All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH 0/5 v5] ext4: Speedup orphan file handling
@ 2021-08-11 10:19 Jan Kara
  2021-08-11 10:19 ` [PATCH 1/5] ext4: Support for checksumming from journal triggers Jan Kara
                   ` (4 more replies)
  0 siblings, 5 replies; 17+ messages in thread
From: Jan Kara @ 2021-08-11 10:19 UTC (permalink / raw)
  To: Ted Tso; +Cc: linux-ext4, Jan Kara

  Hello,

Here is a fourth version of my series to speed up orphan inode handling in
ext4.

Orphan inode handling in ext4 is a bottleneck for workloads which heavily
excercise truncate / unlink of small files as they contend on global
s_orphan_mutex (when you have fast enough storage). This patch set implements
new way of handling orphan inodes - instead of using a linked list, we store
inode numbers of orphaned inodes in a file which is possible to implement in a
more scalable manner than linked list manipulations. See description of patch
3/5 for more details.

The patch set achieves significant gains both for a micro benchmark stressing
orphan inode handling (truncating file byte-by-byte, several threads in
parallel) and for reaim creat_clo workload. I'm happy for any review, thoughts,
ideas about the patches. I have also implemented full support in e2fsprogs
which I'll send separately.

								Honza

[1] https://lore.kernel.org/lkml/20210227120804.GB22871@xsang-OptiPlex-9020/

Changes since v4:
* Rebased on top of v5.14-rc5
* Updated commit message of patch 1/5
* Added Reviewed-by tags from Ted

Changes since v3:
* Added documentation about on-disk format changes
* Add physical block number into orphan block checksum
* Improve some sanity checks, handling of corrupted orphan file
* Improved some changelogs

Changes since v2:
* Updated some comments
* Rebased onto 5.13-rc5
* Change orphan file inode from a fixed inode number to inode number stored
  in the superblock

Changes since v1:
* orphan blocks have now magic numbers
* split out orphan handling to a separate source file
* some smaller updates according to review

Previous versions:
Link: https://lore.kernel.org/linux-ext4/20210712154009.9290-1-jack@suse.cz/ #v4
Link: https://lore.kernel.org/linux-ext4/20210616105655.5129-1-jack@suse.cz/ #v3
Link: https://lore.kernel.org/linux-ext4/1432293717-24010-1-git-send-email-jack@suse.cz/ #v2

^ permalink raw reply	[flat|nested] 17+ messages in thread
* [PATCH 0/5 v7] ext4: Speedup orphan file handling
@ 2021-08-16  9:57 Jan Kara
  2021-08-16  9:57 ` [PATCH 5/5] ext4: Improve scalability of ext4 " Jan Kara
  0 siblings, 1 reply; 17+ messages in thread
From: Jan Kara @ 2021-08-16  9:57 UTC (permalink / raw)
  To: Ted Tso; +Cc: linux-ext4, Jan Kara

  Hello,

Here is the seventh version of my series to speed up orphan inode handling in
ext4. I've forgot to add a check that orphan file is not exposed in directory
hierarchy. The only change in this version is addition of that fix.

Orphan inode handling in ext4 is a bottleneck for workloads which heavily
excercise truncate / unlink of small files as they contend on global
s_orphan_mutex (when you have fast enough storage). This patch set implements
new way of handling orphan inodes - instead of using a linked list, we store
inode numbers of orphaned inodes in a file which is possible to implement in a
more scalable manner than linked list manipulations. See description of patch
3/5 for more details.

The patch set achieves significant gains both for a micro benchmark stressing
orphan inode handling (truncating file byte-by-byte, several threads in
parallel) and for reaim creat_clo workload. I'm happy for any review, thoughts,
ideas about the patches. I have also implemented full support in e2fsprogs
which I'll send separately.

								Honza

[1] https://lore.kernel.org/lkml/20210227120804.GB22871@xsang-OptiPlex-9020/

Changes since v6:
* Rebased on top of v5.14-rc6 + patches fixing exposure of hidden inodes in
  directory hierarchy
* Added check orphan file cannot be linked from directory hierarchy
* Get orphan file inode with EXT4_IGET_SPECIAL flag

Changes since v5:
* Added Reviewed-by tags from Ted
* Fixed up sparse warning spotted by 0-day
* Fixed error handling path in ext4_orphan_add() to not leak orphan entry

Changes since v4:
* Rebased on top of v5.14-rc5
* Updated commit message of patch 1/5
* Added Reviewed-by tags from Ted

Changes since v3:
* Added documentation about on-disk format changes
* Add physical block number into orphan block checksum
* Improve some sanity checks, handling of corrupted orphan file
* Improved some changelogs

Changes since v2:
* Updated some comments
* Rebased onto 5.13-rc5
* Change orphan file inode from a fixed inode number to inode number stored
  in the superblock

Changes since v1:
* orphan blocks have now magic numbers
* split out orphan handling to a separate source file
* some smaller updates according to review

Previous versions:
Link: http://lore.kernel.org/r/20210816091810.16994-1-jack@suse.cz # v6
Link: http://lore.kernel.org/r/20210811101006.2033-1-jack@suse.cz # v5
Link: https://lore.kernel.org/linux-ext4/20210712154009.9290-1-jack@suse.cz/ #v4
Link: https://lore.kernel.org/linux-ext4/20210616105655.5129-1-jack@suse.cz/ #v3
Link: https://lore.kernel.org/linux-ext4/1432293717-24010-1-git-send-email-jack@suse.cz/ #v2

^ permalink raw reply	[flat|nested] 17+ messages in thread
* [PATCH 0/5 v6] ext4: Speedup orphan file handling
@ 2021-08-16  9:22 Jan Kara
  2021-08-16  9:23 ` [PATCH 5/5] ext4: Improve scalability of ext4 " Jan Kara
  0 siblings, 1 reply; 17+ messages in thread
From: Jan Kara @ 2021-08-16  9:22 UTC (permalink / raw)
  To: Ted Tso; +Cc: linux-ext4, Jan Kara

  Hello,

Here is a fourth version of my series to speed up orphan inode handling in
ext4.

Orphan inode handling in ext4 is a bottleneck for workloads which heavily
excercise truncate / unlink of small files as they contend on global
s_orphan_mutex (when you have fast enough storage). This patch set implements
new way of handling orphan inodes - instead of using a linked list, we store
inode numbers of orphaned inodes in a file which is possible to implement in a
more scalable manner than linked list manipulations. See description of patch
3/5 for more details.

The patch set achieves significant gains both for a micro benchmark stressing
orphan inode handling (truncating file byte-by-byte, several threads in
parallel) and for reaim creat_clo workload. I'm happy for any review, thoughts,
ideas about the patches. I have also implemented full support in e2fsprogs
which I'll send separately.

								Honza

[1] https://lore.kernel.org/lkml/20210227120804.GB22871@xsang-OptiPlex-9020/

Changes since v5:
* Added Reviewed-by tags from Ted
* Fixed up sparse warning spotted by 0-day
* Fixed error handling path in ext4_orphan_add() to not leak orphan entry

Changes since v4:
* Rebased on top of v5.14-rc5
* Updated commit message of patch 1/5
* Added Reviewed-by tags from Ted

Changes since v3:
* Added documentation about on-disk format changes
* Add physical block number into orphan block checksum
* Improve some sanity checks, handling of corrupted orphan file
* Improved some changelogs

Changes since v2:
* Updated some comments
* Rebased onto 5.13-rc5
* Change orphan file inode from a fixed inode number to inode number stored
  in the superblock

Changes since v1:
* orphan blocks have now magic numbers
* split out orphan handling to a separate source file
* some smaller updates according to review

Previous versions:
Link: http://lore.kernel.org/r/20210811101006.2033-1-jack@suse.cz # v5
Link: https://lore.kernel.org/linux-ext4/20210712154009.9290-1-jack@suse.cz/ #v4
Link: https://lore.kernel.org/linux-ext4/20210616105655.5129-1-jack@suse.cz/ #v3
Link: https://lore.kernel.org/linux-ext4/1432293717-24010-1-git-send-email-jack@suse.cz/ #v2

^ permalink raw reply	[flat|nested] 17+ messages in thread
* [PATCH 0/5 v4] ext4: Speedup orphan file handling
@ 2021-07-12 15:40 Jan Kara
  2021-07-12 15:40 ` [PATCH 5/5] ext4: Improve scalability of ext4 " Jan Kara
  0 siblings, 1 reply; 17+ messages in thread
From: Jan Kara @ 2021-07-12 15:40 UTC (permalink / raw)
  To: Ted Tso; +Cc: linux-ext4, Jan Kara

  Hello,

Here is a fourth version of my series to speed up orphan inode handling in
ext4.

Orphan inode handling in ext4 is a bottleneck for workloads which heavily
excercise truncate / unlink of small files as they contend on global
s_orphan_mutex (when you have fast enough storage). This patch set implements
new way of handling orphan inodes - instead of using a linked list, we store
inode numbers of orphaned inodes in a file which is possible to implement in a
more scalable manner than linked list manipulations. See description of patch
3/5 for more details.

The patch set achieves significant gains both for a micro benchmark stressing
orphan inode handling (truncating file byte-by-byte, several threads in
parallel) and for reaim creat_clo workload. I'm happy for any review, thoughts,
ideas about the patches. I have also implemented full support in e2fsprogs
which I'll send separately.

								Honza

[1] https://lore.kernel.org/lkml/20210227120804.GB22871@xsang-OptiPlex-9020/

Changes since v3:
* Added documentation about on-disk format changes
* Add physical block number into orphan block checksum
* Improve some sanity checks, handling of corrupted orphan file
* Improved some changelogs

Changes since v2:
* Updated some comments
* Rebased onto 5.13-rc5
* Change orphan file inode from a fixed inode number to inode number stored
  in the superblock

Changes since v1:
* orphan blocks have now magic numbers
* split out orphan handling to a separate source file
* some smaller updates according to review

^ permalink raw reply	[flat|nested] 17+ messages in thread

end of thread, other threads:[~2021-08-16  9:57 UTC | newest]

Thread overview: 17+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-08-11 10:19 [PATCH 0/5 v5] ext4: Speedup orphan file handling Jan Kara
2021-08-11 10:19 ` [PATCH 1/5] ext4: Support for checksumming from journal triggers Jan Kara
2021-08-11 10:19 ` [PATCH 2/5] ext4: Move orphan inode handling into a separate file Jan Kara
2021-08-11 10:19 ` [PATCH 3/5] ext4: Speedup ext4 orphan inode handling Jan Kara
2021-08-12 15:01   ` Theodore Ts'o
2021-08-13 12:34     ` Jan Kara
2021-08-13 15:26       ` Theodore Ts'o
2021-08-16  9:23         ` Jan Kara
2021-08-11 10:19 ` [PATCH 4/5] ext4: Orphan file documentation Jan Kara
2021-08-12 15:02   ` Theodore Ts'o
2021-08-11 10:19 ` [PATCH 5/5] ext4: Improve scalability of ext4 orphan file handling Jan Kara
2021-08-11 18:19   ` kernel test robot
2021-08-12 12:59     ` Jan Kara
2021-08-12 16:48   ` Theodore Ts'o
  -- strict thread matches above, loose matches on Subject: below --
2021-08-16  9:57 [PATCH 0/5 v7] ext4: Speedup " Jan Kara
2021-08-16  9:57 ` [PATCH 5/5] ext4: Improve scalability of ext4 " Jan Kara
2021-08-16  9:22 [PATCH 0/5 v6] ext4: Speedup " Jan Kara
2021-08-16  9:23 ` [PATCH 5/5] ext4: Improve scalability of ext4 " Jan Kara
2021-07-12 15:40 [PATCH 0/5 v4] ext4: Speedup " Jan Kara
2021-07-12 15:40 ` [PATCH 5/5] ext4: Improve scalability of ext4 " Jan Kara

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.