Revert "Add SSM support to Direct Sockets API (services/network and Blink layers)" [chromium/src : main]

0 views

Skip to first unread message

Yuki Shiino (Gerrit)

unread,

Mar 10, 2026, 9:23:03 PMMar 10

to Yuki Shiino, Rubber Stamper, Muyao Xu, Tom Sepez, Chromium LUCI CQ, Kartik Hegde, Reilly Grant, Omar Ramadan, Jeremy Roman, Andrew Rayskiy, Jordan Bayles, Adam Rice, Vlad Krot, AyeAye, Raphael Kubo da Costa, Simon Hangl, Ian Kilpatrick, chromium...@chromium.org, Nikhil Nayunigari, Chromium IPC Reviews, Kentaro Hara, extension...@chromium.org, chadduffin+w...@chromium.org, bnc+...@chromium.org, blink-revie...@chromium.org, jiajunz+wa...@google.com, jmedle...@chromium.org, jonmann+wa...@chromium.org, net-r...@chromium.org, hsuregan+wa...@chromium.org, stevenjb+wa...@chromium.org, fenced-fra...@chromium.org, jackshira+w...@google.com, blink-...@chromium.org, chromium-a...@chromium.org, khorimoto+w...@chromium.org, ejcaruso+wa...@chromium.org, network-ser...@chromium.org

Attention needed from Adam Rice, Andrew Rayskiy, Jeremy Roman, Jordan Bayles, Kartik Hegde, Muyao Xu, Omar Ramadan, Reilly Grant, Tom Sepez and Vlad Krot

Yuki Shiino voted Owners-Override+1

Owners-Override

Open in Gerrit

Related details

Attention is currently required from:

Adam Rice
Andrew Rayskiy
Jeremy Roman
Jordan Bayles
Kartik Hegde
Muyao Xu
Omar Ramadan
Reilly Grant
Tom Sepez
Vlad Krot

Submit Requirements:

Code-Coverage
Code-Owners
Code-Review
Review-Enforcement

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

satisfied_requirement

unsatisfied_requirement

open

diffy

Rubber Stamper (Gerrit)

unread,

Mar 10, 2026, 9:23:15 PMMar 10

to Yuki Shiino, Muyao Xu, Tom Sepez, Chromium LUCI CQ, Kartik Hegde, Reilly Grant, Omar Ramadan, Jeremy Roman, Andrew Rayskiy, Jordan Bayles, Adam Rice, Vlad Krot, AyeAye, Raphael Kubo da Costa, Simon Hangl, Ian Kilpatrick, chromium...@chromium.org, Nikhil Nayunigari, Chromium IPC Reviews, Kentaro Hara, extension...@chromium.org, chadduffin+w...@chromium.org, bnc+...@chromium.org, blink-revie...@chromium.org, jiajunz+wa...@google.com, jmedle...@chromium.org, jonmann+wa...@chromium.org, net-r...@chromium.org, hsuregan+wa...@chromium.org, stevenjb+wa...@chromium.org, fenced-fra...@chromium.org, jackshira+w...@google.com, blink-...@chromium.org, chromium-a...@chromium.org, khorimoto+w...@chromium.org, ejcaruso+wa...@chromium.org, network-ser...@chromium.org

Attention needed from Adam Rice, Andrew Rayskiy, Jeremy Roman, Jordan Bayles, Kartik Hegde, Muyao Xu, Omar Ramadan, Reilly Grant, Tom Sepez and Vlad Krot

Rubber Stamper voted

Bot-Commit	+1
Commit-Queue	+2

Open in Gerrit

Related details

Attention is currently required from:

Adam Rice
Andrew Rayskiy
Jeremy Roman
Jordan Bayles
Kartik Hegde
Muyao Xu
Omar Ramadan
Reilly Grant
Tom Sepez
Vlad Krot

Submit Requirements:

Code-Coverage
Code-Owners

Code-Review
Review-Enforcement

satisfied_requirement

open

diffy

Chromium LUCI CQ (Gerrit)

unread,

Mar 10, 2026, 9:23:38 PMMar 10

to Yuki Shiino, Rubber Stamper, Muyao Xu, Tom Sepez, Kartik Hegde, Reilly Grant, Omar Ramadan, Jeremy Roman, Andrew Rayskiy, Jordan Bayles, Adam Rice, Vlad Krot, AyeAye, Raphael Kubo da Costa, Simon Hangl, Ian Kilpatrick, chromium...@chromium.org, Nikhil Nayunigari, Chromium IPC Reviews, Kentaro Hara, extension...@chromium.org, chadduffin+w...@chromium.org, bnc+...@chromium.org, blink-revie...@chromium.org, jiajunz+wa...@google.com, jmedle...@chromium.org, jonmann+wa...@chromium.org, net-r...@chromium.org, hsuregan+wa...@chromium.org, stevenjb+wa...@chromium.org, fenced-fra...@chromium.org, jackshira+w...@google.com, blink-...@chromium.org, chromium-a...@chromium.org, khorimoto+w...@chromium.org, ejcaruso+wa...@chromium.org, network-ser...@chromium.org

Chromium LUCI CQ submitted the change

Change information

Commit message:

Revert "Add SSM support to Direct Sockets API (services/network and Blink layers)"

This reverts commit 030237772913cd7805c13962d9a50b145c7ba90d.

Reason for revert: Suspicious about causing test failures of
ChromeDirectSocketsUdpIsolatedWebAppMulticastTest.MulticastJoinGroupSSMSameGroupDifferentSources
https://ci.chromium.org/ui/p/chromium/builders/ci/mac15-x64-rel-tests/15107/overview

Original change's description:
> Add SSM support to Direct Sockets API (services/network and Blink layers)
>
> https://chromestatus.com/feature/6208452397498368
>
> Intent to prototype: https://groups.google.com/a/chromium.org/d/msgid/blink-dev/692dffe2.050a0220.17ec37.0673.GAE%40google.com
>
> Implement Source-Specific Multicast (SSM) support in the Network Service
> and Blink renderer layers, building on the net/socket foundation to
> expose SSM capabilities to isolated web applications through the Direct
> Sockets API.
>
> Network Service Layer (services/network):
> - Extend the RestrictedUDPSocket Mojo interface with an optional source_address
> - Update UDPSocket and RestrictedUDPSocket to pass source filtering parameters
> - Add SSM support to test mocks (TestRestrictedUDPSocket)
>
> Blink Renderer Layer (third_party/blink):
> - Add MulticastController interface for managing group memberships
> - Implement joinGroup()/leaveGroup() with optional source filtering
> - Add MulticastMembership dictionary for SSM (group, source) pairs
> - Validate multicast addresses (IPv4: 224.0.0.0/4, IPv6: ff00::/8)
> - Prevent mixing ASM and SSM for the same group
> - Track active memberships with automatic cleanup on socket close
> - Add joinedGroups() attribute for introspection
>
> Testing:
> - Add 15 MulticastController unit tests (8 ASM + 7 SSM)
>
> ---
>
> Previous CI failures and fixes (PS99 → current PS):
>
> Platforms: fuchsia-x64-cast-receiver-rel, linux-chromeos-rel, linux-rel,
> linux_chromium_asan_rel_ng, linux_chromium_tsan_rel_ng, mac-rel, win-rel
>
> Root cause: FakeRestrictedUDPSocket in multicast_controller_unittest.cc did
> not track source_address in JoinGroup(), so last_join_source_address()
> .has_value() was always false.
>
> Fix: FakeRestrictedUDPSocket::JoinGroup now stores source_address in
> last_join_source_address_. Full Blink->Mojo plumbing:
> - multicast_controller.idl/.h/.cc: joinGroup() accepts
>   MulticastGroupOptions.sourceAddress, parses it, passes it to
>   RestrictedUDPSocket::JoinGroup
> - restricted_udp_socket.mojom: JoinGroup gains net.IPAddress? source_address
> - restricted_udp_socket.cc: forwards source_address to UDPSocket::JoinGroup
> - udp_socket.cc: calls JoinSourceGroup() (SSM) vs JoinGroup() (ASM) based
>   on source_address
>
> Tests: MulticastJoinGroupSSMSameGroupDifferentSources, MulticastJoinLeaveGroupSSM
> Platforms: linux-chromeos-rel, linux-rel, linux-js-coverage-rel
>
> Root cause: MCAST_JOIN_SOURCE_GROUP / MCAST_LEAVE_SOURCE_GROUP were not in
> the BPF setsockopt allowlist for the network service sandbox. Any SSM
> setsockopt call hit CrashSIGSYSSockopt(), killing the process.
>
> Fix: crrev.com/c/7641869 -- MCAST_JOIN_SOURCE_GROUP and
> MCAST_LEAVE_SOURCE_GROUP added to both the IPv4 and IPv6 setsockopt case
> lists in RestrictSetSockoptForNetworkService().
>
> Tests: JoinGroupSSM, JoinGroupSSMSameGroupDifferentSources,
> JoinGroupSSMTwiceWithSameSource, LeaveGroupSSMMustMatchSource
> Platforms: linux-chromeos-rel, linux-rel
>
> Root cause: Same seccomp crash as above. content_browsertests reports the
> process kill as "failed" rather than "crashed".
>
> Fix: Same -- crrev.com/c/7641869.
>
> Root cause: Old patchset used a non-locally-routable address (192.0.2.1,
> TEST-NET-1) as the second source. macOS's GetInterfaceForDestination uses
> connect() to find the outgoing interface; with a non-routable destination it
> returns interface index 0, causing MCAST_JOIN_SOURCE_GROUP to fail.
>
> Fix: direct_sockets_udp_browsertest.cc -- source2 is now constructed by
> toggling the last byte of source1 (a real local address), giving a
> same-subnet address that IS routable via the same interface. The stale
>
> Platforms: linux-chromeos-rel, linux-rel, linux_chromium_asan_rel_ng,
> mac-rel, win-rel
>
> Root cause: Old patchset added ERR_INVALID_ARGUMENT -> NotSupportedError to
> CreateDOMExceptionCodeAndMessageFromNetErrorCode in socket.cc (intended for
> SSM IP-version mismatch). Port 0 is rejected by MockUDPSocket with
> ERR_INVALID_ARGUMENT, so the test received "NotSupportedError" but expected
> HasSubstr("NetworkError").
>
> Fix: socket.cc -- the ERR_INVALID_ARGUMENT case is removed. Port 0 now
> falls through to default -> NetworkError. IP-version mismatch for SSM is
> caught at the Blink validation layer in multicast_controller.cc before
> reaching the network service, so the removed case was effectively dead.
>
> Bug: 461262401
> Change-Id: Ia399ebcdbae425b1b89089eb5bcebbc5e8fa8d45
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7186747
> Reviewed-by: Reilly Grant <rei...@chromium.org>
> Reviewed-by: Tom Sepez <tse...@chromium.org>
> Reviewed-by: Muyao Xu <muy...@google.com>
> Commit-Queue: Reilly Grant <rei...@chromium.org>
> Reviewed-by: Andrew Rayskiy <green...@google.com>
> Reviewed-by: Jordan Bayles <jop...@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#1597329}

Bug: 461262401

No-Presubmit: true

No-Tree-Checks: true

No-Try: true

Change-Id: I2476b2d2f529f91fefcf98d8db5dbebfeadddab8

Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7651295

Commit-Queue: Rubber Stamper <rubber-...@appspot.gserviceaccount.com>

Bot-Commit: Rubber Stamper <rubber-...@appspot.gserviceaccount.com>

Auto-Submit: Yuki Shiino <yukis...@chromium.org>

Owners-Override: Yuki Shiino <yukis...@chromium.org>

Cr-Commit-Position: refs/heads/main@{#1597454}

Files:

M chrome/browser/ash/net/network_diagnostics/fake_udp_socket.cc
M chrome/browser/ash/net/network_diagnostics/fake_udp_socket.h
M chrome/browser/direct_sockets/direct_sockets_apitest.cc
M chrome/test/BUILD.gn
M components/openscreen_platform/udp_socket.cc
M content/browser/direct_sockets/direct_sockets_service_impl.cc
M content/browser/direct_sockets/direct_sockets_test_utils.cc
M content/browser/direct_sockets/direct_sockets_test_utils.h
M content/browser/direct_sockets/direct_sockets_udp_browsertest.cc
M content/test/data/direct_sockets/udp.js
M extensions/browser/api/socket/udp_socket.cc
M net/base/ip_address.cc
M net/base/ip_address.h
M net/base/ip_address_unittest.cc
M services/network/network_context.cc
M services/network/network_context.h
M services/network/network_context_unittest.cc
M services/network/public/mojom/network_context.mojom
M services/network/public/mojom/restricted_udp_socket.mojom
M services/network/public/mojom/udp_socket.mojom
M services/network/restricted_udp_socket.cc
M services/network/restricted_udp_socket.h
M services/network/socket_factory.cc
M services/network/socket_factory.h
M services/network/test/test_network_context.h
M services/network/test/test_restricted_udp_socket.cc
M services/network/test/test_restricted_udp_socket.h
M services/network/test/test_udp_socket.cc
M services/network/test/test_udp_socket.h
M services/network/test/udp_socket_test_util.cc
M services/network/test/udp_socket_test_util.h
M services/network/udp_socket.cc
M services/network/udp_socket.h
M services/network/udp_socket_unittest.cc
M third_party/blink/renderer/bindings/generated_in_modules.gni
M third_party/blink/renderer/modules/direct_sockets/multicast_controller.cc
M third_party/blink/renderer/modules/direct_sockets/multicast_controller.h
M third_party/blink/renderer/modules/direct_sockets/multicast_controller.idl
M third_party/blink/renderer/modules/direct_sockets/multicast_controller_unittest.cc
M third_party/blink/renderer/modules/direct_sockets/socket.cc
M third_party/blink/renderer/modules/direct_sockets/udp_readable_stream_wrapper_unittest.cc
M third_party/blink/renderer/modules/direct_sockets/udp_writable_stream_wrapper_unittest.cc
M third_party/blink/tools/blinkpy/presubmit/audit_non_blink_usage.py

Change size: XL

Delta: 43 files changed, 153 insertions(+), 1499 deletions(-)

Branch: refs/heads/main

Submit Requirements:

Code-Review: Bot-Commit+1 by Rubber Stamper

Open in Gerrit

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

open

diffy

satisfied_requirement

Reply all

Reply to author

Forward

0 new messages