r535_gsp_retain_logging(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

🌸 r535_gsp_retain_logging β€” the manual page, but make it make sense ✨

okay bestie, so this page is about r535_gsp_retain_logging, and it lives in the Kernel API Manual, section 9 (that's kernel-land, iykyk). It's a lil function with big "save the receipts" energy πŸ’…


🌈 SYNOPSIS

void r535_gsp_retain_logging (struct nvkm_gsp *gsp );

returns nothing, takes one thing. short and sweet, periodt.


πŸŽ€ ARGUMENTS

  • gsp β€” the gsp pointer 🫢 that's it, that's the whole argument list.

πŸ’– DESCRIPTION

okay so here's the vibe check: if keep_gsp_logging is enabled, then we want to preserve the GSP-RM logging buffers and their debugfs entries. but plot twist 😭 β€” all those objects would normally get deleted if GSP-RM fails to load. so we swoop in and save them before that happens. main character energy fr.

to preserve the logging buffers, we need to do these things:

  1. Allocate new buffers and copy the logs into them, so that the original DMA buffers can be released. (like moving your notes into a new notebook so you can hand back the old one πŸ“)
  2. Preserve the directories. We don't need to save single dentries because we're going to delete the parent when the *β€” and yeah, the sentence literally trails off like that in the source πŸ™ˆ not me finishing it for you, i refuse to invent lore.
⚠️ and here's the important bit: if anything fails in this process, then all the dentries need to be deleted. We don't need to deallocate the original logging buffers, because the caller will do that regardless. so like, don't double-free, bestie πŸ’€ the caller's already got it.

πŸ¦‹ SEE ALSO

like, that is a whole friend group right there. go read them too πŸ‘‘


and that's the page! source path got cut off at ./drivers/gpu/drm/nouveau/nvkm... (they really said "you'll figure it out" 😭), the date on this one is October 2026, and it's from the Kernel API Manual. mwah, go ship it πŸ’œβœ¨

πŸ“¦ where this page comes from