Repository navigation
tun: allocate GRO flow capacity on demand - #104
Open
unisequence wants to merge 1 commit into
Open
unisequence wants to merge 1 commit into
unisequence wants to merge 1 commit into
Conversation
Signed-off-by: Murad Rabadanov <the21.21@mail.ru>
unisequence
marked this pull request as ready for review
September 26, 2026 15:25
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The Linux GRO tables reserve a full batch of flow entries and iov slots at construction, even when a TUN device is idle. Under the opt-in
ts_lowmembuild tag, start those slices at one flow and one packet plus header; append still grows to the existing batch limit as traffic requires. Default builds retain their original capacities.On linux/amd64,
BenchmarkLowMemoryGROInitmeasured 2,294,006–2,294,019 B/op in the default build and 69,872 B/op withts_lowmemacross three runs. This measures construction allocations, not steady-state daemon RSS. The tradeoff is additional allocations while previously cold flows first grow under load; grown backing arrays can remain allocated afterward. Full-batch growth and reset tests cover both TCP and UDP, including many distinct flows.Validation:
go test -tags ts_lowmem ./tun; defaultgo test ./tun.