[RESEND,v1,3/5] mm: Fix failure to unmap pte on highmem systems

Message ID 20230511132113.80196-4-ryan.roberts@arm.com
State New
Headers
Series Encapsulate PTE contents from non-arch code |

Commit Message

Ryan Roberts May 11, 2023, 1:21 p.m. UTC
  The loser of a race to service a pte for a device private entry in the
swap path previously unlocked the ptl, but failed to unmap the pte. This
only affects highmem systems since unmapping a pte is a noop on
non-highmem systems.

Fixes: 16ce101db85d ("mm/memory.c: fix race when faulting a device private page")
Signed-off-by: Ryan Roberts <ryan.roberts@arm.com>
---
 mm/memory.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

--
2.25.1
  

Comments

Zi Yan May 11, 2023, 3:01 p.m. UTC | #1
On 11 May 2023, at 9:21, Ryan Roberts wrote:

> The loser of a race to service a pte for a device private entry in the
> swap path previously unlocked the ptl, but failed to unmap the pte. This
> only affects highmem systems since unmapping a pte is a noop on
> non-highmem systems.
>
> Fixes: 16ce101db85d ("mm/memory.c: fix race when faulting a device private page")
> Signed-off-by: Ryan Roberts <ryan.roberts@arm.com>
> ---
>  mm/memory.c | 6 ++----
>  1 file changed, 2 insertions(+), 4 deletions(-)
LGTM. Reviewed-by: Zi Yan <ziy@nvidia.com>

--
Best Regards,
Yan, Zi
  

Patch

diff --git a/mm/memory.c b/mm/memory.c
index f69fbc251198..ed429e20a1bb 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -3728,10 +3728,8 @@  vm_fault_t do_swap_page(struct vm_fault *vmf)
 			vmf->page = pfn_swap_entry_to_page(entry);
 			vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd,
 					vmf->address, &vmf->ptl);
-			if (unlikely(!pte_same(*vmf->pte, vmf->orig_pte))) {
-				spin_unlock(vmf->ptl);
-				goto out;
-			}
+			if (unlikely(!pte_same(*vmf->pte, vmf->orig_pte)))
+				goto unlock;

 			/*
 			 * Get a page reference while we know the page can't be