public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* [PATCH 1/7] habanalabs/gaudi2: print RAZWI info upon PCIe access error
@ 2022-09-18 11:37 Oded Gabbay
  2022-09-18 11:37 ` [PATCH 2/7] habanalabs/gaudi2: increase hard-reset sleep time to 2 sec Oded Gabbay
                   ` (5 more replies)
  0 siblings, 6 replies; 7+ messages in thread
From: Oded Gabbay @ 2022-09-18 11:37 UTC (permalink / raw)
  To: linux-kernel; +Cc: Tomer Tayar

From: Tomer Tayar <ttayar@habana.ai>

Add the dump of the RAZWI information when a PCIe access is blocked by
RR.

Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/gaudi2/gaudi2.c       | 52 +++++++++++++++++--
 drivers/misc/habanalabs/gaudi2/gaudi2_masks.h | 13 +++++
 2 files changed, 61 insertions(+), 4 deletions(-)

diff --git a/drivers/misc/habanalabs/gaudi2/gaudi2.c b/drivers/misc/habanalabs/gaudi2/gaudi2.c
index 5761ca5d50ae..c040e01adafe 100644
--- a/drivers/misc/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/misc/habanalabs/gaudi2/gaudi2.c
@@ -7963,14 +7963,58 @@ static void gaudi2_handle_dma_core_event(struct hl_device *hdev, u64 intr_cause_
 						gaudi2_dma_core_interrupts_cause[i]);
 }
 
+static void gaudi2_print_pcie_mstr_rr_mstr_if_razwi_info(struct hl_device *hdev)
+{
+	u32 mstr_if_base_addr = mmPCIE_MSTR_RR_MSTR_IF_RR_SHRD_HBW_BASE, razwi_happened_addr;
+
+	razwi_happened_addr = mstr_if_base_addr + RR_SHRD_HBW_AW_RAZWI_HAPPENED;
+	if (RREG32(razwi_happened_addr)) {
+		gaudi2_razwi_rr_hbw_shared_printf_info(hdev, mstr_if_base_addr, true, "PCIE", true,
+							NULL);
+		WREG32(razwi_happened_addr, 0x1);
+	}
+
+	razwi_happened_addr = mstr_if_base_addr + RR_SHRD_HBW_AR_RAZWI_HAPPENED;
+	if (RREG32(razwi_happened_addr)) {
+		gaudi2_razwi_rr_hbw_shared_printf_info(hdev, mstr_if_base_addr, false, "PCIE", true,
+							NULL);
+		WREG32(razwi_happened_addr, 0x1);
+	}
+
+	razwi_happened_addr = mstr_if_base_addr + RR_SHRD_LBW_AW_RAZWI_HAPPENED;
+	if (RREG32(razwi_happened_addr)) {
+		gaudi2_razwi_rr_lbw_shared_printf_info(hdev, mstr_if_base_addr, true, "PCIE", true,
+							NULL);
+		WREG32(razwi_happened_addr, 0x1);
+	}
+
+	razwi_happened_addr = mstr_if_base_addr + RR_SHRD_LBW_AR_RAZWI_HAPPENED;
+	if (RREG32(razwi_happened_addr)) {
+		gaudi2_razwi_rr_lbw_shared_printf_info(hdev, mstr_if_base_addr, false, "PCIE", true,
+							NULL);
+		WREG32(razwi_happened_addr, 0x1);
+	}
+}
+
 static void gaudi2_print_pcie_addr_dec_info(struct hl_device *hdev, u64 intr_cause_data)
 {
 	int i;
 
-	for (i = 0 ; i < GAUDI2_NUM_OF_PCIE_ADDR_DEC_ERR_CAUSE; i++)
-		if (intr_cause_data & BIT_ULL(i))
-			dev_err_ratelimited(hdev->dev, "PCIE ADDR DEC Error: %s\n",
-						gaudi2_pcie_addr_dec_error_cause[i]);
+	for (i = 0 ; i < GAUDI2_NUM_OF_PCIE_ADDR_DEC_ERR_CAUSE ; i++) {
+		if (!(intr_cause_data & BIT_ULL(i)))
+			continue;
+
+		dev_err_ratelimited(hdev->dev, "PCIE ADDR DEC Error: %s\n",
+					gaudi2_pcie_addr_dec_error_cause[i]);
+
+		switch (intr_cause_data & BIT_ULL(i)) {
+		case PCIE_WRAP_PCIE_IC_SEI_INTR_IND_AXI_LBW_ERR_INTR_MASK:
+			break;
+		case PCIE_WRAP_PCIE_IC_SEI_INTR_IND_BAD_ACCESS_INTR_MASK:
+			gaudi2_print_pcie_mstr_rr_mstr_if_razwi_info(hdev);
+			break;
+		}
+	}
 }
 
 static void gaudi2_handle_pif_fatal(struct hl_device *hdev, u64 intr_cause_data)
diff --git a/drivers/misc/habanalabs/gaudi2/gaudi2_masks.h b/drivers/misc/habanalabs/gaudi2/gaudi2_masks.h
index 0239d118abc5..e9ac87828221 100644
--- a/drivers/misc/habanalabs/gaudi2/gaudi2_masks.h
+++ b/drivers/misc/habanalabs/gaudi2/gaudi2_masks.h
@@ -144,4 +144,17 @@
 #define DCORE0_SYNC_MNGR_OBJS_SOB_OBJ_SIGN_SHIFT	15
 #define DCORE0_SYNC_MNGR_OBJS_SOB_OBJ_SIGN_MASK		0x8000
 
+#define PCIE_WRAP_PCIE_IC_SEI_INTR_IND_AXI_ERR_INTR_SHIFT		0
+#define PCIE_WRAP_PCIE_IC_SEI_INTR_IND_AXI_ERR_INTR_MASK		0x1
+#define PCIE_WRAP_PCIE_IC_SEI_INTR_IND_AXI_LBW_ERR_INTR_SHIFT		1
+#define PCIE_WRAP_PCIE_IC_SEI_INTR_IND_AXI_LBW_ERR_INTR_MASK		0x2
+#define PCIE_WRAP_PCIE_IC_SEI_INTR_IND_BAD_ACCESS_INTR_SHIFT		2
+#define PCIE_WRAP_PCIE_IC_SEI_INTR_IND_BAD_ACCESS_INTR_MASK		0x4
+#define PCIE_WRAP_PCIE_IC_SEI_INTR_IND_AXI_ERR_INTR_MASK_SHIFT		3
+#define PCIE_WRAP_PCIE_IC_SEI_INTR_IND_AXI_ERR_INTR_MASK_MASK		0x8
+#define PCIE_WRAP_PCIE_IC_SEI_INTR_IND_AXI_LBW_ERR_INTR_MASK_SHIFT	4
+#define PCIE_WRAP_PCIE_IC_SEI_INTR_IND_AXI_LBW_ERR_INTR_MASK_MASK	0x10
+#define PCIE_WRAP_PCIE_IC_SEI_INTR_IND_BAD_ACCESS_INTR_MASK_SHIFT	5
+#define PCIE_WRAP_PCIE_IC_SEI_INTR_IND_BAD_ACCESS_INTR_MASK_MASK	0x20
+
 #endif /* GAUDI2_MASKS_H_ */
-- 
2.25.1


^ permalink raw reply related	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2022-09-18 11:38 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2022-09-18 11:37 [PATCH 1/7] habanalabs/gaudi2: print RAZWI info upon PCIe access error Oded Gabbay
2022-09-18 11:37 ` [PATCH 2/7] habanalabs/gaudi2: increase hard-reset sleep time to 2 sec Oded Gabbay
2022-09-18 11:37 ` [PATCH 3/7] habanalabs/gaudi2: get f/w reset status register dynamically Oded Gabbay
2022-09-18 11:37 ` [PATCH 4/7] habanalabs: rename error info structure Oded Gabbay
2022-09-18 11:37 ` [PATCH 5/7] habanalabs/gaudi: change TPC Assert to use TPC DEC instead of QMAN err Oded Gabbay
2022-09-18 11:37 ` [PATCH 6/7] habanalabs/gaudi2: add handling to pmmu events in eqe handler Oded Gabbay
2022-09-18 11:37 ` [PATCH 7/7] habanalabs/gaudi2: add secured attestation info uapi Oded Gabbay

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox