habanalabs: extend process wait timeout in device fine

Processes that use our device are likely to use at the same time other
devices such as remote storage.

In case our device is removed and a user process is still using the
device, we need to kill the user process. However, if that process
has a thread waiting for i/o to complete on remote storage, for example,
the process won't terminate.

Let's give it enough time to terminate before giving up.

Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Reviewed-by: Tomer Tayar <ttayar@habana.ai>

diff --git a/drivers/misc/habanalabs/common/device.c b/drivers/misc/habanalabs/common/device.c
index 0650e51..63d0cb7 100644 (file)
--- a/drivers/misc/habanalabs/common/device.c
+++ b/drivers/misc/habanalabs/common/device.c
@@ -2300,14 +2300,16 @@ void hl_device_fini(struct hl_device *hdev)
         */
        dev_info(hdev->dev,
                "Waiting for all processes to exit (timeout of %u seconds)",
-               HL_PENDING_RESET_LONG_SEC);
+               HL_WAIT_PROCESS_KILL_ON_DEVICE_FINI);
 
-       rc = device_kill_open_processes(hdev, HL_PENDING_RESET_LONG_SEC, false);
+       hdev->process_kill_trial_cnt = 0;
+       rc = device_kill_open_processes(hdev, HL_WAIT_PROCESS_KILL_ON_DEVICE_FINI, false);
        if (rc) {
                dev_crit(hdev->dev, "Failed to kill all open processes\n");
                device_disable_open_processes(hdev, false);
        }
 
+       hdev->process_kill_trial_cnt = 0;
        rc = device_kill_open_processes(hdev, 0, true);
        if (rc) {
                dev_crit(hdev->dev, "Failed to kill all control device open processes\n");

diff --git a/drivers/misc/habanalabs/common/habanalabs.h b/drivers/misc/habanalabs/common/habanalabs.h
index 0781b86..e7f8986 100644 (file)
--- a/drivers/misc/habanalabs/common/habanalabs.h
+++ b/drivers/misc/habanalabs/common/habanalabs.h
@@ -50,9 +50,14 @@ struct hl_fpriv;
 #define HL_MMAP_OFFSET_VALUE_MASK      (0x1FFFFFFFFFFFull >> PAGE_SHIFT)
 #define HL_MMAP_OFFSET_VALUE_GET(off)  (off & HL_MMAP_OFFSET_VALUE_MASK)
 
-#define HL_PENDING_RESET_PER_SEC       10
-#define HL_PENDING_RESET_MAX_TRIALS    60 /* 10 minutes */
-#define HL_PENDING_RESET_LONG_SEC      60
+#define HL_PENDING_RESET_PER_SEC               10
+#define HL_PENDING_RESET_MAX_TRIALS            60 /* 10 minutes */
+#define HL_PENDING_RESET_LONG_SEC              60
+/*
+ * In device fini, wait 10 minutes for user processes to be terminated after we kill them.
+ * This is needed to prevent situation of clearing resources while user processes are still alive.
+ */
+#define HL_WAIT_PROCESS_KILL_ON_DEVICE_FINI    600
 
 #define HL_HARD_RESET_MAX_TIMEOUT      120
 #define HL_PLDM_HARD_RESET_MAX_TIMEOUT (HL_HARD_RESET_MAX_TIMEOUT * 3)