xe_lmtt_drop_pages(9)
β¨ xe_lmtt_drop_pages(9) β Kernel API Manual energy β¨
okay so bestie, welcome to xe_lmtt_drop_pages, straight out of the Kernel API Manual π this is the page where we learn how to yeet a VF's LMTT pages into the void. no cap, it's a whole vibe.
π NAME
xe_lmtt_drop_pages β Remove VF's LMTT Pages.
that's it. that's the whole job. it drops pages. it's giving cleanup era π§Ήβ¨
πΈ SYNOPSIS
void xe_lmtt_drop_pages (struct xe_lmtt *lmtt , unsigned int vfid );
returns nothing (void, iconic, no receipts), takes two things, and expects you to have read the rest of this page first. lowkey non-negotiable π
π ARGUMENTS
lmttβ thexe_lmttto update π¦ this is the struct you're operating on, the main character of the operation.vfidβ the VF identifier, and it's 1-based fr fr (so like, not zero-based, don't come for me, that's just how it is π).
β DESCRIPTION
so what does it actually do? this function removes all LMTT Page Tables that were prepared by xe_lmtt_prepare_pages. every single one. gone. slay π
β οΈ okay this part matters, bestie: this function shall be called only after successful LMTT initialization. see xe_lmtt_init. like, if init didn't succeed, do not pass go, do not collect $200, do not call this. that's a whole mess you don't want π
that's the entire behavior. no sneaky extras, no surprise side quests. it drops pages, and it's obsessed with doing exactly that π₯
π SEE ALSO
- Kernel file
./drivers/gpu/drm/xe/xe_lmtt.c - Local Memory Translation Table(9)
xe_lmtt_page_size(9)xe_lmtt_init(9)xe_lmtt_init_hw(9)xe_lmtt_invalidate_hw(9)xe_lmtt_prepare_pages(9)xe_lmtt_populate_pages(9)xe_lmtt_estimate_pt_size(9)
like, read the whole friend group, they all make sense together π«Ά
and that's the page!! mwah π this came from ./drivers/gpu/drm/xe, dated October 2026, and yes the docs really did name it xe_lmtt_drop_pages(9) twice (title + footer) because consistency is a lifestyle β¨ go drop some pages responsibly, ily π
π¦ where this page comes from
- nixpkgs unstable: linux-manual 7.3.1
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on : a language model's rewrite of the original, under its terms