AI Trajectory


Trajectory: (Expand/Collapse All)
Seq Timestamp Type Name Duration
0/0 2026/09/29 05:51 flow patch-triage
0m Args:
null

Results:
{
  "EnableConfigs": null,
  "FocusSymbols": [],
  "KMSANReasoning": "",
  "NeedsKMSAN": false,
  "Reasoning": "The patch exclusively modifies the Broadcom NetXtreme-C/E (bnxt) driver (drivers/net/ethernet/broadcom/bnxt/). This is a vendor-specific PCIe NIC driver requiring physical hardware not emulated in standard virtualized test environments (e.g., QEMU or GCE), making the modified code structurally unreachable for fuzzing.",
  "WorthFuzzing": false
}

1/1 2026/09/29 05:51 action read-patch-diff
0m Args:
null

Results:
{
  "PatchDiff": "commit da0eaeb5346d003e1f19ba5c7171208f952592e9\nAuthor: syz-cluster \u003ctriage@syzkaller.com\u003e\nDate:   Tue Sep 29 05:51:52 2026 +0000\n\n    syz-cluster: applied patch under review\n\ndiff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c\nindex d7728d0c5b6e6..810219d9cae21 100644\n--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c\n+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c\n@@ -8163,17 +8163,53 @@ static void bnxt_copy_reserved_rings(struct bnxt *bp, struct bnxt_hw_rings *hwr)\n \t}\n }\n \n+static bool bnxt_nitro_a0_rings_ok(struct bnxt *bp, int rx)\n+{\n+\tif (BNXT_CHIP_TYPE_NITRO_A0(bp) \u0026\u0026 rx \u003c 2)\n+\t\treturn false;\n+\treturn true;\n+}\n+\n static bool bnxt_rings_ok(struct bnxt *bp, struct bnxt_hw_rings *hwr)\n {\n-\treturn hwr-\u003etx \u0026\u0026 hwr-\u003erx \u0026\u0026 hwr-\u003ecp \u0026\u0026 hwr-\u003egrp \u0026\u0026 hwr-\u003evnic \u0026\u0026\n-\t       hwr-\u003estat \u0026\u0026 (hwr-\u003ecp_p5 || !(bp-\u003eflags \u0026 BNXT_FLAG_CHIP_P5_PLUS));\n+\tint min_tx = bp-\u003enum_tc ? bp-\u003enum_tc : 1;\n+\tint min_rx = 1;\n+\tint min_cp = 1;\n+\tint tx_cp;\n+\n+\tif (bp-\u003etx_nr_rings_xdp)\n+\t\tmin_tx++;\n+\n+\tif (!(bp-\u003eflags \u0026 BNXT_FLAG_SHARED_RINGS)) {\n+\t\ttx_cp = bnxt_num_tx_to_cp(bp, min_tx);\n+\t\tmin_cp = tx_cp + min_rx;\n+\t}\n+\treturn hwr-\u003etx \u003e= min_tx \u0026\u0026 hwr-\u003erx \u003e= min_rx \u0026\u0026 hwr-\u003ecp \u003e= min_cp \u0026\u0026\n+\t       hwr-\u003egrp \u0026\u0026 hwr-\u003evnic \u0026\u0026 hwr-\u003estat \u0026\u0026\n+\t       (hwr-\u003ecp_p5 || !(bp-\u003eflags \u0026 BNXT_FLAG_CHIP_P5_PLUS));\n+}\n+\n+/* Check to see if we need to reset the user configured RSS table\n+ * (default RSS context) because of change in RX rings from\n+ * old_rx to new_rx.\n+ */\n+static void bnxt_check_rss_tbl_lost(struct bnxt *bp, int old_rx, int new_rx)\n+{\n+\tif (netif_is_rxfh_configured(bp-\u003edev) \u0026\u0026\n+\t    (bnxt_get_nr_rss_ctxs(bp, old_rx) !=\n+\t     bnxt_get_nr_rss_ctxs(bp, new_rx) ||\n+\t     bnxt_get_max_rss_ring(bp) \u003e= new_rx)) {\n+\t\tethtool_rxfh_indir_lost(bp-\u003edev);\n+\t}\n }\n \n static int bnxt_get_avail_msix(struct bnxt *bp, int num);\n+static void bnxt_adj_rings(struct bnxt *bp);\n \n static int __bnxt_reserve_rings(struct bnxt *bp)\n {\n \tstruct bnxt_en_dev *edev = bp-\u003eedev[BNXT_AUXDEV_RDMA];\n+\tint req_rx_rings = bp-\u003erx_nr_rings;\n \tstruct bnxt_hw_rings hwr = {0};\n \tint rx_rings, old_rx_rings, rc;\n \tint cp = bp-\u003ecp_nr_rings;\n@@ -8241,54 +8277,68 @@ static int __bnxt_reserve_rings(struct bnxt *bp)\n \t\thwr.stat -= bnxt_get_ulp_stat_ctxs(bp);\n \thwr.cp = min_t(int, hwr.cp, hwr.stat);\n \trc = bnxt_trim_rings(bp, \u0026rx_rings, \u0026hwr.tx, hwr.cp, sh);\n+\tif (rc)\n+\t\tgoto reserve_rings_exit;\n+\n+\tif (bp-\u003ebnapi \u0026\u0026 (rx_rings \u003c bp-\u003erx_nr_rings ||\n+\t\t\t  hwr.tx \u003c bp-\u003etx_nr_rings)) {\n+\t\tnetdev_err(bp-\u003edev, \"Unexpected ring shrinkage %d/%d RX/TX to %d/%d\\n\",\n+\t\t\t   bp-\u003erx_nr_rings, bp-\u003etx_nr_rings, rx_rings, hwr.tx);\n+\t\trc = -ENOSPC;\n+\t\tgoto reserve_rings_exit;\n+\t}\n \tif (bp-\u003eflags \u0026 BNXT_FLAG_AGG_RINGS)\n \t\thwr.rx = rx_rings \u003c\u003c 1;\n \ttx_cp = bnxt_num_tx_to_cp(bp, hwr.tx);\n \thwr.cp = sh ? max_t(int, tx_cp, rx_rings) : tx_cp + rx_rings;\n+\n+\tif (!bnxt_rings_ok(bp, \u0026hwr)) {\n+\t\trc = -ENOMEM;\n+\t\tgoto reserve_rings_exit;\n+\t}\n+\n \tif (hwr.tx != bp-\u003etx_nr_rings) {\n \t\tnetdev_warn(bp-\u003edev,\n \t\t\t    \"Able to reserve only %d out of %d requested TX rings\\n\",\n \t\t\t    hwr.tx, bp-\u003etx_nr_rings);\n \t}\n \tbp-\u003etx_nr_rings = hwr.tx;\n+\tbp-\u003erx_nr_rings = rx_rings;\n+\tbp-\u003ecp_nr_rings = hwr.cp;\n+\n+\tif (!bp-\u003ebnapi)\n+\t\tbnxt_adj_rings(bp);\n \n \t/* If we cannot reserve all the RX rings, reset the RSS map only\n \t * if absolutely necessary\n \t */\n-\tif (rx_rings != bp-\u003erx_nr_rings) {\n-\t\tnetdev_warn(bp-\u003edev, \"Able to reserve only %d out of %d requested RX rings\\n\",\n-\t\t\t    rx_rings, bp-\u003erx_nr_rings);\n-\t\tif (netif_is_rxfh_configured(bp-\u003edev) \u0026\u0026\n-\t\t    (bnxt_get_nr_rss_ctxs(bp, bp-\u003erx_nr_rings) !=\n-\t\t     bnxt_get_nr_rss_ctxs(bp, rx_rings) ||\n-\t\t     bnxt_get_max_rss_ring(bp) \u003e= rx_rings)) {\n-\t\t\tethtool_rxfh_indir_lost(bp-\u003edev);\n-\t\t}\n+\tif (req_rx_rings != bp-\u003erx_nr_rings) {\n+\t\tnetdev_warn(bp-\u003edev, \"RX rings reduced from %d to %d\\n\",\n+\t\t\t    req_rx_rings, bp-\u003erx_nr_rings);\n+\t\tbnxt_check_rss_tbl_lost(bp, req_rx_rings, bp-\u003erx_nr_rings);\n \t}\n-\tbp-\u003erx_nr_rings = rx_rings;\n-\tbp-\u003ecp_nr_rings = hwr.cp;\n \n \t/* Fall back if we cannot reserve enough HW RSS contexts */\n \tif ((bp-\u003erss_cap \u0026 BNXT_RSS_CAP_LARGE_RSS_CTX) \u0026\u0026\n \t    hwr.rss_ctx \u003c bnxt_get_total_rss_ctxs(bp, \u0026hwr))\n \t\tbp-\u003erss_cap \u0026= ~BNXT_RSS_CAP_LARGE_RSS_CTX;\n \n-\tif (!bnxt_rings_ok(bp, \u0026hwr))\n-\t\treturn -ENOMEM;\n-\n-\tif (old_rx_rings != bp-\u003ehw_resc.resv_rx_rings \u0026\u0026\n+\tif ((old_rx_rings != bp-\u003ehw_resc.resv_rx_rings ||\n+\t     req_rx_rings != bp-\u003erx_nr_rings) \u0026\u0026\n \t    !netif_is_rxfh_configured(bp-\u003edev))\n \t\tbnxt_set_dflt_rss_indir_tbl(bp, NULL);\n \n+reserve_rings_exit:\n \tif (!bnxt_ulp_registered(edev) \u0026\u0026 BNXT_NEW_RM(bp)) {\n \t\tint resv_msix, resv_ctx, ulp_ctxs;\n \t\tstruct bnxt_hw_resc *hw_resc;\n \n \t\thw_resc = \u0026bp-\u003ehw_resc;\n-\t\tresv_msix = hw_resc-\u003eresv_irqs - bp-\u003ecp_nr_rings;\n+\t\tresv_msix = max_t(int, hw_resc-\u003eresv_irqs - bp-\u003ecp_nr_rings, 0);\n \t\tulp_msix = min_t(int, resv_msix, ulp_msix);\n \t\tbnxt_set_ulp_msix_num(bp, ulp_msix);\n-\t\tresv_ctx = hw_resc-\u003eresv_stat_ctxs  - bp-\u003ecp_nr_rings;\n+\t\tresv_ctx = max_t(int, hw_resc-\u003eresv_stat_ctxs - bp-\u003ecp_nr_rings,\n+\t\t\t\t 0);\n \t\tulp_ctxs = min(resv_ctx, bnxt_get_ulp_stat_ctxs(bp));\n \t\tbnxt_set_ulp_stat_ctxs(bp, ulp_ctxs);\n \t}\n@@ -11511,6 +11561,12 @@ static int bnxt_change_msix(struct bnxt *bp, int total)\n \tstruct msi_map map;\n \tint i;\n \n+\tif (!bp-\u003eirq_tbl)\n+\t\treturn 0;\n+\n+\tif (total \u003e bp-\u003eirq_tbl_size)\n+\t\treturn bp-\u003etotal_irqs;\n+\n \t/* add MSIX to the end if needed */\n \tfor (i = bp-\u003etotal_irqs; i \u003c total; i++) {\n \t\tmap = pci_msix_alloc_irq_at(bp-\u003epdev, i, NULL);\n@@ -11630,7 +11686,9 @@ static int bnxt_get_num_msix(struct bnxt *bp)\n \n static int bnxt_init_int_mode(struct bnxt *bp)\n {\n-\tint i, total_vecs, max, rc, min = 1, ulp_msix, tx_cp, tbl_size;\n+\tint i, total_vecs, max, rc, min = 1, ulp_msix, tbl_size, req_rx_rings;\n+\tint min_tx, req_tx_rings, req_cp_rings, req_tx_per_tc, req_tx_xdp;\n+\tint tcs = bp-\u003enum_tc ? bp-\u003enum_tc : 1;\n \n \ttotal_vecs = bnxt_get_num_msix(bp);\n \tmax = bnxt_get_max_func_irqs(bp);\n@@ -11653,34 +11711,74 @@ static int bnxt_init_int_mode(struct bnxt *bp)\n \n \ttbl_size = total_vecs;\n \tif (pci_msix_can_alloc_dyn(bp-\u003epdev))\n-\t\ttbl_size = max;\n+\t\ttbl_size = bp-\u003ehw_resc.max_irqs;\n \tbp-\u003eirq_tbl = kzalloc_objs(*bp-\u003eirq_tbl, tbl_size);\n \tif (!bp-\u003eirq_tbl) {\n \t\trc = -ENOMEM;\n \t\tgoto msix_setup_exit;\n \t}\n+\tbp-\u003eirq_tbl_size = tbl_size;\n \n \tfor (i = 0; i \u003c total_vecs; i++)\n \t\tbp-\u003eirq_tbl[i].vector = pci_irq_vector(bp-\u003epdev, i);\n \n \tbp-\u003etotal_irqs = total_vecs;\n+\treq_tx_rings = bp-\u003etx_nr_rings;\n+\treq_tx_per_tc = bp-\u003etx_nr_rings_per_tc;\n+\treq_tx_xdp = bp-\u003etx_nr_rings_xdp;\n+\treq_rx_rings = bp-\u003erx_nr_rings;\n+\treq_cp_rings = bp-\u003ecp_nr_rings;\n+\n \t/* Trim rings based upon num of vectors allocated */\n \trc = bnxt_trim_rings(bp, \u0026bp-\u003erx_nr_rings, \u0026bp-\u003etx_nr_rings,\n \t\t\t     total_vecs - ulp_msix, min == 1);\n \tif (rc)\n \t\tgoto msix_setup_exit;\n \n-\ttx_cp = bnxt_num_tx_to_cp(bp, bp-\u003etx_nr_rings);\n-\tbp-\u003ecp_nr_rings = (min == 1) ?\n-\t\t\t  max_t(int, tx_cp, bp-\u003erx_nr_rings) :\n-\t\t\t  tx_cp + bp-\u003erx_nr_rings;\n+\tmin_tx = bp-\u003etx_nr_rings_xdp ? tcs + 1 : tcs;\n+\tif (bp-\u003etx_nr_rings \u003c min_tx) {\n+\t\tnetdev_err(bp-\u003edev, \"Not enough MSI-X to satisfy min. TX rings\\n\");\n+\t\trc = -ENOMEM;\n+\t\tgoto msix_setup_exit_restore;\n+\t}\n+\n+\tbnxt_adj_rings(bp);\n \n+\tif (!bnxt_nitro_a0_rings_ok(bp, bp-\u003erx_nr_rings)) {\n+\t\tnetdev_err(bp-\u003edev, \"Not enough MSI-X to satisfy min. RX rings\\n\");\n+\t\trc = -ENOMEM;\n+\t\tgoto msix_setup_exit_restore;\n+\t}\n+\n+\tif (bp-\u003ebnapi \u0026\u0026 (req_tx_rings != bp-\u003etx_nr_rings ||\n+\t\t\t  req_rx_rings != bp-\u003erx_nr_rings)) {\n+\t\tnetdev_err(bp-\u003edev, \"Cannot shrink rings once NAPI is allocated\\n\");\n+\t\trc = -ENOSPC;\n+\t\tgoto msix_setup_exit_restore;\n+\t}\n+\n+\tif (req_rx_rings != bp-\u003erx_nr_rings) {\n+\t\tnetdev_warn(bp-\u003edev, \"RX rings reduced from %d to %d\\n\",\n+\t\t\t    req_rx_rings, bp-\u003erx_nr_rings);\n+\t\tbnxt_check_rss_tbl_lost(bp, req_rx_rings, bp-\u003erx_nr_rings);\n+\t\tif (!netif_is_rxfh_configured(bp-\u003edev))\n+\t\t\tbnxt_set_dflt_rss_indir_tbl(bp, NULL);\n+\t}\n \treturn 0;\n \n+msix_setup_exit_restore:\n+\tbp-\u003etx_nr_rings = req_tx_rings;\n+\tbp-\u003etx_nr_rings_per_tc = req_tx_per_tc;\n+\tbp-\u003etx_nr_rings_xdp = req_tx_xdp;\n+\tbp-\u003erx_nr_rings = req_rx_rings;\n+\tbp-\u003ecp_nr_rings = req_cp_rings;\n+\n msix_setup_exit:\n \tnetdev_err(bp-\u003edev, \"bnxt_init_int_mode err: %x\\n\", rc);\n \tkfree(bp-\u003eirq_tbl);\n \tbp-\u003eirq_tbl = NULL;\n+\tbp-\u003etotal_irqs = 0;\n+\tbp-\u003eirq_tbl_size = 0;\n \tpci_free_irq_vectors(bp-\u003epdev);\n \treturn rc;\n }\n@@ -11691,33 +11789,36 @@ static void bnxt_clear_int_mode(struct bnxt *bp)\n \n \tkfree(bp-\u003eirq_tbl);\n \tbp-\u003eirq_tbl = NULL;\n+\tbp-\u003eirq_tbl_size = 0;\n }\n \n-int bnxt_reserve_rings(struct bnxt *bp, bool irq_re_init)\n+static int bnxt_irqs_required(struct bnxt *bp)\n {\n \tstruct bnxt_en_dev *edev = bp-\u003eedev[BNXT_AUXDEV_RDMA];\n-\tbool irq_cleared = false;\n-\tbool irq_change = false;\n-\tint tcs = bp-\u003enum_tc;\n-\tint irqs_required;\n-\tint rc;\n-\n-\tif (!bnxt_need_reserve_rings(bp))\n-\t\treturn 0;\n \n \tif (BNXT_NEW_RM(bp) \u0026\u0026 !bnxt_ulp_registered(edev)) {\n \t\tint ulp_msix = bnxt_get_avail_msix(bp, bp-\u003eulp_num_msix_want);\n \n \t\tif (ulp_msix \u003e bp-\u003eulp_num_msix_want)\n \t\t\tulp_msix = bp-\u003eulp_num_msix_want;\n-\t\tirqs_required = ulp_msix + bp-\u003ecp_nr_rings;\n-\t} else {\n-\t\tirqs_required = bnxt_get_num_msix(bp);\n+\t\treturn ulp_msix + bp-\u003ecp_nr_rings;\n \t}\n+\treturn bnxt_get_num_msix(bp);\n+}\n+\n+int bnxt_reserve_rings(struct bnxt *bp, bool irq_re_init)\n+{\n+\tbool irq_cleared = false;\n+\tint irqs_required;\n+\tint rc;\n+\n+\tif (!bnxt_need_reserve_rings(bp))\n+\t\treturn 0;\n+\n+\tirqs_required = bnxt_irqs_required(bp);\n \n \tif (irq_re_init \u0026\u0026 BNXT_NEW_RM(bp) \u0026\u0026 irqs_required != bp-\u003etotal_irqs) {\n-\t\tirq_change = true;\n-\t\tif (!pci_msix_can_alloc_dyn(bp-\u003epdev)) {\n+\t\tif (!pci_msix_can_alloc_dyn(bp-\u003epdev) || !bp-\u003eirq_tbl) {\n \t\t\tbnxt_ulp_irq_stop(bp);\n \t\t\tbnxt_clear_int_mode(bp);\n \t\t\tirq_cleared = true;\n@@ -11728,25 +11829,17 @@ int bnxt_reserve_rings(struct bnxt *bp, bool irq_re_init)\n \t\tif (!rc)\n \t\t\trc = bnxt_init_int_mode(bp);\n \t\tbnxt_ulp_irq_restart(bp, rc);\n-\t} else if (irq_change \u0026\u0026 !rc) {\n-\t\tif (bnxt_change_msix(bp, irqs_required) != irqs_required)\n+\t} else if (irq_re_init \u0026\u0026 BNXT_NEW_RM(bp) \u0026\u0026 !rc) {\n+\t\t/* __bnxt_reserve_rings() may have shrunk the rings */\n+\t\tirqs_required = bnxt_irqs_required(bp);\n+\t\tif (irqs_required != bp-\u003etotal_irqs \u0026\u0026\n+\t\t    bnxt_change_msix(bp, irqs_required) != irqs_required)\n \t\t\trc = -ENOSPC;\n \t}\n \tif (rc) {\n \t\tnetdev_err(bp-\u003edev, \"ring reservation/IRQ init failure rc: %d\\n\", rc);\n \t\treturn rc;\n \t}\n-\tif (tcs \u0026\u0026 (bp-\u003etx_nr_rings_per_tc * tcs !=\n-\t\t    bp-\u003etx_nr_rings - bp-\u003etx_nr_rings_xdp)) {\n-\t\tnetdev_err(bp-\u003edev, \"tx ring reservation failure\\n\");\n-\t\tnetdev_reset_tc(bp-\u003edev);\n-\t\tbp-\u003enum_tc = 0;\n-\t\tif (bp-\u003etx_nr_rings_xdp)\n-\t\t\tbp-\u003etx_nr_rings_per_tc = bp-\u003etx_nr_rings_xdp;\n-\t\telse\n-\t\t\tbp-\u003etx_nr_rings_per_tc = bp-\u003etx_nr_rings;\n-\t\treturn -ENOMEM;\n-\t}\n \treturn 0;\n }\n \n@@ -13251,13 +13344,29 @@ static void bnxt_set_xdp_tx_rings(struct bnxt *bp)\n \tbp-\u003etx_nr_rings += bp-\u003etx_nr_rings_xdp;\n }\n \n-static void bnxt_adj_tx_rings(struct bnxt *bp)\n+static void bnxt_adj_rings(struct bnxt *bp)\n {\n-\t/* Make adjustments if reserved TX rings are less than requested */\n-\tbp-\u003etx_nr_rings -= bp-\u003etx_nr_rings_xdp;\n-\tbp-\u003etx_nr_rings_per_tc = bnxt_tx_nr_rings_per_tc(bp);\n-\tif (bp-\u003etx_nr_rings_xdp)\n-\t\tbnxt_set_xdp_tx_rings(bp);\n+\tint tcs = bp-\u003enum_tc ? bp-\u003enum_tc : 1;\n+\tint rx = bp-\u003erx_nr_rings;\n+\n+\t/* Make adjustments to rings based on TC/XDP/shared rings policies */\n+\tif (bp-\u003etx_nr_rings_xdp) {\n+\t\ttcs++;\n+\t\tbp-\u003etx_nr_rings_per_tc = bp-\u003etx_nr_rings / tcs;\n+\t\tbp-\u003etx_nr_rings_xdp = bp-\u003etx_nr_rings_per_tc;\n+\t} else {\n+\t\tbp-\u003etx_nr_rings_per_tc = bnxt_tx_nr_rings_per_tc(bp);\n+\t}\n+\n+\tif (bp-\u003eflags \u0026 BNXT_FLAG_SHARED_RINGS) {\n+\t\tbp-\u003etx_nr_rings_per_tc = min(bp-\u003etx_nr_rings_per_tc, rx);\n+\t\tbp-\u003erx_nr_rings = bp-\u003etx_nr_rings_per_tc;\n+\t\tif (bp-\u003etx_nr_rings_xdp)\n+\t\t\tbp-\u003etx_nr_rings_xdp = bp-\u003etx_nr_rings_per_tc;\n+\t}\n+\n+\tbp-\u003etx_nr_rings = bp-\u003etx_nr_rings_per_tc * tcs;\n+\tbnxt_set_cp_rings(bp, bp-\u003eflags \u0026 BNXT_FLAG_SHARED_RINGS);\n }\n \n static int __bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init)\n@@ -13277,7 +13386,6 @@ static int __bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init)\n \tif (rc)\n \t\treturn rc;\n \n-\tbnxt_adj_tx_rings(bp);\n \trc = bnxt_alloc_mem(bp, irq_re_init);\n \tif (rc) {\n \t\tnetdev_err(bp-\u003edev, \"bnxt_alloc_mem err: %x\\n\", rc);\n@@ -14081,7 +14189,8 @@ static int bnxt_reinit_features(struct bnxt *bp, bool irq_re_init,\n \t\t\t\tbool link_re_init, u32 flags, bool update_tpa)\n {\n \tbnxt_close_nic(bp, irq_re_init, link_re_init);\n-\tbp-\u003eflags = flags;\n+\tbp-\u003eflags = (bp-\u003eflags \u0026 ~BNXT_FLAG_ALL_CONFIG_FEATS) |\n+\t\t    (flags \u0026 BNXT_FLAG_ALL_CONFIG_FEATS);\n \tif (update_tpa)\n \t\tbnxt_set_ring_params(bp);\n \treturn bnxt_open_nic(bp, irq_re_init, link_re_init);\n@@ -14091,10 +14200,11 @@ static int bnxt_set_features(struct net_device *dev, netdev_features_t features)\n {\n \tbool update_tpa = false, update_ntuple = false;\n \tstruct bnxt *bp = netdev_priv(dev);\n+\tbool irq_re_init = false;\n \tu32 flags = bp-\u003eflags;\n+\tbool re_init = false;\n \tu32 changes;\n \tint rc = 0;\n-\tbool re_init = false;\n \n \tbp-\u003etx_wake_thresh = max_t(int, bp-\u003etx_ring_size / 2,\n \t\t\t\t   bnxt_min_tx_desc_cnt(bp, features));\n@@ -14121,8 +14231,12 @@ static int bnxt_set_features(struct net_device *dev, netdev_features_t features)\n \t\tupdate_tpa = true;\n \t\tif ((bp-\u003eflags \u0026 BNXT_FLAG_TPA) == 0 ||\n \t\t    (flags \u0026 BNXT_FLAG_TPA) == 0 ||\n-\t\t    (bp-\u003eflags \u0026 BNXT_FLAG_CHIP_P5_PLUS))\n+\t\t    (bp-\u003eflags \u0026 BNXT_FLAG_CHIP_P5_PLUS)) {\n \t\t\tre_init = true;\n+\t\t\tif (!(bp-\u003eflags \u0026 BNXT_FLAG_AGG_RINGS) \u0026\u0026\n+\t\t\t    (flags \u0026 BNXT_FLAG_AGG_RINGS))\n+\t\t\t\tirq_re_init = true;\n+\t\t}\n \t}\n \n \tif (changes \u0026 ~BNXT_FLAG_TPA)\n@@ -14145,7 +14259,7 @@ static int bnxt_set_features(struct net_device *dev, netdev_features_t features)\n \t\t\treturn bnxt_reinit_features(bp, true, false, flags, update_tpa);\n \n \t\tif (re_init)\n-\t\t\treturn bnxt_reinit_features(bp, false, false, flags, update_tpa);\n+\t\t\treturn bnxt_reinit_features(bp, irq_re_init, false, flags, update_tpa);\n \n \t\tif (update_tpa) {\n \t\t\tbp-\u003eflags = flags;\n@@ -15081,7 +15195,7 @@ int bnxt_check_rings(struct bnxt *bp, int tx, int rx, bool sh, int tcs,\n \t\t\thwr.cp += bnxt_get_ulp_msix_num(bp);\n \t\t\thwr.cp = min_t(int, hwr.cp, bnxt_get_max_func_irqs(bp));\n \t\t}\n-\t\tif (hwr.cp \u003e bp-\u003etotal_irqs) {\n+\t\tif (bp-\u003eirq_tbl \u0026\u0026 hwr.cp \u003e bp-\u003etotal_irqs) {\n \t\t\tint total_msix = bnxt_change_msix(bp, hwr.cp);\n \n \t\t\tif (total_msix \u003c hwr.cp) {\n@@ -16937,7 +17051,6 @@ static int bnxt_set_dflt_rings(struct bnxt *bp, bool sh)\n \tif (rc \u0026\u0026 rc != -ENODEV)\n \t\tnetdev_warn(bp-\u003edev, \"Unable to reserve tx rings\\n\");\n \n-\tbnxt_adj_tx_rings(bp);\n \tif (sh)\n \t\tbnxt_adj_dflt_rings(bp, true);\n \n@@ -16946,7 +17059,6 @@ static int bnxt_set_dflt_rings(struct bnxt *bp, bool sh)\n \t\trc = __bnxt_reserve_rings(bp);\n \t\tif (rc \u0026\u0026 rc != -ENODEV)\n \t\t\tnetdev_warn(bp-\u003edev, \"2nd rings reservation failed.\\n\");\n-\t\tbnxt_adj_tx_rings(bp);\n \t}\n \tif (BNXT_CHIP_TYPE_NITRO_A0(bp)) {\n \t\tbp-\u003erx_nr_rings++;\n@@ -16980,8 +17092,6 @@ static int bnxt_init_dflt_ring_mode(struct bnxt *bp)\n \tif (rc)\n \t\tgoto init_dflt_ring_err;\n \n-\tbnxt_adj_tx_rings(bp);\n-\n \tbnxt_set_dflt_rfs(bp);\n \n init_dflt_ring_err:\n@@ -17111,6 +17221,50 @@ void bnxt_print_device_info(struct bnxt *bp)\n \tpcie_print_link_status(bp-\u003epdev);\n }\n \n+static void bnxt_clear_bars(struct pci_dev *pdev)\n+{\n+\tint off;\n+\n+\tfor (off = PCI_BASE_ADDRESS_0; off \u003c= PCI_BASE_ADDRESS_5; off += 4)\n+\t\tpci_write_config_dword(pdev, off, 0);\n+}\n+\n+/* Clear any pending DMA transactions from crash kernel while loading driver in\n+ * capture kernel.\n+ */\n+static void bnxt_kdump_reset(struct pci_dev *pdev)\n+{\n+\tu16 cmd;\n+\tint rc;\n+\n+\tpci_read_config_word(pdev, PCI_COMMAND, \u0026cmd);\n+\tcmd \u0026= ~(PCI_COMMAND_MASTER | PCI_COMMAND_MEMORY);\n+\tpci_write_config_word(pdev, PCI_COMMAND, cmd);\n+\n+\tif (pci_save_state(pdev))\n+\t\tdev_warn(\u0026pdev-\u003edev, \"Failed to save PCI state, PCI restore may be incomplete\\n\");\n+\n+\t/* Since it is kdump kernel, try to continue in the best possible way\n+\t * even if the device is unresponsive after FLR.  The device may\n+\t * eventually respond to HWRM_VER_GET later in the init sequence.\n+\t */\n+\trc = pcie_flr(pdev);\n+\tif (rc)\n+\t\tdev_warn(\u0026pdev-\u003edev, \"pcie_flr() failed (rc: %d), trying to continue\\n\",\n+\t\t\t rc);\n+\n+\t/* A complete or partial reset has been done.  Clear the BARs\n+\t * if the device is responsive.\n+\t */\n+\tpci_read_config_word(pdev, PCI_COMMAND, \u0026cmd);\n+\tif (PCI_POSSIBLE_ERROR(cmd))\n+\t\tdev_warn(\u0026pdev-\u003edev, \"PCI config space inaccessible after FLR, not clearing BARs\\n\");\n+\telse\n+\t\tbnxt_clear_bars(pdev);\n+\n+\tpci_restore_state(pdev);\n+}\n+\n static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)\n {\n \tstruct bnxt_hw_resc *hw_resc;\n@@ -17126,15 +17280,11 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)\n \t\treturn -ENODEV;\n \t}\n \n-\t/* Clear any pending DMA transactions from crash kernel\n-\t * while loading driver in capture kernel.\n-\t */\n-\tif (is_kdump_kernel()) {\n-\t\tpci_clear_master(pdev);\n-\t\tpcie_flr(pdev);\n-\t}\n-\n \tmax_irqs = bnxt_get_max_irq(pdev);\n+\n+\tif (is_kdump_kernel())\n+\t\tbnxt_kdump_reset(pdev);\n+\n \tdev = alloc_etherdev_mqs(sizeof(*bp), max_irqs * BNXT_MAX_QUEUE,\n \t\t\t\t max_irqs);\n \tif (!dev)\n@@ -17325,11 +17475,6 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)\n \tif (rc)\n \t\tgoto init_err_pci_clean;\n \n-\t/* No TC has been set yet and rings may have been trimmed due to\n-\t * limited MSIX, so we re-initialize the TX rings per TC.\n-\t */\n-\tbp-\u003etx_nr_rings_per_tc = bp-\u003etx_nr_rings;\n-\n \tif (BNXT_PF(bp)) {\n \t\tif (!bnxt_pf_wq) {\n \t\t\tbnxt_pf_wq =\n@@ -17565,10 +17710,8 @@ static pci_ers_result_t bnxt_io_error_detected(struct pci_dev *pdev,\n \t * so we disable bus master to prevent any potential bad DMAs before\n \t * freeing kernel memory.\n \t */\n-\tif (state == pci_channel_io_frozen) {\n-\t\tset_bit(BNXT_STATE_PCI_CHANNEL_IO_FROZEN, \u0026bp-\u003estate);\n+\tif (state == pci_channel_io_frozen)\n \t\tbnxt_fw_fatal_close(bp);\n-\t}\n \n \tif (netif_running(netdev))\n \t\t__bnxt_close_nic(bp, true, true);\n@@ -17598,69 +17741,80 @@ static pci_ers_result_t bnxt_io_slot_reset(struct pci_dev *pdev)\n \tstruct bnxt *bp = netdev_priv(netdev);\n \tint retry = 0;\n \tint err = 0;\n-\tint off;\n+\tu16 cmd;\n \n \tnetdev_info(bp-\u003edev, \"PCI Slot Reset\\n\");\n \n-\tif (test_bit(BNXT_STATE_PCI_CHANNEL_IO_FROZEN, \u0026bp-\u003estate)) {\n-\t\t/* After DPC, the chip should return CRS when the vendor ID\n-\t\t * config register is read until it is ready.  On all chips,\n-\t\t * this is not happening reliably so add a 5-second delay as a\n-\t\t * workaround.\n-\t\t */\n-\t\tmsleep(5000);\n-\t}\n+\t/* After a PCIe hot reset, the chip should return CRS when the\n+\t * vendor ID config register is read until it is ready.  On all\n+\t * chips, this is not happening reliably so add a 5-second delay\n+\t * as a workaround.\n+\t */\n+\tmsleep(5000);\n \n \tnetdev_lock(netdev);\n \n-\tif (pci_enable_device(pdev)) {\n+\tpci_read_config_word(pdev, PCI_COMMAND, \u0026cmd);\n+\tif (PCI_POSSIBLE_ERROR(cmd)) {\n \t\tdev_err(\u0026pdev-\u003edev,\n-\t\t\t\"Cannot re-enable PCI device after reset.\\n\");\n-\t} else {\n-\t\tpci_set_master(pdev);\n-\t\t/* Upon fatal error, our device internal logic that latches to\n-\t\t * BAR value is getting reset and will restore only upon\n-\t\t * rewriting the BARs.\n-\t\t *\n-\t\t * As pci_restore_state() does not re-write the BARs if the\n-\t\t * value is same as saved value earlier, driver needs to\n-\t\t * write the BARs to 0 to force restore, in case of fatal error.\n-\t\t */\n-\t\tif (test_and_clear_bit(BNXT_STATE_PCI_CHANNEL_IO_FROZEN,\n-\t\t\t\t       \u0026bp-\u003estate)) {\n-\t\t\tfor (off = PCI_BASE_ADDRESS_0;\n-\t\t\t     off \u003c= PCI_BASE_ADDRESS_5; off += 4)\n-\t\t\t\tpci_write_config_dword(bp-\u003epdev, off, 0);\n-\t\t}\n-\t\tpci_restore_state(pdev);\n+\t\t\t\"PCI config space inaccessible after reset\\n\");\n+\t\tgoto reset_exit;\n+\t}\n \n-\t\tbnxt_inv_fw_health_reg(bp);\n-\t\tbnxt_try_map_fw_health_reg(bp);\n+\t/* Upon PCIe error, our device internal logic that latches to\n+\t * BAR value is getting reset and will restore only upon\n+\t * rewriting the BARs.\n+\t *\n+\t * As pci_restore_state() does not re-write the BARs if the\n+\t * value is same as saved value earlier, driver needs to\n+\t * write the BARs to 0 to force restore.\n+\t */\n+\tpci_clear_master(pdev);\n+\tpci_read_config_word(pdev, PCI_COMMAND, \u0026cmd);\n+\tcmd \u0026= ~PCI_COMMAND_MEMORY;\n+\tpci_write_config_word(pdev, PCI_COMMAND, cmd);\n \n-\t\t/* In some PCIe AER scenarios, firmware may take up to\n-\t\t * 10 seconds to become ready in the worst case.\n-\t\t */\n-\t\tdo {\n-\t\t\terr = bnxt_try_recover_fw(bp);\n-\t\t\tif (!err)\n-\t\t\t\tbreak;\n-\t\t\tretry++;\n-\t\t} while (retry \u003c BNXT_FW_SLOT_RESET_RETRY);\n+\tbnxt_clear_bars(pdev);\n+\tpci_restore_state(pdev);\n \n-\t\tif (err) {\n-\t\t\tdev_err(\u0026pdev-\u003edev, \"Firmware not ready\\n\");\n-\t\t\tgoto reset_exit;\n+\tif (!pci_is_enabled(pdev) \u0026\u0026 pci_enable_device(pdev)) {\n+\t\tdev_err(\u0026pdev-\u003edev,\n+\t\t\t\"Cannot re-enable PCI device after reset.\\n\");\n+\t\tpci_read_config_word(pdev, PCI_COMMAND, \u0026cmd);\n+\t\tif (!PCI_POSSIBLE_ERROR(cmd)) {\n+\t\t\tcmd \u0026= ~(PCI_COMMAND_MASTER | PCI_COMMAND_MEMORY);\n+\t\t\tpci_write_config_word(pdev, PCI_COMMAND, cmd);\n \t\t}\n+\t\tgoto reset_exit;\n+\t}\n+\tpci_set_master(pdev);\n \n-\t\terr = bnxt_hwrm_func_reset(bp);\n+\tbnxt_inv_fw_health_reg(bp);\n+\tbnxt_try_map_fw_health_reg(bp);\n+\n+\t/* In some PCIe AER scenarios, firmware may take up to\n+\t * 10 seconds to become ready in the worst case.\n+\t */\n+\tdo {\n+\t\terr = bnxt_try_recover_fw(bp);\n \t\tif (!err)\n-\t\t\tresult = PCI_ERS_RESULT_RECOVERED;\n+\t\t\tbreak;\n+\t\tretry++;\n+\t} while (retry \u003c BNXT_FW_SLOT_RESET_RETRY);\n \n-\t\t/* IRQ will be initialized later in bnxt_io_resume */\n-\t\tbnxt_ulp_irq_stop(bp);\n-\t\tbnxt_clear_int_mode(bp);\n+\tif (err) {\n+\t\tdev_err(\u0026pdev-\u003edev, \"Firmware not ready\\n\");\n+\t\tgoto reset_exit;\n \t}\n \n+\terr = bnxt_hwrm_func_reset(bp);\n+\tif (!err)\n+\t\tresult = PCI_ERS_RESULT_RECOVERED;\n+\n+\t/* IRQ will be initialized later in bnxt_io_resume */\n+\tbnxt_ulp_irq_stop(bp);\n+\tbnxt_clear_int_mode(bp);\n+\n reset_exit:\n \tclear_bit(BNXT_STATE_IN_FW_RESET, \u0026bp-\u003estate);\n \tbnxt_clear_reservations(bp, true);\n@@ -17692,7 +17846,7 @@ static void bnxt_io_resume(struct pci_dev *pdev)\n \t\t\terr = bnxt_open(netdev);\n \t\t} else {\n \t\t\terr = bnxt_reserve_rings(bp, true);\n-\t\t\tif (!err)\n+\t\t\tif (!err \u0026\u0026 !bp-\u003eirq_tbl)\n \t\t\t\terr = bnxt_init_int_mode(bp);\n \t\t}\n \t}\ndiff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h\nindex c673b2ce4a0d2..75428a585577d 100644\n--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h\n+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h\n@@ -2468,7 +2468,6 @@ struct bnxt {\n #define BNXT_STATE_ABORT_ERR\t5\n #define BNXT_STATE_FW_FATAL_COND\t6\n #define BNXT_STATE_DRV_REGISTERED\t7\n-#define BNXT_STATE_PCI_CHANNEL_IO_FROZEN\t8\n #define BNXT_STATE_NAPI_DISABLED\t9\n #define BNXT_STATE_FW_ACTIVATE\t\t11\n #define BNXT_STATE_RECOVER\t\t12\n@@ -2490,6 +2489,7 @@ struct bnxt {\n \t */\n \tunsigned long\t\t*ring_affinity_set;\n \tint\t\t\tmax_irqs;\n+\tint\t\t\tirq_tbl_size;\n \tint\t\t\ttotal_irqs;\n \tint\t\t\tulp_num_msix_want;\n \tu8\t\t\tmac_addr[ETH_ALEN];\ndiff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c\nindex 622e89587e5db..5c9e770960e6d 100644\n--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c\n+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c\n@@ -853,6 +853,7 @@ static int bnxt_set_ringparam(struct net_device *dev,\n {\n \tu8 tcp_data_split = kernel_ering-\u003etcp_data_split;\n \tstruct bnxt *bp = netdev_priv(dev);\n+\tbool irq_re_init = false;\n \tu8 hds_config_mod;\n \tint rc;\n \n@@ -876,8 +877,13 @@ static int bnxt_set_ringparam(struct net_device *dev,\n \t\treturn -EINVAL;\n \t}\n \n+\tif (hds_config_mod \u0026\u0026\n+\t    tcp_data_split == ETHTOOL_TCP_DATA_SPLIT_ENABLED \u0026\u0026\n+\t    !(bp-\u003eflags \u0026 BNXT_FLAG_AGG_RINGS))\n+\t\tirq_re_init = true;\n+\n \tif (netif_running(dev))\n-\t\tbnxt_close_nic(bp, false, false);\n+\t\tbnxt_close_nic(bp, irq_re_init, false);\n \n \tif (hds_config_mod) {\n \t\tif (tcp_data_split == ETHTOOL_TCP_DATA_SPLIT_ENABLED)\n@@ -891,7 +897,7 @@ static int bnxt_set_ringparam(struct net_device *dev,\n \tbnxt_set_ring_params(bp);\n \n \tif (netif_running(dev)) {\n-\t\trc = bnxt_open_nic(bp, false, false);\n+\t\trc = bnxt_open_nic(bp, irq_re_init, false);\n \t\tif (rc)\n \t\t\treturn rc;\n \t}\n"
}

2/1 2026/09/29 05:51 agent patch-evaluator
0m Model:
core

Args:
null

Results:
{
  "EnableConfigs": null,
  "FocusSymbols": [],
  "Reasoning": "The patch exclusively modifies the Broadcom NetXtreme-C/E (bnxt) driver (drivers/net/ethernet/broadcom/bnxt/). This is a vendor-specific PCIe NIC driver requiring physical hardware not emulated in standard virtualized test environments (e.g., QEMU or GCE), making the modified code structurally unreachable for fuzzing.",
  "WorthFuzzing": false
}

Instruction:
You are an expert Linux kernel maintainer and security engineer.
Your job is to review a provided patch series and evaluate whether it warrants fuzzing with syzkaller.

IMPORTANT: The changes have ALREADY been applied and committed as the HEAD commit in
your workspace. Do NOT rely on internal assumptions. You must actively use your code access
tools to inspect the actual source code, callers, and surrounding context.

================================================================================
1. CORE TRIAGE PHILOSOPHY
================================================================================
The goal of patch fuzzing is to discover crashes, regressions, exposed latent bugs,
and newly triggered assertions introduced by the patch series.

- REACHABILITY IS THE PRIMARY GATE:
  Fuzzing can only discover bugs in code that can actually execute in standard virtualized
  environments (GCE or QEMU, utilizing software-emulated devices like USB gadgets, netdev, tun/tap).
  If the modified code is structurally unreachable (see Section 2), it MUST NOT be fuzzed,
  regardless of whether it adds assertions or complex logic.

- DO NOT BLINDLY TRUST "NO FUNCTIONAL CHANGE" (NFCI) OR "REFACTORING" CLAIMS:
  Patch authors routinely label changes as "cleanups", "refactorings", or state
  "No functional change intended". Do NOT take these claims at face value.
  Code refactorings that rearrange logic, introduce helper functions, or alter state management
  in core subsystems frequently introduce subtle semantic shifts or uncover latent kernel bugs.
  If reachable executable code is modified or refactored, it MUST be fuzzed.

- NEW OR MODIFIED ASSERTIONS IN REACHABLE CODE MUST BE FUZZED:
  When a patch introduces or modifies runtime checks or assertions (e.g., WARN_ON*, VM_WARN_ON*,
  BUG_ON*, lockdep_assert*) in reachable code paths, it enforces new or stricter invariants.
  Even if the author believes the invariant always holds, fuzzing is essential to verify whether
  an unusual sequence of operations can violate it.

================================================================================
2. WHEN TO RETURN WorthFuzzing=false (NEGATIVE CRITERIA)
================================================================================
Return WorthFuzzing=false ONLY IF all modified code falls strictly into one or more of these categories:

- Non-kernel and non-executable changes:
  * Modifications to Documentation/, comments, or spelling fixes.
  * User-space directories, self-tests, samples, or scripts (e.g., tools/, samples/, scripts/, usr/)
    that do not affect the compiled kernel image (vmlinux) or kernel modules.
  * Purely decorative logging (e.g., message strings in pr_err, printk, dev_info) or tracepoints
    that do not alter control flow or data structures.
  * Build system or Kconfig changes that do not alter compiled C logic.
- Structurally unreachable hardware:
  * Vendor-specific PCIe switches, SmartNICs, or GPU drivers (e.g., mlxsw, pds_core, qed,
    ionic, amdgpu) requiring physical ASIC/PCIe cards not emulated in standard QEMU.
- Unreachable execution paths:
  * Driver teardown callbacks (.remove, .shutdown, pci_unregister_driver) executed only during
    physical PCI hot-unplug or manual sysfs driver unbinding.
  * Code paths exclusive to architectures other than the target architecture.

================================================================================
3. WHEN TO RETURN WorthFuzzing=true (POSITIVE CRITERIA)
================================================================================
Return WorthFuzzing=true whenever the patch touches reachable executable code, including:
- Core Subsystems:
  * Any logic modifications in memory management (mm/), synchronization/locking (kernel/locking/),
    BPF, scheduler, core networking, VFS, or syscall handling.
- Refactorings and Code Cleanups:
  * Any restructuring of reachable data structures, helper abstractions, or algorithm flows.
- Runtime Assertions and Defensive Checks:
  * Any introduction or alteration of assertions (WARN_ON*, VM_WARN_ON*, BUG_ON*, etc.) in reachable paths.
- Reachable Drivers and Protocols:
  * Drivers accessible via virtual buses (virtio, USB gadget, loopback, netlink, binder, sockets, etc.).

================================================================================
4. EXTRACTING FocusSymbols (PREVENTING DILUTION)
================================================================================
When WorthFuzzing=true, you must extract specific kernel functions into FocusSymbols to guide the fuzzer:

- AVOID UBIQUITOUS LIFECYCLE HOT-PATHS:
  Do NOT list generic, ubiquitous functions called by almost every program in the corpus
  (including, but not limited to: general memory allocators and deallocators, page fault
  and trap handlers, or core synchronization primitives; this is not an exhaustive list).
  Listing ubiquitous functions causes the fuzzer to classify thousands of unrelated tests as "focused",
  which severely dilutes fuzzing effort away from the actual changes.

- TARGET SPECIFIC FEATURE LOGIC AND ENTRYPOINTS:
  List functions that specifically implement the logic being added or altered, or direct API entrypoints
  for the subsystem feature under review.

- HANDLING STATIC INLINE FUNCTIONS IN HEADERS (.h):
  Compiler-inlined static functions (such as static inlines in mm/*.h or include/linux/*.h) lack
  distinct symbol addresses in vmlinux and cannot be targeted directly by symbol coverage filters.
  If the changes are primarily in static inline helpers, identify non-static, feature-specific caller
  functions in .c files that exercise them (avoiding ubiquitous lifecycle wrappers).

================================================================================
5. IDENTIFYING EnableConfigs
================================================================================
Identify any specific CONFIG_ options required to properly compile and reach the modified code:
- Inspect Kconfig files and #ifdef guards; do not make assumptions.
- Check "depends on" lines in Kconfig to include any non-standard parent subsystem configs needed.
- Strip any 'CONFIG_' prefix (e.g., return "NET_IPV4" instead of "CONFIG_NET_IPV4").
Prefer calling several tools at the same time to save round-trips.


Use set-results tool to provide results of the analysis.
It must be called exactly once before the final reply.
Ignore results of this tool.

Prompt:
Target architecture: amd64

For your convenience, here is the diff of the changes:
commit da0eaeb5346d003e1f19ba5c7171208f952592e9
Author: syz-cluster <triage@syzkaller.com>
Date:   Tue Sep 29 05:51:52 2026 +0000

    syz-cluster: applied patch under review

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index d7728d0c5b6e6..810219d9cae21 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -8163,17 +8163,53 @@ static void bnxt_copy_reserved_rings(struct bnxt *bp, struct bnxt_hw_rings *hwr)
 	}
 }
 
+static bool bnxt_nitro_a0_rings_ok(struct bnxt *bp, int rx)
+{
+	if (BNXT_CHIP_TYPE_NITRO_A0(bp) && rx < 2)
+		return false;
+	return true;
+}
+
 static bool bnxt_rings_ok(struct bnxt *bp, struct bnxt_hw_rings *hwr)
 {
-	return hwr->tx && hwr->rx && hwr->cp && hwr->grp && hwr->vnic &&
-	       hwr->stat && (hwr->cp_p5 || !(bp->flags & BNXT_FLAG_CHIP_P5_PLUS));
+	int min_tx = bp->num_tc ? bp->num_tc : 1;
+	int min_rx = 1;
+	int min_cp = 1;
+	int tx_cp;
+
+	if (bp->tx_nr_rings_xdp)
+		min_tx++;
+
+	if (!(bp->flags & BNXT_FLAG_SHARED_RINGS)) {
+		tx_cp = bnxt_num_tx_to_cp(bp, min_tx);
+		min_cp = tx_cp + min_rx;
+	}
+	return hwr->tx >= min_tx && hwr->rx >= min_rx && hwr->cp >= min_cp &&
+	       hwr->grp && hwr->vnic && hwr->stat &&
+	       (hwr->cp_p5 || !(bp->flags & BNXT_FLAG_CHIP_P5_PLUS));
+}
+
+/* Check to see if we need to reset the user configured RSS table
+ * (default RSS context) because of change in RX rings from
+ * old_rx to new_rx.
+ */
+static void bnxt_check_rss_tbl_lost(struct bnxt *bp, int old_rx, int new_rx)
+{
+	if (netif_is_rxfh_configured(bp->dev) &&
+	    (bnxt_get_nr_rss_ctxs(bp, old_rx) !=
+	     bnxt_get_nr_rss_ctxs(bp, new_rx) ||
+	     bnxt_get_max_rss_ring(bp) >= new_rx)) {
+		ethtool_rxfh_indir_lost(bp->dev);
+	}
 }
 
 static int bnxt_get_avail_msix(struct bnxt *bp, int num);
+static void bnxt_adj_rings(struct bnxt *bp);
 
 static int __bnxt_reserve_rings(struct bnxt *bp)
 {
 	struct bnxt_en_dev *edev = bp->edev[BNXT_AUXDEV_RDMA];
+	int req_rx_rings = bp->rx_nr_rings;
 	struct bnxt_hw_rings hwr = {0};
 	int rx_rings, old_rx_rings, rc;
 	int cp = bp->cp_nr_rings;
@@ -8241,54 +8277,68 @@ static int __bnxt_reserve_rings(struct bnxt *bp)
 		hwr.stat -= bnxt_get_ulp_stat_ctxs(bp);
 	hwr.cp = min_t(int, hwr.cp, hwr.stat);
 	rc = bnxt_trim_rings(bp, &rx_rings, &hwr.tx, hwr.cp, sh);
+	if (rc)
+		goto reserve_rings_exit;
+
+	if (bp->bnapi && (rx_rings < bp->rx_nr_rings ||
+			  hwr.tx < bp->tx_nr_rings)) {
+		netdev_err(bp->dev, "Unexpected ring shrinkage %d/%d RX/TX to %d/%d\n",
+			   bp->rx_nr_rings, bp->tx_nr_rings, rx_rings, hwr.tx);
+		rc = -ENOSPC;
+		goto reserve_rings_exit;
+	}
 	if (bp->flags & BNXT_FLAG_AGG_RINGS)
 		hwr.rx = rx_rings << 1;
 	tx_cp = bnxt_num_tx_to_cp(bp, hwr.tx);
 	hwr.cp = sh ? max_t(int, tx_cp, rx_rings) : tx_cp + rx_rings;
+
+	if (!bnxt_rings_ok(bp, &hwr)) {
+		rc = -ENOMEM;
+		goto reserve_rings_exit;
+	}
+
 	if (hwr.tx != bp->tx_nr_rings) {
 		netdev_warn(bp->dev,
 			    "Able to reserve only %d out of %d requested TX rings\n",
 			    hwr.tx, bp->tx_nr_rings);
 	}
 	bp->tx_nr_rings = hwr.tx;
+	bp->rx_nr_rings = rx_rings;
+	bp->cp_nr_rings = hwr.cp;
+
+	if (!bp->bnapi)
+		bnxt_adj_rings(bp);
 
 	/* If we cannot reserve all the RX rings, reset the RSS map only
 	 * if absolutely necessary
 	 */
-	if (rx_rings != bp->rx_nr_rings) {
-		netdev_warn(bp->dev, "Able to reserve only %d out of %d requested RX rings\n",
-			    rx_rings, bp->rx_nr_rings);
-		if (netif_is_rxfh_configured(bp->dev) &&
-		    (bnxt_get_nr_rss_ctxs(bp, bp->rx_nr_rings) !=
-		     bnxt_get_nr_rss_ctxs(bp, rx_rings) ||
-		     bnxt_get_max_rss_ring(bp) >= rx_rings)) {
-			ethtool_rxfh_indir_lost(bp->dev);
-		}
+	if (req_rx_rings != bp->rx_nr_rings) {
+		netdev_warn(bp->dev, "RX rings reduced from %d to %d\n",
+			    req_rx_rings, bp->rx_nr_rings);
+		bnxt_check_rss_tbl_lost(bp, req_rx_rings, bp->rx_nr_rings);
 	}
-	bp->rx_nr_rings = rx_rings;
-	bp->cp_nr_rings = hwr.cp;
 
 	/* Fall back if we cannot reserve enough HW RSS contexts */
 	if ((bp->rss_cap & BNXT_RSS_CAP_LARGE_RSS_CTX) &&
 	    hwr.rss_ctx < bnxt_get_total_rss_ctxs(bp, &hwr))
 		bp->rss_cap &= ~BNXT_RSS_CAP_LARGE_RSS_CTX;
 
-	if (!bnxt_rings_ok(bp, &hwr))
-		return -ENOMEM;
-
-	if (old_rx_rings != bp->hw_resc.resv_rx_rings &&
+	if ((old_rx_rings != bp->hw_resc.resv_rx_rings ||
+	     req_rx_rings != bp->rx_nr_rings) &&
 	    !netif_is_rxfh_configured(bp->dev))
 		bnxt_set_dflt_rss_indir_tbl(bp, NULL);
 
+reserve_rings_exit:
 	if (!bnxt_ulp_registered(edev) && BNXT_NEW_RM(bp)) {
 		int resv_msix, resv_ctx, ulp_ctxs;
 		struct bnxt_hw_resc *hw_resc;
 
 		hw_resc = &bp->hw_resc;
-		resv_msix = hw_resc->resv_irqs - bp->cp_nr_rings;
+		resv_msix = max_t(int, hw_resc->resv_irqs - bp->cp_nr_rings, 0);
 		ulp_msix = min_t(int, resv_msix, ulp_msix);
 		bnxt_set_ulp_msix_num(bp, ulp_msix);
-		resv_ctx = hw_resc->resv_stat_ctxs  - bp->cp_nr_rings;
+		resv_ctx = max_t(int, hw_resc->resv_stat_ctxs - bp->cp_nr_rings,
+				 0);
 		ulp_ctxs = min(resv_ctx, bnxt_get_ulp_stat_ctxs(bp));
 		bnxt_set_ulp_stat_ctxs(bp, ulp_ctxs);
 	}
@@ -11511,6 +11561,12 @@ static int bnxt_change_msix(struct bnxt *bp, int total)
 	struct msi_map map;
 	int i;
 
+	if (!bp->irq_tbl)
+		return 0;
+
+	if (total > bp->irq_tbl_size)
+		return bp->total_irqs;
+
 	/* add MSIX to the end if needed */
 	for (i = bp->total_irqs; i < total; i++) {
 		map = pci_msix_alloc_irq_at(bp->pdev, i, NULL);
@@ -11630,7 +11686,9 @@ static int bnxt_get_num_msix(struct bnxt *bp)
 
 static int bnxt_init_int_mode(struct bnxt *bp)
 {
-	int i, total_vecs, max, rc, min = 1, ulp_msix, tx_cp, tbl_size;
+	int i, total_vecs, max, rc, min = 1, ulp_msix, tbl_size, req_rx_rings;
+	int min_tx, req_tx_rings, req_cp_rings, req_tx_per_tc, req_tx_xdp;
+	int tcs = bp->num_tc ? bp->num_tc : 1;
 
 	total_vecs = bnxt_get_num_msix(bp);
 	max = bnxt_get_max_func_irqs(bp);
@@ -11653,34 +11711,74 @@ static int bnxt_init_int_mode(struct bnxt *bp)
 
 	tbl_size = total_vecs;
 	if (pci_msix_can_alloc_dyn(bp->pdev))
-		tbl_size = max;
+		tbl_size = bp->hw_resc.max_irqs;
 	bp->irq_tbl = kzalloc_objs(*bp->irq_tbl, tbl_size);
 	if (!bp->irq_tbl) {
 		rc = -ENOMEM;
 		goto msix_setup_exit;
 	}
+	bp->irq_tbl_size = tbl_size;
 
 	for (i = 0; i < total_vecs; i++)
 		bp->irq_tbl[i].vector = pci_irq_vector(bp->pdev, i);
 
 	bp->total_irqs = total_vecs;
+	req_tx_rings = bp->tx_nr_rings;
+	req_tx_per_tc = bp->tx_nr_rings_per_tc;
+	req_tx_xdp = bp->tx_nr_rings_xdp;
+	req_rx_rings = bp->rx_nr_rings;
+	req_cp_rings = bp->cp_nr_rings;
+
 	/* Trim rings based upon num of vectors allocated */
 	rc = bnxt_trim_rings(bp, &bp->rx_nr_rings, &bp->tx_nr_rings,
 			     total_vecs - ulp_msix, min == 1);
 	if (rc)
 		goto msix_setup_exit;
 
-	tx_cp = bnxt_num_tx_to_cp(bp, bp->tx_nr_rings);
-	bp->cp_nr_rings = (min == 1) ?
-			  max_t(int, tx_cp, bp->rx_nr_rings) :
-			  tx_cp + bp->rx_nr_rings;
+	min_tx = bp->tx_nr_rings_xdp ? tcs + 1 : tcs;
+	if (bp->tx_nr_rings < min_tx) {
+		netdev_err(bp->dev, "Not enough MSI-X to satisfy min. TX rings\n");
+		rc = -ENOMEM;
+		goto msix_setup_exit_restore;
+	}
+
+	bnxt_adj_rings(bp);
 
+	if (!bnxt_nitro_a0_rings_ok(bp, bp->rx_nr_rings)) {
+		netdev_err(bp->dev, "Not enough MSI-X to satisfy min. RX rings\n");
+		rc = -ENOMEM;
+		goto msix_setup_exit_restore;
+	}
+
+	if (bp->bnapi && (req_tx_rings != bp->tx_nr_rings ||
+			  req_rx_rings != bp->rx_nr_rings)) {
+		netdev_err(bp->dev, "Cannot shrink rings once NAPI is allocated\n");
+		rc = -ENOSPC;
+		goto msix_setup_exit_restore;
+	}
+
+	if (req_rx_rings != bp->rx_nr_rings) {
+		netdev_warn(bp->dev, "RX rings reduced from %d to %d\n",
+			    req_rx_rings, bp->rx_nr_rings);
+		bnxt_check_rss_tbl_lost(bp, req_rx_rings, bp->rx_nr_rings);
+		if (!netif_is_rxfh_configured(bp->dev))
+			bnxt_set_dflt_rss_indir_tbl(bp, NULL);
+	}
 	return 0;
 
+msix_setup_exit_restore:
+	bp->tx_nr_rings = req_tx_rings;
+	bp->tx_nr_rings_per_tc = req_tx_per_tc;
+	bp->tx_nr_rings_xdp = req_tx_xdp;
+	bp->rx_nr_rings = req_rx_rings;
+	bp->cp_nr_rings = req_cp_rings;
+
 msix_setup_exit:
 	netdev_err(bp->dev, "bnxt_init_int_mode err: %x\n", rc);
 	kfree(bp->irq_tbl);
 	bp->irq_tbl = NULL;
+	bp->total_irqs = 0;
+	bp->irq_tbl_size = 0;
 	pci_free_irq_vectors(bp->pdev);
 	return rc;
 }
@@ -11691,33 +11789,36 @@ static void bnxt_clear_int_mode(struct bnxt *bp)
 
 	kfree(bp->irq_tbl);
 	bp->irq_tbl = NULL;
+	bp->irq_tbl_size = 0;
 }
 
-int bnxt_reserve_rings(struct bnxt *bp, bool irq_re_init)
+static int bnxt_irqs_required(struct bnxt *bp)
 {
 	struct bnxt_en_dev *edev = bp->edev[BNXT_AUXDEV_RDMA];
-	bool irq_cleared = false;
-	bool irq_change = false;
-	int tcs = bp->num_tc;
-	int irqs_required;
-	int rc;
-
-	if (!bnxt_need_reserve_rings(bp))
-		return 0;
 
 	if (BNXT_NEW_RM(bp) && !bnxt_ulp_registered(edev)) {
 		int ulp_msix = bnxt_get_avail_msix(bp, bp->ulp_num_msix_want);
 
 		if (ulp_msix > bp->ulp_num_msix_want)
 			ulp_msix = bp->ulp_num_msix_want;
-		irqs_required = ulp_msix + bp->cp_nr_rings;
-	} else {
-		irqs_required = bnxt_get_num_msix(bp);
+		return ulp_msix + bp->cp_nr_rings;
 	}
+	return bnxt_get_num_msix(bp);
+}
+
+int bnxt_reserve_rings(struct bnxt *bp, bool irq_re_init)
+{
+	bool irq_cleared = false;
+	int irqs_required;
+	int rc;
+
+	if (!bnxt_need_reserve_rings(bp))
+		return 0;
+
+	irqs_required = bnxt_irqs_required(bp);
 
 	if (irq_re_init && BNXT_NEW_RM(bp) && irqs_required != bp->total_irqs) {
-		irq_change = true;
-		if (!pci_msix_can_alloc_dyn(bp->pdev)) {
+		if (!pci_msix_can_alloc_dyn(bp->pdev) || !bp->irq_tbl) {
 			bnxt_ulp_irq_stop(bp);
 			bnxt_clear_int_mode(bp);
 			irq_cleared = true;
@@ -11728,25 +11829,17 @@ int bnxt_reserve_rings(struct bnxt *bp, bool irq_re_init)
 		if (!rc)
 			rc = bnxt_init_int_mode(bp);
 		bnxt_ulp_irq_restart(bp, rc);
-	} else if (irq_change && !rc) {
-		if (bnxt_change_msix(bp, irqs_required) != irqs_required)
+	} else if (irq_re_init && BNXT_NEW_RM(bp) && !rc) {
+		/* __bnxt_reserve_rings() may have shrunk the rings */
+		irqs_required = bnxt_irqs_required(bp);
+		if (irqs_required != bp->total_irqs &&
+		    bnxt_change_msix(bp, irqs_required) != irqs_required)
 			rc = -ENOSPC;
 	}
 	if (rc) {
 		netdev_err(bp->dev, "ring reservation/IRQ init failure rc: %d\n", rc);
 		return rc;
 	}
-	if (tcs && (bp->tx_nr_rings_per_tc * tcs !=
-		    bp->tx_nr_rings - bp->tx_nr_rings_xdp)) {
-		netdev_err(bp->dev, "tx ring reservation failure\n");
-		netdev_reset_tc(bp->dev);
-		bp->num_tc = 0;
-		if (bp->tx_nr_rings_xdp)
-			bp->tx_nr_rings_per_tc = bp->tx_nr_rings_xdp;
-		else
-			bp->tx_nr_rings_per_tc = bp->tx_nr_rings;
-		return -ENOMEM;
-	}
 	return 0;
 }
 
@@ -13251,13 +13344,29 @@ static void bnxt_set_xdp_tx_rings(struct bnxt *bp)
 	bp->tx_nr_rings += bp->tx_nr_rings_xdp;
 }
 
-static void bnxt_adj_tx_rings(struct bnxt *bp)
+static void bnxt_adj_rings(struct bnxt *bp)
 {
-	/* Make adjustments if reserved TX rings are less than requested */
-	bp->tx_nr_rings -= bp->tx_nr_rings_xdp;
-	bp->tx_nr_rings_per_tc = bnxt_tx_nr_rings_per_tc(bp);
-	if (bp->tx_nr_rings_xdp)
-		bnxt_set_xdp_tx_rings(bp);
+	int tcs = bp->num_tc ? bp->num_tc : 1;
+	int rx = bp->rx_nr_rings;
+
+	/* Make adjustments to rings based on TC/XDP/shared rings policies */
+	if (bp->tx_nr_rings_xdp) {
+		tcs++;
+		bp->tx_nr_rings_per_tc = bp->tx_nr_rings / tcs;
+		bp->tx_nr_rings_xdp = bp->tx_nr_rings_per_tc;
+	} else {
+		bp->tx_nr_rings_per_tc = bnxt_tx_nr_rings_per_tc(bp);
+	}
+
+	if (bp->flags & BNXT_FLAG_SHARED_RINGS) {
+		bp->tx_nr_rings_per_tc = min(bp->tx_nr_rings_per_tc, rx);
+		bp->rx_nr_rings = bp->tx_nr_rings_per_tc;
+		if (bp->tx_nr_rings_xdp)
+			bp->tx_nr_rings_xdp = bp->tx_nr_rings_per_tc;
+	}
+
+	bp->tx_nr_rings = bp->tx_nr_rings_per_tc * tcs;
+	bnxt_set_cp_rings(bp, bp->flags & BNXT_FLAG_SHARED_RINGS);
 }
 
 static int __bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init)
@@ -13277,7 +13386,6 @@ static int __bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init)
 	if (rc)
 		return rc;
 
-	bnxt_adj_tx_rings(bp);
 	rc = bnxt_alloc_mem(bp, irq_re_init);
 	if (rc) {
 		netdev_err(bp->dev, "bnxt_alloc_mem err: %x\n", rc);
@@ -14081,7 +14189,8 @@ static int bnxt_reinit_features(struct bnxt *bp, bool irq_re_init,
 				bool link_re_init, u32 flags, bool update_tpa)
 {
 	bnxt_close_nic(bp, irq_re_init, link_re_init);
-	bp->flags = flags;
+	bp->flags = (bp->flags & ~BNXT_FLAG_ALL_CONFIG_FEATS) |
+		    (flags & BNXT_FLAG_ALL_CONFIG_FEATS);
 	if (update_tpa)
 		bnxt_set_ring_params(bp);
 	return bnxt_open_nic(bp, irq_re_init, link_re_init);
@@ -14091,10 +14200,11 @@ static int bnxt_set_features(struct net_device *dev, netdev_features_t features)
 {
 	bool update_tpa = false, update_ntuple = false;
 	struct bnxt *bp = netdev_priv(dev);
+	bool irq_re_init = false;
 	u32 flags = bp->flags;
+	bool re_init = false;
 	u32 changes;
 	int rc = 0;
-	bool re_init = false;
 
 	bp->tx_wake_thresh = max_t(int, bp->tx_ring_size / 2,
 				   bnxt_min_tx_desc_cnt(bp, features));
@@ -14121,8 +14231,12 @@ static int bnxt_set_features(struct net_device *dev, netdev_features_t features)
 		update_tpa = true;
 		if ((bp->flags & BNXT_FLAG_TPA) == 0 ||
 		    (flags & BNXT_FLAG_TPA) == 0 ||
-		    (bp->flags & BNXT_FLAG_CHIP_P5_PLUS))
+		    (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)) {
 			re_init = true;
+			if (!(bp->flags & BNXT_FLAG_AGG_RINGS) &&
+			    (flags & BNXT_FLAG_AGG_RINGS))
+				irq_re_init = true;
+		}
 	}
 
 	if (changes & ~BNXT_FLAG_TPA)
@@ -14145,7 +14259,7 @@ static int bnxt_set_features(struct net_device *dev, netdev_features_t features)
 			return bnxt_reinit_features(bp, true, false, flags, update_tpa);
 
 		if (re_init)
-			return bnxt_reinit_features(bp, false, false, flags, update_tpa);
+			return bnxt_reinit_features(bp, irq_re_init, false, flags, update_tpa);
 
 		if (update_tpa) {
 			bp->flags = flags;
@@ -15081,7 +15195,7 @@ int bnxt_check_rings(struct bnxt *bp, int tx, int rx, bool sh, int tcs,
 			hwr.cp += bnxt_get_ulp_msix_num(bp);
 			hwr.cp = min_t(int, hwr.cp, bnxt_get_max_func_irqs(bp));
 		}
-		if (hwr.cp > bp->total_irqs) {
+		if (bp->irq_tbl && hwr.cp > bp->total_irqs) {
 			int total_msix = bnxt_change_msix(bp, hwr.cp);
 
 			if (total_msix < hwr.cp) {
@@ -16937,7 +17051,6 @@ static int bnxt_set_dflt_rings(struct bnxt *bp, bool sh)
 	if (rc && rc != -ENODEV)
 		netdev_warn(bp->dev, "Unable to reserve tx rings\n");
 
-	bnxt_adj_tx_rings(bp);
 	if (sh)
 		bnxt_adj_dflt_rings(bp, true);
 
@@ -16946,7 +17059,6 @@ static int bnxt_set_dflt_rings(struct bnxt *bp, bool sh)
 		rc = __bnxt_reserve_rings(bp);
 		if (rc && rc != -ENODEV)
 			netdev_warn(bp->dev, "2nd rings reservation failed.\n");
-		bnxt_adj_tx_rings(bp);
 	}
 	if (BNXT_CHIP_TYPE_NITRO_A0(bp)) {
 		bp->rx_nr_rings++;
@@ -16980,8 +17092,6 @@ static int bnxt_init_dflt_ring_mode(struct bnxt *bp)
 	if (rc)
 		goto init_dflt_ring_err;
 
-	bnxt_adj_tx_rings(bp);
-
 	bnxt_set_dflt_rfs(bp);
 
 init_dflt_ring_err:
@@ -17111,6 +17221,50 @@ void bnxt_print_device_info(struct bnxt *bp)
 	pcie_print_link_status(bp->pdev);
 }
 
+static void bnxt_clear_bars(struct pci_dev *pdev)
+{
+	int off;
+
+	for (off = PCI_BASE_ADDRESS_0; off <= PCI_BASE_ADDRESS_5; off += 4)
+		pci_write_config_dword(pdev, off, 0);
+}
+
+/* Clear any pending DMA transactions from crash kernel while loading driver in
+ * capture kernel.
+ */
+static void bnxt_kdump_reset(struct pci_dev *pdev)
+{
+	u16 cmd;
+	int rc;
+
+	pci_read_config_word(pdev, PCI_COMMAND, &cmd);
+	cmd &= ~(PCI_COMMAND_MASTER | PCI_COMMAND_MEMORY);
+	pci_write_config_word(pdev, PCI_COMMAND, cmd);
+
+	if (pci_save_state(pdev))
+		dev_warn(&pdev->dev, "Failed to save PCI state, PCI restore may be incomplete\n");
+
+	/* Since it is kdump kernel, try to continue in the best possible way
+	 * even if the device is unresponsive after FLR.  The device may
+	 * eventually respond to HWRM_VER_GET later in the init sequence.
+	 */
+	rc = pcie_flr(pdev);
+	if (rc)
+		dev_warn(&pdev->dev, "pcie_flr() failed (rc: %d), trying to continue\n",
+			 rc);
+
+	/* A complete or partial reset has been done.  Clear the BARs
+	 * if the device is responsive.
+	 */
+	pci_read_config_word(pdev, PCI_COMMAND, &cmd);
+	if (PCI_POSSIBLE_ERROR(cmd))
+		dev_warn(&pdev->dev, "PCI config space inaccessible after FLR, not clearing BARs\n");
+	else
+		bnxt_clear_bars(pdev);
+
+	pci_restore_state(pdev);
+}
+
 static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
 	struct bnxt_hw_resc *hw_resc;
@@ -17126,15 +17280,11 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		return -ENODEV;
 	}
 
-	/* Clear any pending DMA transactions from crash kernel
-	 * while loading driver in capture kernel.
-	 */
-	if (is_kdump_kernel()) {
-		pci_clear_master(pdev);
-		pcie_flr(pdev);
-	}
-
 	max_irqs = bnxt_get_max_irq(pdev);
+
+	if (is_kdump_kernel())
+		bnxt_kdump_reset(pdev);
+
 	dev = alloc_etherdev_mqs(sizeof(*bp), max_irqs * BNXT_MAX_QUEUE,
 				 max_irqs);
 	if (!dev)
@@ -17325,11 +17475,6 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (rc)
 		goto init_err_pci_clean;
 
-	/* No TC has been set yet and rings may have been trimmed due to
-	 * limited MSIX, so we re-initialize the TX rings per TC.
-	 */
-	bp->tx_nr_rings_per_tc = bp->tx_nr_rings;
-
 	if (BNXT_PF(bp)) {
 		if (!bnxt_pf_wq) {
 			bnxt_pf_wq =
@@ -17565,10 +17710,8 @@ static pci_ers_result_t bnxt_io_error_detected(struct pci_dev *pdev,
 	 * so we disable bus master to prevent any potential bad DMAs before
 	 * freeing kernel memory.
 	 */
-	if (state == pci_channel_io_frozen) {
-		set_bit(BNXT_STATE_PCI_CHANNEL_IO_FROZEN, &bp->state);
+	if (state == pci_channel_io_frozen)
 		bnxt_fw_fatal_close(bp);
-	}
 
 	if (netif_running(netdev))
 		__bnxt_close_nic(bp, true, true);
@@ -17598,69 +17741,80 @@ static pci_ers_result_t bnxt_io_slot_reset(struct pci_dev *pdev)
 	struct bnxt *bp = netdev_priv(netdev);
 	int retry = 0;
 	int err = 0;
-	int off;
+	u16 cmd;
 
 	netdev_info(bp->dev, "PCI Slot Reset\n");
 
-	if (test_bit(BNXT_STATE_PCI_CHANNEL_IO_FROZEN, &bp->state)) {
-		/* After DPC, the chip should return CRS when the vendor ID
-		 * config register is read until it is ready.  On all chips,
-		 * this is not happening reliably so add a 5-second delay as a
-		 * workaround.
-		 */
-		msleep(5000);
-	}
+	/* After a PCIe hot reset, the chip should return CRS when the
+	 * vendor ID config register is read until it is ready.  On all
+	 * chips, this is not happening reliably so add a 5-second delay
+	 * as a workaround.
+	 */
+	msleep(5000);
 
 	netdev_lock(netdev);
 
-	if (pci_enable_device(pdev)) {
+	pci_read_config_word(pdev, PCI_COMMAND, &cmd);
+	if (PCI_POSSIBLE_ERROR(cmd)) {
 		dev_err(&pdev->dev,
-			"Cannot re-enable PCI device after reset.\n");
-	} else {
-		pci_set_master(pdev);
-		/* Upon fatal error, our device internal logic that latches to
-		 * BAR value is getting reset and will restore only upon
-		 * rewriting the BARs.
-		 *
-		 * As pci_restore_state() does not re-write the BARs if the
-		 * value is same as saved value earlier, driver needs to
-		 * write the BARs to 0 to force restore, in case of fatal error.
-		 */
-		if (test_and_clear_bit(BNXT_STATE_PCI_CHANNEL_IO_FROZEN,
-				       &bp->state)) {
-			for (off = PCI_BASE_ADDRESS_0;
-			     off <= PCI_BASE_ADDRESS_5; off += 4)
-				pci_write_config_dword(bp->pdev, off, 0);
-		}
-		pci_restore_state(pdev);
+			"PCI config space inaccessible after reset\n");
+		goto reset_exit;
+	}
 
-		bnxt_inv_fw_health_reg(bp);
-		bnxt_try_map_fw_health_reg(bp);
+	/* Upon PCIe error, our device internal logic that latches to
+	 * BAR value is getting reset and will restore only upon
+	 * rewriting the BARs.
+	 *
+	 * As pci_restore_state() does not re-write the BARs if the
+	 * value is same as saved value earlier, driver needs to
+	 * write the BARs to 0 to force restore.
+	 */
+	pci_clear_master(pdev);
+	pci_read_config_word(pdev, PCI_COMMAND, &cmd);
+	cmd &= ~PCI_COMMAND_MEMORY;
+	pci_write_config_word(pdev, PCI_COMMAND, cmd);
 
-		/* In some PCIe AER scenarios, firmware may take up to
-		 * 10 seconds to become ready in the worst case.
-		 */
-		do {
-			err = bnxt_try_recover_fw(bp);
-			if (!err)
-				break;
-			retry++;
-		} while (retry < BNXT_FW_SLOT_RESET_RETRY);
+	bnxt_clear_bars(pdev);
+	pci_restore_state(pdev);
 
-		if (err) {
-			dev_err(&pdev->dev, "Firmware not ready\n");
-			goto reset_exit;
+	if (!pci_is_enabled(pdev) && pci_enable_device(pdev)) {
+		dev_err(&pdev->dev,
+			"Cannot re-enable PCI device after reset.\n");
+		pci_read_config_word(pdev, PCI_COMMAND, &cmd);
+		if (!PCI_POSSIBLE_ERROR(cmd)) {
+			cmd &= ~(PCI_COMMAND_MASTER | PCI_COMMAND_MEMORY);
+			pci_write_config_word(pdev, PCI_COMMAND, cmd);
 		}
+		goto reset_exit;
+	}
+	pci_set_master(pdev);
 
-		err = bnxt_hwrm_func_reset(bp);
+	bnxt_inv_fw_health_reg(bp);
+	bnxt_try_map_fw_health_reg(bp);
+
+	/* In some PCIe AER scenarios, firmware may take up to
+	 * 10 seconds to become ready in the worst case.
+	 */
+	do {
+		err = bnxt_try_recover_fw(bp);
 		if (!err)
-			result = PCI_ERS_RESULT_RECOVERED;
+			break;
+		retry++;
+	} while (retry < BNXT_FW_SLOT_RESET_RETRY);
 
-		/* IRQ will be initialized later in bnxt_io_resume */
-		bnxt_ulp_irq_stop(bp);
-		bnxt_clear_int_mode(bp);
+	if (err) {
+		dev_err(&pdev->dev, "Firmware not ready\n");
+		goto reset_exit;
 	}
 
+	err = bnxt_hwrm_func_reset(bp);
+	if (!err)
+		result = PCI_ERS_RESULT_RECOVERED;
+
+	/* IRQ will be initialized later in bnxt_io_resume */
+	bnxt_ulp_irq_stop(bp);
+	bnxt_clear_int_mode(bp);
+
 reset_exit:
 	clear_bit(BNXT_STATE_IN_FW_RESET, &bp->state);
 	bnxt_clear_reservations(bp, true);
@@ -17692,7 +17846,7 @@ static void bnxt_io_resume(struct pci_dev *pdev)
 			err = bnxt_open(netdev);
 		} else {
 			err = bnxt_reserve_rings(bp, true);
-			if (!err)
+			if (!err && !bp->irq_tbl)
 				err = bnxt_init_int_mode(bp);
 		}
 	}
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index c673b2ce4a0d2..75428a585577d 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -2468,7 +2468,6 @@ struct bnxt {
 #define BNXT_STATE_ABORT_ERR	5
 #define BNXT_STATE_FW_FATAL_COND	6
 #define BNXT_STATE_DRV_REGISTERED	7
-#define BNXT_STATE_PCI_CHANNEL_IO_FROZEN	8
 #define BNXT_STATE_NAPI_DISABLED	9
 #define BNXT_STATE_FW_ACTIVATE		11
 #define BNXT_STATE_RECOVER		12
@@ -2490,6 +2489,7 @@ struct bnxt {
 	 */
 	unsigned long		*ring_affinity_set;
 	int			max_irqs;
+	int			irq_tbl_size;
 	int			total_irqs;
 	int			ulp_num_msix_want;
 	u8			mac_addr[ETH_ALEN];
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index 622e89587e5db..5c9e770960e6d 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -853,6 +853,7 @@ static int bnxt_set_ringparam(struct net_device *dev,
 {
 	u8 tcp_data_split = kernel_ering->tcp_data_split;
 	struct bnxt *bp = netdev_priv(dev);
+	bool irq_re_init = false;
 	u8 hds_config_mod;
 	int rc;
 
@@ -876,8 +877,13 @@ static int bnxt_set_ringparam(struct net_device *dev,
 		return -EINVAL;
 	}
 
+	if (hds_config_mod &&
+	    tcp_data_split == ETHTOOL_TCP_DATA_SPLIT_ENABLED &&
+	    !(bp->flags & BNXT_FLAG_AGG_RINGS))
+		irq_re_init = true;
+
 	if (netif_running(dev))
-		bnxt_close_nic(bp, false, false);
+		bnxt_close_nic(bp, irq_re_init, false);
 
 	if (hds_config_mod) {
 		if (tcp_data_split == ETHTOOL_TCP_DATA_SPLIT_ENABLED)
@@ -891,7 +897,7 @@ static int bnxt_set_ringparam(struct net_device *dev,
 	bnxt_set_ring_params(bp);
 
 	if (netif_running(dev)) {
-		rc = bnxt_open_nic(bp, false, false);
+		rc = bnxt_open_nic(bp, irq_re_init, false);
 		if (rc)
 			return rc;
 	}

3/2 2026/09/29 05:51 llm patch-evaluator
0m Model:
gemini-3.8-flash

Args:
null

Results:
null

Tokens:
							input: 13589
							output: 36
							thoughts: 385
						

Thoughts:
**Fuzzing Evaluation for bnxt Driver Patches**

Okay, let's take a look at this patch series and determine if it's a good candidate for syzkaller fuzzing. The instructions are clear: reachability is the key. I can't waste time on unreachable code. It needs to *execute* in a standard virtualized environment like GCE or QEMU, using software-emulated devices. Vendor-specific PCIe hardware that's not emulated, like SmartNICs or GPUs, is a non-starter right off the bat. And of course, anything unreachable, like driver teardown code or architecture-specific paths, is out.

Now, what are these patches actually touching? I see `drivers/net/ethernet/broadcom/bnxt/bnxt.c`, `.h`, and `.ethtool.c`.  Ah, `bnxt`... that's Broadcom's NetXtreme-C or NetXtreme-E BCM573xx/BCM574xx/BCM575xx driver.  The big question is, and this is crucial, does QEMU or syzkaller have reasonable emulation for the Broadcom bnxt (NetXtreme-E) NIC?  I need to check if this driver relies on physical hardware or if it's something that can run in a virtualized environment. Time to dig into the driver's Kconfig or any available driver documentation for clues. That's the critical information needed to make this assessment.



4/2 2026/09/29 05:52 tool grepper
0m Error:
no matches

Args:
{
  "Expression": "config BNXT",
  "PathPrefix": "drivers/net/ethernet/broadcom/bnxt/"
}

Results:
{
  "error": "no matches"
}

5/2 2026/09/29 05:52 llm patch-evaluator
0m Model:
gemini-3.8-flash

Args:
null

Results:
null

Tokens:
							input: 14348
							output: 32
							thoughts: 10
						

Thoughts:
Okay, here's my attempt at summarizing that thought process, tailored for an expert audience:

**Hunting for Broadcom's Kconfig**

Alright, so the task at hand is to locate the Kconfig configuration files within the Broadcom ecosystem. That's the starting point. I need to get my hands on those configs. Presumably, they'll be key to understanding the board's configuration, and the build process. I'm thinking, "Let's find Kconfig in Broadcom." That's the core directive. Now it's a matter of where to look – depending on the specific target, there could be variations. This means I'll want to think about the different Broadcom products I might have to deal with, and how to track down this information across them.



6/2 2026/09/29 05:52 tool grepper
0m Args:
{
  "Expression": "BNXT",
  "PathPrefix": "drivers/net/ethernet/broadcom/"
}

Results:
{
  "Output": "\nFull output is too long, showing 500 out of 9905 lines.\nUse more precise expression if possible.\n\ndrivers/net/ethernet/broadcom/Kconfig=195=config SYSTEMPORT\n--\ndrivers/net/ethernet/broadcom/Kconfig-206-\ndrivers/net/ethernet/broadcom/Kconfig:207:config BNXT\ndrivers/net/ethernet/broadcom/Kconfig-208-\ttristate \"Broadcom NetXtreme-C/E support\"\n--\ndrivers/net/ethernet/broadcom/Kconfig-221-\ndrivers/net/ethernet/broadcom/Kconfig:222:config BNXT_SRIOV\ndrivers/net/ethernet/broadcom/Kconfig-223-\tbool \"Broadcom NetXtreme-C/E SR-IOV support\"\ndrivers/net/ethernet/broadcom/Kconfig:224:\tdepends on BNXT \u0026\u0026 PCI_IOV\ndrivers/net/ethernet/broadcom/Kconfig-225-\tdefault y\n--\ndrivers/net/ethernet/broadcom/Kconfig-230-\ndrivers/net/ethernet/broadcom/Kconfig:231:config BNXT_FLOWER_OFFLOAD\ndrivers/net/ethernet/broadcom/Kconfig-232-\tbool \"TC Flower offload support for NetXtreme-C/E\"\ndrivers/net/ethernet/broadcom/Kconfig:233:\tdepends on BNXT\ndrivers/net/ethernet/broadcom/Kconfig-234-\tdefault y\n--\ndrivers/net/ethernet/broadcom/Kconfig-239-\ndrivers/net/ethernet/broadcom/Kconfig:240:config BNXT_DCB\ndrivers/net/ethernet/broadcom/Kconfig-241-\tbool \"Data Center Bridging (DCB) Support\"\ndrivers/net/ethernet/broadcom/Kconfig-242-\tdefault n\ndrivers/net/ethernet/broadcom/Kconfig:243:\tdepends on BNXT \u0026\u0026 DCB\ndrivers/net/ethernet/broadcom/Kconfig-244-\thelp\n--\ndrivers/net/ethernet/broadcom/Kconfig-249-\ndrivers/net/ethernet/broadcom/Kconfig:250:config BNXT_HWMON\ndrivers/net/ethernet/broadcom/Kconfig-251-\tbool \"Broadcom NetXtreme-C/E HWMON support\"\ndrivers/net/ethernet/broadcom/Kconfig-252-\tdefault y\ndrivers/net/ethernet/broadcom/Kconfig:253:\tdepends on BNXT \u0026\u0026 HWMON \u0026\u0026 !(BNXT=y \u0026\u0026 HWMON=m)\ndrivers/net/ethernet/broadcom/Kconfig-254-\thelp\n--\ndrivers/net/ethernet/broadcom/Makefile=18=obj-$(CONFIG_SYSTEMPORT) += bcmsysport.o\ndrivers/net/ethernet/broadcom/Makefile:19:obj-$(CONFIG_BNXT) += bnxt/\ndrivers/net/ethernet/broadcom/Makefile-20-obj-$(CONFIG_BCMASP) += asp2/\n--\ndrivers/net/ethernet/broadcom/bnxt/Makefile-1-# SPDX-License-Identifier: GPL-2.0-only\ndrivers/net/ethernet/broadcom/bnxt/Makefile:2:obj-$(CONFIG_BNXT) += bnxt_en.o\ndrivers/net/ethernet/broadcom/bnxt/Makefile-3-\ndrivers/net/ethernet/broadcom/bnxt/Makefile=4=bnxt_en-y := bnxt.o bnxt_hwrm.o bnxt_sriov.o bnxt_ethtool.o bnxt_dcb.o bnxt_ulp.o bnxt_xdp.o bnxt_ptp.o bnxt_vfr.o bnxt_devlink.o bnxt_dim.o bnxt_coredump...\ndrivers/net/ethernet/broadcom/bnxt/Makefile:5:bnxt_en-$(CONFIG_BNXT_FLOWER_OFFLOAD) += bnxt_tc.o\ndrivers/net/ethernet/broadcom/bnxt/Makefile-6-bnxt_en-$(CONFIG_DEBUG_FS) += bnxt_debugfs.o\ndrivers/net/ethernet/broadcom/bnxt/Makefile:7:bnxt_en-$(CONFIG_BNXT_HWMON) += bnxt_hwmon.o\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-79-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:80:#define BNXT_TX_TIMEOUT\t\t(5 * HZ)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:81:#define BNXT_DEF_MSG_ENABLE\t(NETIF_MSG_DRV | NETIF_MSG_HW | \\\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-82-\t\t\t\t NETIF_MSG_TX_ERR)\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=86=MODULE_DESCRIPTION(\"Broadcom NetXtreme network driver\");\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-87-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:88:#define BNXT_RX_OFFSET (NET_SKB_PAD + NET_IP_ALIGN)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:89:#define BNXT_RX_DMA_OFFSET NET_SKB_PAD\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-90-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:91:#define BNXT_TX_PUSH_THRESH 164\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-92-\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=150=static const struct pci_device_id bnxt_pci_tbl[] = {\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-200-\t{ PCI_VDEVICE(BROADCOM, 0xd804), .driver_data = BCM58804 },\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:201:#ifdef CONFIG_BNXT_SRIOV\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-202-\t{ PCI_VDEVICE(BROADCOM, 0x1606), .driver_data = NETXTREME_E_VF },\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=258=const u16 bnxt_bstore_to_trace[] = {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:259:\t[BNXT_CTX_SRT]\t\t= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_SRT_TRACE,\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:260:\t[BNXT_CTX_SRT2]\t\t= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_SRT2_TRACE,\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:261:\t[BNXT_CTX_CRT]\t\t= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_CRT_TRACE,\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:262:\t[BNXT_CTX_CRT2]\t\t= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_CRT2_TRACE,\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:263:\t[BNXT_CTX_RIGP0]\t= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_RIGP0_TRACE,\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:264:\t[BNXT_CTX_L2HWRM]\t= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_L2_HWRM_TRACE,\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:265:\t[BNXT_CTX_REHWRM]\t= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_ROCE_HWRM_TRACE,\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:266:\t[BNXT_CTX_CA0]\t\t= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_CA0_TRACE,\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:267:\t[BNXT_CTX_CA1]\t\t= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_CA1_TRACE,\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:268:\t[BNXT_CTX_CA2]\t\t= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_CA2_TRACE,\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:269:\t[BNXT_CTX_RIGP1]\t= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_RIGP1_TRACE,\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:270:\t[BNXT_CTX_KONG]\t\t= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_AFM_KONG_HWRM_TRACE,\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:271:\t[BNXT_CTX_QPC]\t\t= DBG_LOG_BUFFER_FLUSH_REQ_TYPE_ERR_QPC_TRACE,\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-272-};\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=274=static struct workqueue_struct *bnxt_pf_wq;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-275-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:276:#define BNXT_IPV6_MASK_ALL {{{ 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, \\\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-277-\t\t\t       0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff }}}\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:278:#define BNXT_IPV6_MASK_NONE {{{ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 }}}\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-279-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:280:const struct bnxt_flow_masks BNXT_FLOW_MASK_NONE = {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-281-\t.ports = {\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-286-\t\t.v6addrs = {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:287:\t\t\t.src = BNXT_IPV6_MASK_NONE,\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:288:\t\t\t.dst = BNXT_IPV6_MASK_NONE,\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-289-\t\t},\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-292-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:293:const struct bnxt_flow_masks BNXT_FLOW_IPV6_MASK_ALL = {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-294-\t.ports = {\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-299-\t\t.v6addrs = {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:300:\t\t\t.src = BNXT_IPV6_MASK_ALL,\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:301:\t\t\t.dst = BNXT_IPV6_MASK_ALL,\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-302-\t\t},\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-305-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:306:const struct bnxt_flow_masks BNXT_FLOW_IPV4_MASK_ALL = {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-307-\t.ports = {\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=319=static bool bnxt_vf_pciid(enum board_idx idx)\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-330-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:331:#define BNXT_DB_CQ(db, idx)\t\t\t\t\t\t\\\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-332-\twritel(DB_CP_FLAGS | DB_RING_IDX(db, idx), (db)-\u003edoorbell)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-333-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:334:#define BNXT_DB_NQ_P5(db, idx)\t\t\t\t\t\t\\\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-335-\tbnxt_writeq(bp, (db)-\u003edb_key64 | DBR_TYPE_NQ | DB_RING_IDX(db, idx),\\\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-337-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:338:#define BNXT_DB_NQ_P7(db, idx)\t\t\t\t\t\t\\\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-339-\tbnxt_writeq(bp, (db)-\u003edb_key64 | DBR_TYPE_NQ_MASK |\t\t\\\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-341-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:342:#define BNXT_DB_CQ_ARM(db, idx)\t\t\t\t\t\t\\\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-343-\twritel(DB_CP_REARM_FLAGS | DB_RING_IDX(db, idx), (db)-\u003edoorbell)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-344-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:345:#define BNXT_DB_NQ_ARM_P5(db, idx)\t\t\t\t\t\\\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-346-\tbnxt_writeq(bp, (db)-\u003edb_key64 | DBR_TYPE_NQ_ARM |\t\t\\\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=349=static void bnxt_db_nq(struct bnxt *bp, struct bnxt_db_info *db, u32 idx)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-350-{\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:351:\tif (bp-\u003eflags \u0026 BNXT_FLAG_CHIP_P7)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:352:\t\tBNXT_DB_NQ_P7(db, idx);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:353:\telse if (bp-\u003eflags \u0026 BNXT_FLAG_CHIP_P5_PLUS)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:354:\t\tBNXT_DB_NQ_P5(db, idx);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-355-\telse\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:356:\t\tBNXT_DB_CQ(db, idx);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-357-}\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=359=static void bnxt_db_nq_arm(struct bnxt *bp, struct bnxt_db_info *db, u32 idx)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-360-{\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:361:\tif (bp-\u003eflags \u0026 BNXT_FLAG_CHIP_P5_PLUS)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:362:\t\tBNXT_DB_NQ_ARM_P5(db, idx);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-363-\telse\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:364:\t\tBNXT_DB_CQ_ARM(db, idx);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-365-}\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=367=static void bnxt_db_cq(struct bnxt *bp, struct bnxt_db_info *db, u32 idx)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-368-{\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:369:\tif (bp-\u003eflags \u0026 BNXT_FLAG_CHIP_P5_PLUS)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-370-\t\tbnxt_writeq(bp, db-\u003edb_key64 | DBR_TYPE_CQ_ARMALL |\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-372-\telse\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:373:\t\tBNXT_DB_CQ(db, idx);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-374-}\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=376=static void bnxt_queue_fw_reset_work(struct bnxt *bp, unsigned long delay)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-377-{\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:378:\tif (!(test_bit(BNXT_STATE_IN_FW_RESET, \u0026bp-\u003estate)))\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-379-\t\treturn;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-380-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:381:\tif (BNXT_PF(bp))\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-382-\t\tqueue_delayed_work(bnxt_pf_wq, \u0026bp-\u003efw_reset_task, delay);\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=387=static void __bnxt_queue_sp_work(struct bnxt *bp)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-388-{\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:389:\tif (BNXT_PF(bp))\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-390-\t\tqueue_work(bnxt_pf_wq, \u0026bp-\u003esp_task);\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=401=static void bnxt_sched_reset_rxr(struct bnxt *bp, struct bnxt_rx_ring_info *rxr)\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-404-\t\trxr-\u003ebnapi-\u003ein_reset = true;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:405:\t\tif (bp-\u003eflags \u0026 BNXT_FLAG_CHIP_P5_PLUS)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:406:\t\t\tset_bit(BNXT_RESET_TASK_SP_EVENT, \u0026bp-\u003esp_event);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-407-\t\telse\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:408:\t\t\tset_bit(BNXT_RST_RING_SP_EVENT, \u0026bp-\u003esp_event);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-409-\t\t__bnxt_queue_sp_work(bp);\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=414=void bnxt_sched_reset_txr(struct bnxt *bp, struct bnxt_tx_ring_info *txr,\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-426-\tbnapi-\u003etx_fault = 1;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:427:\tbnxt_queue_sp_work(bp, BNXT_RESET_TASK_SP_EVENT);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-428-}\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=481=static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-519-\t    (skb_shinfo(skb)-\u003egso_type \u0026 SKB_GSO_UDP_L4) \u0026\u0026\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:520:\t    !(bp-\u003eflags \u0026 BNXT_FLAG_UDP_GSO_CAP)) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-521-\t\tint rc = bnxt_sw_udp_gso_xmit(bp, txr, txq, skb);\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-573-\t    ptp-\u003etx_tstamp_en) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:574:\t\tif (bp-\u003efw_cap \u0026 BNXT_FW_CAP_TX_TS_CMP) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-575-\t\t\tlflags |= cpu_to_le32(TX_BD_FLAGS_STAMP);\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-674-normal_tx:\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:675:\tif (length \u003c BNXT_MIN_PKT_SIZE) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:676:\t\tpad = BNXT_MIN_PKT_SIZE - length;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-677-\t\tif (skb_pad(skb, pad))\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-679-\t\t\tgoto tx_kick_pending;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:680:\t\tlength = BNXT_MIN_PKT_SIZE;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-681-\t}\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-820-tx_kick_pending:\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:821:\tif (BNXT_TX_PTP_IS_SET(lflags)) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-822-\t\ttxr-\u003etx_buf_ring[RING_TX(bp, txr-\u003etx_prod)].is_ts_pkt = 0;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-823-\t\tatomic64_inc(\u0026bp-\u003eptp_cfg-\u003estats.ts_err);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:824:\t\tif (!(bp-\u003efw_cap \u0026 BNXT_FW_CAP_TX_TS_CMP))\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-825-\t\t\t/* set SKB to err so PTP worker will clean up */\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=836=static bool __bnxt_tx_int(struct bnxt *bp, struct bnxt_tx_ring_info *txr,\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-864-\t\tis_ts_pkt = tx_buf-\u003eis_ts_pkt;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:865:\t\tif (is_ts_pkt \u0026\u0026 (bp-\u003efw_cap \u0026 BNXT_FW_CAP_TX_TS_CMP)) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-866-\t\t\trc = true;\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-907-\t\t\tWRITE_ONCE(txr-\u003etx_inline_cons, inline_cons);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:908:\t\t\tif (head_buf-\u003eis_sw_gso == BNXT_SW_GSO_LAST) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-909-\t\t\t\ttso_dma_map_complete(\u0026pdev-\u003edev,\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-919-\t\tif (unlikely(is_ts_pkt)) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:920:\t\t\tif (BNXT_CHIP_P5(bp)) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-921-\t\t\t\t/* PTP worker takes ownership of the skb */\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-936-\t\t\t\t   bnxt_tx_avail(bp, txr), bp-\u003etx_wake_thresh,\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:937:\t\t\t\t   READ_ONCE(txr-\u003edev_state) == BNXT_DEV_STATE_CLOSING);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-938-\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=942=static void bnxt_tx_int(struct bnxt *bp, struct bnxt_napi *bnapi, int budget)\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-952-\tif (!more)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:953:\t\tbnapi-\u003eevents \u0026= ~BNXT_TX_CMP_EVENT;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-954-}\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=1019=int bnxt_alloc_rx_data(struct bnxt *bp, struct bnxt_rx_ring_info *rxr,\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1025-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1026:\tif (BNXT_RX_PAGE_MODE(bp)) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1027-\t\tunsigned int offset;\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=1137=static void bnxt_reuse_rx_agg_bufs(struct bnxt_cp_ring_info *cpr, u16 idx,\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1147-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1148:\tif ((bp-\u003eflags \u0026 BNXT_FLAG_CHIP_P5_PLUS) \u0026\u0026 tpa)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1149-\t\tp5_tpa = true;\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=1306=static u32 __bnxt_rx_agg_netmems(struct bnxt *bp,\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1321-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1322:\tif ((bp-\u003eflags \u0026 BNXT_FLAG_CHIP_P5_PLUS) \u0026\u0026 tpa)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1323-\t\tp5_tpa = true;\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=1507=static int bnxt_discard_rx(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1522-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1523:\t\tif (bp-\u003eflags \u0026 BNXT_FLAG_CHIP_P5_PLUS)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1524-\t\t\treturn 0;\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=1599=static void bnxt_tpa_start(struct bnxt *bp, struct bnxt_rx_ring_info *rxr,\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1608-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1609:\tif (bp-\u003eflags \u0026 BNXT_FLAG_CHIP_P5_PLUS) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1610-\t\tagg_id = TPA_START_AGG_ID_P5(tpa_start);\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1659-\t\t/* RSS profiles 1 and 3 with extract code 0 for inner 4-tuple */\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1660:\t\telse if (!BNXT_CHIP_P4_PLUS(bp) \u0026\u0026\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1661-\t\t\t TPA_START_HASH_TYPE(tpa_start) == 3)\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=1719=static struct sk_buff *bnxt_gro_func_5731x(struct bnxt_tpa_info *tpa_info,\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1729-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1730:\tinner_ip_off = BNXT_TPA_INNER_L3_OFF(hdr_info);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1731:\tinner_mac_off = BNXT_TPA_INNER_L2_OFF(hdr_info);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1732:\touter_ip_off = BNXT_TPA_OUTER_L3_OFF(hdr_info);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1733-\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=1785=static struct sk_buff *bnxt_gro_func_5750x(struct bnxt_tpa_info *tpa_info,\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1793-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1794:\tinner_ip_off = BNXT_TPA_INNER_L3_OFF(hdr_info);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1795:\tinner_mac_off = BNXT_TPA_INNER_L2_OFF(hdr_info);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1796:\touter_ip_off = BNXT_TPA_OUTER_L3_OFF(hdr_info);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1797-\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1813-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1814:#define BNXT_IPV4_HDR_SIZE\t(sizeof(struct iphdr) + sizeof(struct tcphdr))\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1815:#define BNXT_IPV6_HDR_SIZE\t(sizeof(struct ipv6hdr) + sizeof(struct tcphdr))\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1816-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=1817=static struct sk_buff *bnxt_gro_func_5730x(struct bnxt_tpa_info *tpa_info,\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1830-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1831:\t\tnw_off = payload_off - BNXT_IPV4_HDR_SIZE - tcp_opt_len -\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1832-\t\t\t ETH_HLEN;\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1841-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1842:\t\tnw_off = payload_off - BNXT_IPV6_HDR_SIZE - tcp_opt_len -\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1843-\t\t\t ETH_HLEN;\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=1861=static inline struct sk_buff *bnxt_gro_skb(struct bnxt *bp,\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1882-\tskb_shinfo(skb)-\u003egso_type = tpa_info-\u003egso_type;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1883:\tif (bp-\u003eflags \u0026 BNXT_FLAG_CHIP_P5_PLUS)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1884-\t\tpayload_off = TPA_END_PAYLOAD_OFF_P5(tpa_end1);\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=1905=static inline struct sk_buff *bnxt_tpa_end(struct bnxt *bp,\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1931-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1932:\tif (bp-\u003eflags \u0026 BNXT_FLAG_CHIP_P5_PLUS) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1933-\t\tagg_id = TPA_END_AGG_ID_P5(tpa_end);\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1942-\t\ttpa_info-\u003eagg_count = 0;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1943:\t\t*event |= BNXT_AGG_EVENT;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1944-\t\tbnxt_free_agg_idx(rxr, agg_id);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1945-\t\tidx = agg_id;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1946:\t\tgro = !!(bp-\u003eflags \u0026 BNXT_FLAG_GRO);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1947-\t} else {\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1955-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:1956:\t\t\t*event |= BNXT_AGG_EVENT;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-1957-\t\t\tidx = NEXT_CMP(idx);\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2030-\tif (tpa_info-\u003evlan_valid \u0026\u0026\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2031:\t    (dev-\u003efeatures \u0026 BNXT_HW_FEATURE_VLAN_ALL_RX)) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2032-\t\t__be16 vlan_proto = htons(tpa_info-\u003emetadata \u003e\u003e\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=2084=static bool bnxt_rx_ts_valid(struct bnxt *bp, u32 flags,\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2088-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2089:\tif (BNXT_PTP_RX_TS_VALID(flags))\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2090-\t\tgoto ts_valid;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2091:\tif (!bp-\u003eptp_all_rx_tstamp || !ts || !BNXT_ALL_RX_TS_VALID(flags))\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2092-\t\treturn false;\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=2148=static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2200-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2201:\t\t*event |= BNXT_RX_EVENT;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2202-\t\tgoto next_rx_no_prod_no_len;\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2216-\t\t}\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2217:\t\t*event |= BNXT_RX_EVENT;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2218-\t\tgoto next_rx_no_prod_no_len;\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2246-\t\tcp_cons = NEXT_CMP(cp_cons);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2247:\t\t*event |= BNXT_AGG_EVENT;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2248-\t}\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2249:\t*event |= BNXT_RX_EVENT;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2250-\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2262-\t\t\tbnapi-\u003ecp_ring.sw_stats-\u003erx.rx_buf_errors++;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2263:\t\t\tif (!(bp-\u003eflags \u0026 BNXT_FLAG_CHIP_P5_PLUS) \u0026\u0026\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2264:\t\t\t    !(bp-\u003efw_cap \u0026 BNXT_FW_CAP_RING_MONITOR)) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2265-\t\t\t\tnetdev_warn_once(bp-\u003edev, \"RX buffer error %x\\n\",\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2378-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2379:\tif (skb-\u003edev-\u003efeatures \u0026 BNXT_HW_FEATURE_VLAN_ALL_RX) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2380-\t\tskb = bnxt_rx_vlan(skb, cmp_type, rxcmp, rxcmp1);\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2398-\tif (bnxt_rx_ts_valid(bp, flags, rxcmp1, \u0026cmpl_ts)) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2399:\t\tif (bp-\u003eflags \u0026 BNXT_FLAG_CHIP_P5_PLUS) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2400-\t\t\tu64 ns, ts;\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=2482=u32 bnxt_fw_health_readl(struct bnxt *bp, int reg_idx)\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2487-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2488:\treg_type = BNXT_FW_HEALTH_REG_TYPE(reg);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2489:\treg_off = BNXT_FW_HEALTH_REG_OFF(reg);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2490-\tswitch (reg_type) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2491:\tcase BNXT_FW_HEALTH_REG_TYPE_CFG:\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2492-\t\tpci_read_config_dword(bp-\u003epdev, reg_off, \u0026val);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2493-\t\tbreak;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2494:\tcase BNXT_FW_HEALTH_REG_TYPE_GRC:\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2495-\t\treg_off = fw_health-\u003emapped_regs[reg_idx];\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2496-\t\tfallthrough;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2497:\tcase BNXT_FW_HEALTH_REG_TYPE_BAR0:\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2498-\t\tval = readl(bp-\u003ebar0 + reg_off);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2499-\t\tbreak;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2500:\tcase BNXT_FW_HEALTH_REG_TYPE_BAR1:\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2501-\t\tval = readl(bp-\u003ebar1 + reg_off);\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2503-\t}\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2504:\tif (reg_idx == BNXT_FW_RESET_INPROG_REG)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2505-\t\tval \u0026= fw_health-\u003efw_reset_inprog_reg_mask;\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=2524=static u16 bnxt_get_force_speed(struct bnxt_link_info *link_info)\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2527-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2528:\tif (bp-\u003ephy_flags \u0026 BNXT_PHY_FL_SPEEDS2)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2529-\t\treturn link_info-\u003eforce_link_speed2;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2530:\tif (link_info-\u003ereq_signal_mode == BNXT_SIG_MODE_PAM4)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2531-\t\treturn link_info-\u003eforce_pam4_link_speed;\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=2535=static void bnxt_set_force_speed(struct bnxt_link_info *link_info)\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2538-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2539:\tif (bp-\u003ephy_flags \u0026 BNXT_PHY_FL_SPEEDS2) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2540-\t\tlink_info-\u003ereq_link_speed = link_info-\u003eforce_link_speed2;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2541:\t\tlink_info-\u003ereq_signal_mode = BNXT_SIG_MODE_NRZ;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2542-\t\tswitch (link_info-\u003ereq_link_speed) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2543:\t\tcase BNXT_LINK_SPEED_50GB_PAM4:\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2544:\t\tcase BNXT_LINK_SPEED_100GB_PAM4:\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2545:\t\tcase BNXT_LINK_SPEED_200GB_PAM4:\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2546:\t\tcase BNXT_LINK_SPEED_400GB_PAM4:\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2547:\t\t\tlink_info-\u003ereq_signal_mode = BNXT_SIG_MODE_PAM4;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2548-\t\t\tbreak;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2549:\t\tcase BNXT_LINK_SPEED_100GB_PAM4_112:\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2550:\t\tcase BNXT_LINK_SPEED_200GB_PAM4_112:\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2551:\t\tcase BNXT_LINK_SPEED_400GB_PAM4_112:\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2552:\t\t\tlink_info-\u003ereq_signal_mode = BNXT_SIG_MODE_PAM4_112;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2553-\t\t\tbreak;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2554-\t\tdefault:\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2555:\t\t\tlink_info-\u003ereq_signal_mode = BNXT_SIG_MODE_NRZ;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2556-\t\t}\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2559-\tlink_info-\u003ereq_link_speed = link_info-\u003eforce_link_speed;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2560:\tlink_info-\u003ereq_signal_mode = BNXT_SIG_MODE_NRZ;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2561-\tif (link_info-\u003eforce_pam4_link_speed) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2562-\t\tlink_info-\u003ereq_link_speed = link_info-\u003eforce_pam4_link_speed;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2563:\t\tlink_info-\u003ereq_signal_mode = BNXT_SIG_MODE_PAM4;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2564-\t}\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=2567=static void bnxt_set_auto_speed(struct bnxt_link_info *link_info)\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2570-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2571:\tif (bp-\u003ephy_flags \u0026 BNXT_PHY_FL_SPEEDS2) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2572-\t\tlink_info-\u003eadvertising = link_info-\u003eauto_link_speeds2;\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=2579=static bool bnxt_force_speed_updated(struct bnxt_link_info *link_info)\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2582-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2583:\tif (bp-\u003ephy_flags \u0026 BNXT_PHY_FL_SPEEDS2) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2584-\t\tif (link_info-\u003ereq_link_speed != link_info-\u003eforce_link_speed2)\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2587-\t}\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2588:\tif (link_info-\u003ereq_signal_mode == BNXT_SIG_MODE_NRZ \u0026\u0026\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2589-\t    link_info-\u003ereq_link_speed != link_info-\u003eforce_link_speed)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2590-\t\treturn true;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2591:\tif (link_info-\u003ereq_signal_mode == BNXT_SIG_MODE_PAM4 \u0026\u0026\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2592-\t    link_info-\u003ereq_link_speed != link_info-\u003eforce_pam4_link_speed)\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=2597=static bool bnxt_auto_speed_updated(struct bnxt_link_info *link_info)\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2600-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2601:\tif (bp-\u003ephy_flags \u0026 BNXT_PHY_FL_SPEEDS2) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2602-\t\tif (link_info-\u003eadvertising != link_info-\u003eauto_link_speeds2)\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=2612=bool bnxt_bs_trace_avail(struct bnxt *bp, u16 type)\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2615-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2616:\treturn (flags \u0026 BNXT_CTX_MEM_TYPE_VALID) \u0026\u0026\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2617:\t\t((flags \u0026 BNXT_CTX_MEM_FW_TRACE) ||\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2618:\t\t (flags \u0026 BNXT_CTX_MEM_FW_BIN_TRACE));\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2619-}\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=2621=static void bnxt_bs_trace_init(struct bnxt *bp, struct bnxt_ctx_mem_type *ctxm)\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2633-\tmem_size = ctxm-\u003emax_entries * ctxm-\u003eentry_size;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2634:\trem_bytes = mem_size % BNXT_PAGE_SIZE;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2635:\tpages = DIV_ROUND_UP(mem_size, BNXT_PAGE_SIZE);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2636-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2637-\tlast_pg = (pages - 1) \u0026 (MAX_CTX_PAGES - 1);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2638:\tmagic_byte_offset = (rem_bytes ? rem_bytes : BNXT_PAGE_SIZE) - 1;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2639-\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2652-\tbs_trace-\u003emagic_byte += magic_byte_offset;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2653:\t*bs_trace-\u003emagic_byte = BNXT_TRACE_BUF_MAGIC_BYTE;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2654-}\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2655-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2656:#define BNXT_EVENT_BUF_PRODUCER_TYPE(data1)\t\t\t\t\\\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2657-\t(((data1) \u0026 ASYNC_EVENT_CMPL_DBG_BUF_PRODUCER_EVENT_DATA1_TYPE_MASK) \u003e\u003e\\\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2659-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2660:#define BNXT_EVENT_BUF_PRODUCER_OFFSET(data2)\t\t\t\t\\\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2661-\t(((data2) \u0026\t\t\t\t\t\t\t\\\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2664-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2665:#define BNXT_EVENT_THERMAL_CURRENT_TEMP(data2)\t\t\t\t\\\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2666-\t((data2) \u0026\t\t\t\t\t\t\t\\\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2668-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2669:#define BNXT_EVENT_THERMAL_THRESHOLD_TEMP(data2)\t\t\t\\\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2670-\t(((data2) \u0026\t\t\t\t\t\t\t\\\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=2684=static bool bnxt_event_error_report(struct bnxt *bp, u32 data1, u32 data2)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2685-{\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2686:\tu32 err_type = BNXT_EVENT_ERROR_REPORT_TYPE(data1);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2687-\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2690-\t\tnetdev_err(bp-\u003edev, \"1PPS: Received invalid signal on pin%lu from the external source. Please fix the signal and reconfigure the pin\\n\",\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2691:\t\t\t   BNXT_EVENT_INVALID_SIGNAL_DATA(data2));\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2692-\t\tbreak;\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2730-\t\tnetdev_warn(bp-\u003edev, \"Temperature (In Celsius), Current: %lu, threshold: %lu\\n\",\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2731:\t\t\t    BNXT_EVENT_THERMAL_CURRENT_TEMP(data2),\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2732:\t\t\t    BNXT_EVENT_THERMAL_THRESHOLD_TEMP(data2));\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2733-\t\tif (notify) {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2734-\t\t\tbp-\u003ethermal_threshold_type = type;\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2735:\t\t\tset_bit(BNXT_THERMAL_THRESHOLD_SP_EVENT, \u0026bp-\u003esp_event);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2736-\t\t\treturn true;\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2750-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:2751:#define BNXT_GET_EVENT_PORT(data)\t\\\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-2752-\t((data) \u0026\t\t\t\\\n\n"
}

7/2 2026/09/29 05:52 llm patch-evaluator
0m Model:
gemini-3.8-flash

Args:
null

Results:
null

Tokens:
							input: 29799
							output: 40
							thoughts: 677
						

Thoughts:
**Analyzing the Fuzzing Worthiness of a `bnxt` Driver Patch**

Okay, let's dive into this. The task is to determine if a patch related to the `bnxt` driver, located at `drivers/net/ethernet/broadcom/bnxt/bnxt.c`, warrants fuzzing. My immediate thought is, "Is this driver even relevant in a fuzzing context?" Let's start with the basics. It's a PCI driver for Broadcom NetXtreme NICs. The initial question then becomes, does QEMU, our primary emulation environment, support these NICs? No, it doesn't. QEMU emulates other NICs but not Broadcom NetXtreme-C or E (BCM57xxx).

However, hold on a second. We have to consider the criteria. The instructions specify scenarios where we *shouldn't* fuzz (`WorthFuzzing=false`), specifically mentioning "Structurally unreachable hardware: Vendor-specific PCIe switches, SmartNICs..." which fits `bnxt` if we're only considering physical PCI cards. *But* it also provides positive criteria: "Reachable Drivers and Protocols: Drivers accessible via virtual buses (virtio, USB gadget, loopback, netlink, binder, sockets, etc.)".

Now, the crucial question. Is `bnxt` *exclusively* a physical PCI device driver, or does it have any potential reach via a virtual bus or software emulation? I need to ascertain if there's any part of this driver or its related infrastructure that might be exposed or exercised through a virtual environment. I need to search the kernel code for any evidence that the `bnxt` driver could be attached to a virtual bus or software construct. A search for `bnxt_pci_driver` or other pertinent references is needed to determine the answer to the above question.



8/2 2026/09/29 05:52 tool grepper
0m Args:
{
  "Expression": "bnxt_pci_driver",
  "PathPrefix": "drivers/net/ethernet/broadcom/bnxt/"
}

Results:
{
  "Output": "drivers/net/ethernet/broadcom/bnxt/bnxt.c=17865=static const struct pci_error_handlers bnxt_err_handler = {\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-17870-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:17871:static struct pci_driver bnxt_pci_driver = {\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-17872-\t.name\t\t= DRV_MODULE_NAME,\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=17884=static int __init bnxt_init(void)\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-17888-\tbnxt_debug_init();\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:17889:\terr = pci_register_driver(\u0026bnxt_pci_driver);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-17890-\tif (err) {\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c=17898=static void __exit bnxt_exit(void)\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-17899-{\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:17900:\tpci_unregister_driver(\u0026bnxt_pci_driver);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-17901-\tif (bnxt_pf_wq)\n"
}

9/2 2026/09/29 05:52 llm patch-evaluator
0m Model:
gemini-3.8-flash

Args:
null

Results:
null

Tokens:
							input: 31355
							output: 41
							thoughts: 46
						

Thoughts:
Okay, here's my understanding, formatted as requested:

**BNXT Driver Initialization**

Alright, so the `bnxt` driver is getting its hooks into the system through `pci_register_driver`.  Specifically, the `&bnxt_pci_driver` pointer is what's being registered, which makes sense. That's the standard entry point for a PCI driver. Now, the next logical step is to dive into the `bnxt.c` file and zero in on how the PCI device IDs are handled. I need to understand which devices are supported. I'll need to examine the `bnxt_pci_tbl` table. That's where the vendor and device ID pairs are probably defined, along with any relevant driver-specific data for those devices. Let's see what's in there.



10/2 2026/09/29 05:52 tool grepper
0m Args:
{
  "Expression": "MODULE_DEVICE_TABLE\\(pci",
  "PathPrefix": "drivers/net/ethernet/broadcom/bnxt/"
}

Results:
{
  "Output": "drivers/net/ethernet/broadcom/bnxt/bnxt.c=150=static const struct pci_device_id bnxt_pci_tbl[] = {\n--\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-228-\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c:229:MODULE_DEVICE_TABLE(pci, bnxt_pci_tbl);\ndrivers/net/ethernet/broadcom/bnxt/bnxt.c-230-\n"
}

11/2 2026/09/29 05:52 llm patch-evaluator
0m Model:
gemini-3.8-flash

Args:
null

Results:
null

Tokens:
							input: 31761
							output: 45
							thoughts: 8
						

Thoughts:
Okay, here's my interpretation of "Let's see the device table," rendered in the first person, assuming a technical expert is the recipient.

**Initial Assessment: Navigating the Device Landscape**

Alright, let's take a look at the device table. That usually means I need to quickly get a lay of the land. I need to understand what specific hardware or software systems are in scope. I'm immediately trying to assess the context; what problem am I trying to solve? Or, am I just performing an audit? I'm asking myself, *what are the key attributes of these devices that I need to know?* Is it their model, operating system, network configuration, associated software, or some other factor entirely? Knowing this will help me focus on the relevant columns and data in the table. This first "look" gives me a broad overview and lets me start forming hypotheses about device behavior and interdependencies.



12/2 2026/09/29 05:52 tool read-file
0m Args:
{
  "File": "drivers/net/ethernet/broadcom/bnxt/bnxt.c",
  "FirstLine": 150,
  "LineCount": 80
}

Results:
{
  "Contents": " 150:\tstatic const struct pci_device_id bnxt_pci_tbl[] = {\n 151:\t\t{ PCI_VDEVICE(BROADCOM, 0x1604), .driver_data = BCM5745x_NPAR },\n 152:\t\t{ PCI_VDEVICE(BROADCOM, 0x1605), .driver_data = BCM5745x_NPAR },\n 153:\t\t{ PCI_VDEVICE(BROADCOM, 0x1614), .driver_data = BCM57454 },\n 154:\t\t{ PCI_VDEVICE(BROADCOM, 0x16c0), .driver_data = BCM57417_NPAR },\n 155:\t\t{ PCI_VDEVICE(BROADCOM, 0x16c8), .driver_data = BCM57301 },\n 156:\t\t{ PCI_VDEVICE(BROADCOM, 0x16c9), .driver_data = BCM57302 },\n 157:\t\t{ PCI_VDEVICE(BROADCOM, 0x16ca), .driver_data = BCM57304 },\n 158:\t\t{ PCI_VDEVICE(BROADCOM, 0x16cc), .driver_data = BCM57417_NPAR },\n 159:\t\t{ PCI_VDEVICE(BROADCOM, 0x16cd), .driver_data = BCM58700 },\n 160:\t\t{ PCI_VDEVICE(BROADCOM, 0x16ce), .driver_data = BCM57311 },\n 161:\t\t{ PCI_VDEVICE(BROADCOM, 0x16cf), .driver_data = BCM57312 },\n 162:\t\t{ PCI_VDEVICE(BROADCOM, 0x16d0), .driver_data = BCM57402 },\n 163:\t\t{ PCI_VDEVICE(BROADCOM, 0x16d1), .driver_data = BCM57404 },\n 164:\t\t{ PCI_VDEVICE(BROADCOM, 0x16d2), .driver_data = BCM57406 },\n 165:\t\t{ PCI_VDEVICE(BROADCOM, 0x16d4), .driver_data = BCM57402_NPAR },\n 166:\t\t{ PCI_VDEVICE(BROADCOM, 0x16d5), .driver_data = BCM57407 },\n 167:\t\t{ PCI_VDEVICE(BROADCOM, 0x16d6), .driver_data = BCM57412 },\n 168:\t\t{ PCI_VDEVICE(BROADCOM, 0x16d7), .driver_data = BCM57414 },\n 169:\t\t{ PCI_VDEVICE(BROADCOM, 0x16d8), .driver_data = BCM57416 },\n 170:\t\t{ PCI_VDEVICE(BROADCOM, 0x16d9), .driver_data = BCM57417 },\n 171:\t\t{ PCI_VDEVICE(BROADCOM, 0x16de), .driver_data = BCM57412_NPAR },\n 172:\t\t{ PCI_VDEVICE(BROADCOM, 0x16df), .driver_data = BCM57314 },\n 173:\t\t{ PCI_VDEVICE(BROADCOM, 0x16e2), .driver_data = BCM57417_SFP },\n 174:\t\t{ PCI_VDEVICE(BROADCOM, 0x16e3), .driver_data = BCM57416_SFP },\n 175:\t\t{ PCI_VDEVICE(BROADCOM, 0x16e7), .driver_data = BCM57404_NPAR },\n 176:\t\t{ PCI_VDEVICE(BROADCOM, 0x16e8), .driver_data = BCM57406_NPAR },\n 177:\t\t{ PCI_VDEVICE(BROADCOM, 0x16e9), .driver_data = BCM57407_SFP },\n 178:\t\t{ PCI_VDEVICE(BROADCOM, 0x16ea), .driver_data = BCM57407_NPAR },\n 179:\t\t{ PCI_VDEVICE(BROADCOM, 0x16eb), .driver_data = BCM57412_NPAR },\n 180:\t\t{ PCI_VDEVICE(BROADCOM, 0x16ec), .driver_data = BCM57414_NPAR },\n 181:\t\t{ PCI_VDEVICE(BROADCOM, 0x16ed), .driver_data = BCM57414_NPAR },\n 182:\t\t{ PCI_VDEVICE(BROADCOM, 0x16ee), .driver_data = BCM57416_NPAR },\n 183:\t\t{ PCI_VDEVICE(BROADCOM, 0x16ef), .driver_data = BCM57416_NPAR },\n 184:\t\t{ PCI_VDEVICE(BROADCOM, 0x16f0), .driver_data = BCM58808 },\n 185:\t\t{ PCI_VDEVICE(BROADCOM, 0x16f1), .driver_data = BCM57452 },\n 186:\t\t{ PCI_VDEVICE(BROADCOM, 0x1750), .driver_data = BCM57508 },\n 187:\t\t{ PCI_VDEVICE(BROADCOM, 0x1751), .driver_data = BCM57504 },\n 188:\t\t{ PCI_VDEVICE(BROADCOM, 0x1752), .driver_data = BCM57502 },\n 189:\t\t{ PCI_VDEVICE(BROADCOM, 0x1760), .driver_data = BCM57608 },\n 190:\t\t{ PCI_VDEVICE(BROADCOM, 0x1761), .driver_data = BCM57604 },\n 191:\t\t{ PCI_VDEVICE(BROADCOM, 0x1762), .driver_data = BCM57602 },\n 192:\t\t{ PCI_VDEVICE(BROADCOM, 0x1763), .driver_data = BCM57601 },\n 193:\t\t{ PCI_VDEVICE(BROADCOM, 0x1800), .driver_data = BCM57502_NPAR },\n 194:\t\t{ PCI_VDEVICE(BROADCOM, 0x1801), .driver_data = BCM57504_NPAR },\n 195:\t\t{ PCI_VDEVICE(BROADCOM, 0x1802), .driver_data = BCM57508_NPAR },\n 196:\t\t{ PCI_VDEVICE(BROADCOM, 0x1803), .driver_data = BCM57502_NPAR },\n 197:\t\t{ PCI_VDEVICE(BROADCOM, 0x1804), .driver_data = BCM57504_NPAR },\n 198:\t\t{ PCI_VDEVICE(BROADCOM, 0x1805), .driver_data = BCM57508_NPAR },\n 199:\t\t{ PCI_VDEVICE(BROADCOM, 0xd802), .driver_data = BCM58802 },\n 200:\t\t{ PCI_VDEVICE(BROADCOM, 0xd804), .driver_data = BCM58804 },\n 201:\t#ifdef CONFIG_BNXT_SRIOV\n 202:\t\t{ PCI_VDEVICE(BROADCOM, 0x1606), .driver_data = NETXTREME_E_VF },\n 203:\t\t{ PCI_VDEVICE(BROADCOM, 0x1607), .driver_data = NETXTREME_E_VF_HV },\n 204:\t\t{ PCI_VDEVICE(BROADCOM, 0x1608), .driver_data = NETXTREME_E_VF_HV },\n 205:\t\t{ PCI_VDEVICE(BROADCOM, 0x1609), .driver_data = NETXTREME_E_VF },\n 206:\t\t{ PCI_VDEVICE(BROADCOM, 0x16bd), .driver_data = NETXTREME_E_VF_HV },\n 207:\t\t{ PCI_VDEVICE(BROADCOM, 0x16c1), .driver_data = NETXTREME_E_VF },\n 208:\t\t{ PCI_VDEVICE(BROADCOM, 0x16c2), .driver_data = NETXTREME_C_VF_HV },\n 209:\t\t{ PCI_VDEVICE(BROADCOM, 0x16c3), .driver_data = NETXTREME_C_VF_HV },\n 210:\t\t{ PCI_VDEVICE(BROADCOM, 0x16c4), .driver_data = NETXTREME_E_VF_HV },\n 211:\t\t{ PCI_VDEVICE(BROADCOM, 0x16c5), .driver_data = NETXTREME_E_VF_HV },\n 212:\t\t{ PCI_VDEVICE(BROADCOM, 0x16cb), .driver_data = NETXTREME_C_VF },\n 213:\t\t{ PCI_VDEVICE(BROADCOM, 0x16d3), .driver_data = NETXTREME_E_VF },\n 214:\t\t{ PCI_VDEVICE(BROADCOM, 0x16dc), .driver_data = NETXTREME_E_VF },\n 215:\t\t{ PCI_VDEVICE(BROADCOM, 0x16e1), .driver_data = NETXTREME_C_VF },\n 216:\t\t{ PCI_VDEVICE(BROADCOM, 0x16e5), .driver_data = NETXTREME_C_VF },\n 217:\t\t{ PCI_VDEVICE(BROADCOM, 0x16e6), .driver_data = NETXTREME_C_VF_HV },\n 218:\t\t{ PCI_VDEVICE(BROADCOM, 0x1806), .driver_data = NETXTREME_E_P5_VF },\n 219:\t\t{ PCI_VDEVICE(BROADCOM, 0x1807), .driver_data = NETXTREME_E_P5_VF },\n 220:\t\t{ PCI_VDEVICE(BROADCOM, 0x1808), .driver_data = NETXTREME_E_P5_VF_HV },\n 221:\t\t{ PCI_VDEVICE(BROADCOM, 0x1809), .driver_data = NETXTREME_E_P5_VF_HV },\n 222:\t\t{ PCI_VDEVICE(BROADCOM, 0x1819), .driver_data = NETXTREME_E_P7_VF },\n 223:\t\t{ PCI_VDEVICE(BROADCOM, 0x181b), .driver_data = NETXTREME_E_P7_VF_HV },\n 224:\t\t{ PCI_VDEVICE(BROADCOM, 0xd800), .driver_data = NETXTREME_S_VF },\n 225:\t#endif\n 226:\t\t{ 0 }\n 227:\t};\n 228:\t\n 229:\tMODULE_DEVICE_TABLE(pci, bnxt_pci_tbl);\n"
}

13/2 2026/09/29 05:52 llm patch-evaluator
0m Model:
gemini-3.8-flash

Args:
null

Results:
null

Tokens:
							input: 34986
							output: 107
							thoughts: 567
						

Thoughts:
**My Assessment: Not Worth Fuzzing**

Okay, let's break this down systematically. I've got a `bnxt_pci_tbl` here, and those are all physical Broadcom NetXtreme PCI IDs, the BCM57xxx and BCM58xxx families. Now, the instructions are very clear: section 2 highlights "Structurally unreachable hardware" as a key factor. We're looking at things like vendor-specific PCIe switches, SmartNICs, or GPU drivers that *need* physical ASIC/PCIe cards that are not emulated in the usual QEMU setups.

The driver in question, `bnxt`, is precisely that – it's the driver for Broadcom NetXtreme-C and NetXtreme-E Ethernet controllers/SmartNICs. It's a vendor-specific PCIe NIC driver, and it relies on physical hardware that isn't replicated in standard virtualized environments. The modified files – `bnxt.c`, `bnxt.h`, and `bnxt_ethtool.c` – are all *within* the `bnxt` driver itself. These aren't touching core networking code, generic netdev code, or any virtual bus or software emulated devices. It's all strictly contained within `bnxt`.

Therefore, based on my understanding, that modified code is inherently unreachable in standard fuzzing environments like QEMU or GCE, because it needs that specific Broadcom NetXtreme hardware.

Let me make sure I'm fully compliant with the instructions: We're to return `WorthFuzzing=false` *only* if all modifications fall into the "structurally unreachable hardware" category. Yes, that's precisely what we have here. The instruction also dictates that when `WorthFuzzing` is `false`: we need to have `FocusSymbols` as an empty list or null, and the same for `EnableConfigs`. We also need concise reasoning.

Here's my rationale: the patch series solely modifies the Broadcom NetXtreme-C/E (`bnxt`) ethernet driver (`bnxt.c`, `bnxt.h`, and `bnxt_ethtool.c`). This driver specifically targets vendor-specific Broadcom PCIe NIC hardware that is not emulated within standard virtualized environments (e.g., QEMU or GCE), and consequently, the modified code is structurally unreachable for syzkaller fuzzing.

I'm ready to call `set-results` now.



14/2 2026/09/29 05:52 tool set-results
0m Args:
{
  "EnableConfigs": [],
  "FocusSymbols": [],
  "Reasoning": "The patch exclusively modifies the Broadcom NetXtreme-C/E (bnxt) driver (drivers/net/ethernet/broadcom/bnxt/). This is a vendor-specific PCIe NIC driver requiring physical hardware not emulated in standard virtualized test environments (e.g., QEMU or GCE), making the modified code structurally unreachable for fuzzing.",
  "WorthFuzzing": false
}

Results:
{
  "EnableConfigs": null,
  "FocusSymbols": [],
  "Reasoning": "The patch exclusively modifies the Broadcom NetXtreme-C/E (bnxt) driver (drivers/net/ethernet/broadcom/bnxt/). This is a vendor-specific PCIe NIC driver requiring physical hardware not emulated in standard virtualized test environments (e.g., QEMU or GCE), making the modified code structurally unreachable for fuzzing.",
  "WorthFuzzing": false
}

15/1 2026/09/29 05:52 action If
0m Args:
{
  "WorthFuzzing": false
}

Results:
null

LLM Calls Summary:
Total Calls Total Tokens Avg Tokens Total Duration (Seconds) Avg Duration (Seconds)
Tool Calls Summary:
Total Calls Total Duration (Seconds) Avg Duration (Seconds)