__dma_fence_is_later(9)

✨ __dma_fence_is_later(9) ✨

okay so bestie, welcome to the Kernel API Manual 🎀 — today we're doing __dma_fence_is_later, and honestly? it's giving tiny function, big main character energy 💅

---

💖 NAME 💖

__dma_fence_is_later — returns whether f1 is chronologically later than f2 ⏰✨

---

🦋 SYNOPSIS 🦋

bool __dma_fence_is_later (struct dma_fence *fence , u64 f1 , u64 f2 );

like, it takes a fence and two u64s and gives you back a bool, that's the whole vibe 🫶

---

🌸 ARGUMENTS 🌸

• fence — the fence in whose context we're doing the comparison, bestie. it's literally the setting of the whole scene 💫
• f1 — the first fence's seqno, ok? 🍓
• f2 — the second fence's seqno, and heads up: it's from the same context 🔥

---

👑 DESCRIPTION 👑

So here's the tea ☕: this returns true if f1 is chronologically later than f2. That's it. That's the function. Slay 💅

⚠️ but listen — both fences MUST be from the same context. A seqno is not common across contexts, so comparing across them is a no-no, no cap 😭 🙈

---

⭐ SEE ALSO ⭐

Kernel file ./include/linux/dma-fence.h 🧚 and the whole friend group:

• struct dma_fence(9)
• struct dma_fence_cb(9)
• struct dma_fence_ops(9)
• dma_fence_was_initialized(9)
• dma_fence_put(9)
• dma_fence_get(9)
• dma_fence_get_rcu(9)
• dma_fence_get_rcu_safe(9)
• dma_fence_spinlock(9)
• dma_fence_lock_irqsave(9)
• dma_fence_unlock_irqrestore(9)
• dma_fence_assert_held(9)
• Safe external access to driver provided object members(9)
• dma_fence_is_signaled_locked(9)
• dma_fence_is_signaled(9)
• dma_fence_is_later(9)
• dma_fence_is_later_or_same(9)
• dma_fence_later(9)
• dma_fence_get_status_locked(9)
• dma_fence_set_error(9)
• dma_fence_timestamp(9)
• dma_fence_wait(9)
• dma_fence_is_array(9)
• dma_fence_is_chain(9)
• dma_fence_is_container(9)

---

and that's the whole page, bestie 🌈 this lil gem lives in ./include/linux, the version is October 2026, and it's filed under __dma_fence_is_later(9) — go forth and compare seqnos responsibly, ilysm 💜🫶✨