Skip to contentSkip to conversation

selftests: drv-net: gro: increase the rcvbuf size

PATCH Applied Jakub Kicinski 3 participants 3 messages Jan 7 – Jan 10

Subsystems (3)
lore
Share & tools
lore

Jakub Kicinski

author maintainer
#0 lore raw

PATCH 1/1 selftests: drv-net: gro: increase the rcvbuf size

The gro.py test (testing software GRO) is slightly flaky when running against fbnic. We see one flake per roughly 20 runs in NIPA, mostly in ipip.large, and always including some EAGAIN:

# Shouldn't coalesce if exceed IP max pkt size: Test succeeded
# Expected {65475 899 }, Total 2 packets
# Received {65475 899 }, Total 2 packets.
# Expected {64576 900 900 }, Total 3 packets
# Received {64576 /home/virtme/testing/wt-24/tools/testing/selftests/drivers/net/gro: could not receive: Resource temporarily unavailable

The test sends 2 large frames (64k + change). Looks like the default packet socket rcvbuf (~200kB) may not be large enough to hold them. Bump the rcvbuf to 1MB.

Add a debug print showing socket statistics to make debugging this issue easier in the future. Without the rcvbuf increase we see:

# Shouldn't coalesce if exceed IP max pkt size: Test succeeded # Expected {65475 899 }, Total 2 packets
# Received {65475 899 }, Total 2 packets.
# Expected {64576 900 900 }, Total 3 packets
# Received {64576 Socket stats: packets=7, drops=3

                    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

# /home/virtme/testing/wt-24/tools/testing/selftests/drivers/net/gro: could not receive: Resource temporarily unavailable

---
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]

---
 tools/testing/selftests/drivers/net/gro.c | 25 ++++++++++++++++++++++-
 1 file changed, 24 insertions(+), 1 deletion(-)
tools/testing/selftests/drivers/net/gro.c +24 -1
index e894037d2e3e..751a8103f408 100644
@@ -926,6 +926,28 @@ static void set_timeout(int fd)
 		error(1, errno, "cannot set timeout, setsockopt failed");
 }
 
+static void set_rcvbuf(int fd)
+{
+	int bufsize = 1 * 1024 * 1024; /* 1 MB */
+
+	if (setsockopt(fd, SOL_SOCKET, SO_RCVBUF, &bufsize, sizeof(bufsize)))
+		error(1, errno, "cannot set rcvbuf size, setsockopt failed");
+}
+
+static void recv_error(int fd, int rcv_errno)
+{
+	struct tpacket_stats stats;
+	socklen_t len;
+
+	len = sizeof(stats);
+	if (getsockopt(fd, SOL_PACKET, PACKET_STATISTICS, &stats, &len))
+		error(1, errno, "can't get stats");
+
+	fprintf(stderr, "Socket stats: packets=%u, drops=%u\n",
+		stats.tp_packets, stats.tp_drops);
+	error(1, rcv_errno, "could not receive");
+}
+
 static void check_recv_pkts(int fd, int *correct_payload,
 			    int correct_num_pkts)
 {
@@ -950,7 +972,7 @@ static void check_recv_pkts(int fd, int *correct_payload,
 		ip_ext_len = 0;
 		pkt_size = recv(fd, buffer, IP_MAXPACKET + ETH_HLEN + 1, 0);
 		if (pkt_size < 0)
-			error(1, errno, "could not receive");
+			recv_error(fd, errno);
 
 		if (iph->version == 4)
 			ip_ext_len = (iph->ihl - 5) * 4;
@@ -1126,6 +1148,7 @@ static void gro_receiver(void)
 		error(1, 0, "socket creation");
 	setup_sock_filter(rxfd);
 	set_timeout(rxfd);
+	set_rcvbuf(rxfd);
 	bind_packetsocket(rxfd);
 
 	ksft_ready();
Signature · 1 line
2.52.0
Reviewed-by on v1 1/1
#1 lore raw
Quoted 22 lines · Jakub Kicinski

The gro.py test (testing software GRO) is slightly flaky when running against fbnic. We see one flake per roughly 20 runs in NIPA, mostly in ipip.large, and always including some EAGAIN:

# Shouldn't coalesce if exceed IP max pkt size: Test succeeded
# Expected {65475 899 }, Total 2 packets
# Received {65475 899 }, Total 2 packets.
# Expected {64576 900 900 }, Total 3 packets
# Received {64576 /home/virtme/testing/wt-24/tools/testing/selftests/drivers/net/gro: could not receive: Resource temporarily unavailable

The test sends 2 large frames (64k + change). Looks like the default packet socket rcvbuf (~200kB) may not be large enough to hold them. Bump the rcvbuf to 1MB.

Add a debug print showing socket statistics to make debugging this issue easier in the future. Without the rcvbuf increase we see:

# Shouldn't coalesce if exceed IP max pkt size: Test succeeded # Expected {65475 899 }, Total 2 packets
# Received {65475 899 }, Total 2 packets.
# Expected {64576 900 900 }, Total 3 packets
# Received {64576 Socket stats: packets=7, drops=3

Quoted 4 lines
                    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

# /home/virtme/testing/wt-24/tools/testing/selftests/drivers/net/gro: could not receive: Resource temporarily unavailable

#2 lore raw
In reply to Jakub Kicinski
This patch was applied to netdev/net-next.git (main) by Jakub Kicinski <[email protected]>:

Hello:

This patch was applied to netdev/net-next.git (main) by Jakub Kicinski <[email protected]>:

Quoted 11 lines · On Wed, 7 Jan 2026 15:25:57 -0800 you wrote:

The gro.py test (testing software GRO) is slightly flaky when running against fbnic. We see one flake per roughly 20 runs in NIPA, mostly in ipip.large, and always including some EAGAIN:

# Shouldn't coalesce if exceed IP max pkt size: Test succeeded
# Expected {65475 899 }, Total 2 packets
# Received {65475 899 }, Total 2 packets.
# Expected {64576 900 900 }, Total 3 packets
# Received {64576 /home/virtme/testing/wt-24/tools/testing/selftests/drivers/net/gro: could not receive: Resource temporarily unavailable

[...]

Here is the summary with links:
- [net-next] selftests: drv-net: gro: increase the rcvbuf size

    https://git.kernel.org/netdev/net-next/c/a0ac0ff38276

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html