[BK] [2.4] Reiserfs changeset 3 out of 4, please apply.

From: Hans Reiser (reiser@namesys.com)
Date: Tue May 07 2002 - 10:05:44 EST


Hello!

  You can get this changeset from bk://thebsh.namesys.com/bk/reiser3-linux-2.4

  This changeset fixes 2 incorrect padding problems and a race with knfsd,
  where file might be attemted to be deleted twice.

Diffstat:
 inode.c | 17 ++++++++++++-----
 namei.c | 2 +-
 tail_conversion.c | 2 +-
 3 files changed, 14 insertions(+), 7 deletions(-)

Plain text patch
# This is a BitKeeper generated patch for the following project:
# Project Name: Linux kernel tree
# This patch format is intended for GNU patch command version 2.5 or higher.
# This patch includes the following deltas:
# ChangeSet 1.383.2.39 -> 1.383.2.40
# fs/reiserfs/namei.c 1.20 -> 1.21
# fs/reiserfs/inode.c 1.33 -> 1.34
# fs/reiserfs/tail_conversion.c 1.11 -> 1.12
#
# The following is the BitKeeper ChangeSet Log
# --------------------------------------------
# 02/04/30 green@angband.namesys.com 1.383.2.40
# inode.c:
# implemented reiserfs_make_bad_inode() function that
# zeroes inode key therefore avoiding races with knfsd
# and other stuff that might access files bypassing directory lookups
# tail_conversion.c:
# old format DIRECT items were incorrectly aligned on new format filesystems
# namei.c:
# old format DIR entries were incorrectly aligned on new format filesystems
# --------------------------------------------
#
diff -Nru a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
--- a/fs/reiserfs/inode.c Tue May 7 17:54:21 2002
+++ b/fs/reiserfs/inode.c Tue May 7 17:54:21 2002
@@ -1128,8 +1128,15 @@
     return;
 }
 
+/* We need to clear inode key in private part of inode to avoid races between
+ blocking iput, knfsd and file deletion with creating of safelinks.*/
+static void reiserfs_make_bad_inode(struct inode *inode) {
+ memset(INODE_PKEY(inode), 0, KEY_SIZE);
+ make_bad_inode(inode);
+}
+
 void reiserfs_read_inode(struct inode *inode) {
- make_bad_inode(inode) ;
+ reiserfs_make_bad_inode(inode) ;
 }
 
 
@@ -1144,7 +1151,7 @@
     int retval;
 
     if (!p) {
- make_bad_inode(inode) ;
+ reiserfs_make_bad_inode(inode) ;
         return;
     }
 
@@ -1164,13 +1171,13 @@
         reiserfs_warning ("vs-13070: reiserfs_read_inode2: "
                     "i/o failure occurred trying to find stat data of %K\n",
                     &key);
- make_bad_inode(inode) ;
+ reiserfs_make_bad_inode(inode) ;
         return;
     }
     if (retval != ITEM_FOUND) {
         /* a stale NFS handle can trigger this without it being an error */
         pathrelse (&path_to_sd);
- make_bad_inode(inode) ;
+ reiserfs_make_bad_inode(inode) ;
         inode->i_nlink = 0;
         return;
     }
@@ -1197,7 +1204,7 @@
                               "dead inode read from disk %K. "
                               "This is likely to be race with knfsd. Ignore\n",
                               &key );
- make_bad_inode( inode );
+ reiserfs_make_bad_inode( inode );
     }
 
     reiserfs_check_path(&path_to_sd) ; /* init inode should be relsing */
diff -Nru a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c
--- a/fs/reiserfs/namei.c Tue May 7 17:54:21 2002
+++ b/fs/reiserfs/namei.c Tue May 7 17:54:21 2002
@@ -388,7 +388,7 @@
     } else
         buffer = small_buf;
 
- paste_size = (old_format_only (dir->i_sb)) ? (DEH_SIZE + namelen) : buflen;
+ paste_size = (get_inode_sd_version (dir) == STAT_DATA_V1) ? (DEH_SIZE + namelen) : buflen;
 
     /* fill buffer : directory entry head, name[, dir objectid | , stat data | ,stat data, dir objectid ] */
     deh = (struct reiserfs_de_head *)buffer;
diff -Nru a/fs/reiserfs/tail_conversion.c b/fs/reiserfs/tail_conversion.c
--- a/fs/reiserfs/tail_conversion.c Tue May 7 17:54:21 2002
+++ b/fs/reiserfs/tail_conversion.c Tue May 7 17:54:21 2002
@@ -213,7 +213,7 @@
     copy_item_head (&s_ih, PATH_PITEM_HEAD(p_s_path));
 
     tail_len = (n_new_file_size & (n_block_size - 1));
- if (!old_format_only (p_s_sb))
+ if (get_inode_sd_version (p_s_inode) == STAT_DATA_V2)
         round_tail_len = ROUND_UP (tail_len);
     else
         round_tail_len = tail_len;
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Tue May 07 2002 - 22:00:30 EST