irdma_send_mpa_request(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… irdma_send_mpa_request β€” from the Kernel API Manual ✨

Okay so bestie, welcome to a section 9 page (that's the kernel API manual, not your regular user-command one 🫢). This is the page for irdma_send_mpa_request, and it's a whole little moment in the connection manager. Let's get into it πŸ’–


🌸 NAME 🌸

irdma_send_mpa_request β€” the active node goes ahead and sends the MPA request over to the passive node πŸ’Œ

Literally that's the whole vibe: active side speaks first, passive side is sitting there waiting for the message. Iconic behaviour, no notes.


✨ SYNOPSIS ✨

int irdma_send_mpa_request (struct irdma_cm_node *cm_node );

πŸŽ€ ARGUMENTS πŸŽ€

  • cm_node β€” the connection's node πŸ¦‹ that's it, that's the whole argument list. One (1) param, and it's the node that holds all the connection lore for this particular connection.

πŸ¦‹ SEE ALSO πŸ¦‹

So this function lives in the kernel file ./drivers/infiniband/hw/irdma/cm.c (that's the connection manager source, the mothership 🌈), and the manpage points you at the whole extended friend group. Here's everyone, in the order the page gives them:

Like I said. Entire friend group. Read them all, they're all doing bits in the same file πŸ’€


Heads up bestie: this page is short on purpose β€” no OPTIONS, no RETURN VALUE, no ERRORS section, no environment variables, nothing else hiding down here. What you see is genuinely the entire page in the original. Don't go looking for lore that isn't there πŸ™ˆ

And that's the whole thing! Per the footer, this one ships from ./drivers/infiniband/hw/irdma and is dated October 2026 β€” so yeah, mark your calendars, the connection manager said see you then πŸŽ€βœ¨ slay, bye! πŸ’œ

πŸ“¦ where this page comes from