[PATCH net-next 1/6] stmmac: intel: set IRQ affinity hint for multi MSI vectors
From: Ong Boon Leong <hidden>
Date: 2021-03-29 13:37:30
Also in:
bpf, linux-arm-kernel, lkml
Subsystem:
networking drivers, stmmac ethernet driver, the rest · Maintainers:
Andrew Lunn, "David S. Miller", Eric Dumazet, Jakub Kicinski, Paolo Abeni, Linus Torvalds
Intel mGBE has independent hardware IRQ resources for TX and RX DMA operation. In preparation to support XDP TX, we add IRQ affinity hint to group both RX and TX queue of the same queue ID to the same CPU. Signed-off-by: Ong Boon Leong <redacted> --- drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c | 7 +++++++ 1 file changed, 7 insertions(+)
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c
index 08b4852eed4c..53a24932a192 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c@@ -810,6 +810,7 @@ static int stmmac_config_multi_msi(struct pci_dev *pdev, struct plat_stmmacenet_data *plat, struct stmmac_resources *res) { + cpumask_t cpu_mask; int ret; int i;
@@ -832,12 +833,18 @@ static int stmmac_config_multi_msi(struct pci_dev *pdev, for (i = 0; i < plat->rx_queues_to_use; i++) { res->rx_irq[i] = pci_irq_vector(pdev, plat->msi_rx_base_vec + i * 2); + cpumask_clear(&cpu_mask); + cpumask_set_cpu(i % num_online_cpus(), &cpu_mask); + irq_set_affinity_hint(res->rx_irq[i], &cpu_mask); } /* For TX MSI */ for (i = 0; i < plat->tx_queues_to_use; i++) { res->tx_irq[i] = pci_irq_vector(pdev, plat->msi_tx_base_vec + i * 2); + cpumask_clear(&cpu_mask); + cpumask_set_cpu(i % num_online_cpus(), &cpu_mask); + irq_set_affinity_hint(res->tx_irq[i], &cpu_mask); } if (plat->msi_mac_vec < STMMAC_MSI_VEC_MAX)
--
2.25.1