rpmh_write_batch(9)

✨ NAME ✨

okay so this is rpmh_write_batch(9) from the Kernel API Manual, and it does exactly what it says on the tin, bestie: write multiple sets of RPMH commands and wait for the batch to finish. 💅

---

🌸 SYNOPSIS 🌸

int rpmh_write_batch (const struct device *dev , enum rpmh_state state ,
       const struct tcs_cmd *cmd , u32 *n );

---

🎀 ARGUMENTS 🎀

• dev — the device making the request 🫶
• state — Active/sleep set ✨
• cmd — the payload data 💖
• n — the array of count of elements in each batch, 0 terminated. periodt.

---

🦋 DESCRIPTION 🦋

So you're writing a request to the RSC controller without caching. No sneaky cache, no little stash, just straight up sending it. 🚀

If the request state is ACTIVE, then the requests are treated as a completion request and sent to the controller immediately. The function waits until all the commands are complete. like, it is literally standing there with its little clipboard waiting for everyone to finish. 😭

If the request was to SLEEP or WAKE_ONLY, then the request is sent as fire-n-forget and no ack is expected. bye bestie, no read receipt, we move. 💀

⚠️ May sleep. Do not call from atomic contexts for ACTIVE_ONLY requests. fr fr, that is not the vibe you want. 🙈

---

🔥 SEE ALSO 🔥

• Kernel file ./drivers/soc/qcom/rpmh.c
• struct cache_req(9)
• struct batch_cache_req(9)
• __rpmh_write(9)
• rpmh_read(9)
• rpmh_write_async(9)
• rpmh_write(9)
• rpmh_flush(9)
• rpmh_invalidate(9)

---

and that's the whole page, bestie 💖 this little cutie comes from the source/version corner at ./drivers/soc/qcom, dated October 2026, and it will forever be known as rpmh_write_batch(9) 🎀✨👑