kernel.git/drivers/net/ethernet/intel/igb, branch linux-4.19.y

igb: Do not bring the device up after non-fatal error

2024-11-08T15:19:18Z

[ Upstream commit 330a699ecbfc9c26ec92c6310686da1230b4e7eb ] Commit 004d25060c78 ("igb: Fix igb_down hung on surprise removal") changed igb_io_error_detected() to ignore non-fatal pcie errors in order to avoid hung task that can happen when igb_down() is called multiple times. This caused an issue when processing transient non-fatal errors. igb_io_resume(), which is called after igb_io_error_detected(), assumes that device is brought down by igb_io_error_detected() if the interface is up. This resulted in panic with stacktrace below. [ T3256] igb 0000:09:00.0 haeth0: igb: haeth0 NIC Link is Down [ T292] pcieport 0000:00:1c.5: AER: Uncorrected (Non-Fatal) error received: 0000:09:00.0 [ T292] igb 0000:09:00.0: PCIe Bus Error: severity=Uncorrected (Non-Fatal), type=Transaction Layer, (Requester ID) [ T292] igb 0000:09:00.0: device [8086:1537] error status/mask=00004000/00000000 [ T292] igb 0000:09:00.0: [14] CmpltTO [ 200.105524,009][ T292] igb 0000:09:00.0: AER: TLP Header: 00000000 00000000 00000000 00000000 [ T292] pcieport 0000:00:1c.5: AER: broadcast error_detected message [ T292] igb 0000:09:00.0: Non-correctable non-fatal error reported. [ T292] pcieport 0000:00:1c.5: AER: broadcast mmio_enabled message [ T292] pcieport 0000:00:1c.5: AER: broadcast resume message [ T292] ------------[ cut here ]------------ [ T292] kernel BUG at net/core/dev.c:6539! [ T292] invalid opcode: 0000 [#1] PREEMPT SMP [ T292] RIP: 0010:napi_enable+0x37/0x40 [ T292] Call Trace: [ T292] [ T292] ? die+0x33/0x90 [ T292] ? do_trap+0xdc/0x110 [ T292] ? napi_enable+0x37/0x40 [ T292] ? do_error_trap+0x70/0xb0 [ T292] ? napi_enable+0x37/0x40 [ T292] ? napi_enable+0x37/0x40 [ T292] ? exc_invalid_op+0x4e/0x70 [ T292] ? napi_enable+0x37/0x40 [ T292] ? asm_exc_invalid_op+0x16/0x20 [ T292] ? napi_enable+0x37/0x40 [ T292] igb_up+0x41/0x150 [ T292] igb_io_resume+0x25/0x70 [ T292] report_resume+0x54/0x70 [ T292] ? report_frozen_detected+0x20/0x20 [ T292] pci_walk_bus+0x6c/0x90 [ T292] ? aer_print_port_info+0xa0/0xa0 [ T292] pcie_do_recovery+0x22f/0x380 [ T292] aer_process_err_devices+0x110/0x160 [ T292] aer_isr+0x1c1/0x1e0 [ T292] ? disable_irq_nosync+0x10/0x10 [ T292] irq_thread_fn+0x1a/0x60 [ T292] irq_thread+0xe3/0x1a0 [ T292] ? irq_set_affinity_notifier+0x120/0x120 [ T292] ? irq_affinity_notify+0x100/0x100 [ T292] kthread+0xe2/0x110 [ T292] ? kthread_complete_and_exit+0x20/0x20 [ T292] ret_from_fork+0x2d/0x50 [ T292] ? kthread_complete_and_exit+0x20/0x20 [ T292] ret_from_fork_asm+0x11/0x20 [ T292] To fix this issue igb_io_resume() checks if the interface is running and the device is not down this means igb_io_error_detected() did not bring the device down and there is no need to bring it up. Signed-off-by: Mohamed Khalfella Reviewed-by: Yuanyuan Zhong Fixes: 004d25060c78 ("igb: Fix igb_down hung on surprise removal") Reviewed-by: Simon Horman Tested-by: Pucha Himasekhar Reddy (A Contingent worker at Intel) Signed-off-by: Tony Nguyen Signed-off-by: Sasha Levin

igb: Fix not clearing TimeSync interrupts for 82580

2024-09-12T09:02:52Z

[ Upstream commit ba8cf80724dbc09825b52498e4efacb563935408 ] 82580 NICs have a hardware bug that makes it necessary to write into the TSICR (TimeSync Interrupt Cause) register to clear it: https://lore.kernel.org/all/CDCB8BE0.1EC2C%25matthew.vick@intel.com/ Add a conditional so only for 82580 we write into the TSICR register, so we don't risk losing events for other models. Without this change, when running ptp4l with an Intel 82580 card, I get the following output: > timed out while polling for tx timestamp increasing tx_timestamp_timeout or > increasing kworker priority may correct this issue, but a driver bug likely > causes it This goes away with this change. This (partially) reverts commit ee14cc9ea19b ("igb: Fix missing time sync events"). Fixes: ee14cc9ea19b ("igb: Fix missing time sync events") Closes: https://lore.kernel.org/intel-wired-lan/CAN0jFd1kO0MMtOh8N2Ztxn6f7vvDKp2h507sMryobkBKe=xk=w@mail.gmail.com/ Tested-by: Daiwei Li Suggested-by: Vinicius Costa Gomes Signed-off-by: Daiwei Li Acked-by: Vinicius Costa Gomes Reviewed-by: Kurt Kanzenbach Tested-by: Pucha Himasekhar Reddy (A Contingent worker at Intel) Signed-off-by: Tony Nguyen Signed-off-by: Sasha Levin

igb: Fix missing time sync events

2024-03-26T22:22:36Z

[ Upstream commit ee14cc9ea19ba9678177e2224a9c58cce5937c73 ] Fix "double" clearing of interrupts, which can cause external events or timestamps to be missed. The E1000_TSIRC Time Sync Interrupt Cause register can be cleared in two ways, by either reading it or by writing '1' into the specific cause bit. This is documented in section 8.16.1. The following flow was used: 1. read E1000_TSIRC into 'tsicr'; 2. handle the interrupts present into 'tsirc' and mark them in 'ack'; 3. write 'ack' into E1000_TSICR; As both (1) and (3) will clear the interrupt cause, if the same interrupt happens again between (1) and (3) it will be ignored, causing events to be missed. Remove the extra clear in (3). Fixes: 00c65578b47b ("igb: enable internal PPS for the i210") Acked-by: Richard Cochran Signed-off-by: Vinicius Costa Gomes Tested-by: Pucha Himasekhar Reddy (A Contingent worker at Intel) Signed-off-by: Tony Nguyen Signed-off-by: Sasha Levin

igb: move PEROUT and EXTTS isr logic to separate functions

2024-03-26T22:22:36Z

[ Upstream commit cf99c1dd7b7729091043374b90807c7a5f9fd9b1 ] Remove code duplication in the tsync interrupt handler function by moving this logic to separate functions. This keeps the interrupt handler readable and allows the new functions to be extended for adapter types other than i210. Signed-off-by: Ruud Bos Tested-by: Gurucharan G Signed-off-by: Tony Nguyen Stable-dep-of: ee14cc9ea19b ("igb: Fix missing time sync events") Signed-off-by: Sasha Levin

igb: Fix potential memory leak in igb_add_ethtool_nfc_entry

2023-11-08T10:22:15Z

[ Upstream commit 8c0b48e01daba5ca58f939a8425855d3f4f2ed14 ] Add check for return of igb_update_ethtool_nfc_entry so that in case of any potential errors the memory alocated for input will be freed. Fixes: 0e71def25281 ("igb: add support of RX network flow classification") Reviewed-by: Wojciech Drewek Signed-off-by: Mateusz Palczewski Tested-by: Arpana Arland (A Contingent worker at Intel) Signed-off-by: Jacob Keller Signed-off-by: David S. Miller Signed-off-by: Sasha Levin

igb: Change IGB_MIN to allow set rx/tx value between 64 and 80

2023-09-23T08:48:14Z

[ Upstream commit 6319685bdc8ad5310890add907b7c42f89302886 ] Change the minimum value of RX/TX descriptors to 64 to enable setting the rx/tx value between 64 and 80. All igb devices can use as low as 64 descriptors. This change will unify igb with other drivers. Based on commit 7b1be1987c1e ("e1000e: lower ring minimum size to 64") Fixes: 9d5c824399de ("igb: PCI-Express 82575 Gigabit Ethernet driver") Signed-off-by: Olga Zaborska Tested-by: Pucha Himasekhar Reddy (A Contingent worker at Intel) Signed-off-by: Tony Nguyen Signed-off-by: Sasha Levin

igb: disable virtualization features on 82580

2023-09-23T08:48:13Z

[ Upstream commit fa09bc40b21a33937872c4c4cf0f266ec9fa4869 ] Disable virtualization features on 82580 just as on i210/i211. This avoids that virt functions are acidentally called on 82850. Fixes: 55cac248caa4 ("igb: Add full support for 82580 devices") Signed-off-by: Corinna Vinschen Reviewed-by: Simon Horman Signed-off-by: David S. Miller Signed-off-by: Sasha Levin

igb: set max size RX buffer when store bad packet is enabled

2023-09-23T08:48:09Z

commit bb5ed01cd2428cd25b1c88a3a9cba87055eb289f upstream. Increase the RX buffer size to 3K when the SBP bit is on. The size of the RX buffer determines the number of pages allocated which may not be sufficient for receive frames larger than the set MTU size. Cc: stable@vger.kernel.org Fixes: 89eaefb61dc9 ("igb: Support RX-ALL feature flag.") Reported-by: Manfred Rudigier Signed-off-by: Radoslaw Tyl Tested-by: Arpana Arland (A Contingent worker at Intel) Signed-off-by: Tony Nguyen Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

igb: Avoid starting unnecessary workqueues

2023-08-30T14:31:54Z

[ Upstream commit b888c510f7b3d64ca75fc0f43b4a4bd1a611312f ] If ptp_clock_register() fails or CONFIG_PTP isn't enabled, avoid starting PTP related workqueues. In this way we can fix this: BUG: unable to handle page fault for address: ffffc9000440b6f8 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 100000067 P4D 100000067 PUD 1001e0067 PMD 107dc5067 PTE 0 Oops: 0000 [#1] PREEMPT SMP [...] Workqueue: events igb_ptp_overflow_check RIP: 0010:igb_rd32+0x1f/0x60 [...] Call Trace: igb_ptp_read_82580+0x20/0x50 timecounter_read+0x15/0x60 igb_ptp_overflow_check+0x1a/0x50 process_one_work+0x1cb/0x3c0 worker_thread+0x53/0x3f0 ? rescuer_thread+0x370/0x370 kthread+0x142/0x160 ? kthread_associate_blkcg+0xc0/0xc0 ret_from_fork+0x1f/0x30 Fixes: 1f6e8178d685 ("igb: Prevent dropped Tx timestamps via work items and interrupts.") Fixes: d339b1331616 ("igb: add PTP Hardware Clock code") Signed-off-by: Alessio Igor Bogani Tested-by: Arpana Arland (A Contingent worker at Intel) Signed-off-by: Tony Nguyen Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20230821171927.2203644-1-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin

igb: Fix igb_down hung on surprise removal

2023-08-11T09:45:25Z

[ Upstream commit 004d25060c78fc31f66da0fa439c544dda1ac9d5 ] In a setup where a Thunderbolt hub connects to Ethernet and a display through USB Type-C, users may experience a hung task timeout when they remove the cable between the PC and the Thunderbolt hub. This is because the igb_down function is called multiple times when the Thunderbolt hub is unplugged. For example, the igb_io_error_detected triggers the first call, and the igb_remove triggers the second call. The second call to igb_down will block at napi_synchronize. Here's the call trace: __schedule+0x3b0/0xddb ? __mod_timer+0x164/0x5d3 schedule+0x44/0xa8 schedule_timeout+0xb2/0x2a4 ? run_local_timers+0x4e/0x4e msleep+0x31/0x38 igb_down+0x12c/0x22a [igb 6615058754948bfde0bf01429257eb59f13030d4] __igb_close+0x6f/0x9c [igb 6615058754948bfde0bf01429257eb59f13030d4] igb_close+0x23/0x2b [igb 6615058754948bfde0bf01429257eb59f13030d4] __dev_close_many+0x95/0xec dev_close_many+0x6e/0x103 unregister_netdevice_many+0x105/0x5b1 unregister_netdevice_queue+0xc2/0x10d unregister_netdev+0x1c/0x23 igb_remove+0xa7/0x11c [igb 6615058754948bfde0bf01429257eb59f13030d4] pci_device_remove+0x3f/0x9c device_release_driver_internal+0xfe/0x1b4 pci_stop_bus_device+0x5b/0x7f pci_stop_bus_device+0x30/0x7f pci_stop_bus_device+0x30/0x7f pci_stop_and_remove_bus_device+0x12/0x19 pciehp_unconfigure_device+0x76/0xe9 pciehp_disable_slot+0x6e/0x131 pciehp_handle_presence_or_link_change+0x7a/0x3f7 pciehp_ist+0xbe/0x194 irq_thread_fn+0x22/0x4d ? irq_thread+0x1fd/0x1fd irq_thread+0x17b/0x1fd ? irq_forced_thread_fn+0x5f/0x5f kthread+0x142/0x153 ? __irq_get_irqchip_state+0x46/0x46 ? kthread_associate_blkcg+0x71/0x71 ret_from_fork+0x1f/0x30 In this case, igb_io_error_detected detaches the network interface and requests a PCIE slot reset, however, the PCIE reset callback is not being invoked and thus the Ethernet connection breaks down. As the PCIE error in this case is a non-fatal one, requesting a slot reset can be avoided. This patch fixes the task hung issue and preserves Ethernet connection by ignoring non-fatal PCIE errors. Signed-off-by: Ying Hsu Tested-by: Pucha Himasekhar Reddy (A Contingent worker at Intel) Signed-off-by: Tony Nguyen Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20230620174732.4145155-1-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin