On Thu, Sep 17, 2015 at 10:58:59AM +0200, Martin Schwidefsky wrote:
> Fixes a regression introduced with commit 179ef71cbc085252
> "mm: save soft-dirty bits on swapped pages"
>
> The maybe_same_pte() function is used to match a swap pte independent
> of the swap software dirty bit set with pte_swp_mksoft_dirty().
>
> For CONFIG_HAVE_ARCH_SOFT_DIRTY=y but CONFIG_MEM_SOFT_DIRTY=n the
> software dirty bit may be set but maybe_same_pte() will not recognize
> a software dirty swap pte. Due to this a 'swapoff -a' will hang.
>
> The straightforward solution is to replace CONFIG_MEM_SOFT_DIRTY
> with HAVE_ARCH_SOFT_DIRTY in maybe_same_pte().
>
> Cc: linux-mm@xxxxxxxxx
> Cc: Cyrill Gorcunov <gorcunov@xxxxxxxxx>
> Cc: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
> Cc: Michal Hocko <mhocko@xxxxxxx>
> Signed-off-by: Martin Schwidefsky <schwidefsky@xxxxxxxxxx>
> ---
> mm/swapfile.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/mm/swapfile.c b/mm/swapfile.c
> index 5887731..bf7da58 100644
> --- a/mm/swapfile.c
> +++ b/mm/swapfile.c
> @@ -1113,7 +1113,7 @@ unsigned int count_swap_pages(int type, int free)
>
> static inline int maybe_same_pte(pte_t pte, pte_t swp_pte)
> {
> -#ifdef CONFIG_MEM_SOFT_DIRTY
> +#ifdef CONFIG_HAVE_ARCH_SOFT_DIRTY
> /*
> * When pte keeps soft dirty bit the pte generated
> * from swap entry does not has it, still it's same
You know, I seem to miss how this might help. If CONFIG_MEM_SOFT_DIRTY=n
then all related helpers are nop'ed.
In particular in the commit you mentioned
+static inline int maybe_same_pte(pte_t pte, pte_t swp_pte)
+{
+#ifdef CONFIG_MEM_SOFT_DIRTY
+ /*
+ * When pte keeps soft dirty bit the pte generated
+ * from swap entry does not has it, still it's same
+ * pte from logical point of view.
+ */
+ pte_t swp_pte_dirty = pte_swp_mksoft_dirty(swp_pte);
+ return pte_same(pte, swp_pte) || pte_same(pte, swp_pte_dirty);
+#else
+ return pte_same(pte, swp_pte);
+#endif
+}
+
...
@@ -892,7 +907,7 @@ static int unuse_pte(struct vm_area_struct *vma, pmd_t *pmd,
}
pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
- if (unlikely(!pte_same(*pte, swp_entry_to_pte(entry)))) {
+ if (unlikely(!maybe_same_pte(*pte, swp_entry_to_pte(entry)))) {
mem_cgroup_cancel_charge_swapin(memcg);
ret = 0;
goto out;
@@ -947,7 +962,7 @@ static int unuse_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
* swapoff spends a _lot_ of time in this loop!
* Test inline before going to call unuse_pte.
*/
- if (unlikely(pte_same(*pte, swp_pte))) {
+ if (unlikely(maybe_same_pte(*pte, swp_pte))) {
pte_unmap(pte);
ret = unuse_pte(vma, pmd, addr, entry, page);
if (ret)
Thus when CONFIG_MEM_SOFT_DIRTY = n, the unuse_pte will be the same
as it were without the patch, calling pte_same.
Now to the bit itself
#ifdef CONFIG_MEM_SOFT_DIRTY
#define _PAGE_SWP_SOFT_DIRTY_PAGE_PSE
#else
#define _PAGE_SWP_SOFT_DIRTY_PAGE_PSE(_AT(pteval_t, 0))
#endif
it's 0 if CONFIG_MEM_SOFT_DIRTY=n, so any setup of this
bit will simply become nop
#ifdef CONFIG_HAVE_ARCH_SOFT_DIRTY
static inline pte_t pte_swp_mksoft_dirty(pte_t pte)
{
return pte_set_flags(pte, _PAGE_SWP_SOFT_DIRTY);
}
static inline int pte_swp_soft_dirty(pte_t pte)
{
return pte_flags(pte) & _PAGE_SWP_SOFT_DIRTY;
}
static inline pte_t pte_swp_clear_soft_dirty(pte_t pte)
{
return pte_clear_flags(pte, _PAGE_SWP_SOFT_DIRTY);
}
#endif
So I fear I'm lost where this "set" of the bit comes from
when CONFIG_MEM_SOFT_DIRTY=n.
Martin, could you please elaborate? Seems I'm missing
something obvious.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>