Pass vmf->address directly instead of ALIGN_DOWN(vmf->address, ...). NUMA interleave is not affected: the ilx calculation in get_vma_policy() shifts addr >> PAGE_SHIFT >> order, which drops sub-page bits regardless of alignment. post_alloc_hook will use the raw address for cache-friendly zeroing via folio_zero_user().
Signed-off-by: Michael S. Tsirkin <[email protected]> Reviewed-by: Gregory Price <[email protected]> Assisted-by: Claude:claude-opus-4-6 Assisted-by: cursor-agent:GPT-5.4-xhigh --- mm/memory.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/mm/memory.c b/mm/memory.c index 86a973119bd4..21f640674c4f 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -5268,8 +5268,7 @@ static struct folio *alloc_anon_folio(struct vm_fault *vmf) /* Try allocating the highest of the remaining orders. */ gfp = vma_thp_gfp_mask(vma); while (orders) { - addr = ALIGN_DOWN(vmf->address, PAGE_SIZE << order); - folio = vma_alloc_folio(gfp, order, vma, addr); + folio = vma_alloc_folio(gfp, order, vma, vmf->address); if (folio) { if (mem_cgroup_charge(folio, vma->vm_mm, gfp)) { count_mthp_stat(order, MTHP_STAT_ANON_FAULT_FALLBACK_CHARGE); -- MST
