vmbus_alloc_buffer(9)

📦 version 7.3.1, in nixpkgs unstable

💖 vmbus_alloc_buffer(9) 💖

okay so this is from the Kernel API Manual bestie, and it's literally about grabbing yourself a host-visible, virtually-contiguous buffer. yes queen, memory allocation but make it fashion ✨

🌸 NAME 🌸

vmbus_alloc_buffer — allocate a host-visible, virtually-contiguous buffer.

🎀 SYNOPSIS 🎀

void * vmbus_alloc_buffer (struct vmbus_channel *channel , u32 size ,
struct page ***chunks_out , u32 *chunk_cnt_out );

✨ ARGUMENTS ✨

  • channel — the channel the buffer will be attached to. it's giving "where this baby lives" 🏠
  • size — requested buffer size in bytes. heads up bestie, it will be rounded up to PAGE_SIZE 🫶
  • chunks_out — on success, set to the array of underlying chunks, or NULL when the buffer was allocated with vzalloc. iykyk 💅
  • chunk_cnt_out — on success, set to the number of chunks. counting era 🔢

💜 DESCRIPTION 💜

so there's two whole vibes here depending on whether your buffer needs decryption or not:

  • buffers not requiring decryption are allocated with vzalloc. simple, clean, iconic ✨
  • buffers requiring decryption are allocated as a series of physically-contiguous chunks, starting at MAX_PAGE_ORDER and falling back to smaller orders on allocation failure. she tries big first and downsizes gracefully, no cap 😭

then each chunk is transitioned to host-visible via set_memory_decrypted on its direct-map address — like a glow-up for the direct map 💫 — and after that, all chunks are combined into a virtually-contiguous range via vmap. main character energy fr fr 🌈

🔥 RETURN 🔥

the buffer's virtual address, or NULL on failure. that's it that's the answer 💀

🦋 SEE ALSO 🦋

Kernel file ./drivers/hv/channel.c and also these besties:

ok that's the whole page, go forth and allocate responsibly 🫶💖

this little masterpiece comes to you from ./drivers/hv, dated October 2026, and it's vmbus_alloc_buffer(9) signing off — slay, periodt, bye bestie 👑🌸

📦 where this page comes from