS_AQ_DEPTH_MASK 0xffff +#define ENA_REGS_AQ_CAPS_AQ_ENTRY_SIZE_SHIFT 16 +#define ENA_REGS_AQ_CAPS_AQ_ENTRY_SIZE_MASK 0xffff0000 + +/* acq_caps register */ +#define ENA_REGS_ACQ_CAPS_ACQ_DEPTH_MASK 0xffff +#define ENA_REGS_ACQ_CAPS_ACQ_ENTRY_SIZE_SHIFT 16 +#define ENA_REGS_ACQ_CAPS_ACQ_ENTRY_SIZE_MASK 0xffff0000 + +/* aenq_caps register */ +#define ENA_REGS_AENQ_CAPS_AENQ_DEPTH_MASK 0xffff +#define ENA_REGS_AENQ_CAPS_AENQ_ENTRY_SIZE_SHIFT 16 +#define ENA_REGS_AENQ_CAPS_AENQ_ENTRY_SIZE_MASK 0xffff0000 + +/* dev_ctl register */ +#define ENA_REGS_DEV_CTL_DEV_RESET_MASK 0x1 +#define ENA_REGS_DEV_CTL_AQ_RESTART_SHIFT 1 +#define ENA_REGS_DEV_CTL_AQ_RESTART_MASK 0x2 +#define ENA_REGS_DEV_CTL_QUIESCENT_SHIFT 2 +#define ENA_REGS_DEV_CTL_QUIESCENT_MASK 0x4 +#define ENA_REGS_DEV_CTL_IO_RESUME_SHIFT 3 +#define ENA_REGS_DEV_CTL_IO_RESUME_MASK 0x8 + +/* dev_sts register */ +#define ENA_REGS_DEV_STS_READY_MASK 0x1 +#define ENA_REGS_DEV_STS_AQ_RESTART_IN_PROGRESS_SHIFT 1 +#define ENA_REGS_DEV_STS_AQ_RESTART_IN_PROGRESS_MASK 0x2 +#define ENA_REGS_DEV_STS_AQ_RESTART_FINISHED_SHIFT 2 +#define ENA_REGS_DEV_STS_AQ_RESTART_FINISHED_MASK 0x4 +#define ENA_REGS_DEV_STS_RESET_IN_PROGRESS_SHIFT 3 +#define ENA_REGS_DEV_STS_RESET_IN_PROGRESS_MASK 0x8 +#define ENA_REGS_DEV_STS_RESET_FINISHED_SHIFT 4 +#define ENA_REGS_DEV_STS_RESET_FINISHED_MASK 0x10 +#define ENA_REGS_DEV_STS_FATAL_ERROR_SHIFT 5 +#define ENA_REGS_DEV_STS_FATAL_ERROR_MASK 0x20 +#define ENA_REGS_DEV_STS_QUIESCENT_STATE_IN_PROGRESS_SHIFT 6 +#define ENA_REGS_DEV_STS_QUIESCENT_STATE_IN_PROGRESS_MASK 0x40 +#define ENA_REGS_DEV_STS_QUIESCENT_STATE_ACHIEVED_SHIFT 7 +#define ENA_REGS_DEV_STS_QUIESCENT_STATE_ACHIEVED_MASK 0x80 + +/* mmio_reg_read register */ +#define ENA_REGS_MMIO_REG_READ_REQ_ID_MASK 0xffff +#define ENA_REGS_MMIO_REG_READ_REG_OFF_SHIFT 16 +#define ENA_REGS_MMIO_REG_READ_REG_OFF_MASK 0xffff0000 + +/* rss_ind_entry_update register */ +#define ENA_REGS_RSS_IND_ENTRY_UPDATE_INDEX_MASK 0xffff +#define ENA_REGS_RSS_IND_ENTRY_UPDATE_CQ_IDX_SHIFT 16 +#define ENA_REGS_RSS_IND_ENTRY_UPDATE_CQ_IDX_MASK 0xffff0000 + +#endif /*_ENA_REGS_H_ */ diff --git a/drivers/net/ethernet/apm/xgene/Kconfig b/drivers/net/ethernet/apm/xgene/Kconfig index 300e3b5c54e0..afccb033177b 100644 --- a/drivers/net/ethernet/apm/xgene/Kconfig +++ b/drivers/net/ethernet/apm/xgene/Kconfig @@ -4,6 +4,7 @@ config NET_XGENE depends on ARCH_XGENE || COMPILE_TEST select PHYLIB select MDIO_XGENE + select GPIOLIB help This is the Ethernet driver for the on-chip ethernet interface on the APM X-Gene SoC. diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_cle.c b/drivers/net/ethernet/apm/xgene/xgene_enet_cle.c index 472c0fb3f4c4..23d72af83d82 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_cle.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_cle.c @@ -32,12 +32,19 @@ static void xgene_cle_sband_to_hw(u8 frag, enum xgene_cle_prot_version ver, SET_VAL(SB_HDRLEN, len); } -static void xgene_cle_idt_to_hw(u32 dstqid, u32 fpsel, +static void xgene_cle_idt_to_hw(struct xgene_enet_pdata *pdata, + u32 dstqid, u32 fpsel, u32 nfpsel, u32 *idt_reg) { - *idt_reg = SET_VAL(IDT_DSTQID, dstqid) | - SET_VAL(IDT_FPSEL, fpsel) | - SET_VAL(IDT_NFPSEL, nfpsel); + if (pdata->enet_id == XGENE_ENET1) { + *idt_reg = SET_VAL(IDT_DSTQID, dstqid) | + SET_VAL(IDT_FPSEL1, fpsel) | + SET_VAL(IDT_NFPSEL1, nfpsel); + } else { + *idt_reg = SET_VAL(IDT_DSTQID, dstqid) | + SET_VAL(IDT_FPSEL, fpsel) | + SET_VAL(IDT_NFPSEL, nfpsel); + } } static void xgene_cle_dbptr_to_hw(struct xgene_enet_pdata *pdata, @@ -344,7 +351,7 @@ static int xgene_cle_set_rss_idt(struct xgene_enet_pdata *pdata) nfpsel = 0; idt_reg = 0; - xgene_cle_idt_to_hw(dstqid, fpsel, nfpsel, &idt_reg); + xgene_cle_idt_to_hw(pdata, dstqid, fpsel, nfpsel, &idt_reg); ret = xgene_cle_dram_wr(&pdata->cle, &idt_reg, 1, i, RSS_IDT, CLE_CMD_WR); if (ret) diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_cle.h b/drivers/net/ethernet/apm/xgene/xgene_enet_cle.h index 33c5f6b25824..9ac9f8e145ec 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_cle.h +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_cle.h @@ -196,9 +196,13 @@ enum xgene_cle_ptree_dbptrs { #define IDT_DSTQID_POS 0 #define IDT_DSTQID_LEN 12 #define IDT_FPSEL_POS 12 -#define IDT_FPSEL_LEN 4 -#define IDT_NFPSEL_POS 16 -#define IDT_NFPSEL_LEN 4 +#define IDT_FPSEL_LEN 5 +#define IDT_NFPSEL_POS 17 +#define IDT_NFPSEL_LEN 5 +#define IDT_FPSEL1_POS 12 +#define IDT_FPSEL1_LEN 4 +#define IDT_NFPSEL1_POS 16 +#define IDT_NFPSEL1_LEN 4 struct xgene_cle_ptree_branch { bool valid; diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c index 18bb9556dd00..321fb197621e 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c @@ -761,18 +761,18 @@ int xgene_enet_phy_connect(struct net_device *ndev) if (dev->of_node) { for (i = 0 ; i < 2; i++) { np = of_parse_phandle(dev->of_node, "phy-handle", i); - if (np) - break; - } - if (!np) { - netdev_dbg(ndev, "No phy-handle found in DT\n"); - return -ENODEV; + if (!np) + continue; + + phy_dev = of_phy_connect(ndev, np, + &xgene_enet_adjust_link, + 0, pdata->phy_mode); + of_node_put(np); + if (phy_dev) + break; } - phy_dev = of_phy_connect(ndev, np, &xgene_enet_adjust_link, - 0, pdata->phy_mode); - of_node_put(np); if (!phy_dev) { netdev_err(ndev, "Could not connect to PHY\n"); return -ENODEV; diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h index 179a44dceb29..8a8d05500894 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h @@ -124,6 +124,12 @@ enum xgene_enet_rm { #define MAC_READ_REG_OFFSET 0x0c #define MAC_COMMAND_DONE_REG_OFFSET 0x10 +#define PCS_ADDR_REG_OFFSET 0x00 +#define PCS_COMMAND_REG_OFFSET 0x04 +#define PCS_WRITE_REG_OFFSET 0x08 +#define PCS_READ_REG_OFFSET 0x0c +#define PCS_COMMAND_DONE_REG_OFFSET 0x10 + #define MII_MGMT_CONFIG_ADDR 0x20 #define MII_MGMT_COMMAND_ADDR 0x24 #define MII_MGMT_ADDRESS_ADDR 0x28 diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c index d1d6b5eeb613..b8b9495e6da6 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c @@ -19,6 +19,7 @@ * along with this program. If not, see <http://www.gnu.org/licenses/>. */ +#include <linux/gpio.h> #include "xgene_enet_main.h" #include "xgene_enet_hw.h" #include "xgene_enet_sgmac.h" @@ -72,7 +73,6 @@ static int xgene_enet_refill_bufpool(struct xgene_enet_desc_ring *buf_pool, skb = netdev_alloc_skb_ip_align(ndev, len); if (unlikely(!skb)) return -ENOMEM; - buf_pool->rx_skb[tail] = skb; dma_addr = dma_map_single(dev, skb->data, len, DMA_FROM_DEVICE); if (dma_mapping_error(dev, dma_addr)) { @@ -81,6 +81,8 @@ static int xgene_enet_refill_bufpool(struct xgene_enet_desc_ring *buf_pool, return -EINVAL; } + buf_pool->rx_skb[tail] = skb; + raw_desc->m1 = cpu_to_le64(SET_VAL(DATAADDR, dma_addr) | SET_VAL(BUFDATALEN, bufdatalen) | SET_BIT(COHERENT)); @@ -102,12 +104,21 @@ static u8 xgene_enet_hdr_len(const void *data) static void xgene_enet_delete_bufpool(struct xgene_enet_desc_ring *buf_pool) { + struct device *dev = ndev_to_dev(buf_pool->ndev); + struct xgene_enet_raw_desc16 *raw_desc; + dma_addr_t dma_addr; int i; /* Free up the buffers held by hardware */ for (i = 0; i < buf_pool->slots; i++) { - if (buf_pool->rx_skb[i]) + if (buf_pool->rx_skb[i]) { dev_kfree_skb_any(buf_pool->rx_skb[i]); + + raw_desc = &buf_pool->raw_desc16[i]; + dma_addr = GET_VAL(DATAADDR, le64_to_cpu(raw_desc->m1)); + dma_unmap_single(dev, dma_addr, XGENE_ENET_MAX_MTU, + DMA_FROM_DEVICE); + } } } @@ -452,7 +463,6 @@ static int xgene_enet_rx_frame(struct xgene_enet_desc_ring *rx_ring, struct xgene_enet_raw_desc *raw_desc) { struct net_device *ndev; - struct xgene_enet_pdata *pdata; struct device *dev; struct xgene_enet_desc_ring *buf_pool; u32 datalen, skb_index; @@ -461,7 +471,6 @@ static int xgene_enet_rx_frame(struct xgene_enet_desc_ring *rx_ring, int ret = 0; ndev = rx_ring->ndev; - pdata = netdev_priv(ndev); dev = ndev_to_dev(rx_ring->ndev); buf_pool = rx_ring->buf_pool; @@ -1312,6 +1321,18 @@ static int xgene_enet_check_phy_handle(struct xgene_enet_pdata *pdata) return 0; } +static void xgene_enet_gpiod_get(struct xgene_enet_pdata *pdata) +{ + struct device *dev = &pdata->pdev->dev; + + if (pdata->phy_mode != PHY_INTERFACE_MODE_XGMII) + return; + + pdata->sfp_rdy = gpiod_get(dev, "rxlos", GPIOD_IN); + if (IS_ERR(pdata->sfp_rdy)) + pdata->sfp_rdy = gpiod_get(dev, "sfp", GPIOD_IN); +} + static int xgene_enet_get_resources(struct xgene_enet_pdata *pdata) { struct platform_device *pdev; @@ -1401,6 +1422,8 @@ static int xgene_enet_get_resources(struct xgene_enet_pdata *pdata) if (ret) return ret; + xgene_enet_gpiod_get(pdata); + pdata->clk = devm_clk_get(&pdev->dev, NULL); if (IS_ERR(pdata->clk)) { /* Firmware may have set up the clock already. */ @@ -1425,6 +1448,7 @@ static int xgene_enet_get_resources(struct xgene_enet_pdata *pdata) } else { pdata->mcx_mac_addr = base_addr + BLOCK_AXG_MAC_OFFSET; pdata->mcx_mac_csr_addr = base_addr + BLOCK_AXG_MAC_CSR_OFFSET; + pdata->pcs_addr = base_addr + BLOCK_PCS_OFFSET; } pdata->rx_buff_cnt = NUM_PKT_BUF; @@ -1454,10 +1478,8 @@ static int xgene_enet_init_hw(struct xgene_enet_pdata *pdata) buf_pool = pdata->rx_ring[i]->buf_pool; xgene_enet_init_bufpool(buf_pool); ret = xgene_enet_refill_bufpool(buf_pool, pdata->rx_buff_cnt); - if (ret) { - xgene_enet_delete_desc_rings(pdata); - return ret; - } + if (ret) + goto err; } dst_ring_num = xgene_enet_dst_ring_num(pdata->rx_ring[0]); @@ -1474,7 +1496,7 @@ static int xgene_enet_init_hw(struct xgene_enet_pdata *pdata) ret = pdata->cle_ops->cle_init(pdata); if (ret) { netdev_err(ndev, "Preclass Tree init error\n"); - return ret; + goto err; } } else { pdata->port_ops->cle_bypass(pdata, dst_ring_num, buf_pool->id); @@ -1484,6 +1506,10 @@ static int xgene_enet_init_hw(struct xgene_enet_pdata *pdata) pdata->mac_ops->init(pdata); return ret; + +err: + xgene_enet_delete_desc_rings(pdata); + return ret; } static void xgene_enet_setup_ops(struct xgene_enet_pdata *pdata) @@ -1631,8 +1657,8 @@ static int xgene_enet_probe(struct platform_device *pdev) } #endif if (!pdata->enet_id) { - free_netdev(ndev); - return -ENODEV; + ret = -ENODEV; + goto err; } ret = xgene_enet_get_resources(pdata); @@ -1655,7 +1681,7 @@ static int xgene_enet_probe(struct platform_device *pdev) ret = xgene_enet_init_hw(pdata); if (ret) - goto err_netdev; + goto err; link_state = pdata->mac_ops->link_state; if (pdata->phy_mode == PHY_INTERFACE_MODE_XGMII) { @@ -1665,21 +1691,32 @@ static int xgene_enet_probe(struct platform_device *pdev) ret = xgene_enet_mdio_config(pdata); else INIT_DELAYED_WORK(&pdata->link_work, link_state); + + if (ret) + goto err1; } - if (ret) - goto err; xgene_enet_napi_add(pdata); ret = register_netdev(ndev); if (ret) { netdev_err(ndev, "Failed to register netdev\n"); - goto err; + goto err2; } return 0; -err_netdev: - unregister_netdev(ndev); +err2: + /* + * If necessary, free_netdev() will call netif_napi_del() and undo + * the effects of xgene_enet_napi_add()'s calls to netif_napi_add(). + */ + + if (pdata->mdio_driver) + xgene_enet_phy_disconnect(pdata); + else if (pdata->phy_mode == PHY_INTERFACE_MODE_RGMII) + xgene_enet_mdio_remove(pdata); +err1: + xgene_enet_delete_desc_rings(pdata); err: free_netdev(ndev); return ret; @@ -1688,11 +1725,9 @@ err: static int xgene_enet_remove(struct platform_device *pdev) { struct xgene_enet_pdata *pdata; - const struct xgene_mac_ops *mac_ops; struct net_device *ndev; pdata = platform_get_drvdata(pdev); - mac_ops = pdata->mac_ops; ndev = pdata->ndev; rtnl_lock(); diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h index 217546e5714a..b339fc1e8841 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h @@ -196,6 +196,7 @@ struct xgene_enet_pdata { void __iomem *mcx_mac_addr; void __iomem *mcx_mac_csr_addr; void __iomem *base_addr; + void __iomem *pcs_addr; void __iomem *ring_csr_addr; void __iomem *ring_cmd_addr; int phy_mode; @@ -216,6 +217,7 @@ struct xgene_enet_pdata { u8 tx_delay; u8 rx_delay; bool mdio_driver; + struct gpio_desc *sfp_rdy; }; struct xgene_indirect_ctl { diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c index 9c6ad0dce00f..d672e71b5a50 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c @@ -18,6 +18,8 @@ * along with this program. If not, see <http://www.gnu.org/licenses/>. */ +#include <linux/of_gpio.h> +#include <linux/gpio.h> #include "xgene_enet_main.h" #include "xgene_enet_hw.h" #include "xgene_enet_xgmac.h" @@ -84,6 +86,21 @@ static void xgene_enet_wr_mac(struct xgene_enet_pdata *pdata, wr_addr); } +static void xgene_enet_wr_pcs(struct xgene_enet_pdata *pdata, + u32 wr_addr, u32 wr_data) +{ + void __iomem *addr, *wr, *cmd, *cmd_done; + + addr = pdata->pcs_addr + PCS_ADDR_REG_OFFSET; + wr = pdata->pcs_addr + PCS_WRITE_REG_OFFSET; + cmd = pdata->pcs_addr + PCS_COMMAND_REG_OFFSET; + cmd_done = pdata->pcs_addr + PCS_COMMAND_DONE_REG_OFFSET; + + if (!xgene_enet_wr_indirect(addr, wr, cmd, cmd_done, wr_addr, wr_data)) + netdev_err(pdata->ndev, "PCS write failed, addr: %04x\n", + wr_addr); +} + static void xgene_enet_rd_csr(struct xgene_enet_pdata *pdata, u32 offset, u32 *val) { @@ -122,6 +139,7 @@ static bool xgene_enet_rd_indirect(void __iomem *addr, void __iomem *rd, return true; } + static void xgene_enet_rd_mac(struct xgene_enet_pdata *pdata, u32 rd_addr, u32 *rd_data) { @@ -137,6 +155,21 @@ static void xgene_enet_rd_mac(struct xgene_enet_pdata *pdata, rd_addr); } +static void xgene_enet_rd_pcs(struct xgene_enet_pdata *pdata, + u32 rd_addr, u32 *rd_data) +{ + void __iomem *addr, *rd, *cmd, *cmd_done; + + addr = pdata->pcs_addr + PCS_ADDR_REG_OFFSET; + rd = pdata->pcs_addr + PCS_READ_REG_OFFSET; + cmd = pdata->pcs_addr + PCS_COMMAND_REG_OFFSET; + cmd_done = pdata->pcs_addr + PCS_COMMAND_DONE_REG_OFFSET; + + if (!xgene_enet_rd_indirect(addr, rd, cmd, cmd_done, rd_addr, rd_data)) + netdev_err(pdata->ndev, "PCS read failed, addr: %04x\n", + rd_addr); +} + static int xgene_enet_ecc_init(struct xgene_enet_pdata *pdata) { struct net_device *ndev = pdata->ndev; @@ -171,6 +204,15 @@ static void xgene_xgmac_reset(struct xgene_enet_pdata *pdata) xgene_enet_wr_mac(pdata, AXGMAC_CONFIG_0, 0); } +static void xgene_pcs_reset(struct xgene_enet_pdata *pdata) +{ + u32 data; + + xgene_enet_rd_pcs(pdata, PCS_CONTROL_1, &data); + xgene_enet_wr_pcs(pdata, PCS_CONTROL_1, data | PCS_CTRL_PCS_RST); + xgene_enet_wr_pcs(pdata, PCS_CONTROL_1, data & ~PCS_CTRL_PCS_RST); +} + static void xgene_xgmac_set_mac_addr(struct xgene_enet_pdata *pdata) { u32 addr0, addr1; @@ -216,12 +258,12 @@ static void xgene_xgmac_init(struct xgene_enet_pdata *pdata) data |= CFG_RSIF_FPBUFF_TIMEOUT_EN; xgene_enet_wr_csr(pdata, XG_RSIF_CONFIG_REG_ADDR, data); - xgene_enet_wr_csr(pdata, XG_CFG_BYPASS_ADDR, RESUME_TX); - xgene_enet_wr_csr(pdata, XGENET_RX_DV_GATE_REG_0_ADDR, 0); xgene_enet_rd_csr(pdata, XG_ENET_SPARE_CFG_REG_ADDR, &data); data |= BIT(12); xgene_enet_wr_csr(pdata, XG_ENET_SPARE_CFG_REG_ADDR, data); xgene_enet_wr_csr(pdata, XG_ENET_SPARE_CFG_REG_1_ADDR, 0x82); + xgene_enet_wr_csr(pdata, XGENET_RX_DV_GATE_REG_0_ADDR, 0); + xgene_enet_wr_csr(pdata, XG_CFG_BYPASS_ADDR, RESUME_TX); } static void xgene_xgmac_rx_enable(struct xgene_enet_pdata *pdata) @@ -359,14 +401,17 @@ static void xgene_enet_link_state(struct work_struct *work) { struct xgene_enet_pdata *pdata = container_of(to_delayed_work(work), struct xgene_enet_pdata, link_work); + struct gpio_desc *sfp_rdy = pdata->sfp_rdy; struct net_device *ndev = pdata->ndev; u32 link_status, poll_interval; link_status = xgene_enet_link_status(pdata); + if (link_status && !IS_ERR(sfp_rdy) && !gpiod_get_value(sfp_rdy)) + link_status = 0; + if (link_status) { if (!netif_carrier_ok(ndev)) { netif_carrier_on(ndev); - xgene_xgmac_init(pdata); xgene_xgmac_rx_enable(pdata); xgene_xgmac_tx_enable(pdata); netdev_info(ndev, "Link is Up - 10Gbps\n"); @@ -380,6 +425,8 @@ static void xgene_enet_link_state(struct work_struct *work) netdev_info(ndev, "Link is Down\n"); } poll_interval = PHY_POLL_LINK_OFF; + + xgene_pcs_reset(pdata); } schedule_delayed_work(&pdata->link_work, poll_interval); diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h index f1ea485f916b..360ccbd95566 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h @@ -24,6 +24,7 @@ #define X2_BLOCK_ETH_MAC_CSR_OFFSET 0x3000 #define BLOCK_AXG_MAC_OFFSET 0x0800 #define BLOCK_AXG_MAC_CSR_OFFSET 0x2000 +#define BLOCK_PCS_OFFSET 0x3800 #define XGENET_CONFIG_REG_ADDR 0x20 #define XGENET_SRST_ADDR 0x00 @@ -72,6 +73,9 @@ #define XG_MCX_ICM_CONFIG0_REG_0_ADDR 0x00e0 #define XG_MCX_ICM_CONFIG2_REG_0_ADDR 0x00e8 +#define PCS_CONTROL_1 0x0000 +#define PCS_CTRL_PCS_RST BIT(15) + extern const struct xgene_mac_ops xgene_xgmac_ops; extern const struct xgene_port_ops xgene_xgport_ops; diff --git a/drivers/net/ethernet/brocade/bna/bnad.c b/drivers/net/ethernet/brocade/bna/bnad.c index 771cc267f217..f9df4b5ae90e 100644 --- a/drivers/net/ethernet/brocade/bna/bnad.c +++ b/drivers/net/ethernet/brocade/bna/bnad.c @@ -54,9 +54,7 @@ MODULE_PARM_DESC(bna_debugfs_enable, "Enables debugfs feature, default=1," * Global variables */ static u32 bnad_rxqs_per_cq = 2; -static u32 bna_id; -static struct mutex bnad_list_mutex; -static LIST_HEAD(bnad_list); +static atomic_t bna_id; static const u8 bnad_bcast_addr[] __aligned(2) = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff }; @@ -76,23 +74,6 @@ do { \ (_res_info)->res_u.mem_info.len = (_size); \ } while (0) -static void -bnad_add_to_list(struct bnad *bnad) -{ - mutex_lock(&bnad_list_mutex); - list_add_tail(&bnad->list_entry, &bnad_list); - bnad->id = bna_id++; - mutex_unlock(&bnad_list_mutex); -} - -static void -bnad_remove_from_list(struct bnad *bnad) -{ - mutex_lock(&bnad_list_mutex); - list_del(&bnad->list_entry); - mutex_unlock(&bnad_list_mutex); -} - /* * Reinitialize completions in CQ, once Rx is taken down */ @@ -3573,14 +3554,12 @@ bnad_lock_init(struct bnad *bnad) { spin_lock_init(&bnad->bna_lock); mutex_init(&bnad->conf_mutex); - mutex_init(&bnad_list_mutex); } static void bnad_lock_uninit(struct bnad *bnad) { mutex_destroy(&bnad->conf_mutex); - mutex_destroy(&bnad_list_mutex); } /* PCI Initialization */ @@ -3653,7 +3632,7 @@ bnad_pci_probe(struct pci_dev *pdev, } bnad = netdev_priv(netdev); bnad_lock_init(bnad); - bnad_add_to_list(bnad); + bnad->id = atomic_inc_return(&bna_id) - 1; mutex_lock(&bnad->conf_mutex); /* @@ -3807,7 +3786,6 @@ pci_uninit: bnad_pci_uninit(pdev); unlock_mutex: mutex_unlock(&bnad->conf_mutex); - bnad_remove_from_list(bnad); bnad_lock_uninit(bnad); free_netdev(netdev); return err; @@ -3845,7 +3823,6 @@ bnad_pci_remove(struct pci_dev *pdev) bnad_disable_msix(bnad); bnad_pci_uninit(pdev); mutex_unlock(&bnad->conf_mutex); - bnad_remove_from_list(bnad); bnad_lock_uninit(bnad); /* Remove the debugfs node for this bnad */ kfree(bnad->regdata); diff --git a/drivers/net/ethernet/brocade/bna/bnad.h b/drivers/net/ethernet/brocade/bna/bnad.h index f4ed816b93ee..46f7b842b39c 100644 --- a/drivers/net/ethernet/brocade/bna/bnad.h +++ b/drivers/net/ethernet/brocade/bna/bnad.h @@ -288,7 +288,6 @@ struct bnad_rx_unmap_q { struct bnad { struct net_device *netdev; u32 id; - struct list_head list_entry; /* Data path */ struct bnad_tx_info tx_info[BNAD_MAX_TX]; diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c index 89c0cfa9719f..dbce938e7d0c 100644 --- a/drivers/net/ethernet/cadence/macb.c +++ b/drivers/net/ethernet/cadence/macb.c @@ -541,6 +541,14 @@ static void macb_tx_unmap(struct macb *bp, struct macb_tx_skb *tx_skb) } } +static inline void macb_set_addr(struct macb_dma_desc *desc, dma_addr_t addr) +{ + desc->addr = (u32)addr; +#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT + desc->addrh = (u32)(addr >> 32); +#endif +} + static void macb_tx_error_task(struct work_struct *work) { struct macb_queue *queue = container_of(work, struct macb_queue, @@ -621,14 +629,17 @@ static void macb_tx_error_task(struct work_struct *work) /* Set end of TX queue */ desc = macb_tx_desc(queue, 0); - desc->addr = 0; + macb_set_addr(desc, 0); desc->ctrl = MACB_BIT(TX_USED); /* Make descriptor updates visible to hardware */ wmb(); /* Reinitialize the TX desc queue */ - queue_writel(queue, TBQP, queue->tx_ring_dma); + queue_writel(queue, TBQP, (u32)(queue->tx_ring_dma)); +#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT + queue_writel(queue, TBQPH, (u32)(queue->tx_ring_dma >> 32)); +#endif /* Make TX ring reflect state of hardware */ queue->tx_head = 0; queue->tx_tail = 0; @@ -750,7 +761,7 @@ static void gem_rx_refill(struct macb *bp) if (entry == RX_RING_SIZE - 1) paddr |= MACB_BIT(RX_WRAP); - bp->rx_ring[entry].addr = paddr; + macb_set_addr(&(bp->rx_ring[entry]), paddr); bp->rx_ring[entry].ctrl = 0; /* properly align Ethernet header */ @@ -798,7 +809,9 @@ static int gem_rx(struct macb *bp, int budget) int count = 0; while (count < budget) { - u32 addr, ctrl; + u32 ctrl; + dma_addr_t addr; + bool rxused; entry = macb_rx_ring_wrap(bp->rx_tail); desc = &bp->rx_ring[entry]; @@ -806,10 +819,14 @@ static int gem_rx(struct macb *bp, int budget) /* Make hw descriptor updates visible to CPU */ rmb(); - addr = desc->addr; + rxused = (desc->addr & MACB_BIT(RX_USED)) ? true : false; + addr = MACB_BF(RX_WADDR, MACB_BFEXT(RX_WADDR, desc->addr)); +#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT + addr |= ((u64)(desc->addrh) << 32); +#endif ctrl = desc->ctrl; - if (!(addr & MACB_BIT(RX_USED))) + if (!rxused) break; bp->rx_tail++; @@ -835,7 +852,6 @@ static int gem_rx(struct macb *bp, int budget) netdev_vdbg(bp->dev, "gem_rx %u (len %u)\n", entry, len); skb_put(skb, len); - addr = MACB_BF(RX_WADDR, MACB_BFEXT(RX_WADDR, addr)); dma_unmap_single(&bp->pdev->dev, addr, bp->rx_buffer_size, DMA_FROM_DEVICE); @@ -1299,7 +1315,7 @@ static unsigned int macb_tx_map(struct macb *bp, ctrl |= MACB_BIT(TX_WRAP); /* Set TX buffer descriptor */ - desc->addr = tx_skb->mapping; + macb_set_addr(desc, tx_skb->mapping); /* desc->addr must be visible to hardware before clearing * 'TX_USED' bit in desc->ctrl. */ @@ -1422,6 +1438,9 @@ static void gem_free_rx_buffers(struct macb *bp) desc = &bp->rx_ring[i]; addr = MACB_BF(RX_WADDR, MACB_BFEXT(RX_WADDR, desc->addr)); +#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT + addr |= ((u64)(desc->addrh) << 32); +#endif dma_unmap_single(&bp->pdev->dev, addr, bp->rx_buffer_size, DMA_FROM_DEVICE); dev_kfree_skb_any(skb); @@ -1547,7 +1566,7 @@ static void gem_init_rings(struct macb *bp) for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) { for (i = 0; i < TX_RING_SIZE; i++) { - queue->tx_ring[i].addr = 0; + macb_set_addr(&(queue->tx_ring[i]), 0); queue->tx_ring[i].ctrl = MACB_BIT(TX_USED); } queue->tx_ring[TX_RING_SIZE - 1].ctrl |= MACB_BIT(TX_WRAP); @@ -1694,6 +1713,10 @@ static void macb_configure_dma(struct macb *bp) dmacfg |= GEM_BIT(TXCOEN); else dmacfg &= ~GEM_BIT(TXCOEN); + +#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT + dmacfg |= GEM_BIT(ADDR64); +#endif netdev_dbg(bp->dev, "Cadence configure DMA with 0x%08x\n", dmacfg); gem_writel(bp, DMACFG, dmacfg); @@ -1739,9 +1762,15 @@ static void macb_init_hw(struct macb *bp) macb_configure_dma(bp); /* Initialize TX and RX buffers */ - macb_writel(bp, RBQP, bp->rx_ring_dma); + macb_writel(bp, RBQP, (u32)(bp->rx_ring_dma)); +#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT + macb_writel(bp, RBQPH, (u32)(bp->rx_ring_dma >> 32)); +#endif for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) { - queue_writel(queue, TBQP, queue->tx_ring_dma); + queue_writel(queue, TBQP, (u32)(queue->tx_ring_dma)); +#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT + queue_writel(queue, TBQPH, (u32)(queue->tx_ring_dma >> 32)); +#endif /* Enable interrupts */ queue_writel(queue, IER, @@ -2379,6 +2408,9 @@ static int macb_init(struct platform_device *pdev) queue->IDR = GEM_IDR(hw_q - 1); queue->IMR = GEM_IMR(hw_q - 1); queue->TBQP = GEM_TBQP(hw_q - 1); +#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT + queue->TBQPH = GEM_TBQPH(hw_q -1); +#endif } else { /* queue0 uses legacy registers */ queue->ISR = MACB_ISR; @@ -2386,6 +2418,9 @@ static int macb_init(struct platform_device *pdev) queue->IDR = MACB_IDR; queue->IMR = MACB_IMR; queue->TBQP = MACB_TBQP; +#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT + queue->TBQPH = MACB_TBQPH; +#endif } /* get irq: here we use the linux queue index, not the hardware @@ -2935,6 +2970,11 @@ static int macb_probe(struct platform_device *pdev) bp->wol |= MACB_WOL_HAS_MAGIC_PACKET; device_init_wakeup(&pdev->dev, bp->wol & MACB_WOL_HAS_MAGIC_PACKET); +#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT + if (GEM_BFEXT(DBWDEF, gem_readl(bp, DCFG1)) > GEM_DBW32) + dma_set_mask(&pdev->dev, DMA_BIT_MASK(44)); +#endif + spin_lock_init(&bp->lock); /* setup capabilities */ @@ -2945,7 +2985,7 @@ static int macb_probe(struct platform_device *pdev) dev->irq = platform_get_irq(pdev, 0); if (dev->irq < 0) { err = dev->irq; - goto err_disable_clocks; + goto err_out_free_netdev; } mac = of_get_mac_address(np); diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h index b6fcf10621b6..aa3aeecc3132 100644 --- a/drivers/net/ethernet/cadence/macb.h +++ b/drivers/net/ethernet/cadence/macb.h @@ -66,6 +66,8 @@ #define MACB_USRIO 0x00c0 #define MACB_WOL 0x00c4 #define MACB_MID 0x00fc +#define MACB_TBQPH 0x04C8 +#define MACB_RBQPH 0x04D4 /* GEM register offsets. */ #define GEM_NCFGR 0x0004 /* Network Config */ @@ -139,6 +141,7 @@ #define GEM_ISR(hw_q) (0x0400 + ((hw_q) << 2)) #define GEM_TBQP(hw_q) (0x0440 + ((hw_q) << 2)) +#define GEM_TBQPH(hw_q) (0x04C8) #define GEM_RBQP(hw_q) (0x0480 + ((hw_q) << 2)) #define GEM_IER(hw_q) (0x0600 + ((hw_q) << 2)) #define GEM_IDR(hw_q) (0x0620 + ((hw_q) << 2)) @@ -249,6 +252,8 @@ #define GEM_RXBS_SIZE 8 #define GEM_DDRP_OFFSET 24 /* disc_when_no_ahb */ #define GEM_DDRP_SIZE 1 +#define GEM_ADDR64_OFFSET 30 /* Address bus width - 64b or 32b */ +#define GEM_ADDR64_SIZE 1 /* Bitfields in NSR */ @@ -474,6 +479,10 @@ struct macb_dma_desc { u32 addr; u32 ctrl; +#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT + u32 addrh; + u32 resvd; +#endif }; /* DMA descriptor bitfields */ @@ -777,6 +786,7 @@ struct macb_queue { unsigned int IDR; unsigned int IMR; unsigned int TBQP; + unsigned int TBQPH; unsigned int tx_head, tx_tail; struct macb_dma_desc *tx_ring; diff --git a/drivers/net/ethernet/cavium/Kconfig b/drivers/net/ethernet/cavium/Kconfig index 0ef232d3331e..e1b78b500309 100644 --- a/drivers/net/ethernet/cavium/Kconfig +++ b/drivers/net/ethernet/cavium/Kconfig @@ -36,10 +36,20 @@ config THUNDER_NIC_BGX depends on 64BIT select PHYLIB select MDIO_THUNDER + select THUNDER_NIC_RGX ---help--- This driver supports programming and controlling of MAC interface from NIC physical function driver. +config THUNDER_NIC_RGX + tristate "Thunder MAC interface driver (RGX)" + depends on 64BIT + select PHYLIB + select MDIO_THUNDER + ---help--- + This driver supports configuring XCV block of RGX interface + present on CN81XX chip. + config LIQUIDIO tristate "Cavium LiquidIO support" depends on 64BIT diff --git a/drivers/net/ethernet/cavium/thunder/Makefile b/drivers/net/ethernet/cavium/thunder/Makefile index 5c4615ccaa14..6b4d4add7353 100644 --- a/drivers/net/ethernet/cavium/thunder/Makefile +++ b/drivers/net/ethernet/cavium/thunder/Makefile @@ -2,6 +2,7 @@ # Makefile for Cavium's Thunder ethernet device # +obj-$(CONFIG_THUNDER_NIC_RGX) += thunder_xcv.o obj-$(CONFIG_THUNDER_NIC_BGX) += thunder_bgx.o obj-$(CONFIG_THUNDER_NIC_PF) += nicpf.o obj-$(CONFIG_THUNDER_NIC_VF) += nicvf.o diff --git a/drivers/net/ethernet/cavium/thunder/nic.h b/drivers/net/ethernet/cavium/thunder/nic.h index 83025bb4737c..dd63f961827a 100644 --- a/drivers/net/ethernet/cavium/thunder/nic.h +++ b/drivers/net/ethernet/cavium/thunder/nic.h @@ -20,6 +20,17 @@ #define PCI_DEVICE_ID_THUNDER_NIC_VF 0xA034 #define PCI_DEVICE_ID_THUNDER_BGX 0xA026 +/* Subsystem device IDs */ +#define PCI_SUBSYS_DEVID_88XX_NIC_PF 0xA11E +#define PCI_SUBSYS_DEVID_81XX_NIC_PF 0xA21E +#define PCI_SUBSYS_DEVID_83XX_NIC_PF 0xA31E + +#define PCI_SUBSYS_DEVID_88XX_PASS1_NIC_VF 0xA11E +#define PCI_SUBSYS_DEVID_88XX_NIC_VF 0xA134 +#define PCI_SUBSYS_DEVID_81XX_NIC_VF 0xA234 +#define PCI_SUBSYS_DEVID_83XX_NIC_VF 0xA334 + + /* PCI BAR nos */ #define PCI_CFG_REG_BAR_NUM 0 #define PCI_MSIX_REG_BAR_NUM 4 @@ -41,40 +52,8 @@ /* Max pkinds */ #define NIC_MAX_PKIND 16 -/* Rx Channels */ -/* Receive channel configuration in TNS bypass mode - * Below is configuration in TNS bypass mode - * BGX0-LMAC0-CHAN0 - VNIC CHAN0 - * BGX0-LMAC1-CHAN0 - VNIC CHAN16 - * ... - * BGX1-LMAC0-CHAN0 - VNIC CHAN128 - * ... - * BGX1-LMAC3-CHAN0 - VNIC CHAN174 - */ -#define NIC_INTF_COUNT 2 /* Interfaces btw VNIC and TNS/BGX */ -#define NIC_CHANS_PER_INF 128 -#define NIC_MAX_CHANS (NIC_INTF_COUNT * NIC_CHANS_PER_INF) -#define NIC_CPI_COUNT 2048 /* No of channel parse indices */ - -/* TNS bypass mode: 1-1 mapping between VNIC and BGX:LMAC */ -#define NIC_MAX_BGX MAX_BGX_PER_CN88XX -#define NIC_CPI_PER_BGX (NIC_CPI_COUNT / NIC_MAX_BGX) -#define NIC_MAX_CPI_PER_LMAC 64 /* Max when CPI_ALG is IP diffserv */ -#define NIC_RSSI_PER_BGX (NIC_RSSI_COUNT / NIC_MAX_BGX) - -/* Tx scheduling */ -#define NIC_MAX_TL4 1024 -#define NIC_MAX_TL4_SHAPERS 256 /* 1 shaper for 4 TL4s */ -#define NIC_MAX_TL3 256 -#define NIC_MAX_TL3_SHAPERS 64 /* 1 shaper for 4 TL3s */ -#define NIC_MAX_TL2 64 -#define NIC_MAX_TL2_SHAPERS 2 /* 1 shaper for 32 TL2s */ -#define NIC_MAX_TL1 2 - -/* TNS bypass mode */ -#define NIC_TL2_PER_BGX 32 -#define NIC_TL4_PER_BGX (NIC_MAX_TL4 / NIC_MAX_BGX) -#define NIC_TL4_PER_LMAC (NIC_MAX_TL4 / NIC_CHANS_PER_INF) +/* Max when CPI_ALG is IP diffserv */ +#define NIC_MAX_CPI_PER_LMAC 64 /* NIC VF Interrupts */ #define NICVF_INTR_CQ 0 @@ -148,7 +127,6 @@ struct nicvf_cq_poll { struct napi_struct napi; }; -#define NIC_RSSI_COUNT 4096 /* Total no of RSS indices */ #define NIC_MAX_RSS_HASH_BITS 8 #define NIC_MAX_RSS_IDR_TBL_SIZE (1 << NIC_MAX_RSS_HASH_BITS) #define RSS_HASH_KEY_SIZE 5 /* 320 bit key */ @@ -273,6 +251,7 @@ struct nicvf { struct net_device *netdev; struct pci_dev *pdev; void __iomem *reg_base; +#define MAX_QUEUES_PER_QSET 8 struct queue_set *qs; struct nicvf_cq_poll *napi[8]; u8 vf_id; @@ -368,6 +347,7 @@ struct nicvf { #define NIC_MBOX_MSG_PNICVF_PTR 0x14 /* Get primary qset nicvf ptr */ #define NIC_MBOX_MSG_SNICVF_PTR 0x15 /* Send sqet nicvf ptr to PVF */ #define NIC_MBOX_MSG_LOOPBACK 0x16 /* Set interface in loopback */ +#define NIC_MBOX_MSG_RESET_STAT_COUNTER 0x17 /* Reset statistics counters */ #define NIC_MBOX_MSG_CFG_DONE 0xF0 /* VF configuration done */ #define NIC_MBOX_MSG_SHUTDOWN 0xF1 /* VF is being shutdown */ @@ -484,6 +464,31 @@ struct set_loopback { bool enable; }; +/* Reset statistics counters */ +struct reset_stat_cfg { + u8 msg; + /* Bitmap to select NIC_PF_VNIC(vf_id)_RX_STAT(0..13) */ + u16 rx_stat_mask; + /* Bitmap to select NIC_PF_VNIC(vf_id)_TX_STAT(0..4) */ + u8 tx_stat_mask; + /* Bitmap to select NIC_PF_QS(0..127)_RQ(0..7)_STAT(0..1) + * bit14, bit15 NIC_PF_QS(vf_id)_RQ7_STAT(0..1) + * bit12, bit13 NIC_PF_QS(vf_id)_RQ6_STAT(0..1) + * .. + * bit2, bit3 NIC_PF_QS(vf_id)_RQ1_STAT(0..1) + * bit0, bit1 NIC_PF_QS(vf_id)_RQ0_STAT(0..1) + */ + u16 rq_stat_mask; + /* Bitmap to select NIC_PF_QS(0..127)_SQ(0..7)_STAT(0..1) + * bit14, bit15 NIC_PF_QS(vf_id)_SQ7_STAT(0..1) + * bit12, bit13 NIC_PF_QS(vf_id)_SQ6_STAT(0..1) + * .. + * bit2, bit3 NIC_PF_QS(vf_id)_SQ1_STAT(0..1) + * bit0, bit1 NIC_PF_QS(vf_id)_SQ0_STAT(0..1) + */ + u16 sq_stat_mask; +}; + /* 128 bit shared memory between PF and each VF */ union nic_mbx { struct { u8 msg; } msg; @@ -501,6 +506,7 @@ union nic_mbx { struct sqs_alloc sqs_alloc; struct nicvf_ptr nicvf; struct set_loopback lbk; + struct reset_stat_cfg reset_stat; }; #define NIC_NODE_ID_MASK 0x03 @@ -514,7 +520,14 @@ static inline int nic_get_node_id(struct pci_dev *pdev) static inline bool pass1_silicon(struct pci_dev *pdev) { - return pdev->revision < 8; + return (pdev->revision < 8) && + (pdev->subsystem_device == PCI_SUBSYS_DEVID_88XX_NIC_PF); +} + +static inline bool pass2_silicon(struct pci_dev *pdev) +{ + return (pdev->revision >= 8) && + (pdev->subsystem_device == PCI_SUBSYS_DEVID_88XX_NIC_PF); } int nicvf_set_real_num_queues(struct net_device *netdev, diff --git a/drivers/net/ethernet/cavium/thunder/nic_main.c b/drivers/net/ethernet/cavium/thunder/nic_main.c index 16ed20357c5c..25618d203931 100644 --- a/drivers/net/ethernet/cavium/thunder/nic_main.c +++ b/drivers/net/ethernet/cavium/thunder/nic_main.c @@ -20,8 +20,25 @@ #define DRV_NAME "thunder-nic" #define DRV_VERSION "1.0" +struct hw_info { + u8 bgx_cnt; + u8 chans_per_lmac; + u8 chans_per_bgx; /* Rx/Tx chans */ + u8 chans_per_rgx; + u8 chans_per_lbk; + u16 cpi_cnt; + u16 rssi_cnt; + u16 rss_ind_tbl_size; + u16 tl4_cnt; + u16 tl3_cnt; + u8 tl2_cnt; + u8 tl1_cnt; + bool tl1_per_bgx; /* TL1 per BGX or per LMAC */ +}; + struct nicpf { struct pci_dev *pdev; + struct hw_info *hw; u8 node; unsigned int flags; u8 num_vf_en; /* No of VF enabled */ @@ -36,22 +53,22 @@ struct nicpf { #define NIC_SET_VF_LMAC_MAP(bgx, lmac) (((bgx & 0xF) << 4) | (lmac & 0xF)) #define NIC_GET_BGX_FROM_VF_LMAC_MAP(map) ((map >> 4) & 0xF) #define NIC_GET_LMAC_FROM_VF_LMAC_MAP(map) (map & 0xF) - u8 vf_lmac_map[MAX_LMAC]; + u8 *vf_lmac_map; struct delayed_work dwork; struct workqueue_struct *check_link; - u8 link[MAX_LMAC]; - u8 duplex[MAX_LMAC]; - u32 speed[MAX_LMAC]; + u8 *link; + u8 *duplex; + u32 *speed; u16 cpi_base[MAX_NUM_VFS_SUPPORTED]; u16 rssi_base[MAX_NUM_VFS_SUPPORTED]; - u16 rss_ind_tbl_size; bool mbx_lock[MAX_NUM_VFS_SUPPORTED]; /* MSI-X */ bool msix_enabled; u8 num_vec; - struct msix_entry msix_entries[NIC_PF_MSIX_VECTORS]; + struct msix_entry *msix_entries; bool irq_allocated[NIC_PF_MSIX_VECTORS]; + char irq_name[NIC_PF_MSIX_VECTORS][20]; }; /* Supported devices */ @@ -89,9 +106,22 @@ static u64 nic_reg_read(struct nicpf *nic, u64 offset) /* PF -> VF mailbox communication APIs */ static void nic_enable_mbx_intr(struct nicpf *nic) { - /* Enable mailbox interrupt for all 128 VFs */ - nic_reg_write(nic, NIC_PF_MAILBOX_ENA_W1S, ~0ull); - nic_reg_write(nic, NIC_PF_MAILBOX_ENA_W1S + sizeof(u64), ~0ull); + int vf_cnt = pci_sriov_get_totalvfs(nic->pdev); + +#define INTR_MASK(vfs) ((vfs < 64) ? (BIT_ULL(vfs) - 1) : (~0ull)) + + /* Clear it, to avoid spurious interrupts (if any) */ + nic_reg_write(nic, NIC_PF_MAILBOX_INT, INTR_MASK(vf_cnt)); + + /* Enable mailbox interrupt for all VFs */ + nic_reg_write(nic, NIC_PF_MAILBOX_ENA_W1S, INTR_MASK(vf_cnt)); + /* One mailbox intr enable reg per 64 VFs */ + if (vf_cnt > 64) { + nic_reg_write(nic, NIC_PF_MAILBOX_INT + sizeof(u64), + INTR_MASK(vf_cnt - 64)); + nic_reg_write(nic, NIC_PF_MAILBOX_ENA_W1S + sizeof(u64), + INTR_MASK(vf_cnt - 64)); + } } static void nic_clear_mbx_intr(struct nicpf *nic, int vf, int mbx_reg) @@ -144,7 +174,7 @@ static void nic_mbx_send_ready(struct nicpf *nic, int vf) mbx.nic_cfg.tns_mode = NIC_TNS_BYPASS_MODE; - if (vf < MAX_LMAC) { + if (vf < nic->num_vf_en) { bgx_idx = NIC_GET_BGX_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vf]); lmac = NIC_GET_LMAC_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vf]); @@ -155,7 +185,7 @@ static void nic_mbx_send_ready(struct nicpf *nic, int vf) mbx.nic_cfg.sqs_mode = (vf >= nic->num_vf_en) ? true : false; mbx.nic_cfg.node_id = nic->node; - mbx.nic_cfg.loopback_supported = vf < MAX_LMAC; + mbx.nic_cfg.loopback_supported = vf < nic->num_vf_en; nic_send_msg_to_vf(nic, vf, &mbx); } @@ -248,14 +278,22 @@ static int nic_update_hw_frs(struct nicpf *nic, int new_frs, int vf) /* Set minimum transmit packet size */ static void nic_set_tx_pkt_pad(struct nicpf *nic, int size) { - int lmac; + int lmac, max_lmac; + u16 sdevid; u64 lmac_cfg; /* Max value that can be set is 60 */ if (size > 60) size = 60; - for (lmac = 0; lmac < (MAX_BGX_PER_CN88XX * MAX_LMAC_PER_BGX); lmac++) { + pci_read_config_word(nic->pdev, PCI_SUBSYSTEM_ID, &sdevid); + /* 81xx's RGX has only one LMAC */ + if (sdevid == PCI_SUBSYS_DEVID_81XX_NIC_PF) + max_lmac = ((nic->hw->bgx_cnt - 1) * MAX_LMAC_PER_BGX) + 1; + else + max_lmac = nic->hw->bgx_cnt * MAX_LMAC_PER_BGX; + + for (lmac = 0; lmac < max_lmac; lmac++) { lmac_cfg = nic_reg_read(nic, NIC_PF_LMAC_0_7_CFG | (lmac << 3)); lmac_cfg &= ~(0xF << 2); lmac_cfg |= ((size / 4) << 2); @@ -275,7 +313,7 @@ static void nic_set_lmac_vf_mapping(struct nicpf *nic) nic->num_vf_en = 0; - for (bgx = 0; bgx < NIC_MAX_BGX; bgx++) { + for (bgx = 0; bgx < nic->hw->bgx_cnt; bgx++) { if (!(bgx_map & (1 << bgx))) continue; lmac_cnt = bgx_get_lmac_count(nic->node, bgx); @@ -295,28 +333,125 @@ static void nic_set_lmac_vf_mapping(struct nicpf *nic) nic_reg_write(nic, NIC_PF_LMAC_0_7_CREDIT + (lmac * 8), lmac_credit); + + /* On CN81XX there are only 8 VFs but max possible no of + * interfaces are 9. + */ + if (nic->num_vf_en >= pci_sriov_get_totalvfs(nic->pdev)) { + nic->num_vf_en = pci_sriov_get_totalvfs(nic->pdev); + break; + } } } +static void nic_free_lmacmem(struct nicpf *nic) +{ + kfree(nic->vf_lmac_map); + kfree(nic->link); + kfree(nic->duplex); + kfree(nic->speed); +} + +static int nic_get_hw_info(struct nicpf *nic) +{ + u8 max_lmac; + u16 sdevid; + struct hw_info *hw = nic->hw; + + pci_read_config_word(nic->pdev, PCI_SUBSYSTEM_ID, &sdevid); + + switch (sdevid) { + case PCI_SUBSYS_DEVID_88XX_NIC_PF: + hw->bgx_cnt = MAX_BGX_PER_CN88XX; + hw->chans_per_lmac = 16; + hw->chans_per_bgx = 128; + hw->cpi_cnt = 2048; + hw->rssi_cnt = 4096; + hw->rss_ind_tbl_size = NIC_MAX_RSS_IDR_TBL_SIZE; + hw->tl3_cnt = 256; + hw->tl2_cnt = 64; + hw->tl1_cnt = 2; + hw->tl1_per_bgx = true; + break; + case PCI_SUBSYS_DEVID_81XX_NIC_PF: + hw->bgx_cnt = MAX_BGX_PER_CN81XX; + hw->chans_per_lmac = 8; + hw->chans_per_bgx = 32; + hw->chans_per_rgx = 8; + hw->chans_per_lbk = 24; + hw->cpi_cnt = 512; + hw->rssi_cnt = 256; + hw->rss_ind_tbl_size = 32; /* Max RSSI / Max interfaces */ + hw->tl3_cnt = 64; + hw->tl2_cnt = 16; + hw->tl1_cnt = 10; + hw->tl1_per_bgx = false; + break; + case PCI_SUBSYS_DEVID_83XX_NIC_PF: + hw->bgx_cnt = MAX_BGX_PER_CN83XX; + hw->chans_per_lmac = 8; + hw->chans_per_bgx = 32; + hw->chans_per_lbk = 64; + hw->cpi_cnt = 2048; + hw->rssi_cnt = 1024; + hw->rss_ind_tbl_size = 64; /* Max RSSI / Max interfaces */ + hw->tl3_cnt = 256; + hw->tl2_cnt = 64; + hw->tl1_cnt = 18; + hw->tl1_per_bgx = false; + break; + } + hw->tl4_cnt = MAX_QUEUES_PER_QSET * pci_sriov_get_totalvfs(nic->pdev); + + /* Allocate memory for LMAC tracking elements */ + max_lmac = hw->bgx_cnt * MAX_LMAC_PER_BGX; + nic->vf_lmac_map = kmalloc_array(max_lmac, sizeof(u8), GFP_KERNEL); + if (!nic->vf_lmac_map) + goto error; + nic->link = kmalloc_array(max_lmac, sizeof(u8), GFP_KERNEL); + if (!nic->link) + goto error; + nic->duplex = kmalloc_array(max_lmac, sizeof(u8), GFP_KERNEL); + if (!nic->duplex) + goto error; + nic->speed = kmalloc_array(max_lmac, sizeof(u32), GFP_KERNEL); + if (!nic->speed) + goto error; + return 0; + +error: + nic_free_lmacmem(nic); + return -ENOMEM; +} + #define BGX0_BLOCK 8 #define BGX1_BLOCK 9 -static void nic_init_hw(struct nicpf *nic) +static int nic_init_hw(struct nicpf *nic) { - int i; + int i, err; u64 cqm_cfg; + /* Get HW capability info */ + err = nic_get_hw_info(nic); + if (err) + return err; + /* Enable NIC HW block */ nic_reg_write(nic, NIC_PF_CFG, 0x3); /* Enable backpressure */ nic_reg_write(nic, NIC_PF_BP_CFG, (1ULL << 6) | 0x03); - /* Disable TNS mode on both interfaces */ - nic_reg_write(nic, NIC_PF_INTF_0_1_SEND_CFG, - (NIC_TNS_BYPASS_MODE << 7) | BGX0_BLOCK); - nic_reg_write(nic, NIC_PF_INTF_0_1_SEND_CFG | (1 << 8), - (NIC_TNS_BYPASS_MODE << 7) | BGX1_BLOCK); + /* TNS and TNS bypass modes are present only on 88xx */ + if (nic->pdev->subsystem_device == PCI_SUBSYS_DEVID_88XX_NIC_PF) { + /* Disable TNS mode on both interfaces */ + nic_reg_write(nic, NIC_PF_INTF_0_1_SEND_CFG, + (NIC_TNS_BYPASS_MODE << 7) | BGX0_BLOCK); + nic_reg_write(nic, NIC_PF_INTF_0_1_SEND_CFG | (1 << 8), + (NIC_TNS_BYPASS_MODE << 7) | BGX1_BLOCK); + } + nic_reg_write(nic, NIC_PF_INTF_0_1_BP_CFG, (1ULL << 63) | BGX0_BLOCK); nic_reg_write(nic, NIC_PF_INTF_0_1_BP_CFG + (1 << 8), @@ -346,11 +481,14 @@ static void nic_init_hw(struct nicpf *nic) cqm_cfg = nic_reg_read(nic, NIC_PF_CQM_CFG); if (cqm_cfg < NICPF_CQM_MIN_DROP_LEVEL) nic_reg_write(nic, NIC_PF_CQM_CFG, NICPF_CQM_MIN_DROP_LEVEL); + + return 0; } /* Channel parse index configuration */ static void nic_config_cpi(struct nicpf *nic, struct cpi_cfg_msg *cfg) { + struct hw_info *hw = nic->hw; u32 vnic, bgx, lmac, chan; u32 padd, cpi_count = 0; u64 cpi_base, cpi, rssi_base, rssi; @@ -360,9 +498,9 @@ static void nic_config_cpi(struct nicpf *nic, struct cpi_cfg_msg *cfg) bgx = NIC_GET_BGX_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vnic]); lmac = NIC_GET_LMAC_FROM_VF_LMAC_MAP(nic->vf_lmac_map[vnic]); - chan = (lmac * MAX_BGX_CHANS_PER_LMAC) + (bgx * NIC_CHANS_PER_INF); - cpi_base = (lmac * NIC_MAX_CPI_PER_LMAC) + (bgx * NIC_CPI_PER_BGX); - rssi_base = (lmac * nic->rss_ind_tbl_size) + (bgx * NIC_RSSI_PER_BGX); + chan = (lmac * hw->chans_per_lmac) + (bgx * hw->chans_per_bgx); + cpi_base = vnic * NIC_MAX_CPI_PER_LMAC; + rssi_base = vnic * hw->rss_ind_tbl_size; /* Rx channel configuration */ nic_reg_write(nic, NIC_PF_CHAN_0_255_RX_BP_CFG | (chan << 3), @@ -434,7 +572,7 @@ static void nic_send_rss_size(struct nicpf *nic, int vf) msg = (u64 *)&mbx; mbx.rss_size.msg = NIC_MBOX_MSG_RSS_SIZE; - mbx.rss_size.ind_tbl_size = nic->rss_ind_tbl_size; + mbx.rss_size.ind_tbl_size = nic->hw->rss_ind_tbl_size; nic_send_msg_to_vf(nic, vf, &mbx); } @@ -481,7 +619,7 @@ static void nic_config_rss(struct nicpf *nic, struct rss_cfg_msg *cfg) /* 4 level transmit side scheduler configutation * for TNS bypass mode * - * Sample configuration for SQ0 + * Sample configuration for SQ0 on 88xx * VNIC0-SQ0 -> TL4(0) -> TL3[0] -> TL2[0] -> TL1[0] -> BGX0 * VNIC1-SQ0 -> TL4(8) -> TL3[2] -> TL2[0] -> TL1[0] -> BGX0 * VNIC2-SQ0 -> TL4(16) -> TL3[4] -> TL2[1] -> TL1[0] -> BGX0 @@ -494,6 +632,7 @@ static void nic_config_rss(struct nicpf *nic, struct rss_cfg_msg *cfg) static void nic_tx_channel_cfg(struct nicpf *nic, u8 vnic, struct sq_cfg_msg *sq) { + struct hw_info *hw = nic->hw; u32 bgx, lmac, chan; u32 tl2, tl3, tl4; u32 rr_quantum; @@ -512,21 +651,28 @@ static void nic_tx_channel_cfg(struct nicpf *nic, u8 vnic, /* 24 bytes for FCS, IPG and preamble */ rr_quantum = ((NIC_HW_MAX_FRS + 24) / 4); - if (!sq->sqs_mode) { - tl4 = (lmac * NIC_TL4_PER_LMAC) + (bgx * NIC_TL4_PER_BGX); - } else { - for (svf = 0; svf < MAX_SQS_PER_VF; svf++) { - if (nic->vf_sqs[pqs_vnic][svf] == vnic) - break; + /* For 88xx 0-511 TL4 transmits via BGX0 and + * 512-1023 TL4s transmit via BGX1. + */ + if (hw->tl1_per_bgx) { + tl4 = bgx * (hw->tl4_cnt / hw->bgx_cnt); + if (!sq->sqs_mode) { + tl4 += (lmac * MAX_QUEUES_PER_QSET); + } else { + for (svf = 0; svf < MAX_SQS_PER_VF; svf++) { + if (nic->vf_sqs[pqs_vnic][svf] == vnic) + break; + } + tl4 += (MAX_LMAC_PER_BGX * MAX_QUEUES_PER_QSET); + tl4 += (lmac * MAX_QUEUES_PER_QSET * MAX_SQS_PER_VF); + tl4 += (svf * MAX_QUEUES_PER_QSET); } - tl4 = (MAX_LMAC_PER_BGX * NIC_TL4_PER_LMAC); - tl4 += (lmac * NIC_TL4_PER_LMAC * MAX_SQS_PER_VF); - tl4 += (svf * NIC_TL4_PER_LMAC); - tl4 += (bgx * NIC_TL4_PER_BGX); + } else { + tl4 = (vnic * MAX_QUEUES_PER_QSET); } tl4 += sq_idx; - tl3 = tl4 / (NIC_MAX_TL4 / NIC_MAX_TL3); + tl3 = tl4 / (hw->tl4_cnt / hw->tl3_cnt); nic_reg_write(nic, NIC_PF_QSET_0_127_SQ_0_7_CFG2 | ((u64)vnic << NIC_QS_ID_SHIFT) | ((u32)sq_idx << NIC_Q_NUM_SHIFT), tl4); @@ -534,8 +680,19 @@ static void nic_tx_channel_cfg(struct nicpf *nic, u8 vnic, ((u64)vnic << 27) | ((u32)sq_idx << 24) | rr_quantum); nic_reg_write(nic, NIC_PF_TL3_0_255_CFG | (tl3 << 3), rr_quantum); - chan = (lmac * MAX_BGX_CHANS_PER_LMAC) + (bgx * NIC_CHANS_PER_INF); - nic_reg_write(nic, NIC_PF_TL3_0_255_CHAN | (tl3 << 3), chan); + + /* On 88xx 0-127 channels are for BGX0 and + * 127-255 channels for BGX1. + * + * On 81xx/83xx TL3_CHAN reg should be configured with channel + * within LMAC i.e 0-7 and not the actual channel number like on 88xx + */ + chan = (lmac * hw->chans_per_lmac) + (bgx * hw->chans_per_bgx); + if (hw->tl1_per_bgx) + nic_reg_write(nic, NIC_PF_TL3_0_255_CHAN | (tl3 << 3), chan); + else + nic_reg_write(nic, NIC_PF_TL3_0_255_CHAN | (tl3 << 3), 0); + /* Enable backpressure on the channel */ nic_reg_write(nic, NIC_PF_CHAN_0_255_TX_CFG | (chan << 3), 1); @@ -544,6 +701,16 @@ static void nic_tx_channel_cfg(struct nicpf *nic, u8 vnic, nic_reg_write(nic, NIC_PF_TL2_0_63_CFG | (tl2 << 3), rr_quantum); /* No priorities as of now */ nic_reg_write(nic, NIC_PF_TL2_0_63_PRI | (tl2 << 3), 0x00); + + /* Unlike 88xx where TL2s 0-31 transmits to TL1 '0' and rest to TL1 '1' + * on 81xx/83xx TL2 needs to be configured to transmit to one of the + * possible LMACs. + * + * This register doesn't exist on 88xx. + */ + if (!hw->tl1_per_bgx) + nic_reg_write(nic, NIC_PF_TL2_LMAC | (tl2 << 3), + lmac + (bgx * MAX_LMAC_PER_BGX)); } /* Send primary nicvf pointer to secondary QS's VF */ @@ -615,7 +782,7 @@ static int nic_config_loopback(struct nicpf *nic, struct set_loopback *lbk) { int bgx_idx, lmac_idx; - if (lbk->vf_id > MAX_LMAC) + if (lbk->vf_id >= nic->num_vf_en) return -1; bgx_idx = NIC_GET_BGX_FROM_VF_LMAC_MAP(nic->vf_lmac_map[lbk->vf_id]); @@ -626,6 +793,67 @@ static int nic_config_loopback(struct nicpf *nic, struct set_loopback *lbk) return 0; } +/* Reset statistics counters */ +static int nic_reset_stat_counters(struct nicpf *nic, + int vf, struct reset_stat_cfg *cfg) +{ + int i, stat, qnum; + u64 reg_addr; + + for (i = 0; i < RX_STATS_ENUM_LAST; i++) { + if (cfg->rx_stat_mask & BIT(i)) { + reg_addr = NIC_PF_VNIC_0_127_RX_STAT_0_13 | + (vf << NIC_QS_ID_SHIFT) | + (i << 3); + nic_reg_write(nic, reg_addr, 0); + } + } + + for (i = 0; i < TX_STATS_ENUM_LAST; i++) { + if (cfg->tx_stat_mask & BIT(i)) { + reg_addr = NIC_PF_VNIC_0_127_TX_STAT_0_4 | + (vf << NIC_QS_ID_SHIFT) | + (i << 3); + nic_reg_write(nic, reg_addr, 0); + } + } + + for (i = 0; i <= 15; i++) { + qnum = i >> 1; + stat = i & 1 ? 1 : 0; + reg_addr = (vf << NIC_QS_ID_SHIFT) | + (qnum << NIC_Q_NUM_SHIFT) | (stat << 3); + if (cfg->rq_stat_mask & BIT(i)) { + reg_addr |= NIC_PF_QSET_0_127_RQ_0_7_STAT_0_1; + nic_reg_write(nic, reg_addr, 0); + } + if (cfg->sq_stat_mask & BIT(i)) { + reg_addr |= NIC_PF_QSET_0_127_SQ_0_7_STAT_0_1; + nic_reg_write(nic, reg_addr, 0); + } + } + return 0; +} + +static void nic_enable_tunnel_parsing(struct nicpf *nic, int vf) +{ + u64 prot_def = (IPV6_PROT << 32) | (IPV4_PROT << 16) | ET_PROT; + u64 vxlan_prot_def = (IPV6_PROT_DEF << 32) | + (IPV4_PROT_DEF) << 16 | ET_PROT_DEF; + + /* Configure tunnel parsing parameters */ + nic_reg_write(nic, NIC_PF_RX_GENEVE_DEF, + (1ULL << 63 | UDP_GENEVE_PORT_NUM)); + nic_reg_write(nic, NIC_PF_RX_GENEVE_PROT_DEF, + ((7ULL << 61) | prot_def)); + nic_reg_write(nic, NIC_PF_RX_NVGRE_PROT_DEF, + ((7ULL << 61) | prot_def)); + nic_reg_write(nic, NIC_PF_RX_VXLAN_DEF_0_1, + ((1ULL << 63) | UDP_VXLAN_PORT_NUM)); + nic_reg_write(nic, NIC_PF_RX_VXLAN_PROT_DEF, + ((0xfULL << 60) | vxlan_prot_def)); +} + static void nic_enable_vf(struct nicpf *nic, int vf, bool enable) { int bgx, lmac; @@ -664,18 +892,17 @@ static void nic_handle_mbx_intr(struct nicpf *nic, int vf) mbx_addr += sizeof(u64); } - dev_dbg(&nic->pdev->dev, "%s: Mailbox msg %d from VF%d\n", + dev_dbg(&nic->pdev->dev, "%s: Mailbox msg 0x%02x from VF%d\n", __func__, mbx.msg.msg, vf); switch (mbx.msg.msg) { case NIC_MBOX_MSG_READY: nic_mbx_send_ready(nic, vf); - if (vf < MAX_LMAC) { + if (vf < nic->num_vf_en) { nic->link[vf] = 0; nic->duplex[vf] = 0; nic->speed[vf] = 0; } - ret = 1; - break; + goto unlock; case NIC_MBOX_MSG_QS_CFG: reg_addr = NIC_PF_QSET_0_127_CFG | (mbx.qs.num << NIC_QS_ID_SHIFT); @@ -693,6 +920,15 @@ static void nic_handle_mbx_intr(struct nicpf *nic, int vf) (mbx.rq.qs_num << NIC_QS_ID_SHIFT) | (mbx.rq.rq_num << NIC_Q_NUM_SHIFT); nic_reg_write(nic, reg_addr, mbx.rq.cfg); + /* Enable CQE_RX2_S extension in CQE_RX descriptor. + * This gets appended by default on 81xx/83xx chips, + * for consistency enabling the same on 88xx pass2 + * where this is introduced. + */ + if (pass2_silicon(nic->pdev)) + nic_reg_write(nic, NIC_PF_RX_CFG, 0x01); + if (!pass1_silicon(nic->pdev)) + nic_enable_tunnel_parsing(nic, vf); break; case NIC_MBOX_MSG_RQ_BP_CFG: reg_addr = NIC_PF_QSET_0_127_RQ_0_7_BP_CFG | @@ -717,8 +953,10 @@ static void nic_handle_mbx_intr(struct nicpf *nic, int vf) nic_tx_channel_cfg(nic, mbx.qs.num, &mbx.sq); break; case NIC_MBOX_MSG_SET_MAC: - if (vf >= nic->num_vf_en) + if (vf >= nic->num_vf_en) { + ret = -1; /* NACK */ break; + } lmac = mbx.mac.vf_id; bgx = NIC_GET_BGX_FROM_VF_LMAC_MAP(nic->vf_lmac_map[lmac]); lmac = NIC_GET_LMAC_FROM_VF_LMAC_MAP(nic->vf_lmac_map[lmac]); @@ -767,25 +1005,38 @@ static void nic_handle_mbx_intr(struct nicpf *nic, int vf) case NIC_MBOX_MSG_LOOPBACK: ret = nic_config_loopback(nic, &mbx.lbk); break; + case NIC_MBOX_MSG_RESET_STAT_COUNTER: + ret = nic_reset_stat_counters(nic, vf, &mbx.reset_stat); + break; default: dev_err(&nic->pdev->dev, "Invalid msg from VF%d, msg 0x%x\n", vf, mbx.msg.msg); break; } - if (!ret) + if (!ret) { nic_mbx_send_ack(nic, vf); - else if (mbx.msg.msg != NIC_MBOX_MSG_READY) + } else if (mbx.msg.msg != NIC_MBOX_MSG_READY) { + dev_err(&nic->pdev->dev, "NACK for MBOX 0x%02x from VF %d\n", + mbx.msg.msg, vf); nic_mbx_send_nack(nic, vf); + } unlock: nic->mbx_lock[vf] = false; } -static void nic_mbx_intr_handler (struct nicpf *nic, int mbx) +static irqreturn_t nic_mbx_intr_handler(int irq, void *nic_irq) { + struct nicpf *nic = (struct nicpf *)nic_irq; + int mbx; u64 intr; u8 vf, vf_per_mbx_reg = 64; + if (irq == nic->msix_entries[NIC_PF_INTR_ID_MBOX0].vector) + mbx = 0; + else + mbx = 1; + intr = nic_reg_read(nic, NIC_PF_MAILBOX_INT + (mbx << 3)); dev_dbg(&nic->pdev->dev, "PF interrupt Mbox%d 0x%llx\n", mbx, intr); for (vf = 0; vf < vf_per_mbx_reg; vf++) { @@ -797,23 +1048,6 @@ static void nic_mbx_intr_handler (struct nicpf *nic, int mbx) nic_clear_mbx_intr(nic, vf, mbx); } } -} - -static irqreturn_t nic_mbx0_intr_handler (int irq, void *nic_irq) -{ - struct nicpf *nic = (struct nicpf *)nic_irq; - - nic_mbx_intr_handler(nic, 0); - - return IRQ_HANDLED; -} - -static irqreturn_t nic_mbx1_intr_handler (int irq, void *nic_irq) -{ - struct nicpf *nic = (struct nicpf *)nic_irq; - - nic_mbx_intr_handler(nic, 1); - return IRQ_HANDLED; } @@ -821,7 +1055,13 @@ static int nic_enable_msix(struct nicpf *nic) { int i, ret; - nic->num_vec = NIC_PF_MSIX_VECTORS; + nic->num_vec = pci_msix_vec_count(nic->pdev); + + nic->msix_entries = kmalloc_array(nic->num_vec, + sizeof(struct msix_entry), + GFP_KERNEL); + if (!nic->msix_entries) + return -ENOMEM; for (i = 0; i < nic->num_vec; i++) nic->msix_entries[i].entry = i; @@ -829,8 +1069,9 @@ static int nic_enable_msix(struct nicpf *nic) ret = pci_enable_msix(nic->pdev, nic->msix_entries, nic->num_vec); if (ret) { dev_err(&nic->pdev->dev, - "Request for #%d msix vectors failed\n", - nic->num_vec); + "Request for #%d msix vectors failed, returned %d\n", + nic->num_vec, ret); + kfree(nic->msix_entries); return ret; } @@ -842,6 +1083,7 @@ static void nic_disable_msix(struct nicpf *nic) { if (nic->msix_enabled) { pci_disable_msix(nic->pdev); + kfree(nic->msix_entries); nic->msix_enabled = 0; nic->num_vec = 0; } @@ -860,27 +1102,26 @@ static void nic_free_all_interrupts(struct nicpf *nic) static int nic_register_interrupts(struct nicpf *nic) { - int ret; + int i, ret; /* Enable MSI-X */ ret = nic_enable_msix(nic); if (ret) return ret; - /* Register mailbox interrupt handlers */ - ret = request_irq(nic->msix_entries[NIC_PF_INTR_ID_MBOX0].vector, - nic_mbx0_intr_handler, 0, "NIC Mbox0", nic); - if (ret) - goto fail; - - nic->irq_allocated[NIC_PF_INTR_ID_MBOX0] = true; + /* Register mailbox interrupt handler */ + for (i = NIC_PF_INTR_ID_MBOX0; i < nic->num_vec; i++) { + sprintf(nic->irq_name[i], + "NICPF Mbox%d", (i - NIC_PF_INTR_ID_MBOX0)); - ret = request_irq(nic->msix_entries[NIC_PF_INTR_ID_MBOX1].vector, - nic_mbx1_intr_handler, 0, "NIC Mbox1", nic); - if (ret) - goto fail; + ret = request_irq(nic->msix_entries[i].vector, + nic_mbx_intr_handler, 0, + nic->irq_name[i], nic); + if (ret) + goto fail; - nic->irq_allocated[NIC_PF_INTR_ID_MBOX1] = true; + nic->irq_allocated[i] = true; + } /* Enable mailbox interrupt */ nic_enable_mbx_intr(nic); @@ -889,6 +1130,7 @@ static int nic_register_interrupts(struct nicpf *nic) fail: dev_err(&nic->pdev->dev, "Request irq failed\n"); nic_free_all_interrupts(nic); + nic_disable_msix(nic); return ret; } @@ -903,6 +1145,12 @@ static int nic_num_sqs_en(struct nicpf *nic, int vf_en) int pos, sqs_per_vf = MAX_SQS_PER_VF_SINGLE_NODE; u16 total_vf; + /* Secondary Qsets are needed only if CPU count is + * morethan MAX_QUEUES_PER_QSET. + */ + if (num_online_cpus() <= MAX_QUEUES_PER_QSET) + return 0; + /* Check if its a multi-node environment */ if (nr_node_ids > 1) sqs_per_vf = MAX_SQS_PER_VF; @@ -1008,6 +1256,12 @@ static int nic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) if (!nic) return -ENOMEM; + nic->hw = devm_kzalloc(dev, sizeof(struct hw_info), GFP_KERNEL); + if (!nic->hw) { + devm_kfree(dev, nic); + return -ENOMEM; + } + pci_set_drvdata(pdev, nic); nic->pdev = pdev; @@ -1047,13 +1301,12 @@ static int nic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) nic->node = nic_get_node_id(pdev); - nic_set_lmac_vf_mapping(nic); - /* Initialize hardware */ - nic_init_hw(nic); + err = nic_init_hw(nic); + if (err) + goto err_release_regions; - /* Set RSS TBL size for each VF */ - nic->rss_ind_tbl_size = NIC_MAX_RSS_IDR_TBL_SIZE; + nic_set_lmac_vf_mapping(nic); /* Register interrupts */ err = nic_register_interrupts(nic); @@ -1086,6 +1339,9 @@ err_unregister_interrupts: err_release_regions: pci_release_regions(pdev); err_disable_device: + nic_free_lmacmem(nic); + devm_kfree(dev, nic->hw); + devm_kfree(dev, nic); pci_disable_device(pdev); pci_set_drvdata(pdev, NULL); return err; @@ -1106,6 +1362,11 @@ static void nic_remove(struct pci_dev *pdev) nic_unregister_interrupts(nic); pci_release_regions(pdev); + + nic_free_lmacmem(nic); + devm_kfree(&pdev->dev, nic->hw); + devm_kfree(&pdev->dev, nic); + pci_disable_device(pdev); pci_set_drvdata(pdev, NULL); } diff --git a/drivers/net/ethernet/cavium/thunder/nic_reg.h b/drivers/net/ethernet/cavium/thunder/nic_reg.h index afb10e326b4f..db9c632d0157 100644 --- a/drivers/net/ethernet/cavium/thunder/nic_reg.h +++ b/drivers/net/ethernet/cavium/thunder/nic_reg.h @@ -36,6 +36,20 @@ #define NIC_PF_MAILBOX_ENA_W1C (0x0450) #define NIC_PF_MAILBOX_ENA_W1S (0x0470) #define NIC_PF_RX_ETYPE_0_7 (0x0500) +#define NIC_PF_RX_GENEVE_DEF (0x0580) +#define UDP_GENEVE_PORT_NUM 0x17C1ULL +#define NIC_PF_RX_GENEVE_PROT_DEF (0x0588) +#define IPV6_PROT 0x86DDULL +#define IPV4_PROT 0x800ULL +#define ET_PROT 0x6558ULL +#define NIC_PF_RX_NVGRE_PROT_DEF (0x0598) +#define NIC_PF_RX_VXLAN_DEF_0_1 (0x05A0) +#define UDP_VXLAN_PORT_NUM 0x12B5 +#define NIC_PF_RX_VXLAN_PROT_DEF (0x05B0) +#define IPV6_PROT_DEF 0x2ULL +#define IPV4_PROT_DEF 0x1ULL +#define ET_PROT_DEF 0x3ULL +#define NIC_PF_RX_CFG (0x05D0) #define NIC_PF_PKIND_0_15_CFG (0x0600) #define NIC_PF_ECC0_FLIP0 (0x1000) #define NIC_PF_ECC1_FLIP0 (0x1008) @@ -103,6 +117,7 @@ #define NIC_PF_SW_SYNC_RX_DONE (0x490008) #define NIC_PF_TL2_0_63_CFG (0x500000) #define NIC_PF_TL2_0_63_PRI (0x520000) +#define NIC_PF_TL2_LMAC (0x540000) #define NIC_PF_TL2_0_63_SH_STATUS (0x580000) #define NIC_PF_TL3A_0_63_CFG (0x5F0000) #define NIC_PF_TL3_0_255_CFG (0x600000) diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c index a19e73f11d73..06c014edf762 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c @@ -29,10 +29,20 @@ static const struct pci_device_id nicvf_id_table[] = { { PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, PCI_DEVICE_ID_THUNDER_NIC_VF, - PCI_VENDOR_ID_CAVIUM, 0xA134) }, + PCI_VENDOR_ID_CAVIUM, + PCI_SUBSYS_DEVID_88XX_NIC_VF) }, { PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, PCI_DEVICE_ID_THUNDER_PASS1_NIC_VF, - PCI_VENDOR_ID_CAVIUM, 0xA11E) }, + PCI_VENDOR_ID_CAVIUM, + PCI_SUBSYS_DEVID_88XX_PASS1_NIC_VF) }, + { PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, + PCI_DEVICE_ID_THUNDER_NIC_VF, + PCI_VENDOR_ID_CAVIUM, + PCI_SUBSYS_DEVID_81XX_NIC_VF) }, + { PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, + PCI_DEVICE_ID_THUNDER_NIC_VF, + PCI_VENDOR_ID_CAVIUM, + PCI_SUBSYS_DEVID_83XX_NIC_VF) }, { 0, } /* end of table */ }; @@ -134,15 +144,19 @@ int nicvf_send_msg_to_pf(struct nicvf *nic, union nic_mbx *mbx) /* Wait for previous message to be acked, timeout 2sec */ while (!nic->pf_acked) { - if (nic->pf_nacked) + if (nic->pf_nacked) { + netdev_err(nic->netdev, + "PF NACK to mbox msg 0x%02x from VF%d\n", + (mbx->msg.msg & 0xFF), nic->vf_id); return -EINVAL; + } msleep(sleep); if (nic->pf_acked) break; timeout -= sleep; if (!timeout) { netdev_err(nic->netdev, - "PF didn't ack to mbox msg %d from VF%d\n", + "PF didn't ACK to mbox msg 0x%02x from VF%d\n", (mbx->msg.msg & 0xFF), nic->vf_id); return -EBUSY; } @@ -352,13 +366,7 @@ static int nicvf_rss_init(struct nicvf *nic) rss->enable = true; - /* Using the HW reset value for now */ - rss->key[0] = 0xFEED0BADFEED0BADULL; - rss->key[1] = 0xFEED0BADFEED0BADULL; - rss->key[2] = 0xFEED0BADFEED0BADULL; - rss->key[3] = 0xFEED0BADFEED0BADULL; - rss->key[4] = 0xFEED0BADFEED0BADULL; - + netdev_rss_key_fill(rss->key, RSS_HASH_KEY_SIZE * sizeof(u64)); nicvf_set_rss_key(nic); rss->cfg = RSS_IP_HASH_ENA | RSS_TCP_HASH_ENA | RSS_UDP_HASH_ENA; @@ -507,7 +515,8 @@ static int nicvf_init_resources(struct nicvf *nic) static void nicvf_snd_pkt_handler(struct net_device *netdev, struct cmp_queue *cq, - struct cqe_send_t *cqe_tx, int cqe_type) + struct cqe_send_t *cqe_tx, + int cqe_type, int budget) { struct sk_buff *skb = NULL; struct nicvf *nic = netdev_priv(netdev); @@ -531,7 +540,7 @@ static void nicvf_snd_pkt_handler(struct net_device *netdev, if (skb) { nicvf_put_sq_desc(sq, hdr->subdesc_cnt + 1); prefetch(skb); - dev_consume_skb_any(skb); + napi_consume_skb(skb, budget); sq->skbuff[cqe_tx->sqe_ptr] = (u64)NULL; } else { /* In case of HW TSO, HW sends a CQE for each segment of a TSO @@ -686,7 +695,8 @@ loop: break; case CQE_TYPE_SEND: nicvf_snd_pkt_handler(netdev, cq, - (void *)cq_desc, CQE_TYPE_SEND); + (void *)cq_desc, CQE_TYPE_SEND, + budget); tx_done++; break; case CQE_TYPE_INVALID: @@ -928,16 +938,19 @@ static int nicvf_register_interrupts(struct nicvf *nic) int vector; for_each_cq_irq(irq) - sprintf(nic->irq_name[irq], "NICVF%d CQ%d", - nic->vf_id, irq); + sprintf(nic->irq_name[irq], "%s-rxtx-%d", + nic->pnicvf->netdev->name, + nicvf_netdev_qidx(nic, irq)); for_each_sq_irq(irq) - sprintf(nic->irq_name[irq], "NICVF%d SQ%d", - nic->vf_id, irq - NICVF_INTR_ID_SQ); + sprintf(nic->irq_name[irq], "%s-sq-%d", + nic->pnicvf->netdev->name, + nicvf_netdev_qidx(nic, irq - NICVF_INTR_ID_SQ)); for_each_rbdr_irq(irq) - sprintf(nic->irq_name[irq], "NICVF%d RBDR%d", - nic->vf_id, irq - NICVF_INTR_ID_RBDR); + sprintf(nic->irq_name[irq], "%s-rbdr-%d", + nic->pnicvf->netdev->name, + nic->sqs_mode ? (nic->sqs_id + 1) : 0); /* Register CQ interrupts */ for (irq = 0; irq < nic->qs->cq_cnt; irq++) { @@ -961,8 +974,9 @@ static int nicvf_register_interrupts(struct nicvf *nic) } /* Register QS error interrupt */ - sprintf(nic->irq_name[NICVF_INTR_ID_QS_ERR], - "NICVF%d Qset error", nic->vf_id); + sprintf(nic->irq_name[NICVF_INTR_ID_QS_ERR], "%s-qset-err-%d", + nic->pnicvf->netdev->name, + nic->sqs_mode ? (nic->sqs_id + 1) : 0); irq = NICVF_INTR_ID_QS_ERR; ret = request_irq(nic->msix_entries[irq].vector, nicvf_qs_err_intr_handler, @@ -1191,7 +1205,7 @@ int nicvf_open(struct net_device *netdev) } /* Check if we got MAC address from PF or else generate a radom MAC */ - if (is_zero_ether_addr(netdev->dev_addr)) { + if (!nic->sqs_mode && is_zero_ether_addr(netdev->dev_addr)) { eth_hw_addr_random(netdev); nicvf_hw_set_mac_addr(nic, netdev); } @@ -1527,14 +1541,13 @@ static int nicvf_probe(struct pci_dev *pdev, const struct pci_device_id *ent) goto err_release_regions; } - qcount = MAX_CMP_QUEUES_PER_QS; + qcount = netif_get_num_default_rss_queues(); /* Restrict multiqset support only for host bound VFs */ if (pdev->is_virtfn) { /* Set max number of queues per VF */ - qcount = roundup(num_online_cpus(), MAX_CMP_QUEUES_PER_QS); - qcount = min(qcount, - (MAX_SQS_PER_VF + 1) * MAX_CMP_QUEUES_PER_QS); + qcount = min_t(int, num_online_cpus(), + (MAX_SQS_PER_VF + 1) * MAX_CMP_QUEUES_PER_QS); } netdev = alloc_etherdev_mqs(sizeof(struct nicvf), qcount, qcount); diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c index 0ff8e60deccb..7d90856c9783 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c @@ -479,6 +479,16 @@ void nicvf_config_vlan_stripping(struct nicvf *nic, netdev_features_t features) NIC_QSET_RQ_GEN_CFG, 0, rq_cfg); } +static void nicvf_reset_rcv_queue_stats(struct nicvf *nic) +{ + union nic_mbx mbx = {}; + + /* Reset all RXQ's stats */ + mbx.reset_stat.msg = NIC_MBOX_MSG_RESET_STAT_COUNTER; + mbx.reset_stat.rq_stat_mask = 0xFFFF; + nicvf_send_msg_to_pf(nic, &mbx); +} + /* Configures receive queue */ static void nicvf_rcv_queue_config(struct nicvf *nic, struct queue_set *qs, int qidx, bool enable) @@ -762,10 +772,10 @@ int nicvf_set_qset_resources(struct nicvf *nic) nic->qs = qs; /* Set count of each queue */ - qs->rbdr_cnt = RBDR_CNT; - qs->rq_cnt = RCV_QUEUE_CNT; - qs->sq_cnt = SND_QUEUE_CNT; - qs->cq_cnt = CMP_QUEUE_CNT; + qs->rbdr_cnt = DEFAULT_RBDR_CNT; + qs->rq_cnt = min_t(u8, MAX_RCV_QUEUES_PER_QS, num_online_cpus()); + qs->sq_cnt = min_t(u8, MAX_SND_QUEUES_PER_QS, num_online_cpus()); + qs->cq_cnt = max_t(u8, qs->rq_cnt, qs->sq_cnt); /* Set queue lengths */ qs->rbdr_len = RCV_BUF_COUNT; @@ -812,6 +822,11 @@ int nicvf_config_data_transfer(struct nicvf *nic, bool enable) nicvf_free_resources(nic); } + /* Reset RXQ's stats. + * SQ's stats will get reset automatically once SQ is reset. + */ + nicvf_reset_rcv_queue_stats(nic); + return 0; } @@ -1184,13 +1199,23 @@ struct sk_buff *nicvf_get_rcv_skb(struct nicvf *nic, struct cqe_rx_t *cqe_rx) int frag; int payload_len = 0; struct sk_buff *skb = NULL; - struct sk_buff *skb_frag = NULL; - struct sk_buff *prev_frag = NULL; + struct page *page; + int offset; u16 *rb_lens = NULL; u64 *rb_ptrs = NULL; rb_lens = (void *)cqe_rx + (3 * sizeof(u64)); - rb_ptrs = (void *)cqe_rx + (6 * sizeof(u64)); + /* Except 88xx pass1 on all other chips CQE_RX2_S is added to + * CQE_RX at word6, hence buffer pointers move by word + * + * Use existing 'hw_tso' flag which will be set for all chips + * except 88xx pass1 instead of a additional cache line + * access (or miss) by using pci dev's revision. + */ + if (!nic->hw_tso) + rb_ptrs = (void *)cqe_rx + (6 * sizeof(u64)); + else + rb_ptrs = (void *)cqe_rx + (7 * sizeof(u64)); netdev_dbg(nic->netdev, "%s rb_cnt %d rb0_ptr %llx rb0_sz %d\n", __func__, cqe_rx->rb_cnt, cqe_rx->rb0_ptr, cqe_rx->rb0_sz); @@ -1208,22 +1233,10 @@ struct sk_buff *nicvf_get_rcv_skb(struct nicvf *nic, struct cqe_rx_t *cqe_rx) skb_put(skb, payload_len); } else { /* Add fragments */ - skb_frag = nicvf_rb_ptr_to_skb(nic, *rb_ptrs, - payload_len); - if (!skb_frag) { - dev_kfree_skb(skb); - return NULL; - } - - if (!skb_shinfo(skb)->frag_list) - skb_shinfo(skb)->frag_list = skb_frag; - else - prev_frag->next = skb_frag; - - prev_frag = skb_frag; - skb->len += payload_len; - skb->data_len += payload_len; - skb_frag->len = payload_len; + page = virt_to_page(phys_to_virt(*rb_ptrs)); + offset = phys_to_virt(*rb_ptrs) - page_address(page); + skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, page, + offset, payload_len, RCV_FRAG_LEN); } /* Next buffer pointer */ rb_ptrs++; diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.h b/drivers/net/ethernet/cavium/thunder/nicvf_queues.h index 6673e1133523..869f3386028b 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.h +++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.h @@ -57,10 +57,7 @@ #define CMP_QUEUE_SIZE6 6ULL /* 64K entries */ /* Default queue count per QS, its lengths and threshold values */ -#define RBDR_CNT 1 -#define RCV_QUEUE_CNT 8 -#define SND_QUEUE_CNT 8 -#define CMP_QUEUE_CNT 8 /* Max of RCV and SND qcount */ +#define DEFAULT_RBDR_CNT 1 #define SND_QSIZE SND_QUEUE_SIZE2 #define SND_QUEUE_LEN (1ULL << (SND_QSIZE + 10)) diff --git a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c index 63a39ac97d53..8bbaedbb7b94 100644 --- a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c +++ b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c @@ -28,6 +28,9 @@ struct lmac { struct bgx *bgx; int dmac; u8 mac[ETH_ALEN]; + u8 lmac_type; + u8 lane_to_sds; + bool use_training; bool link_up; int lmacid; /* ID within BGX */ int lmacid_bd; /* ID on board */ @@ -43,14 +46,13 @@ struct lmac { struct bgx { u8 bgx_id; - u8 qlm_mode; struct lmac lmac[MAX_LMAC_PER_BGX]; int lmac_count; - int lmac_type; - int lane_to_sds; - int use_training; + u8 max_lmac; void __iomem *reg_base; struct pci_dev *pdev; + bool is_dlm; + bool is_rgx; }; static struct bgx *bgx_vnic[MAX_BGX_THUNDER]; @@ -61,6 +63,7 @@ static int bgx_xaui_check_link(struct lmac *lmac); /* Supported devices */ static const struct pci_device_id bgx_id_table[] = { { PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, PCI_DEVICE_ID_THUNDER_BGX) }, + { PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, PCI_DEVICE_ID_THUNDER_RGX) }, { 0, } /* end of table */ }; @@ -124,8 +127,8 @@ unsigned bgx_get_map(int node) int i; unsigned map = 0; - for (i = 0; i < MAX_BGX_PER_CN88XX; i++) { - if (bgx_vnic[(node * MAX_BGX_PER_CN88XX) + i]) + for (i = 0; i < MAX_BGX_PER_NODE; i++) { + if (bgx_vnic[(node * MAX_BGX_PER_NODE) + i]) map |= (1 << i); } @@ -138,7 +141,7 @@ int bgx_get_lmac_count(int node, int bgx_idx) { struct bgx *bgx; - bgx = bgx_vnic[(node * MAX_BGX_PER_CN88XX) + bgx_idx]; + bgx = bgx_vnic[(node * MAX_BGX_PER_NODE) + bgx_idx]; if (bgx) return bgx->lmac_count; @@ -153,7 +156,7 @@ void bgx_get_lmac_link_state(int node, int bgx_idx, int lmacid, void *status) struct bgx *bgx; struct lmac *lmac; - bgx = bgx_vnic[(node * MAX_BGX_PER_CN88XX) + bgx_idx]; + bgx = bgx_vnic[(node * MAX_BGX_PER_NODE) + bgx_idx]; if (!bgx) return; @@ -166,7 +169,7 @@ EXPORT_SYMBOL(bgx_get_lmac_link_state); const u8 *bgx_get_lmac_mac(int node, int bgx_idx, int lmacid) { - struct bgx *bgx = bgx_vnic[(node * MAX_BGX_PER_CN88XX) + bgx_idx]; + struct bgx *bgx = bgx_vnic[(node * MAX_BGX_PER_NODE) + bgx_idx]; if (bgx) return bgx->lmac[lmacid].mac; @@ -177,7 +180,7 @@ EXPORT_SYMBOL(bgx_get_lmac_mac); void bgx_set_lmac_mac(int node, int bgx_idx, int lmacid, const u8 *mac) { - struct bgx *bgx = bgx_vnic[(node * MAX_BGX_PER_CN88XX) + bgx_idx]; + struct bgx *bgx = bgx_vnic[(node * MAX_BGX_PER_NODE) + bgx_idx]; if (!bgx) return; @@ -188,11 +191,13 @@ EXPORT_SYMBOL(bgx_set_lmac_mac); void bgx_lmac_rx_tx_enable(int node, int bgx_idx, int lmacid, bool enable) { - struct bgx *bgx = bgx_vnic[(node * MAX_BGX_PER_CN88XX) + bgx_idx]; + struct bgx *bgx = bgx_vnic[(node * MAX_BGX_PER_NODE) + bgx_idx]; + struct lmac *lmac; u64 cfg; if (!bgx) return; + lmac = &bgx->lmac[lmacid]; cfg = bgx_reg_read(bgx, lmacid, BGX_CMRX_CFG); if (enable) @@ -200,6 +205,9 @@ void bgx_lmac_rx_tx_enable(int node, int bgx_idx, int lmacid, bool enable) else cfg &= ~(CMR_PKT_RX_EN | CMR_PKT_TX_EN); bgx_reg_write(bgx, lmacid, BGX_CMRX_CFG, cfg); + + if (bgx->is_rgx) + xcv_setup_link(enable ? lmac->link_up : 0, lmac->last_speed); } EXPORT_SYMBOL(bgx_lmac_rx_tx_enable); @@ -266,9 +274,12 @@ static void bgx_sgmii_change_link_state(struct lmac *lmac) port_cfg = bgx_reg_read(bgx, lmac->lmacid, BGX_GMP_GMI_PRTX_CFG); - /* renable lmac */ + /* Re-enable lmac */ cmr_cfg |= CMR_EN; bgx_reg_write(bgx, lmac->lmacid, BGX_CMRX_CFG, cmr_cfg); + + if (bgx->is_rgx && (cmr_cfg & (CMR_PKT_RX_EN | CMR_PKT_TX_EN))) + xcv_setup_link(lmac->link_up, lmac->last_speed); } static void bgx_lmac_handler(struct net_device *netdev) @@ -314,7 +325,7 @@ u64 bgx_get_rx_stats(int node, int bgx_idx, int lmac, int idx) { struct bgx *bgx; - bgx = bgx_vnic[(node * MAX_BGX_PER_CN88XX) + bgx_idx]; + bgx = bgx_vnic[(node * MAX_BGX_PER_NODE) + bgx_idx]; if (!bgx) return 0; @@ -328,7 +339,7 @@ u64 bgx_get_tx_stats(int node, int bgx_idx, int lmac, int idx) { struct bgx *bgx; - bgx = bgx_vnic[(node * MAX_BGX_PER_CN88XX) + bgx_idx]; + bgx = bgx_vnic[(node * MAX_BGX_PER_NODE) + bgx_idx]; if (!bgx) return 0; @@ -356,7 +367,7 @@ void bgx_lmac_internal_loopback(int node, int bgx_idx, struct lmac *lmac; u64 cfg; - bgx = bgx_vnic[(node * MAX_BGX_PER_CN88XX) + bgx_idx]; + bgx = bgx_vnic[(node * MAX_BGX_PER_NODE) + bgx_idx]; if (!bgx) return; @@ -379,8 +390,9 @@ void bgx_lmac_internal_loopback(int node, int bgx_idx, } EXPORT_SYMBOL(bgx_lmac_internal_loopback); -static int bgx_lmac_sgmii_init(struct bgx *bgx, int lmacid) +static int bgx_lmac_sgmii_init(struct bgx *bgx, struct lmac *lmac) { + int lmacid = lmac->lmacid; u64 cfg; bgx_reg_modify(bgx, lmacid, BGX_GMP_GMI_TXX_THRESH, 0x30); @@ -409,18 +421,29 @@ static int bgx_lmac_sgmii_init(struct bgx *bgx, int lmacid) cfg |= (PCS_MRX_CTL_RST_AN | PCS_MRX_CTL_AN_EN); bgx_reg_write(bgx, lmacid, BGX_GMP_PCS_MRX_CTL, cfg); - if (bgx_poll_reg(bgx, lmacid, BGX_GMP_PCS_MRX_STATUS, - PCS_MRX_STATUS_AN_CPT, false)) { - dev_err(&bgx->pdev->dev, "BGX AN_CPT not completed\n"); - return -1; + if (lmac->lmac_type == BGX_MODE_QSGMII) { + /* Disable disparity check for QSGMII */ + cfg = bgx_reg_read(bgx, lmacid, BGX_GMP_PCS_MISCX_CTL); + cfg &= ~PCS_MISC_CTL_DISP_EN; + bgx_reg_write(bgx, lmacid, BGX_GMP_PCS_MISCX_CTL, cfg); + return 0; + } + + if (lmac->lmac_type == BGX_MODE_SGMII) { + if (bgx_poll_reg(bgx, lmacid, BGX_GMP_PCS_MRX_STATUS, + PCS_MRX_STATUS_AN_CPT, false)) { + dev_err(&bgx->pdev->dev, "BGX AN_CPT not completed\n"); + return -1; + } } return 0; } -static int bgx_lmac_xaui_init(struct bgx *bgx, int lmacid, int lmac_type) +static int bgx_lmac_xaui_init(struct bgx *bgx, struct lmac *lmac) { u64 cfg; + int lmacid = lmac->lmacid; /* Reset SPU */ bgx_reg_modify(bgx, lmacid, BGX_SPUX_CONTROL1, SPU_CTL_RESET); @@ -436,12 +459,14 @@ static int bgx_lmac_xaui_init(struct bgx *bgx, int lmacid, int lmac_type) bgx_reg_modify(bgx, lmacid, BGX_SPUX_CONTROL1, SPU_CTL_LOW_POWER); /* Set interleaved running disparity for RXAUI */ - if (bgx->lmac_type != BGX_MODE_RXAUI) - bgx_reg_modify(bgx, lmacid, - BGX_SPUX_MISC_CONTROL, SPU_MISC_CTL_RX_DIS); - else + if (lmac->lmac_type == BGX_MODE_RXAUI) bgx_reg_modify(bgx, lmacid, BGX_SPUX_MISC_CONTROL, - SPU_MISC_CTL_RX_DIS | SPU_MISC_CTL_INTLV_RDISP); + SPU_MISC_CTL_INTLV_RDISP); + + /* Clear receive packet disable */ + cfg = bgx_reg_read(bgx, lmacid, BGX_SPUX_MISC_CONTROL); + cfg &= ~SPU_MISC_CTL_RX_DIS; + bgx_reg_write(bgx, lmacid, BGX_SPUX_MISC_CONTROL, cfg); /* clear all interrupts */ cfg = bgx_reg_read(bgx, lmacid, BGX_SMUX_RX_INT); @@ -451,7 +476,7 @@ static int bgx_lmac_xaui_init(struct bgx *bgx, int lmacid, int lmac_type) cfg = bgx_reg_read(bgx, lmacid, BGX_SPUX_INT); bgx_reg_write(bgx, lmacid, BGX_SPUX_INT, cfg); - if (bgx->use_training) { + if (lmac->use_training) { bgx_reg_write(bgx, lmacid, BGX_SPUX_BR_PMD_LP_CUP, 0x00); bgx_reg_write(bgx, lmacid, BGX_SPUX_BR_PMD_LD_CUP, 0x00); bgx_reg_write(bgx, lmacid, BGX_SPUX_BR_PMD_LD_REP, 0x00); @@ -474,9 +499,9 @@ static int bgx_lmac_xaui_init(struct bgx *bgx, int lmacid, int lmac_type) bgx_reg_write(bgx, lmacid, BGX_SPUX_AN_CONTROL, cfg); cfg = bgx_reg_read(bgx, lmacid, BGX_SPUX_AN_ADV); - if (bgx->lmac_type == BGX_MODE_10G_KR) + if (lmac->lmac_type == BGX_MODE_10G_KR) cfg |= (1 << 23); - else if (bgx->lmac_type == BGX_MODE_40G_KR) + else if (lmac->lmac_type == BGX_MODE_40G_KR) cfg |= (1 << 24); else cfg &= ~((1 << 23) | (1 << 24)); @@ -511,11 +536,10 @@ static int bgx_xaui_check_link(struct lmac *lmac) { struct bgx *bgx = lmac->bgx; int lmacid = lmac->lmacid; - int lmac_type = bgx->lmac_type; + int lmac_type = lmac->lmac_type; u64 cfg; - bgx_reg_modify(bgx, lmacid, BGX_SPUX_MISC_CONTROL, SPU_MISC_CTL_RX_DIS); - if (bgx->use_training) { + if (lmac->use_training) { cfg = bgx_reg_read(bgx, lmacid, BGX_SPUX_INT); if (!(cfg & (1ull << 13))) { cfg = (1ull << 13) | (1ull << 14); @@ -556,7 +580,7 @@ static int bgx_xaui_check_link(struct lmac *lmac) BGX_SPUX_STATUS2, SPU_STATUS2_RCVFLT); if (bgx_reg_read(bgx, lmacid, BGX_SPUX_STATUS2) & SPU_STATUS2_RCVFLT) { dev_err(&bgx->pdev->dev, "Receive fault, retry training\n"); - if (bgx->use_training) { + if (lmac->use_training) { cfg = bgx_reg_read(bgx, lmacid, BGX_SPUX_INT); if (!(cfg & (1ull << 13))) { cfg = (1ull << 13) | (1ull << 14); @@ -584,11 +608,6 @@ static int bgx_xaui_check_link(struct lmac *lmac) return -1; } - /* Clear receive packet disable */ - cfg = bgx_reg_read(bgx, lmacid, BGX_SPUX_MISC_CONTROL); - cfg &= ~SPU_MISC_CTL_RX_DIS; - bgx_reg_write(bgx, lmacid, BGX_SPUX_MISC_CONTROL, cfg); - /* Check for MAC RX faults */ cfg = bgx_reg_read(bgx, lmacid, BGX_SMUX_RX_CTL); /* 0 - Link is okay, 1 - Local fault, 2 - Remote fault */ @@ -599,7 +618,7 @@ static int bgx_xaui_check_link(struct lmac *lmac) /* Rx local/remote fault seen. * Do lmac reinit to see if condition recovers */ - bgx_lmac_xaui_init(bgx, lmacid, bgx->lmac_type); + bgx_lmac_xaui_init(bgx, lmac); return -1; } @@ -623,7 +642,7 @@ static void bgx_poll_for_link(struct work_struct *work) if ((spu_link & SPU_STATUS1_RCV_LNK) && !(smu_link & SMU_RX_CTL_STATUS)) { lmac->link_up = 1; - if (lmac->bgx->lmac_type == BGX_MODE_XLAUI) + if (lmac->lmac_type == BGX_MODE_XLAUI) lmac->last_speed = 40000; else lmac->last_speed = 10000; @@ -649,6 +668,16 @@ static void bgx_poll_for_link(struct work_struct *work) queue_delayed_work(lmac->check_link, &lmac->dwork, HZ * 2); } +static int phy_interface_mode(u8 lmac_type) +{ + if (lmac_type == BGX_MODE_QSGMII) + return PHY_INTERFACE_MODE_QSGMII; + if (lmac_type == BGX_MODE_RGMII) + return PHY_INTERFACE_MODE_RGMII; + + return PHY_INTERFACE_MODE_SGMII; +} + static int bgx_lmac_enable(struct bgx *bgx, u8 lmacid) { struct lmac *lmac; @@ -657,13 +686,15 @@ static int bgx_lmac_enable(struct bgx *bgx, u8 lmacid) lmac = &bgx->lmac[lmacid]; lmac->bgx = bgx; - if (bgx->lmac_type == BGX_MODE_SGMII) { + if ((lmac->lmac_type == BGX_MODE_SGMII) || + (lmac->lmac_type == BGX_MODE_QSGMII) || + (lmac->lmac_type == BGX_MODE_RGMII)) { lmac->is_sgmii = 1; - if (bgx_lmac_sgmii_init(bgx, lmacid)) + if (bgx_lmac_sgmii_init(bgx, lmac)) return -1; } else { lmac->is_sgmii = 0; - if (bgx_lmac_xaui_init(bgx, lmacid, bgx->lmac_type)) + if (bgx_lmac_xaui_init(bgx, lmac)) return -1; } @@ -685,10 +716,10 @@ static int bgx_lmac_enable(struct bgx *bgx, u8 lmacid) /* Restore default cfg, incase low level firmware changed it */ bgx_reg_write(bgx, lmacid, BGX_CMRX_RX_DMAC_CTL, 0x03); - if ((bgx->lmac_type != BGX_MODE_XFI) && - (bgx->lmac_type != BGX_MODE_XLAUI) && - (bgx->lmac_type != BGX_MODE_40G_KR) && - (bgx->lmac_type != BGX_MODE_10G_KR)) { + if ((lmac->lmac_type != BGX_MODE_XFI) && + (lmac->lmac_type != BGX_MODE_XLAUI) && + (lmac->lmac_type != BGX_MODE_40G_KR) && + (lmac->lmac_type != BGX_MODE_10G_KR)) { if (!lmac->phydev) return -ENODEV; @@ -696,7 +727,7 @@ static int bgx_lmac_enable(struct bgx *bgx, u8 lmacid) if (phy_connect_direct(&lmac->netdev, lmac->phydev, bgx_lmac_handler, - PHY_INTERFACE_MODE_SGMII)) + phy_interface_mode(lmac->lmac_type))) return -ENODEV; phy_start_aneg(lmac->phydev); @@ -753,76 +784,19 @@ static void bgx_lmac_disable(struct bgx *bgx, u8 lmacid) bgx_flush_dmac_addrs(bgx, lmacid); - if ((bgx->lmac_type != BGX_MODE_XFI) && - (bgx->lmac_type != BGX_MODE_XLAUI) && - (bgx->lmac_type != BGX_MODE_40G_KR) && - (bgx->lmac_type != BGX_MODE_10G_KR) && lmac->phydev) + if ((lmac->lmac_type != BGX_MODE_XFI) && + (lmac->lmac_type != BGX_MODE_XLAUI) && + (lmac->lmac_type != BGX_MODE_40G_KR) && + (lmac->lmac_type != BGX_MODE_10G_KR) && lmac->phydev) phy_disconnect(lmac->phydev); lmac->phydev = NULL; } -static void bgx_set_num_ports(struct bgx *bgx) -{ - u64 lmac_count; - - switch (bgx->qlm_mode) { - case QLM_MODE_SGMII: - bgx->lmac_count = 4; - bgx->lmac_type = BGX_MODE_SGMII; - bgx->lane_to_sds = 0; - break; - case QLM_MODE_XAUI_1X4: - bgx->lmac_count = 1; - bgx->lmac_type = BGX_MODE_XAUI; - bgx->lane_to_sds = 0xE4; - break; - case QLM_MODE_RXAUI_2X2: - bgx->lmac_count = 2; - bgx->lmac_type = BGX_MODE_RXAUI; - bgx->lane_to_sds = 0xE4; - break; - case QLM_MODE_XFI_4X1: - bgx->lmac_count = 4; - bgx->lmac_type = BGX_MODE_XFI; - bgx->lane_to_sds = 0; - break; - case QLM_MODE_XLAUI_1X4: - bgx->lmac_count = 1; - bgx->lmac_type = BGX_MODE_XLAUI; - bgx->lane_to_sds = 0xE4; - break; - case QLM_MODE_10G_KR_4X1: - bgx->lmac_count = 4; - bgx->lmac_type = BGX_MODE_10G_KR; - bgx->lane_to_sds = 0; - bgx->use_training = 1; - break; - case QLM_MODE_40G_KR4_1X4: - bgx->lmac_count = 1; - bgx->lmac_type = BGX_MODE_40G_KR; - bgx->lane_to_sds = 0xE4; - bgx->use_training = 1; - break; - default: - bgx->lmac_count = 0; - break; - } - - /* Check if low level firmware has programmed LMAC count - * based on board type, if yes consider that otherwise - * the default static values - */ - lmac_count = bgx_reg_read(bgx, 0, BGX_CMR_RX_LMACS) & 0x7; - if (lmac_count != 4) - bgx->lmac_count = lmac_count; -} - static void bgx_init_hw(struct bgx *bgx) { int i; - - bgx_set_num_ports(bgx); + struct lmac *lmac; bgx_reg_modify(bgx, 0, BGX_CMR_GLOBAL_CFG, CMR_GLOBAL_CFG_FCS_STRIP); if (bgx_reg_read(bgx, 0, BGX_CMR_BIST_STATUS)) @@ -830,17 +804,9 @@ static void bgx_init_hw(struct bgx *bgx) /* Set lmac type and lane2serdes mapping */ for (i = 0; i < bgx->lmac_count; i++) { - if (bgx->lmac_type == BGX_MODE_RXAUI) { - if (i) - bgx->lane_to_sds = 0x0e; - else - bgx->lane_to_sds = 0x04; - bgx_reg_write(bgx, i, BGX_CMRX_CFG, - (bgx->lmac_type << 8) | bgx->lane_to_sds); - continue; - } + lmac = &bgx->lmac[i]; bgx_reg_write(bgx, i, BGX_CMRX_CFG, - (bgx->lmac_type << 8) | (bgx->lane_to_sds + i)); + (lmac->lmac_type << 8) | lmac->lane_to_sds); bgx->lmac[i].lmacid_bd = lmac_count; lmac_count++; } @@ -863,55 +829,212 @@ static void bgx_init_hw(struct bgx *bgx) bgx_reg_write(bgx, 0, BGX_CMR_RX_STREERING + (i * 8), 0x00); } -static void bgx_get_qlm_mode(struct bgx *bgx) +static u8 bgx_get_lane2sds_cfg(struct bgx *bgx, struct lmac *lmac) +{ + return (u8)(bgx_reg_read(bgx, lmac->lmacid, BGX_CMRX_CFG) & 0xFF); +} + +static void bgx_print_qlm_mode(struct bgx *bgx, u8 lmacid) { struct device *dev = &bgx->pdev->dev; - int lmac_type; - int train_en; + struct lmac *lmac; + char str[20]; + u8 dlm; - /* Read LMAC0 type to figure out QLM mode - * This is configured by low level firmware - */ - lmac_type = bgx_reg_read(bgx, 0, BGX_CMRX_CFG); - lmac_type = (lmac_type >> 8) & 0x07; + if (lmacid > bgx->max_lmac) + return; - train_en = bgx_reg_read(bgx, 0, BGX_SPUX_BR_PMD_CRTL) & - SPU_PMD_CRTL_TRAIN_EN; + lmac = &bgx->lmac[lmacid]; + dlm = (lmacid / 2) + (bgx->bgx_id * 2); + if (!bgx->is_dlm) + sprintf(str, "BGX%d QLM mode", bgx->bgx_id); + else + sprintf(str, "BGX%d DLM%d mode", bgx->bgx_id, dlm); - switch (lmac_type) { + switch (lmac->lmac_type) { case BGX_MODE_SGMII: - bgx->qlm_mode = QLM_MODE_SGMII; - dev_info(dev, "BGX%d QLM mode: SGMII\n", bgx->bgx_id); + dev_info(dev, "%s: SGMII\n", (char *)str); break; case BGX_MODE_XAUI: - bgx->qlm_mode = QLM_MODE_XAUI_1X4; - dev_info(dev, "BGX%d QLM mode: XAUI\n", bgx->bgx_id); + dev_info(dev, "%s: XAUI\n", (char *)str); break; case BGX_MODE_RXAUI: - bgx->qlm_mode = QLM_MODE_RXAUI_2X2; - dev_info(dev, "BGX%d QLM mode: RXAUI\n", bgx->bgx_id); + dev_info(dev, "%s: RXAUI\n", (char *)str); break; case BGX_MODE_XFI: - if (!train_en) { - bgx->qlm_mode = QLM_MODE_XFI_4X1; - dev_info(dev, "BGX%d QLM mode: XFI\n", bgx->bgx_id); - } else { - bgx->qlm_mode = QLM_MODE_10G_KR_4X1; - dev_info(dev, "BGX%d QLM mode: 10G_KR\n", bgx->bgx_id); - } + if (!lmac->use_training) + dev_info(dev, "%s: XFI\n", (char *)str); + else + dev_info(dev, "%s: 10G_KR\n", (char *)str); break; case BGX_MODE_XLAUI: - if (!train_en) { - bgx->qlm_mode = QLM_MODE_XLAUI_1X4; - dev_info(dev, "BGX%d QLM mode: XLAUI\n", bgx->bgx_id); - } else { - bgx->qlm_mode = QLM_MODE_40G_KR4_1X4; - dev_info(dev, "BGX%d QLM mode: 40G_KR4\n", bgx->bgx_id); - } + if (!lmac->use_training) + dev_info(dev, "%s: XLAUI\n", (char *)str); + else + dev_info(dev, "%s: 40G_KR4\n", (char *)str); + break; + case BGX_MODE_QSGMII: + if ((lmacid == 0) && + (bgx_get_lane2sds_cfg(bgx, lmac) != lmacid)) + return; + if ((lmacid == 2) && + (bgx_get_lane2sds_cfg(bgx, lmac) == lmacid)) + return; + dev_info(dev, "%s: QSGMII\n", (char *)str); + break; + case BGX_MODE_RGMII: + dev_info(dev, "%s: RGMII\n", (char *)str); + break; + case BGX_MODE_INVALID: + /* Nothing to do */ + break; + } +} + +static void lmac_set_lane2sds(struct bgx *bgx, struct lmac *lmac) +{ + switch (lmac->lmac_type) { + case BGX_MODE_SGMII: + case BGX_MODE_XFI: + lmac->lane_to_sds = lmac->lmacid; + break; + case BGX_MODE_XAUI: + case BGX_MODE_XLAUI: + case BGX_MODE_RGMII: + lmac->lane_to_sds = 0xE4; + break; + case BGX_MODE_RXAUI: + lmac->lane_to_sds = (lmac->lmacid) ? 0xE : 0x4; + break; + case BGX_MODE_QSGMII: + /* There is no way to determine if DLM0/2 is QSGMII or + * DLM1/3 is configured to QSGMII as bootloader will + * configure all LMACs, so take whatever is configured + * by low level firmware. + */ + lmac->lane_to_sds = bgx_get_lane2sds_cfg(bgx, lmac); break; default: - bgx->qlm_mode = QLM_MODE_SGMII; - dev_info(dev, "BGX%d QLM default mode: SGMII\n", bgx->bgx_id); + lmac->lane_to_sds = 0; + break; + } +} + +static void lmac_set_training(struct bgx *bgx, struct lmac *lmac, int lmacid) +{ + if ((lmac->lmac_type != BGX_MODE_10G_KR) && + (lmac->lmac_type != BGX_MODE_40G_KR)) { + lmac->use_training = 0; + return; + } + + lmac->use_training = bgx_reg_read(bgx, lmacid, BGX_SPUX_BR_PMD_CRTL) & + SPU_PMD_CRTL_TRAIN_EN; +} + +static void bgx_set_lmac_config(struct bgx *bgx, u8 idx) +{ + struct lmac *lmac; + struct lmac *olmac; + u64 cmr_cfg; + u8 lmac_type; + u8 lane_to_sds; + + lmac = &bgx->lmac[idx]; + + if (!bgx->is_dlm || bgx->is_rgx) { + /* Read LMAC0 type to figure out QLM mode + * This is configured by low level firmware + */ + cmr_cfg = bgx_reg_read(bgx, 0, BGX_CMRX_CFG); + lmac->lmac_type = (cmr_cfg >> 8) & 0x07; + if (bgx->is_rgx) + lmac->lmac_type = BGX_MODE_RGMII; + lmac_set_training(bgx, lmac, 0); + lmac_set_lane2sds(bgx, lmac); + return; + } + + /* On 81xx BGX can be split across 2 DLMs + * firmware programs lmac_type of LMAC0 and LMAC2 + */ + if ((idx == 0) || (idx == 2)) { + cmr_cfg = bgx_reg_read(bgx, idx, BGX_CMRX_CFG); + lmac_type = (u8)((cmr_cfg >> 8) & 0x07); + lane_to_sds = (u8)(cmr_cfg & 0xFF); + /* Check if config is not reset value */ + if ((lmac_type == 0) && (lane_to_sds == 0xE4)) + lmac->lmac_type = BGX_MODE_INVALID; + else + lmac->lmac_type = lmac_type; + lmac_set_training(bgx, lmac, lmac->lmacid); + lmac_set_lane2sds(bgx, lmac); + + /* Set LMAC type of other lmac on same DLM i.e LMAC 1/3 */ + olmac = &bgx->lmac[idx + 1]; + olmac->lmac_type = lmac->lmac_type; + lmac_set_training(bgx, olmac, olmac->lmacid); + lmac_set_lane2sds(bgx, olmac); + } +} + +static bool is_dlm0_in_bgx_mode(struct bgx *bgx) +{ + struct lmac *lmac; + + if (!bgx->is_dlm) + return true; + + lmac = &bgx->lmac[0]; + if (lmac->lmac_type == BGX_MODE_INVALID) + return false; + + return true; +} + +static void bgx_get_qlm_mode(struct bgx *bgx) +{ + struct lmac *lmac; + struct lmac *lmac01; + struct lmac *lmac23; + u8 idx; + + /* Init all LMAC's type to invalid */ + for (idx = 0; idx < bgx->max_lmac; idx++) { + lmac = &bgx->lmac[idx]; + lmac->lmacid = idx; + lmac->lmac_type = BGX_MODE_INVALID; + lmac->use_training = false; + } + + /* It is assumed that low level firmware sets this value */ + bgx->lmac_count = bgx_reg_read(bgx, 0, BGX_CMR_RX_LMACS) & 0x7; + if (bgx->lmac_count > bgx->max_lmac) + bgx->lmac_count = bgx->max_lmac; + + for (idx = 0; idx < bgx->max_lmac; idx++) + bgx_set_lmac_config(bgx, idx); + + if (!bgx->is_dlm || bgx->is_rgx) { + bgx_print_qlm_mode(bgx, 0); + return; + } + + if (bgx->lmac_count) { + bgx_print_qlm_mode(bgx, 0); + bgx_print_qlm_mode(bgx, 2); + } + + /* If DLM0 is not in BGX mode then LMAC0/1 have + * to be configured with serdes lanes of DLM1 + */ + if (is_dlm0_in_bgx_mode(bgx) || (bgx->lmac_count > 2)) + return; + for (idx = 0; idx < bgx->lmac_count; idx++) { + lmac01 = &bgx->lmac[idx]; + lmac23 = &bgx->lmac[idx + 2]; + lmac01->lmac_type = lmac23->lmac_type; + lmac01->lane_to_sds = lmac23->lane_to_sds; } } @@ -1042,7 +1165,7 @@ static int bgx_init_of_phy(struct bgx *bgx) } lmac++; - if (lmac == MAX_LMAC_PER_BGX) { + if (lmac == bgx->max_lmac) { of_node_put(node); break; } @@ -1087,6 +1210,7 @@ static int bgx_probe(struct pci_dev *pdev, const struct pci_device_id *ent) struct device *dev = &pdev->dev; struct bgx *bgx = NULL; u8 lmac; + u16 sdevid; bgx = devm_kzalloc(dev, sizeof(*bgx), GFP_KERNEL); if (!bgx) @@ -1115,10 +1239,30 @@ static int bgx_probe(struct pci_dev *pdev, const struct pci_device_id *ent) err = -ENOMEM; goto err_release_regions; } - bgx->bgx_id = (pci_resource_start(pdev, PCI_CFG_REG_BAR_NUM) >> 24) & 1; - bgx->bgx_id += nic_get_node_id(pdev) * MAX_BGX_PER_CN88XX; - bgx_vnic[bgx->bgx_id] = bgx; + pci_read_config_word(pdev, PCI_DEVICE_ID, &sdevid); + if (sdevid != PCI_DEVICE_ID_THUNDER_RGX) { + bgx->bgx_id = + (pci_resource_start(pdev, PCI_CFG_REG_BAR_NUM) >> 24) & 1; + bgx->bgx_id += nic_get_node_id(pdev) * MAX_BGX_PER_NODE; + bgx->max_lmac = MAX_LMAC_PER_BGX; + bgx_vnic[bgx->bgx_id] = bgx; + } else { + bgx->is_rgx = true; + bgx->max_lmac = 1; + bgx->bgx_id = MAX_BGX_PER_CN81XX - 1; + bgx_vnic[bgx->bgx_id] = bgx; + xcv_init_hw(); + } + + /* On 81xx all are DLMs and on 83xx there are 3 BGX QLMs and one + * BGX i.e BGX2 can be split across 2 DLMs. + */ + pci_read_config_word(pdev, PCI_SUBSYSTEM_ID, &sdevid); + if ((sdevid == PCI_SUBSYS_DEVID_81XX_BGX) || + ((sdevid == PCI_SUBSYS_DEVID_83XX_BGX) && (bgx->bgx_id == 2))) + bgx->is_dlm = true; + bgx_get_qlm_mode(bgx); err = bgx_init_phy(bgx); @@ -1133,6 +1277,8 @@ static int bgx_probe(struct pci_dev *pdev, const struct pci_device_id *ent) if (err) { dev_err(dev, "BGX%d failed to enable lmac%d\n", bgx->bgx_id, lmac); + while (lmac) + bgx_lmac_disable(bgx, --lmac); goto err_enable; } } diff --git a/drivers/net/ethernet/cavium/thunder/thunder_bgx.h b/drivers/net/ethernet/cavium/thunder/thunder_bgx.h index 42010d2e5ddf..d59c71e4a000 100644 --- a/drivers/net/ethernet/cavium/thunder/thunder_bgx.h +++ b/drivers/net/ethernet/cavium/thunder/thunder_bgx.h @@ -9,8 +9,20 @@ #ifndef THUNDER_BGX_H #define THUNDER_BGX_H -#define MAX_BGX_THUNDER 8 /* Max 4 nodes, 2 per node */ +/* PCI device ID */ +#define PCI_DEVICE_ID_THUNDER_BGX 0xA026 +#define PCI_DEVICE_ID_THUNDER_RGX 0xA054 + +/* Subsystem device IDs */ +#define PCI_SUBSYS_DEVID_88XX_BGX 0xA126 +#define PCI_SUBSYS_DEVID_81XX_BGX 0xA226 +#define PCI_SUBSYS_DEVID_83XX_BGX 0xA326 + +#define MAX_BGX_THUNDER 8 /* Max 2 nodes, 4 per node */ #define MAX_BGX_PER_CN88XX 2 +#define MAX_BGX_PER_CN81XX 3 /* 2 BGXs + 1 RGX */ +#define MAX_BGX_PER_CN83XX 4 +#define MAX_BGX_PER_NODE 4 #define MAX_LMAC_PER_BGX 4 #define MAX_BGX_CHANS_PER_LMAC 16 #define MAX_DMAC_PER_LMAC 8 @@ -18,8 +30,6 @@ #define MAX_DMAC_PER_LMAC_TNS_BYPASS_MODE 2 -#define MAX_LMAC (MAX_BGX_PER_CN88XX * MAX_LMAC_PER_BGX) - /* Registers */ #define BGX_CMRX_CFG 0x00 #define CMR_PKT_TX_EN BIT_ULL(13) @@ -136,6 +146,7 @@ #define BGX_GMP_PCS_ANX_AN_RESULTS 0x30020 #define BGX_GMP_PCS_SGM_AN_ADV 0x30068 #define BGX_GMP_PCS_MISCX_CTL 0x30078 +#define PCS_MISC_CTL_DISP_EN BIT_ULL(13) #define PCS_MISC_CTL_GMX_ENO BIT_ULL(11) #define PCS_MISC_CTL_SAMP_PT_MASK 0x7Full #define BGX_GMP_GMI_PRTX_CFG 0x38020 @@ -194,6 +205,9 @@ void bgx_set_lmac_mac(int node, int bgx_idx, int lmacid, const u8 *mac); void bgx_get_lmac_link_state(int node, int bgx_idx, int lmacid, void *status); void bgx_lmac_internal_loopback(int node, int bgx_idx, int lmac_idx, bool enable); +void xcv_init_hw(void); +void xcv_setup_link(bool link_up, int link_speed); + u64 bgx_get_rx_stats(int node, int bgx_idx, int lmac, int idx); u64 bgx_get_tx_stats(int node, int bgx_idx, int lmac, int idx); #define BGX_RX_STATS_COUNT 11 @@ -213,16 +227,9 @@ enum LMAC_TYPE { BGX_MODE_XLAUI = 4, /* 4 lanes, 10.3125 Gbaud */ BGX_MODE_10G_KR = 3,/* 1 lane, 10.3125 Gbaud */ BGX_MODE_40G_KR = 4,/* 4 lanes, 10.3125 Gbaud */ -}; - -enum qlm_mode { - QLM_MODE_SGMII, /* SGMII, each lane independent */ - QLM_MODE_XAUI_1X4, /* 1 XAUI or DXAUI, 4 lanes */ - QLM_MODE_RXAUI_2X2, /* 2 RXAUI, 2 lanes each */ - QLM_MODE_XFI_4X1, /* 4 XFI, 1 lane each */ - QLM_MODE_XLAUI_1X4, /* 1 XLAUI, 4 lanes each */ - QLM_MODE_10G_KR_4X1, /* 4 10GBASE-KR, 1 lane each */ - QLM_MODE_40G_KR4_1X4, /* 1 40GBASE-KR4, 4 lanes each */ + BGX_MODE_RGMII = 5, + BGX_MODE_QSGMII = 6, + BGX_MODE_INVALID = 7, }; #endif /* THUNDER_BGX_H */ diff --git a/drivers/net/ethernet/cavium/thunder/thunder_xcv.c b/drivers/net/ethernet/cavium/thunder/thunder_xcv.c new file mode 100644 index 000000000000..67befedef709 --- /dev/null +++ b/drivers/net/ethernet/cavium/thunder/thunder_xcv.c @@ -0,0 +1,235 @@ +/* + * Copyright (C) 2016 Cavium, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of version 2 of the GNU General Public License + * as published by the Free Software Foundation. + */ + +#include <linux/acpi.h> +#include <linux/module.h> +#include <linux/interrupt.h> +#include <linux/pci.h> +#include <linux/netdevice.h> +#include <linux/etherdevice.h> +#include <linux/phy.h> +#include <linux/of.h> +#include <linux/of_mdio.h> +#include <linux/of_net.h> + +#include "nic.h" +#include "thunder_bgx.h" + +#define DRV_NAME "thunder-xcv" +#define DRV_VERSION "1.0" + +/* Register offsets */ +#define XCV_RESET 0x00 +#define PORT_EN BIT_ULL(63) +#define CLK_RESET BIT_ULL(15) +#define DLL_RESET BIT_ULL(11) +#define COMP_EN BIT_ULL(7) +#define TX_PKT_RESET BIT_ULL(3) +#define TX_DATA_RESET BIT_ULL(2) +#define RX_PKT_RESET BIT_ULL(1) +#define RX_DATA_RESET BIT_ULL(0) +#define XCV_DLL_CTL 0x10 +#define CLKRX_BYP BIT_ULL(23) +#define CLKTX_BYP BIT_ULL(15) +#define XCV_COMP_CTL 0x20 +#define DRV_BYP BIT_ULL(63) +#define XCV_CTL 0x30 +#define XCV_INT 0x40 +#define XCV_INT_W1S 0x48 +#define XCV_INT_ENA_W1C 0x50 +#define XCV_INT_ENA_W1S 0x58 +#define XCV_INBND_STATUS 0x80 +#define XCV_BATCH_CRD_RET 0x100 + +struct xcv { + void __iomem *reg_base; + struct pci_dev *pdev; +}; + +static struct xcv *xcv; + +/* Supported devices */ +static const struct pci_device_id xcv_id_table[] = { + { PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, 0xA056) }, + { 0, } /* end of table */ +}; + +MODULE_AUTHOR("Cavium Inc"); +MODULE_DESCRIPTION("Cavium Thunder RGX/XCV Driver"); +MODULE_LICENSE("GPL v2"); +MODULE_VERSION(DRV_VERSION); +MODULE_DEVICE_TABLE(pci, xcv_id_table); + +void xcv_init_hw(void) +{ + u64 cfg; + + /* Take DLL out of reset */ + cfg = readq_relaxed(xcv->reg_base + XCV_RESET); + cfg &= ~DLL_RESET; + writeq_relaxed(cfg, xcv->reg_base + XCV_RESET); + + /* Take clock tree out of reset */ + cfg = readq_relaxed(xcv->reg_base + XCV_RESET); + cfg &= ~CLK_RESET; + writeq_relaxed(cfg, xcv->reg_base + XCV_RESET); + /* Wait for DLL to lock */ + msleep(1); + + /* Configure DLL - enable or bypass + * TX no bypass, RX bypass + */ + cfg = readq_relaxed(xcv->reg_base + XCV_DLL_CTL); + cfg &= ~0xFF03; + cfg |= CLKRX_BYP; + writeq_relaxed(cfg, xcv->reg_base + XCV_DLL_CTL); + + /* Enable compensation controller and force the + * write to be visible to HW by readig back. + */ + cfg = readq_relaxed(xcv->reg_base + XCV_RESET); + cfg |= COMP_EN; + writeq_relaxed(cfg, xcv->reg_base + XCV_RESET); + readq_relaxed(xcv->reg_base + XCV_RESET); + /* Wait for compensation state machine to lock */ + msleep(10); + + /* enable the XCV block */ + cfg = readq_relaxed(xcv->reg_base + XCV_RESET); + cfg |= PORT_EN; + writeq_relaxed(cfg, xcv->reg_base + XCV_RESET); + + cfg = readq_relaxed(xcv->reg_base + XCV_RESET); + cfg |= CLK_RESET; + writeq_relaxed(cfg, xcv->reg_base + XCV_RESET); +} +EXPORT_SYMBOL(xcv_init_hw); + +void xcv_setup_link(bool link_up, int link_speed) +{ + u64 cfg; + int speed = 2; + + if (!xcv) { + dev_err(&xcv->pdev->dev, + "XCV init not done, probe may have failed\n"); + return; + } + + if (link_speed == 100) + speed = 1; + else if (link_speed == 10) + speed = 0; + + if (link_up) { + /* set operating speed */ + cfg = readq_relaxed(xcv->reg_base + XCV_CTL); + cfg &= ~0x03; + cfg |= speed; + writeq_relaxed(cfg, xcv->reg_base + XCV_CTL); + + /* Reset datapaths */ + cfg = readq_relaxed(xcv->reg_base + XCV_RESET); + cfg |= TX_DATA_RESET | RX_DATA_RESET; + writeq_relaxed(cfg, xcv->reg_base + XCV_RESET); + + /* Enable the packet flow */ + cfg = readq_relaxed(xcv->reg_base + XCV_RESET); + cfg |= TX_PKT_RESET | RX_PKT_RESET; + writeq_relaxed(cfg, xcv->reg_base + XCV_RESET); + + /* Return credits to RGX */ + writeq_relaxed(0x01, xcv->reg_base + XCV_BATCH_CRD_RET); + } else { + /* Disable packet flow */ + cfg = readq_relaxed(xcv->reg_base + XCV_RESET); + cfg &= ~(TX_PKT_RESET | RX_PKT_RESET); + writeq_relaxed(cfg, xcv->reg_base + XCV_RESET); + readq_relaxed(xcv->reg_base + XCV_RESET); + } +} +EXPORT_SYMBOL(xcv_setup_link); + +static int xcv_probe(struct pci_dev *pdev, const struct pci_device_id *ent) +{ + int err; + struct device *dev = &pdev->dev; + + xcv = devm_kzalloc(dev, sizeof(struct xcv), GFP_KERNEL); + if (!xcv) + return -ENOMEM; + xcv->pdev = pdev; + + pci_set_drvdata(pdev, xcv); + + err = pci_enable_device(pdev); + if (err) { + dev_err(dev, "Failed to enable PCI device\n"); + goto err_kfree; + } + + err = pci_request_regions(pdev, DRV_NAME); + if (err) { + dev_err(dev, "PCI request regions failed 0x%x\n", err); + goto err_disable_device; + } + + /* MAP configuration registers */ + xcv->reg_base = pcim_iomap(pdev, PCI_CFG_REG_BAR_NUM, 0); + if (!xcv->reg_base) { + dev_err(dev, "XCV: Cannot map CSR memory space, aborting\n"); + err = -ENOMEM; + goto err_release_regions; + } + + return 0; + +err_release_regions: + pci_release_regions(pdev); +err_disable_device: + pci_disable_device(pdev); +err_kfree: + devm_kfree(dev, xcv); + xcv = NULL; + return err; +} + +static void xcv_remove(struct pci_dev *pdev) +{ + struct device *dev = &pdev->dev; + + if (xcv) { + devm_kfree(dev, xcv); + xcv = NULL; + } + + pci_release_regions(pdev); + pci_disable_device(pdev); +} + +static struct pci_driver xcv_driver = { + .name = DRV_NAME, + .id_table = xcv_id_table, + .probe = xcv_probe, + .remove = xcv_remove, +}; + +static int __init xcv_init_module(void) +{ + pr_info("%s, ver %s\n", DRV_NAME, DRV_VERSION); + + return pci_register_driver(&xcv_driver); +} + +static void __exit xcv_cleanup_module(void) +{ + pci_unregister_driver(&xcv_driver); +} + +module_init(xcv_init_module); +module_exit(xcv_cleanup_module); diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h index 2e2aa9fec9bb..bcfa51226b46 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h @@ -1521,4 +1521,7 @@ void t4_idma_monitor_init(struct adapter *adapter, void t4_idma_monitor(struct adapter *adapter, struct sge_idma_monitor_state *idma, int hz, int ticks); +int t4_set_vf_mac_acl(struct adapter *adapter, unsigned int vf, + unsigned int naddr, u8 *addr); + #endif /* __CXGB4_H__ */ diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index c45de49dc963..2bb804c93688 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -3078,6 +3078,26 @@ static int cxgb_change_mtu(struct net_device *dev, int new_mtu) return ret; } +#ifdef CONFIG_PCI_IOV +static int cxgb_set_vf_mac(struct net_device *dev, int vf, u8 *mac) +{ + struct port_info *pi = netdev_priv(dev); + struct adapter *adap = pi->adapter; + + /* verify MAC addr is valid */ + if (!is_valid_ether_addr(mac)) { + dev_err(pi->adapter->pdev_dev, + "Invalid Ethernet address %pM for VF %d\n", + mac, vf); + return -EINVAL; + } + + dev_info(pi->adapter->pdev_dev, + "Setting MAC %pM on VF %d\n", mac, vf); + return t4_set_vf_mac_acl(adap, vf + 1, 1, mac); +} +#endif + static int cxgb_set_mac_addr(struct net_device *dev, void *p) { int ret; @@ -3136,7 +3156,27 @@ static const struct net_device_ops cxgb4_netdev_ops = { #ifdef CONFIG_NET_RX_BUSY_POLL .ndo_busy_poll = cxgb_busy_poll, #endif +}; +static const struct net_device_ops cxgb4_mgmt_netdev_ops = { +#ifdef CONFIG_PCI_IOV + .ndo_set_vf_mac = cxgb_set_vf_mac, +#endif +}; + +static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) +{ + struct adapter *adapter = netdev2adap(dev); + + strlcpy(info->driver, cxgb4_driver_name, sizeof(info->driver)); + strlcpy(info->version, cxgb4_driver_version, + sizeof(info->version)); + strlcpy(info->bus_info, pci_name(adapter->pdev), + sizeof(info->bus_info)); +} + +static const struct ethtool_ops cxgb4_mgmt_ethtool_ops = { + .get_drvinfo = get_drvinfo, }; void t4_fatal_err(struct adapter *adap) @@ -4836,19 +4876,12 @@ static int get_chip_type(struct pci_dev *pdev, u32 pl_rev) #ifdef CONFIG_PCI_IOV static int cxgb4_iov_configure(struct pci_dev *pdev, int num_vfs) { + struct adapter *adap = pci_get_drvdata(pdev); int err = 0; int current_vfs = pci_num_vf(pdev); u32 pcie_fw; - void __iomem *regs; - regs = pci_ioremap_bar(pdev, 0); - if (!regs) { - dev_err(&pdev->dev, "cannot map device registers\n"); - return -ENOMEM; - } - - pcie_fw = readl(regs + PCIE_FW_A); - iounmap(regs); + pcie_fw = readl(adap->regs + PCIE_FW_A); /* Check if cxgb4 is the MASTER and fw is initialized */ if (!(pcie_fw & PCIE_FW_INIT_F) || !(pcie_fw & PCIE_FW_MASTER_VLD_F) || @@ -4875,6 +4908,8 @@ static int cxgb4_iov_configure(struct pci_dev *pdev, int num_vfs) */ if (!num_vfs) { pci_disable_sriov(pdev); + if (adap->port[0]->reg_state == NETREG_REGISTERED) + unregister_netdev(adap->port[0]); return num_vfs; } @@ -4882,6 +4917,12 @@ static int cxgb4_iov_configure(struct pci_dev *pdev, int num_vfs) err = pci_enable_sriov(pdev, num_vfs); if (err) return err; + + if (adap->port[0]->reg_state == NETREG_UNINITIALIZED) { + err = register_netdev(adap->port[0]); + if (err < 0) + pr_info("Unable to register VF mgmt netdev\n"); + } } return num_vfs; } @@ -4893,9 +4934,14 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) struct port_info *pi; bool highdma = false; struct adapter *adapter = NULL; + struct net_device *netdev; +#ifdef CONFIG_PCI_IOV + char name[IFNAMSIZ]; +#endif void __iomem *regs; u32 whoami, pl_rev; enum chip_type chip; + static int adap_idx = 1; printk_once(KERN_INFO "%s - version %s\n", DRV_DESC, DRV_VERSION); @@ -4930,7 +4976,9 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) func = CHELSIO_CHIP_VERSION(chip) <= CHELSIO_T5 ? SOURCEPF_G(whoami) : T6_SOURCEPF_G(whoami); if (func != ent->driver_data) { +#ifndef CONFIG_PCI_IOV iounmap(regs); +#endif pci_disable_device(pdev); pci_save_state(pdev); /* to restore SR-IOV later */ goto sriov; @@ -4962,6 +5010,7 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) err = -ENOMEM; goto out_unmap_bar0; } + adap_idx++; adapter->workq = create_singlethread_workqueue("cxgb4"); if (!adapter->workq) { @@ -5048,8 +5097,6 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) T6_STATMODE_V(0))); for_each_port(adapter, i) { - struct net_device *netdev; - netdev = alloc_etherdev_mq(sizeof(struct port_info), MAX_ETH_QSETS); if (!netdev) { @@ -5217,6 +5264,7 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) attach_ulds(adapter); print_adapter_info(adapter); + return 0; sriov: #ifdef CONFIG_PCI_IOV @@ -5230,8 +5278,57 @@ sriov: "instantiated %u virtual functions\n", num_vf[func]); } -#endif + + adapter = kzalloc(sizeof(*adapter), GFP_KERNEL); + if (!adapter) { + err = -ENOMEM; + goto free_pci_region; + } + + snprintf(name, IFNAMSIZ, "mgmtpf%d%d", adap_idx, func); + netdev = alloc_netdev(0, name, NET_NAME_UNKNOWN, ether_setup); + if (!netdev) { + err = -ENOMEM; + goto free_adapter; + } + + adapter->pdev = pdev; + adapter->pdev_dev = &pdev->dev; + adapter->name = pci_name(pdev); + adapter->mbox = func; + adapter->pf = func; + adapter->regs = regs; + adapter->mbox_log = kzalloc(sizeof(*adapter->mbox_log) + + (sizeof(struct mbox_cmd) * + T4_OS_LOG_MBOX_CMDS), + GFP_KERNEL); + if (!adapter->mbox_log) { + err = -ENOMEM; + goto free_netdevice; + } + pi = netdev_priv(netdev); + pi->adapter = adapter; + SET_NETDEV_DEV(netdev, &pdev->dev); + pci_set_drvdata(pdev, adapter); + + adapter->port[0] = netdev; + netdev->netdev_ops = &cxgb4_mgmt_netdev_ops; + netdev->ethtool_ops = &cxgb4_mgmt_ethtool_ops; + + return 0; + + free_netdevice: + free_netdev(adapter->port[0]); + free_adapter: + kfree(adapter); + free_pci_region: + iounmap(regs); + pci_disable_sriov(pdev); + pci_release_regions(pdev); + return err; +#else return 0; +#endif out_free_dev: free_some_resources(adapter); @@ -5258,12 +5355,12 @@ static void remove_one(struct pci_dev *pdev) { struct adapter *adapter = pci_get_drvdata(pdev); -#ifdef CONFIG_PCI_IOV - pci_disable_sriov(pdev); - -#endif + if (!adapter) { + pci_release_regions(pdev); + return; + } - if (adapter) { + if (adapter->pf == 4) { int i; /* Tear down per-adapter Work Queue first since it can contain @@ -5312,8 +5409,18 @@ static void remove_one(struct pci_dev *pdev) kfree(adapter->mbox_log); synchronize_rcu(); kfree(adapter); - } else + } +#ifdef CONFIG_PCI_IOV + else { + if (adapter->port[0]->reg_state == NETREG_REGISTERED) + unregister_netdev(adapter->port[0]); + free_netdev(adapter->port[0]); + iounmap(adapter->regs); + kfree(adapter); + pci_disable_sriov(pdev); pci_release_regions(pdev); + } +#endif } static struct pci_driver cxgb4_driver = { diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c index dc92c80a75f4..2a476cc4e073 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c @@ -8264,3 +8264,44 @@ void t4_idma_monitor(struct adapter *adapter, t4_sge_decode_idma_state(adapter, idma->idma_state[i]); } } + +/** + * t4_set_vf_mac - Set MAC address for the specified VF + * @adapter: The adapter + * @vf: one of the VFs instantiated by the specified PF + * @naddr: the number of MAC addresses + * @addr: the MAC address(es) to be set to the specified VF + */ +int t4_set_vf_mac_acl(struct adapter *adapter, unsigned int vf, + unsigned int naddr, u8 *addr) +{ + struct fw_acl_mac_cmd cmd; + + memset(&cmd, 0, sizeof(cmd)); + cmd.op_to_vfn = cpu_to_be32(FW_CMD_OP_V(FW_ACL_MAC_CMD) | + FW_CMD_REQUEST_F | + FW_CMD_WRITE_F | + FW_ACL_MAC_CMD_PFN_V(adapter->pf) | + FW_ACL_MAC_CMD_VFN_V(vf)); + + /* Note: Do not enable the ACL */ + cmd.en_to_len16 = cpu_to_be32((unsigned int)FW_LEN16(cmd)); + cmd.nmac = naddr; + + switch (adapter->pf) { + case 3: + memcpy(cmd.macaddr3, addr, sizeof(cmd.macaddr3)); + break; + case 2: + memcpy(cmd.macaddr2, addr, sizeof(cmd.macaddr2)); + break; + case 1: + memcpy(cmd.macaddr1, addr, sizeof(cmd.macaddr1)); + break; + case 0: + memcpy(cmd.macaddr0, addr, sizeof(cmd.macaddr0)); + break; + } + + return t4_wr_mbox(adapter, adapter->mbox, &cmd, sizeof(cmd), &cmd); +} diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c index e116bb8d1729..f2951bf68992 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c @@ -2777,6 +2777,7 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev, struct adapter *adapter; struct port_info *pi; struct net_device *netdev; + unsigned int pf; /* * Print our driver banner the first time we're called to initialize a @@ -2903,8 +2904,11 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev, * Allocate our "adapter ports" and stitch everything together. */ pmask = adapter->params.vfres.pmask; + pf = t4vf_get_pf_from_vf(adapter); for_each_port(adapter, pidx) { int port_id, viid; + u8 mac[ETH_ALEN]; + unsigned int naddr = 1; /* * We simplistically allocate our virtual interfaces @@ -2975,6 +2979,26 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev, pidx); goto err_free_dev; } + + err = t4vf_get_vf_mac_acl(adapter, pf, &naddr, mac); + if (err) { + dev_err(&pdev->dev, + "unable to determine MAC ACL address, " + "continuing anyway.. (status %d)\n", err); + } else if (naddr && adapter->params.vfres.nvi == 1) { + struct sockaddr addr; + + ether_addr_copy(addr.sa_data, mac); + err = cxgb4vf_set_mac_addr(netdev, &addr); + if (err) { + dev_err(&pdev->dev, + "unable to set MAC address %pM\n", + mac); + goto err_free_dev; + } + dev_info(&pdev->dev, + "Using assigned MAC ACL: %pM\n", mac); + } } /* See what interrupts we'll be using. If we've been configured to diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h index 8ee541431e8b..8067424ad4a8 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h +++ b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h @@ -347,6 +347,7 @@ int t4vf_bar2_sge_qregs(struct adapter *adapter, u64 *pbar2_qoffset, unsigned int *pbar2_qid); +unsigned int t4vf_get_pf_from_vf(struct adapter *); int t4vf_get_sge_params(struct adapter *); int t4vf_get_vpd_params(struct adapter *); int t4vf_get_dev_params(struct adapter *); @@ -381,5 +382,7 @@ int t4vf_eth_eq_free(struct adapter *, unsigned int); int t4vf_handle_fw_rpl(struct adapter *, const __be64 *); int t4vf_prep_adapter(struct adapter *); +int t4vf_get_vf_mac_acl(struct adapter *adapter, unsigned int pf, + unsigned int *naddr, u8 *addr); #endif /* __T4VF_COMMON_H__ */ diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c index 427bfa71388b..879f4c52b3d5 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c +++ b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c @@ -639,6 +639,15 @@ int t4vf_bar2_sge_qregs(struct adapter *adapter, return 0; } +unsigned int t4vf_get_pf_from_vf(struct adapter *adapter) +{ + u32 whoami; + + whoami = t4_read_reg(adapter, T4VF_PL_BASE_ADDR + PL_VF_WHOAMI_A); + return (CHELSIO_CHIP_VERSION(adapter->params.chip) <= CHELSIO_T5 ? + SOURCEPF_G(whoami) : T6_SOURCEPF_G(whoami)); +} + /** * t4vf_get_sge_params - retrieve adapter Scatter gather Engine parameters * @adapter: the adapter @@ -716,7 +725,6 @@ int t4vf_get_sge_params(struct adapter *adapter) * read. */ if (!is_t4(adapter->params.chip)) { - u32 whoami; unsigned int pf, s_hps, s_qpp; params[0] = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_REG) | @@ -740,11 +748,7 @@ int t4vf_get_sge_params(struct adapter *adapter) * register we just read. Do it once here so other code in * the driver can just use it. */ - whoami = t4_read_reg(adapter, - T4VF_PL_BASE_ADDR + PL_VF_WHOAMI_A); - pf = CHELSIO_CHIP_VERSION(adapter->params.chip) <= CHELSIO_T5 ? - SOURCEPF_G(whoami) : T6_SOURCEPF_G(whoami); - + pf = t4vf_get_pf_from_vf(adapter); s_hps = (HOSTPAGESIZEPF0_S + (HOSTPAGESIZEPF1_S - HOSTPAGESIZEPF0_S) * pf); sge_params->sge_vf_hps = @@ -1807,3 +1811,50 @@ int t4vf_prep_adapter(struct adapter *adapter) return 0; } + +/** + * t4vf_get_vf_mac_acl - Get the MAC address to be set to + * the VI of this VF. + * @adapter: The adapter + * @pf: The pf associated with vf + * @naddr: the number of ACL MAC addresses returned in addr + * @addr: Placeholder for MAC addresses + * + * Find the MAC address to be set to the VF's VI. The requested MAC address + * is from the host OS via callback in the PF driver. + */ +int t4vf_get_vf_mac_acl(struct adapter *adapter, unsigned int pf, + unsigned int *naddr, u8 *addr) +{ + struct fw_acl_mac_cmd cmd; + int ret; + + memset(&cmd, 0, sizeof(cmd)); + cmd.op_to_vfn = cpu_to_be32(FW_CMD_OP_V(FW_ACL_MAC_CMD) | + FW_CMD_REQUEST_F | + FW_CMD_READ_F); + cmd.en_to_len16 = cpu_to_be32((unsigned int)FW_LEN16(cmd)); + ret = t4vf_wr_mbox(adapter, &cmd, sizeof(cmd), &cmd); + if (ret) + return ret; + + if (cmd.nmac < *naddr) + *naddr = cmd.nmac; + + switch (pf) { + case 3: + memcpy(addr, cmd.macaddr3, sizeof(cmd.macaddr3)); + break; + case 2: + memcpy(addr, cmd.macaddr2, sizeof(cmd.macaddr2)); + break; + case 1: + memcpy(addr, cmd.macaddr1, sizeof(cmd.macaddr1)); + break; + case 0: + memcpy(addr, cmd.macaddr0, sizeof(cmd.macaddr0)); + break; + } + + return ret; +} diff --git a/drivers/net/ethernet/dec/tulip/de4x5.c b/drivers/net/ethernet/dec/tulip/de4x5.c index f0e9e2ef62a0..6620fc861c47 100644 --- a/drivers/net/ethernet/dec/tulip/de4x5.c +++ b/drivers/net/ethernet/dec/tulip/de4x5.c @@ -1966,7 +1966,7 @@ SetMulticastFilter(struct net_device *dev) } else if (lp->setup_f == HASH_PERF) { /* Hash Filtering */ netdev_for_each_mc_addr(ha, dev) { crc = ether_crc_le(ETH_ALEN, ha->addr); - hashcode = crc & HASH_BITS; /* hashcode is 9 LSb of CRC */ + hashcode = crc & DE4X5_HASH_BITS; /* hashcode is 9 LSb of CRC */ byte = hashcode >> 3; /* bit[3-8] -> byte in filter */ bit = 1 << (hashcode & 0x07);/* bit[0-2] -> bit in byte */ @@ -5043,7 +5043,7 @@ build_setup_frame(struct net_device *dev, int mode) *(pa + i) = dev->dev_addr[i]; /* Host address */ if (i & 0x01) pa += 2; } - *(lp->setup_frame + (HASH_TABLE_LEN >> 3) - 3) = 0x80; + *(lp->setup_frame + (DE4X5_HASH_TABLE_LEN >> 3) - 3) = 0x80; } else { for (i=0; i<ETH_ALEN; i++) { /* Host address */ *(pa + (i&1)) = dev->dev_addr[i]; diff --git a/drivers/net/ethernet/dec/tulip/de4x5.h b/drivers/net/ethernet/dec/tulip/de4x5.h index ec756eba397b..1bfdc9b117f6 100644 --- a/drivers/net/ethernet/dec/tulip/de4x5.h +++ b/drivers/net/ethernet/dec/tulip/de4x5.h @@ -860,8 +860,8 @@ #define PCI 0 #define EISA 1 -#define HASH_TABLE_LEN 512 /* Bits */ -#define HASH_BITS 0x01ff /* 9 LS bits */ +#define DE4X5_HASH_TABLE_LEN 512 /* Bits */ +#define DE4X5_HASH_BITS 0x01ff /* 9 LS bits */ #define SETUP_FRAME_LEN 192 /* Bytes */ #define IMPERF_PA_OFFSET 156 /* Bytes */ diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h index 4555e041ef69..86780b5c40ef 100644 --- a/drivers/net/ethernet/emulex/benet/be.h +++ b/drivers/net/ethernet/emulex/benet/be.h @@ -508,6 +508,10 @@ struct be_wrb_params { u16 lso_mss; /* MSS for LSO */ }; +struct be_eth_addr { + unsigned char mac[ETH_ALEN]; +}; + struct be_adapter { struct pci_dev *pdev; struct net_device *netdev; @@ -523,7 +527,7 @@ struct be_adapter { struct be_dma_mem mbox_mem_alloced; struct be_mcc_obj mcc_obj; - spinlock_t mcc_lock; /* For serializing mcc cmds to BE card */ + struct mutex mcc_lock; /* For serializing mcc cmds to BE card */ spinlock_t mcc_cq_lock; u16 cfg_num_rx_irqs; /* configured via set-channels */ @@ -570,9 +574,15 @@ struct be_adapter { int if_handle; /* Used to configure filtering */ u32 if_flags; /* Interface filtering flags */ u32 *pmac_id; /* MAC addr handle used by BE card */ + struct be_eth_addr *uc_list;/* list of uc-addrs programmed (not perm) */ u32 uc_macs; /* Count of secondary UC MAC programmed */ + struct be_eth_addr *mc_list;/* list of mcast addrs programmed */ + u32 mc_count; unsigned long vids[BITS_TO_LONGS(VLAN_N_VID)]; u16 vlans_added; + bool update_uc_list; + bool update_mc_list; + struct mutex rx_filter_lock;/* For protecting vids[] & mc/uc_list[] */ u32 beacon_state; /* for set_phys_id */ @@ -626,6 +636,15 @@ struct be_adapter { u8 phy_state; /* state of sfp optics (functional, faulted, etc.,) */ }; +/* Used for defered FW config cmds. Add fields to this struct as reqd */ +struct be_cmd_work { + struct work_struct work; + struct be_adapter *adapter; + union { + __be16 vxlan_port; + } info; +}; + #define be_physfn(adapter) (!adapter->virtfn) #define be_virtfn(adapter) (adapter->virtfn) #define sriov_enabled(adapter) (adapter->flags & \ diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c index 2cc11756859f..fa11a5a8c354 100644 --- a/drivers/net/ethernet/emulex/benet/be_cmds.c +++ b/drivers/net/ethernet/emulex/benet/be_cmds.c @@ -571,7 +571,7 @@ int be_process_mcc(struct be_adapter *adapter) /* Wait till no more pending mcc requests are present */ static int be_mcc_wait_compl(struct be_adapter *adapter) { -#define mcc_timeout 120000 /* 12s timeout */ +#define mcc_timeout 12000 /* 12s timeout */ int i, status = 0; struct be_mcc_obj *mcc_obj = &adapter->mcc_obj; @@ -585,7 +585,7 @@ static int be_mcc_wait_compl(struct be_adapter *adapter) if (atomic_read(&mcc_obj->q.used) == 0) break; - udelay(100); + usleep_range(500, 1000); } if (i == mcc_timeout) { dev_err(&adapter->pdev->dev, "FW not responding\n"); @@ -863,7 +863,7 @@ static bool use_mcc(struct be_adapter *adapter) static int be_cmd_lock(struct be_adapter *adapter) { if (use_mcc(adapter)) { - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); return 0; } else { return mutex_lock_interruptible(&adapter->mbox_lock); @@ -874,7 +874,7 @@ static int be_cmd_lock(struct be_adapter *adapter) static void be_cmd_unlock(struct be_adapter *adapter) { if (use_mcc(adapter)) - spin_unlock_bh(&adapter->mcc_lock); + return mutex_unlock(&adapter->mcc_lock); else return mutex_unlock(&adapter->mbox_lock); } @@ -1044,7 +1044,7 @@ int be_cmd_mac_addr_query(struct be_adapter *adapter, u8 *mac_addr, struct be_cmd_req_mac_query *req; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -1073,7 +1073,7 @@ int be_cmd_mac_addr_query(struct be_adapter *adapter, u8 *mac_addr, } err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -1085,7 +1085,7 @@ int be_cmd_pmac_add(struct be_adapter *adapter, u8 *mac_addr, struct be_cmd_req_pmac_add *req; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -1110,7 +1110,7 @@ int be_cmd_pmac_add(struct be_adapter *adapter, u8 *mac_addr, } err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); if (status == MCC_STATUS_UNAUTHORIZED_REQUEST) status = -EPERM; @@ -1128,7 +1128,7 @@ int be_cmd_pmac_del(struct be_adapter *adapter, u32 if_id, int pmac_id, u32 dom) if (pmac_id == -1) return 0; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -1148,7 +1148,7 @@ int be_cmd_pmac_del(struct be_adapter *adapter, u32 if_id, int pmac_id, u32 dom) status = be_mcc_notify_wait(adapter); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -1411,7 +1411,7 @@ int be_cmd_rxq_create(struct be_adapter *adapter, struct be_dma_mem *q_mem = &rxq->dma_mem; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -1441,7 +1441,7 @@ int be_cmd_rxq_create(struct be_adapter *adapter, } err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -1505,7 +1505,7 @@ int be_cmd_rxq_destroy(struct be_adapter *adapter, struct be_queue_info *q) struct be_cmd_req_q_destroy *req; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -1522,7 +1522,7 @@ int be_cmd_rxq_destroy(struct be_adapter *adapter, struct be_queue_info *q) q->created = false; err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -1590,7 +1590,7 @@ int be_cmd_get_stats(struct be_adapter *adapter, struct be_dma_mem *nonemb_cmd) struct be_cmd_req_hdr *hdr; int status = 0; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -1618,7 +1618,7 @@ int be_cmd_get_stats(struct be_adapter *adapter, struct be_dma_mem *nonemb_cmd) adapter->stats_cmd_sent = true; err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -1634,7 +1634,7 @@ int lancer_cmd_get_pport_stats(struct be_adapter *adapter, CMD_SUBSYSTEM_ETH)) return -EPERM; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -1657,7 +1657,7 @@ int lancer_cmd_get_pport_stats(struct be_adapter *adapter, adapter->stats_cmd_sent = true; err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -1694,7 +1694,7 @@ int be_cmd_link_status_query(struct be_adapter *adapter, u16 *link_speed, struct be_cmd_req_link_status *req; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); if (link_status) *link_status = LINK_DOWN; @@ -1733,7 +1733,7 @@ int be_cmd_link_status_query(struct be_adapter *adapter, u16 *link_speed, } err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -1744,7 +1744,7 @@ int be_cmd_get_die_temperature(struct be_adapter *adapter) struct be_cmd_req_get_cntl_addnl_attribs *req; int status = 0; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -1759,7 +1759,7 @@ int be_cmd_get_die_temperature(struct be_adapter *adapter) status = be_mcc_notify(adapter); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -1808,7 +1808,7 @@ int be_cmd_get_fat_dump(struct be_adapter *adapter, u32 buf_len, void *buf) if (!get_fat_cmd.va) return -ENOMEM; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); while (total_size) { buf_size = min(total_size, (u32)60*1024); @@ -1848,7 +1848,7 @@ int be_cmd_get_fat_dump(struct be_adapter *adapter, u32 buf_len, void *buf) err: dma_free_coherent(&adapter->pdev->dev, get_fat_cmd.size, get_fat_cmd.va, get_fat_cmd.dma); - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -1859,7 +1859,7 @@ int be_cmd_get_fw_ver(struct be_adapter *adapter) struct be_cmd_req_get_fw_version *req; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -1882,7 +1882,7 @@ int be_cmd_get_fw_ver(struct be_adapter *adapter) sizeof(adapter->fw_on_flash)); } err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -1896,7 +1896,7 @@ static int __be_cmd_modify_eqd(struct be_adapter *adapter, struct be_cmd_req_modify_eq_delay *req; int status = 0, i; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -1919,7 +1919,7 @@ static int __be_cmd_modify_eqd(struct be_adapter *adapter, status = be_mcc_notify(adapter); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -1946,7 +1946,7 @@ int be_cmd_vlan_config(struct be_adapter *adapter, u32 if_id, u16 *vtag_array, struct be_cmd_req_vlan_config *req; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -1968,7 +1968,7 @@ int be_cmd_vlan_config(struct be_adapter *adapter, u32 if_id, u16 *vtag_array, status = be_mcc_notify_wait(adapter); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -1979,7 +1979,7 @@ static int __be_cmd_rx_filter(struct be_adapter *adapter, u32 flags, u32 value) struct be_cmd_req_rx_filter *req = mem->va; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -1996,8 +1996,7 @@ static int __be_cmd_rx_filter(struct be_adapter *adapter, u32 flags, u32 value) req->if_flags = (value == ON) ? req->if_flags_mask : 0; if (flags & BE_IF_FLAGS_MULTICAST) { - struct netdev_hw_addr *ha; - int i = 0; + int i; /* Reset mcast promisc mode if already set by setting mask * and not setting flags field @@ -2005,14 +2004,15 @@ static int __be_cmd_rx_filter(struct be_adapter *adapter, u32 flags, u32 value) req->if_flags_mask |= cpu_to_le32(BE_IF_FLAGS_MCAST_PROMISCUOUS & be_if_cap_flags(adapter)); - req->mcast_num = cpu_to_le32(netdev_mc_count(adapter->netdev)); - netdev_for_each_mc_addr(ha, adapter->netdev) - memcpy(req->mcast_mac[i++].byte, ha->addr, ETH_ALEN); + req->mcast_num = cpu_to_le32(adapter->mc_count); + for (i = 0; i < adapter->mc_count; i++) + ether_addr_copy(req->mcast_mac[i].byte, + adapter->mc_list[i].mac); } status = be_mcc_notify_wait(adapter); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -2043,7 +2043,7 @@ int be_cmd_set_flow_control(struct be_adapter *adapter, u32 tx_fc, u32 rx_fc) CMD_SUBSYSTEM_COMMON)) return -EPERM; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -2063,7 +2063,7 @@ int be_cmd_set_flow_control(struct be_adapter *adapter, u32 tx_fc, u32 rx_fc) status = be_mcc_notify_wait(adapter); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); if (base_status(status) == MCC_STATUS_FEATURE_NOT_SUPPORTED) return -EOPNOTSUPP; @@ -2082,7 +2082,7 @@ int be_cmd_get_flow_control(struct be_adapter *adapter, u32 *tx_fc, u32 *rx_fc) CMD_SUBSYSTEM_COMMON)) return -EPERM; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -2105,7 +2105,7 @@ int be_cmd_get_flow_control(struct be_adapter *adapter, u32 *tx_fc, u32 *rx_fc) } err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -2186,7 +2186,7 @@ int be_cmd_rss_config(struct be_adapter *adapter, u8 *rsstable, if (!(be_if_cap_flags(adapter) & BE_IF_FLAGS_RSS)) return 0; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -2211,7 +2211,7 @@ int be_cmd_rss_config(struct be_adapter *adapter, u8 *rsstable, status = be_mcc_notify_wait(adapter); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -2223,7 +2223,7 @@ int be_cmd_set_beacon_state(struct be_adapter *adapter, u8 port_num, struct be_cmd_req_enable_disable_beacon *req; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -2244,7 +2244,7 @@ int be_cmd_set_beacon_state(struct be_adapter *adapter, u8 port_num, status = be_mcc_notify_wait(adapter); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -2255,7 +2255,7 @@ int be_cmd_get_beacon_state(struct be_adapter *adapter, u8 port_num, u32 *state) struct be_cmd_req_get_beacon_state *req; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -2279,7 +2279,7 @@ int be_cmd_get_beacon_state(struct be_adapter *adapter, u8 port_num, u32 *state) } err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -2303,7 +2303,7 @@ int be_cmd_read_port_transceiver_data(struct be_adapter *adapter, return -ENOMEM; } - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -2325,7 +2325,7 @@ int be_cmd_read_port_transceiver_data(struct be_adapter *adapter, memcpy(data, resp->page_data, PAGE_DATA_LEN); } err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); dma_free_coherent(&adapter->pdev->dev, cmd.size, cmd.va, cmd.dma); return status; } @@ -2342,7 +2342,7 @@ static int lancer_cmd_write_object(struct be_adapter *adapter, void *ctxt = NULL; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); adapter->flash_status = 0; wrb = wrb_from_mccq(adapter); @@ -2384,7 +2384,7 @@ static int lancer_cmd_write_object(struct be_adapter *adapter, if (status) goto err_unlock; - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); if (!wait_for_completion_timeout(&adapter->et_cmd_compl, msecs_to_jiffies(60000))) @@ -2403,7 +2403,7 @@ static int lancer_cmd_write_object(struct be_adapter *adapter, return status; err_unlock: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -2457,7 +2457,7 @@ static int lancer_cmd_delete_object(struct be_adapter *adapter, struct be_mcc_wrb *wrb; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -2475,7 +2475,7 @@ static int lancer_cmd_delete_object(struct be_adapter *adapter, status = be_mcc_notify_wait(adapter); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -2488,7 +2488,7 @@ int lancer_cmd_read_object(struct be_adapter *adapter, struct be_dma_mem *cmd, struct lancer_cmd_resp_read_object *resp; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -2522,7 +2522,7 @@ int lancer_cmd_read_object(struct be_adapter *adapter, struct be_dma_mem *cmd, } err_unlock: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -2534,7 +2534,7 @@ static int be_cmd_write_flashrom(struct be_adapter *adapter, struct be_cmd_write_flashrom *req; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); adapter->flash_status = 0; wrb = wrb_from_mccq(adapter); @@ -2559,7 +2559,7 @@ static int be_cmd_write_flashrom(struct be_adapter *adapter, if (status) goto err_unlock; - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); if (!wait_for_completion_timeout(&adapter->et_cmd_compl, msecs_to_jiffies(40000))) @@ -2570,7 +2570,7 @@ static int be_cmd_write_flashrom(struct be_adapter *adapter, return status; err_unlock: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -2581,7 +2581,7 @@ static int be_cmd_get_flash_crc(struct be_adapter *adapter, u8 *flashed_crc, struct be_mcc_wrb *wrb; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -2608,7 +2608,7 @@ static int be_cmd_get_flash_crc(struct be_adapter *adapter, u8 *flashed_crc, memcpy(flashed_crc, req->crc, 4); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -3192,7 +3192,7 @@ int be_cmd_enable_magic_wol(struct be_adapter *adapter, u8 *mac, struct be_cmd_req_acpi_wol_magic_config *req; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -3209,7 +3209,7 @@ int be_cmd_enable_magic_wol(struct be_adapter *adapter, u8 *mac, status = be_mcc_notify_wait(adapter); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -3224,7 +3224,7 @@ int be_cmd_set_loopback(struct be_adapter *adapter, u8 port_num, CMD_SUBSYSTEM_LOWLEVEL)) return -EPERM; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -3247,7 +3247,7 @@ int be_cmd_set_loopback(struct be_adapter *adapter, u8 port_num, if (status) goto err_unlock; - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); if (!wait_for_completion_timeout(&adapter->et_cmd_compl, msecs_to_jiffies(SET_LB_MODE_TIMEOUT))) @@ -3256,7 +3256,7 @@ int be_cmd_set_loopback(struct be_adapter *adapter, u8 port_num, return status; err_unlock: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -3273,7 +3273,7 @@ int be_cmd_loopback_test(struct be_adapter *adapter, u32 port_num, CMD_SUBSYSTEM_LOWLEVEL)) return -EPERM; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -3299,7 +3299,7 @@ int be_cmd_loopback_test(struct be_adapter *adapter, u32 port_num, if (status) goto err; - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); wait_for_completion(&adapter->et_cmd_compl); resp = embedded_payload(wrb); @@ -3307,7 +3307,7 @@ int be_cmd_loopback_test(struct be_adapter *adapter, u32 port_num, return status; err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -3323,7 +3323,7 @@ int be_cmd_ddr_dma_test(struct be_adapter *adapter, u64 pattern, CMD_SUBSYSTEM_LOWLEVEL)) return -EPERM; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -3357,7 +3357,7 @@ int be_cmd_ddr_dma_test(struct be_adapter *adapter, u64 pattern, } err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -3368,7 +3368,7 @@ int be_cmd_get_seeprom_data(struct be_adapter *adapter, struct be_cmd_req_seeprom_read *req; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -3384,7 +3384,7 @@ int be_cmd_get_seeprom_data(struct be_adapter *adapter, status = be_mcc_notify_wait(adapter); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -3399,7 +3399,7 @@ int be_cmd_get_phy_info(struct be_adapter *adapter) CMD_SUBSYSTEM_COMMON)) return -EPERM; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -3444,7 +3444,7 @@ int be_cmd_get_phy_info(struct be_adapter *adapter) } dma_free_coherent(&adapter->pdev->dev, cmd.size, cmd.va, cmd.dma); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -3454,7 +3454,7 @@ static int be_cmd_set_qos(struct be_adapter *adapter, u32 bps, u32 domain) struct be_cmd_req_set_qos *req; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -3474,7 +3474,7 @@ static int be_cmd_set_qos(struct be_adapter *adapter, u32 bps, u32 domain) status = be_mcc_notify_wait(adapter); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -3581,7 +3581,7 @@ int be_cmd_get_fn_privileges(struct be_adapter *adapter, u32 *privilege, struct be_cmd_req_get_fn_privileges *req; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -3613,7 +3613,7 @@ int be_cmd_get_fn_privileges(struct be_adapter *adapter, u32 *privilege, } err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -3625,7 +3625,7 @@ int be_cmd_set_fn_privileges(struct be_adapter *adapter, u32 privileges, struct be_cmd_req_set_fn_privileges *req; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -3645,7 +3645,7 @@ int be_cmd_set_fn_privileges(struct be_adapter *adapter, u32 privileges, status = be_mcc_notify_wait(adapter); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -3677,7 +3677,7 @@ int be_cmd_get_mac_from_list(struct be_adapter *adapter, u8 *mac, return -ENOMEM; } - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -3741,7 +3741,7 @@ int be_cmd_get_mac_from_list(struct be_adapter *adapter, u8 *mac, } out: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); dma_free_coherent(&adapter->pdev->dev, get_mac_list_cmd.size, get_mac_list_cmd.va, get_mac_list_cmd.dma); return status; @@ -3801,7 +3801,7 @@ int be_cmd_set_mac_list(struct be_adapter *adapter, u8 *mac_array, if (!cmd.va) return -ENOMEM; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -3823,7 +3823,7 @@ int be_cmd_set_mac_list(struct be_adapter *adapter, u8 *mac_array, err: dma_free_coherent(&adapter->pdev->dev, cmd.size, cmd.va, cmd.dma); - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -3859,7 +3859,7 @@ int be_cmd_set_hsw_config(struct be_adapter *adapter, u16 pvid, CMD_SUBSYSTEM_COMMON)) return -EPERM; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -3900,7 +3900,7 @@ int be_cmd_set_hsw_config(struct be_adapter *adapter, u16 pvid, status = be_mcc_notify_wait(adapter); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -3914,7 +3914,7 @@ int be_cmd_get_hsw_config(struct be_adapter *adapter, u16 *pvid, int status; u16 vid; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -3961,7 +3961,7 @@ int be_cmd_get_hsw_config(struct be_adapter *adapter, u16 *pvid, } err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -4156,7 +4156,7 @@ int be_cmd_set_ext_fat_capabilites(struct be_adapter *adapter, struct be_cmd_req_set_ext_fat_caps *req; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -4172,7 +4172,7 @@ int be_cmd_set_ext_fat_capabilites(struct be_adapter *adapter, status = be_mcc_notify_wait(adapter); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -4650,7 +4650,7 @@ int be_cmd_manage_iface(struct be_adapter *adapter, u32 iface, u8 op) if (iface == 0xFFFFFFFF) return -1; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -4667,7 +4667,7 @@ int be_cmd_manage_iface(struct be_adapter *adapter, u32 iface, u8 op) status = be_mcc_notify_wait(adapter); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -4701,7 +4701,7 @@ int be_cmd_get_if_id(struct be_adapter *adapter, struct be_vf_cfg *vf_cfg, struct be_cmd_resp_get_iface_list *resp; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -4722,7 +4722,7 @@ int be_cmd_get_if_id(struct be_adapter *adapter, struct be_vf_cfg *vf_cfg, } err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -4816,7 +4816,7 @@ int be_cmd_enable_vf(struct be_adapter *adapter, u8 domain) if (BEx_chip(adapter)) return 0; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -4834,7 +4834,7 @@ int be_cmd_enable_vf(struct be_adapter *adapter, u8 domain) req->enable = 1; status = be_mcc_notify_wait(adapter); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -4905,7 +4905,7 @@ int __be_cmd_set_logical_link_config(struct be_adapter *adapter, struct be_cmd_req_set_ll_link *req; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -4931,7 +4931,7 @@ int __be_cmd_set_logical_link_config(struct be_adapter *adapter, status = be_mcc_notify_wait(adapter); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } @@ -4964,7 +4964,7 @@ int be_roce_mcc_cmd(void *netdev_handle, void *wrb_payload, struct be_cmd_resp_hdr *resp; int status; - spin_lock_bh(&adapter->mcc_lock); + mutex_lock(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); if (!wrb) { @@ -4987,7 +4987,7 @@ int be_roce_mcc_cmd(void *netdev_handle, void *wrb_payload, memcpy(wrb_payload, resp, sizeof(*resp) + resp->response_length); be_dws_le_to_cpu(wrb_payload, sizeof(*resp) + resp->response_length); err: - spin_unlock_bh(&adapter->mcc_lock); + mutex_unlock(&adapter->mcc_lock); return status; } EXPORT_SYMBOL(be_roce_mcc_cmd); diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index 874c7539a79d..f7584d4139ff 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -53,6 +53,10 @@ static const struct pci_device_id be_dev_ids[] = { { 0 } }; MODULE_DEVICE_TABLE(pci, be_dev_ids); + +/* Workqueue used by all functions for defering cmd calls to the adapter */ +struct workqueue_struct *be_wq; + /* UE Status Low CSR */ static const char * const ue_status_low_desc[] = { "CEV", @@ -1420,13 +1424,18 @@ static int be_vid_config(struct be_adapter *adapter) u16 num = 0, i = 0; int status = 0; - /* No need to further configure vids if in promiscuous mode */ - if (be_in_all_promisc(adapter)) + /* No need to change the VLAN state if the I/F is in promiscuous */ + if (adapter->netdev->flags & IFF_PROMISC) return 0; if (adapter->vlans_added > be_max_vlans(adapter)) return be_set_vlan_promisc(adapter); + if (adapter->if_flags & BE_IF_FLAGS_VLAN_PROMISCUOUS) { + status = be_clear_vlan_promisc(adapter); + if (status) + return status; + } /* Construct VLAN Table to give to HW */ for_each_set_bit(i, adapter->vids, VLAN_N_VID) vids[num++] = cpu_to_le16(i); @@ -1439,8 +1448,6 @@ static int be_vid_config(struct be_adapter *adapter) addl_status(status) == MCC_ADDL_STATUS_INSUFFICIENT_RESOURCES) return be_set_vlan_promisc(adapter); - } else if (adapter->if_flags & BE_IF_FLAGS_VLAN_PROMISCUOUS) { - status = be_clear_vlan_promisc(adapter); } return status; } @@ -1450,46 +1457,45 @@ static int be_vlan_add_vid(struct net_device *netdev, __be16 proto, u16 vid) struct be_adapter *adapter = netdev_priv(netdev); int status = 0; + mutex_lock(&adapter->rx_filter_lock); + /* Packets with VID 0 are always received by Lancer by default */ if (lancer_chip(adapter) && vid == 0) - return status; + goto done; if (test_bit(vid, adapter->vids)) - return status; + goto done; set_bit(vid, adapter->vids); adapter->vlans_added++; status = be_vid_config(adapter); - if (status) { - adapter->vlans_added--; - clear_bit(vid, adapter->vids); - } - +done: + mutex_unlock(&adapter->rx_filter_lock); return status; } static int be_vlan_rem_vid(struct net_device *netdev, __be16 proto, u16 vid) { struct be_adapter *adapter = netdev_priv(netdev); + int status = 0; + + mutex_lock(&adapter->rx_filter_lock); /* Packets with VID 0 are always received by Lancer by default */ if (lancer_chip(adapter) && vid == 0) - return 0; + goto done; if (!test_bit(vid, adapter->vids)) - return 0; + goto done; clear_bit(vid, adapter->vids); adapter->vlans_added--; - return be_vid_config(adapter); -} - -static void be_clear_all_promisc(struct be_adapter *adapter) -{ - be_cmd_rx_filter(adapter, BE_IF_FLAGS_ALL_PROMISCUOUS, OFF); - adapter->if_flags &= ~BE_IF_FLAGS_ALL_PROMISCUOUS; + status = be_vid_config(adapter); +done: + mutex_unlock(&adapter->rx_filter_lock); + return status; } static void be_set_all_promisc(struct be_adapter *adapter) @@ -1510,75 +1516,207 @@ static void be_set_mc_promisc(struct be_adapter *adapter) adapter->if_flags |= BE_IF_FLAGS_MCAST_PROMISCUOUS; } -static void be_set_mc_list(struct be_adapter *adapter) +static void be_set_uc_promisc(struct be_adapter *adapter) { int status; - status = be_cmd_rx_filter(adapter, BE_IF_FLAGS_MULTICAST, ON); + if (adapter->if_flags & BE_IF_FLAGS_PROMISCUOUS) + return; + + status = be_cmd_rx_filter(adapter, BE_IF_FLAGS_PROMISCUOUS, ON); if (!status) - adapter->if_flags &= ~BE_IF_FLAGS_MCAST_PROMISCUOUS; - else + adapter->if_flags |= BE_IF_FLAGS_PROMISCUOUS; +} + +static void be_clear_uc_promisc(struct be_adapter *adapter) +{ + int status; + + if (!(adapter->if_flags & BE_IF_FLAGS_PROMISCUOUS)) + return; + + status = be_cmd_rx_filter(adapter, BE_IF_FLAGS_PROMISCUOUS, OFF); + if (!status) + adapter->if_flags &= ~BE_IF_FLAGS_PROMISCUOUS; +} + +/* The below 2 functions are the callback args for __dev_mc_sync/dev_uc_sync(). + * We use a single callback function for both sync and unsync. We really don't + * add/remove addresses through this callback. But, we use it to detect changes + * to the uc/mc lists. The entire uc/mc list is programmed in be_set_rx_mode(). + */ +static int be_uc_list_update(struct net_device *netdev, + const unsigned char *addr) +{ + struct be_adapter *adapter = netdev_priv(netdev); + + adapter->update_uc_list = true; + return 0; +} + +static int be_mc_list_update(struct net_device *netdev, + const unsigned char *addr) +{ + struct be_adapter *adapter = netdev_priv(netdev); + + adapter->update_mc_list = true; + return 0; +} + +static void be_set_mc_list(struct be_adapter *adapter) +{ + struct net_device *netdev = adapter->netdev; + struct netdev_hw_addr *ha; + bool mc_promisc = false; + int status; + + netif_addr_lock_bh(netdev); + __dev_mc_sync(netdev, be_mc_list_update, be_mc_list_update); + + if (netdev->flags & IFF_PROMISC) { + adapter->update_mc_list = false; + } else if (netdev->flags & IFF_ALLMULTI || + netdev_mc_count(netdev) > be_max_mc(adapter)) { + /* Enable multicast promisc if num configured exceeds + * what we support + */ + mc_promisc = true; + adapter->update_mc_list = false; + } else if (adapter->if_flags & BE_IF_FLAGS_MCAST_PROMISCUOUS) { + /* Update mc-list unconditionally if the iface was previously + * in mc-promisc mode and now is out of that mode. + */ + adapter->update_mc_list = true; + } + + if (adapter->update_mc_list) { + int i = 0; + + /* cache the mc-list in adapter */ + netdev_for_each_mc_addr(ha, netdev) { + ether_addr_copy(adapter->mc_list[i].mac, ha->addr); + i++; + } + adapter->mc_count = netdev_mc_count(netdev); + } + netif_addr_unlock_bh(netdev); + + if (mc_promisc) { be_set_mc_promisc(adapter); + } else if (adapter->update_mc_list) { + status = be_cmd_rx_filter(adapter, BE_IF_FLAGS_MULTICAST, ON); + if (!status) + adapter->if_flags &= ~BE_IF_FLAGS_MCAST_PROMISCUOUS; + else + be_set_mc_promisc(adapter); + + adapter->update_mc_list = false; + } +} + +static void be_clear_mc_list(struct be_adapter *adapter) +{ + struct net_device *netdev = adapter->netdev; + + __dev_mc_unsync(netdev, NULL); + be_cmd_rx_filter(adapter, BE_IF_FLAGS_MULTICAST, OFF); + adapter->mc_count = 0; } static void be_set_uc_list(struct be_adapter *adapter) { + struct net_device *netdev = adapter->netdev; struct netdev_hw_addr *ha; - int i = 1; /* First slot is claimed by the Primary MAC */ + bool uc_promisc = false; + int curr_uc_macs = 0, i; - for (; adapter->uc_macs > 0; adapter->uc_macs--, i++) - be_cmd_pmac_del(adapter, adapter->if_handle, - adapter->pmac_id[i], 0); + netif_addr_lock_bh(netdev); + __dev_uc_sync(netdev, be_uc_list_update, be_uc_list_update); - if (netdev_uc_count(adapter->netdev) > be_max_uc(adapter)) { - be_set_all_promisc(adapter); - return; + if (netdev->flags & IFF_PROMISC) { + adapter->update_uc_list = false; + } else if (netdev_uc_count(netdev) > (be_max_uc(adapter) - 1)) { + uc_promisc = true; + adapter->update_uc_list = false; + } else if (adapter->if_flags & BE_IF_FLAGS_PROMISCUOUS) { + /* Update uc-list unconditionally if the iface was previously + * in uc-promisc mode and now is out of that mode. + */ + adapter->update_uc_list = true; } - netdev_for_each_uc_addr(ha, adapter->netdev) { - adapter->uc_macs++; /* First slot is for Primary MAC */ - be_cmd_pmac_add(adapter, (u8 *)ha->addr, adapter->if_handle, - &adapter->pmac_id[adapter->uc_macs], 0); + if (adapter->update_uc_list) { + i = 1; /* First slot is claimed by the Primary MAC */ + + /* cache the uc-list in adapter array */ + netdev_for_each_uc_addr(ha, netdev) { + ether_addr_copy(adapter->uc_list[i].mac, ha->addr); + i++; + } + curr_uc_macs = netdev_uc_count(netdev); + } + netif_addr_unlock_bh(netdev); + + if (uc_promisc) { + be_set_uc_promisc(adapter); + } else if (adapter->update_uc_list) { + be_clear_uc_promisc(adapter); + + for (i = 0; i < adapter->uc_macs; i++) + be_cmd_pmac_del(adapter, adapter->if_handle, + adapter->pmac_id[i + 1], 0); + + for (i = 0; i < curr_uc_macs; i++) + be_cmd_pmac_add(adapter, adapter->uc_list[i].mac, + adapter->if_handle, + &adapter->pmac_id[i + 1], 0); + adapter->uc_macs = curr_uc_macs; + adapter->update_uc_list = false; } } static void be_clear_uc_list(struct be_adapter *adapter) { + struct net_device *netdev = adapter->netdev; int i; - for (i = 1; i < (adapter->uc_macs + 1); i++) + __dev_uc_unsync(netdev, NULL); + for (i = 0; i < adapter->uc_macs; i++) be_cmd_pmac_del(adapter, adapter->if_handle, - adapter->pmac_id[i], 0); + adapter->pmac_id[i + 1], 0); adapter->uc_macs = 0; } -static void be_set_rx_mode(struct net_device *netdev) +static void __be_set_rx_mode(struct be_adapter *adapter) { - struct be_adapter *adapter = netdev_priv(netdev); + struct net_device *netdev = adapter->netdev; + + mutex_lock(&adapter->rx_filter_lock); if (netdev->flags & IFF_PROMISC) { - be_set_all_promisc(adapter); - return; + if (!be_in_all_promisc(adapter)) + be_set_all_promisc(adapter); + } else if (be_in_all_promisc(adapter)) { + /* We need to re-program the vlan-list or clear + * vlan-promisc mode (if needed) when the interface + * comes out of promisc mode. + */ + be_vid_config(adapter); } - /* Interface was previously in promiscuous mode; disable it */ - if (be_in_all_promisc(adapter)) { - be_clear_all_promisc(adapter); - if (adapter->vlans_added) - be_vid_config(adapter); - } + be_set_uc_list(adapter); + be_set_mc_list(adapter); - /* Enable multicast promisc if num configured exceeds what we support */ - if (netdev->flags & IFF_ALLMULTI || - netdev_mc_count(netdev) > be_max_mc(adapter)) { - be_set_mc_promisc(adapter); - return; - } + mutex_unlock(&adapter->rx_filter_lock); +} - if (netdev_uc_count(netdev) != adapter->uc_macs) - be_set_uc_list(adapter); +static void be_work_set_rx_mode(struct work_struct *work) +{ + struct be_cmd_work *cmd_work = + container_of(work, struct be_cmd_work, work); - be_set_mc_list(adapter); + __be_set_rx_mode(cmd_work->adapter); + kfree(cmd_work); } static int be_set_vf_mac(struct net_device *netdev, int vf, u8 *mac) @@ -3429,6 +3567,7 @@ static void be_disable_if_filters(struct be_adapter *adapter) adapter->pmac_id[0], 0); be_clear_uc_list(adapter); + be_clear_mc_list(adapter); /* The IFACE flags are enabled in the open path and cleared * in the close path. When a VF gets detached from the host and @@ -3462,6 +3601,11 @@ static int be_close(struct net_device *netdev) if (!(adapter->flags & BE_FLAGS_SETUP_DONE)) return 0; + /* Before attempting cleanup ensure all the pending cmds in the + * config_wq have finished execution + */ + flush_workqueue(be_wq); + be_disable_if_filters(adapter); if (adapter->flags & BE_FLAGS_NAPI_ENABLED) { @@ -3586,7 +3730,7 @@ static int be_enable_if_filters(struct be_adapter *adapter) if (adapter->vlans_added) be_vid_config(adapter); - be_set_rx_mode(adapter->netdev); + __be_set_rx_mode(adapter); return 0; } @@ -3860,6 +4004,20 @@ static void be_calculate_vf_res(struct be_adapter *adapter, u16 num_vfs, vft_res->max_mcc_count = res.max_mcc_count / (num_vfs + 1); } +static void be_if_destroy(struct be_adapter *adapter) +{ + be_cmd_if_destroy(adapter, adapter->if_handle, 0); + + kfree(adapter->pmac_id); + adapter->pmac_id = NULL; + + kfree(adapter->mc_list); + adapter->mc_list = NULL; + + kfree(adapter->uc_list); + adapter->uc_list = NULL; +} + static int be_clear(struct be_adapter *adapter) { struct pci_dev *pdev = adapter->pdev; @@ -3867,6 +4025,8 @@ static int be_clear(struct be_adapter *adapter) be_cancel_worker(adapter); + flush_workqueue(be_wq); + if (sriov_enabled(adapter)) be_vf_clear(adapter); @@ -3884,10 +4044,8 @@ static int be_clear(struct be_adapter *adapter) } be_disable_vxlan_offloads(adapter); - kfree(adapter->pmac_id); - adapter->pmac_id = NULL; - be_cmd_if_destroy(adapter, adapter->if_handle, 0); + be_if_destroy(adapter); be_clear_queues(adapter); @@ -4341,7 +4499,7 @@ static int be_mac_setup(struct be_adapter *adapter) static void be_schedule_worker(struct be_adapter *adapter) { - schedule_delayed_work(&adapter->work, msecs_to_jiffies(1000)); + queue_delayed_work(be_wq, &adapter->work, msecs_to_jiffies(1000)); adapter->flags |= BE_FLAGS_WORKER_SCHEDULED; } @@ -4393,6 +4551,22 @@ static int be_if_create(struct be_adapter *adapter) u32 cap_flags = be_if_cap_flags(adapter); int status; + /* alloc required memory for other filtering fields */ + adapter->pmac_id = kcalloc(be_max_uc(adapter), + sizeof(*adapter->pmac_id), GFP_KERNEL); + if (!adapter->pmac_id) + return -ENOMEM; + + adapter->mc_list = kcalloc(be_max_mc(adapter), + sizeof(*adapter->mc_list), GFP_KERNEL); + if (!adapter->mc_list) + return -ENOMEM; + + adapter->uc_list = kcalloc(be_max_uc(adapter), + sizeof(*adapter->uc_list), GFP_KERNEL); + if (!adapter->uc_list) + return -ENOMEM; + if (adapter->cfg_num_rx_irqs == 1) cap_flags &= ~(BE_IF_FLAGS_DEFQ_RSS | BE_IF_FLAGS_RSS); @@ -4401,7 +4575,10 @@ static int be_if_create(struct be_adapter *adapter) status = be_cmd_if_create(adapter, be_if_cap_flags(adapter), en_flags, &adapter->if_handle, 0); - return status; + if (status) + return status; + + return 0; } int be_update_queues(struct be_adapter *adapter) @@ -4530,11 +4707,6 @@ static int be_setup(struct be_adapter *adapter) if (status) goto err; - adapter->pmac_id = kcalloc(be_max_uc(adapter), - sizeof(*adapter->pmac_id), GFP_KERNEL); - if (!adapter->pmac_id) - return -ENOMEM; - status = be_msix_enable(adapter); if (status) goto err; @@ -4728,6 +4900,23 @@ static int be_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, 0, 0, nlflags, filter_mask, NULL); } +static struct be_cmd_work *be_alloc_work(struct be_adapter *adapter, + void (*func)(struct work_struct *)) +{ + struct be_cmd_work *work; + + work = kzalloc(sizeof(*work), GFP_ATOMIC); + if (!work) { + dev_err(&adapter->pdev->dev, + "be_work memory allocation failed\n"); + return NULL; + } + + INIT_WORK(&work->work, func); + work->adapter = adapter; + return work; +} + /* VxLAN offload Notes: * * The stack defines tunnel offload flags (hw_enc_features) for IP and doesn't @@ -4742,23 +4931,19 @@ static int be_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, * adds more than one port, disable offloads and don't re-enable them again * until after all the tunnels are removed. */ -static void be_add_vxlan_port(struct net_device *netdev, - struct udp_tunnel_info *ti) +static void be_work_add_vxlan_port(struct work_struct *work) { - struct be_adapter *adapter = netdev_priv(netdev); + struct be_cmd_work *cmd_work = + container_of(work, struct be_cmd_work, work); + struct be_adapter *adapter = cmd_work->adapter; + struct net_device *netdev = adapter->netdev; struct device *dev = &adapter->pdev->dev; - __be16 port = ti->port; + __be16 port = cmd_work->info.vxlan_port; int status; - if (ti->type != UDP_TUNNEL_TYPE_VXLAN) - return; - - if (lancer_chip(adapter) || BEx_chip(adapter) || be_is_mc(adapter)) - return; - if (adapter->vxlan_port == port && adapter->vxlan_port_count) { adapter->vxlan_port_aliases++; - return; + goto done; } if (adapter->flags & BE_FLAGS_VXLAN_OFFLOADS) { @@ -4770,7 +4955,7 @@ static void be_add_vxlan_port(struct net_device *netdev, } if (adapter->vxlan_port_count++ >= 1) - return; + goto done; status = be_cmd_manage_iface(adapter, adapter->if_handle, OP_CONVERT_NORMAL_TO_TUNNEL); @@ -4795,29 +4980,26 @@ static void be_add_vxlan_port(struct net_device *netdev, dev_info(dev, "Enabled VxLAN offloads for UDP port %d\n", be16_to_cpu(port)); - return; + goto done; err: be_disable_vxlan_offloads(adapter); +done: + kfree(cmd_work); } -static void be_del_vxlan_port(struct net_device *netdev, - struct udp_tunnel_info *ti) +static void be_work_del_vxlan_port(struct work_struct *work) { - struct be_adapter *adapter = netdev_priv(netdev); - __be16 port = ti->port; - - if (ti->type != UDP_TUNNEL_TYPE_VXLAN) - return; - - if (lancer_chip(adapter) || BEx_chip(adapter) || be_is_mc(adapter)) - return; + struct be_cmd_work *cmd_work = + container_of(work, struct be_cmd_work, work); + struct be_adapter *adapter = cmd_work->adapter; + __be16 port = cmd_work->info.vxlan_port; if (adapter->vxlan_port != port) goto done; if (adapter->vxlan_port_aliases) { adapter->vxlan_port_aliases--; - return; + goto out; } be_disable_vxlan_offloads(adapter); @@ -4827,6 +5009,40 @@ static void be_del_vxlan_port(struct net_device *netdev, be16_to_cpu(port)); done: adapter->vxlan_port_count--; +out: + kfree(cmd_work); +} + +static void be_cfg_vxlan_port(struct net_device *netdev, + struct udp_tunnel_info *ti, + void (*func)(struct work_struct *)) +{ + struct be_adapter *adapter = netdev_priv(netdev); + struct be_cmd_work *cmd_work; + + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + + if (lancer_chip(adapter) || BEx_chip(adapter) || be_is_mc(adapter)) + return; + + cmd_work = be_alloc_work(adapter, func); + if (cmd_work) { + cmd_work->info.vxlan_port = ti->port; + queue_work(be_wq, &cmd_work->work); + } +} + +static void be_del_vxlan_port(struct net_device *netdev, + struct udp_tunnel_info *ti) +{ + be_cfg_vxlan_port(netdev, ti, be_work_del_vxlan_port); +} + +static void be_add_vxlan_port(struct net_device *netdev, + struct udp_tunnel_info *ti) +{ + be_cfg_vxlan_port(netdev, ti, be_work_add_vxlan_port); } static netdev_features_t be_features_check(struct sk_buff *skb, @@ -4891,6 +5107,16 @@ static int be_get_phys_port_id(struct net_device *dev, return 0; } +static void be_set_rx_mode(struct net_device *dev) +{ + struct be_adapter *adapter = netdev_priv(dev); + struct be_cmd_work *work; + + work = be_alloc_work(adapter, be_work_set_rx_mode); + if (work) + queue_work(be_wq, &work->work); +} + static const struct net_device_ops be_netdev_ops = { .ndo_open = be_open, .ndo_stop = be_close, @@ -5116,7 +5342,7 @@ static void be_worker(struct work_struct *work) reschedule: adapter->work_counter++; - schedule_delayed_work(&adapter->work, msecs_to_jiffies(1000)); + queue_delayed_work(be_wq, &adapter->work, msecs_to_jiffies(1000)); } static void be_unmap_pci_bars(struct be_adapter *adapter) @@ -5256,7 +5482,8 @@ static int be_drv_init(struct be_adapter *adapter) } mutex_init(&adapter->mbox_lock); - spin_lock_init(&adapter->mcc_lock); + mutex_init(&adapter->mcc_lock); + mutex_init(&adapter->rx_filter_lock); spin_lock_init(&adapter->mcc_cq_lock); init_completion(&adapter->et_cmd_compl); @@ -5712,6 +5939,12 @@ static int __init be_init_module(void) pr_info(DRV_NAME " : Use sysfs method to enable VFs\n"); } + be_wq = create_singlethread_workqueue("be_wq"); + if (!be_wq) { + pr_warn(DRV_NAME "workqueue creation failed\n"); + return -1; + } + return pci_register_driver(&be_driver); } module_init(be_init_module); @@ -5719,5 +5952,8 @@ module_init(be_init_module); static void __exit be_exit_module(void) { pci_unregister_driver(&be_driver); + + if (be_wq) + destroy_workqueue(be_wq); } module_exit(be_exit_module); diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 01f7e811739b..fb5c63881340 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -2887,7 +2887,7 @@ fec_enet_close(struct net_device *ndev) * this kind of feature?). */ -#define HASH_BITS 6 /* #bits in hash */ +#define FEC_HASH_BITS 6 /* #bits in hash */ #define CRC32_POLY 0xEDB88320 static void set_multicast_list(struct net_device *ndev) @@ -2935,10 +2935,10 @@ static void set_multicast_list(struct net_device *ndev) } } - /* only upper 6 bits (HASH_BITS) are used + /* only upper 6 bits (FEC_HASH_BITS) are used * which point to specific bit in he hash registers */ - hash = (crc >> (32 - HASH_BITS)) & 0x3f; + hash = (crc >> (32 - FEC_HASH_BITS)) & 0x3f; if (hash > 31) { tmp = readl(fep->hwp + FEC_GRP_HASH_TABLE_HIGH); diff --git a/drivers/net/ethernet/freescale/xgmac_mdio.c b/drivers/net/ethernet/freescale/xgmac_mdio.c index 7b8fe866f603..e03b30c60dcf 100644 --- a/drivers/net/ethernet/freescale/xgmac_mdio.c +++ b/drivers/net/ethernet/freescale/xgmac_mdio.c @@ -271,11 +271,8 @@ static int xgmac_mdio_probe(struct platform_device *pdev) goto err_ioremap; } - if (of_get_property(pdev->dev.of_node, - "little-endian", NULL)) - priv->is_little_endian = true; - else - priv->is_little_endian = false; + priv->is_little_endian = of_property_read_bool(pdev->dev.of_node, + "little-endian"); ret = of_mdiobus_register(bus, np); if (ret) { diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c index 5c8afe1a5ccb..a834774fdb02 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c @@ -684,8 +684,7 @@ hns_mac_register_phydev(struct mii_bus *mdio, struct hns_mac_cb *mac_cb, if (!phy || IS_ERR(phy)) return -EIO; - if (mdio->irq) - phy->irq = mdio->irq[addr]; + phy->irq = mdio->irq[addr]; /* All data is now stored in the phy struct; * register it diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c index d41c28d00b57..8e4252dd9a9d 100644 --- a/drivers/net/ethernet/marvell/mvneta.c +++ b/drivers/net/ethernet/marvell/mvneta.c @@ -399,7 +399,6 @@ struct mvneta_port { u16 rx_ring_size; struct mii_bus *mii_bus; - struct phy_device *phy_dev; phy_interface_t phy_interface; struct device_node *phy_node; unsigned int link; @@ -2651,6 +2650,7 @@ static int mvneta_poll(struct napi_struct *napi, int budget) u32 cause_rx_tx; int rx_queue; struct mvneta_port *pp = netdev_priv(napi->dev); + struct net_device *ndev = pp->dev; struct mvneta_pcpu_port *port = this_cpu_ptr(pp->ports); if (!netif_running(pp->dev)) { @@ -2668,7 +2668,7 @@ static int mvneta_poll(struct napi_struct *napi, int budget) (MVNETA_CAUSE_PHY_STATUS_CHANGE | MVNETA_CAUSE_LINK_CHANGE | MVNETA_CAUSE_PSC_SYNC_CHANGE))) { - mvneta_fixed_link_update(pp, pp->phy_dev); + mvneta_fixed_link_update(pp, ndev->phydev); } } @@ -2963,6 +2963,7 @@ static int mvneta_setup_txqs(struct mvneta_port *pp) static void mvneta_start_dev(struct mvneta_port *pp) { int cpu; + struct net_device *ndev = pp->dev; mvneta_max_rx_size_set(pp, pp->pkt_size); mvneta_txq_max_tx_size_set(pp, pp->pkt_size); @@ -2985,15 +2986,16 @@ static void mvneta_start_dev(struct mvneta_port *pp) MVNETA_CAUSE_LINK_CHANGE | MVNETA_CAUSE_PSC_SYNC_CHANGE); - phy_start(pp->phy_dev); + phy_start(ndev->phydev); netif_tx_start_all_queues(pp->dev); } static void mvneta_stop_dev(struct mvneta_port *pp) { unsigned int cpu; + struct net_device *ndev = pp->dev; - phy_stop(pp->phy_dev); + phy_stop(ndev->phydev); for_each_online_cpu(cpu) { struct mvneta_pcpu_port *port = per_cpu_ptr(pp->ports, cpu); @@ -3166,7 +3168,7 @@ static int mvneta_set_mac_addr(struct net_device *dev, void *addr) static void mvneta_adjust_link(struct net_device *ndev) { struct mvneta_port *pp = netdev_priv(ndev); - struct phy_device *phydev = pp->phy_dev; + struct phy_device *phydev = ndev->phydev; int status_change = 0; if (phydev->link) { @@ -3244,7 +3246,6 @@ static int mvneta_mdio_probe(struct mvneta_port *pp) phy_dev->supported &= PHY_GBIT_FEATURES; phy_dev->advertising = phy_dev->supported; - pp->phy_dev = phy_dev; pp->link = 0; pp->duplex = 0; pp->speed = 0; @@ -3254,8 +3255,9 @@ static int mvneta_mdio_probe(struct mvneta_port *pp) static void mvneta_mdio_remove(struct mvneta_port *pp) { - phy_disconnect(pp->phy_dev); - pp->phy_dev = NULL; + struct net_device *ndev = pp->dev; + + phy_disconnect(ndev->phydev); } /* Electing a CPU must be done in an atomic way: it should be done @@ -3495,42 +3497,30 @@ static int mvneta_stop(struct net_device *dev) static int mvneta_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) { - struct mvneta_port *pp = netdev_priv(dev); - - if (!pp->phy_dev) + if (!dev->phydev) return -ENOTSUPP; - return phy_mii_ioctl(pp->phy_dev, ifr, cmd); + return phy_mii_ioctl(dev->phydev, ifr, cmd); } /* Ethtool methods */ -/* Get settings (phy address, speed) for ethtools */ -int mvneta_ethtool_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) +/* Set link ksettings (phy address, speed) for ethtools */ +int mvneta_ethtool_set_link_ksettings(struct net_device *ndev, + const struct ethtool_link_ksettings *cmd) { - struct mvneta_port *pp = netdev_priv(dev); - - if (!pp->phy_dev) - return -ENODEV; - - return phy_ethtool_gset(pp->phy_dev, cmd); -} - -/* Set settings (phy address, speed) for ethtools */ -int mvneta_ethtool_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct mvneta_port *pp = netdev_priv(dev); - struct phy_device *phydev = pp->phy_dev; + struct mvneta_port *pp = netdev_priv(ndev); + struct phy_device *phydev = ndev->phydev; if (!phydev) return -ENODEV; - if ((cmd->autoneg == AUTONEG_ENABLE) != pp->use_inband_status) { + if ((cmd->base.autoneg == AUTONEG_ENABLE) != pp->use_inband_status) { u32 val; - mvneta_set_autoneg(pp, cmd->autoneg == AUTONEG_ENABLE); + mvneta_set_autoneg(pp, cmd->base.autoneg == AUTONEG_ENABLE); - if (cmd->autoneg == AUTONEG_DISABLE) { + if (cmd->base.autoneg == AUTONEG_DISABLE) { val = mvreg_read(pp, MVNETA_GMAC_AUTONEG_CONFIG); val &= ~(MVNETA_GMAC_CONFIG_MII_SPEED | MVNETA_GMAC_CONFIG_GMII_SPEED | @@ -3547,17 +3537,17 @@ int mvneta_ethtool_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG, val); } - pp->use_inband_status = (cmd->autoneg == AUTONEG_ENABLE); + pp->use_inband_status = (cmd->base.autoneg == AUTONEG_ENABLE); netdev_info(pp->dev, "autoneg status set to %i\n", pp->use_inband_status); - if (netif_running(dev)) { + if (netif_running(ndev)) { mvneta_port_down(pp); mvneta_port_up(pp); } } - return phy_ethtool_sset(pp->phy_dev, cmd); + return phy_ethtool_ksettings_set(ndev->phydev, cmd); } /* Set interrupt coalescing for ethtools */ @@ -3821,8 +3811,6 @@ static const struct net_device_ops mvneta_netdev_ops = { const struct ethtool_ops mvneta_eth_tool_ops = { .get_link = ethtool_op_get_link, - .get_settings = mvneta_ethtool_get_settings, - .set_settings = mvneta_ethtool_set_settings, .set_coalesce = mvneta_ethtool_set_coalesce, .get_coalesce = mvneta_ethtool_get_coalesce, .get_drvinfo = mvneta_ethtool_get_drvinfo, @@ -3835,6 +3823,8 @@ const struct ethtool_ops mvneta_eth_tool_ops = { .get_rxnfc = mvneta_ethtool_get_rxnfc, .get_rxfh = mvneta_ethtool_get_rxfh, .set_rxfh = mvneta_ethtool_set_rxfh, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = mvneta_ethtool_set_link_ksettings, }; /* Initialize hw */ diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index b57ae3afb994..994232e03380 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -353,18 +353,17 @@ static int mtk_set_mac_address(struct net_device *dev, void *p) int ret = eth_mac_addr(dev, p); struct mtk_mac *mac = netdev_priv(dev); const char *macaddr = dev->dev_addr; - unsigned long flags; if (ret) return ret; - spin_lock_irqsave(&mac->hw->page_lock, flags); + spin_lock_bh(&mac->hw->page_lock); mtk_w32(mac->hw, (macaddr[0] << 8) | macaddr[1], MTK_GDMA_MAC_ADRH(mac->id)); mtk_w32(mac->hw, (macaddr[2] << 24) | (macaddr[3] << 16) | (macaddr[4] << 8) | macaddr[5], MTK_GDMA_MAC_ADRL(mac->id)); - spin_unlock_irqrestore(&mac->hw->page_lock, flags); + spin_unlock_bh(&mac->hw->page_lock); return 0; } @@ -748,7 +747,6 @@ static int mtk_start_xmit(struct sk_buff *skb, struct net_device *dev) struct mtk_eth *eth = mac->hw; struct mtk_tx_ring *ring = ð->tx_ring; struct net_device_stats *stats = &dev->stats; - unsigned long flags; bool gso = false; int tx_num; @@ -756,14 +754,14 @@ static int mtk_start_xmit(struct sk_buff *skb, struct net_device *dev) * however we have 2 queues running on the same ring so we need to lock * the ring access */ - spin_lock_irqsave(ð->page_lock, flags); + spin_lock(ð->page_lock); tx_num = mtk_cal_txd_req(skb); if (unlikely(atomic_read(&ring->free_count) <= tx_num)) { mtk_stop_queue(eth); netif_err(eth, tx_queued, dev, "Tx Ring full when queue awake!\n"); - spin_unlock_irqrestore(ð->page_lock, flags); + spin_unlock(ð->page_lock); return NETDEV_TX_BUSY; } @@ -788,12 +786,12 @@ static int mtk_start_xmit(struct sk_buff *skb, struct net_device *dev) if (unlikely(atomic_read(&ring->free_count) <= ring->thresh)) mtk_stop_queue(eth); - spin_unlock_irqrestore(ð->page_lock, flags); + spin_unlock(ð->page_lock); return NETDEV_TX_OK; drop: - spin_unlock_irqrestore(ð->page_lock, flags); + spin_unlock(ð->page_lock); stats->tx_dropped++; dev_kfree_skb(skb); return NETDEV_TX_OK; @@ -1347,16 +1345,15 @@ static int mtk_open(struct net_device *dev) static void mtk_stop_dma(struct mtk_eth *eth, u32 glo_cfg) { - unsigned long flags; u32 val; int i; /* stop the dma engine */ - spin_lock_irqsave(ð->page_lock, flags); + spin_lock_bh(ð->page_lock); val = mtk_r32(eth, glo_cfg); mtk_w32(eth, val & ~(MTK_TX_WB_DDONE | MTK_RX_DMA_EN | MTK_TX_DMA_EN), glo_cfg); - spin_unlock_irqrestore(ð->page_lock, flags); + spin_unlock_bh(ð->page_lock); /* wait for dma stop */ for (i = 0; i < 10; i++) { @@ -1895,7 +1892,6 @@ static int mtk_remove(struct platform_device *pdev) netif_napi_del(ð->tx_napi); netif_napi_del(ð->rx_napi); mtk_cleanup(eth); - platform_set_drvdata(pdev, NULL); return 0; } diff --git a/drivers/net/ethernet/qlogic/qed/qed_cxt.c b/drivers/net/ethernet/qlogic/qed/qed_cxt.c index 1c35f376143e..547692759d06 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_cxt.c +++ b/drivers/net/ethernet/qlogic/qed/qed_cxt.c @@ -377,9 +377,8 @@ static void qed_cxt_set_proto_cid_count(struct qed_hwfn *p_hwfn, } } -u32 qed_cxt_get_proto_cid_count(struct qed_hwfn *p_hwfn, - enum protocol_type type, - u32 *vf_cid) +u32 qed_cxt_get_proto_cid_count(struct qed_hwfn *p_hwfn, + enum protocol_type type, u32 *vf_cid) { if (vf_cid) *vf_cid = p_hwfn->p_cxt_mngr->conn_cfg[type].cids_per_vf; @@ -405,10 +404,10 @@ u32 qed_cxt_get_proto_tid_count(struct qed_hwfn *p_hwfn, return cnt; } -static void -qed_cxt_set_proto_tid_count(struct qed_hwfn *p_hwfn, - enum protocol_type proto, - u8 seg, u8 seg_type, u32 count, bool has_fl) +static void qed_cxt_set_proto_tid_count(struct qed_hwfn *p_hwfn, + enum protocol_type proto, + u8 seg, + u8 seg_type, u32 count, bool has_fl) { struct qed_cxt_mngr *p_mngr = p_hwfn->p_cxt_mngr; struct qed_tid_seg *p_seg = &p_mngr->conn_cfg[proto].tid_seg[seg]; @@ -420,8 +419,7 @@ qed_cxt_set_proto_tid_count(struct qed_hwfn *p_hwfn, static void qed_ilt_cli_blk_fill(struct qed_ilt_client_cfg *p_cli, struct qed_ilt_cli_blk *p_blk, - u32 start_line, u32 total_size, - u32 elem_size) + u32 start_line, u32 total_size, u32 elem_size) { u32 ilt_size = ILT_PAGE_IN_BYTES(p_cli->p_size.val); @@ -448,8 +446,7 @@ static void qed_ilt_cli_adv_line(struct qed_hwfn *p_hwfn, p_cli->first.val = *p_line; p_cli->active = true; - *p_line += DIV_ROUND_UP(p_blk->total_size, - p_blk->real_size_in_page); + *p_line += DIV_ROUND_UP(p_blk->total_size, p_blk->real_size_in_page); p_cli->last.val = *p_line - 1; DP_VERBOSE(p_hwfn, QED_MSG_ILT, @@ -926,12 +923,9 @@ static int qed_ilt_blk_alloc(struct qed_hwfn *p_hwfn, void *p_virt; u32 size; - size = min_t(u32, sz_left, - p_blk->real_size_in_page); + size = min_t(u32, sz_left, p_blk->real_size_in_page); p_virt = dma_alloc_coherent(&p_hwfn->cdev->pdev->dev, - size, - &p_phys, - GFP_KERNEL); + size, &p_phys, GFP_KERNEL); if (!p_virt) return -ENOMEM; memset(p_virt, 0, size); @@ -976,7 +970,7 @@ static int qed_ilt_shadow_alloc(struct qed_hwfn *p_hwfn) for (j = 0; j < ILT_CLI_PF_BLOCKS; j++) { p_blk = &clients[i].pf_blks[j]; rc = qed_ilt_blk_alloc(p_hwfn, p_blk, i, 0); - if (rc != 0) + if (rc) goto ilt_shadow_fail; } for (k = 0; k < p_mngr->vf_count; k++) { @@ -985,7 +979,7 @@ static int qed_ilt_shadow_alloc(struct qed_hwfn *p_hwfn) p_blk = &clients[i].vf_blks[j]; rc = qed_ilt_blk_alloc(p_hwfn, p_blk, i, lines); - if (rc != 0) + if (rc) goto ilt_shadow_fail; } } @@ -1672,7 +1666,7 @@ static void qed_tm_init_pf(struct qed_hwfn *p_hwfn) p_hwfn->rel_pf_id * NUM_TASK_PF_SEGMENTS + i); STORE_RT_REG_AGG(p_hwfn, rt_reg, cfg_word); - active_seg_mask |= (tm_iids.pf_tids[i] ? (1 << i) : 0); + active_seg_mask |= (tm_iids.pf_tids[i] ? BIT(i) : 0); tm_offset += tm_iids.pf_tids[i]; } @@ -1702,8 +1696,7 @@ void qed_cxt_hw_init_pf(struct qed_hwfn *p_hwfn) } int qed_cxt_acquire_cid(struct qed_hwfn *p_hwfn, - enum protocol_type type, - u32 *p_cid) + enum protocol_type type, u32 *p_cid) { struct qed_cxt_mngr *p_mngr = p_hwfn->p_cxt_mngr; u32 rel_cid; @@ -1717,8 +1710,7 @@ int qed_cxt_acquire_cid(struct qed_hwfn *p_hwfn, p_mngr->acquired[type].max_count); if (rel_cid >= p_mngr->acquired[type].max_count) { - DP_NOTICE(p_hwfn, "no CID available for protocol %d\n", - type); + DP_NOTICE(p_hwfn, "no CID available for protocol %d\n", type); return -EINVAL; } @@ -1730,8 +1722,7 @@ int qed_cxt_acquire_cid(struct qed_hwfn *p_hwfn, } static bool qed_cxt_test_cid_acquired(struct qed_hwfn *p_hwfn, - u32 cid, - enum protocol_type *p_type) + u32 cid, enum protocol_type *p_type) { struct qed_cxt_mngr *p_mngr = p_hwfn->p_cxt_mngr; struct qed_cid_acquired_map *p_map; @@ -1763,8 +1754,7 @@ static bool qed_cxt_test_cid_acquired(struct qed_hwfn *p_hwfn, return true; } -void qed_cxt_release_cid(struct qed_hwfn *p_hwfn, - u32 cid) +void qed_cxt_release_cid(struct qed_hwfn *p_hwfn, u32 cid) { struct qed_cxt_mngr *p_mngr = p_hwfn->p_cxt_mngr; enum protocol_type type; @@ -1781,8 +1771,7 @@ void qed_cxt_release_cid(struct qed_hwfn *p_hwfn, __clear_bit(rel_cid, p_mngr->acquired[type].cid_map); } -int qed_cxt_get_cid_info(struct qed_hwfn *p_hwfn, - struct qed_cxt_info *p_info) +int qed_cxt_get_cid_info(struct qed_hwfn *p_hwfn, struct qed_cxt_info *p_info) { struct qed_cxt_mngr *p_mngr = p_hwfn->p_cxt_mngr; u32 conn_cxt_size, hw_p_size, cxts_per_p, line; diff --git a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c index d0dc28f93c0e..d2f202eae6b2 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c @@ -1855,6 +1855,7 @@ static int qed_dcbnl_get_ieee_pfc(struct qed_dev *cdev, if (!dcbx_info->operational.ieee) { DP_INFO(hwfn, "DCBX is not enabled/operational in IEEE mode\n"); + kfree(dcbx_info); return -EINVAL; } diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c index 0e4f4a9306b5..8117ddff501b 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c @@ -35,8 +35,7 @@ #include "qed_sriov.h" #include "qed_vf.h" -static spinlock_t qm_lock; -static bool qm_lock_init = false; +static DEFINE_SPINLOCK(qm_lock); /* API common to all protocols */ enum BAR_ID { @@ -44,8 +43,7 @@ enum BAR_ID { BAR_ID_1 /* Used for doorbells */ }; -static u32 qed_hw_bar_size(struct qed_hwfn *p_hwfn, - enum BAR_ID bar_id) +static u32 qed_hw_bar_size(struct qed_hwfn *p_hwfn, enum BAR_ID bar_id) { u32 bar_reg = (bar_id == BAR_ID_0 ? PGLUE_B_REG_PF_BAR0_SIZE : PGLUE_B_REG_PF_BAR1_SIZE); @@ -70,8 +68,7 @@ static u32 qed_hw_bar_size(struct qed_hwfn *p_hwfn, } } -void qed_init_dp(struct qed_dev *cdev, - u32 dp_module, u8 dp_level) +void qed_init_dp(struct qed_dev *cdev, u32 dp_module, u8 dp_level) { u32 i; @@ -543,8 +540,7 @@ int qed_resc_alloc(struct qed_dev *cdev) cdev->reset_stats = kzalloc(sizeof(*cdev->reset_stats), GFP_KERNEL); if (!cdev->reset_stats) { DP_NOTICE(cdev, "Failed to allocate reset statistics\n"); - rc = -ENOMEM; - goto alloc_err; + goto alloc_no_mem; } return 0; @@ -605,9 +601,8 @@ int qed_final_cleanup(struct qed_hwfn *p_hwfn, /* Make sure notification is not set before initiating final cleanup */ if (REG_RD(p_hwfn, addr)) { - DP_NOTICE( - p_hwfn, - "Unexpected; Found final cleanup notification before initiating final cleanup\n"); + DP_NOTICE(p_hwfn, + "Unexpected; Found final cleanup notification before initiating final cleanup\n"); REG_WR(p_hwfn, addr, 0); } @@ -701,17 +696,14 @@ static void qed_init_cau_rt_data(struct qed_dev *cdev) continue; qed_init_cau_sb_entry(p_hwfn, &sb_entry, - p_block->function_id, - 0, 0); - STORE_RT_REG_AGG(p_hwfn, offset + sb_id * 2, - sb_entry); + p_block->function_id, 0, 0); + STORE_RT_REG_AGG(p_hwfn, offset + sb_id * 2, sb_entry); } } } static int qed_hw_init_common(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - int hw_mode) + struct qed_ptt *p_ptt, int hw_mode) { struct qed_qm_info *qm_info = &p_hwfn->qm_info; struct qed_qm_common_rt_init_params params; @@ -759,7 +751,7 @@ static int qed_hw_init_common(struct qed_hwfn *p_hwfn, qed_port_unpretend(p_hwfn, p_ptt); rc = qed_init_run(p_hwfn, p_ptt, PHASE_ENGINE, ANY_PHASE_ID, hw_mode); - if (rc != 0) + if (rc) return rc; qed_wr(p_hwfn, p_ptt, PSWRQ2_REG_L2P_VALIDATE_VFID, 0); @@ -788,13 +780,12 @@ static int qed_hw_init_common(struct qed_hwfn *p_hwfn, } static int qed_hw_init_port(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - int hw_mode) + struct qed_ptt *p_ptt, int hw_mode) { int rc = 0; rc = qed_init_run(p_hwfn, p_ptt, PHASE_PORT, p_hwfn->port_id, hw_mode); - if (rc != 0) + if (rc) return rc; if (hw_mode & (1 << MODE_MF_SI)) { @@ -848,7 +839,7 @@ static int qed_hw_init_pf(struct qed_hwfn *p_hwfn, qed_int_igu_init_rt(p_hwfn); /* Set VLAN in NIG if needed */ - if (hw_mode & (1 << MODE_MF_SD)) { + if (hw_mode & BIT(MODE_MF_SD)) { DP_VERBOSE(p_hwfn, NETIF_MSG_HW, "Configuring LLH_FUNC_TAG\n"); STORE_RT_REG(p_hwfn, NIG_REG_LLH_FUNC_TAG_EN_RT_OFFSET, 1); STORE_RT_REG(p_hwfn, NIG_REG_LLH_FUNC_TAG_VALUE_RT_OFFSET, @@ -856,7 +847,7 @@ static int qed_hw_init_pf(struct qed_hwfn *p_hwfn, } /* Enable classification by MAC if needed */ - if (hw_mode & (1 << MODE_MF_SI)) { + if (hw_mode & BIT(MODE_MF_SI)) { DP_VERBOSE(p_hwfn, NETIF_MSG_HW, "Configuring TAGMAC_CLS_TYPE\n"); STORE_RT_REG(p_hwfn, @@ -871,7 +862,7 @@ static int qed_hw_init_pf(struct qed_hwfn *p_hwfn, /* Cleanup chip from previous driver if such remains exist */ rc = qed_final_cleanup(p_hwfn, p_ptt, rel_pf_id, false); - if (rc != 0) + if (rc) return rc; /* PF Init sequence */ @@ -950,8 +941,7 @@ static void qed_reset_mb_shadow(struct qed_hwfn *p_hwfn, /* Read shadow of current MFW mailbox */ qed_mcp_read_mb(p_hwfn, p_main_ptt); memcpy(p_hwfn->mcp_info->mfw_mb_shadow, - p_hwfn->mcp_info->mfw_mb_cur, - p_hwfn->mcp_info->mfw_mb_length); + p_hwfn->mcp_info->mfw_mb_cur, p_hwfn->mcp_info->mfw_mb_length); } int qed_hw_init(struct qed_dev *cdev, @@ -971,7 +961,7 @@ int qed_hw_init(struct qed_dev *cdev, if (IS_PF(cdev)) { rc = qed_init_fw_data(cdev, bin_fw_data); - if (rc != 0) + if (rc) return rc; } @@ -988,8 +978,7 @@ int qed_hw_init(struct qed_dev *cdev, qed_calc_hw_mode(p_hwfn); - rc = qed_mcp_load_req(p_hwfn, p_hwfn->p_main_ptt, - &load_code); + rc = qed_mcp_load_req(p_hwfn, p_hwfn->p_main_ptt, &load_code); if (rc) { DP_NOTICE(p_hwfn, "Failed sending LOAD_REQ command\n"); return rc; @@ -1004,11 +993,6 @@ int qed_hw_init(struct qed_dev *cdev, p_hwfn->first_on_engine = (load_code == FW_MSG_CODE_DRV_LOAD_ENGINE); - if (!qm_lock_init) { - spin_lock_init(&qm_lock); - qm_lock_init = true; - } - switch (load_code) { case FW_MSG_CODE_DRV_LOAD_ENGINE: rc = qed_hw_init_common(p_hwfn, p_hwfn->p_main_ptt, @@ -1071,9 +1055,8 @@ int qed_hw_init(struct qed_dev *cdev, } #define QED_HW_STOP_RETRY_LIMIT (10) -static inline void qed_hw_timers_stop(struct qed_dev *cdev, - struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) +static void qed_hw_timers_stop(struct qed_dev *cdev, + struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { int i; @@ -1084,8 +1067,7 @@ static inline void qed_hw_timers_stop(struct qed_dev *cdev, for (i = 0; i < QED_HW_STOP_RETRY_LIMIT; i++) { if ((!qed_rd(p_hwfn, p_ptt, TM_REG_PF_SCAN_ACTIVE_CONN)) && - (!qed_rd(p_hwfn, p_ptt, - TM_REG_PF_SCAN_ACTIVE_TASK))) + (!qed_rd(p_hwfn, p_ptt, TM_REG_PF_SCAN_ACTIVE_TASK))) break; /* Dependent on number of connection/tasks, possibly @@ -1190,8 +1172,7 @@ void qed_hw_stop_fastpath(struct qed_dev *cdev) } DP_VERBOSE(p_hwfn, - NETIF_MSG_IFDOWN, - "Shutting down the fastpath\n"); + NETIF_MSG_IFDOWN, "Shutting down the fastpath\n"); qed_wr(p_hwfn, p_ptt, NIG_REG_RX_LLH_BRB_GATE_DNTFWD_PERPF, 0x1); @@ -1219,14 +1200,13 @@ void qed_hw_start_fastpath(struct qed_hwfn *p_hwfn) NIG_REG_RX_LLH_BRB_GATE_DNTFWD_PERPF, 0x0); } -static int qed_reg_assert(struct qed_hwfn *hwfn, - struct qed_ptt *ptt, u32 reg, - bool expected) +static int qed_reg_assert(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, u32 reg, bool expected) { - u32 assert_val = qed_rd(hwfn, ptt, reg); + u32 assert_val = qed_rd(p_hwfn, p_ptt, reg); if (assert_val != expected) { - DP_NOTICE(hwfn, "Value at address 0x%x != 0x%08x\n", + DP_NOTICE(p_hwfn, "Value at address 0x%08x != 0x%08x\n", reg, expected); return -EINVAL; } @@ -1306,8 +1286,7 @@ static void qed_hw_hwfn_prepare(struct qed_hwfn *p_hwfn) /* Clean Previous errors if such exist */ qed_wr(p_hwfn, p_hwfn->p_main_ptt, - PGLUE_B_REG_WAS_ERROR_PF_31_0_CLR, - 1 << p_hwfn->abs_pf_id); + PGLUE_B_REG_WAS_ERROR_PF_31_0_CLR, 1 << p_hwfn->abs_pf_id); /* enable internal target-read */ qed_wr(p_hwfn, p_hwfn->p_main_ptt, @@ -1317,7 +1296,8 @@ static void qed_hw_hwfn_prepare(struct qed_hwfn *p_hwfn) static void get_function_id(struct qed_hwfn *p_hwfn) { /* ME Register */ - p_hwfn->hw_info.opaque_fid = (u16)REG_RD(p_hwfn, PXP_PF_ME_OPAQUE_ADDR); + p_hwfn->hw_info.opaque_fid = (u16) REG_RD(p_hwfn, + PXP_PF_ME_OPAQUE_ADDR); p_hwfn->hw_info.concrete_fid = REG_RD(p_hwfn, PXP_PF_ME_CONCRETE_ADDR); @@ -1326,6 +1306,10 @@ static void get_function_id(struct qed_hwfn *p_hwfn) PXP_CONCRETE_FID_PFID); p_hwfn->port_id = GET_FIELD(p_hwfn->hw_info.concrete_fid, PXP_CONCRETE_FID_PORT); + + DP_VERBOSE(p_hwfn, NETIF_MSG_PROBE, + "Read ME register: Concrete 0x%08x Opaque 0x%04x\n", + p_hwfn->hw_info.concrete_fid, p_hwfn->hw_info.opaque_fid); } static void qed_hw_set_feat(struct qed_hwfn *p_hwfn) @@ -1417,8 +1401,7 @@ static int qed_hw_get_resc(struct qed_hwfn *p_hwfn) return 0; } -static int qed_hw_get_nvm_info(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) +static int qed_hw_get_nvm_info(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { u32 nvm_cfg1_offset, mf_mode, addr, generic_cont0, core_cfg; u32 port_cfg_addr, link_temp, nvm_cfg_addr, device_capabilities; @@ -1472,8 +1455,7 @@ static int qed_hw_get_nvm_info(struct qed_hwfn *p_hwfn, p_hwfn->hw_info.port_mode = QED_PORT_MODE_DE_1X25G; break; default: - DP_NOTICE(p_hwfn, "Unknown port mode in 0x%08x\n", - core_cfg); + DP_NOTICE(p_hwfn, "Unknown port mode in 0x%08x\n", core_cfg); break; } @@ -1484,11 +1466,11 @@ static int qed_hw_get_nvm_info(struct qed_hwfn *p_hwfn, link_temp = qed_rd(p_hwfn, p_ptt, port_cfg_addr + offsetof(struct nvm_cfg1_port, speed_cap_mask)); - link->speed.advertised_speeds = - link_temp & NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_MASK; + link_temp &= NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_MASK; + link->speed.advertised_speeds = link_temp; - p_hwfn->mcp_info->link_capabilities.speed_capabilities = - link->speed.advertised_speeds; + link_temp = link->speed.advertised_speeds; + p_hwfn->mcp_info->link_capabilities.speed_capabilities = link_temp; link_temp = qed_rd(p_hwfn, p_ptt, port_cfg_addr + @@ -1517,8 +1499,7 @@ static int qed_hw_get_nvm_info(struct qed_hwfn *p_hwfn, link->speed.forced_speed = 100000; break; default: - DP_NOTICE(p_hwfn, "Unknown Speed in 0x%08x\n", - link_temp); + DP_NOTICE(p_hwfn, "Unknown Speed in 0x%08x\n", link_temp); } link_temp &= NVM_CFG1_PORT_DRV_FLOW_CONTROL_MASK; @@ -1628,10 +1609,10 @@ static void qed_get_num_funcs(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) DP_VERBOSE(p_hwfn, NETIF_MSG_PROBE, - "PF [rel_id %d, abs_id %d] within the %d enabled functions on the engine\n", + "PF [rel_id %d, abs_id %d] occupies index %d within the %d enabled functions on the engine\n", p_hwfn->rel_pf_id, p_hwfn->abs_pf_id, - p_hwfn->num_funcs_on_engine); + p_hwfn->enabled_func_idx, p_hwfn->num_funcs_on_engine); } static int @@ -1703,10 +1684,9 @@ static int qed_get_dev_info(struct qed_dev *cdev) u32 tmp; /* Read Vendor Id / Device Id */ - pci_read_config_word(cdev->pdev, PCI_VENDOR_ID, - &cdev->vendor_id); - pci_read_config_word(cdev->pdev, PCI_DEVICE_ID, - &cdev->device_id); + pci_read_config_word(cdev->pdev, PCI_VENDOR_ID, &cdev->vendor_id); + pci_read_config_word(cdev->pdev, PCI_DEVICE_ID, &cdev->device_id); + cdev->chip_num = (u16)qed_rd(p_hwfn, p_hwfn->p_main_ptt, MISCS_REG_CHIP_NUM); cdev->chip_rev = (u16)qed_rd(p_hwfn, p_hwfn->p_main_ptt, @@ -1782,7 +1762,7 @@ static int qed_hw_prepare_single(struct qed_hwfn *p_hwfn, /* First hwfn learns basic information, e.g., number of hwfns */ if (!p_hwfn->my_id) { rc = qed_get_dev_info(p_hwfn->cdev); - if (rc != 0) + if (rc) goto err1; } @@ -2183,8 +2163,7 @@ int qed_fw_l2_queue(struct qed_hwfn *p_hwfn, u16 src_id, u16 *dst_id) return 0; } -int qed_fw_vport(struct qed_hwfn *p_hwfn, - u8 src_id, u8 *dst_id) +int qed_fw_vport(struct qed_hwfn *p_hwfn, u8 src_id, u8 *dst_id) { if (src_id >= RESC_NUM(p_hwfn, QED_VPORT)) { u8 min, max; @@ -2203,8 +2182,7 @@ int qed_fw_vport(struct qed_hwfn *p_hwfn, return 0; } -int qed_fw_rss_eng(struct qed_hwfn *p_hwfn, - u8 src_id, u8 *dst_id) +int qed_fw_rss_eng(struct qed_hwfn *p_hwfn, u8 src_id, u8 *dst_id) { if (src_id >= RESC_NUM(p_hwfn, QED_RSS_ENG)) { u8 min, max; @@ -2386,8 +2364,7 @@ static void qed_disable_wfq_for_all_vports(struct qed_hwfn *p_hwfn, * 3. total_req_min_rate [all vports min rate sum] shouldn't exceed min_pf_rate. */ static int qed_init_wfq_param(struct qed_hwfn *p_hwfn, - u16 vport_id, u32 req_rate, - u32 min_pf_rate) + u16 vport_id, u32 req_rate, u32 min_pf_rate) { u32 total_req_min_rate = 0, total_left_rate = 0, left_rate_per_vp = 0; int non_requested_count = 0, req_count = 0, i, num_vports; @@ -2471,7 +2448,7 @@ static int __qed_configure_vport_wfq(struct qed_hwfn *p_hwfn, rc = qed_init_wfq_param(p_hwfn, vp_id, rate, p_link->min_pf_rate); - if (rc == 0) + if (!rc) qed_configure_wfq_for_all_vports(p_hwfn, p_ptt, p_link->min_pf_rate); else diff --git a/drivers/net/ethernet/qlogic/qed/qed_hw.c b/drivers/net/ethernet/qlogic/qed/qed_hw.c index e17885321faf..8ebdc79b3850 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hw.c +++ b/drivers/net/ethernet/qlogic/qed/qed_hw.c @@ -44,8 +44,7 @@ struct qed_ptt_pool { int qed_ptt_pool_alloc(struct qed_hwfn *p_hwfn) { - struct qed_ptt_pool *p_pool = kmalloc(sizeof(*p_pool), - GFP_KERNEL); + struct qed_ptt_pool *p_pool = kmalloc(sizeof(*p_pool), GFP_KERNEL); int i; if (!p_pool) @@ -113,16 +112,14 @@ struct qed_ptt *qed_ptt_acquire(struct qed_hwfn *p_hwfn) return NULL; } -void qed_ptt_release(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) +void qed_ptt_release(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { spin_lock_bh(&p_hwfn->p_ptt_pool->lock); list_add(&p_ptt->list_entry, &p_hwfn->p_ptt_pool->free_list); spin_unlock_bh(&p_hwfn->p_ptt_pool->lock); } -u32 qed_ptt_get_hw_addr(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) +u32 qed_ptt_get_hw_addr(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { /* The HW is using DWORDS and we need to translate it to Bytes */ return le32_to_cpu(p_ptt->pxp.offset) << 2; @@ -141,8 +138,7 @@ u32 qed_ptt_get_bar_addr(struct qed_ptt *p_ptt) } void qed_ptt_set_win(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - u32 new_hw_addr) + struct qed_ptt *p_ptt, u32 new_hw_addr) { u32 prev_hw_addr; @@ -166,8 +162,7 @@ void qed_ptt_set_win(struct qed_hwfn *p_hwfn, } static u32 qed_set_ptt(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - u32 hw_addr) + struct qed_ptt *p_ptt, u32 hw_addr) { u32 win_hw_addr = qed_ptt_get_hw_addr(p_hwfn, p_ptt); u32 offset; @@ -224,10 +219,7 @@ u32 qed_rd(struct qed_hwfn *p_hwfn, static void qed_memcpy_hw(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, - void *addr, - u32 hw_addr, - size_t n, - bool to_device) + void *addr, u32 hw_addr, size_t n, bool to_device) { u32 dw_count, *host_addr, hw_offset; size_t quota, done = 0; @@ -259,8 +251,7 @@ static void qed_memcpy_hw(struct qed_hwfn *p_hwfn, } void qed_memcpy_from(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - void *dest, u32 hw_addr, size_t n) + struct qed_ptt *p_ptt, void *dest, u32 hw_addr, size_t n) { DP_VERBOSE(p_hwfn, NETIF_MSG_HW, "hw_addr 0x%x, dest %p hw_addr 0x%x, size %lu\n", @@ -270,8 +261,7 @@ void qed_memcpy_from(struct qed_hwfn *p_hwfn, } void qed_memcpy_to(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - u32 hw_addr, void *src, size_t n) + struct qed_ptt *p_ptt, u32 hw_addr, void *src, size_t n) { DP_VERBOSE(p_hwfn, NETIF_MSG_HW, "hw_addr 0x%x, hw_addr 0x%x, src %p size %lu\n", @@ -280,9 +270,7 @@ void qed_memcpy_to(struct qed_hwfn *p_hwfn, qed_memcpy_hw(p_hwfn, p_ptt, src, hw_addr, n, true); } -void qed_fid_pretend(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - u16 fid) +void qed_fid_pretend(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u16 fid) { u16 control = 0; @@ -309,8 +297,7 @@ void qed_fid_pretend(struct qed_hwfn *p_hwfn, } void qed_port_pretend(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - u8 port_id) + struct qed_ptt *p_ptt, u8 port_id) { u16 control = 0; @@ -326,8 +313,7 @@ void qed_port_pretend(struct qed_hwfn *p_hwfn, *(u32 *)&p_ptt->pxp.pretend); } -void qed_port_unpretend(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) +void qed_port_unpretend(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { u16 control = 0; @@ -429,28 +415,27 @@ u32 qed_dmae_idx_to_go_cmd(u8 idx) return DMAE_REG_GO_C0 + (idx << 2); } -static int -qed_dmae_post_command(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) +static int qed_dmae_post_command(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt) { - struct dmae_cmd *command = p_hwfn->dmae_info.p_dmae_cmd; + struct dmae_cmd *p_command = p_hwfn->dmae_info.p_dmae_cmd; u8 idx_cmd = p_hwfn->dmae_info.channel, i; int qed_status = 0; /* verify address is not NULL */ - if ((((command->dst_addr_lo == 0) && (command->dst_addr_hi == 0)) || - ((command->src_addr_lo == 0) && (command->src_addr_hi == 0)))) { + if ((((!p_command->dst_addr_lo) && (!p_command->dst_addr_hi)) || + ((!p_command->src_addr_lo) && (!p_command->src_addr_hi)))) { DP_NOTICE(p_hwfn, "source or destination address 0 idx_cmd=%d\n" "opcode = [0x%08x,0x%04x] len=0x%x src=0x%x:%x dst=0x%x:%x\n", - idx_cmd, - le32_to_cpu(command->opcode), - le16_to_cpu(command->opcode_b), - le16_to_cpu(command->length_dw), - le32_to_cpu(command->src_addr_hi), - le32_to_cpu(command->src_addr_lo), - le32_to_cpu(command->dst_addr_hi), - le32_to_cpu(command->dst_addr_lo)); + idx_cmd, + le32_to_cpu(p_command->opcode), + le16_to_cpu(p_command->opcode_b), + le16_to_cpu(p_command->length_dw), + le32_to_cpu(p_command->src_addr_hi), + le32_to_cpu(p_command->src_addr_lo), + le32_to_cpu(p_command->dst_addr_hi), + le32_to_cpu(p_command->dst_addr_lo)); return -EINVAL; } @@ -459,13 +444,13 @@ qed_dmae_post_command(struct qed_hwfn *p_hwfn, NETIF_MSG_HW, "Posting DMAE command [idx %d]: opcode = [0x%08x,0x%04x] len=0x%x src=0x%x:%x dst=0x%x:%x\n", idx_cmd, - le32_to_cpu(command->opcode), - le16_to_cpu(command->opcode_b), - le16_to_cpu(command->length_dw), - le32_to_cpu(command->src_addr_hi), - le32_to_cpu(command->src_addr_lo), - le32_to_cpu(command->dst_addr_hi), - le32_to_cpu(command->dst_addr_lo)); + le32_to_cpu(p_command->opcode), + le16_to_cpu(p_command->opcode_b), + le16_to_cpu(p_command->length_dw), + le32_to_cpu(p_command->src_addr_hi), + le32_to_cpu(p_command->src_addr_lo), + le32_to_cpu(p_command->dst_addr_hi), + le32_to_cpu(p_command->dst_addr_lo)); /* Copy the command to DMAE - need to do it before every call * for source/dest address no reset. @@ -475,7 +460,7 @@ qed_dmae_post_command(struct qed_hwfn *p_hwfn, */ for (i = 0; i < DMAE_CMD_SIZE; i++) { u32 data = (i < DMAE_CMD_SIZE_TO_FILL) ? - *(((u32 *)command) + i) : 0; + *(((u32 *)p_command) + i) : 0; qed_wr(p_hwfn, p_ptt, DMAE_REG_CMD_MEM + @@ -483,9 +468,7 @@ qed_dmae_post_command(struct qed_hwfn *p_hwfn, (i * sizeof(u32)), data); } - qed_wr(p_hwfn, p_ptt, - qed_dmae_idx_to_go_cmd(idx_cmd), - DMAE_GO_VALUE); + qed_wr(p_hwfn, p_ptt, qed_dmae_idx_to_go_cmd(idx_cmd), DMAE_GO_VALUE); return qed_status; } @@ -498,9 +481,7 @@ int qed_dmae_info_alloc(struct qed_hwfn *p_hwfn) u32 **p_comp = &p_hwfn->dmae_info.p_completion_word; *p_comp = dma_alloc_coherent(&p_hwfn->cdev->pdev->dev, - sizeof(u32), - p_addr, - GFP_KERNEL); + sizeof(u32), p_addr, GFP_KERNEL); if (!*p_comp) { DP_NOTICE(p_hwfn, "Failed to allocate `p_completion_word'\n"); goto err; @@ -543,8 +524,7 @@ void qed_dmae_info_free(struct qed_hwfn *p_hwfn) p_phys = p_hwfn->dmae_info.completion_word_phys_addr; dma_free_coherent(&p_hwfn->cdev->pdev->dev, sizeof(u32), - p_hwfn->dmae_info.p_completion_word, - p_phys); + p_hwfn->dmae_info.p_completion_word, p_phys); p_hwfn->dmae_info.p_completion_word = NULL; } @@ -552,8 +532,7 @@ void qed_dmae_info_free(struct qed_hwfn *p_hwfn) p_phys = p_hwfn->dmae_info.dmae_cmd_phys_addr; dma_free_coherent(&p_hwfn->cdev->pdev->dev, sizeof(struct dmae_cmd), - p_hwfn->dmae_info.p_dmae_cmd, - p_phys); + p_hwfn->dmae_info.p_dmae_cmd, p_phys); p_hwfn->dmae_info.p_dmae_cmd = NULL; } @@ -571,9 +550,7 @@ void qed_dmae_info_free(struct qed_hwfn *p_hwfn) static int qed_dmae_operation_wait(struct qed_hwfn *p_hwfn) { - u32 wait_cnt = 0; - u32 wait_cnt_limit = 10000; - + u32 wait_cnt_limit = 10000, wait_cnt = 0; int qed_status = 0; barrier(); @@ -606,7 +583,7 @@ static int qed_dmae_execute_sub_operation(struct qed_hwfn *p_hwfn, u64 dst_addr, u8 src_type, u8 dst_type, - u32 length) + u32 length_dw) { dma_addr_t phys = p_hwfn->dmae_info.intermediate_buffer_phys_addr; struct dmae_cmd *cmd = p_hwfn->dmae_info.p_dmae_cmd; @@ -624,7 +601,7 @@ static int qed_dmae_execute_sub_operation(struct qed_hwfn *p_hwfn, cmd->src_addr_lo = cpu_to_le32(lower_32_bits(phys)); memcpy(&p_hwfn->dmae_info.p_intermediate_buffer[0], (void *)(uintptr_t)src_addr, - length * sizeof(u32)); + length_dw * sizeof(u32)); break; default: return -EINVAL; @@ -645,7 +622,7 @@ static int qed_dmae_execute_sub_operation(struct qed_hwfn *p_hwfn, return -EINVAL; } - cmd->length_dw = cpu_to_le16((u16)length); + cmd->length_dw = cpu_to_le16((u16)length_dw); qed_dmae_post_command(p_hwfn, p_ptt); @@ -654,16 +631,14 @@ static int qed_dmae_execute_sub_operation(struct qed_hwfn *p_hwfn, if (qed_status) { DP_NOTICE(p_hwfn, "qed_dmae_host2grc: Wait Failed. source_addr 0x%llx, grc_addr 0x%llx, size_in_dwords 0x%x\n", - src_addr, - dst_addr, - length); + src_addr, dst_addr, length_dw); return qed_status; } if (dst_type == QED_DMAE_ADDRESS_HOST_VIRT) memcpy((void *)(uintptr_t)(dst_addr), &p_hwfn->dmae_info.p_intermediate_buffer[0], - length * sizeof(u32)); + length_dw * sizeof(u32)); return 0; } @@ -730,10 +705,7 @@ static int qed_dmae_execute_command(struct qed_hwfn *p_hwfn, if (qed_status) { DP_NOTICE(p_hwfn, "qed_dmae_execute_sub_operation Failed with error 0x%x. source_addr 0x%llx, destination addr 0x%llx, size_in_dwords 0x%x\n", - qed_status, - src_addr, - dst_addr, - length_cur); + qed_status, src_addr, dst_addr, length_cur); break; } } @@ -743,10 +715,7 @@ static int qed_dmae_execute_command(struct qed_hwfn *p_hwfn, int qed_dmae_host2grc(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, - u64 source_addr, - u32 grc_addr, - u32 size_in_dwords, - u32 flags) + u64 source_addr, u32 grc_addr, u32 size_in_dwords, u32 flags) { u32 grc_addr_in_dw = grc_addr / sizeof(u32); struct qed_dmae_params params; @@ -768,9 +737,10 @@ int qed_dmae_host2grc(struct qed_hwfn *p_hwfn, return rc; } -int -qed_dmae_grc2host(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u32 grc_addr, - dma_addr_t dest_addr, u32 size_in_dwords, u32 flags) +int qed_dmae_grc2host(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + u32 grc_addr, + dma_addr_t dest_addr, u32 size_in_dwords, u32 flags) { u32 grc_addr_in_dw = grc_addr / sizeof(u32); struct qed_dmae_params params; @@ -791,12 +761,11 @@ qed_dmae_grc2host(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u32 grc_addr, return rc; } -int -qed_dmae_host2host(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - dma_addr_t source_addr, - dma_addr_t dest_addr, - u32 size_in_dwords, struct qed_dmae_params *p_params) +int qed_dmae_host2host(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + dma_addr_t source_addr, + dma_addr_t dest_addr, + u32 size_in_dwords, struct qed_dmae_params *p_params) { int rc; diff --git a/drivers/net/ethernet/qlogic/qed/qed_init_ops.c b/drivers/net/ethernet/qlogic/qed/qed_init_ops.c index 9866a20d2128..b7a4b27ebdbd 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_init_ops.c +++ b/drivers/net/ethernet/qlogic/qed/qed_init_ops.c @@ -59,17 +59,14 @@ void qed_init_clear_rt_data(struct qed_hwfn *p_hwfn) p_hwfn->rt_data.b_valid[i] = false; } -void qed_init_store_rt_reg(struct qed_hwfn *p_hwfn, - u32 rt_offset, - u32 val) +void qed_init_store_rt_reg(struct qed_hwfn *p_hwfn, u32 rt_offset, u32 val) { p_hwfn->rt_data.init_val[rt_offset] = val; p_hwfn->rt_data.b_valid[rt_offset] = true; } void qed_init_store_rt_agg(struct qed_hwfn *p_hwfn, - u32 rt_offset, u32 *p_val, - size_t size) + u32 rt_offset, u32 *p_val, size_t size) { size_t i; @@ -81,10 +78,7 @@ void qed_init_store_rt_agg(struct qed_hwfn *p_hwfn, static int qed_init_rt(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, - u32 addr, - u16 rt_offset, - u16 size, - bool b_must_dmae) + u32 addr, u16 rt_offset, u16 size, bool b_must_dmae) { u32 *p_init_val = &p_hwfn->rt_data.init_val[rt_offset]; bool *p_valid = &p_hwfn->rt_data.b_valid[rt_offset]; @@ -102,8 +96,7 @@ static int qed_init_rt(struct qed_hwfn *p_hwfn, * simply write the data instead of using dmae. */ if (!b_must_dmae) { - qed_wr(p_hwfn, p_ptt, addr + (i << 2), - p_init_val[i]); + qed_wr(p_hwfn, p_ptt, addr + (i << 2), p_init_val[i]); continue; } @@ -115,7 +108,7 @@ static int qed_init_rt(struct qed_hwfn *p_hwfn, rc = qed_dmae_host2grc(p_hwfn, p_ptt, (uintptr_t)(p_init_val + i), addr + (i << 2), segment, 0); - if (rc != 0) + if (rc) return rc; /* Jump over the entire segment, including invalid entry */ @@ -182,9 +175,7 @@ static int qed_init_array_dmae(struct qed_hwfn *p_hwfn, static int qed_init_fill_dmae(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, - u32 addr, - u32 fill, - u32 fill_count) + u32 addr, u32 fill, u32 fill_count) { static u32 zero_buffer[DMAE_MAX_RW_SIZE]; @@ -199,15 +190,12 @@ static int qed_init_fill_dmae(struct qed_hwfn *p_hwfn, return qed_dmae_host2grc(p_hwfn, p_ptt, (uintptr_t)(&zero_buffer[0]), - addr, fill_count, - QED_DMAE_FLAG_RW_REPL_SRC); + addr, fill_count, QED_DMAE_FLAG_RW_REPL_SRC); } static void qed_init_fill(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, - u32 addr, - u32 fill, - u32 fill_count) + u32 addr, u32 fill, u32 fill_count) { u32 i; @@ -218,12 +206,12 @@ static void qed_init_fill(struct qed_hwfn *p_hwfn, static int qed_init_cmd_array(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, struct init_write_op *cmd, - bool b_must_dmae, - bool b_can_dmae) + bool b_must_dmae, bool b_can_dmae) { + u32 dmae_array_offset = le32_to_cpu(cmd->args.array_offset); u32 data = le32_to_cpu(cmd->data); u32 addr = GET_FIELD(data, INIT_WRITE_OP_ADDRESS) << 2; - u32 dmae_array_offset = le32_to_cpu(cmd->args.array_offset); + u32 offset, output_len, input_len, max_size; struct qed_dev *cdev = p_hwfn->cdev; union init_array_hdr *hdr; @@ -233,8 +221,7 @@ static int qed_init_cmd_array(struct qed_hwfn *p_hwfn, array_data = cdev->fw_data->arr_data; - hdr = (union init_array_hdr *)(array_data + - dmae_array_offset); + hdr = (union init_array_hdr *)(array_data + dmae_array_offset); data = le32_to_cpu(hdr->raw.data); switch (GET_FIELD(data, INIT_ARRAY_RAW_HDR_TYPE)) { case INIT_ARR_ZIPPED: @@ -290,13 +277,12 @@ static int qed_init_cmd_array(struct qed_hwfn *p_hwfn, /* init_ops write command */ static int qed_init_cmd_wr(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, - struct init_write_op *cmd, - bool b_can_dmae) + struct init_write_op *p_cmd, bool b_can_dmae) { - u32 data = le32_to_cpu(cmd->data); - u32 addr = GET_FIELD(data, INIT_WRITE_OP_ADDRESS) << 2; + u32 data = le32_to_cpu(p_cmd->data); bool b_must_dmae = GET_FIELD(data, INIT_WRITE_OP_WIDE_BUS); - union init_write_args *arg = &cmd->args; + u32 addr = GET_FIELD(data, INIT_WRITE_OP_ADDRESS) << 2; + union init_write_args *arg = &p_cmd->args; int rc = 0; /* Sanitize */ @@ -309,20 +295,18 @@ static int qed_init_cmd_wr(struct qed_hwfn *p_hwfn, switch (GET_FIELD(data, INIT_WRITE_OP_SOURCE)) { case INIT_SRC_INLINE: - qed_wr(p_hwfn, p_ptt, addr, - le32_to_cpu(arg->inline_val)); + data = le32_to_cpu(p_cmd->args.inline_val); + qed_wr(p_hwfn, p_ptt, addr, data); break; case INIT_SRC_ZEROS: - if (b_must_dmae || - (b_can_dmae && (le32_to_cpu(arg->zeros_count) >= 64))) - rc = qed_init_fill_dmae(p_hwfn, p_ptt, addr, 0, - le32_to_cpu(arg->zeros_count)); + data = le32_to_cpu(p_cmd->args.zeros_count); + if (b_must_dmae || (b_can_dmae && (data >= 64))) + rc = qed_init_fill_dmae(p_hwfn, p_ptt, addr, 0, data); else - qed_init_fill(p_hwfn, p_ptt, addr, 0, - le32_to_cpu(arg->zeros_count)); + qed_init_fill(p_hwfn, p_ptt, addr, 0, data); break; case INIT_SRC_ARRAY: - rc = qed_init_cmd_array(p_hwfn, p_ptt, cmd, + rc = qed_init_cmd_array(p_hwfn, p_ptt, p_cmd, b_must_dmae, b_can_dmae); break; case INIT_SRC_RUNTIME: @@ -353,8 +337,7 @@ static inline bool comp_or(u32 val, u32 expected_val) /* init_ops read/poll commands */ static void qed_init_cmd_rd(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - struct init_read_op *cmd) + struct qed_ptt *p_ptt, struct init_read_op *cmd) { bool (*comp_check)(u32 val, u32 expected_val); u32 delay = QED_INIT_POLL_PERIOD_US, val; @@ -412,35 +395,33 @@ static void qed_init_cmd_cb(struct qed_hwfn *p_hwfn, } static u8 qed_init_cmd_mode_match(struct qed_hwfn *p_hwfn, - u16 *offset, - int modes) + u16 *p_offset, int modes) { struct qed_dev *cdev = p_hwfn->cdev; const u8 *modes_tree_buf; u8 arg1, arg2, tree_val; modes_tree_buf = cdev->fw_data->modes_tree_buf; - tree_val = modes_tree_buf[(*offset)++]; + tree_val = modes_tree_buf[(*p_offset)++]; switch (tree_val) { case INIT_MODE_OP_NOT: - return qed_init_cmd_mode_match(p_hwfn, offset, modes) ^ 1; + return qed_init_cmd_mode_match(p_hwfn, p_offset, modes) ^ 1; case INIT_MODE_OP_OR: - arg1 = qed_init_cmd_mode_match(p_hwfn, offset, modes); - arg2 = qed_init_cmd_mode_match(p_hwfn, offset, modes); + arg1 = qed_init_cmd_mode_match(p_hwfn, p_offset, modes); + arg2 = qed_init_cmd_mode_match(p_hwfn, p_offset, modes); return arg1 | arg2; case INIT_MODE_OP_AND: - arg1 = qed_init_cmd_mode_match(p_hwfn, offset, modes); - arg2 = qed_init_cmd_mode_match(p_hwfn, offset, modes); + arg1 = qed_init_cmd_mode_match(p_hwfn, p_offset, modes); + arg2 = qed_init_cmd_mode_match(p_hwfn, p_offset, modes); return arg1 & arg2; default: tree_val -= MAX_INIT_MODE_OPS; - return (modes & (1 << tree_val)) ? 1 : 0; + return (modes & BIT(tree_val)) ? 1 : 0; } } static u32 qed_init_cmd_mode(struct qed_hwfn *p_hwfn, - struct init_if_mode_op *p_cmd, - int modes) + struct init_if_mode_op *p_cmd, int modes) { u16 offset = le16_to_cpu(p_cmd->modes_buf_offset); @@ -453,8 +434,7 @@ static u32 qed_init_cmd_mode(struct qed_hwfn *p_hwfn, static u32 qed_init_cmd_phase(struct qed_hwfn *p_hwfn, struct init_if_phase_op *p_cmd, - u32 phase, - u32 phase_id) + u32 phase, u32 phase_id) { u32 data = le32_to_cpu(p_cmd->phase_data); u32 op_data = le32_to_cpu(p_cmd->op_data); @@ -468,10 +448,7 @@ static u32 qed_init_cmd_phase(struct qed_hwfn *p_hwfn, } int qed_init_run(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - int phase, - int phase_id, - int modes) + struct qed_ptt *p_ptt, int phase, int phase_id, int modes) { struct qed_dev *cdev = p_hwfn->cdev; u32 cmd_num, num_init_ops; diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.c b/drivers/net/ethernet/qlogic/qed/qed_int.c index 8fa50fa23c8d..61ec973a06c7 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_int.c +++ b/drivers/net/ethernet/qlogic/qed/qed_int.c @@ -1775,10 +1775,9 @@ struct qed_sb_attn_info { }; static inline u16 qed_attn_update_idx(struct qed_hwfn *p_hwfn, - struct qed_sb_attn_info *p_sb_desc) + struct qed_sb_attn_info *p_sb_desc) { - u16 rc = 0; - u16 index; + u16 rc = 0, index; /* Make certain HW write took affect */ mmiowb(); @@ -1802,15 +1801,13 @@ static inline u16 qed_attn_update_idx(struct qed_hwfn *p_hwfn, * @param asserted_bits newly asserted bits * @return int */ -static int qed_int_assertion(struct qed_hwfn *p_hwfn, - u16 asserted_bits) +static int qed_int_assertion(struct qed_hwfn *p_hwfn, u16 asserted_bits) { struct qed_sb_attn_info *sb_attn_sw = p_hwfn->p_sb_attn; u32 igu_mask; /* Mask the source of the attention in the IGU */ - igu_mask = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, - IGU_REG_ATTENTION_ENABLE); + igu_mask = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, IGU_REG_ATTENTION_ENABLE); DP_VERBOSE(p_hwfn, NETIF_MSG_INTR, "IGU mask: 0x%08x --> 0x%08x\n", igu_mask, igu_mask & ~(asserted_bits & ATTN_BITS_MASKABLE)); igu_mask &= ~(asserted_bits & ATTN_BITS_MASKABLE); @@ -2041,7 +2038,7 @@ static int qed_int_deassertion(struct qed_hwfn *p_hwfn, struct aeu_invert_reg_bit *p_bit = &p_aeu->bits[j]; if ((p_bit->flags & ATTENTION_PARITY) && - !!(parities & (1 << bit_idx))) + !!(parities & BIT(bit_idx))) qed_int_deassertion_parity(p_hwfn, p_bit, bit_idx); @@ -2114,8 +2111,7 @@ static int qed_int_deassertion(struct qed_hwfn *p_hwfn, ~((u32)deasserted_bits)); /* Unmask deasserted attentions in IGU */ - aeu_mask = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, - IGU_REG_ATTENTION_ENABLE); + aeu_mask = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, IGU_REG_ATTENTION_ENABLE); aeu_mask |= (deasserted_bits & ATTN_BITS_MASKABLE); qed_wr(p_hwfn, p_hwfn->p_dpc_ptt, IGU_REG_ATTENTION_ENABLE, aeu_mask); @@ -2160,8 +2156,7 @@ static int qed_int_attentions(struct qed_hwfn *p_hwfn) index, attn_bits, attn_acks, asserted_bits, deasserted_bits, p_sb_attn_sw->known_attn); } else if (asserted_bits == 0x100) { - DP_INFO(p_hwfn, - "MFW indication via attention\n"); + DP_INFO(p_hwfn, "MFW indication via attention\n"); } else { DP_VERBOSE(p_hwfn, NETIF_MSG_INTR, "MFW indication [deassertion]\n"); @@ -2173,18 +2168,14 @@ static int qed_int_attentions(struct qed_hwfn *p_hwfn) return rc; } - if (deasserted_bits) { + if (deasserted_bits) rc = qed_int_deassertion(p_hwfn, deasserted_bits); - if (rc) - return rc; - } return rc; } static void qed_sb_ack_attn(struct qed_hwfn *p_hwfn, - void __iomem *igu_addr, - u32 ack_cons) + void __iomem *igu_addr, u32 ack_cons) { struct igu_prod_cons_update igu_ack = { 0 }; @@ -2242,9 +2233,8 @@ void qed_int_sp_dpc(unsigned long hwfn_cookie) /* Gather Interrupts/Attentions information */ if (!sb_info->sb_virt) { - DP_ERR( - p_hwfn->cdev, - "Interrupt Status block is NULL - cannot check for new interrupts!\n"); + DP_ERR(p_hwfn->cdev, + "Interrupt Status block is NULL - cannot check for new interrupts!\n"); } else { u32 tmp_index = sb_info->sb_ack; @@ -2255,9 +2245,8 @@ void qed_int_sp_dpc(unsigned long hwfn_cookie) } if (!sb_attn || !sb_attn->sb_attn) { - DP_ERR( - p_hwfn->cdev, - "Attentions Status block is NULL - cannot check for new attentions!\n"); + DP_ERR(p_hwfn->cdev, + "Attentions Status block is NULL - cannot check for new attentions!\n"); } else { u16 tmp_index = sb_attn->index; @@ -2313,8 +2302,7 @@ static void qed_int_sb_attn_free(struct qed_hwfn *p_hwfn) if (p_sb->sb_attn) dma_free_coherent(&p_hwfn->cdev->pdev->dev, SB_ATTN_ALIGNED_SIZE(p_hwfn), - p_sb->sb_attn, - p_sb->sb_phys); + p_sb->sb_attn, p_sb->sb_phys); kfree(p_sb); } @@ -2337,8 +2325,7 @@ static void qed_int_sb_attn_setup(struct qed_hwfn *p_hwfn, static void qed_int_sb_attn_init(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, - void *sb_virt_addr, - dma_addr_t sb_phy_addr) + void *sb_virt_addr, dma_addr_t sb_phy_addr) { struct qed_sb_attn_info *sb_info = p_hwfn->p_sb_attn; int i, j, k; @@ -2378,8 +2365,8 @@ static int qed_int_sb_attn_alloc(struct qed_hwfn *p_hwfn, { struct qed_dev *cdev = p_hwfn->cdev; struct qed_sb_attn_info *p_sb; - void *p_virt; dma_addr_t p_phys = 0; + void *p_virt; /* SB struct */ p_sb = kmalloc(sizeof(*p_sb), GFP_KERNEL); @@ -2412,9 +2399,7 @@ static int qed_int_sb_attn_alloc(struct qed_hwfn *p_hwfn, void qed_init_cau_sb_entry(struct qed_hwfn *p_hwfn, struct cau_sb_entry *p_sb_entry, - u8 pf_id, - u16 vf_number, - u8 vf_valid) + u8 pf_id, u16 vf_number, u8 vf_valid) { struct qed_dev *cdev = p_hwfn->cdev; u32 cau_state; @@ -2428,12 +2413,6 @@ void qed_init_cau_sb_entry(struct qed_hwfn *p_hwfn, SET_FIELD(p_sb_entry->params, CAU_SB_ENTRY_SB_TIMESET0, 0x7F); SET_FIELD(p_sb_entry->params, CAU_SB_ENTRY_SB_TIMESET1, 0x7F); - /* setting the time resultion to a fixed value ( = 1) */ - SET_FIELD(p_sb_entry->params, CAU_SB_ENTRY_TIMER_RES0, - QED_CAU_DEF_RX_TIMER_RES); - SET_FIELD(p_sb_entry->params, CAU_SB_ENTRY_TIMER_RES1, - QED_CAU_DEF_TX_TIMER_RES); - cau_state = CAU_HC_DISABLE_STATE; if (cdev->int_coalescing_mode == QED_COAL_MODE_ENABLE) { @@ -2468,9 +2447,7 @@ void qed_init_cau_sb_entry(struct qed_hwfn *p_hwfn, void qed_int_cau_conf_sb(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, dma_addr_t sb_phys, - u16 igu_sb_id, - u16 vf_number, - u8 vf_valid) + u16 igu_sb_id, u16 vf_number, u8 vf_valid) { struct cau_sb_entry sb_entry; @@ -2514,8 +2491,7 @@ void qed_int_cau_conf_sb(struct qed_hwfn *p_hwfn, timer_res = 2; timeset = (u8)(p_hwfn->cdev->rx_coalesce_usecs >> timer_res); qed_int_cau_conf_pi(p_hwfn, p_ptt, igu_sb_id, RX_PI, - QED_COAL_RX_STATE_MACHINE, - timeset); + QED_COAL_RX_STATE_MACHINE, timeset); if (p_hwfn->cdev->tx_coalesce_usecs <= 0x7F) timer_res = 0; @@ -2541,8 +2517,7 @@ void qed_int_cau_conf_pi(struct qed_hwfn *p_hwfn, u8 timeset) { struct cau_pi_entry pi_entry; - u32 sb_offset; - u32 pi_offset; + u32 sb_offset, pi_offset; if (IS_VF(p_hwfn->cdev)) return; @@ -2569,8 +2544,7 @@ void qed_int_cau_conf_pi(struct qed_hwfn *p_hwfn, } void qed_int_sb_setup(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - struct qed_sb_info *sb_info) + struct qed_ptt *p_ptt, struct qed_sb_info *sb_info) { /* zero status block and ack counter */ sb_info->sb_ack = 0; @@ -2590,8 +2564,7 @@ void qed_int_sb_setup(struct qed_hwfn *p_hwfn, * * @return u16 */ -static u16 qed_get_igu_sb_id(struct qed_hwfn *p_hwfn, - u16 sb_id) +static u16 qed_get_igu_sb_id(struct qed_hwfn *p_hwfn, u16 sb_id) { u16 igu_sb_id; @@ -2603,8 +2576,12 @@ static u16 qed_get_igu_sb_id(struct qed_hwfn *p_hwfn, else igu_sb_id = qed_vf_get_igu_sb_id(p_hwfn, sb_id); - DP_VERBOSE(p_hwfn, NETIF_MSG_INTR, "SB [%s] index is 0x%04x\n", - (sb_id == QED_SP_SB_ID) ? "DSB" : "non-DSB", igu_sb_id); + if (sb_id == QED_SP_SB_ID) + DP_VERBOSE(p_hwfn, NETIF_MSG_INTR, + "Slowpath SB index in IGU is 0x%04x\n", igu_sb_id); + else + DP_VERBOSE(p_hwfn, NETIF_MSG_INTR, + "SB [%04x] <--> IGU SB [%04x]\n", sb_id, igu_sb_id); return igu_sb_id; } @@ -2612,9 +2589,7 @@ static u16 qed_get_igu_sb_id(struct qed_hwfn *p_hwfn, int qed_int_sb_init(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, struct qed_sb_info *sb_info, - void *sb_virt_addr, - dma_addr_t sb_phy_addr, - u16 sb_id) + void *sb_virt_addr, dma_addr_t sb_phy_addr, u16 sb_id) { sb_info->sb_virt = sb_virt_addr; sb_info->sb_phys = sb_phy_addr; @@ -2650,8 +2625,7 @@ int qed_int_sb_init(struct qed_hwfn *p_hwfn, } int qed_int_sb_release(struct qed_hwfn *p_hwfn, - struct qed_sb_info *sb_info, - u16 sb_id) + struct qed_sb_info *sb_info, u16 sb_id) { if (sb_id == QED_SP_SB_ID) { DP_ERR(p_hwfn, "Do Not free sp sb using this function"); @@ -2685,8 +2659,7 @@ static void qed_int_sp_sb_free(struct qed_hwfn *p_hwfn) kfree(p_sb); } -static int qed_int_sp_sb_alloc(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) +static int qed_int_sp_sb_alloc(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { struct qed_sb_sp_info *p_sb; dma_addr_t p_phys = 0; @@ -2721,9 +2694,7 @@ static int qed_int_sp_sb_alloc(struct qed_hwfn *p_hwfn, int qed_int_register_cb(struct qed_hwfn *p_hwfn, qed_int_comp_cb_t comp_cb, - void *cookie, - u8 *sb_idx, - __le16 **p_fw_cons) + void *cookie, u8 *sb_idx, __le16 **p_fw_cons) { struct qed_sb_sp_info *p_sp_sb = p_hwfn->p_sp_sb; int rc = -ENOMEM; @@ -2764,8 +2735,7 @@ u16 qed_int_get_sp_sb_id(struct qed_hwfn *p_hwfn) } void qed_int_igu_enable_int(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - enum qed_int_mode int_mode) + struct qed_ptt *p_ptt, enum qed_int_mode int_mode) { u32 igu_pf_conf = IGU_PF_CONF_FUNC_EN | IGU_PF_CONF_ATTN_BIT_EN; @@ -2809,7 +2779,7 @@ int qed_int_igu_enable(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, qed_wr(p_hwfn, p_ptt, MISC_REG_AEU_MASK_ATTN_IGU, 0xff); if ((int_mode != QED_INT_MODE_INTA) || IS_LEAD_HWFN(p_hwfn)) { rc = qed_slowpath_irq_req(p_hwfn); - if (rc != 0) { + if (rc) { DP_NOTICE(p_hwfn, "Slowpath IRQ request failed\n"); return -EINVAL; } @@ -2822,8 +2792,7 @@ int qed_int_igu_enable(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, return rc; } -void qed_int_igu_disable_int(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) +void qed_int_igu_disable_int(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { p_hwfn->b_int_enabled = 0; @@ -2950,13 +2919,11 @@ void qed_int_igu_init_pure_rt(struct qed_hwfn *p_hwfn, p_hwfn->hw_info.opaque_fid, b_set); } -static u32 qed_int_igu_read_cam_block(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - u16 sb_id) +static u32 qed_int_igu_read_cam_block(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, u16 sb_id) { u32 val = qed_rd(p_hwfn, p_ptt, - IGU_REG_MAPPING_MEMORY + - sizeof(u32) * sb_id); + IGU_REG_MAPPING_MEMORY + sizeof(u32) * sb_id); struct qed_igu_block *p_block; p_block = &p_hwfn->hw_info.p_igu_info->igu_map.igu_blocks[sb_id]; @@ -2983,8 +2950,7 @@ out: return val; } -int qed_int_igu_read_cam(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) +int qed_int_igu_read_cam(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { struct qed_igu_info *p_igu_info; u32 val, min_vf = 0, max_vf = 0; @@ -3104,22 +3070,19 @@ int qed_int_igu_read_cam(struct qed_hwfn *p_hwfn, */ void qed_int_igu_init_rt(struct qed_hwfn *p_hwfn) { - u32 igu_pf_conf = 0; - - igu_pf_conf |= IGU_PF_CONF_FUNC_EN; + u32 igu_pf_conf = IGU_PF_CONF_FUNC_EN; STORE_RT_REG(p_hwfn, IGU_REG_PF_CONFIGURATION_RT_OFFSET, igu_pf_conf); } u64 qed_int_igu_read_sisr_reg(struct qed_hwfn *p_hwfn) { - u64 intr_status = 0; - u32 intr_status_lo = 0; - u32 intr_status_hi = 0; u32 lsb_igu_cmd_addr = IGU_REG_SISR_MDPC_WMASK_LSB_UPPER - IGU_CMD_INT_ACK_BASE; u32 msb_igu_cmd_addr = IGU_REG_SISR_MDPC_WMASK_MSB_UPPER - IGU_CMD_INT_ACK_BASE; + u32 intr_status_hi = 0, intr_status_lo = 0; + u64 intr_status = 0; intr_status_lo = REG_RD(p_hwfn, GTT_BAR0_MAP_REG_IGU_CMD + @@ -3153,8 +3116,7 @@ static void qed_int_sp_dpc_free(struct qed_hwfn *p_hwfn) kfree(p_hwfn->sp_dpc); } -int qed_int_alloc(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) +int qed_int_alloc(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { int rc = 0; @@ -3169,10 +3131,9 @@ int qed_int_alloc(struct qed_hwfn *p_hwfn, return rc; } rc = qed_int_sb_attn_alloc(p_hwfn, p_ptt); - if (rc) { + if (rc) DP_ERR(p_hwfn->cdev, "Failed to allocate sb attn mem\n"); - return rc; - } + return rc; } @@ -3183,8 +3144,7 @@ void qed_int_free(struct qed_hwfn *p_hwfn) qed_int_sp_dpc_free(p_hwfn); } -void qed_int_setup(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) +void qed_int_setup(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { qed_int_sb_setup(p_hwfn, p_ptt, &p_hwfn->p_sp_sb->sb_info); qed_int_sb_attn_setup(p_hwfn, p_ptt); diff --git a/drivers/net/ethernet/qlogic/qed/qed_l2.c b/drivers/net/ethernet/qlogic/qed/qed_l2.c index 401e738543b5..c823c46fd979 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_l2.c +++ b/drivers/net/ethernet/qlogic/qed/qed_l2.c @@ -52,7 +52,7 @@ int qed_sp_eth_vport_start(struct qed_hwfn *p_hwfn, u16 rx_mode = 0; rc = qed_fw_vport(p_hwfn, p_params->vport_id, &abs_vport_id); - if (rc != 0) + if (rc) return rc; memset(&init_data, 0, sizeof(init_data)); @@ -80,8 +80,7 @@ int qed_sp_eth_vport_start(struct qed_hwfn *p_hwfn, p_ramrod->rx_mode.state = cpu_to_le16(rx_mode); /* TPA related fields */ - memset(&p_ramrod->tpa_param, 0, - sizeof(struct eth_vport_tpa_param)); + memset(&p_ramrod->tpa_param, 0, sizeof(struct eth_vport_tpa_param)); p_ramrod->tpa_param.max_buff_num = p_params->max_buffers_per_cqe; @@ -306,14 +305,14 @@ qed_sp_update_mcast_bin(struct qed_hwfn *p_hwfn, memset(&p_ramrod->approx_mcast.bins, 0, sizeof(p_ramrod->approx_mcast.bins)); - if (p_params->update_approx_mcast_flg) { - p_ramrod->common.update_approx_mcast_flg = 1; - for (i = 0; i < ETH_MULTICAST_MAC_BINS_IN_REGS; i++) { - u32 *p_bins = (u32 *)p_params->bins; - __le32 val = cpu_to_le32(p_bins[i]); + if (!p_params->update_approx_mcast_flg) + return; - p_ramrod->approx_mcast.bins[i] = val; - } + p_ramrod->common.update_approx_mcast_flg = 1; + for (i = 0; i < ETH_MULTICAST_MAC_BINS_IN_REGS; i++) { + u32 *p_bins = (u32 *)p_params->bins; + + p_ramrod->approx_mcast.bins[i] = cpu_to_le32(p_bins[i]); } } @@ -336,7 +335,7 @@ int qed_sp_vport_update(struct qed_hwfn *p_hwfn, } rc = qed_fw_vport(p_hwfn, p_params->vport_id, &abs_vport_id); - if (rc != 0) + if (rc) return rc; memset(&init_data, 0, sizeof(init_data)); @@ -361,8 +360,8 @@ int qed_sp_vport_update(struct qed_hwfn *p_hwfn, p_cmn->tx_active_flg = p_params->vport_active_tx_flg; p_cmn->update_tx_active_flg = p_params->update_vport_active_tx_flg; p_cmn->accept_any_vlan = p_params->accept_any_vlan; - p_cmn->update_accept_any_vlan_flg = - p_params->update_accept_any_vlan_flg; + val = p_params->update_accept_any_vlan_flg; + p_cmn->update_accept_any_vlan_flg = val; p_cmn->inner_vlan_removal_en = p_params->inner_vlan_removal_flg; val = p_params->update_inner_vlan_removal_flg; @@ -411,7 +410,7 @@ int qed_sp_vport_stop(struct qed_hwfn *p_hwfn, u16 opaque_fid, u8 vport_id) return qed_vf_pf_vport_stop(p_hwfn); rc = qed_fw_vport(p_hwfn, vport_id, &abs_vport_id); - if (rc != 0) + if (rc) return rc; memset(&init_data, 0, sizeof(init_data)); @@ -476,7 +475,7 @@ static int qed_filter_accept_cmd(struct qed_dev *cdev, rc = qed_sp_vport_update(p_hwfn, &vport_update_params, comp_mode, p_comp_data); - if (rc != 0) { + if (rc) { DP_ERR(cdev, "Update rx_mode failed %d\n", rc); return rc; } @@ -511,7 +510,7 @@ static int qed_sp_release_queue_cid( int qed_sp_eth_rxq_start_ramrod(struct qed_hwfn *p_hwfn, u16 opaque_fid, u32 cid, - struct qed_queue_start_common_params *params, + struct qed_queue_start_common_params *p_params, u8 stats_id, u16 bd_max_bytes, dma_addr_t bd_chain_phys_addr, @@ -526,23 +525,23 @@ int qed_sp_eth_rxq_start_ramrod(struct qed_hwfn *p_hwfn, int rc = -EINVAL; /* Store information for the stop */ - p_rx_cid = &p_hwfn->p_rx_cids[params->queue_id]; - p_rx_cid->cid = cid; - p_rx_cid->opaque_fid = opaque_fid; - p_rx_cid->vport_id = params->vport_id; + p_rx_cid = &p_hwfn->p_rx_cids[p_params->queue_id]; + p_rx_cid->cid = cid; + p_rx_cid->opaque_fid = opaque_fid; + p_rx_cid->vport_id = p_params->vport_id; - rc = qed_fw_vport(p_hwfn, params->vport_id, &abs_vport_id); - if (rc != 0) + rc = qed_fw_vport(p_hwfn, p_params->vport_id, &abs_vport_id); + if (rc) return rc; - rc = qed_fw_l2_queue(p_hwfn, params->queue_id, &abs_rx_q_id); - if (rc != 0) + rc = qed_fw_l2_queue(p_hwfn, p_params->queue_id, &abs_rx_q_id); + if (rc) return rc; DP_VERBOSE(p_hwfn, QED_MSG_SP, "opaque_fid=0x%x, cid=0x%x, rx_qid=0x%x, vport_id=0x%x, sb_id=0x%x\n", - opaque_fid, cid, params->queue_id, params->vport_id, - params->sb); + opaque_fid, + cid, p_params->queue_id, p_params->vport_id, p_params->sb); /* Get SPQ entry */ memset(&init_data, 0, sizeof(init_data)); @@ -558,24 +557,25 @@ int qed_sp_eth_rxq_start_ramrod(struct qed_hwfn *p_hwfn, p_ramrod = &p_ent->ramrod.rx_queue_start; - p_ramrod->sb_id = cpu_to_le16(params->sb); - p_ramrod->sb_index = params->sb_idx; - p_ramrod->vport_id = abs_vport_id; - p_ramrod->stats_counter_id = stats_id; - p_ramrod->rx_queue_id = cpu_to_le16(abs_rx_q_id); - p_ramrod->complete_cqe_flg = 0; - p_ramrod->complete_event_flg = 1; + p_ramrod->sb_id = cpu_to_le16(p_params->sb); + p_ramrod->sb_index = p_params->sb_idx; + p_ramrod->vport_id = abs_vport_id; + p_ramrod->stats_counter_id = stats_id; + p_ramrod->rx_queue_id = cpu_to_le16(abs_rx_q_id); + p_ramrod->complete_cqe_flg = 0; + p_ramrod->complete_event_flg = 1; - p_ramrod->bd_max_bytes = cpu_to_le16(bd_max_bytes); + p_ramrod->bd_max_bytes = cpu_to_le16(bd_max_bytes); DMA_REGPAIR_LE(p_ramrod->bd_base, bd_chain_phys_addr); - p_ramrod->num_of_pbl_pages = cpu_to_le16(cqe_pbl_size); + p_ramrod->num_of_pbl_pages = cpu_to_le16(cqe_pbl_size); DMA_REGPAIR_LE(p_ramrod->cqe_pbl_addr, cqe_pbl_addr); - p_ramrod->vf_rx_prod_index = params->vf_qid; - if (params->vf_qid) + p_ramrod->vf_rx_prod_index = p_params->vf_qid; + if (p_params->vf_qid) DP_VERBOSE(p_hwfn, QED_MSG_SP, - "Queue is meant for VF rxq[%04x]\n", params->vf_qid); + "Queue is meant for VF rxq[%04x]\n", + p_params->vf_qid); return qed_spq_post(p_hwfn, p_ent, NULL); } @@ -583,7 +583,7 @@ int qed_sp_eth_rxq_start_ramrod(struct qed_hwfn *p_hwfn, static int qed_sp_eth_rx_queue_start(struct qed_hwfn *p_hwfn, u16 opaque_fid, - struct qed_queue_start_common_params *params, + struct qed_queue_start_common_params *p_params, u16 bd_max_bytes, dma_addr_t bd_chain_phys_addr, dma_addr_t cqe_pbl_addr, @@ -597,20 +597,20 @@ qed_sp_eth_rx_queue_start(struct qed_hwfn *p_hwfn, if (IS_VF(p_hwfn->cdev)) { return qed_vf_pf_rxq_start(p_hwfn, - params->queue_id, - params->sb, - params->sb_idx, + p_params->queue_id, + p_params->sb, + (u8)p_params->sb_idx, bd_max_bytes, bd_chain_phys_addr, cqe_pbl_addr, cqe_pbl_size, pp_prod); } - rc = qed_fw_l2_queue(p_hwfn, params->queue_id, &abs_l2_queue); - if (rc != 0) + rc = qed_fw_l2_queue(p_hwfn, p_params->queue_id, &abs_l2_queue); + if (rc) return rc; - rc = qed_fw_vport(p_hwfn, params->vport_id, &abs_stats_id); - if (rc != 0) + rc = qed_fw_vport(p_hwfn, p_params->vport_id, &abs_stats_id); + if (rc) return rc; *pp_prod = (u8 __iomem *)p_hwfn->regview + @@ -622,9 +622,8 @@ qed_sp_eth_rx_queue_start(struct qed_hwfn *p_hwfn, (u32 *)(&init_prod_val)); /* Allocate a CID for the queue */ - p_rx_cid = &p_hwfn->p_rx_cids[params->queue_id]; - rc = qed_cxt_acquire_cid(p_hwfn, PROTOCOLID_ETH, - &p_rx_cid->cid); + p_rx_cid = &p_hwfn->p_rx_cids[p_params->queue_id]; + rc = qed_cxt_acquire_cid(p_hwfn, PROTOCOLID_ETH, &p_rx_cid->cid); if (rc) { DP_NOTICE(p_hwfn, "Failed to acquire cid\n"); return rc; @@ -634,14 +633,14 @@ qed_sp_eth_rx_queue_start(struct qed_hwfn *p_hwfn, rc = qed_sp_eth_rxq_start_ramrod(p_hwfn, opaque_fid, p_rx_cid->cid, - params, + p_params, abs_stats_id, bd_max_bytes, bd_chain_phys_addr, cqe_pbl_addr, cqe_pbl_size); - if (rc != 0) + if (rc) qed_sp_release_queue_cid(p_hwfn, p_rx_cid); return rc; @@ -788,21 +787,20 @@ int qed_sp_eth_txq_start_ramrod(struct qed_hwfn *p_hwfn, if (rc) return rc; - p_ramrod = &p_ent->ramrod.tx_queue_start; - p_ramrod->vport_id = abs_vport_id; + p_ramrod = &p_ent->ramrod.tx_queue_start; + p_ramrod->vport_id = abs_vport_id; - p_ramrod->sb_id = cpu_to_le16(p_params->sb); - p_ramrod->sb_index = p_params->sb_idx; - p_ramrod->stats_counter_id = stats_id; + p_ramrod->sb_id = cpu_to_le16(p_params->sb); + p_ramrod->sb_index = p_params->sb_idx; + p_ramrod->stats_counter_id = stats_id; - p_ramrod->queue_zone_id = cpu_to_le16(abs_tx_q_id); - p_ramrod->pbl_size = cpu_to_le16(pbl_size); + p_ramrod->queue_zone_id = cpu_to_le16(abs_tx_q_id); + + p_ramrod->pbl_size = cpu_to_le16(pbl_size); DMA_REGPAIR_LE(p_ramrod->pbl_base_addr, pbl_addr); - pq_id = qed_get_qm_pq(p_hwfn, - PROTOCOLID_ETH, - p_pq_params); - p_ramrod->qm_pq_id = cpu_to_le16(pq_id); + pq_id = qed_get_qm_pq(p_hwfn, PROTOCOLID_ETH, p_pq_params); + p_ramrod->qm_pq_id = cpu_to_le16(pq_id); return qed_spq_post(p_hwfn, p_ent, NULL); } @@ -836,8 +834,7 @@ qed_sp_eth_tx_queue_start(struct qed_hwfn *p_hwfn, memset(&pq_params, 0, sizeof(pq_params)); /* Allocate a CID for the queue */ - rc = qed_cxt_acquire_cid(p_hwfn, PROTOCOLID_ETH, - &p_tx_cid->cid); + rc = qed_cxt_acquire_cid(p_hwfn, PROTOCOLID_ETH, &p_tx_cid->cid); if (rc) { DP_NOTICE(p_hwfn, "Failed to acquire cid\n"); return rc; @@ -896,8 +893,7 @@ int qed_sp_eth_tx_queue_stop(struct qed_hwfn *p_hwfn, u16 tx_queue_id) return qed_sp_release_queue_cid(p_hwfn, p_tx_cid); } -static enum eth_filter_action -qed_filter_action(enum qed_filter_opcode opcode) +static enum eth_filter_action qed_filter_action(enum qed_filter_opcode opcode) { enum eth_filter_action action = MAX_ETH_FILTER_ACTION; @@ -1033,19 +1029,19 @@ qed_filter_ucast_common(struct qed_hwfn *p_hwfn, p_first_filter->vni = cpu_to_le32(p_filter_cmd->vni); if (p_filter_cmd->opcode == QED_FILTER_MOVE) { - p_second_filter->type = p_first_filter->type; - p_second_filter->mac_msb = p_first_filter->mac_msb; - p_second_filter->mac_mid = p_first_filter->mac_mid; - p_second_filter->mac_lsb = p_first_filter->mac_lsb; - p_second_filter->vlan_id = p_first_filter->vlan_id; - p_second_filter->vni = p_first_filter->vni; + p_second_filter->type = p_first_filter->type; + p_second_filter->mac_msb = p_first_filter->mac_msb; + p_second_filter->mac_mid = p_first_filter->mac_mid; + p_second_filter->mac_lsb = p_first_filter->mac_lsb; + p_second_filter->vlan_id = p_first_filter->vlan_id; + p_second_filter->vni = p_first_filter->vni; p_first_filter->action = ETH_FILTER_ACTION_REMOVE; p_first_filter->vport_id = vport_to_remove_from; - p_second_filter->action = ETH_FILTER_ACTION_ADD; - p_second_filter->vport_id = vport_to_add_to; + p_second_filter->action = ETH_FILTER_ACTION_ADD; + p_second_filter->vport_id = vport_to_add_to; } else if (p_filter_cmd->opcode == QED_FILTER_REPLACE) { p_first_filter->vport_id = vport_to_add_to; memcpy(p_second_filter, p_first_filter, @@ -1086,7 +1082,7 @@ int qed_sp_eth_filter_ucast(struct qed_hwfn *p_hwfn, rc = qed_filter_ucast_common(p_hwfn, opaque_fid, p_filter_cmd, &p_ramrod, &p_ent, comp_mode, p_comp_data); - if (rc != 0) { + if (rc) { DP_ERR(p_hwfn, "Uni. filter command failed %d\n", rc); return rc; } @@ -1094,10 +1090,8 @@ int qed_sp_eth_filter_ucast(struct qed_hwfn *p_hwfn, p_header->assert_on_error = p_filter_cmd->assert_on_error; rc = qed_spq_post(p_hwfn, p_ent, NULL); - if (rc != 0) { - DP_ERR(p_hwfn, - "Unicast filter ADD command failed %d\n", - rc); + if (rc) { + DP_ERR(p_hwfn, "Unicast filter ADD command failed %d\n", rc); return rc; } @@ -1136,15 +1130,10 @@ int qed_sp_eth_filter_ucast(struct qed_hwfn *p_hwfn, * Return: ******************************************************************************/ static u32 qed_calc_crc32c(u8 *crc32_packet, - u32 crc32_length, - u32 crc32_seed, - u8 complement) + u32 crc32_length, u32 crc32_seed, u8 complement) { - u32 byte = 0; - u32 bit = 0; - u8 msb = 0; - u8 current_byte = 0; - u32 crc32_result = crc32_seed; + u32 byte = 0, bit = 0, crc32_result = crc32_seed; + u8 msb = 0, current_byte = 0; if ((!crc32_packet) || (crc32_length == 0) || @@ -1164,9 +1153,7 @@ static u32 qed_calc_crc32c(u8 *crc32_packet, return crc32_result; } -static inline u32 qed_crc32c_le(u32 seed, - u8 *mac, - u32 len) +static u32 qed_crc32c_le(u32 seed, u8 *mac, u32 len) { u32 packet_buf[2] = { 0 }; @@ -1196,17 +1183,14 @@ qed_sp_eth_filter_mcast(struct qed_hwfn *p_hwfn, u8 abs_vport_id = 0; int rc, i; - if (p_filter_cmd->opcode == QED_FILTER_ADD) { + if (p_filter_cmd->opcode == QED_FILTER_ADD) rc = qed_fw_vport(p_hwfn, p_filter_cmd->vport_to_add_to, &abs_vport_id); - if (rc) - return rc; - } else { + else rc = qed_fw_vport(p_hwfn, p_filter_cmd->vport_to_remove_from, &abs_vport_id); - if (rc) - return rc; - } + if (rc) + return rc; /* Get SPQ entry */ memset(&init_data, 0, sizeof(init_data)); @@ -1244,11 +1228,11 @@ qed_sp_eth_filter_mcast(struct qed_hwfn *p_hwfn, /* Convert to correct endianity */ for (i = 0; i < ETH_MULTICAST_MAC_BINS_IN_REGS; i++) { + struct vport_update_ramrod_mcast *p_ramrod_bins; u32 *p_bins = (u32 *)bins; - struct vport_update_ramrod_mcast *approx_mcast; - approx_mcast = &p_ramrod->approx_mcast; - approx_mcast->bins[i] = cpu_to_le32(p_bins[i]); + p_ramrod_bins = &p_ramrod->approx_mcast; + p_ramrod_bins->bins[i] = cpu_to_le32(p_bins[i]); } } @@ -1286,8 +1270,7 @@ static int qed_filter_mcast_cmd(struct qed_dev *cdev, rc = qed_sp_eth_filter_mcast(p_hwfn, opaque_fid, p_filter_cmd, - comp_mode, - p_comp_data); + comp_mode, p_comp_data); } return rc; } @@ -1314,9 +1297,8 @@ static int qed_filter_ucast_cmd(struct qed_dev *cdev, rc = qed_sp_eth_filter_ucast(p_hwfn, opaque_fid, p_filter_cmd, - comp_mode, - p_comp_data); - if (rc != 0) + comp_mode, p_comp_data); + if (rc) break; } @@ -1590,8 +1572,7 @@ out: } } -void qed_get_vport_stats(struct qed_dev *cdev, - struct qed_eth_stats *stats) +void qed_get_vport_stats(struct qed_dev *cdev, struct qed_eth_stats *stats) { u32 i; @@ -1766,8 +1747,7 @@ static int qed_start_vport(struct qed_dev *cdev, return 0; } -static int qed_stop_vport(struct qed_dev *cdev, - u8 vport_id) +static int qed_stop_vport(struct qed_dev *cdev, u8 vport_id) { int rc, i; @@ -1775,8 +1755,7 @@ static int qed_stop_vport(struct qed_dev *cdev, struct qed_hwfn *p_hwfn = &cdev->hwfns[i]; rc = qed_sp_vport_stop(p_hwfn, - p_hwfn->hw_info.opaque_fid, - vport_id); + p_hwfn->hw_info.opaque_fid, vport_id); if (rc) { DP_ERR(cdev, "Failed to stop VPORT\n"); @@ -1801,10 +1780,8 @@ static int qed_update_vport(struct qed_dev *cdev, /* Translate protocol params into sp params */ sp_params.vport_id = params->vport_id; - sp_params.update_vport_active_rx_flg = - params->update_vport_active_flg; - sp_params.update_vport_active_tx_flg = - params->update_vport_active_flg; + sp_params.update_vport_active_rx_flg = params->update_vport_active_flg; + sp_params.update_vport_active_tx_flg = params->update_vport_active_flg; sp_params.vport_active_rx_flg = params->vport_active_flg; sp_params.vport_active_tx_flg = params->vport_active_flg; sp_params.update_tx_switching_flg = params->update_tx_switching_flg; @@ -1817,8 +1794,7 @@ static int qed_update_vport(struct qed_dev *cdev, * We need to re-fix the rss values per engine for CMT. */ if (cdev->num_hwfns > 1 && params->update_rss_flg) { - struct qed_update_vport_rss_params *rss = - ¶ms->rss_params; + struct qed_update_vport_rss_params *rss = ¶ms->rss_params; int k, max = 0; /* Find largest entry, since it's possible RSS needs to @@ -1861,8 +1837,8 @@ static int qed_update_vport(struct qed_dev *cdev, QED_RSS_IND_TABLE_SIZE * sizeof(u16)); memcpy(sp_rss_params.rss_key, params->rss_params.rss_key, QED_RSS_KEY_SIZE * sizeof(u32)); + sp_params.rss_params = &sp_rss_params; } - sp_params.rss_params = &sp_rss_params; for_each_hwfn(cdev, i) { struct qed_hwfn *p_hwfn = &cdev->hwfns[i]; @@ -1893,8 +1869,8 @@ static int qed_start_rxq(struct qed_dev *cdev, u16 cqe_pbl_size, void __iomem **pp_prod) { - int rc, hwfn_index; struct qed_hwfn *p_hwfn; + int rc, hwfn_index; hwfn_index = params->rss_id % cdev->num_hwfns; p_hwfn = &cdev->hwfns[hwfn_index]; @@ -1935,8 +1911,7 @@ static int qed_stop_rxq(struct qed_dev *cdev, rc = qed_sp_eth_rx_queue_stop(p_hwfn, params->rx_queue_id / cdev->num_hwfns, - params->eq_completion_only, - false); + params->eq_completion_only, false); if (rc) { DP_ERR(cdev, "Failed to stop RXQ#%d\n", params->rx_queue_id); return rc; @@ -2047,11 +2022,11 @@ static int qed_configure_filter_rx_mode(struct qed_dev *cdev, memset(&accept_flags, 0, sizeof(accept_flags)); - accept_flags.update_rx_mode_config = 1; - accept_flags.update_tx_mode_config = 1; - accept_flags.rx_accept_filter = QED_ACCEPT_UCAST_MATCHED | - QED_ACCEPT_MCAST_MATCHED | - QED_ACCEPT_BCAST; + accept_flags.update_rx_mode_config = 1; + accept_flags.update_tx_mode_config = 1; + accept_flags.rx_accept_filter = QED_ACCEPT_UCAST_MATCHED | + QED_ACCEPT_MCAST_MATCHED | + QED_ACCEPT_BCAST; accept_flags.tx_accept_filter = QED_ACCEPT_UCAST_MATCHED | QED_ACCEPT_MCAST_MATCHED | QED_ACCEPT_BCAST; @@ -2072,9 +2047,8 @@ static int qed_configure_filter_ucast(struct qed_dev *cdev, struct qed_filter_ucast ucast; if (!params->vlan_valid && !params->mac_valid) { - DP_NOTICE( - cdev, - "Tried configuring a unicast filter, but both MAC and VLAN are not set\n"); + DP_NOTICE(cdev, + "Tried configuring a unicast filter, but both MAC and VLAN are not set\n"); return -EINVAL; } @@ -2135,8 +2109,7 @@ static int qed_configure_filter_mcast(struct qed_dev *cdev, for (i = 0; i < mcast.num_mc_addrs; i++) ether_addr_copy(mcast.mac[i], params->mac[i]); - return qed_filter_mcast_cmd(cdev, &mcast, - QED_SPQ_MODE_CB, NULL); + return qed_filter_mcast_cmd(cdev, &mcast, QED_SPQ_MODE_CB, NULL); } static int qed_configure_filter(struct qed_dev *cdev, @@ -2153,15 +2126,13 @@ static int qed_configure_filter(struct qed_dev *cdev, accept_flags = params->filter.accept_flags; return qed_configure_filter_rx_mode(cdev, accept_flags); default: - DP_NOTICE(cdev, "Unknown filter type %d\n", - (int)params->type); + DP_NOTICE(cdev, "Unknown filter type %d\n", (int)params->type); return -EINVAL; } } static int qed_fp_cqe_completion(struct qed_dev *dev, - u8 rss_id, - struct eth_slow_path_rx_cqe *cqe) + u8 rss_id, struct eth_slow_path_rx_cqe *cqe) { return qed_eth_cqe_completion(&dev->hwfns[rss_id % dev->num_hwfns], cqe); diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c index c7dc34bfdd0a..54976cc2747d 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_main.c +++ b/drivers/net/ethernet/qlogic/qed/qed_main.c @@ -51,8 +51,6 @@ MODULE_FIRMWARE(QED_FW_FILE_NAME); static int __init qed_init(void) { - pr_notice("qed_init called\n"); - pr_info("%s", version); return 0; @@ -106,8 +104,7 @@ static void qed_free_pci(struct qed_dev *cdev) /* Performs PCI initializations as well as initializing PCI-related parameters * in the device structrue. Returns 0 in case of success. */ -static int qed_init_pci(struct qed_dev *cdev, - struct pci_dev *pdev) +static int qed_init_pci(struct qed_dev *cdev, struct pci_dev *pdev) { u8 rev_id; int rc; @@ -263,8 +260,7 @@ static struct qed_dev *qed_alloc_cdev(struct pci_dev *pdev) } /* Sets the requested power state */ -static int qed_set_power_state(struct qed_dev *cdev, - pci_power_t state) +static int qed_set_power_state(struct qed_dev *cdev, pci_power_t state) { if (!cdev) return -ENODEV; @@ -366,8 +362,8 @@ static int qed_enable_msix(struct qed_dev *cdev, DP_NOTICE(cdev, "Trying to enable MSI-X with less vectors (%d out of %d)\n", cnt, int_params->in.num_vectors); - rc = pci_enable_msix_exact(cdev->pdev, - int_params->msix_table, cnt); + rc = pci_enable_msix_exact(cdev->pdev, int_params->msix_table, + cnt); if (!rc) rc = cnt; } @@ -439,6 +435,11 @@ static int qed_set_int_mode(struct qed_dev *cdev, bool force_mode) } out: + if (!rc) + DP_INFO(cdev, "Using %s interrupts\n", + int_params->out.int_mode == QED_INT_MODE_INTA ? + "INTa" : int_params->out.int_mode == QED_INT_MODE_MSI ? + "MSI" : "MSIX"); cdev->int_coalescing_mode = QED_COAL_MODE_ENABLE; return rc; @@ -514,19 +515,18 @@ static irqreturn_t qed_single_int(int irq, void *dev_instance) int qed_slowpath_irq_req(struct qed_hwfn *hwfn) { struct qed_dev *cdev = hwfn->cdev; + u32 int_mode; int rc = 0; u8 id; - if (cdev->int_params.out.int_mode == QED_INT_MODE_MSIX) { + int_mode = cdev->int_params.out.int_mode; + if (int_mode == QED_INT_MODE_MSIX) { id = hwfn->my_id; snprintf(hwfn->name, NAME_SIZE, "sp-%d-%02x:%02x.%02x", id, cdev->pdev->bus->number, PCI_SLOT(cdev->pdev->devfn), hwfn->abs_pf_id); rc = request_irq(cdev->int_params.msix_table[id].vector, qed_msix_sp_int, 0, hwfn->name, hwfn->sp_dpc); - if (!rc) - DP_VERBOSE(hwfn, (NETIF_MSG_INTR | QED_MSG_SP), - "Requested slowpath MSI-X\n"); } else { unsigned long flags = 0; @@ -541,6 +541,13 @@ int qed_slowpath_irq_req(struct qed_hwfn *hwfn) flags, cdev->name, cdev); } + if (rc) + DP_NOTICE(cdev, "request_irq failed, rc = %d\n", rc); + else + DP_VERBOSE(hwfn, (NETIF_MSG_INTR | QED_MSG_SP), + "Requested slowpath %s\n", + (int_mode == QED_INT_MODE_MSIX) ? "MSI-X" : "IRQ"); + return rc; } @@ -974,8 +981,7 @@ static u32 qed_sb_init(struct qed_dev *cdev, } static u32 qed_sb_release(struct qed_dev *cdev, - struct qed_sb_info *sb_info, - u16 sb_id) + struct qed_sb_info *sb_info, u16 sb_id) { struct qed_hwfn *p_hwfn; int hwfn_index; @@ -1025,20 +1031,23 @@ static int qed_set_link(struct qed_dev *cdev, struct qed_link_params *params) link_params->speed.autoneg = params->autoneg; if (params->override_flags & QED_LINK_OVERRIDE_SPEED_ADV_SPEEDS) { link_params->speed.advertised_speeds = 0; - if ((params->adv_speeds & SUPPORTED_1000baseT_Half) || - (params->adv_speeds & SUPPORTED_1000baseT_Full)) + if ((params->adv_speeds & QED_LM_1000baseT_Half_BIT) || + (params->adv_speeds & QED_LM_1000baseT_Full_BIT)) + link_params->speed.advertised_speeds |= + NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_1G; + if (params->adv_speeds & QED_LM_10000baseKR_Full_BIT) link_params->speed.advertised_speeds |= - NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_1G; - if (params->adv_speeds & SUPPORTED_10000baseKR_Full) + NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_10G; + if (params->adv_speeds & QED_LM_25000baseKR_Full_BIT) link_params->speed.advertised_speeds |= - NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_10G; - if (params->adv_speeds & SUPPORTED_40000baseLR4_Full) + NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_25G; + if (params->adv_speeds & QED_LM_40000baseLR4_Full_BIT) link_params->speed.advertised_speeds |= - NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_40G; - if (params->adv_speeds & 0) + NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_40G; + if (params->adv_speeds & QED_LM_50000baseKR2_Full_BIT) link_params->speed.advertised_speeds |= - NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_50G; - if (params->adv_speeds & 0) + NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_50G; + if (params->adv_speeds & QED_LM_100000baseKR4_Full_BIT) link_params->speed.advertised_speeds |= NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_BB_100G; } @@ -1168,50 +1177,56 @@ static void qed_fill_link(struct qed_hwfn *hwfn, if_link->link_up = true; /* TODO - at the moment assume supported and advertised speed equal */ - if_link->supported_caps = SUPPORTED_FIBRE; + if_link->supported_caps = QED_LM_FIBRE_BIT; if (params.speed.autoneg) - if_link->supported_caps |= SUPPORTED_Autoneg; + if_link->supported_caps |= QED_LM_Autoneg_BIT; if (params.pause.autoneg || (params.pause.forced_rx && params.pause.forced_tx)) - if_link->supported_caps |= SUPPORTED_Asym_Pause; + if_link->supported_caps |= QED_LM_Asym_Pause_BIT; if (params.pause.autoneg || params.pause.forced_rx || params.pause.forced_tx) - if_link->supported_caps |= SUPPORTED_Pause; + if_link->supported_caps |= QED_LM_Pause_BIT; if_link->advertised_caps = if_link->supported_caps; if (params.speed.advertised_speeds & NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_1G) - if_link->advertised_caps |= SUPPORTED_1000baseT_Half | - SUPPORTED_1000baseT_Full; + if_link->advertised_caps |= QED_LM_1000baseT_Half_BIT | + QED_LM_1000baseT_Full_BIT; if (params.speed.advertised_speeds & NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_10G) - if_link->advertised_caps |= SUPPORTED_10000baseKR_Full; + if_link->advertised_caps |= QED_LM_10000baseKR_Full_BIT; if (params.speed.advertised_speeds & - NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_40G) - if_link->advertised_caps |= SUPPORTED_40000baseLR4_Full; + NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_25G) + if_link->advertised_caps |= QED_LM_25000baseKR_Full_BIT; if (params.speed.advertised_speeds & - NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_50G) - if_link->advertised_caps |= 0; + NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_40G) + if_link->advertised_caps |= QED_LM_40000baseLR4_Full_BIT; + if (params.speed.advertised_speeds & + NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_50G) + if_link->advertised_caps |= QED_LM_50000baseKR2_Full_BIT; if (params.speed.advertised_speeds & NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_BB_100G) - if_link->advertised_caps |= 0; + if_link->advertised_caps |= QED_LM_100000baseKR4_Full_BIT; if (link_caps.speed_capabilities & NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_1G) - if_link->supported_caps |= SUPPORTED_1000baseT_Half | - SUPPORTED_1000baseT_Full; + if_link->supported_caps |= QED_LM_1000baseT_Half_BIT | + QED_LM_1000baseT_Full_BIT; if (link_caps.speed_capabilities & NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_10G) - if_link->supported_caps |= SUPPORTED_10000baseKR_Full; + if_link->supported_caps |= QED_LM_10000baseKR_Full_BIT; + if (link_caps.speed_capabilities & + NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_25G) + if_link->supported_caps |= QED_LM_25000baseKR_Full_BIT; if (link_caps.speed_capabilities & - NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_40G) - if_link->supported_caps |= SUPPORTED_40000baseLR4_Full; + NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_40G) + if_link->supported_caps |= QED_LM_40000baseLR4_Full_BIT; if (link_caps.speed_capabilities & - NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_50G) - if_link->supported_caps |= 0; + NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_50G) + if_link->supported_caps |= QED_LM_50000baseKR2_Full_BIT; if (link_caps.speed_capabilities & NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_BB_100G) - if_link->supported_caps |= 0; + if_link->supported_caps |= QED_LM_100000baseKR4_Full_BIT; if (link.link_up) if_link->speed = link.speed; @@ -1231,33 +1246,29 @@ static void qed_fill_link(struct qed_hwfn *hwfn, if_link->pause_config |= QED_LINK_PAUSE_TX_ENABLE; /* Link partner capabilities */ - if (link.partner_adv_speed & - QED_LINK_PARTNER_SPEED_1G_HD) - if_link->lp_caps |= SUPPORTED_1000baseT_Half; - if (link.partner_adv_speed & - QED_LINK_PARTNER_SPEED_1G_FD) - if_link->lp_caps |= SUPPORTED_1000baseT_Full; - if (link.partner_adv_speed & - QED_LINK_PARTNER_SPEED_10G) - if_link->lp_caps |= SUPPORTED_10000baseKR_Full; - if (link.partner_adv_speed & - QED_LINK_PARTNER_SPEED_40G) - if_link->lp_caps |= SUPPORTED_40000baseLR4_Full; - if (link.partner_adv_speed & - QED_LINK_PARTNER_SPEED_50G) - if_link->lp_caps |= 0; - if (link.partner_adv_speed & - QED_LINK_PARTNER_SPEED_100G) - if_link->lp_caps |= 0; + if (link.partner_adv_speed & QED_LINK_PARTNER_SPEED_1G_HD) + if_link->lp_caps |= QED_LM_1000baseT_Half_BIT; + if (link.partner_adv_speed & QED_LINK_PARTNER_SPEED_1G_FD) + if_link->lp_caps |= QED_LM_1000baseT_Full_BIT; + if (link.partner_adv_speed & QED_LINK_PARTNER_SPEED_10G) + if_link->lp_caps |= QED_LM_10000baseKR_Full_BIT; + if (link.partner_adv_speed & QED_LINK_PARTNER_SPEED_25G) + if_link->lp_caps |= QED_LM_25000baseKR_Full_BIT; + if (link.partner_adv_speed & QED_LINK_PARTNER_SPEED_40G) + if_link->lp_caps |= QED_LM_40000baseLR4_Full_BIT; + if (link.partner_adv_speed & QED_LINK_PARTNER_SPEED_50G) + if_link->lp_caps |= QED_LM_50000baseKR2_Full_BIT; + if (link.partner_adv_speed & QED_LINK_PARTNER_SPEED_100G) + if_link->lp_caps |= QED_LM_100000baseKR4_Full_BIT; if (link.an_complete) - if_link->lp_caps |= SUPPORTED_Autoneg; + if_link->lp_caps |= QED_LM_Autoneg_BIT; if (link.partner_adv_pause) - if_link->lp_caps |= SUPPORTED_Pause; + if_link->lp_caps |= QED_LM_Pause_BIT; if (link.partner_adv_pause == QED_LINK_PARTNER_ASYMMETRIC_PAUSE || link.partner_adv_pause == QED_LINK_PARTNER_BOTH_PAUSE) - if_link->lp_caps |= SUPPORTED_Asym_Pause; + if_link->lp_caps |= QED_LM_Asym_Pause_BIT; } static void qed_get_current_link(struct qed_dev *cdev, diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c index a240f26344a4..88b448bb2949 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c +++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c @@ -54,8 +54,7 @@ bool qed_mcp_is_init(struct qed_hwfn *p_hwfn) return true; } -void qed_mcp_cmd_port_init(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) +void qed_mcp_cmd_port_init(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { u32 addr = SECTION_OFFSIZE_ADDR(p_hwfn->mcp_info->public_base, PUBLIC_PORT); @@ -68,8 +67,7 @@ void qed_mcp_cmd_port_init(struct qed_hwfn *p_hwfn, p_hwfn->mcp_info->port_addr, MFW_PORT(p_hwfn)); } -void qed_mcp_read_mb(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) +void qed_mcp_read_mb(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { u32 length = MFW_DRV_MSG_MAX_DWORDS(p_hwfn->mcp_info->mfw_mb_length); u32 tmp, i; @@ -99,8 +97,7 @@ int qed_mcp_free(struct qed_hwfn *p_hwfn) return 0; } -static int qed_load_mcp_offsets(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) +static int qed_load_mcp_offsets(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { struct qed_mcp_info *p_info = p_hwfn->mcp_info; u32 drv_mb_offsize, mfw_mb_offsize; @@ -143,8 +140,7 @@ static int qed_load_mcp_offsets(struct qed_hwfn *p_hwfn, return 0; } -int qed_mcp_cmd_init(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) +int qed_mcp_cmd_init(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { struct qed_mcp_info *p_info; u32 size; @@ -165,9 +161,7 @@ int qed_mcp_cmd_init(struct qed_hwfn *p_hwfn, size = MFW_DRV_MSG_MAX_DWORDS(p_info->mfw_mb_length) * sizeof(u32); p_info->mfw_mb_cur = kzalloc(size, GFP_KERNEL); - p_info->mfw_mb_shadow = - kzalloc(sizeof(u32) * MFW_DRV_MSG_MAX_DWORDS( - p_info->mfw_mb_length), GFP_KERNEL); + p_info->mfw_mb_shadow = kzalloc(size, GFP_KERNEL); if (!p_info->mfw_mb_shadow || !p_info->mfw_mb_addr) goto err; @@ -189,8 +183,7 @@ err: * access is achieved by setting a blocking flag, which will fail other * competing contexts to send their mailboxes. */ -static int qed_mcp_mb_lock(struct qed_hwfn *p_hwfn, - u32 cmd) +static int qed_mcp_mb_lock(struct qed_hwfn *p_hwfn, u32 cmd) { spin_lock_bh(&p_hwfn->mcp_info->lock); @@ -221,15 +214,13 @@ static int qed_mcp_mb_lock(struct qed_hwfn *p_hwfn, return 0; } -static void qed_mcp_mb_unlock(struct qed_hwfn *p_hwfn, - u32 cmd) +static void qed_mcp_mb_unlock(struct qed_hwfn *p_hwfn, u32 cmd) { if (cmd != DRV_MSG_CODE_LOAD_REQ && cmd != DRV_MSG_CODE_UNLOAD_REQ) spin_unlock_bh(&p_hwfn->mcp_info->lock); } -int qed_mcp_reset(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) +int qed_mcp_reset(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { u32 seq = ++p_hwfn->mcp_info->drv_mb_seq; u8 delay = CHIP_MCP_RESP_ITER_US; @@ -326,7 +317,8 @@ static int qed_do_mcp_cmd(struct qed_hwfn *p_hwfn, *o_mcp_param = DRV_MB_RD(p_hwfn, p_ptt, fw_mb_param); } else { /* FW BUG! */ - DP_ERR(p_hwfn, "MFW failed to respond!\n"); + DP_ERR(p_hwfn, "MFW failed to respond [cmd 0x%x param 0x%x]\n", + cmd, param); *o_mcp_resp = 0; rc = -EAGAIN; } @@ -342,7 +334,7 @@ static int qed_mcp_cmd_and_union(struct qed_hwfn *p_hwfn, /* MCP not initialized */ if (!qed_mcp_is_init(p_hwfn)) { - DP_NOTICE(p_hwfn, "MFW is not initialized !\n"); + DP_NOTICE(p_hwfn, "MFW is not initialized!\n"); return -EBUSY; } @@ -399,8 +391,7 @@ int qed_mcp_cmd(struct qed_hwfn *p_hwfn, } int qed_mcp_load_req(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - u32 *p_load_code) + struct qed_ptt *p_ptt, u32 *p_load_code) { struct qed_dev *cdev = p_hwfn->cdev; struct qed_mcp_mb_params mb_params; @@ -527,8 +518,7 @@ static void qed_mcp_handle_transceiver_change(struct qed_hwfn *p_hwfn, "Received transceiver state update [0x%08x] from mfw [Addr 0x%x]\n", transceiver_state, (u32)(p_hwfn->mcp_info->port_addr + - offsetof(struct public_port, - transceiver_data))); + offsetof(struct public_port, transceiver_data))); transceiver_state = GET_FIELD(transceiver_state, ETH_TRANSCEIVER_STATE); @@ -540,8 +530,7 @@ static void qed_mcp_handle_transceiver_change(struct qed_hwfn *p_hwfn, } static void qed_mcp_handle_link_change(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - bool b_reset) + struct qed_ptt *p_ptt, bool b_reset) { struct qed_mcp_link_state *p_link; u8 max_bw, min_bw; @@ -557,8 +546,7 @@ static void qed_mcp_handle_link_change(struct qed_hwfn *p_hwfn, "Received link update [0x%08x] from mfw [Addr 0x%x]\n", status, (u32)(p_hwfn->mcp_info->port_addr + - offsetof(struct public_port, - link_status))); + offsetof(struct public_port, link_status))); } else { DP_VERBOSE(p_hwfn, NETIF_MSG_LINK, "Resetting link indications\n"); @@ -635,6 +623,9 @@ static void qed_mcp_handle_link_change(struct qed_hwfn *p_hwfn, (status & LINK_STATUS_LINK_PARTNER_20G_CAPABLE) ? QED_LINK_PARTNER_SPEED_20G : 0; p_link->partner_adv_speed |= + (status & LINK_STATUS_LINK_PARTNER_25G_CAPABLE) ? + QED_LINK_PARTNER_SPEED_25G : 0; + p_link->partner_adv_speed |= (status & LINK_STATUS_LINK_PARTNER_40G_CAPABLE) ? QED_LINK_PARTNER_SPEED_40G : 0; p_link->partner_adv_speed |= @@ -752,8 +743,7 @@ static void qed_read_pf_bandwidth(struct qed_hwfn *p_hwfn, static u32 qed_mcp_get_shmem_func(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, - struct public_func *p_data, - int pfid) + struct public_func *p_data, int pfid) { u32 addr = SECTION_OFFSIZE_ADDR(p_hwfn->mcp_info->public_base, PUBLIC_FUNC); @@ -763,8 +753,7 @@ static u32 qed_mcp_get_shmem_func(struct qed_hwfn *p_hwfn, memset(p_data, 0, sizeof(*p_data)); - size = min_t(u32, sizeof(*p_data), - QED_SECTION_SIZE(mfw_path_offsize)); + size = min_t(u32, sizeof(*p_data), QED_SECTION_SIZE(mfw_path_offsize)); for (i = 0; i < size / sizeof(u32); i++) ((u32 *)p_data)[i] = qed_rd(p_hwfn, p_ptt, func_addr + (i << 2)); @@ -799,15 +788,13 @@ int qed_hw_init_first_eth(struct qed_hwfn *p_hwfn, return -EINVAL; } -static void qed_mcp_update_bw(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) +static void qed_mcp_update_bw(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { struct qed_mcp_function_info *p_info; struct public_func shmem_info; u32 resp = 0, param = 0; - qed_mcp_get_shmem_func(p_hwfn, p_ptt, &shmem_info, - MCP_PF_ID(p_hwfn)); + qed_mcp_get_shmem_func(p_hwfn, p_ptt, &shmem_info, MCP_PF_ID(p_hwfn)); qed_read_pf_bandwidth(p_hwfn, &shmem_info); @@ -940,8 +927,7 @@ int qed_mcp_get_mfw_ver(struct qed_hwfn *p_hwfn, return 0; } -int qed_mcp_get_media_type(struct qed_dev *cdev, - u32 *p_media_type) +int qed_mcp_get_media_type(struct qed_dev *cdev, u32 *p_media_type) { struct qed_hwfn *p_hwfn = &cdev->hwfns[0]; struct qed_ptt *p_ptt; @@ -950,7 +936,7 @@ int qed_mcp_get_media_type(struct qed_dev *cdev, return -EINVAL; if (!qed_mcp_is_init(p_hwfn)) { - DP_NOTICE(p_hwfn, "MFW is not initialized !\n"); + DP_NOTICE(p_hwfn, "MFW is not initialized!\n"); return -EBUSY; } @@ -1003,15 +989,13 @@ int qed_mcp_fill_shmem_func_info(struct qed_hwfn *p_hwfn, struct qed_mcp_function_info *info; struct public_func shmem_info; - qed_mcp_get_shmem_func(p_hwfn, p_ptt, &shmem_info, - MCP_PF_ID(p_hwfn)); + qed_mcp_get_shmem_func(p_hwfn, p_ptt, &shmem_info, MCP_PF_ID(p_hwfn)); info = &p_hwfn->mcp_info->func_info; info->pause_on_host = (shmem_info.config & FUNC_MF_CFG_PAUSE_ON_HOST_RING) ? 1 : 0; - if (qed_mcp_get_shmem_proto(p_hwfn, &shmem_info, - &info->protocol)) { + if (qed_mcp_get_shmem_proto(p_hwfn, &shmem_info, &info->protocol)) { DP_ERR(p_hwfn, "Unknown personality %08x\n", (u32)(shmem_info.config & FUNC_MF_CFG_PROTOCOL_MASK)); return -EINVAL; @@ -1072,15 +1056,13 @@ struct qed_mcp_link_capabilities return &p_hwfn->mcp_info->link_capabilities; } -int qed_mcp_drain(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) +int qed_mcp_drain(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { u32 resp = 0, param = 0; int rc; rc = qed_mcp_cmd(p_hwfn, p_ptt, - DRV_MSG_CODE_NIG_DRAIN, 1000, - &resp, ¶m); + DRV_MSG_CODE_NIG_DRAIN, 1000, &resp, ¶m); /* Wait for the drain to complete before returning */ msleep(1020); @@ -1089,8 +1071,7 @@ int qed_mcp_drain(struct qed_hwfn *p_hwfn, } int qed_mcp_get_flash_size(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - u32 *p_flash_size) + struct qed_ptt *p_ptt, u32 *p_flash_size) { u32 flash_size; @@ -1168,8 +1149,8 @@ qed_mcp_send_drv_version(struct qed_hwfn *p_hwfn, return rc; } -int qed_mcp_set_led(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, - enum qed_led_mode mode) +int qed_mcp_set_led(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, enum qed_led_mode mode) { u32 resp = 0, param = 0, drv_mb_param; int rc; diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.h b/drivers/net/ethernet/qlogic/qed/qed_mcp.h index 7f319aa1b229..013d1b92ed63 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_mcp.h +++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.h @@ -60,9 +60,10 @@ struct qed_mcp_link_state { #define QED_LINK_PARTNER_SPEED_1G_FD BIT(1) #define QED_LINK_PARTNER_SPEED_10G BIT(2) #define QED_LINK_PARTNER_SPEED_20G BIT(3) -#define QED_LINK_PARTNER_SPEED_40G BIT(4) -#define QED_LINK_PARTNER_SPEED_50G BIT(5) -#define QED_LINK_PARTNER_SPEED_100G BIT(6) +#define QED_LINK_PARTNER_SPEED_25G BIT(4) +#define QED_LINK_PARTNER_SPEED_40G BIT(5) +#define QED_LINK_PARTNER_SPEED_50G BIT(6) +#define QED_LINK_PARTNER_SPEED_100G BIT(7) u32 partner_adv_speed; bool partner_tx_flow_ctrl_en; diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c index a52f3fc051f5..2888eb0628f8 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c @@ -25,9 +25,7 @@ int qed_sp_init_request(struct qed_hwfn *p_hwfn, struct qed_spq_entry **pp_ent, - u8 cmd, - u8 protocol, - struct qed_sp_init_data *p_data) + u8 cmd, u8 protocol, struct qed_sp_init_data *p_data) { u32 opaque_cid = p_data->opaque_fid << 16 | p_data->cid; struct qed_spq_entry *p_ent = NULL; @@ -38,7 +36,7 @@ int qed_sp_init_request(struct qed_hwfn *p_hwfn, rc = qed_spq_get_entry(p_hwfn, pp_ent); - if (rc != 0) + if (rc) return rc; p_ent = *pp_ent; @@ -321,8 +319,7 @@ int qed_sp_pf_start(struct qed_hwfn *p_hwfn, rc = qed_sp_init_request(p_hwfn, &p_ent, COMMON_RAMROD_PF_START, - PROTOCOLID_COMMON, - &init_data); + PROTOCOLID_COMMON, &init_data); if (rc) return rc; @@ -356,8 +353,7 @@ int qed_sp_pf_start(struct qed_hwfn *p_hwfn, DMA_REGPAIR_LE(p_ramrod->consolid_q_pbl_addr, p_hwfn->p_consq->chain.pbl.p_phys_table); - qed_tunn_set_pf_start_params(p_hwfn, p_tunn, - &p_ramrod->tunnel_config); + qed_tunn_set_pf_start_params(p_hwfn, p_tunn, &p_ramrod->tunnel_config); if (IS_MF_SI(p_hwfn)) p_ramrod->allow_npar_tx_switching = allow_npar_tx_switch; @@ -389,8 +385,7 @@ int qed_sp_pf_start(struct qed_hwfn *p_hwfn, DP_VERBOSE(p_hwfn, QED_MSG_SPQ, "Setting event_ring_sb [id %04x index %02x], outer_tag [%d]\n", - sb, sb_index, - p_ramrod->outer_tag); + sb, sb_index, p_ramrod->outer_tag); rc = qed_spq_post(p_hwfn, p_ent, NULL); diff --git a/drivers/net/ethernet/qlogic/qed/qed_spq.c b/drivers/net/ethernet/qlogic/qed/qed_spq.c index d73456eab1d7..0265a32c8681 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_spq.c +++ b/drivers/net/ethernet/qlogic/qed/qed_spq.c @@ -41,8 +41,7 @@ ***************************************************************************/ static void qed_spq_blocking_cb(struct qed_hwfn *p_hwfn, void *cookie, - union event_ring_data *data, - u8 fw_return_code) + union event_ring_data *data, u8 fw_return_code) { struct qed_spq_comp_done *comp_done; @@ -109,9 +108,8 @@ static int qed_spq_block(struct qed_hwfn *p_hwfn, /*************************************************************************** * SPQ entries inner API ***************************************************************************/ -static int -qed_spq_fill_entry(struct qed_hwfn *p_hwfn, - struct qed_spq_entry *p_ent) +static int qed_spq_fill_entry(struct qed_hwfn *p_hwfn, + struct qed_spq_entry *p_ent) { p_ent->flags = 0; @@ -189,8 +187,7 @@ static void qed_spq_hw_initialize(struct qed_hwfn *p_hwfn, } static int qed_spq_hw_post(struct qed_hwfn *p_hwfn, - struct qed_spq *p_spq, - struct qed_spq_entry *p_ent) + struct qed_spq *p_spq, struct qed_spq_entry *p_ent) { struct qed_chain *p_chain = &p_hwfn->p_spq->chain; u16 echo = qed_chain_get_prod_idx(p_chain); @@ -255,8 +252,7 @@ qed_async_event_completion(struct qed_hwfn *p_hwfn, /*************************************************************************** * EQ API ***************************************************************************/ -void qed_eq_prod_update(struct qed_hwfn *p_hwfn, - u16 prod) +void qed_eq_prod_update(struct qed_hwfn *p_hwfn, u16 prod) { u32 addr = GTT_BAR0_MAP_REG_USDM_RAM + USTORM_EQE_CONS_OFFSET(p_hwfn->rel_pf_id); @@ -267,9 +263,7 @@ void qed_eq_prod_update(struct qed_hwfn *p_hwfn, mmiowb(); } -int qed_eq_completion(struct qed_hwfn *p_hwfn, - void *cookie) - +int qed_eq_completion(struct qed_hwfn *p_hwfn, void *cookie) { struct qed_eq *p_eq = cookie; struct qed_chain *p_chain = &p_eq->chain; @@ -323,8 +317,7 @@ int qed_eq_completion(struct qed_hwfn *p_hwfn, return rc; } -struct qed_eq *qed_eq_alloc(struct qed_hwfn *p_hwfn, - u16 num_elem) +struct qed_eq *qed_eq_alloc(struct qed_hwfn *p_hwfn, u16 num_elem) { struct qed_eq *p_eq; @@ -348,11 +341,8 @@ struct qed_eq *qed_eq_alloc(struct qed_hwfn *p_hwfn, } /* register EQ completion on the SP SB */ - qed_int_register_cb(p_hwfn, - qed_eq_completion, - p_eq, - &p_eq->eq_sb_index, - &p_eq->p_fw_cons); + qed_int_register_cb(p_hwfn, qed_eq_completion, + p_eq, &p_eq->eq_sb_index, &p_eq->p_fw_cons); return p_eq; @@ -361,14 +351,12 @@ eq_allocate_fail: return NULL; } -void qed_eq_setup(struct qed_hwfn *p_hwfn, - struct qed_eq *p_eq) +void qed_eq_setup(struct qed_hwfn *p_hwfn, struct qed_eq *p_eq) { qed_chain_reset(&p_eq->chain); } -void qed_eq_free(struct qed_hwfn *p_hwfn, - struct qed_eq *p_eq) +void qed_eq_free(struct qed_hwfn *p_hwfn, struct qed_eq *p_eq) { if (!p_eq) return; @@ -379,10 +367,9 @@ void qed_eq_free(struct qed_hwfn *p_hwfn, /*************************************************************************** * CQE API - manipulate EQ functionality ***************************************************************************/ -static int qed_cqe_completion( - struct qed_hwfn *p_hwfn, - struct eth_slow_path_rx_cqe *cqe, - enum protocol_type protocol) +static int qed_cqe_completion(struct qed_hwfn *p_hwfn, + struct eth_slow_path_rx_cqe *cqe, + enum protocol_type protocol) { if (IS_VF(p_hwfn->cdev)) return 0; @@ -463,8 +450,7 @@ int qed_spq_alloc(struct qed_hwfn *p_hwfn) u32 capacity; /* SPQ struct */ - p_spq = - kzalloc(sizeof(struct qed_spq), GFP_KERNEL); + p_spq = kzalloc(sizeof(struct qed_spq), GFP_KERNEL); if (!p_spq) { DP_NOTICE(p_hwfn, "Failed to allocate `struct qed_spq'\n"); return -ENOMEM; @@ -525,9 +511,7 @@ void qed_spq_free(struct qed_hwfn *p_hwfn) kfree(p_spq); } -int -qed_spq_get_entry(struct qed_hwfn *p_hwfn, - struct qed_spq_entry **pp_ent) +int qed_spq_get_entry(struct qed_hwfn *p_hwfn, struct qed_spq_entry **pp_ent) { struct qed_spq *p_spq = p_hwfn->p_spq; struct qed_spq_entry *p_ent = NULL; @@ -538,14 +522,15 @@ qed_spq_get_entry(struct qed_hwfn *p_hwfn, if (list_empty(&p_spq->free_pool)) { p_ent = kzalloc(sizeof(*p_ent), GFP_ATOMIC); if (!p_ent) { + DP_NOTICE(p_hwfn, + "Failed to allocate an SPQ entry for a pending ramrod\n"); rc = -ENOMEM; goto out_unlock; } p_ent->queue = &p_spq->unlimited_pending; } else { p_ent = list_first_entry(&p_spq->free_pool, - struct qed_spq_entry, - list); + struct qed_spq_entry, list); list_del(&p_ent->list); p_ent->queue = &p_spq->pending; } @@ -564,8 +549,7 @@ static void __qed_spq_return_entry(struct qed_hwfn *p_hwfn, list_add_tail(&p_ent->list, &p_hwfn->p_spq->free_pool); } -void qed_spq_return_entry(struct qed_hwfn *p_hwfn, - struct qed_spq_entry *p_ent) +void qed_spq_return_entry(struct qed_hwfn *p_hwfn, struct qed_spq_entry *p_ent) { spin_lock_bh(&p_hwfn->p_spq->lock); __qed_spq_return_entry(p_hwfn, p_ent); @@ -586,10 +570,9 @@ void qed_spq_return_entry(struct qed_hwfn *p_hwfn, * * @return int */ -static int -qed_spq_add_entry(struct qed_hwfn *p_hwfn, - struct qed_spq_entry *p_ent, - enum spq_priority priority) +static int qed_spq_add_entry(struct qed_hwfn *p_hwfn, + struct qed_spq_entry *p_ent, + enum spq_priority priority) { struct qed_spq *p_spq = p_hwfn->p_spq; @@ -604,8 +587,7 @@ qed_spq_add_entry(struct qed_hwfn *p_hwfn, struct qed_spq_entry *p_en2; p_en2 = list_first_entry(&p_spq->free_pool, - struct qed_spq_entry, - list); + struct qed_spq_entry, list); list_del(&p_en2->list); /* Copy the ring element physical pointer to the new @@ -655,8 +637,7 @@ u32 qed_spq_get_cid(struct qed_hwfn *p_hwfn) * Posting new Ramrods ***************************************************************************/ static int qed_spq_post_list(struct qed_hwfn *p_hwfn, - struct list_head *head, - u32 keep_reserve) + struct list_head *head, u32 keep_reserve) { struct qed_spq *p_spq = p_hwfn->p_spq; int rc; @@ -690,8 +671,7 @@ static int qed_spq_pend_post(struct qed_hwfn *p_hwfn) break; p_ent = list_first_entry(&p_spq->unlimited_pending, - struct qed_spq_entry, - list); + struct qed_spq_entry, list); if (!p_ent) return -EINVAL; @@ -705,8 +685,7 @@ static int qed_spq_pend_post(struct qed_hwfn *p_hwfn) } int qed_spq_post(struct qed_hwfn *p_hwfn, - struct qed_spq_entry *p_ent, - u8 *fw_return_code) + struct qed_spq_entry *p_ent, u8 *fw_return_code) { int rc = 0; struct qed_spq *p_spq = p_hwfn ? p_hwfn->p_spq : NULL; @@ -803,8 +782,7 @@ int qed_spq_completion(struct qed_hwfn *p_hwfn, return -EINVAL; spin_lock_bh(&p_spq->lock); - list_for_each_entry_safe(p_ent, tmp, &p_spq->completion_pending, - list) { + list_for_each_entry_safe(p_ent, tmp, &p_spq->completion_pending, list) { if (p_ent->elem.hdr.echo == echo) { u16 pos = le16_to_cpu(echo) % SPQ_RING_SIZE; @@ -846,15 +824,22 @@ int qed_spq_completion(struct qed_hwfn *p_hwfn, if (!found) { DP_NOTICE(p_hwfn, - "Failed to find an entry this EQE completes\n"); + "Failed to find an entry this EQE [echo %04x] completes\n", + le16_to_cpu(echo)); return -EEXIST; } - DP_VERBOSE(p_hwfn, QED_MSG_SPQ, "Complete: func %p cookie %p)\n", + DP_VERBOSE(p_hwfn, QED_MSG_SPQ, + "Complete EQE [echo %04x]: func %p cookie %p)\n", + le16_to_cpu(echo), p_ent->comp_cb.function, p_ent->comp_cb.cookie); if (found->comp_cb.function) found->comp_cb.function(p_hwfn, found->comp_cb.cookie, p_data, fw_return_code); + else + DP_VERBOSE(p_hwfn, + QED_MSG_SPQ, + "Got a completion without a callback function\n"); if ((found->comp_mode != QED_SPQ_MODE_EBLOCK) || (found->queue == &p_spq->unlimited_pending)) @@ -901,14 +886,12 @@ consq_allocate_fail: return NULL; } -void qed_consq_setup(struct qed_hwfn *p_hwfn, - struct qed_consq *p_consq) +void qed_consq_setup(struct qed_hwfn *p_hwfn, struct qed_consq *p_consq) { qed_chain_reset(&p_consq->chain); } -void qed_consq_free(struct qed_hwfn *p_hwfn, - struct qed_consq *p_consq) +void qed_consq_free(struct qed_hwfn *p_hwfn, struct qed_consq *p_consq) { if (!p_consq) return; diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.c b/drivers/net/ethernet/qlogic/qed/qed_sriov.c index 15399da268d9..51e4c906833f 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.c @@ -699,7 +699,7 @@ static void qed_iov_config_perm_table(struct qed_hwfn *p_hwfn, &qzone_id); reg_addr = PSWHST_REG_ZONE_PERMISSION_TABLE + qzone_id * 4; - val = enable ? (vf->abs_vf_id | (1 << 8)) : 0; + val = enable ? (vf->abs_vf_id | BIT(8)) : 0; qed_wr(p_hwfn, p_ptt, reg_addr, val); } } @@ -1090,13 +1090,13 @@ static u16 qed_iov_prep_vp_update_resp_tlvs(struct qed_hwfn *p_hwfn, /* Prepare response for all extended tlvs if they are found by PF */ for (i = 0; i < QED_IOV_VP_UPDATE_MAX; i++) { - if (!(tlvs_mask & (1 << i))) + if (!(tlvs_mask & BIT(i))) continue; resp = qed_add_tlv(p_hwfn, &p_mbx->offset, qed_iov_vport_to_tlv(p_hwfn, i), size); - if (tlvs_accepted & (1 << i)) + if (tlvs_accepted & BIT(i)) resp->hdr.status = status; else resp->hdr.status = PFVF_STATUS_NOT_SUPPORTED; @@ -1334,8 +1334,7 @@ static void qed_iov_vf_mbx_acquire(struct qed_hwfn *p_hwfn, pfdev_info->fw_minor = FW_MINOR_VERSION; pfdev_info->fw_rev = FW_REVISION_VERSION; pfdev_info->fw_eng = FW_ENGINEERING_VERSION; - pfdev_info->minor_fp_hsi = min_t(u8, - ETH_HSI_VER_MINOR, + pfdev_info->minor_fp_hsi = min_t(u8, ETH_HSI_VER_MINOR, req->vfdev_info.eth_fp_hsi_minor); pfdev_info->os_type = VFPF_ACQUIRE_OS_LINUX; qed_mcp_get_mfw_ver(p_hwfn, p_ptt, &pfdev_info->mfw_ver, NULL); @@ -1438,14 +1437,11 @@ static int qed_iov_reconfigure_unicast_vlan(struct qed_hwfn *p_hwfn, filter.type = QED_FILTER_VLAN; filter.vlan = p_vf->shadow_config.vlans[i].vid; - DP_VERBOSE(p_hwfn, - QED_MSG_IOV, + DP_VERBOSE(p_hwfn, QED_MSG_IOV, "Reconfiguring VLAN [0x%04x] for VF [%04x]\n", filter.vlan, p_vf->relative_vf_id); - rc = qed_sp_eth_filter_ucast(p_hwfn, - p_vf->opaque_fid, - &filter, - QED_SPQ_MODE_CB, NULL); + rc = qed_sp_eth_filter_ucast(p_hwfn, p_vf->opaque_fid, + &filter, QED_SPQ_MODE_CB, NULL); if (rc) { DP_NOTICE(p_hwfn, "Failed to configure VLAN [%04x] to VF [%04x]\n", @@ -1463,7 +1459,7 @@ qed_iov_reconfigure_unicast_shadow(struct qed_hwfn *p_hwfn, { int rc = 0; - if ((events & (1 << VLAN_ADDR_FORCED)) && + if ((events & BIT(VLAN_ADDR_FORCED)) && !(p_vf->configured_features & (1 << VLAN_ADDR_FORCED))) rc = qed_iov_reconfigure_unicast_vlan(p_hwfn, p_vf); @@ -1479,7 +1475,7 @@ static int qed_iov_configure_vport_forced(struct qed_hwfn *p_hwfn, if (!p_vf->vport_instance) return -EINVAL; - if (events & (1 << MAC_ADDR_FORCED)) { + if (events & BIT(MAC_ADDR_FORCED)) { /* Since there's no way [currently] of removing the MAC, * we can always assume this means we need to force it. */ @@ -1502,7 +1498,7 @@ static int qed_iov_configure_vport_forced(struct qed_hwfn *p_hwfn, p_vf->configured_features |= 1 << MAC_ADDR_FORCED; } - if (events & (1 << VLAN_ADDR_FORCED)) { + if (events & BIT(VLAN_ADDR_FORCED)) { struct qed_sp_vport_update_params vport_update; u8 removal; int i; @@ -1572,7 +1568,7 @@ static int qed_iov_configure_vport_forced(struct qed_hwfn *p_hwfn, if (filter.vlan) p_vf->configured_features |= 1 << VLAN_ADDR_FORCED; else - p_vf->configured_features &= ~(1 << VLAN_ADDR_FORCED); + p_vf->configured_features &= ~BIT(VLAN_ADDR_FORCED); } /* If forced features are terminated, we need to configure the shadow @@ -1619,8 +1615,7 @@ static void qed_iov_vf_mbx_start_vport(struct qed_hwfn *p_hwfn, qed_int_cau_conf_sb(p_hwfn, p_ptt, start->sb_addr[sb_id], - vf->igu_sbs[sb_id], - vf->abs_vf_id, 1); + vf->igu_sbs[sb_id], vf->abs_vf_id, 1); } qed_iov_enable_vf_traffic(p_hwfn, p_ptt, vf); @@ -1632,7 +1627,7 @@ static void qed_iov_vf_mbx_start_vport(struct qed_hwfn *p_hwfn, * vfs that would still be fine, since they passed '0' as padding]. */ p_bitmap = &vf_info->bulletin.p_virt->valid_bitmap; - if (!(*p_bitmap & (1 << VFPF_BULLETIN_UNTAGGED_DEFAULT_FORCED))) { + if (!(*p_bitmap & BIT(VFPF_BULLETIN_UNTAGGED_DEFAULT_FORCED))) { u8 vf_req = start->only_untagged; vf_info->bulletin.p_virt->default_only_untagged = vf_req; @@ -1652,7 +1647,7 @@ static void qed_iov_vf_mbx_start_vport(struct qed_hwfn *p_hwfn, params.mtu = vf->mtu; rc = qed_sp_eth_vport_start(p_hwfn, ¶ms); - if (rc != 0) { + if (rc) { DP_ERR(p_hwfn, "qed_iov_vf_mbx_start_vport returned error %d\n", rc); status = PFVF_STATUS_FAILURE; @@ -1679,7 +1674,7 @@ static void qed_iov_vf_mbx_stop_vport(struct qed_hwfn *p_hwfn, vf->spoof_chk = false; rc = qed_sp_vport_stop(p_hwfn, vf->opaque_fid, vf->vport_id); - if (rc != 0) { + if (rc) { DP_ERR(p_hwfn, "qed_iov_vf_mbx_stop_vport returned error %d\n", rc); status = PFVF_STATUS_FAILURE; @@ -2045,7 +2040,7 @@ qed_iov_vp_update_vlan_param(struct qed_hwfn *p_hwfn, p_vf->shadow_config.inner_vlan_removal = p_vlan_tlv->remove_vlan; /* Ignore the VF request if we're forcing a vlan */ - if (!(p_vf->configured_features & (1 << VLAN_ADDR_FORCED))) { + if (!(p_vf->configured_features & BIT(VLAN_ADDR_FORCED))) { p_data->update_inner_vlan_removal_flg = 1; p_data->inner_vlan_removal_flg = p_vlan_tlv->remove_vlan; } @@ -2340,7 +2335,7 @@ static int qed_iov_vf_update_vlan_shadow(struct qed_hwfn *p_hwfn, /* In forced mode, we're willing to remove entries - but we don't add * new ones. */ - if (p_vf->bulletin.p_virt->valid_bitmap & (1 << VLAN_ADDR_FORCED)) + if (p_vf->bulletin.p_virt->valid_bitmap & BIT(VLAN_ADDR_FORCED)) return 0; if (p_params->opcode == QED_FILTER_ADD || @@ -2374,7 +2369,7 @@ static int qed_iov_vf_update_mac_shadow(struct qed_hwfn *p_hwfn, int i; /* If we're in forced-mode, we don't allow any change */ - if (p_vf->bulletin.p_virt->valid_bitmap & (1 << MAC_ADDR_FORCED)) + if (p_vf->bulletin.p_virt->valid_bitmap & BIT(MAC_ADDR_FORCED)) return 0; /* First remove entries and then add new ones */ @@ -2509,7 +2504,7 @@ static void qed_iov_vf_mbx_ucast_filter(struct qed_hwfn *p_hwfn, } /* Determine if the unicast filtering is acceptible by PF */ - if ((p_bulletin->valid_bitmap & (1 << VLAN_ADDR_FORCED)) && + if ((p_bulletin->valid_bitmap & BIT(VLAN_ADDR_FORCED)) && (params.type == QED_FILTER_VLAN || params.type == QED_FILTER_MAC_VLAN)) { /* Once VLAN is forced or PVID is set, do not allow @@ -2521,7 +2516,7 @@ static void qed_iov_vf_mbx_ucast_filter(struct qed_hwfn *p_hwfn, goto out; } - if ((p_bulletin->valid_bitmap & (1 << MAC_ADDR_FORCED)) && + if ((p_bulletin->valid_bitmap & BIT(MAC_ADDR_FORCED)) && (params.type == QED_FILTER_MAC || params.type == QED_FILTER_MAC_VLAN)) { if (!ether_addr_equal(p_bulletin->mac, params.mac) || @@ -2749,7 +2744,7 @@ cleanup: /* Mark VF for ack and clean pending state */ if (p_vf->state == VF_RESET) p_vf->state = VF_STOPPED; - ack_vfs[vfid / 32] |= (1 << (vfid % 32)); + ack_vfs[vfid / 32] |= BIT((vfid % 32)); p_hwfn->pf_iov_info->pending_flr[rel_vf_id / 64] &= ~(1ULL << (rel_vf_id % 64)); p_hwfn->pf_iov_info->pending_events[rel_vf_id / 64] &= @@ -2805,7 +2800,7 @@ int qed_iov_mark_vf_flr(struct qed_hwfn *p_hwfn, u32 *p_disabled_vfs) continue; vfid = p_vf->abs_vf_id; - if ((1 << (vfid % 32)) & p_disabled_vfs[vfid / 32]) { + if (BIT((vfid % 32)) & p_disabled_vfs[vfid / 32]) { u64 *p_flr = p_hwfn->pf_iov_info->pending_flr; u16 rel_vf_id = p_vf->relative_vf_id; @@ -3064,8 +3059,7 @@ static void qed_iov_bulletin_set_forced_mac(struct qed_hwfn *p_hwfn, vf_info->bulletin.p_virt->valid_bitmap |= feature; /* Forced MAC will disable MAC_ADDR */ - vf_info->bulletin.p_virt->valid_bitmap &= - ~(1 << VFPF_BULLETIN_MAC_ADDR); + vf_info->bulletin.p_virt->valid_bitmap &= ~BIT(VFPF_BULLETIN_MAC_ADDR); qed_iov_configure_vport_forced(p_hwfn, vf_info, feature); } @@ -3163,7 +3157,7 @@ static u8 *qed_iov_bulletin_get_forced_mac(struct qed_hwfn *p_hwfn, if (!p_vf || !p_vf->bulletin.p_virt) return NULL; - if (!(p_vf->bulletin.p_virt->valid_bitmap & (1 << MAC_ADDR_FORCED))) + if (!(p_vf->bulletin.p_virt->valid_bitmap & BIT(MAC_ADDR_FORCED))) return NULL; return p_vf->bulletin.p_virt->mac; @@ -3177,7 +3171,7 @@ u16 qed_iov_bulletin_get_forced_vlan(struct qed_hwfn *p_hwfn, u16 rel_vf_id) if (!p_vf || !p_vf->bulletin.p_virt) return 0; - if (!(p_vf->bulletin.p_virt->valid_bitmap & (1 << VLAN_ADDR_FORCED))) + if (!(p_vf->bulletin.p_virt->valid_bitmap & BIT(VLAN_ADDR_FORCED))) return 0; return p_vf->bulletin.p_virt->pvid; diff --git a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c index f8492cac9290..427e043a033f 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c +++ b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c @@ -249,78 +249,150 @@ static u32 qede_get_priv_flags(struct net_device *dev) return (!!(edev->dev_info.common.num_hwfns > 1)) << QEDE_PRI_FLAG_CMT; } -static int qede_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) +struct qede_link_mode_mapping { + u32 qed_link_mode; + u32 ethtool_link_mode; +}; + +static const struct qede_link_mode_mapping qed_lm_map[] = { + {QED_LM_FIBRE_BIT, ETHTOOL_LINK_MODE_FIBRE_BIT}, + {QED_LM_Autoneg_BIT, ETHTOOL_LINK_MODE_Autoneg_BIT}, + {QED_LM_Asym_Pause_BIT, ETHTOOL_LINK_MODE_Asym_Pause_BIT}, + {QED_LM_Pause_BIT, ETHTOOL_LINK_MODE_Pause_BIT}, + {QED_LM_1000baseT_Half_BIT, ETHTOOL_LINK_MODE_1000baseT_Half_BIT}, + {QED_LM_1000baseT_Full_BIT, ETHTOOL_LINK_MODE_1000baseT_Full_BIT}, + {QED_LM_10000baseKR_Full_BIT, ETHTOOL_LINK_MODE_10000baseKR_Full_BIT}, + {QED_LM_25000baseKR_Full_BIT, ETHTOOL_LINK_MODE_25000baseKR_Full_BIT}, + {QED_LM_40000baseLR4_Full_BIT, ETHTOOL_LINK_MODE_40000baseLR4_Full_BIT}, + {QED_LM_50000baseKR2_Full_BIT, ETHTOOL_LINK_MODE_50000baseKR2_Full_BIT}, + {QED_LM_100000baseKR4_Full_BIT, + ETHTOOL_LINK_MODE_100000baseKR4_Full_BIT}, +}; + +#define QEDE_DRV_TO_ETHTOOL_CAPS(caps, lk_ksettings, name) \ +{ \ + int i; \ + \ + for (i = 0; i < QED_LM_COUNT; i++) { \ + if ((caps) & (qed_lm_map[i].qed_link_mode)) \ + __set_bit(qed_lm_map[i].ethtool_link_mode,\ + lk_ksettings->link_modes.name); \ + } \ +} + +#define QEDE_ETHTOOL_TO_DRV_CAPS(caps, lk_ksettings, name) \ +{ \ + int i; \ + \ + for (i = 0; i < QED_LM_COUNT; i++) { \ + if (test_bit(qed_lm_map[i].ethtool_link_mode, \ + lk_ksettings->link_modes.name)) \ + caps |= qed_lm_map[i].qed_link_mode; \ + } \ +} + +static int qede_get_link_ksettings(struct net_device *dev, + struct ethtool_link_ksettings *cmd) { + struct ethtool_link_settings *base = &cmd->base; struct qede_dev *edev = netdev_priv(dev); struct qed_link_output current_link; memset(¤t_link, 0, sizeof(current_link)); edev->ops->common->get_link(edev->cdev, ¤t_link); - cmd->supported = current_link.supported_caps; - cmd->advertising = current_link.advertised_caps; + ethtool_link_ksettings_zero_link_mode(cmd, supported); + QEDE_DRV_TO_ETHTOOL_CAPS(current_link.supported_caps, cmd, supported) + + ethtool_link_ksettings_zero_link_mode(cmd, advertising); + QEDE_DRV_TO_ETHTOOL_CAPS(current_link.advertised_caps, cmd, advertising) + + ethtool_link_ksettings_zero_link_mode(cmd, lp_advertising); + QEDE_DRV_TO_ETHTOOL_CAPS(current_link.lp_caps, cmd, lp_advertising) + if ((edev->state == QEDE_STATE_OPEN) && (current_link.link_up)) { - ethtool_cmd_speed_set(cmd, current_link.speed); - cmd->duplex = current_link.duplex; + base->speed = current_link.speed; + base->duplex = current_link.duplex; } else { - cmd->duplex = DUPLEX_UNKNOWN; - ethtool_cmd_speed_set(cmd, SPEED_UNKNOWN); + base->speed = SPEED_UNKNOWN; + base->duplex = DUPLEX_UNKNOWN; } - cmd->port = current_link.port; - cmd->autoneg = (current_link.autoneg) ? AUTONEG_ENABLE : - AUTONEG_DISABLE; - cmd->lp_advertising = current_link.lp_caps; + base->port = current_link.port; + base->autoneg = (current_link.autoneg) ? AUTONEG_ENABLE : + AUTONEG_DISABLE; return 0; } -static int qede_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) +static int qede_set_link_ksettings(struct net_device *dev, + const struct ethtool_link_ksettings *cmd) { + const struct ethtool_link_settings *base = &cmd->base; struct qede_dev *edev = netdev_priv(dev); struct qed_link_output current_link; struct qed_link_params params; - u32 speed; if (!edev->ops || !edev->ops->common->can_link_change(edev->cdev)) { - DP_INFO(edev, - "Link settings are not allowed to be changed\n"); + DP_INFO(edev, "Link settings are not allowed to be changed\n"); return -EOPNOTSUPP; } - memset(¤t_link, 0, sizeof(current_link)); memset(¶ms, 0, sizeof(params)); edev->ops->common->get_link(edev->cdev, ¤t_link); - speed = ethtool_cmd_speed(cmd); params.override_flags |= QED_LINK_OVERRIDE_SPEED_ADV_SPEEDS; params.override_flags |= QED_LINK_OVERRIDE_SPEED_AUTONEG; - if (cmd->autoneg == AUTONEG_ENABLE) { + if (base->autoneg == AUTONEG_ENABLE) { params.autoneg = true; params.forced_speed = 0; - params.adv_speeds = cmd->advertising; - } else { /* forced speed */ + QEDE_ETHTOOL_TO_DRV_CAPS(params.adv_speeds, cmd, advertising) + } else { /* forced speed */ params.override_flags |= QED_LINK_OVERRIDE_SPEED_FORCED_SPEED; params.autoneg = false; - params.forced_speed = speed; - switch (speed) { + params.forced_speed = base->speed; + switch (base->speed) { case SPEED_10000: if (!(current_link.supported_caps & - SUPPORTED_10000baseKR_Full)) { + QED_LM_10000baseKR_Full_BIT)) { DP_INFO(edev, "10G speed not supported\n"); return -EINVAL; } - params.adv_speeds = SUPPORTED_10000baseKR_Full; + params.adv_speeds = QED_LM_10000baseKR_Full_BIT; + break; + case SPEED_25000: + if (!(current_link.supported_caps & + QED_LM_25000baseKR_Full_BIT)) { + DP_INFO(edev, "25G speed not supported\n"); + return -EINVAL; + } + params.adv_speeds = QED_LM_25000baseKR_Full_BIT; break; case SPEED_40000: if (!(current_link.supported_caps & - SUPPORTED_40000baseLR4_Full)) { + QED_LM_40000baseLR4_Full_BIT)) { DP_INFO(edev, "40G speed not supported\n"); return -EINVAL; } - params.adv_speeds = SUPPORTED_40000baseLR4_Full; + params.adv_speeds = QED_LM_40000baseLR4_Full_BIT; + break; + case 0xdead: + if (!(current_link.supported_caps & + QED_LM_50000baseKR2_Full_BIT)) { + DP_INFO(edev, "50G speed not supported\n"); + return -EINVAL; + } + params.adv_speeds = QED_LM_50000baseKR2_Full_BIT; + break; + case 0xbeef: + if (!(current_link.supported_caps & + QED_LM_100000baseKR4_Full_BIT)) { + DP_INFO(edev, "100G speed not supported\n"); + return -EINVAL; + } + params.adv_speeds = QED_LM_100000baseKR4_Full_BIT; break; default: - DP_INFO(edev, "Unsupported speed %u\n", speed); + DP_INFO(edev, "Unsupported speed %u\n", base->speed); return -EINVAL; } } @@ -368,8 +440,7 @@ static u32 qede_get_msglevel(struct net_device *ndev) { struct qede_dev *edev = netdev_priv(ndev); - return ((u32)edev->dp_level << QED_LOG_LEVEL_SHIFT) | - edev->dp_module; + return ((u32)edev->dp_level << QED_LOG_LEVEL_SHIFT) | edev->dp_module; } static void qede_set_msglevel(struct net_device *ndev, u32 level) @@ -393,8 +464,7 @@ static int qede_nway_reset(struct net_device *dev) struct qed_link_params link_params; if (!edev->ops || !edev->ops->common->can_link_change(edev->cdev)) { - DP_INFO(edev, - "Link settings are not allowed to be changed\n"); + DP_INFO(edev, "Link settings are not allowed to be changed\n"); return -EOPNOTSUPP; } @@ -1228,8 +1298,8 @@ static int qede_get_tunable(struct net_device *dev, } static const struct ethtool_ops qede_ethtool_ops = { - .get_settings = qede_get_settings, - .set_settings = qede_set_settings, + .get_link_ksettings = qede_get_link_ksettings, + .set_link_ksettings = qede_set_link_ksettings, .get_drvinfo = qede_get_drvinfo, .get_msglevel = qede_get_msglevel, .set_msglevel = qede_set_msglevel, @@ -1260,7 +1330,7 @@ static const struct ethtool_ops qede_ethtool_ops = { }; static const struct ethtool_ops qede_vf_ethtool_ops = { - .get_settings = qede_get_settings, + .get_link_ksettings = qede_get_link_ksettings, .get_drvinfo = qede_get_drvinfo, .get_msglevel = qede_get_msglevel, .set_msglevel = qede_set_msglevel, diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index e4bd02e46e57..5ce8a3c9ed6f 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -222,7 +222,7 @@ int __init qede_init(void) { int ret; - pr_notice("qede_init: %s\n", version); + pr_info("qede_init: %s\n", version); qed_ops = qed_get_eth_ops(); if (!qed_ops) { @@ -253,7 +253,8 @@ int __init qede_init(void) static void __exit qede_cleanup(void) { - pr_notice("qede_cleanup called\n"); + if (debug & QED_LOG_INFO_MASK) + pr_info("qede_cleanup called\n"); unregister_netdevice_notifier(&qede_netdev_notifier); pci_unregister_driver(&qede_pci_driver); @@ -270,8 +271,7 @@ module_exit(qede_cleanup); /* Unmap the data and free skb */ static int qede_free_tx_pkt(struct qede_dev *edev, - struct qede_tx_queue *txq, - int *len) + struct qede_tx_queue *txq, int *len) { u16 idx = txq->sw_tx_cons & NUM_TX_BDS_MAX; struct sk_buff *skb = txq->sw_tx_ring[idx].skb; @@ -329,8 +329,7 @@ static int qede_free_tx_pkt(struct qede_dev *edev, static void qede_free_failed_tx_pkt(struct qede_dev *edev, struct qede_tx_queue *txq, struct eth_tx_1st_bd *first_bd, - int nbd, - bool data_split) + int nbd, bool data_split) { u16 idx = txq->sw_tx_prod & NUM_TX_BDS_MAX; struct sk_buff *skb = txq->sw_tx_ring[idx].skb; @@ -339,8 +338,7 @@ static void qede_free_failed_tx_pkt(struct qede_dev *edev, /* Return prod to its position before this skb was handled */ qed_chain_set_prod(&txq->tx_pbl, - le16_to_cpu(txq->tx_db.data.bd_prod), - first_bd); + le16_to_cpu(txq->tx_db.data.bd_prod), first_bd); first_bd = (struct eth_tx_1st_bd *)qed_chain_produce(&txq->tx_pbl); @@ -366,8 +364,7 @@ static void qede_free_failed_tx_pkt(struct qede_dev *edev, /* Return again prod to its position before this skb was handled */ qed_chain_set_prod(&txq->tx_pbl, - le16_to_cpu(txq->tx_db.data.bd_prod), - first_bd); + le16_to_cpu(txq->tx_db.data.bd_prod), first_bd); /* Free skb */ dev_kfree_skb_any(skb); @@ -376,8 +373,7 @@ static void qede_free_failed_tx_pkt(struct qede_dev *edev, } static u32 qede_xmit_type(struct qede_dev *edev, - struct sk_buff *skb, - int *ipv6_ext) + struct sk_buff *skb, int *ipv6_ext) { u32 rc = XMIT_L4_CSUM; __be16 l3_proto; @@ -434,15 +430,13 @@ static void qede_set_params_for_ipv6_ext(struct sk_buff *skb, } static int map_frag_to_bd(struct qede_dev *edev, - skb_frag_t *frag, - struct eth_tx_bd *bd) + skb_frag_t *frag, struct eth_tx_bd *bd) { dma_addr_t mapping; /* Map skb non-linear frag data for DMA */ mapping = skb_frag_dma_map(&edev->pdev->dev, frag, 0, - skb_frag_size(frag), - DMA_TO_DEVICE); + skb_frag_size(frag), DMA_TO_DEVICE); if (unlikely(dma_mapping_error(&edev->pdev->dev, mapping))) { DP_NOTICE(edev, "Unable to map frag - dropping packet\n"); return -ENOMEM; @@ -504,9 +498,8 @@ static inline void qede_update_tx_producer(struct qede_tx_queue *txq) } /* Main transmit function */ -static -netdev_tx_t qede_start_xmit(struct sk_buff *skb, - struct net_device *ndev) +static netdev_tx_t qede_start_xmit(struct sk_buff *skb, + struct net_device *ndev) { struct qede_dev *edev = netdev_priv(ndev); struct netdev_queue *netdev_txq; @@ -530,8 +523,7 @@ netdev_tx_t qede_start_xmit(struct sk_buff *skb, txq = QEDE_TX_QUEUE(edev, txq_index); netdev_txq = netdev_get_tx_queue(ndev, txq_index); - WARN_ON(qed_chain_get_elem_left(&txq->tx_pbl) < - (MAX_SKB_FRAGS + 1)); + WARN_ON(qed_chain_get_elem_left(&txq->tx_pbl) < (MAX_SKB_FRAGS + 1)); xmit_type = qede_xmit_type(edev, skb, &ipv6_ext); @@ -761,8 +753,7 @@ int qede_txq_has_work(struct qede_tx_queue *txq) return hw_bd_cons != qed_chain_get_cons_idx(&txq->tx_pbl); } -static int qede_tx_int(struct qede_dev *edev, - struct qede_tx_queue *txq) +static int qede_tx_int(struct qede_dev *edev, struct qede_tx_queue *txq) { struct netdev_queue *netdev_txq; u16 hw_bd_cons; @@ -960,8 +951,7 @@ static inline void qede_update_rx_prod(struct qede_dev *edev, static u32 qede_get_rxhash(struct qede_dev *edev, u8 bitfields, - __le32 rss_hash, - enum pkt_hash_types *rxhash_type) + __le32 rss_hash, enum pkt_hash_types *rxhash_type) { enum rss_hash_type htype; @@ -990,12 +980,10 @@ static void qede_set_skb_csum(struct sk_buff *skb, u8 csum_flag) static inline void qede_skb_receive(struct qede_dev *edev, struct qede_fastpath *fp, - struct sk_buff *skb, - u16 vlan_tag) + struct sk_buff *skb, u16 vlan_tag) { if (vlan_tag) - __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), - vlan_tag); + __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vlan_tag); napi_gro_receive(&fp->napi, skb); } @@ -1018,8 +1006,7 @@ static void qede_set_gro_params(struct qede_dev *edev, static int qede_fill_frag_skb(struct qede_dev *edev, struct qede_rx_queue *rxq, - u8 tpa_agg_index, - u16 len_on_bd) + u8 tpa_agg_index, u16 len_on_bd) { struct sw_rx_data *current_bd = &rxq->sw_rx_ring[rxq->sw_rx_cons & NUM_RX_BDS_MAX]; @@ -1467,7 +1454,7 @@ alloc_skb: skb = netdev_alloc_skb(edev->ndev, QEDE_RX_HDR_SIZE); if (unlikely(!skb)) { DP_NOTICE(edev, - "Build_skb failed, dropping incoming packet\n"); + "skb allocation failed, dropping incoming packet\n"); qede_recycle_rx_bd_ring(rxq, edev, fp_cqe->bd_num); rxq->rx_alloc_errors++; goto next_cqe; @@ -1575,8 +1562,7 @@ alloc_skb: skb->protocol = eth_type_trans(skb, edev->ndev); rx_hash = qede_get_rxhash(edev, fp_cqe->bitfields, - fp_cqe->rss_hash, - &rxhash_type); + fp_cqe->rss_hash, &rxhash_type); skb_set_hash(skb, rx_hash, rxhash_type); @@ -1787,9 +1773,9 @@ void qede_fill_by_demand_stats(struct qede_dev *edev) edev->stats.tx_mac_ctrl_frames = stats.tx_mac_ctrl_frames; } -static struct rtnl_link_stats64 *qede_get_stats64( - struct net_device *dev, - struct rtnl_link_stats64 *stats) +static +struct rtnl_link_stats64 *qede_get_stats64(struct net_device *dev, + struct rtnl_link_stats64 *stats) { struct qede_dev *edev = netdev_priv(dev); @@ -2103,8 +2089,7 @@ static void qede_vlan_mark_nonconfigured(struct qede_dev *edev) } DP_VERBOSE(edev, NETIF_MSG_IFDOWN, - "marked vlan %d as non-configured\n", - vlan->vid); + "marked vlan %d as non-configured\n", vlan->vid); } edev->accept_any_vlan = false; @@ -2146,7 +2131,7 @@ static void qede_udp_tunnel_add(struct net_device *dev, edev->vxlan_dst_port = t_port; - DP_VERBOSE(edev, QED_MSG_DEBUG, "Added vxlan port=%d", + DP_VERBOSE(edev, QED_MSG_DEBUG, "Added vxlan port=%d\n", t_port); set_bit(QEDE_SP_VXLAN_PORT_CONFIG, &edev->sp_flags); @@ -2157,7 +2142,7 @@ static void qede_udp_tunnel_add(struct net_device *dev, edev->geneve_dst_port = t_port; - DP_VERBOSE(edev, QED_MSG_DEBUG, "Added geneve port=%d", + DP_VERBOSE(edev, QED_MSG_DEBUG, "Added geneve port=%d\n", t_port); set_bit(QEDE_SP_GENEVE_PORT_CONFIG, &edev->sp_flags); break; @@ -2181,7 +2166,7 @@ static void qede_udp_tunnel_del(struct net_device *dev, edev->vxlan_dst_port = 0; - DP_VERBOSE(edev, QED_MSG_DEBUG, "Deleted vxlan port=%d", + DP_VERBOSE(edev, QED_MSG_DEBUG, "Deleted vxlan port=%d\n", t_port); set_bit(QEDE_SP_VXLAN_PORT_CONFIG, &edev->sp_flags); @@ -2192,7 +2177,7 @@ static void qede_udp_tunnel_del(struct net_device *dev, edev->geneve_dst_port = 0; - DP_VERBOSE(edev, QED_MSG_DEBUG, "Deleted geneve port=%d", + DP_VERBOSE(edev, QED_MSG_DEBUG, "Deleted geneve port=%d\n", t_port); set_bit(QEDE_SP_GENEVE_PORT_CONFIG, &edev->sp_flags); break; @@ -2237,15 +2222,13 @@ static const struct net_device_ops qede_netdev_ops = { static struct qede_dev *qede_alloc_etherdev(struct qed_dev *cdev, struct pci_dev *pdev, struct qed_dev_eth_info *info, - u32 dp_module, - u8 dp_level) + u32 dp_module, u8 dp_level) { struct net_device *ndev; struct qede_dev *edev; ndev = alloc_etherdev_mqs(sizeof(*edev), - info->num_queues, - info->num_queues); + info->num_queues, info->num_queues); if (!ndev) { pr_err("etherdev allocation failed\n"); return NULL; @@ -2261,6 +2244,9 @@ static struct qede_dev *qede_alloc_etherdev(struct qed_dev *cdev, edev->q_num_rx_buffers = NUM_RX_BDS_DEF; edev->q_num_tx_buffers = NUM_TX_BDS_DEF; + DP_INFO(edev, "Allocated netdev with %d tx queues and %d rx queues\n", + info->num_queues, info->num_queues); + SET_NETDEV_DEV(ndev, &pdev->dev); memset(&edev->stats, 0, sizeof(edev->stats)); @@ -2453,7 +2439,7 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, bool is_vf, enum qede_probe_mode mode) { struct qed_probe_params probe_params; - struct qed_slowpath_params params; + struct qed_slowpath_params sp_params; struct qed_dev_eth_info dev_info; struct qede_dev *edev; struct qed_dev *cdev; @@ -2476,14 +2462,14 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, qede_update_pf_params(cdev); /* Start the Slowpath-process */ - memset(¶ms, 0, sizeof(struct qed_slowpath_params)); - params.int_mode = QED_INT_MODE_MSIX; - params.drv_major = QEDE_MAJOR_VERSION; - params.drv_minor = QEDE_MINOR_VERSION; - params.drv_rev = QEDE_REVISION_VERSION; - params.drv_eng = QEDE_ENGINEERING_VERSION; - strlcpy(params.name, "qede LAN", QED_DRV_VER_STR_SIZE); - rc = qed_ops->common->slowpath_start(cdev, ¶ms); + memset(&sp_params, 0, sizeof(sp_params)); + sp_params.int_mode = QED_INT_MODE_MSIX; + sp_params.drv_major = QEDE_MAJOR_VERSION; + sp_params.drv_minor = QEDE_MINOR_VERSION; + sp_params.drv_rev = QEDE_REVISION_VERSION; + sp_params.drv_eng = QEDE_ENGINEERING_VERSION; + strlcpy(sp_params.name, "qede LAN", QED_DRV_VER_STR_SIZE); + rc = qed_ops->common->slowpath_start(cdev, &sp_params); if (rc) { pr_notice("Cannot start slowpath\n"); goto err1; @@ -2586,7 +2572,7 @@ static void __qede_remove(struct pci_dev *pdev, enum qede_remove_mode mode) qed_ops->common->slowpath_stop(cdev); qed_ops->common->remove(cdev); - pr_notice("Ending successfully qede_remove\n"); + dev_info(&pdev->dev, "Ending qede_remove successfully\n"); } static void qede_remove(struct pci_dev *pdev) @@ -2634,16 +2620,14 @@ static void qede_free_mem_sb(struct qede_dev *edev, /* This function allocates fast-path status block memory */ static int qede_alloc_mem_sb(struct qede_dev *edev, - struct qed_sb_info *sb_info, - u16 sb_id) + struct qed_sb_info *sb_info, u16 sb_id) { struct status_block *sb_virt; dma_addr_t sb_phys; int rc; sb_virt = dma_alloc_coherent(&edev->pdev->dev, - sizeof(*sb_virt), - &sb_phys, GFP_KERNEL); + sizeof(*sb_virt), &sb_phys, GFP_KERNEL); if (!sb_virt) { DP_ERR(edev, "Status block allocation failed\n"); return -ENOMEM; @@ -2675,16 +2659,15 @@ static void qede_free_rx_buffers(struct qede_dev *edev, data = rx_buf->data; dma_unmap_page(&edev->pdev->dev, - rx_buf->mapping, - PAGE_SIZE, DMA_FROM_DEVICE); + rx_buf->mapping, PAGE_SIZE, DMA_FROM_DEVICE); rx_buf->data = NULL; __free_page(data); } } -static void qede_free_sge_mem(struct qede_dev *edev, - struct qede_rx_queue *rxq) { +static void qede_free_sge_mem(struct qede_dev *edev, struct qede_rx_queue *rxq) +{ int i; if (edev->gro_disable) @@ -2703,8 +2686,7 @@ static void qede_free_sge_mem(struct qede_dev *edev, } } -static void qede_free_mem_rxq(struct qede_dev *edev, - struct qede_rx_queue *rxq) +static void qede_free_mem_rxq(struct qede_dev *edev, struct qede_rx_queue *rxq) { qede_free_sge_mem(edev, rxq); @@ -2726,9 +2708,6 @@ static int qede_alloc_rx_buffer(struct qede_dev *edev, struct eth_rx_bd *rx_bd; dma_addr_t mapping; struct page *data; - u16 rx_buf_size; - - rx_buf_size = rxq->rx_buf_size; data = alloc_pages(GFP_ATOMIC, 0); if (unlikely(!data)) { @@ -2763,8 +2742,7 @@ static int qede_alloc_rx_buffer(struct qede_dev *edev, return 0; } -static int qede_alloc_sge_mem(struct qede_dev *edev, - struct qede_rx_queue *rxq) +static int qede_alloc_sge_mem(struct qede_dev *edev, struct qede_rx_queue *rxq) { dma_addr_t mapping; int i; @@ -2811,15 +2789,14 @@ err: } /* This function allocates all memory needed per Rx queue */ -static int qede_alloc_mem_rxq(struct qede_dev *edev, - struct qede_rx_queue *rxq) +static int qede_alloc_mem_rxq(struct qede_dev *edev, struct qede_rx_queue *rxq) { int i, rc, size; rxq->num_rx_buffers = edev->q_num_rx_buffers; - rxq->rx_buf_size = NET_IP_ALIGN + ETH_OVERHEAD + - edev->ndev->mtu; + rxq->rx_buf_size = NET_IP_ALIGN + ETH_OVERHEAD + edev->ndev->mtu; + if (rxq->rx_buf_size > PAGE_SIZE) rxq->rx_buf_size = PAGE_SIZE; @@ -2873,8 +2850,7 @@ err: return rc; } -static void qede_free_mem_txq(struct qede_dev *edev, - struct qede_tx_queue *txq) +static void qede_free_mem_txq(struct qede_dev *edev, struct qede_tx_queue *txq) { /* Free the parallel SW ring */ kfree(txq->sw_tx_ring); @@ -2884,8 +2860,7 @@ static void qede_free_mem_txq(struct qede_dev *edev, } /* This function allocates all memory needed per Tx queue */ -static int qede_alloc_mem_txq(struct qede_dev *edev, - struct qede_tx_queue *txq) +static int qede_alloc_mem_txq(struct qede_dev *edev, struct qede_tx_queue *txq) { int size, rc; union eth_tx_bd_types *p_virt; @@ -2917,8 +2892,7 @@ err: } /* This function frees all memory of a single fp */ -static void qede_free_mem_fp(struct qede_dev *edev, - struct qede_fastpath *fp) +static void qede_free_mem_fp(struct qede_dev *edev, struct qede_fastpath *fp) { int tc; @@ -2933,8 +2907,7 @@ static void qede_free_mem_fp(struct qede_dev *edev, /* This function allocates all memory needed for a single fp (i.e. an entity * which contains status block, one rx queue and multiple per-TC tx queues. */ -static int qede_alloc_mem_fp(struct qede_dev *edev, - struct qede_fastpath *fp) +static int qede_alloc_mem_fp(struct qede_dev *edev, struct qede_fastpath *fp) { int rc, tc; @@ -3146,8 +3119,7 @@ static int qede_setup_irqs(struct qede_dev *edev) } static int qede_drain_txq(struct qede_dev *edev, - struct qede_tx_queue *txq, - bool allow_drain) + struct qede_tx_queue *txq, bool allow_drain) { int rc, cnt = 1000; diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index 1e1cc0fad17f..d4809adeb048 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -1876,6 +1876,20 @@ static int ravb_set_gti(struct net_device *ndev) return 0; } +static void ravb_set_config_mode(struct net_device *ndev) +{ + struct ravb_private *priv = netdev_priv(ndev); + + if (priv->chip_id == RCAR_GEN2) { + ravb_modify(ndev, CCC, CCC_OPC, CCC_OPC_CONFIG); + /* Set CSEL value */ + ravb_modify(ndev, CCC, CCC_CSEL, CCC_CSEL_HPB); + } else { + ravb_modify(ndev, CCC, CCC_OPC, CCC_OPC_CONFIG | + CCC_GAC | CCC_CSEL_HPB); + } +} + static int ravb_probe(struct platform_device *pdev) { struct device_node *np = pdev->dev.of_node; @@ -1978,14 +1992,7 @@ static int ravb_probe(struct platform_device *pdev) ndev->ethtool_ops = &ravb_ethtool_ops; /* Set AVB config mode */ - if (chip_id == RCAR_GEN2) { - ravb_modify(ndev, CCC, CCC_OPC, CCC_OPC_CONFIG); - /* Set CSEL value */ - ravb_modify(ndev, CCC, CCC_CSEL, CCC_CSEL_HPB); - } else { - ravb_modify(ndev, CCC, CCC_OPC, CCC_OPC_CONFIG | - CCC_GAC | CCC_CSEL_HPB); - } + ravb_set_config_mode(ndev); /* Set GTI value */ error = ravb_set_gti(ndev); @@ -2097,6 +2104,54 @@ static int ravb_remove(struct platform_device *pdev) } #ifdef CONFIG_PM +static int ravb_suspend(struct device *dev) +{ + struct net_device *ndev = dev_get_drvdata(dev); + int ret = 0; + + if (netif_running(ndev)) { + netif_device_detach(ndev); + ret = ravb_close(ndev); + } + + return ret; +} + +static int ravb_resume(struct device *dev) +{ + struct net_device *ndev = dev_get_drvdata(dev); + struct ravb_private *priv = netdev_priv(ndev); + int ret = 0; + + /* All register have been reset to default values. + * Restore all registers which where setup at probe time and + * reopen device if it was running before system suspended. + */ + + /* Set AVB config mode */ + ravb_set_config_mode(ndev); + + /* Set GTI value */ + ret = ravb_set_gti(ndev); + if (ret) + return ret; + + /* Request GTI loading */ + ravb_modify(ndev, GCCR, GCCR_LTI, GCCR_LTI); + + /* Restore descriptor base address table */ + ravb_write(ndev, priv->desc_bat_dma, DBAT); + + if (netif_running(ndev)) { + ret = ravb_open(ndev); + if (ret < 0) + return ret; + netif_device_attach(ndev); + } + + return ret; +} + static int ravb_runtime_nop(struct device *dev) { /* Runtime PM callback shared between ->runtime_suspend() @@ -2110,6 +2165,7 @@ static int ravb_runtime_nop(struct device *dev) } static const struct dev_pm_ops ravb_dev_pm_ops = { + SET_SYSTEM_SLEEP_PM_OPS(ravb_suspend, ravb_resume) SET_RUNTIME_PM_OPS(ravb_runtime_nop, ravb_runtime_nop, NULL) }; diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c index 799d58d86e6d..1f8240aec086 100644 --- a/drivers/net/ethernet/renesas/sh_eth.c +++ b/drivers/net/ethernet/renesas/sh_eth.c @@ -1723,7 +1723,7 @@ out: static void sh_eth_adjust_link(struct net_device *ndev) { struct sh_eth_private *mdp = netdev_priv(ndev); - struct phy_device *phydev = mdp->phydev; + struct phy_device *phydev = ndev->phydev; int new_state = 0; if (phydev->link) { @@ -1800,51 +1800,48 @@ static int sh_eth_phy_init(struct net_device *ndev) phy_attached_info(phydev); - mdp->phydev = phydev; - return 0; } /* PHY control start function */ static int sh_eth_phy_start(struct net_device *ndev) { - struct sh_eth_private *mdp = netdev_priv(ndev); int ret; ret = sh_eth_phy_init(ndev); if (ret) return ret; - phy_start(mdp->phydev); + phy_start(ndev->phydev); return 0; } -static int sh_eth_get_settings(struct net_device *ndev, - struct ethtool_cmd *ecmd) +static int sh_eth_get_link_ksettings(struct net_device *ndev, + struct ethtool_link_ksettings *cmd) { struct sh_eth_private *mdp = netdev_priv(ndev); unsigned long flags; int ret; - if (!mdp->phydev) + if (!ndev->phydev) return -ENODEV; spin_lock_irqsave(&mdp->lock, flags); - ret = phy_ethtool_gset(mdp->phydev, ecmd); + ret = phy_ethtool_ksettings_get(ndev->phydev, cmd); spin_unlock_irqrestore(&mdp->lock, flags); return ret; } -static int sh_eth_set_settings(struct net_device *ndev, - struct ethtool_cmd *ecmd) +static int sh_eth_set_link_ksettings(struct net_device *ndev, + const struct ethtool_link_ksettings *cmd) { struct sh_eth_private *mdp = netdev_priv(ndev); unsigned long flags; int ret; - if (!mdp->phydev) + if (!ndev->phydev) return -ENODEV; spin_lock_irqsave(&mdp->lock, flags); @@ -1852,11 +1849,11 @@ static int sh_eth_set_settings(struct net_device *ndev, /* disable tx and rx */ sh_eth_rcv_snd_disable(ndev); - ret = phy_ethtool_sset(mdp->phydev, ecmd); + ret = phy_ethtool_ksettings_set(ndev->phydev, cmd); if (ret) goto error_exit; - if (ecmd->duplex == DUPLEX_FULL) + if (cmd->base.duplex == DUPLEX_FULL) mdp->duplex = 1; else mdp->duplex = 0; @@ -2067,11 +2064,11 @@ static int sh_eth_nway_reset(struct net_device *ndev) unsigned long flags; int ret; - if (!mdp->phydev) + if (!ndev->phydev) return -ENODEV; spin_lock_irqsave(&mdp->lock, flags); - ret = phy_start_aneg(mdp->phydev); + ret = phy_start_aneg(ndev->phydev); spin_unlock_irqrestore(&mdp->lock, flags); return ret; @@ -2198,8 +2195,6 @@ static int sh_eth_set_ringparam(struct net_device *ndev, } static const struct ethtool_ops sh_eth_ethtool_ops = { - .get_settings = sh_eth_get_settings, - .set_settings = sh_eth_set_settings, .get_regs_len = sh_eth_get_regs_len, .get_regs = sh_eth_get_regs, .nway_reset = sh_eth_nway_reset, @@ -2211,6 +2206,8 @@ static const struct ethtool_ops sh_eth_ethtool_ops = { .get_sset_count = sh_eth_get_sset_count, .get_ringparam = sh_eth_get_ringparam, .set_ringparam = sh_eth_set_ringparam, + .get_link_ksettings = sh_eth_get_link_ksettings, + .set_link_ksettings = sh_eth_set_link_ksettings, }; /* network device open function */ @@ -2408,10 +2405,9 @@ static int sh_eth_close(struct net_device *ndev) sh_eth_dev_exit(ndev); /* PHY Disconnect */ - if (mdp->phydev) { - phy_stop(mdp->phydev); - phy_disconnect(mdp->phydev); - mdp->phydev = NULL; + if (ndev->phydev) { + phy_stop(ndev->phydev); + phy_disconnect(ndev->phydev); } free_irq(ndev->irq, ndev); @@ -2429,8 +2425,7 @@ static int sh_eth_close(struct net_device *ndev) /* ioctl to device function */ static int sh_eth_do_ioctl(struct net_device *ndev, struct ifreq *rq, int cmd) { - struct sh_eth_private *mdp = netdev_priv(ndev); - struct phy_device *phydev = mdp->phydev; + struct phy_device *phydev = ndev->phydev; if (!netif_running(ndev)) return -EINVAL; diff --git a/drivers/net/ethernet/renesas/sh_eth.h b/drivers/net/ethernet/renesas/sh_eth.h index c62380e34a1d..d050f37f3e0f 100644 --- a/drivers/net/ethernet/renesas/sh_eth.h +++ b/drivers/net/ethernet/renesas/sh_eth.h @@ -518,7 +518,6 @@ struct sh_eth_private { /* MII transceiver section. */ u32 phy_id; /* PHY ID */ struct mii_bus *mii_bus; /* MDIO bus control */ - struct phy_device *phydev; /* PHY device control */ int link; phy_interface_t phy_interface; int msg_enable; diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index f658fee74f18..b8c9f1884a15 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -177,7 +177,7 @@ static int efx_ef10_get_vf_index(struct efx_nic *efx) static int efx_ef10_init_datapath_caps(struct efx_nic *efx) { - MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_CAPABILITIES_OUT_LEN); + MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_CAPABILITIES_V2_OUT_LEN); struct efx_ef10_nic_data *nic_data = efx->nic_data; size_t outlen; int rc; @@ -188,7 +188,7 @@ static int efx_ef10_init_datapath_caps(struct efx_nic *efx) outbuf, sizeof(outbuf), &outlen); if (rc) return rc; - if (outlen < sizeof(outbuf)) { + if (outlen < MC_CMD_GET_CAPABILITIES_OUT_LEN) { netif_err(efx, drv, efx->net_dev, "unable to read datapath firmware capabilities\n"); return -EIO; @@ -197,6 +197,12 @@ static int efx_ef10_init_datapath_caps(struct efx_nic *efx) nic_data->datapath_caps = MCDI_DWORD(outbuf, GET_CAPABILITIES_OUT_FLAGS1); + if (outlen >= MC_CMD_GET_CAPABILITIES_V2_OUT_LEN) + nic_data->datapath_caps2 = MCDI_DWORD(outbuf, + GET_CAPABILITIES_V2_OUT_FLAGS2); + else + nic_data->datapath_caps2 = 0; + /* record the DPCPU firmware IDs to determine VEB vswitching support. */ nic_data->rx_dpcpu_fw_id = @@ -227,6 +233,116 @@ static int efx_ef10_get_sysclk_freq(struct efx_nic *efx) return rc > 0 ? rc : -ERANGE; } +static int efx_ef10_get_timer_workarounds(struct efx_nic *efx) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + unsigned int implemented; + unsigned int enabled; + int rc; + + nic_data->workaround_35388 = false; + nic_data->workaround_61265 = false; + + rc = efx_mcdi_get_workarounds(efx, &implemented, &enabled); + + if (rc == -ENOSYS) { + /* Firmware without GET_WORKAROUNDS - not a problem. */ + rc = 0; + } else if (rc == 0) { + /* Bug61265 workaround is always enabled if implemented. */ + if (enabled & MC_CMD_GET_WORKAROUNDS_OUT_BUG61265) + nic_data->workaround_61265 = true; + + if (enabled & MC_CMD_GET_WORKAROUNDS_OUT_BUG35388) { + nic_data->workaround_35388 = true; + } else if (implemented & MC_CMD_GET_WORKAROUNDS_OUT_BUG35388) { + /* Workaround is implemented but not enabled. + * Try to enable it. + */ + rc = efx_mcdi_set_workaround(efx, + MC_CMD_WORKAROUND_BUG35388, + true, NULL); + if (rc == 0) + nic_data->workaround_35388 = true; + /* If we failed to set the workaround just carry on. */ + rc = 0; + } + } + + netif_dbg(efx, probe, efx->net_dev, + "workaround for bug 35388 is %sabled\n", + nic_data->workaround_35388 ? "en" : "dis"); + netif_dbg(efx, probe, efx->net_dev, + "workaround for bug 61265 is %sabled\n", + nic_data->workaround_61265 ? "en" : "dis"); + + return rc; +} + +static void efx_ef10_process_timer_config(struct efx_nic *efx, + const efx_dword_t *data) +{ + unsigned int max_count; + + if (EFX_EF10_WORKAROUND_61265(efx)) { + efx->timer_quantum_ns = MCDI_DWORD(data, + GET_EVQ_TMR_PROPERTIES_OUT_MCDI_TMR_STEP_NS); + efx->timer_max_ns = MCDI_DWORD(data, + GET_EVQ_TMR_PROPERTIES_OUT_MCDI_TMR_MAX_NS); + } else if (EFX_EF10_WORKAROUND_35388(efx)) { + efx->timer_quantum_ns = MCDI_DWORD(data, + GET_EVQ_TMR_PROPERTIES_OUT_BUG35388_TMR_NS_PER_COUNT); + max_count = MCDI_DWORD(data, + GET_EVQ_TMR_PROPERTIES_OUT_BUG35388_TMR_MAX_COUNT); + efx->timer_max_ns = max_count * efx->timer_quantum_ns; + } else { + efx->timer_quantum_ns = MCDI_DWORD(data, + GET_EVQ_TMR_PROPERTIES_OUT_TMR_REG_NS_PER_COUNT); + max_count = MCDI_DWORD(data, + GET_EVQ_TMR_PROPERTIES_OUT_TMR_REG_MAX_COUNT); + efx->timer_max_ns = max_count * efx->timer_quantum_ns; + } + + netif_dbg(efx, probe, efx->net_dev, + "got timer properties from MC: quantum %u ns; max %u ns\n", + efx->timer_quantum_ns, efx->timer_max_ns); +} + +static int efx_ef10_get_timer_config(struct efx_nic *efx) +{ + MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_LEN); + int rc; + + rc = efx_ef10_get_timer_workarounds(efx); + if (rc) + return rc; + + rc = efx_mcdi_rpc_quiet(efx, MC_CMD_GET_EVQ_TMR_PROPERTIES, NULL, 0, + outbuf, sizeof(outbuf), NULL); + + if (rc == 0) { + efx_ef10_process_timer_config(efx, outbuf); + } else if (rc == -ENOSYS || rc == -EPERM) { + /* Not available - fall back to Huntington defaults. */ + unsigned int quantum; + + rc = efx_ef10_get_sysclk_freq(efx); + if (rc < 0) + return rc; + + quantum = 1536000 / rc; /* 1536 cycles */ + efx->timer_quantum_ns = quantum; + efx->timer_max_ns = efx->type->timer_period_max * quantum; + rc = 0; + } else { + efx_mcdi_display_error(efx, MC_CMD_GET_EVQ_TMR_PROPERTIES, + MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_LEN, + NULL, 0, rc); + } + + return rc; +} + static int efx_ef10_get_mac_address_pf(struct efx_nic *efx, u8 *mac_address) { MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_MAC_ADDRESSES_OUT_LEN); @@ -527,33 +643,11 @@ static int efx_ef10_probe(struct efx_nic *efx) if (rc) goto fail5; - rc = efx_ef10_get_sysclk_freq(efx); + rc = efx_ef10_get_timer_config(efx); if (rc < 0) goto fail5; efx->timer_quantum_ns = 1536000 / rc; /* 1536 cycles */ - /* Check whether firmware supports bug 35388 workaround. - * First try to enable it, then if we get EPERM, just - * ask if it's already enabled - */ - rc = efx_mcdi_set_workaround(efx, MC_CMD_WORKAROUND_BUG35388, true, NULL); - if (rc == 0) { - nic_data->workaround_35388 = true; - } else if (rc == -EPERM) { - unsigned int enabled; - - rc = efx_mcdi_get_workarounds(efx, NULL, &enabled); - if (rc) - goto fail3; - nic_data->workaround_35388 = enabled & - MC_CMD_GET_WORKAROUNDS_OUT_BUG35388; - } else if (rc != -ENOSYS && rc != -ENOENT) { - goto fail5; - } - netif_dbg(efx, probe, efx->net_dev, - "workaround for bug 35388 is %sabled\n", - nic_data->workaround_35388 ? "en" : "dis"); - rc = efx_mcdi_mon_probe(efx); if (rc && rc != -EPERM) goto fail5; @@ -1743,27 +1837,43 @@ static size_t efx_ef10_update_stats_vf(struct efx_nic *efx, u64 *full_stats, static void efx_ef10_push_irq_moderation(struct efx_channel *channel) { struct efx_nic *efx = channel->efx; - unsigned int mode, value; + unsigned int mode, usecs; efx_dword_t timer_cmd; - if (channel->irq_moderation) { + if (channel->irq_moderation_us) { mode = 3; - value = channel->irq_moderation - 1; + usecs = channel->irq_moderation_us; } else { mode = 0; - value = 0; + usecs = 0; } - if (EFX_EF10_WORKAROUND_35388(efx)) { + if (EFX_EF10_WORKAROUND_61265(efx)) { + MCDI_DECLARE_BUF(inbuf, MC_CMD_SET_EVQ_TMR_IN_LEN); + unsigned int ns = usecs * 1000; + + MCDI_SET_DWORD(inbuf, SET_EVQ_TMR_IN_INSTANCE, + channel->channel); + MCDI_SET_DWORD(inbuf, SET_EVQ_TMR_IN_TMR_LOAD_REQ_NS, ns); + MCDI_SET_DWORD(inbuf, SET_EVQ_TMR_IN_TMR_RELOAD_REQ_NS, ns); + MCDI_SET_DWORD(inbuf, SET_EVQ_TMR_IN_TMR_MODE, mode); + + efx_mcdi_rpc_async(efx, MC_CMD_SET_EVQ_TMR, + inbuf, sizeof(inbuf), 0, NULL, 0); + } else if (EFX_EF10_WORKAROUND_35388(efx)) { + unsigned int ticks = efx_usecs_to_ticks(efx, usecs); + EFX_POPULATE_DWORD_3(timer_cmd, ERF_DD_EVQ_IND_TIMER_FLAGS, EFE_DD_EVQ_IND_TIMER_FLAGS, ERF_DD_EVQ_IND_TIMER_MODE, mode, - ERF_DD_EVQ_IND_TIMER_VAL, value); + ERF_DD_EVQ_IND_TIMER_VAL, ticks); efx_writed_page(efx, &timer_cmd, ER_DD_EVQ_INDIRECT, channel->channel); } else { + unsigned int ticks = efx_usecs_to_ticks(efx, usecs); + EFX_POPULATE_DWORD_2(timer_cmd, ERF_DZ_TC_TIMER_MODE, mode, - ERF_DZ_TC_TIMER_VAL, value); + ERF_DZ_TC_TIMER_VAL, ticks); efx_writed_page(efx, &timer_cmd, ER_DZ_EVQ_TMR, channel->channel); } @@ -2535,13 +2645,12 @@ fail: static int efx_ef10_ev_init(struct efx_channel *channel) { MCDI_DECLARE_BUF(inbuf, - MC_CMD_INIT_EVQ_IN_LEN(EFX_MAX_EVQ_SIZE * 8 / - EFX_BUF_SIZE)); - MCDI_DECLARE_BUF(outbuf, MC_CMD_INIT_EVQ_OUT_LEN); + MC_CMD_INIT_EVQ_V2_IN_LEN(EFX_MAX_EVQ_SIZE * 8 / + EFX_BUF_SIZE)); + MCDI_DECLARE_BUF(outbuf, MC_CMD_INIT_EVQ_V2_OUT_LEN); size_t entries = channel->eventq.buf.len / EFX_BUF_SIZE; struct efx_nic *efx = channel->efx; struct efx_ef10_nic_data *nic_data; - bool supports_rx_merge; size_t inlen, outlen; unsigned int enabled, implemented; dma_addr_t dma_addr; @@ -2549,9 +2658,6 @@ static int efx_ef10_ev_init(struct efx_channel *channel) int i; nic_data = efx->nic_data; - supports_rx_merge = - !!(nic_data->datapath_caps & - 1 << MC_CMD_GET_CAPABILITIES_OUT_RX_BATCHING_LBN); /* Fill event queue with all ones (i.e. empty events) */ memset(channel->eventq.buf.addr, 0xff, channel->eventq.buf.len); @@ -2560,11 +2666,6 @@ static int efx_ef10_ev_init(struct efx_channel *channel) MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_INSTANCE, channel->channel); /* INIT_EVQ expects index in vector table, not absolute */ MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_IRQ_NUM, channel->channel); - MCDI_POPULATE_DWORD_4(inbuf, INIT_EVQ_IN_FLAGS, - INIT_EVQ_IN_FLAG_INTERRUPTING, 1, - INIT_EVQ_IN_FLAG_RX_MERGE, 1, - INIT_EVQ_IN_FLAG_TX_MERGE, 1, - INIT_EVQ_IN_FLAG_CUT_THRU, !supports_rx_merge); MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_TMR_MODE, MC_CMD_INIT_EVQ_IN_TMR_MODE_DIS); MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_TMR_LOAD, 0); @@ -2573,6 +2674,27 @@ static int efx_ef10_ev_init(struct efx_channel *channel) MC_CMD_INIT_EVQ_IN_COUNT_MODE_DIS); MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_COUNT_THRSHLD, 0); + if (nic_data->datapath_caps2 & + 1 << MC_CMD_GET_CAPABILITIES_V2_OUT_INIT_EVQ_V2_LBN) { + /* Use the new generic approach to specifying event queue + * configuration, requesting lower latency or higher throughput. + * The options that actually get used appear in the output. + */ + MCDI_POPULATE_DWORD_2(inbuf, INIT_EVQ_V2_IN_FLAGS, + INIT_EVQ_V2_IN_FLAG_INTERRUPTING, 1, + INIT_EVQ_V2_IN_FLAG_TYPE, + MC_CMD_INIT_EVQ_V2_IN_FLAG_TYPE_AUTO); + } else { + bool cut_thru = !(nic_data->datapath_caps & + 1 << MC_CMD_GET_CAPABILITIES_OUT_RX_BATCHING_LBN); + + MCDI_POPULATE_DWORD_4(inbuf, INIT_EVQ_IN_FLAGS, + INIT_EVQ_IN_FLAG_INTERRUPTING, 1, + INIT_EVQ_IN_FLAG_RX_MERGE, 1, + INIT_EVQ_IN_FLAG_TX_MERGE, 1, + INIT_EVQ_IN_FLAG_CUT_THRU, cut_thru); + } + dma_addr = channel->eventq.buf.dma_addr; for (i = 0; i < entries; ++i) { MCDI_SET_ARRAY_QWORD(inbuf, INIT_EVQ_IN_DMA_ADDR, i, dma_addr); @@ -2583,6 +2705,13 @@ static int efx_ef10_ev_init(struct efx_channel *channel) rc = efx_mcdi_rpc(efx, MC_CMD_INIT_EVQ, inbuf, inlen, outbuf, sizeof(outbuf), &outlen); + + if (outlen >= MC_CMD_INIT_EVQ_V2_OUT_LEN) + netif_dbg(efx, drv, efx->net_dev, + "Channel %d using event queue flags %08x\n", + channel->channel, + MCDI_DWORD(outbuf, INIT_EVQ_V2_OUT_FLAGS)); + /* IRQ return is ignored */ if (channel->channel || rc) return rc; @@ -2590,8 +2719,8 @@ static int efx_ef10_ev_init(struct efx_channel *channel) /* Successfully created event queue on channel 0 */ rc = efx_mcdi_get_workarounds(efx, &implemented, &enabled); if (rc == -ENOSYS) { - /* GET_WORKAROUNDS was implemented before the bug26807 - * workaround, thus the latter must be unavailable in this fw + /* GET_WORKAROUNDS was implemented before this workaround, + * thus it must be unavailable in this firmware. */ nic_data->workaround_26807 = false; rc = 0; diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c index 14b821b1c880..f3826ae28bac 100644 --- a/drivers/net/ethernet/sfc/efx.c +++ b/drivers/net/ethernet/sfc/efx.c @@ -281,6 +281,27 @@ static int efx_process_channel(struct efx_channel *channel, int budget) * NAPI guarantees serialisation of polls of the same device, which * provides the guarantee required by efx_process_channel(). */ +static void efx_update_irq_mod(struct efx_nic *efx, struct efx_channel *channel) +{ + int step = efx->irq_mod_step_us; + + if (channel->irq_mod_score < irq_adapt_low_thresh) { + if (channel->irq_moderation_us > step) { + channel->irq_moderation_us -= step; + efx->type->push_irq_moderation(channel); + } + } else if (channel->irq_mod_score > irq_adapt_high_thresh) { + if (channel->irq_moderation_us < + efx->irq_rx_moderation_us) { + channel->irq_moderation_us += step; + efx->type->push_irq_moderation(channel); + } + } + + channel->irq_count = 0; + channel->irq_mod_score = 0; +} + static int efx_poll(struct napi_struct *napi, int budget) { struct efx_channel *channel = @@ -301,22 +322,7 @@ static int efx_poll(struct napi_struct *napi, int budget) if (efx_channel_has_rx_queue(channel) && efx->irq_rx_adaptive && unlikely(++channel->irq_count == 1000)) { - if (unlikely(channel->irq_mod_score < - irq_adapt_low_thresh)) { - if (channel->irq_moderation > 1) { - channel->irq_moderation -= 1; - efx->type->push_irq_moderation(channel); - } - } else if (unlikely(channel->irq_mod_score > - irq_adapt_high_thresh)) { - if (channel->irq_moderation < - efx->irq_rx_moderation) { - channel->irq_moderation += 1; - efx->type->push_irq_moderation(channel); - } - } - channel->irq_count = 0; - channel->irq_mod_score = 0; + efx_update_irq_mod(efx, channel); } efx_filter_rfs_expire(channel); @@ -1703,6 +1709,7 @@ static int efx_probe_nic(struct efx_nic *efx) netif_set_real_num_rx_queues(efx->net_dev, efx->n_rx_channels); /* Initialise the interrupt moderation settings */ + efx->irq_mod_step_us = DIV_ROUND_UP(efx->timer_quantum_ns, 1000); efx_init_irq_moderation(efx, tx_irq_mod_usec, rx_irq_mod_usec, true, true); @@ -1949,14 +1956,21 @@ static void efx_remove_all(struct efx_nic *efx) * Interrupt moderation * **************************************************************************/ - -static unsigned int irq_mod_ticks(unsigned int usecs, unsigned int quantum_ns) +unsigned int efx_usecs_to_ticks(struct efx_nic *efx, unsigned int usecs) { if (usecs == 0) return 0; - if (usecs * 1000 < quantum_ns) + if (usecs * 1000 < efx->timer_quantum_ns) return 1; /* never round down to 0 */ - return usecs * 1000 / quantum_ns; + return usecs * 1000 / efx->timer_quantum_ns; +} + +unsigned int efx_ticks_to_usecs(struct efx_nic *efx, unsigned int ticks) +{ + /* We must round up when converting ticks to microseconds + * because we round down when converting the other way. + */ + return DIV_ROUND_UP(ticks * efx->timer_quantum_ns, 1000); } /* Set interrupt moderation parameters */ @@ -1965,21 +1979,16 @@ int efx_init_irq_moderation(struct efx_nic *efx, unsigned int tx_usecs, bool rx_may_override_tx) { struct efx_channel *channel; - unsigned int irq_mod_max = DIV_ROUND_UP(efx->type->timer_period_max * - efx->timer_quantum_ns, - 1000); - unsigned int tx_ticks; - unsigned int rx_ticks; + unsigned int timer_max_us; EFX_ASSERT_RESET_SERIALISED(efx); - if (tx_usecs > irq_mod_max || rx_usecs > irq_mod_max) - return -EINVAL; + timer_max_us = efx->timer_max_ns / 1000; - tx_ticks = irq_mod_ticks(tx_usecs, efx->timer_quantum_ns); - rx_ticks = irq_mod_ticks(rx_usecs, efx->timer_quantum_ns); + if (tx_usecs > timer_max_us || rx_usecs > timer_max_us) + return -EINVAL; - if (tx_ticks != rx_ticks && efx->tx_channel_offset == 0 && + if (tx_usecs != rx_usecs && efx->tx_channel_offset == 0 && !rx_may_override_tx) { netif_err(efx, drv, efx->net_dev, "Channels are shared. " "RX and TX IRQ moderation must be equal\n"); @@ -1987,12 +1996,12 @@ int efx_init_irq_moderation(struct efx_nic *efx, unsigned int tx_usecs, } efx->irq_rx_adaptive = rx_adaptive; - efx->irq_rx_moderation = rx_ticks; + efx->irq_rx_moderation_us = rx_usecs; efx_for_each_channel(channel, efx) { if (efx_channel_has_rx_queue(channel)) - channel->irq_moderation = rx_ticks; + channel->irq_moderation_us = rx_usecs; else if (efx_channel_has_tx_queues(channel)) - channel->irq_moderation = tx_ticks; + channel->irq_moderation_us = tx_usecs; } return 0; @@ -2001,26 +2010,21 @@ int efx_init_irq_moderation(struct efx_nic *efx, unsigned int tx_usecs, void efx_get_irq_moderation(struct efx_nic *efx, unsigned int *tx_usecs, unsigned int *rx_usecs, bool *rx_adaptive) { - /* We must round up when converting ticks to microseconds - * because we round down when converting the other way. - */ - *rx_adaptive = efx->irq_rx_adaptive; - *rx_usecs = DIV_ROUND_UP(efx->irq_rx_moderation * - efx->timer_quantum_ns, - 1000); + *rx_usecs = efx->irq_rx_moderation_us; /* If channels are shared between RX and TX, so is IRQ * moderation. Otherwise, IRQ moderation is the same for all * TX channels and is not adaptive. */ - if (efx->tx_channel_offset == 0) + if (efx->tx_channel_offset == 0) { *tx_usecs = *rx_usecs; - else - *tx_usecs = DIV_ROUND_UP( - efx->channel[efx->tx_channel_offset]->irq_moderation * - efx->timer_quantum_ns, - 1000); + } else { + struct efx_channel *tx_channel; + + tx_channel = efx->channel[efx->tx_channel_offset]; + *tx_usecs = tx_channel->irq_moderation_us; + } } /************************************************************************** diff --git a/drivers/net/ethernet/sfc/efx.h b/drivers/net/ethernet/sfc/efx.h index c3ae739e9c7a..342ae16e1f2d 100644 --- a/drivers/net/ethernet/sfc/efx.h +++ b/drivers/net/ethernet/sfc/efx.h @@ -204,6 +204,8 @@ int efx_try_recovery(struct efx_nic *efx); /* Global */ void efx_schedule_reset(struct efx_nic *efx, enum reset_type type); +unsigned int efx_usecs_to_ticks(struct efx_nic *efx, unsigned int usecs); +unsigned int efx_ticks_to_usecs(struct efx_nic *efx, unsigned int ticks); int efx_init_irq_moderation(struct efx_nic *efx, unsigned int tx_usecs, unsigned int rx_usecs, bool rx_adaptive, bool rx_may_override_tx); diff --git a/drivers/net/ethernet/sfc/falcon.c b/drivers/net/ethernet/sfc/falcon.c index d790cb8d9db3..1a7092602aec 100644 --- a/drivers/net/ethernet/sfc/falcon.c +++ b/drivers/net/ethernet/sfc/falcon.c @@ -378,12 +378,15 @@ static void falcon_push_irq_moderation(struct efx_channel *channel) struct efx_nic *efx = channel->efx; /* Set timer register */ - if (channel->irq_moderation) { + if (channel->irq_moderation_us) { + unsigned int ticks; + + ticks = efx_usecs_to_ticks(efx, channel->irq_moderation_us); EFX_POPULATE_DWORD_2(timer_cmd, FRF_AB_TC_TIMER_MODE, FFE_BB_TIMER_MODE_INT_HLDOFF, FRF_AB_TC_TIMER_VAL, - channel->irq_moderation - 1); + ticks - 1); } else { EFX_POPULATE_DWORD_2(timer_cmd, FRF_AB_TC_TIMER_MODE, @@ -2373,6 +2376,8 @@ static int falcon_probe_nic(struct efx_nic *efx) EFX_MAX_CHANNELS); efx->max_tx_channels = efx->max_channels; efx->timer_quantum_ns = 4968; /* 621 cycles */ + efx->timer_max_ns = efx->type->timer_period_max * + efx->timer_quantum_ns; /* Initialise I2C adapter */ board = falcon_board(efx); diff --git a/drivers/net/ethernet/sfc/mcdi.c b/drivers/net/ethernet/sfc/mcdi.c index d28e7dd8fa3c..9fbc12a8f80c 100644 --- a/drivers/net/ethernet/sfc/mcdi.c +++ b/drivers/net/ethernet/sfc/mcdi.c @@ -548,7 +548,10 @@ static bool efx_mcdi_complete_async(struct efx_mcdi_iface *mcdi, bool timeout) efx_mcdi_display_error(efx, async->cmd, async->inlen, errbuf, err_len, rc); } - async->complete(efx, async->cookie, rc, outbuf, data_len); + + if (async->complete) + async->complete(efx, async->cookie, rc, outbuf, + min(async->outlen, data_len)); kfree(async); efx_mcdi_release(mcdi); diff --git a/drivers/net/ethernet/sfc/mcdi_pcol.h b/drivers/net/ethernet/sfc/mcdi_pcol.h index c9a5b003caaf..ccceafc15896 100644 --- a/drivers/net/ethernet/sfc/mcdi_pcol.h +++ b/drivers/net/ethernet/sfc/mcdi_pcol.h @@ -2645,16 +2645,20 @@ #define MC_CMD_POLL_BIST_MEM_BUS_MC 0x0 /* enum: CSR IREG bus. */ #define MC_CMD_POLL_BIST_MEM_BUS_CSR 0x1 -/* enum: RX DPCPU bus. */ +/* enum: RX0 DPCPU bus. */ #define MC_CMD_POLL_BIST_MEM_BUS_DPCPU_RX 0x2 /* enum: TX0 DPCPU bus. */ #define MC_CMD_POLL_BIST_MEM_BUS_DPCPU_TX0 0x3 /* enum: TX1 DPCPU bus. */ #define MC_CMD_POLL_BIST_MEM_BUS_DPCPU_TX1 0x4 -/* enum: RX DICPU bus. */ +/* enum: RX0 DICPU bus. */ #define MC_CMD_POLL_BIST_MEM_BUS_DICPU_RX 0x5 /* enum: TX DICPU bus. */ #define MC_CMD_POLL_BIST_MEM_BUS_DICPU_TX 0x6 +/* enum: RX1 DPCPU bus. */ +#define MC_CMD_POLL_BIST_MEM_BUS_DPCPU_RX1 0x7 +/* enum: RX1 DICPU bus. */ +#define MC_CMD_POLL_BIST_MEM_BUS_DICPU_RX1 0x8 /* Pattern written to RAM / register */ #define MC_CMD_POLL_BIST_OUT_MEM_EXPECT_OFST 16 /* Actual value read from RAM / register */ @@ -3612,6 +3616,8 @@ #define MC_CMD_NVRAM_INFO_OUT_PROTECTED_WIDTH 1 #define MC_CMD_NVRAM_INFO_OUT_TLV_LBN 1 #define MC_CMD_NVRAM_INFO_OUT_TLV_WIDTH 1 +#define MC_CMD_NVRAM_INFO_OUT_CMAC_LBN 6 +#define MC_CMD_NVRAM_INFO_OUT_CMAC_WIDTH 1 #define MC_CMD_NVRAM_INFO_OUT_A_B_LBN 7 #define MC_CMD_NVRAM_INFO_OUT_A_B_WIDTH 1 #define MC_CMD_NVRAM_INFO_OUT_PHYSDEV_OFST 16 @@ -4389,6 +4395,8 @@ * the command will fail with MC_CMD_ERR_FILTERS_PRESENT. */ #define MC_CMD_WORKAROUND_BUG26807 0x6 +/* enum: Bug 61265 work around (broken EVQ TMR writes). */ +#define MC_CMD_WORKAROUND_BUG61265 0x7 /* 0 = disable the workaround indicated by TYPE; any non-zero value = enable * the workaround */ @@ -4413,7 +4421,6 @@ * (GET_PHY_CFG_OUT_MEDIA_TYPE); the valid 'page number' input values, and the * output data, are interpreted on a per-type basis. For SFP+: PAGE=0 or 1 * returns a 128-byte block read from module I2C address 0xA0 offset 0 or 0x80. - * Anything else: currently undefined. Locks required: None. Return code: 0. */ #define MC_CMD_GET_PHY_MEDIA_INFO 0x4b @@ -5479,6 +5486,8 @@ #define LICENSED_V3_FEATURES_TX_SNIFF_WIDTH 1 #define LICENSED_V3_FEATURES_PROXY_FILTER_OPS_LBN 8 #define LICENSED_V3_FEATURES_PROXY_FILTER_OPS_WIDTH 1 +#define LICENSED_V3_FEATURES_EVENT_CUT_THROUGH_LBN 9 +#define LICENSED_V3_FEATURES_EVENT_CUT_THROUGH_WIDTH 1 #define LICENSED_V3_FEATURES_MASK_LBN 0 #define LICENSED_V3_FEATURES_MASK_WIDTH 64 @@ -5634,6 +5643,109 @@ /* Only valid if INTRFLAG was true */ #define MC_CMD_INIT_EVQ_OUT_IRQ_OFST 0 +/* MC_CMD_INIT_EVQ_V2_IN msgrequest */ +#define MC_CMD_INIT_EVQ_V2_IN_LENMIN 44 +#define MC_CMD_INIT_EVQ_V2_IN_LENMAX 548 +#define MC_CMD_INIT_EVQ_V2_IN_LEN(num) (36+8*(num)) +/* Size, in entries */ +#define MC_CMD_INIT_EVQ_V2_IN_SIZE_OFST 0 +/* Desired instance. Must be set to a specific instance, which is a function + * local queue index. + */ +#define MC_CMD_INIT_EVQ_V2_IN_INSTANCE_OFST 4 +/* The initial timer value. The load value is ignored if the timer mode is DIS. + */ +#define MC_CMD_INIT_EVQ_V2_IN_TMR_LOAD_OFST 8 +/* The reload value is ignored in one-shot modes */ +#define MC_CMD_INIT_EVQ_V2_IN_TMR_RELOAD_OFST 12 +/* tbd */ +#define MC_CMD_INIT_EVQ_V2_IN_FLAGS_OFST 16 +#define MC_CMD_INIT_EVQ_V2_IN_FLAG_INTERRUPTING_LBN 0 +#define MC_CMD_INIT_EVQ_V2_IN_FLAG_INTERRUPTING_WIDTH 1 +#define MC_CMD_INIT_EVQ_V2_IN_FLAG_RPTR_DOS_LBN 1 +#define MC_CMD_INIT_EVQ_V2_IN_FLAG_RPTR_DOS_WIDTH 1 +#define MC_CMD_INIT_EVQ_V2_IN_FLAG_INT_ARMD_LBN 2 +#define MC_CMD_INIT_EVQ_V2_IN_FLAG_INT_ARMD_WIDTH 1 +#define MC_CMD_INIT_EVQ_V2_IN_FLAG_CUT_THRU_LBN 3 +#define MC_CMD_INIT_EVQ_V2_IN_FLAG_CUT_THRU_WIDTH 1 +#define MC_CMD_INIT_EVQ_V2_IN_FLAG_RX_MERGE_LBN 4 +#define MC_CMD_INIT_EVQ_V2_IN_FLAG_RX_MERGE_WIDTH 1 +#define MC_CMD_INIT_EVQ_V2_IN_FLAG_TX_MERGE_LBN 5 +#define MC_CMD_INIT_EVQ_V2_IN_FLAG_TX_MERGE_WIDTH 1 +#define MC_CMD_INIT_EVQ_V2_IN_FLAG_USE_TIMER_LBN 6 +#define MC_CMD_INIT_EVQ_V2_IN_FLAG_USE_TIMER_WIDTH 1 +#define MC_CMD_INIT_EVQ_V2_IN_FLAG_TYPE_LBN 7 +#define MC_CMD_INIT_EVQ_V2_IN_FLAG_TYPE_WIDTH 4 +/* enum: All initialisation flags specified by host. */ +#define MC_CMD_INIT_EVQ_V2_IN_FLAG_TYPE_MANUAL 0x0 +/* enum: MEDFORD only. Certain initialisation flags specified by host may be + * over-ridden by firmware based on licenses and firmware variant in order to + * provide the lowest latency achievable. See + * MC_CMD_INIT_EVQ_V2/MC_CMD_INIT_EVQ_V2_OUT/FLAGS for list of affected flags. + */ +#define MC_CMD_INIT_EVQ_V2_IN_FLAG_TYPE_LOW_LATENCY 0x1 +/* enum: MEDFORD only. Certain initialisation flags specified by host may be + * over-ridden by firmware based on licenses and firmware variant in order to + * provide the best throughput achievable. See + * MC_CMD_INIT_EVQ_V2/MC_CMD_INIT_EVQ_V2_OUT/FLAGS for list of affected flags. + */ +#define MC_CMD_INIT_EVQ_V2_IN_FLAG_TYPE_THROUGHPUT 0x2 +/* enum: MEDFORD only. Certain initialisation flags may be over-ridden by + * firmware based on licenses and firmware variant. See + * MC_CMD_INIT_EVQ_V2/MC_CMD_INIT_EVQ_V2_OUT/FLAGS for list of affected flags. + */ +#define MC_CMD_INIT_EVQ_V2_IN_FLAG_TYPE_AUTO 0x3 +#define MC_CMD_INIT_EVQ_V2_IN_TMR_MODE_OFST 20 +/* enum: Disabled */ +#define MC_CMD_INIT_EVQ_V2_IN_TMR_MODE_DIS 0x0 +/* enum: Immediate */ +#define MC_CMD_INIT_EVQ_V2_IN_TMR_IMMED_START 0x1 +/* enum: Triggered */ +#define MC_CMD_INIT_EVQ_V2_IN_TMR_TRIG_START 0x2 +/* enum: Hold-off */ +#define MC_CMD_INIT_EVQ_V2_IN_TMR_INT_HLDOFF 0x3 +/* Target EVQ for wakeups if in wakeup mode. */ +#define MC_CMD_INIT_EVQ_V2_IN_TARGET_EVQ_OFST 24 +/* Target interrupt if in interrupting mode (note union with target EVQ). Use + * MC_CMD_RESOURCE_INSTANCE_ANY unless a specific one required for test + * purposes. + */ +#define MC_CMD_INIT_EVQ_V2_IN_IRQ_NUM_OFST 24 +/* Event Counter Mode. */ +#define MC_CMD_INIT_EVQ_V2_IN_COUNT_MODE_OFST 28 +/* enum: Disabled */ +#define MC_CMD_INIT_EVQ_V2_IN_COUNT_MODE_DIS 0x0 +/* enum: Disabled */ +#define MC_CMD_INIT_EVQ_V2_IN_COUNT_MODE_RX 0x1 +/* enum: Disabled */ +#define MC_CMD_INIT_EVQ_V2_IN_COUNT_MODE_TX 0x2 +/* enum: Disabled */ +#define MC_CMD_INIT_EVQ_V2_IN_COUNT_MODE_RXTX 0x3 +/* Event queue packet count threshold. */ +#define MC_CMD_INIT_EVQ_V2_IN_COUNT_THRSHLD_OFST 32 +/* 64-bit address of 4k of 4k-aligned host memory buffer */ +#define MC_CMD_INIT_EVQ_V2_IN_DMA_ADDR_OFST 36 +#define MC_CMD_INIT_EVQ_V2_IN_DMA_ADDR_LEN 8 +#define MC_CMD_INIT_EVQ_V2_IN_DMA_ADDR_LO_OFST 36 +#define MC_CMD_INIT_EVQ_V2_IN_DMA_ADDR_HI_OFST 40 +#define MC_CMD_INIT_EVQ_V2_IN_DMA_ADDR_MINNUM 1 +#define MC_CMD_INIT_EVQ_V2_IN_DMA_ADDR_MAXNUM 64 + +/* MC_CMD_INIT_EVQ_V2_OUT msgresponse */ +#define MC_CMD_INIT_EVQ_V2_OUT_LEN 8 +/* Only valid if INTRFLAG was true */ +#define MC_CMD_INIT_EVQ_V2_OUT_IRQ_OFST 0 +/* Actual configuration applied on the card */ +#define MC_CMD_INIT_EVQ_V2_OUT_FLAGS_OFST 4 +#define MC_CMD_INIT_EVQ_V2_OUT_FLAG_CUT_THRU_LBN 0 +#define MC_CMD_INIT_EVQ_V2_OUT_FLAG_CUT_THRU_WIDTH 1 +#define MC_CMD_INIT_EVQ_V2_OUT_FLAG_RX_MERGE_LBN 1 +#define MC_CMD_INIT_EVQ_V2_OUT_FLAG_RX_MERGE_WIDTH 1 +#define MC_CMD_INIT_EVQ_V2_OUT_FLAG_TX_MERGE_LBN 2 +#define MC_CMD_INIT_EVQ_V2_OUT_FLAG_TX_MERGE_WIDTH 1 +#define MC_CMD_INIT_EVQ_V2_OUT_FLAG_RXQ_FORCE_EV_MERGING_LBN 3 +#define MC_CMD_INIT_EVQ_V2_OUT_FLAG_RXQ_FORCE_EV_MERGING_WIDTH 1 + /* QUEUE_CRC_MODE structuredef */ #define QUEUE_CRC_MODE_LEN 1 #define QUEUE_CRC_MODE_MODE_LBN 0 @@ -5697,8 +5809,8 @@ #define MC_CMD_INIT_RXQ_IN_FLAG_PREFIX_WIDTH 1 #define MC_CMD_INIT_RXQ_IN_FLAG_DISABLE_SCATTER_LBN 9 #define MC_CMD_INIT_RXQ_IN_FLAG_DISABLE_SCATTER_WIDTH 1 -#define MC_CMD_INIT_RXQ_IN_FLAG_FORCE_EV_MERGING_LBN 10 -#define MC_CMD_INIT_RXQ_IN_FLAG_FORCE_EV_MERGING_WIDTH 1 +#define MC_CMD_INIT_RXQ_IN_UNUSED_LBN 10 +#define MC_CMD_INIT_RXQ_IN_UNUSED_WIDTH 1 /* Owner ID to use if in buffer mode (zero if physical) */ #define MC_CMD_INIT_RXQ_IN_OWNER_ID_OFST 20 /* The port ID associated with the v-adaptor which should contain this DMAQ. */ @@ -7854,6 +7966,20 @@ #define MC_CMD_GET_CAPABILITIES_V2_OUT_EVENT_CUT_THROUGH_WIDTH 1 #define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_CUT_THROUGH_LBN 4 #define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_CUT_THROUGH_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_VFIFO_ULL_MODE_LBN 5 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_VFIFO_ULL_MODE_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_MAC_STATS_40G_TX_SIZE_BINS_LBN 6 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_MAC_STATS_40G_TX_SIZE_BINS_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_INIT_EVQ_V2_LBN 7 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_INIT_EVQ_V2_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_MAC_TIMESTAMPING_LBN 8 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_MAC_TIMESTAMPING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TIMESTAMP_LBN 9 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TIMESTAMP_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_SNIFF_LBN 10 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_SNIFF_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_SNIFF_LBN 11 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_SNIFF_WIDTH 1 /* Number of FATSOv2 contexts per datapath supported by this NIC. Not present * on older firmware (check the length). */ @@ -7910,6 +8036,288 @@ #define MC_CMD_GET_CAPABILITIES_V2_OUT_SIZE_PIO_BUFF_OFST 70 #define MC_CMD_GET_CAPABILITIES_V2_OUT_SIZE_PIO_BUFF_LEN 2 +/* MC_CMD_GET_CAPABILITIES_V3_OUT msgresponse */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_LEN 73 +/* First word of flags. */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_FLAGS1_OFST 0 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_VPORT_RECONFIGURE_LBN 3 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_VPORT_RECONFIGURE_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_STRIPING_LBN 4 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_STRIPING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_VADAPTOR_QUERY_LBN 5 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_VADAPTOR_QUERY_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_EVB_PORT_VLAN_RESTRICT_LBN 6 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_EVB_PORT_VLAN_RESTRICT_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_DRV_ATTACH_PREBOOT_LBN 7 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_DRV_ATTACH_PREBOOT_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_FORCE_EVENT_MERGING_LBN 8 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_FORCE_EVENT_MERGING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_SET_MAC_ENHANCED_LBN 9 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_SET_MAC_ENHANCED_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_UNKNOWN_UCAST_DST_FILTER_ALWAYS_MULTI_RECIPIENT_LBN 10 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_UNKNOWN_UCAST_DST_FILTER_ALWAYS_MULTI_RECIPIENT_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_VADAPTOR_PERMIT_SET_MAC_WHEN_FILTERS_INSTALLED_LBN 11 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_VADAPTOR_PERMIT_SET_MAC_WHEN_FILTERS_INSTALLED_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_MAC_SECURITY_FILTERING_LBN 12 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_MAC_SECURITY_FILTERING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_ADDITIONAL_RSS_MODES_LBN 13 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_ADDITIONAL_RSS_MODES_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_QBB_LBN 14 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_QBB_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_PACKED_STREAM_VAR_BUFFERS_LBN 15 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_PACKED_STREAM_VAR_BUFFERS_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_RSS_LIMITED_LBN 16 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_RSS_LIMITED_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_PACKED_STREAM_LBN 17 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_PACKED_STREAM_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_INCLUDE_FCS_LBN 18 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_INCLUDE_FCS_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_VLAN_INSERTION_LBN 19 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_VLAN_INSERTION_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_VLAN_STRIPPING_LBN 20 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_VLAN_STRIPPING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_TSO_LBN 21 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_TSO_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_PREFIX_LEN_0_LBN 22 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_PREFIX_LEN_0_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_PREFIX_LEN_14_LBN 23 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_PREFIX_LEN_14_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_TIMESTAMP_LBN 24 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_TIMESTAMP_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_BATCHING_LBN 25 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_BATCHING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_MCAST_FILTER_CHAINING_LBN 26 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_MCAST_FILTER_CHAINING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_PM_AND_RXDP_COUNTERS_LBN 27 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_PM_AND_RXDP_COUNTERS_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_DISABLE_SCATTER_LBN 28 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_DISABLE_SCATTER_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_MCAST_UDP_LOOPBACK_LBN 29 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_MCAST_UDP_LOOPBACK_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_EVB_LBN 30 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_EVB_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_VXLAN_NVGRE_LBN 31 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_VXLAN_NVGRE_WIDTH 1 +/* RxDPCPU firmware id. */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_DPCPU_FW_ID_OFST 4 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_DPCPU_FW_ID_LEN 2 +/* enum: Standard RXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP 0x0 +/* enum: Low latency RXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_LOW_LATENCY 0x1 +/* enum: Packed stream RXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_PACKED_STREAM 0x2 +/* enum: BIST RXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_BIST 0x10a +/* enum: RXDP Test firmware image 1 */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_TEST_FW_TO_MC_CUT_THROUGH 0x101 +/* enum: RXDP Test firmware image 2 */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_TEST_FW_TO_MC_STORE_FORWARD 0x102 +/* enum: RXDP Test firmware image 3 */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_TEST_FW_TO_MC_STORE_FORWARD_FIRST 0x103 +/* enum: RXDP Test firmware image 4 */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_TEST_EVERY_EVENT_BATCHABLE 0x104 +/* enum: RXDP Test firmware image 5 */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_TEST_BACKPRESSURE 0x105 +/* enum: RXDP Test firmware image 6 */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_TEST_FW_PACKET_EDITS 0x106 +/* enum: RXDP Test firmware image 7 */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_TEST_FW_RX_HDR_SPLIT 0x107 +/* enum: RXDP Test firmware image 8 */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_TEST_FW_DISABLE_DL 0x108 +/* enum: RXDP Test firmware image 9 */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_TEST_FW_DOORBELL_DELAY 0x10b +/* TxDPCPU firmware id. */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_DPCPU_FW_ID_OFST 6 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_DPCPU_FW_ID_LEN 2 +/* enum: Standard TXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXDP 0x0 +/* enum: Low latency TXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXDP_LOW_LATENCY 0x1 +/* enum: High packet rate TXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXDP_HIGH_PACKET_RATE 0x3 +/* enum: BIST TXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXDP_BIST 0x12d +/* enum: TXDP Test firmware image 1 */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXDP_TEST_FW_TSO_EDIT 0x101 +/* enum: TXDP Test firmware image 2 */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXDP_TEST_FW_PACKET_EDITS 0x102 +/* enum: TXDP CSR bus test firmware */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXDP_TEST_FW_CSR 0x103 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_VERSION_OFST 8 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_VERSION_LEN 2 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_VERSION_REV_LBN 0 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_VERSION_REV_WIDTH 12 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_VERSION_TYPE_LBN 12 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_VERSION_TYPE_WIDTH 4 +/* enum: reserved value - do not use (may indicate alternative interpretation + * of REV field in future) + */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_RESERVED 0x0 +/* enum: Trivial RX PD firmware for early Huntington development (Huntington + * development only) + */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_FIRST_PKT 0x1 +/* enum: RX PD firmware with approximately Siena-compatible behaviour + * (Huntington development only) + */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_SIENA_COMPAT 0x2 +/* enum: Virtual switching (full feature) RX PD production firmware */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_VSWITCH 0x3 +/* enum: siena_compat variant RX PD firmware using PM rather than MAC + * (Huntington development only) + */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_SIENA_COMPAT_PM 0x4 +/* enum: Low latency RX PD production firmware */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_LOW_LATENCY 0x5 +/* enum: Packed stream RX PD production firmware */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_PACKED_STREAM 0x6 +/* enum: RX PD firmware handling layer 2 only for high packet rate performance + * tests (Medford development only) + */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_LAYER2_PERF 0x7 +/* enum: Rules engine RX PD production firmware */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_RULES_ENGINE 0x8 +/* enum: RX PD firmware for GUE parsing prototype (Medford development only) */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_TESTFW_GUE_PROTOTYPE 0xe +/* enum: RX PD firmware parsing but not filtering network overlay tunnel + * encapsulations (Medford development only) + */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_TESTFW_ENCAP_PARSING_ONLY 0xf +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_VERSION_OFST 10 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_VERSION_LEN 2 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_VERSION_REV_LBN 0 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_VERSION_REV_WIDTH 12 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_VERSION_TYPE_LBN 12 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_VERSION_TYPE_WIDTH 4 +/* enum: reserved value - do not use (may indicate alternative interpretation + * of REV field in future) + */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_TYPE_RESERVED 0x0 +/* enum: Trivial TX PD firmware for early Huntington development (Huntington + * development only) + */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_TYPE_FIRST_PKT 0x1 +/* enum: TX PD firmware with approximately Siena-compatible behaviour + * (Huntington development only) + */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_TYPE_SIENA_COMPAT 0x2 +/* enum: Virtual switching (full feature) TX PD production firmware */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_TYPE_VSWITCH 0x3 +/* enum: siena_compat variant TX PD firmware using PM rather than MAC + * (Huntington development only) + */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_TYPE_SIENA_COMPAT_PM 0x4 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_TYPE_LOW_LATENCY 0x5 /* enum */ +/* enum: TX PD firmware handling layer 2 only for high packet rate performance + * tests (Medford development only) + */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_TYPE_LAYER2_PERF 0x7 +/* enum: Rules engine TX PD production firmware */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_TYPE_RULES_ENGINE 0x8 +/* enum: RX PD firmware for GUE parsing prototype (Medford development only) */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_TYPE_TESTFW_GUE_PROTOTYPE 0xe +/* Hardware capabilities of NIC */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_HW_CAPABILITIES_OFST 12 +/* Licensed capabilities */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_LICENSE_CAPABILITIES_OFST 16 +/* Second word of flags. Not present on older firmware (check the length). */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_FLAGS2_OFST 20 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_TSO_V2_LBN 0 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_TSO_V2_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_TSO_V2_ENCAP_LBN 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_TSO_V2_ENCAP_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_EVQ_TIMER_CTRL_LBN 2 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_EVQ_TIMER_CTRL_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_EVENT_CUT_THROUGH_LBN 3 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_EVENT_CUT_THROUGH_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_CUT_THROUGH_LBN 4 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_CUT_THROUGH_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_VFIFO_ULL_MODE_LBN 5 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_VFIFO_ULL_MODE_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_MAC_STATS_40G_TX_SIZE_BINS_LBN 6 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_MAC_STATS_40G_TX_SIZE_BINS_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_INIT_EVQ_V2_LBN 7 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_INIT_EVQ_V2_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_MAC_TIMESTAMPING_LBN 8 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_MAC_TIMESTAMPING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_TIMESTAMP_LBN 9 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_TIMESTAMP_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_SNIFF_LBN 10 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_SNIFF_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_SNIFF_LBN 11 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_SNIFF_WIDTH 1 +/* Number of FATSOv2 contexts per datapath supported by this NIC. Not present + * on older firmware (check the length). + */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_TSO_V2_N_CONTEXTS_OFST 24 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_TSO_V2_N_CONTEXTS_LEN 2 +/* One byte per PF containing the number of the external port assigned to this + * PF, indexed by PF number. Special values indicate that a PF is either not + * present or not assigned. + */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_PFS_TO_PORTS_ASSIGNMENT_OFST 26 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_PFS_TO_PORTS_ASSIGNMENT_LEN 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_PFS_TO_PORTS_ASSIGNMENT_NUM 16 +/* enum: The caller is not permitted to access information on this PF. */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_ACCESS_NOT_PERMITTED 0xff +/* enum: PF does not exist. */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_PF_NOT_PRESENT 0xfe +/* enum: PF does exist but is not assigned to any external port. */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_PF_NOT_ASSIGNED 0xfd +/* enum: This value indicates that PF is assigned, but it cannot be expressed + * in this field. It is intended for a possible future situation where a more + * complex scheme of PFs to ports mapping is being used. The future driver + * should look for a new field supporting the new scheme. The current/old + * driver should treat this value as PF_NOT_ASSIGNED. + */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_INCOMPATIBLE_ASSIGNMENT 0xfc +/* One byte per PF containing the number of its VFs, indexed by PF number. A + * special value indicates that a PF is not present. + */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_NUM_VFS_PER_PF_OFST 42 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_NUM_VFS_PER_PF_LEN 1 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_NUM_VFS_PER_PF_NUM 16 +/* enum: The caller is not permitted to access information on this PF. */ +/* MC_CMD_GET_CAPABILITIES_V3_OUT_ACCESS_NOT_PERMITTED 0xff */ +/* enum: PF does not exist. */ +/* MC_CMD_GET_CAPABILITIES_V3_OUT_PF_NOT_PRESENT 0xfe */ +/* Number of VIs available for each external port */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_NUM_VIS_PER_PORT_OFST 58 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_NUM_VIS_PER_PORT_LEN 2 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_NUM_VIS_PER_PORT_NUM 4 +/* Size of RX descriptor cache expressed as binary logarithm The actual size + * equals (2 ^ RX_DESC_CACHE_SIZE) + */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_DESC_CACHE_SIZE_OFST 66 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_DESC_CACHE_SIZE_LEN 1 +/* Size of TX descriptor cache expressed as binary logarithm The actual size + * equals (2 ^ TX_DESC_CACHE_SIZE) + */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_DESC_CACHE_SIZE_OFST 67 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_DESC_CACHE_SIZE_LEN 1 +/* Total number of available PIO buffers */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_NUM_PIO_BUFFS_OFST 68 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_NUM_PIO_BUFFS_LEN 2 +/* Size of a single PIO buffer */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_SIZE_PIO_BUFF_OFST 70 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_SIZE_PIO_BUFF_LEN 2 +/* On chips later than Medford the amount of address space assigned to each VI + * is configurable. This is a global setting that the driver must query to + * discover the VI to address mapping. Cut-through PIO (CTPIO) in not available + * with 8k VI windows. + */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_VI_WINDOW_MODE_OFST 72 +#define MC_CMD_GET_CAPABILITIES_V3_OUT_VI_WINDOW_MODE_LEN 1 +/* enum: Each VI occupies 8k as on Huntington and Medford. PIO is at offset 4k. + * CTPIO is not mapped. + */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_VI_WINDOW_MODE_8K 0x0 +/* enum: Each VI occupies 16k. PIO is at offset 4k. CTPIO is at offset 12k. */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_VI_WINDOW_MODE_16K 0x1 +/* enum: Each VI occupies 64k. PIO is at offset 4k. CTPIO is at offset 12k. */ +#define MC_CMD_GET_CAPABILITIES_V3_OUT_VI_WINDOW_MODE_64K 0x2 + /***********************************/ /* MC_CMD_V2_EXTN @@ -9026,7 +9434,7 @@ */ #define MC_CMD_GET_RXDP_CONFIG 0xc2 -#define MC_CMD_0xc2_PRIVILEGE_CTG SRIOV_CTG_ADMIN +#define MC_CMD_0xc2_PRIVILEGE_CTG SRIOV_CTG_GENERAL /* MC_CMD_GET_RXDP_CONFIG_IN msgrequest */ #define MC_CMD_GET_RXDP_CONFIG_IN_LEN 0 @@ -10125,7 +10533,9 @@ * that this operation returns a zero-length response */ #define MC_CMD_LICENSING_V3_IN_OP_UPDATE_LICENSE 0x0 -/* enum: report counts of installed licenses */ +/* enum: report counts of installed licenses Returns EAGAIN if license + * processing (updating) has been started but not yet completed. + */ #define MC_CMD_LICENSING_V3_IN_OP_REPORT_LICENSE 0x1 /* MC_CMD_LICENSING_V3_OUT msgresponse */ @@ -10763,6 +11173,8 @@ #define MC_CMD_GET_WORKAROUNDS_OUT_BUG42008 0x20 /* enum: Bug 26807 features present in firmware (multicast filter chaining) */ #define MC_CMD_GET_WORKAROUNDS_OUT_BUG26807 0x40 +/* enum: Bug 61265 work around (broken EVQ TMR writes). */ +#define MC_CMD_GET_WORKAROUNDS_OUT_BUG61265 0x80 /***********************************/ @@ -11280,22 +11692,110 @@ #define MC_CMD_0x118_PRIVILEGE_CTG SRIOV_CTG_ADMIN /* MC_CMD_RX_BALANCING_IN msgrequest */ -#define MC_CMD_RX_BALANCING_IN_LEN 4 +#define MC_CMD_RX_BALANCING_IN_LEN 16 /* The RX port whose upconverter table will be modified */ #define MC_CMD_RX_BALANCING_IN_PORT_OFST 0 -#define MC_CMD_RX_BALANCING_IN_PORT_LEN 1 /* The VLAN priority associated to the table index and vFIFO */ -#define MC_CMD_RX_BALANCING_IN_PRIORITY_OFST 1 -#define MC_CMD_RX_BALANCING_IN_PRIORITY_LEN 1 +#define MC_CMD_RX_BALANCING_IN_PRIORITY_OFST 4 /* The resulting bit of SRC^DST for indexing the table */ -#define MC_CMD_RX_BALANCING_IN_SRC_DST_OFST 2 -#define MC_CMD_RX_BALANCING_IN_SRC_DST_LEN 1 +#define MC_CMD_RX_BALANCING_IN_SRC_DST_OFST 8 /* The RX engine to which the vFIFO in the table entry will point to */ -#define MC_CMD_RX_BALANCING_IN_ENG_OFST 3 -#define MC_CMD_RX_BALANCING_IN_ENG_LEN 1 +#define MC_CMD_RX_BALANCING_IN_ENG_OFST 12 /* MC_CMD_RX_BALANCING_OUT msgresponse */ #define MC_CMD_RX_BALANCING_OUT_LEN 0 +/***********************************/ +/* MC_CMD_SET_EVQ_TMR + * Update the timer load, timer reload and timer mode values for a given EVQ. + * The requested timer values (in TMR_LOAD_REQ_NS and TMR_RELOAD_REQ_NS) will + * be rounded up to the granularity supported by the hardware, then truncated + * to the range supported by the hardware. The resulting value after the + * rounding and truncation will be returned to the caller (in TMR_LOAD_ACT_NS + * and TMR_RELOAD_ACT_NS). + */ +#define MC_CMD_SET_EVQ_TMR 0x120 + +#define MC_CMD_0x120_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_SET_EVQ_TMR_IN msgrequest */ +#define MC_CMD_SET_EVQ_TMR_IN_LEN 16 +/* Function-relative queue instance */ +#define MC_CMD_SET_EVQ_TMR_IN_INSTANCE_OFST 0 +/* Requested value for timer load (in nanoseconds) */ +#define MC_CMD_SET_EVQ_TMR_IN_TMR_LOAD_REQ_NS_OFST 4 +/* Requested value for timer reload (in nanoseconds) */ +#define MC_CMD_SET_EVQ_TMR_IN_TMR_RELOAD_REQ_NS_OFST 8 +/* Timer mode. Meanings as per EVQ_TMR_REG.TC_TIMER_VAL */ +#define MC_CMD_SET_EVQ_TMR_IN_TMR_MODE_OFST 12 +#define MC_CMD_SET_EVQ_TMR_IN_TIMER_MODE_DIS 0x0 /* enum */ +#define MC_CMD_SET_EVQ_TMR_IN_TIMER_MODE_IMMED_START 0x1 /* enum */ +#define MC_CMD_SET_EVQ_TMR_IN_TIMER_MODE_TRIG_START 0x2 /* enum */ +#define MC_CMD_SET_EVQ_TMR_IN_TIMER_MODE_INT_HLDOFF 0x3 /* enum */ + +/* MC_CMD_SET_EVQ_TMR_OUT msgresponse */ +#define MC_CMD_SET_EVQ_TMR_OUT_LEN 8 +/* Actual value for timer load (in nanoseconds) */ +#define MC_CMD_SET_EVQ_TMR_OUT_TMR_LOAD_ACT_NS_OFST 0 +/* Actual value for timer reload (in nanoseconds) */ +#define MC_CMD_SET_EVQ_TMR_OUT_TMR_RELOAD_ACT_NS_OFST 4 + + +/***********************************/ +/* MC_CMD_GET_EVQ_TMR_PROPERTIES + * Query properties about the event queue timers. + */ +#define MC_CMD_GET_EVQ_TMR_PROPERTIES 0x122 + +#define MC_CMD_0x122_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_GET_EVQ_TMR_PROPERTIES_IN msgrequest */ +#define MC_CMD_GET_EVQ_TMR_PROPERTIES_IN_LEN 0 + +/* MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT msgresponse */ +#define MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_LEN 36 +/* Reserved for future use. */ +#define MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_FLAGS_OFST 0 +/* For timers updated via writes to EVQ_TMR_REG, this is the time interval (in + * nanoseconds) for each increment of the timer load/reload count. The + * requested duration of a timer is this value multiplied by the timer + * load/reload count. + */ +#define MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_TMR_REG_NS_PER_COUNT_OFST 4 +/* For timers updated via writes to EVQ_TMR_REG, this is the maximum value + * allowed for timer load/reload counts. + */ +#define MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_TMR_REG_MAX_COUNT_OFST 8 +/* For timers updated via writes to EVQ_TMR_REG, timer load/reload counts not a + * multiple of this step size will be rounded in an implementation defined + * manner. + */ +#define MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_TMR_REG_STEP_OFST 12 +/* Maximum timer duration (in nanoseconds) for timers updated via MCDI. Only + * meaningful if MC_CMD_SET_EVQ_TMR is implemented. + */ +#define MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_MCDI_TMR_MAX_NS_OFST 16 +/* Timer durations requested via MCDI that are not a multiple of this step size + * will be rounded up. Only meaningful if MC_CMD_SET_EVQ_TMR is implemented. + */ +#define MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_MCDI_TMR_STEP_NS_OFST 20 +/* For timers updated using the bug35388 workaround, this is the time interval + * (in nanoseconds) for each increment of the timer load/reload count. The + * requested duration of a timer is this value multiplied by the timer + * load/reload count. This field is only meaningful if the bug35388 workaround + * is enabled. + */ +#define MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_BUG35388_TMR_NS_PER_COUNT_OFST 24 +/* For timers updated using the bug35388 workaround, this is the maximum value + * allowed for timer load/reload counts. This field is only meaningful if the + * bug35388 workaround is enabled. + */ +#define MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_BUG35388_TMR_MAX_COUNT_OFST 28 +/* For timers updated using the bug35388 workaround, timer load/reload counts + * not a multiple of this step size will be rounded in an implementation + * defined manner. This field is only meaningful if the bug35388 workaround is + * enabled. + */ +#define MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_BUG35388_TMR_STEP_OFST 32 #endif /* MCDI_PCOL_H */ diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h index 9ff062a36ea8..13b7f52e6724 100644 --- a/drivers/net/ethernet/sfc/net_driver.h +++ b/drivers/net/ethernet/sfc/net_driver.h @@ -392,7 +392,7 @@ enum efx_sync_events_state { * @eventq_init: Event queue initialised flag * @enabled: Channel enabled indicator * @irq: IRQ number (MSI and MSI-X only) - * @irq_moderation: IRQ moderation value (in hardware ticks) + * @irq_moderation_us: IRQ moderation value (in microseconds) * @napi_dev: Net device used with NAPI * @napi_str: NAPI control structure * @state: state for NAPI vs busy polling @@ -433,7 +433,7 @@ struct efx_channel { bool eventq_init; bool enabled; int irq; - unsigned int irq_moderation; + unsigned int irq_moderation_us; struct net_device *napi_dev; struct napi_struct napi_str; #ifdef CONFIG_NET_RX_BUSY_POLL @@ -810,8 +810,10 @@ struct vfdi_status; * @membase: Memory BAR value * @interrupt_mode: Interrupt mode * @timer_quantum_ns: Interrupt timer quantum, in nanoseconds + * @timer_max_ns: Interrupt timer maximum value, in nanoseconds * @irq_rx_adaptive: Adaptive IRQ moderation enabled for RX event queues - * @irq_rx_moderation: IRQ moderation time for RX event queues + * @irq_rx_mod_step_us: Step size for IRQ moderation for RX event queues + * @irq_rx_moderation_us: IRQ moderation time for RX event queues * @msg_enable: Log message enable flags * @state: Device state number (%STATE_*). Serialised by the rtnl_lock. * @reset_pending: Bitmask for pending resets @@ -940,8 +942,10 @@ struct efx_nic { enum efx_int_mode interrupt_mode; unsigned int timer_quantum_ns; + unsigned int timer_max_ns; bool irq_rx_adaptive; - unsigned int irq_rx_moderation; + unsigned int irq_mod_step_us; + unsigned int irq_rx_moderation_us; u32 msg_enable; enum nic_state state; diff --git a/drivers/net/ethernet/sfc/nic.h b/drivers/net/ethernet/sfc/nic.h index 96944c3c9d14..d8b1694638cd 100644 --- a/drivers/net/ethernet/sfc/nic.h +++ b/drivers/net/ethernet/sfc/nic.h @@ -507,10 +507,13 @@ enum { * @stats: Hardware statistics * @workaround_35388: Flag: firmware supports workaround for bug 35388 * @workaround_26807: Flag: firmware supports workaround for bug 26807 + * @workaround_61265: Flag: firmware supports workaround for bug 61265 * @must_check_datapath_caps: Flag: @datapath_caps needs to be revalidated * after MC reboot * @datapath_caps: Capabilities of datapath firmware (FLAGS1 field of * %MC_CMD_GET_CAPABILITIES response) + * @datapath_caps2: Further Capabilities of datapath firmware (FLAGS2 field of + * %MC_CMD_GET_CAPABILITIES response) * @rx_dpcpu_fw_id: Firmware ID of the RxDPCPU * @tx_dpcpu_fw_id: Firmware ID of the TxDPCPU * @vport_id: The function's vport ID, only relevant for PFs @@ -540,8 +543,10 @@ struct efx_ef10_nic_data { u64 stats[EF10_STAT_COUNT]; bool workaround_35388; bool workaround_26807; + bool workaround_61265; bool must_check_datapath_caps; u32 datapath_caps; + u32 datapath_caps2; unsigned int rx_dpcpu_fw_id; unsigned int tx_dpcpu_fw_id; unsigned int vport_id; diff --git a/drivers/net/ethernet/sfc/ptp.c b/drivers/net/ethernet/sfc/ptp.c index c771e0af4e06..dd204d9704c6 100644 --- a/drivers/net/ethernet/sfc/ptp.c +++ b/drivers/net/ethernet/sfc/ptp.c @@ -1306,7 +1306,7 @@ static int efx_ptp_probe_channel(struct efx_channel *channel) { struct efx_nic *efx = channel->efx; - channel->irq_moderation = 0; + channel->irq_moderation_us = 0; channel->rx_queue.core_index = 0; return efx_ptp_probe(efx, channel); diff --git a/drivers/net/ethernet/sfc/siena.c b/drivers/net/ethernet/sfc/siena.c index 2219b5424d2b..04ed1b4c7cd9 100644 --- a/drivers/net/ethernet/sfc/siena.c +++ b/drivers/net/ethernet/sfc/siena.c @@ -34,19 +34,24 @@ static void siena_init_wol(struct efx_nic *efx); static void siena_push_irq_moderation(struct efx_channel *channel) { + struct efx_nic *efx = channel->efx; efx_dword_t timer_cmd; - if (channel->irq_moderation) + if (channel->irq_moderation_us) { + unsigned int ticks; + + ticks = efx_usecs_to_ticks(efx, channel->irq_moderation_us); EFX_POPULATE_DWORD_2(timer_cmd, FRF_CZ_TC_TIMER_MODE, FFE_CZ_TIMER_MODE_INT_HLDOFF, FRF_CZ_TC_TIMER_VAL, - channel->irq_moderation - 1); - else + ticks - 1); + } else { EFX_POPULATE_DWORD_2(timer_cmd, FRF_CZ_TC_TIMER_MODE, FFE_CZ_TIMER_MODE_DIS, FRF_CZ_TC_TIMER_VAL, 0); + } efx_writed_page_locked(channel->efx, &timer_cmd, FR_BZ_TIMER_COMMAND_P0, channel->channel); } @@ -222,6 +227,9 @@ static int siena_probe_nvconfig(struct efx_nic *efx) efx->timer_quantum_ns = (caps & (1 << MC_CMD_CAPABILITIES_TURBO_ACTIVE_LBN)) ? 3072 : 6144; /* 768 cycles */ + efx->timer_max_ns = efx->type->timer_period_max * + efx->timer_quantum_ns; + return rc; } diff --git a/drivers/net/ethernet/sfc/workarounds.h b/drivers/net/ethernet/sfc/workarounds.h index 2310b75d4ec2..351cd14cb9f9 100644 --- a/drivers/net/ethernet/sfc/workarounds.h +++ b/drivers/net/ethernet/sfc/workarounds.h @@ -50,4 +50,8 @@ #define EFX_WORKAROUND_35388(efx) \ (efx_nic_rev(efx) == EFX_REV_HUNT_A0 && EFX_EF10_WORKAROUND_35388(efx)) +/* Moderation timer access must go through MCDI */ +#define EFX_EF10_WORKAROUND_61265(efx) \ + (((struct efx_ef10_nic_data *)efx->nic_data)->workaround_61265) + #endif /* EFX_WORKAROUNDS_H */ diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index c51f34693eae..b4d3b410c32a 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -140,9 +140,10 @@ do { \ #define CPSW_CMINTMAX_INTVL (1000 / CPSW_CMINTMIN_CNT) #define CPSW_CMINTMIN_INTVL ((1000 / CPSW_CMINTMAX_CNT) + 1) -#define cpsw_slave_index(priv) \ - ((priv->data.dual_emac) ? priv->emac_port : \ - priv->data.active_slave) +#define cpsw_slave_index(cpsw, priv) \ + ((cpsw->data.dual_emac) ? priv->emac_port : \ + cpsw->data.active_slave) +#define IRQ_NUM 2 static int debug_level; module_param(debug_level, int, 0); @@ -363,38 +364,39 @@ static inline void slave_write(struct cpsw_slave *slave, u32 val, u32 offset) __raw_writel(val, slave->regs + offset); } -struct cpsw_priv { - struct platform_device *pdev; - struct net_device *ndev; - struct napi_struct napi_rx; - struct napi_struct napi_tx; +struct cpsw_common { struct device *dev; struct cpsw_platform_data data; + struct napi_struct napi_rx; + struct napi_struct napi_tx; struct cpsw_ss_regs __iomem *regs; struct cpsw_wr_regs __iomem *wr_regs; u8 __iomem *hw_stats; struct cpsw_host_regs __iomem *host_port_regs; - u32 msg_enable; u32 version; u32 coal_intvl; u32 bus_freq_mhz; int rx_packet_max; - struct clk *clk; - u8 mac_addr[ETH_ALEN]; struct cpsw_slave *slaves; struct cpdma_ctlr *dma; struct cpdma_chan *txch, *rxch; struct cpsw_ale *ale; - bool rx_pause; - bool tx_pause; bool quirk_irq; bool rx_irq_disabled; bool tx_irq_disabled; - /* snapshot of IRQ numbers */ - u32 irqs_table[4]; - u32 num_irqs; - struct cpts *cpts; + u32 irqs_table[IRQ_NUM]; + struct cpts *cpts; +}; + +struct cpsw_priv { + struct net_device *ndev; + struct device *dev; + u32 msg_enable; + u8 mac_addr[ETH_ALEN]; + bool rx_pause; + bool tx_pause; u32 emac_port; + struct cpsw_common *cpsw; }; struct cpsw_stats { @@ -485,78 +487,71 @@ static const struct cpsw_stats cpsw_gstrings_stats[] = { #define CPSW_STATS_LEN ARRAY_SIZE(cpsw_gstrings_stats) -#define napi_to_priv(napi) container_of(napi, struct cpsw_priv, napi) +#define ndev_to_cpsw(ndev) (((struct cpsw_priv *)netdev_priv(ndev))->cpsw) +#define napi_to_cpsw(napi) container_of(napi, struct cpsw_common, napi) #define for_each_slave(priv, func, arg...) \ do { \ struct cpsw_slave *slave; \ + struct cpsw_common *cpsw = (priv)->cpsw; \ int n; \ - if (priv->data.dual_emac) \ - (func)((priv)->slaves + priv->emac_port, ##arg);\ + if (cpsw->data.dual_emac) \ + (func)((cpsw)->slaves + priv->emac_port, ##arg);\ else \ - for (n = (priv)->data.slaves, \ - slave = (priv)->slaves; \ + for (n = cpsw->data.slaves, \ + slave = cpsw->slaves; \ n; n--) \ (func)(slave++, ##arg); \ } while (0) -#define cpsw_get_slave_ndev(priv, __slave_no__) \ - ((__slave_no__ < priv->data.slaves) ? \ - priv->slaves[__slave_no__].ndev : NULL) -#define cpsw_get_slave_priv(priv, __slave_no__) \ - (((__slave_no__ < priv->data.slaves) && \ - (priv->slaves[__slave_no__].ndev)) ? \ - netdev_priv(priv->slaves[__slave_no__].ndev) : NULL) \ - -#define cpsw_dual_emac_src_port_detect(status, priv, ndev, skb) \ + +#define cpsw_dual_emac_src_port_detect(cpsw, status, ndev, skb) \ do { \ - if (!priv->data.dual_emac) \ + if (!cpsw->data.dual_emac) \ break; \ if (CPDMA_RX_SOURCE_PORT(status) == 1) { \ - ndev = cpsw_get_slave_ndev(priv, 0); \ - priv = netdev_priv(ndev); \ + ndev = cpsw->slaves[0].ndev; \ skb->dev = ndev; \ } else if (CPDMA_RX_SOURCE_PORT(status) == 2) { \ - ndev = cpsw_get_slave_ndev(priv, 1); \ - priv = netdev_priv(ndev); \ + ndev = cpsw->slaves[1].ndev; \ skb->dev = ndev; \ } \ } while (0) -#define cpsw_add_mcast(priv, addr) \ +#define cpsw_add_mcast(cpsw, priv, addr) \ do { \ - if (priv->data.dual_emac) { \ - struct cpsw_slave *slave = priv->slaves + \ + if (cpsw->data.dual_emac) { \ + struct cpsw_slave *slave = cpsw->slaves + \ priv->emac_port; \ - int slave_port = cpsw_get_slave_port(priv, \ + int slave_port = cpsw_get_slave_port( \ slave->slave_num); \ - cpsw_ale_add_mcast(priv->ale, addr, \ + cpsw_ale_add_mcast(cpsw->ale, addr, \ 1 << slave_port | ALE_PORT_HOST, \ ALE_VLAN, slave->port_vlan, 0); \ } else { \ - cpsw_ale_add_mcast(priv->ale, addr, \ + cpsw_ale_add_mcast(cpsw->ale, addr, \ ALE_ALL_PORTS, \ 0, 0, 0); \ } \ } while (0) -static inline int cpsw_get_slave_port(struct cpsw_priv *priv, u32 slave_num) +static inline int cpsw_get_slave_port(u32 slave_num) { return slave_num + 1; } static void cpsw_set_promiscious(struct net_device *ndev, bool enable) { - struct cpsw_priv *priv = netdev_priv(ndev); - struct cpsw_ale *ale = priv->ale; + struct cpsw_common *cpsw = ndev_to_cpsw(ndev); + struct cpsw_ale *ale = cpsw->ale; int i; - if (priv->data.dual_emac) { + if (cpsw->data.dual_emac) { bool flag = false; /* Enabling promiscuous mode for one interface will be * common for both the interface as the interface shares * the same hardware resource. */ - for (i = 0; i < priv->data.slaves; i++) - if (priv->slaves[i].ndev->flags & IFF_PROMISC) + for (i = 0; i < cpsw->data.slaves; i++) + if (cpsw->slaves[i].ndev->flags & IFF_PROMISC) flag = true; if (!enable && flag) { @@ -579,7 +574,7 @@ static void cpsw_set_promiscious(struct net_device *ndev, bool enable) unsigned long timeout = jiffies + HZ; /* Disable Learn for all ports (host is port 0 and slaves are port 1 and up */ - for (i = 0; i <= priv->data.slaves; i++) { + for (i = 0; i <= cpsw->data.slaves; i++) { cpsw_ale_control_set(ale, i, ALE_PORT_NOLEARN, 1); cpsw_ale_control_set(ale, i, @@ -606,7 +601,7 @@ static void cpsw_set_promiscious(struct net_device *ndev, bool enable) cpsw_ale_control_set(ale, 0, ALE_P0_UNI_FLOOD, 0); /* Enable Learn for all ports (host is port 0 and slaves are port 1 and up */ - for (i = 0; i <= priv->data.slaves; i++) { + for (i = 0; i <= cpsw->data.slaves; i++) { cpsw_ale_control_set(ale, i, ALE_PORT_NOLEARN, 0); cpsw_ale_control_set(ale, i, @@ -620,17 +615,18 @@ static void cpsw_set_promiscious(struct net_device *ndev, bool enable) static void cpsw_ndo_set_rx_mode(struct net_device *ndev) { struct cpsw_priv *priv = netdev_priv(ndev); + struct cpsw_common *cpsw = priv->cpsw; int vid; - if (priv->data.dual_emac) - vid = priv->slaves[priv->emac_port].port_vlan; + if (cpsw->data.dual_emac) + vid = cpsw->slaves[priv->emac_port].port_vlan; else - vid = priv->data.default_vlan; + vid = cpsw->data.default_vlan; if (ndev->flags & IFF_PROMISC) { /* Enable promiscuous mode */ cpsw_set_promiscious(ndev, true); - cpsw_ale_set_allmulti(priv->ale, IFF_ALLMULTI); + cpsw_ale_set_allmulti(cpsw->ale, IFF_ALLMULTI); return; } else { /* Disable promiscuous mode */ @@ -638,36 +634,36 @@ static void cpsw_ndo_set_rx_mode(struct net_device *ndev) } /* Restore allmulti on vlans if necessary */ - cpsw_ale_set_allmulti(priv->ale, priv->ndev->flags & IFF_ALLMULTI); + cpsw_ale_set_allmulti(cpsw->ale, priv->ndev->flags & IFF_ALLMULTI); /* Clear all mcast from ALE */ - cpsw_ale_flush_multicast(priv->ale, ALE_ALL_PORTS, vid); + cpsw_ale_flush_multicast(cpsw->ale, ALE_ALL_PORTS, vid); if (!netdev_mc_empty(ndev)) { struct netdev_hw_addr *ha; /* program multicast address list into ALE register */ netdev_for_each_mc_addr(ha, ndev) { - cpsw_add_mcast(priv, (u8 *)ha->addr); + cpsw_add_mcast(cpsw, priv, (u8 *)ha->addr); } } } -static void cpsw_intr_enable(struct cpsw_priv *priv) +static void cpsw_intr_enable(struct cpsw_common *cpsw) { - __raw_writel(0xFF, &priv->wr_regs->tx_en); - __raw_writel(0xFF, &priv->wr_regs->rx_en); + __raw_writel(0xFF, &cpsw->wr_regs->tx_en); + __raw_writel(0xFF, &cpsw->wr_regs->rx_en); - cpdma_ctlr_int_ctrl(priv->dma, true); + cpdma_ctlr_int_ctrl(cpsw->dma, true); return; } -static void cpsw_intr_disable(struct cpsw_priv *priv) +static void cpsw_intr_disable(struct cpsw_common *cpsw) { - __raw_writel(0, &priv->wr_regs->tx_en); - __raw_writel(0, &priv->wr_regs->rx_en); + __raw_writel(0, &cpsw->wr_regs->tx_en); + __raw_writel(0, &cpsw->wr_regs->rx_en); - cpdma_ctlr_int_ctrl(priv->dma, false); + cpdma_ctlr_int_ctrl(cpsw->dma, false); return; } @@ -675,14 +671,14 @@ static void cpsw_tx_handler(void *token, int len, int status) { struct sk_buff *skb = token; struct net_device *ndev = skb->dev; - struct cpsw_priv *priv = netdev_priv(ndev); + struct cpsw_common *cpsw = ndev_to_cpsw(ndev); /* Check whether the queue is stopped due to stalled tx dma, if the * queue is stopped then start the queue as we have free desc for tx */ if (unlikely(netif_queue_stopped(ndev))) netif_wake_queue(ndev); - cpts_tx_timestamp(priv->cpts, skb); + cpts_tx_timestamp(cpsw->cpts, skb); ndev->stats.tx_packets++; ndev->stats.tx_bytes += len; dev_kfree_skb_any(skb); @@ -693,19 +689,19 @@ static void cpsw_rx_handler(void *token, int len, int status) struct sk_buff *skb = token; struct sk_buff *new_skb; struct net_device *ndev = skb->dev; - struct cpsw_priv *priv = netdev_priv(ndev); int ret = 0; + struct cpsw_common *cpsw = ndev_to_cpsw(ndev); - cpsw_dual_emac_src_port_detect(status, priv, ndev, skb); + cpsw_dual_emac_src_port_detect(cpsw, status, ndev, skb); if (unlikely(status < 0) || unlikely(!netif_running(ndev))) { bool ndev_status = false; - struct cpsw_slave *slave = priv->slaves; + struct cpsw_slave *slave = cpsw->slaves; int n; - if (priv->data.dual_emac) { + if (cpsw->data.dual_emac) { /* In dual emac mode check for all interfaces */ - for (n = priv->data.slaves; n; n--, slave++) + for (n = cpsw->data.slaves; n; n--, slave++) if (netif_running(slave->ndev)) ndev_status = true; } @@ -726,10 +722,10 @@ static void cpsw_rx_handler(void *token, int len, int status) return; } - new_skb = netdev_alloc_skb_ip_align(ndev, priv->rx_packet_max); + new_skb = netdev_alloc_skb_ip_align(ndev, cpsw->rx_packet_max); if (new_skb) { skb_put(skb, len); - cpts_rx_timestamp(priv->cpts, skb); + cpts_rx_timestamp(cpsw->cpts, skb); skb->protocol = eth_type_trans(skb, ndev); netif_receive_skb(skb); ndev->stats.rx_bytes += len; @@ -740,83 +736,77 @@ static void cpsw_rx_handler(void *token, int len, int status) } requeue: - ret = cpdma_chan_submit(priv->rxch, new_skb, new_skb->data, - skb_tailroom(new_skb), 0); + ret = cpdma_chan_submit(cpsw->rxch, new_skb, new_skb->data, + skb_tailroom(new_skb), 0); if (WARN_ON(ret < 0)) dev_kfree_skb_any(new_skb); } static irqreturn_t cpsw_tx_interrupt(int irq, void *dev_id) { - struct cpsw_priv *priv = dev_id; + struct cpsw_common *cpsw = dev_id; - writel(0, &priv->wr_regs->tx_en); - cpdma_ctlr_eoi(priv->dma, CPDMA_EOI_TX); + writel(0, &cpsw->wr_regs->tx_en); + cpdma_ctlr_eoi(cpsw->dma, CPDMA_EOI_TX); - if (priv->quirk_irq) { - disable_irq_nosync(priv->irqs_table[1]); - priv->tx_irq_disabled = true; + if (cpsw->quirk_irq) { + disable_irq_nosync(cpsw->irqs_table[1]); + cpsw->tx_irq_disabled = true; } - napi_schedule(&priv->napi_tx); + napi_schedule(&cpsw->napi_tx); return IRQ_HANDLED; } static irqreturn_t cpsw_rx_interrupt(int irq, void *dev_id) { - struct cpsw_priv *priv = dev_id; + struct cpsw_common *cpsw = dev_id; - cpdma_ctlr_eoi(priv->dma, CPDMA_EOI_RX); - writel(0, &priv->wr_regs->rx_en); + cpdma_ctlr_eoi(cpsw->dma, CPDMA_EOI_RX); + writel(0, &cpsw->wr_regs->rx_en); - if (priv->quirk_irq) { - disable_irq_nosync(priv->irqs_table[0]); - priv->rx_irq_disabled = true; + if (cpsw->quirk_irq) { + disable_irq_nosync(cpsw->irqs_table[0]); + cpsw->rx_irq_disabled = true; } - napi_schedule(&priv->napi_rx); + napi_schedule(&cpsw->napi_rx); return IRQ_HANDLED; } static int cpsw_tx_poll(struct napi_struct *napi_tx, int budget) { - struct cpsw_priv *priv = napi_to_priv(napi_tx); + struct cpsw_common *cpsw = napi_to_cpsw(napi_tx); int num_tx; - num_tx = cpdma_chan_process(priv->txch, budget); + num_tx = cpdma_chan_process(cpsw->txch, budget); if (num_tx < budget) { napi_complete(napi_tx); - writel(0xff, &priv->wr_regs->tx_en); - if (priv->quirk_irq && priv->tx_irq_disabled) { - priv->tx_irq_disabled = false; - enable_irq(priv->irqs_table[1]); + writel(0xff, &cpsw->wr_regs->tx_en); + if (cpsw->quirk_irq && cpsw->tx_irq_disabled) { + cpsw->tx_irq_disabled = false; + enable_irq(cpsw->irqs_table[1]); } } - if (num_tx) - cpsw_dbg(priv, intr, "poll %d tx pkts\n", num_tx); - return num_tx; } static int cpsw_rx_poll(struct napi_struct *napi_rx, int budget) { - struct cpsw_priv *priv = napi_to_priv(napi_rx); + struct cpsw_common *cpsw = napi_to_cpsw(napi_rx); int num_rx; - num_rx = cpdma_chan_process(priv->rxch, budget); + num_rx = cpdma_chan_process(cpsw->rxch, budget); if (num_rx < budget) { napi_complete(napi_rx); - writel(0xff, &priv->wr_regs->rx_en); - if (priv->quirk_irq && priv->rx_irq_disabled) { - priv->rx_irq_disabled = false; - enable_irq(priv->irqs_table[0]); + writel(0xff, &cpsw->wr_regs->rx_en); + if (cpsw->quirk_irq && cpsw->rx_irq_disabled) { + cpsw->rx_irq_disabled = false; + enable_irq(cpsw->irqs_table[0]); } } - if (num_rx) - cpsw_dbg(priv, intr, "poll %d rx pkts\n", num_rx); - return num_rx; } @@ -849,17 +839,18 @@ static void _cpsw_adjust_link(struct cpsw_slave *slave, struct phy_device *phy = slave->phy; u32 mac_control = 0; u32 slave_port; + struct cpsw_common *cpsw = priv->cpsw; if (!phy) return; - slave_port = cpsw_get_slave_port(priv, slave->slave_num); + slave_port = cpsw_get_slave_port(slave->slave_num); if (phy->link) { - mac_control = priv->data.mac_control; + mac_control = cpsw->data.mac_control; /* enable forwarding */ - cpsw_ale_control_set(priv->ale, slave_port, + cpsw_ale_control_set(cpsw->ale, slave_port, ALE_PORT_STATE, ALE_PORT_STATE_FORWARD); if (phy->speed == 1000) @@ -883,7 +874,7 @@ static void _cpsw_adjust_link(struct cpsw_slave *slave, } else { mac_control = 0; /* disable forwarding */ - cpsw_ale_control_set(priv->ale, slave_port, + cpsw_ale_control_set(cpsw->ale, slave_port, ALE_PORT_STATE, ALE_PORT_STATE_DISABLE); } @@ -915,9 +906,9 @@ static void cpsw_adjust_link(struct net_device *ndev) static int cpsw_get_coalesce(struct net_device *ndev, struct ethtool_coalesce *coal) { - struct cpsw_priv *priv = netdev_priv(ndev); + struct cpsw_common *cpsw = ndev_to_cpsw(ndev); - coal->rx_coalesce_usecs = priv->coal_intvl; + coal->rx_coalesce_usecs = cpsw->coal_intvl; return 0; } @@ -930,11 +921,12 @@ static int cpsw_set_coalesce(struct net_device *ndev, u32 prescale = 0; u32 addnl_dvdr = 1; u32 coal_intvl = 0; + struct cpsw_common *cpsw = priv->cpsw; coal_intvl = coal->rx_coalesce_usecs; - int_ctrl = readl(&priv->wr_regs->int_control); - prescale = priv->bus_freq_mhz * 4; + int_ctrl = readl(&cpsw->wr_regs->int_control); + prescale = cpsw->bus_freq_mhz * 4; if (!coal->rx_coalesce_usecs) { int_ctrl &= ~(CPSW_INTPRESCALE_MASK | CPSW_INTPACEEN); @@ -962,27 +954,18 @@ static int cpsw_set_coalesce(struct net_device *ndev, } num_interrupts = (1000 * addnl_dvdr) / coal_intvl; - writel(num_interrupts, &priv->wr_regs->rx_imax); - writel(num_interrupts, &priv->wr_regs->tx_imax); + writel(num_interrupts, &cpsw->wr_regs->rx_imax); + writel(num_interrupts, &cpsw->wr_regs->tx_imax); int_ctrl |= CPSW_INTPACEEN; int_ctrl &= (~CPSW_INTPRESCALE_MASK); int_ctrl |= (prescale & CPSW_INTPRESCALE_MASK); update_return: - writel(int_ctrl, &priv->wr_regs->int_control); + writel(int_ctrl, &cpsw->wr_regs->int_control); cpsw_notice(priv, timer, "Set coalesce to %d usecs.\n", coal_intvl); - if (priv->data.dual_emac) { - int i; - - for (i = 0; i < priv->data.slaves; i++) { - priv = netdev_priv(priv->slaves[i].ndev); - priv->coal_intvl = coal_intvl; - } - } else { - priv->coal_intvl = coal_intvl; - } + cpsw->coal_intvl = coal_intvl; return 0; } @@ -1016,21 +999,21 @@ static void cpsw_get_strings(struct net_device *ndev, u32 stringset, u8 *data) static void cpsw_get_ethtool_stats(struct net_device *ndev, struct ethtool_stats *stats, u64 *data) { - struct cpsw_priv *priv = netdev_priv(ndev); struct cpdma_chan_stats rx_stats; struct cpdma_chan_stats tx_stats; u32 val; u8 *p; int i; + struct cpsw_common *cpsw = ndev_to_cpsw(ndev); /* Collect Davinci CPDMA stats for Rx and Tx Channel */ - cpdma_chan_get_stats(priv->rxch, &rx_stats); - cpdma_chan_get_stats(priv->txch, &tx_stats); + cpdma_chan_get_stats(cpsw->rxch, &rx_stats); + cpdma_chan_get_stats(cpsw->txch, &tx_stats); for (i = 0; i < CPSW_STATS_LEN; i++) { switch (cpsw_gstrings_stats[i].type) { case CPSW_STATS: - val = readl(priv->hw_stats + + val = readl(cpsw->hw_stats + cpsw_gstrings_stats[i].stat_offset); data[i] = val; break; @@ -1050,52 +1033,48 @@ static void cpsw_get_ethtool_stats(struct net_device *ndev, } } -static int cpsw_common_res_usage_state(struct cpsw_priv *priv) +static int cpsw_common_res_usage_state(struct cpsw_common *cpsw) { u32 i; u32 usage_count = 0; - if (!priv->data.dual_emac) + if (!cpsw->data.dual_emac) return 0; - for (i = 0; i < priv->data.slaves; i++) - if (priv->slaves[i].open_stat) + for (i = 0; i < cpsw->data.slaves; i++) + if (cpsw->slaves[i].open_stat) usage_count++; return usage_count; } -static inline int cpsw_tx_packet_submit(struct net_device *ndev, - struct cpsw_priv *priv, struct sk_buff *skb) +static inline int cpsw_tx_packet_submit(struct cpsw_priv *priv, + struct sk_buff *skb) { - if (!priv->data.dual_emac) - return cpdma_chan_submit(priv->txch, skb, skb->data, - skb->len, 0); + struct cpsw_common *cpsw = priv->cpsw; - if (ndev == cpsw_get_slave_ndev(priv, 0)) - return cpdma_chan_submit(priv->txch, skb, skb->data, - skb->len, 1); - else - return cpdma_chan_submit(priv->txch, skb, skb->data, - skb->len, 2); + return cpdma_chan_submit(cpsw->txch, skb, skb->data, skb->len, + priv->emac_port + cpsw->data.dual_emac); } static inline void cpsw_add_dual_emac_def_ale_entries( struct cpsw_priv *priv, struct cpsw_slave *slave, u32 slave_port) { + struct cpsw_common *cpsw = priv->cpsw; u32 port_mask = 1 << slave_port | ALE_PORT_HOST; - if (priv->version == CPSW_VERSION_1) + if (cpsw->version == CPSW_VERSION_1) slave_write(slave, slave->port_vlan, CPSW1_PORT_VLAN); else slave_write(slave, slave->port_vlan, CPSW2_PORT_VLAN); - cpsw_ale_add_vlan(priv->ale, slave->port_vlan, port_mask, + cpsw_ale_add_vlan(cpsw->ale, slave->port_vlan, port_mask, port_mask, port_mask, 0); - cpsw_ale_add_mcast(priv->ale, priv->ndev->broadcast, + cpsw_ale_add_mcast(cpsw->ale, priv->ndev->broadcast, port_mask, ALE_VLAN, slave->port_vlan, 0); - cpsw_ale_add_ucast(priv->ale, priv->mac_addr, - HOST_PORT_NUM, ALE_VLAN | ALE_SECURE, slave->port_vlan); + cpsw_ale_add_ucast(cpsw->ale, priv->mac_addr, + HOST_PORT_NUM, ALE_VLAN | + ALE_SECURE, slave->port_vlan); } static void soft_reset_slave(struct cpsw_slave *slave) @@ -1109,13 +1088,14 @@ static void soft_reset_slave(struct cpsw_slave *slave) static void cpsw_slave_open(struct cpsw_slave *slave, struct cpsw_priv *priv) { u32 slave_port; + struct cpsw_common *cpsw = priv->cpsw; soft_reset_slave(slave); /* setup priority mapping */ __raw_writel(RX_PRIORITY_MAPPING, &slave->sliver->rx_pri_map); - switch (priv->version) { + switch (cpsw->version) { case CPSW_VERSION_1: slave_write(slave, TX_PRIORITY_MAPPING, CPSW1_TX_PRI_MAP); break; @@ -1127,17 +1107,17 @@ static void cpsw_slave_open(struct cpsw_slave *slave, struct cpsw_priv *priv) } /* setup max packet size, and mac address */ - __raw_writel(priv->rx_packet_max, &slave->sliver->rx_maxlen); + __raw_writel(cpsw->rx_packet_max, &slave->sliver->rx_maxlen); cpsw_set_slave_mac(slave, priv); slave->mac_control = 0; /* no link yet */ - slave_port = cpsw_get_slave_port(priv, slave->slave_num); + slave_port = cpsw_get_slave_port(slave->slave_num); - if (priv->data.dual_emac) + if (cpsw->data.dual_emac) cpsw_add_dual_emac_def_ale_entries(priv, slave, slave_port); else - cpsw_ale_add_mcast(priv->ale, priv->ndev->broadcast, + cpsw_ale_add_mcast(cpsw->ale, priv->ndev->broadcast, 1 << slave_port, 0, 0, ALE_MCAST_FWD_2); if (slave->data->phy_node) { @@ -1167,81 +1147,83 @@ static void cpsw_slave_open(struct cpsw_slave *slave, struct cpsw_priv *priv) phy_start(slave->phy); /* Configure GMII_SEL register */ - cpsw_phy_sel(&priv->pdev->dev, slave->phy->interface, slave->slave_num); + cpsw_phy_sel(cpsw->dev, slave->phy->interface, slave->slave_num); } static inline void cpsw_add_default_vlan(struct cpsw_priv *priv) { - const int vlan = priv->data.default_vlan; + struct cpsw_common *cpsw = priv->cpsw; + const int vlan = cpsw->data.default_vlan; u32 reg; int i; int unreg_mcast_mask; - reg = (priv->version == CPSW_VERSION_1) ? CPSW1_PORT_VLAN : + reg = (cpsw->version == CPSW_VERSION_1) ? CPSW1_PORT_VLAN : CPSW2_PORT_VLAN; - writel(vlan, &priv->host_port_regs->port_vlan); + writel(vlan, &cpsw->host_port_regs->port_vlan); - for (i = 0; i < priv->data.slaves; i++) - slave_write(priv->slaves + i, vlan, reg); + for (i = 0; i < cpsw->data.slaves; i++) + slave_write(cpsw->slaves + i, vlan, reg); if (priv->ndev->flags & IFF_ALLMULTI) unreg_mcast_mask = ALE_ALL_PORTS; else unreg_mcast_mask = ALE_PORT_1 | ALE_PORT_2; - cpsw_ale_add_vlan(priv->ale, vlan, ALE_ALL_PORTS, + cpsw_ale_add_vlan(cpsw->ale, vlan, ALE_ALL_PORTS, ALE_ALL_PORTS, ALE_ALL_PORTS, unreg_mcast_mask); } static void cpsw_init_host_port(struct cpsw_priv *priv) { - u32 control_reg; u32 fifo_mode; + u32 control_reg; + struct cpsw_common *cpsw = priv->cpsw; /* soft reset the controller and initialize ale */ - soft_reset("cpsw", &priv->regs->soft_reset); - cpsw_ale_start(priv->ale); + soft_reset("cpsw", &cpsw->regs->soft_reset); + cpsw_ale_start(cpsw->ale); /* switch to vlan unaware mode */ - cpsw_ale_control_set(priv->ale, HOST_PORT_NUM, ALE_VLAN_AWARE, + cpsw_ale_control_set(cpsw->ale, HOST_PORT_NUM, ALE_VLAN_AWARE, CPSW_ALE_VLAN_AWARE); - control_reg = readl(&priv->regs->control); + control_reg = readl(&cpsw->regs->control); control_reg |= CPSW_VLAN_AWARE; - writel(control_reg, &priv->regs->control); - fifo_mode = (priv->data.dual_emac) ? CPSW_FIFO_DUAL_MAC_MODE : + writel(control_reg, &cpsw->regs->control); + fifo_mode = (cpsw->data.dual_emac) ? CPSW_FIFO_DUAL_MAC_MODE : CPSW_FIFO_NORMAL_MODE; - writel(fifo_mode, &priv->host_port_regs->tx_in_ctl); + writel(fifo_mode, &cpsw->host_port_regs->tx_in_ctl); /* setup host port priority mapping */ __raw_writel(CPDMA_TX_PRIORITY_MAP, - &priv->host_port_regs->cpdma_tx_pri_map); - __raw_writel(0, &priv->host_port_regs->cpdma_rx_chan_map); + &cpsw->host_port_regs->cpdma_tx_pri_map); + __raw_writel(0, &cpsw->host_port_regs->cpdma_rx_chan_map); - cpsw_ale_control_set(priv->ale, HOST_PORT_NUM, + cpsw_ale_control_set(cpsw->ale, HOST_PORT_NUM, ALE_PORT_STATE, ALE_PORT_STATE_FORWARD); - if (!priv->data.dual_emac) { - cpsw_ale_add_ucast(priv->ale, priv->mac_addr, HOST_PORT_NUM, + if (!cpsw->data.dual_emac) { + cpsw_ale_add_ucast(cpsw->ale, priv->mac_addr, HOST_PORT_NUM, 0, 0); - cpsw_ale_add_mcast(priv->ale, priv->ndev->broadcast, + cpsw_ale_add_mcast(cpsw->ale, priv->ndev->broadcast, ALE_PORT_HOST, 0, 0, ALE_MCAST_FWD_2); } } -static void cpsw_slave_stop(struct cpsw_slave *slave, struct cpsw_priv *priv) +static void cpsw_slave_stop(struct cpsw_slave *slave, struct cpsw_common *cpsw) { u32 slave_port; - slave_port = cpsw_get_slave_port(priv, slave->slave_num); + slave_port = cpsw_get_slave_port(slave->slave_num); if (!slave->phy) return; phy_stop(slave->phy); phy_disconnect(slave->phy); slave->phy = NULL; - cpsw_ale_control_set(priv->ale, slave_port, + cpsw_ale_control_set(cpsw->ale, slave_port, ALE_PORT_STATE, ALE_PORT_STATE_DISABLE); soft_reset_slave(slave); } @@ -1249,78 +1231,78 @@ static void cpsw_slave_stop(struct cpsw_slave *slave, struct cpsw_priv *priv) static int cpsw_ndo_open(struct net_device *ndev) { struct cpsw_priv *priv = netdev_priv(ndev); + struct cpsw_common *cpsw = priv->cpsw; int i, ret; u32 reg; - ret = pm_runtime_get_sync(&priv->pdev->dev); + ret = pm_runtime_get_sync(cpsw->dev); if (ret < 0) { - pm_runtime_put_noidle(&priv->pdev->dev); + pm_runtime_put_noidle(cpsw->dev); return ret; } - if (!cpsw_common_res_usage_state(priv)) - cpsw_intr_disable(priv); + if (!cpsw_common_res_usage_state(cpsw)) + cpsw_intr_disable(cpsw); netif_carrier_off(ndev); - reg = priv->version; + reg = cpsw->version; dev_info(priv->dev, "initializing cpsw version %d.%d (%d)\n", CPSW_MAJOR_VERSION(reg), CPSW_MINOR_VERSION(reg), CPSW_RTL_VERSION(reg)); /* initialize host and slave ports */ - if (!cpsw_common_res_usage_state(priv)) + if (!cpsw_common_res_usage_state(cpsw)) cpsw_init_host_port(priv); for_each_slave(priv, cpsw_slave_open, priv); /* Add default VLAN */ - if (!priv->data.dual_emac) + if (!cpsw->data.dual_emac) cpsw_add_default_vlan(priv); else - cpsw_ale_add_vlan(priv->ale, priv->data.default_vlan, + cpsw_ale_add_vlan(cpsw->ale, cpsw->data.default_vlan, ALE_ALL_PORTS, ALE_ALL_PORTS, 0, 0); - if (!cpsw_common_res_usage_state(priv)) { - struct cpsw_priv *priv_sl0 = cpsw_get_slave_priv(priv, 0); + if (!cpsw_common_res_usage_state(cpsw)) { int buf_num; /* setup tx dma to fixed prio and zero offset */ - cpdma_control_set(priv->dma, CPDMA_TX_PRIO_FIXED, 1); - cpdma_control_set(priv->dma, CPDMA_RX_BUFFER_OFFSET, 0); + cpdma_control_set(cpsw->dma, CPDMA_TX_PRIO_FIXED, 1); + cpdma_control_set(cpsw->dma, CPDMA_RX_BUFFER_OFFSET, 0); /* disable priority elevation */ - __raw_writel(0, &priv->regs->ptype); + __raw_writel(0, &cpsw->regs->ptype); /* enable statistics collection only on all ports */ - __raw_writel(0x7, &priv->regs->stat_port_en); + __raw_writel(0x7, &cpsw->regs->stat_port_en); /* Enable internal fifo flow control */ - writel(0x7, &priv->regs->flow_control); + writel(0x7, &cpsw->regs->flow_control); - napi_enable(&priv_sl0->napi_rx); - napi_enable(&priv_sl0->napi_tx); + napi_enable(&cpsw->napi_rx); + napi_enable(&cpsw->napi_tx); - if (priv_sl0->tx_irq_disabled) { - priv_sl0->tx_irq_disabled = false; - enable_irq(priv->irqs_table[1]); + if (cpsw->tx_irq_disabled) { + cpsw->tx_irq_disabled = false; + enable_irq(cpsw->irqs_table[1]); } - if (priv_sl0->rx_irq_disabled) { - priv_sl0->rx_irq_disabled = false; - enable_irq(priv->irqs_table[0]); + if (cpsw->rx_irq_disabled) { + cpsw->rx_irq_disabled = false; + enable_irq(cpsw->irqs_table[0]); } - buf_num = cpdma_chan_get_rx_buf_num(priv->dma); + buf_num = cpdma_chan_get_rx_buf_num(cpsw->dma); for (i = 0; i < buf_num; i++) { struct sk_buff *skb; ret = -ENOMEM; skb = __netdev_alloc_skb_ip_align(priv->ndev, - priv->rx_packet_max, GFP_KERNEL); + cpsw->rx_packet_max, GFP_KERNEL); if (!skb) goto err_cleanup; - ret = cpdma_chan_submit(priv->rxch, skb, skb->data, - skb_tailroom(skb), 0); + ret = cpdma_chan_submit(cpsw->rxch, skb, skb->data, + skb_tailroom(skb), 0); if (ret < 0) { kfree_skb(skb); goto err_cleanup; @@ -1331,32 +1313,32 @@ static int cpsw_ndo_open(struct net_device *ndev) */ cpsw_info(priv, ifup, "submitted %d rx descriptors\n", i); - if (cpts_register(&priv->pdev->dev, priv->cpts, - priv->data.cpts_clock_mult, - priv->data.cpts_clock_shift)) + if (cpts_register(cpsw->dev, cpsw->cpts, + cpsw->data.cpts_clock_mult, + cpsw->data.cpts_clock_shift)) dev_err(priv->dev, "error registering cpts device\n"); } /* Enable Interrupt pacing if configured */ - if (priv->coal_intvl != 0) { + if (cpsw->coal_intvl != 0) { struct ethtool_coalesce coal; - coal.rx_coalesce_usecs = priv->coal_intvl; + coal.rx_coalesce_usecs = cpsw->coal_intvl; cpsw_set_coalesce(ndev, &coal); } - cpdma_ctlr_start(priv->dma); - cpsw_intr_enable(priv); + cpdma_ctlr_start(cpsw->dma); + cpsw_intr_enable(cpsw); - if (priv->data.dual_emac) - priv->slaves[priv->emac_port].open_stat = true; + if (cpsw->data.dual_emac) + cpsw->slaves[priv->emac_port].open_stat = true; return 0; err_cleanup: - cpdma_ctlr_stop(priv->dma); - for_each_slave(priv, cpsw_slave_stop, priv); - pm_runtime_put_sync(&priv->pdev->dev); + cpdma_ctlr_stop(cpsw->dma); + for_each_slave(priv, cpsw_slave_stop, cpsw); + pm_runtime_put_sync(cpsw->dev); netif_carrier_off(priv->ndev); return ret; } @@ -1364,25 +1346,24 @@ err_cleanup: static int cpsw_ndo_stop(struct net_device *ndev) { struct cpsw_priv *priv = netdev_priv(ndev); + struct cpsw_common *cpsw = priv->cpsw; cpsw_info(priv, ifdown, "shutting down cpsw device\n"); netif_stop_queue(priv->ndev); netif_carrier_off(priv->ndev); - if (cpsw_common_res_usage_state(priv) <= 1) { - struct cpsw_priv *priv_sl0 = cpsw_get_slave_priv(priv, 0); - - napi_disable(&priv_sl0->napi_rx); - napi_disable(&priv_sl0->napi_tx); - cpts_unregister(priv->cpts); - cpsw_intr_disable(priv); - cpdma_ctlr_stop(priv->dma); - cpsw_ale_stop(priv->ale); - } - for_each_slave(priv, cpsw_slave_stop, priv); - pm_runtime_put_sync(&priv->pdev->dev); - if (priv->data.dual_emac) - priv->slaves[priv->emac_port].open_stat = false; + if (cpsw_common_res_usage_state(cpsw) <= 1) { + napi_disable(&cpsw->napi_rx); + napi_disable(&cpsw->napi_tx); + cpts_unregister(cpsw->cpts); + cpsw_intr_disable(cpsw); + cpdma_ctlr_stop(cpsw->dma); + cpsw_ale_stop(cpsw->ale); + } + for_each_slave(priv, cpsw_slave_stop, cpsw); + pm_runtime_put_sync(cpsw->dev); + if (cpsw->data.dual_emac) + cpsw->slaves[priv->emac_port].open_stat = false; return 0; } @@ -1391,6 +1372,7 @@ static netdev_tx_t cpsw_ndo_start_xmit(struct sk_buff *skb, { struct cpsw_priv *priv = netdev_priv(ndev); int ret; + struct cpsw_common *cpsw = priv->cpsw; netif_trans_update(ndev); @@ -1401,12 +1383,12 @@ static netdev_tx_t cpsw_ndo_start_xmit(struct sk_buff *skb, } if (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP && - priv->cpts->tx_enable) + cpsw->cpts->tx_enable) skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; skb_tx_timestamp(skb); - ret = cpsw_tx_packet_submit(ndev, priv, skb); + ret = cpsw_tx_packet_submit(priv, skb); if (unlikely(ret != 0)) { cpsw_err(priv, tx_err, "desc submit failed\n"); goto fail; @@ -1415,7 +1397,7 @@ static netdev_tx_t cpsw_ndo_start_xmit(struct sk_buff *skb, /* If there is no more tx desc left free then we need to * tell the kernel to stop sending us tx frames. */ - if (unlikely(!cpdma_check_free_tx_desc(priv->txch))) + if (unlikely(!cpdma_check_free_tx_desc(cpsw->txch))) netif_stop_queue(ndev); return NETDEV_TX_OK; @@ -1427,12 +1409,12 @@ fail: #ifdef CONFIG_TI_CPTS -static void cpsw_hwtstamp_v1(struct cpsw_priv *priv) +static void cpsw_hwtstamp_v1(struct cpsw_common *cpsw) { - struct cpsw_slave *slave = &priv->slaves[priv->data.active_slave]; + struct cpsw_slave *slave = &cpsw->slaves[cpsw->data.active_slave]; u32 ts_en, seq_id; - if (!priv->cpts->tx_enable && !priv->cpts->rx_enable) { + if (!cpsw->cpts->tx_enable && !cpsw->cpts->rx_enable) { slave_write(slave, 0, CPSW1_TS_CTL); return; } @@ -1440,10 +1422,10 @@ static void cpsw_hwtstamp_v1(struct cpsw_priv *priv) seq_id = (30 << CPSW_V1_SEQ_ID_OFS_SHIFT) | ETH_P_1588; ts_en = EVENT_MSG_BITS << CPSW_V1_MSG_TYPE_OFS; - if (priv->cpts->tx_enable) + if (cpsw->cpts->tx_enable) ts_en |= CPSW_V1_TS_TX_EN; - if (priv->cpts->rx_enable) + if (cpsw->cpts->rx_enable) ts_en |= CPSW_V1_TS_RX_EN; slave_write(slave, ts_en, CPSW1_TS_CTL); @@ -1453,32 +1435,33 @@ static void cpsw_hwtstamp_v1(struct cpsw_priv *priv) static void cpsw_hwtstamp_v2(struct cpsw_priv *priv) { struct cpsw_slave *slave; + struct cpsw_common *cpsw = priv->cpsw; u32 ctrl, mtype; - if (priv->data.dual_emac) - slave = &priv->slaves[priv->emac_port]; + if (cpsw->data.dual_emac) + slave = &cpsw->slaves[priv->emac_port]; else - slave = &priv->slaves[priv->data.active_slave]; + slave = &cpsw->slaves[cpsw->data.active_slave]; ctrl = slave_read(slave, CPSW2_CONTROL); - switch (priv->version) { + switch (cpsw->version) { case CPSW_VERSION_2: ctrl &= ~CTRL_V2_ALL_TS_MASK; - if (priv->cpts->tx_enable) + if (cpsw->cpts->tx_enable) ctrl |= CTRL_V2_TX_TS_BITS; - if (priv->cpts->rx_enable) + if (cpsw->cpts->rx_enable) ctrl |= CTRL_V2_RX_TS_BITS; break; case CPSW_VERSION_3: default: ctrl &= ~CTRL_V3_ALL_TS_MASK; - if (priv->cpts->tx_enable) + if (cpsw->cpts->tx_enable) ctrl |= CTRL_V3_TX_TS_BITS; - if (priv->cpts->rx_enable) + if (cpsw->cpts->rx_enable) ctrl |= CTRL_V3_RX_TS_BITS; break; } @@ -1487,18 +1470,19 @@ static void cpsw_hwtstamp_v2(struct cpsw_priv *priv) slave_write(slave, mtype, CPSW2_TS_SEQ_MTYPE); slave_write(slave, ctrl, CPSW2_CONTROL); - __raw_writel(ETH_P_1588, &priv->regs->ts_ltype); + __raw_writel(ETH_P_1588, &cpsw->regs->ts_ltype); } static int cpsw_hwtstamp_set(struct net_device *dev, struct ifreq *ifr) { struct cpsw_priv *priv = netdev_priv(dev); - struct cpts *cpts = priv->cpts; struct hwtstamp_config cfg; + struct cpsw_common *cpsw = priv->cpsw; + struct cpts *cpts = cpsw->cpts; - if (priv->version != CPSW_VERSION_1 && - priv->version != CPSW_VERSION_2 && - priv->version != CPSW_VERSION_3) + if (cpsw->version != CPSW_VERSION_1 && + cpsw->version != CPSW_VERSION_2 && + cpsw->version != CPSW_VERSION_3) return -EOPNOTSUPP; if (copy_from_user(&cfg, ifr->ifr_data, sizeof(cfg))) @@ -1538,9 +1522,9 @@ static int cpsw_hwtstamp_set(struct net_device *dev, struct ifreq *ifr) cpts->tx_enable = cfg.tx_type == HWTSTAMP_TX_ON; - switch (priv->version) { + switch (cpsw->version) { case CPSW_VERSION_1: - cpsw_hwtstamp_v1(priv); + cpsw_hwtstamp_v1(cpsw); break; case CPSW_VERSION_2: case CPSW_VERSION_3: @@ -1555,13 +1539,13 @@ static int cpsw_hwtstamp_set(struct net_device *dev, struct ifreq *ifr) static int cpsw_hwtstamp_get(struct net_device *dev, struct ifreq *ifr) { - struct cpsw_priv *priv = netdev_priv(dev); - struct cpts *cpts = priv->cpts; + struct cpsw_common *cpsw = ndev_to_cpsw(dev); + struct cpts *cpts = cpsw->cpts; struct hwtstamp_config cfg; - if (priv->version != CPSW_VERSION_1 && - priv->version != CPSW_VERSION_2 && - priv->version != CPSW_VERSION_3) + if (cpsw->version != CPSW_VERSION_1 && + cpsw->version != CPSW_VERSION_2 && + cpsw->version != CPSW_VERSION_3) return -EOPNOTSUPP; cfg.flags = 0; @@ -1577,7 +1561,8 @@ static int cpsw_hwtstamp_get(struct net_device *dev, struct ifreq *ifr) static int cpsw_ndo_ioctl(struct net_device *dev, struct ifreq *req, int cmd) { struct cpsw_priv *priv = netdev_priv(dev); - int slave_no = cpsw_slave_index(priv); + struct cpsw_common *cpsw = priv->cpsw; + int slave_no = cpsw_slave_index(cpsw, priv); if (!netif_running(dev)) return -EINVAL; @@ -1591,27 +1576,29 @@ static int cpsw_ndo_ioctl(struct net_device *dev, struct ifreq *req, int cmd) #endif } - if (!priv->slaves[slave_no].phy) + if (!cpsw->slaves[slave_no].phy) return -EOPNOTSUPP; - return phy_mii_ioctl(priv->slaves[slave_no].phy, req, cmd); + return phy_mii_ioctl(cpsw->slaves[slave_no].phy, req, cmd); } static void cpsw_ndo_tx_timeout(struct net_device *ndev) { struct cpsw_priv *priv = netdev_priv(ndev); + struct cpsw_common *cpsw = priv->cpsw; cpsw_err(priv, tx_err, "transmit timeout, restarting dma\n"); ndev->stats.tx_errors++; - cpsw_intr_disable(priv); - cpdma_chan_stop(priv->txch); - cpdma_chan_start(priv->txch); - cpsw_intr_enable(priv); + cpsw_intr_disable(cpsw); + cpdma_chan_stop(cpsw->txch); + cpdma_chan_start(cpsw->txch); + cpsw_intr_enable(cpsw); } static int cpsw_ndo_set_mac_address(struct net_device *ndev, void *p) { struct cpsw_priv *priv = netdev_priv(ndev); struct sockaddr *addr = (struct sockaddr *)p; + struct cpsw_common *cpsw = priv->cpsw; int flags = 0; u16 vid = 0; int ret; @@ -1619,27 +1606,27 @@ static int cpsw_ndo_set_mac_address(struct net_device *ndev, void *p) if (!is_valid_ether_addr(addr->sa_data)) return -EADDRNOTAVAIL; - ret = pm_runtime_get_sync(&priv->pdev->dev); + ret = pm_runtime_get_sync(cpsw->dev); if (ret < 0) { - pm_runtime_put_noidle(&priv->pdev->dev); + pm_runtime_put_noidle(cpsw->dev); return ret; } - if (priv->data.dual_emac) { - vid = priv->slaves[priv->emac_port].port_vlan; + if (cpsw->data.dual_emac) { + vid = cpsw->slaves[priv->emac_port].port_vlan; flags = ALE_VLAN; } - cpsw_ale_del_ucast(priv->ale, priv->mac_addr, HOST_PORT_NUM, + cpsw_ale_del_ucast(cpsw->ale, priv->mac_addr, HOST_PORT_NUM, flags, vid); - cpsw_ale_add_ucast(priv->ale, addr->sa_data, HOST_PORT_NUM, + cpsw_ale_add_ucast(cpsw->ale, addr->sa_data, HOST_PORT_NUM, flags, vid); memcpy(priv->mac_addr, addr->sa_data, ETH_ALEN); memcpy(ndev->dev_addr, priv->mac_addr, ETH_ALEN); for_each_slave(priv, cpsw_set_slave_mac, priv); - pm_runtime_put(&priv->pdev->dev); + pm_runtime_put(cpsw->dev); return 0; } @@ -1647,12 +1634,12 @@ static int cpsw_ndo_set_mac_address(struct net_device *ndev, void *p) #ifdef CONFIG_NET_POLL_CONTROLLER static void cpsw_ndo_poll_controller(struct net_device *ndev) { - struct cpsw_priv *priv = netdev_priv(ndev); + struct cpsw_common *cpsw = ndev_to_cpsw(ndev); - cpsw_intr_disable(priv); - cpsw_rx_interrupt(priv->irqs_table[0], priv); - cpsw_tx_interrupt(priv->irqs_table[1], priv); - cpsw_intr_enable(priv); + cpsw_intr_disable(cpsw); + cpsw_rx_interrupt(cpsw->irqs_table[0], cpsw); + cpsw_tx_interrupt(cpsw->irqs_table[1], cpsw); + cpsw_intr_enable(cpsw); } #endif @@ -1662,8 +1649,9 @@ static inline int cpsw_add_vlan_ale_entry(struct cpsw_priv *priv, int ret; int unreg_mcast_mask = 0; u32 port_mask; + struct cpsw_common *cpsw = priv->cpsw; - if (priv->data.dual_emac) { + if (cpsw->data.dual_emac) { port_mask = (1 << (priv->emac_port + 1)) | ALE_PORT_HOST; if (priv->ndev->flags & IFF_ALLMULTI) @@ -1677,27 +1665,27 @@ static inline int cpsw_add_vlan_ale_entry(struct cpsw_priv *priv, unreg_mcast_mask = ALE_PORT_1 | ALE_PORT_2; } - ret = cpsw_ale_add_vlan(priv->ale, vid, port_mask, 0, port_mask, + ret = cpsw_ale_add_vlan(cpsw->ale, vid, port_mask, 0, port_mask, unreg_mcast_mask); if (ret != 0) return ret; - ret = cpsw_ale_add_ucast(priv->ale, priv->mac_addr, + ret = cpsw_ale_add_ucast(cpsw->ale, priv->mac_addr, HOST_PORT_NUM, ALE_VLAN, vid); if (ret != 0) goto clean_vid; - ret = cpsw_ale_add_mcast(priv->ale, priv->ndev->broadcast, + ret = cpsw_ale_add_mcast(cpsw->ale, priv->ndev->broadcast, port_mask, ALE_VLAN, vid, 0); if (ret != 0) goto clean_vlan_ucast; return 0; clean_vlan_ucast: - cpsw_ale_del_ucast(priv->ale, priv->mac_addr, + cpsw_ale_del_ucast(cpsw->ale, priv->mac_addr, HOST_PORT_NUM, ALE_VLAN, vid); clean_vid: - cpsw_ale_del_vlan(priv->ale, vid, 0); + cpsw_ale_del_vlan(cpsw->ale, vid, 0); return ret; } @@ -1705,26 +1693,27 @@ static int cpsw_ndo_vlan_rx_add_vid(struct net_device *ndev, __be16 proto, u16 vid) { struct cpsw_priv *priv = netdev_priv(ndev); + struct cpsw_common *cpsw = priv->cpsw; int ret; - if (vid == priv->data.default_vlan) + if (vid == cpsw->data.default_vlan) return 0; - ret = pm_runtime_get_sync(&priv->pdev->dev); + ret = pm_runtime_get_sync(cpsw->dev); if (ret < 0) { - pm_runtime_put_noidle(&priv->pdev->dev); + pm_runtime_put_noidle(cpsw->dev); return ret; } - if (priv->data.dual_emac) { + if (cpsw->data.dual_emac) { /* In dual EMAC, reserved VLAN id should not be used for * creating VLAN interfaces as this can break the dual * EMAC port separation */ int i; - for (i = 0; i < priv->data.slaves; i++) { - if (vid == priv->slaves[i].port_vlan) + for (i = 0; i < cpsw->data.slaves; i++) { + if (vid == cpsw->slaves[i].port_vlan) return -EINVAL; } } @@ -1732,7 +1721,7 @@ static int cpsw_ndo_vlan_rx_add_vid(struct net_device *ndev, dev_info(priv->dev, "Adding vlanid %d to vlan filter\n", vid); ret = cpsw_add_vlan_ale_entry(priv, vid); - pm_runtime_put(&priv->pdev->dev); + pm_runtime_put(cpsw->dev); return ret; } @@ -1740,39 +1729,40 @@ static int cpsw_ndo_vlan_rx_kill_vid(struct net_device *ndev, __be16 proto, u16 vid) { struct cpsw_priv *priv = netdev_priv(ndev); + struct cpsw_common *cpsw = priv->cpsw; int ret; - if (vid == priv->data.default_vlan) + if (vid == cpsw->data.default_vlan) return 0; - ret = pm_runtime_get_sync(&priv->pdev->dev); + ret = pm_runtime_get_sync(cpsw->dev); if (ret < 0) { - pm_runtime_put_noidle(&priv->pdev->dev); + pm_runtime_put_noidle(cpsw->dev); return ret; } - if (priv->data.dual_emac) { + if (cpsw->data.dual_emac) { int i; - for (i = 0; i < priv->data.slaves; i++) { - if (vid == priv->slaves[i].port_vlan) + for (i = 0; i < cpsw->data.slaves; i++) { + if (vid == cpsw->slaves[i].port_vlan) return -EINVAL; } } dev_info(priv->dev, "removing vlanid %d from vlan filter\n", vid); - ret = cpsw_ale_del_vlan(priv->ale, vid, 0); + ret = cpsw_ale_del_vlan(cpsw->ale, vid, 0); if (ret != 0) return ret; - ret = cpsw_ale_del_ucast(priv->ale, priv->mac_addr, + ret = cpsw_ale_del_ucast(cpsw->ale, priv->mac_addr, HOST_PORT_NUM, ALE_VLAN, vid); if (ret != 0) return ret; - ret = cpsw_ale_del_mcast(priv->ale, priv->ndev->broadcast, + ret = cpsw_ale_del_mcast(cpsw->ale, priv->ndev->broadcast, 0, ALE_VLAN, vid); - pm_runtime_put(&priv->pdev->dev); + pm_runtime_put(cpsw->dev); return ret; } @@ -1795,31 +1785,32 @@ static const struct net_device_ops cpsw_netdev_ops = { static int cpsw_get_regs_len(struct net_device *ndev) { - struct cpsw_priv *priv = netdev_priv(ndev); + struct cpsw_common *cpsw = ndev_to_cpsw(ndev); - return priv->data.ale_entries * ALE_ENTRY_WORDS * sizeof(u32); + return cpsw->data.ale_entries * ALE_ENTRY_WORDS * sizeof(u32); } static void cpsw_get_regs(struct net_device *ndev, struct ethtool_regs *regs, void *p) { - struct cpsw_priv *priv = netdev_priv(ndev); u32 *reg = p; + struct cpsw_common *cpsw = ndev_to_cpsw(ndev); /* update CPSW IP version */ - regs->version = priv->version; + regs->version = cpsw->version; - cpsw_ale_dump(priv->ale, reg); + cpsw_ale_dump(cpsw->ale, reg); } static void cpsw_get_drvinfo(struct net_device *ndev, struct ethtool_drvinfo *info) { - struct cpsw_priv *priv = netdev_priv(ndev); + struct cpsw_common *cpsw = ndev_to_cpsw(ndev); + struct platform_device *pdev = to_platform_device(cpsw->dev); strlcpy(info->driver, "cpsw", sizeof(info->driver)); strlcpy(info->version, "1.0", sizeof(info->version)); - strlcpy(info->bus_info, priv->pdev->name, sizeof(info->bus_info)); + strlcpy(info->bus_info, pdev->name, sizeof(info->bus_info)); } static u32 cpsw_get_msglevel(struct net_device *ndev) @@ -1838,7 +1829,7 @@ static int cpsw_get_ts_info(struct net_device *ndev, struct ethtool_ts_info *info) { #ifdef CONFIG_TI_CPTS - struct cpsw_priv *priv = netdev_priv(ndev); + struct cpsw_common *cpsw = ndev_to_cpsw(ndev); info->so_timestamping = SOF_TIMESTAMPING_TX_HARDWARE | @@ -1847,7 +1838,7 @@ static int cpsw_get_ts_info(struct net_device *ndev, SOF_TIMESTAMPING_RX_SOFTWARE | SOF_TIMESTAMPING_SOFTWARE | SOF_TIMESTAMPING_RAW_HARDWARE; - info->phc_index = priv->cpts->phc_index; + info->phc_index = cpsw->cpts->phc_index; info->tx_types = (1 << HWTSTAMP_TX_OFF) | (1 << HWTSTAMP_TX_ON); @@ -1870,10 +1861,11 @@ static int cpsw_get_settings(struct net_device *ndev, struct ethtool_cmd *ecmd) { struct cpsw_priv *priv = netdev_priv(ndev); - int slave_no = cpsw_slave_index(priv); + struct cpsw_common *cpsw = priv->cpsw; + int slave_no = cpsw_slave_index(cpsw, priv); - if (priv->slaves[slave_no].phy) - return phy_ethtool_gset(priv->slaves[slave_no].phy, ecmd); + if (cpsw->slaves[slave_no].phy) + return phy_ethtool_gset(cpsw->slaves[slave_no].phy, ecmd); else return -EOPNOTSUPP; } @@ -1881,10 +1873,11 @@ static int cpsw_get_settings(struct net_device *ndev, static int cpsw_set_settings(struct net_device *ndev, struct ethtool_cmd *ecmd) { struct cpsw_priv *priv = netdev_priv(ndev); - int slave_no = cpsw_slave_index(priv); + struct cpsw_common *cpsw = priv->cpsw; + int slave_no = cpsw_slave_index(cpsw, priv); - if (priv->slaves[slave_no].phy) - return phy_ethtool_sset(priv->slaves[slave_no].phy, ecmd); + if (cpsw->slaves[slave_no].phy) + return phy_ethtool_sset(cpsw->slaves[slave_no].phy, ecmd); else return -EOPNOTSUPP; } @@ -1892,22 +1885,24 @@ static int cpsw_set_settings(struct net_device *ndev, struct ethtool_cmd *ecmd) static void cpsw_get_wol(struct net_device *ndev, struct ethtool_wolinfo *wol) { struct cpsw_priv *priv = netdev_priv(ndev); - int slave_no = cpsw_slave_index(priv); + struct cpsw_common *cpsw = priv->cpsw; + int slave_no = cpsw_slave_index(cpsw, priv); wol->supported = 0; wol->wolopts = 0; - if (priv->slaves[slave_no].phy) - phy_ethtool_get_wol(priv->slaves[slave_no].phy, wol); + if (cpsw->slaves[slave_no].phy) + phy_ethtool_get_wol(cpsw->slaves[slave_no].phy, wol); } static int cpsw_set_wol(struct net_device *ndev, struct ethtool_wolinfo *wol) { struct cpsw_priv *priv = netdev_priv(ndev); - int slave_no = cpsw_slave_index(priv); + struct cpsw_common *cpsw = priv->cpsw; + int slave_no = cpsw_slave_index(cpsw, priv); - if (priv->slaves[slave_no].phy) - return phy_ethtool_set_wol(priv->slaves[slave_no].phy, wol); + if (cpsw->slaves[slave_no].phy) + return phy_ethtool_set_wol(cpsw->slaves[slave_no].phy, wol); else return -EOPNOTSUPP; } @@ -1938,12 +1933,13 @@ static int cpsw_set_pauseparam(struct net_device *ndev, static int cpsw_ethtool_op_begin(struct net_device *ndev) { struct cpsw_priv *priv = netdev_priv(ndev); + struct cpsw_common *cpsw = priv->cpsw; int ret; - ret = pm_runtime_get_sync(&priv->pdev->dev); + ret = pm_runtime_get_sync(cpsw->dev); if (ret < 0) { cpsw_err(priv, drv, "ethtool begin failed %d\n", ret); - pm_runtime_put_noidle(&priv->pdev->dev); + pm_runtime_put_noidle(cpsw->dev); } return ret; @@ -1954,7 +1950,7 @@ static void cpsw_ethtool_op_complete(struct net_device *ndev) struct cpsw_priv *priv = netdev_priv(ndev); int ret; - ret = pm_runtime_put(&priv->pdev->dev); + ret = pm_runtime_put(priv->cpsw->dev); if (ret < 0) cpsw_err(priv, drv, "ethtool complete failed %d\n", ret); } @@ -1982,12 +1978,12 @@ static const struct ethtool_ops cpsw_ethtool_ops = { .complete = cpsw_ethtool_op_complete, }; -static void cpsw_slave_init(struct cpsw_slave *slave, struct cpsw_priv *priv, +static void cpsw_slave_init(struct cpsw_slave *slave, struct cpsw_common *cpsw, u32 slave_reg_ofs, u32 sliver_reg_ofs) { - void __iomem *regs = priv->regs; + void __iomem *regs = cpsw->regs; int slave_num = slave->slave_num; - struct cpsw_slave_data *data = priv->data.slave_data + slave_num; + struct cpsw_slave_data *data = cpsw->data.slave_data + slave_num; slave->data = data; slave->regs = regs + slave_reg_ofs; @@ -2158,71 +2154,50 @@ no_phy_slave: return 0; } -static int cpsw_probe_dual_emac(struct platform_device *pdev, - struct cpsw_priv *priv) +static int cpsw_probe_dual_emac(struct cpsw_priv *priv) { - struct cpsw_platform_data *data = &priv->data; + struct cpsw_common *cpsw = priv->cpsw; + struct cpsw_platform_data *data = &cpsw->data; struct net_device *ndev; struct cpsw_priv *priv_sl2; - int ret = 0, i; + int ret = 0; ndev = alloc_etherdev(sizeof(struct cpsw_priv)); if (!ndev) { - dev_err(&pdev->dev, "cpsw: error allocating net_device\n"); + dev_err(cpsw->dev, "cpsw: error allocating net_device\n"); return -ENOMEM; } priv_sl2 = netdev_priv(ndev); - priv_sl2->data = *data; - priv_sl2->pdev = pdev; + priv_sl2->cpsw = cpsw; priv_sl2->ndev = ndev; priv_sl2->dev = &ndev->dev; priv_sl2->msg_enable = netif_msg_init(debug_level, CPSW_DEBUG); - priv_sl2->rx_packet_max = max(rx_packet_max, 128); if (is_valid_ether_addr(data->slave_data[1].mac_addr)) { memcpy(priv_sl2->mac_addr, data->slave_data[1].mac_addr, ETH_ALEN); - dev_info(&pdev->dev, "cpsw: Detected MACID = %pM\n", priv_sl2->mac_addr); + dev_info(cpsw->dev, "cpsw: Detected MACID = %pM\n", + priv_sl2->mac_addr); } else { random_ether_addr(priv_sl2->mac_addr); - dev_info(&pdev->dev, "cpsw: Random MACID = %pM\n", priv_sl2->mac_addr); + dev_info(cpsw->dev, "cpsw: Random MACID = %pM\n", + priv_sl2->mac_addr); } memcpy(ndev->dev_addr, priv_sl2->mac_addr, ETH_ALEN); - priv_sl2->slaves = priv->slaves; - priv_sl2->clk = priv->clk; - - priv_sl2->coal_intvl = 0; - priv_sl2->bus_freq_mhz = priv->bus_freq_mhz; - - priv_sl2->regs = priv->regs; - priv_sl2->host_port_regs = priv->host_port_regs; - priv_sl2->wr_regs = priv->wr_regs; - priv_sl2->hw_stats = priv->hw_stats; - priv_sl2->dma = priv->dma; - priv_sl2->txch = priv->txch; - priv_sl2->rxch = priv->rxch; - priv_sl2->ale = priv->ale; priv_sl2->emac_port = 1; - priv->slaves[1].ndev = ndev; - priv_sl2->cpts = priv->cpts; - priv_sl2->version = priv->version; - - for (i = 0; i < priv->num_irqs; i++) { - priv_sl2->irqs_table[i] = priv->irqs_table[i]; - priv_sl2->num_irqs = priv->num_irqs; - } + cpsw->slaves[1].ndev = ndev; ndev->features |= NETIF_F_HW_VLAN_CTAG_FILTER; ndev->netdev_ops = &cpsw_netdev_ops; ndev->ethtool_ops = &cpsw_ethtool_ops; /* register the network device */ - SET_NETDEV_DEV(ndev, &pdev->dev); + SET_NETDEV_DEV(ndev, cpsw->dev); ret = register_netdev(ndev); if (ret) { - dev_err(&pdev->dev, "cpsw: error registering net device\n"); + dev_err(cpsw->dev, "cpsw: error registering net device\n"); free_netdev(ndev); ret = -ENODEV; } @@ -2270,6 +2245,7 @@ MODULE_DEVICE_TABLE(of, cpsw_of_mtable); static int cpsw_probe(struct platform_device *pdev) { + struct clk *clk; struct cpsw_platform_data *data; struct net_device *ndev; struct cpsw_priv *priv; @@ -2280,9 +2256,13 @@ static int cpsw_probe(struct platform_device *pdev) const struct of_device_id *of_id; struct gpio_descs *mode; u32 slave_offset, sliver_offset, slave_size; + struct cpsw_common *cpsw; int ret = 0, i; int irq; + cpsw = devm_kzalloc(&pdev->dev, sizeof(struct cpsw_common), GFP_KERNEL); + cpsw->dev = &pdev->dev; + ndev = alloc_etherdev(sizeof(struct cpsw_priv)); if (!ndev) { dev_err(&pdev->dev, "error allocating net_device\n"); @@ -2291,13 +2271,13 @@ static int cpsw_probe(struct platform_device *pdev) platform_set_drvdata(pdev, ndev); priv = netdev_priv(ndev); - priv->pdev = pdev; + priv->cpsw = cpsw; priv->ndev = ndev; priv->dev = &ndev->dev; priv->msg_enable = netif_msg_init(debug_level, CPSW_DEBUG); - priv->rx_packet_max = max(rx_packet_max, 128); - priv->cpts = devm_kzalloc(&pdev->dev, sizeof(struct cpts), GFP_KERNEL); - if (!priv->cpts) { + cpsw->rx_packet_max = max(rx_packet_max, 128); + cpsw->cpts = devm_kzalloc(&pdev->dev, sizeof(struct cpts), GFP_KERNEL); + if (!cpsw->cpts) { dev_err(&pdev->dev, "error allocating cpts\n"); ret = -ENOMEM; goto clean_ndev_ret; @@ -2318,12 +2298,12 @@ static int cpsw_probe(struct platform_device *pdev) /* Select default pin state */ pinctrl_pm_select_default_state(&pdev->dev); - if (cpsw_probe_dt(&priv->data, pdev)) { + if (cpsw_probe_dt(&cpsw->data, pdev)) { dev_err(&pdev->dev, "cpsw: platform data missing\n"); ret = -ENODEV; goto clean_runtime_disable_ret; } - data = &priv->data; + data = &cpsw->data; if (is_valid_ether_addr(data->slave_data[0].mac_addr)) { memcpy(priv->mac_addr, data->slave_data[0].mac_addr, ETH_ALEN); @@ -2335,27 +2315,26 @@ static int cpsw_probe(struct platform_device *pdev) memcpy(ndev->dev_addr, priv->mac_addr, ETH_ALEN); - priv->slaves = devm_kzalloc(&pdev->dev, + cpsw->slaves = devm_kzalloc(&pdev->dev, sizeof(struct cpsw_slave) * data->slaves, GFP_KERNEL); - if (!priv->slaves) { + if (!cpsw->slaves) { ret = -ENOMEM; goto clean_runtime_disable_ret; } for (i = 0; i < data->slaves; i++) - priv->slaves[i].slave_num = i; + cpsw->slaves[i].slave_num = i; - priv->slaves[0].ndev = ndev; + cpsw->slaves[0].ndev = ndev; priv->emac_port = 0; - priv->clk = devm_clk_get(&pdev->dev, "fck"); - if (IS_ERR(priv->clk)) { + clk = devm_clk_get(&pdev->dev, "fck"); + if (IS_ERR(clk)) { dev_err(priv->dev, "fck is not found\n"); ret = -ENODEV; goto clean_runtime_disable_ret; } - priv->coal_intvl = 0; - priv->bus_freq_mhz = clk_get_rate(priv->clk) / 1000000; + cpsw->bus_freq_mhz = clk_get_rate(clk) / 1000000; ss_res = platform_get_resource(pdev, IORESOURCE_MEM, 0); ss_regs = devm_ioremap_resource(&pdev->dev, ss_res); @@ -2363,7 +2342,7 @@ static int cpsw_probe(struct platform_device *pdev) ret = PTR_ERR(ss_regs); goto clean_runtime_disable_ret; } - priv->regs = ss_regs; + cpsw->regs = ss_regs; /* Need to enable clocks with runtime PM api to access module * registers @@ -2373,24 +2352,24 @@ static int cpsw_probe(struct platform_device *pdev) pm_runtime_put_noidle(&pdev->dev); goto clean_runtime_disable_ret; } - priv->version = readl(&priv->regs->id_ver); + cpsw->version = readl(&cpsw->regs->id_ver); pm_runtime_put_sync(&pdev->dev); res = platform_get_resource(pdev, IORESOURCE_MEM, 1); - priv->wr_regs = devm_ioremap_resource(&pdev->dev, res); - if (IS_ERR(priv->wr_regs)) { - ret = PTR_ERR(priv->wr_regs); + cpsw->wr_regs = devm_ioremap_resource(&pdev->dev, res); + if (IS_ERR(cpsw->wr_regs)) { + ret = PTR_ERR(cpsw->wr_regs); goto clean_runtime_disable_ret; } memset(&dma_params, 0, sizeof(dma_params)); memset(&ale_params, 0, sizeof(ale_params)); - switch (priv->version) { + switch (cpsw->version) { case CPSW_VERSION_1: - priv->host_port_regs = ss_regs + CPSW1_HOST_PORT_OFFSET; - priv->cpts->reg = ss_regs + CPSW1_CPTS_OFFSET; - priv->hw_stats = ss_regs + CPSW1_HW_STATS; + cpsw->host_port_regs = ss_regs + CPSW1_HOST_PORT_OFFSET; + cpsw->cpts->reg = ss_regs + CPSW1_CPTS_OFFSET; + cpsw->hw_stats = ss_regs + CPSW1_HW_STATS; dma_params.dmaregs = ss_regs + CPSW1_CPDMA_OFFSET; dma_params.txhdp = ss_regs + CPSW1_STATERAM_OFFSET; ale_params.ale_regs = ss_regs + CPSW1_ALE_OFFSET; @@ -2402,9 +2381,9 @@ static int cpsw_probe(struct platform_device *pdev) case CPSW_VERSION_2: case CPSW_VERSION_3: case CPSW_VERSION_4: - priv->host_port_regs = ss_regs + CPSW2_HOST_PORT_OFFSET; - priv->cpts->reg = ss_regs + CPSW2_CPTS_OFFSET; - priv->hw_stats = ss_regs + CPSW2_HW_STATS; + cpsw->host_port_regs = ss_regs + CPSW2_HOST_PORT_OFFSET; + cpsw->cpts->reg = ss_regs + CPSW2_CPTS_OFFSET; + cpsw->hw_stats = ss_regs + CPSW2_HW_STATS; dma_params.dmaregs = ss_regs + CPSW2_CPDMA_OFFSET; dma_params.txhdp = ss_regs + CPSW2_STATERAM_OFFSET; ale_params.ale_regs = ss_regs + CPSW2_ALE_OFFSET; @@ -2415,13 +2394,14 @@ static int cpsw_probe(struct platform_device *pdev) (u32 __force) ss_res->start + CPSW2_BD_OFFSET; break; default: - dev_err(priv->dev, "unknown version 0x%08x\n", priv->version); + dev_err(priv->dev, "unknown version 0x%08x\n", cpsw->version); ret = -ENODEV; goto clean_runtime_disable_ret; } - for (i = 0; i < priv->data.slaves; i++) { - struct cpsw_slave *slave = &priv->slaves[i]; - cpsw_slave_init(slave, priv, slave_offset, sliver_offset); + for (i = 0; i < cpsw->data.slaves; i++) { + struct cpsw_slave *slave = &cpsw->slaves[i]; + + cpsw_slave_init(slave, cpsw, slave_offset, sliver_offset); slave_offset += slave_size; sliver_offset += SLIVER_SIZE; } @@ -2441,19 +2421,19 @@ static int cpsw_probe(struct platform_device *pdev) dma_params.has_ext_regs = true; dma_params.desc_hw_addr = dma_params.desc_mem_phys; - priv->dma = cpdma_ctlr_create(&dma_params); - if (!priv->dma) { + cpsw->dma = cpdma_ctlr_create(&dma_params); + if (!cpsw->dma) { dev_err(priv->dev, "error initializing dma\n"); ret = -ENOMEM; goto clean_runtime_disable_ret; } - priv->txch = cpdma_chan_create(priv->dma, tx_chan_num(0), + cpsw->txch = cpdma_chan_create(cpsw->dma, tx_chan_num(0), cpsw_tx_handler); - priv->rxch = cpdma_chan_create(priv->dma, rx_chan_num(0), + cpsw->rxch = cpdma_chan_create(cpsw->dma, rx_chan_num(0), cpsw_rx_handler); - if (WARN_ON(!priv->txch || !priv->rxch)) { + if (WARN_ON(!cpsw->txch || !cpsw->rxch)) { dev_err(priv->dev, "error initializing dma channels\n"); ret = -ENOMEM; goto clean_dma_ret; @@ -2464,8 +2444,8 @@ static int cpsw_probe(struct platform_device *pdev) ale_params.ale_entries = data->ale_entries; ale_params.ale_ports = data->slaves; - priv->ale = cpsw_ale_create(&ale_params); - if (!priv->ale) { + cpsw->ale = cpsw_ale_create(&ale_params); + if (!cpsw->ale) { dev_err(priv->dev, "error initializing ale engine\n"); ret = -ENODEV; goto clean_dma_ret; @@ -2482,7 +2462,7 @@ static int cpsw_probe(struct platform_device *pdev) if (of_id) { pdev->id_entry = of_id->data; if (pdev->id_entry->driver_data) - priv->quirk_irq = true; + cpsw->quirk_irq = true; } /* Grab RX and TX IRQs. Note that we also have RX_THRESHOLD and @@ -2500,9 +2480,9 @@ static int cpsw_probe(struct platform_device *pdev) goto clean_ale_ret; } - priv->irqs_table[0] = irq; + cpsw->irqs_table[0] = irq; ret = devm_request_irq(&pdev->dev, irq, cpsw_rx_interrupt, - 0, dev_name(&pdev->dev), priv); + 0, dev_name(&pdev->dev), cpsw); if (ret < 0) { dev_err(priv->dev, "error attaching irq (%d)\n", ret); goto clean_ale_ret; @@ -2515,21 +2495,20 @@ static int cpsw_probe(struct platform_device *pdev) goto clean_ale_ret; } - priv->irqs_table[1] = irq; + cpsw->irqs_table[1] = irq; ret = devm_request_irq(&pdev->dev, irq, cpsw_tx_interrupt, - 0, dev_name(&pdev->dev), priv); + 0, dev_name(&pdev->dev), cpsw); if (ret < 0) { dev_err(priv->dev, "error attaching irq (%d)\n", ret); goto clean_ale_ret; } - priv->num_irqs = 2; ndev->features |= NETIF_F_HW_VLAN_CTAG_FILTER; ndev->netdev_ops = &cpsw_netdev_ops; ndev->ethtool_ops = &cpsw_ethtool_ops; - netif_napi_add(ndev, &priv->napi_rx, cpsw_rx_poll, CPSW_POLL_WEIGHT); - netif_tx_napi_add(ndev, &priv->napi_tx, cpsw_tx_poll, CPSW_POLL_WEIGHT); + netif_napi_add(ndev, &cpsw->napi_rx, cpsw_rx_poll, CPSW_POLL_WEIGHT); + netif_tx_napi_add(ndev, &cpsw->napi_tx, cpsw_tx_poll, CPSW_POLL_WEIGHT); /* register the network device */ SET_NETDEV_DEV(ndev, &pdev->dev); @@ -2543,8 +2522,8 @@ static int cpsw_probe(struct platform_device *pdev) cpsw_notice(priv, probe, "initialized device (regs %pa, irq %d)\n", &ss_res->start, ndev->irq); - if (priv->data.dual_emac) { - ret = cpsw_probe_dual_emac(pdev, priv); + if (cpsw->data.dual_emac) { + ret = cpsw_probe_dual_emac(priv); if (ret) { cpsw_err(priv, probe, "error probe slave 2 emac interface\n"); goto clean_ale_ret; @@ -2554,9 +2533,9 @@ static int cpsw_probe(struct platform_device *pdev) return 0; clean_ale_ret: - cpsw_ale_destroy(priv->ale); + cpsw_ale_destroy(cpsw->ale); clean_dma_ret: - cpdma_ctlr_destroy(priv->dma); + cpdma_ctlr_destroy(cpsw->dma); clean_runtime_disable_ret: pm_runtime_disable(&pdev->dev); clean_ndev_ret: @@ -2567,7 +2546,7 @@ clean_ndev_ret: static int cpsw_remove(struct platform_device *pdev) { struct net_device *ndev = platform_get_drvdata(pdev); - struct cpsw_priv *priv = netdev_priv(ndev); + struct cpsw_common *cpsw = ndev_to_cpsw(ndev); int ret; ret = pm_runtime_get_sync(&pdev->dev); @@ -2576,17 +2555,17 @@ static int cpsw_remove(struct platform_device *pdev) return ret; } - if (priv->data.dual_emac) - unregister_netdev(cpsw_get_slave_ndev(priv, 1)); + if (cpsw->data.dual_emac) + unregister_netdev(cpsw->slaves[1].ndev); unregister_netdev(ndev); - cpsw_ale_destroy(priv->ale); - cpdma_ctlr_destroy(priv->dma); + cpsw_ale_destroy(cpsw->ale); + cpdma_ctlr_destroy(cpsw->dma); of_platform_depopulate(&pdev->dev); pm_runtime_put_sync(&pdev->dev); pm_runtime_disable(&pdev->dev); - if (priv->data.dual_emac) - free_netdev(cpsw_get_slave_ndev(priv, 1)); + if (cpsw->data.dual_emac) + free_netdev(cpsw->slaves[1].ndev); free_netdev(ndev); return 0; } @@ -2596,14 +2575,14 @@ static int cpsw_suspend(struct device *dev) { struct platform_device *pdev = to_platform_device(dev); struct net_device *ndev = platform_get_drvdata(pdev); - struct cpsw_priv *priv = netdev_priv(ndev); + struct cpsw_common *cpsw = ndev_to_cpsw(ndev); - if (priv->data.dual_emac) { + if (cpsw->data.dual_emac) { int i; - for (i = 0; i < priv->data.slaves; i++) { - if (netif_running(priv->slaves[i].ndev)) - cpsw_ndo_stop(priv->slaves[i].ndev); + for (i = 0; i < cpsw->data.slaves; i++) { + if (netif_running(cpsw->slaves[i].ndev)) + cpsw_ndo_stop(cpsw->slaves[i].ndev); } } else { if (netif_running(ndev)) @@ -2611,7 +2590,7 @@ static int cpsw_suspend(struct device *dev) } /* Select sleep pin state */ - pinctrl_pm_select_sleep_state(&pdev->dev); + pinctrl_pm_select_sleep_state(dev); return 0; } @@ -2620,17 +2599,17 @@ static int cpsw_resume(struct device *dev) { struct platform_device *pdev = to_platform_device(dev); struct net_device *ndev = platform_get_drvdata(pdev); - struct cpsw_priv *priv = netdev_priv(ndev); + struct cpsw_common *cpsw = netdev_priv(ndev); /* Select default pin state */ - pinctrl_pm_select_default_state(&pdev->dev); + pinctrl_pm_select_default_state(dev); - if (priv->data.dual_emac) { + if (cpsw->data.dual_emac) { int i; - for (i = 0; i < priv->data.slaves; i++) { - if (netif_running(priv->slaves[i].ndev)) - cpsw_ndo_open(priv->slaves[i].ndev); + for (i = 0; i < cpsw->data.slaves; i++) { + if (netif_running(cpsw->slaves[i].ndev)) + cpsw_ndo_open(cpsw->slaves[i].ndev); } } else { if (netif_running(ndev)) diff --git a/drivers/net/ethernet/ti/davinci_cpdma.c b/drivers/net/ethernet/ti/davinci_cpdma.c index 19e5f32a8a64..cf72b3390d85 100644 --- a/drivers/net/ethernet/ti/davinci_cpdma.c +++ b/drivers/net/ethernet/ti/davinci_cpdma.c @@ -86,7 +86,7 @@ struct cpdma_desc_pool { void __iomem *iomap; /* ioremap map */ void *cpumap; /* dma_alloc map */ int desc_size, mem_size; - int num_desc, used_desc; + int num_desc; struct device *dev; struct gen_pool *gen_pool; }; @@ -148,7 +148,10 @@ static void cpdma_desc_pool_destroy(struct cpdma_desc_pool *pool) if (!pool) return; - WARN_ON(pool->used_desc); + WARN(gen_pool_size(pool->gen_pool) != gen_pool_avail(pool->gen_pool), + "cpdma_desc_pool size %d != avail %d", + gen_pool_size(pool->gen_pool), + gen_pool_avail(pool->gen_pool)); if (pool->cpumap) dma_free_coherent(pool->dev, pool->mem_size, pool->cpumap, pool->phys); @@ -232,21 +235,14 @@ desc_from_phys(struct cpdma_desc_pool *pool, dma_addr_t dma) static struct cpdma_desc __iomem * cpdma_desc_alloc(struct cpdma_desc_pool *pool) { - struct cpdma_desc __iomem *desc = NULL; - - desc = (struct cpdma_desc __iomem *)gen_pool_alloc(pool->gen_pool, - pool->desc_size); - if (desc) - pool->used_desc++; - - return desc; + return (struct cpdma_desc __iomem *) + gen_pool_alloc(pool->gen_pool, pool->desc_size); } static void cpdma_desc_free(struct cpdma_desc_pool *pool, struct cpdma_desc __iomem *desc, int num_desc) { gen_pool_free(pool->gen_pool, (unsigned long)desc, pool->desc_size); - pool->used_desc--; } struct cpdma_ctlr *cpdma_ctlr_create(struct cpdma_params *params) diff --git a/drivers/net/ethernet/ti/davinci_emac.c b/drivers/net/ethernet/ti/davinci_emac.c index 727a79f3c7dd..2d6fc9a0fb21 100644 --- a/drivers/net/ethernet/ti/davinci_emac.c +++ b/drivers/net/ethernet/ti/davinci_emac.c @@ -597,14 +597,14 @@ static u32 hash_get(u8 *addr) } /** - * hash_add - Hash function to add mac addr from hash table + * emac_hash_add - Hash function to add mac addr from hash table * @priv: The DaVinci EMAC private adapter structure * @mac_addr: mac address to delete from hash table * * Adds mac address to the internal hash table * */ -static int hash_add(struct emac_priv *priv, u8 *mac_addr) +static int emac_hash_add(struct emac_priv *priv, u8 *mac_addr) { struct device *emac_dev = &priv->ndev->dev; u32 rc = 0; @@ -613,7 +613,7 @@ static int hash_add(struct emac_priv *priv, u8 *mac_addr) if (hash_value >= EMAC_NUM_MULTICAST_BITS) { if (netif_msg_drv(priv)) { - dev_err(emac_dev, "DaVinci EMAC: hash_add(): Invalid "\ + dev_err(emac_dev, "DaVinci EMAC: emac_hash_add(): Invalid "\ "Hash %08x, should not be greater than %08x", hash_value, (EMAC_NUM_MULTICAST_BITS - 1)); } @@ -639,14 +639,14 @@ static int hash_add(struct emac_priv *priv, u8 *mac_addr) } /** - * hash_del - Hash function to delete mac addr from hash table + * emac_hash_del - Hash function to delete mac addr from hash table * @priv: The DaVinci EMAC private adapter structure * @mac_addr: mac address to delete from hash table * * Removes mac address from the internal hash table * */ -static int hash_del(struct emac_priv *priv, u8 *mac_addr) +static int emac_hash_del(struct emac_priv *priv, u8 *mac_addr) { u32 hash_value; u32 hash_bit; @@ -696,10 +696,10 @@ static void emac_add_mcast(struct emac_priv *priv, u32 action, u8 *mac_addr) switch (action) { case EMAC_MULTICAST_ADD: - update = hash_add(priv, mac_addr); + update = emac_hash_add(priv, mac_addr); break; case EMAC_MULTICAST_DEL: - update = hash_del(priv, mac_addr); + update = emac_hash_del(priv, mac_addr); break; case EMAC_ALL_MULTI_SET: update = 1; diff --git a/drivers/net/fjes/fjes_main.c b/drivers/net/fjes/fjes_main.c index 9006877c53f2..e46b1ebbbff4 100644 --- a/drivers/net/fjes/fjes_main.c +++ b/drivers/net/fjes/fjes_main.c @@ -97,7 +97,6 @@ static struct acpi_driver fjes_acpi_driver = { static struct platform_driver fjes_driver = { .driver = { .name = DRV_NAME, - .owner = THIS_MODULE, }, .probe = fjes_probe, .remove = fjes_remove, diff --git a/drivers/net/hyperv/hyperv_net.h b/drivers/net/hyperv/hyperv_net.h index 467fb8b4d083..c1403fda874c 100644 --- a/drivers/net/hyperv/hyperv_net.h +++ b/drivers/net/hyperv/hyperv_net.h @@ -490,6 +490,7 @@ struct nvsp_2_vsc_capability { u64 sriov:1; u64 ieee8021q:1; u64 correlation_id:1; + u64 teaming:1; }; }; } __packed; diff --git a/drivers/net/hyperv/netvsc.c b/drivers/net/hyperv/netvsc.c index 20e09174ff62..36a7994f2811 100644 --- a/drivers/net/hyperv/netvsc.c +++ b/drivers/net/hyperv/netvsc.c @@ -472,9 +472,13 @@ static int negotiate_nvsp_ver(struct hv_device *device, init_packet->msg.v2_msg.send_ndis_config.mtu = ndev->mtu + ETH_HLEN; init_packet->msg.v2_msg.send_ndis_config.capability.ieee8021q = 1; - if (nvsp_ver >= NVSP_PROTOCOL_VERSION_5) + if (nvsp_ver >= NVSP_PROTOCOL_VERSION_5) { init_packet->msg.v2_msg.send_ndis_config.capability.sriov = 1; + /* Teaming bit is needed to receive link speed updates */ + init_packet->msg.v2_msg.send_ndis_config.capability.teaming = 1; + } + ret = vmbus_sendpacket(device->channel, init_packet, sizeof(struct nvsp_message), (unsigned long)init_packet, diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c index 41bd952cc28d..82c8f6d47283 100644 --- a/drivers/net/hyperv/netvsc_drv.c +++ b/drivers/net/hyperv/netvsc_drv.c @@ -579,19 +579,32 @@ void netvsc_linkstatus_callback(struct hv_device *device_obj, struct netvsc_reconfig *event; unsigned long flags; - /* Handle link change statuses only */ + net = hv_get_drvdata(device_obj); + + if (!net) + return; + + ndev_ctx = netdev_priv(net); + + /* Update the physical link speed when changing to another vSwitch */ + if (indicate->status == RNDIS_STATUS_LINK_SPEED_CHANGE) { + u32 speed; + + speed = *(u32 *)((void *)indicate + indicate-> + status_buf_offset) / 10000; + ndev_ctx->speed = speed; + return; + } + + /* Handle these link change statuses below */ if (indicate->status != RNDIS_STATUS_NETWORK_CHANGE && indicate->status != RNDIS_STATUS_MEDIA_CONNECT && indicate->status != RNDIS_STATUS_MEDIA_DISCONNECT) return; - net = hv_get_drvdata(device_obj); - - if (!net || net->reg_state != NETREG_REGISTERED) + if (net->reg_state != NETREG_REGISTERED) return; - ndev_ctx = netdev_priv(net); - event = kzalloc(sizeof(*event), GFP_ATOMIC); if (!event) return; @@ -1352,6 +1365,8 @@ static int netvsc_probe(struct hv_device *dev, netif_carrier_off(net); + netvsc_init_settings(net); + net_device_ctx = netdev_priv(net); net_device_ctx->device_ctx = dev; net_device_ctx->msg_enable = netif_msg_init(debug, default_msg); @@ -1410,8 +1425,6 @@ static int netvsc_probe(struct hv_device *dev, netif_set_real_num_tx_queues(net, nvdev->num_chn); netif_set_real_num_rx_queues(net, nvdev->num_chn); - netvsc_init_settings(net); - ret = register_netdev(net); if (ret != 0) { pr_err("Unable to register netdev.\n"); diff --git a/drivers/net/hyperv/rndis_filter.c b/drivers/net/hyperv/rndis_filter.c index 8e830f741d47..dd3b3352a950 100644 --- a/drivers/net/hyperv/rndis_filter.c +++ b/drivers/net/hyperv/rndis_filter.c @@ -752,6 +752,28 @@ static int rndis_filter_query_device_link_status(struct rndis_device *dev) return ret; } +static int rndis_filter_query_link_speed(struct rndis_device *dev) +{ + u32 size = sizeof(u32); + u32 link_speed; + struct net_device_context *ndc; + int ret; + + ret = rndis_filter_query_device(dev, RNDIS_OID_GEN_LINK_SPEED, + &link_speed, &size); + + if (!ret) { + ndc = netdev_priv(dev->ndev); + + /* The link speed reported from host is in 100bps unit, so + * we convert it to Mbps here. + */ + ndc->speed = link_speed / 10000; + } + + return ret; +} + int rndis_filter_set_packet_filter(struct rndis_device *dev, u32 new_filter) { struct rndis_request *request; @@ -1044,6 +1066,8 @@ int rndis_filter_device_add(struct hv_device *dev, if (net_device->nvsp_version < NVSP_PROTOCOL_VERSION_5) return 0; + rndis_filter_query_link_speed(rndis_device); + /* vRSS setup */ memset(&rsscap, 0, rsscap_size); ret = rndis_filter_query_device(rndis_device, diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig index 47a64342cc16..d66133bf3eb5 100644 --- a/drivers/net/phy/Kconfig +++ b/drivers/net/phy/Kconfig @@ -307,6 +307,18 @@ config MDIO_XGENE This module provides a driver for the MDIO busses found in the APM X-Gene SoC's. +config MICROSEMI_PHY + tristate "Drivers for the Microsemi PHYs" + ---help--- + Currently supports the VSC8531 and VSC8541 PHYs + +config XILINX_GMII2RGMII + tristate "Xilinx GMII2RGMII converter driver" + ---help--- + This driver support xilinx GMII to RGMII IP core it provides + the Reduced Gigabit Media Independent Interface(RGMII) between + Ethernet physical media devices and the Gigabit Ethernet controller. + endif # PHYLIB config MICREL_KS8995MA diff --git a/drivers/net/phy/Makefile b/drivers/net/phy/Makefile index 534dfa74d5a2..73d65ce04454 100644 --- a/drivers/net/phy/Makefile +++ b/drivers/net/phy/Makefile @@ -11,6 +11,7 @@ obj-$(CONFIG_CICADA_PHY) += cicada.o obj-$(CONFIG_LXT_PHY) += lxt.o obj-$(CONFIG_QSEMI_PHY) += qsemi.o obj-$(CONFIG_SMSC_PHY) += smsc.o +obj-$(CONFIG_MICROSEMI_PHY) += mscc.o obj-$(CONFIG_TERANETICS_PHY) += teranetics.o obj-$(CONFIG_VITESSE_PHY) += vitesse.o obj-$(CONFIG_BCM_NET_PHYLIB) += bcm-phy-lib.o @@ -49,3 +50,4 @@ obj-$(CONFIG_MDIO_BCM_IPROC) += mdio-bcm-iproc.o obj-$(CONFIG_INTEL_XWAY_PHY) += intel-xway.o obj-$(CONFIG_MDIO_HISI_FEMAC) += mdio-hisi-femac.o obj-$(CONFIG_MDIO_XGENE) += mdio-xgene.o +obj-$(CONFIG_XILINX_GMII2RGMII) += xilinx_gmii2rgmii.o diff --git a/drivers/net/phy/mscc.c b/drivers/net/phy/mscc.c new file mode 100644 index 000000000000..ad33390b382a --- /dev/null +++ b/drivers/net/phy/mscc.c @@ -0,0 +1,161 @@ +/* + * Driver for Microsemi VSC85xx PHYs + * + * Author: Nagaraju Lakkaraju + * License: Dual MIT/GPL + * Copyright (c) 2016 Microsemi Corporation + */ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/mdio.h> +#include <linux/mii.h> +#include <linux/phy.h> + +enum rgmii_rx_clock_delay { + RGMII_RX_CLK_DELAY_0_2_NS = 0, + RGMII_RX_CLK_DELAY_0_8_NS = 1, + RGMII_RX_CLK_DELAY_1_1_NS = 2, + RGMII_RX_CLK_DELAY_1_7_NS = 3, + RGMII_RX_CLK_DELAY_2_0_NS = 4, + RGMII_RX_CLK_DELAY_2_3_NS = 5, + RGMII_RX_CLK_DELAY_2_6_NS = 6, + RGMII_RX_CLK_DELAY_3_4_NS = 7 +}; + +#define MII_VSC85XX_INT_MASK 25 +#define MII_VSC85XX_INT_MASK_MASK 0xa000 +#define MII_VSC85XX_INT_STATUS 26 + +#define MSCC_EXT_PAGE_ACCESS 31 +#define MSCC_PHY_PAGE_STANDARD 0x0000 /* Standard registers */ +#define MSCC_PHY_PAGE_EXTENDED_2 0x0002 /* Extended reg - page 2 */ + +/* Extended Page 2 Registers */ +#define MSCC_PHY_RGMII_CNTL 20 +#define RGMII_RX_CLK_DELAY_MASK 0x0070 +#define RGMII_RX_CLK_DELAY_POS 4 + +/* Microsemi PHY ID's */ +#define PHY_ID_VSC8531 0x00070570 +#define PHY_ID_VSC8541 0x00070770 + +static int vsc85xx_phy_page_set(struct phy_device *phydev, u8 page) +{ + int rc; + + rc = phy_write(phydev, MSCC_EXT_PAGE_ACCESS, page); + return rc; +} + +static int vsc85xx_default_config(struct phy_device *phydev) +{ + int rc; + u16 reg_val; + + mutex_lock(&phydev->lock); + rc = vsc85xx_phy_page_set(phydev, MSCC_PHY_PAGE_EXTENDED_2); + if (rc != 0) + goto out_unlock; + + reg_val = phy_read(phydev, MSCC_PHY_RGMII_CNTL); + reg_val &= ~(RGMII_RX_CLK_DELAY_MASK); + reg_val |= (RGMII_RX_CLK_DELAY_1_1_NS << RGMII_RX_CLK_DELAY_POS); + phy_write(phydev, MSCC_PHY_RGMII_CNTL, reg_val); + rc = vsc85xx_phy_page_set(phydev, MSCC_PHY_PAGE_STANDARD); + +out_unlock: + mutex_unlock(&phydev->lock); + + return rc; +} + +static int vsc85xx_config_init(struct phy_device *phydev) +{ + int rc; + + rc = vsc85xx_default_config(phydev); + if (rc) + return rc; + rc = genphy_config_init(phydev); + + return rc; +} + +static int vsc85xx_ack_interrupt(struct phy_device *phydev) +{ + int rc = 0; + + if (phydev->interrupts == PHY_INTERRUPT_ENABLED) + rc = phy_read(phydev, MII_VSC85XX_INT_STATUS); + + return (rc < 0) ? rc : 0; +} + +static int vsc85xx_config_intr(struct phy_device *phydev) +{ + int rc; + + if (phydev->interrupts == PHY_INTERRUPT_ENABLED) { + rc = phy_write(phydev, MII_VSC85XX_INT_MASK, + MII_VSC85XX_INT_MASK_MASK); + } else { + rc = phy_write(phydev, MII_VSC85XX_INT_MASK, 0); + if (rc < 0) + return rc; + rc = phy_read(phydev, MII_VSC85XX_INT_STATUS); + } + + return rc; +} + +/* Microsemi VSC85xx PHYs */ +static struct phy_driver vsc85xx_driver[] = { +{ + .phy_id = PHY_ID_VSC8531, + .name = "Microsemi VSC8531", + .phy_id_mask = 0xfffffff0, + .features = PHY_GBIT_FEATURES, + .flags = PHY_HAS_INTERRUPT, + .soft_reset = &genphy_soft_reset, + .config_init = &vsc85xx_config_init, + .config_aneg = &genphy_config_aneg, + .aneg_done = &genphy_aneg_done, + .read_status = &genphy_read_status, + .ack_interrupt = &vsc85xx_ack_interrupt, + .config_intr = &vsc85xx_config_intr, + .suspend = &genphy_suspend, + .resume = &genphy_resume, +}, +{ + .phy_id = PHY_ID_VSC8541, + .name = "Microsemi VSC8541 SyncE", + .phy_id_mask = 0xfffffff0, + .features = PHY_GBIT_FEATURES, + .flags = PHY_HAS_INTERRUPT, + .soft_reset = &genphy_soft_reset, + .config_init = &vsc85xx_config_init, + .config_aneg = &genphy_config_aneg, + .aneg_done = &genphy_aneg_done, + .read_status = &genphy_read_status, + .ack_interrupt = &vsc85xx_ack_interrupt, + .config_intr = &vsc85xx_config_intr, + .suspend = &genphy_suspend, + .resume = &genphy_resume, +} + +}; + +module_phy_driver(vsc85xx_driver); + +static struct mdio_device_id __maybe_unused vsc85xx_tbl[] = { + { PHY_ID_VSC8531, 0xfffffff0, }, + { PHY_ID_VSC8541, 0xfffffff0, }, + { } +}; + +MODULE_DEVICE_TABLE(mdio, vsc85xx_tbl); + +MODULE_DESCRIPTION("Microsemi VSC85xx PHY driver"); +MODULE_AUTHOR("Nagaraju Lakkaraju"); +MODULE_LICENSE("Dual MIT/GPL"); diff --git a/drivers/net/phy/xilinx_gmii2rgmii.c b/drivers/net/phy/xilinx_gmii2rgmii.c new file mode 100644 index 000000000000..cad6e19724e2 --- /dev/null +++ b/drivers/net/phy/xilinx_gmii2rgmii.c @@ -0,0 +1,109 @@ +/* Xilinx GMII2RGMII Converter driver + * + * Copyright (C) 2016 Xilinx, Inc. + * + * Author: Kedareswara rao Appana <appanad@xilinx.com> + * + * Description: + * This driver is developed for Xilinx GMII2RGMII Converter + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/mii.h> +#include <linux/mdio.h> +#include <linux/phy.h> +#include <linux/of_mdio.h> + +#define XILINX_GMII2RGMII_REG 0x10 +#define XILINX_GMII2RGMII_SPEED_MASK (BMCR_SPEED1000 | BMCR_SPEED100) + +struct gmii2rgmii { + struct phy_device *phy_dev; + struct phy_driver *phy_drv; + struct phy_driver conv_phy_drv; + int addr; +}; + +static int xgmiitorgmii_read_status(struct phy_device *phydev) +{ + struct gmii2rgmii *priv = phydev->priv; + u16 val = 0; + + priv->phy_drv->read_status(phydev); + + val = mdiobus_read(phydev->mdio.bus, priv->addr, XILINX_GMII2RGMII_REG); + val &= XILINX_GMII2RGMII_SPEED_MASK; + + if (phydev->speed == SPEED_1000) + val |= BMCR_SPEED1000; + else if (phydev->speed == SPEED_100) + val |= BMCR_SPEED100; + else + val |= BMCR_SPEED10; + + mdiobus_write(phydev->mdio.bus, priv->addr, XILINX_GMII2RGMII_REG, val); + + return 0; +} + +int xgmiitorgmii_probe(struct mdio_device *mdiodev) +{ + struct device *dev = &mdiodev->dev; + struct device_node *np = dev->of_node, *phy_node; + struct gmii2rgmii *priv; + + priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL); + if (!priv) + return -ENOMEM; + + phy_node = of_parse_phandle(np, "phy-handle", 0); + if (!phy_node) { + dev_err(dev, "Couldn't parse phy-handle\n"); + return -ENODEV; + } + + priv->phy_dev = of_phy_find_device(phy_node); + if (!priv->phy_dev) { + dev_info(dev, "Couldn't find phydev\n"); + return -EPROBE_DEFER; + } + + priv->addr = mdiodev->addr; + priv->phy_drv = priv->phy_dev->drv; + memcpy(&priv->conv_phy_drv, priv->phy_dev->drv, + sizeof(struct phy_driver)); + priv->conv_phy_drv.read_status = xgmiitorgmii_read_status; + priv->phy_dev->priv = priv; + priv->phy_dev->drv = &priv->conv_phy_drv; + + return 0; +} + +static const struct of_device_id xgmiitorgmii_of_match[] = { + { .compatible = "xlnx,gmii-to-rgmii-1.0" }, + {}, +}; +MODULE_DEVICE_TABLE(of, xgmiitorgmii_of_match); + +static struct mdio_driver xgmiitorgmii_driver = { + .probe = xgmiitorgmii_probe, + .mdiodrv.driver = { + .name = "xgmiitorgmii", + .of_match_table = xgmiitorgmii_of_match, + }, +}; + +mdio_module_driver(xgmiitorgmii_driver); + +MODULE_DESCRIPTION("Xilinx GMII2RGMII converter driver"); +MODULE_LICENSE("GPL"); diff --git a/drivers/net/ppp/ppp_generic.c b/drivers/net/ppp/ppp_generic.c index f226db4616b7..70cfa06ccd40 100644 --- a/drivers/net/ppp/ppp_generic.c +++ b/drivers/net/ppp/ppp_generic.c @@ -1103,6 +1103,15 @@ static int ppp_nl_newlink(struct net *src_net, struct net_device *dev, } conf.file = file; + + /* Don't use device name generated by the rtnetlink layer when ifname + * isn't specified. Let ppp_dev_configure() set the device name using + * the PPP unit identifer as suffix (i.e. ppp<unit_id>). This allows + * userspace to infer the device name using to the PPPIOCGUNIT ioctl. + */ + if (!tb[IFLA_IFNAME]) + conf.ifname_is_set = false; + err = ppp_dev_configure(src_net, dev, &conf); out_unlock: diff --git a/drivers/net/ppp/pptp.c b/drivers/net/ppp/pptp.c index ae0905ed4a32..1951b1085cb8 100644 --- a/drivers/net/ppp/pptp.c +++ b/drivers/net/ppp/pptp.c @@ -37,6 +37,7 @@ #include <net/icmp.h> #include <net/route.h> #include <net/gre.h> +#include <net/pptp.h> #include <linux/uaccess.h> @@ -53,41 +54,6 @@ static struct proto pptp_sk_proto __read_mostly; static const struct ppp_channel_ops pptp_chan_ops; static const struct proto_ops pptp_ops; -#define PPP_LCP_ECHOREQ 0x09 -#define PPP_LCP_ECHOREP 0x0A -#define SC_RCV_BITS (SC_RCV_B7_1|SC_RCV_B7_0|SC_RCV_ODDP|SC_RCV_EVNP) - -#define MISSING_WINDOW 20 -#define WRAPPED(curseq, lastseq)\ - ((((curseq) & 0xffffff00) == 0) &&\ - (((lastseq) & 0xffffff00) == 0xffffff00)) - -#define PPTP_GRE_PROTO 0x880B -#define PPTP_GRE_VER 0x1 - -#define PPTP_GRE_FLAG_C 0x80 -#define PPTP_GRE_FLAG_R 0x40 -#define PPTP_GRE_FLAG_K 0x20 -#define PPTP_GRE_FLAG_S 0x10 -#define PPTP_GRE_FLAG_A 0x80 - -#define PPTP_GRE_IS_C(f) ((f)&PPTP_GRE_FLAG_C) -#define PPTP_GRE_IS_R(f) ((f)&PPTP_GRE_FLAG_R) -#define PPTP_GRE_IS_K(f) ((f)&PPTP_GRE_FLAG_K) -#define PPTP_GRE_IS_S(f) ((f)&PPTP_GRE_FLAG_S) -#define PPTP_GRE_IS_A(f) ((f)&PPTP_GRE_FLAG_A) - -#define PPTP_HEADER_OVERHEAD (2+sizeof(struct pptp_gre_header)) -struct pptp_gre_header { - u8 flags; - u8 ver; - __be16 protocol; - __be16 payload_len; - __be16 call_id; - __be32 seq; - __be32 ack; -} __packed; - static struct pppox_sock *lookup_chan(u16 call_id, __be32 s_addr) { struct pppox_sock *sock; @@ -240,16 +206,14 @@ static int pptp_xmit(struct ppp_channel *chan, struct sk_buff *skb) skb_push(skb, header_len); hdr = (struct pptp_gre_header *)(skb->data); - hdr->flags = PPTP_GRE_FLAG_K; - hdr->ver = PPTP_GRE_VER; - hdr->protocol = htons(PPTP_GRE_PROTO); - hdr->call_id = htons(opt->dst_addr.call_id); + hdr->gre_hd.flags = GRE_KEY | GRE_VERSION_1 | GRE_SEQ; + hdr->gre_hd.protocol = GRE_PROTO_PPP; + hdr->call_id = htons(opt->dst_addr.call_id); - hdr->flags |= PPTP_GRE_FLAG_S; - hdr->seq = htonl(++opt->seq_sent); + hdr->seq = htonl(++opt->seq_sent); if (opt->ack_sent != seq_recv) { /* send ack with this message */ - hdr->ver |= PPTP_GRE_FLAG_A; + hdr->gre_hd.flags |= GRE_ACK; hdr->ack = htonl(seq_recv); opt->ack_sent = seq_recv; } @@ -312,7 +276,7 @@ static int pptp_rcv_core(struct sock *sk, struct sk_buff *skb) headersize = sizeof(*header); /* test if acknowledgement present */ - if (PPTP_GRE_IS_A(header->ver)) { + if (GRE_IS_ACK(header->gre_hd.flags)) { __u32 ack; if (!pskb_may_pull(skb, headersize)) @@ -320,7 +284,7 @@ static int pptp_rcv_core(struct sock *sk, struct sk_buff *skb) header = (struct pptp_gre_header *)(skb->data); /* ack in different place if S = 0 */ - ack = PPTP_GRE_IS_S(header->flags) ? header->ack : header->seq; + ack = GRE_IS_SEQ(header->gre_hd.flags) ? header->ack : header->seq; ack = ntohl(ack); @@ -333,7 +297,7 @@ static int pptp_rcv_core(struct sock *sk, struct sk_buff *skb) headersize -= sizeof(header->ack); } /* test if payload present */ - if (!PPTP_GRE_IS_S(header->flags)) + if (!GRE_IS_SEQ(header->gre_hd.flags)) goto drop; payload_len = ntohs(header->payload_len); @@ -394,11 +358,11 @@ static int pptp_rcv(struct sk_buff *skb) header = (struct pptp_gre_header *)skb->data; - if (ntohs(header->protocol) != PPTP_GRE_PROTO || /* PPTP-GRE protocol for PPTP */ - PPTP_GRE_IS_C(header->flags) || /* flag C should be clear */ - PPTP_GRE_IS_R(header->flags) || /* flag R should be clear */ - !PPTP_GRE_IS_K(header->flags) || /* flag K should be set */ - (header->flags&0xF) != 0) /* routing and recursion ctrl = 0 */ + if (header->gre_hd.protocol != GRE_PROTO_PPP || /* PPTP-GRE protocol for PPTP */ + GRE_IS_CSUM(header->gre_hd.flags) || /* flag CSUM should be clear */ + GRE_IS_ROUTING(header->gre_hd.flags) || /* flag ROUTING should be clear */ + !GRE_IS_KEY(header->gre_hd.flags) || /* flag KEY should be set */ + (header->gre_hd.flags & GRE_FLAGS)) /* flag Recursion Ctrl should be clear */ /* if invalid, discard this packet */ goto drop; diff --git a/drivers/net/usb/hso.c b/drivers/net/usb/hso.c index 4b4458616693..c5544d36c54f 100644 --- a/drivers/net/usb/hso.c +++ b/drivers/net/usb/hso.c @@ -2300,10 +2300,8 @@ static int hso_serial_common_create(struct hso_serial *serial, int num_urbs, serial->rx_data_length = rx_size; for (i = 0; i < serial->num_rx_urbs; i++) { serial->rx_urb[i] = usb_alloc_urb(0, GFP_KERNEL); - if (!serial->rx_urb[i]) { - dev_err(dev, "Could not allocate urb?\n"); + if (!serial->rx_urb[i]) goto exit; - } serial->rx_urb[i]->transfer_buffer = NULL; serial->rx_urb[i]->transfer_buffer_length = 0; serial->rx_data[i] = kzalloc(serial->rx_data_length, @@ -2314,10 +2312,8 @@ static int hso_serial_common_create(struct hso_serial *serial, int num_urbs, /* TX, allocate urb and initialize */ serial->tx_urb = usb_alloc_urb(0, GFP_KERNEL); - if (!serial->tx_urb) { - dev_err(dev, "Could not allocate urb?\n"); + if (!serial->tx_urb) goto exit; - } serial->tx_urb->transfer_buffer = NULL; serial->tx_urb->transfer_buffer_length = 0; /* prepare our TX buffer */ @@ -2555,20 +2551,16 @@ static struct hso_device *hso_create_net_device(struct usb_interface *interface, /* start allocating */ for (i = 0; i < MUX_BULK_RX_BUF_COUNT; i++) { hso_net->mux_bulk_rx_urb_pool[i] = usb_alloc_urb(0, GFP_KERNEL); - if (!hso_net->mux_bulk_rx_urb_pool[i]) { - dev_err(&interface->dev, "Could not allocate rx urb\n"); + if (!hso_net->mux_bulk_rx_urb_pool[i]) goto exit; - } hso_net->mux_bulk_rx_buf_pool[i] = kzalloc(MUX_BULK_RX_BUF_SIZE, GFP_KERNEL); if (!hso_net->mux_bulk_rx_buf_pool[i]) goto exit; } hso_net->mux_bulk_tx_urb = usb_alloc_urb(0, GFP_KERNEL); - if (!hso_net->mux_bulk_tx_urb) { - dev_err(&interface->dev, "Could not allocate tx urb\n"); + if (!hso_net->mux_bulk_tx_urb) goto exit; - } hso_net->mux_bulk_tx_buf = kzalloc(MUX_BULK_TX_BUF_SIZE, GFP_KERNEL); if (!hso_net->mux_bulk_tx_buf) goto exit; @@ -2787,10 +2779,8 @@ struct hso_shared_int *hso_create_shared_int(struct usb_interface *interface) } mux->shared_intr_urb = usb_alloc_urb(0, GFP_KERNEL); - if (!mux->shared_intr_urb) { - dev_err(&interface->dev, "Could not allocate intr urb?\n"); + if (!mux->shared_intr_urb) goto exit; - } mux->shared_intr_buf = kzalloc(le16_to_cpu(mux->intr_endp->wMaxPacketSize), GFP_KERNEL); diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c index 6a9d474b08b2..432b8a3ae354 100644 --- a/drivers/net/usb/lan78xx.c +++ b/drivers/net/usb/lan78xx.c @@ -3002,10 +3002,8 @@ static void lan78xx_tx_bh(struct lan78xx_net *dev) gso_skb: urb = usb_alloc_urb(0, GFP_ATOMIC); - if (!urb) { - netif_dbg(dev, tx_err, dev->net, "no urb\n"); + if (!urb) goto drop; - } entry = (struct skb_data *)skb->cb; entry->urb = urb; diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c index 3bfb59209326..d5071e364d40 100644 --- a/drivers/net/usb/usbnet.c +++ b/drivers/net/usb/usbnet.c @@ -2062,11 +2062,8 @@ int usbnet_write_cmd_async(struct usbnet *dev, u8 cmd, u8 reqtype, cmd, reqtype, value, index, size); urb = usb_alloc_urb(0, GFP_ATOMIC); - if (!urb) { - netdev_err(dev->net, "Error allocating URB in" - " %s!\n", __func__); + if (!urb) goto fail; - } if (data) { buf = kmemdup(data, size, GFP_ATOMIC); diff --git a/drivers/net/wimax/i2400m/usb-notif.c b/drivers/net/wimax/i2400m/usb-notif.c index fc1355d98bc6..5d429f816125 100644 --- a/drivers/net/wimax/i2400m/usb-notif.c +++ b/drivers/net/wimax/i2400m/usb-notif.c @@ -206,7 +206,6 @@ int i2400mu_notification_setup(struct i2400mu *i2400mu) i2400mu->notif_urb = usb_alloc_urb(0, GFP_KERNEL); if (!i2400mu->notif_urb) { ret = -ENOMEM; - dev_err(dev, "notification: cannot allocate URB\n"); goto error_alloc_urb; } epd = usb_get_epd(i2400mu->usb_iface, diff --git a/drivers/net/wireless/ath/ar5523/ar5523.c b/drivers/net/wireless/ath/ar5523/ar5523.c index 8aded24bcdf4..7a60d2e652da 100644 --- a/drivers/net/wireless/ath/ar5523/ar5523.c +++ b/drivers/net/wireless/ath/ar5523/ar5523.c @@ -706,10 +706,8 @@ static int ar5523_alloc_rx_bufs(struct ar5523 *ar) data->ar = ar; data->urb = usb_alloc_urb(0, GFP_KERNEL); - if (!data->urb) { - ar5523_err(ar, "could not allocate rx data urb\n"); + if (!data->urb) goto err; - } list_add_tail(&data->list, &ar->rx_data_free); atomic_inc(&ar->rx_data_free_cnt); } @@ -824,7 +822,6 @@ static void ar5523_tx_work_locked(struct ar5523 *ar) urb = usb_alloc_urb(0, GFP_KERNEL); if (!urb) { - ar5523_err(ar, "Failed to allocate TX urb\n"); ieee80211_free_txskb(ar->hw, skb); continue; } @@ -949,10 +946,8 @@ static int ar5523_alloc_tx_cmd(struct ar5523 *ar) init_completion(&cmd->done); cmd->urb_tx = usb_alloc_urb(0, GFP_KERNEL); - if (!cmd->urb_tx) { - ar5523_err(ar, "could not allocate urb\n"); + if (!cmd->urb_tx) return -ENOMEM; - } cmd->buf_tx = usb_alloc_coherent(ar->dev, AR5523_MAX_TXCMDSZ, GFP_KERNEL, &cmd->urb_tx->transfer_dma); diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c index 98b15a9a2779..fa26619a7945 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c @@ -1099,15 +1099,11 @@ struct brcmf_usbdev *brcmf_usb_attach(struct brcmf_usbdev_info *devinfo, devinfo->tx_freecount = ntxq; devinfo->ctl_urb = usb_alloc_urb(0, GFP_ATOMIC); - if (!devinfo->ctl_urb) { - brcmf_err("usb_alloc_urb (ctl) failed\n"); + if (!devinfo->ctl_urb) goto error; - } devinfo->bulk_urb = usb_alloc_urb(0, GFP_ATOMIC); - if (!devinfo->bulk_urb) { - brcmf_err("usb_alloc_urb (bulk) failed\n"); + if (!devinfo->bulk_urb) goto error; - } return &devinfo->bus_pub; diff --git a/drivers/net/wireless/intersil/orinoco/orinoco_usb.c b/drivers/net/wireless/intersil/orinoco/orinoco_usb.c index 56f109bc8394..bca6935a94db 100644 --- a/drivers/net/wireless/intersil/orinoco/orinoco_usb.c +++ b/drivers/net/wireless/intersil/orinoco/orinoco_usb.c @@ -1613,10 +1613,8 @@ static int ezusb_probe(struct usb_interface *interface, } upriv->read_urb = usb_alloc_urb(0, GFP_KERNEL); - if (!upriv->read_urb) { - err("No free urbs available"); + if (!upriv->read_urb) goto error; - } if (le16_to_cpu(ep->wMaxPacketSize) != 64) pr_warn("bulk in: wMaxPacketSize!= 64\n"); if (ep->bEndpointAddress != (2 | USB_DIR_IN)) diff --git a/drivers/net/wireless/marvell/libertas_tf/if_usb.c b/drivers/net/wireless/marvell/libertas_tf/if_usb.c index 799a2efe5793..e0ade40d9497 100644 --- a/drivers/net/wireless/marvell/libertas_tf/if_usb.c +++ b/drivers/net/wireless/marvell/libertas_tf/if_usb.c @@ -198,22 +198,16 @@ static int if_usb_probe(struct usb_interface *intf, } cardp->rx_urb = usb_alloc_urb(0, GFP_KERNEL); - if (!cardp->rx_urb) { - lbtf_deb_usbd(&udev->dev, "Rx URB allocation failed\n"); + if (!cardp->rx_urb) goto dealloc; - } cardp->tx_urb = usb_alloc_urb(0, GFP_KERNEL); - if (!cardp->tx_urb) { - lbtf_deb_usbd(&udev->dev, "Tx URB allocation failed\n"); + if (!cardp->tx_urb) goto dealloc; - } cardp->cmd_urb = usb_alloc_urb(0, GFP_KERNEL); - if (!cardp->cmd_urb) { - lbtf_deb_usbd(&udev->dev, "Cmd URB allocation failed\n"); + if (!cardp->cmd_urb) goto dealloc; - } cardp->ep_out_buf = kmalloc(MRVDRV_ETH_TX_PACKET_BUFFER_SIZE, GFP_KERNEL); diff --git a/drivers/net/wireless/marvell/mwifiex/usb.c b/drivers/net/wireless/marvell/mwifiex/usb.c index 0857575c5c39..3bd04f52f369 100644 --- a/drivers/net/wireless/marvell/mwifiex/usb.c +++ b/drivers/net/wireless/marvell/mwifiex/usb.c @@ -657,11 +657,8 @@ static int mwifiex_usb_tx_init(struct mwifiex_adapter *adapter) card->tx_cmd.ep = card->tx_cmd_ep; card->tx_cmd.urb = usb_alloc_urb(0, GFP_KERNEL); - if (!card->tx_cmd.urb) { - mwifiex_dbg(adapter, ERROR, - "tx_cmd.urb allocation failed\n"); + if (!card->tx_cmd.urb) return -ENOMEM; - } for (i = 0; i < MWIFIEX_TX_DATA_PORT; i++) { port = &card->port[i]; @@ -677,11 +674,8 @@ static int mwifiex_usb_tx_init(struct mwifiex_adapter *adapter) port->tx_data_list[j].ep = port->tx_data_ep; port->tx_data_list[j].urb = usb_alloc_urb(0, GFP_KERNEL); - if (!port->tx_data_list[j].urb) { - mwifiex_dbg(adapter, ERROR, - "urb allocation failed\n"); + if (!port->tx_data_list[j].urb) return -ENOMEM; - } } } @@ -697,10 +691,8 @@ static int mwifiex_usb_rx_init(struct mwifiex_adapter *adapter) card->rx_cmd.ep = card->rx_cmd_ep; card->rx_cmd.urb = usb_alloc_urb(0, GFP_KERNEL); - if (!card->rx_cmd.urb) { - mwifiex_dbg(adapter, ERROR, "rx_cmd.urb allocation failed\n"); + if (!card->rx_cmd.urb) return -ENOMEM; - } card->rx_cmd.skb = dev_alloc_skb(MWIFIEX_RX_CMD_BUF_SIZE); if (!card->rx_cmd.skb) @@ -714,11 +706,8 @@ static int mwifiex_usb_rx_init(struct mwifiex_adapter *adapter) card->rx_data_list[i].ep = card->rx_data_ep; card->rx_data_list[i].urb = usb_alloc_urb(0, GFP_KERNEL); - if (!card->rx_data_list[i].urb) { - mwifiex_dbg(adapter, ERROR, - "rx_data_list[] urb allocation failed\n"); + if (!card->rx_data_list[i].urb) return -1; - } if (mwifiex_usb_submit_rx_urb(&card->rx_data_list[i], MWIFIEX_RX_DATA_BUF_SIZE)) return -1; diff --git a/drivers/net/wireless/realtek/rtlwifi/usb.c b/drivers/net/wireless/realtek/rtlwifi/usb.c index 41617b7b0822..32aa5c1d070a 100644 --- a/drivers/net/wireless/realtek/rtlwifi/usb.c +++ b/drivers/net/wireless/realtek/rtlwifi/usb.c @@ -739,11 +739,8 @@ static int _rtl_usb_receive(struct ieee80211_hw *hw) for (i = 0; i < rtlusb->rx_urb_num; i++) { err = -ENOMEM; urb = usb_alloc_urb(0, GFP_KERNEL); - if (!urb) { - RT_TRACE(rtlpriv, COMP_USB, DBG_EMERG, - "Failed to alloc URB!!\n"); + if (!urb) goto err_out; - } err = _rtl_prep_rx_urb(hw, rtlusb, urb, GFP_KERNEL); if (err < 0) { @@ -907,15 +904,12 @@ static void _rtl_tx_complete(struct urb *urb) static struct urb *_rtl_usb_tx_urb_setup(struct ieee80211_hw *hw, struct sk_buff *skb, u32 ep_num) { - struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_usb *rtlusb = rtl_usbdev(rtl_usbpriv(hw)); struct urb *_urb; WARN_ON(NULL == skb); _urb = usb_alloc_urb(0, GFP_ATOMIC); if (!_urb) { - RT_TRACE(rtlpriv, COMP_USB, DBG_EMERG, - "Can't allocate URB for bulk out!\n"); kfree_skb(skb); return NULL; } diff --git a/drivers/staging/octeon/ethernet-mdio.c b/drivers/staging/octeon/ethernet-mdio.c index e13a4ab46977..1fde9c824948 100644 --- a/drivers/staging/octeon/ethernet-mdio.c +++ b/drivers/staging/octeon/ethernet-mdio.c @@ -34,48 +34,23 @@ static void cvm_oct_get_drvinfo(struct net_device *dev, strlcpy(info->bus_info, "Builtin", sizeof(info->bus_info)); } -static int cvm_oct_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct octeon_ethernet *priv = netdev_priv(dev); - - if (priv->phydev) - return phy_ethtool_gset(priv->phydev, cmd); - - return -EINVAL; -} - -static int cvm_oct_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct octeon_ethernet *priv = netdev_priv(dev); - - if (!capable(CAP_NET_ADMIN)) - return -EPERM; - - if (priv->phydev) - return phy_ethtool_sset(priv->phydev, cmd); - - return -EINVAL; -} - static int cvm_oct_nway_reset(struct net_device *dev) { - struct octeon_ethernet *priv = netdev_priv(dev); - if (!capable(CAP_NET_ADMIN)) return -EPERM; - if (priv->phydev) - return phy_start_aneg(priv->phydev); + if (dev->phydev) + return phy_start_aneg(dev->phydev); return -EINVAL; } const struct ethtool_ops cvm_oct_ethtool_ops = { .get_drvinfo = cvm_oct_get_drvinfo, - .get_settings = cvm_oct_get_settings, - .set_settings = cvm_oct_set_settings, .nway_reset = cvm_oct_nway_reset, .get_link = ethtool_op_get_link, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; /** @@ -88,15 +63,13 @@ const struct ethtool_ops cvm_oct_ethtool_ops = { */ int cvm_oct_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { - struct octeon_ethernet *priv = netdev_priv(dev); - if (!netif_running(dev)) return -EINVAL; - if (!priv->phydev) + if (!dev->phydev) return -EINVAL; - return phy_mii_ioctl(priv->phydev, rq, cmd); + return phy_mii_ioctl(dev->phydev, rq, cmd); } void cvm_oct_note_carrier(struct octeon_ethernet *priv, @@ -119,9 +92,9 @@ void cvm_oct_adjust_link(struct net_device *dev) cvmx_helper_link_info_t link_info; link_info.u64 = 0; - link_info.s.link_up = priv->phydev->link ? 1 : 0; - link_info.s.full_duplex = priv->phydev->duplex ? 1 : 0; - link_info.s.speed = priv->phydev->speed; + link_info.s.link_up = dev->phydev->link ? 1 : 0; + link_info.s.full_duplex = dev->phydev->duplex ? 1 : 0; + link_info.s.speed = dev->phydev->speed; priv->link_info = link_info.u64; /* @@ -130,8 +103,8 @@ void cvm_oct_adjust_link(struct net_device *dev) if (priv->poll) priv->poll(dev); - if (priv->last_link != priv->phydev->link) { - priv->last_link = priv->phydev->link; + if (priv->last_link != dev->phydev->link) { + priv->last_link = dev->phydev->link; cvmx_helper_link_set(priv->port, link_info); cvm_oct_note_carrier(priv, link_info); } @@ -151,9 +124,8 @@ int cvm_oct_common_stop(struct net_device *dev) priv->poll = NULL; - if (priv->phydev) - phy_disconnect(priv->phydev); - priv->phydev = NULL; + if (dev->phydev) + phy_disconnect(dev->phydev); if (priv->last_link) { link_info.u64 = 0; @@ -176,6 +148,7 @@ int cvm_oct_phy_setup_device(struct net_device *dev) { struct octeon_ethernet *priv = netdev_priv(dev); struct device_node *phy_node; + struct phy_device *phydev = NULL; if (!priv->of_node) goto no_phy; @@ -193,14 +166,14 @@ int cvm_oct_phy_setup_device(struct net_device *dev) if (!phy_node) goto no_phy; - priv->phydev = of_phy_connect(dev, phy_node, cvm_oct_adjust_link, 0, - PHY_INTERFACE_MODE_GMII); + phydev = of_phy_connect(dev, phy_node, cvm_oct_adjust_link, 0, + PHY_INTERFACE_MODE_GMII); - if (!priv->phydev) + if (!phydev) return -ENODEV; priv->last_link = 0; - phy_start_aneg(priv->phydev); + phy_start_aneg(phydev); return 0; no_phy: diff --git a/drivers/staging/octeon/ethernet-rgmii.c b/drivers/staging/octeon/ethernet-rgmii.c index 91b148cfcbdb..48846dffc8e1 100644 --- a/drivers/staging/octeon/ethernet-rgmii.c +++ b/drivers/staging/octeon/ethernet-rgmii.c @@ -145,7 +145,7 @@ int cvm_oct_rgmii_open(struct net_device *dev) if (ret) return ret; - if (priv->phydev) { + if (dev->phydev) { /* * In phydev mode, we need still periodic polling for the * preamble error checking, and we also need to call this diff --git a/drivers/staging/octeon/ethernet.c b/drivers/staging/octeon/ethernet.c index e9cd5f242921..45d576361319 100644 --- a/drivers/staging/octeon/ethernet.c +++ b/drivers/staging/octeon/ethernet.c @@ -457,10 +457,8 @@ int cvm_oct_common_init(struct net_device *dev) void cvm_oct_common_uninit(struct net_device *dev) { - struct octeon_ethernet *priv = netdev_priv(dev); - - if (priv->phydev) - phy_disconnect(priv->phydev); + if (dev->phydev) + phy_disconnect(dev->phydev); } int cvm_oct_common_open(struct net_device *dev, @@ -484,10 +482,10 @@ int cvm_oct_common_open(struct net_device *dev, if (octeon_is_simulation()) return 0; - if (priv->phydev) { - int r = phy_read_status(priv->phydev); + if (dev->phydev) { + int r = phy_read_status(dev->phydev); - if (r == 0 && priv->phydev->link == 0) + if (r == 0 && dev->phydev->link == 0) netif_carrier_off(dev); cvm_oct_adjust_link(dev); } else { diff --git a/drivers/staging/octeon/octeon-ethernet.h b/drivers/staging/octeon/octeon-ethernet.h index 6275c15e0035..d533aefe085a 100644 --- a/drivers/staging/octeon/octeon-ethernet.h +++ b/drivers/staging/octeon/octeon-ethernet.h @@ -40,7 +40,6 @@ struct octeon_ethernet { struct sk_buff_head tx_free_list[16]; /* Device statistics */ struct net_device_stats stats; - struct phy_device *phydev; unsigned int last_speed; unsigned int last_link; /* Last negotiated link state */ diff --git a/fs/proc/generic.c b/fs/proc/generic.c index c633476616e0..bca66d83a765 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -390,6 +390,8 @@ static struct proc_dir_entry *__proc_create(struct proc_dir_entry **parent, atomic_set(&ent->count, 1); spin_lock_init(&ent->pde_unload_lock); INIT_LIST_HEAD(&ent->pde_openers); + proc_set_user(ent, (*parent)->uid, (*parent)->gid); + out: return ent; } diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c index c8bbc68cdb05..7ae6b1da7cab 100644 --- a/fs/proc/proc_net.c +++ b/fs/proc/proc_net.c @@ -21,6 +21,7 @@ #include <linux/bitops.h> #include <linux/mount.h> #include <linux/nsproxy.h> +#include <linux/uidgid.h> #include <net/net_namespace.h> #include <linux/seq_file.h> @@ -185,6 +186,8 @@ const struct file_operations proc_net_operations = { static __net_init int proc_net_ns_init(struct net *net) { struct proc_dir_entry *netd, *net_statd; + kuid_t uid; + kgid_t gid; int err; err = -ENOMEM; @@ -199,6 +202,16 @@ static __net_init int proc_net_ns_init(struct net *net) netd->parent = &proc_root; memcpy(netd->name, "net", 4); + uid = make_kuid(net->user_ns, 0); + if (!uid_valid(uid)) + uid = netd->uid; + + gid = make_kgid(net->user_ns, 0); + if (!gid_valid(gid)) + gid = netd->gid; + + proc_set_user(netd, uid, gid); + err = -EEXIST; net_statd = proc_net_mkdir(net, "stat", netd); if (!net_statd) diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index b59db94d2ff4..62d8c6975d34 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -430,6 +430,7 @@ static int sysctl_perm(struct ctl_table_header *head, struct ctl_table *table, i static struct inode *proc_sys_make_inode(struct super_block *sb, struct ctl_table_header *head, struct ctl_table *table) { + struct ctl_table_root *root = head->root; struct inode *inode; struct proc_inode *ei; @@ -457,6 +458,10 @@ static struct inode *proc_sys_make_inode(struct super_block *sb, if (is_empty_dir(head)) make_empty_dir_inode(inode); } + + if (root->set_ownership) + root->set_ownership(head, table, &inode->i_uid, &inode->i_gid); + out: return inode; } diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index 984f73b719a9..a4414a11eea7 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -497,6 +497,23 @@ static inline bool cgroup_is_descendant(struct cgroup *cgrp, return cgrp->ancestor_ids[ancestor->level] == ancestor->id; } +/** + * task_under_cgroup_hierarchy - test task's membership of cgroup ancestry + * @task: the task to be tested + * @ancestor: possible ancestor of @task's cgroup + * + * Tests whether @task's default cgroup hierarchy is a descendant of @ancestor. + * It follows all the same rules as cgroup_is_descendant, and only applies + * to the default hierarchy. + */ +static inline bool task_under_cgroup_hierarchy(struct task_struct *task, + struct cgroup *ancestor) +{ + struct css_set *cset = task_css_set(task); + + return cgroup_is_descendant(cset->dfl_cgrp, ancestor); +} + /* no synchronization, the result can only be used as a hint */ static inline bool cgroup_is_populated(struct cgroup *cgrp) { @@ -557,6 +574,7 @@ static inline void pr_cont_cgroup_path(struct cgroup *cgrp) #else /* !CONFIG_CGROUPS */ struct cgroup_subsys_state; +struct cgroup; static inline void css_put(struct cgroup_subsys_state *css) {} static inline int cgroup_attach_task_all(struct task_struct *from, @@ -574,6 +592,11 @@ static inline void cgroup_free(struct task_struct *p) {} static inline int cgroup_init_early(void) { return 0; } static inline int cgroup_init(void) { return 0; } +static inline bool task_under_cgroup_hierarchy(struct task_struct *task, + struct cgroup *ancestor) +{ + return true; +} #endif /* !CONFIG_CGROUPS */ /* diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 076df5360ba5..96e0b6cd964e 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -52,6 +52,7 @@ #include <uapi/linux/netdevice.h> #include <uapi/linux/if_bonding.h> #include <uapi/linux/pkt_cls.h> +#include <linux/hashtable.h> struct netpoll_info; struct device; @@ -1800,6 +1801,9 @@ struct net_device { unsigned int num_tx_queues; unsigned int real_num_tx_queues; struct Qdisc *qdisc; +#ifdef CONFIG_NET_SCHED + DECLARE_HASHTABLE (qdisc_hash, 4); +#endif unsigned long tx_queue_len; spinlock_t tx_global_lock; int watchdog_timeo; diff --git a/include/linux/qed/qed_if.h b/include/linux/qed/qed_if.h index b1e3c57c7117..737fc4c8db49 100644 --- a/include/linux/qed/qed_if.h +++ b/include/linux/qed/qed_if.h @@ -268,6 +268,21 @@ enum qed_protocol { QED_PROTOCOL_ISCSI, }; +enum qed_link_mode_bits { + QED_LM_FIBRE_BIT = BIT(0), + QED_LM_Autoneg_BIT = BIT(1), + QED_LM_Asym_Pause_BIT = BIT(2), + QED_LM_Pause_BIT = BIT(3), + QED_LM_1000baseT_Half_BIT = BIT(4), + QED_LM_1000baseT_Full_BIT = BIT(5), + QED_LM_10000baseKR_Full_BIT = BIT(6), + QED_LM_25000baseKR_Full_BIT = BIT(7), + QED_LM_40000baseLR4_Full_BIT = BIT(8), + QED_LM_50000baseKR2_Full_BIT = BIT(9), + QED_LM_100000baseKR4_Full_BIT = BIT(10), + QED_LM_COUNT = 11 +}; + struct qed_link_params { bool link_up; diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 697e160c78d0..d82cb6011e77 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -25,6 +25,7 @@ #include <linux/rcupdate.h> #include <linux/wait.h> #include <linux/rbtree.h> +#include <linux/uidgid.h> #include <uapi/linux/sysctl.h> /* For the /proc/sys support */ @@ -157,6 +158,9 @@ struct ctl_table_root { struct ctl_table_set default_set; struct ctl_table_set *(*lookup)(struct ctl_table_root *root, struct nsproxy *namespaces); + void (*set_ownership)(struct ctl_table_header *head, + struct ctl_table *table, + kuid_t *uid, kgid_t *gid); int (*permissions)(struct ctl_table_header *head, struct ctl_table *table); }; diff --git a/include/net/gre.h b/include/net/gre.h index 7a54a31d1d4c..8962e1e68449 100644 --- a/include/net/gre.h +++ b/include/net/gre.h @@ -7,7 +7,15 @@ struct gre_base_hdr { __be16 flags; __be16 protocol; -}; +} __packed; + +struct gre_full_hdr { + struct gre_base_hdr fixed_header; + __be16 csum; + __be16 reserved1; + __be32 key; + __be32 seq; +} __packed; #define GRE_HEADER_SECTION 4 #define GREPROTO_CISCO 0 diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index d061ffeb1e71..7adf4386ac8f 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -40,7 +40,6 @@ struct netns_ipv4 { #ifdef CONFIG_IP_MULTIPLE_TABLES struct fib_rules_ops *rules_ops; bool fib_has_custom_rules; - struct fib_table __rcu *fib_local; struct fib_table __rcu *fib_main; struct fib_table __rcu *fib_default; #endif diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h index 7caa99b482c6..cd334c9584e9 100644 --- a/include/net/pkt_sched.h +++ b/include/net/pkt_sched.h @@ -90,8 +90,8 @@ int unregister_qdisc(struct Qdisc_ops *qops); void qdisc_get_default(char *id, size_t len); int qdisc_set_default(const char *id); -void qdisc_list_add(struct Qdisc *q); -void qdisc_list_del(struct Qdisc *q); +void qdisc_hash_add(struct Qdisc *q); +void qdisc_hash_del(struct Qdisc *q); struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle); struct Qdisc *qdisc_lookup_class(struct net_device *dev, u32 handle); struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r, diff --git a/include/net/pptp.h b/include/net/pptp.h new file mode 100644 index 000000000000..92e9f1fe2628 --- /dev/null +++ b/include/net/pptp.h @@ -0,0 +1,23 @@ +#ifndef _NET_PPTP_H +#define _NET_PPTP_H + +#define PPP_LCP_ECHOREQ 0x09 +#define PPP_LCP_ECHOREP 0x0A +#define SC_RCV_BITS (SC_RCV_B7_1|SC_RCV_B7_0|SC_RCV_ODDP|SC_RCV_EVNP) + +#define MISSING_WINDOW 20 +#define WRAPPED(curseq, lastseq)\ + ((((curseq) & 0xffffff00) == 0) &&\ + (((lastseq) & 0xffffff00) == 0xffffff00)) + +#define PPTP_HEADER_OVERHEAD (2+sizeof(struct pptp_gre_header)) +struct pptp_gre_header { + struct gre_base_hdr gre_hd; + __be16 payload_len; + __be16 call_id; + __be32 seq; + __be32 ack; +} __packed; + + +#endif diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 909aff2db2b3..0d501779cc68 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -61,7 +61,7 @@ struct Qdisc { u32 limit; const struct Qdisc_ops *ops; struct qdisc_size_table __rcu *stab; - struct list_head list; + struct hlist_node hash; u32 handle; u32 parent; void *u32_node; diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index da218fec6056..bea0c4e2830a 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -375,6 +375,17 @@ enum bpf_func_id { */ BPF_FUNC_probe_write_user, + /** + * bpf_current_task_under_cgroup(map, index) - Check cgroup2 membership of current task + * @map: pointer to bpf_map in BPF_MAP_TYPE_CGROUP_ARRAY type + * @index: index of the cgroup in the bpf_map + * Return: + * == 0 current failed the cgroup2 descendant test + * == 1 current succeeded the cgroup2 descendant test + * < 0 error + */ + BPF_FUNC_current_task_under_cgroup, + __BPF_FUNC_MAX_ID, }; diff --git a/include/uapi/linux/if_tunnel.h b/include/uapi/linux/if_tunnel.h index 1046f5515174..361b9f0f20d7 100644 --- a/include/uapi/linux/if_tunnel.h +++ b/include/uapi/linux/if_tunnel.h @@ -24,9 +24,22 @@ #define GRE_SEQ __cpu_to_be16(0x1000) #define GRE_STRICT __cpu_to_be16(0x0800) #define GRE_REC __cpu_to_be16(0x0700) -#define GRE_FLAGS __cpu_to_be16(0x00F8) +#define GRE_ACK __cpu_to_be16(0x0080) +#define GRE_FLAGS __cpu_to_be16(0x0078) #define GRE_VERSION __cpu_to_be16(0x0007) +#define GRE_IS_CSUM(f) ((f) & GRE_CSUM) +#define GRE_IS_ROUTING(f) ((f) & GRE_ROUTING) +#define GRE_IS_KEY(f) ((f) & GRE_KEY) +#define GRE_IS_SEQ(f) ((f) & GRE_SEQ) +#define GRE_IS_STRICT(f) ((f) & GRE_STRICT) +#define GRE_IS_REC(f) ((f) & GRE_REC) +#define GRE_IS_ACK(f) ((f) & GRE_ACK) + +#define GRE_VERSION_1 __cpu_to_be16(0x0001) +#define GRE_PROTO_PPP __cpu_to_be16(0x880b) +#define GRE_PPTP_KEY_MASK __cpu_to_be32(0xffff) + struct ip_tunnel_parm { char name[IFNAMSIZ]; int link; diff --git a/include/uapi/linux/mii.h b/include/uapi/linux/mii.h index 237fac4bc17b..15d8510cdae0 100644 --- a/include/uapi/linux/mii.h +++ b/include/uapi/linux/mii.h @@ -48,6 +48,7 @@ #define BMCR_SPEED100 0x2000 /* Select 100Mbps */ #define BMCR_LOOPBACK 0x4000 /* TXD loopback bits */ #define BMCR_RESET 0x8000 /* Reset to default state */ +#define BMCR_SPEED10 0x0000 /* Select 10Mbps */ /* Basic mode status register. */ #define BMSR_ERCAP 0x0001 /* Ext-reg capability */ diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index 633a650d7aeb..a2ac051c342f 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -538,7 +538,7 @@ static int __init register_perf_event_array_map(void) } late_initcall(register_perf_event_array_map); -#ifdef CONFIG_SOCK_CGROUP_DATA +#ifdef CONFIG_CGROUPS static void *cgroup_fd_array_get_ptr(struct bpf_map *map, struct file *map_file /* not used */, int fd) diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 7094c69ac199..0cc46f1df358 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -930,14 +930,14 @@ static int check_func_arg(struct verifier_env *env, u32 regno, enum bpf_arg_type arg_type, struct bpf_call_arg_meta *meta) { - struct reg_state *reg = env->cur_state.regs + regno; - enum bpf_reg_type expected_type; + struct reg_state *regs = env->cur_state.regs, *reg = ®s[regno]; + enum bpf_reg_type expected_type, type = reg->type; int err = 0; if (arg_type == ARG_DONTCARE) return 0; - if (reg->type == NOT_INIT) { + if (type == NOT_INIT) { verbose("R%d !read_ok\n", regno); return -EACCES; } @@ -950,16 +950,29 @@ static int check_func_arg(struct verifier_env *env, u32 regno, return 0; } + if (type == PTR_TO_PACKET && !may_write_pkt_data(env->prog->type)) { + verbose("helper access to the packet is not allowed for clsact\n"); + return -EACCES; + } + if (arg_type == ARG_PTR_TO_MAP_KEY || arg_type == ARG_PTR_TO_MAP_VALUE) { expected_type = PTR_TO_STACK; + if (type != PTR_TO_PACKET && type != expected_type) + goto err_type; } else if (arg_type == ARG_CONST_STACK_SIZE || arg_type == ARG_CONST_STACK_SIZE_OR_ZERO) { expected_type = CONST_IMM; + if (type != expected_type) + goto err_type; } else if (arg_type == ARG_CONST_MAP_PTR) { expected_type = CONST_PTR_TO_MAP; + if (type != expected_type) + goto err_type; } else if (arg_type == ARG_PTR_TO_CTX) { expected_type = PTR_TO_CTX; + if (type != expected_type) + goto err_type; } else if (arg_type == ARG_PTR_TO_STACK || arg_type == ARG_PTR_TO_RAW_STACK) { expected_type = PTR_TO_STACK; @@ -967,20 +980,16 @@ static int check_func_arg(struct verifier_env *env, u32 regno, * passed in as argument, it's a CONST_IMM type. Final test * happens during stack boundary checking. */ - if (reg->type == CONST_IMM && reg->imm == 0) - expected_type = CONST_IMM; + if (type == CONST_IMM && reg->imm == 0) + /* final test in check_stack_boundary() */; + else if (type != PTR_TO_PACKET && type != expected_type) + goto err_type; meta->raw_mode = arg_type == ARG_PTR_TO_RAW_STACK; } else { verbose("unsupported arg_type %d\n", arg_type); return -EFAULT; } - if (reg->type != expected_type) { - verbose("R%d type=%s expected=%s\n", regno, - reg_type_str[reg->type], reg_type_str[expected_type]); - return -EACCES; - } - if (arg_type == ARG_CONST_MAP_PTR) { /* bpf_map_xxx(map_ptr) call: remember that map_ptr */ meta->map_ptr = reg->map_ptr; @@ -998,8 +1007,13 @@ static int check_func_arg(struct verifier_env *env, u32 regno, verbose("invalid map_ptr to access map->key\n"); return -EACCES; } - err = check_stack_boundary(env, regno, meta->map_ptr->key_size, - false, NULL); + if (type == PTR_TO_PACKET) + err = check_packet_access(env, regno, 0, + meta->map_ptr->key_size); + else + err = check_stack_boundary(env, regno, + meta->map_ptr->key_size, + false, NULL); } else if (arg_type == ARG_PTR_TO_MAP_VALUE) { /* bpf_map_xxx(..., map_ptr, ..., value) call: * check [value, value + map->value_size) validity @@ -1009,9 +1023,13 @@ static int check_func_arg(struct verifier_env *env, u32 regno, verbose("invalid map_ptr to access map->value\n"); return -EACCES; } - err = check_stack_boundary(env, regno, - meta->map_ptr->value_size, - false, NULL); + if (type == PTR_TO_PACKET) + err = check_packet_access(env, regno, 0, + meta->map_ptr->value_size); + else + err = check_stack_boundary(env, regno, + meta->map_ptr->value_size, + false, NULL); } else if (arg_type == ARG_CONST_STACK_SIZE || arg_type == ARG_CONST_STACK_SIZE_OR_ZERO) { bool zero_size_allowed = (arg_type == ARG_CONST_STACK_SIZE_OR_ZERO); @@ -1025,11 +1043,18 @@ static int check_func_arg(struct verifier_env *env, u32 regno, verbose("ARG_CONST_STACK_SIZE cannot be first argument\n"); return -EACCES; } - err = check_stack_boundary(env, regno - 1, reg->imm, - zero_size_allowed, meta); + if (regs[regno - 1].type == PTR_TO_PACKET) + err = check_packet_access(env, regno - 1, 0, reg->imm); + else + err = check_stack_boundary(env, regno - 1, reg->imm, + zero_size_allowed, meta); } return err; +err_type: + verbose("R%d type=%s expected=%s\n", regno, + reg_type_str[type], reg_type_str[expected_type]); + return -EACCES; } static int check_map_func_compatibility(struct bpf_map *map, int func_id) @@ -1053,7 +1078,8 @@ static int check_map_func_compatibility(struct bpf_map *map, int func_id) goto error; break; case BPF_MAP_TYPE_CGROUP_ARRAY: - if (func_id != BPF_FUNC_skb_in_cgroup) + if (func_id != BPF_FUNC_skb_in_cgroup && + func_id != BPF_FUNC_current_task_under_cgroup) goto error; break; default: @@ -1075,6 +1101,7 @@ static int check_map_func_compatibility(struct bpf_map *map, int func_id) if (map->map_type != BPF_MAP_TYPE_STACK_TRACE) goto error; break; + case BPF_FUNC_current_task_under_cgroup: case BPF_FUNC_skb_in_cgroup: if (map->map_type != BPF_MAP_TYPE_CGROUP_ARRAY) goto error; diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index b20438fdb029..ad35213b8405 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -376,6 +376,34 @@ static const struct bpf_func_proto bpf_get_current_task_proto = { .ret_type = RET_INTEGER, }; +static u64 bpf_current_task_under_cgroup(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) +{ + struct bpf_map *map = (struct bpf_map *)(long)r1; + struct bpf_array *array = container_of(map, struct bpf_array, map); + struct cgroup *cgrp; + u32 idx = (u32)r2; + + if (unlikely(in_interrupt())) + return -EINVAL; + + if (unlikely(idx >= array->map.max_entries)) + return -E2BIG; + + cgrp = READ_ONCE(array->ptrs[idx]); + if (unlikely(!cgrp)) + return -EAGAIN; + + return task_under_cgroup_hierarchy(current, cgrp); +} + +static const struct bpf_func_proto bpf_current_task_under_cgroup_proto = { + .func = bpf_current_task_under_cgroup, + .gpl_only = false, + .ret_type = RET_INTEGER, + .arg1_type = ARG_CONST_MAP_PTR, + .arg2_type = ARG_ANYTHING, +}; + static const struct bpf_func_proto *tracing_func_proto(enum bpf_func_id func_id) { switch (func_id) { @@ -407,6 +435,10 @@ static const struct bpf_func_proto *tracing_func_proto(enum bpf_func_id func_id) return &bpf_perf_event_read_proto; case BPF_FUNC_probe_write_user: return bpf_get_probe_write_proto(); + case BPF_FUNC_current_task_under_cgroup: + return &bpf_current_task_under_cgroup_proto; + case BPF_FUNC_get_prandom_u32: + return &bpf_get_prandom_u32_proto; default: return NULL; } diff --git a/net/core/dev.c b/net/core/dev.c index 4ce07dc25573..936ea0054f57 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -7629,6 +7629,9 @@ struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name, INIT_LIST_HEAD(&dev->all_adj_list.lower); INIT_LIST_HEAD(&dev->ptype_all); INIT_LIST_HEAD(&dev->ptype_specific); +#ifdef CONFIG_NET_SCHED + hash_init(dev->qdisc_hash); +#endif dev->priv_flags = IFF_XMIT_DST_RELEASE | IFF_XMIT_DST_RELEASE_PERM; setup(dev); diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c index 61ad43f61c5e..91028ae2fb01 100644 --- a/net/core/flow_dissector.c +++ b/net/core/flow_dissector.c @@ -6,6 +6,8 @@ #include <linux/if_vlan.h> #include <net/ip.h> #include <net/ipv6.h> +#include <net/gre.h> +#include <net/pptp.h> #include <linux/igmp.h> #include <linux/icmp.h> #include <linux/sctp.h> @@ -338,32 +340,42 @@ mpls: ip_proto_again: switch (ip_proto) { case IPPROTO_GRE: { - struct gre_hdr { - __be16 flags; - __be16 proto; - } *hdr, _hdr; + struct gre_base_hdr *hdr, _hdr; + u16 gre_ver; + int offset = 0; hdr = __skb_header_pointer(skb, nhoff, sizeof(_hdr), data, hlen, &_hdr); if (!hdr) goto out_bad; - /* - * Only look inside GRE if version zero and no - * routing - */ - if (hdr->flags & (GRE_VERSION | GRE_ROUTING)) + + /* Only look inside GRE without routing */ + if (hdr->flags & GRE_ROUTING) break; - proto = hdr->proto; - nhoff += 4; + /* Only look inside GRE for version 0 and 1 */ + gre_ver = ntohs(hdr->flags & GRE_VERSION); + if (gre_ver > 1) + break; + + proto = hdr->protocol; + if (gre_ver) { + /* Version1 must be PPTP, and check the flags */ + if (!(proto == GRE_PROTO_PPP && (hdr->flags & GRE_KEY))) + break; + } + + offset += sizeof(struct gre_base_hdr); + if (hdr->flags & GRE_CSUM) - nhoff += 4; + offset += sizeof(((struct gre_full_hdr *)0)->csum) + + sizeof(((struct gre_full_hdr *)0)->reserved1); + if (hdr->flags & GRE_KEY) { const __be32 *keyid; __be32 _keyid; - keyid = __skb_header_pointer(skb, nhoff, sizeof(_keyid), + keyid = __skb_header_pointer(skb, nhoff + offset, sizeof(_keyid), data, hlen, &_keyid); - if (!keyid) goto out_bad; @@ -372,32 +384,65 @@ ip_proto_again: key_keyid = skb_flow_dissector_target(flow_dissector, FLOW_DISSECTOR_KEY_GRE_KEYID, target_container); - key_keyid->keyid = *keyid; + if (gre_ver == 0) + key_keyid->keyid = *keyid; + else + key_keyid->keyid = *keyid & GRE_PPTP_KEY_MASK; } - nhoff += 4; + offset += sizeof(((struct gre_full_hdr *)0)->key); } + if (hdr->flags & GRE_SEQ) - nhoff += 4; - if (proto == htons(ETH_P_TEB)) { - const struct ethhdr *eth; - struct ethhdr _eth; - - eth = __skb_header_pointer(skb, nhoff, - sizeof(_eth), - data, hlen, &_eth); - if (!eth) + offset += sizeof(((struct pptp_gre_header *)0)->seq); + + if (gre_ver == 0) { + if (proto == htons(ETH_P_TEB)) { + const struct ethhdr *eth; + struct ethhdr _eth; + + eth = __skb_header_pointer(skb, nhoff + offset, + sizeof(_eth), + data, hlen, &_eth); + if (!eth) + goto out_bad; + proto = eth->h_proto; + offset += sizeof(*eth); + + /* Cap headers that we access via pointers at the + * end of the Ethernet header as our maximum alignment + * at that point is only 2 bytes. + */ + if (NET_IP_ALIGN) + hlen = (nhoff + offset); + } + } else { /* version 1, must be PPTP */ + u8 _ppp_hdr[PPP_HDRLEN]; + u8 *ppp_hdr; + + if (hdr->flags & GRE_ACK) + offset += sizeof(((struct pptp_gre_header *)0)->ack); + + ppp_hdr = skb_header_pointer(skb, nhoff + offset, + sizeof(_ppp_hdr), _ppp_hdr); + if (!ppp_hdr) goto out_bad; - proto = eth->h_proto; - nhoff += sizeof(*eth); - - /* Cap headers that we access via pointers at the - * end of the Ethernet header as our maximum alignment - * at that point is only 2 bytes. - */ - if (NET_IP_ALIGN) - hlen = nhoff; + + switch (PPP_PROTOCOL(ppp_hdr)) { + case PPP_IP: + proto = htons(ETH_P_IP); + break; + case PPP_IPV6: + proto = htons(ETH_P_IPV6); + break; + default: + /* Could probably catch some more like MPLS */ + break; + } + + offset += PPP_HDRLEN; } + nhoff += offset; key_control->flags |= FLOW_DIS_ENCAPSULATION; if (flags & FLOW_DISSECTOR_F_STOP_AT_ENCAP) goto out_good; diff --git a/net/core/neighbour.c b/net/core/neighbour.c index cf26e04c4046..2ae929f9bd06 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -1148,7 +1148,8 @@ int neigh_update(struct neighbour *neigh, const u8 *lladdr, u8 new, } else goto out; } else { - if (lladdr == neigh->ha && new == NUD_STALE) + if (lladdr == neigh->ha && new == NUD_STALE && + !(flags & NEIGH_UPDATE_F_ADMIN)) new = old; } } diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index 2c2eb1b629b1..1fe58167d39a 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -37,6 +37,8 @@ struct net init_net = { }; EXPORT_SYMBOL(init_net); +static bool init_net_initialized; + #define INITIAL_NET_GEN_PTRS 13 /* +1 for len +2 for rcu_head */ static unsigned int max_gen_ptrs = INITIAL_NET_GEN_PTRS; @@ -750,6 +752,8 @@ static int __init net_ns_init(void) if (setup_net(&init_net, &init_user_ns)) panic("Could not setup the initial network namespace"); + init_net_initialized = true; + rtnl_lock(); list_add_tail_rcu(&init_net.list, &net_namespace_list); rtnl_unlock(); @@ -811,15 +815,24 @@ static void __unregister_pernet_operations(struct pernet_operations *ops) static int __register_pernet_operations(struct list_head *list, struct pernet_operations *ops) { + if (!init_net_initialized) { + list_add_tail(&ops->list, list); + return 0; + } + return ops_init(ops, &init_net); } static void __unregister_pernet_operations(struct pernet_operations *ops) { - LIST_HEAD(net_exit_list); - list_add(&init_net.exit_list, &net_exit_list); - ops_exit_list(ops, &net_exit_list); - ops_free_list(ops, &net_exit_list); + if (!init_net_initialized) { + list_del(&ops->list); + } else { + LIST_HEAD(net_exit_list); + list_add(&init_net.exit_list, &net_exit_list); + ops_exit_list(ops, &net_exit_list); + ops_free_list(ops, &net_exit_list); + } } #endif /* CONFIG_NET_NS */ diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index ef2ebeb89d0f..317c31939732 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -93,9 +93,6 @@ struct fib_table *fib_new_table(struct net *net, u32 id) return NULL; switch (id) { - case RT_TABLE_LOCAL: - rcu_assign_pointer(net->ipv4.fib_local, tb); - break; case RT_TABLE_MAIN: rcu_assign_pointer(net->ipv4.fib_main, tb); break; @@ -137,9 +134,6 @@ static void fib_replace_table(struct net *net, struct fib_table *old, { #ifdef CONFIG_IP_MULTIPLE_TABLES switch (new->tb_id) { - case RT_TABLE_LOCAL: - rcu_assign_pointer(net->ipv4.fib_local, new); - break; case RT_TABLE_MAIN: rcu_assign_pointer(net->ipv4.fib_main, new); break; @@ -1249,7 +1243,6 @@ static void ip_fib_net_exit(struct net *net) rtnl_lock(); #ifdef CONFIG_IP_MULTIPLE_TABLES - RCU_INIT_POINTER(net->ipv4.fib_local, NULL); RCU_INIT_POINTER(net->ipv4.fib_main, NULL); RCU_INIT_POINTER(net->ipv4.fib_default, NULL); #endif diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index 9b4ca87f70ba..606cc3e85d2b 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -472,6 +472,15 @@ static struct sk_buff *add_grec(struct sk_buff *skb, struct ip_mc_list *pmc, continue; } + /* Based on RFC3376 5.1. Should not send source-list change + * records when there is a filter mode change. + */ + if (((gdeleted && pmc->sfmode == MCAST_EXCLUDE) || + (!gdeleted && pmc->crcount)) && + (type == IGMPV3_ALLOW_NEW_SOURCES || + type == IGMPV3_BLOCK_OLD_SOURCES) && psf->sf_crcount) + goto decrease_sf_crcount; + /* clear marks on query responses */ if (isquery) psf->sf_gsresp = 0; @@ -499,6 +508,7 @@ static struct sk_buff *add_grec(struct sk_buff *skb, struct ip_mc_list *pmc, scount++; stotal++; if ((type == IGMPV3_ALLOW_NEW_SOURCES || type == IGMPV3_BLOCK_OLD_SOURCES) && psf->sf_crcount) { +decrease_sf_crcount: psf->sf_crcount--; if ((sdeleted || gdeleted) && psf->sf_crcount == 0) { if (psf_prev) diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c index 1d71c40eaaf3..66c2fe602810 100644 --- a/net/ipv4/ipconfig.c +++ b/net/ipv4/ipconfig.c @@ -85,7 +85,6 @@ /* Define the timeout for waiting for a DHCP/BOOTP/RARP reply */ #define CONF_OPEN_RETRIES 2 /* (Re)open devices twice */ #define CONF_SEND_RETRIES 6 /* Send six requests per open */ -#define CONF_INTER_TIMEOUT (HZ) /* Inter-device timeout: 1 second */ #define CONF_BASE_TIMEOUT (HZ*2) /* Initial timeout: 2 seconds */ #define CONF_TIMEOUT_RANDOM (HZ) /* Maximum amount of randomization */ #define CONF_TIMEOUT_MULT *7/4 /* Rate of timeout growth */ @@ -188,7 +187,7 @@ struct ic_device { }; static struct ic_device *ic_first_dev __initdata; /* List of open device */ -static struct net_device *ic_dev __initdata; /* Selected device */ +static struct ic_device *ic_dev __initdata; /* Selected device */ static bool __init ic_is_init_dev(struct net_device *dev) { @@ -307,7 +306,7 @@ static void __init ic_close_devs(void) while ((d = next)) { next = d->next; dev = d->dev; - if (dev != ic_dev && !netdev_uses_dsa(dev)) { + if (dev != ic_dev->dev && !netdev_uses_dsa(dev)) { pr_debug("IP-Config: Downing %s\n", dev->name); dev_change_flags(dev, d->flags); } @@ -372,7 +371,7 @@ static int __init ic_setup_if(void) int err; memset(&ir, 0, sizeof(ir)); - strcpy(ir.ifr_ifrn.ifrn_name, ic_dev->name); + strcpy(ir.ifr_ifrn.ifrn_name, ic_dev->dev->name); set_sockaddr(sin, ic_myaddr, 0); if ((err = ic_devinet_ioctl(SIOCSIFADDR, &ir)) < 0) { pr_err("IP-Config: Unable to set interface address (%d)\n", @@ -396,7 +395,7 @@ static int __init ic_setup_if(void) * out, we'll try to muddle along. */ if (ic_dev_mtu != 0) { - strcpy(ir.ifr_name, ic_dev->name); + strcpy(ir.ifr_name, ic_dev->dev->name); ir.ifr_mtu = ic_dev_mtu; if ((err = ic_dev_ioctl(SIOCSIFMTU, &ir)) < 0) pr_err("IP-Config: Unable to set interface mtu to %d (%d)\n", @@ -568,7 +567,7 @@ ic_rarp_recv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt goto drop_unlock; /* We have a winner! */ - ic_dev = dev; + ic_dev = d; if (ic_myaddr == NONE) ic_myaddr = tip; ic_servaddr = sip; @@ -655,8 +654,6 @@ static struct packet_type bootp_packet_type __initdata = { .func = ic_bootp_recv, }; -static __be32 ic_dev_xid; /* Device under configuration */ - /* * Initialize DHCP/BOOTP extension fields in the request. */ @@ -666,14 +663,14 @@ static const u8 ic_bootp_cookie[4] = { 99, 130, 83, 99 }; #ifdef IPCONFIG_DHCP static void __init -ic_dhcp_init_options(u8 *options) +ic_dhcp_init_options(u8 *options, struct ic_device *d) { u8 mt = ((ic_servaddr == NONE) ? DHCPDISCOVER : DHCPREQUEST); u8 *e = options; int len; - pr_debug("DHCP: Sending message type %d\n", mt); + pr_debug("DHCP: Sending message type %d (%s)\n", mt, d->dev->name); memcpy(e, ic_bootp_cookie, 4); /* RFC1048 Magic Cookie */ e += 4; @@ -857,7 +854,7 @@ static void __init ic_bootp_send_if(struct ic_device *d, unsigned long jiffies_d /* add DHCP options or BOOTP extensions */ #ifdef IPCONFIG_DHCP if (ic_proto_enabled & IC_USE_DHCP) - ic_dhcp_init_options(b->exten); + ic_dhcp_init_options(b->exten, d); else #endif ic_bootp_init_ext(b->exten); @@ -1033,14 +1030,8 @@ static int __init ic_bootp_recv(struct sk_buff *skb, struct net_device *dev, str /* Is it a reply to our BOOTP request? */ if (b->op != BOOTP_REPLY || b->xid != d->xid) { - net_err_ratelimited("DHCP/BOOTP: Reply not for us, op[%x] xid[%x]\n", - b->op, b->xid); - goto drop_unlock; - } - - /* Is it a reply for the device we are configuring? */ - if (b->xid != ic_dev_xid) { - net_err_ratelimited("DHCP/BOOTP: Ignoring delayed packet\n"); + net_err_ratelimited("DHCP/BOOTP: Reply not for us on %s, op[%x] xid[%x]\n", + d->dev->name, b->op, b->xid); goto drop_unlock; } @@ -1075,7 +1066,7 @@ static int __init ic_bootp_recv(struct sk_buff *skb, struct net_device *dev, str } } - pr_debug("DHCP: Got message type %d\n", mt); + pr_debug("DHCP: Got message type %d (%s)\n", mt, d->dev->name); switch (mt) { case DHCPOFFER: @@ -1130,7 +1121,7 @@ static int __init ic_bootp_recv(struct sk_buff *skb, struct net_device *dev, str } /* We have a winner! */ - ic_dev = dev; + ic_dev = d; ic_myaddr = b->your_ip; ic_servaddr = b->server_ip; ic_addrservaddr = b->iph.saddr; @@ -1225,9 +1216,6 @@ static int __init ic_dynamic(void) timeout = CONF_BASE_TIMEOUT + (timeout % (unsigned int) CONF_TIMEOUT_RANDOM); for (;;) { #ifdef IPCONFIG_BOOTP - /* Track the device we are configuring */ - ic_dev_xid = d->xid; - if (do_bootp && (d->able & IC_BOOTP)) ic_bootp_send_if(d, jiffies - start_jiffies); #endif @@ -1236,15 +1224,19 @@ static int __init ic_dynamic(void) ic_rarp_send_if(d); #endif - jiff = jiffies + (d->next ? CONF_INTER_TIMEOUT : timeout); - while (time_before(jiffies, jiff) && !ic_got_reply) - schedule_timeout_uninterruptible(1); + if (!d->next) { + jiff = jiffies + timeout; + while (time_before(jiffies, jiff) && !ic_got_reply) + schedule_timeout_uninterruptible(1); + } #ifdef IPCONFIG_DHCP /* DHCP isn't done until we get a DHCPACK. */ if ((ic_got_reply & IC_BOOTP) && (ic_proto_enabled & IC_USE_DHCP) && ic_dhcp_msgtype != DHCPACK) { ic_got_reply = 0; + /* continue on device that got the reply */ + d = ic_dev; pr_cont(","); continue; } @@ -1487,7 +1479,7 @@ static int __init ip_auto_config(void) #endif /* IPCONFIG_DYNAMIC */ } else { /* Device selected manually or only one device -> use it */ - ic_dev = ic_first_dev->dev; + ic_dev = ic_first_dev; } addr = root_nfs_parse_addr(root_server_path); @@ -1501,14 +1493,6 @@ static int __init ip_auto_config(void) return -1; /* - * Close all network devices except the device we've - * autoconfigured and set up routes. - */ - ic_close_devs(); - if (ic_setup_if() < 0 || ic_setup_routes() < 0) - return -1; - - /* * Record which protocol was actually used. */ #ifdef IPCONFIG_DYNAMIC @@ -1522,7 +1506,7 @@ static int __init ip_auto_config(void) pr_info("IP-Config: Complete:\n"); pr_info(" device=%s, hwaddr=%*phC, ipaddr=%pI4, mask=%pI4, gw=%pI4\n", - ic_dev->name, ic_dev->addr_len, ic_dev->dev_addr, + ic_dev->dev->name, ic_dev->dev->addr_len, ic_dev->dev->dev_addr, &ic_myaddr, &ic_netmask, &ic_gateway); pr_info(" host=%s, domain=%s, nis-domain=%s\n", utsname()->nodename, ic_domain, utsname()->domainname); @@ -1542,6 +1526,15 @@ static int __init ip_auto_config(void) pr_cont("\n"); #endif /* !SILENT */ + /* + * Close all network devices except the device we've + * autoconfigured and set up routes. + */ + ic_close_devs(); + if (ic_setup_if() < 0 || ic_setup_routes() < 0) + return -1; + + return 0; } diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c index 776d145113e1..b375b5addb76 100644 --- a/net/ipv6/ip6_gre.c +++ b/net/ipv6/ip6_gre.c @@ -61,12 +61,12 @@ static bool log_ecn_error = true; module_param(log_ecn_error, bool, 0644); MODULE_PARM_DESC(log_ecn_error, "Log packets received with corrupted ECN"); -#define HASH_SIZE_SHIFT 5 -#define HASH_SIZE (1 << HASH_SIZE_SHIFT) +#define IP6_GRE_HASH_SIZE_SHIFT 5 +#define IP6_GRE_HASH_SIZE (1 << IP6_GRE_HASH_SIZE_SHIFT) static int ip6gre_net_id __read_mostly; struct ip6gre_net { - struct ip6_tnl __rcu *tunnels[4][HASH_SIZE]; + struct ip6_tnl __rcu *tunnels[4][IP6_GRE_HASH_SIZE]; struct net_device *fb_tunnel_dev; }; @@ -96,12 +96,12 @@ static void ip6gre_tnl_link_config(struct ip6_tnl *t, int set_mtu); will match fallback tunnel. */ -#define HASH_KEY(key) (((__force u32)key^((__force u32)key>>4))&(HASH_SIZE - 1)) +#define HASH_KEY(key) (((__force u32)key^((__force u32)key>>4))&(IP6_GRE_HASH_SIZE - 1)) static u32 HASH_ADDR(const struct in6_addr *addr) { u32 hash = ipv6_addr_hash(addr); - return hash_32(hash, HASH_SIZE_SHIFT); + return hash_32(hash, IP6_GRE_HASH_SIZE_SHIFT); } #define tunnels_r_l tunnels[3] @@ -1089,7 +1089,7 @@ static void ip6gre_destroy_tunnels(struct net *net, struct list_head *head) for (prio = 0; prio < 4; prio++) { int h; - for (h = 0; h < HASH_SIZE; h++) { + for (h = 0; h < IP6_GRE_HASH_SIZE; h++) { struct ip6_tnl *t; t = rtnl_dereference(ign->tunnels[prio][h]); diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 7b0481e3738f..2050217df565 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -64,8 +64,8 @@ MODULE_LICENSE("GPL"); MODULE_ALIAS_RTNL_LINK("ip6tnl"); MODULE_ALIAS_NETDEV("ip6tnl0"); -#define HASH_SIZE_SHIFT 5 -#define HASH_SIZE (1 << HASH_SIZE_SHIFT) +#define IP6_TUNNEL_HASH_SIZE_SHIFT 5 +#define IP6_TUNNEL_HASH_SIZE (1 << IP6_TUNNEL_HASH_SIZE_SHIFT) static bool log_ecn_error = true; module_param(log_ecn_error, bool, 0644); @@ -75,7 +75,7 @@ static u32 HASH(const struct in6_addr *addr1, const struct in6_addr *addr2) { u32 hash = ipv6_addr_hash(addr1) ^ ipv6_addr_hash(addr2); - return hash_32(hash, HASH_SIZE_SHIFT); + return hash_32(hash, IP6_TUNNEL_HASH_SIZE_SHIFT); } static int ip6_tnl_dev_init(struct net_device *dev); @@ -87,7 +87,7 @@ struct ip6_tnl_net { /* the IPv6 tunnel fallback device */ struct net_device *fb_tnl_dev; /* lists for storing tunnels in use */ - struct ip6_tnl __rcu *tnls_r_l[HASH_SIZE]; + struct ip6_tnl __rcu *tnls_r_l[IP6_TUNNEL_HASH_SIZE]; struct ip6_tnl __rcu *tnls_wc[1]; struct ip6_tnl __rcu **tnls[2]; }; @@ -2031,7 +2031,7 @@ static void __net_exit ip6_tnl_destroy_tunnels(struct net *net) if (dev->rtnl_link_ops == &ip6_link_ops) unregister_netdevice_queue(dev, &list); - for (h = 0; h < HASH_SIZE; h++) { + for (h = 0; h < IP6_TUNNEL_HASH_SIZE; h++) { t = rtnl_dereference(ip6n->tnls_r_l[h]); while (t) { /* If dev is in the same netns, it has already diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c index d90a11f14040..cc7e05898307 100644 --- a/net/ipv6/ip6_vti.c +++ b/net/ipv6/ip6_vti.c @@ -50,14 +50,14 @@ #include <net/net_namespace.h> #include <net/netns/generic.h> -#define HASH_SIZE_SHIFT 5 -#define HASH_SIZE (1 << HASH_SIZE_SHIFT) +#define IP6_VTI_HASH_SIZE_SHIFT 5 +#define IP6_VTI_HASH_SIZE (1 << IP6_VTI_HASH_SIZE_SHIFT) static u32 HASH(const struct in6_addr *addr1, const struct in6_addr *addr2) { u32 hash = ipv6_addr_hash(addr1) ^ ipv6_addr_hash(addr2); - return hash_32(hash, HASH_SIZE_SHIFT); + return hash_32(hash, IP6_VTI_HASH_SIZE_SHIFT); } static int vti6_dev_init(struct net_device *dev); @@ -69,7 +69,7 @@ struct vti6_net { /* the vti6 tunnel fallback device */ struct net_device *fb_tnl_dev; /* lists for storing tunnels in use */ - struct ip6_tnl __rcu *tnls_r_l[HASH_SIZE]; + struct ip6_tnl __rcu *tnls_r_l[IP6_VTI_HASH_SIZE]; struct ip6_tnl __rcu *tnls_wc[1]; struct ip6_tnl __rcu **tnls[2]; }; @@ -1040,7 +1040,7 @@ static void __net_exit vti6_destroy_tunnels(struct vti6_net *ip6n) struct ip6_tnl *t; LIST_HEAD(list); - for (h = 0; h < HASH_SIZE; h++) { + for (h = 0; h < IP6_VTI_HASH_SIZE; h++) { t = rtnl_dereference(ip6n->tnls_r_l[h]); while (t) { unregister_netdevice_queue(t->dev, &list); diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index d64ee7e83664..75c1fc54f188 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -1739,6 +1739,15 @@ static struct sk_buff *add_grec(struct sk_buff *skb, struct ifmcaddr6 *pmc, continue; } + /* Based on RFC3810 6.1. Should not send source-list change + * records when there is a filter mode change. + */ + if (((gdeleted && pmc->mca_sfmode == MCAST_EXCLUDE) || + (!gdeleted && pmc->mca_crcount)) && + (type == MLD2_ALLOW_NEW_SOURCES || + type == MLD2_BLOCK_OLD_SOURCES) && psf->sf_crcount) + goto decrease_sf_crcount; + /* clear marks on query responses */ if (isquery) psf->sf_gsresp = 0; @@ -1766,6 +1775,7 @@ static struct sk_buff *add_grec(struct sk_buff *skb, struct ifmcaddr6 *pmc, scount++; stotal++; if ((type == MLD2_ALLOW_NEW_SOURCES || type == MLD2_BLOCK_OLD_SOURCES) && psf->sf_crcount) { +decrease_sf_crcount: psf->sf_crcount--; if ((sdeleted || gdeleted) && psf->sf_crcount == 0) { if (psf_prev) diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 182b6a9be29d..b1cdf8009d29 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -62,7 +62,7 @@ For comments look at net/ipv4/ip_gre.c --ANK */ -#define HASH_SIZE 16 +#define IP6_SIT_HASH_SIZE 16 #define HASH(addr) (((__force u32)addr^((__force u32)addr>>4))&0xF) static bool log_ecn_error = true; @@ -78,9 +78,9 @@ static struct rtnl_link_ops sit_link_ops __read_mostly; static int sit_net_id __read_mostly; struct sit_net { - struct ip_tunnel __rcu *tunnels_r_l[HASH_SIZE]; - struct ip_tunnel __rcu *tunnels_r[HASH_SIZE]; - struct ip_tunnel __rcu *tunnels_l[HASH_SIZE]; + struct ip_tunnel __rcu *tunnels_r_l[IP6_SIT_HASH_SIZE]; + struct ip_tunnel __rcu *tunnels_r[IP6_SIT_HASH_SIZE]; + struct ip_tunnel __rcu *tunnels_l[IP6_SIT_HASH_SIZE]; struct ip_tunnel __rcu *tunnels_wc[1]; struct ip_tunnel __rcu **tunnels[4]; @@ -1126,7 +1126,7 @@ static int ipip6_tunnel_update_6rd(struct ip_tunnel *t, } #endif -bool ipip6_valid_ip_proto(u8 ipproto) +static bool ipip6_valid_ip_proto(u8 ipproto) { return ipproto == IPPROTO_IPV6 || ipproto == IPPROTO_IPIP || @@ -1783,7 +1783,7 @@ static void __net_exit sit_destroy_tunnels(struct net *net, for (prio = 1; prio < 4; prio++) { int h; - for (h = 0; h < HASH_SIZE; h++) { + for (h = 0; h < IP6_SIT_HASH_SIZE; h++) { struct ip_tunnel *t; t = rtnl_dereference(sitn->tunnels[prio][h]); diff --git a/net/mac80211/driver-ops.h b/net/mac80211/driver-ops.h index ba5fc1f01e53..42a41ae405ba 100644 --- a/net/mac80211/driver-ops.h +++ b/net/mac80211/driver-ops.h @@ -1088,13 +1088,13 @@ static inline void drv_leave_ibss(struct ieee80211_local *local, } static inline u32 drv_get_expected_throughput(struct ieee80211_local *local, - struct ieee80211_sta *sta) + struct sta_info *sta) { u32 ret = 0; - trace_drv_get_expected_throughput(sta); - if (local->ops->get_expected_throughput) - ret = local->ops->get_expected_throughput(&local->hw, sta); + trace_drv_get_expected_throughput(&sta->sta); + if (local->ops->get_expected_throughput && sta->uploaded) + ret = local->ops->get_expected_throughput(&local->hw, &sta->sta); trace_drv_return_u32(local, ret); return ret; diff --git a/net/mac80211/mesh_hwmp.c b/net/mac80211/mesh_hwmp.c index 8f9c3bde835f..fa7d37cf0351 100644 --- a/net/mac80211/mesh_hwmp.c +++ b/net/mac80211/mesh_hwmp.c @@ -326,22 +326,33 @@ static u32 airtime_link_metric_get(struct ieee80211_local *local, u32 tx_time, estimated_retx; u64 result; - if (sta->mesh->fail_avg >= 100) - return MAX_METRIC; + /* Try to get rate based on HW/SW RC algorithm. + * Rate is returned in units of Kbps, correct this + * to comply with airtime calculation units + * Round up in case we get rate < 100Kbps + */ + rate = DIV_ROUND_UP(sta_get_expected_throughput(sta), 100); - sta_set_rate_info_tx(sta, &sta->tx_stats.last_rate, &rinfo); - rate = cfg80211_calculate_bitrate(&rinfo); - if (WARN_ON(!rate)) - return MAX_METRIC; + if (rate) { + err = 0; + } else { + if (sta->mesh->fail_avg >= 100) + return MAX_METRIC; - err = (sta->mesh->fail_avg << ARITH_SHIFT) / 100; + sta_set_rate_info_tx(sta, &sta->tx_stats.last_rate, &rinfo); + rate = cfg80211_calculate_bitrate(&rinfo); + if (WARN_ON(!rate)) + return MAX_METRIC; + + err = (sta->mesh->fail_avg << ARITH_SHIFT) / 100; + } /* bitrate is in units of 100 Kbps, while we need rate in units of * 1Mbps. This will be corrected on tx_time computation. */ tx_time = (device_constant + 10 * test_frame_len / rate); estimated_retx = ((1 << (2 * ARITH_SHIFT)) / (s_unit - err)); - result = (tx_time * estimated_retx) >> (2 * ARITH_SHIFT) ; + result = (tx_time * estimated_retx) >> (2 * ARITH_SHIFT); return (u32)result; } diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index 76b737dcc36f..19f14c907d74 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -2279,11 +2279,7 @@ void sta_set_sinfo(struct sta_info *sta, struct station_info *sinfo) if (test_sta_flag(sta, WLAN_STA_TDLS_PEER)) sinfo->sta_flags.set |= BIT(NL80211_STA_FLAG_TDLS_PEER); - /* check if the driver has a SW RC implementation */ - if (ref && ref->ops->get_expected_throughput) - thr = ref->ops->get_expected_throughput(sta->rate_ctrl_priv); - else - thr = drv_get_expected_throughput(local, &sta->sta); + thr = sta_get_expected_throughput(sta); if (thr != 0) { sinfo->filled |= BIT(NL80211_STA_INFO_EXPECTED_THROUGHPUT); @@ -2291,6 +2287,25 @@ void sta_set_sinfo(struct sta_info *sta, struct station_info *sinfo) } } +u32 sta_get_expected_throughput(struct sta_info *sta) +{ + struct ieee80211_sub_if_data *sdata = sta->sdata; + struct ieee80211_local *local = sdata->local; + struct rate_control_ref *ref = NULL; + u32 thr = 0; + + if (test_sta_flag(sta, WLAN_STA_RATE_CONTROL)) + ref = local->rate_ctrl; + + /* check if the driver has a SW RC implementation */ + if (ref && ref->ops->get_expected_throughput) + thr = ref->ops->get_expected_throughput(sta->rate_ctrl_priv); + else + thr = drv_get_expected_throughput(local, sta); + + return thr; +} + unsigned long ieee80211_sta_last_active(struct sta_info *sta) { struct ieee80211_sta_rx_stats *stats = sta_get_last_rx_stats(sta); diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h index 78b0ef32dddd..0556be3e3628 100644 --- a/net/mac80211/sta_info.h +++ b/net/mac80211/sta_info.h @@ -712,6 +712,8 @@ void sta_set_rate_info_tx(struct sta_info *sta, struct rate_info *rinfo); void sta_set_sinfo(struct sta_info *sta, struct station_info *sinfo); +u32 sta_get_expected_throughput(struct sta_info *sta); + void ieee80211_sta_expire(struct ieee80211_sub_if_data *sdata, unsigned long exp_time); u8 sta_info_tx_streams(struct sta_info *sta); diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 502396694f47..1d0746dfea57 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -2334,7 +2334,6 @@ static struct sk_buff *ieee80211_build_hdr(struct ieee80211_sub_if_data *sdata, struct mesh_path __maybe_unused *mppath = NULL, *mpath = NULL; const u8 *encaps_data; int encaps_len, skip_header_bytes; - int nh_pos, h_pos; bool wme_sta = false, authorized = false; bool tdls_peer; bool multicast; @@ -2640,13 +2639,7 @@ static struct sk_buff *ieee80211_build_hdr(struct ieee80211_sub_if_data *sdata, encaps_len = 0; } - nh_pos = skb_network_header(skb) - skb->data; - h_pos = skb_transport_header(skb) - skb->data; - skb_pull(skb, skip_header_bytes); - nh_pos -= skip_header_bytes; - h_pos -= skip_header_bytes; - head_need = hdrlen + encaps_len + meshhdrlen - skb_headroom(skb); /* @@ -2672,18 +2665,12 @@ static struct sk_buff *ieee80211_build_hdr(struct ieee80211_sub_if_data *sdata, } } - if (encaps_data) { + if (encaps_data) memcpy(skb_push(skb, encaps_len), encaps_data, encaps_len); - nh_pos += encaps_len; - h_pos += encaps_len; - } #ifdef CONFIG_MAC80211_MESH - if (meshhdrlen > 0) { + if (meshhdrlen > 0) memcpy(skb_push(skb, meshhdrlen), &mesh_hdr, meshhdrlen); - nh_pos += meshhdrlen; - h_pos += meshhdrlen; - } #endif if (ieee80211_is_data_qos(fc)) { @@ -2699,15 +2686,7 @@ static struct sk_buff *ieee80211_build_hdr(struct ieee80211_sub_if_data *sdata, } else memcpy(skb_push(skb, hdrlen), &hdr, hdrlen); - nh_pos += hdrlen; - h_pos += hdrlen; - - /* Update skb pointers to various headers since this modified frame - * is going to go through Linux networking code that may potentially - * need things like pointer to IP header. */ skb_reset_mac_header(skb); - skb_set_network_header(skb, nh_pos); - skb_set_transport_header(skb, h_pos); info = IEEE80211_SKB_CB(skb); memset(info, 0, sizeof(*info)); @@ -4390,9 +4369,6 @@ void __ieee80211_tx_skb_tid_band(struct ieee80211_sub_if_data *sdata, int ac = ieee802_1d_to_ac[tid & 7]; skb_reset_mac_header(skb); - skb_reset_network_header(skb); - skb_reset_transport_header(skb); - skb_set_queue_mapping(skb, ac); skb->priority = tid; diff --git a/net/rds/ib.h b/net/rds/ib.h index 046f7508c06b..45ac8e8e58f4 100644 --- a/net/rds/ib.h +++ b/net/rds/ib.h @@ -333,6 +333,7 @@ void rds_ib_conn_path_shutdown(struct rds_conn_path *cp); void rds_ib_state_change(struct sock *sk); int rds_ib_listen_init(void); void rds_ib_listen_stop(void); +__printf(2, 3) void __rds_ib_conn_error(struct rds_connection *conn, const char *, ...); int rds_ib_cm_handle_connect(struct rdma_cm_id *cm_id, struct rdma_cm_event *event); diff --git a/net/rds/rds.h b/net/rds/rds.h index b2d17f0fafa8..fd0bccb2f9f9 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -688,6 +688,7 @@ void __rds_conn_error(struct rds_connection *conn, const char *, ...); #define rds_conn_error(conn, fmt...) \ __rds_conn_error(conn, KERN_WARNING "RDS: " fmt) +__printf(2, 3) void __rds_conn_path_error(struct rds_conn_path *cp, const char *, ...); #define rds_conn_path_error(cp, fmt...) \ __rds_conn_path_error(cp, KERN_WARNING "RDS: " fmt) diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 12ebde845523..25aada7b095c 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -29,6 +29,7 @@ #include <linux/hrtimer.h> #include <linux/lockdep.h> #include <linux/slab.h> +#include <linux/hashtable.h> #include <net/net_namespace.h> #include <net/sock.h> @@ -263,33 +264,33 @@ static struct Qdisc *qdisc_match_from_root(struct Qdisc *root, u32 handle) root->handle == handle) return root; - list_for_each_entry_rcu(q, &root->list, list) { + hash_for_each_possible_rcu(qdisc_dev(root)->qdisc_hash, q, hash, handle) { if (q->handle == handle) return q; } return NULL; } -void qdisc_list_add(struct Qdisc *q) +void qdisc_hash_add(struct Qdisc *q) { if ((q->parent != TC_H_ROOT) && !(q->flags & TCQ_F_INGRESS)) { struct Qdisc *root = qdisc_dev(q)->qdisc; WARN_ON_ONCE(root == &noop_qdisc); ASSERT_RTNL(); - list_add_tail_rcu(&q->list, &root->list); + hash_add_rcu(qdisc_dev(q)->qdisc_hash, &q->hash, q->handle); } } -EXPORT_SYMBOL(qdisc_list_add); +EXPORT_SYMBOL(qdisc_hash_add); -void qdisc_list_del(struct Qdisc *q) +void qdisc_hash_del(struct Qdisc *q) { if ((q->parent != TC_H_ROOT) && !(q->flags & TCQ_F_INGRESS)) { ASSERT_RTNL(); - list_del_rcu(&q->list); + hash_del_rcu(&q->hash); } } -EXPORT_SYMBOL(qdisc_list_del); +EXPORT_SYMBOL(qdisc_hash_del); struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle) { @@ -998,7 +999,7 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue, goto err_out4; } - qdisc_list_add(sch); + qdisc_hash_add(sch); return sch; } @@ -1435,6 +1436,7 @@ static int tc_dump_qdisc_root(struct Qdisc *root, struct sk_buff *skb, { int ret = 0, q_idx = *q_idx_p; struct Qdisc *q; + int b; if (!root) return 0; @@ -1449,7 +1451,7 @@ static int tc_dump_qdisc_root(struct Qdisc *root, struct sk_buff *skb, goto done; q_idx++; } - list_for_each_entry(q, &root->list, list) { + hash_for_each(qdisc_dev(root)->qdisc_hash, b, q, hash) { if (q_idx < s_q_idx) { q_idx++; continue; @@ -1765,6 +1767,7 @@ static int tc_dump_tclass_root(struct Qdisc *root, struct sk_buff *skb, int *t_p, int s_t) { struct Qdisc *q; + int b; if (!root) return 0; @@ -1772,7 +1775,7 @@ static int tc_dump_tclass_root(struct Qdisc *root, struct sk_buff *skb, if (tc_dump_tclass_qdisc(root, skb, tcm, cb, t_p, s_t) < 0) return -1; - list_for_each_entry(q, &root->list, list) { + hash_for_each(qdisc_dev(root)->qdisc_hash, b, q, hash) { if (tc_dump_tclass_qdisc(q, skb, tcm, cb, t_p, s_t) < 0) return -1; } diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index e95b67cd5718..18faecc3f13e 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -423,7 +423,6 @@ struct Qdisc noop_qdisc = { .dequeue = noop_dequeue, .flags = TCQ_F_BUILTIN, .ops = &noop_qdisc_ops, - .list = LIST_HEAD_INIT(noop_qdisc.list), .q.lock = __SPIN_LOCK_UNLOCKED(noop_qdisc.q.lock), .dev_queue = &noop_netdev_queue, .running = SEQCNT_ZERO(noop_qdisc.running), @@ -613,7 +612,6 @@ struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue, sch = (struct Qdisc *) QDISC_ALIGN((unsigned long) p); sch->padded = (char *) sch - (char *) p; } - INIT_LIST_HEAD(&sch->list); skb_queue_head_init(&sch->q); spin_lock_init(&sch->busylock); @@ -700,7 +698,7 @@ void qdisc_destroy(struct Qdisc *qdisc) return; #ifdef CONFIG_NET_SCHED - qdisc_list_del(qdisc); + qdisc_hash_del(qdisc); qdisc_put_stab(rtnl_dereference(qdisc->stab)); #endif @@ -788,6 +786,10 @@ static void attach_default_qdiscs(struct net_device *dev) qdisc->ops->attach(qdisc); } } +#ifdef CONFIG_NET_SCHED + if (dev->qdisc) + qdisc_hash_add(dev->qdisc); +#endif } static void transition_one_qdisc(struct net_device *dev, diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index 3ddc7bd74ecb..000f1d36128e 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -142,8 +142,6 @@ struct hfsc_class { link-sharing, max(myf, cfmin) */ u64 cl_myf; /* my fit-time (calculated from this class's own upperlimit curve) */ - u64 cl_myfadj; /* my fit-time adjustment (to cancel - history dependence) */ u64 cl_cfmin; /* earliest children's fit-time (used with cl_myf to obtain cl_f) */ u64 cl_cvtmin; /* minimal virtual time among the @@ -151,11 +149,8 @@ struct hfsc_class { (monotonic within a period) */ u64 cl_vtadj; /* intra-period cumulative vt adjustment */ - u64 cl_vtoff; /* inter-period cumulative vt offset */ - u64 cl_cvtmax; /* max child's vt in the last period */ - u64 cl_cvtoff; /* cumulative cvtmax of all periods */ - u64 cl_pcvtoff; /* parent's cvtoff at initialization - time */ + u64 cl_cvtoff; /* largest virtual time seen among + the children */ struct internal_sc cl_rsc; /* internal real-time service curve */ struct internal_sc cl_fsc; /* internal fair service curve */ @@ -701,28 +696,16 @@ init_vf(struct hfsc_class *cl, unsigned int len) } else { /* * first child for a new parent backlog period. - * add parent's cvtmax to cvtoff to make a new - * vt (vtoff + vt) larger than the vt in the - * last period for all children. + * initialize cl_vt to the highest value seen + * among the siblings. this is analogous to + * what cur_time would provide in realtime case. */ - vt = cl->cl_parent->cl_cvtmax; - cl->cl_parent->cl_cvtoff += vt; - cl->cl_parent->cl_cvtmax = 0; + cl->cl_vt = cl->cl_parent->cl_cvtoff; cl->cl_parent->cl_cvtmin = 0; - cl->cl_vt = 0; } - cl->cl_vtoff = cl->cl_parent->cl_cvtoff - - cl->cl_pcvtoff; - /* update the virtual curve */ - vt = cl->cl_vt + cl->cl_vtoff; - rtsc_min(&cl->cl_virtual, &cl->cl_fsc, vt, - cl->cl_total); - if (cl->cl_virtual.x == vt) { - cl->cl_virtual.x -= cl->cl_vtoff; - cl->cl_vtoff = 0; - } + rtsc_min(&cl->cl_virtual, &cl->cl_fsc, cl->cl_vt, cl->cl_total); cl->cl_vtadj = 0; cl->cl_vtperiod++; /* increment vt period */ @@ -745,7 +728,6 @@ init_vf(struct hfsc_class *cl, unsigned int len) /* compute myf */ cl->cl_myf = rtsc_y2x(&cl->cl_ulimit, cl->cl_total); - cl->cl_myfadj = 0; } } @@ -779,8 +761,7 @@ update_vf(struct hfsc_class *cl, unsigned int len, u64 cur_time) go_passive = 0; /* update vt */ - cl->cl_vt = rtsc_y2x(&cl->cl_virtual, cl->cl_total) - - cl->cl_vtoff + cl->cl_vtadj; + cl->cl_vt = rtsc_y2x(&cl->cl_virtual, cl->cl_total) + cl->cl_vtadj; /* * if vt of the class is smaller than cvtmin, @@ -795,9 +776,9 @@ update_vf(struct hfsc_class *cl, unsigned int len, u64 cur_time) if (go_passive) { /* no more active child, going passive */ - /* update cvtmax of the parent class */ - if (cl->cl_vt > cl->cl_parent->cl_cvtmax) - cl->cl_parent->cl_cvtmax = cl->cl_vt; + /* update cvtoff of the parent class */ + if (cl->cl_vt > cl->cl_parent->cl_cvtoff) + cl->cl_parent->cl_cvtoff = cl->cl_vt; /* remove this class from the vt tree */ vttree_remove(cl); @@ -813,9 +794,10 @@ update_vf(struct hfsc_class *cl, unsigned int len, u64 cur_time) /* update f */ if (cl->cl_flags & HFSC_USC) { + cl->cl_myf = rtsc_y2x(&cl->cl_ulimit, cl->cl_total); +#if 0 cl->cl_myf = cl->cl_myfadj + rtsc_y2x(&cl->cl_ulimit, cl->cl_total); -#if 0 /* * This code causes classes to stay way under their * limit when multiple classes are used at gigabit @@ -940,7 +922,7 @@ static void hfsc_change_fsc(struct hfsc_class *cl, struct tc_service_curve *fsc) { sc2isc(fsc, &cl->cl_fsc); - rtsc_init(&cl->cl_virtual, &cl->cl_fsc, cl->cl_vtoff + cl->cl_vt, cl->cl_total); + rtsc_init(&cl->cl_virtual, &cl->cl_fsc, cl->cl_vt, cl->cl_total); cl->cl_flags |= HFSC_FSC; } @@ -1094,7 +1076,6 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid, if (parent->level == 0) hfsc_purge_queue(sch, parent); hfsc_adjust_levels(parent); - cl->cl_pcvtoff = parent->cl_cvtoff; sch_tree_unlock(sch); qdisc_class_hash_grow(sch, &q->clhash); @@ -1482,16 +1463,12 @@ hfsc_reset_class(struct hfsc_class *cl) cl->cl_e = 0; cl->cl_vt = 0; cl->cl_vtadj = 0; - cl->cl_vtoff = 0; cl->cl_cvtmin = 0; - cl->cl_cvtmax = 0; cl->cl_cvtoff = 0; - cl->cl_pcvtoff = 0; cl->cl_vtperiod = 0; cl->cl_parentperiod = 0; cl->cl_f = 0; cl->cl_myf = 0; - cl->cl_myfadj = 0; cl->cl_cfmin = 0; cl->cl_nactive = 0; diff --git a/net/sched/sch_mq.c b/net/sched/sch_mq.c index b9439827c172..2bc8d7f8df16 100644 --- a/net/sched/sch_mq.c +++ b/net/sched/sch_mq.c @@ -88,7 +88,7 @@ static void mq_attach(struct Qdisc *sch) qdisc_destroy(old); #ifdef CONFIG_NET_SCHED if (ntx < dev->real_num_tx_queues) - qdisc_list_add(qdisc); + qdisc_hash_add(qdisc); #endif } diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c index 549c66359924..b5c502c78143 100644 --- a/net/sched/sch_mqprio.c +++ b/net/sched/sch_mqprio.c @@ -182,7 +182,7 @@ static void mqprio_attach(struct Qdisc *sch) if (old) qdisc_destroy(old); if (ntx < dev->real_num_tx_queues) - qdisc_list_add(qdisc); + qdisc_hash_add(qdisc); } kfree(priv->qdiscs); priv->qdiscs = NULL; diff --git a/net/switchdev/switchdev.c b/net/switchdev/switchdev.c index a5fc9dd24aa9..9e9012956993 100644 --- a/net/switchdev/switchdev.c +++ b/net/switchdev/switchdev.c @@ -1306,6 +1306,7 @@ bool switchdev_port_same_parent_id(struct net_device *a, return netdev_phys_item_id_same(&a_attr.u.ppid, &b_attr.u.ppid); } +EXPORT_SYMBOL_GPL(switchdev_port_same_parent_id); static u32 switchdev_port_fwd_mark_get(struct net_device *dev, struct net_device *group_dev) @@ -1323,7 +1324,6 @@ static u32 switchdev_port_fwd_mark_get(struct net_device *dev, return dev->ifindex; } -EXPORT_SYMBOL_GPL(switchdev_port_same_parent_id); static void switchdev_port_fwd_mark_reset(struct net_device *group_dev, u32 old_mark, u32 *reset_mark) diff --git a/net/sysctl_net.c b/net/sysctl_net.c index 46a71c701e7c..5bc1a3d57401 100644 --- a/net/sysctl_net.c +++ b/net/sysctl_net.c @@ -42,26 +42,37 @@ static int net_ctl_permissions(struct ctl_table_header *head, struct ctl_table *table) { struct net *net = container_of(head->set, struct net, sysctls); - kuid_t root_uid = make_kuid(net->user_ns, 0); - kgid_t root_gid = make_kgid(net->user_ns, 0); /* Allow network administrator to have same access as root. */ - if (ns_capable_noaudit(net->user_ns, CAP_NET_ADMIN) || - uid_eq(root_uid, current_euid())) { + if (ns_capable(net->user_ns, CAP_NET_ADMIN)) { int mode = (table->mode >> 6) & 7; return (mode << 6) | (mode << 3) | mode; } - /* Allow netns root group to have the same access as the root group */ - if (in_egroup_p(root_gid)) { - int mode = (table->mode >> 3) & 7; - return (mode << 3) | mode; - } + return table->mode; } +static void net_ctl_set_ownership(struct ctl_table_header *head, + struct ctl_table *table, + kuid_t *uid, kgid_t *gid) +{ + struct net *net = container_of(head->set, struct net, sysctls); + kuid_t ns_root_uid; + kgid_t ns_root_gid; + + ns_root_uid = make_kuid(net->user_ns, 0); + if (uid_valid(ns_root_uid)) + *uid = ns_root_uid; + + ns_root_gid = make_kgid(net->user_ns, 0); + if (gid_valid(ns_root_gid)) + *gid = ns_root_gid; +} + static struct ctl_table_root net_sysctl_root = { .lookup = net_ctl_header_lookup, .permissions = net_ctl_permissions, + .set_ownership = net_ctl_set_ownership, }; static int __net_init sysctl_net_init(struct net *net) diff --git a/net/wireless/core.c b/net/wireless/core.c index 7645e97362c0..2029b49a1df3 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -906,6 +906,8 @@ void cfg80211_unregister_wdev(struct wireless_dev *wdev) if (WARN_ON(wdev->netdev)) return; + nl80211_notify_iface(rdev, wdev, NL80211_CMD_DEL_INTERFACE); + list_del_rcu(&wdev->list); rdev->devlist_generation++; @@ -1079,6 +1081,8 @@ static int cfg80211_netdev_notifier_call(struct notifier_block *nb, wdev->iftype == NL80211_IFTYPE_P2P_CLIENT || wdev->iftype == NL80211_IFTYPE_ADHOC) && !wdev->use_4addr) dev->priv_flags |= IFF_DONT_BRIDGE; + + nl80211_notify_iface(rdev, wdev, NL80211_CMD_NEW_INTERFACE); break; case NETDEV_GOING_DOWN: cfg80211_leave(rdev, wdev); @@ -1157,6 +1161,8 @@ static int cfg80211_netdev_notifier_call(struct notifier_block *nb, * remove and clean it up. */ if (!list_empty(&wdev->list)) { + nl80211_notify_iface(rdev, wdev, + NL80211_CMD_DEL_INTERFACE); sysfs_remove_link(&dev->dev.kobj, "phy80211"); list_del_rcu(&wdev->list); rdev->devlist_generation++; diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index f02653a08993..499785778983 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -2751,7 +2751,7 @@ static int nl80211_new_interface(struct sk_buff *skb, struct genl_info *info) struct cfg80211_registered_device *rdev = info->user_ptr[0]; struct vif_params params; struct wireless_dev *wdev; - struct sk_buff *msg, *event; + struct sk_buff *msg; int err; enum nl80211_iftype type = NL80211_IFTYPE_UNSPECIFIED; u32 flags; @@ -2855,20 +2855,15 @@ static int nl80211_new_interface(struct sk_buff *skb, struct genl_info *info) return -ENOBUFS; } - event = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); - if (event) { - if (nl80211_send_iface(event, 0, 0, 0, - rdev, wdev, false) < 0) { - nlmsg_free(event); - goto out; - } - - genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), - event, 0, NL80211_MCGRP_CONFIG, - GFP_KERNEL); - } + /* + * For wdevs which have no associated netdev object (e.g. of type + * NL80211_IFTYPE_P2P_DEVICE), emit the NEW_INTERFACE event here. + * For all other types, the event will be generated from the + * netdev notifier + */ + if (!wdev->netdev) + nl80211_notify_iface(rdev, wdev, NL80211_CMD_NEW_INTERFACE); -out: return genlmsg_reply(msg, info); } @@ -2876,18 +2871,10 @@ static int nl80211_del_interface(struct sk_buff *skb, struct genl_info *info) { struct cfg80211_registered_device *rdev = info->user_ptr[0]; struct wireless_dev *wdev = info->user_ptr[1]; - struct sk_buff *msg; - int status; if (!rdev->ops->del_virtual_intf) return -EOPNOTSUPP; - msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); - if (msg && nl80211_send_iface(msg, 0, 0, 0, rdev, wdev, true) < 0) { - nlmsg_free(msg); - msg = NULL; - } - /* * If we remove a wireless device without a netdev then clear * user_ptr[1] so that nl80211_post_doit won't dereference it @@ -2898,15 +2885,7 @@ static int nl80211_del_interface(struct sk_buff *skb, struct genl_info *info) if (!wdev->netdev) info->user_ptr[1] = NULL; - status = rdev_del_virtual_intf(rdev, wdev); - if (status >= 0 && msg) - genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), - msg, 0, NL80211_MCGRP_CONFIG, - GFP_KERNEL); - else - nlmsg_free(msg); - - return status; + return rdev_del_virtual_intf(rdev, wdev); } static int nl80211_set_noack_map(struct sk_buff *skb, struct genl_info *info) @@ -5374,6 +5353,18 @@ static int nl80211_check_s32(const struct nlattr *nla, s32 min, s32 max, s32 *ou return 0; } +static int nl80211_check_power_mode(const struct nlattr *nla, + enum nl80211_mesh_power_mode min, + enum nl80211_mesh_power_mode max, + enum nl80211_mesh_power_mode *out) +{ + u32 val = nla_get_u32(nla); + if (val < min || val > max) + return -EINVAL; + *out = val; + return 0; +} + static int nl80211_parse_mesh_config(struct genl_info *info, struct mesh_config *cfg, u32 *mask_out) @@ -5518,7 +5509,7 @@ do { \ NL80211_MESH_POWER_ACTIVE, NL80211_MESH_POWER_MAX, mask, NL80211_MESHCONF_POWER_MODE, - nl80211_check_u32); + nl80211_check_power_mode); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshAwakeWindowDuration, 0, 65535, mask, NL80211_MESHCONF_AWAKE_WINDOW, nl80211_check_u16); @@ -7773,12 +7764,13 @@ static int nl80211_join_ibss(struct sk_buff *skb, struct genl_info *info) ibss.beacon_interval = 100; - if (info->attrs[NL80211_ATTR_BEACON_INTERVAL]) { + if (info->attrs[NL80211_ATTR_BEACON_INTERVAL]) ibss.beacon_interval = nla_get_u32(info->attrs[NL80211_ATTR_BEACON_INTERVAL]); - if (ibss.beacon_interval < 1 || ibss.beacon_interval > 10000) - return -EINVAL; - } + + err = cfg80211_validate_beacon_int(rdev, ibss.beacon_interval); + if (err) + return err; if (!rdev->ops->join_ibss) return -EOPNOTSUPP; @@ -9252,9 +9244,10 @@ static int nl80211_join_mesh(struct sk_buff *skb, struct genl_info *info) if (info->attrs[NL80211_ATTR_BEACON_INTERVAL]) { setup.beacon_interval = nla_get_u32(info->attrs[NL80211_ATTR_BEACON_INTERVAL]); - if (setup.beacon_interval < 10 || - setup.beacon_interval > 10000) - return -EINVAL; + + err = cfg80211_validate_beacon_int(rdev, setup.beacon_interval); + if (err) + return err; } if (info->attrs[NL80211_ATTR_DTIM_PERIOD]) { @@ -11847,6 +11840,29 @@ void nl80211_notify_wiphy(struct cfg80211_registered_device *rdev, NL80211_MCGRP_CONFIG, GFP_KERNEL); } +void nl80211_notify_iface(struct cfg80211_registered_device *rdev, + struct wireless_dev *wdev, + enum nl80211_commands cmd) +{ + struct sk_buff *msg; + + WARN_ON(cmd != NL80211_CMD_NEW_INTERFACE && + cmd != NL80211_CMD_DEL_INTERFACE); + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) + return; + + if (nl80211_send_iface(msg, 0, 0, 0, rdev, wdev, + cmd == NL80211_CMD_DEL_INTERFACE) < 0) { + nlmsg_free(msg); + return; + } + + genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0, + NL80211_MCGRP_CONFIG, GFP_KERNEL); +} + static int nl80211_add_scan_req(struct sk_buff *msg, struct cfg80211_registered_device *rdev) { diff --git a/net/wireless/nl80211.h b/net/wireless/nl80211.h index a63f402b10b7..7e3821d7fcc5 100644 --- a/net/wireless/nl80211.h +++ b/net/wireless/nl80211.h @@ -7,6 +7,9 @@ int nl80211_init(void); void nl80211_exit(void); void nl80211_notify_wiphy(struct cfg80211_registered_device *rdev, enum nl80211_commands cmd); +void nl80211_notify_iface(struct cfg80211_registered_device *rdev, + struct wireless_dev *wdev, + enum nl80211_commands cmd); void nl80211_send_scan_start(struct cfg80211_registered_device *rdev, struct wireless_dev *wdev); struct sk_buff *nl80211_build_scan_msg(struct cfg80211_registered_device *rdev, diff --git a/net/wireless/util.c b/net/wireless/util.c index b7d1592bd5b8..0675f513e7b9 100644 --- a/net/wireless/util.c +++ b/net/wireless/util.c @@ -1559,7 +1559,7 @@ int cfg80211_validate_beacon_int(struct cfg80211_registered_device *rdev, struct wireless_dev *wdev; int res = 0; - if (!beacon_int) + if (beacon_int < 10 || beacon_int > 10000) return -EINVAL; list_for_each_entry(wdev, &rdev->wiphy.wdev_list, list) { diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile index 90ebf7d35c07..eb582c6264c3 100644 --- a/samples/bpf/Makefile +++ b/samples/bpf/Makefile @@ -24,6 +24,7 @@ hostprogs-y += test_overhead hostprogs-y += test_cgrp2_array_pin hostprogs-y += xdp1 hostprogs-y += xdp2 +hostprogs-y += test_current_task_under_cgroup test_verifier-objs := test_verifier.o libbpf.o test_maps-objs := test_maps.o libbpf.o @@ -49,6 +50,8 @@ test_cgrp2_array_pin-objs := libbpf.o test_cgrp2_array_pin.o xdp1-objs := bpf_load.o libbpf.o xdp1_user.o # reuse xdp1 source intentionally xdp2-objs := bpf_load.o libbpf.o xdp1_user.o +test_current_task_under_cgroup-objs := bpf_load.o libbpf.o \ + test_current_task_under_cgroup_user.o # Tell kbuild to always build the programs always := $(hostprogs-y) @@ -74,6 +77,7 @@ always += parse_varlen.o parse_simple.o parse_ldabs.o always += test_cgrp2_tc_kern.o always += xdp1_kern.o always += xdp2_kern.o +always += test_current_task_under_cgroup_kern.o HOSTCFLAGS += -I$(objtree)/usr/include @@ -97,6 +101,7 @@ HOSTLOADLIBES_map_perf_test += -lelf -lrt HOSTLOADLIBES_test_overhead += -lelf -lrt HOSTLOADLIBES_xdp1 += -lelf HOSTLOADLIBES_xdp2 += -lelf +HOSTLOADLIBES_test_current_task_under_cgroup += -lelf # Allows pointing LLC/CLANG to a LLVM backend with bpf support, redefine on cmdline: # make samples/bpf/ LLC=~/git/llvm/build/bin/llc CLANG=~/git/llvm/build/bin/clang diff --git a/samples/bpf/bpf_helpers.h b/samples/bpf/bpf_helpers.h index 217c8d507f2e..5e4c41e256b8 100644 --- a/samples/bpf/bpf_helpers.h +++ b/samples/bpf/bpf_helpers.h @@ -37,12 +37,16 @@ static int (*bpf_clone_redirect)(void *ctx, int ifindex, int flags) = (void *) BPF_FUNC_clone_redirect; static int (*bpf_redirect)(int ifindex, int flags) = (void *) BPF_FUNC_redirect; -static int (*bpf_perf_event_output)(void *ctx, void *map, int index, void *data, int size) = +static int (*bpf_perf_event_output)(void *ctx, void *map, + unsigned long long flags, void *data, + int size) = (void *) BPF_FUNC_perf_event_output; static int (*bpf_get_stackid)(void *ctx, void *map, int flags) = (void *) BPF_FUNC_get_stackid; static int (*bpf_probe_write_user)(void *dst, void *src, int size) = (void *) BPF_FUNC_probe_write_user; +static int (*bpf_current_task_under_cgroup)(void *map, int index) = + (void *) BPF_FUNC_current_task_under_cgroup; /* llvm builtin functions that eBPF C program may use to * emit BPF_LD_ABS and BPF_LD_IND instructions diff --git a/samples/bpf/test_current_task_under_cgroup_kern.c b/samples/bpf/test_current_task_under_cgroup_kern.c new file mode 100644 index 000000000000..86b28d7d6c99 --- /dev/null +++ b/samples/bpf/test_current_task_under_cgroup_kern.c @@ -0,0 +1,43 @@ +/* Copyright (c) 2016 Sargun Dhillon <sargun@sargun.me> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + */ + +#include <linux/ptrace.h> +#include <uapi/linux/bpf.h> +#include <linux/version.h> +#include "bpf_helpers.h" +#include <uapi/linux/utsname.h> + +struct bpf_map_def SEC("maps") cgroup_map = { + .type = BPF_MAP_TYPE_CGROUP_ARRAY, + .key_size = sizeof(u32), + .value_size = sizeof(u32), + .max_entries = 1, +}; + +struct bpf_map_def SEC("maps") perf_map = { + .type = BPF_MAP_TYPE_ARRAY, + .key_size = sizeof(u32), + .value_size = sizeof(u64), + .max_entries = 1, +}; + +/* Writes the last PID that called sync to a map at index 0 */ +SEC("kprobe/sys_sync") +int bpf_prog1(struct pt_regs *ctx) +{ + u64 pid = bpf_get_current_pid_tgid(); + int idx = 0; + + if (!bpf_current_task_under_cgroup(&cgroup_map, 0)) + return 0; + + bpf_map_update_elem(&perf_map, &idx, &pid, BPF_ANY); + return 0; +} + +char _license[] SEC("license") = "GPL"; +u32 _version SEC("version") = LINUX_VERSION_CODE; diff --git a/samples/bpf/test_current_task_under_cgroup_user.c b/samples/bpf/test_current_task_under_cgroup_user.c new file mode 100644 index 000000000000..30b0bce884f9 --- /dev/null +++ b/samples/bpf/test_current_task_under_cgroup_user.c @@ -0,0 +1,145 @@ +/* Copyright (c) 2016 Sargun Dhillon <sargun@sargun.me> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + */ + +#define _GNU_SOURCE +#include <stdio.h> +#include <linux/bpf.h> +#include <unistd.h> +#include "libbpf.h" +#include "bpf_load.h" +#include <string.h> +#include <fcntl.h> +#include <errno.h> +#include <linux/bpf.h> +#include <sched.h> +#include <sys/mount.h> +#include <sys/stat.h> +#include <sys/types.h> +#include <linux/limits.h> + +#define CGROUP_MOUNT_PATH "/mnt" +#define CGROUP_PATH "/mnt/my-cgroup" + +#define clean_errno() (errno == 0 ? "None" : strerror(errno)) +#define log_err(MSG, ...) fprintf(stderr, "(%s:%d: errno: %s) " MSG "\n", \ + __FILE__, __LINE__, clean_errno(), ##__VA_ARGS__) + +static int join_cgroup(char *path) +{ + int fd, rc = 0; + pid_t pid = getpid(); + char cgroup_path[PATH_MAX + 1]; + + snprintf(cgroup_path, sizeof(cgroup_path), "%s/cgroup.procs", path); + + fd = open(cgroup_path, O_WRONLY); + if (fd < 0) { + log_err("Opening Cgroup"); + return 1; + } + + if (dprintf(fd, "%d\n", pid) < 0) { + log_err("Joining Cgroup"); + rc = 1; + } + close(fd); + return rc; +} + +int main(int argc, char **argv) +{ + char filename[256]; + int cg2, idx = 0; + pid_t remote_pid, local_pid = getpid(); + + snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]); + if (load_bpf_file(filename)) { + printf("%s", bpf_log_buf); + return 1; + } + + /* + * This is to avoid interfering with existing cgroups. Unfortunately, + * most people don't have cgroupv2 enabled at this point in time. + * It's easier to create our own mount namespace and manage it + * ourselves. + */ + if (unshare(CLONE_NEWNS)) { + log_err("unshare"); + return 1; + } + + if (mount("none", "/", NULL, MS_REC | MS_PRIVATE, NULL)) { + log_err("mount fakeroot"); + return 1; + } + + if (mount("none", CGROUP_MOUNT_PATH, "cgroup2", 0, NULL)) { + log_err("mount cgroup2"); + return 1; + } + + if (mkdir(CGROUP_PATH, 0777) && errno != EEXIST) { + log_err("mkdir cgroup"); + return 1; + } + + cg2 = open(CGROUP_PATH, O_RDONLY); + if (cg2 < 0) { + log_err("opening target cgroup"); + goto cleanup_cgroup_err; + } + + if (bpf_update_elem(map_fd[0], &idx, &cg2, BPF_ANY)) { + log_err("Adding target cgroup to map"); + goto cleanup_cgroup_err; + } + if (join_cgroup("/mnt/my-cgroup")) { + log_err("Leaving target cgroup"); + goto cleanup_cgroup_err; + } + + /* + * The installed helper program catched the sync call, and should + * write it to the map. + */ + + sync(); + bpf_lookup_elem(map_fd[1], &idx, &remote_pid); + + if (local_pid != remote_pid) { + fprintf(stderr, + "BPF Helper didn't write correct PID to map, but: %d\n", + remote_pid); + goto leave_cgroup_err; + } + + /* Verify the negative scenario; leave the cgroup */ + if (join_cgroup(CGROUP_MOUNT_PATH)) + goto leave_cgroup_err; + + remote_pid = 0; + bpf_update_elem(map_fd[1], &idx, &remote_pid, BPF_ANY); + + sync(); + bpf_lookup_elem(map_fd[1], &idx, &remote_pid); + + if (local_pid == remote_pid) { + fprintf(stderr, "BPF cgroup negative test did not work\n"); + goto cleanup_cgroup_err; + } + + rmdir(CGROUP_PATH); + return 0; + + /* Error condition, cleanup */ +leave_cgroup_err: + join_cgroup(CGROUP_MOUNT_PATH); +cleanup_cgroup_err: + rmdir(CGROUP_PATH); + return 1; +} diff --git a/samples/bpf/test_verifier.c b/samples/bpf/test_verifier.c index fe2fcec98c1f..78c6f131d94f 100644 --- a/samples/bpf/test_verifier.c +++ b/samples/bpf/test_verifier.c @@ -1449,7 +1449,7 @@ static struct bpf_test tests[] = { .prog_type = BPF_PROG_TYPE_SCHED_CLS, }, { - "pkt: test1", + "direct packet access: test1", .insns = { BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct __sk_buff, data)), @@ -1466,7 +1466,7 @@ static struct bpf_test tests[] = { .prog_type = BPF_PROG_TYPE_SCHED_CLS, }, { - "pkt: test2", + "direct packet access: test2", .insns = { BPF_MOV64_IMM(BPF_REG_0, 1), BPF_LDX_MEM(BPF_W, BPF_REG_4, BPF_REG_1, @@ -1499,7 +1499,7 @@ static struct bpf_test tests[] = { .prog_type = BPF_PROG_TYPE_SCHED_CLS, }, { - "pkt: test3", + "direct packet access: test3", .insns = { BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct __sk_buff, data)), @@ -1511,7 +1511,7 @@ static struct bpf_test tests[] = { .prog_type = BPF_PROG_TYPE_SOCKET_FILTER, }, { - "pkt: test4", + "direct packet access: test4", .insns = { BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, offsetof(struct __sk_buff, data)), @@ -1528,6 +1528,112 @@ static struct bpf_test tests[] = { .result = REJECT, .prog_type = BPF_PROG_TYPE_SCHED_CLS, }, + { + "helper access to packet: test1, valid packet_ptr range", + .insns = { + BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, + offsetof(struct xdp_md, data)), + BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, + offsetof(struct xdp_md, data_end)), + BPF_MOV64_REG(BPF_REG_1, BPF_REG_2), + BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, 8), + BPF_JMP_REG(BPF_JGT, BPF_REG_1, BPF_REG_3, 5), + BPF_LD_MAP_FD(BPF_REG_1, 0), + BPF_MOV64_REG(BPF_REG_3, BPF_REG_2), + BPF_MOV64_IMM(BPF_REG_4, 0), + BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_update_elem), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + }, + .fixup = {5}, + .result_unpriv = ACCEPT, + .result = ACCEPT, + .prog_type = BPF_PROG_TYPE_XDP, + }, + { + "helper access to packet: test2, unchecked packet_ptr", + .insns = { + BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, + offsetof(struct xdp_md, data)), + BPF_LD_MAP_FD(BPF_REG_1, 0), + BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + }, + .fixup = {1}, + .result = REJECT, + .errstr = "invalid access to packet", + .prog_type = BPF_PROG_TYPE_XDP, + }, + { + "helper access to packet: test3, variable add", + .insns = { + BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, + offsetof(struct xdp_md, data)), + BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, + offsetof(struct xdp_md, data_end)), + BPF_MOV64_REG(BPF_REG_4, BPF_REG_2), + BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, 8), + BPF_JMP_REG(BPF_JGT, BPF_REG_4, BPF_REG_3, 10), + BPF_LDX_MEM(BPF_B, BPF_REG_5, BPF_REG_2, 0), + BPF_MOV64_REG(BPF_REG_4, BPF_REG_2), + BPF_ALU64_REG(BPF_ADD, BPF_REG_4, BPF_REG_5), + BPF_MOV64_REG(BPF_REG_5, BPF_REG_4), + BPF_ALU64_IMM(BPF_ADD, BPF_REG_5, 8), + BPF_JMP_REG(BPF_JGT, BPF_REG_5, BPF_REG_3, 4), + BPF_LD_MAP_FD(BPF_REG_1, 0), + BPF_MOV64_REG(BPF_REG_2, BPF_REG_4), + BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + }, + .fixup = {11}, + .result = ACCEPT, + .prog_type = BPF_PROG_TYPE_XDP, + }, + { + "helper access to packet: test4, packet_ptr with bad range", + .insns = { + BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, + offsetof(struct xdp_md, data)), + BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, + offsetof(struct xdp_md, data_end)), + BPF_MOV64_REG(BPF_REG_4, BPF_REG_2), + BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, 4), + BPF_JMP_REG(BPF_JGT, BPF_REG_4, BPF_REG_3, 2), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + BPF_LD_MAP_FD(BPF_REG_1, 0), + BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + }, + .fixup = {7}, + .result = REJECT, + .errstr = "invalid access to packet", + .prog_type = BPF_PROG_TYPE_XDP, + }, + { + "helper access to packet: test5, packet_ptr with too short range", + .insns = { + BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, + offsetof(struct xdp_md, data)), + BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1, + offsetof(struct xdp_md, data_end)), + BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, 1), + BPF_MOV64_REG(BPF_REG_4, BPF_REG_2), + BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, 7), + BPF_JMP_REG(BPF_JGT, BPF_REG_4, BPF_REG_3, 3), + BPF_LD_MAP_FD(BPF_REG_1, 0), + BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem), + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + }, + .fixup = {6}, + .result = REJECT, + .errstr = "invalid access to packet", + .prog_type = BPF_PROG_TYPE_XDP, + }, }; static int probe_filter_length(struct bpf_insn *fp) |