E2EE: allow setting key_ring_size and key_derivation_algorithm by onestacked · Pull Request #921 · livekit/rust-sdks

onestacked · 2026-02-28T20:33:57Z

This PR uses this webrtc-sdk PR to configure the KDF, instead of just overwriting the derived key afterwards. This should also now properly support KeyRatcheting.

I've tested this with https://codeberg.org/esoteric_programmer/matrix-jukebox and it is compatible with Element Call.

Fixes: #796

onestacked · 2026-03-05T22:39:50Z

Now based on webrtc-sdk/webrtc#224

onestacked · 2026-03-06T10:08:32Z

webrtc-sdk/webrtc#224 is now merged, this will only need a new webrtc-sdk build and update to that. (once thats done I'll look into why node fails.

alexander-potemkin · 2026-03-07T22:32:32Z

Please, keep going, @onestacked !! :)

onestacked · 2026-03-08T10:06:16Z

This is mosly waiting for review and the WebRTC build to be done.
From my side there isn'tuch to do anymore, just fixing the CI (which needs something for node) and maybe adding some tests. Most of the CI failures are because this uses a local WebRTC build which of course doesn't exust in CI.

alexander-potemkin · 2026-03-09T14:22:28Z

@theomonnom , @typester , I hope you would forgive me for pinging you directly. I'm just a person who wants to help a very needed work from @onestacked to make to your product.

maxinflection · 2026-03-12T01:48:51Z

We've been running a production deployment with an equivalent patch set for about three weeks now — a Matrix bot that joins Element Call rooms via LiveKit, decrypts E2EE audio, and transcribes it. Our patches target the same layers (webrtc C++ → webrtc-sys → livekit crates → livekit-ffi) and the approach is functionally identical: branch on a KeyDerivationAlgorithm enum in DeriveKeys(), call BoringSSL's HKDF() with EVP_sha256(), and thread the option through the Rust/FFI stack.
Can confirm:
• HKDF-SHA256 with salt "LKFrameEncryptionKey" and 128 zero-byte info produces correct AES-128-GCM keys matching Element Call's JS SDK
• Key ratcheting still works correctly when left on PBKDF2 (default path unchanged)
• Tested against Element Call clients on Element Web and Element X

Happy to help test once the new webrtc-sdk release unblocks this. Looking forward to dropping our custom .so build.

onestacked · 2026-03-12T07:10:38Z

Webrtc-sdk has just been rebased to a new upstream release: webrtc-sdk/webrtc#217 so livekit will probably update fairly soon.

fkwp · 2026-03-12T16:06:46Z

We've been running a production deployment with an equivalent patch set for about three weeks now — a Matrix bot that joins Element Call rooms via LiveKit, decrypts E2EE audio, and transcribes it. Our patches target the same layers (webrtc C++ → webrtc-sys → livekit crates → livekit-ffi) and the approach is functionally identical: branch on a KeyDerivationAlgorithm enum in DeriveKeys(), call BoringSSL's HKDF() with EVP_sha256(), and thread the option through the Rust/FFI stack. Can confirm: • HKDF-SHA256 with salt "LKFrameEncryptionKey" and 128 zero-byte info produces correct AES-128-GCM keys matching Element Call's JS SDK • Key ratcheting still works correctly when left on PBKDF2 (default path unchanged) • Tested against Element Call clients on Element Web and Element X

Happy to help test once the new webrtc-sdk release unblocks this. Looking forward to dropping our custom .so build.

that's nice @maxinflection that MatrixRTC is getting traction around the matrix rust-sdk. I think the community would be really interested in your project. So yes please share your stuff :-)

onestacked · 2026-03-18T11:30:09Z

I'm not sure why the FFI Proto fails on checkout?

I don't think any change should effect this.

onestacked · 2026-03-18T11:32:12Z

Also let me know if you want my PR split, as it now does ~~three~~ two things (key_ring_size and key_derivation_algorithm ~~and webrtc update~~)

…lgorithm.

cloudwebrtc

lgtm!

@onestacked

> [!IMPORTANT] > Merging this pull request will create these releases # libwebrtc 0.3.27 (2026-03-22) ## Features ### E2EE: allow setting key_ring_size and key_derivation_algorithm, update webrtc to m144 #921 by @onestacked This PR uses [this webrtc-sdk PR](webrtc-sdk/webrtc#224) to configure the KDF. I've tested this with https://codeberg.org/esoteric_programmer/matrix-jukebox and it is compatible with Element Call. Fixed: #796 ## Fixes - Fix H.264 codec matching ### add bounded buffer to audio_stream, and use 10 frames as the default #945 by @xianshijing-lk ### fix clang build issue from zed patches (#949) #950 by @cloudwebrtc * webrtc-sys: Use clang instead of gcc * Debug CI output for aarch64-linux * ci: Install lld for aarch64-linux FFI builders * webrtc-sys: Disable CREL # livekit 0.7.34 (2026-03-22) ## Features ### E2EE: allow setting key_ring_size and key_derivation_algorithm, update webrtc to m144 #921 by @onestacked This PR uses [this webrtc-sdk PR](webrtc-sdk/webrtc#224) to configure the KDF. I've tested this with https://codeberg.org/esoteric_programmer/matrix-jukebox and it is compatible with Element Call. Fixed: #796 ## Fixes - Add disconnectReason to Room::close - End-to-end testing for video streams - Fix H.264 codec matching ### add bounded buffer to audio_stream, and use 10 frames as the default #945 by @xianshijing-lk ### fix PC timeout when connecting with can_subscribe=false #955 by @s-hamdananwar When a participant connects with `canSubscribe=false` in their token, the server sends `subscriber_primary=false` in the JoinResponse and does not send a subscriber offer. This results in `wait_pc_connection` timing out as it is expecting a subscriber PC even when the publisher PC is primary. This PR will skip waiting for subscriber PC when `subscriber_primary=false`. ### Send client os and os_version from rust #952 by @MaxHeimbrock Adds [os_info](https://crates.io/crates/os_info) crate as dependency and sends the data for client connections. # livekit-ffi 0.12.50 (2026-03-22) ## Features ### E2EE: allow setting key_ring_size and key_derivation_algorithm, update webrtc to m144 #921 by @onestacked This PR uses [this webrtc-sdk PR](webrtc-sdk/webrtc#224) to configure the KDF. I've tested this with https://codeberg.org/esoteric_programmer/matrix-jukebox and it is compatible with Element Call. Fixed: #796 ## Fixes - Fix H.264 codec matching ### add bounded buffer to audio_stream, and use 10 frames as the default #945 by @xianshijing-lk ### Send client os and os_version from rust #952 by @MaxHeimbrock Adds [os_info](https://crates.io/crates/os_info) crate as dependency and sends the data for client connections. # webrtc-sys 0.3.25 (2026-03-22) ## Fixes - fix: enable AGC2 adaptive digital controller - Fix H.264 codec matching ### E2EE: allow setting key_ring_size and key_derivation_algorithm, update webrtc to m144 #921 by @onestacked This PR uses [this webrtc-sdk PR](webrtc-sdk/webrtc#224) to configure the KDF. I've tested this with https://codeberg.org/esoteric_programmer/matrix-jukebox and it is compatible with Element Call. Fixed: #796 ### fix clang build issue from zed patches (#949) #950 by @cloudwebrtc * webrtc-sys: Use clang instead of gcc * Debug CI output for aarch64-linux * ci: Install lld for aarch64-linux FFI builders * webrtc-sys: Disable CREL # livekit-datatrack 0.1.1 (2026-03-22) ## Features - Initial release. # webrtc-sys-build 0.3.14 (2026-03-22) ## Fixes ### fix: Bump webrtc build to fix build for Android JNI prefixed. #954 by @cloudwebrtc ### fix clang build issue from zed patches (#949) #950 by @cloudwebrtc * webrtc-sys: Use clang instead of gcc * Debug CI output for aarch64-linux * ci: Install lld for aarch64-linux FFI builders * webrtc-sys: Disable CREL # livekit-protocol 0.7.2 (2026-03-22) ## Fixes - Add disconnectReason to Room::close # livekit-api 0.4.16 (2026-03-22) ## Fixes ### Send client os and os_version from rust #952 by @MaxHeimbrock Adds [os_info](https://crates.io/crates/os_info) crate as dependency and sends the data for client connections. Co-authored-by: knope-bot[bot] <152252888+knope-bot[bot]@users.noreply.github.com>

livekit/rust-sdks#796 is resolved (livekit/rust-sdks#921, present in the locked livekit 0.7.48): KeyProviderOptions exposes key_derivation_algorithm and key_ring_size, and set_key imports per-participant key material. Bridges built with MediaKeyBridge::with_provider forward keys into a provider configured for MSC4195 (HKDF, per-participant mode); new() stays record-only. The native ParticipantKeyHandler aborts the process on any key index at or past the allocated ring size, and rings above 255 are accepted but not allocated. The wire index is a u8, so index 255 is unrepresentable; the bridge rejects out-of-range indices before the FFI boundary. Remaining: surface set_key failures, validate decryption against Element Call via demo/backend.

This was referenced Feb 28, 2026

E2EE: Allow setting key_ring_size and key_derivation_function (And add HKDF as a kdf) #904

Closed

Allow using custom KDF in frame_crypto_transformer webrtc-sdk/webrtc#223

Closed

E2EE HKDF & keyring_size support #796

Closed

onestacked force-pushed the EC-compat-changes-webrtc-change branch from a15edf5 to 379961f Compare March 5, 2026 22:37

onestacked changed the title ~~E2EE: Alternative implementation: allow setting key_ring_size and key_derivation_function (And add HKDF as a kdf)~~ E2EE: allow setting key_ring_size and key_derivation_function (And add HKDF as a kdf) Mar 6, 2026

onestacked changed the title ~~E2EE: allow setting key_ring_size and key_derivation_function (And add HKDF as a kdf)~~ E2EE: allow setting key_ring_size and key_derivation_algorithm Mar 6, 2026

onestacked force-pushed the EC-compat-changes-webrtc-change branch from a775d90 to 303b040 Compare March 18, 2026 09:08

onestacked changed the title ~~E2EE: allow setting key_ring_size and key_derivation_algorithm~~ E2EE: allow setting key_ring_size and key_derivation_algorithm, update webrtc to m144 Mar 18, 2026

onestacked force-pushed the EC-compat-changes-webrtc-change branch from 303b040 to 229e2bb Compare March 18, 2026 11:26

onestacked added 4 commits March 18, 2026 12:47

Allow setting keyring-size from Rust

330375c

Implement using HKDF as the KDF.

8189aa9

Reexport KeyDerivationAlgorithm, use const for default KeyDerivationA…

a3604b2

…lgorithm.

Update webrtc

2b5ec8e

onestacked force-pushed the EC-compat-changes-webrtc-change branch from 229e2bb to 6f225c3 Compare March 18, 2026 11:48

Add changeset

6856cd4

onestacked force-pushed the EC-compat-changes-webrtc-change branch from 6f225c3 to 6856cd4 Compare March 18, 2026 12:06

cloudwebrtc requested changes Mar 19, 2026

View reviewed changes

Comment thread webrtc-sys/build/src/lib.rs Outdated

Comment thread webrtc-sys/libwebrtc/.gclient

onestacked changed the title ~~E2EE: allow setting key_ring_size and key_derivation_algorithm, update webrtc to m144~~ E2EE: allow setting key_ring_size and key_derivation_algorithm Mar 20, 2026

Don't wrongly to m144 change, update webrtc to webrtc-24f6822-2

3af7821

onestacked requested a review from cloudwebrtc March 20, 2026 09:18

cloudwebrtc approved these changes Mar 20, 2026

View reviewed changes

cloudwebrtc merged commit c863be7 into livekit:main Mar 20, 2026
18 of 22 checks passed

knope-bot Bot mentioned this pull request Mar 20, 2026

Release packages #948

Merged

onestacked deleted the EC-compat-changes-webrtc-change branch March 20, 2026 10:10

onestacked mentioned this pull request Mar 20, 2026

Fixup: Remove accidental additional ' in .gclient #957

Merged

xianshijing-lk mentioned this pull request Apr 15, 2026

Fix E2EE connection failure caused by missing required protobuf fields key_ring_size and key_derivation_function in KeyProviderOptions. livekit/python-sdks#628

Merged

This was referenced Jul 21, 2026

E2EE key hand-off: started? BillCarsonFr/matrix-rust-rtc#6

Closed

feat(livekit): import signalled keys into a LiveKit KeyProvider BillCarsonFr/matrix-rust-rtc#7

Merged

Navigation Menu

E2EE: allow setting key_ring_size and key_derivation_algorithm - #921

E2EE: allow setting key_ring_size and key_derivation_algorithm#921
cloudwebrtc merged 6 commits into
livekit:mainfrom
onestacked:EC-compat-changes-webrtc-change

onestacked commented Feb 28, 2026 •

edited

Loading

Uh oh!

onestacked commented Mar 5, 2026

Uh oh!

onestacked commented Mar 6, 2026

Uh oh!

alexander-potemkin commented Mar 7, 2026

Uh oh!

onestacked commented Mar 8, 2026 •

edited

Loading

Uh oh!

alexander-potemkin commented Mar 9, 2026

Uh oh!

maxinflection commented Mar 12, 2026

Uh oh!

onestacked commented Mar 12, 2026 •

edited

Loading

Uh oh!

fkwp commented Mar 12, 2026

Uh oh!

onestacked commented Mar 18, 2026

Uh oh!

onestacked commented Mar 18, 2026 •

edited

Loading

Uh oh!

Uh oh!

Uh oh!

cloudwebrtc left a comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

Conversation

onestacked commented Feb 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

onestacked commented Mar 5, 2026

Uh oh!

onestacked commented Mar 6, 2026

Uh oh!

alexander-potemkin commented Mar 7, 2026

Uh oh!

onestacked commented Mar 8, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

alexander-potemkin commented Mar 9, 2026

Uh oh!

maxinflection commented Mar 12, 2026

Uh oh!

onestacked commented Mar 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

fkwp commented Mar 12, 2026

Uh oh!

onestacked commented Mar 18, 2026

Uh oh!

onestacked commented Mar 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Uh oh!

cloudwebrtc left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

onestacked commented Feb 28, 2026 •

edited

Loading

onestacked commented Mar 8, 2026 •

edited

Loading

onestacked commented Mar 12, 2026 •

edited

Loading

onestacked commented Mar 18, 2026 •

edited

Loading