This is a note to let you know that I've just added the patch titled net: hns3: change affinity_mask to numa node range to the 5.10-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary The filename of the patch is: net-hns3-change-affinity_mask-to-numa-node-range.patch and it can be found in the queue-5.10 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable@xxxxxxxxxxxxxxx> know about it. >From 1dc839ec09d3ab2a4156dc98328b8bc3586f2b70 Mon Sep 17 00:00:00 2001 From: Yufeng Mo <moyufeng@xxxxxxxxxx> Date: Mon, 13 Sep 2021 21:08:22 +0800 Subject: net: hns3: change affinity_mask to numa node range From: Yufeng Mo <moyufeng@xxxxxxxxxx> commit 1dc839ec09d3ab2a4156dc98328b8bc3586f2b70 upstream. Currently, affinity_mask is set to a single cpu. As a result, irqbalance becomes invalid in SUBSET or EXACT mode. To solve this problem, change affinity_mask to numa node range. In this way, irqbalance can be performed on the cpu of the numa node. Fixes: 0812545487ec ("net: hns3: add interrupt affinity support for misc interrupt") Signed-off-by: Yufeng Mo <moyufeng@xxxxxxxxxx> Signed-off-by: Guangbin Huang <huangguangbin2@xxxxxxxxxx> Signed-off-by: David S. Miller <davem@xxxxxxxxxxxxx> Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> --- drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -1463,9 +1463,10 @@ static void hclge_init_kdump_kernel_conf static int hclge_configure(struct hclge_dev *hdev) { + const struct cpumask *cpumask = cpu_online_mask; struct hclge_cfg cfg; unsigned int i; - int ret; + int node, ret; ret = hclge_get_cfg(hdev, &cfg); if (ret) @@ -1526,11 +1527,12 @@ static int hclge_configure(struct hclge_ hclge_init_kdump_kernel_config(hdev); - /* Set the init affinity based on pci func number */ - i = cpumask_weight(cpumask_of_node(dev_to_node(&hdev->pdev->dev))); - i = i ? PCI_FUNC(hdev->pdev->devfn) % i : 0; - cpumask_set_cpu(cpumask_local_spread(i, dev_to_node(&hdev->pdev->dev)), - &hdev->affinity_mask); + /* Set the affinity based on numa node */ + node = dev_to_node(&hdev->pdev->dev); + if (node != NUMA_NO_NODE) + cpumask = cpumask_of_node(node); + + cpumask_copy(&hdev->affinity_mask, cpumask); return ret; } Patches currently in stable-queue which might be from moyufeng@xxxxxxxxxx are queue-5.10/net-hns3-pad-the-short-tunnel-frame-before-sending-to-hardware.patch queue-5.10/net-hns3-disable-mac-in-flr-process.patch queue-5.10/net-hns3-change-affinity_mask-to-numa-node-range.patch