Skip to contentSkip to conversation

bnxt_en: Bug fixes.

PATCH In review Michael Chan 2 participants 7 messages Jun 29, 2019 – Jun 30, 2019

Subsystems (3)
lore
Share & tools
lore

Michael Chan

author maintainer
#0 lore raw

cover bnxt_en: Bug fixes.

Miscellaneous bug fix patches, including two resource handling fixes for the RDMA driver, a PCI shutdown patch to add pci_disable_device(), a patch to fix ethtool selftest crash, and the last one suppresses an unnecessry error message.

Please also queue patches 1, 2, and 3 for -stable. Thanks.

Michael Chan (5):
bnxt_en: Disable bus master during PCI shutdown and driver unload.
bnxt_en: Fix ethtool selftest crash under error conditions.
bnxt_en: Fix statistics context reservation logic for RDMA driver.
bnxt_en: Cap the returned MSIX vectors to the RDMA driver.
bnxt_en: Suppress error messages when querying DSCP DCB capabilities.

 drivers/net/ethernet/broadcom/bnxt/bnxt.c         | 20 +++++++++++++-------
 drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c     |  2 +-
 drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c |  6 +++---
 drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c     |  4 +++-
 4 files changed, 20 insertions(+), 12 deletions(-)
Signature · 1 line
2.5.1
author maintainer
#1 lore raw

PATCH 1/5 bnxt_en: Disable bus master during PCI shutdown and driver unload.

Some chips with older firmware can continue to perform DMA read from context memory even after the memory has been freed. In the PCI shutdown method, we need to call pci_disable_device() to shutdown DMA to prevent this DMA before we put the device into D3hot. DMA memory request in D3hot state will generate PCI fatal error. Similarly, in the driver remove method, the context memory should only be freed after DMA has been shutdown for correctness.

Fixes 98f04cf0f1fc ("bnxt_en: Check context memory requirements from firmware.")
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/ethernet/broadcom/bnxt/bnxt.c +2 -1
index f758b2e..b9bc829 100644
@@ -10262,10 +10262,10 @@ static void bnxt_remove_one(struct pci_dev *pdev)
 	bnxt_dcb_free(bp);
 	kfree(bp->edev);
 	bp->edev = NULL;
+	bnxt_cleanup_pci(bp);
 	bnxt_free_ctx_mem(bp);
 	kfree(bp->ctx);
 	bp->ctx = NULL;
-	bnxt_cleanup_pci(bp);
 	bnxt_free_port_stats(bp);
 	free_netdev(dev);
 }
@@ -10859,6 +10859,7 @@ static void bnxt_shutdown(struct pci_dev *pdev)
 
 	if (system_state == SYSTEM_POWER_OFF) {
 		bnxt_clear_int_mode(bp);
+		pci_disable_device(pdev);
 		pci_wake_from_d3(pdev, bp->wol);
 		pci_set_power_state(pdev, PCI_D3hot);
 	}
Signature · 1 line
2.5.1
author maintainer
#2 lore raw

PATCH 2/5 bnxt_en: Fix ethtool selftest crash under error conditions.

In reply to Michael Chan

After ethtool loopback packet tests, we re-open the nic for the next IRQ test. If the open fails, we must not proceed with the IRQ test or we will crash with NULL pointer dereference. Fix it by checking the bnxt_open_nic() return code before proceeding.

Fixes 67fea463fd87 ("bnxt_en: Add interrupt test to ethtool -t selftest.")
---
 drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +3 -3
index a6c7baf..ec68707 100644
@@ -2842,7 +2842,7 @@ static void bnxt_self_test(struct net_device *dev, struct ethtool_test *etest,
 	bool offline = false;
 	u8 test_results = 0;
 	u8 test_mask = 0;
-	int rc, i;
+	int rc = 0, i;
 
 	if (!bp->num_tests || !BNXT_SINGLE_PF(bp))
 		return;
@@ -2913,9 +2913,9 @@ static void bnxt_self_test(struct net_device *dev, struct ethtool_test *etest,
 		}
 		bnxt_hwrm_phy_loopback(bp, false, false);
 		bnxt_half_close_nic(bp);
-		bnxt_open_nic(bp, false, true);
+		rc = bnxt_open_nic(bp, false, true);
 	}
-	if (bnxt_test_irq(bp)) {
+	if (rc || bnxt_test_irq(bp)) {
 		buf[BNXT_IRQ_TEST_IDX] = 1;
 		etest->flags |= ETH_TEST_FL_FAILED;
 	}
Signature · 1 line
2.5.1
author maintainer
#3 lore raw

PATCH 3/5 bnxt_en: Fix statistics context reservation logic for RDMA driver.

In reply to Michael Chan

The current logic assumes that the RDMA driver uses one statistics context adjacent to the ones used by the network driver. This assumption is not true and the statistics context used by the RDMA driver is tied to its MSIX base vector. This wrong assumption can cause RDMA driver failure after changing ethtool rings on the network side. Fix the statistics reservation logic accordingly.

Fixes 780baad44f0f ("bnxt_en: Reserve 1 stat_ctx for RDMA driver.")
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)
drivers/net/ethernet/broadcom/bnxt/bnxt.c +11 -6
index b9bc829..9090c79 100644
@@ -5508,7 +5508,16 @@ static int bnxt_cp_rings_in_use(struct bnxt *bp)
 
 static int bnxt_get_func_stat_ctxs(struct bnxt *bp)
 {
-	return bp->cp_nr_rings + bnxt_get_ulp_stat_ctxs(bp);
+	int ulp_stat = bnxt_get_ulp_stat_ctxs(bp);
+	int cp = bp->cp_nr_rings;
+
+	if (!ulp_stat)
+		return cp;
+
+	if (bnxt_nq_rings_in_use(bp) > cp + bnxt_get_ulp_msix_num(bp))
+		return bnxt_get_ulp_msix_base(bp) + ulp_stat;
+
+	return cp + ulp_stat;
 }
 
 static bool bnxt_need_reserve_rings(struct bnxt *bp)
@@ -7477,11 +7486,7 @@ unsigned int bnxt_get_avail_cp_rings_for_en(struct bnxt *bp)
 
 unsigned int bnxt_get_avail_stat_ctxs_for_en(struct bnxt *bp)
 {
-	unsigned int stat;
-
-	stat = bnxt_get_max_func_stat_ctxs(bp) - bnxt_get_ulp_stat_ctxs(bp);
-	stat -= bp->cp_nr_rings;
-	return stat;
+	return bnxt_get_max_func_stat_ctxs(bp) - bnxt_get_func_stat_ctxs(bp);
 }
 
 int bnxt_get_avail_msix(struct bnxt *bp, int num)
Signature · 1 line
2.5.1
author maintainer
#4 lore raw

PATCH 4/5 bnxt_en: Cap the returned MSIX vectors to the RDMA driver.

In reply to Michael Chan

In an earlier commit to improve NQ reservations on 57500 chips, we set the resv_irqs on the 57500 VFs to the fixed value assigned by the PF regardless of how many are actually used. The current code assumes that resv_irqs minus the ones used by the network driver must be the ones for the RDMA driver. This is no longer true and we may return more MSIX vectors than requested, causing inconsistency. Fix it by capping the value.

Fixes 01989c6b69d9 ("bnxt_en: Improve NQ reservations.")
---
 drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c +3 -1
index bfa342a..fc77caf 100644
@@ -157,8 +157,10 @@ static int bnxt_req_msix_vecs(struct bnxt_en_dev *edev, int ulp_id,
 
 	if (BNXT_NEW_RM(bp)) {
 		struct bnxt_hw_resc *hw_resc = &bp->hw_resc;
+		int resv_msix;
 
-		avail_msix = hw_resc->resv_irqs - bp->cp_nr_rings;
+		resv_msix = hw_resc->resv_irqs - bp->cp_nr_rings;
+		avail_msix = min_t(int, resv_msix, avail_msix);
 		edev->ulp_tbl[ulp_id].msix_requested = avail_msix;
 	}
 	bnxt_fill_msix_vecs(bp, ent);
Signature · 1 line
2.5.1
author maintainer
#5 lore raw

PATCH 5/5 bnxt_en: Suppress error messages when querying DSCP DCB capabilities.

In reply to Michael Chan

Some firmware versions do not support this so use the silent variant to send the message to firmware to suppress the harmless error. This error message is unnecessarily alarming the user.

Fixes afdc8a84844a ("bnxt_en: Add DCBNL DSCP application protocol support.")
---
 drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c +1 -1
index 7077515..07301cb 100644
@@ -396,7 +396,7 @@ static int bnxt_hwrm_queue_dscp_qcaps(struct bnxt *bp)
 
 	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_QUEUE_DSCP_QCAPS, -1, -1);
 	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = _hwrm_send_message_silent(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
 	if (!rc) {
 		bp->max_dscp_value = (1 << resp->num_dscp_bits) - 1;
 		if (bp->max_dscp_value < 0x3f)
Signature · 1 line
2.5.1
maintainer
#6 lore raw
In reply to Michael Chan

From: Michael Chan <[email protected]>
Date: Sat, 29 Jun 2019 11:16:43 -0400

Quoted 4 lines

Miscellaneous bug fix patches, including two resource handling fixes for the RDMA driver, a PCI shutdown patch to add pci_disable_device(), a patch to fix ethtool selftest crash, and the last one suppresses an unnecessry error message.

Series applied.

Quoted 1 line

Please also queue patches 1, 2, and 3 for -stable. Thanks.

Queued up.