From: Naveen Mamindlapalli <naveenm@marvell.com>
To: <davem@davemloft.net>, <edumazet@google.com>, <kuba@kernel.org>,
<pabeni@redhat.com>, <netdev@vger.kernel.org>,
<linux-kernel@vger.kernel.org>, <sgoutham@marvell.com>
Cc: Kiran Kumar K <kirankumark@marvell.com>,
Geetha sowjanya <gakula@marvell.com>,
Naveen Mamindlapalli <naveenm@marvell.com>
Subject: [net-next PATCH v2 2/6] octeontx2-af: extend RSS supported offload types
Date: Mon, 12 Jun 2023 11:34:20 +0530 [thread overview]
Message-ID: <20230612060424.1427-3-naveenm@marvell.com> (raw)
In-Reply-To: <20230612060424.1427-1-naveenm@marvell.com>
From: Kiran Kumar K <kirankumark@marvell.com>
Add support to select L3 SRC or DST only, L4 SRC or DST only for RSS
calculation.
AF consumer may have requirement as we can select only SRC or DST data for
RSS calculation in L3, L4 layers. With this requirement there will be
following combinations, IPV[4,6]_SRC_ONLY, IPV[4,6]_DST_ONLY,
[TCP,UDP,SCTP]_SRC_ONLY, [TCP,UDP,SCTP]_DST_ONLY. So, instead of creating
a bit for each combination, we are using upper 4 bits (31:28) in the
flow_key_cfg to represent the SRC, DST selection. 31 => L3_SRC,
30 => L3_DST, 29 => L4_SRC, 28 => L4_DST. These won't be part of flow_cfg,
so that we don't need to change the existing ABI.
Signed-off-by: Kiran Kumar K <kirankumark@marvell.com>
Signed-off-by: Geetha sowjanya <gakula@marvell.com>
Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
---
.../net/ethernet/marvell/octeontx2/af/mbox.h | 6 ++
.../ethernet/marvell/octeontx2/af/rvu_nix.c | 57 +++++++++++++++++++
2 files changed, 63 insertions(+)
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
index 6389ed83637d..671fcf86ed87 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
@@ -1080,6 +1080,8 @@ struct nix_vtag_config_rsp {
*/
};
+#define NIX_FLOW_KEY_TYPE_L3_L4_MASK (~(0xf << 28))
+
struct nix_rss_flowkey_cfg {
struct mbox_msghdr hdr;
int mcam_index; /* MCAM entry index to modify */
@@ -1105,6 +1107,10 @@ struct nix_rss_flowkey_cfg {
#define NIX_FLOW_KEY_TYPE_IPV4_PROTO BIT(21)
#define NIX_FLOW_KEY_TYPE_AH BIT(22)
#define NIX_FLOW_KEY_TYPE_ESP BIT(23)
+#define NIX_FLOW_KEY_TYPE_L4_DST_ONLY BIT(28)
+#define NIX_FLOW_KEY_TYPE_L4_SRC_ONLY BIT(29)
+#define NIX_FLOW_KEY_TYPE_L3_DST_ONLY BIT(30)
+#define NIX_FLOW_KEY_TYPE_L3_SRC_ONLY BIT(31)
u32 flowkey_cfg; /* Flowkey types selected */
u8 group; /* RSS context or group */
};
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
index 79ed7af0b0a4..ee52b86c061f 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
@@ -3353,6 +3353,7 @@ static int set_flowkey_fields(struct nix_rx_flowkey_alg *alg, u32 flow_cfg)
struct nix_rx_flowkey_alg *field;
struct nix_rx_flowkey_alg tmp;
u32 key_type, valid_key;
+ u32 l3_l4_src_dst;
int l4_key_offset = 0;
if (!alg)
@@ -3380,6 +3381,15 @@ static int set_flowkey_fields(struct nix_rx_flowkey_alg *alg, u32 flow_cfg)
* group_member - Enabled when protocol is part of a group.
*/
+ /* Last 4 bits (31:28) are reserved to specify SRC, DST
+ * selection for L3, L4 i.e IPV[4,6]_SRC, IPV[4,6]_DST,
+ * [TCP,UDP,SCTP]_SRC, [TCP,UDP,SCTP]_DST
+ * 31 => L3_SRC, 30 => L3_DST, 29 => L4_SRC, 28 => L4_DST
+ */
+ l3_l4_src_dst = flow_cfg;
+ /* Reset these 4 bits, so that these won't be part of key */
+ flow_cfg &= NIX_FLOW_KEY_TYPE_L3_L4_MASK;
+
keyoff_marker = 0; max_key_off = 0; group_member = 0;
nr_field = 0; key_off = 0; field_marker = 1;
field = &tmp; max_bit_pos = fls(flow_cfg);
@@ -3417,6 +3427,22 @@ static int set_flowkey_fields(struct nix_rx_flowkey_alg *alg, u32 flow_cfg)
}
field->hdr_offset = 12; /* SIP offset */
field->bytesm1 = 7; /* SIP + DIP, 8 bytes */
+
+ /* Only SIP */
+ if (l3_l4_src_dst & NIX_FLOW_KEY_TYPE_L3_SRC_ONLY)
+ field->bytesm1 = 3; /* SIP, 4 bytes */
+
+ if (l3_l4_src_dst & NIX_FLOW_KEY_TYPE_L3_DST_ONLY) {
+ /* Both SIP + DIP */
+ if (field->bytesm1 == 3) {
+ field->bytesm1 = 7; /* SIP + DIP, 8B */
+ } else {
+ /* Only DIP */
+ field->hdr_offset = 16; /* DIP off */
+ field->bytesm1 = 3; /* DIP, 4 bytes */
+ }
+ }
+
field->ltype_mask = 0xF; /* Match only IPv4 */
keyoff_marker = false;
break;
@@ -3430,6 +3456,22 @@ static int set_flowkey_fields(struct nix_rx_flowkey_alg *alg, u32 flow_cfg)
}
field->hdr_offset = 8; /* SIP offset */
field->bytesm1 = 31; /* SIP + DIP, 32 bytes */
+
+ /* Only SIP */
+ if (l3_l4_src_dst & NIX_FLOW_KEY_TYPE_L3_SRC_ONLY)
+ field->bytesm1 = 15; /* SIP, 16 bytes */
+
+ if (l3_l4_src_dst & NIX_FLOW_KEY_TYPE_L3_DST_ONLY) {
+ /* Both SIP + DIP */
+ if (field->bytesm1 == 15) {
+ /* SIP + DIP, 32 bytes */
+ field->bytesm1 = 31;
+ } else {
+ /* Only DIP */
+ field->hdr_offset = 24; /* DIP off */
+ field->bytesm1 = 15; /* DIP,16 bytes */
+ }
+ }
field->ltype_mask = 0xF; /* Match only IPv6 */
break;
case NIX_FLOW_KEY_TYPE_TCP:
@@ -3445,6 +3487,21 @@ static int set_flowkey_fields(struct nix_rx_flowkey_alg *alg, u32 flow_cfg)
field->lid = NPC_LID_LH;
field->bytesm1 = 3; /* Sport + Dport, 4 bytes */
+ if (l3_l4_src_dst & NIX_FLOW_KEY_TYPE_L4_SRC_ONLY)
+ field->bytesm1 = 1; /* SRC, 2 bytes */
+
+ if (l3_l4_src_dst & NIX_FLOW_KEY_TYPE_L4_DST_ONLY) {
+ /* Both SRC + DST */
+ if (field->bytesm1 == 1) {
+ /* SRC + DST, 4 bytes */
+ field->bytesm1 = 3;
+ } else {
+ /* Only DIP */
+ field->hdr_offset = 2; /* DST off */
+ field->bytesm1 = 1; /* DST, 2 bytes */
+ }
+ }
+
/* Enum values for NPC_LID_LD and NPC_LID_LG are same,
* so no need to change the ltype_match, just change
* the lid for inner protocols
--
2.39.0.198.ga38d39a4c5
next prev parent reply other threads:[~2023-06-12 6:05 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-06-12 6:04 [net-next PATCH v2 0/6] RVU NIX AF driver updates Naveen Mamindlapalli
2023-06-12 6:04 ` [net-next PATCH v2 1/6] octeontx2-af: Add devlink option to adjust mcam high prio zone entries Naveen Mamindlapalli
2023-06-12 6:04 ` Naveen Mamindlapalli [this message]
2023-06-12 6:04 ` [net-next PATCH v2 3/6] octeontx2-af: cn10k: Set NIX DWRR MTU for CN10KB silicon Naveen Mamindlapalli
2023-06-12 6:04 ` [net-next PATCH v2 4/6] octeontx2-af: Enable LBK links only when switch mode is on Naveen Mamindlapalli
2023-06-12 6:04 ` [net-next PATCH v2 5/6] octeontx2-af: add option to toggle DROP_RE enable in rx cfg Naveen Mamindlapalli
2023-06-12 6:04 ` [net-next PATCH v2 6/6] octeontx2-af: Set XOFF on other child transmit schedulers during SMQ flush Naveen Mamindlapalli
2023-06-13 9:00 ` [net-next PATCH v2 0/6] RVU NIX AF driver updates patchwork-bot+netdevbpf
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230612060424.1427-3-naveenm@marvell.com \
--to=naveenm@marvell.com \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=gakula@marvell.com \
--cc=kirankumark@marvell.com \
--cc=kuba@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=sgoutham@marvell.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®