ENUM MDDEV_FLAGS(9)

👑 enum mddev_flags(9) — the Kernel API Manual, but make it cute

okay so bestie, this is the page for enum mddev_flags(9) — yeah, the whole flag buffet for md devices. it lives in the Kernel API Manual and it's basically the master list of vibes an md array can be in. let's get into it 💅

---

🌸 NAME 🌸

enum mddev_flags — md device flags. literally that's it, that's the whole name. iconic brevity ✨

---

🦋 SYNOPSIS 🦋

enum mddev_flags {
    MD_ARRAY_FIRST_USE ,
    MD_CLOSING ,
    MD_JOURNAL_CLEAN ,
    MD_HAS_JOURNAL ,
    MD_CLUSTER_RESYNC_LOCKED ,
    MD_FAILFAST_SUPPORTED ,
    MD_HAS_PPL ,
    MD_HAS_MULTIPLE_PPLS ,
    MD_NOT_READY ,
    MD_BROKEN ,
    MD_DO_DELETE ,
    MD_DELETED ,
    MD_HAS_SUPERBLOCK ,
    MD_FAILLAST_DEV ,
    MD_SERIALIZE_POLICY ,
    MD_DM_SUSPENDING

};

look at her. sixteen constants, all lined up, all serving 💖

---

🎀 Constants 🎀

the whole cast of flags, one by one

• MD_ARRAY_FIRST_USE — first use of the array, so it needs initialization. baby's first boot energy 🌱
• MD_CLOSING — if this is set, we're closing the array, so don't go opening it then. read the room, fr fr 🙈
• MD_JOURNAL_CLEAN — a raid with journal is already clean. spotless. ate and left no crumbs ✨
• MD_HAS_JOURNAL — the raid array has the journal feature set. she journals 📓
• MD_CLUSTER_RESYNC_LOCKED — cluster raid only, and it means the node already took the resync lock, so the lock needs to be released. team effort, no cap 🫶
• MD_FAILFAST_SUPPORTED — using MD_FAILFAST on metadata writes is supported, because calls to md_error will never cause the array to become failed. safety net: confirmed 😌
• MD_HAS_PPL — the raid array has the PPL feature set.
• MD_HAS_MULTIPLE_PPLS — the raid array has the multiple PPLs feature set. yes, plural, she's extra 💅
• MD_NOT_READY — do_md_run is active, so array_state must not report that the array is ready yet. hold on bestie, we're still loading ⏳
• MD_BROKEN — this one is used to stop writes and mark the array as failed. it's giving intervention 💀
• MD_DO_DELETE — -- undescribed --. the manpage really said no thoughts, head empty 😭 (that's genuinely all the source gives us)
• MD_DELETED — this device is being deleted. bye bestie 👋
• MD_HAS_SUPERBLOCK — there is persistence sb in member disks.
• MD_FAILLAST_DEV — allow the last rdev to be removed. one last ride 🔥
• MD_SERIALIZE_POLICY — enforce that write IO is not reordered; just used by raid1. no cutting the line 🚫
• MD_DM_SUSPENDING — this DM raid device is suspending. brb, going dormant 🧚

---

🌈 Description 🌈

so the actual instruction here is: change UNSUPPORTED_MDDEV_FLAGS for each array type if a new flag is added. that's the whole description. short, sweet, and honestly kind of a threat 👑

---

⭐ SEE ALSO ⭐

• Kernel file ./drivers/md/md.h — go read the source, iykyk 📖

---

and that's the page!! sign-off time: this one ships from ./drivers/md, dated October 2026, no cap. go forth and flag responsibly 🫶💜🎀