hfi1_num_netdev_contexts(9)

✨ hfi1_num_netdev_contexts ✨

okay bestie, welcome to hfi1_num_netdev_contexts — that's a kernel API manual page, section 9, and it is literally the function that decides how many netdev receive contexts you get. yes. that one. 🎀

---

💖 NAME 💖

hfi1_num_netdev_contexts — count of netdev recv contexts to use. that's it, that's the vibe. 🦋

---

🌸 SYNOPSIS 🌸

u32 hfi1_num_netdev_contexts (struct hfi1_devdata *dd , u32
available_contexts , struct cpumask *cpu_mask );

look at her. three arguments, one u32 back out. main character energy fr. 👑

---

💅 ARGUMENTS 💅

• dd — the device you're allocating netdev contexts on. like, which device are we doing this for, bestie. 🫶
• available_contexts — count of available receive contexts. that's the number of recv contexts you've got lying around to play with. ✨
• cpu_mask — mask of possible cpus to include for contexts. so like, which cpus are invited to this party. 🌈

---

🔥 RETURN 🔥

so what comes back? the count of physical cores on a node, OR the remaining available recv contexts for netdev recv context usage — but only up to the maximum of HFI1_MAX_NETDEV_CTXTS. that cap is non-negotiable, no cap. 💀

and heads up, because this part matters: a value of 0 can be returned when acceleration is explicitly turned off, when a memory allocation error occurs, or when there are no available contexts. so if you get a 0, that's not a glitch in the matrix, that's one of those three things happening. 😭

---

🦋 SEE ALSO 🦋

• Kernel file ./drivers/infiniband/hw/hfi1/netdev_rx.c
• hfi1_netdev_rx_init(9)
• hfi1_netdev_rx_destroy(9)
• hfi1_alloc_rx(9)
• hfi1_netdev_enable_queues(9)
• hfi1_netdev_add_data(9)
• hfi1_netdev_remove_data(9)
• hfi1_netdev_get_data(9)
• hfi1_netdev_get_first_data(9)

go read those too, iykyk. 📚✨

---

that's the page, bestie 💜 — brought to you from ./drivers/infiniband/hw/hfi1, dated October 2026, slaying under the hfi1_num_netdev_contexts(9) name. mwah. 💋🎀