drbd_submit_peer_request(9)

okay so welcome to drbd_submit_peer_request(9), straight out of the Kernel API Manual — Unix bestie, this one is short but lowkey important 💅✨

🌸 NAME 🌸

drbd_submit_peer_request — and bestie, the descriptor after the dash is literally blank in the source 💀. No extra tagline, no little summary, just the function name being the main character.

---

✨ SYNOPSIS ✨

Catch the signature, no cap:

int drbd_submit_peer_request (struct drbd_peer_request *peer_req );

---

🫶 ARGUMENTS 🫶

• peer_req — peer request. That's the whole note. Short, sweet, and absolutely not going to elaborate further 🎀

---

💅 DESCRIPTION 💅

So the way this works: it may spread the pages to multiple bios, depending on bio_add_page restrictions. It's giving "one request, possibly many bios," iykyk 🦋

Here's the return-value vibe check:

• Returns 0 if all the bios have been submitted — slay, we love a clean finish ✨
• Returns -ENOMEM if we could not allocate enough bios 😭
• Returns -ENOSPC — and the docs literally add “(any better suggestion?)” right there, which is such a mood 💀 — if we have not been able to bio_add_page a single page to an empty bio.

That -ENOSPC case should never happen and likely indicates that the lower level IO stack is in some way broken. And yes, this has been observed on certain Xen deployments. Not a drill, bestie 🙈🔥

---

👑 SEE ALSO 👑

• Kernel file ./drivers/block/drbd/drbd_receiver.c
• drbd_alloc_pages(9)
• drbd_socket_okay(9)
• drbd_may_finish_epoch(9)
• convert_state(9)

---

💖 cute little footer sign-off: this page comes from the source/version field ./drivers/block/drbd, dated October 2026, and it's all about drbd_submit_peer_request(9). bye bestie, go submit those peer requests responsibly 🎀✨