Merge tag 'misc-habanalabs-fixes-2020-02-11' of git://people.freedesktop.org/~gabbayo/linux into char-misc-linus

Oded writes:

This tag contains the following fixes:

- Two fixes to the reset process of the ASIC. Without these fixes, the
  reset process might take a long time and produce a kernel panic.
  Alternatively, the ASIC could get stuck.

- Fix to reference counting of a command buffer object. It was kref_put
  one more time than it should have been.

* tag 'misc-habanalabs-fixes-2020-02-11' of git://people.freedesktop.org/~gabbayo/linux:
  habanalabs: patched cb equals user cb in device memset
  habanalabs: do not halt CoreSight during hard reset
  habanalabs: halt the engines before hard-reset

2 files changed, 46 insertions, 3 deletions

diff --git a/drivers/misc/habanalabs/device.c b/drivers/misc/habanalabs/device.c
index b155e9549076..b680b0caa69b 100644
--- a/drivers/misc/habanalabs/device.c
+++ b/drivers/misc/habanalabs/device.c
@@ -598,7 +598,9 @@ int hl_device_set_debug_mode(struct hl_device *hdev, bool enable)
 			goto out;
 		}
 
-		hdev->asic_funcs->halt_coresight(hdev);
+		if (!hdev->hard_reset_pending)
+			hdev->asic_funcs->halt_coresight(hdev);
+
 		hdev->in_debug = 0;
 
 		goto out;
@@ -1189,6 +1191,7 @@ int hl_device_init(struct hl_device *hdev, struct class *hclass)
 	if (hdev->asic_funcs->get_hw_state(hdev) == HL_DEVICE_HW_STATE_DIRTY) {
 		dev_info(hdev->dev,
 			"H/W state is dirty, must reset before initializing\n");
+		hdev->asic_funcs->halt_engines(hdev, true);
 		hdev->asic_funcs->hw_fini(hdev, true);
 	}
 
diff --git a/drivers/misc/habanalabs/goya/goya.c b/drivers/misc/habanalabs/goya/goya.c
index 7344e8a222ae..b8a8de24aaf7 100644
--- a/drivers/misc/habanalabs/goya/goya.c
+++ b/drivers/misc/habanalabs/goya/goya.c
@@ -895,6 +895,11 @@ void goya_init_dma_qmans(struct hl_device *hdev)
  */
 static void goya_disable_external_queues(struct hl_device *hdev)
 {
+	struct goya_device *goya = hdev->asic_specific;
+
+	if (!(goya->hw_cap_initialized & HW_CAP_DMA))
+		return;
+
 	WREG32(mmDMA_QM_0_GLBL_CFG0, 0);
 	WREG32(mmDMA_QM_1_GLBL_CFG0, 0);
 	WREG32(mmDMA_QM_2_GLBL_CFG0, 0);
@@ -956,6 +961,11 @@ static int goya_stop_external_queues(struct hl_device *hdev)
 {
 	int rc, retval = 0;
 
+	struct goya_device *goya = hdev->asic_specific;
+
+	if (!(goya->hw_cap_initialized & HW_CAP_DMA))
+		return retval;
+
 	rc = goya_stop_queue(hdev,
 			mmDMA_QM_0_GLBL_CFG1,
 			mmDMA_QM_0_CP_STS,
@@ -1744,9 +1754,18 @@ void goya_init_tpc_qmans(struct hl_device *hdev)
  */
 static void goya_disable_internal_queues(struct hl_device *hdev)
 {
+	struct goya_device *goya = hdev->asic_specific;
+
+	if (!(goya->hw_cap_initialized & HW_CAP_MME))
+		goto disable_tpc;
+
 	WREG32(mmMME_QM_GLBL_CFG0, 0);
 	WREG32(mmMME_CMDQ_GLBL_CFG0, 0);
 
+disable_tpc:
+	if (!(goya->hw_cap_initialized & HW_CAP_TPC))
+		return;
+
 	WREG32(mmTPC0_QM_GLBL_CFG0, 0);
 	WREG32(mmTPC0_CMDQ_GLBL_CFG0, 0);
 
@@ -1782,8 +1801,12 @@ static void goya_disable_internal_queues(struct hl_device *hdev)
  */
 static int goya_stop_internal_queues(struct hl_device *hdev)
 {
+	struct goya_device *goya = hdev->asic_specific;
 	int rc, retval = 0;
 
+	if (!(goya->hw_cap_initialized & HW_CAP_MME))
+		goto stop_tpc;
+
 	/*
 	 * Each queue (QMAN) is a separate H/W logic. That means that each
 	 * QMAN can be stopped independently and failure to stop one does NOT
@@ -1810,6 +1833,10 @@ static int goya_stop_internal_queues(struct hl_device *hdev)
 		retval = -EIO;
 	}
 
+stop_tpc:
+	if (!(goya->hw_cap_initialized & HW_CAP_TPC))
+		return retval;
+
 	rc = goya_stop_queue(hdev,
 			mmTPC0_QM_GLBL_CFG1,
 			mmTPC0_QM_CP_STS,
@@ -1975,6 +2002,11 @@ static int goya_stop_internal_queues(struct hl_device *hdev)
 
 static void goya_dma_stall(struct hl_device *hdev)
 {
+	struct goya_device *goya = hdev->asic_specific;
+
+	if (!(goya->hw_cap_initialized & HW_CAP_DMA))
+		return;
+
 	WREG32(mmDMA_QM_0_GLBL_CFG1, 1 << DMA_QM_0_GLBL_CFG1_DMA_STOP_SHIFT);
 	WREG32(mmDMA_QM_1_GLBL_CFG1, 1 << DMA_QM_1_GLBL_CFG1_DMA_STOP_SHIFT);
 	WREG32(mmDMA_QM_2_GLBL_CFG1, 1 << DMA_QM_2_GLBL_CFG1_DMA_STOP_SHIFT);
@@ -1984,6 +2016,11 @@ static void goya_dma_stall(struct hl_device *hdev)
 
 static void goya_tpc_stall(struct hl_device *hdev)
 {
+	struct goya_device *goya = hdev->asic_specific;
+
+	if (!(goya->hw_cap_initialized & HW_CAP_TPC))
+		return;
+
 	WREG32(mmTPC0_CFG_TPC_STALL, 1 << TPC0_CFG_TPC_STALL_V_SHIFT);
 	WREG32(mmTPC1_CFG_TPC_STALL, 1 << TPC1_CFG_TPC_STALL_V_SHIFT);
 	WREG32(mmTPC2_CFG_TPC_STALL, 1 << TPC2_CFG_TPC_STALL_V_SHIFT);
@@ -1996,6 +2033,11 @@ static void goya_tpc_stall(struct hl_device *hdev)
 
 static void goya_mme_stall(struct hl_device *hdev)
 {
+	struct goya_device *goya = hdev->asic_specific;
+
+	if (!(goya->hw_cap_initialized & HW_CAP_MME))
+		return;
+
 	WREG32(mmMME_STALL, 0xFFFFFFFF);
 }
 
@@ -4648,8 +4690,6 @@ static int goya_memset_device_memory(struct hl_device *hdev, u64 addr, u64 size,
 
 	rc = goya_send_job_on_qman0(hdev, job);
 
-	hl_cb_put(job->patched_cb);
-
 	hl_debugfs_remove_job(hdev, job);
 	kfree(job);
 	cb->cs_cnt--;