Merge branch 'x86-mm-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull x86 mm updates from Thomas Gleixner:

 - Make lazy TLB mode even lazier to avoid pointless switch_mm()
   operations, which reduces CPU load by 1-2% for memcache workloads

 - Small cleanups and improvements all over the place

* 'x86-mm-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/mm: Remove redundant check for kmem_cache_create()
  arm/asm/tlb.h: Fix build error implicit func declaration
  x86/mm/tlb: Make clear_asid_other() static
  x86/mm/tlb: Skip atomic operations for 'init_mm' in switch_mm_irqs_off()
  x86/mm/tlb: Always use lazy TLB mode
  x86/mm/tlb: Only send page table free TLB flush to lazy TLB CPUs
  x86/mm/tlb: Make lazy TLB mode lazier
  x86/mm/tlb: Restructure switch_mm_irqs_off()
  x86/mm/tlb: Leave lazy TLB mode at page table free time
  mm: Allocate the mm_cpumask (mm->cpu_bitmap[]) dynamically based on nr_cpu_ids
  x86/mm: Add TLB purge to free pmd/pte page interfaces
  ioremap: Update pgtable free interfaces with addr
  x86/mm: Disable ioremap free page handling on x86-PAE

arch/arm/include/asm/tlb.h

@@ -292,5 +292,13 @@ static inline void tlb_remove_check_page_size_change(struct mmu_gather *tlb,
 {
 }
 
+static inline void tlb_flush_remove_tables(struct mm_struct *mm)
+{
+}
+
+static inline void tlb_flush_remove_tables_local(void *arg)
+{
+}
+
 #endif /* CONFIG_MMU */
 #endif

arch/arm64/mm/mmu.c

@@ -977,12 +977,12 @@ int pmd_clear_huge(pmd_t *pmdp)
 	return 1;
 }
 
-int pud_free_pmd_page(pud_t *pud)
+int pud_free_pmd_page(pud_t *pud, unsigned long addr)
 {
 	return pud_none(*pud);
 }
 
-int pmd_free_pte_page(pmd_t *pmd)
+int pmd_free_pte_page(pmd_t *pmd, unsigned long addr)
 {
 	return pmd_none(*pmd);
 }

arch/x86/include/asm/tlbflush.h

@@ -148,22 +148,6 @@ static inline unsigned long build_cr3_noflush(pgd_t *pgd, u16 asid)
 #define __flush_tlb_one_user(addr) __native_flush_tlb_one_user(addr)
 #endif
 
-static inline bool tlb_defer_switch_to_init_mm(void)
-{
-	/*
-	 * If we have PCID, then switching to init_mm is reasonably
-	 * fast.  If we don't have PCID, then switching to init_mm is
-	 * quite slow, so we try to defer it in the hopes that we can
-	 * avoid it entirely.  The latter approach runs the risk of
-	 * receiving otherwise unnecessary IPIs.
-	 *
-	 * This choice is just a heuristic.  The tlb code can handle this
-	 * function returning true or false regardless of whether we have
-	 * PCID.
-	 */
-	return !static_cpu_has(X86_FEATURE_PCID);
-}
-
 struct tlb_context {
 	u64 ctx_id;
 	u64 tlb_gen;
@@ -554,4 +538,9 @@ extern void arch_tlbbatch_flush(struct arch_tlbflush_unmap_batch *batch);
 	native_flush_tlb_others(mask, info)
 #endif
 
+extern void tlb_flush_remove_tables(struct mm_struct *mm);
+extern void tlb_flush_remove_tables_local(void *arg);
+
+#define HAVE_TLB_FLUSH_REMOVE_TABLES
+
 #endif /* _ASM_X86_TLBFLUSH_H */

arch/x86/mm/pgtable.c

@@ -329,9 +329,6 @@ static int __init pgd_cache_init(void)
 	 */
 	pgd_cache = kmem_cache_create("pgd_cache", PGD_SIZE, PGD_ALIGN,
 				      SLAB_PANIC, NULL);
-	if (!pgd_cache)
-		return -ENOMEM;
-
 	return 0;
 }
 core_initcall(pgd_cache_init);
@@ -719,28 +716,50 @@ int pmd_clear_huge(pmd_t *pmd)
 	return 0;
 }
 
+#ifdef CONFIG_X86_64
 /**
  * pud_free_pmd_page - Clear pud entry and free pmd page.
  * @pud: Pointer to a PUD.
+ * @addr: Virtual address associated with pud.
  *
- * Context: The pud range has been unmaped and TLB purged.
+ * Context: The pud range has been unmapped and TLB purged.
  * Return: 1 if clearing the entry succeeded. 0 otherwise.
+ *
+ * NOTE: Callers must allow a single page allocation.
  */
-int pud_free_pmd_page(pud_t *pud)
+int pud_free_pmd_page(pud_t *pud, unsigned long addr)
 {
-	pmd_t *pmd;
+	pmd_t *pmd, *pmd_sv;
+	pte_t *pte;
 	int i;
 
 	if (pud_none(*pud))
 		return 1;
 
 	pmd = (pmd_t *)pud_page_vaddr(*pud);
+	pmd_sv = (pmd_t *)__get_free_page(GFP_KERNEL);
+	if (!pmd_sv)
+		return 0;
 
-	for (i = 0; i < PTRS_PER_PMD; i++)
-		if (!pmd_free_pte_page(&pmd[i]))
-			return 0;
+	for (i = 0; i < PTRS_PER_PMD; i++) {
+		pmd_sv[i] = pmd[i];
+		if (!pmd_none(pmd[i]))
+			pmd_clear(&pmd[i]);
+	}
 
 	pud_clear(pud);
+
+	/* INVLPG to clear all paging-structure caches */
+	flush_tlb_kernel_range(addr, addr + PAGE_SIZE-1);
+
+	for (i = 0; i < PTRS_PER_PMD; i++) {
+		if (!pmd_none(pmd_sv[i])) {
+			pte = (pte_t *)pmd_page_vaddr(pmd_sv[i]);
+			free_page((unsigned long)pte);
+		}
+	}
+
+	free_page((unsigned long)pmd_sv);
 	free_page((unsigned long)pmd);
 
 	return 1;
@@ -749,11 +768,12 @@ int pud_free_pmd_page(pud_t *pud)
 /**
  * pmd_free_pte_page - Clear pmd entry and free pte page.
  * @pmd: Pointer to a PMD.
+ * @addr: Virtual address associated with pmd.
  *
- * Context: The pmd range has been unmaped and TLB purged.
+ * Context: The pmd range has been unmapped and TLB purged.
  * Return: 1 if clearing the entry succeeded. 0 otherwise.
  */
-int pmd_free_pte_page(pmd_t *pmd)
+int pmd_free_pte_page(pmd_t *pmd, unsigned long addr)
 {
 	pte_t *pte;
 
@@ -762,8 +782,30 @@ int pmd_free_pte_page(pmd_t *pmd)
 
 	pte = (pte_t *)pmd_page_vaddr(*pmd);
 	pmd_clear(pmd);
+
+	/* INVLPG to clear all paging-structure caches */
+	flush_tlb_kernel_range(addr, addr + PAGE_SIZE-1);
+
 	free_page((unsigned long)pte);
 
 	return 1;
 }
+
+#else /* !CONFIG_X86_64 */
+
+int pud_free_pmd_page(pud_t *pud, unsigned long addr)
+{
+	return pud_none(*pud);
+}
+
+/*
+ * Disable free page handling on x86-PAE. This assures that ioremap()
+ * does not update sync'd pmd entries. See vmalloc_sync_one().
+ */
+int pmd_free_pte_page(pmd_t *pmd, unsigned long addr)
+{
+	return pmd_none(*pmd);
+}
+
+#endif /* CONFIG_X86_64 */
 #endif	/* CONFIG_HAVE_ARCH_HUGE_VMAP */

drivers/firmware/efi/efi.c

@@ -82,6 +82,7 @@ struct mm_struct efi_mm = {
 	.mmap_sem		= __RWSEM_INITIALIZER(efi_mm.mmap_sem),
 	.page_table_lock	= __SPIN_LOCK_UNLOCKED(efi_mm.page_table_lock),
 	.mmlist			= LIST_HEAD_INIT(efi_mm.mmlist),
+	.cpu_bitmap		= { [BITS_TO_LONGS(NR_CPUS)] = 0},
 };
 
 struct workqueue_struct *efi_rts_wq;

include/asm-generic/pgtable.h

@@ -1019,8 +1019,8 @@ int pud_set_huge(pud_t *pud, phys_addr_t addr, pgprot_t prot);
 int pmd_set_huge(pmd_t *pmd, phys_addr_t addr, pgprot_t prot);
 int pud_clear_huge(pud_t *pud);
 int pmd_clear_huge(pmd_t *pmd);
-int pud_free_pmd_page(pud_t *pud);
-int pmd_free_pte_page(pmd_t *pmd);
+int pud_free_pmd_page(pud_t *pud, unsigned long addr);
+int pmd_free_pte_page(pmd_t *pmd, unsigned long addr);
 #else	/* !CONFIG_HAVE_ARCH_HUGE_VMAP */
 static inline int p4d_set_huge(p4d_t *p4d, phys_addr_t addr, pgprot_t prot)
 {
@@ -1046,11 +1046,11 @@ static inline int pmd_clear_huge(pmd_t *pmd)
 {
 	return 0;
 }
-static inline int pud_free_pmd_page(pud_t *pud)
+static inline int pud_free_pmd_page(pud_t *pud, unsigned long addr)
 {
 	return 0;
 }
-static inline int pmd_free_pte_page(pmd_t *pmd)
+static inline int pmd_free_pte_page(pmd_t *pmd, unsigned long addr)
 {
 	return 0;
 }

include/asm-generic/tlb.h

@@ -303,4 +303,14 @@ static inline void tlb_remove_check_page_size_change(struct mmu_gather *tlb,
 
 #define tlb_migrate_finish(mm) do {} while (0)
 
+/*
+ * Used to flush the TLB when page tables are removed, when lazy
+ * TLB mode may cause a CPU to retain intermediate translations
+ * pointing to about-to-be-freed page table memory.
+ */
+#ifndef HAVE_TLB_FLUSH_REMOVE_TABLES
+#define tlb_flush_remove_tables(mm) do {} while (0)
+#define tlb_flush_remove_tables_local(mm) do {} while (0)
+#endif
+
 #endif /* _ASM_GENERIC__TLB_H */

kernel/fork.c

@@ -2276,6 +2276,8 @@ static void sighand_ctor(void *data)
 
 void __init proc_caches_init(void)
 {
+	unsigned int mm_size;
+
 	sighand_cachep = kmem_cache_create("sighand_cache",
 			sizeof(struct sighand_struct), 0,
 			SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_TYPESAFE_BY_RCU|
@@ -2292,15 +2294,16 @@ void __init proc_caches_init(void)
 			sizeof(struct fs_struct), 0,
 			SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_ACCOUNT,
 			NULL);
+
 	/*
-	 * FIXME! The "sizeof(struct mm_struct)" currently includes the
-	 * whole struct cpumask for the OFFSTACK case. We could change
-	 * this to *only* allocate as much of it as required by the
-	 * maximum number of CPU's we can ever have.  The cpumask_allocation
-	 * is at the end of the structure, exactly for that reason.
+	 * The mm_cpumask is located at the end of mm_struct, and is
+	 * dynamically sized based on the maximum CPU number this system
+	 * can have, taking hotplug into account (nr_cpu_ids).
 	 */
+	mm_size = sizeof(struct mm_struct) + cpumask_size();
+
 	mm_cachep = kmem_cache_create_usercopy("mm_struct",
-			sizeof(struct mm_struct), ARCH_MIN_MMSTRUCT_ALIGN,
+			mm_size, ARCH_MIN_MMSTRUCT_ALIGN,
 			SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_ACCOUNT,
 			offsetof(struct mm_struct, saved_auxv),
 			sizeof_field(struct mm_struct, saved_auxv),

lib/ioremap.c

@@ -92,7 +92,7 @@ static inline int ioremap_pmd_range(pud_t *pud, unsigned long addr,
 		if (ioremap_pmd_enabled() &&
 		    ((next - addr) == PMD_SIZE) &&
 		    IS_ALIGNED(phys_addr + addr, PMD_SIZE) &&
-		    pmd_free_pte_page(pmd)) {
+		    pmd_free_pte_page(pmd, addr)) {
 			if (pmd_set_huge(pmd, phys_addr + addr, prot))
 				continue;
 		}
@@ -119,7 +119,7 @@ static inline int ioremap_pud_range(p4d_t *p4d, unsigned long addr,
 		if (ioremap_pud_enabled() &&
 		    ((next - addr) == PUD_SIZE) &&
 		    IS_ALIGNED(phys_addr + addr, PUD_SIZE) &&
-		    pud_free_pmd_page(pud)) {
+		    pud_free_pmd_page(pud, addr)) {
 			if (pud_set_huge(pud, phys_addr + addr, prot))
 				continue;
 		}

mm/init-mm.c

@@ -15,6 +15,16 @@
 #define INIT_MM_CONTEXT(name)
 #endif
 
+/*
+ * For dynamically allocated mm_structs, there is a dynamically sized cpumask
+ * at the end of the structure, the size of which depends on the maximum CPU
+ * number the system can see. That way we allocate only as much memory for
+ * mm_cpumask() as needed for the hundreds, or thousands of processes that
+ * a system typically runs.
+ *
+ * Since there is only one init_mm in the entire system, keep it simple
+ * and size this cpu_bitmask to NR_CPUS.
+ */
 struct mm_struct init_mm = {
 	.mm_rb		= RB_ROOT,
 	.pgd		= swapper_pg_dir,
@@ -25,5 +35,6 @@ struct mm_struct init_mm = {
 	.arg_lock	=  __SPIN_LOCK_UNLOCKED(init_mm.arg_lock),
 	.mmlist		= LIST_HEAD_INIT(init_mm.mmlist),
 	.user_ns	= &init_user_ns,
+	.cpu_bitmap	= { [BITS_TO_LONGS(NR_CPUS)] = 0},
 	INIT_MM_CONTEXT(init_mm)
 };

mm/memory.c

@@ -326,16 +326,20 @@ bool __tlb_remove_page_size(struct mmu_gather *tlb, struct page *page, int page_
 
 #ifdef CONFIG_HAVE_RCU_TABLE_FREE
 
-/*
- * See the comment near struct mmu_table_batch.
- */
-
 static void tlb_remove_table_smp_sync(void *arg)
 {
-	/* Simply deliver the interrupt */
+	struct mm_struct __maybe_unused *mm = arg;
+	/*
+	 * On most architectures this does nothing. Simply delivering the
+	 * interrupt is enough to prevent races with software page table
+	 * walking like that done in get_user_pages_fast.
+	 *
+	 * See the comment near struct mmu_table_batch.
+	 */
+	tlb_flush_remove_tables_local(mm);
 }
 
-static void tlb_remove_table_one(void *table)
+static void tlb_remove_table_one(void *table, struct mmu_gather *tlb)
 {
 	/*
 	 * This isn't an RCU grace period and hence the page-tables cannot be
@@ -344,7 +348,7 @@ static void tlb_remove_table_one(void *table)
 	 * It is however sufficient for software page-table walkers that rely on
 	 * IRQ disabling. See the comment near struct mmu_table_batch.
 	 */
-	smp_call_function(tlb_remove_table_smp_sync, NULL, 1);
+	smp_call_function(tlb_remove_table_smp_sync, tlb->mm, 1);
 	__tlb_remove_table(table);
 }
 
@@ -365,6 +369,8 @@ void tlb_table_flush(struct mmu_gather *tlb)
 {
 	struct mmu_table_batch **batch = &tlb->batch;
 
+	tlb_flush_remove_tables(tlb->mm);
+
 	if (*batch) {
 		call_rcu_sched(&(*batch)->rcu, tlb_remove_table_rcu);
 		*batch = NULL;
@@ -387,7 +393,7 @@ void tlb_remove_table(struct mmu_gather *tlb, void *table)
 	if (*batch == NULL) {
 		*batch = (struct mmu_table_batch *)__get_free_page(GFP_NOWAIT | __GFP_NOWARN);
 		if (*batch == NULL) {
-			tlb_remove_table_one(table);
+			tlb_remove_table_one(table, tlb);
 			return;
 		}
 		(*batch)->nr = 0;