powerpc: add support for folded p4d page tables

Implement primitives necessary for the 4th level folding, add walks of p4d level where appropriate and replace 5level-fixup.h with pgtable-nop4d.h. [rppt@linux.ibm.com: powerpc/xmon: drop unused pgdir varialble in show_pte() function] Link: http://lkml.kernel.org/r/20200519181454.GI1059226@linux.ibm.com [rppt@linux.ibm.com; build fix] Link: http://lkml.kernel.org/r/20200423141845.GI13521@linux.ibm.com Signed-off-by: Mike Rapoport <rppt@linux.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Tested-by: Christophe Leroy <christophe.leroy@c-s.fr> # 8xx and 83xx Cc: Arnd Bergmann <arnd@arndb.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Brian Cain <bcain@codeaurora.org> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Geert Uytterhoeven <geert+renesas@glider.be> Cc: Guan Xuetao <gxt@pku.edu.cn> Cc: James Morse <james.morse@arm.com> Cc: Jonas Bonn <jonas@southpole.se> Cc: Julien Thierry <julien.thierry.kdev@gmail.com> Cc: Ley Foon Tan <ley.foon.tan@intel.com> Cc: Marc Zyngier <maz@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Paul Mackerras <paulus@samba.org> Cc: Rich Felker <dalias@libc.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Stafford Horne <shorne@gmail.com> Cc: Stefan Kristiansson <stefan.kristiansson@saunalahti.fi> Cc: Suzuki K Poulose <suzuki.poulose@arm.com> Cc: Tony Luck <tony.luck@intel.com> Cc: Will Deacon <will@kernel.org> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Link: http://lkml.kernel.org/r/20200414153455.21744-9-rppt@kernel.org Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-06-04 16:46:44 -07:00
--- a/arch/powerpc/mm/book3s64/hash_pgtable.c
+++ b/arch/powerpc/mm/book3s64/hash_pgtable.c
@@ -148,6 +148,7 @@ void hash__vmemmap_remove_mapping(unsigned long start,
 int hash__map_kernel_page(unsigned long ea, unsigned long pa, pgprot_t prot)
 {
 	pgd_t *pgdp;
+	p4d_t *p4dp;
 	pud_t *pudp;
 	pmd_t *pmdp;
 	pte_t *ptep;
@@ -155,7 +156,8 @@ int hash__map_kernel_page(unsigned long ea, unsigned long pa, pgprot_t prot)
 	BUILD_BUG_ON(TASK_SIZE_USER64 > H_PGTABLE_RANGE);
 	if (slab_is_available()) {
 		pgdp = pgd_offset_k(ea);
-		pudp = pud_alloc(&init_mm, pgdp, ea);
+		p4dp = p4d_offset(pgdp, ea);
+		pudp = pud_alloc(&init_mm, p4dp, ea);
 		if (!pudp)
 			return -ENOMEM;
 		pmdp = pmd_alloc(&init_mm, pudp, ea);
--- a/arch/powerpc/mm/book3s64/radix_pgtable.c
+++ b/arch/powerpc/mm/book3s64/radix_pgtable.c
@@ -65,17 +65,19 @@ static int early_map_kernel_page(unsigned long ea, unsigned long pa,
 {
 	unsigned long pfn = pa >> PAGE_SHIFT;
 	pgd_t *pgdp;
+	p4d_t *p4dp;
 	pud_t *pudp;
 	pmd_t *pmdp;
 	pte_t *ptep;

 	pgdp = pgd_offset_k(ea);
-	if (pgd_none(*pgdp)) {
+	p4dp = p4d_offset(pgdp, ea);
+	if (p4d_none(*p4dp)) {
 		pudp = early_alloc_pgtable(PUD_TABLE_SIZE, nid,
 						region_start, region_end);
-		pgd_populate(&init_mm, pgdp, pudp);
+		p4d_populate(&init_mm, p4dp, pudp);
 	}
-	pudp = pud_offset(pgdp, ea);
+	pudp = pud_offset(p4dp, ea);
 	if (map_page_size == PUD_SIZE) {
 		ptep = (pte_t *)pudp;
 		goto set_the_pte;
@@ -115,6 +117,7 @@ static int __map_kernel_page(unsigned long ea, unsigned long pa,
 {
 	unsigned long pfn = pa >> PAGE_SHIFT;
 	pgd_t *pgdp;
+	p4d_t *p4dp;
 	pud_t *pudp;
 	pmd_t *pmdp;
 	pte_t *ptep;
@@ -137,7 +140,8 @@ static int __map_kernel_page(unsigned long ea, unsigned long pa,
 	 * boot.
 	 */
 	pgdp = pgd_offset_k(ea);
-	pudp = pud_alloc(&init_mm, pgdp, ea);
+	p4dp = p4d_offset(pgdp, ea);
+	pudp = pud_alloc(&init_mm, p4dp, ea);
 	if (!pudp)
 		return -ENOMEM;
 	if (map_page_size == PUD_SIZE) {
@@ -174,6 +178,7 @@ void radix__change_memory_range(unsigned long start, unsigned long end,
 {
 	unsigned long idx;
 	pgd_t *pgdp;
+	p4d_t *p4dp;
 	pud_t *pudp;
 	pmd_t *pmdp;
 	pte_t *ptep;
@@ -186,7 +191,8 @@ void radix__change_memory_range(unsigned long start, unsigned long end,

 	for (idx = start; idx < end; idx += PAGE_SIZE) {
 		pgdp = pgd_offset_k(idx);
-		pudp = pud_alloc(&init_mm, pgdp, idx);
+		p4dp = p4d_offset(pgdp, idx);
+		pudp = pud_alloc(&init_mm, p4dp, idx);
 		if (!pudp)
 			continue;
 		if (pud_is_leaf(*pudp)) {
@@ -850,6 +856,7 @@ static void __meminit remove_pagetable(unsigned long start, unsigned long end)
 	unsigned long addr, next;
 	pud_t *pud_base;
 	pgd_t *pgd;
+	p4d_t *p4d;

 	spin_lock(&init_mm.page_table_lock);

@@ -857,15 +864,16 @@ static void __meminit remove_pagetable(unsigned long start, unsigned long end)
 		next = pgd_addr_end(addr, end);

 		pgd = pgd_offset_k(addr);
-		if (!pgd_present(*pgd))
+		p4d = p4d_offset(pgd, addr);
+		if (!p4d_present(*p4d))
 			continue;

-		if (pgd_is_leaf(*pgd)) {
-			split_kernel_mapping(addr, end, PGDIR_SIZE, (pte_t *)pgd);
+		if (p4d_is_leaf(*p4d)) {
+			split_kernel_mapping(addr, end, P4D_SIZE, (pte_t *)p4d);
 			continue;
 		}

-		pud_base = (pud_t *)pgd_page_vaddr(*pgd);
+		pud_base = (pud_t *)p4d_page_vaddr(*p4d);
 		remove_pud_table(pud_base, addr, next);
 	}

--- a/arch/powerpc/mm/book3s64/subpage_prot.c
+++ b/arch/powerpc/mm/book3s64/subpage_prot.c
@@ -54,15 +54,17 @@ static void hpte_flush_range(struct mm_struct *mm, unsigned long addr,
 			     int npages)
 {
 	pgd_t *pgd;
+	p4d_t *p4d;
 	pud_t *pud;
 	pmd_t *pmd;
 	pte_t *pte;
 	spinlock_t *ptl;

 	pgd = pgd_offset(mm, addr);
-	if (pgd_none(*pgd))
+	p4d = p4d_offset(pgd, addr);
+	if (p4d_none(*p4d))
 		return;
-	pud = pud_offset(pgd, addr);
+	pud = pud_offset(p4d, addr);
 	if (pud_none(*pud))
 		return;
 	pmd = pmd_offset(pud, addr);
--- a/arch/powerpc/mm/hugetlbpage.c
+++ b/arch/powerpc/mm/hugetlbpage.c
@@ -119,6 +119,7 @@ static int __hugepte_alloc(struct mm_struct *mm, hugepd_t *hpdp,
 pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr, unsigned long sz)
 {
 	pgd_t *pg;
+	p4d_t *p4;
 	pud_t *pu;
 	pmd_t *pm;
 	hugepd_t *hpdp = NULL;
@@ -128,20 +129,21 @@ pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr, unsigned long sz

 	addr &= ~(sz-1);
 	pg = pgd_offset(mm, addr);
+	p4 = p4d_offset(pg, addr);

 #ifdef CONFIG_PPC_BOOK3S_64
 	if (pshift == PGDIR_SHIFT)
 		/* 16GB huge page */
-		return (pte_t *) pg;
+		return (pte_t *) p4;
 	else if (pshift > PUD_SHIFT) {
 		/*
 		 * We need to use hugepd table
 		 */
 		ptl = &mm->page_table_lock;
-		hpdp = (hugepd_t *)pg;
+		hpdp = (hugepd_t *)p4;
 	} else {
 		pdshift = PUD_SHIFT;
-		pu = pud_alloc(mm, pg, addr);
+		pu = pud_alloc(mm, p4, addr);
 		if (!pu)
 			return NULL;
 		if (pshift == PUD_SHIFT)
@@ -166,10 +168,10 @@ pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr, unsigned long sz
 #else
 	if (pshift >= PGDIR_SHIFT) {
 		ptl = &mm->page_table_lock;
-		hpdp = (hugepd_t *)pg;
+		hpdp = (hugepd_t *)p4;
 	} else {
 		pdshift = PUD_SHIFT;
-		pu = pud_alloc(mm, pg, addr);
+		pu = pud_alloc(mm, p4, addr);
 		if (!pu)
 			return NULL;
 		if (pshift >= PUD_SHIFT) {
@@ -390,7 +392,7 @@ static void hugetlb_free_pmd_range(struct mmu_gather *tlb, pud_t *pud,
 	mm_dec_nr_pmds(tlb->mm);
 }

-static void hugetlb_free_pud_range(struct mmu_gather *tlb, pgd_t *pgd,
+static void hugetlb_free_pud_range(struct mmu_gather *tlb, p4d_t *p4d,
 				   unsigned long addr, unsigned long end,
 				   unsigned long floor, unsigned long ceiling)
 {
@@ -400,7 +402,7 @@ static void hugetlb_free_pud_range(struct mmu_gather *tlb, pgd_t *pgd,

 	start = addr;
 	do {
-		pud = pud_offset(pgd, addr);
+		pud = pud_offset(p4d, addr);
 		next = pud_addr_end(addr, end);
 		if (!is_hugepd(__hugepd(pud_val(*pud)))) {
 			if (pud_none_or_clear_bad(pud))
@@ -435,8 +437,8 @@ static void hugetlb_free_pud_range(struct mmu_gather *tlb, pgd_t *pgd,
 	if (end - 1 > ceiling - 1)
 		return;

-	pud = pud_offset(pgd, start);
-	pgd_clear(pgd);
+	pud = pud_offset(p4d, start);
+	p4d_clear(p4d);
 	pud_free_tlb(tlb, pud, start);
 	mm_dec_nr_puds(tlb->mm);
 }
@@ -449,6 +451,7 @@ void hugetlb_free_pgd_range(struct mmu_gather *tlb,
 			    unsigned long floor, unsigned long ceiling)
 {
 	pgd_t *pgd;
+	p4d_t *p4d;
 	unsigned long next;

 	/*
@@ -471,10 +474,11 @@ void hugetlb_free_pgd_range(struct mmu_gather *tlb,
 	do {
 		next = pgd_addr_end(addr, end);
 		pgd = pgd_offset(tlb->mm, addr);
+		p4d = p4d_offset(pgd, addr);
 		if (!is_hugepd(__hugepd(pgd_val(*pgd)))) {
-			if (pgd_none_or_clear_bad(pgd))
+			if (p4d_none_or_clear_bad(p4d))
 				continue;
-			hugetlb_free_pud_range(tlb, pgd, addr, next, floor, ceiling);
+			hugetlb_free_pud_range(tlb, p4d, addr, next, floor, ceiling);
 		} else {
 			unsigned long more;
 			/*
@@ -487,7 +491,7 @@ void hugetlb_free_pgd_range(struct mmu_gather *tlb,
 			if (more > next)
 				next = more;

-			free_hugepd_range(tlb, (hugepd_t *)pgd, PGDIR_SHIFT,
+			free_hugepd_range(tlb, (hugepd_t *)p4d, PGDIR_SHIFT,
 					  addr, next, floor, ceiling);
 		}
 	} while (addr = next, addr != end);
--- a/arch/powerpc/mm/kasan/kasan_init_32.c
+++ b/arch/powerpc/mm/kasan/kasan_init_32.c
@@ -121,7 +121,7 @@ static void __init kasan_unmap_early_shadow_vmalloc(void)
 	phys_addr_t pa = __pa(kasan_early_shadow_page);

 	for (k_cur = k_start & PAGE_MASK; k_cur < k_end; k_cur += PAGE_SIZE) {
-		pmd_t *pmd = pmd_offset(pud_offset(pgd_offset_k(k_cur), k_cur), k_cur);
+		pmd_t *pmd = pmd_ptr_k(k_cur);
 		pte_t *ptep = pte_offset_kernel(pmd, k_cur);

 		if ((pte_val(*ptep) & PTE_RPN_MASK) != pa)
--- a/arch/powerpc/mm/nohash/book3e_pgtable.c
+++ b/arch/powerpc/mm/nohash/book3e_pgtable.c
@@ -73,6 +73,7 @@ static void __init *early_alloc_pgtable(unsigned long size)
 int __ref map_kernel_page(unsigned long ea, unsigned long pa, pgprot_t prot)
 {
 	pgd_t *pgdp;
+	p4d_t *p4dp;
 	pud_t *pudp;
 	pmd_t *pmdp;
 	pte_t *ptep;
@@ -80,7 +81,8 @@ int __ref map_kernel_page(unsigned long ea, unsigned long pa, pgprot_t prot)
 	BUILD_BUG_ON(TASK_SIZE_USER64 > PGTABLE_RANGE);
 	if (slab_is_available()) {
 		pgdp = pgd_offset_k(ea);
-		pudp = pud_alloc(&init_mm, pgdp, ea);
+		p4dp = p4d_offset(pgdp, ea);
+		pudp = pud_alloc(&init_mm, p4dp, ea);
 		if (!pudp)
 			return -ENOMEM;
 		pmdp = pmd_alloc(&init_mm, pudp, ea);
@@ -91,13 +93,12 @@ int __ref map_kernel_page(unsigned long ea, unsigned long pa, pgprot_t prot)
 			return -ENOMEM;
 	} else {
 		pgdp = pgd_offset_k(ea);
-#ifndef __PAGETABLE_PUD_FOLDED
-		if (pgd_none(*pgdp)) {
-			pudp = early_alloc_pgtable(PUD_TABLE_SIZE);
-			pgd_populate(&init_mm, pgdp, pudp);
+		p4dp = p4d_offset(pgdp, ea);
+		if (p4d_none(*p4dp)) {
+			pmdp = early_alloc_pgtable(PMD_TABLE_SIZE);
+			p4d_populate(&init_mm, p4dp, pmdp);
 		}
-#endif /* !__PAGETABLE_PUD_FOLDED */
-		pudp = pud_offset(pgdp, ea);
+		pudp = pud_offset(p4dp, ea);
 		if (pud_none(*pudp)) {
 			pmdp = early_alloc_pgtable(PMD_TABLE_SIZE);
 			pud_populate(&init_mm, pudp, pmdp);
--- a/arch/powerpc/mm/pgtable.c
+++ b/arch/powerpc/mm/pgtable.c
@@ -265,6 +265,7 @@ int huge_ptep_set_access_flags(struct vm_area_struct *vma,
 void assert_pte_locked(struct mm_struct *mm, unsigned long addr)
 {
 	pgd_t *pgd;
+	p4d_t *p4d;
 	pud_t *pud;
 	pmd_t *pmd;

@@ -272,7 +273,9 @@ void assert_pte_locked(struct mm_struct *mm, unsigned long addr)
 		return;
 	pgd = mm->pgd + pgd_index(addr);
 	BUG_ON(pgd_none(*pgd));
-	pud = pud_offset(pgd, addr);
+	p4d = p4d_offset(pgd, addr);
+	BUG_ON(p4d_none(*p4d));
+	pud = pud_offset(p4d, addr);
 	BUG_ON(pud_none(*pud));
 	pmd = pmd_offset(pud, addr);
 	/*
@@ -312,12 +315,13 @@ EXPORT_SYMBOL_GPL(vmalloc_to_phys);
 pte_t *__find_linux_pte(pgd_t *pgdir, unsigned long ea,
 			bool *is_thp, unsigned *hpage_shift)
 {
-	pgd_t pgd, *pgdp;
+	pgd_t *pgdp;
+	p4d_t p4d, *p4dp;
 	pud_t pud, *pudp;
 	pmd_t pmd, *pmdp;
 	pte_t *ret_pte;
 	hugepd_t *hpdp = NULL;
-	unsigned pdshift = PGDIR_SHIFT;
+	unsigned pdshift;

 	if (hpage_shift)
 		*hpage_shift = 0;
@@ -325,24 +329,28 @@ pte_t *__find_linux_pte(pgd_t *pgdir, unsigned long ea,
 	if (is_thp)
 		*is_thp = false;

-	pgdp = pgdir + pgd_index(ea);
-	pgd  = READ_ONCE(*pgdp);
 	/*
 	 * Always operate on the local stack value. This make sure the
 	 * value don't get updated by a parallel THP split/collapse,
 	 * page fault or a page unmap. The return pte_t * is still not
 	 * stable. So should be checked there for above conditions.
+	 * Top level is an exception because it is folded into p4d.
 	 */
-	if (pgd_none(pgd))
+	pgdp = pgdir + pgd_index(ea);
+	p4dp = p4d_offset(pgdp, ea);
+	p4d  = READ_ONCE(*p4dp);
+	pdshift = P4D_SHIFT;
+
+	if (p4d_none(p4d))
 		return NULL;

-	if (pgd_is_leaf(pgd)) {
-		ret_pte = (pte_t *)pgdp;
+	if (p4d_is_leaf(p4d)) {
+		ret_pte = (pte_t *)p4dp;
 		goto out;
 	}

-	if (is_hugepd(__hugepd(pgd_val(pgd)))) {
-		hpdp = (hugepd_t *)&pgd;
+	if (is_hugepd(__hugepd(p4d_val(p4d)))) {
+		hpdp = (hugepd_t *)&p4d;
 		goto out_huge;
 	}

@@ -352,7 +360,7 @@ pte_t *__find_linux_pte(pgd_t *pgdir, unsigned long ea,
 	 * irq disabled
 	 */
 	pdshift = PUD_SHIFT;
-	pudp = pud_offset(&pgd, ea);
+	pudp = pud_offset(&p4d, ea);
 	pud  = READ_ONCE(*pudp);

 	if (pud_none(pud))
--- a/arch/powerpc/mm/pgtable_64.c
+++ b/arch/powerpc/mm/pgtable_64.c
@@ -101,13 +101,13 @@ EXPORT_SYMBOL(__pte_frag_size_shift);

 #ifndef __PAGETABLE_PUD_FOLDED
 /* 4 level page table */
-struct page *pgd_page(pgd_t pgd)
+struct page *p4d_page(p4d_t p4d)
 {
-	if (pgd_is_leaf(pgd)) {
-		VM_WARN_ON(!pgd_huge(pgd));
-		return pte_page(pgd_pte(pgd));
+	if (p4d_is_leaf(p4d)) {
+		VM_WARN_ON(!p4d_huge(p4d));
+		return pte_page(p4d_pte(p4d));
 	}
-	return virt_to_page(pgd_page_vaddr(pgd));
+	return virt_to_page(p4d_page_vaddr(p4d));
 }
 #endif

--- a/arch/powerpc/mm/ptdump/hashpagetable.c
+++ b/arch/powerpc/mm/ptdump/hashpagetable.c
@@ -417,9 +417,9 @@ static void walk_pmd(struct pg_state *st, pud_t *pud, unsigned long start)
 	}
 }

-static void walk_pud(struct pg_state *st, pgd_t *pgd, unsigned long start)
+static void walk_pud(struct pg_state *st, p4d_t *p4d, unsigned long start)
 {
-	pud_t *pud = pud_offset(pgd, 0);
+	pud_t *pud = pud_offset(p4d, 0);
 	unsigned long addr;
 	unsigned int i;

@@ -431,6 +431,20 @@ static void walk_pud(struct pg_state *st, pgd_t *pgd, unsigned long start)
 	}
 }

+static void walk_p4d(struct pg_state *st, pgd_t *pgd, unsigned long start)
+{
+	p4d_t *p4d = p4d_offset(pgd, 0);
+	unsigned long addr;
+	unsigned int i;
+
+	for (i = 0; i < PTRS_PER_P4D; i++, p4d++) {
+		addr = start + i * P4D_SIZE;
+		if (!p4d_none(*p4d))
+			/* p4d exists */
+			walk_pud(st, p4d, addr);
+	}
+}
+
 static void walk_pagetables(struct pg_state *st)
 {
 	pgd_t *pgd = pgd_offset_k(0UL);
@@ -445,7 +459,7 @@ static void walk_pagetables(struct pg_state *st)
 		addr = KERN_VIRT_START + i * PGDIR_SIZE;
 		if (!pgd_none(*pgd))
 			/* pgd exists */
-			walk_pud(st, pgd, addr);
+			walk_p4d(st, pgd, addr);
 	}
 }

--- a/arch/powerpc/mm/ptdump/ptdump.c
+++ b/arch/powerpc/mm/ptdump/ptdump.c
@@ -277,9 +277,9 @@ static void walk_pmd(struct pg_state *st, pud_t *pud, unsigned long start)
 	}
 }

-static void walk_pud(struct pg_state *st, pgd_t *pgd, unsigned long start)
+static void walk_pud(struct pg_state *st, p4d_t *p4d, unsigned long start)
 {
-	pud_t *pud = pud_offset(pgd, 0);
+	pud_t *pud = pud_offset(p4d, 0);
 	unsigned long addr;
 	unsigned int i;

@@ -304,11 +304,13 @@ static void walk_pagetables(struct pg_state *st)
 	 * the hash pagetable.
 	 */
 	for (i = pgd_index(addr); i < PTRS_PER_PGD; i++, pgd++, addr += PGDIR_SIZE) {
-		if (!pgd_none(*pgd) && !pgd_is_leaf(*pgd))
+		p4d_t *p4d = p4d_offset(pgd, 0);
+
+		if (!p4d_none(*p4d) && !p4d_is_leaf(*p4d))
 			/* pgd exists */
-			walk_pud(st, pgd, addr);
+			walk_pud(st, p4d, addr);
 		else
-			note_page(st, addr, 1, pgd_val(*pgd));
+			note_page(st, addr, 1, p4d_val(*p4d));
 	}
 }