omap2_prm_de...rt_hardreset(9)

💅 omap2_prm_deassert_hardreset — the Kernel API Manual, bestie edition 🎀

okay so real talk, this page lives in the Kernel API Manual (omap2_prm_de...rt_hardreset(9)), and it's about pulling a submodule out of hard reset and then being patient about it. she waits. she's polite. periodt. ✨

---

🌸 NAME 🌸

omap2_prm_deassert_hardreset — deassert a submodule hardreset line and wait 🫶

like literally: un-reset the thing, then wait for the hardware to confirm it actually happened before coming back to you.

---

🦋 SYNOPSIS 🦋

int omap2_prm_deassert_hardreset (u8 rst_shift , u8 st_shift , u8 part ,
       s16 prm_mod , u16 rst_offset , u16 st_offset );

six arguments, one int back. she's giving "small function, strong opinions." 💅

---

⭐ ARGUMENTS ⭐

• rst_shift — the register bit shift that corresponds to the reset line you're deasserting. this is your line, the one you came here for. 👑
• st_shift — the register bit shift for the status of the deasserted submodule. so like, how you check on it afterwards. 🩺
• part — the PRM partition. warning: not used for OMAP2, so on OMAP2 she's just… there. vibes only. 🙈
• prm_mod — the PRM submodule base (e.g. CORE_MOD). this one actually matters. 🔥
• rst_offset — reset register offset. not used for OMAP2 either, bestie. 💀
• st_offset — reset status register offset. also not used for OMAP2. three for three on the OMAP2 no-shows. 😭

---

💖 DESCRIPTION 💖

Some IPs — think dsp or iva — contain processors that require a HW reset line to be asserted / deasserted in order to fully enable the IP. iconic hardware behaviour, honestly. these modules may have multiple hard-reset lines that reset different 'submodules' inside the IP block. so many little guys, so many reset lines. 🧚

This function takes the submodule out of reset and waits until the PRCM indicates that the reset has completed before returning. like, she doesn't just fire and forget — she confirms. main character energy with follow-through. ✨

🎀 Return values 🎀

• returns 0 upon success 🎉
• -EINVAL upon an argument error (you gave her something she can't work with, ngl) 🙈
• -EEXIST if the submodule was already out of reset (she got there first, lowkey awkward) 💀
• -EBUSY if the submodule did not exit reset promptly (she waited, it ghosted her) 😭

no cap: -EBUSY means the hardware didn't cooperate in time. don't take it personally, but definitely do handle it. 🫶

---

🌈 SEE ALSO 🌈

• Kernel file ./arch/arm/mach-omap2/prm2xxx_3xxx.c — the source, go read it, iykyk 📖
• omap2_prm_is_hardreset_asserted(9) — the "is it still in reset tho" check ✅
• omap2_prm_assert_hardreset(9) — the opposite direction, the assertive sibling 💅

---

and that's the whole page, bestie 💜 source lives at ./arch/arm/mach-omap2, version date October 2026 — slay, go deassert something responsibly ✨🎀🦋