Large store support for ext2fs

This is a revised version of the large store patch for ext2fs, written
by Ognyan Kulev. It provides support for stores larger than 2 GiB.

* ext2fs/balloc.c: Use the new disk_cache_block_ref and disk_cache_block_deref
functions to access blocks from the disk cache.
* ext2fs/ext2fs.c (main): Update initialization call to pokel_init, and
call map_hypermetadata instead of get_hypermetadata.
* ext2fs/ext2fs.h: Include <hurd/store.h> and <hurd/ihash.h>.
(DISK_CACHE_BLOCKS): New macro.
(DC_INCORE): Likewise.
(DC_UNTOUCHED): Likewise.
(DC_FIXED): Likewise.
(DC_DONT_REUSE): Likewise.
(DC_NO_BLOCK): Likewise.
(DISK_CACHE_LAST_READ_XOR) [!NDEBUG]: Likewise.
(struct disk_cache_info): New structure.
(disk_cache): New external variable.
(disk_cache_size): Likewise.
(disk_cache_blocks): Likewise.
(disk_cache_bptr): Likewise.
(disk_cache_info): Likewise.
(disk_cache_lock): Likewise.
(disk_cache_reassociation): Likewise.
(disk_cache_block_ref): New declaration.
(disk_cache_block_ref_ptr): Likewise.
(disk_cache_block_deref): Likewise.
(disk_cache_block_is_ref): Likewise.
(map_hypermetadata): Likewise.
(trunc_block): Cast to off_t.
(round_block): Likewise.
(boffs): Likewise.
(bptr_index): New macro.
(boffs_ptr): Rewrite as an inline function to make it look up a block from
the disk cache.
(bptr_offs): Likewise.
(dino): Remove function, replaced with ...
(dino_ref): ... this one, which adds a reference to the inode block.
(dino_deref): New inline function.
(record_global_poke): Make sure block is referenced.
(record_indir_poke): Likewise.
(sync_global_ptr): Remove block reference, and adjust call to
pager_sync_some.
(sync_global): Add debug call to print wait parameter.
* ext2fs/getblk.c: Use the new disk_cache_block_ref and disk_cache_block_deref
functions to access blocks from the disk cache.
* ext2fs/hyper.c (get_hypermetadata): Read the superblock from the store
now that it's not directly mapped in memory. Move the initialization of
zeroblock here from ...
(map_hypermetadata): ... here. Also, set the superblock pointer.
(diskfs_set_hypermetadata): Add a reference to the superblock.
(diskfs_readonly_changed): Update call to mprotect.
* ext2fs/ialloc.c: Use the new disk_cache_block_ref, disk_cache_block_ref_ptr
and disk_cache_block_deref functions to access blocks from the disk cache.
* ext2fs/inode.c: Update calls that used the disk image to use the disk cache,
and use the new reference handling functions where appropriate.
* ext2fs/pager.c: Include <unistd.h> and "../libpager/priv.h".
(disk_image): Remove global variable.
(disk_pager_read_page): Update cache information.
(disk_pager_write_page): Likewise.
(disk_pager_notify_evict): New function.
(pager_notify_evict): Call disk_pager_notify_evict appropriately.
(disk_cache): New global variable.
(disk_cache_size): Likewise.
(disk_cache_blocks): Likewise.
(disk_cache_bptr): Likewise.
(disk_cache_info): Likewise.
(disk_cache_hint): Likewise.
(disk_cache_lock): Likewise.
(disk_cache_reassociation): Likewise.
(disk_cache_init): New function.
(disk_cache_return_unused): Likewise.
(disk_cache_block_ref): Likewise.
(disk_cache_block_ref_ptr): Likewise.
(disk_cache_block_deref): Likewise.
(disk_cache_block_is_ref): Likewise.
(create_disk_pager): Update initialization of the disk pager.
* ext2fs/pokel.c (pokel_add): Drop block references with disk_cache_block_deref.
(_pokel_exec): Likewise.
* ext2fs/truncate.c (trunc_indirect): Use the new disk_cache_block_ref and
disk_cache_block_deref functions to access blocks from the disk cache.

1 files changed, 31 insertions, 9 deletions

diff --git a/ext2fs/inode.c b/ext2fs/inode.c
index 2c442795..e75c63f9 100644
--- a/ext2fs/inode.c
+++ b/ext2fs/inode.c
@@ -92,7 +92,7 @@ diskfs_cached_lookup (ino_t inum, struct node **npp)
   dn->dir_idx = 0;
   dn->pager = 0;
   pthread_rwlock_init (&dn->alloc_lock, NULL);
-  pokel_init (&dn->indir_pokel, diskfs_disk_pager, disk_image);
+  pokel_init (&dn->indir_pokel, diskfs_disk_pager, disk_cache);
 
   /* Create the new node.  */
   np = diskfs_make_node (dn);
@@ -201,13 +201,17 @@ read_node (struct node *np)
   error_t err;
   struct stat *st = &np->dn_stat;
   struct disknode *dn = np->dn;
-  struct ext2_inode *di = dino (np->cache_id);
+  struct ext2_inode *di;
   struct ext2_inode_info *info = &dn->info;
 
+  ext2_debug ("(%llu)", np->cache_id);
+
   err = diskfs_catch_exception ();
   if (err)
     return err;
 
+  di = dino_ref (np->cache_id);
+
   st->st_fstype = FSTYPE_EXT2FS;
   st->st_fsid = getpid ();	/* This call is very cheap.  */
   st->st_ino = np->cache_id;
@@ -285,7 +289,9 @@ read_node (struct node *np)
       info->i_high_size = di->i_size_high;
       if (info->i_high_size)	/* XXX */
 	{
+	  dino_deref (di);
 	  ext2_warning ("cannot handle large file inode %Ld", np->cache_id);
+	  diskfs_end_catch_exception ();
 	  return EFBIG;
 	}
     }
@@ -307,6 +313,7 @@ read_node (struct node *np)
     }
   dn->info_i_translator = di->i_translator;
 
+  dino_deref (di);
   diskfs_end_catch_exception ();
 
   if (S_ISREG (st->st_mode) || S_ISDIR (st->st_mode)
@@ -408,7 +415,9 @@ write_node (struct node *np)
 {
   error_t err;
   struct stat *st = &np->dn_stat;
-  struct ext2_inode *di = dino (np->cache_id);
+  struct ext2_inode *di;
+
+  ext2_debug ("(%llu)", np->cache_id);
 
   if (np->dn->info.i_prealloc_count)
     ext2_discard_prealloc (np);
@@ -425,6 +434,8 @@ write_node (struct node *np)
       if (err)
 	return NULL;
 
+      di = dino_ref (np->cache_id);
+
       di->i_generation = st->st_gen;
 
       /* We happen to know that the stat mode bits are the same
@@ -505,6 +516,7 @@ write_node (struct node *np)
       diskfs_end_catch_exception ();
       np->dn_stat_dirty = 0;
 
+      /* Leave invoking dino_deref (di) to the caller.  */
       return di;
     }
   else
@@ -674,7 +686,7 @@ diskfs_set_translator (struct node *np, const char *name, unsigned namelen,
   if (err)
     return err;
 
-  di = dino (np->cache_id);
+  di = dino_ref (np->cache_id);
   blkno = di->i_translator;
 
   if (namelen && !blkno)
@@ -687,6 +699,7 @@ diskfs_set_translator (struct node *np, const char *name, unsigned namelen,
 			0, 0, 0);
       if (blkno == 0)
 	{
+	  dino_deref (di);
 	  diskfs_end_catch_exception ();
 	  return ENOSPC;
 	}
@@ -710,15 +723,20 @@ diskfs_set_translator (struct node *np, const char *name, unsigned namelen,
       np->dn_stat.st_mode &= ~S_IPTRANS;
       np->dn_set_ctime = 1;
     }
+  else
+    dino_deref (di);
 
   if (namelen)
     {
+      void *blkptr;
+
       buf[0] = namelen & 0xFF;
       buf[1] = (namelen >> 8) & 0xFF;
       bcopy (name, buf + 2, namelen);
 
-      bcopy (buf, bptr (blkno), block_size);
-      record_global_poke (bptr (blkno));
+      blkptr = disk_cache_block_ref (blkno);
+      memcpy (blkptr, buf, block_size);
+      record_global_poke (blkptr);
 
       np->dn_stat.st_mode |= S_IPTRANS;
       np->dn_set_ctime = 1;
@@ -736,7 +754,8 @@ diskfs_get_translator (struct node *np, char **namep, unsigned *namelen)
   error_t err = 0;
   daddr_t blkno;
   unsigned datalen;
-  const void *transloc;
+  void *transloc;
+  struct ext2_inode *di;
 
   assert (sblock->s_creator_os == EXT2_OS_HURD);
 
@@ -744,9 +763,11 @@ diskfs_get_translator (struct node *np, char **namep, unsigned *namelen)
   if (err)
     return err;
 
-  blkno = (dino (np->cache_id))->i_translator;
+  di = dino_ref (np->cache_id);
+  blkno = di->i_translator;
+  dino_deref (di);
   assert (blkno);
-  transloc = bptr (blkno);
+  transloc = disk_cache_block_ref (blkno);
 
   datalen =
     ((unsigned char *)transloc)[0] + (((unsigned char *)transloc)[1] << 8);
@@ -761,6 +782,7 @@ diskfs_get_translator (struct node *np, char **namep, unsigned *namelen)
 	memcpy (*namep, transloc + 2, datalen);
     }
 
+  disk_cache_block_deref (transloc);
   diskfs_end_catch_exception ();
 
   *namelen = datalen;