[PATCH] x86_64: pageattr remove __put_page
authorNick Piggin <npiggin@suse.de>
Wed, 22 Mar 2006 08:08:33 +0000 (00:08 -0800)
committerLinus Torvalds <torvalds@g5.osdl.org>
Wed, 22 Mar 2006 15:54:01 +0000 (07:54 -0800)
Remove page_count and __put_page from x86-64 pageattr

Signed-off-by: Nick Piggin <npiggin@suse.de>
Acked-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
arch/x86_64/mm/pageattr.c

index efe5af14c7db1f9df8900ecf6f85b0419a89bb6c..531ad21447b1c95e4e5a9d9bc020f13784b01d7f 100644 (file)
@@ -45,6 +45,13 @@ static struct page *split_large_page(unsigned long address, pgprot_t prot,
        pte_t *pbase;
        if (!base) 
                return NULL;
+       /*
+        * page_private is used to track the number of entries in
+        * the page table page have non standard attributes.
+        */
+       SetPagePrivate(base);
+       page_private(base) = 0;
+
        address = __pa(address);
        addr = address & LARGE_PAGE_MASK; 
        pbase = (pte_t *)page_address(base);
@@ -124,8 +131,8 @@ __change_page_attr(unsigned long address, unsigned long pfn, pgprot_t prot,
                        set_pte(kpte, pfn_pte(pfn, prot));
                } else {
                        /*
-                        * split_large_page will take the reference for this change_page_attr
-                        * on the split page.
+                        * split_large_page will take the reference for this
+                        * change_page_attr on the split page.
                         */
 
                        struct page *split;
@@ -137,23 +144,20 @@ __change_page_attr(unsigned long address, unsigned long pfn, pgprot_t prot,
                        set_pte(kpte,mk_pte(split, ref_prot2));
                        kpte_page = split;
                }       
-               get_page(kpte_page);
+               page_private(kpte_page)++;
        } else if ((kpte_flags & _PAGE_PSE) == 0) { 
                set_pte(kpte, pfn_pte(pfn, ref_prot));
-               __put_page(kpte_page);
+               BUG_ON(page_private(kpte_page) == 0);
+               page_private(kpte_page)--;
        } else
                BUG();
 
        /* on x86-64 the direct mapping set at boot is not using 4k pages */
        BUG_ON(PageReserved(kpte_page));
 
-       switch (page_count(kpte_page)) {
-       case 1:
+       if (page_private(kpte_page) == 0) {
                save_page(kpte_page);
                revert_page(address, ref_prot);
-               break;
-       case 0:
-               BUG(); /* memleak and failed 2M page regeneration */
        }
        return 0;
 } 
@@ -216,6 +220,7 @@ void global_flush_tlb(void)
        while (dpage) {
                struct page *tmp = dpage;
                dpage = (struct page *)dpage->lru.next;
+               ClearPagePrivate(tmp);
                __free_page(tmp);
        } 
 }