xe_lmtt_estimate_pt_size(9)
π xe_lmtt_estimate_pt_size(9) π
okay so bestie, welcome to the Kernel API Manual β yes we're reading kernel docs for fun now, this is the era we're in, and honestly? iconic of us. β¨
π NAME π
xe_lmtt_estimate_pt_size β literally just estimates the size of LMTT PT allocations. that's the whole job. no cap. π
π¦ SYNOPSIS π¦
u64 xe_lmtt_estimate_pt_size (struct xe_lmtt *lmtt , u64 size );
look at that signature. clean. pristine. u64 in, u64 out, struct pointer in the middle. we love a return type that means business. π
πΈ ARGUMENTS πΈ
lmttβ thexe_lmtt. THE one. the main character. π«Άsizeβ the size of the LMEM that's gonna be mapped over LMTT, and yes that includes any offset, don't forget the offset, i'm begging you π
β DESCRIPTION β
okay so real talk: this function shall be called only by PF. that's it, that's the rule. not by VF, not by your friend, not by you "just testing something". PF only. periodt. π
fr fr, the "shall" is doing a lot of heavy lifting here β this is a hard requirement, not a suggestion π¨
π₯ RETURN π₯
returns the size of the PT allocation(s) needed to support the given LMEM size. that's the vibe check result: a u64 telling you how much PT space you gotta set aside. slay. β¨
π SEE ALSO π
Kernel file ./drivers/gpu/drm/xe/xe_lmtt.cβ the source, the origin story, the birthplace π§Local Memory Translation Table(9)β the whole LMTT situation, go read it bestiexe_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_drop_pages(9)
that's a whole friend group right there, they all hang out in the same file and they all get their own man page because they're all that girl π
and that's the page!! brought to you by ./drivers/gpu/drm/xe (the source, the moment, the legend π) β October 2026 edition, xe_lmtt_estimate_pt_size(9) out. mwah π bye bestie, go estimate some page tables ππΈ
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