Don't reset on HW fault, but take the interface down.

There is currently no interface to try again, other than detaching and
re-attaching the line discipline. It wouldn't make much sense either,
unless we've run into an unexpected state in the ELM327's firmware.

diff --git a/module/elmcan.c b/module/elmcan.c
index 6ae52975fa2ff100a3520bdd7ff6cb6bec575d1a..fee83c528d174b4a88feb68582d1aa53d4d8f795 100644 (file)
--- a/module/elmcan.c
+++ b/module/elmcan.c
@@ -93,6 +93,11 @@ struct elmcan {
         */
        atomic_t                refcount;
 
+       /* Stop the channel on hardware failure.
+        * Once this is true, nothing will be sent to the TTY.
+        */
+       bool                    hw_failure;
+
        /* TTY TX helpers */
        struct work_struct      tx_work;        /* Flushes TTY TX buffer   */
        unsigned char           txbuf[32];
@@ -149,6 +154,10 @@ static void elm327_send(struct elmcan *elm, const void *buf, size_t len)
 {
        int actual;
 
+       if (elm->hw_failure) {
+               return;
+       }
+
        memcpy(elm->txbuf, buf, len);
 
        /* Order of next two lines is *very* important.
@@ -327,14 +336,18 @@ static inline void elm327_hw_failure(struct elmcan *elm)
 {
        struct can_frame frame = {0};
 
-       frame.can_id = CAN_ERR_FLAG | CAN_ERR_RESTARTED;
+       frame.can_id = CAN_ERR_FLAG;
        frame.can_dlc = CAN_ERR_DLC;
+       frame.data[5] = 'R';
+       frame.data[6] = 'I';
+       frame.data[7] = 'P';
        elm327_feed_frame_to_netdev(elm, &frame);
 
-       pr_err("ELM327 misbehaved. Re-initializing.\n");
+       netdev_err(elm->dev, "ELM327 misbehaved. "
+                       "Blocking further communication.\n");
 
-       elm->can.can_stats.restarts++;
-       elm327_init(elm);
+       elm->hw_failure = true;
+       can_bus_off(elm->dev);
 }
 
 
@@ -714,6 +727,13 @@ static int elmcan_netdev_open(struct net_device *dev)
        int err;
 
        spin_lock_bh(&elm->lock);
+       if (elm->hw_failure) {
+               netdev_err(elm->dev, "Refusing to open interface after "
+                               "a hardware fault has been detected.\n");
+               spin_unlock_bh(&elm->lock);
+               return -EIO;
+       }
+
        if (elm->tty == NULL) {
                spin_unlock_bh(&elm->lock);
                return -ENODEV;
@@ -787,6 +807,12 @@ static netdev_tx_t elmcan_netdev_start_xmit(struct sk_buff *skb, struct net_devi
         * See Documentation/networking/netdevices.txt
         */
        spin_lock(&elm->lock);
+
+       /* We shouldn't get here after a hardware fault:
+        * can_bus_off() calls netif_carrier_off()
+        */
+       BUG_ON(elm->hw_failure);
+
        if (elm->tty == NULL
                || elm->can.ctrlmode & CAN_CTRLMODE_LISTENONLY) {
                spin_unlock(&elm->lock);
@@ -933,7 +959,11 @@ static void elmcan_ldisc_tx_worker(struct work_struct *work)
        ssize_t actual;
 
        spin_lock_bh(&elm->lock);
-       /* First make sure we're connected. */
+       if (elm->hw_failure) {
+               spin_unlock_bh(&elm->lock);
+               return;
+       }
+
        if (!elm->tty || !netif_running(elm->dev)) {
                spin_unlock_bh(&elm->lock);
                return;

diff --git a/readme.rst b/readme.rst
index df78721e711f285f0cb2d3b7592909cdb9d7deac..d9eb5c17bd63ad1bb5d15bb5484a53f76b4495fc 100644 (file)
--- a/readme.rst
+++ b/readme.rst
@@ -260,8 +260,6 @@ incomplete) data frame::
 To Do list for future development
 ----------------------------------
 
-- No auto-restart in ``elm327_panic()``?
-
 - Stop current function when in ``elm327_panic()``
 
 - DMA capable rx/tx buffers