* [PATCH net-next v2 0/3] selftests/net: packetdrill: netns and two imports
@ 2024-09-12 0:52 Willem de Bruijn
2024-09-12 0:52 ` [PATCH net-next v2 1/3] selftests/net: packetdrill: run in netns and expand config Willem de Bruijn
` (4 more replies)
0 siblings, 5 replies; 9+ messages in thread
From: Willem de Bruijn @ 2024-09-12 0:52 UTC (permalink / raw)
To: netdev
Cc: davem, kuba, edumazet, pabeni, ncardwell, shuah, linux-kselftest,
matttbe, Willem de Bruijn
From: Willem de Bruijn <willemb@google.com>
1/3: run in nets, as discussed, and add missing CONFIGs
2/3: import tcp/zerocopy
3/3: import tcp/slow_start
Willem de Bruijn (3):
selftests/net: packetdrill: run in netns and expand config
selftests/net: packetdrill: import tcp/zerocopy
selftests/net: packetdrill: import tcp/slow_start
.../selftests/net/packetdrill/Makefile | 1 +
.../testing/selftests/net/packetdrill/config | 6 +
.../selftests/net/packetdrill/ksft_runner.sh | 4 +-
.../selftests/net/packetdrill/set_sysctls.py | 38 ++++++
...tcp_slow_start_slow-start-ack-per-1pkt.pkt | 56 +++++++++
...tart_slow-start-ack-per-2pkt-send-5pkt.pkt | 33 +++++
...tart_slow-start-ack-per-2pkt-send-6pkt.pkt | 34 +++++
...tcp_slow_start_slow-start-ack-per-2pkt.pkt | 42 +++++++
...tcp_slow_start_slow-start-ack-per-4pkt.pkt | 35 ++++++
.../tcp_slow_start_slow-start-after-idle.pkt | 39 ++++++
...slow_start_slow-start-after-win-update.pkt | 50 ++++++++
...t_slow-start-app-limited-9-packets-out.pkt | 38 ++++++
.../tcp_slow_start_slow-start-app-limited.pkt | 36 ++++++
..._slow_start_slow-start-fq-ack-per-2pkt.pkt | 63 ++++++++++
.../net/packetdrill/tcp_zerocopy_basic.pkt | 55 ++++++++
.../net/packetdrill/tcp_zerocopy_batch.pkt | 41 ++++++
.../net/packetdrill/tcp_zerocopy_client.pkt | 30 +++++
.../net/packetdrill/tcp_zerocopy_closed.pkt | 44 +++++++
.../packetdrill/tcp_zerocopy_epoll_edge.pkt | 61 +++++++++
.../tcp_zerocopy_epoll_exclusive.pkt | 63 ++++++++++
.../tcp_zerocopy_epoll_oneshot.pkt | 66 ++++++++++
.../tcp_zerocopy_fastopen-client.pkt | 56 +++++++++
.../tcp_zerocopy_fastopen-server.pkt | 44 +++++++
.../net/packetdrill/tcp_zerocopy_maxfrags.pkt | 118 ++++++++++++++++++
.../net/packetdrill/tcp_zerocopy_small.pkt | 57 +++++++++
25 files changed, 1108 insertions(+), 2 deletions(-)
create mode 100755 tools/testing/selftests/net/packetdrill/set_sysctls.py
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-1pkt.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-2pkt-send-5pkt.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-2pkt-send-6pkt.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-2pkt.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-4pkt.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-after-idle.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-after-win-update.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-app-limited-9-packets-out.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-app-limited.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-fq-ack-per-2pkt.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_basic.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_batch.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_client.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_closed.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_edge.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_exclusive.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_oneshot.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_fastopen-client.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_fastopen-server.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_maxfrags.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_small.pkt
--
2.46.0.598.g6f2099f65c-goog
^ permalink raw reply [flat|nested] 9+ messages in thread
* [PATCH net-next v2 1/3] selftests/net: packetdrill: run in netns and expand config
2024-09-12 0:52 [PATCH net-next v2 0/3] selftests/net: packetdrill: netns and two imports Willem de Bruijn
@ 2024-09-12 0:52 ` Willem de Bruijn
2024-09-12 0:52 ` [PATCH net-next v2 2/3] selftests/net: packetdrill: import tcp/zerocopy Willem de Bruijn
` (3 subsequent siblings)
4 siblings, 0 replies; 9+ messages in thread
From: Willem de Bruijn @ 2024-09-12 0:52 UTC (permalink / raw)
To: netdev
Cc: davem, kuba, edumazet, pabeni, ncardwell, shuah, linux-kselftest,
matttbe, Willem de Bruijn
From: Willem de Bruijn <willemb@google.com>
Run packetdrill tests inside netns.
They may change system settings, such as sysctl.
Also expand config with a few more needed CONFIGs.
Link: https://lore.kernel.org/netdev/20240910152640.429920be@kernel.org/
Signed-off-by: Willem de Bruijn <willemb@google.com>
Acked-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
---
tools/testing/selftests/net/packetdrill/config | 5 +++++
tools/testing/selftests/net/packetdrill/ksft_runner.sh | 4 ++--
2 files changed, 7 insertions(+), 2 deletions(-)
diff --git a/tools/testing/selftests/net/packetdrill/config b/tools/testing/selftests/net/packetdrill/config
index 0d402830f18d8..a7877819081f5 100644
--- a/tools/testing/selftests/net/packetdrill/config
+++ b/tools/testing/selftests/net/packetdrill/config
@@ -1,5 +1,10 @@
CONFIG_IPV6=y
+CONFIG_HZ_1000=y
+CONFIG_HZ=1000
+CONFIG_NET_NS=y
CONFIG_NET_SCH_FIFO=y
CONFIG_PROC_SYSCTL=y
+CONFIG_SYN_COOKIES=y
+CONFIG_TCP_CONG_CUBIC=y
CONFIG_TCP_MD5SIG=y
CONFIG_TUN=y
diff --git a/tools/testing/selftests/net/packetdrill/ksft_runner.sh b/tools/testing/selftests/net/packetdrill/ksft_runner.sh
index 2f62caccbbbc5..7478c0c0c9aac 100755
--- a/tools/testing/selftests/net/packetdrill/ksft_runner.sh
+++ b/tools/testing/selftests/net/packetdrill/ksft_runner.sh
@@ -33,9 +33,9 @@ fi
ktap_print_header
ktap_set_plan 2
-packetdrill ${ipv4_args[@]} $(basename $script) > /dev/null \
+unshare -n packetdrill ${ipv4_args[@]} $(basename $script) > /dev/null \
&& ktap_test_pass "ipv4" || ktap_test_fail "ipv4"
-packetdrill ${ipv6_args[@]} $(basename $script) > /dev/null \
+unshare -n packetdrill ${ipv6_args[@]} $(basename $script) > /dev/null \
&& ktap_test_pass "ipv6" || ktap_test_fail "ipv6"
ktap_finished
--
2.46.0.598.g6f2099f65c-goog
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [PATCH net-next v2 2/3] selftests/net: packetdrill: import tcp/zerocopy
2024-09-12 0:52 [PATCH net-next v2 0/3] selftests/net: packetdrill: netns and two imports Willem de Bruijn
2024-09-12 0:52 ` [PATCH net-next v2 1/3] selftests/net: packetdrill: run in netns and expand config Willem de Bruijn
@ 2024-09-12 0:52 ` Willem de Bruijn
2024-09-12 0:52 ` [PATCH net-next v2 3/3] selftests/net: packetdrill: import tcp/slow_start Willem de Bruijn
` (2 subsequent siblings)
4 siblings, 0 replies; 9+ messages in thread
From: Willem de Bruijn @ 2024-09-12 0:52 UTC (permalink / raw)
To: netdev
Cc: davem, kuba, edumazet, pabeni, ncardwell, shuah, linux-kselftest,
matttbe, Willem de Bruijn
From: Willem de Bruijn <willemb@google.com>
Same as initial tests, import verbatim from
github.com/google/packetdrill, aside from:
- update `source ./defaults.sh` path to adjust for flat dir
- add SPDX headers
- remove author statements if any
- drop blank lines at EOF (new)
Also import set_sysctls.py, which many scripts depend on to set
sysctls and then restore them later. This is no longer strictly needed
for namespacified sysctl. But not all sysctls are namespacified, and
doesn't hurt if they are.
Signed-off-by: Willem de Bruijn <willemb@google.com>
---
changes v1->v2:
- add set_sysctls.py to TEST_INCLUDES
- drop blank lines at EOF
---
.../selftests/net/packetdrill/Makefile | 1 +
.../selftests/net/packetdrill/set_sysctls.py | 38 ++++++
.../net/packetdrill/tcp_zerocopy_basic.pkt | 55 ++++++++
.../net/packetdrill/tcp_zerocopy_batch.pkt | 41 ++++++
.../net/packetdrill/tcp_zerocopy_client.pkt | 30 +++++
.../net/packetdrill/tcp_zerocopy_closed.pkt | 44 +++++++
.../packetdrill/tcp_zerocopy_epoll_edge.pkt | 61 +++++++++
.../tcp_zerocopy_epoll_exclusive.pkt | 63 ++++++++++
.../tcp_zerocopy_epoll_oneshot.pkt | 66 ++++++++++
.../tcp_zerocopy_fastopen-client.pkt | 56 +++++++++
.../tcp_zerocopy_fastopen-server.pkt | 44 +++++++
.../net/packetdrill/tcp_zerocopy_maxfrags.pkt | 118 ++++++++++++++++++
.../net/packetdrill/tcp_zerocopy_small.pkt | 57 +++++++++
13 files changed, 674 insertions(+)
create mode 100755 tools/testing/selftests/net/packetdrill/set_sysctls.py
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_basic.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_batch.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_client.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_closed.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_edge.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_exclusive.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_oneshot.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_fastopen-client.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_fastopen-server.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_maxfrags.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_zerocopy_small.pkt
diff --git a/tools/testing/selftests/net/packetdrill/Makefile b/tools/testing/selftests/net/packetdrill/Makefile
index 870f7258dc8d7..31cfb666ba8b3 100644
--- a/tools/testing/selftests/net/packetdrill/Makefile
+++ b/tools/testing/selftests/net/packetdrill/Makefile
@@ -2,6 +2,7 @@
TEST_INCLUDES := ksft_runner.sh \
defaults.sh \
+ set_sysctls.py \
../../kselftest/ktap_helpers.sh
TEST_PROGS := $(wildcard *.pkt)
diff --git a/tools/testing/selftests/net/packetdrill/set_sysctls.py b/tools/testing/selftests/net/packetdrill/set_sysctls.py
new file mode 100755
index 0000000000000..5ddf456ae973a
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/set_sysctls.py
@@ -0,0 +1,38 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: GPL-2.0
+
+"""Sets sysctl values and writes a file that restores them.
+
+The arguments are of the form "<proc-file>=<val>" separated by spaces.
+The program first reads the current value of the proc-file and creates
+a shell script named "/tmp/sysctl_restore_${PACKETDRILL_PID}.sh" which
+restores the values when executed. It then sets the new values.
+
+PACKETDRILL_PID is set by packetdrill to the pid of itself, so a .pkt
+file could restore sysctls by running `/tmp/sysctl_restore_${PPID}.sh`
+at the end.
+"""
+
+import os
+import subprocess
+import sys
+
+filename = '/tmp/sysctl_restore_%s.sh' % os.environ['PACKETDRILL_PID']
+
+# Open file for restoring sysctl values
+restore_file = open(filename, 'w')
+print('#!/bin/bash', file=restore_file)
+
+for a in sys.argv[1:]:
+ sysctl = a.split('=')
+ # sysctl[0] contains the proc-file name, sysctl[1] the new value
+
+ # read current value and add restore command to file
+ cur_val = subprocess.check_output(['cat', sysctl[0]], universal_newlines=True)
+ print('echo "%s" > %s' % (cur_val.strip(), sysctl[0]), file=restore_file)
+
+ # set new value
+ cmd = 'echo "%s" > %s' % (sysctl[1], sysctl[0])
+ os.system(cmd)
+
+os.system('chmod u+x %s' % filename)
diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_basic.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_basic.pkt
new file mode 100644
index 0000000000000..a82c8899d36bf
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_basic.pkt
@@ -0,0 +1,55 @@
+// SPDX-License-Identifier: GPL-2.0
+// basic zerocopy test:
+//
+// send a packet with MSG_ZEROCOPY and receive the notification ID
+// repeat and verify IDs are consecutive
+
+`./defaults.sh`
+
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+ +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
+ +0 setsockopt(3, SOL_SOCKET, SO_ZEROCOPY, [1], 4) = 0
+ +0 bind(3, ..., ...) = 0
+ +0 listen(3, 1) = 0
+
+ +0 < S 0:0(0) win 32792 <mss 1000,sackOK,nop,nop,nop,wscale 7>
+ +0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 8>
+ +0 < . 1:1(0) ack 1 win 257
+
+ +0 accept(3, ..., ...) = 4
+
+ +0 send(4, ..., 4000, MSG_ZEROCOPY) = 4000
+ +0 > P. 1:4001(4000) ack 1
+ +0 < . 1:1(0) ack 4001 win 257
+
+ +0 recvmsg(4, {msg_name(...)=...,
+ msg_iov(1)=[{...,0}],
+ msg_flags=MSG_ERRQUEUE,
+ msg_control=[
+ {cmsg_level=CMSG_LEVEL_IP,
+ cmsg_type=CMSG_TYPE_RECVERR,
+ cmsg_data={ee_errno=0,
+ ee_origin=SO_EE_ORIGIN_ZEROCOPY,
+ ee_type=0,
+ ee_code=SO_EE_CODE_ZEROCOPY_COPIED,
+ ee_info=0,
+ ee_data=0}}
+ ]}, MSG_ERRQUEUE) = 0
+
+ +0 send(4, ..., 4000, MSG_ZEROCOPY) = 4000
+ +0 > P. 4001:8001(4000) ack 1
+ +0 < . 1:1(0) ack 8001 win 257
+
+ +0 recvmsg(4, {msg_name(...)=...,
+ msg_iov(1)=[{...,0}],
+ msg_flags=MSG_ERRQUEUE,
+ msg_control=[
+ {cmsg_level=CMSG_LEVEL_IP,
+ cmsg_type=CMSG_TYPE_RECVERR,
+ cmsg_data={ee_errno=0,
+ ee_origin=SO_EE_ORIGIN_ZEROCOPY,
+ ee_type=0,
+ ee_code=SO_EE_CODE_ZEROCOPY_COPIED,
+ ee_info=1,
+ ee_data=1}}
+ ]}, MSG_ERRQUEUE) = 0
diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_batch.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_batch.pkt
new file mode 100644
index 0000000000000..c01915e7f4a15
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_batch.pkt
@@ -0,0 +1,41 @@
+// SPDX-License-Identifier: GPL-2.0
+// batch zerocopy test:
+//
+// send multiple packets, then read one range of all notifications.
+
+`./defaults.sh`
+
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+ +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
+ +0 setsockopt(3, SOL_SOCKET, SO_ZEROCOPY, [1], 4) = 0
+ +0 bind(3, ..., ...) = 0
+ +0 listen(3, 1) = 0
+
+ +0 < S 0:0(0) win 32792 <mss 1000,sackOK,nop,nop,nop,wscale 7>
+ +0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 8>
+ +0 < . 1:1(0) ack 1 win 257
+
+ +0 accept(3, ..., ...) = 4
+ +0 setsockopt(4, SOL_SOCKET, SO_MARK, [666], 4) = 0
+
+ +0 send(4, ..., 4000, MSG_ZEROCOPY) = 4000
+ +0 > P. 1:4001(4000) ack 1
+ +0 < . 1:1(0) ack 4001 win 257
+
+ +0 send(4, ..., 4000, MSG_ZEROCOPY) = 4000
+ +0 > P. 4001:8001(4000) ack 1
+ +0 < . 1:1(0) ack 8001 win 257
+
+ +0 recvmsg(4, {msg_name(...)=...,
+ msg_iov(1)=[{...,0}],
+ msg_flags=MSG_ERRQUEUE,
+ msg_control=[
+ {cmsg_level=CMSG_LEVEL_IP,
+ cmsg_type=CMSG_TYPE_RECVERR,
+ cmsg_data={ee_errno=0,
+ ee_origin=SO_EE_ORIGIN_ZEROCOPY,
+ ee_type=0,
+ ee_code=SO_EE_CODE_ZEROCOPY_COPIED,
+ ee_info=0,
+ ee_data=1}}
+ ]}, MSG_ERRQUEUE) = 0
diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_client.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_client.pkt
new file mode 100644
index 0000000000000..6509882932e91
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_client.pkt
@@ -0,0 +1,30 @@
+// SPDX-License-Identifier: GPL-2.0
+// Minimal client-side zerocopy test
+
+`./defaults.sh`
+
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 4
+ +0 setsockopt(4, SOL_SOCKET, SO_ZEROCOPY, [1], 4) = 0
+ +0...0 connect(4, ..., ...) = 0
+
+ +0 > S 0:0(0) <mss 1460,sackOK,TS val 0 ecr 0,nop,wscale 8>
+ +0 < S. 0:0(0) ack 1 win 32792 <mss 1000,sackOK,nop,nop,nop,wscale 7>
+ +0 > . 1:1(0) ack 1
+
+ +0 send(4, ..., 4000, MSG_ZEROCOPY) = 4000
+ +0 > P. 1:4001(4000) ack 1
+ +0 < . 1:1(0) ack 4001 win 257
+
+ +0 recvmsg(4, {msg_name(...)=...,
+ msg_iov(1)=[{...,0}],
+ msg_flags=MSG_ERRQUEUE,
+ msg_control=[
+ {cmsg_level=CMSG_LEVEL_IP,
+ cmsg_type=CMSG_TYPE_RECVERR,
+ cmsg_data={ee_errno=0,
+ ee_origin=SO_EE_ORIGIN_ZEROCOPY,
+ ee_type=0,
+ ee_code=SO_EE_CODE_ZEROCOPY_COPIED,
+ ee_info=0,
+ ee_data=0}}
+ ]}, MSG_ERRQUEUE) = 0
diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_closed.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_closed.pkt
new file mode 100644
index 0000000000000..2cd78755cb2ac
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_closed.pkt
@@ -0,0 +1,44 @@
+// SPDX-License-Identifier: GPL-2.0
+// send with MSG_ZEROCOPY on a non-established socket
+//
+// verify that a send in state TCP_CLOSE correctly aborts the zerocopy
+// operation, specifically it does not increment the zerocopy counter.
+//
+// First send on a closed socket and wait for (absent) notification.
+// Then connect and send and verify that notification nr. is zero.
+
+`./defaults.sh`
+
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 4
+ +0 setsockopt(4, SOL_SOCKET, SO_ZEROCOPY, [1], 4) = 0
+
+ +0 send(4, ..., 4000, MSG_ZEROCOPY) = -1 EPIPE (Broken pipe)
+
+ +0.1 recvmsg(4, {msg_name(...)=...,
+ msg_iov(1)=[{...,0}],
+ msg_flags=MSG_ERRQUEUE,
+ msg_control=[]}, MSG_ERRQUEUE) = -1 EAGAIN (Resource temporarily unavailable)
+
+ +0...0 connect(4, ..., ...) = 0
+
+ +0 > S 0:0(0) <mss 1460,sackOK,TS val 0 ecr 0,nop,wscale 8>
+ +0 < S. 0:0(0) ack 1 win 32792 <mss 1000,sackOK,nop,nop,nop,wscale 7>
+ +0 > . 1:1(0) ack 1
+
+ +0 send(4, ..., 4000, MSG_ZEROCOPY) = 4000
+ +0 > P. 1:4001(4000) ack 1
+ +0 < . 1:1(0) ack 4001 win 257
+
+ +0 recvmsg(4, {msg_name(...)=...,
+ msg_iov(1)=[{...,0}],
+ msg_flags=MSG_ERRQUEUE,
+ msg_control=[
+ {cmsg_level=CMSG_LEVEL_IP,
+ cmsg_type=CMSG_TYPE_RECVERR,
+ cmsg_data={ee_errno=0,
+ ee_origin=SO_EE_ORIGIN_ZEROCOPY,
+ ee_type=0,
+ ee_code=SO_EE_CODE_ZEROCOPY_COPIED,
+ ee_info=0,
+ ee_data=0}}
+ ]}, MSG_ERRQUEUE) = 0
diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_edge.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_edge.pkt
new file mode 100644
index 0000000000000..7671c20e01cf6
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_edge.pkt
@@ -0,0 +1,61 @@
+// SPDX-License-Identifier: GPL-2.0
+// epoll zerocopy test:
+//
+// EPOLLERR is known to be not edge-triggered unlike EPOLLIN and EPOLLOUT but
+// it is not level-triggered either.
+//
+// fire two sends with MSG_ZEROCOPY and receive the acks. confirm that EPOLLERR
+// is correctly fired only once, when EPOLLET is set. send another packet with
+// MSG_ZEROCOPY. confirm that EPOLLERR is correctly fired again only once.
+`./defaults.sh`
+
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+ +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
+ +0 setsockopt(3, SOL_SOCKET, SO_ZEROCOPY, [1], 4) = 0
+ +0 bind(3, ..., ...) = 0
+ +0 listen(3, 1) = 0
+
+ +0 < S 0:0(0) win 32792 <mss 1000,sackOK,nop,nop,nop,wscale 7>
+ +0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 8>
+ +0 < . 1:1(0) ack 1 win 257
+
+ +0 accept(3, ..., ...) = 4
+
+ +0 fcntl(4, F_SETFL, O_RDWR|O_NONBLOCK) = 0
+ +0 epoll_create(1) = 5
+ +0 epoll_ctl(5, EPOLL_CTL_ADD, 4, {events=EPOLLOUT|EPOLLET, fd=4}) = 0
+ +0 epoll_wait(5, {events=EPOLLOUT, fd=4}, 1, 0) = 1
+
+ +0 send(4, ..., 4000, MSG_ZEROCOPY) = 4000
+ +0 > P. 1:4001(4000) ack 1
+ +0 < . 1:1(0) ack 4001 win 257
+
+ +0 send(4, ..., 4000, MSG_ZEROCOPY) = 4000
+ +0 > P. 4001:8001(4000) ack 1
+ +0 < . 1:1(0) ack 8001 win 257
+
+// receive only one EPOLLERR for the two sends above.
+ +0 epoll_wait(5, {events=EPOLLERR|EPOLLOUT, fd=4}, 1, 0) = 1
+ +0 epoll_wait(5, {events=0, ptr=0}, 1, 0) = 0
+
+ +0 send(4, ..., 4000, MSG_ZEROCOPY) = 4000
+ +0 > P. 8001:12001(4000) ack 1
+ +0 < . 1:1(0) ack 12001 win 257
+
+// receive only one EPOLLERR for the third send above.
+ +0 epoll_wait(5, {events=EPOLLERR|EPOLLOUT, fd=4}, 1, 0) = 1
+ +0 epoll_wait(5, {events=0, ptr=0}, 1, 0) = 0
+
+ +0 recvmsg(4, {msg_name(...)=...,
+ msg_iov(1)=[{...,0}],
+ msg_flags=MSG_ERRQUEUE,
+ msg_control=[
+ {cmsg_level=CMSG_LEVEL_IP,
+ cmsg_type=CMSG_TYPE_RECVERR,
+ cmsg_data={ee_errno=0,
+ ee_origin=SO_EE_ORIGIN_ZEROCOPY,
+ ee_type=0,
+ ee_code=SO_EE_CODE_ZEROCOPY_COPIED,
+ ee_info=0,
+ ee_data=2}}
+ ]}, MSG_ERRQUEUE) = 0
diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_exclusive.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_exclusive.pkt
new file mode 100644
index 0000000000000..fadc480fdb7fe
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_exclusive.pkt
@@ -0,0 +1,63 @@
+// SPDX-License-Identifier: GPL-2.0
+// epoll zerocopy test:
+//
+// EPOLLERR is known to be not edge-triggered unlike EPOLLIN and EPOLLOUT but
+// it is not level-triggered either. this tests verify that the same behavior is
+// maintained when we have EPOLLEXCLUSIVE.
+//
+// fire two sends with MSG_ZEROCOPY and receive the acks. confirm that EPOLLERR
+// is correctly fired only once, when EPOLLET is set. send another packet with
+// MSG_ZEROCOPY. confirm that EPOLLERR is correctly fired again only once.
+`./defaults.sh`
+
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+ +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
+ +0 setsockopt(3, SOL_SOCKET, SO_ZEROCOPY, [1], 4) = 0
+ +0 bind(3, ..., ...) = 0
+ +0 listen(3, 1) = 0
+
+ +0 < S 0:0(0) win 32792 <mss 1000,sackOK,nop,nop,nop,wscale 7>
+ +0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 8>
+ +0 < . 1:1(0) ack 1 win 257
+
+ +0 accept(3, ..., ...) = 4
+
+ +0 fcntl(4, F_SETFL, O_RDWR|O_NONBLOCK) = 0
+ +0 epoll_create(1) = 5
+ +0 epoll_ctl(5, EPOLL_CTL_ADD, 4,
+ {events=EPOLLOUT|EPOLLET|EPOLLEXCLUSIVE, fd=4}) = 0
+ +0 epoll_wait(5, {events=EPOLLOUT, fd=4}, 1, 0) = 1
+
+ +0 send(4, ..., 4000, MSG_ZEROCOPY) = 4000
+ +0 > P. 1:4001(4000) ack 1
+ +0 < . 1:1(0) ack 4001 win 257
+
+ +0 send(4, ..., 4000, MSG_ZEROCOPY) = 4000
+ +0 > P. 4001:8001(4000) ack 1
+ +0 < . 1:1(0) ack 8001 win 257
+
+// receive only one EPOLLERR for the two sends above.
+ +0 epoll_wait(5, {events=EPOLLERR|EPOLLOUT, fd=4}, 1, 0) = 1
+ +0 epoll_wait(5, {events=0, ptr=0}, 1, 0) = 0
+
+ +0 send(4, ..., 4000, MSG_ZEROCOPY) = 4000
+ +0 > P. 8001:12001(4000) ack 1
+ +0 < . 1:1(0) ack 12001 win 257
+
+// receive only one EPOLLERR for the third send above.
+ +0 epoll_wait(5, {events=EPOLLERR|EPOLLOUT, fd=4}, 1, 0) = 1
+ +0 epoll_wait(5, {events=0, ptr=0}, 1, 0) = 0
+
+ +0 recvmsg(4, {msg_name(...)=...,
+ msg_iov(1)=[{...,0}],
+ msg_flags=MSG_ERRQUEUE,
+ msg_control=[
+ {cmsg_level=CMSG_LEVEL_IP,
+ cmsg_type=CMSG_TYPE_RECVERR,
+ cmsg_data={ee_errno=0,
+ ee_origin=SO_EE_ORIGIN_ZEROCOPY,
+ ee_type=0,
+ ee_code=SO_EE_CODE_ZEROCOPY_COPIED,
+ ee_info=0,
+ ee_data=2}}
+ ]}, MSG_ERRQUEUE) = 0
diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_oneshot.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_oneshot.pkt
new file mode 100644
index 0000000000000..5bfa0d1d2f4a3
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_epoll_oneshot.pkt
@@ -0,0 +1,66 @@
+// SPDX-License-Identifier: GPL-2.0
+// epoll zerocopy test:
+//
+// This is a test to confirm that EPOLLERR is only fired once for an FD when
+// EPOLLONESHOT is set.
+//
+// fire two sends with MSG_ZEROCOPY and receive the acks. confirm that EPOLLERR
+// is correctly fired only once, when EPOLLONESHOT is set. send another packet
+// with MSG_ZEROCOPY. confirm that EPOLLERR is not fired. Rearm the FD and
+// confirm that EPOLLERR is correctly set.
+`./defaults.sh`
+
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+ +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
+ +0 setsockopt(3, SOL_SOCKET, SO_ZEROCOPY, [1], 4) = 0
+ +0 bind(3, ..., ...) = 0
+ +0 listen(3, 1) = 0
+
+ +0 < S 0:0(0) win 32792 <mss 1000,sackOK,nop,nop,nop,wscale 7>
+ +0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 8>
+ +0 < . 1:1(0) ack 1 win 257
+
+ +0 accept(3, ..., ...) = 4
+
+ +0 fcntl(4, F_SETFL, O_RDWR|O_NONBLOCK) = 0
+ +0 epoll_create(1) = 5
+ +0 epoll_ctl(5, EPOLL_CTL_ADD, 4,
+ {events=EPOLLOUT|EPOLLET|EPOLLONESHOT, fd=4}) = 0
+
+ +0 send(4, ..., 4000, MSG_ZEROCOPY) = 4000
+ +0 > P. 1:4001(4000) ack 1
+ +0 < . 1:1(0) ack 4001 win 257
+
+ +0 send(4, ..., 4000, MSG_ZEROCOPY) = 4000
+ +0 > P. 4001:8001(4000) ack 1
+ +0 < . 1:1(0) ack 8001 win 257
+
+// receive only one EPOLLERR for the two sends above.
+ +0 epoll_wait(5, {events=EPOLLERR|EPOLLOUT, fd=4}, 1, 0) = 1
+ +0 epoll_wait(5, {events=0, ptr=0}, 1, 0) = 0
+
+ +0 send(4, ..., 4000, MSG_ZEROCOPY) = 4000
+ +0 > P. 8001:12001(4000) ack 1
+ +0 < . 1:1(0) ack 12001 win 257
+
+// receive no EPOLLERR for the third send above.
+ +0 epoll_wait(5, {events=0, ptr=0}, 1, 0) = 0
+
+// rearm the FD and verify the EPOLLERR is fired again.
+ +0 epoll_ctl(5, EPOLL_CTL_MOD, 4, {events=EPOLLOUT|EPOLLONESHOT, fd=4}) = 0
+ +0 epoll_wait(5, {events=EPOLLERR|EPOLLOUT, fd=4}, 1, 0) = 1
+ +0 epoll_wait(5, {events=0, ptr=0}, 1, 0) = 0
+
+ +0 recvmsg(4, {msg_name(...)=...,
+ msg_iov(1)=[{...,0}],
+ msg_flags=MSG_ERRQUEUE,
+ msg_control=[
+ {cmsg_level=CMSG_LEVEL_IP,
+ cmsg_type=CMSG_TYPE_RECVERR,
+ cmsg_data={ee_errno=0,
+ ee_origin=SO_EE_ORIGIN_ZEROCOPY,
+ ee_type=0,
+ ee_code=SO_EE_CODE_ZEROCOPY_COPIED,
+ ee_info=0,
+ ee_data=2}}
+ ]}, MSG_ERRQUEUE) = 0
diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_fastopen-client.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_fastopen-client.pkt
new file mode 100644
index 0000000000000..4a73bbf469610
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_fastopen-client.pkt
@@ -0,0 +1,56 @@
+// SPDX-License-Identifier: GPL-2.0
+// Fastopen client zerocopy test:
+//
+// send data with MSG_FASTOPEN | MSG_ZEROCOPY and verify that the
+// kernel returns the notification ID.
+//
+// Fastopen requires a stored cookie. Create two sockets. The first
+// one will have no data in the initial send. On return 0 the
+// zerocopy notification counter is not incremented. Verify this too.
+
+`./defaults.sh`
+
+// Send a FastOpen request, no cookie yet so no data in SYN
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+ +0 fcntl(3, F_SETFL, O_RDWR|O_NONBLOCK) = 0
+ +0 setsockopt(3, SOL_SOCKET, SO_ZEROCOPY, [1], 4) = 0
+ +0 sendto(3, ..., 500, MSG_FASTOPEN|MSG_ZEROCOPY, ..., ...) = -1 EINPROGRESS (Operation now in progress)
+ +0 > S 0:0(0) <mss 1460,sackOK,TS val 1000 ecr 0,nop,wscale 8,FO,nop,nop>
+ +.01 < S. 123:123(0) ack 1 win 14600 <mss 940,TS val 2000 ecr 1000,sackOK,nop,wscale 6, FO abcd1234,nop,nop>
+ +0 > . 1:1(0) ack 1 <nop,nop,TS val 1001 ecr 2000>
+
+// Read from error queue: no zerocopy notification
+ +1 recvmsg(3, {msg_name(...)=...,
+ msg_iov(1)=[{...,0}],
+ msg_flags=MSG_ERRQUEUE,
+ msg_control=[]}, MSG_ERRQUEUE) = -1 EAGAIN (Resource temporarily unavailable)
+
+ +.01 close(3) = 0
+ +0 > F. 1:1(0) ack 1 <nop,nop,TS val 1002 ecr 2000>
+ +.01 < F. 1:1(0) ack 2 win 92 <nop,nop,TS val 2001 ecr 1002>
+ +0 > . 2:2(0) ack 2 <nop,nop,TS val 1003 ecr 2001>
+
+// Send another Fastopen request, now SYN will have data
+ +.07 `sysctl -q net.ipv4.tcp_timestamps=0`
+ +.1 socket(..., SOCK_STREAM, IPPROTO_TCP) = 5
+ +0 fcntl(5, F_SETFL, O_RDWR|O_NONBLOCK) = 0
+ +0 setsockopt(5, SOL_SOCKET, SO_ZEROCOPY, [1], 4) = 0
+ +0 sendto(5, ..., 500, MSG_FASTOPEN|MSG_ZEROCOPY, ..., ...) = 500
+ +0 > S 0:500(500) <mss 1460,nop,nop,sackOK,nop,wscale 8,FO abcd1234,nop,nop>
+ +.05 < S. 5678:5678(0) ack 501 win 14600 <mss 1460,nop,nop,sackOK,nop,wscale 6>
+ +0 > . 501:501(0) ack 1
+
+// Read from error queue: now has first zerocopy notification
+ +0.5 recvmsg(5, {msg_name(...)=...,
+ msg_iov(1)=[{...,0}],
+ msg_flags=MSG_ERRQUEUE,
+ msg_control=[
+ {cmsg_level=CMSG_LEVEL_IP,
+ cmsg_type=CMSG_TYPE_RECVERR,
+ cmsg_data={ee_errno=0,
+ ee_origin=SO_EE_ORIGIN_ZEROCOPY,
+ ee_type=0,
+ ee_code=SO_EE_CODE_ZEROCOPY_COPIED,
+ ee_info=0,
+ ee_data=0}}
+ ]}, MSG_ERRQUEUE) = 0
diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_fastopen-server.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_fastopen-server.pkt
new file mode 100644
index 0000000000000..36086c5877ce7
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_fastopen-server.pkt
@@ -0,0 +1,44 @@
+// SPDX-License-Identifier: GPL-2.0
+// Fastopen server zerocopy test:
+//
+// send data with MSG_FASTOPEN | MSG_ZEROCOPY and verify that the
+// kernel returns the notification ID.
+
+`./defaults.sh
+ ./set_sysctls.py /proc/sys/net/ipv4/tcp_fastopen=0x207`
+
+// Set up a TFO server listening socket.
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+ +.1 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
+ +0 bind(3, ..., ...) = 0
+ +0 listen(3, 1) = 0
+ +0 setsockopt(3, SOL_TCP, TCP_FASTOPEN, [2], 4) = 0
+ +0 setsockopt(3, SOL_SOCKET, SO_ZEROCOPY, [1], 4) = 0
+
+// Client sends a SYN with data.
+ +.1 < S 0:1000(1000) win 32792 <mss 1460,sackOK,nop,nop>
+ +0 > S. 0:0(0) ack 1001 <mss 1460,nop,nop,sackOK>
+
+// Server accepts and replies with data.
++.005 accept(3, ..., ...) = 4
+ +0 read(4, ..., 1024) = 1000
+ +0 sendto(4, ..., 1000, MSG_ZEROCOPY, ..., ...) = 1000
+ +0 > P. 1:1001(1000) ack 1001
+ +.05 < . 1001:1001(0) ack 1001 win 32792
+
+// Read from error queue: now has first zerocopy notification
+ +0.1 recvmsg(4, {msg_name(...)=...,
+ msg_iov(1)=[{...,0}],
+ msg_flags=MSG_ERRQUEUE,
+ msg_control=[
+ {cmsg_level=CMSG_LEVEL_IP,
+ cmsg_type=CMSG_TYPE_RECVERR,
+ cmsg_data={ee_errno=0,
+ ee_origin=SO_EE_ORIGIN_ZEROCOPY,
+ ee_type=0,
+ ee_code=SO_EE_CODE_ZEROCOPY_COPIED,
+ ee_info=0,
+ ee_data=0}}
+ ]}, MSG_ERRQUEUE) = 0
+
+`/tmp/sysctl_restore_${PPID}.sh`
diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_maxfrags.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_maxfrags.pkt
new file mode 100644
index 0000000000000..672f817faca0d
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_maxfrags.pkt
@@ -0,0 +1,118 @@
+// SPDX-License-Identifier: GPL-2.0
+// tcp_MAX_SKB_FRAGS test
+//
+// Verify that sending an iovec of tcp_MAX_SKB_FRAGS + 1 elements will
+// 1) fit in a single packet without zerocopy
+// 2) spill over into a second packet with zerocopy,
+// because each iovec element becomes a frag
+// 3) the PSH bit is set on an skb when it runs out of fragments
+
+`./defaults.sh`
+
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+ +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
+ +0 setsockopt(3, SOL_SOCKET, SO_ZEROCOPY, [1], 4) = 0
+
+ // Each pinned zerocopy page is fully accounted to skb->truesize.
+ // This test generates a worst case packet with each frag storing
+ // one byte, but increasing truesize with a page (64KB on PPC).
+ +0 setsockopt(3, SOL_SOCKET, SO_SNDBUF, [2000000], 4) = 0
+
+ +0 bind(3, ..., ...) = 0
+ +0 listen(3, 1) = 0
+
+ +0 < S 0:0(0) win 32792 <mss 1000,sackOK,nop,nop,nop,wscale 7>
+ +0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 8>
+ +0 < . 1:1(0) ack 1 win 257
+
+ +0 accept(3, ..., ...) = 4
+
+ // send an iov of 18 elements: just becomes a linear skb
+ +0 sendmsg(4, {msg_name(...)=...,
+ msg_iov(18)=[{..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}],
+ msg_flags=0}, 0) = 18
+
+ +0 > P. 1:19(18) ack 1
+ +0 < . 1:1(0) ack 19 win 257
+
+ // send a zerocopy iov of 18 elements:
+ +1 sendmsg(4, {msg_name(...)=...,
+ msg_iov(18)=[{..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}],
+ msg_flags=0}, MSG_ZEROCOPY) = 18
+
+ // verify that it is split in one skb of 17 frags + 1 of 1 frag
+ // verify that both have the PSH bit set
+ +0 > P. 19:36(17) ack 1
+ +0 < . 1:1(0) ack 36 win 257
+
+ +0 > P. 36:37(1) ack 1
+ +0 < . 1:1(0) ack 37 win 257
+
+ +1 recvmsg(4, {msg_name(...)=...,
+ msg_iov(1)=[{...,0}],
+ msg_flags=MSG_ERRQUEUE,
+ msg_control=[
+ {cmsg_level=CMSG_LEVEL_IP,
+ cmsg_type=CMSG_TYPE_RECVERR,
+ cmsg_data={ee_errno=0,
+ ee_origin=SO_EE_ORIGIN_ZEROCOPY,
+ ee_type=0,
+ ee_code=SO_EE_CODE_ZEROCOPY_COPIED,
+ ee_info=0,
+ ee_data=0}}
+ ]}, MSG_ERRQUEUE) = 0
+
+ // send a zerocopy iov of 64 elements:
+ +0 sendmsg(4, {msg_name(...)=...,
+ msg_iov(64)=[{..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1},
+ {..., 1}, {..., 1}, {..., 1}, {..., 1}],
+ msg_flags=0}, MSG_ZEROCOPY) = 64
+
+ // verify that it is split in skbs with 17 frags
+ +0 > P. 37:54(17) ack 1
+ +0 < . 1:1(0) ack 54 win 257
+
+ +0 > P. 54:71(17) ack 1
+ +0 < . 1:1(0) ack 71 win 257
+
+ +0 > P. 71:88(17) ack 1
+ +0 < . 1:1(0) ack 88 win 257
+
+ +0 > P. 88:101(13) ack 1
+ +0 < . 1:1(0) ack 101 win 257
+
+ +1 recvmsg(4, {msg_name(...)=...,
+ msg_iov(1)=[{...,0}],
+ msg_flags=MSG_ERRQUEUE,
+ msg_control=[
+ {cmsg_level=CMSG_LEVEL_IP,
+ cmsg_type=CMSG_TYPE_RECVERR,
+ cmsg_data={ee_errno=0,
+ ee_origin=SO_EE_ORIGIN_ZEROCOPY,
+ ee_type=0,
+ ee_code=SO_EE_CODE_ZEROCOPY_COPIED,
+ ee_info=1,
+ ee_data=1}}
+ ]}, MSG_ERRQUEUE) = 0
diff --git a/tools/testing/selftests/net/packetdrill/tcp_zerocopy_small.pkt b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_small.pkt
new file mode 100644
index 0000000000000..a9a1ac0aea4f4
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_zerocopy_small.pkt
@@ -0,0 +1,57 @@
+// SPDX-License-Identifier: GPL-2.0
+// small packet zerocopy test:
+//
+// verify that SO_EE_CODE_ZEROCOPY_COPIED is set on zerocopy
+// packets of all sizes, including the smallest payload, 1B.
+
+`./defaults.sh`
+
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+ +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
+ +0 setsockopt(3, SOL_SOCKET, SO_ZEROCOPY, [1], 4) = 0
+ +0 bind(3, ..., ...) = 0
+ +0 listen(3, 1) = 0
+
+ +0 < S 0:0(0) win 32792 <mss 1000,sackOK,nop,nop,nop,wscale 7>
+ +0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 8>
+ +0 < . 1:1(0) ack 1 win 257
+
+ +0 accept(3, ..., ...) = 4
+
+ // send 1B
+ +0 send(4, ..., 1, MSG_ZEROCOPY) = 1
+ +0 > P. 1:2(1) ack 1
+ +0 < . 1:1(0) ack 2 win 257
+
+ +1 recvmsg(4, {msg_name(...)=...,
+ msg_iov(1)=[{...,0}],
+ msg_flags=MSG_ERRQUEUE,
+ msg_control=[
+ {cmsg_level=CMSG_LEVEL_IP,
+ cmsg_type=CMSG_TYPE_RECVERR,
+ cmsg_data={ee_errno=0,
+ ee_origin=SO_EE_ORIGIN_ZEROCOPY,
+ ee_type=0,
+ ee_code=SO_EE_CODE_ZEROCOPY_COPIED,
+ ee_info=0,
+ ee_data=0}}
+ ]}, MSG_ERRQUEUE) = 0
+
+ // send 1B again
+ +0 send(4, ..., 1, MSG_ZEROCOPY) = 1
+ +0 > P. 2:3(1) ack 1
+ +0 < . 1:1(0) ack 3 win 257
+
+ +1 recvmsg(4, {msg_name(...)=...,
+ msg_iov(1)=[{...,0}],
+ msg_flags=MSG_ERRQUEUE,
+ msg_control=[
+ {cmsg_level=CMSG_LEVEL_IP,
+ cmsg_type=CMSG_TYPE_RECVERR,
+ cmsg_data={ee_errno=0,
+ ee_origin=SO_EE_ORIGIN_ZEROCOPY,
+ ee_type=0,
+ ee_code=SO_EE_CODE_ZEROCOPY_COPIED,
+ ee_info=1,
+ ee_data=1}}
+ ]}, MSG_ERRQUEUE) = 0
--
2.46.0.598.g6f2099f65c-goog
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [PATCH net-next v2 3/3] selftests/net: packetdrill: import tcp/slow_start
2024-09-12 0:52 [PATCH net-next v2 0/3] selftests/net: packetdrill: netns and two imports Willem de Bruijn
2024-09-12 0:52 ` [PATCH net-next v2 1/3] selftests/net: packetdrill: run in netns and expand config Willem de Bruijn
2024-09-12 0:52 ` [PATCH net-next v2 2/3] selftests/net: packetdrill: import tcp/zerocopy Willem de Bruijn
@ 2024-09-12 0:52 ` Willem de Bruijn
2024-09-12 11:22 ` [PATCH net-next v2 0/3] selftests/net: packetdrill: netns and two imports Matthieu Baerts
2024-09-13 2:20 ` patchwork-bot+netdevbpf
4 siblings, 0 replies; 9+ messages in thread
From: Willem de Bruijn @ 2024-09-12 0:52 UTC (permalink / raw)
To: netdev
Cc: davem, kuba, edumazet, pabeni, ncardwell, shuah, linux-kselftest,
matttbe, Willem de Bruijn
From: Willem de Bruijn <willemb@google.com>
Same import process as previous tests.
Also add CONFIG_NET_SCH_FQ to config, as one test uses that.
Same test process as previous tests. Both with and without debug mode.
Recording the steps once:
make mrproper
vng --build \
--config tools/testing/selftests/net/packetdrill/config \
--config kernel/configs/debug.config
vng -v --run . --user root --cpus 4 -- \
make -C tools/testing/selftests TARGETS=net/packetdrill run_tests
Link: https://github.com/linux-netdev/nipa/wiki/How-to-run-netdev-selftests-CI-style#how-to-build
Signed-off-by: Willem de Bruijn <willemb@google.com>
---
.../testing/selftests/net/packetdrill/config | 1 +
...tcp_slow_start_slow-start-ack-per-1pkt.pkt | 56 +++++++++++++++++
...tart_slow-start-ack-per-2pkt-send-5pkt.pkt | 33 ++++++++++
...tart_slow-start-ack-per-2pkt-send-6pkt.pkt | 34 ++++++++++
...tcp_slow_start_slow-start-ack-per-2pkt.pkt | 42 +++++++++++++
...tcp_slow_start_slow-start-ack-per-4pkt.pkt | 35 +++++++++++
.../tcp_slow_start_slow-start-after-idle.pkt | 39 ++++++++++++
...slow_start_slow-start-after-win-update.pkt | 50 +++++++++++++++
...t_slow-start-app-limited-9-packets-out.pkt | 38 +++++++++++
.../tcp_slow_start_slow-start-app-limited.pkt | 36 +++++++++++
..._slow_start_slow-start-fq-ack-per-2pkt.pkt | 63 +++++++++++++++++++
11 files changed, 427 insertions(+)
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-1pkt.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-2pkt-send-5pkt.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-2pkt-send-6pkt.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-2pkt.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-4pkt.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-after-idle.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-after-win-update.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-app-limited-9-packets-out.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-app-limited.pkt
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-fq-ack-per-2pkt.pkt
diff --git a/tools/testing/selftests/net/packetdrill/config b/tools/testing/selftests/net/packetdrill/config
index a7877819081f5..0237ed98f3c0d 100644
--- a/tools/testing/selftests/net/packetdrill/config
+++ b/tools/testing/selftests/net/packetdrill/config
@@ -3,6 +3,7 @@ CONFIG_HZ_1000=y
CONFIG_HZ=1000
CONFIG_NET_NS=y
CONFIG_NET_SCH_FIFO=y
+CONFIG_NET_SCH_FQ=y
CONFIG_PROC_SYSCTL=y
CONFIG_SYN_COOKIES=y
CONFIG_TCP_CONG_CUBIC=y
diff --git a/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-1pkt.pkt b/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-1pkt.pkt
new file mode 100644
index 0000000000000..795c476d222d9
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-1pkt.pkt
@@ -0,0 +1,56 @@
+// SPDX-License-Identifier: GPL-2.0
+// Test of slow start when not application-limited, so that
+// the cwnd continues to grow.
+// In this variant, the receiver ACKs every packet.
+
+// Set up config. To keep things simple, disable the
+// mechanism that defers sending in order to send bigger TSO packets.
+`./defaults.sh
+sysctl -q net.ipv4.tcp_tso_win_divisor=100`
+
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+ +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
+ +0 bind(3, ..., ...) = 0
+ +0 listen(3, 1) = 0
+
+ +.1 < S 0:0(0) win 32792 <mss 1000,sackOK,nop,nop,nop,wscale 7>
+ +0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 8>
+ +.1 < . 1:1(0) ack 1 win 257
+ +0 accept(3, ..., ...) = 4
+ +0 setsockopt(4, SOL_SOCKET, SO_SNDBUF, [200000], 4) = 0
+
+ +0 write(4, ..., 30000) = 30000
+ +0 > P. 1:10001(10000) ack 1
+ +0 %{ assert tcpi_snd_cwnd == 10, tcpi_snd_cwnd }%
+
++.105 < . 1:1(0) ack 1001 win 257
+ +0 > P. 10001:12001(2000) ack 1
+
+ +0 < . 1:1(0) ack 2001 win 257
+ +0 > P. 12001:14001(2000) ack 1
+
++.005 < . 1:1(0) ack 3001 win 257
+ +0 > P. 14001:16001(2000) ack 1
+
+ +0 < . 1:1(0) ack 4001 win 257
+ +0 > P. 16001:18001(2000) ack 1
+
++.005 < . 1:1(0) ack 5001 win 257
+ +0 > P. 18001:20001(2000) ack 1
+
+ +0 < . 1:1(0) ack 6001 win 257
+ +0 > P. 20001:22001(2000) ack 1
+
++.005 < . 1:1(0) ack 7001 win 257
+ +0 > P. 22001:24001(2000) ack 1
+
+ +0 < . 1:1(0) ack 8001 win 257
+ +0 > P. 24001:26001(2000) ack 1
+
++.005 < . 1:1(0) ack 9001 win 257
+ +0 > P. 26001:28001(2000) ack 1
+
+ +0 < . 1:1(0) ack 10001 win 257
+ +0 > P. 28001:30001(2000) ack 1
+
+ +0 %{ assert tcpi_snd_cwnd == 20, tcpi_snd_cwnd }%
diff --git a/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-2pkt-send-5pkt.pkt b/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-2pkt-send-5pkt.pkt
new file mode 100644
index 0000000000000..9212ae1fd0f2f
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-2pkt-send-5pkt.pkt
@@ -0,0 +1,33 @@
+// SPDX-License-Identifier: GPL-2.0
+// Test of slow start when an outstanding flight of packets is
+// less than the current cwnd, and not big enough to bump up cwnd.
+//
+// In this variant, the receiver ACKs every other packet,
+// approximating standard delayed ACKs.
+
+// Set up config.
+`./defaults.sh`
+
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+ +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
+ +0 bind(3, ..., ...) = 0
+ +0 listen(3, 1) = 0
+
+ +0 < S 0:0(0) win 32792 <mss 1000,sackOK,nop,nop,nop,wscale 7>
+ +0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 8>
+ +0 < . 1:1(0) ack 1 win 257
+ +0 accept(3, ..., ...) = 4
+
+// Only send 5 packets.
+ +0 write(4, ..., 5000) = 5000
+ +0 > P. 1:5001(5000) ack 1
+ +0 %{ assert tcpi_snd_cwnd == 10, tcpi_snd_cwnd }%
+
+ +0 < . 1:1(0) ack 2001 win 257
+ +0 %{ assert tcpi_snd_cwnd == 10, 'cwnd=%d' % tcpi_snd_cwnd }%
+
+ +0 < . 1:1(0) ack 4001 win 257
+ +0 %{ assert tcpi_snd_cwnd == 10, 'cwnd=%d' % tcpi_snd_cwnd }%
+
+ +0 < . 1:1(0) ack 5001 win 257
+ +0 %{ assert tcpi_snd_cwnd == 10, 'cwnd=%d' % tcpi_snd_cwnd }%
diff --git a/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-2pkt-send-6pkt.pkt b/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-2pkt-send-6pkt.pkt
new file mode 100644
index 0000000000000..416c901ddf518
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-2pkt-send-6pkt.pkt
@@ -0,0 +1,34 @@
+// SPDX-License-Identifier: GPL-2.0
+// Test of slow start when an outstanding flight of packets is
+// less than the current cwnd, but still big enough that in slow
+// start we want to increase our cwnd a little.
+//
+// In this variant, the receiver ACKs every other packet,
+// approximating standard delayed ACKs.
+
+// Set up config.
+`./defaults.sh`
+
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+ +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
+ +0 bind(3, ..., ...) = 0
+ +0 listen(3, 1) = 0
+
+ +0 < S 0:0(0) win 32792 <mss 1000,sackOK,nop,nop,nop,wscale 7>
+ +0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 8>
+ +0 < . 1:1(0) ack 1 win 257
+ +0 accept(3, ..., ...) = 4
+
+// Only send 6 packets.
+ +0 write(4, ..., 6000) = 6000
+ +0 > P. 1:6001(6000) ack 1
+ +0 %{ assert tcpi_snd_cwnd == 10, tcpi_snd_cwnd }%
+
+ +0 < . 1:1(0) ack 2001 win 257
+ +0 %{ assert tcpi_snd_cwnd == 12, 'cwnd=%d' % tcpi_snd_cwnd }%
+
+ +0 < . 1:1(0) ack 4001 win 257
+ +0 %{ assert tcpi_snd_cwnd == 12, 'cwnd=%d' % tcpi_snd_cwnd }%
+
+ +0 < . 1:1(0) ack 6001 win 257
+ +0 %{ assert tcpi_snd_cwnd == 12, 'cwnd=%d' % tcpi_snd_cwnd }%
diff --git a/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-2pkt.pkt b/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-2pkt.pkt
new file mode 100644
index 0000000000000..a894b7d4559c1
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-2pkt.pkt
@@ -0,0 +1,42 @@
+// SPDX-License-Identifier: GPL-2.0
+// Test of slow start when not application-limited, so that
+// the cwnd continues to grow.
+// In this variant, the receiver ACKs every other packet,
+// approximating standard delayed ACKs.
+
+// Set up config. To keep things simple, disable the
+// mechanism that defers sending in order to send bigger TSO packets.
+`./defaults.sh
+sysctl -q net.ipv4.tcp_tso_win_divisor=100`
+
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+ +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
+ +0 bind(3, ..., ...) = 0
+ +0 listen(3, 1) = 0
+
+ +.1 < S 0:0(0) win 32792 <mss 1000,sackOK,nop,nop,nop,wscale 7>
+ +0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 8>
+ +.1 < . 1:1(0) ack 1 win 257
+ +0 accept(3, ..., ...) = 4
+ +0 setsockopt(4, SOL_SOCKET, SO_SNDBUF, [200000], 4) = 0
+
+ +0 write(4, ..., 30000) = 30000
+ +0 > P. 1:10001(10000) ack 1
+ +0 %{ assert tcpi_snd_cwnd == 10, tcpi_snd_cwnd }%
+
++.105 < . 1:1(0) ack 2001 win 257
+ +0 > P. 10001:14001(4000) ack 1
+
++.005 < . 1:1(0) ack 4001 win 257
+ +0 > P. 14001:18001(4000) ack 1
+
++.005 < . 1:1(0) ack 6001 win 257
+ +0 > P. 18001:22001(4000) ack 1
+
++.005 < . 1:1(0) ack 8001 win 257
+ +0 > P. 22001:26001(4000) ack 1
+
++.005 < . 1:1(0) ack 10001 win 257
+ +0 > P. 26001:30001(4000) ack 1
+
+ +0 %{ assert tcpi_snd_cwnd == 20, tcpi_snd_cwnd }%
diff --git a/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-4pkt.pkt b/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-4pkt.pkt
new file mode 100644
index 0000000000000..065fae9e9abd7
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-ack-per-4pkt.pkt
@@ -0,0 +1,35 @@
+// SPDX-License-Identifier: GPL-2.0
+// Test of slow start when not application-limited, so that
+// the cwnd continues to grow.
+// In this variant, the receiver sends one ACK per 4 packets.
+
+// Set up config. To keep things simple, disable the
+// mechanism that defers sending in order to send bigger TSO packets.
+`./defaults.sh
+sysctl -q net.ipv4.tcp_tso_win_divisor=100`
+
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+ +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
+ +0 bind(3, ..., ...) = 0
+ +0 listen(3, 1) = 0
+
+ +.1 < S 0:0(0) win 32792 <mss 1000,sackOK,nop,nop,nop,wscale 7>
+ +0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 8>
+ +.1 < . 1:1(0) ack 1 win 257
+ +0 accept(3, ..., ...) = 4
+ +0 setsockopt(4, SOL_SOCKET, SO_SNDBUF, [200000], 4) = 0
+
+ +0 write(4, ..., 30000) = 30000
+ +0 > P. 1:10001(10000) ack 1
+ +0 %{ assert tcpi_snd_cwnd == 10, tcpi_snd_cwnd }%
+
+ +.11 < . 1:1(0) ack 4001 win 257
+ +0 > P. 10001:18001(8000) ack 1
+
+ +.01 < . 1:1(0) ack 8001 win 257
+ +0 > P. 18001:26001(8000) ack 1
+
++.005 < . 1:1(0) ack 10001 win 257
+ +0 > P. 26001:30001(4000) ack 1
+
+ +0 %{ assert tcpi_snd_cwnd == 20, tcpi_snd_cwnd }%
diff --git a/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-after-idle.pkt b/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-after-idle.pkt
new file mode 100644
index 0000000000000..11b213be11384
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-after-idle.pkt
@@ -0,0 +1,39 @@
+// SPDX-License-Identifier: GPL-2.0
+// Test of slow start after idle
+// This test expects tso size to be at least initial cwnd * mss
+
+`./defaults.sh
+./set_sysctls.py /proc/sys/net/ipv4/tcp_slow_start_after_idle=1 \
+ /proc/sys/net/ipv4/tcp_min_tso_segs=10`
+
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+ +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
+ +0 bind(3, ..., ...) = 0
+ +0 listen(3, 1) = 0
+
+ +0 < S 0:0(0) win 65535 <mss 1000,sackOK,nop,nop,nop,wscale 7>
+ +0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 8>
+ +.1 < . 1:1(0) ack 1 win 511
+ +0 accept(3, ..., ...) = 4
+ +0 setsockopt(4, SOL_SOCKET, SO_SNDBUF, [200000], 4) = 0
+
+ +0 write(4, ..., 26000) = 26000
+ +0 > P. 1:5001(5000) ack 1
+ +0 > P. 5001:10001(5000) ack 1
+ +0 %{ assert tcpi_snd_cwnd == 10, tcpi_snd_cwnd }%
+
+ +.1 < . 1:1(0) ack 10001 win 511
+ +0 %{ assert tcpi_snd_cwnd == 20, tcpi_snd_cwnd }%
+ +0 > P. 10001:20001(10000) ack 1
+ +0 > P. 20001:26001(6000) ack 1
+
+ +.1 < . 1:1(0) ack 26001 win 511
+ +0 %{ assert tcpi_snd_cwnd == 36, tcpi_snd_cwnd }%
+
+ +2 write(4, ..., 20000) = 20000
+// If slow start after idle works properly, we should send 5 MSS here (cwnd/2)
+ +0 > P. 26001:31001(5000) ack 1
+ +0 %{ assert tcpi_snd_cwnd == 10, tcpi_snd_cwnd }%
+
+// Reset sysctls
+`/tmp/sysctl_restore_${PPID}.sh`
diff --git a/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-after-win-update.pkt b/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-after-win-update.pkt
new file mode 100644
index 0000000000000..577ed8c8852cd
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-after-win-update.pkt
@@ -0,0 +1,50 @@
+// SPDX-License-Identifier: GPL-2.0
+// Test of slow start after window update
+// This test expects tso size to be at least initial cwnd * mss
+
+`./defaults.sh
+./set_sysctls.py /proc/sys/net/ipv4/tcp_slow_start_after_idle=1 \
+ /proc/sys/net/ipv4/tcp_min_tso_segs=10`
+
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+ +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
+ +0 bind(3, ..., ...) = 0
+ +0 listen(3, 1) = 0
+
+ +0 < S 0:0(0) win 65535 <mss 1000,sackOK,nop,nop,nop,wscale 7>
+ +0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 8>
+ +.1 < . 1:1(0) ack 1 win 511
+ +0 accept(3, ..., ...) = 4
+ +0 setsockopt(4, SOL_SOCKET, SO_SNDBUF, [200000], 4) = 0
+
+ +0 write(4, ..., 26000) = 26000
+ +0 > P. 1:5001(5000) ack 1
+ +0 > P. 5001:10001(5000) ack 1
+ +0 %{ assert tcpi_snd_cwnd == 10, tcpi_snd_cwnd }%
+
+ +.1 < . 1:1(0) ack 10001 win 511
+ +0 %{ assert tcpi_snd_cwnd == 20, tcpi_snd_cwnd }%
+ +0 > P. 10001:20001(10000) ack 1
+ +0 > P. 20001:26001(6000) ack 1
+
+ +.1 < . 1:1(0) ack 26001 win 0
+ +0 %{ assert tcpi_snd_cwnd == 36, tcpi_snd_cwnd }%
+
+ +0 write(4, ..., 20000) = 20000
+// 1st win0 probe
++.3~+.310 > . 26000:26000(0) ack 1
+ +0 %{ assert tcpi_snd_cwnd == 36, tcpi_snd_cwnd }%
+
+// 2nd win0 probe
++.6~+.620 > . 26000:26000(0) ack 1
+ +0 %{ assert tcpi_snd_cwnd == 36, tcpi_snd_cwnd }%
+
+// 3rd win0 probe
++1.2~+1.240 > . 26000:26000(0) ack 1
+ +0 %{ assert tcpi_snd_cwnd == 36, tcpi_snd_cwnd }%
+
+ +.9 < . 1:1(0) ack 26001 win 511
+ +0 > P. 26001:31001(5000) ack 1
+
+// Reset sysctls
+`/tmp/sysctl_restore_${PPID}.sh`
diff --git a/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-app-limited-9-packets-out.pkt b/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-app-limited-9-packets-out.pkt
new file mode 100644
index 0000000000000..869f32c35a2ab
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-app-limited-9-packets-out.pkt
@@ -0,0 +1,38 @@
+// SPDX-License-Identifier: GPL-2.0
+// Test of slow start when application-limited: in this case,
+// with IW10, if we don't fully use our cwnd but instead
+// send just 9 packets, then cwnd should grow to twice that
+// value, or 18 packets.
+
+// Set up config.
+`./defaults.sh`
+
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+ +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
+ +0 bind(3, ..., ...) = 0
+ +0 listen(3, 1) = 0
+
+ +.1 < S 0:0(0) win 32792 <mss 1000,sackOK,nop,nop,nop,wscale 7>
+ +0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 8>
+ +.1 < . 1:1(0) ack 1 win 257
+ +0 accept(3, ..., ...) = 4
+ +0 setsockopt(4, SOL_SOCKET, SO_SNDBUF, [200000], 4) = 0
+
+ +0 write(4, ..., 9000) = 9000
+ +0 > P. 1:9001(9000) ack 1
+ +0 %{ assert tcpi_snd_cwnd == 10, tcpi_snd_cwnd }%
+
++.105 < . 1:1(0) ack 2001 win 257
+ +0 %{ assert tcpi_snd_cwnd == 12, tcpi_snd_cwnd }%
+
++.005 < . 1:1(0) ack 4001 win 257
+ +0 %{ assert tcpi_snd_cwnd == 14, tcpi_snd_cwnd }%
+
++.005 < . 1:1(0) ack 6001 win 257
+ +0 %{ assert tcpi_snd_cwnd == 16, tcpi_snd_cwnd }%
+
++.005 < . 1:1(0) ack 8001 win 257
+ +0 %{ assert tcpi_snd_cwnd == 18, tcpi_snd_cwnd }%
+
++.005 < . 1:1(0) ack 9001 win 257
+ +0 %{ assert tcpi_snd_cwnd == 18, tcpi_snd_cwnd }%
diff --git a/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-app-limited.pkt b/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-app-limited.pkt
new file mode 100644
index 0000000000000..0f77b7955db6e
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-app-limited.pkt
@@ -0,0 +1,36 @@
+// SPDX-License-Identifier: GPL-2.0
+// Test of slow start when application-limited: in this case,
+// with IW10, if we send exactly 10 packets then cwnd should grow to 20.
+
+// Set up config.
+`./defaults.sh`
+
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+ +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
+ +0 bind(3, ..., ...) = 0
+ +0 listen(3, 1) = 0
+
+ +.1 < S 0:0(0) win 32792 <mss 1000,sackOK,nop,nop,nop,wscale 7>
+ +0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 8>
+ +.1 < . 1:1(0) ack 1 win 257
+ +0 accept(3, ..., ...) = 4
+ +0 setsockopt(4, SOL_SOCKET, SO_SNDBUF, [200000], 4) = 0
+
+ +0 write(4, ..., 10000) = 10000
+ +0 > P. 1:10001(10000) ack 1
+ +0 %{ assert tcpi_snd_cwnd == 10, tcpi_snd_cwnd }%
+
++.105 < . 1:1(0) ack 2001 win 257
+ +0 %{ assert tcpi_snd_cwnd == 12, tcpi_snd_cwnd }%
+
++.005 < . 1:1(0) ack 4001 win 257
+ +0 %{ assert tcpi_snd_cwnd == 14, tcpi_snd_cwnd }%
+
++.005 < . 1:1(0) ack 6001 win 257
+ +0 %{ assert tcpi_snd_cwnd == 16, tcpi_snd_cwnd }%
+
++.005 < . 1:1(0) ack 8001 win 257
+ +0 %{ assert tcpi_snd_cwnd == 18, tcpi_snd_cwnd }%
+
++.005 < . 1:1(0) ack 10001 win 257
+ +0 %{ assert tcpi_snd_cwnd == 20, tcpi_snd_cwnd }%
diff --git a/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-fq-ack-per-2pkt.pkt b/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-fq-ack-per-2pkt.pkt
new file mode 100644
index 0000000000000..7e9c83d617c25
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_slow_start_slow-start-fq-ack-per-2pkt.pkt
@@ -0,0 +1,63 @@
+// SPDX-License-Identifier: GPL-2.0
+// Test of slow start when not application-limited, so that
+// the cwnd continues to grow, even if TSQ triggers.
+// In this variant, the receiver ACKs every other packet,
+// approximating standard delayed ACKs.
+
+// Note we use FQ/pacing to check if TCP Small Queues is not hurting
+
+`./defaults.sh
+tc qdisc replace dev tun0 root fq
+sysctl -q net/ipv4/tcp_pacing_ss_ratio=200
+sysctl -e -q net.ipv4.tcp_min_tso_segs=2`
+
+ 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+ +0 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
+ +0 bind(3, ..., ...) = 0
+ +0 listen(3, 1) = 0
+
+ +.1 < S 0:0(0) win 32792 <mss 1460,sackOK,nop,nop,nop,wscale 7>
+ +0 > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 8>
+ +.1 < . 1:1(0) ack 1 win 500
+ +0 accept(3, ..., ...) = 4
+ +0 setsockopt(4, SOL_SOCKET, SO_SNDBUF, [200000], 4) = 0
+
+ +0 write(4, ..., 40000) = 40000
+// This might change if we cook the initial packet with 10 MSS.
+ +0 > P. 1:2921(2920) ack 1
+ +0 > P. 2921:5841(2920) ack 1
+ +0 > P. 5841:8761(2920) ack 1
+ +0 > P. 8761:11681(2920) ack 1
+ +0 > P. 11681:14601(2920) ack 1
+ +0 %{ assert tcpi_snd_cwnd == 10, tcpi_snd_cwnd }%
+
++.105 < . 1:1(0) ack 2921 win 500
+ +0 %{ assert tcpi_snd_cwnd == 12, tcpi_snd_cwnd }%
+
+// Note: after this commit : "net_sched: sch_fq: account for schedule/timers drifts"
+// FQ notices that this packet missed the 'time to send next packet' computed
+// when prior packet (11681:14601(2920)) was sent.
+// So FQ will allow following packet to be sent a bit earlier (quantum/2)
+// (FQ commit allows an application/cwnd limited flow to get at most quantum/2 extra credit)
+ +0 > P. 14601:17521(2920) ack 1
+
++.003 < . 1:1(0) ack 5841 win 500
+ +0 %{ assert tcpi_snd_cwnd == 14, tcpi_snd_cwnd }%
+
++.001 > P. 17521:20441(2920) ack 1
+
++.001 < . 1:1(0) ack 8761 win 500
+ +0 %{ assert tcpi_snd_cwnd == 16, tcpi_snd_cwnd }%
+
+// remaining packets are delivered at a constant rate.
++.007 > P. 20441:23361(2920) ack 1
+
++.002 < . 1:1(0) ack 11681 win 500
+ +0 %{ assert tcpi_snd_cwnd == 18, tcpi_snd_cwnd }%
++.001 < . 1:1(0) ack 14601 win 500
+
++.004 > P. 23361:26281(2920) ack 1
+
++.007 > P. 26281:29201(2920) ack 1
+
+ +0 %{ assert tcpi_snd_cwnd == 20, 'cwnd=%d' % tcpi_snd_cwnd }%
--
2.46.0.598.g6f2099f65c-goog
^ permalink raw reply related [flat|nested] 9+ messages in thread
* Re: [PATCH net-next v2 0/3] selftests/net: packetdrill: netns and two imports
2024-09-12 0:52 [PATCH net-next v2 0/3] selftests/net: packetdrill: netns and two imports Willem de Bruijn
` (2 preceding siblings ...)
2024-09-12 0:52 ` [PATCH net-next v2 3/3] selftests/net: packetdrill: import tcp/slow_start Willem de Bruijn
@ 2024-09-12 11:22 ` Matthieu Baerts
2024-09-12 12:10 ` Willem de Bruijn
2024-09-13 2:20 ` patchwork-bot+netdevbpf
4 siblings, 1 reply; 9+ messages in thread
From: Matthieu Baerts @ 2024-09-12 11:22 UTC (permalink / raw)
To: Willem de Bruijn, netdev
Cc: davem, kuba, edumazet, pabeni, ncardwell, shuah, linux-kselftest,
Willem de Bruijn
Hi Willem,
On 12/09/2024 02:52, Willem de Bruijn wrote:
> From: Willem de Bruijn <willemb@google.com>
>
> 1/3: run in nets, as discussed, and add missing CONFIGs
> 2/3: import tcp/zerocopy
> 3/3: import tcp/slow_start
Thank you for the v2. This new version looks good to me:
Acked-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
I didn't pay too much attention to the new tests, because they look
good, heavily tested I suppose, and I guess the goal is not to diverge
from the original ones for the moment. Still, please note that the CI
reported some timing issues with tcp_zerocopy_closed.pkt when using a
debug kernel config, e.g.
> tcp_zerocopy_closed.pkt:22: timing error: expected system call return at 0.100596 sec but happened at 0.109564 sec; tolerance 0.004000 sec
https://netdev.bots.linux.dev/contest.html?executor=vmksft-packetdrill-dbg&test=tcp-zerocopy-closed-pkt
Cheers,
Matt
--
Sponsored by the NGI0 Core fund.
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [PATCH net-next v2 0/3] selftests/net: packetdrill: netns and two imports
2024-09-12 11:22 ` [PATCH net-next v2 0/3] selftests/net: packetdrill: netns and two imports Matthieu Baerts
@ 2024-09-12 12:10 ` Willem de Bruijn
2024-09-16 16:48 ` Stanislav Fomichev
0 siblings, 1 reply; 9+ messages in thread
From: Willem de Bruijn @ 2024-09-12 12:10 UTC (permalink / raw)
To: Matthieu Baerts, Willem de Bruijn, netdev
Cc: davem, kuba, edumazet, pabeni, ncardwell, shuah, linux-kselftest,
Willem de Bruijn
Matthieu Baerts wrote:
> Hi Willem,
>
> On 12/09/2024 02:52, Willem de Bruijn wrote:
> > From: Willem de Bruijn <willemb@google.com>
> >
> > 1/3: run in nets, as discussed, and add missing CONFIGs
> > 2/3: import tcp/zerocopy
> > 3/3: import tcp/slow_start
>
> Thank you for the v2. This new version looks good to me:
>
> Acked-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
>
>
> I didn't pay too much attention to the new tests, because they look
> good, heavily tested I suppose, and I guess the goal is not to diverge
> from the original ones for the moment. Still, please note that the CI
> reported some timing issues with tcp_zerocopy_closed.pkt when using a
> debug kernel config, e.g.
>
> > tcp_zerocopy_closed.pkt:22: timing error: expected system call return at 0.100596 sec but happened at 0.109564 sec; tolerance 0.004000 sec
>
> https://netdev.bots.linux.dev/contest.html?executor=vmksft-packetdrill-dbg&test=tcp-zerocopy-closed-pkt
Thanks Matthieu. I did not run the dbg variant often enough to observe
that. Note to self to run more times before I submit.
It seems to fail 2/10 times on the dbg spinner. I don't have an
explanation for the failure yet. The line itself has no expected delay
# script packet: 0.113203 S 0:0(0) <mss 1460,sackOK,TS val 0 ecr 0,nop,wscale 8>
# actual packet: 0.107191 S 0:0(0) win 65535 <mss 1460,sackOK,TS val 0 ecr 0,nop,wscale 8>
+0.1 recvmsg(4, {msg_name(...)=...,
msg_iov(1)=[{...,0}],
msg_flags=MSG_ERRQUEUE,
msg_control=[]}, MSG_ERRQUEUE) = -1 EAGAIN (Resource temporarily unavailable)
+0...0 connect(4, ..., ...) = 0
+0 > S 0:0(0) <mss 1460,sackOK,TS val 0 ecr 0,nop,wscale 8>
I guess the expectation includes the +0.1 delay before calling recvmsg, and that
timer fired a bit early.
I previously shared a draft patch to adjust --tolerance_usecs in dbg runs.
May have to send that after all.
https://lore.kernel.org/netdev/66da5b8b27259_27bb41294c@willemb.c.googlers.com.notmuch/
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [PATCH net-next v2 0/3] selftests/net: packetdrill: netns and two imports
2024-09-12 0:52 [PATCH net-next v2 0/3] selftests/net: packetdrill: netns and two imports Willem de Bruijn
` (3 preceding siblings ...)
2024-09-12 11:22 ` [PATCH net-next v2 0/3] selftests/net: packetdrill: netns and two imports Matthieu Baerts
@ 2024-09-13 2:20 ` patchwork-bot+netdevbpf
4 siblings, 0 replies; 9+ messages in thread
From: patchwork-bot+netdevbpf @ 2024-09-13 2:20 UTC (permalink / raw)
To: Willem de Bruijn
Cc: netdev, davem, kuba, edumazet, pabeni, ncardwell, shuah,
linux-kselftest, matttbe, willemb
Hello:
This series was applied to netdev/net-next.git (main)
by Jakub Kicinski <kuba@kernel.org>:
On Wed, 11 Sep 2024 20:52:39 -0400 you wrote:
> From: Willem de Bruijn <willemb@google.com>
>
> 1/3: run in nets, as discussed, and add missing CONFIGs
> 2/3: import tcp/zerocopy
> 3/3: import tcp/slow_start
>
> Willem de Bruijn (3):
> selftests/net: packetdrill: run in netns and expand config
> selftests/net: packetdrill: import tcp/zerocopy
> selftests/net: packetdrill: import tcp/slow_start
>
> [...]
Here is the summary with links:
- [net-next,v2,1/3] selftests/net: packetdrill: run in netns and expand config
https://git.kernel.org/netdev/net-next/c/cded7e0479c9
- [net-next,v2,2/3] selftests/net: packetdrill: import tcp/zerocopy
https://git.kernel.org/netdev/net-next/c/1e42f73fd3c2
- [net-next,v2,3/3] selftests/net: packetdrill: import tcp/slow_start
https://git.kernel.org/netdev/net-next/c/e874be276ee4
You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [PATCH net-next v2 0/3] selftests/net: packetdrill: netns and two imports
2024-09-12 12:10 ` Willem de Bruijn
@ 2024-09-16 16:48 ` Stanislav Fomichev
2024-09-16 20:54 ` Willem de Bruijn
0 siblings, 1 reply; 9+ messages in thread
From: Stanislav Fomichev @ 2024-09-16 16:48 UTC (permalink / raw)
To: Willem de Bruijn
Cc: Matthieu Baerts, netdev, davem, kuba, edumazet, pabeni, ncardwell,
shuah, linux-kselftest, Willem de Bruijn
On 09/12, Willem de Bruijn wrote:
> Matthieu Baerts wrote:
> > Hi Willem,
> >
> > On 12/09/2024 02:52, Willem de Bruijn wrote:
> > > From: Willem de Bruijn <willemb@google.com>
> > >
> > > 1/3: run in nets, as discussed, and add missing CONFIGs
> > > 2/3: import tcp/zerocopy
> > > 3/3: import tcp/slow_start
> >
> > Thank you for the v2. This new version looks good to me:
> >
> > Acked-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
> >
> >
> > I didn't pay too much attention to the new tests, because they look
> > good, heavily tested I suppose, and I guess the goal is not to diverge
> > from the original ones for the moment. Still, please note that the CI
> > reported some timing issues with tcp_zerocopy_closed.pkt when using a
> > debug kernel config, e.g.
> >
> > > tcp_zerocopy_closed.pkt:22: timing error: expected system call return at 0.100596 sec but happened at 0.109564 sec; tolerance 0.004000 sec
> >
> > https://netdev.bots.linux.dev/contest.html?executor=vmksft-packetdrill-dbg&test=tcp-zerocopy-closed-pkt
>
> Thanks Matthieu. I did not run the dbg variant often enough to observe
> that. Note to self to run more times before I submit.
>
> It seems to fail 2/10 times on the dbg spinner. I don't have an
> explanation for the failure yet. The line itself has no expected delay
>
> # script packet: 0.113203 S 0:0(0) <mss 1460,sackOK,TS val 0 ecr 0,nop,wscale 8>
> # actual packet: 0.107191 S 0:0(0) win 65535 <mss 1460,sackOK,TS val 0 ecr 0,nop,wscale 8>
>
> +0.1 recvmsg(4, {msg_name(...)=...,
> msg_iov(1)=[{...,0}],
> msg_flags=MSG_ERRQUEUE,
> msg_control=[]}, MSG_ERRQUEUE) = -1 EAGAIN (Resource temporarily unavailable)
>
> +0...0 connect(4, ..., ...) = 0
>
> +0 > S 0:0(0) <mss 1460,sackOK,TS val 0 ecr 0,nop,wscale 8>
>
> I guess the expectation includes the +0.1 delay before calling recvmsg, and that
> timer fired a bit early.
>
> I previously shared a draft patch to adjust --tolerance_usecs in dbg runs.
> May have to send that after all.
>
> https://lore.kernel.org/netdev/66da5b8b27259_27bb41294c@willemb.c.googlers.com.notmuch/
Not sure you've seen, but tcp_slow_start_slow-start-after-win-update.pkt
also just popped up on the dashboard for dbg:
# tcp_slow_start_slow-start-after-win-update.pkt:39: error handling packet: timing error: expected outbound packet in relative time range +0.600000~+0.620000
https://netdev-3.bots.linux.dev/vmksft-packetdrill-dbg/results/774981/1-tcp-slow-start-slow-start-after-win-update-pkt/stdout
Do we want to follow up with that '--tolerance_usecs=10000' you've
mentioned above?
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [PATCH net-next v2 0/3] selftests/net: packetdrill: netns and two imports
2024-09-16 16:48 ` Stanislav Fomichev
@ 2024-09-16 20:54 ` Willem de Bruijn
0 siblings, 0 replies; 9+ messages in thread
From: Willem de Bruijn @ 2024-09-16 20:54 UTC (permalink / raw)
To: Stanislav Fomichev, Willem de Bruijn
Cc: Matthieu Baerts, netdev, davem, kuba, edumazet, pabeni, ncardwell,
shuah, linux-kselftest, Willem de Bruijn
Stanislav Fomichev wrote:
> On 09/12, Willem de Bruijn wrote:
> > Matthieu Baerts wrote:
> > > Hi Willem,
> > >
> > > On 12/09/2024 02:52, Willem de Bruijn wrote:
> > > > From: Willem de Bruijn <willemb@google.com>
> > > >
> > > > 1/3: run in nets, as discussed, and add missing CONFIGs
> > > > 2/3: import tcp/zerocopy
> > > > 3/3: import tcp/slow_start
> > >
> > > Thank you for the v2. This new version looks good to me:
> > >
> > > Acked-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
> > >
> > >
> > > I didn't pay too much attention to the new tests, because they look
> > > good, heavily tested I suppose, and I guess the goal is not to diverge
> > > from the original ones for the moment. Still, please note that the CI
> > > reported some timing issues with tcp_zerocopy_closed.pkt when using a
> > > debug kernel config, e.g.
> > >
> > > > tcp_zerocopy_closed.pkt:22: timing error: expected system call return at 0.100596 sec but happened at 0.109564 sec; tolerance 0.004000 sec
> > >
> > > https://netdev.bots.linux.dev/contest.html?executor=vmksft-packetdrill-dbg&test=tcp-zerocopy-closed-pkt
> >
> > Thanks Matthieu. I did not run the dbg variant often enough to observe
> > that. Note to self to run more times before I submit.
> >
> > It seems to fail 2/10 times on the dbg spinner. I don't have an
> > explanation for the failure yet. The line itself has no expected delay
> >
> > # script packet: 0.113203 S 0:0(0) <mss 1460,sackOK,TS val 0 ecr 0,nop,wscale 8>
> > # actual packet: 0.107191 S 0:0(0) win 65535 <mss 1460,sackOK,TS val 0 ecr 0,nop,wscale 8>
> >
> > +0.1 recvmsg(4, {msg_name(...)=...,
> > msg_iov(1)=[{...,0}],
> > msg_flags=MSG_ERRQUEUE,
> > msg_control=[]}, MSG_ERRQUEUE) = -1 EAGAIN (Resource temporarily unavailable)
> >
> > +0...0 connect(4, ..., ...) = 0
> >
> > +0 > S 0:0(0) <mss 1460,sackOK,TS val 0 ecr 0,nop,wscale 8>
> >
> > I guess the expectation includes the +0.1 delay before calling recvmsg, and that
> > timer fired a bit early.
> >
> > I previously shared a draft patch to adjust --tolerance_usecs in dbg runs.
> > May have to send that after all.
> >
> > https://lore.kernel.org/netdev/66da5b8b27259_27bb41294c@willemb.c.googlers.com.notmuch/
>
> Not sure you've seen, but tcp_slow_start_slow-start-after-win-update.pkt
> also just popped up on the dashboard for dbg:
>
> # tcp_slow_start_slow-start-after-win-update.pkt:39: error handling packet: timing error: expected outbound packet in relative time range +0.600000~+0.620000
>
> https://netdev-3.bots.linux.dev/vmksft-packetdrill-dbg/results/774981/1-tcp-slow-start-slow-start-after-win-update-pkt/stdout
>
> Do we want to follow up with that '--tolerance_usecs=10000' you've
> mentioned above?
And more tests coming. Looks like it. I'll finish it up. Thanks for
the pointer.
^ permalink raw reply [flat|nested] 9+ messages in thread
end of thread, other threads:[~2024-09-16 20:54 UTC | newest]
Thread overview: 9+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-09-12 0:52 [PATCH net-next v2 0/3] selftests/net: packetdrill: netns and two imports Willem de Bruijn
2024-09-12 0:52 ` [PATCH net-next v2 1/3] selftests/net: packetdrill: run in netns and expand config Willem de Bruijn
2024-09-12 0:52 ` [PATCH net-next v2 2/3] selftests/net: packetdrill: import tcp/zerocopy Willem de Bruijn
2024-09-12 0:52 ` [PATCH net-next v2 3/3] selftests/net: packetdrill: import tcp/slow_start Willem de Bruijn
2024-09-12 11:22 ` [PATCH net-next v2 0/3] selftests/net: packetdrill: netns and two imports Matthieu Baerts
2024-09-12 12:10 ` Willem de Bruijn
2024-09-16 16:48 ` Stanislav Fomichev
2024-09-16 20:54 ` Willem de Bruijn
2024-09-13 2:20 ` patchwork-bot+netdevbpf
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).