dma_fence_work_commit_imm(9)
β¨ dma_fence_work_commit_imm(9) β¨
okay bestie, welcome to the Kernel API Manual π this one is short but it's giving βtiny function, very specific rules,β and honestly? iconic.
πΈ NAME πΈ
dma_fence_work_commit_imm β commit the fence, and if possible execute locally. slay.
π SYNOPSIS π
void dma_fence_work_commit_imm (struct dma_fence_work *f );
π¦ ARGUMENTS π¦
fβ the fenced worker. that's the whole vibe π«Ά
π DESCRIPTION π
So instead of always scheduling a worker to execute the callback (see dma_fence_work_commit), this function tries to execute the callback immediately in the local context. no cap, that's the whole point.
β οΈ but listen: it is required that the fence be committed before it is published, and that no other threads try to tamper with the number of asynchronous waits on the fence. if they do? the callback will be executed in the wrong context β i.e. not the callers. and that's not cute at all π
like, this is not a βmaybe try it and seeβ situation. it's a βrespect the rules or the context gets messyβ situation. periodt.
π SEE ALSO π
- Kernel file
./drivers/gpu/drm/i915/i915_sw_fence_work.hπ
mwah π this little page came from ./drivers/gpu/drm/i915, dated October 2026 β stay fenced, stay local, and don't let the wrong context catch you lacking π§β¨
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