similar to: [PATCH] Btrfs: fix crash in log replay with qgroups enabled

When making btrfs filesystem. we firstly write root leaf to specified filed, and then we recow the root. If we don''t recow, some trees are not in the correct block group. Steps to reproduce: dd if=/dev/zero of=test.img bs=1M count=100 mkfs.btrfs -f test.img btrfs-debug-tree test.img extent tree key (EXTENT_TREE ROOT_ITEM 0) leaf 4210688 items 10 free space 3349 generation 4 owner 2

I can get btrfs to throw a kernel bug easily by running btrfs fi defrag on some files in 3.9.0: May 7 01:57:33 caper kernel: [ 0.000000] Linux version 3.9.0-030900-generic (apw@gomeisa) (gcc version 4.6.3 (Ubuntu/Linaro 4.6.3-1ubuntu5) ) #201304291257 SMP Mon Apr 29 16:58:15 UTC 2013 ... May 7 02:09:21 caper kernel: [ 726.745485] ------------[ cut here ]------------ May 7 02:09:21 caper

While testing raid-auto-repair patches I''m going to send out later, I just found the very last bug in my current scrub patch series: Changelog v4->v5: - fixed a deadlock when fixup is taking longer while scrub is about to end Original message follows: ------------------------ This patch set introduces two new features for scrub. They share the backref iteration code which is the

We do log replay in a single transaction, so it''s not good to do unbound operations during replaying log. This patch makes orphan inodes cleanup executed after replaying log. It also avoid doing other unbound operations such as truncating a file during replaying log. These unbound operations are postponed to the orphan inode cleanup stage. Signed-off-by: Yan Zheng

The following has been done as a part of cleanup - o Eliminated bin_search() by replacing with btrfs_bin_search() o Eliminated unused return value in fixup_low_keys() o Eliminated additional variable (sret) in setup_nodes_for_search() Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.de> --- ctree.c | 71 ++++++++++++++++++---------------------------------------------- 1 file changed, 20

Hi, The following patches add support for ext4 and btrfs to PV-GRUB. These patches are taken nearly verbatim from those provided by Fedora and Gentoo. We''ve been using these patches for the PV-GRUB images available in EC2 for some time now with no problems. Changes from v1: - Makefile has been changed to check the exit code from patch - The btrfs patch has been rebased to apply

Balance will create reloc_root for each fs root, and it''s going to record last_snapshot to filter shared blocks. The side effect of setting last_snapshot is to break nocow attributes of files. So it turns out that checking last_snapshot does not always ensure that a node/leaf/file_extent is shared. That''s why shared node/leaf needs to search extent tree for number of

We at Wiseguys are looking into GlusterFS to run our Internet Archive. The archive stores webpages collected by our spiders. The test setup consists of three data machines, each exporting a volume of about 3.7TB and one nameserver machine. File layout is such that each host has it's own directory, for example the GlusterFS website would be located in:

The kernel side for rescan, which is needed if you want to enable qgroup tracking on a non-empty volume. The first patch splits btrfs_qgroup_account_ref into readable ans reusable units. The second patch adds the rescan implementation (refer to its commit message for a description of the algorithm). The third patch starts an automatic rescan when qgroups are enabled. It is only separated to

Hello everyone. This is the support for default single-device configuration. Subvolumes are _allowed_. Transparent compression is not allowed. Leafsize != 4K is not allowed. Nodesize != 4K is not allowed. Grub-0.97+btrfs doesn''t understand exotic things like preallocated extents (e.g. files composed by fallocate(2), posix_fallocate(3)). Also grub-0.97 doesn''t check btrfs

This is a first draft of a subvolume quota implementation. It is possible to limit subvolumes and any group of subvolumes and also to track the amount of space that will get freed when deleting snapshots. The current version is functionally incomplete, with the main missing feature being the initial scan and rescan of an existing filesystem. I put some effort into writing an introduction into

Write dirty block groups may allocate new block, thus may add new delayed back ref. btrfs_run_delayed_refs may make some block groups dirty. commit_cowonly_roots does not handle the recursion properly, so some dirty blocks can be left unwritten at commit time. This patch move btrfs_run_delayed_refs into the loop that writes dirty block groups, and make the code not break out of the loop until

This patch introduce extent buffer cache for every i-node. By this way, we can save the search time and reduce the lock contention of the root because we needn''t search the item from the root of b+ tree. Implementation: - add two pointers of extent buffer into btrfs_inode struct, one for nodes/leaves of fs/file tree, the other for nodes/leaves of the log tree. - add a variant to tell

-----BEGIN PGP SIGNED MESSAGE----- > Date: Fri, 2 May 1997 12:33:00 -0500 > From: "Thomas H. Ptacek" <tqbf@ENTERACT.COM> > On almost all Unix operating systems, having superuser access in a > chroot() jail is still dangerous. In some recent revisions of 4.4BSD > operating systems, root can trivially escape chroot(), as well. I was thinking about possible attacks

The reason the deadlock is that: Task Btrfs-cleaner umount() down_write(&s->s_umount) sync_filesystem() do auto-defragment and produce lots of dirty pages close_ctree() wait for the end of btrfs-cleaner start_transaction reserve space shrink_delalloc() writeback_inodes_sb_nr_if_idle()

I made an attempt to partly decrease fragment by using a preallocated area of multiple size of a std blocksize for a tree block when it is COWed. The basic idea is that if any a tree block need to be created, we offer, say, 2 or 4 multiples of a std blocksize for it, then use the first block in the continuous blocks. When this block need a cow in the future, a new free block in these continuous

Hello, I want to throw this out here now that I''ve got most of the heavy lifting done for this code to make sure what I''m doing is ok for now. I''ve added an ORPHAN_DIR item key to have a hidden dir per root. Right now it just does it for whatever the default root is on mount, but I''m going to fix that to do the orphan dir check/creation on lookup of a

When the binary search returns 0 (exact match), the target key will necessarily be at slot 0 of all nodes below the current one, so in this case the binary search is not needed because it will always return 0, and we waste time doing it, holding node locks for longer than necessary, etc. Below follow histograms with the times spent on the current approach of doing a binary search when the

Hi, This series adds support to btrfs qgroup related commands, inclduing quota commands, and two leftover of subvolume commands. Regards, Hu Hu Tao (8): New API: btrfs_subvolume_get_default New API: btrfs_subvolume_show New API: btrfs_quota_enable New API: btrfs_quota_disable New API: btrfs_quota_rescan New API: btrfs_qgroup_limit New API: btrfs_qgroup_create New API:

Hi, This is v3 series to add support to btrfs qgroup related commands, inclduing quota commands, and two leftover of subvolume commands. Regards, Hu changes: v3: - don't intialize fs_buf (patch 1) - check the return value of sysroot_path (patch 1) - check fs_buf rather than fs (patch 1) - fprintf (stderr,...) -> reply_with_error() v2: - add tests for new APIs - combine