linux-4.4.1/__pte_alloc() の変更点

• 追加された行はこの色です。
• 削除された行はこの色です。
• linux-4.4.1/__pte_alloc() へ行く。
• linux-4.4.1/__pte_alloc() の差分を削除

*参照元 [#w8d569df]
#backlinks

*説明 [#t96f1f13]
-パス: [[linux-4.4.1/mm/memory.c]]

-FIXME: これは何？
--説明


**引数 [#k90c29a7]
-struct mm_struct *mm
--
--[[linux-4.4.1/mm_struct]]
-struct vm_area_struct *vma
--
--[[linux-4.4.1/vm_area_struct]]
-pmd_t *pmd
--
--[[linux-4.4.1/pmd_t]]
-unsigned long address
--


**返り値 [#u2274928]
-int
--


**参考 [#cbd8e90f]


*実装 [#gfec7a44]
 int __pte_alloc(struct mm_struct *mm, struct vm_area_struct *vma,
                 pmd_t *pmd, unsigned long address)
 {
         spinlock_t *ptl;
         pgtable_t new = pte_alloc_one(mm, address);
         int wait_split_huge_page;
         if (!new)
                 return -ENOMEM;
 
-
--[[linux-4.4.1/spinlock_t]]
--[[linux-4.4.1/pgtable_t]]
--[[linux-4.4.1/pte_alloc_one()]]

         /*
          * Ensure all pte setup (eg. pte page lock and page clearing) are
          * visible before the pte is made visible to other CPUs by being
          * put into page tables.
          *
          * The other side of the story is the pointer chasing in the page
          * table walking code (when walking the page table without locking;
          * ie. most of the time). Fortunately, these data accesses consist
          * of a chain of data-dependent loads, meaning most CPUs (alpha
          * being the notable exception) will already guarantee loads are
          * seen in-order. See the alpha page table accessors for the
          * smp_read_barrier_depends() barriers in page table walking code.
          */
         smp_wmb(); /* Could be smp_wmb__xxx(before|after)_spin_lock */
 
-
--[[linux-4.4.1/smp_wmb()]]

         ptl = pmd_lock(mm, pmd);
         wait_split_huge_page = 0;
-
--[[linux-4.4.1/pmd_lock()]]

         if (likely(pmd_none(*pmd))) {   /* Has another populated it ? */
                 atomic_long_inc(&mm->nr_ptes);
                 pmd_populate(mm, pmd, new);
                 new = NULL;
-
--[[linux-4.4.1/likely()]]
--[[linux-4.4.1/pmd_none()]]
--[[linux-4.4.1/atomic_long_inc()]]
--[[linux-4.4.1/pmd_populate()]]

         } else if (unlikely(pmd_trans_splitting(*pmd)))
                 wait_split_huge_page = 1;
-
--[[linux-4.4.1/unlikely()]]
--[[linux-4.4.1/pmd_trans_splitting()]]

         spin_unlock(ptl);
-
--[[linux-4.4.1/spin_unlock()]]

         if (new)
                 pte_free(mm, new);
-
--[[linux-4.4.1/pte_free()]]

         if (wait_split_huge_page)
                 wait_split_huge_page(vma->anon_vma, pmd);
-
--[[linux-4.4.1/wait_split_huge_page()]]

         return 0;
 }


*コメント [#qc949721]