starred/qemu
1
0
Fork 0
mirror of https://github.com/qemu/qemu.git synced 2026-09-23 23:14:15 +00:00
Code Issues 380 Packages Projects Releases Wiki Activity

[PR #13] fix QED flush: QED did not perform proper flush #26

New Issue
Open
opened 2026-01-31 21:29:25 +00:00 by claunia · 0 comments
claunia commented 2026-01-31 21:29:25 +00:00
Owner
Copy Link

Original Pull Request: https://github.com/qemu/qemu/pull/13

State: closed
Merged: No

Sorry I misunderstood qed... Please ignore this request!!

fix QED flush: QED did not perform proper flush on synchronized write or flush, at least one flush method should be provided (common/co/aio)'

I was doing "sync write" benchmarking on varies qemu supported formats (on ssd). I was surprised that QED shows significant high performance over several other formats (qcow2/vmdk/vdi..). In some testcases QED even runs a little bit faster than raw format!

After careful reading of the code I found QED skipped the necessary flush by providing nothing on flush api. unlike read/write which require at least one type of implementation, flush API is not mandatory in API because your format could guarantee that synchronization is done in write path. However that's not the case in QED because data is buffered in the backing file and need proper flush to make it safe.

However I'm not a skilled qemu developer. Above is what I got from my thinking and study. Please correct me if I misunderstood the system.

Thanks!
Xingbo
(It's quite a hard work to go through the long callback chain, though much easier that tcg's heavy use of glue macros...)

**Original Pull Request:** https://github.com/qemu/qemu/pull/13 **State:** closed **Merged:** No --- Sorry I misunderstood qed... Please ignore this request!! fix QED flush: QED did not perform proper flush on synchronized write or flush, at least one flush method should be provided (common/co/aio)' I was doing "sync write" benchmarking on varies qemu supported formats (on ssd). I was surprised that QED shows significant high performance over several other formats (qcow2/vmdk/vdi..). In some testcases QED even runs a little bit faster than raw format! After careful reading of the code I found QED skipped the necessary flush by providing nothing on _flush_ api. unlike read/write which require at least one type of implementation, flush API is not mandatory in API because your format could guarantee that synchronization is done in write path. However that's not the case in QED because data is buffered in the backing file and need proper flush to make it safe. However I'm not a skilled qemu developer. Above is what I got from my thinking and study. Please correct me if I misunderstood the system. Thanks! Xingbo (It's quite a hard work to go through the long callback chain, though much easier that tcg's heavy use of glue macros...)
claunia added the pull-request label 2026-01-31 21:29:25 +00:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
pull-request
Mirrored from GitHub Pull Request
No Label pull-request
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
claunia
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: starred/qemu#26
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?