This driver currently doesn't support any control flags.

Use flow_rule_match_has_control_flags() to check for control flags,
such as can be set through `tc flower ... ip_flags frag`.

In case any control flags are masked, flow_rule_match_has_control_flags()
sets a NL extended error message, and we return -EOPNOTSUPP.

Only compile-tested.

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
---
 drivers/net/ethernet/ti/cpsw_priv.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/ti/cpsw_priv.c b/drivers/net/ethernet/ti/cpsw_priv.c
index 764ed298b570..6fe4edabba44 100644
--- a/drivers/net/ethernet/ti/cpsw_priv.c
+++ b/drivers/net/ethernet/ti/cpsw_priv.c
@@ -1404,6 +1404,9 @@ static int cpsw_qos_clsflower_add_policer(struct cpsw_priv *priv,
 		return -EOPNOTSUPP;
 	}
 
+	if (flow_rule_match_has_control_flags(rule, extack))
+		return -EOPNOTSUPP;
+
 	if (!flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ETH_ADDRS)) {
 		NL_SET_ERR_MSG_MOD(extack, "Not matching on eth address");
 		return -EOPNOTSUPP;
-- 
2.43.0

Mon, Apr 22, 2024 at 05:26:55PM CEST, ast@fiberby.net wrote:
>This driver currently doesn't support any control flags.
>
>Use flow_rule_match_has_control_flags() to check for control flags,
>such as can be set through `tc flower ... ip_flags frag`.
>
>In case any control flags are masked, flow_rule_match_has_control_flags()
>sets a NL extended error message, and we return -EOPNOTSUPP.
>
>Only compile-tested.
>
>Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>

Reviewed-by: Jiri Pirko <jiri@nvidia.com>

Next time, could you please bundle similar/related patches into
a patchset?
Thanks!

Hi Jiri,

Thank you for the reviews.

On 4/23/24 12:57 PM, Jiri Pirko wrote:
> Next time, could you please bundle similar/related patches into
> a patchset?

Sure, I did it per driver to make it easier to handle differences in
get_maintainer output and be able to integrate feedback into later patches
for other drivers, and since the patches could be applied individually.
When I started I through that more drivers had their own trees, wasn't
aware that it was only Intel, and that mlx5-next was dead.
For instance the feedback from Jianbo on mlx5 patch, also went into the
octeontx2-pf patch.

I only have one patch left in the queue now (for qede).

-- 
Best regards
Asbjørn Sloth Tønnesen
Network Engineer
Fiberby - AS42541