linux-ext4 - [PATCH v5 00/10] ext4: support adding multi-delalloc blocks

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Date: Fri, 17 May 2024 20:39:55 +0800
From: Zhang Yi <yi.zhang@...weicloud.com>
To: linux-ext4@...r.kernel.org
Cc: linux-fsdevel@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	tytso@....edu,
	adilger.kernel@...ger.ca,
	jack@...e.cz,
	ritesh.list@...il.com,
	yi.zhang@...wei.com,
	yi.zhang@...weicloud.com,
	chengzhihao1@...wei.com,
	yukuai3@...wei.com
Subject: [PATCH v5 00/10] ext4: support adding multi-delalloc blocks

Changes since v4:
 - In patch 3, switch to check EXT4_ERROR_FS instead of
   ext4_forced_shutdown() to prevent warning on errors=continue mode as
   Jan suggested.
 - In patch 8, rename ext4_da_check_clu_allocated() to
   ext4_clu_alloc_state() and change the return value according to the
   cluster allocation state as Jan suggested.
 - In patch 9, do some appropriate logic changes since
   the ext4_clu_alloc_state() has been changed in patch 8, so I remove
   the reviewed-by tag from Jan, please take a look again.

Changes since v3:
 - Fix two commit message grammatical issues in patch 2 and 4.

Changes since v2:
 - Improve the commit message in patch 2,4,6 as Ritesh and Jan
   suggested, makes the changes more clear.
 - Add patch 3, add a warning if the delalloc counters are still not
   zero on inactive.
 - In patch 6, add a WARN in ext4_es_insert_delayed_extent(), strictly
   requires the end_allocated parameter to be set to false if the
   inserting extent belongs to one cluster.
 - In patch 9, modify the reserve blocks math formula as Jan suggested,
   prevent the count going to be negative.
 - In patch 10, update the stale ext4_da_map_blocks() function comments.

Hello!

This patch series is the part 2 prepartory changes of the buffered IO
iomap conversion, I picked them out from my buffered IO iomap conversion
RFC series v3[1], add a fix for an issue found in current ext4 code, and
also add bigalloc feature support. Please look the following patches for
details.

The first 3 patches fix an incorrect delalloc reserved blocks count
issue and add a warning to make it easy to detect, the second 6 patches
make ext4_insert_delayed_block() call path support inserting
multi-delalloc blocks once a time, and the last patch makes
ext4_da_map_blocks() buffer_head unaware, prepared for iomap.

This patch set has been passed 'kvm-xfstests -g auto' tests, I hope it
could be reviewed and merged first.

[1] https://lore.kernel.org/linux-ext4/20240127015825.1608160-1-yi.zhang@huaweicloud.com/

Thanks,
Yi.

---
v2: https://lore.kernel.org/linux-ext4/20240410034203.2188357-1-yi.zhang@huaweicloud.com/
v3: https://lore.kernel.org/linux-ext4/20240508061220.967970-1-yi.zhang@huaweicloud.com/

Zhang Yi (10):
  ext4: factor out a common helper to query extent map
  ext4: check the extent status again before inserting delalloc block
  ext4: warn if delalloc counters are not zero on inactive
  ext4: trim delalloc extent
  ext4: drop iblock parameter
  ext4: make ext4_es_insert_delayed_block() insert multi-blocks
  ext4: make ext4_da_reserve_space() reserve multi-clusters
  ext4: factor out a helper to check the cluster allocation state
  ext4: make ext4_insert_delayed_block() insert multi-blocks
  ext4: make ext4_da_map_blocks() buffer_head unaware

 fs/ext4/extents_status.c    |  70 +++++++---
 fs/ext4/extents_status.h    |   5 +-
 fs/ext4/inode.c             | 250 +++++++++++++++++++++++-------------
 fs/ext4/super.c             |   6 +-
 include/trace/events/ext4.h |  26 ++--
 5 files changed, 234 insertions(+), 123 deletions(-)

-- 
2.39.2