Extract the NUMA node lookup from read_page_owner() into a shared page_owner_get_nid() helper, and expose it to BPF programs as bpf_page_owner_get_nid().
The nid bitfield layout is build-time configuration, so BPF programs cannot accurately derive it from page->flags. Signed-off-by: Zhen Ni <[email protected]> --- kernel/bpf/helpers.c | 1 + mm/page_owner.c | 26 +++++++++++++++----------- mm/page_owner.h | 1 + mm/page_owner_iter.c | 5 +++++ 4 files changed, 22 insertions(+), 11 deletions(-) diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c index 23a2d2d71f4d..3e8afd19a1f4 100644 --- a/kernel/bpf/helpers.c +++ b/kernel/bpf/helpers.c @@ -4951,6 +4951,7 @@ BTF_ID_FLAGS(func, bpf_iter_page_owner_new, KF_ITER_NEW | KF_SLEEPABLE) BTF_ID_FLAGS(func, bpf_iter_page_owner_next, KF_ITER_NEXT | KF_RET_NULL | KF_SLEEPABLE) BTF_ID_FLAGS(func, bpf_iter_page_owner_destroy, KF_ITER_DESTROY | KF_SLEEPABLE) BTF_ID_FLAGS(func, bpf_page_owner_stack_snprint, KF_SLEEPABLE) +BTF_ID_FLAGS(func, bpf_page_owner_get_nid, KF_SLEEPABLE) #endif BTF_ID_FLAGS(func, __bpf_trap) BTF_ID_FLAGS(func, bpf_strcmp, KF_PERFMON); diff --git a/mm/page_owner.c b/mm/page_owner.c index eb420803a434..43a722678acc 100644 --- a/mm/page_owner.c +++ b/mm/page_owner.c @@ -711,6 +711,19 @@ void __dump_page_owner(const struct page *page) page_ext_put(page_ext); } +/* + * NUMA node of @page, bypassing PF_POISONED_CHECK() in page_to_nid() to + * avoid VM_BUG_ON on poisoned pages; -1 for poisoned pages. + */ +int page_owner_get_nid(struct page *page) +{ + memdesc_flags_t page_flags = READ_ONCE(page->flags); + + if (page_flags.f == PAGE_POISON_PATTERN) + return -1; + return memdesc_nid(&page_flags); +} + /* * Advance @scan to the next page eligible for page_owner output. * On a hit, fill scan->pfn/page/po_snap and return true; the caller @@ -810,18 +823,9 @@ read_page_owner(struct file *file, char __user *buf, size_t count, loff_t *ppos) /* Find an allocated page */ while (page_owner_next_eligible(&scan)) { if (state->nid_filter_enabled) { - int nid; - memdesc_flags_t page_flags = - READ_ONCE(scan.page->flags); + int nid = page_owner_get_nid(scan.page); - /* - * Bypass PF_POISONED_CHECK() in page_to_nid() to avoid - * VM_BUG_ON when accessing poisoned pages. - */ - if (page_flags.f == PAGE_POISON_PATTERN) - goto skip_continue; - nid = memdesc_nid(&page_flags); - if (!node_isset(nid, state->nid_filter)) + if (nid < 0 || !node_isset(nid, state->nid_filter)) goto skip_continue; } diff --git a/mm/page_owner.h b/mm/page_owner.h index fda83be69275..c57859d18dc3 100644 --- a/mm/page_owner.h +++ b/mm/page_owner.h @@ -39,6 +39,7 @@ struct page_owner_scan { }; bool page_owner_next_eligible(struct page_owner_scan *scan); +int page_owner_get_nid(struct page *page); /* * mm/page_owner_iter.c diff --git a/mm/page_owner_iter.c b/mm/page_owner_iter.c index 80693d417407..907517a29144 100644 --- a/mm/page_owner_iter.c +++ b/mm/page_owner_iter.c @@ -165,4 +165,9 @@ __bpf_kfunc int bpf_page_owner_stack_snprint(depot_stack_handle_t handle, return stack_depot_snprint(handle, buf, buf_size, 0); } +__bpf_kfunc int bpf_page_owner_get_nid(struct page_owner_scan *scan) +{ + return page_owner_get_nid(scan->page); +} + __bpf_kfunc_end_defs(); -- 2.20.1
