dma_fence_set_deadline(9)
π dma_fence_set_deadline(9) β straight outta the Kernel API Manual
Ok so bestie, picture this: it's the Kernel API Manual, and this one lil function is literally just a hint-dropper. Let's get into it π«Ά
β¨ NAME β¨
dma_fence_set_deadline β set desired fence-wait deadline hint
Yeah. That's the whole vibe. It sets a deadline hint. Iconic, lowkey.
πΈ SYNOPSIS πΈ
void dma_fence_set_deadline (struct dma_fence *fence , ktime_t deadline
);
Two args, zero return, void all the way through. No cap. π
π¦ ARGUMENTS π¦
fenceβ the fence that is to be waited on. This is the main character here, the thing you're waiting for πdeadlineβ the time by which the waiter hopes for the fence to be signaled. It's giving "pls be done by then bestie, I'm begging" β°
π₯ DESCRIPTION π₯
Okay so here's the tea: this function gives the fence signaler a hint about an upcoming deadline β things like vblank β by which point the waiter would prefer the fence to be signaled by. It's not a demand, it's a heads-up. A little "hey, just so you know π" π«Ά
And why do we care? Because it's intended to give feedback to the fence signaler to aid in power management decisions β for example, boosting GPU frequency if a periodic vblank deadline is approaching but the fence is not yet signaled. The way that the GPU gets that little nudge and goes "oh word, bet" and cranks up so the deadline actually gets hit? Obsessed. Literally the plot of this function π
To be super clear (ngl this part matters): it's a hint, delivered to the signaler, to help power management make good choices. That's the whole deal. Periodt π
π SEE ALSO π
Everybody in the friend group shows up here, and yes, we're listing all of them because nobody gets left out:
- Kernel file
./drivers/dma-buf/dma-fence.c DMA fences overview(9)fence cross-driver contract(9)dma_fence_get_stub(9)dma_fence_allocate_private_stub(9)dma_fence_context_alloc(9)fence signalling annotation(9)dma_fence_begin_signalling(9)dma_fence_end_signalling(9)dma_fence_signal_timestamp_locked(9)dma_fence_signal_timestamp(9)dma_fence_signal_locked(9)dma_fence_check_and_signal_locked(9)dma_fence_check_and_signal(9)dma_fence_signal(9)dma_fence_wait_timeout(9)dma_fence_release(9)dma_fence_free(9)dma_fence_enable_signaling(9)dma_fence_add_callback(9)dma_fence_get_status(9)dma_fence_remove_callback(9)dma_fence_default_wait(9)dma_fence_wait_any_timeout(9)deadline hints(9)dma_fence_describe(9)dma_fence_init(9)dma_fence_init64(9)dma_fence_driver_name(9)dma_fence_timeline_name(9)
The whole squad fr fr ππ
Quick reminder bestie: this is about hints, not guarantees. Don't go treating it like a promise the fence will 100% be signaled by deadline β the page never said that, and neither am I π
And that's the page!! Sourced from ./drivers/dma-buf, dated October 2026, and it's a dma_fence_set_deadline(9) situation from top to bottom. Slay, go set some deadlines, mwah π§πβ
π¦ where this page comes from
- nixpkgs unstable: linux-manual 7.3.1
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on : a language model's rewrite of the original, under its terms