habanalabs: handle events during soft-reset

Driver should handle events during soft-reset as F/W is not
going through reset and it keeps sending events towards host.

Signed-off-by: Ofir Bitton <obitton@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 822d9cec5aaf65b08ff55d394d53f15a4b4665fe..720eea0b7e9c7193612d54ae3b5971e6505a80c2 100644 (file)
--- a/drivers/misc/habanalabs/common/device.c
+++ b/drivers/misc/habanalabs/common/device.c
@@ -1019,6 +1019,8 @@ do_reset:
 
                handle_reset_trigger(hdev, flags);
 
+               hdev->is_in_soft_reset = !hard_reset;
+
                /* This also blocks future CS/VM/JOB completion operations */
                hdev->disabled = true;
 
@@ -1171,6 +1173,7 @@ kill_processes:
         * is required for the initialization itself
         */
        hdev->disabled = false;
+       hdev->is_in_soft_reset = false;
 
        rc = hdev->asic_funcs->hw_init(hdev);
        if (rc) {
@@ -1242,6 +1245,7 @@ kill_processes:
 
 out_err:
        hdev->disabled = true;
+       hdev->is_in_soft_reset = false;
 
        if (hard_reset) {
                dev_err(hdev->dev, "Failed to reset! Device is NOT usable\n");

diff --git a/drivers/misc/habanalabs/common/habanalabs.h b/drivers/misc/habanalabs/common/habanalabs.h
index a465b4a5f31dcff05bcd58ea79039f358b12eaf3..c2129c9fe9e4239fec8a9baa93b7b58a1919d114 100644 (file)
--- a/drivers/misc/habanalabs/common/habanalabs.h
+++ b/drivers/misc/habanalabs/common/habanalabs.h
@@ -2591,6 +2591,7 @@ struct last_error_session_info {
  *                        protocol will throw an error. Relevant only for
  *                        cases where Linux was not loaded to device CPU
  * @supports_wait_for_multi_cs: true if wait for multi CS is supported
+ * @is_in_soft_reset: Device is currently in soft reset process.
  */
 struct hl_device {
        struct pci_dev                  *pdev;
@@ -2719,6 +2720,7 @@ struct hl_device {
        u8                              device_cpu_is_halted;
        u8                              supports_wait_for_multi_cs;
        u8                              stream_master_qid_arr_size;
+       u8                              is_in_soft_reset;
 
        /* Parameters for bring-up */
        u64                             nic_ports_mask;

diff --git a/drivers/misc/habanalabs/common/irq.c b/drivers/misc/habanalabs/common/irq.c
index 9fd4c18e274eecbb5cf91d64a7f224bf1d354835..64e0d9de21bd94eb61a2b8e418a2d4aba8ed5096 100644 (file)
--- a/drivers/misc/habanalabs/common/irq.c
+++ b/drivers/misc/habanalabs/common/irq.c
@@ -245,7 +245,7 @@ irqreturn_t hl_irq_handler_eq(int irq, void *arg)
                 */
                dma_rmb();
 
-               if (hdev->disabled) {
+               if (hdev->disabled && !hdev->is_in_soft_reset) {
                        dev_warn(hdev->dev, "Device disabled but received an EQ event\n");
                        goto skip_irq;
                }