From f77bf1ebf8ff6301ccdbc346f7b52db928f9cbf8 Mon Sep 17 00:00:00 2001 From: Michal Swiatkowski Date: Thu, 22 May 2025 10:52:06 +0200 Subject: [PATCH 1/3] idpf: return 0 size for RSS key if not supported Returning -EOPNOTSUPP from function returning u32 is leading to cast and invalid size value as a result. -EOPNOTSUPP as a size probably will lead to allocation fail. Command: ethtool -x eth0 It is visible on all devices that don't have RSS caps set. [ 136.615917] Call Trace: [ 136.615921] [ 136.615927] ? __warn+0x89/0x130 [ 136.615942] ? __alloc_frozen_pages_noprof+0x322/0x330 [ 136.615953] ? report_bug+0x164/0x190 [ 136.615968] ? handle_bug+0x58/0x90 [ 136.615979] ? exc_invalid_op+0x17/0x70 [ 136.615987] ? asm_exc_invalid_op+0x1a/0x20 [ 136.616001] ? rss_prepare_get.constprop.0+0xb9/0x170 [ 136.616016] ? __alloc_frozen_pages_noprof+0x322/0x330 [ 136.616028] __alloc_pages_noprof+0xe/0x20 [ 136.616038] ___kmalloc_large_node+0x80/0x110 [ 136.616072] __kmalloc_large_node_noprof+0x1d/0xa0 [ 136.616081] __kmalloc_noprof+0x32c/0x4c0 [ 136.616098] ? rss_prepare_get.constprop.0+0xb9/0x170 [ 136.616105] rss_prepare_get.constprop.0+0xb9/0x170 [ 136.616114] ethnl_default_doit+0x107/0x3d0 [ 136.616131] genl_family_rcv_msg_doit+0x100/0x160 [ 136.616147] genl_rcv_msg+0x1b8/0x2c0 [ 136.616156] ? __pfx_ethnl_default_doit+0x10/0x10 [ 136.616168] ? __pfx_genl_rcv_msg+0x10/0x10 [ 136.616176] netlink_rcv_skb+0x58/0x110 [ 136.616186] genl_rcv+0x28/0x40 [ 136.616195] netlink_unicast+0x19b/0x290 [ 136.616206] netlink_sendmsg+0x222/0x490 [ 136.616215] __sys_sendto+0x1fd/0x210 [ 136.616233] __x64_sys_sendto+0x24/0x30 [ 136.616242] do_syscall_64+0x82/0x160 [ 136.616252] ? __sys_recvmsg+0x83/0xe0 [ 136.616265] ? syscall_exit_to_user_mode+0x10/0x210 [ 136.616275] ? do_syscall_64+0x8e/0x160 [ 136.616282] ? __count_memcg_events+0xa1/0x130 [ 136.616295] ? count_memcg_events.constprop.0+0x1a/0x30 [ 136.616306] ? handle_mm_fault+0xae/0x2d0 [ 136.616319] ? do_user_addr_fault+0x379/0x670 [ 136.616328] ? clear_bhb_loop+0x45/0xa0 [ 136.616340] ? clear_bhb_loop+0x45/0xa0 [ 136.616349] ? clear_bhb_loop+0x45/0xa0 [ 136.616359] entry_SYSCALL_64_after_hwframe+0x76/0x7e [ 136.616369] RIP: 0033:0x7fd30ba7b047 [ 136.616376] Code: 0c 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b8 0f 1f 00 f3 0f 1e fa 80 3d bd d5 0c 00 00 41 89 ca 74 10 b8 2c 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 71 c3 55 48 83 ec 30 44 89 4c 24 2c 4c 89 44 [ 136.616381] RSP: 002b:00007ffde1796d68 EFLAGS: 00000202 ORIG_RAX: 000000000000002c [ 136.616388] RAX: ffffffffffffffda RBX: 000055d7bd89f2a0 RCX: 00007fd30ba7b047 [ 136.616392] RDX: 0000000000000028 RSI: 000055d7bd89f3b0 RDI: 0000000000000003 [ 136.616396] RBP: 00007ffde1796e10 R08: 00007fd30bb4e200 R09: 000000000000000c [ 136.616399] R10: 0000000000000000 R11: 0000000000000202 R12: 000055d7bd89f340 [ 136.616403] R13: 000055d7bd89f3b0 R14: 000055d78943f200 R15: 0000000000000000 Fixes: 02cbfba1add5 ("idpf: add ethtool callbacks") Reviewed-by: Ahmed Zaki Signed-off-by: Michal Swiatkowski Reviewed-by: Simon Horman Tested-by: Samuel Salin Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/idpf/idpf_ethtool.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/intel/idpf/idpf_ethtool.c b/drivers/net/ethernet/intel/idpf/idpf_ethtool.c index 9bdb309b668e..eaf7a2606faa 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_ethtool.c +++ b/drivers/net/ethernet/intel/idpf/idpf_ethtool.c @@ -47,7 +47,7 @@ static u32 idpf_get_rxfh_key_size(struct net_device *netdev) struct idpf_vport_user_config_data *user_config; if (!idpf_is_cap_ena_all(np->adapter, IDPF_RSS_CAPS, IDPF_CAP_RSS)) - return -EOPNOTSUPP; + return 0; user_config = &np->adapter->vport_config[np->vport_idx]->user_config; @@ -66,7 +66,7 @@ static u32 idpf_get_rxfh_indir_size(struct net_device *netdev) struct idpf_vport_user_config_data *user_config; if (!idpf_is_cap_ena_all(np->adapter, IDPF_RSS_CAPS, IDPF_CAP_RSS)) - return -EOPNOTSUPP; + return 0; user_config = &np->adapter->vport_config[np->vport_idx]->user_config; From b2beb5bb2cd90d7939e470ed4da468683f41baa3 Mon Sep 17 00:00:00 2001 From: Ahmed Zaki Date: Fri, 23 May 2025 14:55:37 -0600 Subject: [PATCH 2/3] idpf: convert control queue mutex to a spinlock With VIRTCHNL2_CAP_MACFILTER enabled, the following warning is generated on module load: [ 324.701677] BUG: sleeping function called from invalid context at kernel/locking/mutex.c:578 [ 324.701684] in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 1582, name: NetworkManager [ 324.701689] preempt_count: 201, expected: 0 [ 324.701693] RCU nest depth: 0, expected: 0 [ 324.701697] 2 locks held by NetworkManager/1582: [ 324.701702] #0: ffffffff9f7be770 (rtnl_mutex){....}-{3:3}, at: rtnl_newlink+0x791/0x21e0 [ 324.701730] #1: ff1100216c380368 (_xmit_ETHER){....}-{2:2}, at: __dev_open+0x3f0/0x870 [ 324.701749] Preemption disabled at: [ 324.701752] [] __dev_open+0x3dd/0x870 [ 324.701765] CPU: 30 UID: 0 PID: 1582 Comm: NetworkManager Not tainted 6.15.0-rc5+ #2 PREEMPT(voluntary) [ 324.701771] Hardware name: Intel Corporation M50FCP2SBSTD/M50FCP2SBSTD, BIOS SE5C741.86B.01.01.0001.2211140926 11/14/2022 [ 324.701774] Call Trace: [ 324.701777] [ 324.701779] dump_stack_lvl+0x5d/0x80 [ 324.701788] ? __dev_open+0x3dd/0x870 [ 324.701793] __might_resched.cold+0x1ef/0x23d <..> [ 324.701818] __mutex_lock+0x113/0x1b80 <..> [ 324.701917] idpf_ctlq_clean_sq+0xad/0x4b0 [idpf] [ 324.701935] ? kasan_save_track+0x14/0x30 [ 324.701941] idpf_mb_clean+0x143/0x380 [idpf] <..> [ 324.701991] idpf_send_mb_msg+0x111/0x720 [idpf] [ 324.702009] idpf_vc_xn_exec+0x4cc/0x990 [idpf] [ 324.702021] ? rcu_is_watching+0x12/0xc0 [ 324.702035] idpf_add_del_mac_filters+0x3ed/0xb50 [idpf] <..> [ 324.702122] __hw_addr_sync_dev+0x1cf/0x300 [ 324.702126] ? find_held_lock+0x32/0x90 [ 324.702134] idpf_set_rx_mode+0x317/0x390 [idpf] [ 324.702152] __dev_open+0x3f8/0x870 [ 324.702159] ? __pfx___dev_open+0x10/0x10 [ 324.702174] __dev_change_flags+0x443/0x650 <..> [ 324.702208] netif_change_flags+0x80/0x160 [ 324.702218] do_setlink.isra.0+0x16a0/0x3960 <..> [ 324.702349] rtnl_newlink+0x12fd/0x21e0 The sequence is as follows: rtnl_newlink()-> __dev_change_flags()-> __dev_open()-> dev_set_rx_mode() - > # disables BH and grabs "dev->addr_list_lock" idpf_set_rx_mode() -> # proceed only if VIRTCHNL2_CAP_MACFILTER is ON __dev_uc_sync() -> idpf_add_mac_filter -> idpf_add_del_mac_filters -> idpf_send_mb_msg() -> idpf_mb_clean() -> idpf_ctlq_clean_sq() # mutex_lock(cq_lock) Fix by converting cq_lock to a spinlock. All operations under the new lock are safe except freeing the DMA memory, which may use vunmap(). Fix by requesting a contiguous physical memory for the DMA mapping. Fixes: a251eee62133 ("idpf: add SRIOV support and other ndo_ops") Reviewed-by: Aleksandr Loktionov Signed-off-by: Ahmed Zaki Reviewed-by: Simon Horman Tested-by: Samuel Salin Signed-off-by: Tony Nguyen --- .../net/ethernet/intel/idpf/idpf_controlq.c | 23 +++++++++---------- .../ethernet/intel/idpf/idpf_controlq_api.h | 2 +- drivers/net/ethernet/intel/idpf/idpf_lib.c | 12 ++++++---- 3 files changed, 20 insertions(+), 17 deletions(-) diff --git a/drivers/net/ethernet/intel/idpf/idpf_controlq.c b/drivers/net/ethernet/intel/idpf/idpf_controlq.c index b28991dd1870..48b8e184f3db 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_controlq.c +++ b/drivers/net/ethernet/intel/idpf/idpf_controlq.c @@ -96,7 +96,7 @@ static void idpf_ctlq_init_rxq_bufs(struct idpf_ctlq_info *cq) */ static void idpf_ctlq_shutdown(struct idpf_hw *hw, struct idpf_ctlq_info *cq) { - mutex_lock(&cq->cq_lock); + spin_lock(&cq->cq_lock); /* free ring buffers and the ring itself */ idpf_ctlq_dealloc_ring_res(hw, cq); @@ -104,8 +104,7 @@ static void idpf_ctlq_shutdown(struct idpf_hw *hw, struct idpf_ctlq_info *cq) /* Set ring_size to 0 to indicate uninitialized queue */ cq->ring_size = 0; - mutex_unlock(&cq->cq_lock); - mutex_destroy(&cq->cq_lock); + spin_unlock(&cq->cq_lock); } /** @@ -173,7 +172,7 @@ int idpf_ctlq_add(struct idpf_hw *hw, idpf_ctlq_init_regs(hw, cq, is_rxq); - mutex_init(&cq->cq_lock); + spin_lock_init(&cq->cq_lock); list_add(&cq->cq_list, &hw->cq_list_head); @@ -272,7 +271,7 @@ int idpf_ctlq_send(struct idpf_hw *hw, struct idpf_ctlq_info *cq, int err = 0; int i; - mutex_lock(&cq->cq_lock); + spin_lock(&cq->cq_lock); /* Ensure there are enough descriptors to send all messages */ num_desc_avail = IDPF_CTLQ_DESC_UNUSED(cq); @@ -332,7 +331,7 @@ int idpf_ctlq_send(struct idpf_hw *hw, struct idpf_ctlq_info *cq, wr32(hw, cq->reg.tail, cq->next_to_use); err_unlock: - mutex_unlock(&cq->cq_lock); + spin_unlock(&cq->cq_lock); return err; } @@ -364,7 +363,7 @@ int idpf_ctlq_clean_sq(struct idpf_ctlq_info *cq, u16 *clean_count, if (*clean_count > cq->ring_size) return -EBADR; - mutex_lock(&cq->cq_lock); + spin_lock(&cq->cq_lock); ntc = cq->next_to_clean; @@ -397,7 +396,7 @@ int idpf_ctlq_clean_sq(struct idpf_ctlq_info *cq, u16 *clean_count, cq->next_to_clean = ntc; - mutex_unlock(&cq->cq_lock); + spin_unlock(&cq->cq_lock); /* Return number of descriptors actually cleaned */ *clean_count = i; @@ -435,7 +434,7 @@ int idpf_ctlq_post_rx_buffs(struct idpf_hw *hw, struct idpf_ctlq_info *cq, if (*buff_count > 0) buffs_avail = true; - mutex_lock(&cq->cq_lock); + spin_lock(&cq->cq_lock); if (tbp >= cq->ring_size) tbp = 0; @@ -524,7 +523,7 @@ int idpf_ctlq_post_rx_buffs(struct idpf_hw *hw, struct idpf_ctlq_info *cq, wr32(hw, cq->reg.tail, cq->next_to_post); } - mutex_unlock(&cq->cq_lock); + spin_unlock(&cq->cq_lock); /* return the number of buffers that were not posted */ *buff_count = *buff_count - i; @@ -552,7 +551,7 @@ int idpf_ctlq_recv(struct idpf_ctlq_info *cq, u16 *num_q_msg, u16 i; /* take the lock before we start messing with the ring */ - mutex_lock(&cq->cq_lock); + spin_lock(&cq->cq_lock); ntc = cq->next_to_clean; @@ -614,7 +613,7 @@ int idpf_ctlq_recv(struct idpf_ctlq_info *cq, u16 *num_q_msg, cq->next_to_clean = ntc; - mutex_unlock(&cq->cq_lock); + spin_unlock(&cq->cq_lock); *num_q_msg = i; if (*num_q_msg == 0) diff --git a/drivers/net/ethernet/intel/idpf/idpf_controlq_api.h b/drivers/net/ethernet/intel/idpf/idpf_controlq_api.h index 9642494a67d8..3414c5f9a831 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_controlq_api.h +++ b/drivers/net/ethernet/intel/idpf/idpf_controlq_api.h @@ -99,7 +99,7 @@ struct idpf_ctlq_info { enum idpf_ctlq_type cq_type; int q_id; - struct mutex cq_lock; /* control queue lock */ + spinlock_t cq_lock; /* control queue lock */ /* used for interrupt processing */ u16 next_to_use; u16 next_to_clean; diff --git a/drivers/net/ethernet/intel/idpf/idpf_lib.c b/drivers/net/ethernet/intel/idpf/idpf_lib.c index 4eb20ec2accb..80382ff4a5fa 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_lib.c +++ b/drivers/net/ethernet/intel/idpf/idpf_lib.c @@ -2314,8 +2314,12 @@ void *idpf_alloc_dma_mem(struct idpf_hw *hw, struct idpf_dma_mem *mem, u64 size) struct idpf_adapter *adapter = hw->back; size_t sz = ALIGN(size, 4096); - mem->va = dma_alloc_coherent(&adapter->pdev->dev, sz, - &mem->pa, GFP_KERNEL); + /* The control queue resources are freed under a spinlock, contiguous + * pages will avoid IOMMU remapping and the use vmap (and vunmap in + * dma_free_*() path. + */ + mem->va = dma_alloc_attrs(&adapter->pdev->dev, sz, &mem->pa, + GFP_KERNEL, DMA_ATTR_FORCE_CONTIGUOUS); mem->size = sz; return mem->va; @@ -2330,8 +2334,8 @@ void idpf_free_dma_mem(struct idpf_hw *hw, struct idpf_dma_mem *mem) { struct idpf_adapter *adapter = hw->back; - dma_free_coherent(&adapter->pdev->dev, mem->size, - mem->va, mem->pa); + dma_free_attrs(&adapter->pdev->dev, mem->size, + mem->va, mem->pa, DMA_ATTR_FORCE_CONTIGUOUS); mem->size = 0; mem->va = NULL; mem->pa = 0; From 0325143b59c6c6d79987afc57d2456e7a20d13b7 Mon Sep 17 00:00:00 2001 From: Vitaly Lifshits Date: Wed, 11 Jun 2025 15:52:54 +0300 Subject: [PATCH 3/3] igc: disable L1.2 PCI-E link substate to avoid performance issue I226 devices advertise support for the PCI-E link L1.2 substate. However, due to a hardware limitation, the exit latency from this low-power state is longer than the packet buffer can tolerate under high traffic conditions. This can lead to packet loss and degraded performance. To mitigate this, disable the L1.2 substate. The increased power draw between L1.1 and L1.2 is insignificant. Fixes: 43546211738e ("igc: Add new device ID's") Link: https://lore.kernel.org/intel-wired-lan/15248b4f-3271-42dd-8e35-02bfc92b25e1@intel.com Signed-off-by: Vitaly Lifshits Reviewed-by: Aleksandr Loktionov Tested-by: Mor Bar-Gabay Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/igc/igc_main.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c index 686793c539f2..031c332f66c4 100644 --- a/drivers/net/ethernet/intel/igc/igc_main.c +++ b/drivers/net/ethernet/intel/igc/igc_main.c @@ -7115,6 +7115,10 @@ static int igc_probe(struct pci_dev *pdev, adapter->port_num = hw->bus.func; adapter->msg_enable = netif_msg_init(debug, DEFAULT_MSG_ENABLE); + /* Disable ASPM L1.2 on I226 devices to avoid packet loss */ + if (igc_is_device_id_i226(hw)) + pci_disable_link_state(pdev, PCIE_LINK_STATE_L1_2); + err = pci_save_state(pdev); if (err) goto err_ioremap; @@ -7500,6 +7504,9 @@ static int __igc_resume(struct device *dev, bool rpm) pci_enable_wake(pdev, PCI_D3hot, 0); pci_enable_wake(pdev, PCI_D3cold, 0); + if (igc_is_device_id_i226(hw)) + pci_disable_link_state(pdev, PCIE_LINK_STATE_L1_2); + if (igc_init_interrupt_scheme(adapter, true)) { netdev_err(netdev, "Unable to allocate memory for queues\n"); return -ENOMEM; @@ -7625,6 +7632,9 @@ static pci_ers_result_t igc_io_slot_reset(struct pci_dev *pdev) pci_enable_wake(pdev, PCI_D3hot, 0); pci_enable_wake(pdev, PCI_D3cold, 0); + if (igc_is_device_id_i226(hw)) + pci_disable_link_state_locked(pdev, PCIE_LINK_STATE_L1_2); + /* In case of PCI error, adapter loses its HW address * so we should re-assign it here. */