Repository navigation
feat(fsmonitor): use recvmmsg for batch event receiving - #283
Merged
deepin-bot[bot] merged 1 commit intoOct 13, 2026
Merged
deepin-bot[bot] merged 1 commit into
deepin-bot[bot] merged 1 commit into
Conversation
Reviewer's GuideReplaces single-message recv() draining with nonblocking recvmmsg() batches of up to 64 fs_event messages, reducing syscall overhead during bursts while retaining the existing event-processing semantics. Sequence diagram for batched filesystem event receivingsequenceDiagram
participant Dispatcher
participant Reader as VfsSocketReader
participant Kernel
participant Watcher
Dispatcher->>Kernel: send fs_event messages
Kernel-->>Dispatcher: socketpair delivery
Reader->>Kernel: recvmmsg(socketFd, msgHeaders, 64, 0, nullptr)
Kernel-->>Reader: up to 64 fs_event messages
loop each received message
Reader->>Reader: resolveAndFilterFullPath(deviceId, event.path)
Reader->>Reader: append pendingBatch
end
Reader->>Watcher: flushPendingBatch()
Reader->>Kernel: recvmmsg(...)
Kernel-->>Reader: EAGAIN
File-Level Changes
Tips and commandsInteracting with Sourcery
Customizing Your ExperienceAccess your dashboard to:
Getting Help
|
Replace single-packet recv() with recvmmsg() to drain up to 64 fs_event messages per syscall, reducing system call overhead during high-event bursts and lowering the risk of dispatcher buffer overflow. The iovec and mmsghdr arrays are pre-allocated as member variables and initialized in the VfsSocketReader constructor. Before each recvmmsg batch call, the iovec pointers are reset because the kernel may advance them while consuming messages. The socket is already non-blocking (SOCK_NONBLOCK set by the dispatcher's socketpair), so flag 0 is used instead of MSG_DONTWAIT. Event handling logic (path resolution, mount refresh deferral, pending batch coalescing) is unchanged.
wangrong1069
force-pushed
the
agent/pms-bug-bot/05976ca37725
branch
from
October 13, 2026 06:06
dd0fa6b to
473dd27
Compare
lzwind
approved these changes
Oct 13, 2026
Contributor
Author
|
/merge |
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: lzwind, wangrong1069 The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
feat(fsmonitor): use recvmmsg for batch event receiving
Replace single-packet recv() with recvmmsg() to drain up to 64
fs_event messages per syscall, reducing system call overhead during
high-event bursts and lowering the risk of dispatcher buffer overflow.
The iovec and mmsghdr arrays are pre-allocated as member variables
and initialized in the VfsSocketReader constructor. Before each
recvmmsg batch call, the iovec pointers are reset because the kernel
may advance them while consuming messages.
The socket is already non-blocking (SOCK_NONBLOCK set by the
dispatcher's socketpair), so flag 0 is used instead of MSG_DONTWAIT.
Event handling logic (path resolution, mount refresh deferral,
pending batch coalescing) is unchanged.
Summary by Sourcery
Enhancements: