Le 19/09/2018 à 04:56, Aneesh Kumar K.V a écrit :

On 9/18/18 10:27 PM, Christophe Leroy wrote:

quoted

There is no point in taking the page table lock as
pte_frag is always NULL when we have only one fragment.

Signed-off-by: Christophe Leroy <redacted>
---
  arch/powerpc/mm/pgtable-frag.c | 3 +++
  1 file changed, 3 insertions(+)

diff --git a/arch/powerpc/mm/pgtable-frag.c 

b/arch/powerpc/mm/pgtable-frag.c
index bc924822dcd6..ab4910e92aaf 100644

--- a/arch/powerpc/mm/pgtable-frag.c
+++ b/arch/powerpc/mm/pgtable-frag.c

@@ -85,6 +85,9 @@ static pte_t *get_pte_from_cache(struct mm_struct *mm)

  {
      void *pte_frag, *ret;

+    if (PTE_FRAG_NR == 1)
+        return NULL;
+
      spin_lock(&mm->page_table_lock);
      ret = mm->context.pte_frag;
      if (ret) {

May be update get_pmd_from_cache too?

Ok, done in v5

Christophe