spum_cipher_req_init(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

spum_cipher_req_init(9) β€” Kernel API Manual ✨

πŸ’– NAME πŸ’–

the way that spum_cipher_req_init builds a SPU request message header, up to and including the BD header β€” iconic behavior, periodt. πŸ’…

🌸 SYNOPSIS 🌸

ngl bestie, here's the signature, copy it exactly:

       u16 spum_cipher_req_init (u8 *spu_hdr , struct spu_cipher_parms
       *cipher_parms );

πŸŽ€ ARGUMENTS πŸŽ€

  • spu_hdr β€” start of the SPU request header (MH). literally the beginning of the header buffer, main character energy. ✨
  • cipher_parms β€” parameters that describe the cipher request. it's giving "all the cipher-request deets," lowkey essential. πŸ’–

πŸ¦‹ DESCRIPTION πŸ¦‹

okay so this function constructs the message starting at spu_hdr. The caller should allocate this buffer in DMA-able memory, and it has to be at least SPU_HEADER_ALLOC_LEN bytes long. highkey this is the vibe check: DMA-able memory, minimum SPU_HEADER_ALLOC_LEN bytes, no cap, fr fr. The function ate that, ngl. 🫢

πŸ”₯ RETURN πŸ”₯

Returns the length of the SPU header in bytes. If an error occurs, it returns 0. That 0 is the "not me getting an error" signal, so don't ignore it β€” obsessed with clear error signals, honestly. 😭

πŸ’œ SEE ALSO πŸ’œ

Kernel file ./drivers/crypto/bcm/spu.c β€” yes, that's the spot. Then go check these besties too (iykyk):


mwah πŸ’‹ this page's source/version bit is ./drivers/crypto/bcm and the date is October 2026 β€” bye bestie, slay your cipher requests! πŸ«ΆπŸ‘‘

πŸ“¦ where this page comes from