1/* 2 * Copyright (c) 2007 Cisco Systems, Inc. All rights reserved. 3 * Copyright (c) 2007, 2008 Mellanox Technologies. All rights reserved. 4 * 5 * This software is available to you under a choice of one of two 6 * licenses. You may choose to be licensed under the terms of the GNU 7 * General Public License (GPL) Version 2, available from the file 8 * COPYING in the main directory of this source tree, or the 9 * OpenIB.org BSD license below: 10 * 11 * Redistribution and use in source and binary forms, with or 12 * without modification, are permitted provided that the following 13 * conditions are met: 14 * 15 * - Redistributions of source code must retain the above 16 * copyright notice, this list of conditions and the following 17 * disclaimer. 18 * 19 * - Redistributions in binary form must reproduce the above 20 * copyright notice, this list of conditions and the following 21 * disclaimer in the documentation and/or other materials 22 * provided with the distribution. 23 * 24 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, 25 * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF 26 * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND 27 * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS 28 * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN 29 * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN 30 * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE 31 * SOFTWARE. 32 */ 33 34#include <linux/workqueue.h> 35 36#include "mlx4.h" 37 38enum { 39 MLX4_CATAS_POLL_INTERVAL = 5 * HZ, 40}; 41 42static DEFINE_SPINLOCK(catas_lock); 43 44static LIST_HEAD(catas_list); 45static struct work_struct catas_work; 46 47static int internal_err_reset = 1; 48module_param(internal_err_reset, int, 0644); 49MODULE_PARM_DESC(internal_err_reset, 50 "Reset device on internal errors if non-zero (default 1)"); 51 52static void dump_err_buf(struct mlx4_dev *dev) 53{ 54 struct mlx4_priv *priv = mlx4_priv(dev); 55 56 int i; 57 58 mlx4_err(dev, "Internal error detected:\n"); 59 for (i = 0; i < priv->fw.catas_size; ++i) 60 mlx4_err(dev, " buf[%02x]: %08x\n", 61 i, swab32(readl(priv->catas_err.map + i))); 62} 63 64static void poll_catas(unsigned long dev_ptr) 65{ 66 struct mlx4_dev *dev = (struct mlx4_dev *) dev_ptr; 67 struct mlx4_priv *priv = mlx4_priv(dev); 68 69 if (readl(priv->catas_err.map)) { 70 dump_err_buf(dev); 71 72 mlx4_dispatch_event(dev, MLX4_DEV_EVENT_CATASTROPHIC_ERROR, 0); 73 74 if (internal_err_reset) { 75 spin_lock(&catas_lock); 76 list_add(&priv->catas_err.list, &catas_list); 77 spin_unlock(&catas_lock); 78 79 queue_work(mlx4_wq, &catas_work); 80 } 81 } else 82 mod_timer(&priv->catas_err.timer, 83 round_jiffies(jiffies + MLX4_CATAS_POLL_INTERVAL)); 84} 85 86static void catas_reset(struct work_struct *work) 87{ 88 struct mlx4_priv *priv, *tmppriv; 89 struct mlx4_dev *dev; 90 91 LIST_HEAD(tlist); 92 int ret; 93 94 spin_lock_irq(&catas_lock); 95 list_splice_init(&catas_list, &tlist); 96 spin_unlock_irq(&catas_lock); 97 98 list_for_each_entry_safe(priv, tmppriv, &tlist, catas_err.list) { 99 struct pci_dev *pdev = priv->dev.pdev; 100 101 ret = mlx4_restart_one(priv->dev.pdev); 102 /* 'priv' now is not valid */ 103 if (ret) 104 pr_err("mlx4 %s: Reset failed (%d)\n", 105 pci_name(pdev), ret); 106 else { 107 dev = pci_get_drvdata(pdev); 108 mlx4_dbg(dev, "Reset succeeded\n"); 109 } 110 } 111} 112 113void mlx4_start_catas_poll(struct mlx4_dev *dev) 114{ 115 struct mlx4_priv *priv = mlx4_priv(dev); 116 unsigned long addr; 117 118 INIT_LIST_HEAD(&priv->catas_err.list); 119 init_timer(&priv->catas_err.timer); 120 priv->catas_err.map = NULL; 121 122 addr = pci_resource_start(dev->pdev, priv->fw.catas_bar) + 123 priv->fw.catas_offset; 124 125 priv->catas_err.map = ioremap(addr, priv->fw.catas_size * 4); 126 if (!priv->catas_err.map) { 127 mlx4_warn(dev, "Failed to map internal error buffer at 0x%lx\n", 128 addr); 129 return; 130 } 131 132 priv->catas_err.timer.data = (unsigned long) dev; 133 priv->catas_err.timer.function = poll_catas; 134 priv->catas_err.timer.expires = 135 round_jiffies(jiffies + MLX4_CATAS_POLL_INTERVAL); 136 add_timer(&priv->catas_err.timer); 137} 138 139void mlx4_stop_catas_poll(struct mlx4_dev *dev) 140{ 141 struct mlx4_priv *priv = mlx4_priv(dev); 142 143 del_timer_sync(&priv->catas_err.timer); 144 145 if (priv->catas_err.map) 146 iounmap(priv->catas_err.map); 147 148 spin_lock_irq(&catas_lock); 149 list_del(&priv->catas_err.list); 150 spin_unlock_irq(&catas_lock); 151} 152 153void __init mlx4_catas_init(void) 154{ 155 INIT_WORK(&catas_work, catas_reset); 156} 157