habanalabs: add hard reset timeout for PLDM

Hard reset flow on PLDM might take more than 2 minutes.
Hence add a dedicated hard reset timeout of 6 minutes for PLDM.

Signed-off-by: Omer Shpigelman <oshpigelman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>

diff --git a/drivers/misc/habanalabs/common/device.c b/drivers/misc/habanalabs/common/device.c
index 46fcab1..cbdf75b 100644 (file)
--- a/drivers/misc/habanalabs/common/device.c
+++ b/drivers/misc/habanalabs/common/device.c
@@ -1501,6 +1501,7 @@ out_disabled:
 void hl_device_fini(struct hl_device *hdev)
 {
        ktime_t timeout;
+       u64 reset_sec;
        int i, rc;
 
        dev_info(hdev->dev, "Removing device\n");
@@ -1508,6 +1509,11 @@ void hl_device_fini(struct hl_device *hdev)
        hdev->device_fini_pending = 1;
        flush_delayed_work(&hdev->device_reset_work.reset_work);
 
+       if (hdev->pldm)
+               reset_sec = HL_PLDM_HARD_RESET_MAX_TIMEOUT;
+       else
+               reset_sec = HL_HARD_RESET_MAX_TIMEOUT;
+
        /*
         * This function is competing with the reset function, so try to
         * take the reset atomic and if we are already in middle of reset,
@@ -1516,8 +1522,7 @@ void hl_device_fini(struct hl_device *hdev)
         * ports, the hard reset could take between 10-30 seconds
         */
 
-       timeout = ktime_add_us(ktime_get(),
-                               HL_HARD_RESET_MAX_TIMEOUT * 1000 * 1000);
+       timeout = ktime_add_us(ktime_get(), reset_sec * 1000 * 1000);
        rc = atomic_cmpxchg(&hdev->in_reset, 0, 1);
        while (rc) {
                usleep_range(50, 200);

diff --git a/drivers/misc/habanalabs/common/habanalabs.h b/drivers/misc/habanalabs/common/habanalabs.h
index bc5a1b4..244fbf2 100644 (file)
--- a/drivers/misc/habanalabs/common/habanalabs.h
+++ b/drivers/misc/habanalabs/common/habanalabs.h
@@ -48,6 +48,7 @@
 #define HL_PENDING_RESET_LONG_SEC      60
 
 #define HL_HARD_RESET_MAX_TIMEOUT      120
+#define HL_PLDM_HARD_RESET_MAX_TIMEOUT (HL_HARD_RESET_MAX_TIMEOUT * 3)
 
 #define HL_DEVICE_TIMEOUT_USEC         1000000 /* 1 s */