Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux into rmobile-latest

7 files changed, 163 insertions, 21 deletions

diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c
index 9564fc779b2..1766def5bc3 100644
--- a/arch/s390/mm/fault.c
+++ b/arch/s390/mm/fault.c
@@ -307,7 +307,7 @@ static inline int do_exception(struct pt_regs *regs, int access,
 
 #ifdef CONFIG_PGSTE
 	if (test_tsk_thread_flag(current, TIF_SIE) && S390_lowcore.gmap) {
-		address = gmap_fault(address,
+		address = __gmap_fault(address,
 				     (struct gmap *) S390_lowcore.gmap);
 		if (address == -EFAULT) {
 			fault = VM_FAULT_BADMAP;
@@ -393,7 +393,7 @@ void __kprobes do_protection_exception(struct pt_regs *regs, long pgm_int_code,
 	int fault;
 
 	/* Protection exception is suppressing, decrement psw address. */
-	regs->psw.addr -= (pgm_int_code >> 16);
+	regs->psw.addr = __rewind_psw(regs->psw, pgm_int_code >> 16);
 	/*
 	 * Check for low-address protection.  This needs to be treated
 	 * as a special case because the translation exception code
@@ -454,7 +454,7 @@ int __handle_fault(unsigned long uaddr, unsigned long pgm_int_code, int write)
 	struct pt_regs regs;
 	int access, fault;
 
-	regs.psw.mask = psw_kernel_bits;
+	regs.psw.mask = psw_kernel_bits | PSW_MASK_DAT | PSW_MASK_MCHECK;
 	if (!irqs_disabled())
 		regs.psw.mask |= PSW_MASK_IO | PSW_MASK_EXT;
 	regs.psw.addr = (unsigned long) __builtin_return_address(0);
diff --git a/arch/s390/mm/gup.c b/arch/s390/mm/gup.c
index 45b405ca256..65cb06e2af4 100644
--- a/arch/s390/mm/gup.c
+++ b/arch/s390/mm/gup.c
@@ -52,7 +52,7 @@ 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;
+	struct page *head, *page, *tail;
 	int refs;
 
 	result = write ? 0 : _SEGMENT_ENTRY_RO;
@@ -64,6 +64,7 @@ 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;
@@ -81,6 +82,17 @@ static inline int gup_huge_pmd(pmd_t *pmdp, pmd_t pmd, unsigned long addr,
 		*nr -= refs;
 		while (refs--)
 			put_page(head);
+		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;
diff --git a/arch/s390/mm/maccess.c b/arch/s390/mm/maccess.c
index 5dbbaa6e594..1cb8427bedf 100644
--- a/arch/s390/mm/maccess.c
+++ b/arch/s390/mm/maccess.c
@@ -11,6 +11,7 @@
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/errno.h>
+#include <linux/gfp.h>
 #include <asm/system.h>
 
 /*
@@ -60,6 +61,9 @@ long probe_kernel_write(void *dst, const void *src, size_t size)
 	return copied < 0 ? -EFAULT : 0;
 }
 
+/*
+ * Copy memory in real mode (kernel to kernel)
+ */
 int memcpy_real(void *dest, void *src, size_t count)
 {
 	register unsigned long _dest asm("2") = (unsigned long) dest;
@@ -101,3 +105,55 @@ void copy_to_absolute_zero(void *dest, void *src, size_t count)
 	__ctl_load(cr0, 0, 0);
 	preempt_enable();
 }
+
+/*
+ * Copy memory from kernel (real) to user (virtual)
+ */
+int copy_to_user_real(void __user *dest, void *src, size_t count)
+{
+	int offs = 0, size, rc;
+	char *buf;
+
+	buf = (char *) __get_free_page(GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+	rc = -EFAULT;
+	while (offs < count) {
+		size = min(PAGE_SIZE, count - offs);
+		if (memcpy_real(buf, src + offs, size))
+			goto out;
+		if (copy_to_user(dest + offs, buf, size))
+			goto out;
+		offs += size;
+	}
+	rc = 0;
+out:
+	free_page((unsigned long) buf);
+	return rc;
+}
+
+/*
+ * Copy memory from user (virtual) to kernel (real)
+ */
+int copy_from_user_real(void *dest, void __user *src, size_t count)
+{
+	int offs = 0, size, rc;
+	char *buf;
+
+	buf = (char *) __get_free_page(GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+	rc = -EFAULT;
+	while (offs < count) {
+		size = min(PAGE_SIZE, count - offs);
+		if (copy_from_user(buf, src + offs, size))
+			goto out;
+		if (memcpy_real(dest + offs, buf, size))
+			goto out;
+		offs += size;
+	}
+	rc = 0;
+out:
+	free_page((unsigned long) buf);
+	return rc;
+}
diff --git a/arch/s390/mm/mmap.c b/arch/s390/mm/mmap.c
index c9a9f7f1818..f09c74881b7 100644
--- a/arch/s390/mm/mmap.c
+++ b/arch/s390/mm/mmap.c
@@ -26,6 +26,7 @@
 
 #include <linux/personality.h>
 #include <linux/mm.h>
+#include <linux/mman.h>
 #include <linux/module.h>
 #include <linux/random.h>
 #include <asm/pgalloc.h>
diff --git a/arch/s390/mm/pageattr.c b/arch/s390/mm/pageattr.c
index d013ed39743..b36537a5f43 100644
--- a/arch/s390/mm/pageattr.c
+++ b/arch/s390/mm/pageattr.c
@@ -5,6 +5,7 @@
 #include <linux/module.h>
 #include <linux/mm.h>
 #include <linux/hugetlb.h>
+#include <asm/cacheflush.h>
 #include <asm/pgtable.h>
 
 static void change_page_attr(unsigned long addr, int numpages,
diff --git a/arch/s390/mm/pgtable.c b/arch/s390/mm/pgtable.c
index 4d1f2bce87b..301c84d3b54 100644
--- a/arch/s390/mm/pgtable.c
+++ b/arch/s390/mm/pgtable.c
@@ -1,5 +1,5 @@
 /*
- *    Copyright IBM Corp. 2007,2009
+ *    Copyright IBM Corp. 2007,2011
  *    Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>
  */
 
@@ -160,6 +160,8 @@ struct gmap *gmap_alloc(struct mm_struct *mm)
 	table = (unsigned long *) page_to_phys(page);
 	crst_table_init(table, _REGION1_ENTRY_EMPTY);
 	gmap->table = table;
+	gmap->asce = _ASCE_TYPE_REGION1 | _ASCE_TABLE_LENGTH |
+		     _ASCE_USER_BITS | __pa(table);
 	list_add(&gmap->list, &mm->context.gmap_list);
 	return gmap;
 
@@ -220,6 +222,7 @@ void gmap_free(struct gmap *gmap)
 
 	/* Free all segment & region tables. */
 	down_read(&gmap->mm->mmap_sem);
+	spin_lock(&gmap->mm->page_table_lock);
 	list_for_each_entry_safe(page, next, &gmap->crst_list, lru) {
 		table = (unsigned long *) page_to_phys(page);
 		if ((*table & _REGION_ENTRY_TYPE_MASK) == 0)
@@ -228,6 +231,7 @@ void gmap_free(struct gmap *gmap)
 				gmap_unlink_segment(gmap, table);
 		__free_pages(page, ALLOC_ORDER);
 	}
+	spin_unlock(&gmap->mm->page_table_lock);
 	up_read(&gmap->mm->mmap_sem);
 	list_del(&gmap->list);
 	kfree(gmap);
@@ -240,10 +244,6 @@ EXPORT_SYMBOL_GPL(gmap_free);
  */
 void gmap_enable(struct gmap *gmap)
 {
-	/* Load primary space page table origin. */
-	S390_lowcore.user_asce = _ASCE_TYPE_REGION1 | _ASCE_TABLE_LENGTH |
-				 _ASCE_USER_BITS | __pa(gmap->table);
-	asm volatile("lctlg 1,1,%0\n" : : "m" (S390_lowcore.user_asce) );
 	S390_lowcore.gmap = (unsigned long) gmap;
 }
 EXPORT_SYMBOL_GPL(gmap_enable);
@@ -254,14 +254,13 @@ EXPORT_SYMBOL_GPL(gmap_enable);
  */
 void gmap_disable(struct gmap *gmap)
 {
-	/* Load primary space page table origin. */
-	S390_lowcore.user_asce =
-		gmap->mm->context.asce_bits | __pa(gmap->mm->pgd);
-	asm volatile("lctlg 1,1,%0\n" : : "m" (S390_lowcore.user_asce) );
 	S390_lowcore.gmap = 0UL;
 }
 EXPORT_SYMBOL_GPL(gmap_disable);
 
+/*
+ * gmap_alloc_table is assumed to be called with mmap_sem held
+ */
 static int gmap_alloc_table(struct gmap *gmap,
 			       unsigned long *table, unsigned long init)
 {
@@ -273,14 +272,12 @@ static int gmap_alloc_table(struct gmap *gmap,
 		return -ENOMEM;
 	new = (unsigned long *) page_to_phys(page);
 	crst_table_init(new, init);
-	down_read(&gmap->mm->mmap_sem);
 	if (*table & _REGION_ENTRY_INV) {
 		list_add(&page->lru, &gmap->crst_list);
 		*table = (unsigned long) new | _REGION_ENTRY_LENGTH |
 			(*table & _REGION_ENTRY_TYPE_MASK);
 	} else
 		__free_pages(page, ALLOC_ORDER);
-	up_read(&gmap->mm->mmap_sem);
 	return 0;
 }
 
@@ -305,19 +302,20 @@ int gmap_unmap_segment(struct gmap *gmap, unsigned long to, unsigned long len)
 
 	flush = 0;
 	down_read(&gmap->mm->mmap_sem);
+	spin_lock(&gmap->mm->page_table_lock);
 	for (off = 0; off < len; off += PMD_SIZE) {
 		/* Walk the guest addr space page table */
 		table = gmap->table + (((to + off) >> 53) & 0x7ff);
 		if (*table & _REGION_ENTRY_INV)
-			return 0;
+			goto out;
 		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
 		table = table + (((to + off) >> 42) & 0x7ff);
 		if (*table & _REGION_ENTRY_INV)
-			return 0;
+			goto out;
 		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
 		table = table + (((to + off) >> 31) & 0x7ff);
 		if (*table & _REGION_ENTRY_INV)
-			return 0;
+			goto out;
 		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
 		table = table + (((to + off) >> 20) & 0x7ff);
 
@@ -325,6 +323,8 @@ int gmap_unmap_segment(struct gmap *gmap, unsigned long to, unsigned long len)
 		flush |= gmap_unlink_segment(gmap, table);
 		*table = _SEGMENT_ENTRY_INV;
 	}
+out:
+	spin_unlock(&gmap->mm->page_table_lock);
 	up_read(&gmap->mm->mmap_sem);
 	if (flush)
 		gmap_flush_tlb(gmap);
@@ -355,6 +355,7 @@ int gmap_map_segment(struct gmap *gmap, unsigned long from,
 
 	flush = 0;
 	down_read(&gmap->mm->mmap_sem);
+	spin_lock(&gmap->mm->page_table_lock);
 	for (off = 0; off < len; off += PMD_SIZE) {
 		/* Walk the gmap address space page table */
 		table = gmap->table + (((to + off) >> 53) & 0x7ff);
@@ -378,19 +379,24 @@ int gmap_map_segment(struct gmap *gmap, unsigned long from,
 		flush |= gmap_unlink_segment(gmap, table);
 		*table = _SEGMENT_ENTRY_INV | _SEGMENT_ENTRY_RO | (from + off);
 	}
+	spin_unlock(&gmap->mm->page_table_lock);
 	up_read(&gmap->mm->mmap_sem);
 	if (flush)
 		gmap_flush_tlb(gmap);
 	return 0;
 
 out_unmap:
+	spin_unlock(&gmap->mm->page_table_lock);
 	up_read(&gmap->mm->mmap_sem);
 	gmap_unmap_segment(gmap, to, len);
 	return -ENOMEM;
 }
 EXPORT_SYMBOL_GPL(gmap_map_segment);
 
-unsigned long gmap_fault(unsigned long address, struct gmap *gmap)
+/*
+ * this function is assumed to be called with mmap_sem held
+ */
+unsigned long __gmap_fault(unsigned long address, struct gmap *gmap)
 {
 	unsigned long *table, vmaddr, segment;
 	struct mm_struct *mm;
@@ -450,16 +456,75 @@ unsigned long gmap_fault(unsigned long address, struct gmap *gmap)
 		page = pmd_page(*pmd);
 		mp = (struct gmap_pgtable *) page->index;
 		rmap->entry = table;
+		spin_lock(&mm->page_table_lock);
 		list_add(&rmap->list, &mp->mapper);
+		spin_unlock(&mm->page_table_lock);
 		/* Set gmap segment table entry to page table. */
 		*table = pmd_val(*pmd) & PAGE_MASK;
 		return vmaddr | (address & ~PMD_MASK);
 	}
 	return -EFAULT;
+}
+
+unsigned long gmap_fault(unsigned long address, struct gmap *gmap)
+{
+	unsigned long rc;
+
+	down_read(&gmap->mm->mmap_sem);
+	rc = __gmap_fault(address, gmap);
+	up_read(&gmap->mm->mmap_sem);
 
+	return rc;
 }
 EXPORT_SYMBOL_GPL(gmap_fault);
 
+void gmap_discard(unsigned long from, unsigned long to, struct gmap *gmap)
+{
+
+	unsigned long *table, address, size;
+	struct vm_area_struct *vma;
+	struct gmap_pgtable *mp;
+	struct page *page;
+
+	down_read(&gmap->mm->mmap_sem);
+	address = from;
+	while (address < to) {
+		/* Walk the gmap address space page table */
+		table = gmap->table + ((address >> 53) & 0x7ff);
+		if (unlikely(*table & _REGION_ENTRY_INV)) {
+			address = (address + PMD_SIZE) & PMD_MASK;
+			continue;
+		}
+		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
+		table = table + ((address >> 42) & 0x7ff);
+		if (unlikely(*table & _REGION_ENTRY_INV)) {
+			address = (address + PMD_SIZE) & PMD_MASK;
+			continue;
+		}
+		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
+		table = table + ((address >> 31) & 0x7ff);
+		if (unlikely(*table & _REGION_ENTRY_INV)) {
+			address = (address + PMD_SIZE) & PMD_MASK;
+			continue;
+		}
+		table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN);
+		table = table + ((address >> 20) & 0x7ff);
+		if (unlikely(*table & _SEGMENT_ENTRY_INV)) {
+			address = (address + PMD_SIZE) & PMD_MASK;
+			continue;
+		}
+		page = pfn_to_page(*table >> PAGE_SHIFT);
+		mp = (struct gmap_pgtable *) page->index;
+		vma = find_vma(gmap->mm, mp->vmaddr);
+		size = min(to - address, PMD_SIZE - (address & ~PMD_MASK));
+		zap_page_range(vma, mp->vmaddr | (address & ~PMD_MASK),
+			       size, NULL);
+		address = (address + PMD_SIZE) & PMD_MASK;
+	}
+	up_read(&gmap->mm->mmap_sem);
+}
+EXPORT_SYMBOL_GPL(gmap_discard);
+
 void gmap_unmap_notifier(struct mm_struct *mm, unsigned long *table)
 {
 	struct gmap_rmap *rmap, *next;
@@ -667,8 +732,9 @@ void page_table_free_rcu(struct mmu_gather *tlb, unsigned long *table)
 
 void __tlb_remove_table(void *_table)
 {
-	void *table = (void *)((unsigned long) _table & PAGE_MASK);
-	unsigned type = (unsigned long) _table & ~PAGE_MASK;
+	const unsigned long mask = (FRAG_MASK << 4) | FRAG_MASK;
+	void *table = (void *)((unsigned long) _table & ~mask);
+	unsigned type = (unsigned long) _table & mask;
 
 	if (type)
 		__page_table_free_rcu(table, type);
diff --git a/arch/s390/mm/vmem.c b/arch/s390/mm/vmem.c
index 781ff516956..4799383e2df 100644
--- a/arch/s390/mm/vmem.c
+++ b/arch/s390/mm/vmem.c
@@ -335,6 +335,9 @@ void __init vmem_map_init(void)
 	ro_start = ((unsigned long)&_stext) & PAGE_MASK;
 	ro_end = PFN_ALIGN((unsigned long)&_eshared);
 	for (i = 0; i < MEMORY_CHUNKS && memory_chunk[i].size > 0; i++) {
+		if (memory_chunk[i].type == CHUNK_CRASHK ||
+		    memory_chunk[i].type == CHUNK_OLDMEM)
+			continue;
 		start = memory_chunk[i].addr;
 		end = memory_chunk[i].addr + memory_chunk[i].size;
 		if (start >= ro_end || end <= ro_start)
@@ -368,6 +371,9 @@ static int __init vmem_convert_memory_chunk(void)
 	for (i = 0; i < MEMORY_CHUNKS; i++) {
 		if (!memory_chunk[i].size)
 			continue;
+		if (memory_chunk[i].type == CHUNK_CRASHK ||
+		    memory_chunk[i].type == CHUNK_OLDMEM)
+			continue;
 		seg = kzalloc(sizeof(*seg), GFP_KERNEL);
 		if (!seg)
 			panic("Out of memory...\n");