[PATCH net-next v2 1/2] selftests: drv-net: rss_ctx: use Netlink for timed reconfig

[PATCH net-next v2 1/2] selftests: drv-net: rss_ctx: use Netlink for timed reconfig

[Jakub Kicinski]

The rss_ctx test has gotten pretty flaky after I increased
the queue count in NIPA 2->3. Not 100% clear why. We get
a lot of failures in the rss_ctx.test_hitless_key_update case.

Looking closer it appears that the failures are mostly due
to startup costs. I measured the following timing for ethtool -X:
 - python cmd(shell=True)  : 150-250msec
 - python cmd(shell=False) :  50- 70msec
 - timed in bash           :  45- 55msec
 - YNL Netlink call        :   2-  4msec
 - .set_rxfh callback      :   1-  2msec

The target in the test was set to 200msec. We were mostly measuring
ethtool startup cost it seems. Switch to YNL since it's 100x faster.

Lower the pass criteria to 150msec, no real science behind this number
but we removed some overhead, drivers which previously passed 200msec
should easily pass 150msec now.

Separately we should probably follow up on defaulting to shell=False,
when script doesn't explicitly ask for True, because the overhead
is rather significant.

Switch from _rss_key_rand() to random.randbytes(), YNL takes a binary
array rather than array of ints.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
v2:
 - increase the threshold to safer 150msec
 - mention change away from _rss_key_rand()
v1: https://lore.kernel.org/20250829220712.327920-1-kuba@kernel.org
---
 tools/testing/selftests/drivers/net/hw/rss_ctx.py | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/drivers/net/hw/rss_ctx.py b/tools/testing/selftests/drivers/net/hw/rss_ctx.py
index 9838b8457e5a..a5562a9f729f 100755
--- a/tools/testing/selftests/drivers/net/hw/rss_ctx.py
+++ b/tools/testing/selftests/drivers/net/hw/rss_ctx.py
@@ -335,19 +335,20 @@ from lib.py import ethtool, ip, defer, GenerateTraffic, CmdExitFailure
     data = get_rss(cfg)
     key_len = len(data['rss-hash-key'])
 
-    key = _rss_key_rand(key_len)
+    ethnl = EthtoolFamily()
+    key = random.randbytes(key_len)
 
     tgen = GenerateTraffic(cfg)
     try:
         errors0, carrier0 = get_drop_err_sum(cfg)
         t0 = datetime.datetime.now()
-        ethtool(f"-X {cfg.ifname} hkey " + _rss_key_str(key))
+        ethnl.rss_set({"header": {"dev-index": cfg.ifindex}, "hkey": key})
         t1 = datetime.datetime.now()
         errors1, carrier1 = get_drop_err_sum(cfg)
     finally:
         tgen.wait_pkts_and_stop(5000)
 
-    ksft_lt((t1 - t0).total_seconds(), 0.2)
+    ksft_lt((t1 - t0).total_seconds(), 0.15)
     ksft_eq(errors1 - errors1, 0)
     ksft_eq(carrier1 - carrier0, 0)
 
-- 
2.51.0

[PATCH net-next v2 2/2] selftests: drv-net: rss_ctx: make the test pass with few queues

[Jakub Kicinski]

rss_ctx.test_rss_key_indir implicitly expects at least 5 queues,
as it checks that the traffic on first 2 queues is lower than
the remaining queues when we use all queues. Special case fewer
queues.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/drivers/net/hw/rss_ctx.py | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/drivers/net/hw/rss_ctx.py b/tools/testing/selftests/drivers/net/hw/rss_ctx.py
index a5562a9f729f..ed7e405682f0 100755
--- a/tools/testing/selftests/drivers/net/hw/rss_ctx.py
+++ b/tools/testing/selftests/drivers/net/hw/rss_ctx.py
@@ -178,8 +178,13 @@ from lib.py import ethtool, ip, defer, GenerateTraffic, CmdExitFailure
     cnts = _get_rx_cnts(cfg)
     GenerateTraffic(cfg).wait_pkts_and_stop(20000)
     cnts = _get_rx_cnts(cfg, prev=cnts)
-    # First two queues get less traffic than all the rest
-    ksft_lt(sum(cnts[:2]), sum(cnts[2:]), "traffic distributed: " + str(cnts))
+    if qcnt > 4:
+        # First two queues get less traffic than all the rest
+        ksft_lt(sum(cnts[:2]), sum(cnts[2:]),
+                "traffic distributed: " + str(cnts))
+    else:
+        # When queue count is low make sure third queue got significant pkts
+        ksft_ge(cnts[2], 3500, "traffic distributed: " + str(cnts))
 
 
 def test_rss_queue_reconfigure(cfg, main_ctx=True):
-- 
2.51.0

Re: [PATCH net-next v2 2/2] selftests: drv-net: rss_ctx: make the test pass with few queues

[Simon Horman]

On Mon, Sep 01, 2025 at 10:31:39AM -0700, Jakub Kicinski wrote:
> rss_ctx.test_rss_key_indir implicitly expects at least 5 queues,
> as it checks that the traffic on first 2 queues is lower than
> the remaining queues when we use all queues. Special case fewer
> queues.
> 
> Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Reviewed-by: Simon Horman <horms@kernel.org>

Re: [PATCH net-next v2 1/2] selftests: drv-net: rss_ctx: use Netlink for timed reconfig

[Simon Horman]

On Mon, Sep 01, 2025 at 10:31:38AM -0700, Jakub Kicinski wrote:
> The rss_ctx test has gotten pretty flaky after I increased
> the queue count in NIPA 2->3. Not 100% clear why. We get
> a lot of failures in the rss_ctx.test_hitless_key_update case.
> 
> Looking closer it appears that the failures are mostly due
> to startup costs. I measured the following timing for ethtool -X:
>  - python cmd(shell=True)  : 150-250msec
>  - python cmd(shell=False) :  50- 70msec
>  - timed in bash           :  45- 55msec
>  - YNL Netlink call        :   2-  4msec
>  - .set_rxfh callback      :   1-  2msec
> 
> The target in the test was set to 200msec. We were mostly measuring
> ethtool startup cost it seems. Switch to YNL since it's 100x faster.
> 
> Lower the pass criteria to 150msec, no real science behind this number
> but we removed some overhead, drivers which previously passed 200msec
> should easily pass 150msec now.
> 
> Separately we should probably follow up on defaulting to shell=False,
> when script doesn't explicitly ask for True, because the overhead
> is rather significant.
> 
> Switch from _rss_key_rand() to random.randbytes(), YNL takes a binary
> array rather than array of ints.
> 
> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
> ---
> v2:
>  - increase the threshold to safer 150msec
>  - mention change away from _rss_key_rand()
> v1: https://lore.kernel.org/20250829220712.327920-1-kuba@kernel.org

Reviewed-by: Simon Horman <horms@kernel.org>

Re: [PATCH net-next v2 1/2] selftests: drv-net: rss_ctx: use Netlink for timed reconfig

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net-next.git (main)
by Jakub Kicinski <kuba@kernel.org>:

On Mon,  1 Sep 2025 10:31:38 -0700 you wrote:
> The rss_ctx test has gotten pretty flaky after I increased
> the queue count in NIPA 2->3. Not 100% clear why. We get
> a lot of failures in the rss_ctx.test_hitless_key_update case.
> 
> Looking closer it appears that the failures are mostly due
> to startup costs. I measured the following timing for ethtool -X:
>  - python cmd(shell=True)  : 150-250msec
>  - python cmd(shell=False) :  50- 70msec
>  - timed in bash           :  45- 55msec
>  - YNL Netlink call        :   2-  4msec
>  - .set_rxfh callback      :   1-  2msec
> 
> [...]

Here is the summary with links:
  - [net-next,v2,1/2] selftests: drv-net: rss_ctx: use Netlink for timed reconfig
    https://git.kernel.org/netdev/net-next/c/4022f92a2e4e
  - [net-next,v2,2/2] selftests: drv-net: rss_ctx: make the test pass with few queues
    https://git.kernel.org/netdev/net-next/c/e2cf2d5baa09

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html