i915_vma_size(9)
β¨ i915_vma_size β¨
okay so real quick β this is i915_vma_size(9) from the Kernel API Manual, and honestly? it's a tiny little function that does exactly one thing and does it slay π
π NAME π
i915_vma_size β literally just: get the va range size of the vma π¦
πΈ SYNOPSIS πΈ
u64 i915_vma_size (const struct i915_vma *vma );
π ARGUMENTS π
vmaβ the vma bestie. that's it. that's the whole argument. the vma in question π
π¦ DESCRIPTION π¦
so here's the tea π β GPU virtual address space can be allocated with padding. like, the address space people be reserving more room than they actually need, no cap. this function hands you back the effective virtual address range size, with that padding subtracted out. so you get the real number, not the inflated one β¨ iconic behavior honestly.
β RETURN β
the effective virtual address range size. that's the whole vibe check. π«Ά
π₯ SEE ALSO π₯
kernel file ./drivers/gpu/drm/i915/i915_vma.h β and yes, the whole friend group is here:
i915_vma_offset(9)i915_vma_pin_iomap(9)i915_vma_unpin_iomap(9)i915_vma_pin_fence(9)i915_vma_unpin_fence(9)for_each_ggtt_vma(9)i915_vma_get_current_resource(9)
ngl this one's short but it ate. go read the header file, iykyk π
and that's the whole page diva π
β brought to you by ./drivers/gpu/drm/i915, dated October 2026, main character energy only ππβ¨
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