-
Notifications
You must be signed in to change notification settings - Fork 1.2k
Comparing changes
Open a pull request
base repository: golang/net
base: v0.38.0
head repository: golang/net
compare: v0.41.0
- 13 commits
- 109 files changed
- 6 contributors
Commits on Apr 7, 2025
-
go.mod: update golang.org/x dependencies
Update golang.org/x dependencies to their latest tagged versions. Change-Id: I9999157ec9379b23f07925ac574003001fdd773c Reviewed-on: https://go-review.googlesource.com/c/net/+/662836 LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Auto-Submit: Gopher Robot <gobot@golang.org> Reviewed-by: Carlos Amedee <carlos@golang.org> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Configuration menu - View commit details
-
Copy full SHA for b8d8877 - Browse repository at this point
Copy the full SHA b8d8877View commit details
Commits on Apr 9, 2025
-
quic: remove go1.21 build constraint
The x/net go.mod now depends on go1.23. Change-Id: I24960949d84753f84f75fbdc97b7dfb92191a4d2 Reviewed-on: https://go-review.googlesource.com/c/net/+/664295 LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Jonathan Amsterdam <jba@google.com> Auto-Submit: Damien Neil <dneil@google.com>
Configuration menu - View commit details
-
Copy full SHA for dd0b200 - Browse repository at this point
Copy the full SHA dd0b200View commit details -
quic: decode packet numbers >255 in tests
Decoding QUIC packet numbers requires keeping track of the largest packet number received so far from the peer. Our tests haven't bothered doing that so far, so tests can't work with packet numbers past 255. Fix that so we can write tests that use more packets. Change-Id: Icb795e5cf69794381c12a3a03b0da6bcf47a69c0 Reviewed-on: https://go-review.googlesource.com/c/net/+/664296 LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Jonathan Amsterdam <jba@google.com> Auto-Submit: Damien Neil <dneil@google.com>
Configuration menu - View commit details
-
Copy full SHA for 22500a6 - Browse repository at this point
Copy the full SHA 22500a6View commit details -
quic: don't re-lose packets when discarding keys
When discarding keys for a packet number space, we remove any in-flight packets in that space from our count of bytes in flight. Skip any packets in the space that were already acked or declared lost. Also skip over acked/lost packets when discarding state in response to a Reset message. This change shouldn't have any effect, since we never declare packets acked/lost until we receive an ACK, and we never handle a Reset after receiving an ACK. Change-Id: I8842f03aa06e57a834a211f1284a48fe5d469db3 Reviewed-on: https://go-review.googlesource.com/c/net/+/664335 LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Jonathan Amsterdam <jba@google.com> Auto-Submit: Damien Neil <dneil@google.com>
Configuration menu - View commit details
-
Copy full SHA for a3b6e77 - Browse repository at this point
Copy the full SHA a3b6e77View commit details -
quic: use an enum for sentPacket state
The sentPacket type tracks the state of a packet sent to the peer. Packets can be in progress, acknowledged, or lost. Track this state with an enum rather than a set of bools, to avoid the possibility of nonsensical states such as a packet being both acknowledged and lost. This also simplifies a following change to add an "unsent" state for intentionally skipped packet numbers. Change-Id: I87c8fc399c72337c033ab7ec5ec8db2c56c732f9 Reviewed-on: https://go-review.googlesource.com/c/net/+/664297 LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Jonathan Amsterdam <jba@google.com> Auto-Submit: Damien Neil <dneil@google.com>
Configuration menu - View commit details
-
Copy full SHA for 3f563d3 - Browse repository at this point
Copy the full SHA 3f563d3View commit details
Commits on Apr 10, 2025
-
quic: skip packet numbers for optimistic ack defense
An "optimistic ACK attack" involves an attacker sending ACKs for packets it hasn't received, causing the victim's congestion controller to improperly send at a higher rate. The standard defense against this attack is to skip the occasional packet number, and to close the connection with an error if the peer ACKs an unsent packet. Implement this defense, increasing the gap between skipped packet numbers as a connection's lifetime grows and correspondingly the amount of work required on the part of the attacker. Change-Id: I01f44f13367821b86af6535ffb69d380e2b4d7b7 Reviewed-on: https://go-review.googlesource.com/c/net/+/664298 LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Jonathan Amsterdam <jba@google.com> Auto-Submit: Damien Neil <dneil@google.com>
Configuration menu - View commit details
-
Copy full SHA for 3e7a445 - Browse repository at this point
Copy the full SHA 3e7a445View commit details
Commits on Apr 16, 2025
-
internal/timeseries: use built-in max/min to simplify the code
Change-Id: I3d408025af072559cac1f2b403a90518995d0a5a Reviewed-on: https://go-review.googlesource.com/c/net/+/664855 Reviewed-by: Damien Neil <dneil@google.com> Auto-Submit: Damien Neil <dneil@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Configuration menu - View commit details
-
Copy full SHA for ea0c1d9 - Browse repository at this point
Copy the full SHA ea0c1d9View commit details
Commits on May 5, 2025
-
go.mod: update golang.org/x dependencies
Update golang.org/x dependencies to their latest tagged versions. Change-Id: Id9bf9fda6785763f418cbe165472193bb23c7401 Reviewed-on: https://go-review.googlesource.com/c/net/+/670096 Reviewed-by: David Chase <drchase@google.com> Reviewed-by: Dmitri Shuralyov <dmitshur@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Auto-Submit: Gopher Robot <gobot@golang.org>
Configuration menu - View commit details
-
Copy full SHA for 7d6e62a - Browse repository at this point
Copy the full SHA 7d6e62aView commit details
Commits on May 12, 2025
-
Change-Id: I4f99953e028a6e397c62605f39b73872abf960bd Reviewed-on: https://go-review.googlesource.com/c/net/+/670995 Reviewed-by: Sean Liao <sean@liao.dev> Auto-Submit: Sean Liao <sean@liao.dev> Reviewed-by: Cherry Mui <cherryyz@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Reviewed-by: Michael Knyszek <mknyszek@google.com>
Configuration menu - View commit details
-
Copy full SHA for bae01a7 - Browse repository at this point
Copy the full SHA bae01a7View commit details -
http2: use an array instead of a map in typeFrameParser
FrameType is a dense integer range, so we can store the frameParsers in an array instead of a map. This should be a very small performance win on all Go http2 servers. For high QPS gRPC services, this function is visible in the Go profiler. For example, it shows up as 0.16% of all CPU time on one production service at Datadog. Change FrameType.String() to use the same pattern. Add a test for testFrameType with unknown FrameTypes. Fixes golang/go#73613 Change-Id: I5f5b523e011a99d6b428cbdbfd97415e488169d1 Reviewed-on: https://go-review.googlesource.com/c/net/+/670415 Reviewed-by: Sean Liao <sean@liao.dev> Reviewed-by: Cherry Mui <cherryyz@google.com> Reviewed-by: Michael Knyszek <mknyszek@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Configuration menu - View commit details
-
Copy full SHA for 919c6bc - Browse repository at this point
Copy the full SHA 919c6bcView commit details
Commits on May 22, 2025
-
internal/http3: use bubbled context in synctest tests
Avoid using T.Context within a synctest bubble. If the Context's Done channel is created outside the bubble, waiting on it within the bubble is durably blocking. If it's created within the bubble, the testing package encounters a panic when closing it after CL 671960. Instead, create our own Context within the bubble and cancel it before the bubble is destroyed. This will be entirely obviated by synctest.Test, which creates a testing.T that returns a properly bubbled context. Change-Id: Iff93c296ccbc1ece8172cb0a60e626ea1bd895ad Reviewed-on: https://go-review.googlesource.com/c/net/+/675615 LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Auto-Submit: Damien Neil <dneil@google.com> Reviewed-by: Jonathan Amsterdam <jba@google.com>
Configuration menu - View commit details
-
Copy full SHA for ef33bc0 - Browse repository at this point
Copy the full SHA ef33bc0View commit details
Commits on May 28, 2025
-
http2: correctly wrap ErrFrameTooLarge in Framer.ReadFrame
In Framer.ReadFrame's frame-too-large check, the code that checks for HTTP/1.1 looking frames accidentally wrapped the unrelated err from the previous readFrameHeader call instead of ErrFrameTooLarge. Fix that. Change-Id: I2237759eaad8c6e06e7195c50410abb5792e57ea Reviewed-on: https://go-review.googlesource.com/c/net/+/676218 Reviewed-by: Chressie Himpel <chressie@google.com> Reviewed-by: Damien Neil <dneil@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Configuration menu - View commit details
-
Copy full SHA for 15f7d40 - Browse repository at this point
Copy the full SHA 15f7d40View commit details
Commits on Jun 5, 2025
-
go.mod: update golang.org/x dependencies
Update golang.org/x dependencies to their latest tagged versions. Change-Id: Ief62c666e731ea5aa7bbf9b6be66c47cb463f867 Reviewed-on: https://go-review.googlesource.com/c/net/+/679315 Reviewed-by: Dmitri Shuralyov <dmitshur@google.com> Reviewed-by: David Chase <drchase@google.com> LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com> Auto-Submit: Gopher Robot <gobot@golang.org>
Configuration menu - View commit details
-
Copy full SHA for 6e41cae - Browse repository at this point
Copy the full SHA 6e41caeView commit details
This comparison is taking too long to generate.
Unfortunately it looks like we can’t render this comparison for you right now. It might be too big, or there might be something weird with your repository.
You can try running this command locally to see the comparison on your machine:
git diff v0.38.0...v0.41.0