David Sterba [Tue, 12 Sep 2017 16:17:51 +0000 (18:17 +0200)]
btrfs-progs: tests: remove temporary loopdev files
Remove files for testing loop devices when using the helper.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Tue, 12 Sep 2017 13:23:05 +0000 (15:23 +0200)]
btrfs-progs: convert: use correct string printing for errcode_t
Signed-off-by: David Sterba <dsterba@suse.com>
Misono, Tomohiro [Tue, 12 Sep 2017 01:04:10 +0000 (10:04 +0900)]
btrfs-progs: cleanup whitespaces of btrfs-completion
Clean up spaces and use only tabs. No functional change.
Signed-off-by: Tomohiro Misono <misono.tomohiro@jp.fujitsu.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Nick Terrell [Mon, 11 Sep 2017 19:27:52 +0000 (21:27 +0200)]
btrfs-progs: tests: add testing image for zstd for btrfs-restore
Adds tests for the new features based on a prebuilt btrfs image with a
zstd compressed file.
Split from the previous patch.
Signed-off-by: Nick Terrell <terrelln@fb.com>
[ with some cleanups ]
Signed-off-by: David Sterba <dsterba@suse.com>
Nick Terrell [Fri, 8 Sep 2017 20:29:23 +0000 (13:29 -0700)]
btrfs-progs: Add zstd support
Adds zstd support to the btrfs program. An optional dependency on libzstd
>= 1.0.0 is added. Autoconf accepts `--enable-zstd' or `--disable-zstd' and
defaults to detecting if libzstd is present using `pkg-config'.
The patch is also available in my fork of btrfs-progs [1], which passes
Travis-CI with the new tests. The prebuilt binary is available there.
I haven't updated Android.mk.
[1] https://github.com/terrelln/btrfs-progs/tree/devel
Signed-off-by: Nick Terrell <terrelln@fb.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Misono, Tomohiro [Mon, 11 Sep 2017 05:22:43 +0000 (14:22 +0900)]
btrfs-progs: update btrfs-completion
This patch updates btrfs-completion:
- add "filesystem du" and "rescure zero-log"
- restrict _btrfs_mnts to show btrfs type only
- add more completion in last case statements
(This file contains both spaces/tabs and may need cleanup.)
Signed-off-by: Tomohiro Misono <misono.tomohiro@jp.fujitsu.com>
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Mon, 11 Sep 2017 18:56:18 +0000 (20:56 +0200)]
btrfs-progs: build: sync Makefile to Android.mk
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Mon, 11 Sep 2017 15:29:03 +0000 (17:29 +0200)]
btrfs-progs: tests: fix run_mustfail in cli-tests/007-check-force
The sanity check in run_mustfail does not work as expected and allowed
the command to continue without the missing parameter
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=196893
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Mon, 11 Sep 2017 15:17:26 +0000 (17:17 +0200)]
btrfs-progs: README: update links, enhance contribution section
Signed-off-by: David Sterba <dsterba@suse.com>
Piotr Pawlow [Fri, 11 Aug 2017 13:47:35 +0000 (15:47 +0200)]
btrfs-progs: image: use CRC32C reversing instead of brute force to find collisions
Author: Piotr Pawlow <pp@siedziba.pl>
Signed-off-by: David Sterba <dsterba@suse.com>
Piotr Pawlow [Fri, 11 Aug 2017 13:31:56 +0000 (15:31 +0200)]
btrfs-progs: image: add a function to check if generated filename suffix is valid
Author: Piotr Pawlow <pp@siedziba.pl>
Signed-off-by: David Sterba <dsterba@suse.com>
Piotr Pawlow [Fri, 11 Aug 2017 13:06:17 +0000 (15:06 +0200)]
btrfs-progs: image: add a function to calculate CRC32C collisions
The function uses the reverse CRC32C table to quickly calculate a
4-byte suffix, that when added to the original data will make it
match desired checksum.
Author: Piotr Pawlow <pp@siedziba.pl>
[ minor adjustments ]
Signed-off-by: David Sterba <dsterba@suse.com>
Piotr Pawlow [Fri, 11 Aug 2017 12:36:32 +0000 (14:36 +0200)]
btrfs-progs: image: add reverse CRC32C table
The table will be used to speed up calculations of CRC32C collisions.
Author: Piotr Pawlow <pp@siedziba.pl>
Signed-off-by: David Sterba <dsterba@suse.com>
Piotr Pawlow [Fri, 11 Aug 2017 11:46:32 +0000 (13:46 +0200)]
btrfs-progs: image: move core find_collision code to a separate function
Author: Piotr Pawlow <pp@siedziba.pl>
Signed-off-by: David Sterba <dsterba@suse.com>
Piotr Pawlow [Fri, 11 Aug 2017 15:11:44 +0000 (17:11 +0200)]
btrfs-progs: image: fix non-printable characters in generated file names
Function find_collision sometimes generated file names with non-printable
DEL characters (code 127), for example file name "|5gp!" would be changed
to "U'2<DEL>y" when using "crc-collisions" sanitize mode.
Author: Piotr Pawlow <pp@siedziba.pl>
Signed-off-by: David Sterba <dsterba@suse.com>
Uli Heller [Sat, 9 Sep 2017 06:02:51 +0000 (08:02 +0200)]
btrfs-progs: Removed missing header file, fixes compilation error
(From pull request #61.)
Author: Uli Heller <uli.heller@daemons-point.com>
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Fri, 8 Sep 2017 14:20:05 +0000 (16:20 +0200)]
Btrfs progs v4.13
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Fri, 28 Jul 2017 13:29:46 +0000 (15:29 +0200)]
btrfs-progs: update CHANGES for v4.13
Signed-off-by: David Sterba <dsterba@suse.com>
Misono, Tomohiro [Thu, 7 Sep 2017 01:48:27 +0000 (10:48 +0900)]
btrfs-progs: test: add new test for inspect-internal rootid
This new test checks inspect-internal rootid
- handle path to subvolume/directory/file as an argument
- get different id for each subvolume
- get the expected id for each file/directory (i.e. the same as
containing subvolume)
Signed-off-by: Tomohiro Misono <misono.tomohiro@jp.fujitsu.com>
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Thu, 7 Sep 2017 15:29:51 +0000 (17:29 +0200)]
btrfs-progs: tests: check for kernel support for reiserfs
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Thu, 7 Sep 2017 12:44:45 +0000 (14:44 +0200)]
btrfs-progs: tests: enhance post-rollback fsck tests
The post-rollback helper still assumes just extN, we need an extra
argument that'll get passed to fsck. Change all callsites at once so the
tests do not fail temporarily.
Signed-off-by: David Sterba <dsterba@suse.com>
Zhang Yu [Thu, 24 Aug 2017 10:38:41 +0000 (18:38 +0800)]
Btrfs-progs: print-tree: check num_stripes in print_chunk
[TEST/fuzz] case: 004-simple-dump-tree
Since the wrong key(DATA_RELOC_TREE CHUNK_ITEM 0) in root tree,
error calling print_chunk(), resulting in num_stripes == 0.
ERROR:
[TEST/fuzz] 004-simple-dump-tree
ctree.h:317: btrfs_chunk_item_size: BUG_ON `num_stripes == 0`
triggered, value 1
failed (ignored, ret=134): /myproject/btrfs-progs/btrfs
inspect-internal dump-tree
/myproject/btrfs-progs/tests/fuzz-tests/images/
bko-155201-wrong-chunk-item-in-root-tree.raw.restored
test failed for case 004-simple-dump-tree
Makefile:288: recipe for target 'test-fuzz' failed
make: *** [test-fuzz] Error 1
So, check on num_stripes in print_chunk
Signed-off-by: Zhang Yu <zhangyu-fnst@cn.fujitsu.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Misono, Tomohiro [Tue, 5 Sep 2017 05:50:03 +0000 (14:50 +0900)]
btrfs-progs: test: fix name generation not to contain trailing spaces
First patch causes test-convert fails. This is because
generate_dataset() creates a name containing trailing spaces for
"slow_symlink" type, and cause getfacl error in convert_test_perm().
(This is not noticed since original run_check_stdout() throws away the
error.)
Fix this by use space for delimiter for cut.
Signed-off-by: Tomohiro Misono <misono.tomohiro@jp.fujitsu.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Misono, Tomohiro [Tue, 5 Sep 2017 05:49:02 +0000 (14:49 +0900)]
btrfs-progs: test: fix run_check_stdout() call _fail()
run_check_stdout() uses "... | tee ... || _fail". However, since tee
won't fail, _fail() is not called even if first command fails.
Fix this by checking PIPESTATUS in the end.
Signed-off-by: Tomohiro Misono <misono.tomohiro@jp.fujitsu.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Qu Wenruo [Tue, 5 Sep 2017 01:24:15 +0000 (10:24 +0900)]
btrfs-progs: tests: Add test case for mkfs --rootdir parameter
Add test case which checks if -r|--rootdir mkfs option can handle
symlink/char/block/fifo files.
Signed-off-by: Qu Wenruo <quwenruo.btrfs@gmx.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Qu Wenruo [Mon, 4 Sep 2017 03:43:19 +0000 (12:43 +0900)]
btrfs-progs: mkfs: Fix wrong file type for dir items and indexes when specifying root directory
[Bug]
If using mkfs.btrfs with "-r" parameter and specified directory has
fifo/socket/char/block special file, then created filesystem can't pass
fsck:
------
checking fs roots
unresolved ref dir 241158 index 3 namelen 9 name S.dirmngr filetype 0 errors 80, filetype mismatch
ERROR: errors found in fs roots
------
[Reason]
Btrfs dir items/indexes records inode type, while "-r" only handles
directories, regular files and symlink, it makes such special files type
to be regular file and caused the problem.
[Fix]
Add missing types for add_directory_items(), so that result of
"mkfs.btrfs -r" can pass mkfs.
Signed-off-by: Qu Wenruo <quwenruo.btrfs@gmx.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Gu Jinxiang [Fri, 25 Aug 2017 07:23:36 +0000 (15:23 +0800)]
btrfs-progs: mkfs: Move the tree root creation to own function
make_btrfs is too long to understand, make creatation of root tree
in a function.
Some of the tree roots are now created in a loop, where the code is just
copypasted. We now make use of the reference_root_table to translate
block index to root objectid.
Signed-off-by: Gu Jinxiang <gujx@cn.fujitsu.com>
[ updated changelog ]
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Mon, 4 Sep 2017 16:33:48 +0000 (18:33 +0200)]
btrfs-progs: tests: missing device and slack space report
Verify that a missing device will not result in reporting a negative
value interpreted as 16EiB.
Signed-off-by: David Sterba <dsterba@suse.com>
Patrik Lundquist [Thu, 31 Aug 2017 11:00:24 +0000 (13:00 +0200)]
btrfs-progs: device usage: don't calculate slack on missing device
Print Device slack: 0.00B
instead of Device slack: 16.00EiB
Signed-off-by: Patrik Lundquist <patrik.lundquist@gmail.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Misono, Tomohiro [Mon, 4 Sep 2017 05:05:34 +0000 (14:05 +0900)]
btrfs-progs: inspect rootid: Allow a file to be specified
Since cmd_inspect_rootid() calls btrfs_open_dir(), it rejects a file to
be specified. But as the document says, a file should be supported.
This patch introduces btrfs_open_file_or_dir(), which is a counterpart
of btrfs_open_dir(), to safely check and open btrfs file or directory.
The original btrfs_open_dir() content is moved to btrfs_open() and shared
by both function.
Signed-off-by: Tomohiro Misono <misono.tomohiro@jp.fujitsu.com>
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Fri, 1 Sep 2017 18:18:20 +0000 (20:18 +0200)]
btrfs-progs: tests: convert misc/011-delete-missing-device to loopdevs
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Fri, 1 Sep 2017 18:14:55 +0000 (20:14 +0200)]
btrfs-progs: tests: convert misc/006-image-on-missing-device to loopdevs
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Fri, 1 Sep 2017 18:14:03 +0000 (20:14 +0200)]
btrfs-progs: tests: move loopdev helpers out of the testcase to common
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Fri, 1 Sep 2017 17:32:16 +0000 (19:32 +0200)]
btrfs-progs: tests: cleanup loop device helpers
Make the loop device helpers a bit more generic before moving them to
the common helpers.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Thu, 31 Aug 2017 15:07:00 +0000 (17:07 +0200)]
btrfs-progs: print-tree: factor out extent_csum dump
Factor out code to own helper and tweak the format so it matches the
rest.
Signed-off-by: David Sterba <dsterba@suse.com>
Josef Bacik [Fri, 25 Aug 2017 15:11:46 +0000 (11:11 -0400)]
btrfs-progs: print the csum length in debug-tree
While looking at a log of a corrupted fs I needed to verify we were
missing csums for a given range. Make this easier by printing out the
range of bytes a csum item covers.
Signed-off-by: Josef Bacik <jbacik@fb.com>
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Fri, 1 Sep 2017 14:45:34 +0000 (16:45 +0200)]
btrfs-progs: tests: add testcase for 'fi du' and empty subvol
Signed-off-by: David Sterba <dsterba@suse.com>
Goffredo Baroncelli [Fri, 18 Aug 2017 07:04:21 +0000 (09:04 +0200)]
btrfs-progs: fi du: don't call lookup_path_rootid for BTRFS_EMPTY_SUBVOL_DIR_OBJECTID
When ino is BTRFS_EMPTY_SUBVOL_DIR_OBJECTID, the item is not referred to
any file-tree. So lookup_path_rootid() doesn't return any meaningful
value.
As was reported, this can be triggered by
$ btrfs sub create test1
$ btrfs sub create test1/test2
$ btrfs sub snap test1 test1.snap
$ btrfs fi du -s test1
Total Exclusive Set shared Filename
0.00B 0.00B 0.00B test1
$ btrfs fi du -s test1.snap
Total Exclusive Set shared Filename
ERROR: cannot check space of 'test1.snap': Inappropriate ioctl for device
Signed-off-by: Goffredo Baroncelli <kreijack@inwind.it>
Signed-off-by: David Sterba <dsterba@suse.com>
Goffredo Baroncelli [Fri, 18 Aug 2017 07:04:20 +0000 (09:04 +0200)]
btrfs-progs: reset the ret value when ignoring an error from du_add_file
In du_walk_dir(), when du_add_file() returns an error it is usually
ignored. However if the error is returned querying the last item, the
error is returned to the caller.
Signed-off-by: Goffredo Baroncelli <kreijack@inwind.it>
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Thu, 31 Aug 2017 16:49:38 +0000 (18:49 +0200)]
btrfs-progs: tests: add test for check --force
Basic test of the --force functionality, on an empty filesystem.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Thu, 31 Aug 2017 16:34:37 +0000 (18:34 +0200)]
btrfs-progs: check: add option to skip mount checks
Sometimes it's needed to do a check on a mounted filesystem. This should
work fine on a quiescent filesystem or a read-only mount. Changes on the
block device done by kernel might confuse the userspace checker and it
might crash when it reads some stale data.
Repair without mount checks is not supported right now.
Signed-off-by: David Sterba <dsterba@suse.cz>
David Sterba [Thu, 31 Aug 2017 15:07:00 +0000 (17:07 +0200)]
btrfs-progs: print-tree: factor out temporary_item dump
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Thu, 31 Aug 2017 15:07:00 +0000 (17:07 +0200)]
btrfs-progs: print-tree: factor out persistent_item dump
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Thu, 31 Aug 2017 15:07:00 +0000 (17:07 +0200)]
btrfs-progs: print-tree: factor out qgroup_limit dump
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Thu, 31 Aug 2017 15:07:00 +0000 (17:07 +0200)]
btrfs-progs: print-tree: factor out qgroup_info dump
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Thu, 31 Aug 2017 15:07:00 +0000 (17:07 +0200)]
btrfs-progs: print-tree: factor out qgroup_status dump
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Thu, 31 Aug 2017 15:07:00 +0000 (17:07 +0200)]
btrfs-progs: print-tree: factor out dev_extent dump
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Thu, 31 Aug 2017 15:07:00 +0000 (17:07 +0200)]
btrfs-progs: print-tree: factor out free_space_info dump
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Thu, 31 Aug 2017 15:07:00 +0000 (17:07 +0200)]
btrfs-progs: print-tree: factor out shared_data_ref dump
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Thu, 31 Aug 2017 15:07:00 +0000 (17:07 +0200)]
btrfs-progs: print-tree: factor out extent_data_ref dump
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Thu, 31 Aug 2017 15:07:00 +0000 (17:07 +0200)]
btrfs-progs: print-tree: factor out block_group_item dump
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Thu, 31 Aug 2017 14:57:47 +0000 (16:57 +0200)]
btrfs-progs: print-tree: don't print the item type for all items
In some cases it's clear from the context which item is being printed,
so we can remove them. If the item has no data, some description is
still desired (eg. orphan or various backrefs).
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Thu, 31 Aug 2017 14:34:15 +0000 (16:34 +0200)]
btrfs-progs: print-tree: rename item callbacks to match the key name
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Wed, 30 Aug 2017 17:47:50 +0000 (19:47 +0200)]
btrfs-progs: check: move fs roots check mode switch to a helper
Simplify main a bit.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Wed, 30 Aug 2017 17:52:44 +0000 (19:52 +0200)]
btrfs-progs: check: cleanup use of fs_info in check_fs_roots
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Wed, 30 Aug 2017 17:50:31 +0000 (19:50 +0200)]
btrfs-progs: check: pass fs_info to check_fs_roots
The root pointer is not used anyway, will be cleaned up next.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Wed, 30 Aug 2017 17:47:50 +0000 (19:47 +0200)]
btrfs-progs: check: move chunk and extent check mode switch to a helper
Simplify main a bit.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Wed, 30 Aug 2017 17:38:02 +0000 (19:38 +0200)]
btrfs-progs: check: pass fs_info to check_chunks_and_extents_v2
The root pointer is set to fs_root as was originally.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Wed, 30 Aug 2017 17:41:15 +0000 (19:41 +0200)]
btrfs-progs: check: use local fs_info in check_chunks_and_extents
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Wed, 30 Aug 2017 17:38:02 +0000 (19:38 +0200)]
btrfs-progs: check: pass fs_info to check_chunks_and_extents
The root pointer is set to fs_root as was originally.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Wed, 30 Aug 2017 17:25:11 +0000 (19:25 +0200)]
btrfs-progs: check: verify critical root pointers before use
The pointers to critical roots must be valid before we start using them,
eg. as the space clearing code.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Wed, 30 Aug 2017 16:57:03 +0000 (18:57 +0200)]
btrfs-progs: check: move more of space cache clearing to a helper
Move the code out of main.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Wed, 30 Aug 2017 14:56:51 +0000 (16:56 +0200)]
btrfs-progs: check: remove old mixed backref code
A code added in 2009 (
95d3f20b51e9b) for a very short-lived change in
the format is no concern to us nowadays.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Wed, 30 Aug 2017 14:56:23 +0000 (16:56 +0200)]
btrfs-progs: tests: add more sanitizer message patterns to log scanner
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Tue, 29 Aug 2017 17:42:10 +0000 (19:42 +0200)]
btrfs-progs: convert: don't print message when fs is not recognized, ext2
Bad magic error means it's not the desired filesystem so the error
message is just noise.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Tue, 29 Aug 2017 16:41:30 +0000 (18:41 +0200)]
btrfs-progs: tests: limit size of log dump from conver tests
The convert tests generate lots of log material, travis CI has limit 4MB
so we don't see anything useful when a late test fails.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Tue, 29 Aug 2017 16:19:03 +0000 (18:19 +0200)]
btrfs-progs: handle failed search in find_search_start better
The warning can pop up frequently on a fuzzed image, the message seems
to be enough. Add a more fitting error code too.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Tue, 29 Aug 2017 15:53:39 +0000 (17:53 +0200)]
btrfs-progs: add more error handling to btrfs_free_block_group
As btrfs_update_block_group fails when the block group is not found in
cache, we can exit btrfs_free_block_group, not much to rollback. The
caller will also exit in turn.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Tue, 29 Aug 2017 15:42:28 +0000 (17:42 +0200)]
btrfs-progs: tests: fix TEST_LOG=dump in convert tests
Test failure in convert tests with log dump does not happen because
_fail is called before that and exits. Other test types are ok.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Tue, 29 Aug 2017 15:22:39 +0000 (17:22 +0200)]
btrfs-progs: handle transaction start failure in close_ctree
Closing the fs will try to commit a pending transaction, but may fail to
do so if the filesystem state is not well defined. This will eg. fail
for some fuzz tests. The data structures are freed but no furhter
attempt to commit is made.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Tue, 29 Aug 2017 15:07:15 +0000 (17:07 +0200)]
btrfs-progs: tests: update README
Wording, runtime dependencies.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Tue, 29 Aug 2017 14:36:15 +0000 (16:36 +0200)]
btrfs-progs: tests: install missing CI packages
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Tue, 29 Aug 2017 14:35:01 +0000 (16:35 +0200)]
btrfs-progs: tests: add more checks for tools used in convert tests
Tools that may not be present in common installations should be checked
in the tests.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Tue, 29 Aug 2017 14:16:44 +0000 (16:16 +0200)]
btrfs-progs: tests: build preparation phases out of script section
The logs of before_install are folded and we don't need to see the
details, unlike the test logs.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Tue, 29 Aug 2017 12:55:37 +0000 (14:55 +0200)]
btrfs-progs: don't start or commit after transaction abort
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Mon, 28 Aug 2017 15:44:41 +0000 (17:44 +0200)]
btrfs-progs: start framework for transaction abort
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Mon, 28 Aug 2017 15:43:10 +0000 (17:43 +0200)]
btrfs-progs: store pointer to fs_info in transaction handle
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Mon, 28 Aug 2017 15:39:26 +0000 (17:39 +0200)]
btrfs-progs: move transaction code out of disk-io
Temporarily export the low-level helpers.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Mon, 28 Aug 2017 15:27:08 +0000 (17:27 +0200)]
btrfs-progs: move transaction implementation out of header
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Mon, 28 Aug 2017 15:21:05 +0000 (17:21 +0200)]
btrfs-progs: use kzalloc to allocate transaction handle
Use the kernel-style allocation helpers and remove redundant zeroing.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Mon, 28 Aug 2017 15:16:51 +0000 (17:16 +0200)]
btrfs-progs: switch fs_info::system_allocs to bit
The value really is 0 or 1, the test in btrfs_reserve_extent can be
simplified.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Mon, 28 Aug 2017 14:54:12 +0000 (16:54 +0200)]
btrfs-progs: return errors from btrfs_start_transaction
Now that all callers will catch errors, we can return them from the
function.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Mon, 28 Aug 2017 14:48:16 +0000 (16:48 +0200)]
btrfs-progs: add crude error handling when transaction start fails
Currently transaction bugs out insided btrfs_start_transaction in case
of error, we want to lift the error handling to the callers. This patch
adds the BUG_ON anywhere it's been missing so far. This is not the best
way of course. Transforming BUG_ON to a proper error handling highly
depends on the caller and should be dealt with case by case.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Mon, 28 Aug 2017 12:28:53 +0000 (14:28 +0200)]
btrfs-progs: tests: pull and build reiserfs lib inside CI
The functionality needed to support reiserfs in convert has been added
recently and is not yet available in the CI images, so we'll download
and build it there.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Mon, 28 Aug 2017 11:54:02 +0000 (13:54 +0200)]
btrfs-progs: tests: add build checks for sanitization features
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Mon, 28 Aug 2017 11:38:32 +0000 (13:38 +0200)]
btrfs-progs: build: fix asan and ubsan build
Probably with a new gcc (7.1.1) I started to see asan/ubsan link failures.
Fixed by explicitly linking the libraries.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Mon, 28 Aug 2017 11:33:19 +0000 (13:33 +0200)]
btrfs-progs: build: fix building libbtrfs.so with tsan
Except libbtrfs.so object, all other tools compile fine. The error is:
ld: send-stream.o: relocation R_X86_64_PC32 against symbol
`stderr@@GLIBC_2.2.5' can not be used when making a shared object;
recompile with -fPIC
Compiling with -fPIC fixes the problem.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Fri, 25 Aug 2017 18:09:34 +0000 (20:09 +0200)]
btrfs-progs: build: properly pass the LDFLAGS during tsan build
The variable LD_FLAGS does not exist and the flags are not used, we need
to use the newly added internal debugging linker flags.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Fri, 25 Aug 2017 18:04:48 +0000 (20:04 +0200)]
btrfs-progs: build: add internal LDFLAGS for the D= features
We might want to pass additional linker flags for various sanitization
features.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Fri, 25 Aug 2017 17:21:01 +0000 (19:21 +0200)]
btrfs-progs: tests: add more fuzzed images from bugzilla
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Fri, 25 Aug 2017 17:08:37 +0000 (19:08 +0200)]
btrfs-progs: handle blockgroup errors when opening filesystem
The call to btrfs_read_block_groups could loop if the metadata are
damaged (reported eg. for an unaligned block), due to lack of error
handling. We have to check for restored images or currently created
filesystems, that do not contain the blockgroups.
Can be reproduced by fuzzed image bko-155551.
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=155551
Reported-by: Lukas Lueg <lukas.lueg@gmail.com>
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Fri, 25 Aug 2017 16:07:15 +0000 (18:07 +0200)]
btrfs-progs: drop blocksize argument from readahead_tree_block
Tree blocks are always nodesize. As readahead is only an optimization,
exact size is not required and is only advisory.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Fri, 25 Aug 2017 15:44:22 +0000 (17:44 +0200)]
btrfs-progs: drop blocksize argument from btrfs_find_create_tree_block
Metadata blocks are always nodesize. When reading the
superblock::sys_array, the actual size of data is fixed to 4k and
smaller than nodesize, but otherwise everything works as before.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Fri, 25 Aug 2017 15:30:34 +0000 (17:30 +0200)]
btrfs-progs: drop redundant check of blocksize in read_tree_block
The tree blocks are supposed to be always of nodesize. Before the
parameter has been dropped, it was unlikely but possible to pass a
misaligned value.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Fri, 25 Aug 2017 15:21:03 +0000 (17:21 +0200)]
btrfs-progs: check: cleanup use of level_size
Nodesize is same for all levels, besides it's been only set and not
used, in root_item_record.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Fri, 25 Aug 2017 14:54:16 +0000 (16:54 +0200)]
btrfs-progs: drop blocksize from read_tree_block
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Fri, 25 Aug 2017 14:20:16 +0000 (16:20 +0200)]
btrfs-progs: drop local blocksize variables if they're nodesize
Prep work so we can drop the blocksize argument from several functions.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Fri, 25 Aug 2017 12:26:51 +0000 (14:26 +0200)]
btrfs-progs: build: fix PIE build
Patch from Marcus Meissner <meissner@suse.com>. The CFLAGS are passed to
the linker and mix up the compilation and linker flags for PIE support.
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Thu, 24 Aug 2017 13:48:48 +0000 (15:48 +0200)]
btrfs-progs: convert: move reiserfs struct definitions to header
Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Tue, 22 Aug 2017 18:43:49 +0000 (20:43 +0200)]
btrfs-progs: convert: move and rename dev_t helpers to common file
Signed-off-by: David Sterba <dsterba@suse.com>