ib_copy_vali..._udata_in_cm(9)

🎀 ib_copy_vali..._udata_in_cm(9) 🎀

okay so bestie, this little legend is from the Kernel API Manual, and it's highkey giving "tiny kernel helper with a very long name" energy, with main character energy for days. The star of the show is ib_copy_validate_udata_in_cm, and the whole vibe is: copy the req structure, then check the comp_mask. Periodt. 💅

✨ NAME ✨

ib_copy_validate_udata_in_cm — copies the req structure and checks the comp_mask. That's it, that's the name, no cap, fr fr. 🫶

🌸 SYNOPSIS 🌸

ib_copy_validate_udata_in_cm (_udata , _req , _end_member , _valid_cm );

🦋 ARGUMENTS 🦋

• _udata — the system call's ib_udata struct. literally the udata situation. 💖
• _req — the name of an on-stack structure that holds the driver data. on-stack bestie, holding the driver data. 🎀
• _end_member — the member in the struct that is the original end of struct from the first kernel to introduce it. not me lowkey loving the original end-of-struct lore. 📜
• _valid_cm — a bitmask of bits permitted in the comp_mask_field. the allowed-bits guest list. 🎟️

🔥 DESCRIPTION 🔥

So what it does: the way that it checks that the udata input request struct is properly formed for this kernel — vibe check — and then copies it into req is honestly iconic. Slay, ngl. That's the whole plot, bestie. ✨

👑 SEE ALSO 👑

• Kernel file: ./include/rdma/uverbs_ioctl.h
• uverbs_attr_get_uobjs_arr(9)
• ib_copy_validate_udata_in(9)
• ib_is_udata_in_empty(9)
• ib_respond_udata(9)
• ib_respond_empty_udata(9)
• ib_no_udata_io(9)

this reference list ate, no crumbs. iykyk. 🍓

Cute sign-off time: source/version is ./include/rdma, date is October 2026, and the page is ib_copy_vali..._udata_in_cm(9) — byeee, love u, mwah 💋🌈