cqhci_crypto_init(9)

💅 cqhci_crypto_init(9) — from the Kernel API Manual bestie, no cap

Okay so this is a page from the Kernel API Manual, and honestly? It's a little one but it does things, so let's get into it 🎀

---

✨ NAME ✨

cqhci_crypto_init — the function that gets CQHCI crypto support up and running, periodt 💫

---

🦋 SYNOPSIS 🦋

int cqhci_crypto_init (struct cqhci_host *cq_host );

---

🌸 ARGUMENTS 🌸

• cq_host — a cqhci host 🫶 that's the one and only thing you hand it

---

💖 DESCRIPTION 💖

So here's the vibe check 💅: if the driver already set MMC_CAP2_CRYPTO, and the CQE declares CQHCI_CAP_CS, then this thing goes ahead and initializes the crypto support ✨

And what does "initializing the crypto support" actually involve? Bestie, I'm so glad you asked:

• reading the crypto capability registers 📖
• initializing the blk_crypto_profile
• clearing all the keyslots 🧹 (yes, all of them, every single one)
• enabling 128-bit task descriptors 👑

It's giving "setup montage"... iconic honestly 🌈

---

🔥 RETURN 🔥

• 0 😌 — crypto was initialized or it just isn't supported. Wondering which one it is? Check whether MMC_CAP2_CRYPTO is still set — that flag is literally the answer key 🔑
• a negative errno value 💀 — on an unexpected error, because sometimes things just don't slay

Lowkey important: 0 doesn't mean "yes crypto happened" — it means "either it happened or it couldn't, and the flag tells you which." Don't get it twisted 🙈

---

⭐ SEE ALSO ⭐

• kernel file ./drivers/mmc/host/cqhci-crypto.c — go read it, iykyk 📚

---

And that's the whole page, bestie 🧚 Signing off from ./drivers/mmc/host, dated October 2026, straight out of the Kernel API Manual 💜 Slay responsibly 🎀