13 files changed, 111 insertions, 49 deletions

diff --git a/arch/x86/kernel/head32.c b/arch/x86/kernel/head32.c
index adedeef1dedc..b2e246037392 100644
--- a/arch/x86/kernel/head32.c
+++ b/arch/x86/kernel/head32.c
@@ -7,6 +7,7 @@
 
 #include <linux/init.h>
 #include <linux/start_kernel.h>
+#include <linux/mm.h>
 
 #include <asm/setup.h>
 #include <asm/sections.h>
@@ -44,9 +45,10 @@ void __init i386_start_kernel(void)
 #ifdef CONFIG_BLK_DEV_INITRD
         /* Reserve INITRD */
         if (boot_params.hdr.type_of_loader && boot_params.hdr.ramdisk_image) {
+                /* Assume only end is not page aligned */
                 u64 ramdisk_image = boot_params.hdr.ramdisk_image;
                 u64 ramdisk_size  = boot_params.hdr.ramdisk_size;
-                u64 ramdisk_end   = ramdisk_image + ramdisk_size;
+                u64 ramdisk_end   = PAGE_ALIGN(ramdisk_image + ramdisk_size);
                 reserve_early(ramdisk_image, ramdisk_end, "RAMDISK");
         }
 #endif
diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
index b5a9896ca1e7..7147143fd614 100644
--- a/arch/x86/kernel/head64.c
+++ b/arch/x86/kernel/head64.c
@@ -103,9 +103,10 @@ void __init x86_64_start_reservations(char *real_mode_data)
 #ifdef CONFIG_BLK_DEV_INITRD
         /* Reserve INITRD */
         if (boot_params.hdr.type_of_loader && boot_params.hdr.ramdisk_image) {
+                /* Assume only end is not page aligned */
                 unsigned long ramdisk_image = boot_params.hdr.ramdisk_image;
                 unsigned long ramdisk_size  = boot_params.hdr.ramdisk_size;
-                unsigned long ramdisk_end   = ramdisk_image + ramdisk_size;
+                unsigned long ramdisk_end   = PAGE_ALIGN(ramdisk_image + ramdisk_size);
                 reserve_early(ramdisk_image, ramdisk_end, "RAMDISK");
         }
 #endif
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 5d7ba1a449bd..d76e18570c60 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -314,16 +314,17 @@ static void __init reserve_brk(void)
 #define MAX_MAP_CHUNK   (NR_FIX_BTMAPS << PAGE_SHIFT)
 static void __init relocate_initrd(void)
 {
-
+        /* Assume only end is not page aligned */
         u64 ramdisk_image = boot_params.hdr.ramdisk_image;
         u64 ramdisk_size  = boot_params.hdr.ramdisk_size;
+        u64 area_size     = PAGE_ALIGN(ramdisk_size);
         u64 end_of_lowmem = max_low_pfn_mapped << PAGE_SHIFT;
         u64 ramdisk_here;
         unsigned long slop, clen, mapaddr;
         char *p, *q;
 
         /* We need to move the initrd down into lowmem */
-        ramdisk_here = find_e820_area(0, end_of_lowmem, ramdisk_size,
+        ramdisk_here = find_e820_area(0, end_of_lowmem, area_size,
                                          PAGE_SIZE);
 
         if (ramdisk_here == -1ULL)
@@ -332,7 +333,7 @@ static void __init relocate_initrd(void)
 
         /* Note: this includes all the lowmem currently occupied by
            the initrd, we rely on that fact to keep the data intact. */
-        reserve_early(ramdisk_here, ramdisk_here + ramdisk_size,
+        reserve_early(ramdisk_here, ramdisk_here + area_size,
                          "NEW RAMDISK");
         initrd_start = ramdisk_here + PAGE_OFFSET;
         initrd_end   = initrd_start + ramdisk_size;
@@ -376,9 +377,10 @@ static void __init relocate_initrd(void)
 
 static void __init reserve_initrd(void)
 {
+        /* Assume only end is not page aligned */
         u64 ramdisk_image = boot_params.hdr.ramdisk_image;
         u64 ramdisk_size  = boot_params.hdr.ramdisk_size;
-        u64 ramdisk_end   = ramdisk_image + ramdisk_size;
+        u64 ramdisk_end   = PAGE_ALIGN(ramdisk_image + ramdisk_size);
         u64 end_of_lowmem = max_low_pfn_mapped << PAGE_SHIFT;
 
         if (!boot_params.hdr.type_of_loader ||
diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S
index 44879df55696..2cc249718c46 100644
--- a/arch/x86/kernel/vmlinux.lds.S
+++ b/arch/x86/kernel/vmlinux.lds.S
@@ -291,8 +291,8 @@ SECTIONS
         .smp_locks : AT(ADDR(.smp_locks) - LOAD_OFFSET) {
                 __smp_locks = .;
                 *(.smp_locks)
-                __smp_locks_end = .;
                 . = ALIGN(PAGE_SIZE);
+                __smp_locks_end = .;
         }
 
 #ifdef CONFIG_X86_64
diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c
index e71c5cbc8f35..452ee5b8f309 100644
--- a/arch/x86/mm/init.c
+++ b/arch/x86/mm/init.c
@@ -331,11 +331,23 @@ int devmem_is_allowed(unsigned long pagenr)
 
 void free_init_pages(char *what, unsigned long begin, unsigned long end)
 {
-        unsigned long addr = begin;
+        unsigned long addr;
+        unsigned long begin_aligned, end_aligned;
 
-        if (addr >= end)
+        /* Make sure boundaries are page aligned */
+        begin_aligned = PAGE_ALIGN(begin);
+        end_aligned   = end & PAGE_MASK;
+
+        if (WARN_ON(begin_aligned != begin || end_aligned != end)) {
+                begin = begin_aligned;
+                end   = end_aligned;
+        }
+
+        if (begin >= end)
                 return;
 
+        addr = begin;
+
         /*
          * If debugging page accesses then do not free this memory but
          * mark them not present - any buggy init-section access will
@@ -343,7 +355,7 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end)
          */
 #ifdef CONFIG_DEBUG_PAGEALLOC
         printk(KERN_INFO "debug: unmapping init memory %08lx..%08lx\n",
-                begin, PAGE_ALIGN(end));
+                begin, end);
         set_memory_np(begin, (end - begin) >> PAGE_SHIFT);
 #else
         /*
@@ -358,8 +370,7 @@ void free_init_pages(char *what, unsigned long begin, unsigned long end)
         for (; addr < end; addr += PAGE_SIZE) {
                 ClearPageReserved(virt_to_page(addr));
                 init_page_count(virt_to_page(addr));
-                memset((void *)(addr & ~(PAGE_SIZE-1)),
-                        POISON_FREE_INITMEM, PAGE_SIZE);
+                memset((void *)addr, POISON_FREE_INITMEM, PAGE_SIZE);
                 free_page(addr);
                 totalram_pages++;
         }
@@ -376,6 +387,15 @@ void free_initmem(void)
 #ifdef CONFIG_BLK_DEV_INITRD
 void free_initrd_mem(unsigned long start, unsigned long end)
 {
-        free_init_pages("initrd memory", start, end);
+        /*
+         * end could be not aligned, and We can not align that,
+         * decompresser could be confused by aligned initrd_end
+         * We already reserve the end partial page before in
+         *   - i386_start_kernel()
+         *   - x86_64_start_kernel()
+         *   - relocate_initrd()
+         * So here We can do PAGE_ALIGN() safely to get partial page to be freed
+         */
+        free_init_pages("initrd memory", start, PAGE_ALIGN(end));
 }
 #endif
diff --git a/fs/logfs/dev_bdev.c b/fs/logfs/dev_bdev.c
index 9718c22f186d..a5d0c56d3ebc 100644
--- a/fs/logfs/dev_bdev.c
+++ b/fs/logfs/dev_bdev.c
@@ -80,6 +80,7 @@ static void writeseg_end_io(struct bio *bio, int err)
                         prefetchw(&bvec->bv_page->flags);
 
                 end_page_writeback(page);
+                page_cache_release(page);
         } while (bvec >= bio->bi_io_vec);
         bio_put(bio);
         if (atomic_dec_and_test(&super->s_pending_writes))
@@ -97,8 +98,10 @@ static int __bdev_writeseg(struct super_block *sb, u64 ofs, pgoff_t index,
         unsigned int max_pages = queue_max_hw_sectors(q) >> (PAGE_SHIFT - 9);
         int i;
 
+        if (max_pages > BIO_MAX_PAGES)
+                max_pages = BIO_MAX_PAGES;
         bio = bio_alloc(GFP_NOFS, max_pages);
-        BUG_ON(!bio); /* FIXME: handle this */
+        BUG_ON(!bio);
 
         for (i = 0; i < nr_pages; i++) {
                 if (i >= max_pages) {
@@ -191,8 +194,10 @@ static int do_erase(struct super_block *sb, u64 ofs, pgoff_t index,
         unsigned int max_pages = queue_max_hw_sectors(q) >> (PAGE_SHIFT - 9);
         int i;
 
+        if (max_pages > BIO_MAX_PAGES)
+                max_pages = BIO_MAX_PAGES;
         bio = bio_alloc(GFP_NOFS, max_pages);
-        BUG_ON(!bio); /* FIXME: handle this */
+        BUG_ON(!bio);
 
         for (i = 0; i < nr_pages; i++) {
                 if (i >= max_pages) {
diff --git a/fs/logfs/dir.c b/fs/logfs/dir.c
index 56a8bfbb0120..c76b4b5c7ff6 100644
--- a/fs/logfs/dir.c
+++ b/fs/logfs/dir.c
@@ -303,12 +303,12 @@ static int __logfs_readdir(struct file *file, void *buf, filldir_t filldir)
                                 (filler_t *)logfs_readpage, NULL);
                 if (IS_ERR(page))
                         return PTR_ERR(page);
-                dd = kmap_atomic(page, KM_USER0);
+                dd = kmap(page);
                 BUG_ON(dd->namelen == 0);
 
                 full = filldir(buf, (char *)dd->name, be16_to_cpu(dd->namelen),
                                 pos, be64_to_cpu(dd->ino), dd->type);
-                kunmap_atomic(dd, KM_USER0);
+                kunmap(page);
                 page_cache_release(page);
                 if (full)
                         break;
diff --git a/fs/logfs/journal.c b/fs/logfs/journal.c
index 6ad30a4c9052..d57c7b07b60b 100644
--- a/fs/logfs/journal.c
+++ b/fs/logfs/journal.c
@@ -800,6 +800,7 @@ void do_logfs_journal_wl_pass(struct super_block *sb)
 {
         struct logfs_super *super = logfs_super(sb);
         struct logfs_area *area = super->s_journal_area;
+        struct btree_head32 *head = &super->s_reserved_segments;
         u32 segno, ec;
         int i, err;
 
@@ -807,6 +808,7 @@ void do_logfs_journal_wl_pass(struct super_block *sb)
         /* Drop old segments */
         journal_for_each(i)
                 if (super->s_journal_seg[i]) {
+                        btree_remove32(head, super->s_journal_seg[i]);
                         logfs_set_segment_unreserved(sb,
                                         super->s_journal_seg[i],
                                         super->s_journal_ec[i]);
@@ -819,8 +821,13 @@ void do_logfs_journal_wl_pass(struct super_block *sb)
                 super->s_journal_seg[i] = segno;
                 super->s_journal_ec[i] = ec;
                 logfs_set_segment_reserved(sb, segno);
+                err = btree_insert32(head, segno, (void *)1, GFP_KERNEL);
+                BUG_ON(err); /* mempool should prevent this */
+                err = logfs_erase_segment(sb, segno, 1);
+                BUG_ON(err); /* FIXME: remount-ro would be nicer */
         }
         /* Manually move journal_area */
+        freeseg(sb, area->a_segno);
         area->a_segno = super->s_journal_seg[0];
         area->a_is_open = 0;
         area->a_used_bytes = 0;
diff --git a/fs/logfs/logfs.h b/fs/logfs/logfs.h
index 129779431373..b84b0eec6024 100644
--- a/fs/logfs/logfs.h
+++ b/fs/logfs/logfs.h
@@ -587,6 +587,7 @@ void move_page_to_btree(struct page *page);
 int logfs_init_mapping(struct super_block *sb);
 void logfs_sync_area(struct logfs_area *area);
 void logfs_sync_segments(struct super_block *sb);
+void freeseg(struct super_block *sb, u32 segno);
 
 /* area handling */
 int logfs_init_areas(struct super_block *sb);
diff --git a/fs/logfs/readwrite.c b/fs/logfs/readwrite.c
index 7a23b3e7c0a7..c3a3a6814b84 100644
--- a/fs/logfs/readwrite.c
+++ b/fs/logfs/readwrite.c
@@ -1594,7 +1594,6 @@ int logfs_delete(struct inode *inode, pgoff_t index,
         return ret;
 }
 
-/* Rewrite cannot mark the inode dirty but has to write it immediatly. */
 int logfs_rewrite_block(struct inode *inode, u64 bix, u64 ofs,
                 gc_level_t gc_level, long flags)
 {
@@ -1611,6 +1610,18 @@ int logfs_rewrite_block(struct inode *inode, u64 bix, u64 ofs,
                 if (level != 0)
                         alloc_indirect_block(inode, page, 0);
                 err = logfs_write_buf(inode, page, flags);
+                if (!err && shrink_level(gc_level) == 0) {
+                        /* Rewrite cannot mark the inode dirty but has to
+                         * write it immediatly.
+                         * Q: Can't we just create an alias for the inode
+                         * instead?  And if not, why not?
+                         */
+                        if (inode->i_ino == LOGFS_INO_MASTER)
+                                logfs_write_anchor(inode->i_sb);
+                        else {
+                                err = __logfs_write_inode(inode, flags);
+                        }
+                }
         }
         logfs_put_write_page(page);
         return err;
diff --git a/fs/logfs/segment.c b/fs/logfs/segment.c
index 1a14f9910d55..0ecd8f07c11e 100644
--- a/fs/logfs/segment.c
+++ b/fs/logfs/segment.c
@@ -93,50 +93,58 @@ void __logfs_buf_write(struct logfs_area *area, u64 ofs, void *buf, size_t len,
         } while (len);
 }
 
-/*
- * bdev_writeseg will write full pages.  Memset the tail to prevent data leaks.
- */
-static void pad_wbuf(struct logfs_area *area, int final)
+static void pad_partial_page(struct logfs_area *area)
 {
         struct super_block *sb = area->a_sb;
-        struct logfs_super *super = logfs_super(sb);
         struct page *page;
         u64 ofs = dev_ofs(sb, area->a_segno, area->a_used_bytes);
         pgoff_t index = ofs >> PAGE_SHIFT;
         long offset = ofs & (PAGE_SIZE-1);
         u32 len = PAGE_SIZE - offset;
 
-        if (len == PAGE_SIZE) {
-                /* The math in this function can surely use some love */
-                len = 0;
-        }
-        if (len) {
-                BUG_ON(area->a_used_bytes >= super->s_segsize);
-
-                page = get_mapping_page(area->a_sb, index, 0);
+        if (len % PAGE_SIZE) {
+                page = get_mapping_page(sb, index, 0);
                 BUG_ON(!page); /* FIXME: reserve a pool */
                 memset(page_address(page) + offset, 0xff, len);
                 SetPagePrivate(page);
                 page_cache_release(page);
         }
+}
 
-        if (!final)
-                return;
+static void pad_full_pages(struct logfs_area *area)
+{
+        struct super_block *sb = area->a_sb;
+        struct logfs_super *super = logfs_super(sb);
+        u64 ofs = dev_ofs(sb, area->a_segno, area->a_used_bytes);
+        u32 len = super->s_segsize - area->a_used_bytes;
+        pgoff_t index = PAGE_CACHE_ALIGN(ofs) >> PAGE_CACHE_SHIFT;
+        pgoff_t no_indizes = len >> PAGE_CACHE_SHIFT;
+        struct page *page;
 
-        area->a_used_bytes += len;
-        for ( ; area->a_used_bytes < super->s_segsize;
-                        area->a_used_bytes += PAGE_SIZE) {
-                /* Memset another page */
-                index++;
-                page = get_mapping_page(area->a_sb, index, 0);
+        while (no_indizes) {
+                page = get_mapping_page(sb, index, 0);
                 BUG_ON(!page); /* FIXME: reserve a pool */
-                memset(page_address(page), 0xff, PAGE_SIZE);
+                SetPageUptodate(page);
+                memset(page_address(page), 0xff, PAGE_CACHE_SIZE);
                 SetPagePrivate(page);
                 page_cache_release(page);
+                index++;
+                no_indizes--;
         }
 }
 
 /*
+ * bdev_writeseg will write full pages.  Memset the tail to prevent data leaks.
+ * Also make sure we allocate (and memset) all pages for final writeout.
+ */
+static void pad_wbuf(struct logfs_area *area, int final)
+{
+        pad_partial_page(area);
+        if (final)
+                pad_full_pages(area);
+}
+
+/*
  * We have to be careful with the alias tree.  Since lookup is done by bix,
  * it needs to be normalized, so 14, 15, 16, etc. all match when dealing with
  * indirect blocks.  So always use it through accessor functions.
@@ -683,7 +691,7 @@ int logfs_segment_delete(struct inode *inode, struct logfs_shadow *shadow)
         return 0;
 }
 
-static void freeseg(struct super_block *sb, u32 segno)
+void freeseg(struct super_block *sb, u32 segno)
 {
         struct logfs_super *super = logfs_super(sb);
         struct address_space *mapping = super->s_mapping_inode->i_mapping;
diff --git a/fs/logfs/super.c b/fs/logfs/super.c
index c66beab78dee..9d856c49afc5 100644
--- a/fs/logfs/super.c
+++ b/fs/logfs/super.c
@@ -277,7 +277,7 @@ static int logfs_recover_sb(struct super_block *sb)
         }
         if (valid0 && valid1 && ds_cmp(ds0, ds1)) {
                 printk(KERN_INFO"Superblocks don't match - fixing.\n");
-                return write_one_sb(sb, super->s_devops->find_last_sb);
+                return logfs_write_sb(sb);
         }
         /* If neither is valid now, something's wrong.  Didn't we properly
          * check them before?!? */
@@ -289,6 +289,10 @@ static int logfs_make_writeable(struct super_block *sb)
 {
         int err;
 
+        err = logfs_open_segfile(sb);
+        if (err)
+                return err;
+
         /* Repair any broken superblock copies */
         err = logfs_recover_sb(sb);
         if (err)
@@ -299,10 +303,6 @@ static int logfs_make_writeable(struct super_block *sb)
         if (err)
                 return err;
 
-        err = logfs_open_segfile(sb);
-        if (err)
-                return err;
-
         /* Do one GC pass before any data gets dirtied */
         logfs_gc_pass(sb);
 
@@ -328,7 +328,7 @@ static int logfs_get_sb_final(struct super_block *sb, struct vfsmount *mnt)
 
         sb->s_root = d_alloc_root(rootdir);
         if (!sb->s_root)
-                goto fail;
+                goto fail2;
 
         super->s_erase_page = alloc_pages(GFP_KERNEL, 0);
         if (!super->s_erase_page)
@@ -572,8 +572,7 @@ int logfs_get_sb_device(struct file_system_type *type, int flags,
         return 0;
 
 err1:
-        up_write(&sb->s_umount);
-        deactivate_super(sb);
+        deactivate_locked_super(sb);
         return err;
 err0:
         kfree(super);
diff --git a/kernel/early_res.c b/kernel/early_res.c
index 3cb2c661bb78..31aa9332ef3f 100644
--- a/kernel/early_res.c
+++ b/kernel/early_res.c
@@ -333,6 +333,12 @@ void __init free_early_partial(u64 start, u64 end)
         struct early_res *r;
         int i;
 
+        if (start == end)
+                return;
+
+        if (WARN_ONCE(start > end, "  wrong range [%#llx, %#llx]\n", start, end))
+                return;
+
 try_next:
         i = find_overlapped_early(start, end);
         if (i >= max_early_res)