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 <[email protected]>
Reviewed-by: Tomer Tayar <[email protected]>
---
drivers/misc/habanalabs/common/device.c | 6 ++++--
drivers/misc/habanalabs/common/habanalabs.h | 11 ++++++++---
2 files changed, 12 insertions(+), 5 deletions(-)
diff --git a/drivers/misc/habanalabs/common/device.c b/drivers/misc/habanalabs/common/device.c
index 0650e511a0f5..63d0cb7087e8 100644
--- 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 0781b8698f74..e7f89868428d 100644
--- 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)
--
2.25.1