From 73856f6778e2316c8d0bcf3f741e1fa9f815e0cc Mon Sep 17 00:00:00 2001
From: Claude <noreply@anthropic.com>
Date: Thu, 7 May 2026 15:14:56 +0000
Subject: [PATCH] fix(quic-interop): bump initial flow-control windows to 32MB
 for longrtt

The longrtt testcase failed at 8s with only 1.2 KB received (out of
3 MB requested). Trace from inspect-testcase:

  handshake completed at t=4502 (3 RTTs at 750ms one-way as expected)
  first stream byte arrived at t=4694, ~200ms after
  test killed at t=8s with code=0x0 (graceful close from us)

After handshake, ~3.5s of transfer time was available before the
docker-compose timeout. With our default
initialMaxStreamDataBidiLocal = 1 MB, the peer sends 1 MB then stalls
until our parser fires a MAX_STREAM_DATA bump. At 1.5s RTT each stall
is one round-trip lost (~1.5s). For a 3 MB file that's 2 stalls = 3s
of pure flow-control idle on top of CC slow-start.

Setting initialMaxData and initialMaxStreamData{BidiLocal,
BidiRemote,Uni} to 32 MB removes flow control as a bottleneck for
the largest interop transfers (longrtt 5 MB, transfercorruption few
MB) and lets the peer's congestion control alone drive the rate.

Doesn't help handshake duration (which is RTT-bound and correct).
Doesn't help slow-start ramp (CC, server-side). Should still close
the gap on longrtt.
---
 .../quic/interop/runner/InteropClient.kt      | 20 ++++++++++++++++++-
 1 file changed, 19 insertions(+), 1 deletion(-)

diff --git a/quic/interop/src/main/kotlin/com/vitorpamplona/quic/interop/runner/InteropClient.kt b/quic/interop/src/main/kotlin/com/vitorpamplona/quic/interop/runner/InteropClient.kt
index 12ed601d6..30759e9e5 100644
--- a/quic/interop/src/main/kotlin/com/vitorpamplona/quic/interop/runner/InteropClient.kt
+++ b/quic/interop/src/main/kotlin/com/vitorpamplona/quic/interop/runner/InteropClient.kt
@@ -285,7 +285,25 @@ private fun runTransferTest(
             val conn =
                 QuicConnection(
                     serverName = host,
-                    config = QuicConnectionConfig(),
+                    config =
+                        QuicConnectionConfig(
+                            // Interop stress sizes — push the receive
+                            // window past the largest single-file transfer
+                            // any testcase exercises (longrtt does 5 MB,
+                            // transferloss / transfercorruption do up to
+                            // a few MB). Without this, the peer sends
+                            // up to initialMaxStreamDataBidiLocal then
+                            // stalls until our parser sends a
+                            // MAX_STREAM_DATA — at high RTT (longrtt
+                            // is 750 ms one-way) each stall is ~1.5 s
+                            // round-trip lost. Setting both connection-
+                            // and stream-level windows to 32 MB lets
+                            // the peer's CC alone determine throughput.
+                            initialMaxData = 32L * 1024 * 1024,
+                            initialMaxStreamDataBidiLocal = 32L * 1024 * 1024,
+                            initialMaxStreamDataBidiRemote = 32L * 1024 * 1024,
+                            initialMaxStreamDataUni = 32L * 1024 * 1024,
+                        ),
                     tlsCertificateValidator = PermissiveCertificateValidator(),
                     alpnList = offeredAlpns.map { it.wireBytes },
                     initialVersion = initialVersion,