aboutsummaryrefslogtreecommitdiff
path: root/mm/memory.c
diff options
context:
space:
mode:
Diffstat (limited to 'mm/memory.c')
-rw-r--r--mm/memory.c28
1 files changed, 24 insertions, 4 deletions
diff --git a/mm/memory.c b/mm/memory.c
index de7fd933415..051e157738d 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1462,6 +1462,16 @@ int zap_vma_ptes(struct vm_area_struct *vma, unsigned long address,
}
EXPORT_SYMBOL_GPL(zap_vma_ptes);
+/*
+ * FOLL_FORCE can write to even unwritable pte's, but only
+ * after we've gone through a COW cycle and they are dirty.
+ */
+static inline bool can_follow_write_pte(pte_t pte, unsigned int flags)
+{
+ return pte_write(pte) ||
+ ((flags & FOLL_FORCE) && (flags & FOLL_COW) && pte_dirty(pte));
+}
+
/**
* follow_page_mask - look up a page descriptor from a user-virtual address
* @vma: vm_area_struct mapping @address
@@ -1569,7 +1579,7 @@ split_fallthrough:
}
if ((flags & FOLL_NUMA) && pte_numa(pte))
goto no_page;
- if ((flags & FOLL_WRITE) && !pte_write(pte))
+ if ((flags & FOLL_WRITE) && !can_follow_write_pte(pte, flags))
goto unlock;
page = vm_normal_page(vma, address, pte);
@@ -1877,7 +1887,7 @@ long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
*/
if ((ret & VM_FAULT_WRITE) &&
!(vma->vm_flags & VM_WRITE))
- foll_flags &= ~FOLL_WRITE;
+ foll_flags |= FOLL_COW;
cond_resched();
}
@@ -3836,8 +3846,18 @@ retry:
if (unlikely(pmd_none(*pmd)) &&
unlikely(__pte_alloc(mm, vma, pmd, address)))
return VM_FAULT_OOM;
- /* if an huge pmd materialized from under us just retry later */
- if (unlikely(pmd_trans_huge(*pmd)))
+ /*
+ * If a huge pmd materialized under us just retry later. Use
+ * pmd_trans_unstable() instead of pmd_trans_huge() to ensure the pmd
+ * didn't become pmd_trans_huge under us and then back to pmd_none, as
+ * a result of MADV_DONTNEED running immediately after a huge pmd fault
+ * in a different thread of this mm, in turn leading to a misleading
+ * pmd_trans_huge() retval. All we have to ensure is that it is a
+ * regular pmd that we can walk with pte_offset_map() and we can do that
+ * through an atomic read in C, which is what pmd_trans_unstable()
+ * provides.
+ */
+ if (unlikely(pmd_trans_unstable(pmd)))
return 0;
/*
* A regular pmd is established and it can't morph into a huge pmd