smi_request_store(9)

📦 version 7.3.1, in nixpkgs unstable

💖 NAME 💖

smi_request_store — and bestie, the manual page literally left the description blank 😭 so we're going with "the store side of the SMI request show" and letting the rest of this page do the talking 💅

This is from the Kernel API Manual, section 9 (kernel API), and it's the sysfs-store handler vibe for the dcdbas Dell driver, iykyk 🦋

✨ SYNOPSIS ✨

ssize_t smi_request_store (struct device *dev , struct device_attribute
*attr , const char *buf , size_t count );

Like, that's the whole signature. Returns an ssize_t, takes four things, no cap 🎀

🫶 ARGUMENTS 🫶

Okay so the original page is out here giving us nothing for these, and I'm not about to invent lore 🙈 — so here's exactly what it says, word for word, which is: not much 💀

  • dev — undescribed 😭
  • attr — undescribed 😭
  • buf — undescribed 😭
  • count — undescribed 😭
Not me filing a bug against whoever wrote this page's argument docs 💀 bestie the fields are just empty. Iconic in a tragic way.

🌸 DESCRIPTION 🌸

This is the part with actual tea ☕ and it's about the valid values you can write in. The valid values are:

  • 0 — zero SMI data buffer 🧹
  • 1 — generate calling interface SMI 📞
  • 2 — generate raw SMI 🔥

That's it, that's the whole list — only those three. Don't be out here writing 3 and expecting main character energy from the driver, because it's not on the list 👑

And the workflow goes like this, bestie, listen up 🫶: the user application writes smi_cmd to smi_data before telling the driver to generate the SMI.

So like — fill the box first, then ring the bell 🔔 periodt.

🦋 SEE ALSO 🦋

Go read these next, they're lowkey the friend group of this page 👯‍♀️

💜 Quick vibe check before you go: this page is from ./drivers/platform/x86/dell and it's dated October 2026, so if you're reading it in the future — hi, hope the SMI's slay 💅

That's the whole page bestie, go write your 0, 1, or 2 and live your truth 🌈✨🎀

📦 where this page comes from