kryo_l2_set_indirect_reg(9)

💅 kryo_l2_set_indirect_reg — from the Kernel API Manual, bestie

okay so we're in the Kernel API Manual today and the star of the show is kryo_l2_set_indirect_reg(9). yes the whole entire name. she's long, she's specific, she's iconic. 🎀

---

✨ NAME ✨

kryo_l2_set_indirect_reg — literally just writes a value to an L2 register. that's it. that's the whole job. no cap. 🦋

---

🌸 SYNOPSIS 🌸

void kryo_l2_set_indirect_reg (u64 reg , u64 val );

returns nothing (void, she gives nothing back and honestly? respect 💅), takes two u64s, and the spacing is exactly as printed — do not come for me about the space before those commas, I'm just the messenger. 😭

---

🧚 ARGUMENTS 🧚

• reg — the address of the L2 register. like, which register. you gotta tell her where to go. 📍
• val — the value to be written to that register. the actual payload. the whole point. 🔥

---

👑 DESCRIPTION 👑

the vibe check here: use the architecturally required barriers for ordering between system register accesses, and also for system registers with respect to device memory. ⭐

like, it's not just "write the number and leave" — the ordering guarantees are the whole personality. barriers are doing the heavy lifting so your system register accesses and your device memory stay in their proper lanes. periodt. 💀

ngl bestie, this is one of those "the ordering matters more than the write" situations. the barriers are architecturally required, so don't go removing them thinking you're clever. 🙈

---

🫶 SEE ALSO 🫶

• the Kernel file: ./drivers/soc/qcom/kryo-l2-accessors.c — go read the source, iykyk 🍓
• kryo_l2_get_indirect_reg(9) — her counterpart, the one who reads. they're a matching set. ✨

---

and that's the page, bestie!! 🌈 this absolute gem lives under ./drivers/soc/qcom and the manual's dated October 2026 — ty for coming to my ted talk, mwah 💖🎀💜