linux/fs/ext2, branch v2.6.30.7

ext2: missing unlock in ext2_quota_write()

2009-04-27T14:49:52Z

The inode->i_mutex should be unlocked. Found by smatch (http://repo.or.cz/w/smatch.git). Compile tested. Signed-off-by: Dan Carpenter Signed-off-by: Jan Kara

ext2: fix data corruption for racing writes

2009-04-13T22:04:33Z

If two writers allocating blocks to file race with each other (e.g. because writepages races with ordinary write or two writepages race with each other), ext2_getblock() can be called on the same inode in parallel. Before we are going to allocate new blocks, we have to recheck the block chain we have obtained so far without holding truncate_mutex. Otherwise we could overwrite the indirect block pointer set by the other writer leading to data loss. The below test program by Ying is able to reproduce the data loss with ext2 on in BRD in a few minutes if the machine is under memory pressure: long kMemSize = 50 << 20; int kPageSize = 4096; int main(int argc, char **argv) { int status; int count = 0; int i; char *fname = "/mnt/test.mmap"; char *mem; unlink(fname); int fd = open(fname, O_CREAT | O_EXCL | O_RDWR, 0600); status = ftruncate(fd, kMemSize); mem = mmap(0, kMemSize, PROT_READ | PROT_WRITE, MAP_SHARED, fd, 0); // Fill the memory with 1s. memset(mem, 1, kMemSize); sleep(2); for (i = 0; i < kMemSize; i++) { int byte_good = mem[i] != 0; if (!byte_good && ((i % kPageSize) == 0)) { //printf("%d ", i / kPageSize); count++; } } munmap(mem, kMemSize); close(fd); unlink(fname); if (count > 0) { printf("Running %d bad page\n", count); return 1; } return 0; } Cc: Ying Han Cc: Nick Piggin Signed-off-by: Jan Kara Cc: Mingming Cao Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

New helper - current_umask()

2009-04-01T03:00:26Z

current->fs->umask is what most of fs_struct users are doing. Put that into a helper function. Signed-off-by: Al Viro

ext2: Zero our b_size in ext2_quota_read()

2009-03-26T01:18:38Z

ext2_quota_read() doesn't initialize tmp_bh.b_size before calling ext2_get_block() where we access it. Since it is a local variable it might contain some garbage. Make sure it is filled with reasonable value before passing. Signed-off-by: Manish Katiyar Signed-off-by: Jan Kara

ext2: Use lowercase names of quota functions

2009-03-26T01:18:36Z

Use lowercase names of quota functions instead of old uppercase ones. Signed-off-by: Jan Kara CC: linux-ext4@vger.kernel.org

ext2/xip: refuse to change xip flag during remount with busy inodes

2009-02-11T22:25:36Z

For a reason that I was unable to understand in three months of debugging, mount ext2 -o remount stopped working properly when remounting from regular operation to xip, or the other way around. According to a git bisect search, the problem was introduced with the VM_MIXEDMAP/PTE_SPECIAL rework in the vm: commit 70688e4dd1647f0ceb502bbd5964fa344c5eb411 Author: Nick Piggin Date: Mon Apr 28 02:13:02 2008 -0700 xip: support non-struct page backed memory In the failing scenario, the filesystem is mounted read only via root= kernel parameter on s390x. During remount (in rc.sysinit), the inodes of the bash binary and its libraries are busy and cannot be invalidated (the bash which is running rc.sysinit resides on subject filesystem). Afterwards, another bash process (running ifup-eth) recurses into a subshell, runs dup_mm (via fork). Some of the mappings in this bash process were created from inodes that could not be invalidated during remount. Both parent and child process crash some time later due to inconsistencies in their address spaces. The issue seems to be timing sensitive, various attempts to recreate it have failed. This patch refuses to change the xip flag during remount in case some inodes cannot be invalidated. This patch keeps users from running into that issue. [akpm@linux-foundation.org: cleanup] Signed-off-by: Carsten Otte Cc: Nick Piggin Cc: Jared Hulbert Cc: Martin Schwidefsky Cc: Heiko Carstens Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

ext2: also update the inode on disk when dir is IS_DIRSYNC

2009-01-16T00:39:42Z

We used to just write changed page for IS_DIRSYNC inodes. But we also have to update the directory inode itself just for the case that we've allocated a new block and changed i_size. [akpm@linux-foundation.org: still sync the data page] Signed-off-by: Jan Kara Tested-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

ext2: tighten restrictions on inode flags

2009-01-08T16:31:00Z

At the moment there are few restrictions on which flags may be set on which inodes. Specifically DIRSYNC may only be set on directories and IMMUTABLE and APPEND may not be set on links. Tighten that to disallow TOPDIR being set on non-directories and only NODUMP and NOATIME to be set on non-regular file, non-directories. Introduces a flags masking function which masks flags based on mode and use it during inode creation and when flags are set via the ioctl to facilitate future consistency. Signed-off-by: Duane Griffin Acked-by: Andreas Dilger Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

ext2: don't inherit inappropriate inode flags from parent

2009-01-08T16:31:00Z

At present BTREE/INDEX is the only flag that new ext2 inodes do NOT inherit from their parent. In addition prevent the flags DIRTY, ECOMPR, INDEX, IMAGIC and TOPDIR from being inherited. List inheritable flags explicitly to prevent future flags from accidentally being inherited. This fixes the TOPDIR flag inheritance bug reported at http://bugzilla.kernel.org/show_bug.cgi?id=9866. Signed-off-by: Duane Griffin Acked-by: Andreas Dilger Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

ext2: allocate ->s_blockgroup_lock separately

2009-01-08T16:31:00Z

As spotted by kmemtrace, struct ext2_sb_info is 17024 bytes on 64-bit which makes it a very bad fit for SLAB allocators. The culprit of the wasted memory is ->s_blockgroup_lock which can be as big as 16 KB when NR_CPUS >= 32. To fix that, allocate ->s_blockgroup_lock, which fits nicely in a order 2 page in the worst case, separately. This shinks down struct ext2_sb_info enough to fit a 1 KB slab cache so now we allocate 16 KB + 1 KB instead of 32 KB saving 15 KB of memory. Acked-by: Andreas Dilger Signed-off-by: Pekka Enberg Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds