rdma_dev_access_netns(9)

📦 version 7.3.1, in nixpkgs unstable

💅 NAME 💅

rdma_dev_access_netns — the little function that tells you whether an rdma device can be accessed from a specified net namespace or not. yeah. that's the whole vibe check. ✨

✨ SYNOPSIS ✨

bool rdma_dev_access_netns (const struct ib_device *dev , const struct net *net );

One call, two pointers, one bool coming back at you bestie 🫶

🎀 ARGUMENTS 🎀

  • dev — a pointer to the rdma device that needs to be checked. like, which device are we even talking about, babes 💖
  • net — a pointer to the net namesapce for which access is to be checked. (yes ok, "namesapce" — that's literally how it's spelled in the original, we don't gatekeep typos here 🙈🫶) so: passing in a net namespace pointer, and this function figures out if you get access or not.

🦋 DESCRIPTION 🦋

okay so here's the tea, and it's honestly not that deep but it matters ☕

When the rdma device is in shared mode, it simply ignores the net namespace. like, no thoughts, head empty, namespace who? 💀 it doesn't care, it doesn't check, it just vibes. 🌈

But when the rdma device is exclusive to a net namespace, that's when the vibe check actually happens — the rdma device's net namespace gets checked against the specified one. so it's giving "are you on the list or not" energy 👑 and the function returns whether that access is a yes or a no.

two modes, one answer. iconic honestly. ✨

🔥 SEE ALSO 🔥

First of all, the kernel file that runs all of this lives at ./drivers/infiniband/core/device.c — a whole main character file, no cap 💅

And then the friend group, aka the entire squad of related calls (go read them, they're all slay, iykyk):

all of them, no skips, no skipping the boring ones either — that's the whole roster bestie, go be obsessed 🧚💜

ngl this one's short but it ate. 💅

okay that's the page!! sourced from ./drivers/infiniband/core, dated October 2026 — mwah, go read your manpages, ily, bye 🎀💖🦋🌸⭐

📦 where this page comes from