ib_umem_get_cq_buf(9)
π ib_umem_get_cq_buf π
okay so this is a Kernel API Manual page and the vibe is "Pin a CQ buffer umem from per-command attributes." bestie, let's go π β¨
πΈ NAME πΈ
ib_umem_get_cq_buf β Pin a CQ buffer umem from per-command attributes.
β¨ SYNOPSIS β¨
struct ib_umem * ib_umem_get_cq_buf (struct ib_device *device , const
struct uverbs_attr_bundle *attrs , size_t size , int access );
π ARGUMENTS π
deviceβ the IB device. main character device energy.attrsβ the uverbs attribute bundle (and yes, it may beNULL, no cap).sizeβ minimum required CQ buffer length. like, the floor, bestie.accessβ IB access flags.
π¦ DESCRIPTION π¦
So it resolves the CQ buffer from the new UMEM attribute or the legacy CQ buffer attributes. That's the whole slay.
π¨ Important vibe check: there is no UHW VA fallback, so when no source is available, the caller must arrange its own backing (typically an in-kernel allocation). not me leaving you to handle that yourself, but yes π
β RETURN β
- caller-owned umem on success
NULLwhen no source supplied a bufferERR_PTR(...)on error
π SEE ALSO π
- Kernel file
./drivers/infiniband/core/umem.c ib_umem_find_best_pgsz(9)ib_umem_get_desc(9)ib_umem_get_attr(9)ib_umem_get_attr_or_va(9)ib_umem_get_cq_buf_or_va(9)ib_umem_release(9)
mwah π this page is from ./drivers/infiniband/core, dated October 2026. go pin that buffer and stay iconic π«Άβ¨
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