[LU-14712] make TRIM state persistent across reboots

• Type: Improvement
• Resolution: Fixed
• Priority: Minor
• Fix Version/s: Lustre 2.16.0, Lustre 2.17.0
• Labels: NVME ldiskfs performance
• Assignee: Dongyang Li
• Reporter: Andreas Dilger

Description

The current "-o discard" mount option for ldiskfs enables on-the-fly TRIM of underlying flash devices (or thinly-provisioned LUNs). However, the current implementation hurts performance because it is done synchronously in the context of the JBD2 commit thread as a commit callback, which blocks later transaction commits.

There are patches by Shilong that make the TRIM state for a block group persistent, so that running TRIM with mke2fs does not also lead to fstrim resubmitting TRIM requests for all of the groups again immediately after mount/remount.

Most recent e2fsprogs patches at:
https://patchwork.ozlabs.org/project/linux-ext4/patch/[e-mail]/
https://patchwork.ozlabs.org/project/linux-ext4/patch/[e-mail]/

ext4 kernel patch:
https://patchwork.ozlabs.org/project/linux-ext4/patch/[e-mail]/
https://patchwork.ozlabs.org/project/linux-ext4/patch/[e-mail]/

I think the combination of these two patches would improve ongoing ldiskfs performance on flash devices significantly.

Issue Links

• is cloned by: LU-17980 improve ldiskfs "-o discard" performance
• is related to: LU-19219 Change in ("fstrim -v") with EXT4_BG_TRIMMED
• is related to: LU-16750 optimize ldiskfs internal metadata allocation for hybrid storage LUNs

Comments

[Andreas Dilger]

dongyang I just noticed that we do not have the patches to support the EXT4_BG_TRIMMED flag in the Lustre e2fsprogs release. Could you please include those patches in the next release.

[Peter Jones]

Merged for 2.16

[Gerrit Updater]

"Oleg Drokin <[e-mail]>" merged in patch https://review.whamcloud.com/c/fs/lustre-release/+/55567/
Subject: LU-14712 ldiskfs: add bg_trimmed_threshold interface
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 7ac01793a3d451700fee3be6b655c4494ee158aa

[Peter Jones]

Seems to be another patch tracked under this ticket

[Gerrit Updater]

"Li Dongyang <[e-mail]>" uploaded a new patch: https://review.whamcloud.com/c/fs/lustre-release/+/55567
Subject: LU-14712 ldiskfs: add bg_trimmed_threshold interface
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 7222197b34c4f2b9dc87805bc3806279041c8fe5

[Andreas Dilger]

nathand I've created LU-17980 to track the additional work to integrate "-o discard" with the persistent EXT4_BG_TRIMMED flag.

Despite repeated attempts, we haven't been able to convince the upstream ext4 maintainers to accept those patches, so we will carry them in the ldiskfs patch series for the near future.

[Nathan Dauchy]

pjones it appears that the landed patches only address the persistent flagging of blocks as trimmed. I don't see updates on the more interesting "-o discard" performance improvements from async trimming on a block basis. If there are references to that work, can they be added here, or the ticket reopened for tracking the future work?

[Peter Jones]

Merged for 2.16

[Gerrit Updater]

"Oleg Drokin <[e-mail]>" merged in patch https://review.whamcloud.com/c/fs/lustre-release/+/51923/
Subject: LU-14712 ldiskfs: introduce EXT4_BG_TRIMMED to optimize fstrim
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: ad30edf91065d733d7da68de755acf1043429101

[Gerrit Updater]

"Andreas Dilger <[e-mail]>" merged in patch https://review.whamcloud.com/c/tools/e2fsprogs/+/51924/
Subject: LU-14712 e2fsprogs: support EXT2_FLAG_BG_TRIMMED and EXT2_FLAGS_TRACK_TRIM
Project: tools/e2fsprogs
Branch: master-lustre
Current Patch Set:
Commit: 3bf271a76463c3d238f342e39f4f2813e7df6442

[Gerrit Updater]

"Li Dongyang <[e-mail]>" uploaded a new patch: https://review.whamcloud.com/c/tools/e2fsprogs/+/51924
Subject: LU-14712 e2fsprogs: support EXT2_FLAG_BG_TRIMMED and EXT2_FLAGS_TRACK_TRIM
Project: tools/e2fsprogs
Branch: master-lustre
Current Patch Set: 1
Commit: c352e90a9d02af8a94f4fb8e28a95016da2f2d68