From: Jan Glauber <jglauber@redhat.com>
Date: Wed, 23 Jan 2008 12:52:40 +0100
Subject: [mm] introduce more huge pte handling functions
Message-id: 1201089160.8618.7.camel@localhost.localdomain
O-Subject: Re: [RHEL5.2 PATCH] 2/3: add huge pte handling functions
Bugzilla: 318951

Summary:     Common Code: Introduce more huge pte handling functions.
Description: This adds two new ARCH_HAS_xxx flags to include/linux/hugetlb.h
             and introduces some architecture-dependent huge pte handling
             functions. This is necessary on s390 because a huge pte needs to
             be handled different than a standard pte.
             Also, two new hooks are added for preparing and releasing huge
             pages. This is needed on s390 to implement shared hugetlbfs page
             tables in software emulation mode.
             All of these changes are a nop on other architectures.

diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index 1e80feb..884467c 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -73,6 +73,24 @@ static inline int prepare_hugepage_range(unsigned long addr, unsigned long len)
 int prepare_hugepage_range(unsigned long addr, unsigned long len);
 #endif
 
+#ifndef ARCH_HAS_HUGE_PTE_TYPE
+#define huge_pte_none(pte)			pte_none(pte)
+#define huge_pte_wrprotect(pte)			pte_wrprotect(pte)
+#define huge_ptep_set_wrprotect(mm, addr, ptep)	\
+	ptep_set_wrprotect(mm, addr, ptep)
+#define huge_ptep_set_access_flags(vma, addr, ptep, pte, dirty)	\
+	ptep_set_access_flags(vma, addr, ptep, pte, dirty)
+#define huge_ptep_get(ptep)			(*ptep)
+#endif
+
+#ifndef ARCH_HAS_PREPARE_HUGEPAGE
+#define arch_prepare_hugepage(page)		0
+#define arch_release_hugepage(page)		do { } while (0)
+#else
+int arch_prepare_hugepage(struct page *page);
+void arch_release_hugepage(struct page *page);
+#endif
+
 #ifndef ARCH_HAS_SETCLEAR_HUGE_PTE
 #define set_huge_pte_at(mm, addr, ptep, pte)	set_pte_at(mm, addr, ptep, pte)
 #define huge_ptep_get_and_clear(mm, addr, ptep) ptep_get_and_clear(mm, addr, ptep)
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index e4157ca..3992662 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -107,6 +107,10 @@ static struct page *alloc_fresh_huge_page_node(int nid)
 	page = alloc_pages_thisnode(nid, GFP_HIGHUSER|__GFP_COMP|__GFP_NOWARN,
 					HUGETLB_PAGE_ORDER);
 	if (page) {
+		if (arch_prepare_hugepage(page)) {
+			__free_pages(page, HUGETLB_PAGE_ORDER);
+			return 0;
+		}
 		set_compound_page_dtor(page, free_huge_page);
 		spin_lock(&hugetlb_lock);
 		nr_huge_pages++;
@@ -218,6 +222,7 @@ static void update_and_free_page(struct page *page)
 	}
 	page[1].lru.next = NULL;
 	set_page_refcounted(page);
+	arch_release_hugepage(page);
 	__free_pages(page, HUGETLB_PAGE_ORDER);
 }
 
@@ -332,7 +337,7 @@ static pte_t make_huge_pte(struct vm_area_struct *vma, struct page *page,
 		entry =
 		    pte_mkwrite(pte_mkdirty(mk_pte(page, vma->vm_page_prot)));
 	} else {
-		entry = pte_wrprotect(mk_pte(page, vma->vm_page_prot));
+		entry = huge_pte_wrprotect(mk_pte(page, vma->vm_page_prot));
 	}
 	entry = pte_mkyoung(entry);
 	entry = pte_mkhuge(entry);
@@ -345,8 +350,8 @@ static void set_huge_ptep_writable(struct vm_area_struct *vma,
 {
 	pte_t entry;
 
-	entry = pte_mkwrite(pte_mkdirty(*ptep));
-	ptep_set_access_flags(vma, address, ptep, entry, 1);
+	entry = pte_mkwrite(pte_mkdirty(huge_ptep_get(ptep)));
+	huge_ptep_set_access_flags(vma, address, ptep, entry, 1);
 	update_mmu_cache(vma, address, entry);
 	lazy_mmu_prot_update(entry);
 }
@@ -374,10 +379,10 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
 			continue;
 		spin_lock(&dst->page_table_lock);
 		spin_lock(&src->page_table_lock);
-		if (!pte_none(*src_pte)) {
+		if (!huge_pte_none(huge_ptep_get(src_pte))) {
 			if (cow)
-				ptep_set_wrprotect(src, addr, src_pte);
-			entry = *src_pte;
+				huge_ptep_set_wrprotect(src, addr, src_pte);
+			entry = huge_ptep_get(src_pte);
 			ptepage = pte_page(entry);
 			get_page(ptepage);
 			set_huge_pte_at(dst, addr, dst_pte, entry);
@@ -414,7 +419,7 @@ void __unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start,
 			continue;
 
 		pte = huge_ptep_get_and_clear(mm, address, ptep);
-		if (pte_none(pte))
+		if (huge_pte_none(pte))
 			continue;
 
 		page = pte_page(pte);
@@ -471,7 +476,7 @@ static int hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma,
 	spin_lock(&mm->page_table_lock);
 
 	ptep = huge_pte_offset(mm, address & HPAGE_MASK);
-	if (likely(pte_same(*ptep, pte))) {
+	if (likely(pte_same(huge_ptep_get(ptep), pte))) {
 		/* Break COW */
 		set_huge_pte_at(mm, address, ptep,
 				make_huge_pte(vma, new_page, 1));
@@ -538,7 +543,7 @@ retry:
 		goto backout;
 
 	ret = VM_FAULT_MINOR;
-	if (!pte_none(*ptep))
+	if (!huge_pte_none(huge_ptep_get(ptep)))
 		goto backout;
 
 	new_pte = make_huge_pte(vma, page, ((vma->vm_flags & VM_WRITE)
@@ -581,8 +586,8 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 	 * the same page in the page cache.
 	 */
 	mutex_lock(&hugetlb_instantiation_mutex);
-	entry = *ptep;
-	if (pte_none(entry)) {
+	entry = huge_ptep_get(ptep);
+	if (huge_pte_none(entry)) {
 		ret = hugetlb_no_page(mm, vma, address, ptep, write_access);
 		mutex_unlock(&hugetlb_instantiation_mutex);
 		return ret;
@@ -592,7 +597,7 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 
 	spin_lock(&mm->page_table_lock);
 	/* Check for a racing update before calling hugetlb_cow */
-	if (likely(pte_same(entry, *ptep)))
+	if (likely(pte_same(entry, huge_ptep_get(ptep))))
 		if (write_access && !pte_write(entry))
 			ret = hugetlb_cow(mm, vma, address, ptep, entry);
 	spin_unlock(&mm->page_table_lock);
@@ -621,7 +626,7 @@ int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		 */
 		pte = huge_pte_offset(mm, vaddr & HPAGE_MASK);
 
-		if (!pte || pte_none(*pte)) {
+		if (!pte || huge_pte_none(huge_ptep_get(pte))) {
 			int ret;
 
 			spin_unlock(&mm->page_table_lock);
@@ -637,7 +642,7 @@ int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		}
 
 		pfn_offset = (vaddr & ~HPAGE_MASK) >> PAGE_SHIFT;
-		page = pte_page(*pte);
+		page = pte_page(huge_ptep_get(pte));
 same_page:
 		if (pages) {
 			get_page(page);
@@ -686,7 +691,7 @@ void hugetlb_change_protection(struct vm_area_struct *vma,
 			continue;
 		if (huge_pmd_unshare(mm, &address, ptep))
 			continue;
-		if (!pte_none(*ptep)) {
+		if (!huge_pte_none(huge_ptep_get(ptep))) {
 			pte = huge_ptep_get_and_clear(mm, address, ptep);
 			pte = pte_mkhuge(pte_modify(pte, newprot));
 			set_huge_pte_at(mm, address, ptep, pte);