Linux-libre 4.5-gnu

8 files changed, 58 insertions, 74 deletions

diff --git a/arch/s390/mm/extmem.c b/arch/s390/mm/extmem.c
index 18fccc303..a1bf4ad89 100644
--- a/arch/s390/mm/extmem.c
+++ b/arch/s390/mm/extmem.c
@@ -94,7 +94,7 @@ static DEFINE_MUTEX(dcss_lock);
 static LIST_HEAD(dcss_list);
 static char *segtype_string[] = { "SW", "EW", "SR", "ER", "SN", "EN", "SC",
 					"EW/EN-MIXED" };
-static int loadshr_scode, loadnsr_scode, findseg_scode;
+static int loadshr_scode, loadnsr_scode;
 static int segext_scode, purgeseg_scode;
 static int scode_set;
 
@@ -130,7 +130,6 @@ dcss_set_subcodes(void)
 		loadshr_scode = DCSS_LOADSHRX;
 		loadnsr_scode = DCSS_LOADNSRX;
 		purgeseg_scode = DCSS_PURGESEG;
-		findseg_scode = DCSS_FINDSEGX;
 		segext_scode = DCSS_SEGEXTX;
 		return 0;
 	}
@@ -138,7 +137,6 @@ dcss_set_subcodes(void)
 	loadshr_scode = DCSS_LOADNOLY;
 	loadnsr_scode = DCSS_LOADNSR;
 	purgeseg_scode = DCSS_PURGESEG;
-	findseg_scode = DCSS_FINDSEG;
 	segext_scode = DCSS_SEGEXT;
 	return 0;
 }
diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c
index ec1a30d0d..791a41460 100644
--- a/arch/s390/mm/fault.c
+++ b/arch/s390/mm/fault.c
@@ -228,7 +228,7 @@ static inline void report_user_fault(struct pt_regs *regs, long signr)
 		return;
 	printk(KERN_ALERT "User process fault: interruption code %04x ilc:%d ",
 	       regs->int_code & 0xffff, regs->int_code >> 17);
-	print_vma_addr(KERN_CONT "in ", regs->psw.addr & PSW_ADDR_INSN);
+	print_vma_addr(KERN_CONT "in ", regs->psw.addr);
 	printk(KERN_CONT "\n");
 	printk(KERN_ALERT "failing address: %016lx TEID: %016lx\n",
 	       regs->int_parm_long & __FAIL_ADDR_MASK, regs->int_parm_long);
@@ -254,12 +254,11 @@ static noinline void do_sigsegv(struct pt_regs *regs, int si_code)
 static noinline void do_no_context(struct pt_regs *regs)
 {
 	const struct exception_table_entry *fixup;
-	unsigned long address;
 
 	/* Are we prepared to handle this kernel fault?  */
-	fixup = search_exception_tables(regs->psw.addr & PSW_ADDR_INSN);
+	fixup = search_exception_tables(regs->psw.addr);
 	if (fixup) {
-		regs->psw.addr = extable_fixup(fixup) | PSW_ADDR_AMODE;
+		regs->psw.addr = extable_fixup(fixup);
 		return;
 	}
 
@@ -267,7 +266,6 @@ static noinline void do_no_context(struct pt_regs *regs)
 	 * Oops. The kernel tried to access some bad page. We'll have to
 	 * terminate things with extreme prejudice.
 	 */
-	address = regs->int_parm_long & __FAIL_ADDR_MASK;
 	if (!user_space_fault(regs))
 		printk(KERN_ALERT "Unable to handle kernel pointer dereference"
 		       " in virtual kernel address space\n");
diff --git a/arch/s390/mm/gup.c b/arch/s390/mm/gup.c
index 12bbf0e84..13dab0c16 100644
--- a/arch/s390/mm/gup.c
+++ b/arch/s390/mm/gup.c
@@ -55,7 +55,7 @@ static inline int gup_huge_pmd(pmd_t *pmdp, pmd_t pmd, unsigned long addr,
 		unsigned long end, int write, struct page **pages, int *nr)
 {
 	unsigned long mask, result;
-	struct page *head, *page, *tail;
+	struct page *head, *page;
 	int refs;
 
 	result = write ? 0 : _SEGMENT_ENTRY_PROTECT;
@@ -67,7 +67,6 @@ static inline int gup_huge_pmd(pmd_t *pmdp, pmd_t pmd, unsigned long addr,
 	refs = 0;
 	head = pmd_page(pmd);
 	page = head + ((addr & ~PMD_MASK) >> PAGE_SHIFT);
-	tail = page;
 	do {
 		VM_BUG_ON(compound_head(page) != head);
 		pages[*nr] = page;
@@ -88,16 +87,6 @@ static inline int gup_huge_pmd(pmd_t *pmdp, pmd_t pmd, unsigned long addr,
 		return 0;
 	}
 
-	/*
-	 * Any tail page need their mapcount reference taken before we
-	 * return.
-	 */
-	while (refs--) {
-		if (PageTail(tail))
-			get_huge_page_tail(tail);
-		tail++;
-	}
-
 	return 1;
 }
 
@@ -116,16 +105,7 @@ static inline int gup_pmd_range(pud_t *pudp, pud_t pud, unsigned long addr,
 		pmd = *pmdp;
 		barrier();
 		next = pmd_addr_end(addr, end);
-		/*
-		 * The pmd_trans_splitting() check below explains why
-		 * pmdp_splitting_flush() has to serialize with
-		 * smp_call_function() against our disabled IRQs, to stop
-		 * this gup-fast code from running while we set the
-		 * splitting bit in the pmd. Returning zero will take
-		 * the slow path that will call wait_split_huge_page()
-		 * if the pmd is still in splitting state.
-		 */
-		if (pmd_none(pmd) || pmd_trans_splitting(pmd))
+		if (pmd_none(pmd))
 			return 0;
 		if (unlikely(pmd_large(pmd))) {
 			/*
@@ -233,6 +213,7 @@ int get_user_pages_fast(unsigned long start, int nr_pages, int write,
 	struct mm_struct *mm = current->mm;
 	int nr, ret;
 
+	might_sleep();
 	start &= PAGE_MASK;
 	nr = __get_user_pages_fast(start, nr_pages, write, pages);
 	if (nr == nr_pages)
diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c
index c722400c7..73e290337 100644
--- a/arch/s390/mm/init.c
+++ b/arch/s390/mm/init.c
@@ -98,7 +98,7 @@ void __init paging_init(void)
 	__ctl_load(S390_lowcore.kernel_asce, 1, 1);
 	__ctl_load(S390_lowcore.kernel_asce, 7, 7);
 	__ctl_load(S390_lowcore.kernel_asce, 13, 13);
-	arch_local_irq_restore(4UL << (BITS_PER_LONG - 8));
+	__arch_local_irq_stosm(0x04);
 
 	sparse_memory_present_with_active_regions(MAX_NUMNODES);
 	sparse_init();
diff --git a/arch/s390/mm/maccess.c b/arch/s390/mm/maccess.c
index 8a993a53f..792f9c63f 100644
--- a/arch/s390/mm/maccess.c
+++ b/arch/s390/mm/maccess.c
@@ -93,15 +93,19 @@ static int __memcpy_real(void *dest, void *src, size_t count)
  */
 int memcpy_real(void *dest, void *src, size_t count)
 {
+	int irqs_disabled, rc;
 	unsigned long flags;
-	int rc;
 
 	if (!count)
 		return 0;
-	local_irq_save(flags);
-	__arch_local_irq_stnsm(0xfbUL);
+	flags = __arch_local_irq_stnsm(0xf8UL);
+	irqs_disabled = arch_irqs_disabled_flags(flags);
+	if (!irqs_disabled)
+		trace_hardirqs_off();
 	rc = __memcpy_real(dest, src, count);
-	local_irq_restore(flags);
+	if (!irqs_disabled)
+		trace_hardirqs_on();
+	__arch_local_irq_ssm(flags);
 	return rc;
 }
 
@@ -163,11 +167,11 @@ static int is_swapped(unsigned long addr)
 	unsigned long lc;
 	int cpu;
 
-	if (addr < sizeof(struct _lowcore))
+	if (addr < sizeof(struct lowcore))
 		return 1;
 	for_each_online_cpu(cpu) {
 		lc = (unsigned long) lowcore_ptr[cpu];
-		if (addr > lc + sizeof(struct _lowcore) - 1 || addr < lc)
+		if (addr > lc + sizeof(struct lowcore) - 1 || addr < lc)
 			continue;
 		return 1;
 	}
diff --git a/arch/s390/mm/mem_detect.c b/arch/s390/mm/mem_detect.c
index e00f0d5d2..d612cc3ee 100644
--- a/arch/s390/mm/mem_detect.c
+++ b/arch/s390/mm/mem_detect.c
@@ -14,8 +14,6 @@
 #include <asm/sclp.h>
 #include <asm/setup.h>
 
-#define ADDR2G (1ULL << 31)
-
 #define CHUNK_READ_WRITE 0
 #define CHUNK_READ_ONLY  1
 
@@ -27,15 +25,14 @@ static inline void memblock_physmem_add(phys_addr_t start, phys_addr_t size)
 
 void __init detect_memory_memblock(void)
 {
-	unsigned long long memsize, rnmax, rzm;
-	unsigned long addr, size;
+	unsigned long memsize, rnmax, rzm, addr, size;
 	int type;
 
 	rzm = sclp.rzm;
 	rnmax = sclp.rnmax;
 	memsize = rzm * rnmax;
 	if (!rzm)
-		rzm = 1ULL << 17;
+		rzm = 1UL << 17;
 	max_physmem_end = memsize;
 	addr = 0;
 	/* keep memblock lists close to the kernel */
diff --git a/arch/s390/mm/mmap.c b/arch/s390/mm/mmap.c
index ea01477b4..45c4daa49 100644
--- a/arch/s390/mm/mmap.c
+++ b/arch/s390/mm/mmap.c
@@ -169,12 +169,12 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0,
 
 int s390_mmap_check(unsigned long addr, unsigned long len, unsigned long flags)
 {
-	if (is_compat_task() || (TASK_SIZE >= (1UL << 53)))
+	if (is_compat_task() || TASK_SIZE >= TASK_MAX_SIZE)
 		return 0;
 	if (!(flags & MAP_FIXED))
 		addr = 0;
 	if ((addr + len) >= TASK_SIZE)
-		return crst_table_upgrade(current->mm, 1UL << 53);
+		return crst_table_upgrade(current->mm, TASK_MAX_SIZE);
 	return 0;
 }
 
@@ -189,9 +189,9 @@ s390_get_unmapped_area(struct file *filp, unsigned long addr,
 	area = arch_get_unmapped_area(filp, addr, len, pgoff, flags);
 	if (!(area & ~PAGE_MASK))
 		return area;
-	if (area == -ENOMEM && !is_compat_task() && TASK_SIZE < (1UL << 53)) {
+	if (area == -ENOMEM && !is_compat_task() && TASK_SIZE < TASK_MAX_SIZE) {
 		/* Upgrade the page table to 4 levels and retry. */
-		rc = crst_table_upgrade(mm, 1UL << 53);
+		rc = crst_table_upgrade(mm, TASK_MAX_SIZE);
 		if (rc)
 			return (unsigned long) rc;
 		area = arch_get_unmapped_area(filp, addr, len, pgoff, flags);
@@ -211,9 +211,9 @@ s390_get_unmapped_area_topdown(struct file *filp, const unsigned long addr,
 	area = arch_get_unmapped_area_topdown(filp, addr, len, pgoff, flags);
 	if (!(area & ~PAGE_MASK))
 		return area;
-	if (area == -ENOMEM && !is_compat_task() && TASK_SIZE < (1UL << 53)) {
+	if (area == -ENOMEM && !is_compat_task() && TASK_SIZE < TASK_MAX_SIZE) {
 		/* Upgrade the page table to 4 levels and retry. */
-		rc = crst_table_upgrade(mm, 1UL << 53);
+		rc = crst_table_upgrade(mm, TASK_MAX_SIZE);
 		if (rc)
 			return (unsigned long) rc;
 		area = arch_get_unmapped_area_topdown(filp, addr, len,
diff --git a/arch/s390/mm/pgtable.c b/arch/s390/mm/pgtable.c
index 54ef3bc01..510982788 100644
--- a/arch/s390/mm/pgtable.c
+++ b/arch/s390/mm/pgtable.c
@@ -55,7 +55,7 @@ int crst_table_upgrade(struct mm_struct *mm, unsigned long limit)
 	unsigned long entry;
 	int flush;
 
-	BUG_ON(limit > (1UL << 53));
+	BUG_ON(limit > TASK_MAX_SIZE);
 	flush = 0;
 repeat:
 	table = crst_table_alloc(mm);
@@ -133,7 +133,7 @@ void crst_table_downgrade(struct mm_struct *mm, unsigned long limit)
 /**
  * gmap_alloc - allocate a guest address space
  * @mm: pointer to the parent mm_struct
- * @limit: maximum size of the gmap address space
+ * @limit: maximum address of the gmap address space
  *
  * Returns a guest address space structure.
  */
@@ -402,7 +402,7 @@ int gmap_map_segment(struct gmap *gmap, unsigned long from,
 	if ((from | to | len) & (PMD_SIZE - 1))
 		return -EINVAL;
 	if (len == 0 || from + len < from || to + len < to ||
-	    from + len > TASK_MAX_SIZE || to + len > gmap->asce_end)
+	    from + len - 1 > TASK_MAX_SIZE || to + len - 1 > gmap->asce_end)
 		return -EINVAL;
 
 	flush = 0;
@@ -578,17 +578,29 @@ int gmap_fault(struct gmap *gmap, unsigned long gaddr,
 {
 	unsigned long vmaddr;
 	int rc;
+	bool unlocked;
 
 	down_read(&gmap->mm->mmap_sem);
+
+retry:
+	unlocked = false;
 	vmaddr = __gmap_translate(gmap, gaddr);
 	if (IS_ERR_VALUE(vmaddr)) {
 		rc = vmaddr;
 		goto out_up;
 	}
-	if (fixup_user_fault(current, gmap->mm, vmaddr, fault_flags)) {
+	if (fixup_user_fault(current, gmap->mm, vmaddr, fault_flags,
+			     &unlocked)) {
 		rc = -EFAULT;
 		goto out_up;
 	}
+	/*
+	 * In the case that fixup_user_fault unlocked the mmap_sem during
+	 * faultin redo __gmap_translate to not race with a map/unmap_segment.
+	 */
+	if (unlocked)
+		goto retry;
+
 	rc = __gmap_link(gmap, gaddr, vmaddr);
 out_up:
 	up_read(&gmap->mm->mmap_sem);
@@ -603,10 +615,7 @@ static void gmap_zap_swap_entry(swp_entry_t entry, struct mm_struct *mm)
 	else if (is_migration_entry(entry)) {
 		struct page *page = migration_entry_to_page(entry);
 
-		if (PageAnon(page))
-			dec_mm_counter(mm, MM_ANONPAGES);
-		else
-			dec_mm_counter(mm, MM_FILEPAGES);
+		dec_mm_counter(mm, mm_counter(page));
 	}
 	free_swap_and_cache(entry);
 }
@@ -717,12 +726,14 @@ int gmap_ipte_notify(struct gmap *gmap, unsigned long gaddr, unsigned long len)
 	spinlock_t *ptl;
 	pte_t *ptep, entry;
 	pgste_t pgste;
+	bool unlocked;
 	int rc = 0;
 
 	if ((gaddr & ~PAGE_MASK) || (len & ~PAGE_MASK))
 		return -EINVAL;
 	down_read(&gmap->mm->mmap_sem);
 	while (len) {
+		unlocked = false;
 		/* Convert gmap address and connect the page tables */
 		addr = __gmap_translate(gmap, gaddr);
 		if (IS_ERR_VALUE(addr)) {
@@ -730,10 +741,14 @@ int gmap_ipte_notify(struct gmap *gmap, unsigned long gaddr, unsigned long len)
 			break;
 		}
 		/* Get the page mapped */
-		if (fixup_user_fault(current, gmap->mm, addr, FAULT_FLAG_WRITE)) {
+		if (fixup_user_fault(current, gmap->mm, addr, FAULT_FLAG_WRITE,
+				     &unlocked)) {
 			rc = -EFAULT;
 			break;
 		}
+		/* While trying to map mmap_sem got unlocked. Let us retry */
+		if (unlocked)
+			continue;
 		rc = __gmap_link(gmap, gaddr, addr);
 		if (rc)
 			break;
@@ -794,9 +809,11 @@ int set_guest_storage_key(struct mm_struct *mm, unsigned long addr,
 	spinlock_t *ptl;
 	pgste_t old, new;
 	pte_t *ptep;
+	bool unlocked;
 
 	down_read(&mm->mmap_sem);
 retry:
+	unlocked = false;
 	ptep = get_locked_pte(mm, addr, &ptl);
 	if (unlikely(!ptep)) {
 		up_read(&mm->mmap_sem);
@@ -805,7 +822,12 @@ retry:
 	if (!(pte_val(*ptep) & _PAGE_INVALID) &&
 	     (pte_val(*ptep) & _PAGE_PROTECT)) {
 		pte_unmap_unlock(ptep, ptl);
-		if (fixup_user_fault(current, mm, addr, FAULT_FLAG_WRITE)) {
+		/*
+		 * We do not really care about unlocked. We will retry either
+		 * way. But this allows fixup_user_fault to enable userfaultfd.
+		 */
+		if (fixup_user_fault(current, mm, addr, FAULT_FLAG_WRITE,
+				     &unlocked)) {
 			up_read(&mm->mmap_sem);
 			return -EFAULT;
 		}
@@ -1308,22 +1330,6 @@ int pmdp_set_access_flags(struct vm_area_struct *vma,
 	return 1;
 }
 
-static void pmdp_splitting_flush_sync(void *arg)
-{
-	/* Simply deliver the interrupt */
-}
-
-void pmdp_splitting_flush(struct vm_area_struct *vma, unsigned long address,
-			  pmd_t *pmdp)
-{
-	VM_BUG_ON(address & ~HPAGE_PMD_MASK);
-	if (!test_and_set_bit(_SEGMENT_ENTRY_SPLIT_BIT,
-			      (unsigned long *) pmdp)) {
-		/* need to serialize against gup-fast (IRQ disabled) */
-		smp_call_function(pmdp_splitting_flush_sync, NULL, 1);
-	}
-}
-
 void pgtable_trans_huge_deposit(struct mm_struct *mm, pmd_t *pmdp,
 				pgtable_t pgtable)
 {