cxl_region_...width_update(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… cxl_region_...width_update(9) β€” like, the abbreviated header bestie, because the real one is cxl_region_shared_upstream_bandwidth_update and it would NOT fit πŸŽ€

okay so this is a page straight out of the Kernel API Manual, and before you ask β€” yes, the header up top literally abbreviates the function name with ... because it's that long. iconic, honestly. anyway. let's get into it ✨


🌸 NAME 🌸

cxl_region_shared_upstream_bandwidth_update β€” recalculates the bandwidth for the region 🫢

✨ SYNOPSIS ✨

void cxl_region_shared_upstream_bandwidth_update (struct cxl_region
*cxlr );

πŸŽ€ ARGUMENTS πŸŽ€

  • cxlr β€” the cxl region you want to recalculate, bestie. that's the whole vibe.

πŸ¦‹ DESCRIPTION πŸ¦‹

so the way this works is it walks the topology from bottom up and calculates the bandwidth. no cap, bottom to top, that's the direction. 🧚

here's the lil journey it goes on, in order:

  1. it starts at the endpoints
  2. then it processes at the switches, if any (if any!! conditional!! πŸ‘€)
  3. then at the rootport level
  4. then at the host bridge level
  5. and finally aggregates at the region πŸ’–

literally just a whole climb up the stack ending in one big aggregate moment. main character energy fr fr.

πŸ”₯ SEE ALSO πŸ”₯

like, go read those too, they're all in the group chat πŸ’…


ngl the footer is giving very much "signed, sealed, delivered" β€” this one lives at ./drivers/cxl/core, and it's dated October 2026 πŸ“ mwah, that's the whole page, mwah πŸ’œ
πŸ“¦ where this page comes from