All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH net 0/2] nfp: fixes for v5.19
@ 2022-06-08  9:28 Simon Horman
  2022-06-08  9:29 ` [PATCH net 1/2] nfp: avoid unnecessary check warnings in nfp_app_get_vf_config Simon Horman
                   ` (2 more replies)
  0 siblings, 3 replies; 4+ messages in thread
From: Simon Horman @ 2022-06-08  9:28 UTC (permalink / raw)
  To: David Miller, Jakub Kicinski, Paolo Abeni
  Cc: netdev, oss-drivers, Fei Qin, Etienne van der Linde

Hi,

this short series includes two fixes for the NFP driver.

1. Restructure GRE+VLAN flower offload to address a miss match
   between the NIC firmware and driver implementation which
   prevented these features from working in combination.

2. Prevent unnecessary warnings regarding rate limiting support.-
   It is expected that this feature to not _always_ be present
   but this was not taken into account when the code to check
   for this feature was added.

Etienne van der Linde (1):
  nfp: flower: restructure flow-key for gre+vlan combination

Fei Qin (1):
  nfp: avoid unnecessary check warnings in nfp_app_get_vf_config

 .../ethernet/netronome/nfp/flower/conntrack.c | 32 +++++++++----------
 .../net/ethernet/netronome/nfp/flower/match.c | 16 +++++-----
 .../ethernet/netronome/nfp/nfp_net_sriov.c    | 28 ++++++++--------
 3 files changed, 39 insertions(+), 37 deletions(-)

-- 
2.30.2


^ permalink raw reply	[flat|nested] 4+ messages in thread
* [PATCH net 1/2] nfp: avoid unnecessary check warnings in nfp_app_get_vf_config
  2022-06-08  9:28 [PATCH net 0/2] nfp: fixes for v5.19 Simon Horman
@ 2022-06-08  9:29 ` Simon Horman
  2022-06-08  9:29 ` [PATCH net 2/2] nfp: flower: restructure flow-key for gre+vlan combination Simon Horman
  2022-06-10  5:20 ` [PATCH net 0/2] nfp: fixes for v5.19 patchwork-bot+netdevbpf
  2 siblings, 0 replies; 4+ messages in thread
From: Simon Horman @ 2022-06-08  9:29 UTC (permalink / raw)
  To: David Miller, Jakub Kicinski, Paolo Abeni
  Cc: netdev, oss-drivers, Fei Qin, Etienne van der Linde

From: Fei Qin <fei.qin@corigine.com>

nfp_net_sriov_check is added in nfp_app_get_vf_config which intends
to ensure ivi->vlan_proto and ivi->max_tx_rate/min_tx_rate can be
read from VF config table only when firmware supports corresponding
capability.

However, "nfp_app_get_vf_config" can be called by commands like
"ip a", "ip link set $DEV up" and "ip link set $DEV vf $NUM vlan
$param" (with VF). When using commands above, many warnings
"ndo_set_vf_<cap_x> not supported" would appear if firmware doesn't
support VF rate limit and 802.1ad VLAN assingment. If more VFs are
created, things could get worse.

Thus, this patch add an extra bool parameter for nfp_net_sriov_check
to enable/disable the cap check warning report. Unnecessary warnings
in nfp_app_get_vf_config can be avoided. Valid warnings in kinds of
vf setting function can be reserved.

Fixes: e0d0e1fdf1ed ("nfp: VF rate limit support")
Fixes: 59359597b010 ("nfp: support 802.1ad VLAN assingment to VF")
Signed-off-by: Fei Qin <fei.qin@corigine.com>
Signed-off-by: Yinjun Zhang <yinjun.zhang@corigine.com>
Signed-off-by: Louis Peens <louis.peens@corigine.com>
Signed-off-by: Simon Horman <simon.horman@corigine.com>
---
 .../ethernet/netronome/nfp/nfp_net_sriov.c    | 28 ++++++++++---------
 1 file changed, 15 insertions(+), 13 deletions(-)

diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_sriov.c b/drivers/net/ethernet/netronome/nfp/nfp_net_sriov.c
index 54af30961351..6eeeb0fda91f 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_sriov.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_sriov.c
@@ -15,7 +15,7 @@
 #include "nfp_net_sriov.h"
 
 static int
-nfp_net_sriov_check(struct nfp_app *app, int vf, u16 cap, const char *msg)
+nfp_net_sriov_check(struct nfp_app *app, int vf, u16 cap, const char *msg, bool warn)
 {
 	u16 cap_vf;
 
@@ -24,12 +24,14 @@ nfp_net_sriov_check(struct nfp_app *app, int vf, u16 cap, const char *msg)
 
 	cap_vf = readw(app->pf->vfcfg_tbl2 + NFP_NET_VF_CFG_MB_CAP);
 	if ((cap_vf & cap) != cap) {
-		nfp_warn(app->pf->cpp, "ndo_set_vf_%s not supported\n", msg);
+		if (warn)
+			nfp_warn(app->pf->cpp, "ndo_set_vf_%s not supported\n", msg);
 		return -EOPNOTSUPP;
 	}
 
 	if (vf < 0 || vf >= app->pf->num_vfs) {
-		nfp_warn(app->pf->cpp, "invalid VF id %d\n", vf);
+		if (warn)
+			nfp_warn(app->pf->cpp, "invalid VF id %d\n", vf);
 		return -EINVAL;
 	}
 
@@ -65,7 +67,7 @@ int nfp_app_set_vf_mac(struct net_device *netdev, int vf, u8 *mac)
 	unsigned int vf_offset;
 	int err;
 
-	err = nfp_net_sriov_check(app, vf, NFP_NET_VF_CFG_MB_CAP_MAC, "mac");
+	err = nfp_net_sriov_check(app, vf, NFP_NET_VF_CFG_MB_CAP_MAC, "mac", true);
 	if (err)
 		return err;
 
@@ -101,7 +103,7 @@ int nfp_app_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, u8 qos,
 	u32 vlan_tag;
 	int err;
 
-	err = nfp_net_sriov_check(app, vf, NFP_NET_VF_CFG_MB_CAP_VLAN, "vlan");
+	err = nfp_net_sriov_check(app, vf, NFP_NET_VF_CFG_MB_CAP_VLAN, "vlan", true);
 	if (err)
 		return err;
 
@@ -115,7 +117,7 @@ int nfp_app_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, u8 qos,
 	}
 
 	/* Check if fw supports or not */
-	err = nfp_net_sriov_check(app, vf, NFP_NET_VF_CFG_MB_CAP_VLAN_PROTO, "vlan_proto");
+	err = nfp_net_sriov_check(app, vf, NFP_NET_VF_CFG_MB_CAP_VLAN_PROTO, "vlan_proto", true);
 	if (err)
 		is_proto_sup = false;
 
@@ -149,7 +151,7 @@ int nfp_app_set_vf_rate(struct net_device *netdev, int vf,
 	u32 vf_offset, ratevalue;
 	int err;
 
-	err = nfp_net_sriov_check(app, vf, NFP_NET_VF_CFG_MB_CAP_RATE, "rate");
+	err = nfp_net_sriov_check(app, vf, NFP_NET_VF_CFG_MB_CAP_RATE, "rate", true);
 	if (err)
 		return err;
 
@@ -181,7 +183,7 @@ int nfp_app_set_vf_spoofchk(struct net_device *netdev, int vf, bool enable)
 	int err;
 
 	err = nfp_net_sriov_check(app, vf, NFP_NET_VF_CFG_MB_CAP_SPOOF,
-				  "spoofchk");
+				  "spoofchk", true);
 	if (err)
 		return err;
 
@@ -205,7 +207,7 @@ int nfp_app_set_vf_trust(struct net_device *netdev, int vf, bool enable)
 	int err;
 
 	err = nfp_net_sriov_check(app, vf, NFP_NET_VF_CFG_MB_CAP_TRUST,
-				  "trust");
+				  "trust", true);
 	if (err)
 		return err;
 
@@ -230,7 +232,7 @@ int nfp_app_set_vf_link_state(struct net_device *netdev, int vf,
 	int err;
 
 	err = nfp_net_sriov_check(app, vf, NFP_NET_VF_CFG_MB_CAP_LINK_STATE,
-				  "link_state");
+				  "link_state", true);
 	if (err)
 		return err;
 
@@ -265,7 +267,7 @@ int nfp_app_get_vf_config(struct net_device *netdev, int vf,
 	u8 flags;
 	int err;
 
-	err = nfp_net_sriov_check(app, vf, 0, "");
+	err = nfp_net_sriov_check(app, vf, 0, "", true);
 	if (err)
 		return err;
 
@@ -285,13 +287,13 @@ int nfp_app_get_vf_config(struct net_device *netdev, int vf,
 
 	ivi->vlan = FIELD_GET(NFP_NET_VF_CFG_VLAN_VID, vlan_tag);
 	ivi->qos = FIELD_GET(NFP_NET_VF_CFG_VLAN_QOS, vlan_tag);
-	if (!nfp_net_sriov_check(app, vf, NFP_NET_VF_CFG_MB_CAP_VLAN_PROTO, "vlan_proto"))
+	if (!nfp_net_sriov_check(app, vf, NFP_NET_VF_CFG_MB_CAP_VLAN_PROTO, "vlan_proto", false))
 		ivi->vlan_proto = htons(FIELD_GET(NFP_NET_VF_CFG_VLAN_PROT, vlan_tag));
 	ivi->spoofchk = FIELD_GET(NFP_NET_VF_CFG_CTRL_SPOOF, flags);
 	ivi->trusted = FIELD_GET(NFP_NET_VF_CFG_CTRL_TRUST, flags);
 	ivi->linkstate = FIELD_GET(NFP_NET_VF_CFG_CTRL_LINK_STATE, flags);
 
-	err = nfp_net_sriov_check(app, vf, NFP_NET_VF_CFG_MB_CAP_RATE, "rate");
+	err = nfp_net_sriov_check(app, vf, NFP_NET_VF_CFG_MB_CAP_RATE, "rate", false);
 	if (!err) {
 		rate = readl(app->pf->vfcfg_tbl2 + vf_offset +
 			     NFP_NET_VF_CFG_RATE);
-- 
2.30.2


^ permalink raw reply related	[flat|nested] 4+ messages in thread
* [PATCH net 2/2] nfp: flower: restructure flow-key for gre+vlan combination
  2022-06-08  9:28 [PATCH net 0/2] nfp: fixes for v5.19 Simon Horman
  2022-06-08  9:29 ` [PATCH net 1/2] nfp: avoid unnecessary check warnings in nfp_app_get_vf_config Simon Horman
@ 2022-06-08  9:29 ` Simon Horman
  2022-06-10  5:20 ` [PATCH net 0/2] nfp: fixes for v5.19 patchwork-bot+netdevbpf
  2 siblings, 0 replies; 4+ messages in thread
From: Simon Horman @ 2022-06-08  9:29 UTC (permalink / raw)
  To: David Miller, Jakub Kicinski, Paolo Abeni
  Cc: netdev, oss-drivers, Fei Qin, Etienne van der Linde

From: Etienne van der Linde <etienne.vanderlinde@corigine.com>

Swap around the GRE and VLAN parts in the flow-key offloaded by
the driver to fit in with other tunnel types and the firmware.
Without this change used cases with GRE+VLAN on the outer header
does not get offloaded as the flow-key mismatches what the
firmware expect.

Fixes: 0d630f58989a ("nfp: flower: add support to offload QinQ match")
Fixes: 5a2b93041646 ("nfp: flower-ct: compile match sections of flow_payload")
Signed-off-by: Etienne van der Linde <etienne.vanderlinde@corigine.com>
Signed-off-by: Louis Peens <louis.peens@corigine.com>
Signed-off-by: Yinjun Zhang <yinjun.zhang@corigine.com>
Signed-off-by: Simon Horman <simon.horman@corigine.com>
---
 .../ethernet/netronome/nfp/flower/conntrack.c | 32 +++++++++----------
 .../net/ethernet/netronome/nfp/flower/match.c | 16 +++++-----
 2 files changed, 24 insertions(+), 24 deletions(-)

diff --git a/drivers/net/ethernet/netronome/nfp/flower/conntrack.c b/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
index 443a5d6eb57b..7c31a46195b2 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/conntrack.c
@@ -507,6 +507,11 @@ nfp_fl_calc_key_layers_sz(struct nfp_fl_key_ls in_key_ls, uint16_t *map)
 		key_size += sizeof(struct nfp_flower_ipv6);
 	}
 
+	if (in_key_ls.key_layer_two & NFP_FLOWER_LAYER2_QINQ) {
+		map[FLOW_PAY_QINQ] = key_size;
+		key_size += sizeof(struct nfp_flower_vlan);
+	}
+
 	if (in_key_ls.key_layer_two & NFP_FLOWER_LAYER2_GRE) {
 		map[FLOW_PAY_GRE] = key_size;
 		if (in_key_ls.key_layer_two & NFP_FLOWER_LAYER2_TUN_IPV6)
@@ -515,11 +520,6 @@ nfp_fl_calc_key_layers_sz(struct nfp_fl_key_ls in_key_ls, uint16_t *map)
 			key_size += sizeof(struct nfp_flower_ipv4_gre_tun);
 	}
 
-	if (in_key_ls.key_layer_two & NFP_FLOWER_LAYER2_QINQ) {
-		map[FLOW_PAY_QINQ] = key_size;
-		key_size += sizeof(struct nfp_flower_vlan);
-	}
-
 	if ((in_key_ls.key_layer & NFP_FLOWER_LAYER_VXLAN) ||
 	    (in_key_ls.key_layer_two & NFP_FLOWER_LAYER2_GENEVE)) {
 		map[FLOW_PAY_UDP_TUN] = key_size;
@@ -758,6 +758,17 @@ static int nfp_fl_ct_add_offload(struct nfp_fl_nft_tc_merge *m_entry)
 		}
 	}
 
+	if (NFP_FLOWER_LAYER2_QINQ & key_layer.key_layer_two) {
+		offset = key_map[FLOW_PAY_QINQ];
+		key = kdata + offset;
+		msk = mdata + offset;
+		for (i = 0; i < _CT_TYPE_MAX; i++) {
+			nfp_flower_compile_vlan((struct nfp_flower_vlan *)key,
+						(struct nfp_flower_vlan *)msk,
+						rules[i]);
+		}
+	}
+
 	if (key_layer.key_layer_two & NFP_FLOWER_LAYER2_GRE) {
 		offset = key_map[FLOW_PAY_GRE];
 		key = kdata + offset;
@@ -798,17 +809,6 @@ static int nfp_fl_ct_add_offload(struct nfp_fl_nft_tc_merge *m_entry)
 		}
 	}
 
-	if (NFP_FLOWER_LAYER2_QINQ & key_layer.key_layer_two) {
-		offset = key_map[FLOW_PAY_QINQ];
-		key = kdata + offset;
-		msk = mdata + offset;
-		for (i = 0; i < _CT_TYPE_MAX; i++) {
-			nfp_flower_compile_vlan((struct nfp_flower_vlan *)key,
-						(struct nfp_flower_vlan *)msk,
-						rules[i]);
-		}
-	}
-
 	if (key_layer.key_layer & NFP_FLOWER_LAYER_VXLAN ||
 	    key_layer.key_layer_two & NFP_FLOWER_LAYER2_GENEVE) {
 		offset = key_map[FLOW_PAY_UDP_TUN];
diff --git a/drivers/net/ethernet/netronome/nfp/flower/match.c b/drivers/net/ethernet/netronome/nfp/flower/match.c
index 193a167a6762..e01430139b6d 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/match.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/match.c
@@ -625,6 +625,14 @@ int nfp_flower_compile_flow_match(struct nfp_app *app,
 		msk += sizeof(struct nfp_flower_ipv6);
 	}
 
+	if (NFP_FLOWER_LAYER2_QINQ & key_ls->key_layer_two) {
+		nfp_flower_compile_vlan((struct nfp_flower_vlan *)ext,
+					(struct nfp_flower_vlan *)msk,
+					rule);
+		ext += sizeof(struct nfp_flower_vlan);
+		msk += sizeof(struct nfp_flower_vlan);
+	}
+
 	if (key_ls->key_layer_two & NFP_FLOWER_LAYER2_GRE) {
 		if (key_ls->key_layer_two & NFP_FLOWER_LAYER2_TUN_IPV6) {
 			struct nfp_flower_ipv6_gre_tun *gre_match;
@@ -660,14 +668,6 @@ int nfp_flower_compile_flow_match(struct nfp_app *app,
 		}
 	}
 
-	if (NFP_FLOWER_LAYER2_QINQ & key_ls->key_layer_two) {
-		nfp_flower_compile_vlan((struct nfp_flower_vlan *)ext,
-					(struct nfp_flower_vlan *)msk,
-					rule);
-		ext += sizeof(struct nfp_flower_vlan);
-		msk += sizeof(struct nfp_flower_vlan);
-	}
-
 	if (key_ls->key_layer & NFP_FLOWER_LAYER_VXLAN ||
 	    key_ls->key_layer_two & NFP_FLOWER_LAYER2_GENEVE) {
 		if (key_ls->key_layer_two & NFP_FLOWER_LAYER2_TUN_IPV6) {
-- 
2.30.2


^ permalink raw reply related	[flat|nested] 4+ messages in thread
* Re: [PATCH net 0/2] nfp: fixes for v5.19
  2022-06-08  9:28 [PATCH net 0/2] nfp: fixes for v5.19 Simon Horman
  2022-06-08  9:29 ` [PATCH net 1/2] nfp: avoid unnecessary check warnings in nfp_app_get_vf_config Simon Horman
  2022-06-08  9:29 ` [PATCH net 2/2] nfp: flower: restructure flow-key for gre+vlan combination Simon Horman
@ 2022-06-10  5:20 ` patchwork-bot+netdevbpf
  2 siblings, 0 replies; 4+ messages in thread
From: patchwork-bot+netdevbpf @ 2022-06-10  5:20 UTC (permalink / raw)
  To: Simon Horman
  Cc: davem, kuba, pabeni, netdev, oss-drivers, fei.qin, etienne.vanderlinde

Hello:

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

On Wed,  8 Jun 2022 11:28:59 +0200 you wrote:
> Hi,
> 
> this short series includes two fixes for the NFP driver.
> 
> 1. Restructure GRE+VLAN flower offload to address a miss match
>    between the NIC firmware and driver implementation which
>    prevented these features from working in combination.
> 
> [...]

Here is the summary with links:
  - [net,1/2] nfp: avoid unnecessary check warnings in nfp_app_get_vf_config
    https://git.kernel.org/netdev/net/c/03d5005ff735
  - [net,2/2] nfp: flower: restructure flow-key for gre+vlan combination
    https://git.kernel.org/netdev/net/c/a0b843340dae

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] 4+ messages in thread
end of thread, other threads:[~2022-06-10  5:20 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-06-08  9:28 [PATCH net 0/2] nfp: fixes for v5.19 Simon Horman
2022-06-08  9:29 ` [PATCH net 1/2] nfp: avoid unnecessary check warnings in nfp_app_get_vf_config Simon Horman
2022-06-08  9:29 ` [PATCH net 2/2] nfp: flower: restructure flow-key for gre+vlan combination Simon Horman
2022-06-10  5:20 ` [PATCH net 0/2] nfp: fixes for v5.19 patchwork-bot+netdevbpf

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.