[PATCH RFC 1/1] numa: fix /proc/<pid>/numa_maps for THP

From: Gerald Schaefer
Date: Fri Feb 05 2016 - 10:34:24 EST


In gather_pte_stats() a THP pmd is cast into a pte, which is wrong because the
layouts may differ depending on the architecture. On s390 this will lead to
inaccurate numap_maps accounting in /proc because of misguided pte_present()
and pte_dirty() checks on the fake pte.

On other architectures pte_present() and pte_dirty() may work by chance, but
there will be an issue with direct-access (dax) mappings w/o underlying struct
pages when HAVE_PTE_SPECIAL is set and THP is available. In vm_normal_page()
the fake pte will be checked with pte_special() and because there is no
"special" bit in a pmd, this will always return false and the VM_PFNMAP |
VM_MIXEDMAP checking will be skipped. On dax mappings w/o struct pages, an
invalid struct page pointer will then be returned that can crash the kernel.

This patch fixes the numa_maps THP handling by introducing new "_pmd" variants
of the can_gather_numa_stats() and vm_normal_page() functions.

Signed-off-by: Gerald Schaefer <gerald.schaefer@xxxxxxxxxx>
Cc: <stable@xxxxxxxxxxxxxxx> [4.3+]
---
fs/proc/task_mmu.c | 29 ++++++++++++++++++++++++++---
include/linux/mm.h | 2 ++
mm/memory.c | 38 ++++++++++++++++++++++++++++++++++++++
3 files changed, 66 insertions(+), 3 deletions(-)

diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index fa95ab2..0cd0eed 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -1504,6 +1504,30 @@ static struct page *can_gather_numa_stats(pte_t pte, struct vm_area_struct *vma,
return page;
}

+static struct page *can_gather_numa_stats_pmd(pmd_t pmd,
+ struct vm_area_struct *vma,
+ unsigned long addr)
+{
+ struct page *page;
+ int nid;
+
+ if (!pmd_present(pmd))
+ return NULL;
+
+ page = vm_normal_page_pmd(vma, addr, pmd);
+ if (!page)
+ return NULL;
+
+ if (PageReserved(page))
+ return NULL;
+
+ nid = page_to_nid(page);
+ if (!node_isset(nid, node_states[N_MEMORY]))
+ return NULL;
+
+ return page;
+}
+
static int gather_pte_stats(pmd_t *pmd, unsigned long addr,
unsigned long end, struct mm_walk *walk)
{
@@ -1515,12 +1539,11 @@ static int gather_pte_stats(pmd_t *pmd, unsigned long addr,

ptl = pmd_trans_huge_lock(pmd, vma);
if (ptl) {
- pte_t huge_pte = *(pte_t *)pmd;
struct page *page;

- page = can_gather_numa_stats(huge_pte, vma, addr);
+ page = can_gather_numa_stats_pmd(*pmd, vma, addr);
if (page)
- gather_stats(page, md, pte_dirty(huge_pte),
+ gather_stats(page, md, pmd_dirty(*pmd),
HPAGE_PMD_SIZE/PAGE_SIZE);
spin_unlock(ptl);
return 0;
diff --git a/include/linux/mm.h b/include/linux/mm.h
index a0ad7af..7809b4f 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1111,6 +1111,8 @@ struct zap_details {

struct page *vm_normal_page(struct vm_area_struct *vma, unsigned long addr,
pte_t pte);
+struct page *vm_normal_page_pmd(struct vm_area_struct *vma, unsigned long addr,
+ pmd_t pmd);

int zap_vma_ptes(struct vm_area_struct *vma, unsigned long address,
unsigned long size);
diff --git a/mm/memory.c b/mm/memory.c
index d279350..ce0a386 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -792,6 +792,44 @@ out:
return pfn_to_page(pfn);
}

+struct page *vm_normal_page_pmd(struct vm_area_struct *vma, unsigned long addr,
+ pmd_t pmd)
+{
+ unsigned long pfn = pmd_pfn(pmd);
+
+ /*
+ * There is no pmd_special() but there may be special pmds, e.g.
+ * in a direct-access (dax) mapping, so let's just replicate the
+ * !HAVE_PTE_SPECIAL case from vm_normal_page() here.
+ */
+ if (unlikely(vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP))) {
+ if (vma->vm_flags & VM_MIXEDMAP) {
+ if (!pfn_valid(pfn))
+ return NULL;
+ goto out;
+ } else {
+ unsigned long off;
+ off = (addr - vma->vm_start) >> PAGE_SHIFT;
+ if (pfn == vma->vm_pgoff + off)
+ return NULL;
+ if (!is_cow_mapping(vma->vm_flags))
+ return NULL;
+ }
+ }
+
+ if (is_zero_pfn(pfn))
+ return NULL;
+ if (unlikely(pfn > highest_memmap_pfn))
+ return NULL;
+
+ /*
+ * NOTE! We still have PageReserved() pages in the page tables.
+ * eg. VDSO mappings can cause them to exist.
+ */
+out:
+ return pfn_to_page(pfn);
+}
+
/*
* copy one vm_area from one task to the other. Assumes the page tables
* already present in the new task to be cleared in the whole range
--
2.3.9