to B_STRING_TYPE for the lower layers - should probably find a better solution
to handle this.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2166 a95241bf-73f2-0310-859d-f6bbb57e9c96
output turned on.
bfs_open_dir() now accepts again all containers, because bfs_open_index_dir()
is using it, too (too bad when you forgot about your own code).
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2122 a95241bf-73f2-0310-859d-f6bbb57e9c96
translated into B_STRING_TYPE for the key comparison (it was translated
but not saved under the new type).
compareKeys() will now return -1 if it didn't know the type to compare.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2121 a95241bf-73f2-0310-859d-f6bbb57e9c96
Most probably these are used by the kernel to identify changes in a shared
library or something like this.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2088 a95241bf-73f2-0310-859d-f6bbb57e9c96
Fixed a bigger race condition when matching against the name index.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2085 a95241bf-73f2-0310-859d-f6bbb57e9c96
The inode is now locked in bfs_close() (when the indices are updated).
Moved the special cases for INODE_NO_CACHE and INODE_CHKBFS_RUNNING into
bfs_free_cookie() - should probably do that with the index update as well.
Updated some comments.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2084 a95241bf-73f2-0310-859d-f6bbb57e9c96
to the provided buffer.
Inode::Name() no longer locks the small_data region anymore.
Added ASSERTs that the small_data region is locked for the methods requiring
that.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2083 a95241bf-73f2-0310-859d-f6bbb57e9c96
bfs_read_vnode() will now wait for half a second at maximum until it returns
the B_BUSY error (so that live queries will like it better).
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2058 a95241bf-73f2-0310-859d-f6bbb57e9c96
instead calls Inode::Free().
If INODE_DONT_FREE_SPACE was set, bfs_remove_vnode() forgot to delete the
inode before returning B_OK.
bfs_read_vnode() will now return the exact error Inode::InitCheck() returned.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2056 a95241bf-73f2-0310-859d-f6bbb57e9c96
to the InodeAllocator class (initialization of the inode, tree creation for
containers).
The InodeAllocator class now frees all the space the inode occupies, including
the B+tree if it's a container (by calling the new Inode::Free() method).
Inode::Free() implementation (logic removed from bfs_remove_vnode()).
Now secures the inode from being loaded by setting the INODE_NOT_READY flag
(completely handled by the InodeAllocator class).
The inode is now removed from its parent if something went wrong after it
had been added.
Utilizes the new BPlusTree::Remove() method where appropriate.
Inode::GetAttribute() now checks if the inode opened is really an attribute.
Inode::InitCheck() now fails with B_BUSY if the INODE_NOT_READY flag is set.
InodeAllocator::Keep() now writes back the inode.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2055 a95241bf-73f2-0310-859d-f6bbb57e9c96
also the numBlocks parameter is ASSERTed - when compiled with DEBUG turned
off, BFS will enter the kernel debugger in this case (through the use of
the DIE() macro).
Same for AllocationBlock::Free() (but the 0xffff mode was already removed
there).
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2053 a95241bf-73f2-0310-859d-f6bbb57e9c96
allocation group as the inode, if possible.
Moved the group used blocks info maintaining code into AllocationGroup::Allocate()
and AllocationGroup::Free() (instead of letting the BlockAllocator class do
the work).
Removed the buggy and useless extra 0xffff mode for AllocationBlock::Allocate()
and AllocationBlock::Free(), both methods now ASSERT their parameter range.
Changed some comments, style cleanups.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2052 a95241bf-73f2-0310-859d-f6bbb57e9c96
of the stream (so the "chkbfs" functionality is complete now).
Fixed wrong block number report in BlockAllocator::CheckBlockRun(), blocks are
now printed out in absolute numbers (not allocation_group relative anymore).
Use Inode::IsContainer() where necessary instead of Inode::IsDirectory().
Fixed a big and nasty bug in AllocationBlock::SetTo(): the number of bits in
a block could be calculated wrong for a partial allocation group which need
more than one block in the bitmap - hopefully, that was the last remaining
big bug in the block allocator.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2032 a95241bf-73f2-0310-859d-f6bbb57e9c96
inode locked if the _inode parameter is passed.
Makes some more use of the new type identificators.
Symlinks are no longer added to the "size" index (compatibility with BFS, and it also
makes much more sense).
Now sets S_STR_INDEX for directories, if no index type was set (again, compatibility
issue with BFS).
Fixed a bug in the Inode::GrowStream() method in the double indirect region.
Some style updates.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2030 a95241bf-73f2-0310-859d-f6bbb57e9c96
Renamed Inode::IsDirectory() to Inode::IsContainer() (since it is also true
for index/attribute directories).
Introduced a new Inode::IsDirectory() that only checks for real and standard
directories. Let's hope I've fixed more bugs with that than introduced new
ones...
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2029 a95241bf-73f2-0310-859d-f6bbb57e9c96
The only functional change is that bfs_rename() now checks for '/' in the
target name (the VFS should do those things, but you never know as long
you haven't written it yourself :).
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2028 a95241bf-73f2-0310-859d-f6bbb57e9c96
via the private ioctls defined in bfs_control.h.
Detects all kinds of errors, and can correct most of them. It doesn't fix
any errors in the b+trees (like the ones created by an earlier version
of the BPlusTree class ;-), and it also can't fix the "blocks are already set"
error, which occurs if more than one file claims a particular block (or
range of blocks) for itself.
The double indirect data range is not yet supported, so use with care, or
run the original chkbfs afterwards to fix any errors because of that.
Might not yet analyze 100% correct (shows way to many errors for my tastes...).
A "chkbfs" command will be made available shortly.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@1994 a95241bf-73f2-0310-859d-f6bbb57e9c96
and atomic_test_and_set() for userland testing purposes.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@1987 a95241bf-73f2-0310-859d-f6bbb57e9c96
the bug where a block_run could get a zero length after merging it...
Updated a comment in Inode::Remove().
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@1986 a95241bf-73f2-0310-859d-f6bbb57e9c96
block_runs can be merged together.
Added a constant for the maximum length of a block_run.
Added a flag constant to the inode that prevents freeing up its space
after it has been removed - this is only useful for the "chkbfs" stuff.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@1985 a95241bf-73f2-0310-859d-f6bbb57e9c96
Cache::Read() != B_OK failed if the underlying class used that function.
We can now boot R5 with VM turned on without any problems (what a stupid bug).
As a side effect, you can now also mount volumes on your BFS disk, which was
also broken because of this bug.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@1930 a95241bf-73f2-0310-859d-f6bbb57e9c96
