Skip to contentSkip to conversation

selftests: xsk: fix rx_full stats test

PATCH Applied Magnus Karlsson 3 participants 3 messages Jan 21, 2022 – Jan 24, 2022

Subsystems (6)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 selftests: xsk: fix rx_full stats test

From: Magnus Karlsson <[email protected]>

Fix the rx_full stats test so that it correctly reports pass even when the fill ring is not full of buffers.

Fixes 872a1184dbf2 ("selftests: xsk: Put the same buffer only once in the fill ring")
---
 tools/testing/selftests/bpf/xdpxceiver.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
tools/testing/selftests/bpf/xdpxceiver.c +4 -1
index 0a5d23da486d..ffa5502ad95e 100644
@@ -906,7 +906,10 @@ static bool rx_stats_are_valid(struct ifobject *ifobject)
 			return true;
 		case STAT_TEST_RX_FULL:
 			xsk_stat = stats.rx_ring_full;
-			expected_stat -= RX_FULL_RXQSIZE;
+			if (ifobject->umem->num_frames < XSK_RING_PROD__DEFAULT_NUM_DESCS)
+				expected_stat = ifobject->umem->num_frames - RX_FULL_RXQSIZE;
+			else
+				expected_stat = XSK_RING_PROD__DEFAULT_NUM_DESCS - RX_FULL_RXQSIZE;
 			break;
 		case STAT_TEST_RX_FILL_EMPTY:
 			xsk_stat = stats.rx_fill_ring_empty_descs;

base-commit: 820e6e227c4053b6b631ae65ef1f65d560cb392b

Signature · 1 line
2.34.1
Tested-by Acked-by on v1 1/1
#1 lore raw
Quoted 7 lines · Magnus Karlsson

From: Magnus Karlsson <[email protected]>

Fix the rx_full stats test so that it correctly reports pass even when the fill ring is not full of buffers.

Fixes 872a1184dbf2 ("selftests: xsk: Put the same buffer only once in the fill ring")

Fixes xdpxceiver hang on my side.

Quoted 25 lines
---
 tools/testing/selftests/bpf/xdpxceiver.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
tools/testing/selftests/bpf/xdpxceiver.c +4 -1
index 0a5d23da486d..ffa5502ad95e 100644
@@ -906,7 +906,10 @@ static bool rx_stats_are_valid(struct ifobject *ifobject)
 			return true;
 		case STAT_TEST_RX_FULL:
 			xsk_stat = stats.rx_ring_full;
-			expected_stat -= RX_FULL_RXQSIZE;
+			if (ifobject->umem->num_frames < XSK_RING_PROD__DEFAULT_NUM_DESCS)
+				expected_stat = ifobject->umem->num_frames - RX_FULL_RXQSIZE;
+			else
+				expected_stat = XSK_RING_PROD__DEFAULT_NUM_DESCS - RX_FULL_RXQSIZE;
 			break;
 		case STAT_TEST_RX_FILL_EMPTY:
 			xsk_stat = stats.rx_fill_ring_empty_descs;

base-commit: 820e6e227c4053b6b631ae65ef1f65d560cb392b

Signature · 1 line
2.34.1
#2 lore raw
In reply to Magnus Karlsson
This patch was applied to bpf/bpf-next.git (master) by Daniel Borkmann <[email protected]>:

Hello:

This patch was applied to bpf/bpf-next.git (master) by Daniel Borkmann <[email protected]>:

Quoted 9 lines · On Fri, 21 Jan 2022 13:35:08 +0100 you wrote:

From: Magnus Karlsson <[email protected]>

Fix the rx_full stats test so that it correctly reports pass even when the fill ring is not full of buffers.

Fixes 872a1184dbf2 ("selftests: xsk: Put the same buffer only once in the fill ring")

[...]

Here is the summary with links:
- [bpf-next] selftests: xsk: fix rx_full stats test

    https://git.kernel.org/bpf/bpf-next/c/b4ec6a192312

You are awesome, thank you!

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