| Commit message (Collapse) | Author | Age |
|
|
|
|
|
|
| |
We don't need to conserve stack space too much unlike kernel, also
remove one error condition.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
|
| |
We don't need to conserve stack space too much unlike kernel, also
remove one error condition.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
|
| |
A leftover when fs_uuid was dynamically allocated, the condition was
always true, no functional change.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
| |
The macro dprintf is defined in stdio.h, rename it.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
|
| |
The DEV_STATS and BALANCE_ITEM have been deprecated in favor of more
general items, print the new names in the dump.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
| |
No need to dynamically allocate the thread pointers.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
|
| |
We don't need to conserve stack space too much unlike kernel, also
remove one error condition.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
|
| |
We don't need to conserve stack space too much unlike kernel, also
remove one error condition.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
|
| |
We don't need to conserve stack space too much unlike kernel, also
remove one error condition.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
|
| |
We don't need to conserve stack space too much unlike kernel, also
remove one error condition.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
|
| |
We don't need to conserve stack space too much unlike kernel, also
remove one error condition.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
|
| |
We don't need to conserve stack space too much unlike kernel, also
remove one error condition.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
|
| |
We use only a few values, reada accessed not in performance critical
contexts.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
|
| |
Size of btrfs_path can be reduced by 32 bytes as we don't use the locks
array, down to 112 from 144 bytes.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
| |
We never modify the members, just point them to other strings.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
|
| |
The structure size went from 56 bytes to 48, which could save some bytes
in larger scale.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
| |
find_good_parent
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
| |
Use matching types for buffer, return value and buffer sizes.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
4K is quite small, increase it to 64K. This reduces number of
context switches and calls to read. Kernel sends us about 50K of data
per read, so increasing the buffer further does not make any
improvement.
Example run on tests/cli-tests/004-send-parent-multi-subvol:
4K:
- ~800 context switches
- ~5000 calls to read
64K:
- ~450 context switches
- ~500 calls to read
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
|
| |
Fixed by "btrfs-progs: send: fix handling of multiple snapshots (-p
option)".
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
|
|
|
|
|
| |
There's lots of code that passes btrfs_item_ptr. Add a helper to extract
the ptr at the beginning, ignoring the type. The print functions take
the correct type, but we ignore it in the conversion macros as it does
not provide the type checking anyway.
If the typed variable is used more than once, it's kept in place.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
|
| |
Counting non-zero bytes is wasteful, besides that it would not recognize
uuids containing a 00 value.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
|
| |
Several functions take item but use just the size, we can factor the
item size and use that.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
|
|
| |
Rename to the structure members, add underscore so it's a single word.
Compression is also printed as raw value.
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
| |
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
| |
[ kernel patch 9f07e1d76eaeeddedcf9651395b4a8f870df31f0 ]
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
| |
[ kernel patch 585a3d0d238dfe42909cb58b0d05f644365fed14 ]
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
|
|
|
| |
[ kernel patch 242e2956e4afde7727fffe21adc0a7b198b77f96 ]
Signed-off-by: David Sterba <dsterba@suse.com>
|