Merge branch 'akpm' (patches from Andrew Morton)

Merge a bunch of fixes from Andrew Morton.

* emailed patches from Andrew Morton <[email protected]>:
  fs/proc/task_mmu.c: fix buffer overflow in add_page_map()
  arch: *: Kconfig: add "kernel/Kconfig.freezer" to "arch/*/Kconfig"
  ocfs2: fix null pointer dereference in ocfs2_dir_foreach_blk_id()
  x86 get_unmapped_area(): use proper mmap base for bottom-up direction
  ocfs2: fix NULL pointer dereference in ocfs2_duplicate_clusters_by_page
  ocfs2: Revert 40bd62e to avoid regression in extended allocation
  drivers/rtc/rtc-stmp3xxx.c: provide timeout for potentially endless loop polling a HW bit
  hugetlb: fix lockdep splat caused by pmd sharing
  aoe: adjust ref of head for compound page tails
  microblaze: fix clone syscall
  mm: save soft-dirty bits on file pages
  mm: save soft-dirty bits on swapped pages
  memcg: don't initialize kmem-cache destroying work for root caches

Author: torvalds

arch/Kconfig

@@ -407,6 +407,12 @@ config CLONE_BACKWARDS2
 	help
 	  Architecture has the first two arguments of clone(2) swapped.
 
+config CLONE_BACKWARDS3
+	bool
+	help
+	  Architecture has tls passed as the 3rd argument of clone(2),
+	  not the 5th one.
+
 config ODD_RT_SIGACTION
 	bool
 	help

arch/hexagon/Kconfig

@@ -158,6 +158,7 @@ source "kernel/Kconfig.hz"
 endmenu
 
 source "init/Kconfig"
+source "kernel/Kconfig.freezer"
 source "drivers/Kconfig"
 source "fs/Kconfig"
 

arch/microblaze/Kconfig

@@ -28,7 +28,7 @@ config MICROBLAZE
 	select GENERIC_CLOCKEVENTS
 	select GENERIC_IDLE_POLL_SETUP
 	select MODULES_USE_ELF_RELA
-	select CLONE_BACKWARDS
+	select CLONE_BACKWARDS3
 
 config SWAP
 	def_bool n

arch/openrisc/Kconfig

@@ -55,6 +55,7 @@ config GENERIC_CSUM
 
 source "init/Kconfig"
 
+source "kernel/Kconfig.freezer"
 
 menu "Processor type and features"
 

arch/score/Kconfig

@@ -87,6 +87,8 @@ config STACKTRACE_SUPPORT
 
 source "init/Kconfig"
 
+source "kernel/Kconfig.freezer"
+
 config MMU
 	def_bool y
 

arch/x86/include/asm/pgtable-2level.h

@@ -55,9 +55,53 @@ static inline pmd_t native_pmdp_get_and_clear(pmd_t *xp)
 #define native_pmdp_get_and_clear(xp) native_local_pmdp_get_and_clear(xp)
 #endif
 
+#ifdef CONFIG_MEM_SOFT_DIRTY
+
+/*
+ * Bits _PAGE_BIT_PRESENT, _PAGE_BIT_FILE, _PAGE_BIT_SOFT_DIRTY and
+ * _PAGE_BIT_PROTNONE are taken, split up the 28 bits of offset
+ * into this range.
+ */
+#define PTE_FILE_MAX_BITS	28
+#define PTE_FILE_SHIFT1		(_PAGE_BIT_PRESENT + 1)
+#define PTE_FILE_SHIFT2		(_PAGE_BIT_FILE + 1)
+#define PTE_FILE_SHIFT3		(_PAGE_BIT_PROTNONE + 1)
+#define PTE_FILE_SHIFT4		(_PAGE_BIT_SOFT_DIRTY + 1)
+#define PTE_FILE_BITS1		(PTE_FILE_SHIFT2 - PTE_FILE_SHIFT1 - 1)
+#define PTE_FILE_BITS2		(PTE_FILE_SHIFT3 - PTE_FILE_SHIFT2 - 1)
+#define PTE_FILE_BITS3		(PTE_FILE_SHIFT4 - PTE_FILE_SHIFT3 - 1)
+
+#define pte_to_pgoff(pte)						\
+	((((pte).pte_low >> (PTE_FILE_SHIFT1))				\
+	  & ((1U << PTE_FILE_BITS1) - 1)))				\
+	+ ((((pte).pte_low >> (PTE_FILE_SHIFT2))			\
+	    & ((1U << PTE_FILE_BITS2) - 1))				\
+	   << (PTE_FILE_BITS1))						\
+	+ ((((pte).pte_low >> (PTE_FILE_SHIFT3))			\
+	    & ((1U << PTE_FILE_BITS3) - 1))				\
+	   << (PTE_FILE_BITS1 + PTE_FILE_BITS2))			\
+	+ ((((pte).pte_low >> (PTE_FILE_SHIFT4)))			\
+	    << (PTE_FILE_BITS1 + PTE_FILE_BITS2 + PTE_FILE_BITS3))
+
+#define pgoff_to_pte(off)						\
+	((pte_t) { .pte_low =						\
+	 ((((off)) & ((1U << PTE_FILE_BITS1) - 1)) << PTE_FILE_SHIFT1)	\
+	 + ((((off) >> PTE_FILE_BITS1)					\
+	     & ((1U << PTE_FILE_BITS2) - 1))				\
+	    << PTE_FILE_SHIFT2)						\
+	 + ((((off) >> (PTE_FILE_BITS1 + PTE_FILE_BITS2))		\
+	     & ((1U << PTE_FILE_BITS3) - 1))				\
+	    << PTE_FILE_SHIFT3)						\
+	 + ((((off) >>							\
+	      (PTE_FILE_BITS1 + PTE_FILE_BITS2 + PTE_FILE_BITS3)))	\
+	    << PTE_FILE_SHIFT4)						\
+	 + _PAGE_FILE })
+
+#else /* CONFIG_MEM_SOFT_DIRTY */
+
 /*
  * Bits _PAGE_BIT_PRESENT, _PAGE_BIT_FILE and _PAGE_BIT_PROTNONE are taken,
- * split up the 29 bits of offset into this range:
+ * split up the 29 bits of offset into this range.
  */
 #define PTE_FILE_MAX_BITS	29
 #define PTE_FILE_SHIFT1		(_PAGE_BIT_PRESENT + 1)
@@ -88,6 +132,8 @@ static inline pmd_t native_pmdp_get_and_clear(pmd_t *xp)
 	    << PTE_FILE_SHIFT3)						\
 	 + _PAGE_FILE })
 
+#endif /* CONFIG_MEM_SOFT_DIRTY */
+
 /* Encode and de-code a swap entry */
 #if _PAGE_BIT_FILE < _PAGE_BIT_PROTNONE
 #define SWP_TYPE_BITS (_PAGE_BIT_FILE - _PAGE_BIT_PRESENT - 1)

arch/x86/include/asm/pgtable-3level.h

@@ -179,6 +179,9 @@ static inline pmd_t native_pmdp_get_and_clear(pmd_t *pmdp)
 /*
  * Bits 0, 6 and 7 are taken in the low part of the pte,
  * put the 32 bits of offset into the high part.
+ *
+ * For soft-dirty tracking 11 bit is taken from
+ * the low part of pte as well.
  */
 #define pte_to_pgoff(pte) ((pte).pte_high)
 #define pgoff_to_pte(off)						\

arch/x86/include/asm/pgtable.h

@@ -314,6 +314,36 @@ static inline pmd_t pmd_mksoft_dirty(pmd_t pmd)
 	return pmd_set_flags(pmd, _PAGE_SOFT_DIRTY);
 }
 
+static inline pte_t pte_swp_mksoft_dirty(pte_t pte)
+{
+	return pte_set_flags(pte, _PAGE_SWP_SOFT_DIRTY);
+}
+
+static inline int pte_swp_soft_dirty(pte_t pte)
+{
+	return pte_flags(pte) & _PAGE_SWP_SOFT_DIRTY;
+}
+
+static inline pte_t pte_swp_clear_soft_dirty(pte_t pte)
+{
+	return pte_clear_flags(pte, _PAGE_SWP_SOFT_DIRTY);
+}
+
+static inline pte_t pte_file_clear_soft_dirty(pte_t pte)
+{
+	return pte_clear_flags(pte, _PAGE_SOFT_DIRTY);
+}
+
+static inline pte_t pte_file_mksoft_dirty(pte_t pte)
+{
+	return pte_set_flags(pte, _PAGE_SOFT_DIRTY);
+}
+
+static inline int pte_file_soft_dirty(pte_t pte)
+{
+	return pte_flags(pte) & _PAGE_SOFT_DIRTY;
+}
+
 /*
  * Mask out unsupported bits in a present pgprot.  Non-present pgprots
  * can use those bits for other purposes, so leave them be.

arch/x86/include/asm/pgtable_types.h

@@ -61,12 +61,27 @@
  * they do not conflict with each other.
  */
 
+#define _PAGE_BIT_SOFT_DIRTY	_PAGE_BIT_HIDDEN
+
 #ifdef CONFIG_MEM_SOFT_DIRTY
-#define _PAGE_SOFT_DIRTY	(_AT(pteval_t, 1) << _PAGE_BIT_HIDDEN)
+#define _PAGE_SOFT_DIRTY	(_AT(pteval_t, 1) << _PAGE_BIT_SOFT_DIRTY)
 #else
 #define _PAGE_SOFT_DIRTY	(_AT(pteval_t, 0))
 #endif
 
+/*
+ * Tracking soft dirty bit when a page goes to a swap is tricky.
+ * We need a bit which can be stored in pte _and_ not conflict
+ * with swap entry format. On x86 bits 6 and 7 are *not* involved
+ * into swap entry computation, but bit 6 is used for nonlinear
+ * file mapping, so we borrow bit 7 for soft dirty tracking.
+ */
+#ifdef CONFIG_MEM_SOFT_DIRTY
+#define _PAGE_SWP_SOFT_DIRTY	_PAGE_PSE
+#else
+#define _PAGE_SWP_SOFT_DIRTY	(_AT(pteval_t, 0))
+#endif
+
 #if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE)
 #define _PAGE_NX	(_AT(pteval_t, 1) << _PAGE_BIT_NX)
 #else

arch/x86/kernel/sys_x86_64.c

@@ -101,7 +101,7 @@ static void find_start_end(unsigned long flags, unsigned long *begin,
 				*begin = new_begin;
 		}
 	} else {
-		*begin = TASK_UNMAPPED_BASE;
+		*begin = mmap_legacy_base();
 		*end = TASK_SIZE;
 	}
 }

arch/x86/mm/mmap.c

@@ -98,7 +98,7 @@ static unsigned long mmap_base(void)
  * Bottom-up (legacy) layout on X86_32 did not support randomization, X86_64
  * does, but not when emulating X86_32
  */
-static unsigned long mmap_legacy_base(void)
+unsigned long mmap_legacy_base(void)
 {
 	if (mmap_is_ia32())
 		return TASK_UNMAPPED_BASE;

drivers/block/aoe/aoecmd.c

@@ -906,16 +906,10 @@ bio_pageinc(struct bio *bio)
 	int i;
 
 	bio_for_each_segment(bv, bio, i) {
-		page = bv->bv_page;
 		/* Non-zero page count for non-head members of
-		 * compound pages is no longer allowed by the kernel,
-		 * but this has never been seen here.
+		 * compound pages is no longer allowed by the kernel.
 		 */
-		if (unlikely(PageCompound(page)))
-			if (compound_trans_head(page) != page) {
-				pr_crit("page tail used for block I/O\n");
-				BUG();
-			}
+		page = compound_trans_head(bv->bv_page);
 		atomic_inc(&page->_count);
 	}
 }
@@ -924,10 +918,13 @@ static void
 bio_pagedec(struct bio *bio)
 {
 	struct bio_vec *bv;
+	struct page *page;
 	int i;
 
-	bio_for_each_segment(bv, bio, i)
-		atomic_dec(&bv->bv_page->_count);
+	bio_for_each_segment(bv, bio, i) {
+		page = compound_trans_head(bv->bv_page);
+		atomic_dec(&page->_count);
+	}
 }
 
 static void

drivers/rtc/rtc-stmp3xxx.c

@@ -23,6 +23,7 @@
 #include <linux/init.h>
 #include <linux/platform_device.h>
 #include <linux/interrupt.h>
+#include <linux/delay.h>
 #include <linux/rtc.h>
 #include <linux/slab.h>
 #include <linux/of_device.h>
@@ -119,24 +120,39 @@ static void stmp3xxx_wdt_register(struct platform_device *rtc_pdev)
 }
 #endif /* CONFIG_STMP3XXX_RTC_WATCHDOG */
 
-static void stmp3xxx_wait_time(struct stmp3xxx_rtc_data *rtc_data)
+static int stmp3xxx_wait_time(struct stmp3xxx_rtc_data *rtc_data)
 {
+	int timeout = 5000; /* 3ms according to i.MX28 Ref Manual */
 	/*
-	 * The datasheet doesn't say which way round the
-	 * NEW_REGS/STALE_REGS bitfields go. In fact it's 0x1=P0,
-	 * 0x2=P1, .., 0x20=P5, 0x40=ALARM, 0x80=SECONDS
+	 * The i.MX28 Applications Processor Reference Manual, Rev. 1, 2010
+	 * states:
+	 * | The order in which registers are updated is
+	 * | Persistent 0, 1, 2, 3, 4, 5, Alarm, Seconds.
+	 * | (This list is in bitfield order, from LSB to MSB, as they would
+	 * | appear in the STALE_REGS and NEW_REGS bitfields of the HW_RTC_STAT
+	 * | register. For example, the Seconds register corresponds to
+	 * | STALE_REGS or NEW_REGS containing 0x80.)
 	 */
-	while (readl(rtc_data->io + STMP3XXX_RTC_STAT) &
-			(0x80 << STMP3XXX_RTC_STAT_STALE_SHIFT))
-		cpu_relax();
+	do {
+		if (!(readl(rtc_data->io + STMP3XXX_RTC_STAT) &
+				(0x80 << STMP3XXX_RTC_STAT_STALE_SHIFT)))
+			return 0;
+		udelay(1);
+	} while (--timeout > 0);
+	return (readl(rtc_data->io + STMP3XXX_RTC_STAT) &
+		(0x80 << STMP3XXX_RTC_STAT_STALE_SHIFT)) ? -ETIME : 0;
 }
 
 /* Time read/write */
 static int stmp3xxx_rtc_gettime(struct device *dev, struct rtc_time *rtc_tm)
 {
+	int ret;
 	struct stmp3xxx_rtc_data *rtc_data = dev_get_drvdata(dev);
 
-	stmp3xxx_wait_time(rtc_data);
+	ret = stmp3xxx_wait_time(rtc_data);
+	if (ret)
+		return ret;
+
 	rtc_time_to_tm(readl(rtc_data->io + STMP3XXX_RTC_SECONDS), rtc_tm);
 	return 0;
 }
@@ -146,8 +162,7 @@ static int stmp3xxx_rtc_set_mmss(struct device *dev, unsigned long t)
 	struct stmp3xxx_rtc_data *rtc_data = dev_get_drvdata(dev);
 
 	writel(t, rtc_data->io + STMP3XXX_RTC_SECONDS);
-	stmp3xxx_wait_time(rtc_data);
-	return 0;
+	return stmp3xxx_wait_time(rtc_data);
 }
 
 /* interrupt(s) handler */

fs/hugetlbfs/inode.c

@@ -463,6 +463,14 @@ static struct inode *hugetlbfs_get_root(struct super_block *sb,
 	return inode;
 }
 
+/*
+ * Hugetlbfs is not reclaimable; therefore its i_mmap_mutex will never
+ * be taken from reclaim -- unlike regular filesystems. This needs an
+ * annotation because huge_pmd_share() does an allocation under
+ * i_mmap_mutex.
+ */
+struct lock_class_key hugetlbfs_i_mmap_mutex_key;
+
 static struct inode *hugetlbfs_get_inode(struct super_block *sb,
 					struct inode *dir,
 					umode_t mode, dev_t dev)
@@ -474,6 +482,8 @@ static struct inode *hugetlbfs_get_inode(struct super_block *sb,
 		struct hugetlbfs_inode_info *info;
 		inode->i_ino = get_next_ino();
 		inode_init_owner(inode, dir, mode);
+		lockdep_set_class(&inode->i_mapping->i_mmap_mutex,
+				&hugetlbfs_i_mmap_mutex_key);
 		inode->i_mapping->a_ops = &hugetlbfs_aops;
 		inode->i_mapping->backing_dev_info =&hugetlbfs_backing_dev_info;
 		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;

fs/ocfs2/aops.c

@@ -1757,7 +1757,7 @@ int ocfs2_write_begin_nolock(struct file *filp,
 		goto out;
 	} else if (ret == 1) {
 		clusters_need = wc->w_clen;
-		ret = ocfs2_refcount_cow(inode, filp, di_bh,
+		ret = ocfs2_refcount_cow(inode, di_bh,
 					 wc->w_cpos, wc->w_clen, UINT_MAX);
 		if (ret) {
 			mlog_errno(ret);

fs/ocfs2/dir.c

@@ -2153,11 +2153,9 @@ int ocfs2_empty_dir(struct inode *inode)
 {
 	int ret;
 	struct ocfs2_empty_dir_priv priv = {
-		.ctx.actor = ocfs2_empty_dir_filldir
+		.ctx.actor = ocfs2_empty_dir_filldir,
 	};
 
-	memset(&priv, 0, sizeof(priv));
-
 	if (ocfs2_dir_indexed(inode)) {
 		ret = ocfs2_empty_dir_dx(inode, &priv);
 		if (ret)

fs/ocfs2/file.c

@@ -370,7 +370,7 @@ static int ocfs2_cow_file_pos(struct inode *inode,
 	if (!(ext_flags & OCFS2_EXT_REFCOUNTED))
 		goto out;
 
-	return ocfs2_refcount_cow(inode, NULL, fe_bh, cpos, 1, cpos+1);
+	return ocfs2_refcount_cow(inode, fe_bh, cpos, 1, cpos+1);
 
 out:
 	return status;
@@ -899,7 +899,7 @@ static int ocfs2_zero_extend_get_range(struct inode *inode,
 		zero_clusters = last_cpos - zero_cpos;
 
 	if (needs_cow) {
-		rc = ocfs2_refcount_cow(inode, NULL, di_bh, zero_cpos,
+		rc = ocfs2_refcount_cow(inode, di_bh, zero_cpos,
 					zero_clusters, UINT_MAX);
 		if (rc) {
 			mlog_errno(rc);
@@ -2078,7 +2078,7 @@ static int ocfs2_prepare_inode_for_refcount(struct inode *inode,
 
 	*meta_level = 1;
 
-	ret = ocfs2_refcount_cow(inode, file, di_bh, cpos, clusters, UINT_MAX);
+	ret = ocfs2_refcount_cow(inode, di_bh, cpos, clusters, UINT_MAX);
 	if (ret)
 		mlog_errno(ret);
 out:

fs/ocfs2/journal.h

@@ -537,7 +537,7 @@ static inline int ocfs2_calc_extend_credits(struct super_block *sb,
 	extent_blocks = 1 + 1 + le16_to_cpu(root_el->l_tree_depth);
 
 	return bitmap_blocks + sysfile_bitmap_blocks + extent_blocks +
-	       ocfs2_quota_trans_credits(sb) + bits_wanted;
+	       ocfs2_quota_trans_credits(sb);
 }
 
 static inline int ocfs2_calc_symlink_credits(struct super_block *sb)

fs/ocfs2/move_extents.c

@@ -69,7 +69,7 @@ static int __ocfs2_move_extent(handle_t *handle,
 	u64 ino = ocfs2_metadata_cache_owner(context->et.et_ci);
 	u64 old_blkno = ocfs2_clusters_to_blocks(inode->i_sb, p_cpos);
 
-	ret = ocfs2_duplicate_clusters_by_page(handle, context->file, cpos,
+	ret = ocfs2_duplicate_clusters_by_page(handle, inode, cpos,
 					       p_cpos, new_p_cpos, len);
 	if (ret) {
 		mlog_errno(ret);