[v7,10/12] s390/mm: hugetlb pages within a gmap can not be freed

Commit Message

Janosch Frank July 17, 2018, 12:44 p.m. UTC

Guests backed by huge pages could theoretically free unused pages via
the diagnose 10 instruction. We currently don't allow that, so we
don't have to refault it once it's needed again.

Signed-off-by: Dominik Dingel <[e-mail]>
Reviewed-by: Martin Schwidefsky <[e-mail]>
Reviewed-by: David Hildenbrand <[e-mail]>
---
 arch/s390/mm/gmap.c | 6 ++++++
 1 file changed, 6 insertions(+)

Patch

diff --git a/arch/s390/mm/gmap.c b/arch/s390/mm/gmap.c
index 46875ea8358d..a5a60afae8df 100644
--- a/arch/s390/mm/gmap.c
+++ b/arch/s390/mm/gmap.c
@@ -707,6 +707,12 @@  void gmap_discard(struct gmap *gmap, unsigned long from, unsigned long to)
 		vmaddr |= gaddr & ~PMD_MASK;
 		/* Find vma in the parent mm */
 		vma = find_vma(gmap->mm, vmaddr);
+		/*
+		 * We do not discard pages that are backed by
+		 * hugetlbfs, so we don't have to refault them.
+		 */
+		if (vma && is_vm_hugetlb_page(vma))
+			continue;
 		size = min(to - gaddr, PMD_SIZE - (gaddr & ~PMD_MASK));
 		zap_page_range(vma, vmaddr, size);
 	}