From patchwork Sat Jan 23 10:46:06 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Coiby Xu X-Patchwork-Id: 369764 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-15.8 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,FREEMAIL_FORGED_FROMDOMAIN,FREEMAIL_FROM, HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER,INCLUDES_PATCH, MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,USER_AGENT_GIT autolearn=unavailable autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id A93CAC433DB for ; Sat, 23 Jan 2021 10:50:01 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 4CFDA23358 for ; Sat, 23 Jan 2021 10:50:01 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726722AbhAWKrl (ORCPT ); Sat, 23 Jan 2021 05:47:41 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:59946 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726560AbhAWKrW (ORCPT ); Sat, 23 Jan 2021 05:47:22 -0500 Received: from mail-pj1-x102c.google.com (mail-pj1-x102c.google.com [IPv6:2607:f8b0:4864:20::102c]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 5D866C06174A; Sat, 23 Jan 2021 02:46:42 -0800 (PST) Received: by mail-pj1-x102c.google.com with SMTP id m5so5461336pjv.5; Sat, 23 Jan 2021 02:46:42 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=from:to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; bh=XS9s6QzYzZnFdCcrnXfNRB7PNo15vgZmoru8aD+ih64=; b=rmODAOSN6TRlXtZtt7ZYKBzzQRk2AYe9IPXFYPNJkUAM1TDDIFj+HtbEDapRJ8Uyh8 f1EJd+DWj12p9MZmtKJBUBXTUgpTjJU53SmJS7ikv/n9YsTK1e9V29RAAxxK0MEWJcZw g/b48Uw/Yna9TX1Ambu6b0HK71ixJIE3wGX+Ey5USIv6N7wuhbBw2VhfmdM95uuZaG4Z y4pxzQw7oKvZM6hx1a0oURL/ner0LNTQxcfQCCqykJ8BfHu2NmaDJgp3FjIISpiGj9JG ZpCX4hKJkI3ZAF3E9lugopzSGTKsIAPv7fGV7IDNQ87w4ahZVkAC/dLsLddHLcanSWe1 JSrA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references:mime-version:content-transfer-encoding; bh=XS9s6QzYzZnFdCcrnXfNRB7PNo15vgZmoru8aD+ih64=; b=a+ByYBBoKRuul9vaH9aim03qFhgZ+b0KLwzOA0EA28hpQ5f+hTaAyarhfVBQ4uCE+p mOmtSORUTlrihTYS0uwuAnh0uIjJ6dLniP4APtIYhdD0jRQvyWKsJwI94AJRk9oLdP0+ VXUhrMZDEuZUzllo86UNX77d55VHpaBC1zYPTU1P0bgmvcOyGD+eI0up+8OHADVE0fhb HP6eTXtW08Gv0o6nAG80G93tkBHURA/iimLy7fofCC49OdHEufXS5JysRdRZXQAtriuE j48i4dFc2Roz2uJn4IruyWJxvYqb+3sR7ueMRHO8wDlY4badgu/gMzULn8MC/44rhTO4 uyhQ== X-Gm-Message-State: AOAM532TrkuJxSZ7JUQffPZZOJk+qtgSoFskazQwYQCTul4ekemjFgc4 BRfB32Ueaogkz7VsS9jT+BUKySyp9kyCOxz2 X-Google-Smtp-Source: ABdhPJzV76bgx0Q+wNsNKpc2KSjFwAB85m6JOup9pg+t55gR8Y041BXpw+jxmGAjY+TwOKaATiIGUA== X-Received: by 2002:a17:902:d2cb:b029:de:757c:1f0c with SMTP id n11-20020a170902d2cbb02900de757c1f0cmr932712plc.40.1611398801123; Sat, 23 Jan 2021 02:46:41 -0800 (PST) Received: from localhost ([2001:e42:102:1532:160:16:113:140]) by smtp.gmail.com with ESMTPSA id j18sm5942180pfc.99.2021.01.23.02.46.24 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Sat, 23 Jan 2021 02:46:40 -0800 (PST) From: Coiby Xu To: devel@driverdev.osuosl.org Cc: Benjamin Poirier , Dan Carpenter , Shung-Hsi Yu , Willem de Bruijn , GR-Linux-NIC-Dev@marvell.com, Manish Chopra , Greg Kroah-Hartman , netdev@vger.kernel.org (open list:QLOGIC QLGE 10Gb ETHERNET DRIVER), linux-kernel@vger.kernel.org (open list) Subject: [PATCH v4 1/8] staging: qlge: use qlge_* prefix to avoid namespace clashes with other qlogic drivers Date: Sat, 23 Jan 2021 18:46:06 +0800 Message-Id: <20210123104613.38359-2-coiby.xu@gmail.com> X-Mailer: git-send-email 2.30.0 In-Reply-To: <20210123104613.38359-1-coiby.xu@gmail.com> References: <20210123104613.38359-1-coiby.xu@gmail.com> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org To avoid namespace clashes with other qlogic drivers and also for the sake of naming consistency, use the "qlge_" prefix as suggested in drivers/staging/qlge/TODO, - For existing ql_ prefix, sed -i "s/ql_/qlge_/g" *.{c,h} - for structs not having a prefix 1. get a list of structs grep "struct.*{" qlge. 2. add qlge_ for each struct, e.g., sed -i "s/ib_ae_iocb_rsp/qlge_ib_ae_iocb_rsp/g" *.{c,h} Link: https://lore.kernel.org/patchwork/patch/1318503/#1516131 Suggested-by: Benjamin Poirier Signed-off-by: Coiby Xu --- drivers/staging/qlge/TODO | 4 - drivers/staging/qlge/qlge.h | 210 ++--- drivers/staging/qlge/qlge_dbg.c | 1075 ++++++++++++----------- drivers/staging/qlge/qlge_ethtool.c | 230 ++--- drivers/staging/qlge/qlge_main.c | 1268 +++++++++++++-------------- drivers/staging/qlge/qlge_mpi.c | 352 ++++---- 6 files changed, 1567 insertions(+), 1572 deletions(-) diff --git a/drivers/staging/qlge/TODO b/drivers/staging/qlge/TODO index f93f7428f5d5..5ac55664c3e2 100644 --- a/drivers/staging/qlge/TODO +++ b/drivers/staging/qlge/TODO @@ -28,10 +28,6 @@ * the driver has a habit of using runtime checks where compile time checks are possible (ex. ql_free_rx_buffers(), ql_alloc_rx_buffers()) * reorder struct members to avoid holes if it doesn't impact performance -* in terms of namespace, the driver uses either qlge_, ql_ (used by - other qlogic drivers, with clashes, ex: ql_sem_spinlock) or nothing (with - clashes, ex: struct ob_mac_iocb_req). Rename everything to use the "qlge_" - prefix. * avoid legacy/deprecated apis (ex. replace pci_dma_*, replace pci_enable_msi, use pci_iomap) * some "while" loops could be rewritten with simple "for", ex. diff --git a/drivers/staging/qlge/qlge.h b/drivers/staging/qlge/qlge.h index 0381f3f56bc7..1ac85f2f770f 100644 --- a/drivers/staging/qlge/qlge.h +++ b/drivers/staging/qlge/qlge.h @@ -1081,7 +1081,7 @@ struct tx_buf_desc { #define OPCODE_IB_MPI_IOCB 0x21 #define OPCODE_IB_AE_IOCB 0x3f -struct ob_mac_iocb_req { +struct qlge_ob_mac_iocb_req { u8 opcode; u8 flags1; #define OB_MAC_IOCB_REQ_OI 0x01 @@ -1104,7 +1104,7 @@ struct ob_mac_iocb_req { struct tx_buf_desc tbd[TX_DESC_PER_IOCB]; } __packed; -struct ob_mac_iocb_rsp { +struct qlge_ob_mac_iocb_rsp { u8 opcode; /* */ u8 flags1; /* */ #define OB_MAC_IOCB_RSP_OI 0x01 /* */ @@ -1121,7 +1121,7 @@ struct ob_mac_iocb_rsp { __le32 reserved[13]; } __packed; -struct ob_mac_tso_iocb_req { +struct qlge_ob_mac_tso_iocb_req { u8 opcode; u8 flags1; #define OB_MAC_TSO_IOCB_OI 0x01 @@ -1149,7 +1149,7 @@ struct ob_mac_tso_iocb_req { struct tx_buf_desc tbd[TX_DESC_PER_IOCB]; } __packed; -struct ob_mac_tso_iocb_rsp { +struct qlge_ob_mac_tso_iocb_rsp { u8 opcode; u8 flags1; #define OB_MAC_TSO_IOCB_RSP_OI 0x01 @@ -1166,7 +1166,7 @@ struct ob_mac_tso_iocb_rsp { __le32 reserved2[13]; } __packed; -struct ib_mac_iocb_rsp { +struct qlge_ib_mac_iocb_rsp { u8 opcode; /* 0x20 */ u8 flags1; #define IB_MAC_IOCB_RSP_OI 0x01 /* Override intr delay */ @@ -1225,7 +1225,7 @@ struct ib_mac_iocb_rsp { __le64 hdr_addr; /* */ } __packed; -struct ib_ae_iocb_rsp { +struct qlge_ib_ae_iocb_rsp { u8 opcode; u8 flags1; #define IB_AE_IOCB_RSP_OI 0x01 @@ -1250,7 +1250,7 @@ struct ib_ae_iocb_rsp { * These three structures are for generic * handling of ib and ob iocbs. */ -struct ql_net_rsp_iocb { +struct qlge_net_rsp_iocb { u8 opcode; u8 flags0; __le16 length; @@ -1258,7 +1258,7 @@ struct ql_net_rsp_iocb { __le32 reserved[14]; } __packed; -struct net_req_iocb { +struct qlge_net_req_iocb { u8 opcode; u8 flags0; __le16 flags1; @@ -1346,7 +1346,7 @@ struct ricb { /* SOFTWARE/DRIVER DATA STRUCTURES. */ -struct oal { +struct qlge_oal { struct tx_buf_desc oal[TX_DESC_PER_OAL]; }; @@ -1357,9 +1357,9 @@ struct map_list { struct tx_ring_desc { struct sk_buff *skb; - struct ob_mac_iocb_req *queue_entry; + struct qlge_ob_mac_iocb_req *queue_entry; u32 index; - struct oal oal; + struct qlge_oal oal; struct map_list map[MAX_SKB_FRAGS + 2]; int map_cnt; struct tx_ring_desc *next; @@ -1388,7 +1388,7 @@ struct tx_ring { spinlock_t lock; atomic_t tx_count; /* counts down for every outstanding IO */ struct delayed_work tx_work; - struct ql_adapter *qdev; + struct qlge_adapter *qdev; u64 tx_packets; u64 tx_bytes; u64 tx_errors; @@ -1469,7 +1469,7 @@ struct rx_ring { dma_addr_t prod_idx_sh_reg_dma; void __iomem *cnsmr_idx_db_reg; /* PCI doorbell mem area + 0 */ u32 cnsmr_idx; /* current sw idx */ - struct ql_net_rsp_iocb *curr_entry; /* next entry on queue */ + struct qlge_net_rsp_iocb *curr_entry; /* next entry on queue */ void __iomem *valid_db_reg; /* PCI doorbell mem area + 0x04 */ /* Large buffer queue elements. */ @@ -1487,7 +1487,7 @@ struct rx_ring { char name[IFNAMSIZ + 5]; struct napi_struct napi; u8 reserved; - struct ql_adapter *qdev; + struct qlge_adapter *qdev; u64 rx_packets; u64 rx_multicast; u64 rx_bytes; @@ -1752,14 +1752,14 @@ enum { #define SHADOW_OFFSET 0xb0000000 #define SHADOW_REG_SHIFT 20 -struct ql_nic_misc { +struct qlge_nic_misc { u32 rx_ring_count; u32 tx_ring_count; u32 intr_count; u32 function; }; -struct ql_reg_dump { +struct qlge_reg_dump { /* segment 0 */ struct mpi_coredump_global_header mpi_global_header; @@ -1769,7 +1769,7 @@ struct ql_reg_dump { /* segment 30 */ struct mpi_coredump_segment_header misc_nic_seg_hdr; - struct ql_nic_misc misc_nic_info; + struct qlge_nic_misc misc_nic_info; /* segment 31 */ /* one interrupt state for each CQ */ @@ -1792,7 +1792,7 @@ struct ql_reg_dump { u32 ets[8 + 2]; }; -struct ql_mpi_coredump { +struct qlge_mpi_coredump { /* segment 0 */ struct mpi_coredump_global_header mpi_global_header; @@ -1914,7 +1914,7 @@ struct ql_mpi_coredump { /* segment 30 */ struct mpi_coredump_segment_header misc_nic_seg_hdr; - struct ql_nic_misc misc_nic_info; + struct qlge_nic_misc misc_nic_info; /* segment 31 */ /* one interrupt state for each CQ */ @@ -1991,7 +1991,7 @@ struct ql_mpi_coredump { * irq environment as a context to the ISR. */ struct intr_context { - struct ql_adapter *qdev; + struct qlge_adapter *qdev; u32 intr; u32 irq_mask; /* Mask of which rings the vector services. */ u32 hooked; @@ -2056,15 +2056,15 @@ enum { }; struct nic_operations { - int (*get_flash)(struct ql_adapter *qdev); - int (*port_initialize)(struct ql_adapter *qdev); + int (*get_flash)(struct qlge_adapter *qdev); + int (*port_initialize)(struct qlge_adapter *qdev); }; /* * The main Adapter structure definition. * This structure has all fields relevant to the hardware. */ -struct ql_adapter { +struct qlge_adapter { struct ricb ricb; unsigned long flags; u32 wol; @@ -2139,7 +2139,7 @@ struct ql_adapter { u32 port_link_up; u32 port_init; u32 link_status; - struct ql_mpi_coredump *mpi_coredump; + struct qlge_mpi_coredump *mpi_coredump; u32 core_is_dumped; u32 link_config; u32 led_config; @@ -2166,7 +2166,7 @@ struct ql_adapter { /* * Typical Register accessor for memory mapped device. */ -static inline u32 ql_read32(const struct ql_adapter *qdev, int reg) +static inline u32 qlge_read32(const struct qlge_adapter *qdev, int reg) { return readl(qdev->reg_base + reg); } @@ -2174,7 +2174,7 @@ static inline u32 ql_read32(const struct ql_adapter *qdev, int reg) /* * Typical Register accessor for memory mapped device. */ -static inline void ql_write32(const struct ql_adapter *qdev, int reg, u32 val) +static inline void qlge_write32(const struct qlge_adapter *qdev, int reg, u32 val) { writel(val, qdev->reg_base + reg); } @@ -2189,7 +2189,7 @@ static inline void ql_write32(const struct ql_adapter *qdev, int reg, u32 val) * 1 4k chunk of memory. The lower half of the space is for outbound * queues. The upper half is for inbound queues. */ -static inline void ql_write_db_reg(u32 val, void __iomem *addr) +static inline void qlge_write_db_reg(u32 val, void __iomem *addr) { writel(val, addr); } @@ -2205,7 +2205,7 @@ static inline void ql_write_db_reg(u32 val, void __iomem *addr) * queues. The upper half is for inbound queues. * Caller has to guarantee ordering. */ -static inline void ql_write_db_reg_relaxed(u32 val, void __iomem *addr) +static inline void qlge_write_db_reg_relaxed(u32 val, void __iomem *addr) { writel_relaxed(val, addr); } @@ -2220,7 +2220,7 @@ static inline void ql_write_db_reg_relaxed(u32 val, void __iomem *addr) * update the relevant index register and then copy the value to the * shadow register in host memory. */ -static inline u32 ql_read_sh_reg(__le32 *addr) +static inline u32 qlge_read_sh_reg(__le32 *addr) { u32 reg; @@ -2233,51 +2233,51 @@ extern char qlge_driver_name[]; extern const char qlge_driver_version[]; extern const struct ethtool_ops qlge_ethtool_ops; -int ql_sem_spinlock(struct ql_adapter *qdev, u32 sem_mask); -void ql_sem_unlock(struct ql_adapter *qdev, u32 sem_mask); -int ql_read_xgmac_reg(struct ql_adapter *qdev, u32 reg, u32 *data); -int ql_get_mac_addr_reg(struct ql_adapter *qdev, u32 type, u16 index, - u32 *value); -int ql_get_routing_reg(struct ql_adapter *qdev, u32 index, u32 *value); -int ql_write_cfg(struct ql_adapter *qdev, void *ptr, int size, u32 bit, - u16 q_id); -void ql_queue_fw_error(struct ql_adapter *qdev); -void ql_mpi_work(struct work_struct *work); -void ql_mpi_reset_work(struct work_struct *work); -void ql_mpi_core_to_log(struct work_struct *work); -int ql_wait_reg_rdy(struct ql_adapter *qdev, u32 reg, u32 bit, u32 ebit); -void ql_queue_asic_error(struct ql_adapter *qdev); -void ql_set_ethtool_ops(struct net_device *ndev); -int ql_read_xgmac_reg64(struct ql_adapter *qdev, u32 reg, u64 *data); -void ql_mpi_idc_work(struct work_struct *work); -void ql_mpi_port_cfg_work(struct work_struct *work); -int ql_mb_get_fw_state(struct ql_adapter *qdev); -int ql_cam_route_initialize(struct ql_adapter *qdev); -int ql_read_mpi_reg(struct ql_adapter *qdev, u32 reg, u32 *data); -int ql_write_mpi_reg(struct ql_adapter *qdev, u32 reg, u32 data); -int ql_unpause_mpi_risc(struct ql_adapter *qdev); -int ql_pause_mpi_risc(struct ql_adapter *qdev); -int ql_hard_reset_mpi_risc(struct ql_adapter *qdev); -int ql_soft_reset_mpi_risc(struct ql_adapter *qdev); -int ql_dump_risc_ram_area(struct ql_adapter *qdev, void *buf, u32 ram_addr, - int word_count); -int ql_core_dump(struct ql_adapter *qdev, struct ql_mpi_coredump *mpi_coredump); -int ql_mb_about_fw(struct ql_adapter *qdev); -int ql_mb_wol_set_magic(struct ql_adapter *qdev, u32 enable_wol); -int ql_mb_wol_mode(struct ql_adapter *qdev, u32 wol); -int ql_mb_set_led_cfg(struct ql_adapter *qdev, u32 led_config); -int ql_mb_get_led_cfg(struct ql_adapter *qdev); -void ql_link_on(struct ql_adapter *qdev); -void ql_link_off(struct ql_adapter *qdev); -int ql_mb_set_mgmnt_traffic_ctl(struct ql_adapter *qdev, u32 control); -int ql_mb_get_port_cfg(struct ql_adapter *qdev); -int ql_mb_set_port_cfg(struct ql_adapter *qdev); -int ql_wait_fifo_empty(struct ql_adapter *qdev); -void ql_get_dump(struct ql_adapter *qdev, void *buff); -netdev_tx_t ql_lb_send(struct sk_buff *skb, struct net_device *ndev); -void ql_check_lb_frame(struct ql_adapter *qdev, struct sk_buff *skb); -int ql_own_firmware(struct ql_adapter *qdev); -int ql_clean_lb_rx_ring(struct rx_ring *rx_ring, int budget); +int qlge_sem_spinlock(struct qlge_adapter *qdev, u32 sem_mask); +void qlge_sem_unlock(struct qlge_adapter *qdev, u32 sem_mask); +int qlge_read_xgmac_reg(struct qlge_adapter *qdev, u32 reg, u32 *data); +int qlge_get_mac_addr_reg(struct qlge_adapter *qdev, u32 type, u16 index, + u32 *value); +int qlge_get_routing_reg(struct qlge_adapter *qdev, u32 index, u32 *value); +int qlge_write_cfg(struct qlge_adapter *qdev, void *ptr, int size, u32 bit, + u16 q_id); +void qlge_queue_fw_error(struct qlge_adapter *qdev); +void qlge_mpi_work(struct work_struct *work); +void qlge_mpi_reset_work(struct work_struct *work); +void qlge_mpi_core_to_log(struct work_struct *work); +int qlge_wait_reg_rdy(struct qlge_adapter *qdev, u32 reg, u32 bit, u32 ebit); +void qlge_queue_asic_error(struct qlge_adapter *qdev); +void qlge_set_ethtool_ops(struct net_device *ndev); +int qlge_read_xgmac_reg64(struct qlge_adapter *qdev, u32 reg, u64 *data); +void qlge_mpi_idc_work(struct work_struct *work); +void qlge_mpi_port_cfg_work(struct work_struct *work); +int qlge_mb_get_fw_state(struct qlge_adapter *qdev); +int qlge_cam_route_initialize(struct qlge_adapter *qdev); +int qlge_read_mpi_reg(struct qlge_adapter *qdev, u32 reg, u32 *data); +int qlge_write_mpi_reg(struct qlge_adapter *qdev, u32 reg, u32 data); +int qlge_unpause_mpi_risc(struct qlge_adapter *qdev); +int qlge_pause_mpi_risc(struct qlge_adapter *qdev); +int qlge_hard_reset_mpi_risc(struct qlge_adapter *qdev); +int qlge_soft_reset_mpi_risc(struct qlge_adapter *qdev); +int qlge_dump_risc_ram_area(struct qlge_adapter *qdev, void *buf, u32 ram_addr, + int word_count); +int qlge_core_dump(struct qlge_adapter *qdev, struct qlge_mpi_coredump *mpi_coredump); +int qlge_mb_about_fw(struct qlge_adapter *qdev); +int qlge_mb_wol_set_magic(struct qlge_adapter *qdev, u32 enable_wol); +int qlge_mb_wol_mode(struct qlge_adapter *qdev, u32 wol); +int qlge_mb_set_led_cfg(struct qlge_adapter *qdev, u32 led_config); +int qlge_mb_get_led_cfg(struct qlge_adapter *qdev); +void qlge_link_on(struct qlge_adapter *qdev); +void qlge_link_off(struct qlge_adapter *qdev); +int qlge_mb_set_mgmnt_traffic_ctl(struct qlge_adapter *qdev, u32 control); +int qlge_mb_get_port_cfg(struct qlge_adapter *qdev); +int qlge_mb_set_port_cfg(struct qlge_adapter *qdev); +int qlge_wait_fifo_empty(struct qlge_adapter *qdev); +void qlge_get_dump(struct qlge_adapter *qdev, void *buff); +netdev_tx_t qlge_lb_send(struct sk_buff *skb, struct net_device *ndev); +void qlge_check_lb_frame(struct qlge_adapter *qdev, struct sk_buff *skb); +int qlge_own_firmware(struct qlge_adapter *qdev); +int qlge_clean_lb_rx_ring(struct rx_ring *rx_ring, int budget); /* #define QL_ALL_DUMP */ /* #define QL_REG_DUMP */ @@ -2287,12 +2287,12 @@ int ql_clean_lb_rx_ring(struct rx_ring *rx_ring, int budget); /* #define QL_OB_DUMP */ #ifdef QL_REG_DUMP -void ql_dump_xgmac_control_regs(struct ql_adapter *qdev); -void ql_dump_routing_entries(struct ql_adapter *qdev); -void ql_dump_regs(struct ql_adapter *qdev); -#define QL_DUMP_REGS(qdev) ql_dump_regs(qdev) -#define QL_DUMP_ROUTE(qdev) ql_dump_routing_entries(qdev) -#define QL_DUMP_XGMAC_CONTROL_REGS(qdev) ql_dump_xgmac_control_regs(qdev) +void qlge_dump_xgmac_control_regs(struct qlge_adapter *qdev); +void qlge_dump_routing_entries(struct qlge_adapter *qdev); +void qlge_dump_regs(struct qlge_adapter *qdev); +#define QL_DUMP_REGS(qdev) qlge_dump_regs(qdev) +#define QL_DUMP_ROUTE(qdev) qlge_dump_routing_entries(qdev) +#define QL_DUMP_XGMAC_CONTROL_REGS(qdev) qlge_dump_xgmac_control_regs(qdev) #else #define QL_DUMP_REGS(qdev) #define QL_DUMP_ROUTE(qdev) @@ -2300,33 +2300,33 @@ void ql_dump_regs(struct ql_adapter *qdev); #endif #ifdef QL_STAT_DUMP -void ql_dump_stat(struct ql_adapter *qdev); -#define QL_DUMP_STAT(qdev) ql_dump_stat(qdev) +void qlge_dump_stat(struct qlge_adapter *qdev); +#define QL_DUMP_STAT(qdev) qlge_dump_stat(qdev) #else #define QL_DUMP_STAT(qdev) #endif #ifdef QL_DEV_DUMP -void ql_dump_qdev(struct ql_adapter *qdev); -#define QL_DUMP_QDEV(qdev) ql_dump_qdev(qdev) +void qlge_dump_qdev(struct qlge_adapter *qdev); +#define QL_DUMP_QDEV(qdev) qlge_dump_qdev(qdev) #else #define QL_DUMP_QDEV(qdev) #endif #ifdef QL_CB_DUMP -void ql_dump_wqicb(struct wqicb *wqicb); -void ql_dump_tx_ring(struct tx_ring *tx_ring); -void ql_dump_ricb(struct ricb *ricb); -void ql_dump_cqicb(struct cqicb *cqicb); -void ql_dump_rx_ring(struct rx_ring *rx_ring); -void ql_dump_hw_cb(struct ql_adapter *qdev, int size, u32 bit, u16 q_id); -#define QL_DUMP_RICB(ricb) ql_dump_ricb(ricb) -#define QL_DUMP_WQICB(wqicb) ql_dump_wqicb(wqicb) -#define QL_DUMP_TX_RING(tx_ring) ql_dump_tx_ring(tx_ring) -#define QL_DUMP_CQICB(cqicb) ql_dump_cqicb(cqicb) -#define QL_DUMP_RX_RING(rx_ring) ql_dump_rx_ring(rx_ring) +void qlge_dump_wqicb(struct wqicb *wqicb); +void qlge_dump_tx_ring(struct tx_ring *tx_ring); +void qlge_dump_ricb(struct ricb *ricb); +void qlge_dump_cqicb(struct cqicb *cqicb); +void qlge_dump_rx_ring(struct rx_ring *rx_ring); +void qlge_dump_hw_cb(struct qlge_adapter *qdev, int size, u32 bit, u16 q_id); +#define QL_DUMP_RICB(ricb) qlge_dump_ricb(ricb) +#define QL_DUMP_WQICB(wqicb) qlge_dump_wqicb(wqicb) +#define QL_DUMP_TX_RING(tx_ring) qlge_dump_tx_ring(tx_ring) +#define QL_DUMP_CQICB(cqicb) qlge_dump_cqicb(cqicb) +#define QL_DUMP_RX_RING(rx_ring) qlge_dump_rx_ring(rx_ring) #define QL_DUMP_HW_CB(qdev, size, bit, q_id) \ - ql_dump_hw_cb(qdev, size, bit, q_id) + qlge_dump_hw_cb(qdev, size, bit, q_id) #else #define QL_DUMP_RICB(ricb) #define QL_DUMP_WQICB(wqicb) @@ -2337,26 +2337,26 @@ void ql_dump_hw_cb(struct ql_adapter *qdev, int size, u32 bit, u16 q_id); #endif #ifdef QL_OB_DUMP -void ql_dump_tx_desc(struct ql_adapter *qdev, struct tx_buf_desc *tbd); -void ql_dump_ob_mac_iocb(struct ql_adapter *qdev, struct ob_mac_iocb_req *ob_mac_iocb); -void ql_dump_ob_mac_rsp(struct ql_adapter *qdev, struct ob_mac_iocb_rsp *ob_mac_rsp); -#define QL_DUMP_OB_MAC_IOCB(qdev, ob_mac_iocb) ql_dump_ob_mac_iocb(qdev, ob_mac_iocb) -#define QL_DUMP_OB_MAC_RSP(qdev, ob_mac_rsp) ql_dump_ob_mac_rsp(qdev, ob_mac_rsp) +void qlge_dump_tx_desc(struct qlge_adapter *qdev, struct tx_buf_desc *tbd); +void qlge_dump_ob_mac_iocb(struct qlge_adapter *qdev, struct qlge_ob_mac_iocb_req *ob_mac_iocb); +void qlge_dump_ob_mac_rsp(struct qlge_adapter *qdev, struct qlge_ob_mac_iocb_rsp *ob_mac_rsp); +#define QL_DUMP_OB_MAC_IOCB(qdev, ob_mac_iocb) qlge_dump_ob_mac_iocb(qdev, ob_mac_iocb) +#define QL_DUMP_OB_MAC_RSP(qdev, ob_mac_rsp) qlge_dump_ob_mac_rsp(qdev, ob_mac_rsp) #else #define QL_DUMP_OB_MAC_IOCB(qdev, ob_mac_iocb) #define QL_DUMP_OB_MAC_RSP(qdev, ob_mac_rsp) #endif #ifdef QL_IB_DUMP -void ql_dump_ib_mac_rsp(struct ql_adapter *qdev, struct ib_mac_iocb_rsp *ib_mac_rsp); -#define QL_DUMP_IB_MAC_RSP(qdev, ib_mac_rsp) ql_dump_ib_mac_rsp(qdev, ib_mac_rsp) +void qlge_dump_ib_mac_rsp(struct qlge_adapter *qdev, struct qlge_ib_mac_iocb_rsp *ib_mac_rsp); +#define QL_DUMP_IB_MAC_RSP(qdev, ib_mac_rsp) qlge_dump_ib_mac_rsp(qdev, ib_mac_rsp) #else #define QL_DUMP_IB_MAC_RSP(qdev, ib_mac_rsp) #endif #ifdef QL_ALL_DUMP -void ql_dump_all(struct ql_adapter *qdev); -#define QL_DUMP_ALL(qdev) ql_dump_all(qdev) +void qlge_dump_all(struct qlge_adapter *qdev); +#define QL_DUMP_ALL(qdev) qlge_dump_all(qdev) #else #define QL_DUMP_ALL(qdev) #endif diff --git a/drivers/staging/qlge/qlge_dbg.c b/drivers/staging/qlge/qlge_dbg.c index 42fd13990f3a..b0d4ea071f32 100644 --- a/drivers/staging/qlge/qlge_dbg.c +++ b/drivers/staging/qlge/qlge_dbg.c @@ -6,8 +6,8 @@ #include "qlge.h" /* Read a NIC register from the alternate function. */ -static u32 ql_read_other_func_reg(struct ql_adapter *qdev, - u32 reg) +static u32 qlge_read_other_func_reg(struct qlge_adapter *qdev, + u32 reg) { u32 register_to_read; u32 reg_val; @@ -17,7 +17,7 @@ static u32 ql_read_other_func_reg(struct ql_adapter *qdev, | MPI_NIC_READ | (qdev->alt_func << MPI_NIC_FUNCTION_SHIFT) | reg; - status = ql_read_mpi_reg(qdev, register_to_read, ®_val); + status = qlge_read_mpi_reg(qdev, register_to_read, ®_val); if (status != 0) return 0xffffffff; @@ -25,8 +25,8 @@ static u32 ql_read_other_func_reg(struct ql_adapter *qdev, } /* Write a NIC register from the alternate function. */ -static int ql_write_other_func_reg(struct ql_adapter *qdev, - u32 reg, u32 reg_val) +static int qlge_write_other_func_reg(struct qlge_adapter *qdev, + u32 reg, u32 reg_val) { u32 register_to_read; @@ -35,17 +35,17 @@ static int ql_write_other_func_reg(struct ql_adapter *qdev, | (qdev->alt_func << MPI_NIC_FUNCTION_SHIFT) | reg; - return ql_write_mpi_reg(qdev, register_to_read, reg_val); + return qlge_write_mpi_reg(qdev, register_to_read, reg_val); } -static int ql_wait_other_func_reg_rdy(struct ql_adapter *qdev, u32 reg, - u32 bit, u32 err_bit) +static int qlge_wait_other_func_reg_rdy(struct qlge_adapter *qdev, u32 reg, + u32 bit, u32 err_bit) { u32 temp; int count; for (count = 10; count; count--) { - temp = ql_read_other_func_reg(qdev, reg); + temp = qlge_read_other_func_reg(qdev, reg); /* check for errors */ if (temp & err_bit) @@ -57,80 +57,80 @@ static int ql_wait_other_func_reg_rdy(struct ql_adapter *qdev, u32 reg, return -1; } -static int ql_read_other_func_serdes_reg(struct ql_adapter *qdev, u32 reg, - u32 *data) +static int qlge_read_other_func_serdes_reg(struct qlge_adapter *qdev, u32 reg, + u32 *data) { int status; /* wait for reg to come ready */ - status = ql_wait_other_func_reg_rdy(qdev, XG_SERDES_ADDR / 4, - XG_SERDES_ADDR_RDY, 0); + status = qlge_wait_other_func_reg_rdy(qdev, XG_SERDES_ADDR / 4, + XG_SERDES_ADDR_RDY, 0); if (status) goto exit; /* set up for reg read */ - ql_write_other_func_reg(qdev, XG_SERDES_ADDR / 4, reg | PROC_ADDR_R); + qlge_write_other_func_reg(qdev, XG_SERDES_ADDR / 4, reg | PROC_ADDR_R); /* wait for reg to come ready */ - status = ql_wait_other_func_reg_rdy(qdev, XG_SERDES_ADDR / 4, - XG_SERDES_ADDR_RDY, 0); + status = qlge_wait_other_func_reg_rdy(qdev, XG_SERDES_ADDR / 4, + XG_SERDES_ADDR_RDY, 0); if (status) goto exit; /* get the data */ - *data = ql_read_other_func_reg(qdev, (XG_SERDES_DATA / 4)); + *data = qlge_read_other_func_reg(qdev, (XG_SERDES_DATA / 4)); exit: return status; } /* Read out the SERDES registers */ -static int ql_read_serdes_reg(struct ql_adapter *qdev, u32 reg, u32 *data) +static int qlge_read_serdes_reg(struct qlge_adapter *qdev, u32 reg, u32 *data) { int status; /* wait for reg to come ready */ - status = ql_wait_reg_rdy(qdev, XG_SERDES_ADDR, XG_SERDES_ADDR_RDY, 0); + status = qlge_wait_reg_rdy(qdev, XG_SERDES_ADDR, XG_SERDES_ADDR_RDY, 0); if (status) goto exit; /* set up for reg read */ - ql_write32(qdev, XG_SERDES_ADDR, reg | PROC_ADDR_R); + qlge_write32(qdev, XG_SERDES_ADDR, reg | PROC_ADDR_R); /* wait for reg to come ready */ - status = ql_wait_reg_rdy(qdev, XG_SERDES_ADDR, XG_SERDES_ADDR_RDY, 0); + status = qlge_wait_reg_rdy(qdev, XG_SERDES_ADDR, XG_SERDES_ADDR_RDY, 0); if (status) goto exit; /* get the data */ - *data = ql_read32(qdev, XG_SERDES_DATA); + *data = qlge_read32(qdev, XG_SERDES_DATA); exit: return status; } -static void ql_get_both_serdes(struct ql_adapter *qdev, u32 addr, - u32 *direct_ptr, u32 *indirect_ptr, - bool direct_valid, bool indirect_valid) +static void qlge_get_both_serdes(struct qlge_adapter *qdev, u32 addr, + u32 *direct_ptr, u32 *indirect_ptr, + bool direct_valid, bool indirect_valid) { unsigned int status; status = 1; if (direct_valid) - status = ql_read_serdes_reg(qdev, addr, direct_ptr); + status = qlge_read_serdes_reg(qdev, addr, direct_ptr); /* Dead fill any failures or invalids. */ if (status) *direct_ptr = 0xDEADBEEF; status = 1; if (indirect_valid) - status = ql_read_other_func_serdes_reg( - qdev, addr, indirect_ptr); + status = qlge_read_other_func_serdes_reg(qdev, addr, + indirect_ptr); /* Dead fill any failures or invalids. */ if (status) *indirect_ptr = 0xDEADBEEF; } -static int ql_get_serdes_regs(struct ql_adapter *qdev, - struct ql_mpi_coredump *mpi_coredump) +static int qlge_get_serdes_regs(struct qlge_adapter *qdev, + struct qlge_mpi_coredump *mpi_coredump) { int status; bool xfi_direct_valid = false, xfi_indirect_valid = false; @@ -140,9 +140,9 @@ static int ql_get_serdes_regs(struct ql_adapter *qdev, u32 *indirect_ptr; /* The XAUI needs to be read out per port */ - status = ql_read_other_func_serdes_reg(qdev, - XG_SERDES_XAUI_HSS_PCS_START, - &temp); + status = qlge_read_other_func_serdes_reg(qdev, + XG_SERDES_XAUI_HSS_PCS_START, + &temp); if (status) temp = XG_SERDES_ADDR_XAUI_PWR_DOWN; @@ -150,7 +150,7 @@ static int ql_get_serdes_regs(struct ql_adapter *qdev, XG_SERDES_ADDR_XAUI_PWR_DOWN) xaui_indirect_valid = false; - status = ql_read_serdes_reg(qdev, XG_SERDES_XAUI_HSS_PCS_START, &temp); + status = qlge_read_serdes_reg(qdev, XG_SERDES_XAUI_HSS_PCS_START, &temp); if (status) temp = XG_SERDES_ADDR_XAUI_PWR_DOWN; @@ -163,7 +163,7 @@ static int ql_get_serdes_regs(struct ql_adapter *qdev, * XFI register is shared so only need to read one * functions and then check the bits. */ - status = ql_read_serdes_reg(qdev, XG_SERDES_ADDR_STS, &temp); + status = qlge_read_serdes_reg(qdev, XG_SERDES_ADDR_STS, &temp); if (status) temp = 0; @@ -198,8 +198,8 @@ static int ql_get_serdes_regs(struct ql_adapter *qdev, } for (i = 0; i <= 0x000000034; i += 4, direct_ptr++, indirect_ptr++) - ql_get_both_serdes(qdev, i, direct_ptr, indirect_ptr, - xaui_direct_valid, xaui_indirect_valid); + qlge_get_both_serdes(qdev, i, direct_ptr, indirect_ptr, + xaui_direct_valid, xaui_indirect_valid); /* Get XAUI_HSS_PCS register block. */ if (qdev->func & 1) { @@ -215,8 +215,8 @@ static int ql_get_serdes_regs(struct ql_adapter *qdev, } for (i = 0x800; i <= 0x880; i += 4, direct_ptr++, indirect_ptr++) - ql_get_both_serdes(qdev, i, direct_ptr, indirect_ptr, - xaui_direct_valid, xaui_indirect_valid); + qlge_get_both_serdes(qdev, i, direct_ptr, indirect_ptr, + xaui_direct_valid, xaui_indirect_valid); /* Get XAUI_XFI_AN register block. */ if (qdev->func & 1) { @@ -228,8 +228,8 @@ static int ql_get_serdes_regs(struct ql_adapter *qdev, } for (i = 0x1000; i <= 0x1034; i += 4, direct_ptr++, indirect_ptr++) - ql_get_both_serdes(qdev, i, direct_ptr, indirect_ptr, - xfi_direct_valid, xfi_indirect_valid); + qlge_get_both_serdes(qdev, i, direct_ptr, indirect_ptr, + xfi_direct_valid, xfi_indirect_valid); /* Get XAUI_XFI_TRAIN register block. */ if (qdev->func & 1) { @@ -243,8 +243,8 @@ static int ql_get_serdes_regs(struct ql_adapter *qdev, } for (i = 0x1050; i <= 0x107c; i += 4, direct_ptr++, indirect_ptr++) - ql_get_both_serdes(qdev, i, direct_ptr, indirect_ptr, - xfi_direct_valid, xfi_indirect_valid); + qlge_get_both_serdes(qdev, i, direct_ptr, indirect_ptr, + xfi_direct_valid, xfi_indirect_valid); /* Get XAUI_XFI_HSS_PCS register block. */ if (qdev->func & 1) { @@ -260,8 +260,8 @@ static int ql_get_serdes_regs(struct ql_adapter *qdev, } for (i = 0x1800; i <= 0x1838; i += 4, direct_ptr++, indirect_ptr++) - ql_get_both_serdes(qdev, i, direct_ptr, indirect_ptr, - xfi_direct_valid, xfi_indirect_valid); + qlge_get_both_serdes(qdev, i, direct_ptr, indirect_ptr, + xfi_direct_valid, xfi_indirect_valid); /* Get XAUI_XFI_HSS_TX register block. */ if (qdev->func & 1) { @@ -275,8 +275,8 @@ static int ql_get_serdes_regs(struct ql_adapter *qdev, mpi_coredump->serdes2_xfi_hss_tx; } for (i = 0x1c00; i <= 0x1c1f; i++, direct_ptr++, indirect_ptr++) - ql_get_both_serdes(qdev, i, direct_ptr, indirect_ptr, - xfi_direct_valid, xfi_indirect_valid); + qlge_get_both_serdes(qdev, i, direct_ptr, indirect_ptr, + xfi_direct_valid, xfi_indirect_valid); /* Get XAUI_XFI_HSS_RX register block. */ if (qdev->func & 1) { @@ -291,8 +291,8 @@ static int ql_get_serdes_regs(struct ql_adapter *qdev, } for (i = 0x1c40; i <= 0x1c5f; i++, direct_ptr++, indirect_ptr++) - ql_get_both_serdes(qdev, i, direct_ptr, indirect_ptr, - xfi_direct_valid, xfi_indirect_valid); + qlge_get_both_serdes(qdev, i, direct_ptr, indirect_ptr, + xfi_direct_valid, xfi_indirect_valid); /* Get XAUI_XFI_HSS_PLL register block. */ if (qdev->func & 1) { @@ -307,33 +307,33 @@ static int ql_get_serdes_regs(struct ql_adapter *qdev, mpi_coredump->serdes2_xfi_hss_pll; } for (i = 0x1e00; i <= 0x1e1f; i++, direct_ptr++, indirect_ptr++) - ql_get_both_serdes(qdev, i, direct_ptr, indirect_ptr, - xfi_direct_valid, xfi_indirect_valid); + qlge_get_both_serdes(qdev, i, direct_ptr, indirect_ptr, + xfi_direct_valid, xfi_indirect_valid); return 0; } -static int ql_read_other_func_xgmac_reg(struct ql_adapter *qdev, u32 reg, - u32 *data) +static int qlge_read_other_func_xgmac_reg(struct qlge_adapter *qdev, u32 reg, + u32 *data) { int status = 0; /* wait for reg to come ready */ - status = ql_wait_other_func_reg_rdy(qdev, XGMAC_ADDR / 4, - XGMAC_ADDR_RDY, XGMAC_ADDR_XME); + status = qlge_wait_other_func_reg_rdy(qdev, XGMAC_ADDR / 4, + XGMAC_ADDR_RDY, XGMAC_ADDR_XME); if (status) goto exit; /* set up for reg read */ - ql_write_other_func_reg(qdev, XGMAC_ADDR / 4, reg | XGMAC_ADDR_R); + qlge_write_other_func_reg(qdev, XGMAC_ADDR / 4, reg | XGMAC_ADDR_R); /* wait for reg to come ready */ - status = ql_wait_other_func_reg_rdy(qdev, XGMAC_ADDR / 4, - XGMAC_ADDR_RDY, XGMAC_ADDR_XME); + status = qlge_wait_other_func_reg_rdy(qdev, XGMAC_ADDR / 4, + XGMAC_ADDR_RDY, XGMAC_ADDR_XME); if (status) goto exit; /* get the data */ - *data = ql_read_other_func_reg(qdev, XGMAC_DATA / 4); + *data = qlge_read_other_func_reg(qdev, XGMAC_DATA / 4); exit: return status; } @@ -341,8 +341,8 @@ static int ql_read_other_func_xgmac_reg(struct ql_adapter *qdev, u32 reg, /* Read the 400 xgmac control/statistics registers * skipping unused locations. */ -static int ql_get_xgmac_regs(struct ql_adapter *qdev, u32 *buf, - unsigned int other_function) +static int qlge_get_xgmac_regs(struct qlge_adapter *qdev, u32 *buf, + unsigned int other_function) { int status = 0; int i; @@ -370,9 +370,9 @@ static int ql_get_xgmac_regs(struct ql_adapter *qdev, u32 *buf, (i > 0x000005c8 && i < 0x00000600)) { if (other_function) status = - ql_read_other_func_xgmac_reg(qdev, i, buf); + qlge_read_other_func_xgmac_reg(qdev, i, buf); else - status = ql_read_xgmac_reg(qdev, i, buf); + status = qlge_read_xgmac_reg(qdev, i, buf); if (status) *buf = 0xdeadbeef; @@ -382,46 +382,46 @@ static int ql_get_xgmac_regs(struct ql_adapter *qdev, u32 *buf, return status; } -static int ql_get_ets_regs(struct ql_adapter *qdev, u32 *buf) +static int qlge_get_ets_regs(struct qlge_adapter *qdev, u32 *buf) { int i; for (i = 0; i < 8; i++, buf++) { - ql_write32(qdev, NIC_ETS, i << 29 | 0x08000000); - *buf = ql_read32(qdev, NIC_ETS); + qlge_write32(qdev, NIC_ETS, i << 29 | 0x08000000); + *buf = qlge_read32(qdev, NIC_ETS); } for (i = 0; i < 2; i++, buf++) { - ql_write32(qdev, CNA_ETS, i << 29 | 0x08000000); - *buf = ql_read32(qdev, CNA_ETS); + qlge_write32(qdev, CNA_ETS, i << 29 | 0x08000000); + *buf = qlge_read32(qdev, CNA_ETS); } return 0; } -static void ql_get_intr_states(struct ql_adapter *qdev, u32 *buf) +static void qlge_get_intr_states(struct qlge_adapter *qdev, u32 *buf) { int i; for (i = 0; i < qdev->rx_ring_count; i++, buf++) { - ql_write32(qdev, INTR_EN, - qdev->intr_context[i].intr_read_mask); - *buf = ql_read32(qdev, INTR_EN); + qlge_write32(qdev, INTR_EN, + qdev->intr_context[i].intr_read_mask); + *buf = qlge_read32(qdev, INTR_EN); } } -static int ql_get_cam_entries(struct ql_adapter *qdev, u32 *buf) +static int qlge_get_cam_entries(struct qlge_adapter *qdev, u32 *buf) { int i, status; u32 value[3]; - status = ql_sem_spinlock(qdev, SEM_MAC_ADDR_MASK); + status = qlge_sem_spinlock(qdev, SEM_MAC_ADDR_MASK); if (status) return status; for (i = 0; i < 16; i++) { - status = ql_get_mac_addr_reg(qdev, - MAC_ADDR_TYPE_CAM_MAC, i, value); + status = qlge_get_mac_addr_reg(qdev, + MAC_ADDR_TYPE_CAM_MAC, i, value); if (status) { netif_err(qdev, drv, qdev->ndev, "Failed read of mac index register\n"); @@ -432,8 +432,8 @@ static int ql_get_cam_entries(struct ql_adapter *qdev, u32 *buf) *buf++ = value[2]; /* output */ } for (i = 0; i < 32; i++) { - status = ql_get_mac_addr_reg(qdev, - MAC_ADDR_TYPE_MULTI_MAC, i, value); + status = qlge_get_mac_addr_reg(qdev, MAC_ADDR_TYPE_MULTI_MAC, + i, value); if (status) { netif_err(qdev, drv, qdev->ndev, "Failed read of mac index register\n"); @@ -443,21 +443,21 @@ static int ql_get_cam_entries(struct ql_adapter *qdev, u32 *buf) *buf++ = value[1]; /* upper Mcast address */ } err: - ql_sem_unlock(qdev, SEM_MAC_ADDR_MASK); + qlge_sem_unlock(qdev, SEM_MAC_ADDR_MASK); return status; } -static int ql_get_routing_entries(struct ql_adapter *qdev, u32 *buf) +static int qlge_get_routing_entries(struct qlge_adapter *qdev, u32 *buf) { int status; u32 value, i; - status = ql_sem_spinlock(qdev, SEM_RT_IDX_MASK); + status = qlge_sem_spinlock(qdev, SEM_RT_IDX_MASK); if (status) return status; for (i = 0; i < 16; i++) { - status = ql_get_routing_reg(qdev, i, &value); + status = qlge_get_routing_reg(qdev, i, &value); if (status) { netif_err(qdev, drv, qdev->ndev, "Failed read of routing index register\n"); @@ -467,23 +467,23 @@ static int ql_get_routing_entries(struct ql_adapter *qdev, u32 *buf) } } err: - ql_sem_unlock(qdev, SEM_RT_IDX_MASK); + qlge_sem_unlock(qdev, SEM_RT_IDX_MASK); return status; } /* Read the MPI Processor shadow registers */ -static int ql_get_mpi_shadow_regs(struct ql_adapter *qdev, u32 *buf) +static int qlge_get_mpi_shadow_regs(struct qlge_adapter *qdev, u32 *buf) { u32 i; int status; for (i = 0; i < MPI_CORE_SH_REGS_CNT; i++, buf++) { - status = ql_write_mpi_reg(qdev, - RISC_124, - (SHADOW_OFFSET | i << SHADOW_REG_SHIFT)); + status = qlge_write_mpi_reg(qdev, + RISC_124, + (SHADOW_OFFSET | i << SHADOW_REG_SHIFT)); if (status) goto end; - status = ql_read_mpi_reg(qdev, RISC_127, buf); + status = qlge_read_mpi_reg(qdev, RISC_127, buf); if (status) goto end; } @@ -492,13 +492,13 @@ static int ql_get_mpi_shadow_regs(struct ql_adapter *qdev, u32 *buf) } /* Read the MPI Processor core registers */ -static int ql_get_mpi_regs(struct ql_adapter *qdev, u32 *buf, - u32 offset, u32 count) +static int qlge_get_mpi_regs(struct qlge_adapter *qdev, u32 *buf, + u32 offset, u32 count) { int i, status = 0; for (i = 0; i < count; i++, buf++) { - status = ql_read_mpi_reg(qdev, offset + i, buf); + status = qlge_read_mpi_reg(qdev, offset + i, buf); if (status) return status; } @@ -506,8 +506,8 @@ static int ql_get_mpi_regs(struct ql_adapter *qdev, u32 *buf, } /* Read the ASIC probe dump */ -static unsigned int *ql_get_probe(struct ql_adapter *qdev, u32 clock, - u32 valid, u32 *buf) +static unsigned int *qlge_get_probe(struct qlge_adapter *qdev, u32 clock, + u32 valid, u32 *buf) { u32 module, mux_sel, probe, lo_val, hi_val; @@ -519,15 +519,15 @@ static unsigned int *ql_get_probe(struct ql_adapter *qdev, u32 clock, | PRB_MX_ADDR_ARE | mux_sel | (module << PRB_MX_ADDR_MOD_SEL_SHIFT); - ql_write32(qdev, PRB_MX_ADDR, probe); - lo_val = ql_read32(qdev, PRB_MX_DATA); + qlge_write32(qdev, PRB_MX_ADDR, probe); + lo_val = qlge_read32(qdev, PRB_MX_DATA); if (mux_sel == 0) { *buf = probe; buf++; } probe |= PRB_MX_ADDR_UP; - ql_write32(qdev, PRB_MX_ADDR, probe); - hi_val = ql_read32(qdev, PRB_MX_DATA); + qlge_write32(qdev, PRB_MX_ADDR, probe); + hi_val = qlge_read32(qdev, PRB_MX_DATA); *buf = lo_val; buf++; *buf = hi_val; @@ -537,23 +537,23 @@ static unsigned int *ql_get_probe(struct ql_adapter *qdev, u32 clock, return buf; } -static int ql_get_probe_dump(struct ql_adapter *qdev, unsigned int *buf) +static int qlge_get_probe_dump(struct qlge_adapter *qdev, unsigned int *buf) { /* First we have to enable the probe mux */ - ql_write_mpi_reg(qdev, MPI_TEST_FUNC_PRB_CTL, MPI_TEST_FUNC_PRB_EN); - buf = ql_get_probe(qdev, PRB_MX_ADDR_SYS_CLOCK, - PRB_MX_ADDR_VALID_SYS_MOD, buf); - buf = ql_get_probe(qdev, PRB_MX_ADDR_PCI_CLOCK, - PRB_MX_ADDR_VALID_PCI_MOD, buf); - buf = ql_get_probe(qdev, PRB_MX_ADDR_XGM_CLOCK, - PRB_MX_ADDR_VALID_XGM_MOD, buf); - buf = ql_get_probe(qdev, PRB_MX_ADDR_FC_CLOCK, - PRB_MX_ADDR_VALID_FC_MOD, buf); + qlge_write_mpi_reg(qdev, MPI_TEST_FUNC_PRB_CTL, MPI_TEST_FUNC_PRB_EN); + buf = qlge_get_probe(qdev, PRB_MX_ADDR_SYS_CLOCK, + PRB_MX_ADDR_VALID_SYS_MOD, buf); + buf = qlge_get_probe(qdev, PRB_MX_ADDR_PCI_CLOCK, + PRB_MX_ADDR_VALID_PCI_MOD, buf); + buf = qlge_get_probe(qdev, PRB_MX_ADDR_XGM_CLOCK, + PRB_MX_ADDR_VALID_XGM_MOD, buf); + buf = qlge_get_probe(qdev, PRB_MX_ADDR_FC_CLOCK, + PRB_MX_ADDR_VALID_FC_MOD, buf); return 0; } /* Read out the routing index registers */ -static int ql_get_routing_index_registers(struct ql_adapter *qdev, u32 *buf) +static int qlge_get_routing_index_registers(struct qlge_adapter *qdev, u32 *buf) { int status; u32 type, index, index_max; @@ -561,7 +561,7 @@ static int ql_get_routing_index_registers(struct ql_adapter *qdev, u32 *buf) u32 result_data; u32 val; - status = ql_sem_spinlock(qdev, SEM_RT_IDX_MASK); + status = qlge_sem_spinlock(qdev, SEM_RT_IDX_MASK); if (status) return status; @@ -574,11 +574,11 @@ static int ql_get_routing_index_registers(struct ql_adapter *qdev, u32 *buf) val = RT_IDX_RS | (type << RT_IDX_TYPE_SHIFT) | (index << RT_IDX_IDX_SHIFT); - ql_write32(qdev, RT_IDX, val); + qlge_write32(qdev, RT_IDX, val); result_index = 0; while ((result_index & RT_IDX_MR) == 0) - result_index = ql_read32(qdev, RT_IDX); - result_data = ql_read32(qdev, RT_DATA); + result_index = qlge_read32(qdev, RT_IDX); + result_data = qlge_read32(qdev, RT_DATA); *buf = type; buf++; *buf = index; @@ -589,12 +589,12 @@ static int ql_get_routing_index_registers(struct ql_adapter *qdev, u32 *buf) buf++; } } - ql_sem_unlock(qdev, SEM_RT_IDX_MASK); + qlge_sem_unlock(qdev, SEM_RT_IDX_MASK); return status; } /* Read out the MAC protocol registers */ -static void ql_get_mac_protocol_registers(struct ql_adapter *qdev, u32 *buf) +static void qlge_get_mac_protocol_registers(struct qlge_adapter *qdev, u32 *buf) { u32 result_index, result_data; u32 type; @@ -657,13 +657,13 @@ static void ql_get_mac_protocol_registers(struct ql_adapter *qdev, u32 *buf) | (type << MAC_ADDR_TYPE_SHIFT) | (index << MAC_ADDR_IDX_SHIFT) | (offset); - ql_write32(qdev, MAC_ADDR_IDX, val); + qlge_write32(qdev, MAC_ADDR_IDX, val); result_index = 0; while ((result_index & MAC_ADDR_MR) == 0) { - result_index = ql_read32(qdev, - MAC_ADDR_IDX); + result_index = qlge_read32(qdev, + MAC_ADDR_IDX); } - result_data = ql_read32(qdev, MAC_ADDR_DATA); + result_data = qlge_read32(qdev, MAC_ADDR_DATA); *buf = result_index; buf++; *buf = result_data; @@ -673,7 +673,7 @@ static void ql_get_mac_protocol_registers(struct ql_adapter *qdev, u32 *buf) } } -static void ql_get_sem_registers(struct ql_adapter *qdev, u32 *buf) +static void qlge_get_sem_registers(struct qlge_adapter *qdev, u32 *buf) { u32 func_num, reg, reg_val; int status; @@ -682,7 +682,7 @@ static void ql_get_sem_registers(struct ql_adapter *qdev, u32 *buf) reg = MPI_NIC_REG_BLOCK | (func_num << MPI_NIC_FUNCTION_SHIFT) | (SEM / 4); - status = ql_read_mpi_reg(qdev, reg, ®_val); + status = qlge_read_mpi_reg(qdev, reg, ®_val); *buf = reg_val; /* if the read failed then dead fill the element. */ if (!status) @@ -692,9 +692,8 @@ static void ql_get_sem_registers(struct ql_adapter *qdev, u32 *buf) } /* Create a coredump segment header */ -static void ql_build_coredump_seg_header( - struct mpi_coredump_segment_header *seg_hdr, - u32 seg_number, u32 seg_size, u8 *desc) +static void qlge_build_coredump_seg_header(struct mpi_coredump_segment_header *seg_hdr, + u32 seg_number, u32 seg_size, u8 *desc) { memset(seg_hdr, 0, sizeof(struct mpi_coredump_segment_header)); seg_hdr->cookie = MPI_COREDUMP_COOKIE; @@ -710,7 +709,7 @@ static void ql_build_coredump_seg_header( * space for this function as well as a coredump structure that * will contain the dump. */ -int ql_core_dump(struct ql_adapter *qdev, struct ql_mpi_coredump *mpi_coredump) +int qlge_core_dump(struct qlge_adapter *qdev, struct qlge_mpi_coredump *mpi_coredump) { int status; int i; @@ -724,9 +723,9 @@ int ql_core_dump(struct ql_adapter *qdev, struct ql_mpi_coredump *mpi_coredump) * it isn't available. If the firmware died it * might be holding the sem. */ - ql_sem_spinlock(qdev, SEM_PROC_REG_MASK); + qlge_sem_spinlock(qdev, SEM_PROC_REG_MASK); - status = ql_pause_mpi_risc(qdev); + status = qlge_pause_mpi_risc(qdev); if (status) { netif_err(qdev, drv, qdev->ndev, "Failed RISC pause. Status = 0x%.08x\n", status); @@ -740,155 +739,155 @@ int ql_core_dump(struct ql_adapter *qdev, struct ql_mpi_coredump *mpi_coredump) mpi_coredump->mpi_global_header.header_size = sizeof(struct mpi_coredump_global_header); mpi_coredump->mpi_global_header.image_size = - sizeof(struct ql_mpi_coredump); + sizeof(struct qlge_mpi_coredump); strncpy(mpi_coredump->mpi_global_header.id_string, "MPI Coredump", sizeof(mpi_coredump->mpi_global_header.id_string)); /* Get generic NIC reg dump */ - ql_build_coredump_seg_header(&mpi_coredump->nic_regs_seg_hdr, - NIC1_CONTROL_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->nic_regs), "NIC1 Registers"); + qlge_build_coredump_seg_header(&mpi_coredump->nic_regs_seg_hdr, + NIC1_CONTROL_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->nic_regs), "NIC1 Registers"); - ql_build_coredump_seg_header(&mpi_coredump->nic2_regs_seg_hdr, - NIC2_CONTROL_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->nic2_regs), "NIC2 Registers"); + qlge_build_coredump_seg_header(&mpi_coredump->nic2_regs_seg_hdr, + NIC2_CONTROL_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->nic2_regs), "NIC2 Registers"); /* Get XGMac registers. (Segment 18, Rev C. step 21) */ - ql_build_coredump_seg_header(&mpi_coredump->xgmac1_seg_hdr, - NIC1_XGMAC_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->xgmac1), "NIC1 XGMac Registers"); + qlge_build_coredump_seg_header(&mpi_coredump->xgmac1_seg_hdr, + NIC1_XGMAC_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->xgmac1), "NIC1 XGMac Registers"); - ql_build_coredump_seg_header(&mpi_coredump->xgmac2_seg_hdr, - NIC2_XGMAC_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->xgmac2), "NIC2 XGMac Registers"); + qlge_build_coredump_seg_header(&mpi_coredump->xgmac2_seg_hdr, + NIC2_XGMAC_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->xgmac2), "NIC2 XGMac Registers"); if (qdev->func & 1) { /* Odd means our function is NIC 2 */ for (i = 0; i < NIC_REGS_DUMP_WORD_COUNT; i++) mpi_coredump->nic2_regs[i] = - ql_read32(qdev, i * sizeof(u32)); + qlge_read32(qdev, i * sizeof(u32)); for (i = 0; i < NIC_REGS_DUMP_WORD_COUNT; i++) mpi_coredump->nic_regs[i] = - ql_read_other_func_reg(qdev, (i * sizeof(u32)) / 4); + qlge_read_other_func_reg(qdev, (i * sizeof(u32)) / 4); - ql_get_xgmac_regs(qdev, &mpi_coredump->xgmac2[0], 0); - ql_get_xgmac_regs(qdev, &mpi_coredump->xgmac1[0], 1); + qlge_get_xgmac_regs(qdev, &mpi_coredump->xgmac2[0], 0); + qlge_get_xgmac_regs(qdev, &mpi_coredump->xgmac1[0], 1); } else { /* Even means our function is NIC 1 */ for (i = 0; i < NIC_REGS_DUMP_WORD_COUNT; i++) mpi_coredump->nic_regs[i] = - ql_read32(qdev, i * sizeof(u32)); + qlge_read32(qdev, i * sizeof(u32)); for (i = 0; i < NIC_REGS_DUMP_WORD_COUNT; i++) mpi_coredump->nic2_regs[i] = - ql_read_other_func_reg(qdev, (i * sizeof(u32)) / 4); + qlge_read_other_func_reg(qdev, (i * sizeof(u32)) / 4); - ql_get_xgmac_regs(qdev, &mpi_coredump->xgmac1[0], 0); - ql_get_xgmac_regs(qdev, &mpi_coredump->xgmac2[0], 1); + qlge_get_xgmac_regs(qdev, &mpi_coredump->xgmac1[0], 0); + qlge_get_xgmac_regs(qdev, &mpi_coredump->xgmac2[0], 1); } /* Rev C. Step 20a */ - ql_build_coredump_seg_header(&mpi_coredump->xaui_an_hdr, - XAUI_AN_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->serdes_xaui_an), - "XAUI AN Registers"); + qlge_build_coredump_seg_header(&mpi_coredump->xaui_an_hdr, + XAUI_AN_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->serdes_xaui_an), + "XAUI AN Registers"); /* Rev C. Step 20b */ - ql_build_coredump_seg_header(&mpi_coredump->xaui_hss_pcs_hdr, - XAUI_HSS_PCS_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->serdes_xaui_hss_pcs), - "XAUI HSS PCS Registers"); - - ql_build_coredump_seg_header(&mpi_coredump->xfi_an_hdr, XFI_AN_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->serdes_xfi_an), - "XFI AN Registers"); - - ql_build_coredump_seg_header(&mpi_coredump->xfi_train_hdr, - XFI_TRAIN_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->serdes_xfi_train), - "XFI TRAIN Registers"); - - ql_build_coredump_seg_header(&mpi_coredump->xfi_hss_pcs_hdr, - XFI_HSS_PCS_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->serdes_xfi_hss_pcs), - "XFI HSS PCS Registers"); - - ql_build_coredump_seg_header(&mpi_coredump->xfi_hss_tx_hdr, - XFI_HSS_TX_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->serdes_xfi_hss_tx), - "XFI HSS TX Registers"); - - ql_build_coredump_seg_header(&mpi_coredump->xfi_hss_rx_hdr, - XFI_HSS_RX_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->serdes_xfi_hss_rx), - "XFI HSS RX Registers"); - - ql_build_coredump_seg_header(&mpi_coredump->xfi_hss_pll_hdr, - XFI_HSS_PLL_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->serdes_xfi_hss_pll), - "XFI HSS PLL Registers"); - - ql_build_coredump_seg_header(&mpi_coredump->xaui2_an_hdr, - XAUI2_AN_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->serdes2_xaui_an), - "XAUI2 AN Registers"); - - ql_build_coredump_seg_header(&mpi_coredump->xaui2_hss_pcs_hdr, - XAUI2_HSS_PCS_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->serdes2_xaui_hss_pcs), - "XAUI2 HSS PCS Registers"); - - ql_build_coredump_seg_header(&mpi_coredump->xfi2_an_hdr, - XFI2_AN_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->serdes2_xfi_an), - "XFI2 AN Registers"); - - ql_build_coredump_seg_header(&mpi_coredump->xfi2_train_hdr, - XFI2_TRAIN_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->serdes2_xfi_train), - "XFI2 TRAIN Registers"); - - ql_build_coredump_seg_header(&mpi_coredump->xfi2_hss_pcs_hdr, - XFI2_HSS_PCS_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->serdes2_xfi_hss_pcs), - "XFI2 HSS PCS Registers"); - - ql_build_coredump_seg_header(&mpi_coredump->xfi2_hss_tx_hdr, - XFI2_HSS_TX_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->serdes2_xfi_hss_tx), - "XFI2 HSS TX Registers"); - - ql_build_coredump_seg_header(&mpi_coredump->xfi2_hss_rx_hdr, - XFI2_HSS_RX_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->serdes2_xfi_hss_rx), - "XFI2 HSS RX Registers"); - - ql_build_coredump_seg_header(&mpi_coredump->xfi2_hss_pll_hdr, - XFI2_HSS_PLL_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->serdes2_xfi_hss_pll), - "XFI2 HSS PLL Registers"); - - status = ql_get_serdes_regs(qdev, mpi_coredump); + qlge_build_coredump_seg_header(&mpi_coredump->xaui_hss_pcs_hdr, + XAUI_HSS_PCS_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->serdes_xaui_hss_pcs), + "XAUI HSS PCS Registers"); + + qlge_build_coredump_seg_header(&mpi_coredump->xfi_an_hdr, XFI_AN_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->serdes_xfi_an), + "XFI AN Registers"); + + qlge_build_coredump_seg_header(&mpi_coredump->xfi_train_hdr, + XFI_TRAIN_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->serdes_xfi_train), + "XFI TRAIN Registers"); + + qlge_build_coredump_seg_header(&mpi_coredump->xfi_hss_pcs_hdr, + XFI_HSS_PCS_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->serdes_xfi_hss_pcs), + "XFI HSS PCS Registers"); + + qlge_build_coredump_seg_header(&mpi_coredump->xfi_hss_tx_hdr, + XFI_HSS_TX_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->serdes_xfi_hss_tx), + "XFI HSS TX Registers"); + + qlge_build_coredump_seg_header(&mpi_coredump->xfi_hss_rx_hdr, + XFI_HSS_RX_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->serdes_xfi_hss_rx), + "XFI HSS RX Registers"); + + qlge_build_coredump_seg_header(&mpi_coredump->xfi_hss_pll_hdr, + XFI_HSS_PLL_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->serdes_xfi_hss_pll), + "XFI HSS PLL Registers"); + + qlge_build_coredump_seg_header(&mpi_coredump->xaui2_an_hdr, + XAUI2_AN_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->serdes2_xaui_an), + "XAUI2 AN Registers"); + + qlge_build_coredump_seg_header(&mpi_coredump->xaui2_hss_pcs_hdr, + XAUI2_HSS_PCS_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->serdes2_xaui_hss_pcs), + "XAUI2 HSS PCS Registers"); + + qlge_build_coredump_seg_header(&mpi_coredump->xfi2_an_hdr, + XFI2_AN_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->serdes2_xfi_an), + "XFI2 AN Registers"); + + qlge_build_coredump_seg_header(&mpi_coredump->xfi2_train_hdr, + XFI2_TRAIN_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->serdes2_xfi_train), + "XFI2 TRAIN Registers"); + + qlge_build_coredump_seg_header(&mpi_coredump->xfi2_hss_pcs_hdr, + XFI2_HSS_PCS_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->serdes2_xfi_hss_pcs), + "XFI2 HSS PCS Registers"); + + qlge_build_coredump_seg_header(&mpi_coredump->xfi2_hss_tx_hdr, + XFI2_HSS_TX_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->serdes2_xfi_hss_tx), + "XFI2 HSS TX Registers"); + + qlge_build_coredump_seg_header(&mpi_coredump->xfi2_hss_rx_hdr, + XFI2_HSS_RX_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->serdes2_xfi_hss_rx), + "XFI2 HSS RX Registers"); + + qlge_build_coredump_seg_header(&mpi_coredump->xfi2_hss_pll_hdr, + XFI2_HSS_PLL_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->serdes2_xfi_hss_pll), + "XFI2 HSS PLL Registers"); + + status = qlge_get_serdes_regs(qdev, mpi_coredump); if (status) { netif_err(qdev, drv, qdev->ndev, "Failed Dump of Serdes Registers. Status = 0x%.08x\n", @@ -896,185 +895,185 @@ int ql_core_dump(struct ql_adapter *qdev, struct ql_mpi_coredump *mpi_coredump) goto err; } - ql_build_coredump_seg_header(&mpi_coredump->core_regs_seg_hdr, - CORE_SEG_NUM, - sizeof(mpi_coredump->core_regs_seg_hdr) + - sizeof(mpi_coredump->mpi_core_regs) + - sizeof(mpi_coredump->mpi_core_sh_regs), - "Core Registers"); + qlge_build_coredump_seg_header(&mpi_coredump->core_regs_seg_hdr, + CORE_SEG_NUM, + sizeof(mpi_coredump->core_regs_seg_hdr) + + sizeof(mpi_coredump->mpi_core_regs) + + sizeof(mpi_coredump->mpi_core_sh_regs), + "Core Registers"); /* Get the MPI Core Registers */ - status = ql_get_mpi_regs(qdev, &mpi_coredump->mpi_core_regs[0], - MPI_CORE_REGS_ADDR, MPI_CORE_REGS_CNT); + status = qlge_get_mpi_regs(qdev, &mpi_coredump->mpi_core_regs[0], + MPI_CORE_REGS_ADDR, MPI_CORE_REGS_CNT); if (status) goto err; /* Get the 16 MPI shadow registers */ - status = ql_get_mpi_shadow_regs(qdev, - &mpi_coredump->mpi_core_sh_regs[0]); + status = qlge_get_mpi_shadow_regs(qdev, + &mpi_coredump->mpi_core_sh_regs[0]); if (status) goto err; /* Get the Test Logic Registers */ - ql_build_coredump_seg_header(&mpi_coredump->test_logic_regs_seg_hdr, - TEST_LOGIC_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->test_logic_regs), - "Test Logic Regs"); - status = ql_get_mpi_regs(qdev, &mpi_coredump->test_logic_regs[0], - TEST_REGS_ADDR, TEST_REGS_CNT); + qlge_build_coredump_seg_header(&mpi_coredump->test_logic_regs_seg_hdr, + TEST_LOGIC_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->test_logic_regs), + "Test Logic Regs"); + status = qlge_get_mpi_regs(qdev, &mpi_coredump->test_logic_regs[0], + TEST_REGS_ADDR, TEST_REGS_CNT); if (status) goto err; /* Get the RMII Registers */ - ql_build_coredump_seg_header(&mpi_coredump->rmii_regs_seg_hdr, - RMII_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->rmii_regs), - "RMII Registers"); - status = ql_get_mpi_regs(qdev, &mpi_coredump->rmii_regs[0], - RMII_REGS_ADDR, RMII_REGS_CNT); + qlge_build_coredump_seg_header(&mpi_coredump->rmii_regs_seg_hdr, + RMII_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->rmii_regs), + "RMII Registers"); + status = qlge_get_mpi_regs(qdev, &mpi_coredump->rmii_regs[0], + RMII_REGS_ADDR, RMII_REGS_CNT); if (status) goto err; /* Get the FCMAC1 Registers */ - ql_build_coredump_seg_header(&mpi_coredump->fcmac1_regs_seg_hdr, - FCMAC1_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->fcmac1_regs), - "FCMAC1 Registers"); - status = ql_get_mpi_regs(qdev, &mpi_coredump->fcmac1_regs[0], - FCMAC1_REGS_ADDR, FCMAC_REGS_CNT); + qlge_build_coredump_seg_header(&mpi_coredump->fcmac1_regs_seg_hdr, + FCMAC1_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->fcmac1_regs), + "FCMAC1 Registers"); + status = qlge_get_mpi_regs(qdev, &mpi_coredump->fcmac1_regs[0], + FCMAC1_REGS_ADDR, FCMAC_REGS_CNT); if (status) goto err; /* Get the FCMAC2 Registers */ - ql_build_coredump_seg_header(&mpi_coredump->fcmac2_regs_seg_hdr, - FCMAC2_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->fcmac2_regs), - "FCMAC2 Registers"); + qlge_build_coredump_seg_header(&mpi_coredump->fcmac2_regs_seg_hdr, + FCMAC2_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->fcmac2_regs), + "FCMAC2 Registers"); - status = ql_get_mpi_regs(qdev, &mpi_coredump->fcmac2_regs[0], - FCMAC2_REGS_ADDR, FCMAC_REGS_CNT); + status = qlge_get_mpi_regs(qdev, &mpi_coredump->fcmac2_regs[0], + FCMAC2_REGS_ADDR, FCMAC_REGS_CNT); if (status) goto err; /* Get the FC1 MBX Registers */ - ql_build_coredump_seg_header(&mpi_coredump->fc1_mbx_regs_seg_hdr, - FC1_MBOX_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->fc1_mbx_regs), - "FC1 MBox Regs"); - status = ql_get_mpi_regs(qdev, &mpi_coredump->fc1_mbx_regs[0], - FC1_MBX_REGS_ADDR, FC_MBX_REGS_CNT); + qlge_build_coredump_seg_header(&mpi_coredump->fc1_mbx_regs_seg_hdr, + FC1_MBOX_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->fc1_mbx_regs), + "FC1 MBox Regs"); + status = qlge_get_mpi_regs(qdev, &mpi_coredump->fc1_mbx_regs[0], + FC1_MBX_REGS_ADDR, FC_MBX_REGS_CNT); if (status) goto err; /* Get the IDE Registers */ - ql_build_coredump_seg_header(&mpi_coredump->ide_regs_seg_hdr, - IDE_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->ide_regs), - "IDE Registers"); - status = ql_get_mpi_regs(qdev, &mpi_coredump->ide_regs[0], - IDE_REGS_ADDR, IDE_REGS_CNT); + qlge_build_coredump_seg_header(&mpi_coredump->ide_regs_seg_hdr, + IDE_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->ide_regs), + "IDE Registers"); + status = qlge_get_mpi_regs(qdev, &mpi_coredump->ide_regs[0], + IDE_REGS_ADDR, IDE_REGS_CNT); if (status) goto err; /* Get the NIC1 MBX Registers */ - ql_build_coredump_seg_header(&mpi_coredump->nic1_mbx_regs_seg_hdr, - NIC1_MBOX_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->nic1_mbx_regs), - "NIC1 MBox Regs"); - status = ql_get_mpi_regs(qdev, &mpi_coredump->nic1_mbx_regs[0], - NIC1_MBX_REGS_ADDR, NIC_MBX_REGS_CNT); + qlge_build_coredump_seg_header(&mpi_coredump->nic1_mbx_regs_seg_hdr, + NIC1_MBOX_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->nic1_mbx_regs), + "NIC1 MBox Regs"); + status = qlge_get_mpi_regs(qdev, &mpi_coredump->nic1_mbx_regs[0], + NIC1_MBX_REGS_ADDR, NIC_MBX_REGS_CNT); if (status) goto err; /* Get the SMBus Registers */ - ql_build_coredump_seg_header(&mpi_coredump->smbus_regs_seg_hdr, - SMBUS_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->smbus_regs), - "SMBus Registers"); - status = ql_get_mpi_regs(qdev, &mpi_coredump->smbus_regs[0], - SMBUS_REGS_ADDR, SMBUS_REGS_CNT); + qlge_build_coredump_seg_header(&mpi_coredump->smbus_regs_seg_hdr, + SMBUS_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->smbus_regs), + "SMBus Registers"); + status = qlge_get_mpi_regs(qdev, &mpi_coredump->smbus_regs[0], + SMBUS_REGS_ADDR, SMBUS_REGS_CNT); if (status) goto err; /* Get the FC2 MBX Registers */ - ql_build_coredump_seg_header(&mpi_coredump->fc2_mbx_regs_seg_hdr, - FC2_MBOX_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->fc2_mbx_regs), - "FC2 MBox Regs"); - status = ql_get_mpi_regs(qdev, &mpi_coredump->fc2_mbx_regs[0], - FC2_MBX_REGS_ADDR, FC_MBX_REGS_CNT); + qlge_build_coredump_seg_header(&mpi_coredump->fc2_mbx_regs_seg_hdr, + FC2_MBOX_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->fc2_mbx_regs), + "FC2 MBox Regs"); + status = qlge_get_mpi_regs(qdev, &mpi_coredump->fc2_mbx_regs[0], + FC2_MBX_REGS_ADDR, FC_MBX_REGS_CNT); if (status) goto err; /* Get the NIC2 MBX Registers */ - ql_build_coredump_seg_header(&mpi_coredump->nic2_mbx_regs_seg_hdr, - NIC2_MBOX_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->nic2_mbx_regs), - "NIC2 MBox Regs"); - status = ql_get_mpi_regs(qdev, &mpi_coredump->nic2_mbx_regs[0], - NIC2_MBX_REGS_ADDR, NIC_MBX_REGS_CNT); + qlge_build_coredump_seg_header(&mpi_coredump->nic2_mbx_regs_seg_hdr, + NIC2_MBOX_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->nic2_mbx_regs), + "NIC2 MBox Regs"); + status = qlge_get_mpi_regs(qdev, &mpi_coredump->nic2_mbx_regs[0], + NIC2_MBX_REGS_ADDR, NIC_MBX_REGS_CNT); if (status) goto err; /* Get the I2C Registers */ - ql_build_coredump_seg_header(&mpi_coredump->i2c_regs_seg_hdr, - I2C_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->i2c_regs), - "I2C Registers"); - status = ql_get_mpi_regs(qdev, &mpi_coredump->i2c_regs[0], - I2C_REGS_ADDR, I2C_REGS_CNT); + qlge_build_coredump_seg_header(&mpi_coredump->i2c_regs_seg_hdr, + I2C_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->i2c_regs), + "I2C Registers"); + status = qlge_get_mpi_regs(qdev, &mpi_coredump->i2c_regs[0], + I2C_REGS_ADDR, I2C_REGS_CNT); if (status) goto err; /* Get the MEMC Registers */ - ql_build_coredump_seg_header(&mpi_coredump->memc_regs_seg_hdr, - MEMC_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->memc_regs), - "MEMC Registers"); - status = ql_get_mpi_regs(qdev, &mpi_coredump->memc_regs[0], - MEMC_REGS_ADDR, MEMC_REGS_CNT); + qlge_build_coredump_seg_header(&mpi_coredump->memc_regs_seg_hdr, + MEMC_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->memc_regs), + "MEMC Registers"); + status = qlge_get_mpi_regs(qdev, &mpi_coredump->memc_regs[0], + MEMC_REGS_ADDR, MEMC_REGS_CNT); if (status) goto err; /* Get the PBus Registers */ - ql_build_coredump_seg_header(&mpi_coredump->pbus_regs_seg_hdr, - PBUS_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->pbus_regs), - "PBUS Registers"); - status = ql_get_mpi_regs(qdev, &mpi_coredump->pbus_regs[0], - PBUS_REGS_ADDR, PBUS_REGS_CNT); + qlge_build_coredump_seg_header(&mpi_coredump->pbus_regs_seg_hdr, + PBUS_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->pbus_regs), + "PBUS Registers"); + status = qlge_get_mpi_regs(qdev, &mpi_coredump->pbus_regs[0], + PBUS_REGS_ADDR, PBUS_REGS_CNT); if (status) goto err; /* Get the MDE Registers */ - ql_build_coredump_seg_header(&mpi_coredump->mde_regs_seg_hdr, - MDE_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->mde_regs), - "MDE Registers"); - status = ql_get_mpi_regs(qdev, &mpi_coredump->mde_regs[0], - MDE_REGS_ADDR, MDE_REGS_CNT); + qlge_build_coredump_seg_header(&mpi_coredump->mde_regs_seg_hdr, + MDE_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->mde_regs), + "MDE Registers"); + status = qlge_get_mpi_regs(qdev, &mpi_coredump->mde_regs[0], + MDE_REGS_ADDR, MDE_REGS_CNT); if (status) goto err; - ql_build_coredump_seg_header(&mpi_coredump->misc_nic_seg_hdr, - MISC_NIC_INFO_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->misc_nic_info), - "MISC NIC INFO"); + qlge_build_coredump_seg_header(&mpi_coredump->misc_nic_seg_hdr, + MISC_NIC_INFO_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->misc_nic_info), + "MISC NIC INFO"); mpi_coredump->misc_nic_info.rx_ring_count = qdev->rx_ring_count; mpi_coredump->misc_nic_info.tx_ring_count = qdev->tx_ring_count; mpi_coredump->misc_nic_info.intr_count = qdev->intr_count; @@ -1082,79 +1081,79 @@ int ql_core_dump(struct ql_adapter *qdev, struct ql_mpi_coredump *mpi_coredump) /* Segment 31 */ /* Get indexed register values. */ - ql_build_coredump_seg_header(&mpi_coredump->intr_states_seg_hdr, - INTR_STATES_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->intr_states), - "INTR States"); - ql_get_intr_states(qdev, &mpi_coredump->intr_states[0]); - - ql_build_coredump_seg_header(&mpi_coredump->cam_entries_seg_hdr, - CAM_ENTRIES_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->cam_entries), - "CAM Entries"); - status = ql_get_cam_entries(qdev, &mpi_coredump->cam_entries[0]); + qlge_build_coredump_seg_header(&mpi_coredump->intr_states_seg_hdr, + INTR_STATES_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->intr_states), + "INTR States"); + qlge_get_intr_states(qdev, &mpi_coredump->intr_states[0]); + + qlge_build_coredump_seg_header(&mpi_coredump->cam_entries_seg_hdr, + CAM_ENTRIES_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->cam_entries), + "CAM Entries"); + status = qlge_get_cam_entries(qdev, &mpi_coredump->cam_entries[0]); if (status) goto err; - ql_build_coredump_seg_header(&mpi_coredump->nic_routing_words_seg_hdr, - ROUTING_WORDS_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->nic_routing_words), - "Routing Words"); - status = ql_get_routing_entries(qdev, - &mpi_coredump->nic_routing_words[0]); + qlge_build_coredump_seg_header(&mpi_coredump->nic_routing_words_seg_hdr, + ROUTING_WORDS_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->nic_routing_words), + "Routing Words"); + status = qlge_get_routing_entries(qdev, + &mpi_coredump->nic_routing_words[0]); if (status) goto err; /* Segment 34 (Rev C. step 23) */ - ql_build_coredump_seg_header(&mpi_coredump->ets_seg_hdr, - ETS_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->ets), - "ETS Registers"); - status = ql_get_ets_regs(qdev, &mpi_coredump->ets[0]); + qlge_build_coredump_seg_header(&mpi_coredump->ets_seg_hdr, + ETS_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->ets), + "ETS Registers"); + status = qlge_get_ets_regs(qdev, &mpi_coredump->ets[0]); if (status) goto err; - ql_build_coredump_seg_header(&mpi_coredump->probe_dump_seg_hdr, - PROBE_DUMP_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->probe_dump), - "Probe Dump"); - ql_get_probe_dump(qdev, &mpi_coredump->probe_dump[0]); - - ql_build_coredump_seg_header(&mpi_coredump->routing_reg_seg_hdr, - ROUTING_INDEX_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->routing_regs), - "Routing Regs"); - status = ql_get_routing_index_registers(qdev, - &mpi_coredump->routing_regs[0]); + qlge_build_coredump_seg_header(&mpi_coredump->probe_dump_seg_hdr, + PROBE_DUMP_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->probe_dump), + "Probe Dump"); + qlge_get_probe_dump(qdev, &mpi_coredump->probe_dump[0]); + + qlge_build_coredump_seg_header(&mpi_coredump->routing_reg_seg_hdr, + ROUTING_INDEX_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->routing_regs), + "Routing Regs"); + status = qlge_get_routing_index_registers(qdev, + &mpi_coredump->routing_regs[0]); if (status) goto err; - ql_build_coredump_seg_header(&mpi_coredump->mac_prot_reg_seg_hdr, - MAC_PROTOCOL_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->mac_prot_regs), - "MAC Prot Regs"); - ql_get_mac_protocol_registers(qdev, &mpi_coredump->mac_prot_regs[0]); + qlge_build_coredump_seg_header(&mpi_coredump->mac_prot_reg_seg_hdr, + MAC_PROTOCOL_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->mac_prot_regs), + "MAC Prot Regs"); + qlge_get_mac_protocol_registers(qdev, &mpi_coredump->mac_prot_regs[0]); /* Get the semaphore registers for all 5 functions */ - ql_build_coredump_seg_header(&mpi_coredump->sem_regs_seg_hdr, - SEM_REGS_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) + - sizeof(mpi_coredump->sem_regs), "Sem Registers"); + qlge_build_coredump_seg_header(&mpi_coredump->sem_regs_seg_hdr, + SEM_REGS_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->sem_regs), "Sem Registers"); - ql_get_sem_registers(qdev, &mpi_coredump->sem_regs[0]); + qlge_get_sem_registers(qdev, &mpi_coredump->sem_regs[0]); /* Prevent the mpi restarting while we dump the memory.*/ - ql_write_mpi_reg(qdev, MPI_TEST_FUNC_RST_STS, MPI_TEST_FUNC_RST_FRC); + qlge_write_mpi_reg(qdev, MPI_TEST_FUNC_RST_STS, MPI_TEST_FUNC_RST_FRC); /* clear the pause */ - status = ql_unpause_mpi_risc(qdev); + status = qlge_unpause_mpi_risc(qdev); if (status) { netif_err(qdev, drv, qdev->ndev, "Failed RISC unpause. Status = 0x%.08x\n", status); @@ -1162,20 +1161,20 @@ int ql_core_dump(struct ql_adapter *qdev, struct ql_mpi_coredump *mpi_coredump) } /* Reset the RISC so we can dump RAM */ - status = ql_hard_reset_mpi_risc(qdev); + status = qlge_hard_reset_mpi_risc(qdev); if (status) { netif_err(qdev, drv, qdev->ndev, "Failed RISC reset. Status = 0x%.08x\n", status); goto err; } - ql_build_coredump_seg_header(&mpi_coredump->code_ram_seg_hdr, - WCS_RAM_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->code_ram), - "WCS RAM"); - status = ql_dump_risc_ram_area(qdev, &mpi_coredump->code_ram[0], - CODE_RAM_ADDR, CODE_RAM_CNT); + qlge_build_coredump_seg_header(&mpi_coredump->code_ram_seg_hdr, + WCS_RAM_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->code_ram), + "WCS RAM"); + status = qlge_dump_risc_ram_area(qdev, &mpi_coredump->code_ram[0], + CODE_RAM_ADDR, CODE_RAM_CNT); if (status) { netif_err(qdev, drv, qdev->ndev, "Failed Dump of CODE RAM. Status = 0x%.08x\n", @@ -1184,13 +1183,13 @@ int ql_core_dump(struct ql_adapter *qdev, struct ql_mpi_coredump *mpi_coredump) } /* Insert the segment header */ - ql_build_coredump_seg_header(&mpi_coredump->memc_ram_seg_hdr, - MEMC_RAM_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->memc_ram), - "MEMC RAM"); - status = ql_dump_risc_ram_area(qdev, &mpi_coredump->memc_ram[0], - MEMC_RAM_ADDR, MEMC_RAM_CNT); + qlge_build_coredump_seg_header(&mpi_coredump->memc_ram_seg_hdr, + MEMC_RAM_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->memc_ram), + "MEMC RAM"); + status = qlge_dump_risc_ram_area(qdev, &mpi_coredump->memc_ram[0], + MEMC_RAM_ADDR, MEMC_RAM_CNT); if (status) { netif_err(qdev, drv, qdev->ndev, "Failed Dump of MEMC RAM. Status = 0x%.08x\n", @@ -1198,13 +1197,13 @@ int ql_core_dump(struct ql_adapter *qdev, struct ql_mpi_coredump *mpi_coredump) goto err; } err: - ql_sem_unlock(qdev, SEM_PROC_REG_MASK); /* does flush too */ + qlge_sem_unlock(qdev, SEM_PROC_REG_MASK); /* does flush too */ return status; } -static void ql_get_core_dump(struct ql_adapter *qdev) +static void qlge_get_core_dump(struct qlge_adapter *qdev) { - if (!ql_own_firmware(qdev)) { + if (!qlge_own_firmware(qdev)) { netif_err(qdev, drv, qdev->ndev, "Don't own firmware!\n"); return; } @@ -1214,11 +1213,11 @@ static void ql_get_core_dump(struct ql_adapter *qdev) "Force Coredump can only be done from interface that is up\n"); return; } - ql_queue_fw_error(qdev); + qlge_queue_fw_error(qdev); } -static void ql_gen_reg_dump(struct ql_adapter *qdev, - struct ql_reg_dump *mpi_coredump) +static void qlge_gen_reg_dump(struct qlge_adapter *qdev, + struct qlge_reg_dump *mpi_coredump) { int i, status; @@ -1228,71 +1227,71 @@ static void ql_gen_reg_dump(struct ql_adapter *qdev, mpi_coredump->mpi_global_header.header_size = sizeof(struct mpi_coredump_global_header); mpi_coredump->mpi_global_header.image_size = - sizeof(struct ql_reg_dump); + sizeof(struct qlge_reg_dump); strncpy(mpi_coredump->mpi_global_header.id_string, "MPI Coredump", sizeof(mpi_coredump->mpi_global_header.id_string)); /* segment 16 */ - ql_build_coredump_seg_header(&mpi_coredump->misc_nic_seg_hdr, - MISC_NIC_INFO_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->misc_nic_info), - "MISC NIC INFO"); + qlge_build_coredump_seg_header(&mpi_coredump->misc_nic_seg_hdr, + MISC_NIC_INFO_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->misc_nic_info), + "MISC NIC INFO"); mpi_coredump->misc_nic_info.rx_ring_count = qdev->rx_ring_count; mpi_coredump->misc_nic_info.tx_ring_count = qdev->tx_ring_count; mpi_coredump->misc_nic_info.intr_count = qdev->intr_count; mpi_coredump->misc_nic_info.function = qdev->func; /* Segment 16, Rev C. Step 18 */ - ql_build_coredump_seg_header(&mpi_coredump->nic_regs_seg_hdr, - NIC1_CONTROL_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->nic_regs), - "NIC Registers"); + qlge_build_coredump_seg_header(&mpi_coredump->nic_regs_seg_hdr, + NIC1_CONTROL_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->nic_regs), + "NIC Registers"); /* Get generic reg dump */ for (i = 0; i < 64; i++) - mpi_coredump->nic_regs[i] = ql_read32(qdev, i * sizeof(u32)); + mpi_coredump->nic_regs[i] = qlge_read32(qdev, i * sizeof(u32)); /* Segment 31 */ /* Get indexed register values. */ - ql_build_coredump_seg_header(&mpi_coredump->intr_states_seg_hdr, - INTR_STATES_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->intr_states), - "INTR States"); - ql_get_intr_states(qdev, &mpi_coredump->intr_states[0]); - - ql_build_coredump_seg_header(&mpi_coredump->cam_entries_seg_hdr, - CAM_ENTRIES_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->cam_entries), - "CAM Entries"); - status = ql_get_cam_entries(qdev, &mpi_coredump->cam_entries[0]); + qlge_build_coredump_seg_header(&mpi_coredump->intr_states_seg_hdr, + INTR_STATES_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->intr_states), + "INTR States"); + qlge_get_intr_states(qdev, &mpi_coredump->intr_states[0]); + + qlge_build_coredump_seg_header(&mpi_coredump->cam_entries_seg_hdr, + CAM_ENTRIES_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->cam_entries), + "CAM Entries"); + status = qlge_get_cam_entries(qdev, &mpi_coredump->cam_entries[0]); if (status) return; - ql_build_coredump_seg_header(&mpi_coredump->nic_routing_words_seg_hdr, - ROUTING_WORDS_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->nic_routing_words), - "Routing Words"); - status = ql_get_routing_entries(qdev, - &mpi_coredump->nic_routing_words[0]); + qlge_build_coredump_seg_header(&mpi_coredump->nic_routing_words_seg_hdr, + ROUTING_WORDS_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->nic_routing_words), + "Routing Words"); + status = qlge_get_routing_entries(qdev, + &mpi_coredump->nic_routing_words[0]); if (status) return; /* Segment 34 (Rev C. step 23) */ - ql_build_coredump_seg_header(&mpi_coredump->ets_seg_hdr, - ETS_SEG_NUM, - sizeof(struct mpi_coredump_segment_header) - + sizeof(mpi_coredump->ets), - "ETS Registers"); - status = ql_get_ets_regs(qdev, &mpi_coredump->ets[0]); + qlge_build_coredump_seg_header(&mpi_coredump->ets_seg_hdr, + ETS_SEG_NUM, + sizeof(struct mpi_coredump_segment_header) + + sizeof(mpi_coredump->ets), + "ETS Registers"); + status = qlge_get_ets_regs(qdev, &mpi_coredump->ets[0]); if (status) return; } -void ql_get_dump(struct ql_adapter *qdev, void *buff) +void qlge_get_dump(struct qlge_adapter *qdev, void *buff) { /* * If the dump has already been taken and is stored @@ -1304,21 +1303,21 @@ void ql_get_dump(struct ql_adapter *qdev, void *buff) */ if (!test_bit(QL_FRC_COREDUMP, &qdev->flags)) { - if (!ql_core_dump(qdev, buff)) - ql_soft_reset_mpi_risc(qdev); + if (!qlge_core_dump(qdev, buff)) + qlge_soft_reset_mpi_risc(qdev); else netif_err(qdev, drv, qdev->ndev, "coredump failed!\n"); } else { - ql_gen_reg_dump(qdev, buff); - ql_get_core_dump(qdev); + qlge_gen_reg_dump(qdev, buff); + qlge_get_core_dump(qdev); } } /* Coredump to messages log file using separate worker thread */ -void ql_mpi_core_to_log(struct work_struct *work) +void qlge_mpi_core_to_log(struct work_struct *work) { - struct ql_adapter *qdev = - container_of(work, struct ql_adapter, mpi_core_to_log.work); + struct qlge_adapter *qdev = + container_of(work, struct qlge_adapter, mpi_core_to_log.work); print_hex_dump(KERN_DEBUG, "Core is dumping to log file!\n", DUMP_PREFIX_OFFSET, 32, 4, qdev->mpi_coredump, @@ -1326,29 +1325,29 @@ void ql_mpi_core_to_log(struct work_struct *work) } #ifdef QL_REG_DUMP -static void ql_dump_intr_states(struct ql_adapter *qdev) +static void qlge_dump_intr_states(struct qlge_adapter *qdev) { int i; u32 value; for (i = 0; i < qdev->intr_count; i++) { - ql_write32(qdev, INTR_EN, qdev->intr_context[i].intr_read_mask); - value = ql_read32(qdev, INTR_EN); + qlge_write32(qdev, INTR_EN, qdev->intr_context[i].intr_read_mask); + value = qlge_read32(qdev, INTR_EN); netdev_err(qdev->ndev, "Interrupt %d is %s\n", i, (value & INTR_EN_EN ? "enabled" : "disabled")); } } #define DUMP_XGMAC(qdev, reg) \ -do { \ - u32 data; \ - ql_read_xgmac_reg(qdev, reg, &data); \ - netdev_err(qdev->ndev, "%s = 0x%.08x\n", #reg, data); \ -} while (0) + do { \ + u32 data; \ + qlge_read_xgmac_reg(qdev, reg, &data); \ + netdev_err(qdev->ndev, "%s = 0x%.08x\n", #reg, data); \ + } while (0) -void ql_dump_xgmac_control_regs(struct ql_adapter *qdev) +void qlge_dump_xgmac_control_regs(struct qlge_adapter *qdev) { - if (ql_sem_spinlock(qdev, qdev->xg_sem_mask)) { + if (qlge_sem_spinlock(qdev, qdev->xg_sem_mask)) { netdev_err(qdev->ndev, "%s: Couldn't get xgmac sem\n", __func__); return; @@ -1370,23 +1369,23 @@ void ql_dump_xgmac_control_regs(struct ql_adapter *qdev) DUMP_XGMAC(qdev, MAC_MGMT_INT); DUMP_XGMAC(qdev, MAC_MGMT_IN_MASK); DUMP_XGMAC(qdev, EXT_ARB_MODE); - ql_sem_unlock(qdev, qdev->xg_sem_mask); + qlge_sem_unlock(qdev, qdev->xg_sem_mask); } -static void ql_dump_ets_regs(struct ql_adapter *qdev) +static void qlge_dump_ets_regs(struct qlge_adapter *qdev) { } -static void ql_dump_cam_entries(struct ql_adapter *qdev) +static void qlge_dump_cam_entries(struct qlge_adapter *qdev) { int i; u32 value[3]; - i = ql_sem_spinlock(qdev, SEM_MAC_ADDR_MASK); + i = qlge_sem_spinlock(qdev, SEM_MAC_ADDR_MASK); if (i) return; for (i = 0; i < 4; i++) { - if (ql_get_mac_addr_reg(qdev, MAC_ADDR_TYPE_CAM_MAC, i, value)) { + if (qlge_get_mac_addr_reg(qdev, MAC_ADDR_TYPE_CAM_MAC, i, value)) { netdev_err(qdev->ndev, "%s: Failed read of mac index register\n", __func__); @@ -1398,7 +1397,7 @@ static void ql_dump_cam_entries(struct ql_adapter *qdev) i, value[1], value[0], value[2]); } for (i = 0; i < 32; i++) { - if (ql_get_mac_addr_reg + if (qlge_get_mac_addr_reg (qdev, MAC_ADDR_TYPE_MULTI_MAC, i, value)) { netdev_err(qdev->ndev, "%s: Failed read of mac index register\n", @@ -1410,20 +1409,20 @@ static void ql_dump_cam_entries(struct ql_adapter *qdev) "MCAST index %d CAM Lookup Lower = 0x%.08x:%.08x\n", i, value[1], value[0]); } - ql_sem_unlock(qdev, SEM_MAC_ADDR_MASK); + qlge_sem_unlock(qdev, SEM_MAC_ADDR_MASK); } -void ql_dump_routing_entries(struct ql_adapter *qdev) +void qlge_dump_routing_entries(struct qlge_adapter *qdev) { int i; u32 value; - i = ql_sem_spinlock(qdev, SEM_RT_IDX_MASK); + i = qlge_sem_spinlock(qdev, SEM_RT_IDX_MASK); if (i) return; for (i = 0; i < 16; i++) { value = 0; - if (ql_get_routing_reg(qdev, i, &value)) { + if (qlge_get_routing_reg(qdev, i, &value)) { netdev_err(qdev->ndev, "%s: Failed read of routing index register\n", __func__); @@ -1434,13 +1433,13 @@ void ql_dump_routing_entries(struct ql_adapter *qdev) "Routing Mask %d = 0x%.08x\n", i, value); } - ql_sem_unlock(qdev, SEM_RT_IDX_MASK); + qlge_sem_unlock(qdev, SEM_RT_IDX_MASK); } #define DUMP_REG(qdev, reg) \ - netdev_err(qdev->ndev, "%-32s= 0x%x\n", #reg, ql_read32(qdev, reg)) + netdev_err(qdev->ndev, "%-32s= 0x%x\n", #reg, qlge_read32(qdev, reg)) -void ql_dump_regs(struct ql_adapter *qdev) +void qlge_dump_regs(struct qlge_adapter *qdev) { netdev_err(qdev->ndev, "reg dump for function #%d\n", qdev->func); DUMP_REG(qdev, SYS); @@ -1496,11 +1495,11 @@ void ql_dump_regs(struct ql_adapter *qdev) DUMP_REG(qdev, XG_SERDES_DATA); DUMP_REG(qdev, PRB_MX_ADDR); DUMP_REG(qdev, PRB_MX_DATA); - ql_dump_intr_states(qdev); - ql_dump_xgmac_control_regs(qdev); - ql_dump_ets_regs(qdev); - ql_dump_cam_entries(qdev); - ql_dump_routing_entries(qdev); + qlge_dump_intr_states(qdev); + qlge_dump_xgmac_control_regs(qdev); + qlge_dump_ets_regs(qdev); + qlge_dump_cam_entries(qdev); + qlge_dump_routing_entries(qdev); } #endif @@ -1510,7 +1509,7 @@ void ql_dump_regs(struct ql_adapter *qdev) netdev_err(qdev->ndev, "%s = %ld\n", #stat, \ (unsigned long)(qdev)->nic_stats.stat) -void ql_dump_stat(struct ql_adapter *qdev) +void qlge_dump_stat(struct qlge_adapter *qdev) { netdev_err(qdev->ndev, "%s: Enter\n", __func__); DUMP_STAT(qdev, tx_pkts); @@ -1567,8 +1566,8 @@ void ql_dump_stat(struct ql_adapter *qdev) (unsigned long long)qdev->field) #define DUMP_QDEV_ARRAY(qdev, type, array, index, field) \ netdev_err(qdev->ndev, "%s[%d].%s = " type "\n", \ - #array, index, #field, (qdev)->array[index].field) -void ql_dump_qdev(struct ql_adapter *qdev) +#array, index, #field, (qdev)->array[index].field) + void qlge_dump_qdev(struct qlge_adapter *qdev) { int i; @@ -1615,10 +1614,10 @@ void ql_dump_qdev(struct ql_adapter *qdev) #endif #ifdef QL_CB_DUMP -void ql_dump_wqicb(struct wqicb *wqicb) +void qlge_dump_wqicb(struct wqicb *wqicb) { struct tx_ring *tx_ring = container_of(wqicb, struct tx_ring, wqicb); - struct ql_adapter *qdev = tx_ring->qdev; + struct qlge_adapter *qdev = tx_ring->qdev; netdev_err(qdev->ndev, "Dumping wqicb stuff...\n"); netdev_err(qdev->ndev, "wqicb->len = 0x%x\n", le16_to_cpu(wqicb->len)); @@ -1633,9 +1632,9 @@ void ql_dump_wqicb(struct wqicb *wqicb) (unsigned long long)le64_to_cpu(wqicb->cnsmr_idx_addr)); } -void ql_dump_tx_ring(struct tx_ring *tx_ring) +void qlge_dump_tx_ring(struct tx_ring *tx_ring) { - struct ql_adapter *qdev = tx_ring->qdev; + struct qlge_adapter *qdev = tx_ring->qdev; netdev_err(qdev->ndev, "===================== Dumping tx_ring %d ===============\n", tx_ring->wq_id); @@ -1645,7 +1644,7 @@ void ql_dump_tx_ring(struct tx_ring *tx_ring) netdev_err(qdev->ndev, "tx_ring->cnsmr_idx_sh_reg, addr = 0x%p, value = %d\n", tx_ring->cnsmr_idx_sh_reg, tx_ring->cnsmr_idx_sh_reg - ? ql_read_sh_reg(tx_ring->cnsmr_idx_sh_reg) : 0); + ? qlge_read_sh_reg(tx_ring->cnsmr_idx_sh_reg) : 0); netdev_err(qdev->ndev, "tx_ring->size = %d\n", tx_ring->wq_size); netdev_err(qdev->ndev, "tx_ring->len = %d\n", tx_ring->wq_len); netdev_err(qdev->ndev, "tx_ring->prod_idx_db_reg = %p\n", tx_ring->prod_idx_db_reg); @@ -1657,11 +1656,11 @@ void ql_dump_tx_ring(struct tx_ring *tx_ring) netdev_err(qdev->ndev, "tx_ring->tx_count = %d\n", atomic_read(&tx_ring->tx_count)); } -void ql_dump_ricb(struct ricb *ricb) +void qlge_dump_ricb(struct ricb *ricb) { int i; - struct ql_adapter *qdev = - container_of(ricb, struct ql_adapter, ricb); + struct qlge_adapter *qdev = + container_of(ricb, struct qlge_adapter, ricb); netdev_err(qdev->ndev, "===================== Dumping ricb ===============\n"); netdev_err(qdev->ndev, "Dumping ricb stuff...\n"); @@ -1689,10 +1688,10 @@ void ql_dump_ricb(struct ricb *ricb) le32_to_cpu(ricb->ipv4_hash_key[i])); } -void ql_dump_cqicb(struct cqicb *cqicb) +void qlge_dump_cqicb(struct cqicb *cqicb) { struct rx_ring *rx_ring = container_of(cqicb, struct rx_ring, cqicb); - struct ql_adapter *qdev = rx_ring->qdev; + struct qlge_adapter *qdev = rx_ring->qdev; netdev_err(qdev->ndev, "Dumping cqicb stuff...\n"); @@ -1723,7 +1722,7 @@ void ql_dump_cqicb(struct cqicb *cqicb) static const char *qlge_rx_ring_type_name(struct rx_ring *rx_ring) { - struct ql_adapter *qdev = rx_ring->qdev; + struct qlge_adapter *qdev = rx_ring->qdev; if (rx_ring->cq_id < qdev->rss_ring_count) return "RX COMPLETION"; @@ -1731,9 +1730,9 @@ static const char *qlge_rx_ring_type_name(struct rx_ring *rx_ring) return "TX COMPLETION"; }; -void ql_dump_rx_ring(struct rx_ring *rx_ring) +void qlge_dump_rx_ring(struct rx_ring *rx_ring) { - struct ql_adapter *qdev = rx_ring->qdev; + struct qlge_adapter *qdev = rx_ring->qdev; netdev_err(qdev->ndev, "===================== Dumping rx_ring %d ===============\n", @@ -1750,7 +1749,7 @@ void ql_dump_rx_ring(struct rx_ring *rx_ring) netdev_err(qdev->ndev, "rx_ring->prod_idx_sh_reg, addr = 0x%p, value = %d\n", rx_ring->prod_idx_sh_reg, - rx_ring->prod_idx_sh_reg ? ql_read_sh_reg(rx_ring->prod_idx_sh_reg) : 0); + rx_ring->prod_idx_sh_reg ? qlge_read_sh_reg(rx_ring->prod_idx_sh_reg) : 0); netdev_err(qdev->ndev, "rx_ring->prod_idx_sh_reg_dma = %llx\n", (unsigned long long)rx_ring->prod_idx_sh_reg_dma); netdev_err(qdev->ndev, "rx_ring->cnsmr_idx_db_reg = %p\n", @@ -1790,7 +1789,7 @@ void ql_dump_rx_ring(struct rx_ring *rx_ring) netdev_err(qdev->ndev, "rx_ring->qdev = %p\n", rx_ring->qdev); } -void ql_dump_hw_cb(struct ql_adapter *qdev, int size, u32 bit, u16 q_id) +void qlge_dump_hw_cb(struct qlge_adapter *qdev, int size, u32 bit, u16 q_id) { void *ptr; @@ -1800,19 +1799,19 @@ void ql_dump_hw_cb(struct ql_adapter *qdev, int size, u32 bit, u16 q_id) if (!ptr) return; - if (ql_write_cfg(qdev, ptr, size, bit, q_id)) { + if (qlge_write_cfg(qdev, ptr, size, bit, q_id)) { netdev_err(qdev->ndev, "%s: Failed to upload control block!\n", __func__); goto fail_it; } switch (bit) { case CFG_DRQ: - ql_dump_wqicb((struct wqicb *)ptr); + qlge_dump_wqicb((struct wqicb *)ptr); break; case CFG_DCQ: - ql_dump_cqicb((struct cqicb *)ptr); + qlge_dump_cqicb((struct cqicb *)ptr); break; case CFG_DR: - ql_dump_ricb((struct ricb *)ptr); + qlge_dump_ricb((struct ricb *)ptr); break; default: netdev_err(qdev->ndev, "%s: Invalid bit value = %x\n", __func__, bit); @@ -1824,7 +1823,7 @@ void ql_dump_hw_cb(struct ql_adapter *qdev, int size, u32 bit, u16 q_id) #endif #ifdef QL_OB_DUMP -void ql_dump_tx_desc(struct ql_adapter *qdev, struct tx_buf_desc *tbd) +void qlge_dump_tx_desc(struct qlge_adapter *qdev, struct tx_buf_desc *tbd) { netdev_err(qdev->ndev, "tbd->addr = 0x%llx\n", le64_to_cpu((u64)tbd->addr)); @@ -1851,10 +1850,10 @@ void ql_dump_tx_desc(struct ql_adapter *qdev, struct tx_buf_desc *tbd) tbd->len & TX_DESC_E ? "E" : "."); } -void ql_dump_ob_mac_iocb(struct ql_adapter *qdev, struct ob_mac_iocb_req *ob_mac_iocb) +void qlge_dump_ob_mac_iocb(struct qlge_adapter *qdev, struct qlge_ob_mac_iocb_req *ob_mac_iocb) { - struct ob_mac_tso_iocb_req *ob_mac_tso_iocb = - (struct ob_mac_tso_iocb_req *)ob_mac_iocb; + struct qlge_ob_mac_tso_iocb_req *ob_mac_tso_iocb = + (struct qlge_ob_mac_tso_iocb_req *)ob_mac_iocb; struct tx_buf_desc *tbd; u16 frame_len; @@ -1894,16 +1893,16 @@ void ql_dump_ob_mac_iocb(struct ql_adapter *qdev, struct ob_mac_iocb_req *ob_mac frame_len = le16_to_cpu(ob_mac_iocb->frame_len); } tbd = &ob_mac_iocb->tbd[0]; - ql_dump_tx_desc(qdev, tbd); + qlge_dump_tx_desc(qdev, tbd); } -void ql_dump_ob_mac_rsp(struct ql_adapter *qdev, struct ob_mac_iocb_rsp *ob_mac_rsp) +void qlge_dump_ob_mac_rsp(struct qlge_adapter *qdev, struct qlge_ob_mac_iocb_rsp *ob_mac_rsp) { netdev_err(qdev->ndev, "%s\n", __func__); netdev_err(qdev->ndev, "opcode = %d\n", ob_mac_rsp->opcode); netdev_err(qdev->ndev, "flags = %s %s %s %s %s %s %s\n", ob_mac_rsp->flags1 & OB_MAC_IOCB_RSP_OI ? - "OI" : ".", ob_mac_rsp->flags1 & OB_MAC_IOCB_RSP_I ? "I" : ".", + "OI" : ".", ob_mac_rsp->flags1 & OB_MAC_IOCB_RSP_I ? "I" : ".", ob_mac_rsp->flags1 & OB_MAC_IOCB_RSP_E ? "E" : ".", ob_mac_rsp->flags1 & OB_MAC_IOCB_RSP_S ? "S" : ".", ob_mac_rsp->flags1 & OB_MAC_IOCB_RSP_L ? "L" : ".", @@ -1914,7 +1913,7 @@ void ql_dump_ob_mac_rsp(struct ql_adapter *qdev, struct ob_mac_iocb_rsp *ob_mac_ #endif #ifdef QL_IB_DUMP -void ql_dump_ib_mac_rsp(struct ql_adapter *qdev, struct ib_mac_iocb_rsp *ib_mac_rsp) +void qlge_dump_ib_mac_rsp(struct qlge_adapter *qdev, struct qlge_ib_mac_iocb_rsp *ib_mac_rsp) { netdev_err(qdev->ndev, "%s\n", __func__); netdev_err(qdev->ndev, "opcode = 0x%x\n", ib_mac_rsp->opcode); @@ -1996,7 +1995,7 @@ void ql_dump_ib_mac_rsp(struct ql_adapter *qdev, struct ib_mac_iocb_rsp *ib_mac_ #endif #ifdef QL_ALL_DUMP -void ql_dump_all(struct ql_adapter *qdev) +void qlge_dump_all(struct qlge_adapter *qdev) { int i; diff --git a/drivers/staging/qlge/qlge_ethtool.c b/drivers/staging/qlge/qlge_ethtool.c index a28f0254cf60..3e577e1bc27c 100644 --- a/drivers/staging/qlge/qlge_ethtool.c +++ b/drivers/staging/qlge/qlge_ethtool.c @@ -34,16 +34,16 @@ #include "qlge.h" -struct ql_stats { +struct qlge_stats { char stat_string[ETH_GSTRING_LEN]; int sizeof_stat; int stat_offset; }; -#define QL_SIZEOF(m) sizeof_field(struct ql_adapter, m) -#define QL_OFF(m) offsetof(struct ql_adapter, m) +#define QL_SIZEOF(m) sizeof_field(struct qlge_adapter, m) +#define QL_OFF(m) offsetof(struct qlge_adapter, m) -static const struct ql_stats ql_gstrings_stats[] = { +static const struct qlge_stats qlge_gstrings_stats[] = { {"tx_pkts", QL_SIZEOF(nic_stats.tx_pkts), QL_OFF(nic_stats.tx_pkts)}, {"tx_bytes", QL_SIZEOF(nic_stats.tx_bytes), QL_OFF(nic_stats.tx_bytes)}, {"tx_mcast_pkts", QL_SIZEOF(nic_stats.tx_mcast_pkts), @@ -175,15 +175,15 @@ static const struct ql_stats ql_gstrings_stats[] = { QL_OFF(nic_stats.rx_nic_fifo_drop)}, }; -static const char ql_gstrings_test[][ETH_GSTRING_LEN] = { +static const char qlge_gstrings_test[][ETH_GSTRING_LEN] = { "Loopback test (offline)" }; -#define QLGE_TEST_LEN (sizeof(ql_gstrings_test) / ETH_GSTRING_LEN) -#define QLGE_STATS_LEN ARRAY_SIZE(ql_gstrings_stats) +#define QLGE_TEST_LEN (sizeof(qlge_gstrings_test) / ETH_GSTRING_LEN) +#define QLGE_STATS_LEN ARRAY_SIZE(qlge_gstrings_stats) #define QLGE_RCV_MAC_ERR_STATS 7 -static int ql_update_ring_coalescing(struct ql_adapter *qdev) +static int qlge_update_ring_coalescing(struct qlge_adapter *qdev) { int i, status = 0; struct rx_ring *rx_ring; @@ -203,10 +203,10 @@ static int ql_update_ring_coalescing(struct ql_adapter *qdev) cqicb = (struct cqicb *)rx_ring; cqicb->irq_delay = cpu_to_le16(qdev->tx_coalesce_usecs); cqicb->pkt_delay = - cpu_to_le16(qdev->tx_max_coalesced_frames); + cpu_to_le16(qdev->tx_max_coalesced_frames); cqicb->flags = FLAGS_LI; - status = ql_write_cfg(qdev, cqicb, sizeof(*cqicb), - CFG_LCQ, rx_ring->cq_id); + status = qlge_write_cfg(qdev, cqicb, sizeof(*cqicb), + CFG_LCQ, rx_ring->cq_id); if (status) { netif_err(qdev, ifup, qdev->ndev, "Failed to load CQICB.\n"); @@ -224,10 +224,10 @@ static int ql_update_ring_coalescing(struct ql_adapter *qdev) cqicb = (struct cqicb *)rx_ring; cqicb->irq_delay = cpu_to_le16(qdev->rx_coalesce_usecs); cqicb->pkt_delay = - cpu_to_le16(qdev->rx_max_coalesced_frames); + cpu_to_le16(qdev->rx_max_coalesced_frames); cqicb->flags = FLAGS_LI; - status = ql_write_cfg(qdev, cqicb, sizeof(*cqicb), - CFG_LCQ, rx_ring->cq_id); + status = qlge_write_cfg(qdev, cqicb, sizeof(*cqicb), + CFG_LCQ, rx_ring->cq_id); if (status) { netif_err(qdev, ifup, qdev->ndev, "Failed to load CQICB.\n"); @@ -239,14 +239,14 @@ static int ql_update_ring_coalescing(struct ql_adapter *qdev) return status; } -static void ql_update_stats(struct ql_adapter *qdev) +static void qlge_update_stats(struct qlge_adapter *qdev) { u32 i; u64 data; u64 *iter = &qdev->nic_stats.tx_pkts; spin_lock(&qdev->stats_lock); - if (ql_sem_spinlock(qdev, qdev->xg_sem_mask)) { + if (qlge_sem_spinlock(qdev, qdev->xg_sem_mask)) { netif_err(qdev, drv, qdev->ndev, "Couldn't get xgmac sem.\n"); goto quit; @@ -255,7 +255,7 @@ static void ql_update_stats(struct ql_adapter *qdev) * Get TX statistics. */ for (i = 0x200; i < 0x280; i += 8) { - if (ql_read_xgmac_reg64(qdev, i, &data)) { + if (qlge_read_xgmac_reg64(qdev, i, &data)) { netif_err(qdev, drv, qdev->ndev, "Error reading status register 0x%.04x.\n", i); @@ -270,7 +270,7 @@ static void ql_update_stats(struct ql_adapter *qdev) * Get RX statistics. */ for (i = 0x300; i < 0x3d0; i += 8) { - if (ql_read_xgmac_reg64(qdev, i, &data)) { + if (qlge_read_xgmac_reg64(qdev, i, &data)) { netif_err(qdev, drv, qdev->ndev, "Error reading status register 0x%.04x.\n", i); @@ -288,7 +288,7 @@ static void ql_update_stats(struct ql_adapter *qdev) * Get Per-priority TX pause frame counter statistics. */ for (i = 0x500; i < 0x540; i += 8) { - if (ql_read_xgmac_reg64(qdev, i, &data)) { + if (qlge_read_xgmac_reg64(qdev, i, &data)) { netif_err(qdev, drv, qdev->ndev, "Error reading status register 0x%.04x.\n", i); @@ -303,7 +303,7 @@ static void ql_update_stats(struct ql_adapter *qdev) * Get Per-priority RX pause frame counter statistics. */ for (i = 0x568; i < 0x5a8; i += 8) { - if (ql_read_xgmac_reg64(qdev, i, &data)) { + if (qlge_read_xgmac_reg64(qdev, i, &data)) { netif_err(qdev, drv, qdev->ndev, "Error reading status register 0x%.04x.\n", i); @@ -317,7 +317,7 @@ static void ql_update_stats(struct ql_adapter *qdev) /* * Get RX NIC FIFO DROP statistics. */ - if (ql_read_xgmac_reg64(qdev, 0x5b8, &data)) { + if (qlge_read_xgmac_reg64(qdev, 0x5b8, &data)) { netif_err(qdev, drv, qdev->ndev, "Error reading status register 0x%.04x.\n", i); goto end; @@ -325,32 +325,32 @@ static void ql_update_stats(struct ql_adapter *qdev) *iter = data; } end: - ql_sem_unlock(qdev, qdev->xg_sem_mask); + qlge_sem_unlock(qdev, qdev->xg_sem_mask); quit: spin_unlock(&qdev->stats_lock); QL_DUMP_STAT(qdev); } -static void ql_get_strings(struct net_device *dev, u32 stringset, u8 *buf) +static void qlge_get_strings(struct net_device *dev, u32 stringset, u8 *buf) { int index; switch (stringset) { case ETH_SS_TEST: - memcpy(buf, *ql_gstrings_test, QLGE_TEST_LEN * ETH_GSTRING_LEN); + memcpy(buf, *qlge_gstrings_test, QLGE_TEST_LEN * ETH_GSTRING_LEN); break; case ETH_SS_STATS: for (index = 0; index < QLGE_STATS_LEN; index++) { memcpy(buf + index * ETH_GSTRING_LEN, - ql_gstrings_stats[index].stat_string, + qlge_gstrings_stats[index].stat_string, ETH_GSTRING_LEN); } break; } } -static int ql_get_sset_count(struct net_device *dev, int sset) +static int qlge_get_sset_count(struct net_device *dev, int sset) { switch (sset) { case ETH_SS_TEST: @@ -363,34 +363,34 @@ static int ql_get_sset_count(struct net_device *dev, int sset) } static void -ql_get_ethtool_stats(struct net_device *ndev, - struct ethtool_stats *stats, u64 *data) +qlge_get_ethtool_stats(struct net_device *ndev, + struct ethtool_stats *stats, u64 *data) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); int index, length; length = QLGE_STATS_LEN; - ql_update_stats(qdev); + qlge_update_stats(qdev); for (index = 0; index < length; index++) { char *p = (char *)qdev + - ql_gstrings_stats[index].stat_offset; - *data++ = (ql_gstrings_stats[index].sizeof_stat == - sizeof(u64)) ? *(u64 *)p : (*(u32 *)p); + qlge_gstrings_stats[index].stat_offset; + *data++ = (qlge_gstrings_stats[index].sizeof_stat == + sizeof(u64)) ? *(u64 *)p : (*(u32 *)p); } } -static int ql_get_link_ksettings(struct net_device *ndev, - struct ethtool_link_ksettings *ecmd) +static int qlge_get_link_ksettings(struct net_device *ndev, + struct ethtool_link_ksettings *ecmd) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); u32 supported, advertising; supported = SUPPORTED_10000baseT_Full; advertising = ADVERTISED_10000baseT_Full; if ((qdev->link_status & STS_LINK_TYPE_MASK) == - STS_LINK_TYPE_10GBASET) { + STS_LINK_TYPE_10GBASET) { supported |= (SUPPORTED_TP | SUPPORTED_Autoneg); advertising |= (ADVERTISED_TP | ADVERTISED_Autoneg); ecmd->base.port = PORT_TP; @@ -412,10 +412,10 @@ static int ql_get_link_ksettings(struct net_device *ndev, return 0; } -static void ql_get_drvinfo(struct net_device *ndev, - struct ethtool_drvinfo *drvinfo) +static void qlge_get_drvinfo(struct net_device *ndev, + struct ethtool_drvinfo *drvinfo) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); strlcpy(drvinfo->driver, qlge_driver_name, sizeof(drvinfo->driver)); strlcpy(drvinfo->version, qlge_driver_version, @@ -429,9 +429,9 @@ static void ql_get_drvinfo(struct net_device *ndev, sizeof(drvinfo->bus_info)); } -static void ql_get_wol(struct net_device *ndev, struct ethtool_wolinfo *wol) +static void qlge_get_wol(struct net_device *ndev, struct ethtool_wolinfo *wol) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); unsigned short ssys_dev = qdev->pdev->subsystem_device; /* WOL is only supported for mezz card. */ @@ -442,9 +442,9 @@ static void ql_get_wol(struct net_device *ndev, struct ethtool_wolinfo *wol) } } -static int ql_set_wol(struct net_device *ndev, struct ethtool_wolinfo *wol) +static int qlge_set_wol(struct net_device *ndev, struct ethtool_wolinfo *wol) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); unsigned short ssys_dev = qdev->pdev->subsystem_device; /* WOL is only supported for mezz card. */ @@ -462,25 +462,25 @@ static int ql_set_wol(struct net_device *ndev, struct ethtool_wolinfo *wol) return 0; } -static int ql_set_phys_id(struct net_device *ndev, - enum ethtool_phys_id_state state) +static int qlge_set_phys_id(struct net_device *ndev, + enum ethtool_phys_id_state state) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); switch (state) { case ETHTOOL_ID_ACTIVE: /* Save the current LED settings */ - if (ql_mb_get_led_cfg(qdev)) + if (qlge_mb_get_led_cfg(qdev)) return -EIO; /* Start blinking */ - ql_mb_set_led_cfg(qdev, QL_LED_BLINK); + qlge_mb_set_led_cfg(qdev, QL_LED_BLINK); return 0; case ETHTOOL_ID_INACTIVE: /* Restore LED settings */ - if (ql_mb_set_led_cfg(qdev, qdev->led_config)) + if (qlge_mb_set_led_cfg(qdev, qdev->led_config)) return -EIO; return 0; @@ -489,7 +489,7 @@ static int ql_set_phys_id(struct net_device *ndev, } } -static int ql_start_loopback(struct ql_adapter *qdev) +static int qlge_start_loopback(struct qlge_adapter *qdev) { if (netif_carrier_ok(qdev->ndev)) { set_bit(QL_LB_LINK_UP, &qdev->flags); @@ -498,21 +498,21 @@ static int ql_start_loopback(struct ql_adapter *qdev) clear_bit(QL_LB_LINK_UP, &qdev->flags); } qdev->link_config |= CFG_LOOPBACK_PCS; - return ql_mb_set_port_cfg(qdev); + return qlge_mb_set_port_cfg(qdev); } -static void ql_stop_loopback(struct ql_adapter *qdev) +static void qlge_stop_loopback(struct qlge_adapter *qdev) { qdev->link_config &= ~CFG_LOOPBACK_PCS; - ql_mb_set_port_cfg(qdev); + qlge_mb_set_port_cfg(qdev); if (test_bit(QL_LB_LINK_UP, &qdev->flags)) { netif_carrier_on(qdev->ndev); clear_bit(QL_LB_LINK_UP, &qdev->flags); } } -static void ql_create_lb_frame(struct sk_buff *skb, - unsigned int frame_size) +static void qlge_create_lb_frame(struct sk_buff *skb, + unsigned int frame_size) { memset(skb->data, 0xFF, frame_size); frame_size &= ~1; @@ -521,8 +521,8 @@ static void ql_create_lb_frame(struct sk_buff *skb, skb->data[frame_size / 2 + 12] = (unsigned char)0xAF; } -void ql_check_lb_frame(struct ql_adapter *qdev, - struct sk_buff *skb) +void qlge_check_lb_frame(struct qlge_adapter *qdev, + struct sk_buff *skb) { unsigned int frame_size = skb->len; @@ -534,7 +534,7 @@ void ql_check_lb_frame(struct ql_adapter *qdev, } } -static int ql_run_loopback_test(struct ql_adapter *qdev) +static int qlge_run_loopback_test(struct qlge_adapter *qdev) { int i; netdev_tx_t rc; @@ -548,33 +548,33 @@ static int ql_run_loopback_test(struct ql_adapter *qdev) skb->queue_mapping = 0; skb_put(skb, size); - ql_create_lb_frame(skb, size); - rc = ql_lb_send(skb, qdev->ndev); + qlge_create_lb_frame(skb, size); + rc = qlge_lb_send(skb, qdev->ndev); if (rc != NETDEV_TX_OK) return -EPIPE; atomic_inc(&qdev->lb_count); } /* Give queue time to settle before testing results. */ msleep(2); - ql_clean_lb_rx_ring(&qdev->rx_ring[0], 128); + qlge_clean_lb_rx_ring(&qdev->rx_ring[0], 128); return atomic_read(&qdev->lb_count) ? -EIO : 0; } -static int ql_loopback_test(struct ql_adapter *qdev, u64 *data) +static int qlge_loopback_test(struct qlge_adapter *qdev, u64 *data) { - *data = ql_start_loopback(qdev); + *data = qlge_start_loopback(qdev); if (*data) goto out; - *data = ql_run_loopback_test(qdev); + *data = qlge_run_loopback_test(qdev); out: - ql_stop_loopback(qdev); + qlge_stop_loopback(qdev); return *data; } -static void ql_self_test(struct net_device *ndev, - struct ethtool_test *eth_test, u64 *data) +static void qlge_self_test(struct net_device *ndev, + struct ethtool_test *eth_test, u64 *data) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); memset(data, 0, sizeof(u64) * QLGE_TEST_LEN); @@ -582,7 +582,7 @@ static void ql_self_test(struct net_device *ndev, set_bit(QL_SELFTEST, &qdev->flags); if (eth_test->flags == ETH_TEST_FL_OFFLINE) { /* Offline tests */ - if (ql_loopback_test(qdev, &data[0])) + if (qlge_loopback_test(qdev, &data[0])) eth_test->flags |= ETH_TEST_FL_FAILED; } else { @@ -601,32 +601,32 @@ static void ql_self_test(struct net_device *ndev, } } -static int ql_get_regs_len(struct net_device *ndev) +static int qlge_get_regs_len(struct net_device *ndev) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); if (!test_bit(QL_FRC_COREDUMP, &qdev->flags)) - return sizeof(struct ql_mpi_coredump); + return sizeof(struct qlge_mpi_coredump); else - return sizeof(struct ql_reg_dump); + return sizeof(struct qlge_reg_dump); } -static void ql_get_regs(struct net_device *ndev, - struct ethtool_regs *regs, void *p) +static void qlge_get_regs(struct net_device *ndev, + struct ethtool_regs *regs, void *p) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); - ql_get_dump(qdev, p); + qlge_get_dump(qdev, p); qdev->core_is_dumped = 0; if (!test_bit(QL_FRC_COREDUMP, &qdev->flags)) - regs->len = sizeof(struct ql_mpi_coredump); + regs->len = sizeof(struct qlge_mpi_coredump); else - regs->len = sizeof(struct ql_reg_dump); + regs->len = sizeof(struct qlge_reg_dump); } -static int ql_get_coalesce(struct net_device *dev, struct ethtool_coalesce *c) +static int qlge_get_coalesce(struct net_device *dev, struct ethtool_coalesce *c) { - struct ql_adapter *qdev = netdev_priv(dev); + struct qlge_adapter *qdev = netdev_priv(dev); c->rx_coalesce_usecs = qdev->rx_coalesce_usecs; c->tx_coalesce_usecs = qdev->tx_coalesce_usecs; @@ -647,14 +647,14 @@ static int ql_get_coalesce(struct net_device *dev, struct ethtool_coalesce *c) return 0; } -static int ql_set_coalesce(struct net_device *ndev, struct ethtool_coalesce *c) +static int qlge_set_coalesce(struct net_device *ndev, struct ethtool_coalesce *c) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); /* Validate user parameters. */ if (c->rx_coalesce_usecs > qdev->rx_ring_size / 2) return -EINVAL; - /* Don't wait more than 10 usec. */ + /* Don't wait more than 10 usec. */ if (c->rx_max_coalesced_frames > MAX_INTER_FRAME_WAIT) return -EINVAL; if (c->tx_coalesce_usecs > qdev->tx_ring_size / 2) @@ -674,25 +674,25 @@ static int ql_set_coalesce(struct net_device *ndev, struct ethtool_coalesce *c) qdev->rx_max_coalesced_frames = c->rx_max_coalesced_frames; qdev->tx_max_coalesced_frames = c->tx_max_coalesced_frames; - return ql_update_ring_coalescing(qdev); + return qlge_update_ring_coalescing(qdev); } -static void ql_get_pauseparam(struct net_device *netdev, - struct ethtool_pauseparam *pause) +static void qlge_get_pauseparam(struct net_device *netdev, + struct ethtool_pauseparam *pause) { - struct ql_adapter *qdev = netdev_priv(netdev); + struct qlge_adapter *qdev = netdev_priv(netdev); - ql_mb_get_port_cfg(qdev); + qlge_mb_get_port_cfg(qdev); if (qdev->link_config & CFG_PAUSE_STD) { pause->rx_pause = 1; pause->tx_pause = 1; } } -static int ql_set_pauseparam(struct net_device *netdev, - struct ethtool_pauseparam *pause) +static int qlge_set_pauseparam(struct net_device *netdev, + struct ethtool_pauseparam *pause) { - struct ql_adapter *qdev = netdev_priv(netdev); + struct qlge_adapter *qdev = netdev_priv(netdev); if ((pause->rx_pause) && (pause->tx_pause)) qdev->link_config |= CFG_PAUSE_STD; @@ -701,19 +701,19 @@ static int ql_set_pauseparam(struct net_device *netdev, else return -EINVAL; - return ql_mb_set_port_cfg(qdev); + return qlge_mb_set_port_cfg(qdev); } -static u32 ql_get_msglevel(struct net_device *ndev) +static u32 qlge_get_msglevel(struct net_device *ndev) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); return qdev->msg_enable; } -static void ql_set_msglevel(struct net_device *ndev, u32 value) +static void qlge_set_msglevel(struct net_device *ndev, u32 value) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); qdev->msg_enable = value; } @@ -721,23 +721,23 @@ static void ql_set_msglevel(struct net_device *ndev, u32 value) const struct ethtool_ops qlge_ethtool_ops = { .supported_coalesce_params = ETHTOOL_COALESCE_USECS | ETHTOOL_COALESCE_MAX_FRAMES, - .get_drvinfo = ql_get_drvinfo, - .get_wol = ql_get_wol, - .set_wol = ql_set_wol, - .get_regs_len = ql_get_regs_len, - .get_regs = ql_get_regs, - .get_msglevel = ql_get_msglevel, - .set_msglevel = ql_set_msglevel, + .get_drvinfo = qlge_get_drvinfo, + .get_wol = qlge_get_wol, + .set_wol = qlge_set_wol, + .get_regs_len = qlge_get_regs_len, + .get_regs = qlge_get_regs, + .get_msglevel = qlge_get_msglevel, + .set_msglevel = qlge_set_msglevel, .get_link = ethtool_op_get_link, - .set_phys_id = ql_set_phys_id, - .self_test = ql_self_test, - .get_pauseparam = ql_get_pauseparam, - .set_pauseparam = ql_set_pauseparam, - .get_coalesce = ql_get_coalesce, - .set_coalesce = ql_set_coalesce, - .get_sset_count = ql_get_sset_count, - .get_strings = ql_get_strings, - .get_ethtool_stats = ql_get_ethtool_stats, - .get_link_ksettings = ql_get_link_ksettings, + .set_phys_id = qlge_set_phys_id, + .self_test = qlge_self_test, + .get_pauseparam = qlge_get_pauseparam, + .set_pauseparam = qlge_set_pauseparam, + .get_coalesce = qlge_get_coalesce, + .set_coalesce = qlge_set_coalesce, + .get_sset_count = qlge_get_sset_count, + .get_strings = qlge_get_strings, + .get_ethtool_stats = qlge_get_ethtool_stats, + .get_link_ksettings = qlge_get_link_ksettings, }; diff --git a/drivers/staging/qlge/qlge_main.c b/drivers/staging/qlge/qlge_main.c index e6b7baa12cd6..4042ea8d36a5 100644 --- a/drivers/staging/qlge/qlge_main.c +++ b/drivers/staging/qlge/qlge_main.c @@ -89,16 +89,16 @@ static const struct pci_device_id qlge_pci_tbl[] = { MODULE_DEVICE_TABLE(pci, qlge_pci_tbl); -static int ql_wol(struct ql_adapter *); +static int qlge_wol(struct qlge_adapter *); static void qlge_set_multicast_list(struct net_device *); -static int ql_adapter_down(struct ql_adapter *); -static int ql_adapter_up(struct ql_adapter *); +static int qlge_adapter_down(struct qlge_adapter *); +static int qlge_adapter_up(struct qlge_adapter *); /* This hardware semaphore causes exclusive access to * resources shared between the NIC driver, MPI firmware, * FCOE firmware and the FC driver. */ -static int ql_sem_trylock(struct ql_adapter *qdev, u32 sem_mask) +static int qlge_sem_trylock(struct qlge_adapter *qdev, u32 sem_mask) { u32 sem_bits = 0; @@ -132,26 +132,26 @@ static int ql_sem_trylock(struct ql_adapter *qdev, u32 sem_mask) return -EINVAL; } - ql_write32(qdev, SEM, sem_bits | sem_mask); - return !(ql_read32(qdev, SEM) & sem_bits); + qlge_write32(qdev, SEM, sem_bits | sem_mask); + return !(qlge_read32(qdev, SEM) & sem_bits); } -int ql_sem_spinlock(struct ql_adapter *qdev, u32 sem_mask) +int qlge_sem_spinlock(struct qlge_adapter *qdev, u32 sem_mask) { unsigned int wait_count = 30; do { - if (!ql_sem_trylock(qdev, sem_mask)) + if (!qlge_sem_trylock(qdev, sem_mask)) return 0; udelay(100); } while (--wait_count); return -ETIMEDOUT; } -void ql_sem_unlock(struct ql_adapter *qdev, u32 sem_mask) +void qlge_sem_unlock(struct qlge_adapter *qdev, u32 sem_mask) { - ql_write32(qdev, SEM, sem_mask); - ql_read32(qdev, SEM); /* flush */ + qlge_write32(qdev, SEM, sem_mask); + qlge_read32(qdev, SEM); /* flush */ } /* This function waits for a specific bit to come ready @@ -159,13 +159,13 @@ void ql_sem_unlock(struct ql_adapter *qdev, u32 sem_mask) * process, but is also used in kernel thread API such as * netdev->set_multi, netdev->set_mac_address, netdev->vlan_rx_add_vid. */ -int ql_wait_reg_rdy(struct ql_adapter *qdev, u32 reg, u32 bit, u32 err_bit) +int qlge_wait_reg_rdy(struct qlge_adapter *qdev, u32 reg, u32 bit, u32 err_bit) { u32 temp; int count; for (count = 0; count < UDELAY_COUNT; count++) { - temp = ql_read32(qdev, reg); + temp = qlge_read32(qdev, reg); /* check for errors */ if (temp & err_bit) { @@ -186,13 +186,13 @@ int ql_wait_reg_rdy(struct ql_adapter *qdev, u32 reg, u32 bit, u32 err_bit) /* The CFG register is used to download TX and RX control blocks * to the chip. This function waits for an operation to complete. */ -static int ql_wait_cfg(struct ql_adapter *qdev, u32 bit) +static int qlge_wait_cfg(struct qlge_adapter *qdev, u32 bit) { int count; u32 temp; for (count = 0; count < UDELAY_COUNT; count++) { - temp = ql_read32(qdev, CFG); + temp = qlge_read32(qdev, CFG); if (temp & CFG_LE) return -EIO; if (!(temp & bit)) @@ -205,8 +205,8 @@ static int ql_wait_cfg(struct ql_adapter *qdev, u32 bit) /* Used to issue init control blocks to hw. Maps control block, * sets address, triggers download, waits for completion. */ -int ql_write_cfg(struct ql_adapter *qdev, void *ptr, int size, u32 bit, - u16 q_id) +int qlge_write_cfg(struct qlge_adapter *qdev, void *ptr, int size, u32 bit, + u16 q_id) { u64 map; int status = 0; @@ -225,38 +225,38 @@ int ql_write_cfg(struct ql_adapter *qdev, void *ptr, int size, u32 bit, return -ENOMEM; } - status = ql_sem_spinlock(qdev, SEM_ICB_MASK); + status = qlge_sem_spinlock(qdev, SEM_ICB_MASK); if (status) goto lock_failed; - status = ql_wait_cfg(qdev, bit); + status = qlge_wait_cfg(qdev, bit); if (status) { netif_err(qdev, ifup, qdev->ndev, "Timed out waiting for CFG to come ready.\n"); goto exit; } - ql_write32(qdev, ICB_L, (u32)map); - ql_write32(qdev, ICB_H, (u32)(map >> 32)); + qlge_write32(qdev, ICB_L, (u32)map); + qlge_write32(qdev, ICB_H, (u32)(map >> 32)); mask = CFG_Q_MASK | (bit << 16); value = bit | (q_id << CFG_Q_SHIFT); - ql_write32(qdev, CFG, (mask | value)); + qlge_write32(qdev, CFG, (mask | value)); /* * Wait for the bit to clear after signaling hw. */ - status = ql_wait_cfg(qdev, bit); + status = qlge_wait_cfg(qdev, bit); exit: - ql_sem_unlock(qdev, SEM_ICB_MASK); /* does flush too */ + qlge_sem_unlock(qdev, SEM_ICB_MASK); /* does flush too */ lock_failed: dma_unmap_single(&qdev->pdev->dev, map, size, direction); return status; } /* Get a specific MAC address from the CAM. Used for debug and reg dump. */ -int ql_get_mac_addr_reg(struct ql_adapter *qdev, u32 type, u16 index, - u32 *value) +int qlge_get_mac_addr_reg(struct qlge_adapter *qdev, u32 type, u16 index, + u32 *value) { u32 offset = 0; int status; @@ -264,46 +264,46 @@ int ql_get_mac_addr_reg(struct ql_adapter *qdev, u32 type, u16 index, switch (type) { case MAC_ADDR_TYPE_MULTI_MAC: case MAC_ADDR_TYPE_CAM_MAC: { - status = ql_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MW, 0); + status = qlge_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MW, 0); if (status) break; - ql_write32(qdev, MAC_ADDR_IDX, - (offset++) | /* offset */ + qlge_write32(qdev, MAC_ADDR_IDX, + (offset++) | /* offset */ (index << MAC_ADDR_IDX_SHIFT) | /* index */ MAC_ADDR_ADR | MAC_ADDR_RS | type); /* type */ - status = ql_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MR, 0); + status = qlge_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MR, 0); if (status) break; - *value++ = ql_read32(qdev, MAC_ADDR_DATA); - status = ql_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MW, 0); + *value++ = qlge_read32(qdev, MAC_ADDR_DATA); + status = qlge_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MW, 0); if (status) break; - ql_write32(qdev, MAC_ADDR_IDX, - (offset++) | /* offset */ + qlge_write32(qdev, MAC_ADDR_IDX, + (offset++) | /* offset */ (index << MAC_ADDR_IDX_SHIFT) | /* index */ MAC_ADDR_ADR | MAC_ADDR_RS | type); /* type */ - status = ql_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MR, 0); + status = qlge_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MR, 0); if (status) break; - *value++ = ql_read32(qdev, MAC_ADDR_DATA); + *value++ = qlge_read32(qdev, MAC_ADDR_DATA); if (type == MAC_ADDR_TYPE_CAM_MAC) { - status = ql_wait_reg_rdy(qdev, MAC_ADDR_IDX, - MAC_ADDR_MW, 0); + status = qlge_wait_reg_rdy(qdev, MAC_ADDR_IDX, + MAC_ADDR_MW, 0); if (status) break; - ql_write32(qdev, MAC_ADDR_IDX, - (offset++) | /* offset */ + qlge_write32(qdev, MAC_ADDR_IDX, + (offset++) | /* offset */ (index << MAC_ADDR_IDX_SHIFT) | /* index */ MAC_ADDR_ADR | MAC_ADDR_RS | type); /* type */ - status = ql_wait_reg_rdy(qdev, MAC_ADDR_IDX, - MAC_ADDR_MR, 0); + status = qlge_wait_reg_rdy(qdev, MAC_ADDR_IDX, + MAC_ADDR_MR, 0); if (status) break; - *value++ = ql_read32(qdev, MAC_ADDR_DATA); + *value++ = qlge_read32(qdev, MAC_ADDR_DATA); } break; } @@ -320,8 +320,8 @@ int ql_get_mac_addr_reg(struct ql_adapter *qdev, u32 type, u16 index, /* Set up a MAC, multicast or VLAN address for the * inbound frame matching. */ -static int ql_set_mac_addr_reg(struct ql_adapter *qdev, u8 *addr, u32 type, - u16 index) +static int qlge_set_mac_addr_reg(struct qlge_adapter *qdev, u8 *addr, u32 type, + u16 index) { u32 offset = 0; int status = 0; @@ -332,22 +332,22 @@ static int ql_set_mac_addr_reg(struct ql_adapter *qdev, u8 *addr, u32 type, u32 lower = (addr[2] << 24) | (addr[3] << 16) | (addr[4] << 8) | (addr[5]); - status = ql_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MW, 0); + status = qlge_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MW, 0); if (status) break; - ql_write32(qdev, MAC_ADDR_IDX, - (offset++) | (index << MAC_ADDR_IDX_SHIFT) | type | + qlge_write32(qdev, MAC_ADDR_IDX, + (offset++) | (index << MAC_ADDR_IDX_SHIFT) | type | MAC_ADDR_E); - ql_write32(qdev, MAC_ADDR_DATA, lower); - status = ql_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MW, 0); + qlge_write32(qdev, MAC_ADDR_DATA, lower); + status = qlge_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MW, 0); if (status) break; - ql_write32(qdev, MAC_ADDR_IDX, - (offset++) | (index << MAC_ADDR_IDX_SHIFT) | type | + qlge_write32(qdev, MAC_ADDR_IDX, + (offset++) | (index << MAC_ADDR_IDX_SHIFT) | type | MAC_ADDR_E); - ql_write32(qdev, MAC_ADDR_DATA, upper); - status = ql_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MW, 0); + qlge_write32(qdev, MAC_ADDR_DATA, upper); + status = qlge_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MW, 0); break; } case MAC_ADDR_TYPE_CAM_MAC: { @@ -355,27 +355,27 @@ static int ql_set_mac_addr_reg(struct ql_adapter *qdev, u8 *addr, u32 type, u32 upper = (addr[0] << 8) | addr[1]; u32 lower = (addr[2] << 24) | (addr[3] << 16) | (addr[4] << 8) | (addr[5]); - status = ql_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MW, 0); + status = qlge_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MW, 0); if (status) break; - ql_write32(qdev, MAC_ADDR_IDX, - (offset++) | /* offset */ + qlge_write32(qdev, MAC_ADDR_IDX, + (offset++) | /* offset */ (index << MAC_ADDR_IDX_SHIFT) | /* index */ type); /* type */ - ql_write32(qdev, MAC_ADDR_DATA, lower); - status = ql_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MW, 0); + qlge_write32(qdev, MAC_ADDR_DATA, lower); + status = qlge_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MW, 0); if (status) break; - ql_write32(qdev, MAC_ADDR_IDX, - (offset++) | /* offset */ + qlge_write32(qdev, MAC_ADDR_IDX, + (offset++) | /* offset */ (index << MAC_ADDR_IDX_SHIFT) | /* index */ type); /* type */ - ql_write32(qdev, MAC_ADDR_DATA, upper); - status = ql_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MW, 0); + qlge_write32(qdev, MAC_ADDR_DATA, upper); + status = qlge_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MW, 0); if (status) break; - ql_write32(qdev, MAC_ADDR_IDX, - (offset) | /* offset */ + qlge_write32(qdev, MAC_ADDR_IDX, + (offset) | /* offset */ (index << MAC_ADDR_IDX_SHIFT) | /* index */ type); /* type */ /* This field should also include the queue id @@ -388,7 +388,7 @@ static int ql_set_mac_addr_reg(struct ql_adapter *qdev, u8 *addr, u32 type, if (qdev->ndev->features & NETIF_F_HW_VLAN_CTAG_RX) cam_output |= CAM_OUT_RV; /* route to NIC core */ - ql_write32(qdev, MAC_ADDR_DATA, cam_output); + qlge_write32(qdev, MAC_ADDR_DATA, cam_output); break; } case MAC_ADDR_TYPE_VLAN: { @@ -398,11 +398,11 @@ static int ql_set_mac_addr_reg(struct ql_adapter *qdev, u8 *addr, u32 type, * addressing. It's either MAC_ADDR_E on or off. * That's bit-27 we're talking about. */ - status = ql_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MW, 0); + status = qlge_wait_reg_rdy(qdev, MAC_ADDR_IDX, MAC_ADDR_MW, 0); if (status) break; - ql_write32(qdev, MAC_ADDR_IDX, - offset | /* offset */ + qlge_write32(qdev, MAC_ADDR_IDX, + offset | /* offset */ (index << MAC_ADDR_IDX_SHIFT) | /* index */ type | /* type */ enable_bit); /* enable/disable */ @@ -421,7 +421,7 @@ static int ql_set_mac_addr_reg(struct ql_adapter *qdev, u8 *addr, u32 type, * have to clear it to prevent wrong frame routing * especially in a bonding environment. */ -static int ql_set_mac_addr(struct ql_adapter *qdev, int set) +static int qlge_set_mac_addr(struct qlge_adapter *qdev, int set) { int status; char zero_mac_addr[ETH_ALEN]; @@ -437,50 +437,50 @@ static int ql_set_mac_addr(struct ql_adapter *qdev, int set) netif_printk(qdev, ifup, KERN_DEBUG, qdev->ndev, "Clearing MAC address\n"); } - status = ql_sem_spinlock(qdev, SEM_MAC_ADDR_MASK); + status = qlge_sem_spinlock(qdev, SEM_MAC_ADDR_MASK); if (status) return status; - status = ql_set_mac_addr_reg(qdev, (u8 *)addr, - MAC_ADDR_TYPE_CAM_MAC, - qdev->func * MAX_CQ); - ql_sem_unlock(qdev, SEM_MAC_ADDR_MASK); + status = qlge_set_mac_addr_reg(qdev, (u8 *)addr, + MAC_ADDR_TYPE_CAM_MAC, + qdev->func * MAX_CQ); + qlge_sem_unlock(qdev, SEM_MAC_ADDR_MASK); if (status) netif_err(qdev, ifup, qdev->ndev, "Failed to init mac address.\n"); return status; } -void ql_link_on(struct ql_adapter *qdev) +void qlge_link_on(struct qlge_adapter *qdev) { netif_err(qdev, link, qdev->ndev, "Link is up.\n"); netif_carrier_on(qdev->ndev); - ql_set_mac_addr(qdev, 1); + qlge_set_mac_addr(qdev, 1); } -void ql_link_off(struct ql_adapter *qdev) +void qlge_link_off(struct qlge_adapter *qdev) { netif_err(qdev, link, qdev->ndev, "Link is down.\n"); netif_carrier_off(qdev->ndev); - ql_set_mac_addr(qdev, 0); + qlge_set_mac_addr(qdev, 0); } /* Get a specific frame routing value from the CAM. * Used for debug and reg dump. */ -int ql_get_routing_reg(struct ql_adapter *qdev, u32 index, u32 *value) +int qlge_get_routing_reg(struct qlge_adapter *qdev, u32 index, u32 *value) { int status = 0; - status = ql_wait_reg_rdy(qdev, RT_IDX, RT_IDX_MW, 0); + status = qlge_wait_reg_rdy(qdev, RT_IDX, RT_IDX_MW, 0); if (status) goto exit; - ql_write32(qdev, RT_IDX, - RT_IDX_TYPE_NICQ | RT_IDX_RS | (index << RT_IDX_IDX_SHIFT)); - status = ql_wait_reg_rdy(qdev, RT_IDX, RT_IDX_MR, 0); + qlge_write32(qdev, RT_IDX, + RT_IDX_TYPE_NICQ | RT_IDX_RS | (index << RT_IDX_IDX_SHIFT)); + status = qlge_wait_reg_rdy(qdev, RT_IDX, RT_IDX_MR, 0); if (status) goto exit; - *value = ql_read32(qdev, RT_DATA); + *value = qlge_read32(qdev, RT_DATA); exit: return status; } @@ -490,8 +490,8 @@ int ql_get_routing_reg(struct ql_adapter *qdev, u32 index, u32 *value) * multicast/error frames to the default queue for slow handling, * and CAM hit/RSS frames to the fast handling queues. */ -static int ql_set_routing_reg(struct ql_adapter *qdev, u32 index, u32 mask, - int enable) +static int qlge_set_routing_reg(struct qlge_adapter *qdev, u32 index, u32 mask, + int enable) { int status = -EINVAL; /* Return error if no mask match. */ u32 value = 0; @@ -577,50 +577,50 @@ static int ql_set_routing_reg(struct ql_adapter *qdev, u32 index, u32 mask, } if (value) { - status = ql_wait_reg_rdy(qdev, RT_IDX, RT_IDX_MW, 0); + status = qlge_wait_reg_rdy(qdev, RT_IDX, RT_IDX_MW, 0); if (status) goto exit; value |= (enable ? RT_IDX_E : 0); - ql_write32(qdev, RT_IDX, value); - ql_write32(qdev, RT_DATA, enable ? mask : 0); + qlge_write32(qdev, RT_IDX, value); + qlge_write32(qdev, RT_DATA, enable ? mask : 0); } exit: return status; } -static void ql_enable_interrupts(struct ql_adapter *qdev) +static void qlge_enable_interrupts(struct qlge_adapter *qdev) { - ql_write32(qdev, INTR_EN, (INTR_EN_EI << 16) | INTR_EN_EI); + qlge_write32(qdev, INTR_EN, (INTR_EN_EI << 16) | INTR_EN_EI); } -static void ql_disable_interrupts(struct ql_adapter *qdev) +static void qlge_disable_interrupts(struct qlge_adapter *qdev) { - ql_write32(qdev, INTR_EN, (INTR_EN_EI << 16)); + qlge_write32(qdev, INTR_EN, (INTR_EN_EI << 16)); } -static void ql_enable_completion_interrupt(struct ql_adapter *qdev, u32 intr) +static void qlge_enable_completion_interrupt(struct qlge_adapter *qdev, u32 intr) { struct intr_context *ctx = &qdev->intr_context[intr]; - ql_write32(qdev, INTR_EN, ctx->intr_en_mask); + qlge_write32(qdev, INTR_EN, ctx->intr_en_mask); } -static void ql_disable_completion_interrupt(struct ql_adapter *qdev, u32 intr) +static void qlge_disable_completion_interrupt(struct qlge_adapter *qdev, u32 intr) { struct intr_context *ctx = &qdev->intr_context[intr]; - ql_write32(qdev, INTR_EN, ctx->intr_dis_mask); + qlge_write32(qdev, INTR_EN, ctx->intr_dis_mask); } -static void ql_enable_all_completion_interrupts(struct ql_adapter *qdev) +static void qlge_enable_all_completion_interrupts(struct qlge_adapter *qdev) { int i; for (i = 0; i < qdev->intr_count; i++) - ql_enable_completion_interrupt(qdev, i); + qlge_enable_completion_interrupt(qdev, i); } -static int ql_validate_flash(struct ql_adapter *qdev, u32 size, const char *str) +static int qlge_validate_flash(struct qlge_adapter *qdev, u32 size, const char *str) { int status, i; u16 csum = 0; @@ -642,31 +642,31 @@ static int ql_validate_flash(struct ql_adapter *qdev, u32 size, const char *str) return csum; } -static int ql_read_flash_word(struct ql_adapter *qdev, int offset, __le32 *data) +static int qlge_read_flash_word(struct qlge_adapter *qdev, int offset, __le32 *data) { int status = 0; /* wait for reg to come ready */ - status = ql_wait_reg_rdy(qdev, - FLASH_ADDR, FLASH_ADDR_RDY, FLASH_ADDR_ERR); + status = qlge_wait_reg_rdy(qdev, + FLASH_ADDR, FLASH_ADDR_RDY, FLASH_ADDR_ERR); if (status) goto exit; /* set up for reg read */ - ql_write32(qdev, FLASH_ADDR, FLASH_ADDR_R | offset); + qlge_write32(qdev, FLASH_ADDR, FLASH_ADDR_R | offset); /* wait for reg to come ready */ - status = ql_wait_reg_rdy(qdev, - FLASH_ADDR, FLASH_ADDR_RDY, FLASH_ADDR_ERR); + status = qlge_wait_reg_rdy(qdev, + FLASH_ADDR, FLASH_ADDR_RDY, FLASH_ADDR_ERR); if (status) goto exit; /* This data is stored on flash as an array of - * __le32. Since ql_read32() returns cpu endian + * __le32. Since qlge_read32() returns cpu endian * we need to swap it back. */ - *data = cpu_to_le32(ql_read32(qdev, FLASH_DATA)); + *data = cpu_to_le32(qlge_read32(qdev, FLASH_DATA)); exit: return status; } -static int ql_get_8000_flash_params(struct ql_adapter *qdev) +static int qlge_get_8000_flash_params(struct qlge_adapter *qdev) { u32 i, size; int status; @@ -682,12 +682,12 @@ static int ql_get_8000_flash_params(struct ql_adapter *qdev) else offset = FUNC1_FLASH_OFFSET / sizeof(u32); - if (ql_sem_spinlock(qdev, SEM_FLASH_MASK)) + if (qlge_sem_spinlock(qdev, SEM_FLASH_MASK)) return -ETIMEDOUT; size = sizeof(struct flash_params_8000) / sizeof(u32); for (i = 0; i < size; i++, p++) { - status = ql_read_flash_word(qdev, i + offset, p); + status = qlge_read_flash_word(qdev, i + offset, p); if (status) { netif_err(qdev, ifup, qdev->ndev, "Error reading flash.\n"); @@ -695,8 +695,8 @@ static int ql_get_8000_flash_params(struct ql_adapter *qdev) } } - status = ql_validate_flash(qdev, - sizeof(struct flash_params_8000) / + status = qlge_validate_flash(qdev, + sizeof(struct flash_params_8000) / sizeof(u16), "8000"); if (status) { @@ -728,11 +728,11 @@ static int ql_get_8000_flash_params(struct ql_adapter *qdev) qdev->ndev->addr_len); exit: - ql_sem_unlock(qdev, SEM_FLASH_MASK); + qlge_sem_unlock(qdev, SEM_FLASH_MASK); return status; } -static int ql_get_8012_flash_params(struct ql_adapter *qdev) +static int qlge_get_8012_flash_params(struct qlge_adapter *qdev) { int i; int status; @@ -746,11 +746,11 @@ static int ql_get_8012_flash_params(struct ql_adapter *qdev) if (qdev->port) offset = size; - if (ql_sem_spinlock(qdev, SEM_FLASH_MASK)) + if (qlge_sem_spinlock(qdev, SEM_FLASH_MASK)) return -ETIMEDOUT; for (i = 0; i < size; i++, p++) { - status = ql_read_flash_word(qdev, i + offset, p); + status = qlge_read_flash_word(qdev, i + offset, p); if (status) { netif_err(qdev, ifup, qdev->ndev, "Error reading flash.\n"); @@ -758,10 +758,10 @@ static int ql_get_8012_flash_params(struct ql_adapter *qdev) } } - status = ql_validate_flash(qdev, - sizeof(struct flash_params_8012) / - sizeof(u16), - "8012"); + status = qlge_validate_flash(qdev, + sizeof(struct flash_params_8012) / + sizeof(u16), + "8012"); if (status) { netif_err(qdev, ifup, qdev->ndev, "Invalid flash.\n"); status = -EINVAL; @@ -778,7 +778,7 @@ static int ql_get_8012_flash_params(struct ql_adapter *qdev) qdev->ndev->addr_len); exit: - ql_sem_unlock(qdev, SEM_FLASH_MASK); + qlge_sem_unlock(qdev, SEM_FLASH_MASK); return status; } @@ -786,18 +786,18 @@ static int ql_get_8012_flash_params(struct ql_adapter *qdev) * register pair. Each read/write requires us to wait for the ready * bit before reading/writing the data. */ -static int ql_write_xgmac_reg(struct ql_adapter *qdev, u32 reg, u32 data) +static int qlge_write_xgmac_reg(struct qlge_adapter *qdev, u32 reg, u32 data) { int status; /* wait for reg to come ready */ - status = ql_wait_reg_rdy(qdev, - XGMAC_ADDR, XGMAC_ADDR_RDY, XGMAC_ADDR_XME); + status = qlge_wait_reg_rdy(qdev, + XGMAC_ADDR, XGMAC_ADDR_RDY, XGMAC_ADDR_XME); if (status) return status; /* write the data to the data reg */ - ql_write32(qdev, XGMAC_DATA, data); + qlge_write32(qdev, XGMAC_DATA, data); /* trigger the write */ - ql_write32(qdev, XGMAC_ADDR, reg); + qlge_write32(qdev, XGMAC_ADDR, reg); return status; } @@ -805,39 +805,39 @@ static int ql_write_xgmac_reg(struct ql_adapter *qdev, u32 reg, u32 data) * register pair. Each read/write requires us to wait for the ready * bit before reading/writing the data. */ -int ql_read_xgmac_reg(struct ql_adapter *qdev, u32 reg, u32 *data) +int qlge_read_xgmac_reg(struct qlge_adapter *qdev, u32 reg, u32 *data) { int status = 0; /* wait for reg to come ready */ - status = ql_wait_reg_rdy(qdev, - XGMAC_ADDR, XGMAC_ADDR_RDY, XGMAC_ADDR_XME); + status = qlge_wait_reg_rdy(qdev, + XGMAC_ADDR, XGMAC_ADDR_RDY, XGMAC_ADDR_XME); if (status) goto exit; /* set up for reg read */ - ql_write32(qdev, XGMAC_ADDR, reg | XGMAC_ADDR_R); + qlge_write32(qdev, XGMAC_ADDR, reg | XGMAC_ADDR_R); /* wait for reg to come ready */ - status = ql_wait_reg_rdy(qdev, - XGMAC_ADDR, XGMAC_ADDR_RDY, XGMAC_ADDR_XME); + status = qlge_wait_reg_rdy(qdev, + XGMAC_ADDR, XGMAC_ADDR_RDY, XGMAC_ADDR_XME); if (status) goto exit; /* get the data */ - *data = ql_read32(qdev, XGMAC_DATA); + *data = qlge_read32(qdev, XGMAC_DATA); exit: return status; } /* This is used for reading the 64-bit statistics regs. */ -int ql_read_xgmac_reg64(struct ql_adapter *qdev, u32 reg, u64 *data) +int qlge_read_xgmac_reg64(struct qlge_adapter *qdev, u32 reg, u64 *data) { int status = 0; u32 hi = 0; u32 lo = 0; - status = ql_read_xgmac_reg(qdev, reg, &lo); + status = qlge_read_xgmac_reg(qdev, reg, &lo); if (status) goto exit; - status = ql_read_xgmac_reg(qdev, reg + 4, &hi); + status = qlge_read_xgmac_reg(qdev, reg + 4, &hi); if (status) goto exit; @@ -847,17 +847,17 @@ int ql_read_xgmac_reg64(struct ql_adapter *qdev, u32 reg, u64 *data) return status; } -static int ql_8000_port_initialize(struct ql_adapter *qdev) +static int qlge_8000_port_initialize(struct qlge_adapter *qdev) { int status; /* * Get MPI firmware version for driver banner * and ethool info. */ - status = ql_mb_about_fw(qdev); + status = qlge_mb_about_fw(qdev); if (status) goto exit; - status = ql_mb_get_fw_state(qdev); + status = qlge_mb_get_fw_state(qdev); if (status) goto exit; /* Wake up a worker to get/set the TX/RX frame sizes. */ @@ -872,18 +872,18 @@ static int ql_8000_port_initialize(struct ql_adapter *qdev) * This functionality may be done in the MPI firmware at a * later date. */ -static int ql_8012_port_initialize(struct ql_adapter *qdev) +static int qlge_8012_port_initialize(struct qlge_adapter *qdev) { int status = 0; u32 data; - if (ql_sem_trylock(qdev, qdev->xg_sem_mask)) { + if (qlge_sem_trylock(qdev, qdev->xg_sem_mask)) { /* Another function has the semaphore, so * wait for the port init bit to come ready. */ netif_info(qdev, link, qdev->ndev, "Another function has the semaphore, so wait for the port init bit to come ready.\n"); - status = ql_wait_reg_rdy(qdev, STS, qdev->port_init, 0); + status = qlge_wait_reg_rdy(qdev, STS, qdev->port_init, 0); if (status) { netif_crit(qdev, link, qdev->ndev, "Port initialize timed out.\n"); @@ -893,11 +893,11 @@ static int ql_8012_port_initialize(struct ql_adapter *qdev) netif_info(qdev, link, qdev->ndev, "Got xgmac semaphore!.\n"); /* Set the core reset. */ - status = ql_read_xgmac_reg(qdev, GLOBAL_CFG, &data); + status = qlge_read_xgmac_reg(qdev, GLOBAL_CFG, &data); if (status) goto end; data |= GLOBAL_CFG_RESET; - status = ql_write_xgmac_reg(qdev, GLOBAL_CFG, data); + status = qlge_write_xgmac_reg(qdev, GLOBAL_CFG, data); if (status) goto end; @@ -906,48 +906,48 @@ static int ql_8012_port_initialize(struct ql_adapter *qdev) data |= GLOBAL_CFG_JUMBO; /* Turn on jumbo. */ data |= GLOBAL_CFG_TX_STAT_EN; data |= GLOBAL_CFG_RX_STAT_EN; - status = ql_write_xgmac_reg(qdev, GLOBAL_CFG, data); + status = qlge_write_xgmac_reg(qdev, GLOBAL_CFG, data); if (status) goto end; /* Enable transmitter, and clear it's reset. */ - status = ql_read_xgmac_reg(qdev, TX_CFG, &data); + status = qlge_read_xgmac_reg(qdev, TX_CFG, &data); if (status) goto end; data &= ~TX_CFG_RESET; /* Clear the TX MAC reset. */ data |= TX_CFG_EN; /* Enable the transmitter. */ - status = ql_write_xgmac_reg(qdev, TX_CFG, data); + status = qlge_write_xgmac_reg(qdev, TX_CFG, data); if (status) goto end; /* Enable receiver and clear it's reset. */ - status = ql_read_xgmac_reg(qdev, RX_CFG, &data); + status = qlge_read_xgmac_reg(qdev, RX_CFG, &data); if (status) goto end; data &= ~RX_CFG_RESET; /* Clear the RX MAC reset. */ data |= RX_CFG_EN; /* Enable the receiver. */ - status = ql_write_xgmac_reg(qdev, RX_CFG, data); + status = qlge_write_xgmac_reg(qdev, RX_CFG, data); if (status) goto end; /* Turn on jumbo. */ status = - ql_write_xgmac_reg(qdev, MAC_TX_PARAMS, MAC_TX_PARAMS_JUMBO | (0x2580 << 16)); + qlge_write_xgmac_reg(qdev, MAC_TX_PARAMS, MAC_TX_PARAMS_JUMBO | (0x2580 << 16)); if (status) goto end; status = - ql_write_xgmac_reg(qdev, MAC_RX_PARAMS, 0x2580); + qlge_write_xgmac_reg(qdev, MAC_RX_PARAMS, 0x2580); if (status) goto end; /* Signal to the world that the port is enabled. */ - ql_write32(qdev, STS, ((qdev->port_init << 16) | qdev->port_init)); + qlge_write32(qdev, STS, ((qdev->port_init << 16) | qdev->port_init)); end: - ql_sem_unlock(qdev, qdev->xg_sem_mask); + qlge_sem_unlock(qdev, qdev->xg_sem_mask); return status; } -static inline unsigned int ql_lbq_block_size(struct ql_adapter *qdev) +static inline unsigned int qlge_lbq_block_size(struct qlge_adapter *qdev) { return PAGE_SIZE << qdev->lbq_buf_order; } @@ -962,8 +962,8 @@ static struct qlge_bq_desc *qlge_get_curr_buf(struct qlge_bq *bq) return bq_desc; } -static struct qlge_bq_desc *ql_get_curr_lchunk(struct ql_adapter *qdev, - struct rx_ring *rx_ring) +static struct qlge_bq_desc *qlge_get_curr_lchunk(struct qlge_adapter *qdev, + struct rx_ring *rx_ring) { struct qlge_bq_desc *lbq_desc = qlge_get_curr_buf(&rx_ring->lbq); @@ -971,17 +971,17 @@ static struct qlge_bq_desc *ql_get_curr_lchunk(struct ql_adapter *qdev, qdev->lbq_buf_size, DMA_FROM_DEVICE); if ((lbq_desc->p.pg_chunk.offset + qdev->lbq_buf_size) == - ql_lbq_block_size(qdev)) { + qlge_lbq_block_size(qdev)) { /* last chunk of the master page */ dma_unmap_page(&qdev->pdev->dev, lbq_desc->dma_addr, - ql_lbq_block_size(qdev), DMA_FROM_DEVICE); + qlge_lbq_block_size(qdev), DMA_FROM_DEVICE); } return lbq_desc; } /* Update an rx ring index. */ -static void ql_update_cq(struct rx_ring *rx_ring) +static void qlge_update_cq(struct rx_ring *rx_ring) { rx_ring->cnsmr_idx++; rx_ring->curr_entry++; @@ -991,9 +991,9 @@ static void ql_update_cq(struct rx_ring *rx_ring) } } -static void ql_write_cq_idx(struct rx_ring *rx_ring) +static void qlge_write_cq_idx(struct rx_ring *rx_ring) { - ql_write_db_reg(rx_ring->cnsmr_idx, rx_ring->cnsmr_idx_db_reg); + qlge_write_db_reg(rx_ring->cnsmr_idx, rx_ring->cnsmr_idx_db_reg); } static const char * const bq_type_name[] = { @@ -1005,7 +1005,7 @@ static const char * const bq_type_name[] = { static int qlge_refill_sb(struct rx_ring *rx_ring, struct qlge_bq_desc *sbq_desc, gfp_t gfp) { - struct ql_adapter *qdev = rx_ring->qdev; + struct qlge_adapter *qdev = rx_ring->qdev; struct sk_buff *skb; if (sbq_desc->p.skb) @@ -1038,7 +1038,7 @@ static int qlge_refill_sb(struct rx_ring *rx_ring, static int qlge_refill_lb(struct rx_ring *rx_ring, struct qlge_bq_desc *lbq_desc, gfp_t gfp) { - struct ql_adapter *qdev = rx_ring->qdev; + struct qlge_adapter *qdev = rx_ring->qdev; struct qlge_page_chunk *master_chunk = &rx_ring->master_chunk; if (!master_chunk->page) { @@ -1049,7 +1049,7 @@ static int qlge_refill_lb(struct rx_ring *rx_ring, if (unlikely(!page)) return -ENOMEM; dma_addr = dma_map_page(&qdev->pdev->dev, page, 0, - ql_lbq_block_size(qdev), + qlge_lbq_block_size(qdev), DMA_FROM_DEVICE); if (dma_mapping_error(&qdev->pdev->dev, dma_addr)) { __free_pages(page, qdev->lbq_buf_order); @@ -1072,7 +1072,7 @@ static int qlge_refill_lb(struct rx_ring *rx_ring, * buffer get. */ master_chunk->offset += qdev->lbq_buf_size; - if (master_chunk->offset == ql_lbq_block_size(qdev)) { + if (master_chunk->offset == qlge_lbq_block_size(qdev)) { master_chunk->page = NULL; } else { master_chunk->va += qdev->lbq_buf_size; @@ -1086,7 +1086,7 @@ static int qlge_refill_lb(struct rx_ring *rx_ring, static int qlge_refill_bq(struct qlge_bq *bq, gfp_t gfp) { struct rx_ring *rx_ring = QLGE_BQ_CONTAINER(bq); - struct ql_adapter *qdev = rx_ring->qdev; + struct qlge_adapter *qdev = rx_ring->qdev; struct qlge_bq_desc *bq_desc; int refill_count; int retval; @@ -1132,7 +1132,7 @@ static int qlge_refill_bq(struct qlge_bq *bq, gfp_t gfp) "ring %u %s: updating prod idx = %d.\n", rx_ring->cq_id, bq_type_name[bq->type], i); - ql_write_db_reg(i, bq->prod_idx_db_reg); + qlge_write_db_reg(i, bq->prod_idx_db_reg); } bq->next_to_use = i; } @@ -1140,8 +1140,8 @@ static int qlge_refill_bq(struct qlge_bq *bq, gfp_t gfp) return retval; } -static void ql_update_buffer_queues(struct rx_ring *rx_ring, gfp_t gfp, - unsigned long delay) +static void qlge_update_buffer_queues(struct rx_ring *rx_ring, gfp_t gfp, + unsigned long delay) { bool sbq_fail, lbq_fail; @@ -1172,7 +1172,7 @@ static void qlge_slow_refill(struct work_struct *work) struct napi_struct *napi = &rx_ring->napi; napi_disable(napi); - ql_update_buffer_queues(rx_ring, GFP_KERNEL, HZ / 2); + qlge_update_buffer_queues(rx_ring, GFP_KERNEL, HZ / 2); napi_enable(napi); local_bh_disable(); @@ -1187,8 +1187,8 @@ static void qlge_slow_refill(struct work_struct *work) /* Unmaps tx buffers. Can be called from send() if a pci mapping * fails at some stage, or from the interrupt when a tx completes. */ -static void ql_unmap_send(struct ql_adapter *qdev, - struct tx_ring_desc *tx_ring_desc, int mapped) +static void qlge_unmap_send(struct qlge_adapter *qdev, + struct tx_ring_desc *tx_ring_desc, int mapped) { int i; @@ -1229,9 +1229,9 @@ static void ql_unmap_send(struct ql_adapter *qdev, /* Map the buffers for this transmit. This will return * NETDEV_TX_BUSY or NETDEV_TX_OK based on success. */ -static int ql_map_send(struct ql_adapter *qdev, - struct ob_mac_iocb_req *mac_iocb_ptr, - struct sk_buff *skb, struct tx_ring_desc *tx_ring_desc) +static int qlge_map_send(struct qlge_adapter *qdev, + struct qlge_ob_mac_iocb_req *mac_iocb_ptr, + struct sk_buff *skb, struct tx_ring_desc *tx_ring_desc) { int len = skb_headlen(skb); dma_addr_t map; @@ -1294,7 +1294,7 @@ static int ql_map_send(struct ql_adapter *qdev, */ /* Tack on the OAL in the eighth segment of IOCB. */ map = dma_map_single(&qdev->pdev->dev, &tx_ring_desc->oal, - sizeof(struct oal), + sizeof(struct qlge_oal), DMA_TO_DEVICE); err = dma_mapping_error(&qdev->pdev->dev, map); if (err) { @@ -1316,7 +1316,7 @@ static int ql_map_send(struct ql_adapter *qdev, dma_unmap_addr_set(&tx_ring_desc->map[map_idx], mapaddr, map); dma_unmap_len_set(&tx_ring_desc->map[map_idx], maplen, - sizeof(struct oal)); + sizeof(struct qlge_oal)); tbd = (struct tx_buf_desc *)&tx_ring_desc->oal; map_idx++; } @@ -1351,13 +1351,13 @@ static int ql_map_send(struct ql_adapter *qdev, * we pass in the number of frags that mapped successfully * so they can be umapped. */ - ql_unmap_send(qdev, tx_ring_desc, map_idx); + qlge_unmap_send(qdev, tx_ring_desc, map_idx); return NETDEV_TX_BUSY; } /* Categorizing receive firmware frame errors */ -static void ql_categorize_rx_err(struct ql_adapter *qdev, u8 rx_err, - struct rx_ring *rx_ring) +static void qlge_categorize_rx_err(struct qlge_adapter *qdev, u8 rx_err, + struct rx_ring *rx_ring) { struct nic_stats *stats = &qdev->nic_stats; @@ -1389,12 +1389,12 @@ static void ql_categorize_rx_err(struct ql_adapter *qdev, u8 rx_err, } /** - * ql_update_mac_hdr_len - helper routine to update the mac header length + * qlge_update_mac_hdr_len - helper routine to update the mac header length * based on vlan tags if present */ -static void ql_update_mac_hdr_len(struct ql_adapter *qdev, - struct ib_mac_iocb_rsp *ib_mac_rsp, - void *page, size_t *len) +static void qlge_update_mac_hdr_len(struct qlge_adapter *qdev, + struct qlge_ib_mac_iocb_rsp *ib_mac_rsp, + void *page, size_t *len) { u16 *tags; @@ -1412,18 +1412,18 @@ static void ql_update_mac_hdr_len(struct ql_adapter *qdev, } /* Process an inbound completion from an rx ring. */ -static void ql_process_mac_rx_gro_page(struct ql_adapter *qdev, - struct rx_ring *rx_ring, - struct ib_mac_iocb_rsp *ib_mac_rsp, - u32 length, u16 vlan_id) +static void qlge_process_mac_rx_gro_page(struct qlge_adapter *qdev, + struct rx_ring *rx_ring, + struct qlge_ib_mac_iocb_rsp *ib_mac_rsp, + u32 length, u16 vlan_id) { struct sk_buff *skb; - struct qlge_bq_desc *lbq_desc = ql_get_curr_lchunk(qdev, rx_ring); + struct qlge_bq_desc *lbq_desc = qlge_get_curr_lchunk(qdev, rx_ring); struct napi_struct *napi = &rx_ring->napi; /* Frame error, so drop the packet. */ if (ib_mac_rsp->flags2 & IB_MAC_IOCB_RSP_ERR_MASK) { - ql_categorize_rx_err(qdev, ib_mac_rsp->flags2, rx_ring); + qlge_categorize_rx_err(qdev, ib_mac_rsp->flags2, rx_ring); put_page(lbq_desc->p.pg_chunk.page); return; } @@ -1458,15 +1458,15 @@ static void ql_process_mac_rx_gro_page(struct ql_adapter *qdev, } /* Process an inbound completion from an rx ring. */ -static void ql_process_mac_rx_page(struct ql_adapter *qdev, - struct rx_ring *rx_ring, - struct ib_mac_iocb_rsp *ib_mac_rsp, - u32 length, u16 vlan_id) +static void qlge_process_mac_rx_page(struct qlge_adapter *qdev, + struct rx_ring *rx_ring, + struct qlge_ib_mac_iocb_rsp *ib_mac_rsp, + u32 length, u16 vlan_id) { struct net_device *ndev = qdev->ndev; struct sk_buff *skb = NULL; void *addr; - struct qlge_bq_desc *lbq_desc = ql_get_curr_lchunk(qdev, rx_ring); + struct qlge_bq_desc *lbq_desc = qlge_get_curr_lchunk(qdev, rx_ring); struct napi_struct *napi = &rx_ring->napi; size_t hlen = ETH_HLEN; @@ -1482,12 +1482,12 @@ static void ql_process_mac_rx_page(struct ql_adapter *qdev, /* Frame error, so drop the packet. */ if (ib_mac_rsp->flags2 & IB_MAC_IOCB_RSP_ERR_MASK) { - ql_categorize_rx_err(qdev, ib_mac_rsp->flags2, rx_ring); + qlge_categorize_rx_err(qdev, ib_mac_rsp->flags2, rx_ring); goto err_out; } /* Update the MAC header length*/ - ql_update_mac_hdr_len(qdev, ib_mac_rsp, addr, &hlen); + qlge_update_mac_hdr_len(qdev, ib_mac_rsp, addr, &hlen); /* The max framesize filter on this chip is set higher than * MTU since FCoE uses 2k frames. @@ -1521,12 +1521,12 @@ static void ql_process_mac_rx_page(struct ql_adapter *qdev, "TCP checksum done!\n"); skb->ip_summed = CHECKSUM_UNNECESSARY; } else if ((ib_mac_rsp->flags2 & IB_MAC_IOCB_RSP_U) && - (ib_mac_rsp->flags3 & IB_MAC_IOCB_RSP_V4)) { + (ib_mac_rsp->flags3 & IB_MAC_IOCB_RSP_V4)) { /* Unfragmented ipv4 UDP frame. */ struct iphdr *iph = (struct iphdr *)((u8 *)addr + hlen); if (!(iph->frag_off & - htons(IP_MF | IP_OFFSET))) { + htons(IP_MF | IP_OFFSET))) { skb->ip_summed = CHECKSUM_UNNECESSARY; netif_printk(qdev, rx_status, KERN_DEBUG, qdev->ndev, @@ -1549,10 +1549,10 @@ static void ql_process_mac_rx_page(struct ql_adapter *qdev, } /* Process an inbound completion from an rx ring. */ -static void ql_process_mac_rx_skb(struct ql_adapter *qdev, - struct rx_ring *rx_ring, - struct ib_mac_iocb_rsp *ib_mac_rsp, - u32 length, u16 vlan_id) +static void qlge_process_mac_rx_skb(struct qlge_adapter *qdev, + struct rx_ring *rx_ring, + struct qlge_ib_mac_iocb_rsp *ib_mac_rsp, + u32 length, u16 vlan_id) { struct qlge_bq_desc *sbq_desc = qlge_get_curr_buf(&rx_ring->sbq); struct net_device *ndev = qdev->ndev; @@ -1576,14 +1576,14 @@ static void ql_process_mac_rx_skb(struct ql_adapter *qdev, /* Frame error, so drop the packet. */ if (ib_mac_rsp->flags2 & IB_MAC_IOCB_RSP_ERR_MASK) { - ql_categorize_rx_err(qdev, ib_mac_rsp->flags2, rx_ring); + qlge_categorize_rx_err(qdev, ib_mac_rsp->flags2, rx_ring); dev_kfree_skb_any(skb); return; } /* loopback self test for ethtool */ if (test_bit(QL_SELFTEST, &qdev->flags)) { - ql_check_lb_frame(qdev, skb); + qlge_check_lb_frame(qdev, skb); dev_kfree_skb_any(skb); return; } @@ -1628,12 +1628,12 @@ static void ql_process_mac_rx_skb(struct ql_adapter *qdev, "TCP checksum done!\n"); skb->ip_summed = CHECKSUM_UNNECESSARY; } else if ((ib_mac_rsp->flags2 & IB_MAC_IOCB_RSP_U) && - (ib_mac_rsp->flags3 & IB_MAC_IOCB_RSP_V4)) { + (ib_mac_rsp->flags3 & IB_MAC_IOCB_RSP_V4)) { /* Unfragmented ipv4 UDP frame. */ struct iphdr *iph = (struct iphdr *)skb->data; if (!(iph->frag_off & - htons(IP_MF | IP_OFFSET))) { + htons(IP_MF | IP_OFFSET))) { skb->ip_summed = CHECKSUM_UNNECESSARY; netif_printk(qdev, rx_status, KERN_DEBUG, qdev->ndev, @@ -1651,7 +1651,7 @@ static void ql_process_mac_rx_skb(struct ql_adapter *qdev, netif_receive_skb(skb); } -static void ql_realign_skb(struct sk_buff *skb, int len) +static void qlge_realign_skb(struct sk_buff *skb, int len) { void *temp_addr = skb->data; @@ -1669,9 +1669,9 @@ static void ql_realign_skb(struct sk_buff *skb, int len) * completion. It will be rewritten for readability in the near * future, but for not it works well. */ -static struct sk_buff *ql_build_rx_skb(struct ql_adapter *qdev, - struct rx_ring *rx_ring, - struct ib_mac_iocb_rsp *ib_mac_rsp) +static struct sk_buff *qlge_build_rx_skb(struct qlge_adapter *qdev, + struct rx_ring *rx_ring, + struct qlge_ib_mac_iocb_rsp *ib_mac_rsp) { u32 length = le32_to_cpu(ib_mac_rsp->data_len); u32 hdr_len = le32_to_cpu(ib_mac_rsp->hdr_len); @@ -1693,7 +1693,7 @@ static struct sk_buff *ql_build_rx_skb(struct ql_adapter *qdev, dma_unmap_single(&qdev->pdev->dev, sbq_desc->dma_addr, SMALL_BUF_MAP_SIZE, DMA_FROM_DEVICE); skb = sbq_desc->p.skb; - ql_realign_skb(skb, hdr_len); + qlge_realign_skb(skb, hdr_len); skb_put(skb, hdr_len); sbq_desc->p.skb = NULL; } @@ -1731,7 +1731,7 @@ static struct sk_buff *ql_build_rx_skb(struct ql_adapter *qdev, length); sbq_desc = qlge_get_curr_buf(&rx_ring->sbq); skb = sbq_desc->p.skb; - ql_realign_skb(skb, length); + qlge_realign_skb(skb, length); skb_put(skb, length); dma_unmap_single(&qdev->pdev->dev, sbq_desc->dma_addr, SMALL_BUF_MAP_SIZE, @@ -1748,7 +1748,7 @@ static struct sk_buff *ql_build_rx_skb(struct ql_adapter *qdev, * chain it to the header buffer's skb and let * it rip. */ - lbq_desc = ql_get_curr_lchunk(qdev, rx_ring); + lbq_desc = qlge_get_curr_lchunk(qdev, rx_ring); netif_printk(qdev, rx_status, KERN_DEBUG, qdev->ndev, "Chaining page at offset = %d, for %d bytes to skb.\n", lbq_desc->p.pg_chunk.offset, length); @@ -1763,7 +1763,7 @@ static struct sk_buff *ql_build_rx_skb(struct ql_adapter *qdev, * copy it to a new skb and let it go. This can happen with * jumbo mtu on a non-TCP/UDP frame. */ - lbq_desc = ql_get_curr_lchunk(qdev, rx_ring); + lbq_desc = qlge_get_curr_lchunk(qdev, rx_ring); skb = netdev_alloc_skb(qdev->ndev, length); if (!skb) { netif_printk(qdev, probe, KERN_DEBUG, qdev->ndev, @@ -1783,9 +1783,9 @@ static struct sk_buff *ql_build_rx_skb(struct ql_adapter *qdev, skb->len += length; skb->data_len += length; skb->truesize += length; - ql_update_mac_hdr_len(qdev, ib_mac_rsp, - lbq_desc->p.pg_chunk.va, - &hlen); + qlge_update_mac_hdr_len(qdev, ib_mac_rsp, + lbq_desc->p.pg_chunk.va, + &hlen); __pskb_pull_tail(skb, hlen); } } else { @@ -1823,7 +1823,7 @@ static struct sk_buff *ql_build_rx_skb(struct ql_adapter *qdev, skb_reserve(skb, NET_IP_ALIGN); } do { - lbq_desc = ql_get_curr_lchunk(qdev, rx_ring); + lbq_desc = qlge_get_curr_lchunk(qdev, rx_ring); size = min(length, qdev->lbq_buf_size); netif_printk(qdev, rx_status, KERN_DEBUG, qdev->ndev, @@ -1838,25 +1838,25 @@ static struct sk_buff *ql_build_rx_skb(struct ql_adapter *qdev, length -= size; i++; } while (length > 0); - ql_update_mac_hdr_len(qdev, ib_mac_rsp, lbq_desc->p.pg_chunk.va, - &hlen); + qlge_update_mac_hdr_len(qdev, ib_mac_rsp, lbq_desc->p.pg_chunk.va, + &hlen); __pskb_pull_tail(skb, hlen); } return skb; } /* Process an inbound completion from an rx ring. */ -static void ql_process_mac_split_rx_intr(struct ql_adapter *qdev, - struct rx_ring *rx_ring, - struct ib_mac_iocb_rsp *ib_mac_rsp, - u16 vlan_id) +static void qlge_process_mac_split_rx_intr(struct qlge_adapter *qdev, + struct rx_ring *rx_ring, + struct qlge_ib_mac_iocb_rsp *ib_mac_rsp, + u16 vlan_id) { struct net_device *ndev = qdev->ndev; struct sk_buff *skb = NULL; QL_DUMP_IB_MAC_RSP(qdev, ib_mac_rsp); - skb = ql_build_rx_skb(qdev, rx_ring, ib_mac_rsp); + skb = qlge_build_rx_skb(qdev, rx_ring, ib_mac_rsp); if (unlikely(!skb)) { netif_printk(qdev, rx_status, KERN_DEBUG, qdev->ndev, "No skb available, drop packet.\n"); @@ -1866,7 +1866,7 @@ static void ql_process_mac_split_rx_intr(struct ql_adapter *qdev, /* Frame error, so drop the packet. */ if (ib_mac_rsp->flags2 & IB_MAC_IOCB_RSP_ERR_MASK) { - ql_categorize_rx_err(qdev, ib_mac_rsp->flags2, rx_ring); + qlge_categorize_rx_err(qdev, ib_mac_rsp->flags2, rx_ring); dev_kfree_skb_any(skb); return; } @@ -1882,7 +1882,7 @@ static void ql_process_mac_split_rx_intr(struct ql_adapter *qdev, /* loopback self test for ethtool */ if (test_bit(QL_SELFTEST, &qdev->flags)) { - ql_check_lb_frame(qdev, skb); + qlge_check_lb_frame(qdev, skb); dev_kfree_skb_any(skb); return; } @@ -1917,12 +1917,12 @@ static void ql_process_mac_split_rx_intr(struct ql_adapter *qdev, "TCP checksum done!\n"); skb->ip_summed = CHECKSUM_UNNECESSARY; } else if ((ib_mac_rsp->flags2 & IB_MAC_IOCB_RSP_U) && - (ib_mac_rsp->flags3 & IB_MAC_IOCB_RSP_V4)) { - /* Unfragmented ipv4 UDP frame. */ + (ib_mac_rsp->flags3 & IB_MAC_IOCB_RSP_V4)) { + /* Unfragmented ipv4 UDP frame. */ struct iphdr *iph = (struct iphdr *)skb->data; if (!(iph->frag_off & - htons(IP_MF | IP_OFFSET))) { + htons(IP_MF | IP_OFFSET))) { skb->ip_summed = CHECKSUM_UNNECESSARY; netif_printk(qdev, rx_status, KERN_DEBUG, qdev->ndev, "TCP checksum done!\n"); @@ -1942,15 +1942,15 @@ static void ql_process_mac_split_rx_intr(struct ql_adapter *qdev, } /* Process an inbound completion from an rx ring. */ -static unsigned long ql_process_mac_rx_intr(struct ql_adapter *qdev, - struct rx_ring *rx_ring, - struct ib_mac_iocb_rsp *ib_mac_rsp) +static unsigned long qlge_process_mac_rx_intr(struct qlge_adapter *qdev, + struct rx_ring *rx_ring, + struct qlge_ib_mac_iocb_rsp *ib_mac_rsp) { u32 length = le32_to_cpu(ib_mac_rsp->data_len); u16 vlan_id = ((ib_mac_rsp->flags2 & IB_MAC_IOCB_RSP_V) && - (qdev->ndev->features & NETIF_F_HW_VLAN_CTAG_RX)) ? - ((le16_to_cpu(ib_mac_rsp->vlan_id) & - IB_MAC_IOCB_RSP_VLAN_MASK)) : 0xffff; + (qdev->ndev->features & NETIF_F_HW_VLAN_CTAG_RX)) ? + ((le16_to_cpu(ib_mac_rsp->vlan_id) & + IB_MAC_IOCB_RSP_VLAN_MASK)) : 0xffff; QL_DUMP_IB_MAC_RSP(qdev, ib_mac_rsp); @@ -1958,43 +1958,43 @@ static unsigned long ql_process_mac_rx_intr(struct ql_adapter *qdev, /* The data and headers are split into * separate buffers. */ - ql_process_mac_split_rx_intr(qdev, rx_ring, ib_mac_rsp, - vlan_id); + qlge_process_mac_split_rx_intr(qdev, rx_ring, ib_mac_rsp, + vlan_id); } else if (ib_mac_rsp->flags3 & IB_MAC_IOCB_RSP_DS) { /* The data fit in a single small buffer. * Allocate a new skb, copy the data and * return the buffer to the free pool. */ - ql_process_mac_rx_skb(qdev, rx_ring, ib_mac_rsp, length, - vlan_id); + qlge_process_mac_rx_skb(qdev, rx_ring, ib_mac_rsp, length, + vlan_id); } else if ((ib_mac_rsp->flags3 & IB_MAC_IOCB_RSP_DL) && - !(ib_mac_rsp->flags1 & IB_MAC_CSUM_ERR_MASK) && - (ib_mac_rsp->flags2 & IB_MAC_IOCB_RSP_T)) { + !(ib_mac_rsp->flags1 & IB_MAC_CSUM_ERR_MASK) && + (ib_mac_rsp->flags2 & IB_MAC_IOCB_RSP_T)) { /* TCP packet in a page chunk that's been checksummed. * Tack it on to our GRO skb and let it go. */ - ql_process_mac_rx_gro_page(qdev, rx_ring, ib_mac_rsp, length, - vlan_id); + qlge_process_mac_rx_gro_page(qdev, rx_ring, ib_mac_rsp, length, + vlan_id); } else if (ib_mac_rsp->flags3 & IB_MAC_IOCB_RSP_DL) { /* Non-TCP packet in a page chunk. Allocate an * skb, tack it on frags, and send it up. */ - ql_process_mac_rx_page(qdev, rx_ring, ib_mac_rsp, length, - vlan_id); + qlge_process_mac_rx_page(qdev, rx_ring, ib_mac_rsp, length, + vlan_id); } else { /* Non-TCP/UDP large frames that span multiple buffers * can be processed corrrectly by the split frame logic. */ - ql_process_mac_split_rx_intr(qdev, rx_ring, ib_mac_rsp, - vlan_id); + qlge_process_mac_split_rx_intr(qdev, rx_ring, ib_mac_rsp, + vlan_id); } return (unsigned long)length; } /* Process an outbound completion from an rx ring. */ -static void ql_process_mac_tx_intr(struct ql_adapter *qdev, - struct ob_mac_iocb_rsp *mac_rsp) +static void qlge_process_mac_tx_intr(struct qlge_adapter *qdev, + struct qlge_ob_mac_iocb_rsp *mac_rsp) { struct tx_ring *tx_ring; struct tx_ring_desc *tx_ring_desc; @@ -2002,7 +2002,7 @@ static void ql_process_mac_tx_intr(struct ql_adapter *qdev, QL_DUMP_OB_MAC_RSP(qdev, mac_rsp); tx_ring = &qdev->tx_ring[mac_rsp->txq_idx]; tx_ring_desc = &tx_ring->q[mac_rsp->tid]; - ql_unmap_send(qdev, tx_ring_desc, tx_ring_desc->map_cnt); + qlge_unmap_send(qdev, tx_ring_desc, tx_ring_desc->map_cnt); tx_ring->tx_bytes += (tx_ring_desc->skb)->len; tx_ring->tx_packets++; dev_kfree_skb(tx_ring_desc->skb); @@ -2033,16 +2033,16 @@ static void ql_process_mac_tx_intr(struct ql_adapter *qdev, } /* Fire up a handler to reset the MPI processor. */ -void ql_queue_fw_error(struct ql_adapter *qdev) +void qlge_queue_fw_error(struct qlge_adapter *qdev) { - ql_link_off(qdev); + qlge_link_off(qdev); queue_delayed_work(qdev->workqueue, &qdev->mpi_reset_work, 0); } -void ql_queue_asic_error(struct ql_adapter *qdev) +void qlge_queue_asic_error(struct qlge_adapter *qdev) { - ql_link_off(qdev); - ql_disable_interrupts(qdev); + qlge_link_off(qdev); + qlge_disable_interrupts(qdev); /* Clear adapter up bit to signal the recovery * process that it shouldn't kill the reset worker * thread @@ -2055,47 +2055,47 @@ void ql_queue_asic_error(struct ql_adapter *qdev) queue_delayed_work(qdev->workqueue, &qdev->asic_reset_work, 0); } -static void ql_process_chip_ae_intr(struct ql_adapter *qdev, - struct ib_ae_iocb_rsp *ib_ae_rsp) +static void qlge_process_chip_ae_intr(struct qlge_adapter *qdev, + struct qlge_ib_ae_iocb_rsp *ib_ae_rsp) { switch (ib_ae_rsp->event) { case MGMT_ERR_EVENT: netif_err(qdev, rx_err, qdev->ndev, "Management Processor Fatal Error.\n"); - ql_queue_fw_error(qdev); + qlge_queue_fw_error(qdev); return; case CAM_LOOKUP_ERR_EVENT: netdev_err(qdev->ndev, "Multiple CAM hits lookup occurred.\n"); netdev_err(qdev->ndev, "This event shouldn't occur.\n"); - ql_queue_asic_error(qdev); + qlge_queue_asic_error(qdev); return; case SOFT_ECC_ERROR_EVENT: netdev_err(qdev->ndev, "Soft ECC error detected.\n"); - ql_queue_asic_error(qdev); + qlge_queue_asic_error(qdev); break; case PCI_ERR_ANON_BUF_RD: netdev_err(qdev->ndev, "PCI error occurred when reading anonymous buffers from rx_ring %d.\n", ib_ae_rsp->q_id); - ql_queue_asic_error(qdev); + qlge_queue_asic_error(qdev); break; default: netif_err(qdev, drv, qdev->ndev, "Unexpected event %d.\n", ib_ae_rsp->event); - ql_queue_asic_error(qdev); + qlge_queue_asic_error(qdev); break; } } -static int ql_clean_outbound_rx_ring(struct rx_ring *rx_ring) +static int qlge_clean_outbound_rx_ring(struct rx_ring *rx_ring) { - struct ql_adapter *qdev = rx_ring->qdev; - u32 prod = ql_read_sh_reg(rx_ring->prod_idx_sh_reg); - struct ob_mac_iocb_rsp *net_rsp = NULL; + struct qlge_adapter *qdev = rx_ring->qdev; + u32 prod = qlge_read_sh_reg(rx_ring->prod_idx_sh_reg); + struct qlge_ob_mac_iocb_rsp *net_rsp = NULL; int count = 0; struct tx_ring *tx_ring; @@ -2105,12 +2105,12 @@ static int ql_clean_outbound_rx_ring(struct rx_ring *rx_ring) "cq_id = %d, prod = %d, cnsmr = %d\n", rx_ring->cq_id, prod, rx_ring->cnsmr_idx); - net_rsp = (struct ob_mac_iocb_rsp *)rx_ring->curr_entry; + net_rsp = (struct qlge_ob_mac_iocb_rsp *)rx_ring->curr_entry; rmb(); switch (net_rsp->opcode) { case OPCODE_OB_MAC_TSO_IOCB: case OPCODE_OB_MAC_IOCB: - ql_process_mac_tx_intr(qdev, net_rsp); + qlge_process_mac_tx_intr(qdev, net_rsp); break; default: netif_printk(qdev, rx_status, KERN_DEBUG, qdev->ndev, @@ -2118,12 +2118,12 @@ static int ql_clean_outbound_rx_ring(struct rx_ring *rx_ring) net_rsp->opcode); } count++; - ql_update_cq(rx_ring); - prod = ql_read_sh_reg(rx_ring->prod_idx_sh_reg); + qlge_update_cq(rx_ring); + prod = qlge_read_sh_reg(rx_ring->prod_idx_sh_reg); } if (!net_rsp) return 0; - ql_write_cq_idx(rx_ring); + qlge_write_cq_idx(rx_ring); tx_ring = &qdev->tx_ring[net_rsp->txq_idx]; if (__netif_subqueue_stopped(qdev->ndev, tx_ring->wq_id)) { if ((atomic_read(&tx_ring->tx_count) > (tx_ring->wq_len / 4))) @@ -2137,11 +2137,11 @@ static int ql_clean_outbound_rx_ring(struct rx_ring *rx_ring) return count; } -static int ql_clean_inbound_rx_ring(struct rx_ring *rx_ring, int budget) +static int qlge_clean_inbound_rx_ring(struct rx_ring *rx_ring, int budget) { - struct ql_adapter *qdev = rx_ring->qdev; - u32 prod = ql_read_sh_reg(rx_ring->prod_idx_sh_reg); - struct ql_net_rsp_iocb *net_rsp; + struct qlge_adapter *qdev = rx_ring->qdev; + u32 prod = qlge_read_sh_reg(rx_ring->prod_idx_sh_reg); + struct qlge_net_rsp_iocb *net_rsp; int count = 0; /* While there are entries in the completion queue. */ @@ -2154,14 +2154,14 @@ static int ql_clean_inbound_rx_ring(struct rx_ring *rx_ring, int budget) rmb(); switch (net_rsp->opcode) { case OPCODE_IB_MAC_IOCB: - ql_process_mac_rx_intr(qdev, rx_ring, - (struct ib_mac_iocb_rsp *) - net_rsp); + qlge_process_mac_rx_intr(qdev, rx_ring, + (struct qlge_ib_mac_iocb_rsp *) + net_rsp); break; case OPCODE_IB_AE_IOCB: - ql_process_chip_ae_intr(qdev, (struct ib_ae_iocb_rsp *) - net_rsp); + qlge_process_chip_ae_intr(qdev, (struct qlge_ib_ae_iocb_rsp *) + net_rsp); break; default: netif_printk(qdev, rx_status, KERN_DEBUG, qdev->ndev, @@ -2170,20 +2170,20 @@ static int ql_clean_inbound_rx_ring(struct rx_ring *rx_ring, int budget) break; } count++; - ql_update_cq(rx_ring); - prod = ql_read_sh_reg(rx_ring->prod_idx_sh_reg); + qlge_update_cq(rx_ring); + prod = qlge_read_sh_reg(rx_ring->prod_idx_sh_reg); if (count == budget) break; } - ql_update_buffer_queues(rx_ring, GFP_ATOMIC, 0); - ql_write_cq_idx(rx_ring); + qlge_update_buffer_queues(rx_ring, GFP_ATOMIC, 0); + qlge_write_cq_idx(rx_ring); return count; } -static int ql_napi_poll_msix(struct napi_struct *napi, int budget) +static int qlge_napi_poll_msix(struct napi_struct *napi, int budget) { struct rx_ring *rx_ring = container_of(napi, struct rx_ring, napi); - struct ql_adapter *qdev = rx_ring->qdev; + struct qlge_adapter *qdev = rx_ring->qdev; struct rx_ring *trx_ring; int i, work_done = 0; struct intr_context *ctx = &qdev->intr_context[rx_ring->cq_id]; @@ -2200,42 +2200,42 @@ static int ql_napi_poll_msix(struct napi_struct *napi, int budget) * it's not empty then service it. */ if ((ctx->irq_mask & (1 << trx_ring->cq_id)) && - (ql_read_sh_reg(trx_ring->prod_idx_sh_reg) != + (qlge_read_sh_reg(trx_ring->prod_idx_sh_reg) != trx_ring->cnsmr_idx)) { netif_printk(qdev, intr, KERN_DEBUG, qdev->ndev, "%s: Servicing TX completion ring %d.\n", __func__, trx_ring->cq_id); - ql_clean_outbound_rx_ring(trx_ring); + qlge_clean_outbound_rx_ring(trx_ring); } } /* * Now service the RSS ring if it's active. */ - if (ql_read_sh_reg(rx_ring->prod_idx_sh_reg) != - rx_ring->cnsmr_idx) { + if (qlge_read_sh_reg(rx_ring->prod_idx_sh_reg) != + rx_ring->cnsmr_idx) { netif_printk(qdev, intr, KERN_DEBUG, qdev->ndev, "%s: Servicing RX completion ring %d.\n", __func__, rx_ring->cq_id); - work_done = ql_clean_inbound_rx_ring(rx_ring, budget); + work_done = qlge_clean_inbound_rx_ring(rx_ring, budget); } if (work_done < budget) { napi_complete_done(napi, work_done); - ql_enable_completion_interrupt(qdev, rx_ring->irq); + qlge_enable_completion_interrupt(qdev, rx_ring->irq); } return work_done; } static void qlge_vlan_mode(struct net_device *ndev, netdev_features_t features) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); if (features & NETIF_F_HW_VLAN_CTAG_RX) { - ql_write32(qdev, NIC_RCV_CFG, NIC_RCV_CFG_VLAN_MASK | - NIC_RCV_CFG_VLAN_MATCH_AND_NON); + qlge_write32(qdev, NIC_RCV_CFG, NIC_RCV_CFG_VLAN_MASK | + NIC_RCV_CFG_VLAN_MATCH_AND_NON); } else { - ql_write32(qdev, NIC_RCV_CFG, NIC_RCV_CFG_VLAN_MASK); + qlge_write32(qdev, NIC_RCV_CFG, NIC_RCV_CFG_VLAN_MASK); } } @@ -2246,12 +2246,12 @@ static void qlge_vlan_mode(struct net_device *ndev, netdev_features_t features) static int qlge_update_hw_vlan_features(struct net_device *ndev, netdev_features_t features) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); int status = 0; bool need_restart = netif_running(ndev); if (need_restart) { - status = ql_adapter_down(qdev); + status = qlge_adapter_down(qdev); if (status) { netif_err(qdev, link, qdev->ndev, "Failed to bring down the adapter\n"); @@ -2263,7 +2263,7 @@ static int qlge_update_hw_vlan_features(struct net_device *ndev, ndev->features = features; if (need_restart) { - status = ql_adapter_up(qdev); + status = qlge_adapter_up(qdev); if (status) { netif_err(qdev, link, qdev->ndev, "Failed to bring up the adapter\n"); @@ -2292,13 +2292,13 @@ static int qlge_set_features(struct net_device *ndev, return 0; } -static int __qlge_vlan_rx_add_vid(struct ql_adapter *qdev, u16 vid) +static int __qlge_vlan_rx_add_vid(struct qlge_adapter *qdev, u16 vid) { u32 enable_bit = MAC_ADDR_E; int err; - err = ql_set_mac_addr_reg(qdev, (u8 *)&enable_bit, - MAC_ADDR_TYPE_VLAN, vid); + err = qlge_set_mac_addr_reg(qdev, (u8 *)&enable_bit, + MAC_ADDR_TYPE_VLAN, vid); if (err) netif_err(qdev, ifup, qdev->ndev, "Failed to init vlan address.\n"); @@ -2307,29 +2307,29 @@ static int __qlge_vlan_rx_add_vid(struct ql_adapter *qdev, u16 vid) static int qlge_vlan_rx_add_vid(struct net_device *ndev, __be16 proto, u16 vid) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); int status; int err; - status = ql_sem_spinlock(qdev, SEM_MAC_ADDR_MASK); + status = qlge_sem_spinlock(qdev, SEM_MAC_ADDR_MASK); if (status) return status; err = __qlge_vlan_rx_add_vid(qdev, vid); set_bit(vid, qdev->active_vlans); - ql_sem_unlock(qdev, SEM_MAC_ADDR_MASK); + qlge_sem_unlock(qdev, SEM_MAC_ADDR_MASK); return err; } -static int __qlge_vlan_rx_kill_vid(struct ql_adapter *qdev, u16 vid) +static int __qlge_vlan_rx_kill_vid(struct qlge_adapter *qdev, u16 vid) { u32 enable_bit = 0; int err; - err = ql_set_mac_addr_reg(qdev, (u8 *)&enable_bit, - MAC_ADDR_TYPE_VLAN, vid); + err = qlge_set_mac_addr_reg(qdev, (u8 *)&enable_bit, + MAC_ADDR_TYPE_VLAN, vid); if (err) netif_err(qdev, ifup, qdev->ndev, "Failed to clear vlan address.\n"); @@ -2338,35 +2338,35 @@ static int __qlge_vlan_rx_kill_vid(struct ql_adapter *qdev, u16 vid) static int qlge_vlan_rx_kill_vid(struct net_device *ndev, __be16 proto, u16 vid) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); int status; int err; - status = ql_sem_spinlock(qdev, SEM_MAC_ADDR_MASK); + status = qlge_sem_spinlock(qdev, SEM_MAC_ADDR_MASK); if (status) return status; err = __qlge_vlan_rx_kill_vid(qdev, vid); clear_bit(vid, qdev->active_vlans); - ql_sem_unlock(qdev, SEM_MAC_ADDR_MASK); + qlge_sem_unlock(qdev, SEM_MAC_ADDR_MASK); return err; } -static void qlge_restore_vlan(struct ql_adapter *qdev) +static void qlge_restore_vlan(struct qlge_adapter *qdev) { int status; u16 vid; - status = ql_sem_spinlock(qdev, SEM_MAC_ADDR_MASK); + status = qlge_sem_spinlock(qdev, SEM_MAC_ADDR_MASK); if (status) return; for_each_set_bit(vid, qdev->active_vlans, VLAN_N_VID) __qlge_vlan_rx_add_vid(qdev, vid); - ql_sem_unlock(qdev, SEM_MAC_ADDR_MASK); + qlge_sem_unlock(qdev, SEM_MAC_ADDR_MASK); } /* MSI-X Multiple Vector Interrupt Handler for inbound completions. */ @@ -2386,7 +2386,7 @@ static irqreturn_t qlge_msix_rx_isr(int irq, void *dev_id) static irqreturn_t qlge_isr(int irq, void *dev_id) { struct rx_ring *rx_ring = dev_id; - struct ql_adapter *qdev = rx_ring->qdev; + struct qlge_adapter *qdev = rx_ring->qdev; struct intr_context *intr_context = &qdev->intr_context[0]; u32 var; int work_done = 0; @@ -2398,18 +2398,18 @@ static irqreturn_t qlge_isr(int irq, void *dev_id) * enable it is not effective. */ if (!test_bit(QL_MSIX_ENABLED, &qdev->flags)) - ql_disable_completion_interrupt(qdev, 0); + qlge_disable_completion_interrupt(qdev, 0); - var = ql_read32(qdev, STS); + var = qlge_read32(qdev, STS); /* * Check for fatal error. */ if (var & STS_FE) { - ql_disable_completion_interrupt(qdev, 0); - ql_queue_asic_error(qdev); + qlge_disable_completion_interrupt(qdev, 0); + qlge_queue_asic_error(qdev); netdev_err(qdev->ndev, "Got fatal error, STS = %x.\n", var); - var = ql_read32(qdev, ERR_STS); + var = qlge_read32(qdev, ERR_STS); netdev_err(qdev->ndev, "Resetting chip. Error Status Register = 0x%x\n", var); return IRQ_HANDLED; } @@ -2418,14 +2418,14 @@ static irqreturn_t qlge_isr(int irq, void *dev_id) * Check MPI processor activity. */ if ((var & STS_PI) && - (ql_read32(qdev, INTR_MASK) & INTR_MASK_PI)) { + (qlge_read32(qdev, INTR_MASK) & INTR_MASK_PI)) { /* * We've got an async event or mailbox completion. * Handle it and clear the source of the interrupt. */ netif_err(qdev, intr, qdev->ndev, "Got MPI processor interrupt.\n"); - ql_write32(qdev, INTR_MASK, (INTR_MASK_PI << 16)); + qlge_write32(qdev, INTR_MASK, (INTR_MASK_PI << 16)); queue_delayed_work_on(smp_processor_id(), qdev->workqueue, &qdev->mpi_work, 0); work_done++; @@ -2436,7 +2436,7 @@ static irqreturn_t qlge_isr(int irq, void *dev_id) * pass. Compare it to the queues that this irq services * and call napi if there's a match. */ - var = ql_read32(qdev, ISR1); + var = qlge_read32(qdev, ISR1); if (var & intr_context->irq_mask) { netif_info(qdev, intr, qdev->ndev, "Waking handler for rx_ring[0].\n"); @@ -2449,13 +2449,13 @@ static irqreturn_t qlge_isr(int irq, void *dev_id) * systematically re-enable the interrupt if we didn't * schedule napi. */ - ql_enable_completion_interrupt(qdev, 0); + qlge_enable_completion_interrupt(qdev, 0); } return work_done ? IRQ_HANDLED : IRQ_NONE; } -static int ql_tso(struct sk_buff *skb, struct ob_mac_tso_iocb_req *mac_iocb_ptr) +static int qlge_tso(struct sk_buff *skb, struct qlge_ob_mac_tso_iocb_req *mac_iocb_ptr) { if (skb_is_gso(skb)) { int err; @@ -2469,11 +2469,11 @@ static int ql_tso(struct sk_buff *skb, struct ob_mac_tso_iocb_req *mac_iocb_ptr) mac_iocb_ptr->flags3 |= OB_MAC_TSO_IOCB_IC; mac_iocb_ptr->frame_len = cpu_to_le32((u32)skb->len); mac_iocb_ptr->total_hdrs_len = - cpu_to_le16(skb_transport_offset(skb) + tcp_hdrlen(skb)); + cpu_to_le16(skb_transport_offset(skb) + tcp_hdrlen(skb)); mac_iocb_ptr->net_trans_offset = - cpu_to_le16(skb_network_offset(skb) | - skb_transport_offset(skb) - << OB_MAC_TRANSPORT_HDR_SHIFT); + cpu_to_le16(skb_network_offset(skb) | + skb_transport_offset(skb) + << OB_MAC_TRANSPORT_HDR_SHIFT); mac_iocb_ptr->mss = cpu_to_le16(skb_shinfo(skb)->gso_size); mac_iocb_ptr->flags2 |= OB_MAC_TSO_IOCB_LSO; if (likely(l3_proto == htons(ETH_P_IP))) { @@ -2488,17 +2488,17 @@ static int ql_tso(struct sk_buff *skb, struct ob_mac_tso_iocb_req *mac_iocb_ptr) } else if (l3_proto == htons(ETH_P_IPV6)) { mac_iocb_ptr->flags1 |= OB_MAC_TSO_IOCB_IP6; tcp_hdr(skb)->check = - ~csum_ipv6_magic(&ipv6_hdr(skb)->saddr, - &ipv6_hdr(skb)->daddr, - 0, IPPROTO_TCP, 0); + ~csum_ipv6_magic(&ipv6_hdr(skb)->saddr, + &ipv6_hdr(skb)->daddr, + 0, IPPROTO_TCP, 0); } return 1; } return 0; } -static void ql_hw_csum_setup(struct sk_buff *skb, - struct ob_mac_tso_iocb_req *mac_iocb_ptr) +static void qlge_hw_csum_setup(struct sk_buff *skb, + struct qlge_ob_mac_tso_iocb_req *mac_iocb_ptr) { int len; struct iphdr *iph = ip_hdr(skb); @@ -2508,7 +2508,7 @@ static void ql_hw_csum_setup(struct sk_buff *skb, mac_iocb_ptr->frame_len = cpu_to_le32((u32)skb->len); mac_iocb_ptr->net_trans_offset = cpu_to_le16(skb_network_offset(skb) | - skb_transport_offset(skb) << OB_MAC_TRANSPORT_HDR_SHIFT); + skb_transport_offset(skb) << OB_MAC_TRANSPORT_HDR_SHIFT); mac_iocb_ptr->flags1 |= OB_MAC_TSO_IOCB_IP4; len = (ntohs(iph->tot_len) - (iph->ihl << 2)); @@ -2516,14 +2516,14 @@ static void ql_hw_csum_setup(struct sk_buff *skb, check = &(tcp_hdr(skb)->check); mac_iocb_ptr->flags2 |= OB_MAC_TSO_IOCB_TC; mac_iocb_ptr->total_hdrs_len = - cpu_to_le16(skb_transport_offset(skb) + - (tcp_hdr(skb)->doff << 2)); + cpu_to_le16(skb_transport_offset(skb) + + (tcp_hdr(skb)->doff << 2)); } else { check = &(udp_hdr(skb)->check); mac_iocb_ptr->flags2 |= OB_MAC_TSO_IOCB_UC; mac_iocb_ptr->total_hdrs_len = - cpu_to_le16(skb_transport_offset(skb) + - sizeof(struct udphdr)); + cpu_to_le16(skb_transport_offset(skb) + + sizeof(struct udphdr)); } *check = ~csum_tcpudp_magic(iph->saddr, iph->daddr, len, iph->protocol, 0); @@ -2532,8 +2532,8 @@ static void ql_hw_csum_setup(struct sk_buff *skb, static netdev_tx_t qlge_send(struct sk_buff *skb, struct net_device *ndev) { struct tx_ring_desc *tx_ring_desc; - struct ob_mac_iocb_req *mac_iocb_ptr; - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_ob_mac_iocb_req *mac_iocb_ptr; + struct qlge_adapter *qdev = netdev_priv(ndev); int tso; struct tx_ring *tx_ring; u32 tx_ring_idx = (u32)skb->queue_mapping; @@ -2571,16 +2571,16 @@ static netdev_tx_t qlge_send(struct sk_buff *skb, struct net_device *ndev) mac_iocb_ptr->flags3 |= OB_MAC_IOCB_V; mac_iocb_ptr->vlan_tci = cpu_to_le16(skb_vlan_tag_get(skb)); } - tso = ql_tso(skb, (struct ob_mac_tso_iocb_req *)mac_iocb_ptr); + tso = qlge_tso(skb, (struct qlge_ob_mac_tso_iocb_req *)mac_iocb_ptr); if (tso < 0) { dev_kfree_skb_any(skb); return NETDEV_TX_OK; } else if (unlikely(!tso) && (skb->ip_summed == CHECKSUM_PARTIAL)) { - ql_hw_csum_setup(skb, - (struct ob_mac_tso_iocb_req *)mac_iocb_ptr); + qlge_hw_csum_setup(skb, + (struct qlge_ob_mac_tso_iocb_req *)mac_iocb_ptr); } - if (ql_map_send(qdev, mac_iocb_ptr, skb, tx_ring_desc) != - NETDEV_TX_OK) { + if (qlge_map_send(qdev, mac_iocb_ptr, skb, tx_ring_desc) != + NETDEV_TX_OK) { netif_err(qdev, tx_queued, qdev->ndev, "Could not map the segments.\n"); tx_ring->tx_errors++; @@ -2592,7 +2592,7 @@ static netdev_tx_t qlge_send(struct sk_buff *skb, struct net_device *ndev) tx_ring->prod_idx = 0; wmb(); - ql_write_db_reg_relaxed(tx_ring->prod_idx, tx_ring->prod_idx_db_reg); + qlge_write_db_reg_relaxed(tx_ring->prod_idx, tx_ring->prod_idx_db_reg); netif_printk(qdev, tx_queued, KERN_DEBUG, qdev->ndev, "tx queued, slot %d, len %d\n", tx_ring->prod_idx, skb->len); @@ -2611,7 +2611,7 @@ static netdev_tx_t qlge_send(struct sk_buff *skb, struct net_device *ndev) return NETDEV_TX_OK; } -static void ql_free_shadow_space(struct ql_adapter *qdev) +static void qlge_free_shadow_space(struct qlge_adapter *qdev) { if (qdev->rx_ring_shadow_reg_area) { dma_free_coherent(&qdev->pdev->dev, @@ -2629,7 +2629,7 @@ static void ql_free_shadow_space(struct ql_adapter *qdev) } } -static int ql_alloc_shadow_space(struct ql_adapter *qdev) +static int qlge_alloc_shadow_space(struct qlge_adapter *qdev) { qdev->rx_ring_shadow_reg_area = dma_alloc_coherent(&qdev->pdev->dev, PAGE_SIZE, @@ -2658,11 +2658,11 @@ static int ql_alloc_shadow_space(struct ql_adapter *qdev) return -ENOMEM; } -static void ql_init_tx_ring(struct ql_adapter *qdev, struct tx_ring *tx_ring) +static void qlge_init_tx_ring(struct qlge_adapter *qdev, struct tx_ring *tx_ring) { struct tx_ring_desc *tx_ring_desc; int i; - struct ob_mac_iocb_req *mac_iocb_ptr; + struct qlge_ob_mac_iocb_req *mac_iocb_ptr; mac_iocb_ptr = tx_ring->wq_base; tx_ring_desc = tx_ring->q; @@ -2676,8 +2676,8 @@ static void ql_init_tx_ring(struct ql_adapter *qdev, struct tx_ring *tx_ring) atomic_set(&tx_ring->tx_count, tx_ring->wq_len); } -static void ql_free_tx_resources(struct ql_adapter *qdev, - struct tx_ring *tx_ring) +static void qlge_free_tx_resources(struct qlge_adapter *qdev, + struct tx_ring *tx_ring) { if (tx_ring->wq_base) { dma_free_coherent(&qdev->pdev->dev, tx_ring->wq_size, @@ -2688,20 +2688,20 @@ static void ql_free_tx_resources(struct ql_adapter *qdev, tx_ring->q = NULL; } -static int ql_alloc_tx_resources(struct ql_adapter *qdev, - struct tx_ring *tx_ring) +static int qlge_alloc_tx_resources(struct qlge_adapter *qdev, + struct tx_ring *tx_ring) { tx_ring->wq_base = - dma_alloc_coherent(&qdev->pdev->dev, tx_ring->wq_size, - &tx_ring->wq_base_dma, GFP_ATOMIC); + dma_alloc_coherent(&qdev->pdev->dev, tx_ring->wq_size, + &tx_ring->wq_base_dma, GFP_ATOMIC); if (!tx_ring->wq_base || tx_ring->wq_base_dma & WQ_ADDR_ALIGN) goto pci_alloc_err; tx_ring->q = - kmalloc_array(tx_ring->wq_len, sizeof(struct tx_ring_desc), - GFP_KERNEL); + kmalloc_array(tx_ring->wq_len, sizeof(struct tx_ring_desc), + GFP_KERNEL); if (!tx_ring->q) goto err; @@ -2715,19 +2715,19 @@ static int ql_alloc_tx_resources(struct ql_adapter *qdev, return -ENOMEM; } -static void ql_free_lbq_buffers(struct ql_adapter *qdev, struct rx_ring *rx_ring) +static void qlge_free_lbq_buffers(struct qlge_adapter *qdev, struct rx_ring *rx_ring) { struct qlge_bq *lbq = &rx_ring->lbq; unsigned int last_offset; - last_offset = ql_lbq_block_size(qdev) - qdev->lbq_buf_size; + last_offset = qlge_lbq_block_size(qdev) - qdev->lbq_buf_size; while (lbq->next_to_clean != lbq->next_to_use) { struct qlge_bq_desc *lbq_desc = &lbq->queue[lbq->next_to_clean]; if (lbq_desc->p.pg_chunk.offset == last_offset) dma_unmap_page(&qdev->pdev->dev, lbq_desc->dma_addr, - ql_lbq_block_size(qdev), + qlge_lbq_block_size(qdev), DMA_FROM_DEVICE); put_page(lbq_desc->p.pg_chunk.page); @@ -2736,13 +2736,13 @@ static void ql_free_lbq_buffers(struct ql_adapter *qdev, struct rx_ring *rx_ring if (rx_ring->master_chunk.page) { dma_unmap_page(&qdev->pdev->dev, rx_ring->chunk_dma_addr, - ql_lbq_block_size(qdev), DMA_FROM_DEVICE); + qlge_lbq_block_size(qdev), DMA_FROM_DEVICE); put_page(rx_ring->master_chunk.page); rx_ring->master_chunk.page = NULL; } } -static void ql_free_sbq_buffers(struct ql_adapter *qdev, struct rx_ring *rx_ring) +static void qlge_free_sbq_buffers(struct qlge_adapter *qdev, struct rx_ring *rx_ring) { int i; @@ -2767,7 +2767,7 @@ static void ql_free_sbq_buffers(struct ql_adapter *qdev, struct rx_ring *rx_ring /* Free all large and small rx buffers associated * with the completion queues for this device. */ -static void ql_free_rx_buffers(struct ql_adapter *qdev) +static void qlge_free_rx_buffers(struct qlge_adapter *qdev) { int i; @@ -2775,25 +2775,25 @@ static void ql_free_rx_buffers(struct ql_adapter *qdev) struct rx_ring *rx_ring = &qdev->rx_ring[i]; if (rx_ring->lbq.queue) - ql_free_lbq_buffers(qdev, rx_ring); + qlge_free_lbq_buffers(qdev, rx_ring); if (rx_ring->sbq.queue) - ql_free_sbq_buffers(qdev, rx_ring); + qlge_free_sbq_buffers(qdev, rx_ring); } } -static void ql_alloc_rx_buffers(struct ql_adapter *qdev) +static void qlge_alloc_rx_buffers(struct qlge_adapter *qdev) { int i; for (i = 0; i < qdev->rss_ring_count; i++) - ql_update_buffer_queues(&qdev->rx_ring[i], GFP_KERNEL, - HZ / 2); + qlge_update_buffer_queues(&qdev->rx_ring[i], GFP_KERNEL, + HZ / 2); } static int qlge_init_bq(struct qlge_bq *bq) { struct rx_ring *rx_ring = QLGE_BQ_CONTAINER(bq); - struct ql_adapter *qdev = rx_ring->qdev; + struct qlge_adapter *qdev = rx_ring->qdev; struct qlge_bq_desc *bq_desc; __le64 *buf_ptr; int i; @@ -2823,8 +2823,8 @@ static int qlge_init_bq(struct qlge_bq *bq) return 0; } -static void ql_free_rx_resources(struct ql_adapter *qdev, - struct rx_ring *rx_ring) +static void qlge_free_rx_resources(struct qlge_adapter *qdev, + struct rx_ring *rx_ring) { /* Free the small buffer queue. */ if (rx_ring->sbq.base) { @@ -2860,15 +2860,15 @@ static void ql_free_rx_resources(struct ql_adapter *qdev, /* Allocate queues and buffers for this completions queue based * on the values in the parameter structure. */ -static int ql_alloc_rx_resources(struct ql_adapter *qdev, - struct rx_ring *rx_ring) +static int qlge_alloc_rx_resources(struct qlge_adapter *qdev, + struct rx_ring *rx_ring) { /* * Allocate the completion queue for this rx_ring. */ rx_ring->cq_base = - dma_alloc_coherent(&qdev->pdev->dev, rx_ring->cq_size, - &rx_ring->cq_base_dma, GFP_ATOMIC); + dma_alloc_coherent(&qdev->pdev->dev, rx_ring->cq_size, + &rx_ring->cq_base_dma, GFP_ATOMIC); if (!rx_ring->cq_base) { netif_err(qdev, ifup, qdev->ndev, "rx_ring alloc failed.\n"); @@ -2877,14 +2877,14 @@ static int ql_alloc_rx_resources(struct ql_adapter *qdev, if (rx_ring->cq_id < qdev->rss_ring_count && (qlge_init_bq(&rx_ring->sbq) || qlge_init_bq(&rx_ring->lbq))) { - ql_free_rx_resources(qdev, rx_ring); + qlge_free_rx_resources(qdev, rx_ring); return -ENOMEM; } return 0; } -static void ql_tx_ring_clean(struct ql_adapter *qdev) +static void qlge_tx_ring_clean(struct qlge_adapter *qdev) { struct tx_ring *tx_ring; struct tx_ring_desc *tx_ring_desc; @@ -2903,8 +2903,8 @@ static void ql_tx_ring_clean(struct ql_adapter *qdev) "Freeing lost SKB %p, from queue %d, index %d.\n", tx_ring_desc->skb, j, tx_ring_desc->index); - ql_unmap_send(qdev, tx_ring_desc, - tx_ring_desc->map_cnt); + qlge_unmap_send(qdev, tx_ring_desc, + tx_ring_desc->map_cnt); dev_kfree_skb(tx_ring_desc->skb); tx_ring_desc->skb = NULL; } @@ -2912,27 +2912,27 @@ static void ql_tx_ring_clean(struct ql_adapter *qdev) } } -static void ql_free_mem_resources(struct ql_adapter *qdev) +static void qlge_free_mem_resources(struct qlge_adapter *qdev) { int i; for (i = 0; i < qdev->tx_ring_count; i++) - ql_free_tx_resources(qdev, &qdev->tx_ring[i]); + qlge_free_tx_resources(qdev, &qdev->tx_ring[i]); for (i = 0; i < qdev->rx_ring_count; i++) - ql_free_rx_resources(qdev, &qdev->rx_ring[i]); - ql_free_shadow_space(qdev); + qlge_free_rx_resources(qdev, &qdev->rx_ring[i]); + qlge_free_shadow_space(qdev); } -static int ql_alloc_mem_resources(struct ql_adapter *qdev) +static int qlge_alloc_mem_resources(struct qlge_adapter *qdev) { int i; /* Allocate space for our shadow registers and such. */ - if (ql_alloc_shadow_space(qdev)) + if (qlge_alloc_shadow_space(qdev)) return -ENOMEM; for (i = 0; i < qdev->rx_ring_count; i++) { - if (ql_alloc_rx_resources(qdev, &qdev->rx_ring[i]) != 0) { + if (qlge_alloc_rx_resources(qdev, &qdev->rx_ring[i]) != 0) { netif_err(qdev, ifup, qdev->ndev, "RX resource allocation failed.\n"); goto err_mem; @@ -2940,7 +2940,7 @@ static int ql_alloc_mem_resources(struct ql_adapter *qdev) } /* Allocate tx queue resources */ for (i = 0; i < qdev->tx_ring_count; i++) { - if (ql_alloc_tx_resources(qdev, &qdev->tx_ring[i]) != 0) { + if (qlge_alloc_tx_resources(qdev, &qdev->tx_ring[i]) != 0) { netif_err(qdev, ifup, qdev->ndev, "TX resource allocation failed.\n"); goto err_mem; @@ -2949,7 +2949,7 @@ static int ql_alloc_mem_resources(struct ql_adapter *qdev) return 0; err_mem: - ql_free_mem_resources(qdev); + qlge_free_mem_resources(qdev); return -ENOMEM; } @@ -2957,7 +2957,7 @@ static int ql_alloc_mem_resources(struct ql_adapter *qdev) * The control block is defined as * "Completion Queue Initialization Control Block", or cqicb. */ -static int ql_start_rx_ring(struct ql_adapter *qdev, struct rx_ring *rx_ring) +static int qlge_start_rx_ring(struct qlge_adapter *qdev, struct rx_ring *rx_ring) { struct cqicb *cqicb = &rx_ring->cqicb; void *shadow_reg = qdev->rx_ring_shadow_reg_area + @@ -2965,7 +2965,7 @@ static int ql_start_rx_ring(struct ql_adapter *qdev, struct rx_ring *rx_ring) u64 shadow_reg_dma = qdev->rx_ring_shadow_reg_dma + (rx_ring->cq_id * RX_RING_SHADOW_SPACE); void __iomem *doorbell_area = - qdev->doorbell_area + (DB_PAGE_SIZE * (128 + rx_ring->cq_id)); + qdev->doorbell_area + (DB_PAGE_SIZE * (128 + rx_ring->cq_id)); int err = 0; u64 tmp; __le64 *base_indirect_ptr; @@ -3012,8 +3012,8 @@ static int ql_start_rx_ring(struct ql_adapter *qdev, struct rx_ring *rx_ring) * Set up the control block load flags. */ cqicb->flags = FLAGS_LC | /* Load queue base address */ - FLAGS_LV | /* Load MSI-X vector */ - FLAGS_LI; /* Load irq delay values */ + FLAGS_LV | /* Load MSI-X vector */ + FLAGS_LI; /* Load irq delay values */ if (rx_ring->cq_id < qdev->rss_ring_count) { cqicb->flags |= FLAGS_LL; /* Load lbq values */ tmp = (u64)rx_ring->lbq.base_dma; @@ -3043,7 +3043,7 @@ static int ql_start_rx_ring(struct ql_adapter *qdev, struct rx_ring *rx_ring) page_entries++; } while (page_entries < MAX_DB_PAGES_PER_BQ(QLGE_BQ_LEN)); cqicb->sbq_addr = - cpu_to_le64(rx_ring->sbq.base_indirect_dma); + cpu_to_le64(rx_ring->sbq.base_indirect_dma); cqicb->sbq_buf_size = cpu_to_le16(SMALL_BUFFER_SIZE); cqicb->sbq_len = cpu_to_le16(QLGE_FIT16(QLGE_BQ_LEN)); rx_ring->sbq.next_to_use = 0; @@ -3053,7 +3053,7 @@ static int ql_start_rx_ring(struct ql_adapter *qdev, struct rx_ring *rx_ring) /* Inbound completion handling rx_rings run in * separate NAPI contexts. */ - netif_napi_add(qdev->ndev, &rx_ring->napi, ql_napi_poll_msix, + netif_napi_add(qdev->ndev, &rx_ring->napi, qlge_napi_poll_msix, 64); cqicb->irq_delay = cpu_to_le16(qdev->rx_coalesce_usecs); cqicb->pkt_delay = cpu_to_le16(qdev->rx_max_coalesced_frames); @@ -3061,8 +3061,8 @@ static int ql_start_rx_ring(struct ql_adapter *qdev, struct rx_ring *rx_ring) cqicb->irq_delay = cpu_to_le16(qdev->tx_coalesce_usecs); cqicb->pkt_delay = cpu_to_le16(qdev->tx_max_coalesced_frames); } - err = ql_write_cfg(qdev, cqicb, sizeof(struct cqicb), - CFG_LCQ, rx_ring->cq_id); + err = qlge_write_cfg(qdev, cqicb, sizeof(struct cqicb), + CFG_LCQ, rx_ring->cq_id); if (err) { netif_err(qdev, ifup, qdev->ndev, "Failed to load CQICB.\n"); return err; @@ -3070,15 +3070,15 @@ static int ql_start_rx_ring(struct ql_adapter *qdev, struct rx_ring *rx_ring) return err; } -static int ql_start_tx_ring(struct ql_adapter *qdev, struct tx_ring *tx_ring) +static int qlge_start_tx_ring(struct qlge_adapter *qdev, struct tx_ring *tx_ring) { struct wqicb *wqicb = (struct wqicb *)tx_ring; void __iomem *doorbell_area = - qdev->doorbell_area + (DB_PAGE_SIZE * tx_ring->wq_id); + qdev->doorbell_area + (DB_PAGE_SIZE * tx_ring->wq_id); void *shadow_reg = qdev->tx_ring_shadow_reg_area + - (tx_ring->wq_id * sizeof(u64)); + (tx_ring->wq_id * sizeof(u64)); u64 shadow_reg_dma = qdev->tx_ring_shadow_reg_dma + - (tx_ring->wq_id * sizeof(u64)); + (tx_ring->wq_id * sizeof(u64)); int err = 0; /* @@ -3105,10 +3105,10 @@ static int ql_start_tx_ring(struct ql_adapter *qdev, struct tx_ring *tx_ring) wqicb->cnsmr_idx_addr = cpu_to_le64(tx_ring->cnsmr_idx_sh_reg_dma); - ql_init_tx_ring(qdev, tx_ring); + qlge_init_tx_ring(qdev, tx_ring); - err = ql_write_cfg(qdev, wqicb, sizeof(*wqicb), CFG_LRQ, - (u16)tx_ring->wq_id); + err = qlge_write_cfg(qdev, wqicb, sizeof(*wqicb), CFG_LRQ, + (u16)tx_ring->wq_id); if (err) { netif_err(qdev, ifup, qdev->ndev, "Failed to load tx_ring.\n"); return err; @@ -3116,7 +3116,7 @@ static int ql_start_tx_ring(struct ql_adapter *qdev, struct tx_ring *tx_ring) return err; } -static void ql_disable_msix(struct ql_adapter *qdev) +static void qlge_disable_msix(struct qlge_adapter *qdev) { if (test_bit(QL_MSIX_ENABLED, &qdev->flags)) { pci_disable_msix(qdev->pdev); @@ -3133,7 +3133,7 @@ static void ql_disable_msix(struct ql_adapter *qdev) * stored in qdev->intr_count. If we don't get that * many then we reduce the count and try again. */ -static void ql_enable_msix(struct ql_adapter *qdev) +static void qlge_enable_msix(struct qlge_adapter *qdev) { int i, err; @@ -3195,7 +3195,7 @@ static void ql_enable_msix(struct ql_adapter *qdev) * and TX completion rings 0,1,2 and 3. Vector 1 would * service RSS ring 1 and TX completion rings 4,5,6 and 7. */ -static void ql_set_tx_vect(struct ql_adapter *qdev) +static void qlge_set_tx_vect(struct qlge_adapter *qdev) { int i, j, vect; u32 tx_rings_per_vector = qdev->tx_ring_count / qdev->intr_count; @@ -3203,7 +3203,7 @@ static void ql_set_tx_vect(struct ql_adapter *qdev) if (likely(test_bit(QL_MSIX_ENABLED, &qdev->flags))) { /* Assign irq vectors to TX rx_rings.*/ for (vect = 0, j = 0, i = qdev->rss_ring_count; - i < qdev->rx_ring_count; i++) { + i < qdev->rx_ring_count; i++) { if (j == tx_rings_per_vector) { vect++; j = 0; @@ -3225,7 +3225,7 @@ static void ql_set_tx_vect(struct ql_adapter *qdev) * rings. This function sets up a bit mask per vector * that indicates which rings it services. */ -static void ql_set_irq_mask(struct ql_adapter *qdev, struct intr_context *ctx) +static void qlge_set_irq_mask(struct qlge_adapter *qdev, struct intr_context *ctx) { int j, vect = ctx->intr; u32 tx_rings_per_vector = qdev->tx_ring_count / qdev->intr_count; @@ -3240,8 +3240,8 @@ static void ql_set_irq_mask(struct ql_adapter *qdev, struct intr_context *ctx) */ for (j = 0; j < tx_rings_per_vector; j++) { ctx->irq_mask |= - (1 << qdev->rx_ring[qdev->rss_ring_count + - (vect * tx_rings_per_vector) + j].cq_id); + (1 << qdev->rx_ring[qdev->rss_ring_count + + (vect * tx_rings_per_vector) + j].cq_id); } } else { /* For single vector we just shift each queue's @@ -3258,7 +3258,7 @@ static void ql_set_irq_mask(struct ql_adapter *qdev, struct intr_context *ctx) * The intr_context structure is used to hook each vector * to possibly different handlers. */ -static void ql_resolve_queues_to_irqs(struct ql_adapter *qdev) +static void qlge_resolve_queues_to_irqs(struct qlge_adapter *qdev) { int i = 0; struct intr_context *intr_context = &qdev->intr_context[0]; @@ -3275,23 +3275,23 @@ static void ql_resolve_queues_to_irqs(struct ql_adapter *qdev) /* Set up this vector's bit-mask that indicates * which queues it services. */ - ql_set_irq_mask(qdev, intr_context); + qlge_set_irq_mask(qdev, intr_context); /* * We set up each vectors enable/disable/read bits so * there's no bit/mask calculations in the critical path. */ intr_context->intr_en_mask = - INTR_EN_TYPE_MASK | INTR_EN_INTR_MASK | - INTR_EN_TYPE_ENABLE | INTR_EN_IHD_MASK | INTR_EN_IHD - | i; + INTR_EN_TYPE_MASK | INTR_EN_INTR_MASK | + INTR_EN_TYPE_ENABLE | INTR_EN_IHD_MASK | INTR_EN_IHD + | i; intr_context->intr_dis_mask = - INTR_EN_TYPE_MASK | INTR_EN_INTR_MASK | - INTR_EN_TYPE_DISABLE | INTR_EN_IHD_MASK | - INTR_EN_IHD | i; + INTR_EN_TYPE_MASK | INTR_EN_INTR_MASK | + INTR_EN_TYPE_DISABLE | INTR_EN_IHD_MASK | + INTR_EN_IHD | i; intr_context->intr_read_mask = - INTR_EN_TYPE_MASK | INTR_EN_INTR_MASK | - INTR_EN_TYPE_READ | INTR_EN_IHD_MASK | INTR_EN_IHD | - i; + INTR_EN_TYPE_MASK | INTR_EN_INTR_MASK | + INTR_EN_TYPE_READ | INTR_EN_IHD_MASK | INTR_EN_IHD | + i; if (i == 0) { /* The first vector/queue handles * broadcast/multicast, fatal errors, @@ -3322,10 +3322,10 @@ static void ql_resolve_queues_to_irqs(struct ql_adapter *qdev) * there's no bit/mask calculations in the critical path. */ intr_context->intr_en_mask = - INTR_EN_TYPE_MASK | INTR_EN_INTR_MASK | INTR_EN_TYPE_ENABLE; + INTR_EN_TYPE_MASK | INTR_EN_INTR_MASK | INTR_EN_TYPE_ENABLE; intr_context->intr_dis_mask = - INTR_EN_TYPE_MASK | INTR_EN_INTR_MASK | - INTR_EN_TYPE_DISABLE; + INTR_EN_TYPE_MASK | INTR_EN_INTR_MASK | + INTR_EN_TYPE_DISABLE; if (test_bit(QL_LEGACY_ENABLED, &qdev->flags)) { /* Experience shows that when using INTx interrupts, * the device does not always auto-mask INTR_EN_EN. @@ -3337,7 +3337,7 @@ static void ql_resolve_queues_to_irqs(struct ql_adapter *qdev) intr_context->intr_dis_mask |= INTR_EN_EI << 16; } intr_context->intr_read_mask = - INTR_EN_TYPE_MASK | INTR_EN_INTR_MASK | INTR_EN_TYPE_READ; + INTR_EN_TYPE_MASK | INTR_EN_INTR_MASK | INTR_EN_TYPE_READ; /* * Single interrupt means one handler for all rings. */ @@ -3348,15 +3348,15 @@ static void ql_resolve_queues_to_irqs(struct ql_adapter *qdev) * a single vector so it will service all RSS and * TX completion rings. */ - ql_set_irq_mask(qdev, intr_context); + qlge_set_irq_mask(qdev, intr_context); } /* Tell the TX completion rings which MSIx vector * they will be using. */ - ql_set_tx_vect(qdev); + qlge_set_tx_vect(qdev); } -static void ql_free_irq(struct ql_adapter *qdev) +static void qlge_free_irq(struct qlge_adapter *qdev) { int i; struct intr_context *intr_context = &qdev->intr_context[0]; @@ -3371,17 +3371,17 @@ static void ql_free_irq(struct ql_adapter *qdev) } } } - ql_disable_msix(qdev); + qlge_disable_msix(qdev); } -static int ql_request_irq(struct ql_adapter *qdev) +static int qlge_request_irq(struct qlge_adapter *qdev) { int i; int status = 0; struct pci_dev *pdev = qdev->pdev; struct intr_context *intr_context = &qdev->intr_context[0]; - ql_resolve_queues_to_irqs(qdev); + qlge_resolve_queues_to_irqs(qdev); for (i = 0; i < qdev->intr_count; i++, intr_context++) { if (test_bit(QL_MSIX_ENABLED, &qdev->flags)) { @@ -3408,11 +3408,11 @@ static int ql_request_irq(struct ql_adapter *qdev) "%s: dev_id = 0x%p.\n", __func__, &qdev->rx_ring[0]); status = - request_irq(pdev->irq, qlge_isr, - test_bit(QL_MSI_ENABLED, &qdev->flags) - ? 0 - : IRQF_SHARED, - intr_context->name, &qdev->rx_ring[0]); + request_irq(pdev->irq, qlge_isr, + test_bit(QL_MSI_ENABLED, &qdev->flags) + ? 0 + : IRQF_SHARED, + intr_context->name, &qdev->rx_ring[0]); if (status) goto err_irq; @@ -3425,11 +3425,11 @@ static int ql_request_irq(struct ql_adapter *qdev) return status; err_irq: netif_err(qdev, ifup, qdev->ndev, "Failed to get the interrupts!!!\n"); - ql_free_irq(qdev); + qlge_free_irq(qdev); return status; } -static int ql_start_rss(struct ql_adapter *qdev) +static int qlge_start_rss(struct qlge_adapter *qdev) { static const u8 init_hash_seed[] = { 0x6d, 0x5a, 0x56, 0xda, 0x25, 0x5b, 0x0e, 0xc2, @@ -3459,7 +3459,7 @@ static int ql_start_rss(struct ql_adapter *qdev) memcpy((void *)&ricb->ipv6_hash_key[0], init_hash_seed, 40); memcpy((void *)&ricb->ipv4_hash_key[0], init_hash_seed, 16); - status = ql_write_cfg(qdev, ricb, sizeof(*ricb), CFG_LR, 0); + status = qlge_write_cfg(qdev, ricb, sizeof(*ricb), CFG_LR, 0); if (status) { netif_err(qdev, ifup, qdev->ndev, "Failed to load RICB.\n"); return status; @@ -3467,55 +3467,55 @@ static int ql_start_rss(struct ql_adapter *qdev) return status; } -static int ql_clear_routing_entries(struct ql_adapter *qdev) +static int qlge_clear_routing_entries(struct qlge_adapter *qdev) { int i, status = 0; - status = ql_sem_spinlock(qdev, SEM_RT_IDX_MASK); + status = qlge_sem_spinlock(qdev, SEM_RT_IDX_MASK); if (status) return status; /* Clear all the entries in the routing table. */ for (i = 0; i < 16; i++) { - status = ql_set_routing_reg(qdev, i, 0, 0); + status = qlge_set_routing_reg(qdev, i, 0, 0); if (status) { netif_err(qdev, ifup, qdev->ndev, "Failed to init routing register for CAM packets.\n"); break; } } - ql_sem_unlock(qdev, SEM_RT_IDX_MASK); + qlge_sem_unlock(qdev, SEM_RT_IDX_MASK); return status; } /* Initialize the frame-to-queue routing. */ -static int ql_route_initialize(struct ql_adapter *qdev) +static int qlge_route_initialize(struct qlge_adapter *qdev) { int status = 0; /* Clear all the entries in the routing table. */ - status = ql_clear_routing_entries(qdev); + status = qlge_clear_routing_entries(qdev); if (status) return status; - status = ql_sem_spinlock(qdev, SEM_RT_IDX_MASK); + status = qlge_sem_spinlock(qdev, SEM_RT_IDX_MASK); if (status) return status; - status = ql_set_routing_reg(qdev, RT_IDX_IP_CSUM_ERR_SLOT, - RT_IDX_IP_CSUM_ERR, 1); + status = qlge_set_routing_reg(qdev, RT_IDX_IP_CSUM_ERR_SLOT, + RT_IDX_IP_CSUM_ERR, 1); if (status) { netif_err(qdev, ifup, qdev->ndev, "Failed to init routing register for IP CSUM error packets.\n"); goto exit; } - status = ql_set_routing_reg(qdev, RT_IDX_TCP_UDP_CSUM_ERR_SLOT, - RT_IDX_TU_CSUM_ERR, 1); + status = qlge_set_routing_reg(qdev, RT_IDX_TCP_UDP_CSUM_ERR_SLOT, + RT_IDX_TU_CSUM_ERR, 1); if (status) { netif_err(qdev, ifup, qdev->ndev, "Failed to init routing register for TCP/UDP CSUM error packets.\n"); goto exit; } - status = ql_set_routing_reg(qdev, RT_IDX_BCAST_SLOT, RT_IDX_BCAST, 1); + status = qlge_set_routing_reg(qdev, RT_IDX_BCAST_SLOT, RT_IDX_BCAST, 1); if (status) { netif_err(qdev, ifup, qdev->ndev, "Failed to init routing register for broadcast packets.\n"); @@ -3525,8 +3525,8 @@ static int ql_route_initialize(struct ql_adapter *qdev) * routing block. */ if (qdev->rss_ring_count > 1) { - status = ql_set_routing_reg(qdev, RT_IDX_RSS_MATCH_SLOT, - RT_IDX_RSS_MATCH, 1); + status = qlge_set_routing_reg(qdev, RT_IDX_RSS_MATCH_SLOT, + RT_IDX_RSS_MATCH, 1); if (status) { netif_err(qdev, ifup, qdev->ndev, "Failed to init routing register for MATCH RSS packets.\n"); @@ -3534,17 +3534,17 @@ static int ql_route_initialize(struct ql_adapter *qdev) } } - status = ql_set_routing_reg(qdev, RT_IDX_CAM_HIT_SLOT, - RT_IDX_CAM_HIT, 1); + status = qlge_set_routing_reg(qdev, RT_IDX_CAM_HIT_SLOT, + RT_IDX_CAM_HIT, 1); if (status) netif_err(qdev, ifup, qdev->ndev, "Failed to init routing register for CAM packets.\n"); exit: - ql_sem_unlock(qdev, SEM_RT_IDX_MASK); + qlge_sem_unlock(qdev, SEM_RT_IDX_MASK); return status; } -int ql_cam_route_initialize(struct ql_adapter *qdev) +int qlge_cam_route_initialize(struct qlge_adapter *qdev) { int status, set; @@ -3552,22 +3552,22 @@ int ql_cam_route_initialize(struct ql_adapter *qdev) * determine if we are setting or clearing * the MAC address in the CAM. */ - set = ql_read32(qdev, STS); + set = qlge_read32(qdev, STS); set &= qdev->port_link_up; - status = ql_set_mac_addr(qdev, set); + status = qlge_set_mac_addr(qdev, set); if (status) { netif_err(qdev, ifup, qdev->ndev, "Failed to init mac address.\n"); return status; } - status = ql_route_initialize(qdev); + status = qlge_route_initialize(qdev); if (status) netif_err(qdev, ifup, qdev->ndev, "Failed to init routing table.\n"); return status; } -static int ql_adapter_initialize(struct ql_adapter *qdev) +static int qlge_adapter_initialize(struct qlge_adapter *qdev) { u32 value, mask; int i; @@ -3578,7 +3578,7 @@ static int ql_adapter_initialize(struct ql_adapter *qdev) */ value = SYS_EFE | SYS_FAE; mask = value << 16; - ql_write32(qdev, SYS, mask | value); + qlge_write32(qdev, SYS, mask | value); /* Set the default queue, and VLAN behavior. */ value = NIC_RCV_CFG_DFQ; @@ -3587,40 +3587,40 @@ static int ql_adapter_initialize(struct ql_adapter *qdev) value |= NIC_RCV_CFG_RV; mask |= (NIC_RCV_CFG_RV << 16); } - ql_write32(qdev, NIC_RCV_CFG, (mask | value)); + qlge_write32(qdev, NIC_RCV_CFG, (mask | value)); /* Set the MPI interrupt to enabled. */ - ql_write32(qdev, INTR_MASK, (INTR_MASK_PI << 16) | INTR_MASK_PI); + qlge_write32(qdev, INTR_MASK, (INTR_MASK_PI << 16) | INTR_MASK_PI); /* Enable the function, set pagesize, enable error checking. */ value = FSC_FE | FSC_EPC_INBOUND | FSC_EPC_OUTBOUND | - FSC_EC | FSC_VM_PAGE_4K; + FSC_EC | FSC_VM_PAGE_4K; value |= SPLT_SETTING; /* Set/clear header splitting. */ mask = FSC_VM_PAGESIZE_MASK | - FSC_DBL_MASK | FSC_DBRST_MASK | (value << 16); - ql_write32(qdev, FSC, mask | value); + FSC_DBL_MASK | FSC_DBRST_MASK | (value << 16); + qlge_write32(qdev, FSC, mask | value); - ql_write32(qdev, SPLT_HDR, SPLT_LEN); + qlge_write32(qdev, SPLT_HDR, SPLT_LEN); /* Set RX packet routing to use port/pci function on which the * packet arrived on in addition to usual frame routing. * This is helpful on bonding where both interfaces can have * the same MAC address. */ - ql_write32(qdev, RST_FO, RST_FO_RR_MASK | RST_FO_RR_RCV_FUNC_CQ); + qlge_write32(qdev, RST_FO, RST_FO_RR_MASK | RST_FO_RR_RCV_FUNC_CQ); /* Reroute all packets to our Interface. * They may have been routed to MPI firmware * due to WOL. */ - value = ql_read32(qdev, MGMT_RCV_CFG); + value = qlge_read32(qdev, MGMT_RCV_CFG); value &= ~MGMT_RCV_CFG_RM; mask = 0xffff0000; /* Sticky reg needs clearing due to WOL. */ - ql_write32(qdev, MGMT_RCV_CFG, mask); - ql_write32(qdev, MGMT_RCV_CFG, mask | value); + qlge_write32(qdev, MGMT_RCV_CFG, mask); + qlge_write32(qdev, MGMT_RCV_CFG, mask | value); /* Default WOL is enable on Mezz cards */ if (qdev->pdev->subsystem_device == 0x0068 || @@ -3629,7 +3629,7 @@ static int ql_adapter_initialize(struct ql_adapter *qdev) /* Start up the rx queues. */ for (i = 0; i < qdev->rx_ring_count; i++) { - status = ql_start_rx_ring(qdev, &qdev->rx_ring[i]); + status = qlge_start_rx_ring(qdev, &qdev->rx_ring[i]); if (status) { netif_err(qdev, ifup, qdev->ndev, "Failed to start rx ring[%d].\n", i); @@ -3641,7 +3641,7 @@ static int ql_adapter_initialize(struct ql_adapter *qdev) * then download a RICB to configure RSS. */ if (qdev->rss_ring_count > 1) { - status = ql_start_rss(qdev); + status = qlge_start_rss(qdev); if (status) { netif_err(qdev, ifup, qdev->ndev, "Failed to start RSS.\n"); return status; @@ -3650,7 +3650,7 @@ static int ql_adapter_initialize(struct ql_adapter *qdev) /* Start up the tx queues. */ for (i = 0; i < qdev->tx_ring_count; i++) { - status = ql_start_tx_ring(qdev, &qdev->tx_ring[i]); + status = qlge_start_tx_ring(qdev, &qdev->tx_ring[i]); if (status) { netif_err(qdev, ifup, qdev->ndev, "Failed to start tx ring[%d].\n", i); @@ -3664,7 +3664,7 @@ static int ql_adapter_initialize(struct ql_adapter *qdev) netif_err(qdev, ifup, qdev->ndev, "Failed to start port.\n"); /* Set up the MAC address and frame routing filter. */ - status = ql_cam_route_initialize(qdev); + status = qlge_cam_route_initialize(qdev); if (status) { netif_err(qdev, ifup, qdev->ndev, "Failed to init CAM/Routing tables.\n"); @@ -3679,14 +3679,14 @@ static int ql_adapter_initialize(struct ql_adapter *qdev) } /* Issue soft reset to chip. */ -static int ql_adapter_reset(struct ql_adapter *qdev) +static int qlge_adapter_reset(struct qlge_adapter *qdev) { u32 value; int status = 0; unsigned long end_jiffies; /* Clear all the entries in the routing table. */ - status = ql_clear_routing_entries(qdev); + status = qlge_clear_routing_entries(qdev); if (status) { netif_err(qdev, ifup, qdev->ndev, "Failed to clear routing bits.\n"); return status; @@ -3697,19 +3697,19 @@ static int ql_adapter_reset(struct ql_adapter *qdev) */ if (!test_bit(QL_ASIC_RECOVERY, &qdev->flags)) { /* Stop management traffic. */ - ql_mb_set_mgmnt_traffic_ctl(qdev, MB_SET_MPI_TFK_STOP); + qlge_mb_set_mgmnt_traffic_ctl(qdev, MB_SET_MPI_TFK_STOP); /* Wait for the NIC and MGMNT FIFOs to empty. */ - ql_wait_fifo_empty(qdev); + qlge_wait_fifo_empty(qdev); } else { clear_bit(QL_ASIC_RECOVERY, &qdev->flags); } - ql_write32(qdev, RST_FO, (RST_FO_FR << 16) | RST_FO_FR); + qlge_write32(qdev, RST_FO, (RST_FO_FR << 16) | RST_FO_FR); end_jiffies = jiffies + usecs_to_jiffies(30); do { - value = ql_read32(qdev, RST_FO); + value = qlge_read32(qdev, RST_FO); if ((value & RST_FO_FR) == 0) break; cpu_relax(); @@ -3722,13 +3722,13 @@ static int ql_adapter_reset(struct ql_adapter *qdev) } /* Resume management traffic. */ - ql_mb_set_mgmnt_traffic_ctl(qdev, MB_SET_MPI_TFK_RESUME); + qlge_mb_set_mgmnt_traffic_ctl(qdev, MB_SET_MPI_TFK_RESUME); return status; } -static void ql_display_dev_info(struct net_device *ndev) +static void qlge_display_dev_info(struct net_device *ndev) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); netif_info(qdev, probe, qdev->ndev, "Function #%d, Port %d, NIC Roll %d, NIC Rev = %d, XG Roll = %d, XG Rev = %d.\n", @@ -3742,7 +3742,7 @@ static void ql_display_dev_info(struct net_device *ndev) "MAC address %pM\n", ndev->dev_addr); } -static int ql_wol(struct ql_adapter *qdev) +static int qlge_wol(struct qlge_adapter *qdev) { int status = 0; u32 wol = MB_WOL_DISABLE; @@ -3755,7 +3755,7 @@ static int ql_wol(struct ql_adapter *qdev) */ if (qdev->wol & (WAKE_ARP | WAKE_MAGICSECURE | WAKE_PHY | WAKE_UCAST | - WAKE_MCAST | WAKE_BCAST)) { + WAKE_MCAST | WAKE_BCAST)) { netif_err(qdev, ifdown, qdev->ndev, "Unsupported WOL parameter. qdev->wol = 0x%x.\n", qdev->wol); @@ -3763,7 +3763,7 @@ static int ql_wol(struct ql_adapter *qdev) } if (qdev->wol & WAKE_MAGIC) { - status = ql_mb_wol_set_magic(qdev, 1); + status = qlge_mb_wol_set_magic(qdev, 1); if (status) { netif_err(qdev, ifdown, qdev->ndev, "Failed to set magic packet on %s.\n", @@ -3779,7 +3779,7 @@ static int ql_wol(struct ql_adapter *qdev) if (qdev->wol) { wol |= MB_WOL_MODE_ON; - status = ql_mb_wol_mode(qdev, wol); + status = qlge_mb_wol_mode(qdev, wol); netif_err(qdev, drv, qdev->ndev, "WOL %s (wol code 0x%x) on %s\n", (status == 0) ? "Successfully set" : "Failed", @@ -3789,7 +3789,7 @@ static int ql_wol(struct ql_adapter *qdev) return status; } -static void ql_cancel_all_work_sync(struct ql_adapter *qdev) +static void qlge_cancel_all_work_sync(struct qlge_adapter *qdev) { /* Don't kill the reset worker thread if we * are in the process of recovery. @@ -3803,54 +3803,54 @@ static void ql_cancel_all_work_sync(struct ql_adapter *qdev) cancel_delayed_work_sync(&qdev->mpi_port_cfg_work); } -static int ql_adapter_down(struct ql_adapter *qdev) +static int qlge_adapter_down(struct qlge_adapter *qdev) { int i, status = 0; - ql_link_off(qdev); + qlge_link_off(qdev); - ql_cancel_all_work_sync(qdev); + qlge_cancel_all_work_sync(qdev); for (i = 0; i < qdev->rss_ring_count; i++) napi_disable(&qdev->rx_ring[i].napi); clear_bit(QL_ADAPTER_UP, &qdev->flags); - ql_disable_interrupts(qdev); + qlge_disable_interrupts(qdev); - ql_tx_ring_clean(qdev); + qlge_tx_ring_clean(qdev); /* Call netif_napi_del() from common point. - */ + */ for (i = 0; i < qdev->rss_ring_count; i++) netif_napi_del(&qdev->rx_ring[i].napi); - status = ql_adapter_reset(qdev); + status = qlge_adapter_reset(qdev); if (status) netif_err(qdev, ifdown, qdev->ndev, "reset(func #%d) FAILED!\n", qdev->func); - ql_free_rx_buffers(qdev); + qlge_free_rx_buffers(qdev); return status; } -static int ql_adapter_up(struct ql_adapter *qdev) +static int qlge_adapter_up(struct qlge_adapter *qdev) { int err = 0; - err = ql_adapter_initialize(qdev); + err = qlge_adapter_initialize(qdev); if (err) { netif_info(qdev, ifup, qdev->ndev, "Unable to initialize adapter.\n"); goto err_init; } set_bit(QL_ADAPTER_UP, &qdev->flags); - ql_alloc_rx_buffers(qdev); + qlge_alloc_rx_buffers(qdev); /* If the port is initialized and the * link is up the turn on the carrier. */ - if ((ql_read32(qdev, STS) & qdev->port_init) && - (ql_read32(qdev, STS) & qdev->port_link_up)) - ql_link_on(qdev); + if ((qlge_read32(qdev, STS) & qdev->port_init) && + (qlge_read32(qdev, STS) & qdev->port_link_up)) + qlge_link_on(qdev); /* Restore rx mode. */ clear_bit(QL_ALLMULTI, &qdev->flags); clear_bit(QL_PROMISCUOUS, &qdev->flags); @@ -3859,34 +3859,34 @@ static int ql_adapter_up(struct ql_adapter *qdev) /* Restore vlan setting. */ qlge_restore_vlan(qdev); - ql_enable_interrupts(qdev); - ql_enable_all_completion_interrupts(qdev); + qlge_enable_interrupts(qdev); + qlge_enable_all_completion_interrupts(qdev); netif_tx_start_all_queues(qdev->ndev); return 0; err_init: - ql_adapter_reset(qdev); + qlge_adapter_reset(qdev); return err; } -static void ql_release_adapter_resources(struct ql_adapter *qdev) +static void qlge_release_adapter_resources(struct qlge_adapter *qdev) { - ql_free_mem_resources(qdev); - ql_free_irq(qdev); + qlge_free_mem_resources(qdev); + qlge_free_irq(qdev); } -static int ql_get_adapter_resources(struct ql_adapter *qdev) +static int qlge_get_adapter_resources(struct qlge_adapter *qdev) { - if (ql_alloc_mem_resources(qdev)) { + if (qlge_alloc_mem_resources(qdev)) { netif_err(qdev, ifup, qdev->ndev, "Unable to allocate memory.\n"); return -ENOMEM; } - return ql_request_irq(qdev); + return qlge_request_irq(qdev); } static int qlge_close(struct net_device *ndev) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); int i; /* If we hit pci_channel_io_perm_failure @@ -3910,12 +3910,12 @@ static int qlge_close(struct net_device *ndev) for (i = 0; i < qdev->rss_ring_count; i++) cancel_delayed_work_sync(&qdev->rx_ring[i].refill_work); - ql_adapter_down(qdev); - ql_release_adapter_resources(qdev); + qlge_adapter_down(qdev); + qlge_release_adapter_resources(qdev); return 0; } -static void qlge_set_lb_size(struct ql_adapter *qdev) +static void qlge_set_lb_size(struct qlge_adapter *qdev) { if (qdev->ndev->mtu <= 1500) qdev->lbq_buf_size = LARGE_BUFFER_MIN_SIZE; @@ -3924,7 +3924,7 @@ static void qlge_set_lb_size(struct ql_adapter *qdev) qdev->lbq_buf_order = get_order(qdev->lbq_buf_size); } -static int ql_configure_rings(struct ql_adapter *qdev) +static int qlge_configure_rings(struct qlge_adapter *qdev) { int i; struct rx_ring *rx_ring; @@ -3933,13 +3933,13 @@ static int ql_configure_rings(struct ql_adapter *qdev) /* In a perfect world we have one RSS ring for each CPU * and each has it's own vector. To do that we ask for - * cpu_cnt vectors. ql_enable_msix() will adjust the + * cpu_cnt vectors. qlge_enable_msix() will adjust the * vector count to what we actually get. We then * allocate an RSS ring for each. * Essentially, we are doing min(cpu_count, msix_vector_count). */ qdev->intr_count = cpu_cnt; - ql_enable_msix(qdev); + qlge_enable_msix(qdev); /* Adjust the RSS ring count to the actual vector count. */ qdev->rss_ring_count = qdev->intr_count; qdev->tx_ring_count = cpu_cnt; @@ -3952,7 +3952,7 @@ static int ql_configure_rings(struct ql_adapter *qdev) tx_ring->wq_id = i; tx_ring->wq_len = qdev->tx_ring_size; tx_ring->wq_size = - tx_ring->wq_len * sizeof(struct ob_mac_iocb_req); + tx_ring->wq_len * sizeof(struct qlge_ob_mac_iocb_req); /* * The completion queue ID for the tx rings start @@ -3973,7 +3973,7 @@ static int ql_configure_rings(struct ql_adapter *qdev) */ rx_ring->cq_len = qdev->rx_ring_size; rx_ring->cq_size = - rx_ring->cq_len * sizeof(struct ql_net_rsp_iocb); + rx_ring->cq_len * sizeof(struct qlge_net_rsp_iocb); rx_ring->lbq.type = QLGE_LB; rx_ring->sbq.type = QLGE_SB; INIT_DELAYED_WORK(&rx_ring->refill_work, @@ -3985,7 +3985,7 @@ static int ql_configure_rings(struct ql_adapter *qdev) /* outbound cq is same size as tx_ring it services. */ rx_ring->cq_len = qdev->tx_ring_size; rx_ring->cq_size = - rx_ring->cq_len * sizeof(struct ql_net_rsp_iocb); + rx_ring->cq_len * sizeof(struct qlge_net_rsp_iocb); } } return 0; @@ -3994,33 +3994,33 @@ static int ql_configure_rings(struct ql_adapter *qdev) static int qlge_open(struct net_device *ndev) { int err = 0; - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); - err = ql_adapter_reset(qdev); + err = qlge_adapter_reset(qdev); if (err) return err; qlge_set_lb_size(qdev); - err = ql_configure_rings(qdev); + err = qlge_configure_rings(qdev); if (err) return err; - err = ql_get_adapter_resources(qdev); + err = qlge_get_adapter_resources(qdev); if (err) goto error_up; - err = ql_adapter_up(qdev); + err = qlge_adapter_up(qdev); if (err) goto error_up; return err; error_up: - ql_release_adapter_resources(qdev); + qlge_release_adapter_resources(qdev); return err; } -static int ql_change_rx_buffers(struct ql_adapter *qdev) +static int qlge_change_rx_buffers(struct qlge_adapter *qdev) { int status; @@ -4041,13 +4041,13 @@ static int ql_change_rx_buffers(struct ql_adapter *qdev) } } - status = ql_adapter_down(qdev); + status = qlge_adapter_down(qdev); if (status) goto error; qlge_set_lb_size(qdev); - status = ql_adapter_up(qdev); + status = qlge_adapter_up(qdev); if (status) goto error; @@ -4062,7 +4062,7 @@ static int ql_change_rx_buffers(struct ql_adapter *qdev) static int qlge_change_mtu(struct net_device *ndev, int new_mtu) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); int status; if (ndev->mtu == 1500 && new_mtu == 9000) @@ -4080,7 +4080,7 @@ static int qlge_change_mtu(struct net_device *ndev, int new_mtu) if (!netif_running(qdev->ndev)) return 0; - status = ql_change_rx_buffers(qdev); + status = qlge_change_rx_buffers(qdev); if (status) { netif_err(qdev, ifup, qdev->ndev, "Changing MTU failed.\n"); @@ -4092,7 +4092,7 @@ static int qlge_change_mtu(struct net_device *ndev, int new_mtu) static struct net_device_stats *qlge_get_stats(struct net_device *ndev) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); struct rx_ring *rx_ring = &qdev->rx_ring[0]; struct tx_ring *tx_ring = &qdev->tx_ring[0]; unsigned long pkts, mcast, dropped, errors, bytes; @@ -4128,11 +4128,11 @@ static struct net_device_stats *qlge_get_stats(struct net_device static void qlge_set_multicast_list(struct net_device *ndev) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); struct netdev_hw_addr *ha; int i, status; - status = ql_sem_spinlock(qdev, SEM_RT_IDX_MASK); + status = qlge_sem_spinlock(qdev, SEM_RT_IDX_MASK); if (status) return; /* @@ -4141,7 +4141,7 @@ static void qlge_set_multicast_list(struct net_device *ndev) */ if (ndev->flags & IFF_PROMISC) { if (!test_bit(QL_PROMISCUOUS, &qdev->flags)) { - if (ql_set_routing_reg + if (qlge_set_routing_reg (qdev, RT_IDX_PROMISCUOUS_SLOT, RT_IDX_VALID, 1)) { netif_err(qdev, hw, qdev->ndev, "Failed to set promiscuous mode.\n"); @@ -4151,7 +4151,7 @@ static void qlge_set_multicast_list(struct net_device *ndev) } } else { if (test_bit(QL_PROMISCUOUS, &qdev->flags)) { - if (ql_set_routing_reg + if (qlge_set_routing_reg (qdev, RT_IDX_PROMISCUOUS_SLOT, RT_IDX_VALID, 0)) { netif_err(qdev, hw, qdev->ndev, "Failed to clear promiscuous mode.\n"); @@ -4168,7 +4168,7 @@ static void qlge_set_multicast_list(struct net_device *ndev) if ((ndev->flags & IFF_ALLMULTI) || (netdev_mc_count(ndev) > MAX_MULTICAST_ENTRIES)) { if (!test_bit(QL_ALLMULTI, &qdev->flags)) { - if (ql_set_routing_reg + if (qlge_set_routing_reg (qdev, RT_IDX_ALLMULTI_SLOT, RT_IDX_MCAST, 1)) { netif_err(qdev, hw, qdev->ndev, "Failed to set all-multi mode.\n"); @@ -4178,7 +4178,7 @@ static void qlge_set_multicast_list(struct net_device *ndev) } } else { if (test_bit(QL_ALLMULTI, &qdev->flags)) { - if (ql_set_routing_reg + if (qlge_set_routing_reg (qdev, RT_IDX_ALLMULTI_SLOT, RT_IDX_MCAST, 0)) { netif_err(qdev, hw, qdev->ndev, "Failed to clear all-multi mode.\n"); @@ -4189,22 +4189,22 @@ static void qlge_set_multicast_list(struct net_device *ndev) } if (!netdev_mc_empty(ndev)) { - status = ql_sem_spinlock(qdev, SEM_MAC_ADDR_MASK); + status = qlge_sem_spinlock(qdev, SEM_MAC_ADDR_MASK); if (status) goto exit; i = 0; netdev_for_each_mc_addr(ha, ndev) { - if (ql_set_mac_addr_reg(qdev, (u8 *)ha->addr, - MAC_ADDR_TYPE_MULTI_MAC, i)) { + if (qlge_set_mac_addr_reg(qdev, (u8 *)ha->addr, + MAC_ADDR_TYPE_MULTI_MAC, i)) { netif_err(qdev, hw, qdev->ndev, "Failed to loadmulticast address.\n"); - ql_sem_unlock(qdev, SEM_MAC_ADDR_MASK); + qlge_sem_unlock(qdev, SEM_MAC_ADDR_MASK); goto exit; } i++; } - ql_sem_unlock(qdev, SEM_MAC_ADDR_MASK); - if (ql_set_routing_reg + qlge_sem_unlock(qdev, SEM_MAC_ADDR_MASK); + if (qlge_set_routing_reg (qdev, RT_IDX_MCAST_MATCH_SLOT, RT_IDX_MCAST_MATCH, 1)) { netif_err(qdev, hw, qdev->ndev, "Failed to set multicast match mode.\n"); @@ -4213,12 +4213,12 @@ static void qlge_set_multicast_list(struct net_device *ndev) } } exit: - ql_sem_unlock(qdev, SEM_RT_IDX_MASK); + qlge_sem_unlock(qdev, SEM_RT_IDX_MASK); } static int qlge_set_mac_address(struct net_device *ndev, void *p) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); struct sockaddr *addr = p; int status; @@ -4228,37 +4228,37 @@ static int qlge_set_mac_address(struct net_device *ndev, void *p) /* Update local copy of current mac address. */ memcpy(qdev->current_mac_addr, ndev->dev_addr, ndev->addr_len); - status = ql_sem_spinlock(qdev, SEM_MAC_ADDR_MASK); + status = qlge_sem_spinlock(qdev, SEM_MAC_ADDR_MASK); if (status) return status; - status = ql_set_mac_addr_reg(qdev, (u8 *)ndev->dev_addr, - MAC_ADDR_TYPE_CAM_MAC, - qdev->func * MAX_CQ); + status = qlge_set_mac_addr_reg(qdev, (u8 *)ndev->dev_addr, + MAC_ADDR_TYPE_CAM_MAC, + qdev->func * MAX_CQ); if (status) netif_err(qdev, hw, qdev->ndev, "Failed to load MAC address.\n"); - ql_sem_unlock(qdev, SEM_MAC_ADDR_MASK); + qlge_sem_unlock(qdev, SEM_MAC_ADDR_MASK); return status; } static void qlge_tx_timeout(struct net_device *ndev, unsigned int txqueue) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); - ql_queue_asic_error(qdev); + qlge_queue_asic_error(qdev); } -static void ql_asic_reset_work(struct work_struct *work) +static void qlge_asic_reset_work(struct work_struct *work) { - struct ql_adapter *qdev = - container_of(work, struct ql_adapter, asic_reset_work.work); + struct qlge_adapter *qdev = + container_of(work, struct qlge_adapter, asic_reset_work.work); int status; rtnl_lock(); - status = ql_adapter_down(qdev); + status = qlge_adapter_down(qdev); if (status) goto error; - status = ql_adapter_up(qdev); + status = qlge_adapter_up(qdev); if (status) goto error; @@ -4279,13 +4279,13 @@ static void ql_asic_reset_work(struct work_struct *work) } static const struct nic_operations qla8012_nic_ops = { - .get_flash = ql_get_8012_flash_params, - .port_initialize = ql_8012_port_initialize, + .get_flash = qlge_get_8012_flash_params, + .port_initialize = qlge_8012_port_initialize, }; static const struct nic_operations qla8000_nic_ops = { - .get_flash = ql_get_8000_flash_params, - .port_initialize = ql_8000_port_initialize, + .get_flash = qlge_get_8000_flash_params, + .port_initialize = qlge_8000_port_initialize, }; /* Find the pcie function number for the other NIC @@ -4295,21 +4295,21 @@ static const struct nic_operations qla8000_nic_ops = { * after a fatal firmware error, or doing a firmware * coredump. */ -static int ql_get_alt_pcie_func(struct ql_adapter *qdev) +static int qlge_get_alt_pcie_func(struct qlge_adapter *qdev) { int status = 0; u32 temp; u32 nic_func1, nic_func2; - status = ql_read_mpi_reg(qdev, MPI_TEST_FUNC_PORT_CFG, - &temp); + status = qlge_read_mpi_reg(qdev, MPI_TEST_FUNC_PORT_CFG, + &temp); if (status) return status; nic_func1 = ((temp >> MPI_TEST_NIC1_FUNC_SHIFT) & - MPI_TEST_NIC_FUNC_MASK); + MPI_TEST_NIC_FUNC_MASK); nic_func2 = ((temp >> MPI_TEST_NIC2_FUNC_SHIFT) & - MPI_TEST_NIC_FUNC_MASK); + MPI_TEST_NIC_FUNC_MASK); if (qdev->func == nic_func1) qdev->alt_func = nic_func2; @@ -4321,16 +4321,16 @@ static int ql_get_alt_pcie_func(struct ql_adapter *qdev) return status; } -static int ql_get_board_info(struct ql_adapter *qdev) +static int qlge_get_board_info(struct qlge_adapter *qdev) { int status; qdev->func = - (ql_read32(qdev, STS) & STS_FUNC_ID_MASK) >> STS_FUNC_ID_SHIFT; + (qlge_read32(qdev, STS) & STS_FUNC_ID_MASK) >> STS_FUNC_ID_SHIFT; if (qdev->func > 3) return -EIO; - status = ql_get_alt_pcie_func(qdev); + status = qlge_get_alt_pcie_func(qdev); if (status) return status; @@ -4348,7 +4348,7 @@ static int ql_get_board_info(struct ql_adapter *qdev) qdev->mailbox_in = PROC_ADDR_MPI_RISC | PROC_ADDR_FUNC0_MBI; qdev->mailbox_out = PROC_ADDR_MPI_RISC | PROC_ADDR_FUNC0_MBO; } - qdev->chip_rev_id = ql_read32(qdev, REV_ID); + qdev->chip_rev_id = qlge_read32(qdev, REV_ID); qdev->device_id = qdev->pdev->device; if (qdev->device_id == QLGE_DEVICE_ID_8012) qdev->nic_ops = &qla8012_nic_ops; @@ -4357,10 +4357,10 @@ static int ql_get_board_info(struct ql_adapter *qdev) return status; } -static void ql_release_all(struct pci_dev *pdev) +static void qlge_release_all(struct pci_dev *pdev) { struct net_device *ndev = pci_get_drvdata(pdev); - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); if (qdev->workqueue) { destroy_workqueue(qdev->workqueue); @@ -4375,10 +4375,10 @@ static void ql_release_all(struct pci_dev *pdev) pci_release_regions(pdev); } -static int ql_init_device(struct pci_dev *pdev, struct net_device *ndev, - int cards_found) +static int qlge_init_device(struct pci_dev *pdev, struct net_device *ndev, + int cards_found) { - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); int err = 0; memset((void *)qdev, 0, sizeof(*qdev)); @@ -4441,7 +4441,7 @@ static int ql_init_device(struct pci_dev *pdev, struct net_device *ndev, goto err_out2; } - err = ql_get_board_info(qdev); + err = qlge_get_board_info(qdev); if (err) { dev_err(&pdev->dev, "Register access failed.\n"); err = -EIO; @@ -4452,7 +4452,7 @@ static int ql_init_device(struct pci_dev *pdev, struct net_device *ndev, if (qlge_mpi_coredump) { qdev->mpi_coredump = - vmalloc(sizeof(struct ql_mpi_coredump)); + vmalloc(sizeof(struct qlge_mpi_coredump)); if (!qdev->mpi_coredump) { err = -ENOMEM; goto err_out2; @@ -4490,12 +4490,12 @@ static int ql_init_device(struct pci_dev *pdev, struct net_device *ndev, goto err_out2; } - INIT_DELAYED_WORK(&qdev->asic_reset_work, ql_asic_reset_work); - INIT_DELAYED_WORK(&qdev->mpi_reset_work, ql_mpi_reset_work); - INIT_DELAYED_WORK(&qdev->mpi_work, ql_mpi_work); - INIT_DELAYED_WORK(&qdev->mpi_port_cfg_work, ql_mpi_port_cfg_work); - INIT_DELAYED_WORK(&qdev->mpi_idc_work, ql_mpi_idc_work); - INIT_DELAYED_WORK(&qdev->mpi_core_to_log, ql_mpi_core_to_log); + INIT_DELAYED_WORK(&qdev->asic_reset_work, qlge_asic_reset_work); + INIT_DELAYED_WORK(&qdev->mpi_reset_work, qlge_mpi_reset_work); + INIT_DELAYED_WORK(&qdev->mpi_work, qlge_mpi_work); + INIT_DELAYED_WORK(&qdev->mpi_port_cfg_work, qlge_mpi_port_cfg_work); + INIT_DELAYED_WORK(&qdev->mpi_idc_work, qlge_mpi_idc_work); + INIT_DELAYED_WORK(&qdev->mpi_core_to_log, qlge_mpi_core_to_log); init_completion(&qdev->ide_completion); mutex_init(&qdev->mpi_mutex); @@ -4506,7 +4506,7 @@ static int ql_init_device(struct pci_dev *pdev, struct net_device *ndev, } return 0; err_out2: - ql_release_all(pdev); + qlge_release_all(pdev); err_out1: pci_disable_device(pdev); return err; @@ -4527,12 +4527,12 @@ static const struct net_device_ops qlge_netdev_ops = { .ndo_vlan_rx_kill_vid = qlge_vlan_rx_kill_vid, }; -static void ql_timer(struct timer_list *t) +static void qlge_timer(struct timer_list *t) { - struct ql_adapter *qdev = from_timer(qdev, t, timer); + struct qlge_adapter *qdev = from_timer(qdev, t, timer); u32 var = 0; - var = ql_read32(qdev, STS); + var = qlge_read32(qdev, STS); if (pci_channel_offline(qdev->pdev)) { netif_err(qdev, ifup, qdev->ndev, "EEH STS = 0x%.08x.\n", var); return; @@ -4545,17 +4545,17 @@ static int qlge_probe(struct pci_dev *pdev, const struct pci_device_id *pci_entry) { struct net_device *ndev = NULL; - struct ql_adapter *qdev = NULL; + struct qlge_adapter *qdev = NULL; static int cards_found; int err = 0; - ndev = alloc_etherdev_mq(sizeof(struct ql_adapter), + ndev = alloc_etherdev_mq(sizeof(struct qlge_adapter), min(MAX_CPUS, netif_get_num_default_rss_queues())); if (!ndev) return -ENOMEM; - err = ql_init_device(pdev, ndev, cards_found); + err = qlge_init_device(pdev, ndev, cards_found); if (err < 0) { free_netdev(ndev); return err; @@ -4564,13 +4564,13 @@ static int qlge_probe(struct pci_dev *pdev, qdev = netdev_priv(ndev); SET_NETDEV_DEV(ndev, &pdev->dev); ndev->hw_features = NETIF_F_SG | - NETIF_F_IP_CSUM | - NETIF_F_TSO | - NETIF_F_TSO_ECN | - NETIF_F_HW_VLAN_CTAG_TX | - NETIF_F_HW_VLAN_CTAG_RX | - NETIF_F_HW_VLAN_CTAG_FILTER | - NETIF_F_RXCSUM; + NETIF_F_IP_CSUM | + NETIF_F_TSO | + NETIF_F_TSO_ECN | + NETIF_F_HW_VLAN_CTAG_TX | + NETIF_F_HW_VLAN_CTAG_RX | + NETIF_F_HW_VLAN_CTAG_FILTER | + NETIF_F_RXCSUM; ndev->features = ndev->hw_features; ndev->vlan_features = ndev->hw_features; /* vlan gets same features (except vlan filter) */ @@ -4601,7 +4601,7 @@ static int qlge_probe(struct pci_dev *pdev, err = register_netdev(ndev); if (err) { dev_err(&pdev->dev, "net device registration failed.\n"); - ql_release_all(pdev); + qlge_release_all(pdev); pci_disable_device(pdev); free_netdev(ndev); return err; @@ -4609,43 +4609,43 @@ static int qlge_probe(struct pci_dev *pdev, /* Start up the timer to trigger EEH if * the bus goes dead */ - timer_setup(&qdev->timer, ql_timer, TIMER_DEFERRABLE); + timer_setup(&qdev->timer, qlge_timer, TIMER_DEFERRABLE); mod_timer(&qdev->timer, jiffies + (5 * HZ)); - ql_link_off(qdev); - ql_display_dev_info(ndev); + qlge_link_off(qdev); + qlge_display_dev_info(ndev); atomic_set(&qdev->lb_count, 0); cards_found++; return 0; } -netdev_tx_t ql_lb_send(struct sk_buff *skb, struct net_device *ndev) +netdev_tx_t qlge_lb_send(struct sk_buff *skb, struct net_device *ndev) { return qlge_send(skb, ndev); } -int ql_clean_lb_rx_ring(struct rx_ring *rx_ring, int budget) +int qlge_clean_lb_rx_ring(struct rx_ring *rx_ring, int budget) { - return ql_clean_inbound_rx_ring(rx_ring, budget); + return qlge_clean_inbound_rx_ring(rx_ring, budget); } static void qlge_remove(struct pci_dev *pdev) { struct net_device *ndev = pci_get_drvdata(pdev); - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); del_timer_sync(&qdev->timer); - ql_cancel_all_work_sync(qdev); + qlge_cancel_all_work_sync(qdev); unregister_netdev(ndev); - ql_release_all(pdev); + qlge_release_all(pdev); pci_disable_device(pdev); free_netdev(ndev); } /* Clean up resources without touching hardware. */ -static void ql_eeh_close(struct net_device *ndev) +static void qlge_eeh_close(struct net_device *ndev) { int i; - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); if (netif_carrier_ok(ndev)) { netif_carrier_off(ndev); @@ -4653,15 +4653,15 @@ static void ql_eeh_close(struct net_device *ndev) } /* Disabling the timer */ - ql_cancel_all_work_sync(qdev); + qlge_cancel_all_work_sync(qdev); for (i = 0; i < qdev->rss_ring_count; i++) netif_napi_del(&qdev->rx_ring[i].napi); clear_bit(QL_ADAPTER_UP, &qdev->flags); - ql_tx_ring_clean(qdev); - ql_free_rx_buffers(qdev); - ql_release_adapter_resources(qdev); + qlge_tx_ring_clean(qdev); + qlge_free_rx_buffers(qdev); + qlge_release_adapter_resources(qdev); } /* @@ -4672,7 +4672,7 @@ static pci_ers_result_t qlge_io_error_detected(struct pci_dev *pdev, pci_channel_state_t state) { struct net_device *ndev = pci_get_drvdata(pdev); - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); switch (state) { case pci_channel_io_normal: @@ -4681,14 +4681,14 @@ static pci_ers_result_t qlge_io_error_detected(struct pci_dev *pdev, netif_device_detach(ndev); del_timer_sync(&qdev->timer); if (netif_running(ndev)) - ql_eeh_close(ndev); + qlge_eeh_close(ndev); pci_disable_device(pdev); return PCI_ERS_RESULT_NEED_RESET; case pci_channel_io_perm_failure: dev_err(&pdev->dev, "%s: pci_channel_io_perm_failure.\n", __func__); del_timer_sync(&qdev->timer); - ql_eeh_close(ndev); + qlge_eeh_close(ndev); set_bit(QL_EEH_FATAL, &qdev->flags); return PCI_ERS_RESULT_DISCONNECT; } @@ -4706,7 +4706,7 @@ static pci_ers_result_t qlge_io_error_detected(struct pci_dev *pdev, static pci_ers_result_t qlge_io_slot_reset(struct pci_dev *pdev) { struct net_device *ndev = pci_get_drvdata(pdev); - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); pdev->error_state = pci_channel_io_normal; @@ -4718,7 +4718,7 @@ static pci_ers_result_t qlge_io_slot_reset(struct pci_dev *pdev) } pci_set_master(pdev); - if (ql_adapter_reset(qdev)) { + if (qlge_adapter_reset(qdev)) { netif_err(qdev, drv, qdev->ndev, "reset FAILED!\n"); set_bit(QL_EEH_FATAL, &qdev->flags); return PCI_ERS_RESULT_DISCONNECT; @@ -4730,7 +4730,7 @@ static pci_ers_result_t qlge_io_slot_reset(struct pci_dev *pdev) static void qlge_io_resume(struct pci_dev *pdev) { struct net_device *ndev = pci_get_drvdata(pdev); - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); int err = 0; if (netif_running(ndev)) { @@ -4757,19 +4757,19 @@ static const struct pci_error_handlers qlge_err_handler = { static int __maybe_unused qlge_suspend(struct device *dev_d) { struct net_device *ndev = dev_get_drvdata(dev_d); - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); int err; netif_device_detach(ndev); del_timer_sync(&qdev->timer); if (netif_running(ndev)) { - err = ql_adapter_down(qdev); + err = qlge_adapter_down(qdev); if (!err) return err; } - ql_wol(qdev); + qlge_wol(qdev); return 0; } @@ -4777,7 +4777,7 @@ static int __maybe_unused qlge_suspend(struct device *dev_d) static int __maybe_unused qlge_resume(struct device *dev_d) { struct net_device *ndev = dev_get_drvdata(dev_d); - struct ql_adapter *qdev = netdev_priv(ndev); + struct qlge_adapter *qdev = netdev_priv(ndev); int err; pci_set_master(to_pci_dev(dev_d)); @@ -4785,7 +4785,7 @@ static int __maybe_unused qlge_resume(struct device *dev_d) device_wakeup_disable(dev_d); if (netif_running(ndev)) { - err = ql_adapter_up(qdev); + err = qlge_adapter_up(qdev); if (err) return err; } diff --git a/drivers/staging/qlge/qlge_mpi.c b/drivers/staging/qlge/qlge_mpi.c index 2ff3db2661a9..2b77995ec76c 100644 --- a/drivers/staging/qlge/qlge_mpi.c +++ b/drivers/staging/qlge/qlge_mpi.c @@ -1,28 +1,28 @@ // SPDX-License-Identifier: GPL-2.0 #include "qlge.h" -int ql_unpause_mpi_risc(struct ql_adapter *qdev) +int qlge_unpause_mpi_risc(struct qlge_adapter *qdev) { u32 tmp; /* Un-pause the RISC */ - tmp = ql_read32(qdev, CSR); + tmp = qlge_read32(qdev, CSR); if (!(tmp & CSR_RP)) return -EIO; - ql_write32(qdev, CSR, CSR_CMD_CLR_PAUSE); + qlge_write32(qdev, CSR, CSR_CMD_CLR_PAUSE); return 0; } -int ql_pause_mpi_risc(struct ql_adapter *qdev) +int qlge_pause_mpi_risc(struct qlge_adapter *qdev) { u32 tmp; int count; /* Pause the RISC */ - ql_write32(qdev, CSR, CSR_CMD_SET_PAUSE); + qlge_write32(qdev, CSR, CSR_CMD_SET_PAUSE); for (count = UDELAY_COUNT; count; count--) { - tmp = ql_read32(qdev, CSR); + tmp = qlge_read32(qdev, CSR); if (tmp & CSR_RP) break; mdelay(UDELAY_DELAY); @@ -30,17 +30,17 @@ int ql_pause_mpi_risc(struct ql_adapter *qdev) return (count == 0) ? -ETIMEDOUT : 0; } -int ql_hard_reset_mpi_risc(struct ql_adapter *qdev) +int qlge_hard_reset_mpi_risc(struct qlge_adapter *qdev) { u32 tmp; int count; /* Reset the RISC */ - ql_write32(qdev, CSR, CSR_CMD_SET_RST); + qlge_write32(qdev, CSR, CSR_CMD_SET_RST); for (count = UDELAY_COUNT; count; count--) { - tmp = ql_read32(qdev, CSR); + tmp = qlge_read32(qdev, CSR); if (tmp & CSR_RR) { - ql_write32(qdev, CSR, CSR_CMD_CLR_RST); + qlge_write32(qdev, CSR, CSR_CMD_CLR_RST); break; } mdelay(UDELAY_DELAY); @@ -48,47 +48,47 @@ int ql_hard_reset_mpi_risc(struct ql_adapter *qdev) return (count == 0) ? -ETIMEDOUT : 0; } -int ql_read_mpi_reg(struct ql_adapter *qdev, u32 reg, u32 *data) +int qlge_read_mpi_reg(struct qlge_adapter *qdev, u32 reg, u32 *data) { int status; /* wait for reg to come ready */ - status = ql_wait_reg_rdy(qdev, PROC_ADDR, PROC_ADDR_RDY, PROC_ADDR_ERR); + status = qlge_wait_reg_rdy(qdev, PROC_ADDR, PROC_ADDR_RDY, PROC_ADDR_ERR); if (status) goto exit; /* set up for reg read */ - ql_write32(qdev, PROC_ADDR, reg | PROC_ADDR_R); + qlge_write32(qdev, PROC_ADDR, reg | PROC_ADDR_R); /* wait for reg to come ready */ - status = ql_wait_reg_rdy(qdev, PROC_ADDR, PROC_ADDR_RDY, PROC_ADDR_ERR); + status = qlge_wait_reg_rdy(qdev, PROC_ADDR, PROC_ADDR_RDY, PROC_ADDR_ERR); if (status) goto exit; /* get the data */ - *data = ql_read32(qdev, PROC_DATA); + *data = qlge_read32(qdev, PROC_DATA); exit: return status; } -int ql_write_mpi_reg(struct ql_adapter *qdev, u32 reg, u32 data) +int qlge_write_mpi_reg(struct qlge_adapter *qdev, u32 reg, u32 data) { int status = 0; /* wait for reg to come ready */ - status = ql_wait_reg_rdy(qdev, PROC_ADDR, PROC_ADDR_RDY, PROC_ADDR_ERR); + status = qlge_wait_reg_rdy(qdev, PROC_ADDR, PROC_ADDR_RDY, PROC_ADDR_ERR); if (status) goto exit; /* write the data to the data reg */ - ql_write32(qdev, PROC_DATA, data); + qlge_write32(qdev, PROC_DATA, data); /* trigger the write */ - ql_write32(qdev, PROC_ADDR, reg); + qlge_write32(qdev, PROC_ADDR, reg); /* wait for reg to come ready */ - status = ql_wait_reg_rdy(qdev, PROC_ADDR, PROC_ADDR_RDY, PROC_ADDR_ERR); + status = qlge_wait_reg_rdy(qdev, PROC_ADDR, PROC_ADDR_RDY, PROC_ADDR_ERR); if (status) goto exit; exit: return status; } -int ql_soft_reset_mpi_risc(struct ql_adapter *qdev) +int qlge_soft_reset_mpi_risc(struct qlge_adapter *qdev) { - return ql_write_mpi_reg(qdev, 0x00001010, 1); + return qlge_write_mpi_reg(qdev, 0x00001010, 1); } /* Determine if we are in charge of the firmware. If @@ -96,7 +96,7 @@ int ql_soft_reset_mpi_risc(struct ql_adapter *qdev) * we are the higher function and the lower function * is not enabled. */ -int ql_own_firmware(struct ql_adapter *qdev) +int qlge_own_firmware(struct qlge_adapter *qdev) { u32 temp; @@ -112,43 +112,43 @@ int ql_own_firmware(struct ql_adapter *qdev) * enabled, then we are responsible for * core dump and firmware reset after an error. */ - temp = ql_read32(qdev, STS); + temp = qlge_read32(qdev, STS); if (!(temp & (1 << (8 + qdev->alt_func)))) return 1; return 0; } -static int ql_get_mb_sts(struct ql_adapter *qdev, struct mbox_params *mbcp) +static int qlge_get_mb_sts(struct qlge_adapter *qdev, struct mbox_params *mbcp) { int i, status; - status = ql_sem_spinlock(qdev, SEM_PROC_REG_MASK); + status = qlge_sem_spinlock(qdev, SEM_PROC_REG_MASK); if (status) return -EBUSY; for (i = 0; i < mbcp->out_count; i++) { status = - ql_read_mpi_reg(qdev, qdev->mailbox_out + i, - &mbcp->mbox_out[i]); + qlge_read_mpi_reg(qdev, qdev->mailbox_out + i, + &mbcp->mbox_out[i]); if (status) { netif_err(qdev, drv, qdev->ndev, "Failed mailbox read.\n"); break; } } - ql_sem_unlock(qdev, SEM_PROC_REG_MASK); /* does flush too */ + qlge_sem_unlock(qdev, SEM_PROC_REG_MASK); /* does flush too */ return status; } /* Wait for a single mailbox command to complete. * Returns zero on success. */ -static int ql_wait_mbx_cmd_cmplt(struct ql_adapter *qdev) +static int qlge_wait_mbx_cmd_cmplt(struct qlge_adapter *qdev) { int count; u32 value; for (count = 100; count; count--) { - value = ql_read32(qdev, STS); + value = qlge_read32(qdev, STS); if (value & STS_PI) return 0; mdelay(UDELAY_DELAY); /* 100ms */ @@ -159,7 +159,7 @@ static int ql_wait_mbx_cmd_cmplt(struct ql_adapter *qdev) /* Execute a single mailbox command. * Caller must hold PROC_ADDR semaphore. */ -static int ql_exec_mb_cmd(struct ql_adapter *qdev, struct mbox_params *mbcp) +static int qlge_exec_mb_cmd(struct qlge_adapter *qdev, struct mbox_params *mbcp) { int i, status; @@ -167,10 +167,10 @@ static int ql_exec_mb_cmd(struct ql_adapter *qdev, struct mbox_params *mbcp) * Make sure there's nothing pending. * This shouldn't happen. */ - if (ql_read32(qdev, CSR) & CSR_HRI) + if (qlge_read32(qdev, CSR) & CSR_HRI) return -EIO; - status = ql_sem_spinlock(qdev, SEM_PROC_REG_MASK); + status = qlge_sem_spinlock(qdev, SEM_PROC_REG_MASK); if (status) return status; @@ -178,17 +178,17 @@ static int ql_exec_mb_cmd(struct ql_adapter *qdev, struct mbox_params *mbcp) * Fill the outbound mailboxes. */ for (i = 0; i < mbcp->in_count; i++) { - status = ql_write_mpi_reg(qdev, qdev->mailbox_in + i, - mbcp->mbox_in[i]); + status = qlge_write_mpi_reg(qdev, qdev->mailbox_in + i, + mbcp->mbox_in[i]); if (status) goto end; } /* * Wake up the MPI firmware. */ - ql_write32(qdev, CSR, CSR_CMD_SET_H2R_INT); + qlge_write32(qdev, CSR, CSR_CMD_SET_H2R_INT); end: - ql_sem_unlock(qdev, SEM_PROC_REG_MASK); + qlge_sem_unlock(qdev, SEM_PROC_REG_MASK); return status; } @@ -199,7 +199,7 @@ static int ql_exec_mb_cmd(struct ql_adapter *qdev, struct mbox_params *mbcp) * to handler processing this since a mailbox command * will need to be sent to ACK the request. */ -static int ql_idc_req_aen(struct ql_adapter *qdev) +static int qlge_idc_req_aen(struct qlge_adapter *qdev) { int status; struct mbox_params *mbcp = &qdev->idc_mbc; @@ -209,17 +209,17 @@ static int ql_idc_req_aen(struct ql_adapter *qdev) * handle the request. */ mbcp->out_count = 4; - status = ql_get_mb_sts(qdev, mbcp); + status = qlge_get_mb_sts(qdev, mbcp); if (status) { netif_err(qdev, drv, qdev->ndev, "Could not read MPI, resetting ASIC!\n"); - ql_queue_asic_error(qdev); + qlge_queue_asic_error(qdev); } else { /* Begin polled mode early so * we don't get another interrupt * when we leave mpi_worker. */ - ql_write32(qdev, INTR_MASK, (INTR_MASK_PI << 16)); + qlge_write32(qdev, INTR_MASK, (INTR_MASK_PI << 16)); queue_delayed_work(qdev->workqueue, &qdev->mpi_idc_work, 0); } return status; @@ -228,17 +228,17 @@ static int ql_idc_req_aen(struct ql_adapter *qdev) /* Process an inter-device event completion. * If good, signal the caller's completion. */ -static int ql_idc_cmplt_aen(struct ql_adapter *qdev) +static int qlge_idc_cmplt_aen(struct qlge_adapter *qdev) { int status; struct mbox_params *mbcp = &qdev->idc_mbc; mbcp->out_count = 4; - status = ql_get_mb_sts(qdev, mbcp); + status = qlge_get_mb_sts(qdev, mbcp); if (status) { netif_err(qdev, drv, qdev->ndev, "Could not read MPI, resetting RISC!\n"); - ql_queue_fw_error(qdev); + qlge_queue_fw_error(qdev); } else { /* Wake up the sleeping mpi_idc_work thread that is * waiting for this event. @@ -248,13 +248,13 @@ static int ql_idc_cmplt_aen(struct ql_adapter *qdev) return status; } -static void ql_link_up(struct ql_adapter *qdev, struct mbox_params *mbcp) +static void qlge_link_up(struct qlge_adapter *qdev, struct mbox_params *mbcp) { int status; mbcp->out_count = 2; - status = ql_get_mb_sts(qdev, mbcp); + status = qlge_get_mb_sts(qdev, mbcp); if (status) { netif_err(qdev, drv, qdev->ndev, "%s: Could not get mailbox status.\n", __func__); @@ -268,7 +268,7 @@ static void ql_link_up(struct ql_adapter *qdev, struct mbox_params *mbcp) * then set up the CAM and frame routing. */ if (test_bit(QL_CAM_RT_SET, &qdev->flags)) { - status = ql_cam_route_initialize(qdev); + status = qlge_cam_route_initialize(qdev); if (status) { netif_err(qdev, ifup, qdev->ndev, "Failed to init CAM/Routing tables.\n"); @@ -288,34 +288,34 @@ static void ql_link_up(struct ql_adapter *qdev, struct mbox_params *mbcp) * we don't get another interrupt * when we leave mpi_worker dpc. */ - ql_write32(qdev, INTR_MASK, (INTR_MASK_PI << 16)); + qlge_write32(qdev, INTR_MASK, (INTR_MASK_PI << 16)); queue_delayed_work(qdev->workqueue, &qdev->mpi_port_cfg_work, 0); } - ql_link_on(qdev); + qlge_link_on(qdev); } -static void ql_link_down(struct ql_adapter *qdev, struct mbox_params *mbcp) +static void qlge_link_down(struct qlge_adapter *qdev, struct mbox_params *mbcp) { int status; mbcp->out_count = 3; - status = ql_get_mb_sts(qdev, mbcp); + status = qlge_get_mb_sts(qdev, mbcp); if (status) netif_err(qdev, drv, qdev->ndev, "Link down AEN broken!\n"); - ql_link_off(qdev); + qlge_link_off(qdev); } -static int ql_sfp_in(struct ql_adapter *qdev, struct mbox_params *mbcp) +static int qlge_sfp_in(struct qlge_adapter *qdev, struct mbox_params *mbcp) { int status; mbcp->out_count = 5; - status = ql_get_mb_sts(qdev, mbcp); + status = qlge_get_mb_sts(qdev, mbcp); if (status) netif_err(qdev, drv, qdev->ndev, "SFP in AEN broken!\n"); else @@ -324,13 +324,13 @@ static int ql_sfp_in(struct ql_adapter *qdev, struct mbox_params *mbcp) return status; } -static int ql_sfp_out(struct ql_adapter *qdev, struct mbox_params *mbcp) +static int qlge_sfp_out(struct qlge_adapter *qdev, struct mbox_params *mbcp) { int status; mbcp->out_count = 1; - status = ql_get_mb_sts(qdev, mbcp); + status = qlge_get_mb_sts(qdev, mbcp); if (status) netif_err(qdev, drv, qdev->ndev, "SFP out AEN broken!\n"); else @@ -339,13 +339,13 @@ static int ql_sfp_out(struct ql_adapter *qdev, struct mbox_params *mbcp) return status; } -static int ql_aen_lost(struct ql_adapter *qdev, struct mbox_params *mbcp) +static int qlge_aen_lost(struct qlge_adapter *qdev, struct mbox_params *mbcp) { int status; mbcp->out_count = 6; - status = ql_get_mb_sts(qdev, mbcp); + status = qlge_get_mb_sts(qdev, mbcp); if (status) { netif_err(qdev, drv, qdev->ndev, "Lost AEN broken!\n"); } else { @@ -360,20 +360,20 @@ static int ql_aen_lost(struct ql_adapter *qdev, struct mbox_params *mbcp) return status; } -static void ql_init_fw_done(struct ql_adapter *qdev, struct mbox_params *mbcp) +static void qlge_init_fw_done(struct qlge_adapter *qdev, struct mbox_params *mbcp) { int status; mbcp->out_count = 2; - status = ql_get_mb_sts(qdev, mbcp); + status = qlge_get_mb_sts(qdev, mbcp); if (status) { netif_err(qdev, drv, qdev->ndev, "Firmware did not initialize!\n"); } else { netif_err(qdev, drv, qdev->ndev, "Firmware Revision = 0x%.08x.\n", mbcp->mbox_out[1]); qdev->fw_rev_id = mbcp->mbox_out[1]; - status = ql_cam_route_initialize(qdev); + status = qlge_cam_route_initialize(qdev); if (status) netif_err(qdev, ifup, qdev->ndev, "Failed to init CAM/Routing tables.\n"); @@ -387,26 +387,26 @@ static void ql_init_fw_done(struct ql_adapter *qdev, struct mbox_params *mbcp) * It also gets called when a mailbox command is polling for * it's completion. */ -static int ql_mpi_handler(struct ql_adapter *qdev, struct mbox_params *mbcp) +static int qlge_mpi_handler(struct qlge_adapter *qdev, struct mbox_params *mbcp) { int status; int orig_count = mbcp->out_count; /* Just get mailbox zero for now. */ mbcp->out_count = 1; - status = ql_get_mb_sts(qdev, mbcp); + status = qlge_get_mb_sts(qdev, mbcp); if (status) { netif_err(qdev, drv, qdev->ndev, "Could not read MPI, resetting ASIC!\n"); - ql_queue_asic_error(qdev); + qlge_queue_asic_error(qdev); goto end; } switch (mbcp->mbox_out[0]) { - /* This case is only active when we arrive here - * as a result of issuing a mailbox command to - * the firmware. - */ + /* This case is only active when we arrive here + * as a result of issuing a mailbox command to + * the firmware. + */ case MB_CMD_STS_INTRMDT: case MB_CMD_STS_GOOD: case MB_CMD_STS_INVLD_CMD: @@ -421,34 +421,34 @@ static int ql_mpi_handler(struct ql_adapter *qdev, struct mbox_params *mbcp) * command completion. */ mbcp->out_count = orig_count; - status = ql_get_mb_sts(qdev, mbcp); + status = qlge_get_mb_sts(qdev, mbcp); return status; - /* We are being asked by firmware to accept - * a change to the port. This is only - * a change to max frame sizes (Tx/Rx), pause - * parameters, or loopback mode. - */ + /* We are being asked by firmware to accept + * a change to the port. This is only + * a change to max frame sizes (Tx/Rx), pause + * parameters, or loopback mode. + */ case AEN_IDC_REQ: - status = ql_idc_req_aen(qdev); + status = qlge_idc_req_aen(qdev); break; - /* Process and inbound IDC event. - * This will happen when we're trying to - * change tx/rx max frame size, change pause - * parameters or loopback mode. - */ + /* Process and inbound IDC event. + * This will happen when we're trying to + * change tx/rx max frame size, change pause + * parameters or loopback mode. + */ case AEN_IDC_CMPLT: case AEN_IDC_EXT: - status = ql_idc_cmplt_aen(qdev); + status = qlge_idc_cmplt_aen(qdev); break; case AEN_LINK_UP: - ql_link_up(qdev, mbcp); + qlge_link_up(qdev, mbcp); break; case AEN_LINK_DOWN: - ql_link_down(qdev, mbcp); + qlge_link_down(qdev, mbcp); break; case AEN_FW_INIT_DONE: @@ -457,48 +457,48 @@ static int ql_mpi_handler(struct ql_adapter *qdev, struct mbox_params *mbcp) */ if (mbcp->mbox_in[0] == MB_CMD_EX_FW) { mbcp->out_count = orig_count; - status = ql_get_mb_sts(qdev, mbcp); + status = qlge_get_mb_sts(qdev, mbcp); mbcp->mbox_out[0] = MB_CMD_STS_GOOD; return status; } - ql_init_fw_done(qdev, mbcp); + qlge_init_fw_done(qdev, mbcp); break; case AEN_AEN_SFP_IN: - ql_sfp_in(qdev, mbcp); + qlge_sfp_in(qdev, mbcp); break; case AEN_AEN_SFP_OUT: - ql_sfp_out(qdev, mbcp); + qlge_sfp_out(qdev, mbcp); break; - /* This event can arrive at boot time or after an - * MPI reset if the firmware failed to initialize. - */ + /* This event can arrive at boot time or after an + * MPI reset if the firmware failed to initialize. + */ case AEN_FW_INIT_FAIL: /* If we're in process on executing the firmware, * then convert the status to normal mailbox status. */ if (mbcp->mbox_in[0] == MB_CMD_EX_FW) { mbcp->out_count = orig_count; - status = ql_get_mb_sts(qdev, mbcp); + status = qlge_get_mb_sts(qdev, mbcp); mbcp->mbox_out[0] = MB_CMD_STS_ERR; return status; } netif_err(qdev, drv, qdev->ndev, "Firmware initialization failed.\n"); status = -EIO; - ql_queue_fw_error(qdev); + qlge_queue_fw_error(qdev); break; case AEN_SYS_ERR: netif_err(qdev, drv, qdev->ndev, "System Error.\n"); - ql_queue_fw_error(qdev); + qlge_queue_fw_error(qdev); status = -EIO; break; case AEN_AEN_LOST: - ql_aen_lost(qdev, mbcp); + qlge_aen_lost(qdev, mbcp); break; case AEN_DCBX_CHG: @@ -510,7 +510,7 @@ static int ql_mpi_handler(struct ql_adapter *qdev, struct mbox_params *mbcp) /* Clear the MPI firmware status. */ } end: - ql_write32(qdev, CSR, CSR_CMD_CLR_R2PCI_INT); + qlge_write32(qdev, CSR, CSR_CMD_CLR_R2PCI_INT); /* Restore the original mailbox count to * what the caller asked for. This can get * changed when a mailbox command is waiting @@ -526,7 +526,7 @@ static int ql_mpi_handler(struct ql_adapter *qdev, struct mbox_params *mbcp) * element in the array contains the value for it's * respective mailbox register. */ -static int ql_mailbox_command(struct ql_adapter *qdev, struct mbox_params *mbcp) +static int qlge_mailbox_command(struct qlge_adapter *qdev, struct mbox_params *mbcp) { int status; unsigned long count; @@ -534,10 +534,10 @@ static int ql_mailbox_command(struct ql_adapter *qdev, struct mbox_params *mbcp) mutex_lock(&qdev->mpi_mutex); /* Begin polled mode for MPI */ - ql_write32(qdev, INTR_MASK, (INTR_MASK_PI << 16)); + qlge_write32(qdev, INTR_MASK, (INTR_MASK_PI << 16)); /* Load the mailbox registers and wake up MPI RISC. */ - status = ql_exec_mb_cmd(qdev, mbcp); + status = qlge_exec_mb_cmd(qdev, mbcp); if (status) goto end; @@ -556,7 +556,7 @@ static int ql_mailbox_command(struct ql_adapter *qdev, struct mbox_params *mbcp) count = jiffies + HZ * MAILBOX_TIMEOUT; do { /* Wait for the interrupt to come in. */ - status = ql_wait_mbx_cmd_cmplt(qdev); + status = qlge_wait_mbx_cmd_cmplt(qdev); if (status) continue; @@ -565,7 +565,7 @@ static int ql_mailbox_command(struct ql_adapter *qdev, struct mbox_params *mbcp) * will be spawned. If it's our completion * we will catch it below. */ - status = ql_mpi_handler(qdev, mbcp); + status = qlge_mpi_handler(qdev, mbcp); if (status) goto end; @@ -574,9 +574,9 @@ static int ql_mailbox_command(struct ql_adapter *qdev, struct mbox_params *mbcp) * completion then get out. */ if (((mbcp->mbox_out[0] & 0x0000f000) == - MB_CMD_STS_GOOD) || - ((mbcp->mbox_out[0] & 0x0000f000) == - MB_CMD_STS_INTRMDT)) + MB_CMD_STS_GOOD) || + ((mbcp->mbox_out[0] & 0x0000f000) == + MB_CMD_STS_INTRMDT)) goto done; } while (time_before(jiffies, count)); @@ -590,17 +590,17 @@ static int ql_mailbox_command(struct ql_adapter *qdev, struct mbox_params *mbcp) /* Now we can clear the interrupt condition * and look at our status. */ - ql_write32(qdev, CSR, CSR_CMD_CLR_R2PCI_INT); + qlge_write32(qdev, CSR, CSR_CMD_CLR_R2PCI_INT); if (((mbcp->mbox_out[0] & 0x0000f000) != - MB_CMD_STS_GOOD) && - ((mbcp->mbox_out[0] & 0x0000f000) != - MB_CMD_STS_INTRMDT)) { + MB_CMD_STS_GOOD) && + ((mbcp->mbox_out[0] & 0x0000f000) != + MB_CMD_STS_INTRMDT)) { status = -EIO; } end: /* End polled mode for MPI */ - ql_write32(qdev, INTR_MASK, (INTR_MASK_PI << 16) | INTR_MASK_PI); + qlge_write32(qdev, INTR_MASK, (INTR_MASK_PI << 16) | INTR_MASK_PI); mutex_unlock(&qdev->mpi_mutex); return status; } @@ -609,7 +609,7 @@ static int ql_mailbox_command(struct ql_adapter *qdev, struct mbox_params *mbcp) * driver banner and for ethtool info. * Returns zero on success. */ -int ql_mb_about_fw(struct ql_adapter *qdev) +int qlge_mb_about_fw(struct qlge_adapter *qdev) { struct mbox_params mbc; struct mbox_params *mbcp = &mbc; @@ -622,7 +622,7 @@ int ql_mb_about_fw(struct ql_adapter *qdev) mbcp->mbox_in[0] = MB_CMD_ABOUT_FW; - status = ql_mailbox_command(qdev, mbcp); + status = qlge_mailbox_command(qdev, mbcp); if (status) return status; @@ -641,7 +641,7 @@ int ql_mb_about_fw(struct ql_adapter *qdev) /* Get functional state for MPI firmware. * Returns zero on success. */ -int ql_mb_get_fw_state(struct ql_adapter *qdev) +int qlge_mb_get_fw_state(struct qlge_adapter *qdev) { struct mbox_params mbc; struct mbox_params *mbcp = &mbc; @@ -654,7 +654,7 @@ int ql_mb_get_fw_state(struct ql_adapter *qdev) mbcp->mbox_in[0] = MB_CMD_GET_FW_STATE; - status = ql_mailbox_command(qdev, mbcp); + status = qlge_mailbox_command(qdev, mbcp); if (status) return status; @@ -680,7 +680,7 @@ int ql_mb_get_fw_state(struct ql_adapter *qdev) /* Send and ACK mailbox command to the firmware to * let it continue with the change. */ -static int ql_mb_idc_ack(struct ql_adapter *qdev) +static int qlge_mb_idc_ack(struct qlge_adapter *qdev) { struct mbox_params mbc; struct mbox_params *mbcp = &mbc; @@ -697,7 +697,7 @@ static int ql_mb_idc_ack(struct ql_adapter *qdev) mbcp->mbox_in[3] = qdev->idc_mbc.mbox_out[3]; mbcp->mbox_in[4] = qdev->idc_mbc.mbox_out[4]; - status = ql_mailbox_command(qdev, mbcp); + status = qlge_mailbox_command(qdev, mbcp); if (status) return status; @@ -712,7 +712,7 @@ static int ql_mb_idc_ack(struct ql_adapter *qdev) * for the current port. * Most likely will block. */ -int ql_mb_set_port_cfg(struct ql_adapter *qdev) +int qlge_mb_set_port_cfg(struct qlge_adapter *qdev) { struct mbox_params mbc; struct mbox_params *mbcp = &mbc; @@ -727,7 +727,7 @@ int ql_mb_set_port_cfg(struct ql_adapter *qdev) mbcp->mbox_in[1] = qdev->link_config; mbcp->mbox_in[2] = qdev->max_frame_size; - status = ql_mailbox_command(qdev, mbcp); + status = qlge_mailbox_command(qdev, mbcp); if (status) return status; @@ -742,8 +742,8 @@ int ql_mb_set_port_cfg(struct ql_adapter *qdev) return status; } -static int ql_mb_dump_ram(struct ql_adapter *qdev, u64 req_dma, u32 addr, - u32 size) +static int qlge_mb_dump_ram(struct qlge_adapter *qdev, u64 req_dma, u32 addr, + u32 size) { int status = 0; struct mbox_params mbc; @@ -764,7 +764,7 @@ static int ql_mb_dump_ram(struct ql_adapter *qdev, u64 req_dma, u32 addr, mbcp->mbox_in[7] = LSW(MSD(req_dma)); mbcp->mbox_in[8] = MSW(addr); - status = ql_mailbox_command(qdev, mbcp); + status = qlge_mailbox_command(qdev, mbcp); if (status) return status; @@ -776,8 +776,8 @@ static int ql_mb_dump_ram(struct ql_adapter *qdev, u64 req_dma, u32 addr, } /* Issue a mailbox command to dump RISC RAM. */ -int ql_dump_risc_ram_area(struct ql_adapter *qdev, void *buf, - u32 ram_addr, int word_count) +int qlge_dump_risc_ram_area(struct qlge_adapter *qdev, void *buf, + u32 ram_addr, int word_count) { int status; char *my_buf; @@ -789,7 +789,7 @@ int ql_dump_risc_ram_area(struct ql_adapter *qdev, void *buf, if (!my_buf) return -EIO; - status = ql_mb_dump_ram(qdev, buf_dma, ram_addr, word_count); + status = qlge_mb_dump_ram(qdev, buf_dma, ram_addr, word_count); if (!status) memcpy(buf, my_buf, word_count * sizeof(u32)); @@ -802,7 +802,7 @@ int ql_dump_risc_ram_area(struct ql_adapter *qdev, void *buf, * for the current port. * Most likely will block. */ -int ql_mb_get_port_cfg(struct ql_adapter *qdev) +int qlge_mb_get_port_cfg(struct qlge_adapter *qdev) { struct mbox_params mbc; struct mbox_params *mbcp = &mbc; @@ -815,7 +815,7 @@ int ql_mb_get_port_cfg(struct ql_adapter *qdev) mbcp->mbox_in[0] = MB_CMD_GET_PORT_CFG; - status = ql_mailbox_command(qdev, mbcp); + status = qlge_mailbox_command(qdev, mbcp); if (status) return status; @@ -832,7 +832,7 @@ int ql_mb_get_port_cfg(struct ql_adapter *qdev) return status; } -int ql_mb_wol_mode(struct ql_adapter *qdev, u32 wol) +int qlge_mb_wol_mode(struct qlge_adapter *qdev, u32 wol) { struct mbox_params mbc; struct mbox_params *mbcp = &mbc; @@ -846,7 +846,7 @@ int ql_mb_wol_mode(struct ql_adapter *qdev, u32 wol) mbcp->mbox_in[0] = MB_CMD_SET_WOL_MODE; mbcp->mbox_in[1] = wol; - status = ql_mailbox_command(qdev, mbcp); + status = qlge_mailbox_command(qdev, mbcp); if (status) return status; @@ -857,7 +857,7 @@ int ql_mb_wol_mode(struct ql_adapter *qdev, u32 wol) return status; } -int ql_mb_wol_set_magic(struct ql_adapter *qdev, u32 enable_wol) +int qlge_mb_wol_set_magic(struct qlge_adapter *qdev, u32 enable_wol) { struct mbox_params mbc; struct mbox_params *mbcp = &mbc; @@ -888,7 +888,7 @@ int ql_mb_wol_set_magic(struct ql_adapter *qdev, u32 enable_wol) mbcp->mbox_in[7] = 0; } - status = ql_mailbox_command(qdev, mbcp); + status = qlge_mailbox_command(qdev, mbcp); if (status) return status; @@ -906,7 +906,7 @@ int ql_mb_wol_set_magic(struct ql_adapter *qdev, u32 enable_wol) * The firmware will complete the request if the other * function doesn't respond. */ -static int ql_idc_wait(struct ql_adapter *qdev) +static int qlge_idc_wait(struct qlge_adapter *qdev) { int status = -ETIMEDOUT; struct mbox_params *mbcp = &qdev->idc_mbc; @@ -947,7 +947,7 @@ static int ql_idc_wait(struct ql_adapter *qdev) return status; } -int ql_mb_set_led_cfg(struct ql_adapter *qdev, u32 led_config) +int qlge_mb_set_led_cfg(struct qlge_adapter *qdev, u32 led_config) { struct mbox_params mbc; struct mbox_params *mbcp = &mbc; @@ -961,7 +961,7 @@ int ql_mb_set_led_cfg(struct ql_adapter *qdev, u32 led_config) mbcp->mbox_in[0] = MB_CMD_SET_LED_CFG; mbcp->mbox_in[1] = led_config; - status = ql_mailbox_command(qdev, mbcp); + status = qlge_mailbox_command(qdev, mbcp); if (status) return status; @@ -974,7 +974,7 @@ int ql_mb_set_led_cfg(struct ql_adapter *qdev, u32 led_config) return status; } -int ql_mb_get_led_cfg(struct ql_adapter *qdev) +int qlge_mb_get_led_cfg(struct qlge_adapter *qdev) { struct mbox_params mbc; struct mbox_params *mbcp = &mbc; @@ -987,7 +987,7 @@ int ql_mb_get_led_cfg(struct ql_adapter *qdev) mbcp->mbox_in[0] = MB_CMD_GET_LED_CFG; - status = ql_mailbox_command(qdev, mbcp); + status = qlge_mailbox_command(qdev, mbcp); if (status) return status; @@ -1001,7 +1001,7 @@ int ql_mb_get_led_cfg(struct ql_adapter *qdev) return status; } -int ql_mb_set_mgmnt_traffic_ctl(struct ql_adapter *qdev, u32 control) +int qlge_mb_set_mgmnt_traffic_ctl(struct qlge_adapter *qdev, u32 control) { struct mbox_params mbc; struct mbox_params *mbcp = &mbc; @@ -1015,7 +1015,7 @@ int ql_mb_set_mgmnt_traffic_ctl(struct ql_adapter *qdev, u32 control) mbcp->mbox_in[0] = MB_CMD_SET_MGMNT_TFK_CTL; mbcp->mbox_in[1] = control; - status = ql_mailbox_command(qdev, mbcp); + status = qlge_mailbox_command(qdev, mbcp); if (status) return status; @@ -1038,7 +1038,7 @@ int ql_mb_set_mgmnt_traffic_ctl(struct ql_adapter *qdev, u32 control) } /* Returns a negative error code or the mailbox command status. */ -static int ql_mb_get_mgmnt_traffic_ctl(struct ql_adapter *qdev, u32 *control) +static int qlge_mb_get_mgmnt_traffic_ctl(struct qlge_adapter *qdev, u32 *control) { struct mbox_params mbc; struct mbox_params *mbcp = &mbc; @@ -1052,7 +1052,7 @@ static int ql_mb_get_mgmnt_traffic_ctl(struct ql_adapter *qdev, u32 *control) mbcp->mbox_in[0] = MB_CMD_GET_MGMNT_TFK_CTL; - status = ql_mailbox_command(qdev, mbcp); + status = qlge_mailbox_command(qdev, mbcp); if (status) return status; @@ -1073,15 +1073,15 @@ static int ql_mb_get_mgmnt_traffic_ctl(struct ql_adapter *qdev, u32 *control) return status; } -int ql_wait_fifo_empty(struct ql_adapter *qdev) +int qlge_wait_fifo_empty(struct qlge_adapter *qdev) { int count; u32 mgmnt_fifo_empty; u32 nic_fifo_empty; for (count = 6; count; count--) { - nic_fifo_empty = ql_read32(qdev, STS) & STS_NFE; - ql_mb_get_mgmnt_traffic_ctl(qdev, &mgmnt_fifo_empty); + nic_fifo_empty = qlge_read32(qdev, STS) & STS_NFE; + qlge_mb_get_mgmnt_traffic_ctl(qdev, &mgmnt_fifo_empty); mgmnt_fifo_empty &= MB_GET_MPI_TFK_FIFO_EMPTY; if (nic_fifo_empty && mgmnt_fifo_empty) return 0; @@ -1093,14 +1093,14 @@ int ql_wait_fifo_empty(struct ql_adapter *qdev) /* API called in work thread context to set new TX/RX * maximum frame size values to match MTU. */ -static int ql_set_port_cfg(struct ql_adapter *qdev) +static int qlge_set_port_cfg(struct qlge_adapter *qdev) { int status; - status = ql_mb_set_port_cfg(qdev); + status = qlge_mb_set_port_cfg(qdev); if (status) return status; - status = ql_idc_wait(qdev); + status = qlge_idc_wait(qdev); return status; } @@ -1112,13 +1112,13 @@ static int ql_set_port_cfg(struct ql_adapter *qdev) * from the firmware and, if necessary, changes them to match * the MTU setting. */ -void ql_mpi_port_cfg_work(struct work_struct *work) +void qlge_mpi_port_cfg_work(struct work_struct *work) { - struct ql_adapter *qdev = - container_of(work, struct ql_adapter, mpi_port_cfg_work.work); + struct qlge_adapter *qdev = + container_of(work, struct qlge_adapter, mpi_port_cfg_work.work); int status; - status = ql_mb_get_port_cfg(qdev); + status = qlge_mb_get_port_cfg(qdev); if (status) { netif_err(qdev, drv, qdev->ndev, "Bug: Failed to get port config data.\n"); @@ -1131,7 +1131,7 @@ void ql_mpi_port_cfg_work(struct work_struct *work) qdev->link_config |= CFG_JUMBO_FRAME_SIZE; qdev->max_frame_size = CFG_DEFAULT_MAX_FRAME_SIZE; - status = ql_set_port_cfg(qdev); + status = qlge_set_port_cfg(qdev); if (status) { netif_err(qdev, drv, qdev->ndev, "Bug: Failed to set port config data.\n"); @@ -1141,7 +1141,7 @@ void ql_mpi_port_cfg_work(struct work_struct *work) clear_bit(QL_PORT_CFG, &qdev->flags); return; err: - ql_queue_fw_error(qdev); + qlge_queue_fw_error(qdev); goto end; } @@ -1151,10 +1151,10 @@ void ql_mpi_port_cfg_work(struct work_struct *work) * has been made and then send a mailbox command ACKing * the change request. */ -void ql_mpi_idc_work(struct work_struct *work) +void qlge_mpi_idc_work(struct work_struct *work) { - struct ql_adapter *qdev = - container_of(work, struct ql_adapter, mpi_idc_work.work); + struct qlge_adapter *qdev = + container_of(work, struct qlge_adapter, mpi_idc_work.work); int status; struct mbox_params *mbcp = &qdev->idc_mbc; u32 aen; @@ -1170,7 +1170,7 @@ void ql_mpi_idc_work(struct work_struct *work) break; case MB_CMD_PORT_RESET: case MB_CMD_STOP_FW: - ql_link_off(qdev); + qlge_link_off(qdev); fallthrough; case MB_CMD_SET_PORT_CFG: /* Signal the resulting link up AEN @@ -1180,7 +1180,7 @@ void ql_mpi_idc_work(struct work_struct *work) set_bit(QL_CAM_RT_SET, &qdev->flags); /* Do ACK if required */ if (timeout) { - status = ql_mb_idc_ack(qdev); + status = qlge_mb_idc_ack(qdev); if (status) netif_err(qdev, drv, qdev->ndev, "Bug: No pending IDC!\n"); @@ -1191,18 +1191,18 @@ void ql_mpi_idc_work(struct work_struct *work) } break; - /* These sub-commands issued by another (FCoE) - * function are requesting to do an operation - * on the shared resource (MPI environment). - * We currently don't issue these so we just - * ACK the request. - */ + /* These sub-commands issued by another (FCoE) + * function are requesting to do an operation + * on the shared resource (MPI environment). + * We currently don't issue these so we just + * ACK the request. + */ case MB_CMD_IOP_RESTART_MPI: case MB_CMD_IOP_PREP_LINK_DOWN: /* Drop the link, reload the routing * table when link comes up. */ - ql_link_off(qdev); + qlge_link_off(qdev); set_bit(QL_CAM_RT_SET, &qdev->flags); fallthrough; case MB_CMD_IOP_DVR_START: @@ -1213,7 +1213,7 @@ void ql_mpi_idc_work(struct work_struct *work) case MB_CMD_IOP_NONE: /* an IDC without params */ /* Do ACK if required */ if (timeout) { - status = ql_mb_idc_ack(qdev); + status = qlge_mb_idc_ack(qdev); if (status) netif_err(qdev, drv, qdev->ndev, "Bug: No pending IDC!\n"); @@ -1226,54 +1226,54 @@ void ql_mpi_idc_work(struct work_struct *work) } } -void ql_mpi_work(struct work_struct *work) +void qlge_mpi_work(struct work_struct *work) { - struct ql_adapter *qdev = - container_of(work, struct ql_adapter, mpi_work.work); + struct qlge_adapter *qdev = + container_of(work, struct qlge_adapter, mpi_work.work); struct mbox_params mbc; struct mbox_params *mbcp = &mbc; int err = 0; mutex_lock(&qdev->mpi_mutex); /* Begin polled mode for MPI */ - ql_write32(qdev, INTR_MASK, (INTR_MASK_PI << 16)); + qlge_write32(qdev, INTR_MASK, (INTR_MASK_PI << 16)); - while (ql_read32(qdev, STS) & STS_PI) { + while (qlge_read32(qdev, STS) & STS_PI) { memset(mbcp, 0, sizeof(struct mbox_params)); mbcp->out_count = 1; /* Don't continue if an async event * did not complete properly. */ - err = ql_mpi_handler(qdev, mbcp); + err = qlge_mpi_handler(qdev, mbcp); if (err) break; } /* End polled mode for MPI */ - ql_write32(qdev, INTR_MASK, (INTR_MASK_PI << 16) | INTR_MASK_PI); + qlge_write32(qdev, INTR_MASK, (INTR_MASK_PI << 16) | INTR_MASK_PI); mutex_unlock(&qdev->mpi_mutex); } -void ql_mpi_reset_work(struct work_struct *work) +void qlge_mpi_reset_work(struct work_struct *work) { - struct ql_adapter *qdev = - container_of(work, struct ql_adapter, mpi_reset_work.work); + struct qlge_adapter *qdev = + container_of(work, struct qlge_adapter, mpi_reset_work.work); cancel_delayed_work_sync(&qdev->mpi_work); cancel_delayed_work_sync(&qdev->mpi_port_cfg_work); cancel_delayed_work_sync(&qdev->mpi_idc_work); /* If we're not the dominant NIC function, * then there is nothing to do. */ - if (!ql_own_firmware(qdev)) { + if (!qlge_own_firmware(qdev)) { netif_err(qdev, drv, qdev->ndev, "Don't own firmware!\n"); return; } - if (qdev->mpi_coredump && !ql_core_dump(qdev, qdev->mpi_coredump)) { + if (qdev->mpi_coredump && !qlge_core_dump(qdev, qdev->mpi_coredump)) { netif_err(qdev, drv, qdev->ndev, "Core is dumped!\n"); qdev->core_is_dumped = 1; queue_delayed_work(qdev->workqueue, &qdev->mpi_core_to_log, 5 * HZ); } - ql_soft_reset_mpi_risc(qdev); + qlge_soft_reset_mpi_risc(qdev); }